0
OREGON
RULE
CO.
1
U.S.A.
2
3
4
5
[OREGON RULE CO.
1 U.S.A.
COLUMBIA LIBRARIES OFFSITE
HEALTH SCIENCES STANDARD
HX640661 50 RD667 Ad5 Primary malignant gr
RECAP
i i ! i 1 I i
:^ ;ii !i 1 1 , !
|
i 1 1=1 Ml! i 1 ! 1 1 |
||
|
lUiuliiluinluiluiUini |
IBBlHiB |
lil!i'>n!!i!il!!!liil!!lll
1^-bGGl
Ad^
Columbia SBmtier^ttp mtljeCitpotBrwgork
CoUege of S^fiv^itims anb ^urgcong
Hibrarp
. J. C- .
'A,^^
^^/cML^U^ ^/' //
/
PRIMARY MALIGNANT GROWTHS
OF THE
LUNGS AND BRONCHI
Digitized by the Internet Archive
in 2010 with funding from
Open Knowledge Commons
http://www.archive.org/details/primarymalignantOOadle
M A r. T (I N A -\
LUNG;:!? AJN ij BiiU.lNl.;H
A PATHOLOGICAL AND CLINICAL STUDY
BY
ADLER. A.M.. M.IK,
:iiiX!di:ji,Qn.
oporiet omnia
ontevipian
LONGMANS, GREE ,
FOURTH AVENUE & 30TH STHEF/i LONDON, BOMBAY,
I'M-
PRIMAEY MALIGNANT GROWTHS
OF THE
LUNGS AND BRONCHI
A PATHOLOGICAL AND CLINICAL STUDY
BY
I. ABLER, A.M., M.D.,
Professor Emeritus at the New York Polyclinic, Consulting Physician to the German, Beth-Israel, Har Moriah, and Peoples Hospitals, and Montefiore Home and Hospital
'Oportet omnia signa contemplari'
LONGMANS, GREEN, AND CO.
FOURTH AVENUE & 30TH STREET, NEW YORK
LONDON, BOMBAY, AND CALCUTTA
1912
COPYRIGHT, 1912, BY I. ABLER, A.M., M.D.
All Bights Reserved
THE'PLIMPTON'PRESS
[ W • D • o] NORWOOD. MASS'U'S -A
TO
MY OLD-TIME TEACHER AND FRIEND
HIS EXCELLENCY
GEH. RAT. PROF. DR. JULIUS ARNOLD
IN HEIDELBERG IN GRATITUDE AND AFFECTION
PEEFACE
T HAD intended that this little monograph on lung ■^ tumors should be handed to Professor Arnold on the occasion of the festival held August 19, 1905, to celebrate the seventieth birthday of the master. The plan as originally conceived could not be carried out, but it is hoped that the delay in bringing out the work may not have been alto- gether valueless in that it made possible a considerable increase in the volume of the material. Great thanks are due to my friends and assistants. Dr. O. Hensel and Dr. O. F. Krehbiel, for their indispensable aid in collecting and sifting the material. I am greatly indebted, as well, to Miss Laura E. Smith, of the Library of the New York Acad- emy of Medicine. I wish also to express my sincere thanks to Dr. H. S. Tienken for her untiring interest, un- selfish devotion, and technical skill in the proper recording and tabulating of the material, and to Dr. A. L. Garbat and to Miss F. H. Fiske for the strenuous work of seeing it through the press. Finally, I would acknowledge my debt to Dr. F. S. Mandlebaum of New York and to Professor S. B. Wolbach of Boston for the very beautiful photographs used here as illustrations.
The author dares to hope for kindly criticism and some renewed interest in the subject.
I. ADLER
New York, Christmas, 1911
vii
CONTENTS
CHAPTER PAGE
I. Introductory 3
Mostly statistical,
II. Introductory {Continued) 13
Remarks on Plan oj Monograph.
III. A Few Historical Notes 16
The Precancerous Influences.
IV. Precancerous Influences (Continued) .... 26
Etiology of Malignant Tumors — Relation of Tuberculosis to Lung Tumors.
V. Pathology 39
Gross Appearance of Lung Tumors — A Minute Study of Sarcoma and Carcinoma.
VI. Pathology (Continued) 55
Histogenesis of Carcinoma — Endothelioma.
VII. Pathology (Continued) 62
Metaplctsia.
VIII. Clinical 68
IX. Clinical (Continued) , . 86
Appendices ■ - . 110
TABLES I. Carcinoma , . 114
II. Sarcoma , . 240
III. Doubtful 278
IV. Miscellaneous 314
LIST OF ILLUSTRATIONS
[The Plates Numbered I to XVI are at the End op the Volume]
Frontispiece. — Section of lung, showing a large tumor originating from the root and destroying the greater part of lung. Communicating cavities and tumor nodules of varying sizes. That portion of lung not infiltrated with tumor, compressed and pushed backward towards the spine.
(From a drawing by H. Becker.)
Plate I. — Transverse section across an infiltrating tumor and adjoining lung. Tumor area sharply defined against lung tissue. Infiltration so dense and complete that only a few vessels and slightly dilated bronchi are visible as remnants of normal structure. (From a drawing by H. Becker.)
Plate II. — Shows destruction of almost entire lung. Pulmonary tissue almost completely replaced by tumor. (From a drawing by H. Becker.)
Plate III. — Section of medullary carcinoma illustrating the occasional impossi- bihty of differentiating between carcinoma and sarcoma.
Plate IV. — Same section as preceding, photographed with high power, ex- hibiting the same indeterminate character.
Plate V. — Section from another portion of the same tumor as shown in Plate IV. In structure and in character of cells plainly suggesting carcinoma.
Plate VI. — Shows section through kidney from same case. An incipient metastatic deposit, consisting of a few genuine epithelial cells just entering Bowman's Capsule, is shown.
Plate VII. — Typical picture of ordinary form of carcinoma. A large alveole is seen, directly injecting a lymph- vessel with cancerous cells.
Plate VIII. — Rapidly proUferating carcinoma, suggesting glandular type. Very little stroma between the alveoles, which latter contain mostly flat and cuboidal epithelial cells.
Plate IX. — Same form of carcinoma. Smaller and more plexiform alveolar structure, more voluminous stroma, injection of lymph-spaces and lymph- vessels from alveoles.
Plate X. — Cancroid with characteristic homy epithelial nests.
xi
xii LIST OF ILLUSTRATIONS
Plate XI. — Cylindrical-celled carcinoma, the epithelium not ciliated. Alveolar structm-e, alveoles varying largely in size. Much mucoid degeneration. Origin from bronchial mucous glands.
Plate XII. — Same type of tumor. To the right, dilated bronchus. In middle bronchial mucous glands, transition to carcinomatous alveoles plainly seen.
Plate XIII. — Similar type of timior. Shows partial destruction of bronchial cartilage and various transitions from normal bronchial mucous glands to cancerous alveoles.
Plate XIV. — Cylindrical-celled carcinoma. Suggestions of alveolar structure. Striking papillary arrangement.
Plate XV. — Besides alveolar structure, shows marked participation of lymph- vessels and spaces in the cancerous proliferation.
Plate XVI. — Shows practically only affection of lymphatic apparatus. Both this plate and the one preceding represent sections taken from tumors which in other localizations show typical carcinomatous structure.
PEIMAEY MALIGNANT GEOWTHS OF THE LUNG
PEIMAEY MALIGNANT GEOWTHS OF THE LUNG
CHAPTER I INTRODUCTORY
IS it worth while to write a monograph on the subject of primary mahgnant tmnors of the Imig? In the com'se of the last two centuries an ever-increasing hteratiu-e has accu- mulated around this subject. But this Hterature is without correlation, much of it buried in dissertations and other out- of-the-way places, and, with but a few notable exceptions, no attempt has been made to study the subject as a whole, either the pathological or the cUnical aspect having been emphasized at the expense of the other, according to the special predilec- tion of the author. On one point, however, there is nearly complete consensus of opinion, and that is that primary mahg- nant neoplasms of the lungs are among the rarest forms of disease. This latter opinion of the extreme rarity of primary timaors has persisted for centuries. Within the last few decades attempts have been made to combat this dogma, but even now the overwhelming majority of medical practitioners rarely, if ever, think of a diagnosis of tumor of the lungs, and the ubiquitous tuberculosis, with its multiform clinical appear- ances and its plastic adaptation to all ages and all conditions of mankind, is ever ready to fxmiish, to all but a very few, a comfortable and satisfactory diagnosis.
Most textbooks hardly notice lung tumors, and if they give the subject any consideration it is, for the most part, insuf- ficient. Thus the well-known and still authoritative textbook on Diseases of the Lungs and Pleurae, including Tuberculosis and Mediastinal Growths, by Sk R. Douglas Powell and
3
4 PRIMARY MALIGNANT GROWTHS OF THE LUNG
P. Horton-Smith Hartley (5th Edition, 1911), while treating at length of thoracic tumors and of mediastinal tumors, etc., has scarcely more than one page to cover the entire subject of carcinoma and sarcoma of the lungs. The excellent book of A. Frankel ^ and the admirable chapters on carcinoma of the lungs in the latest edition of Wolff, ^ as well as a few other publications,^ attempt a more comprehensive presentation of this type of tumor, but they seldom get into the hands of the medical public at large, and so it happens that the general practitioner is not in a position to diagnosticate a primary lung tumor as often as might be, and the belief in the extreme rarity of these cases is still maintained. To add to these difficulties, even the diagnoses made on the autopsy table are not always reliable. There are still careless or insuf- ficiently trained persons called upon to do this rather dehcate work. It may happen also that the most careful and search- ing autopsy will not furnish the true diagnosis until a thorough microscopical examination has been made. Take for example the case of Walter Kretschmar; * also of Morelli,^ This latter case is remarkable for a number of unusual features: the youth of the patient, — a female aged twenty-eight, — the sudden onset after cold, with fever and cough, the clinical symptoms of a pneumonic consoUdation in right base with pleural effusion and endocarditis. The sputum showed diplo- cocci. On autopsy both lungs showed white nodules, corre- sponding to blood vessels, and connective tissue strands not infrequently seen after pneumonic processes. No tumor could be recognized, and only upon microscopic examination were nests of epithelial cells discovered in the lymph spaces
1 Spezielle Pathologie u. Therapie der Lungenkrankheiten, 1904.
2 Die Lehre von der Krebskrankheit, Vol. II, pp. 803 ff., Jena, 1911.
' Credit must be given here to Alfred v. Sokolowski, Klinik der Brustkrank- heiten, Vol. I, Berlin, 1906, and his study of primary malignant and non-malig- nant neoplasms of the bronchi and lungs. He seems to consider bronchial carcinoma extremely rare, — much more rare than primary tumors of the lung. He has a chapter of about fifteen pages devoted to lung tumors, citing several cases of his own experience. He goes rather quickly over the pathology and diagnosis of carcinoma and in the same way hurries over sarcoma without bringing in anything notably new.
^ tjber das primare Bronchial- imd Lungencarcinom, Diss. Leipzig, 1904.
B Table I, No. 201.
INTRODUCTORY 5
of the fibrous tissue, and epithelial clusters in the alveoles and in the alveolar septa.
Furthermore, v. Hansemann ^ relates that in his experi- ence at the Friedrichshain Hospital there were 711 carcino- mata out of 7790 autopsies, of which 156, or 21.94%, were not diagnosticated during life, not even as tumors. Among these 156 cases there were sixteen bronchial and pul- monary tumors. Is it not somewhat humiliating to realize that the difficulties of diagnosis are still so great as to pre- vent the best and most experienced medical men, with all the advantages of a large hospital, from discovering almost one-fifth of all the carcinomata that come before them? If these figures hold good generally, about one-fifth more car- cinoma cases should be added to our ordinary statistics. Another important addition to the difficulties to be contended with lies in the fact that in many countries, as for example our own, justly claiming an advanced stage of civilization, the overwhelmingly great majority of the dead are not sub- jected to any post-mortem examination, and the death certificates on which burial permits are officially given are often ludicrously insufficient. For this reason the United States Census is entirely useless for our purposes. As an example of the misleading diagnoses and insufficient observa- tion which hamper one in getting up the literature of this subject, look up the following: Two Cases of Melanotic Tumors in the Lungs.^ Reliable autopsies, in the majority of cases, there are not, and many autopsy notes that have been recorded are so insufficient in their data and descriptions that a conclusive opinion on the case cannot be formed. The same applies to the clinical notes. It is therefore impossible to say, from the figures given by the United States Census concerning causes of death, how many persons mentioned as having died from tuberculosis, pneumonia, or kindred diseases, may not really have died from lung tumors.
Considering all this, it seems primarily necessary to
^ Riechelmann, Eine Krebsstatistik vom pathologisch-anatomischen Stand- punkt, Berl. Klin. Woch., 1902, N. 31 and 32, pp. 728 ff. 2 Journal A. M. A., 1888, p. 53.
6 PRIMARY MALIGNANT GROWTHS OF THE LUNG
procure enlightenment on the question : Are malignant tumors of the lung as rare as has been supposed? And if they are not so rare, is their more frequent occurrence due to a supposed general increase in the incidence of malignant growths? WilUams/ an enthusiastic exponent of the increase of car- cinoma as a whole and the corresponding decrease of tuberculosis, supports his view with a great mass of statis- tical figures, of which some few are quoted here.
1840
Incidence in England and Wales 1905
2786, a proportion to total number of deaths of 1:129, or 177 per million living.
30221, a proportion to total number of deaths of 1:17, or 885 per million living.
As to Newsholme's contention ^ that the registered increase is only apparent, being actually due to improved methods of diagnosis and death certification, WiUiams's answer is that (1) the uniformity in increase is too marked to be due to im- proved diagnosis, and (2) the very improvements cited have also caused subtractions from the cancer total, since many diseases formerly erroneously called cancer are now given their true names. Nencki is quoted in this connection * as giving the increase in cancer death-rate in Switzerland from 114 in 1889 to 132 in 1898 (per 100,000 living). WilHams gives the following figures for other countries:
Deaths fbom Cancer
Paris, France 1865 84
1900 120
Germany 1872 59
1900 71
Berlin 1870-1882 57
1899 109
Italy 1880 21
1905 58
|
United States |
||
|
(per 100,000 Uving) |
1850.... |
.... 9 |
|
1900 . . . |
...43 |
|
|
New York |
1864 . . . |
...32 |
|
1900 . . . |
...63 |
|
|
Boston |
1863 . . . |
...28 |
|
1903 . . . |
...85 |
|
|
New Orleans |
1864 . . . |
...15 |
|
1903 . . . |
...82 |
|
|
San Francisco |
1856 . . . |
...16 |
|
1900 . . . |
...112 |
* Natural History of Cancer, New York, 1908. » Proceedings of the Royal Society, 1893, Vol. LFV, p. 209. ' Die Frequenz und Verteilung des Krebses in der Schweiz, etc., Zeitschr. f. schw. Statistik, 1900, Vol. II, p. 332.
INTRODUCTORY 7
Other important statistical work to be consulted is that of Robert Behla/ the great standard work, in four volumes, of Juliusburger,2 and the work of Newsholme.^ Looking carefully over these statistics, it is the writer's opinion that the statistics of Williams, as well as all statistical material thus far collected, with a great deal of care and labor, have not succeeded in proving conclusively that there is a real increase in the incidence of cancer and a corresponding decrease in the incidence of tuberculosis. The fact may turn out to be so, but at this writing can by no means be considered as proven. The only figiu-es which in the course of time will give us the means of definitely solving problems such as this will be those obtained from hospitals, where the material is more uniform, where the best modem methods of observation and diagnosis are applied, and where finally the autopsies and microscopical examinations are done with the utmost care. Reports of life insurance officers, statis- tics taken from books of registrars and recorders, where only the causes of death are mentioned, cannot be effectively utilized.
It has been shown, especially by the researches of Behla just quoted, that some sort of influence of occupation or trade may possibly be considered a factor in the incidence of carcinoma. If so, this factor is of very slight significance and may, at least for the study of lung tumors, be entirely disregarded.
It is the conviction of the writer, and he shares this belief with many others, — the mention of whose names and criti- cism of whose work need not be entered upon here, — that there is no absolute increase in the incidence of carcinoma. Nevertheless, the incidence of malignant neoplasms of the lungs seems to show a decided increase. It has been stated that statistical research in this direction is beset with many difficulties. It may be hoped that in the course of a few
^ Krebs und Tuberkulose in beruflicher Beziehung vom Standpunkte der vergleichenden internationalen Statistik, Berlin, 1910.
^ Die Krankheits- und Sterblichkeitsverhaltnisse in der Ortskrankenkasse fiir Leipzig und Umgegend.
' The Statistics of Cancer, The Practitioner, April, 1899.
8 PRIMARY MALIGNANT GROWTHS OF THE LUNG
years accurate and reliable figures will be available. In the meantime, however, the following table, founded on figures collected by Karrenstein ^ and considerably amended and enlarged, will at least serve to show, not the causes, but the fact of the apparent increase. It is very significant that in
Primart Carcinoma of the Lungs and Bronchi
|
I |
II |
in |
IV |
V |
VI |
VII |
|
Time |
Place |
% of all |
Total |
% of aU |
Total |
Author |
|
Carci- |
No. |
Autop- |
No. of |
|||
|
noma |
Carci- noma Cases |
sies |
Autop- sies |
|||
|
1. 1852-67 |
Stadtkrankenhaus, Dresden |
0.91 |
8716 |
Reinhardt^ |
||
|
2. 1852-1908 |
Patholog. Institut, Wurzburg |
15 or 0.93 |
1607 |
Fockler ' |
||
|
3. 1854-85 |
Stadtkrankenhaus, Mtinchen |
8 cases |
0.065 |
12307 |
Fuchs* |
|
|
4. 1870-88 |
Patholog. Institut der Universit . Kolozsvar |
0 |
145 |
Buday ^ |
||
|
5. 1872-89 |
Patholog. Institut, Bern |
2 0.42 |
474 |
0.059 |
3363 |
C. Miillers |
|
6. 1872-98 |
Reichsgesundheits- amt, Hamburg |
84 0.70 |
11930 |
0.02 |
336486 |
Reiche' |
|
7. 1873-87 |
Patholog. Institut, Kiel |
0 |
Danielsen ^ |
|||
|
8. 1877-84 |
Stadtkrankenhaus, Dresden |
9 cases |
0.22 |
4712 |
Wolfs |
|
|
9. 1881-94 |
Patholog. Institut, Breslau |
1.83 |
870 |
9246 |
Passler^" |
|
|
10. 1885-94 |
Stadtkrankenhaus, Dresden |
31 cases |
0.43 |
7728 |
Wolf 11 |
1 Charit^-Annalen, Berlin, 1908.
2 Reinhardt, Der primare Lungenkrebs, Arch. f. Heilkunde, XIX, 1878.-2. ^ Fockler, Krebsstatistik nach den Befunden des patholog. Instituts zu
Wurzburg, Diss. Wiirzburg, 1909.
^ Fiichs, Beitr. zur Kenntnis der Geschwiilstbildungen in der Lunge, Diss. Miinchen, 1886.
^ Buday, Statistik der im patholog. -anatom. Institut der Universitat Koloz- svar usw. Zeitschr. f. Krebsforschung, Vol. VI, S. 7.
' Miiller, C, Beitrag zur Statistik der malignen Tumoren, Diss. Bern, 1890.
^ Reiche, Beitrage zur Statistik des Carcinoms, Deut. Med. Woch., 1900, N. 7, p. 120 ff.
8 Danielsen, Quoted from Schlereth, 2 FaUe von primarem Lungenkrebs, Diss. Kiel, 1888.
9 Wolf, Fortschritte der Medizin, 1895. 10 Passler, s. S. 315, No. 5.
" Wolf, loc. cit.
INTRODLXTORY 9
Primary Carcinoma of the Lungs and Bronchi — Continued
|
I |
II |
III |
IV |
V |
VI |
VII |
|
Time |
Place |
%■ of all |
Total |
%of aU |
Total |
Author |
|
Carci- |
No. |
Autop- |
No. of |
|||
|
noma |
Carci- noma Cases |
sies |
Autop- sies |
|||
|
11. 1886-96 |
Krankenhaus, Munchen |
9 1.2 |
706 |
0.10 |
8727 |
Periitzi |
|
12. 1887-1906 |
Patholog. Institut, Wien |
68 |
0.17 |
40000 |
Haberfeld2 |
|
|
13. 1888-97 |
Patholog. Institut, Greifswald |
1.78 |
Kaminski^ |
|||
|
14. 1888-1905 |
Patholog. Institut, Universit. Kolozsvar |
10 4.5 |
221 |
Buday * |
||
|
15. 1895-1901 |
Friedrichshain, Berlin |
711 |
7790 |
Riechelmann ^ |
||
|
16. 1899-1903 |
Patholog. Lab. Lubarsch, Posen |
3 1.2 |
159 |
0.17 |
1741 |
Sehrte |
|
17. Vor 1900 |
Patholog. Institut am Urban-BerUn |
4 |
100 |
0.4 |
Feilchenfeldt^ |
|
|
18. 1899-1904 |
Patholog. Institut am Urban-Berlin |
0.6 |
Benda^ |
|||
|
19. Zeitraum |
Patholog. Institut, |
20 |
Rieck 9 |
|||
|
V. 10 Jahr. |
Univ. Miinchen |
1.92 |
||||
|
20. |
6 1.3 |
447 |
Lebertio |
|||
|
21. 1900 |
Patholog. Institut, Charit^Berlin |
2.91 3 cases |
103 |
0.23 |
1300 |
Karrenstein i^ |
|
22. 1900-05 |
Urban-Berlin |
31 0.61 |
496 |
0.6 |
5002 |
Redlichi2 |
|
23. 1901 |
Patholog. Institut, Charity-Berlin |
8.86 7 cases |
79 |
0.53 |
1310 |
Karrenstein" |
^ Perutz, Zur Histogenesis des primaren Lungenkarzinoms, Diss. Miinchen, 1897.
^Haberfeld, Carcinom des Magens, der Gallenblase und Bronchien. Z'tschrift f. Krebsforsch., Vol. VII, I. Fasc, p. 204.
3 Kaminski, s. S. 315, No. 6.
* Buday, loc. cit.
^ Riechelmann, Eine Krebsstatistik von path.-anatom. Standpunkt, Berl. klin. Woch., 1902, N. 31 and 32, pp. 728 ff.
^ Sehrt, Beitrage zur Kenntnis des primaren Lungenkarzinoms, Diss. Leip- zig, 1904.
7 Feilchenfeldt, Quoted from Benda, Deut. Med. Woch., 1904, S. 1454. Beitrage zur Statistik und Kasuistik des Karzinoms, Diss. Leipzig, 1901 (after Redlich).
8 Benda, loc. cit., S. 1453.
5 Rieck, Krebsstathstik nach den Befunden des patholog. Instituts zu Miinchen, Diss. Munchen, 1904.
1" Lebert, Traits pratique des Maladies cancereuses.
" Karrenstein, Charite-Annalen, XXXII Jahrg., Berlin, 1908.
12 Redlich, Die Sektions-Statistik des Carcinoms, etc., am Stadt-Kranken- haus am Urban, 1900-1905, Diss. Berlin, 1907.
10 PRIMARY MALIGNANT GROWTHS OF THE LUNG
Primary Carcinoma of the Lungs and Bronchi — Continued
|
I |
II |
III |
IV |
V |
VI |
VII |
|
Time |
Place |
% of all |
Total |
% of aU |
Total |
Author |
|
Carci- |
No. |
Autop- |
No. of |
|||
|
noma |
Carci- noma Cases |
sies |
Autop- sies |
|||
|
24. 1902 |
Patholog. Institut, Charity-Berlin |
3.23 3 cases |
93 |
0.31 |
999 |
Karrenstein^ |
|
25. 1903 |
Patholog. Institut, Charite-BerUn |
3.19 3 cases |
94 |
0.24 |
1272 |
Karrenstein ^ |
|
26. 1904 |
Patholog. Institut, Charite-Berlin |
2.67 4 cases |
150 |
0.28 |
1399 |
Karrenstein ^ |
|
27. 1905 |
Patholog. Institut, Charite-Berlin |
0.71 1 case |
140 |
0.08 |
1313 |
Karrenstein^ |
|
28. 1906 |
Patholog. Institut, Charite-Berlin |
4.84 6 cases |
124 |
0.46 |
1319 |
Karrenstein^ |
|
29. 1906-08 |
Krankenhaus, r. d. I., Manchen |
174 |
0.18 |
945 |
Forstner^ |
|
|
30. 1907 |
Patholog. Institut, Charite-Berlin |
3.31 5 cases |
151 |
0.37 |
1360 |
Karrenstein 1 |
|
31. 1908 |
Stadtkrankanstalten, Hamburg |
11 1.2 |
920 |
Korber' |
||
|
32. 1908-09 |
Patholog. Institut Krankenhaus, Miinchen |
1.8 |
212 |
0.29 |
1371 |
Nobiling^ |
|
33. |
Basel |
1.76 |
Kauf mann ^ |
|||
|
34. 1910-11 |
Charity- Annalen, Berlin |
0.76 |
141 |
0.05 |
185 |
Orth6 |
1900 the Pathological Institute of the Charite in Berlin recorded only three cases of lung tumor, while in 1906 and 1907 five and six cases respectively, were recorded. It is more significant still when the reports of the Pathological Institute of Kolozsvar from 1870 to 1880 and from 1888 to 1905 respectively, are compared. It is to be remembered that this table is made up mainly from records of pathological laboratories of fairly high standing.
There seems hardly room for doubt that the increase in the percentage of lung tumors is to be attributed mainly to
^ Karrenstein, Charitl-Annalen, XXXII Jahrg., Berlin, 1908.
2 Forstner, tjber maligne Tumoren, Diss. Miinchen, 1908.
' Korber, Die Ergebnisse der Hamburgischen Krebsforschung im Jahre 1908. Mitt. Hamburgischen Staatskrankenanstalten, Vol. IX, Supp., 1908.
* Nobihng, Z'tschrift f . Krebsforsch. patholog. Institut Krankenhaus, Miinchen, r. d. I., 1908-1909.
' Kaufmann, Lehrbuch der Spec. Path. Anatomie, Basel, 1909.
6 Orth, Charit6-Annalen, Berlin, XXXV Jahrg., 1911.
INTRODUCTORY 11
the increased attention paid to these types of tumor and the greater care and more extensive microscopic investigation with which autopsies are carried out at present. As early as 1837, Stokes ^ had aheady remarked that in his experience lung tumors are by no means as rare, either in England or in Ireland, as was generally assumed, and Boyd ^ even goes so far as to assert that primary cancer is more frequent in lungs than secondary cancer, an assertion which he explains as follows : ''A case of maUgnant deposit in the bronchial glands, infiltrating the lung, ending in ulceration and the formation of cavities, is frequently set down as one of hopeless phthisis, a post-mortem on which would be of no interest, and all record of the frequency of the disease is in consequence entirely lost." This utterance of Boyd's is probably some- what of an exaggeration, for while it has just been shown that the behef in the extreme rarity of lung tumors, a lusus naturae, as it were, can no longer be maintained, it must be conceded that these tumors belong to the class of rarer neoplasms and their incidence is out of all proportion to the frequency of occurrence of other malignant neoplasms, as for example of the female breast or the stomach.
Seeing, thus, that lung tumors are to be reckoned with more often than was formerly believed, it is to be expected that nimierous problems, both pathological and clinical, will present themselves. Besides these problems of purely theo- retical interest to the pathologist and the clinician, there is the great importance to the patient of a correct diagnosis. It cannot be a matter of indifference to the unfortunate sufferer whether his case be diagnosticated as tuberculosis or as tumor. If tuberculosis, he will be sent from one climate and one sanitarium to another, he and his family possibly deluded with false hopes, until finally secondary symptoms have cleared up the case and death has brought relief. The grave prognosis which is an integral part of the diagnosis of tumor may be of paramount importance to the patient as well as to his relatives. At all events, so much is certain, that if
1 Diseases of the Chest, London, 1837.
2 Table I, No. 46.
12 PRIMARY MALIGNANT GROWTHS OF THE LUNG
the diagnosis of lung tumors is to be developed so as to render it more precise, and if any reasonable attempt is to be made to convert the present desperate prognosis into one less hopeless, this great result can only be achieved if the internist shall work hand in hand and shoulder to shoulder with the surgeon. The internist must be able to furnish as early and as accurate a diagnosis as possible, so that the surgeon under favorable conditions may develop his technique as early as possible. With these few introductory words, the initial question, it is dared to hope, is answered affirmatively.
CHAPTER II
IN TROD UCTOR Y {Continued)
IN undertaking to write this monograph, it is proposed to present the subject and the problems connected therewith in as comprehensive and at the same time as concise a manner as possible. Not only carcinoma, but the other malignant tumors of the lung are to be presented, both from a broad pathological, as well as from a clinical point of view.
As the first step toward the accomplishment of this end, it was found necessary to collect a very large material from the literature. Thus far, but comparatively few cases had been picked up. Passler,^ after much sifting, managed to collect about seventy-four cases of undoubted primary car- cinoma of the lungs. This was in 1896, just fifteen years ago. The latest publication ^ casually remarks that about one hundred cases may now be found in literature. The difficulties of collecting cases in point have already been hinted at. It is extremely trying to delve into all sorts of doctor-dissertations, obscure and forgotten publications of all kinds and in all languages, to be frequently rewarded by finding that, after all, the case is secondary, or is not a case in point at all, or that no autopsy was made, or that no micro- scopic examination was reported. Again, no clinical history is given, and the pathological diagnosis, though modern and very good, is not sufficiently supported by clinical observa- tions. The collection of cases from modern times has been simplified by the introduction of the microscope into pathol- ogy and the nomenclature of tumors based on microscopic
lyirch. Arch., Vol. 145, 1896, p. 191.
2 Edward Boecker, Zur Kenntnis der primaren Lungenkarzinome, Dies. Gottingen, Berlin, 1910.
13
14 PRIMARY MALIGNANT GROWTHS OF THE LUNG
study, which latter, though not fulfilling all demands, is most helpful. But even within the last two years, reports have been pubhshed where there is no autopsy at all, or one that is very insufficient, and the microscopic examination is either absent or summarized in such general terms as "simple carcinoma," etc. Nevertheless, though it has taken several years in the compiling, 374 cases of carcinoma have been collected. It was thought best to make full abstracts wherever possible, so that the principal data of each case, both clinical and pathological, may be at the disposal of the reader, enabling him to use his own judgment and form his own deductions. The same has been done for sarcoma, though the latter is infinitely more difi&cult to get at than carcinoma, — not only because sarcoma is so much rarer, as will be seen, but because very many cases are published without sufficient autopsy, and even if autopsied the almost intolerable confusion in the nomenclature makes the diagnosis from the printed case wellnigh impossible.
A third collection has been made which contains cases desig- nated as doubtful, though many of them may be authentic and valuable. They have been classed as doubtful for various reasons, sometimes because the autopsy was lacking, though the clinical observations pointed almost with certainty to a tumor diagnosis, or it was impossible to decide whether the case was one of carcinoma or sarcoma, etc.
A few other cases have been assembled which, properly speaking, do not belong to the subject in hand, but which may in their symptoms during life so closely resemble primary growths of the lung that it was thought wise to place them here for warning and for comparison.
The reader should well understand that no claim is made for absolute completeness. Many cases were not taken into our collection either because they were not obtainable, or were written in a language that could not be readily trans- lated, or for other reasons. Besides this, too, it was imprac- ticable to continue collecting material indefinitely, and since the collection of material has been discontinued numerous cases have been published, which could not appear in the present collection. It may be stated also that, with the
INTRODUCTORY (Continued) 15
exception of but comparatively few, the references were read and excerpted personally. This rather bulky collection is printed in the form of tables, the first and largest being of carcinoma cases; the second, sarcoma; the third, doubtful; and the fourth, a few miscellaneous cases.
CHAPTER III
A FEW HISTORICAL NOTES
Precancerous Influences
OUR knowledge of lung tinnors dates from comparatively recent times, and the history of its development can be sketched in a very few words. It may aptly be divided into several periods. In the first and longest period, lung tumors were absolutely unknown. This period includes all of ancient and mediaeval medicine until Morgagni ^ (1682-1772) laid the foundations of pathological anatomy. It is most interesting and significant that Morgagni himself was prob- ably the first to publish the results of several autopsies on lungs that might be diagnosticated as cancerous, and were so interpreted by him. It is probable that the first of the cases which he published as cancer of the lungs was really a case of primary lung tumor. In this case he describes the disease of a man sixty years old, which was accompanied by cough and copious expectoration of a yellowish, rather crude material, rarely, but then distinctly, stained by streaks of blood. At autopsy the lung was foimd extremely hard, adhesions to pleura and mediastina, and nothing else but an "ulcus cancrosiun" in the right lung.^ The oft-quoted observations of Lieutaud ^ deal probably with tuberculosis or diseased pleura, and not with tumor. The cases mentioned by Van Swieten * must also be considered extremely doubtful.
1 De Sedibus et Causis Morborum per Anatomen indigatis.
2 Loc. cit.
' Historia anatomico-medica, etc., Paris, 1767, Lib. II. * Comment, ad Boerhaavi Aphorism, Vol. II, 1747. 16
A FEW HISTORICAL NOTES 17
There are a number of French authors about this time ^ who pubhshed cases as cancerous that cannot be distinguished with certainty from tuberculosis. G. L. Bayle^ pubhshed thi'ee cases which he had carefully studied clinically and equally carefully after death, and he is the author of the phrase "phthisie cancereuse" which caused so much discus- sion. The first case he reports may possibly be a primary- tumor, although this is doubtful. The second case is cer- tainly secondary after amputation of the arm. The third case was that of a man seventy-two years old, in which there were found at autopsy, at the root of the lung, shining white encephaloid cancerous masses, which were associated with masses of tuberculosis. It is unnecessary to go into all the clinical and pathological details and theories on which Bayle bases his conclusions. There is some merit in his insistence that cancer and tuberculosis may exist together, although the tubercles, according to him, are the effect of an acid, and cancer the effect of an alkali. No clear idea, however, can be obtained of what he means by cancer and what by tubercu- losis, and it consequently happened altogether too frequently that his followers accepted true tubercular cavities as can- cerous, and vice versa, so that finally great confusion arose as between tubercular phthisis and cancerous phthisis. His contention that cancer of the lungs may exist for a very long time without any symptoms has been corroborated by modem medicine. On the other hand, he makes no distinction between primary and secondary tumor.
Besides the French, a number of German authors have worked on fines similar to those of Bayle, and though the name "phthisie cancereuse" could not maintain itself for a very long period, the name "fungus hsematodes," or simply "fungus of the lung," — especially among German writers, — was used for all pulmonary neoplasms that bore a suspicion of mafignancy. Those seeking further information of these
1 Le Dran, Mem. de I'Acad. royale de Chir., Vol. Ill, p. 28, Obs. 22. Also J. F. Senaux, fils.
2 Journal de Medicine, Tome 73, 1787. Also Recherches sur la Phthisie pulmonaire, Paris, 1810, p. 299. Also Diet, de Science m6d., Paris, 1810.
3
18 PRIMARY MALIGNANT GROWTHS OF THE LUNG
historical questions are referred to the EngUsh classics, espe- cially Stokes/ Graves,^ and Walshe;^ and also to the, for that period, very complete and thorough works of Reinhold Kohler,^ and among modern authors, J. Wolff. ^
With Bayle and his followers ends the second period, and we enter upon the third, characterized by the study of lung tumors by purely clinical methods, reenforced by gross pathological anatomy. This period is introduced by Laennec, the author of TAuscultation Mediate, who, with his great authority and keen mind, took up the combat against Bayle and his after all not very progressive theories of the "phthisie cancereuse" and successfully differentiated the carcinoma of the lungs, whether primary or secondary, from any form of phthisical process, even though cavities should be found coimected with the tumor. He described tumor of the lung in the clearest terms, under the designation ''encephaloid." The use of this term, appUed promiscuously to all sorts of tumors, caused considerable confusion imtil Virchow worked out a rational classification.
Since the time of Laennec, his lifework, the practice and perfection of the methods of auscultation and percussion, has been assiduously continued and by these means a compara- tively large number of lung tumors has been diagnosticated and reported. For a long time the necessary distinction between primary and secondary tumors was not upheld, and a number of cases were insufficiently observed and carelessly reported, but still progress in the diagnosis of primary tumor of the lungs was certainly made. J. Bell ^ is said to have been the first to diagnosticate with certainty a primary tumor, which was undoubtedly sarcoma of the lung. The real founder of this school is Stokes, who, together with Graves, Walshe, Hughes, and others, laid the foundations of our present clinical and pathological knowledge of primary lung
' Loc. cit.
2 Clinical Lectures on the Practice of Medicine, London, New Sydenham Soc, 2d Ed., Dublin, 1848, by J. Moore Neligan.
^ A Practical Treatise on Diseases of the Lung, etc., 4th Ed., London, 1871.
* tJber den Lungenkrebs, Diss. Tubingen, 1847, and Die Krebs- und Schein- krebskrankheit des Menschen, Stuttgart, 1853.
6 Loc. cit. 6 Table II, No. 3.
A FEW HISTORICAL NOTES 19
tumors. Following upon this period of purely clinical and gross pathological observation, there comes the time when, after the fundamental discovery of Schwann, histology be- comes the main factor in pathological research. After the great work of Rokitansky,^ in gathering together a very large material which led to a general cleaning-up and reclassifying of pathological anatomy, it is above all the name and work of Virchow that dominate this entire epoch. He was the first to demand that medicine be lifted out of a maze of hypotheses and more or less plausible theories to become one of the natural sciences, based on critical observation and experiment. The "cellular pathology," with its battle-cry of "Omnis cellula e cellula," exercised great influence on the study of tumors. The entire onkology was taken up again and rearranged in the light of the fact that every cell origi- nated, not from blastema, not from plastic lymph, not from diatheses or other exogenic processes, but from cells alone. 2 The present time is still a part of this period, and the study of lung tumors must be continued along these lines.
Notwithstanding the great amount of work that, as has just been shown, has been done and is still going on, Williams^ is probably correct when he makes the somewhat brusque statement that "it is necessary at the outset to refer thus pointedly to the crudeness and immaturity of medical knowl- edge, because nowhere do these qualities find more striking exemplification than in the terrible welter of disjointed facts and contradictory hypotheses that constitute such a large part of modern Humor science.'" There cannot be any intention to discuss here the multitude of questions and prob- lems concerning the etiology and the true natm^e of malignant growths in general. The many questions of fundamental import, the attempts into the field of etiology, the innumerable
^ Lehrbuch der pathol. Anatomie, 1844.
2 Thiersch, Der Epithelialkrebs namentlich der Haut, Leipzig, 1865; Wal- deyer, Uber den Krebs, Volkmanns Samml., 1873, No. 33; Bard, La Specificity cellulaire et I'Histologie chez I'embryo, Arch, de Phys. normal, et path., 3 Ser., 7, p. 406, the author of the aphorism: "Omnis cellula e cellula ejuedem generis." ' Loc. cit.
20 PRIMARY MALIGNANT GROWTHS OF THE LUNG
theories, and above all, the enormous experimental work that has been done within recent years, — all this is obviously beyond the scope of this little monograph, which is to be devoted solely to the study of lung tumors.
Nearly all the types of malignant neoplasms that occur in other parts of the body are also to be found among the primary growths of the lung, but before taking up the direct study of these tumors, some attention should be given to the conditions which have long been called "predisposing causes," but which latterly and more significantly are termed '^ precancerous conditions and affections."^
First, the influence of race on carcinoma. According to the latest statistics, race and geographical distribution seem to have a decided influence on the incidence of malignant growths.2 In the very thorough work of Dr. Levin,^ sufficient proof appears to be found that there is less cancer among the American Indians and American negroes than among the whites. Tuberculosis decimates the American Indians, while they are almost immune to cancer. This seems to contradict the statistical conclusions arrived at by Behla.^ Levin notes, too, that it is usually sarcoma or epithelioma of the different external parts of the body, which are neces- sarily more exposed to mechanical irritations, that affect the primitive races. In civilized nations there is a prevalence of carcinoma of the internal, parenchymatous organs. The following sentence, quoted from Levin, is important: "Thus the conclusion is forced on one's mind that, while every human being may carry within himself the X which may develop into cancer, it is the modern civilization and the conditions created by it that give rise to the mediate causes which produce the disease." The facts, indeed, at present available, support the conclusion that the white races,
1 All these data and figures have evidently been worked out principally for carcinoma, sarcoma being brought in now and then incidentally only, probably because of its rarity, possibly because no difference was made between the two.
2 Carl Lewin, Die Bosartigen Geschwiilste, Leipzig, 1909. Also Williams, loc. cit.
' I. Levin, Cancer among the American Indians, Zeitschr. f. Krebsforsch., Vol. X, Heft II, 1911. ^ Loc. cit.
A FEW HISTORICAL NOTES 21
especially in Europe and the United States, can claim the greatest mortality from malignant growths, and there is only China, perhaps, that can compete with them in this respect. It is reasonable to suppose that this applies also to lung tumors, though there are no special statistics.
Next, the question of heredity. This has always been considered a very potent factor in the etiology of malignant neoplasms in general. Josefson and Pfannenstill ^ have already noticed, however, that this does not apply to lung tumors. They have found only one case of accredited hered- ity among their seventy cases. According to Table I, in 290 cases of carcinoma heredity is not mentioned. As many of these cases are very superficially reported, and as in many others no clinical history is given, but the cases are simply introduced as pathological specimens, it is likely that among these 290 cases there may be many where the factor of hered- ity was simply overlooked. In twelve cases only it was posi- tively stated that there was a hereditary strain of cancer in the family, and in sixty-eight instances it was asserted that no hereditary strain could be discovered. According to the German Sammelforschung, in 9% of the males and 10.3% of the females hereditary predisposition for cancer was found.^ The experimental study of tumors has thus far not furnished any decided proof of the value of heredity as a causal factor, and Bashford is inclined to deny its influence altogether. It follows, — though the figures are very uncertain, — that the incidence of malignant growths of the lungs does not appear to be seriously affected by hereditary strain.
The influence of sex. M. Askanazy* maintains that there is a distinct connection between premature sexual develop- ment and the development of malignant growths. Among tumors of other kinds he quotes also Linser,* who reported the case of a boy thirteen years of age with a complete
1 Primary Cancer of Lungs, Nov. Med. Arch., Stockholm, 1897, N. F. VIII, Festband, Axel Key; and Lubarsch and Ostertag, Ergebnisse, Wiesbaden, 1904, Vol. VIII, 1902.
2 Quoted from Lewin, loc. cit.
3 t)ber Sexuelle Friihreife, Zeitschr. f. Krebsforsch., Vol. X, Heft. Ill, 1910.
4 Virch. Archiv., 1899, Vol. 157, S. 281.
22 PRIMARY MALIGNANT GROWTHS OF THE LUNG
development of hair such as is seen after development of puberty. He died of a tumor in the left pleural cavity and mediastinmn which, on examination, showed absence of elastic fibres, in stroma, no cihated epitheha, the epithehal cells in certain places still stratified. The natural history of these evidently congenital tumors is as yet entirely obscure.
It has always been maintained that males are by far more frequently subject to lung tumors than females. Tables I and II corroborate this. Among the 374 cases of carcinoma of the limgs, there are 269 males, or 71.9%; ninety-three females, or 24.8%; twelve in which the sex is not stated. In the same way, among ninety-four sarcoma cases, sixty- three, or 67%, are males; twenty-eight, or 29.7%, females; three where sex is not stated.' The domestic life led by women, with their consequent retirement and immunity from the irritations and traumatisms which must be frequent in the more unprotected life of men (the abuse of tobacco and alcohol, the many trades and vocations which are accom- panied by irritations of the respiratory organs, etc.) has been adduced in explanation of this fact. The entire subject is not yet ready for final judgment.
The age of the patient. It is indisputable that age has a certain influence upon the incidence of both carcinoma and sarcoma. Statistics seem to show that carcinoma, roughly speaking, is a disease of that period of life which follows puberty after its completion, while, on the other hand, sar- coma as a rule is a disease of the earher years of hfe. But there are exceptions, and no age is entirely exempt from either type of tumor. The following figures, gathered from Tables I and II, clearly illustrate this. It is evident from this that the majority of carcinoma cases hes beyond the age of forty and attains its maximum between the ages of fifty and sixty. Descending slowly there are still two cases remaining between eighty and ninety, while the majority of sarcoma cases lies below the age of forty, cUmbing up slowly from the decade between ten and twenty, reaching the maximum between thirty and forty, declining again, slowly, and there are still five cases between seventy and eighty.
A FEW HISTORICAL NOTES 23
The first decade, from birth to ten years, seems to be kommie from carcinoma (without counting, of course, the few cases of congenital tumor).
|
Carcinoma |
Sarcoma |
||
|
Age not stated |
18 |
Age not stated |
9 |
|
1-10 |
0 |
1-10 |
6 |
|
10-20 |
6 |
10-20 |
12 |
|
20-30 |
10 |
20-30 |
14 |
|
30-40 |
30 |
30-40 |
19 |
|
40-50 |
78 |
40-50 |
14 |
|
50-60 |
113 |
50-60 |
12 |
|
60-70 |
94 |
60-70 |
3 |
|
70-80 |
23 |
70-80 |
5 |
|
80-90 |
2 374 |
94 |
These figures tally satisfactorily with the age tables given by many authors, for instance Fuchs.^
The question of the influence of age upon the incidence of maUgnant neoplasms is one that is intimately connected with certain problems that have of late years been thor- oughly studied and widely discussed, — the problems of growth and of senility in their physiological and pathological bearings. The older theories, such as those of Thiersch ^ and others, that as the body grows older the interstitial tissue undergoes a change, the equilibrium between this and the epitheUum is impaired, in consequence of which the epithelial tissue proUferates and tends to form carcinoma, while, on the other hand, in youth the connective tissue group is apt to overstep the bounds set to it and thus sarcoma and similar tumors may be formed — these theories no longer hold good. It has just been shown that no age is absolutely immune from the formation of neoplasms and that even in intra-uterine life tumors of all kinds may be developed. These facts seem to lead to the unavoidable conclusion that deeper and more complex principles are involved. It is altogether foreign to the purpose of this study, and would require a book by itseK, to go into details concerning the modem theories of growth and senility. It will suffice to say
^ Beitrage zur Kasuistik des prim. Lungencarzinoms, Diss. Leipzig, 1890. * Log. cit.
24 PRIMARY MALIGNANT GROWTHS OF THE LUNG
that developmental energy of a high degree becomes active as soon as the sperma enters the ovum. After that, until the organism is fully grown, there is a continuous balancing of energies as manifested in highly compHcated chemical and physical processes. Immediately with the completion of growth, the changes begin which lead to senescence and final destruction of the body. The study of the intricate chemistry and physics of growth, regeneration, and senes- cence is by no means concluded, but has in reaUty only just begun. The relation of these problems to the formation and development of neoplasms is as yet sufficiently obscure, but many a single ray of light shed here and there justifies the hope of further enlightenment in the near future.
It is of special interest in this connection to study the work of Rossle,'^ from which only a few conclusions may be quoted. It appears to him as certain that hypersemia is able to produce a considerable increase in the number of those cells which are organically an integral part of the matrix, and for that reason are subject to the laws of nutrition specific to the latter. Hypersemia, however, cannot produce those biological alterations in the cells in consequence of which endless proliferation is caused. Rossle agrees, also, that hypersemia alone cannot account for the development of tumor, but must be associated with many other factors, among others, probably senescence. His aphorisms con- cerning senility are also most plausible and interesting. There may be senescence of the entire organism or of individual organs only. SeniHty does not attack different parts of the body simultaneously. While one part may long ago have become senescent, other organs may as yet be quite youthful. According to Rossle, the general law may probably be that the more intense the function, the sooner the cell grows old. It is doubtful if, with all their plausibihty, these theories will stand before more than a superficial investigation. Rossle further asserts that epithelium in general retains its juvenile status approximately during the entire life of the individual
1 Die RoUe der Hyperamie und des Alters in der Geschwulstentstehung, Munch. Med. Woch., 1904, p. 1330.
A FEW HISTORICAL NOTES 25
and can be rejuvenated by karyokinesis and regeneration. The earlier in the course of the life of an organism a tissue becomes senile the earher it will be possible for tumors to be developed from this tissue, for according to Rossle it is not those cells and tissues which have become senile, but those which have remained youthful and capable of reproduction and regeneration, which form the origins of these tumors.
CHAPTER IV
PRECANCEROUS INFLUENCES {Continued)
AS all these questions are most intimately connected with the question of the etiology of tumors, it will be best to say a few words in this place on the subject of etiology, at present the centre of so much discussion and labor. The despairing exclamation of Heyf elder, ^ — "Je passe sous silence Fetiologie et le traitement de cette maladie qui, jusqu'a present, est hors du domaine de Tart," — is for- tunately no longer true in its entirety. But still it must be confessed that, with all the colossal labor expended on the question of the etiology of tumors in the last half-century, the fundamental cause, the unknown X, that lies at the very bottom of all these manifold processes, is still entirely obscure and there is as yet not even a sufficient basis for an intelhgent statement of the question that would seem to promise any result. What we know to-day of the physiology, the chemis- try, and physics of growth and senihty seems to suggest that mahgnant neoplasms might in general be accounted for in either one of two ways, and the discussions as to etiology actually do gravitate about these two points. Firstly, one might suppose, seeing that the greatest energy and the foundations for its proper balance are put out in early foetal life, that neoplasms are based ultimately on some earUer or later intra-uterine disturbance. This is, indeed, the theory that was furnished and elaborated by Cohnheim and his followers. 2 Cohnheim, however, did not look upon all this
^ Du Cancer du Poumon, Arch. Gen. de Med., Vol. 14, 2d Series, 1837, p. 345.
^ Many years before Cohnheim, in the paper by Langstaff (Table II, No. 49) in 1818, that author says ^p. 345) that he has noticed "pulpy tumors in the lungs of adult persons who had not been affected during their lives with the least symptoms of pulmonic disorder and who died of active disease of a
26
PRECANCEROUS INFLUENCES (Continued) 27
from the mere standpoint of general physiology and of chem- istry, but assmned remnants of embryonal tissue in this or that organ which, left over, as it were, and endowed with proliferative energy, might under favorable conditions become active and produce tumors.
This theory of Cohnheim, which, for reasons not necessary to state here, seemed untenable, was again revived, though in a much modified form, by Borst ^ and his followers. Borst assumed, as the necessary foundation for the formation of neo- plasms, early disturbances in the intra-uterine development, the nature of which is not as yet accurately known. Accord- ing to him, it is not necessary to assume the bodily presence of actual embryonal remnants. He remarks that, according to his view, it is highly probable that each organ has its own peculiar onkology. A true carcinoma is not developed out of any, no matter how irregular, form of inflammation, no trans- formation into carcinoma is effected when short, glandular, cuboid cells happen to be turned into high cylindrical cells of entirely different structure or when high cylindrical cells happen to be changed into others, again of different structure and of different function, or when single layers of pavement epithelium become stratified into numerous layers of epidermal cells. All these and many more deformations of epithelium might be mentioned which, according to Borst's view, would in no wise transform the particular growth in hand into a carcinoma. What Borst does require, and requires without exception, is just that transformation of an epithelial cell into one of cancerous character, on the details of which so many express differing opinions, and the character of which is so difficult to describe and yet is so readily accepted as a matter of belief.
different description in other viscera." He is inclined to think that fungus haematodes and cancer and scrofula "have their origin perhaps with the formation and development of the natural parts of the foetus in utero and that they remain, after the birth of the individual, in some instances dormant or inactive for a series of years, and in all only require a peciiliar morbid excitement to occasion this increase and destructiveness."
^ Die Lehre von den Geschwlilsten, Wiesbaden, 1902. Uber atypische Epithelwucherung und Krebs, Verhand. Deutsch. Path. Ges., Vol. 6-7, 1903- 1904, p. 110.
28 PRIAIARY MALIGNANT GROWTHS OF THE LUNG
It would be most interesting to continue in detail the history of the various theories and speculations which have led to the present state of our knowledge of mahgnant tumors. This is impossible, because the subject of this essay is tumors of the lung, and not mahgnant growths in general. The necessity of closely adhering to this special subject is still more imperative because of the enormous material on tumors in general pubhshed from year to year, a few examples of which have abeady been mentioned, as Willams,^ Borst,^ the various writings of Ribbert and espe- cially his latest.^ But even a simple catalogue of the more important writings on these subjects, with only carcinoma as a subject, would be enough to fill a small book. Does it not after all seem as if one theory were as good as another and might, by some clever reasoning, be selected according to the subjective taste of the author who elects to defend it? In the writer's opinion, the best evidence appears to be on the side of Borst and his followers. Be that as it may, one can only reiterate again and again that, with all the labor and time spent on these questions by workers in many separate fields of research, and especially the tremendous amount of experi- mental work that has of late years been done by Ehrlich and his school, by Bashf ord and many others, — while it has added much that is valuable to our general knowledge and has been of immense service to our better understanding of many medical and biological problems, especially of onkology, — in spite of all this, no light has been shed upon the ultimate etiology of tumors, and the words of Kraske ^ are in the main still true, — ''We know no more to-day of cancer than did our grandfathers."
That cases of tuberculosis the world over, thanks to the preventive work done everywhere, are steadily diminishing in number seems indubitable. There is, as we have seen, a great deal of legitimate doubt as to the increase of carcino- sis. Behla ^ has pointed out that by adequate disinfection of
^ Loc. cit. ^ Loc. cit.
^ Das Karzinom des Menschen, etc., Hugo Ribbert, Bonn, 1911.
« Naturforschen^ersammlung in Freiburg, Marz, 1902.
^ Loc. cit., p. 177.
PRECANCEROUS INFLUENCES (Continued) 29
tubercular sputum, ulcers, and numerous other places where tubercle baciUi may be found or suspected, by proper isolation and proper sanitaria, etc., the progress of tuber- culosis can to some extent be arrested and that a much greater advance in the arresting of this scourge of mankind may be hoped for in the future. It is quite different with carcinoma. There is as yet no known primary cause for malignant growths. Among the multitude of contagions that we know at the present day, none has been found that seems to have any connection, causative or otherwise, with carcinoma or sarcoma. Carcinomatosis, therefore, does not show any similarity with the contagious character of tuber- culosis. It does not seem to spread infection from individual to individual. It is more than doubtful whether environment, as some authors maintain, plays any active part in the development of mahgnant growths. Behla has not suc- ceeded in proving that special forms of vocation, trade, occupation, etc., or calling of any kind, have any active part in the causation of lung tumors. It is true enough that certain kinds of work are apt to produce inflammatory conditions (bronchitis acute or chronic, anthracosis, siderosis, chronic indurative pneumonia, and others), and the locaU- zation of tuberculosis may possibly be determined by such factors. But it has never been proven that any increased tendency toward the development of mahgnant tumors is caused thereby.^
It may be convenient in this connection to refer briefly to the so-called cancer of the lungs as occurring in the mines of Schneeberg, Silesia, Germany.^ It was thought that here at least was proof positive of the production of mahgnant growths solely by the injurious effects of purely exogenic influences as furnished by irritating occupations. In this small Silesian
^ Conf. the work of Williams, loc. cit.; Karl Kolb, Der Einfluss des Berufes auf die Haufigkeit des Krebses, Zeitschr. f. Krebsforsch., Vol. IX, Heft III, Berlin, 1910; Behla, loc. cit., and many others.
2 Hesse, Das Vorkommen von primarem Lungenkrebs, die Bergkrankheit in den Schneeberger Gruben. Vierteljahrschrift f. gerichtliche Medizin, 1879, pp. 296 ff. Also Ancke, Lungenkrebs der Schneeberger Erzarbeiter, Diss. Miinchen, 1884. Also Komer, Munch. Med. Woch,, 1888, No. 11.
30 PRIMARY MALIGNANT GROWTHS OF THE LUNG
town there were eight mines extending to a depth of fifteen hundred yards, from which cobalt, nickel, and bismuth were obtained. There were from six to seven hundred men employed in the mines, and of these the yearly mortaUty, excluding accidents and the like, was about twenty-eight to thirty-two, of which twenty-one to twenty-four were from carcinoma of the lungs, so that a total of seventy-five per cent of all miners in this town died from this disease. The worker was never affected until after twenty years of mine work, usually later, while the worker who siurvived fifty years of mine work was generally immune. Heredity can be excluded, for only those who worked in the mines, and worked steadily, were afflicted. Those who did not work continu- ously in the mines, or who had other occupations besides mining, or who lived better on the whole, might live to be seventy years or over. The symptoms need not be described here. The autopsies showed that the disease always com- menced from the root of the lung where the lymph nodes were involved and enlarged, ranging from the size of a walnut to that of a fist. Sometimes secondary tumors in the subcutis of the thorax, visible from without, occurred. The timiors were examined frequently, especially by E. Wagner,^ who found the nodules to be true Ijmapho-sarcoma. Cohnheim ^ had already hinted at the likelihood of these tumors not being real tumors at all, but products of some infection. The ques- tion was studied in all directions. It was found that only those who did actual mining, and for a considerable number of years, were attacked by the malady; that there was no local irritation caused by the nickel or cobalt or bismuth particles, but that it was a form of poisoning due to the arsenic found in some quantity in those ores. In other mines of cobalt, nickel, etc., in Sweden, Hungary, and the Tyrol, where the ore contained no arsenic, the disease did not occur. Since the authorities have sufficiently ventilated the mines and have properly regulated the lives of the miners, nothing has been heard of the ^'Schneeberger Lungenkrebs."
^ Eulenberg's Vierteljahrschr. f. Gerichtl. Medizin. 2 Vorlesungen, Vol. I» p. 718.
PRECANCEROUS INFLUENCES (Continued) 31
Trauma. Much stress has been laid on traumatism as an important factor in the development of malignant neoplasms. By "traumatism" is meant here the injuries of the grosser kind, like severe contusions by blows, falls, and similar occurrences. It is always claimed that these severer forms of traumatism have some intimate and direct relations with the development and growth of maUgnant tumors; in fact are the growth-determining element. Statistics, however, do not seem to bear this out. Among the material col- lected in Table I dealing with carcinoma, there are but six cases in which traumatism in the ordinary larger sense is recorded.^
The really effective action of traumatism has for a long time been considered, as displayed in the development of sarcoma. Among the ninety cases tabulated on Table II, there are only two cases (Nos. 15 and 51) in which trauma is recorded. This seems to eliminate once and for all the idea that traimiatism of the grosser kind, at least, has any part in the development either of sarcoma or of carcinoma. Granted that the figures are very uncertain and clinical history and careful observations lacking, the small percentage of cases in which trauma is associated with the formation of tumors can only be due to a coincidence. It might, of course, be claimed that the tumor, — carcinoma or sarcoma, — had been latent before trauma, and that the trauma merely hastened the growth of the tumor. This is capable neither of proof nor of disproof and must remain for the present a matter of beUef and not of knowledge. Experimentally, so far as can be seen, convincing testimony has not been brought forward in either direction, but, as we must constantly keep in mind, no experimentation of any kind has as yet been able to produce an experimental case of malignant growth. The question of traumatism is, of coiu-se, still much discussed and it is surprising to note the lengths to which some authors are prepared to go. Herzfeld,^ for instance, concludes his work with the sentence, ^'Ohne Trauma, kein Tumor" (No tumor
1 Nos. 81, 104, 115, 158, 161, and 177.
' Tumor and Trauma, Zeitschr. f . Krebsforsch., Vol. 3, 1905, p. 73.
32 PRIMARY MALIGNANT GROWTHS OF THE LUNG
without trauma). One interesting case is reported by Schoppler/ in which a fall down stairs with severe contusion of the left mamma was supposed to have given rise to a carcinoma, that portion of the breast having been, supposedly, healthy before trauma. It was quickly operated and the diagnosis corroborated by the microscope. The author considers this a convincing proof of the development of a carcinoma from a single traumatism. The writer does not think that he has proved his case, since, in order to have absolute proof, it would be necessary to have demonstrated, microscopically and otherwise, before the fall, that the portion of the breast affected had been entirely healthy. One must coincide with Bostrom^ in so far as he, with other authors, claims that no malignant tumor can be de- veloped after a single traumatism, from tissue previously healthy. It is not possible, however, to accept uncondi- tionally his further statements, that these large traumatisms may act as coincidental irritants and causes of mahgnant growths.
Besides these blows and contusions, falls and all the grosser forms of traimiatisms, those smaller irritations which lead to chronic infianmiations and indurations, to hyperplasia, and often to hj^ersecretion and hyposecretion of the tissues, must be considered under the general head of traumatism. On this subject there is also a very large literature which cannot be mentioned here. A part of it will be found in Schoppler.' Besides the usual standard works, there are also the publica- tions of Brosch,^ Schuchhardt,^ and Ropke.^
Chronic irritations affecting the respiratory organs are numerous and are supposed by many to play a very active
iZeitschr. f. Krebsforsch., Vol. 10, No. 2, 1911, p. 219. Einmaliges Trauma und Carcinom.
2 Traumaticismus und Parasitismus als Ursachen der Geschwiilste, Giessen, 1902. » Loc. cit.
* Theoretische und experimentelle Untersuchungen zur Pathogenese u. Histogenese der malignen Geschwiilste. Quoted after Wolff, loc. cit.
* Beitrage z, Entstehung des Carcinoms aus chronischentzundlichen Zu- standen der Hautdecken und Schleimhaute, Volkmanns Samml. klin. Vortr., No. 257, 1885.
6 Arch, f . Klin. Chirurgie, Bd. 78, 1905, H. II.
PRECANCEROUS INFLUENCES (Continued) 33
part in the causation of tumors of the lung. Such causes are supposed to account for the predominance of males over females in the occurrence of tumors.^ It is very generally stated that the right side is the favorite localization of car- cinoma of the lung, and this is supposed to be in consequence of the anatomical and physiological conditions. The right bronchus is shorter and wider than the left, its course is considerably straighter, and it seems natural enough that irritating substances, both chemical and mechanical, are aspirated more easily into the right than into the left bronchus. The following figures calculated from Tables I and II seem to show that for carcinoma there is a pre- dominance in favor of the right side amounting to thirty- one cases. For sarcoma, on the other hand, there seems to be a predominance in favor of the left side. The figures calculated from Table III show no predominance of either side.
Carcinoma Sarcoma
Right side 188 Right side 36
left 157 left 51
both 18 both 2
doubtful 3 not stated 5
not stated 8 94
374
Comparison of these figures shows results so inconstant and differences so slight that it would not be wise to build any theories thereon. A. FrankeP comes to a similar conclusion, though based on a much smaller material.
Tuberculosis. The authority of Rokitansky for a long time sustained the dogma that carcinoma and tuberculosis are incompatible diseases; in other words, that where tubercu- losis is found a cancer cannot develop. Another view, at one time popular, is expressed by an aphorism of Crazet^ — "The cancerous easily become tuberculous, but the tuber- culous do not easily become subject to cancer." Actual
1 Conf. p. 22, Chap. III. " Loc. cit.
3 Coincidence et rapport du tuberculose avec le cancer, These de Paris,
1878.
4
34 PRIMARY MALIGNANT GROWTHS OF THE LUNG
experience has since shown, not only that carcinoma, espe- cially of the cancroid variety, is sometimes found in a tuber- culous cavity, but that ordinary pulmonary tuberculosis, with breaking down of tissue and formation of cavities, as well as miUary tuberculosis and locahzed tuberculosis in other organs, may be associated with pulmonary neoplasms. In some cases the diagnosis of associated pulmonary neoplasm and tuberculosis has been made during life. A selection of cases taken from the collected material will serve to illustrate the association of mahgnant growths and tuberculosis. Tumor was present in every case, whether expressly men- tioned or not.
Table I
|
54 Cohn |
Autopsy |
Tuberculous cicatrix in right apex and in Bau- hini's valve |
|
87 Friedlander |
Autopsy |
Cancer in left bronchus and tuberculous cavity left lung |
|
98 Gougerot |
Clinical |
Pulmonary tuberculosis of old standing |
|
106 Harbitz |
Clinical |
Tuberculous family history |
|
257 Perrone |
Sputum |
No tubercle bacilli |
|
Autopsy |
Tubercular cavity at left apex, wall of cavity penetrated by tumor |
|
|
295 Sehrt |
Autopsy |
Carcinoma right bronchus, extensive ulcerative tuberculosis |
|
343 Wolf |
Clinical |
Chronic phthisis |
|
Autopsy |
Tubercular cavity left lung and tumor |
|
|
344 WoK |
Clinical |
Chronic phthisis |
|
Autopsy |
Tubercular cavity right lobe and tumor |
|
|
346 WoK |
Clinical |
Signs of pulmonary phthisis |
|
Autopsy |
Tumor left apex, mihary tubercles over right pleura |
|
|
348 Wolf |
Autopsy |
Tumor of right upper lobe surrounded by fresh miUary tubercles, both suprarenals tubercu- lous, tuberculous ulcer in ileum |
|
349 Wolf |
Autopsy |
Nodules root of right lung, excrescences on membrane of larger bronchi, bifurcation sur- rounded by large tumor, fresh miUary tuber- culosis of both lungs |
|
350 Wolf |
Autopsy |
Tuberculous lobe, tuberculous pleuritis |
|
356 Wolf |
Autopsy |
Carcinoma of main bronchus, miliary tubercles in liver |
|
359 Wolf |
Clinical |
Anorexia and emaciation followed by signs of right pulmonary phthisis |
|
365 Wolf |
Clinical |
Pulmonary phthisis |
|
373 Wolf |
Clinical |
Symptoms of tuberculosis with bacilli |
|
Autopsy |
Lesions of old and more recent phthisis |
|
|
374 Wolf |
Clinical |
Diagnosis first as tuberculosis, then as ss^jhilis |
PRECANCEROUS INFLUENCES (Continued) 35
Table II
36 Hildebrand Tubercle bacilli in sputum
79 Schnick Tubercle bacilli in sputum
The cases will probably be much more nmnerous m future, m proportion to the increasing attention given to this subject at autopsies and microscopic examinations. Some authors appear to take a somewhat extreme stand regarding the relation between tuberculosis and tumors generally, and of tumors of the limg especially. For in- stance, Aronson ^ cites twenty-two cases of his own practice in which tuberculous patients had one parent or both suffer- ing from carcinoma. He even goes so far as to suggest the possibility that the tubercle bacillus under favorable con- ditions might produce carcinoma, and refers to the lupus carcinoma as the connecting link between tuberculosis and carcinoma. It is sufficient to quote the following sentence: ''The phthisical diathesis is not only inherited from parents suffering from tubercular phthisis, but also from those suffer- ing from carcinoma. Etiologically considered, carcinoma, lupus, tuberculosis, all these belong most probably to a single family." As a counterpart to these exaggerated statements, Bayha^ describes the so-called lupus epithelioma and declares this form of epithelial proliferation in no wise cancerous or malignant. He shows that genuine carcinoma develops much oftener on active and fresh lupus than on lupus scars. The proclivity of carcinoma to develop from lupus, and especially from lupus scars, has been mentioned so often as a fact beyond dispute that it is important to note the results of Bayha's investigation. He says distinctly that there is no direct transition from lupus to carcinoma, but that the malignant epithelium prohferates into the interpapillary depressions. WilUams ^ reiterates his view that as tubercu- losis declines, carcinoma necessarily increases. It is also his belief that the systemic depreciation that follows as a conse-
1 Beziehungen zwischen Tuberculose und Krebs, Deut. Med. Woch., 1902, No. 37, p. 842.
^ Uber Lupus Carcinom, Bruns, Beitrage zur Klin. Chir., Vol. Ill, 1888, p. 1. » Loo. cit., pp. 337 ff.
36 PRIMARY MALIGNANT GROWTHS OF THE LUNG
quence of fresh tuberculosis, and even of tuberculosis only recently healed, is an undoubted factor in the etiology of cancer. On the other hand, he readily agrees to the fact that while a considerable amount of old, healed, calcified tuber- culous products may be found associated with neoplasm in the lungs, this association has no further meaning than that, cicatrized tuberculosis being so extremely common, the ordi- nary percentage is also found in the cancerous. Furthermore, F. P. Weber and many others suggest that old, quiescent tuberculous foci, not yet completely cicatrized, may be again started into activity by the local as well as systemic effect of the cancer, which naturally tends in a great measure to lower the patient's vitality. This, however, is a speculation of which we know nothing.
The subject of tuberculosis in its relations to carcinoma should not be closed without mentioning the theories of Kurt Wolf."^ Wolf distinguishes closely between bronchial carcinoma and carcinoma of the lung proper. Of the latter he reports nine cases, of carcinoma of the bronchus twenty- two. ^ He points out that bronchial carcinomata are nearly always found in those places which are most subjected to slight, but chronic, irritations, especially on the right side and more particularly near the bifurcations. He does not so much refer to the tracheal bifurcation, but more to the bifur- cations of the second, third, fourth, and following orders. NatiKally, all the irritations of aspiration, of dust, tobacco, and so on, as well as coughs, are apt to centre about these points. It is there that Wolf most frequently finds very small melanotic lymph nodes which, even at a very early stage, are tuberculous. Sooner or later a minute perforation into the bronchus takes place, into which the melanotic contents of the Uttle node are discharged ("Pigmentdurchbruch"). The lymph nodes on the down track toward the hilus of the lung, and of the hilus itself, become enlarged in the course of the process. It is Wolf's contention that these little melanotic lymph nodes are apt to be tuberculous; that
1 Wolf, Der Primare Lungenkrebs, Fort. d. Med., 1893, Vol. 13, Nos. 18 and 19. 2 Conf . Table I.
PRECANCEROUS INFLUENCES (Continued) 37
when penetrating into the bronchus or developing at the root of the lung they act as a chronic irritant at the locali- zations most exposed. This "Pigmentdurchbruch,"i Wolf claims, is sufficient, in persons hereditarily predisposed, to start the development of malignant growth. This malignant neoplasm then proliferates in the bronchus first affected, travels along the ramifications of the bronchial tree, pene- trates into the lungs, and forms more or less extensive timiors. This theory of Wolf has been the subject of some discussion, but has not been generally adopted. The presence of the tubercle bacillus or any active tuberculous process has never been definitely demonstrated in these minute lymph nodes or their further development. He finds, out of the thirty- one cases which he reports, eleven cases which exhibit, not cicatrized and inactive, but mostly fresh and active tuber- culous processes, by the side of indubitable primary malig- nant neoplasms in the lungs. This, however, does not suffice to prove his ingenious theory.
That carcinoma does occur on various cicatrizations, especially of the skin or mucous membrane, is a fact. It is only necessary to refer to the carcinoma on lupus, previously mentioned in this connection, on ulcer of the stomach, on leukoplakia, gall bladder, etc. This form of precancerous affection evidently is not concerned in limg tumors, unless we except the theories of Wolf, just briefly outhned, or of some other authors, who find in tuberculous cicatrizations or tuberculous ulcers a formative irritant for the development of carcinoma.
An attempt has been made to obtain some knowledge of the duration of carcinomatous disease from Table I. Reliable values are, however, not easily obtainable, and it is possible to give only an approximate and very defective notion of the duration of primary carcinoma of the lung. The reasons for this are obvious. Many authors neglect to give any data from which the duration might be deduced, and the patients themselves are often so little self-observant and so careless
> This "Pigmentdurchbruch," so fax as the writer knows, has been demon- strated only a single time.
38 PRIMARY MALIGNANT GROWTHS OF THE LUNG
of their physical condition that they seek medical aid long after the first appearance of symptoms, the date of which, therefore, can no longer be fixed. Finally, the first appear- ance of symptoms does not necessarily coincide with the beginning of the disease. Among the 374 cases tabulated in Table I, there are no means of calculating the duration in 230 cases. The longest duration given is five years, the shortest two weeks. ^
^ For details, see Appendix A.
CHAPTER V
PATHOLOGY
THERE is an old aphorism saying that those organs most disposed to secondary tmnors are least disposed to the formation of primary neoplasms. The limgs are undoubtedly a favorite locaUzation for secondary tumors, but primary neoplasms are by no means rare. All the types of tumors represented in the onkology of other organs may also be found in the limgs.
The gross appearance is not uniform or characteristic. It differs according to the peculiarities in each individual case. For carcinoma of the lungs, the older writers distinguish only between encephaloid, or what they called medullary, cancer (" Markschwamm " and fungus hsematodes) and the infiltrated form, the names being given merely to indicate external differences. Jaccoud ^ mentions that primary cancer of the lung is nearly always of the encephaloid variety and is seen either "en masse" or in a more infiltrated form. He considers the "cancer en masse" as the more frequent. It is not easy to determine just what kind of tumor, — sarcoma or carcinoma, — Jaccoud had before him. A much greater variety in gross appearance of this class of tumors is now recognized.
One form that occurs occasionally is that of a single nodule, usually quite small, surrounded perhaps by a few minute miliary nodules deeply buried in the lung tissue of one lobe, producing only very slight or possibly no symptoms during life, and as a rule discovered by mere accident at autopsy. These cases are rare. The writer has seen two.
There is the so-called mihary form of carcinosis, which in
^ Jaccoud, Legons de Clinique m^dicale, 1871-72, p. 454, Cancer de pou- mon; Traits de pathologic interne, Vol. 2, p. 120.
39
40 PRIMARY MALIGNANT GROWTHS OF THE LUNG
the gross resembles very nearly an eruption of miliary tuber- cles.^ There is perhaps this difference, that the little nodules are somewhat larger than the tubercles and have not the peculiar grayish translucent appearance, but are more whitish and generally distributed along the lymphatics.^ The reader is referred, for a history and description of the acute miliary carcinosis in general, to J. Wolff. ^ As for the lungs, there seems to be no doubt that a miUary carcinosis actually exists, as Rokitansky* and Elisberg^ hold, but it is probable that these cases are not always primary. It is very much more likely that they are secondary to some small tumor that — possibly owing to lack of symptoms, possibly because hidden away in the depths of some bulky organ — was not detected. The nodular form of primary carcinoma of the lung as a rule involves in its beginnings only a portion of one lung, while metastatic carcinomatous nodules in the lungs are apt to be distributed throughout both lungs. The nodules are found of varying sizes, from that of a cherry pit or walnut to that of an egg, small apple, or even a human fist. They are not usually confluent, but are separated from each other by lung tissue. The boundary between the tumor and the lung is sharply defined. As the process continues, the lung tissue intervening between nodules often becomes involved in secondary inflammatory and degenerative conditions, and the nodules, as they increase in size, may merge one into the other. Jaccoud,® and since his time others, have been of opinion that cavities and breaking down of tissue within these nodular carcinomata do not occur, or at all events are very rare. On the contrary, however, the material col- lected in Table I will show that the formation of irregular cavities, especially in the larger nodulated tumors, is a common occurrence. The gross appearance on section of these nodules varies according to the kind of tumor and the condition in which it happens to be, and it is therefore not
^This form was first described by Demme, Schweiz. Monatschrift f. prakt. Medizin, Jahrg. Ill, 1858, No. VI.
2 Conf. Wunderlichs Archiv., 1857. ' Table I, No. 80.
» Loc. cit.. Vol. II, pp. 398 S. " Loc. cit.
« Loc. cit., 1856, Vol. I, p. 255.
PATHOLOGY 41
possible to present a uniform and generally applicable description. One may be sure, however, that besides the usual grayish-white or yellowish or pinkish-white tumor material there may be found pathologically altered bronchi and vessels, bronchiectatic dilatations, and, as has been said, occasional cavities. The cavities have ragged, irregular walls, consisting of tumor. Stumps of vessels and bronchi often protrude into them from the walls. The cavities usually contain detritus from tumor material, old or fresh blood, mucus, and so on.
The infiltrating form. This form is very common. Sepa- rate nodules, large and small, are rare. The tumor, usually starting from a bronchus, penetrates the bronchial wall and infiltrates the lung along the bronchial as well as the venous, arterial, lymphatic, and even nerve ramifications.^ This type is subject to many variations, according as the infiltra- tion happens to proliferate mainly along the preformed track of the bronchial ramifications or extends down to the root of the lung, involving not only larger bronchi but also the bron- chial, tracheal, and mediastinal glands. It thus forms, besides extensive pulmonary infiltrations, considerable masses of tumor at the root which, in their effect upon larger bronchi, trachea, large vessels, and other mediastinal organs, cause bronchiectatic dilatations, atelectatic areas, even gangrene, in the lungs, and all those symptoms, to be discussed later, which pertain to intra- thoracic growths in general. ^
There is another type of infiltrating tumor affecting only a portion of a lobe. This starts as a rule from smaller bronchi or bronchioli; the infiltration is sharply defined against the normal lung tissue, and is so dense that within the region of the tumor scarcely any lung tissue can be found. The entire area is taken up by tumor in which only a few arteries and veins and some slight dilated bronchi are visible.'
In Plate 2 the destruction of almost the entire lung, from top to bottom, is well shown. There is little healthy lung tissue, for nearly the entire lung is gone and the pulmo- nary tissue replaced by tumor, at first creeping along and » Stilling, Table I, No. 310. * Conf. Frontispiece. » Plate 1.
42 PRIMARY MALIGNANT GROWTHS OF THE LUNG
infiltrating the lung tissue, then degenerating and breaking down iato cavities, etc., as described.
The gross forms thus far described apply in general only to carcinoma of the lungs. The rare cases of sarcoma may assume similar macroscopic forms and it will then become difficult to distinguish sarcoma from carcinoma with- out the aid of the microscope. There is one gross form, however, that is, to all intents and purposes, pecuUar to sarcoma. This form appears as very large tumors with fairly homogeneous structure, sometimes containing cavities, but comparatively rarely, and never when the tumor is a lymphosarcoma. These growths may become so large as to occupy the entire half, or more, of the chest. That portion of the lung which is not destroyed and replaced by tumor remains as a mere shell around this growth. Heart, dia- phragm, mediastinal contents may be extensively displaced.
This very brief and necessarily incomplete sketch of the mere gross appearances will suffice to show how varied and comphcated, how difficult of interpretation, are the post- mortem pictures presented by lung tumors. Sometimes the picture as seen by the naked eye cannot be recognized as tumor at all, and the lesions as shown at autopsy will be interpreted as inflammatory or degenerative processes, — for instance, as chronic, indurative, or pneumonic lesions. It follows from this that at every autopsy, even at those where there is no reason to suspect the presence of tumor, a microscopic examination according to modern methods is necessary for every portion of the lungs that does not appear absolutely sound and healthy.
Passing from the macroscopic to the microscopic study of primary maUgnant neoplasms of the limg, manifold difficulties in determining the histological structure of the tumor, its interpretation and classification, are encountered. As the simpler group of these tumors, and presenting fewer of these difficulties, sarcoma will be first discussed. Hertz ^ goes so far as to deny the existence of primary sarcoma of the lung, claiming that every sarcoma found in that organ is
* Neubildungen der Lungen in Ziemssens Handbuch, 1874, Vol. 5.
PATHOLOGY 43
secondary. It must be admitted that primary sarcoma of the lung is a great rarity. The writer has not had the good fortune to observe a single case. Nevertheless, it has been attempted here to show that the relation of primary sarcoma of the limg to primary carcinoma of that organ does not differ from the relation which sarcoma bears to carcinoma in general.^ This conclusion is based on a collection of ninety- four cases from the hterature on the subject, ninety of which have been listed in Table II. It is quite possible that a num- ber of those set down as doubtful in Table III are genuine sarcoma. It is possible also, and very probable, that a great many cases have not been recognized and therefore not recorded. 2 As more attention is paid to this subject, reports of cases are pubhshed in greater number than would have been thought possible some years ago. It would have been easy to increase the number of cases on Table II to more than one hundred. All this shows that the beUef in the extreme rarity of sarcoma has been somewhat exaggerated.
It has been shown above that the gross pictures presented by sarcoma may differ so slightly from those offered by carcinoma that microscopic examination alone would serve to differentiate between the two. It may, however, be said roughly that sarcoma has a greater tendency to spread toward the root of the lung, and involve from there the mediastinal lymph nodes and other organs, than has carci- noma. Melanotic sarcoma is extremely rare, — there is, in fact, some doubt in the writer's mind that it occurs at all. The dark anthracotic pigmentation of lungs and bronchial glands, pathologically more prominent perhaps, may erro- neously lead to the suspicion of melanosis. The very large and massive tumors occupying a great portion of the chest have just been referred to. They are occasionally subject
1 According to Williams (loc. cit., p. 377), 54.5% of all tumors are car- cinoma, 9.4% sarcoma, 24.7% non-malignant, and 11.4% cysts. These figures corroborate the above statement.
2 A quotation from Menetrier (Lubin, These de Paris, 1909, Contributions k I'Etude du Sarcome primitif du Poumon) seems apt enough in this connec- tion: "Le cancer n'est pas une forme morbide primitive; c'est un aboutissant d'etats pathologiques multiples, anterieurs et preparatoires."
44 PRIMARY MALIGNANT GROWTHS OF THE LUNG
to osseous and especially to calcareous degeneration.^ A scrutiny of Table II shows that about half of the cases tabulated are of this massive type. Between these and the more infiltrating forms there are, of course, all manner of transitions. An especially interesting case came to hand after the Tables were finished. In this case the entire left half of the chest was filled by a voluminous mass, dislocating the heart, impinging on the right lung, and depressing the liver. The left lung was almost completely replaced by a huge tumor which pushed the remnants of the pulmonary tissue upward. The tumor contained a cavity in the midst of soft tumor material. The duration of the disease was almost three and a half years. ^ A most interesting case, also, is that reported^ of a male thirty-three years old, who entered the hospital in July, 1896. He had been sick since the previous December with cough, haemoptyses, pains in right chest, and in addition bronzed skin and bluish sclerse. In February, 1896, he was seized with a severe pain in the right leg, especially in the knee, which lasted until death. The entire right side was more painful than the left; no pig- mentation in the mouth; percussion absolutely flat over entire right anterior chest, and resistance much greater than normal; some cavernous breathing below the right clavicle, otherwise absolute silence over the whole right posterior lung; sputum contained nothing characteristic. The autopsy showed an enormous sarcoma of the right lung, many metastases of liver, pancreas, etc. Microscopically, a giant celled sarcoma of mixed type. A diagnosis of primary tumor of the lung had been made during life, but at autopsy the authors were inclined to consider the lung tumor secondary and the tumor in the femur as primary; in the first place on account of its microscopic structure, — the mixed giant celled sarcoma, — the giant cell being more common in
1 Chiari, Table III, No. 4.
2 Heilbron et Sezary, Sarcome primitif du poiimon, Bull, et Mem. de la Soc. Anatom. de Paris, Ann^e 85, No. 7, p. 758.
3 Packard and Steele, Case of Sarcoma of the Lungs, with symptoms of Addison's disease with involvement of suprarenal capsules. Med. News, 1897, No. 11.
PATHOLOGY 45
bone; furthermore, the advanced condition of degeneration in the femur beyond that of the lung. For this reason the authors claim the tumor in the lung as secondary. This may be correct, but the true facts cannot be obtained with certainty. If it is secondary in the lungs, we have the very unusual, as far as the writer knows, the unique, occurrence of a secondary sarcomatous deposit involving only a single lung and assuming such huge proportions as almost to occupy the entire lung. It might be interesting to refer here also to a publication by Eckersdorff.^ According to his statistics 1.5 per mille of all autopsies are primary sarcoma of the lungs. Eckersdorff finds up to the year 1908 only four cases of primary sarcoma of the lungs. He publishes two cases, one of a man fifty years old living rather a wild life. In November, 1902, in joke, a friend gave him a blow between the shoulder-blades which led to a strong desire to urinate. Next day he felt still much affected, but on second day entirely well again. Soon thereafter he began to be hoarse, had pains in region of heart and intermittency of pulse. The most interesting part of the later history is the rapid change when, after considerable dyspnoea, irregular and rapid pulse, urine without albumen, enor- mous thirst, the patient would suddenly get better. It was not until late in the course of the disease that total dulness of left lung with abolished breathing sounds was discovered. This dulness disappeared quickly with the exception of one place. Later on there was a sudden dis- appearance of the pains. Death February 7th in collapse. The diagnosis during life was: probable neoplasm in the lung. The anatomical diagnosis, an annular carcinoma of the left main bronchus with obstruction of this and the formation of metastatic deposits in the lymph nodes and on the heart, oedoema of both lungs, pneumonia of the left lower lobe, and dilatation of both ventricles of the heart. Microscopical examination showed that it was not a car- cinoma, but a sarcoma of small round cell type. The
' Zwei Falle von primarem Sarkom der Lunge, Centralbl. f. allg. Path., Vol. 17, 1906, p. 355.
46 PRIT^IARY MALIGNANT GROWTHS OF THE LUNG
histogenesis cannot with certainty be determined. The author thinks that the connective tissue of the bronchial mucosa is the place of origin. He does not express a positive opinion as to the causal effect of the blow. In a second case the origin is referred to the interalveolar septa. The author expresses the hope that in future the sputum may be studied more carefully in such cases.
Another case which appeared after the Tables were finished may be mentioned here, though not a sarcoma, the interest- ing feature of it being the observation of the blood. Haemo- globin is not mentioned, but in the first blood count the red cells are reduced to 3,886,100 and the leucocytes are increased to 19,840, of which the polynuclears are seventy- nine per cent. A second blood count also does not give the haemoglobin. The red cells have dropped down to 2,926,400, the whites have increased to 24,800, and the poly- nuclears are now eighty-six per cent. A large tumor is found with cavities supposed to involve the larger bronchi and the hilus. The microscopical analysis shows a cancroid. Origin from the bronchus is nevertheless assumed.
The frequent occurrence of primary sarcoma of the lungs in the form of huge and ponderous tumors is also corrob- orated by Duran.i Schech^ states that when in the right lung, the favorite seat of the tumor is the upper lobe, while in the left lung the favorite seat of tumor is the lower lobe, and that he has seen the tumor primary in both lungs only twice. Looking over Table II in regard to this point, one will find that there is no such difference, but that tumor in the right upper or left lower lobe, and the converse, occurs with equal frequency. There are five cases cited in the Table where both lungs are affected. The duration of sarcoma of the lungs does not seem to differ very materially from that of carcinoma. There are fifty-two cases out of the ninety in Table II from which some approximation as to their possible duration may be reached. Among these fifty- two^ the shortest period of duration is one month and the
^ Du sarcome primitif du poumon, Th^se de Paris, 1893. « Table II, No. 78.
PATHOLOGY 47
longest six years, the average being about four and a half months, as compared to that of carcinoma, the average for which is two and a third months. It is evident that these averages have no real significance, and the only legitimate deduction from the figures is that primary carcinoma and sarcoma of the lungs are of indefinite duration, running at times a very rapid course and again assuming the character of chronic disease and lasting for many years. ^
The histology of primary sarcoma of the lungs offers in the main nothing peculiar or characteristic, but practically corresponds with the histology of sarcoma of other organs. It has been said^ that the spindle cells occur more frequently than any other type of cell. Examination of Table II in regard to this point shows only sixty-eight cases available, as in the remaining twenty-three there was no clear state- ment as to the character of the cells. Out of these sixty- eight cases just half were of the typical uncomplicated round celled variety, fourteen only were spindle celled, seven uncomplicated lympho-sarcoma, and there were also a few mixed tumors, such as lympho-sarcoma with small round cells, with spindle cells, etc. It seems, therefore, that round celled, and not spindle celled, sarcomata are by far the most frequent. Occasionally, giant cells are found.^ There are found, also, the usual combinations, such as myxo-sarcoma, fibro-sarcoma, and others; various degenera- tions, as mucoid, colloid, more frequently fatty, and also calcareous and osseous, attributable principally to the stroma; occasionally there are cystic forms.
The histogenesis is still obscure. It seems certain that a great many of the pulmonary sarcomata take their origin from the root of the lung, probably in one or the other of the smaller or smallest of the peribronchial glands, growing from there, as mentioned before, along the track of the bron- chi, and at an early period penetrating a larger or smaller
^ For further details regarding duration of primary sarcoma of limgs, see Appendix B.
* Schech, loc. cit.
' Packard and Steele, loc. cit. Also Colomiatti, Table II, No. 14. Also Klemm, Table IV, No. 10.
48 PMIvLmY MALIGNANT GROWTHS OF THE LUNG
bronchus, obstructing it, and thus continuing in its course through the lungs, the tissue of which it destroys on its way. It may also, it is said, penetrate through the pores of the septa directly into the alveoles. The large massive tumors almost invariably start at the hilus. It is assmned by many, though not yet conceded by all, that sarcoma may develop from the interalveolar septa in the lung itself. The septa, at one or several spots becoming sarcomatous, may compress the pulmonary alveoles and fill with tumor material what is left of the air-vesicles, thus forming nodules of vary- ing size which, again merging into similar nodules, can form considerable tumors. The lung tissue in the immediate en- vironment of these nodular tumors is usually quite healthy, or evidences only minor changes. Microscopic examination may show remains of septa or the latter may have been de- stroyed altogether. As a rule there is no open communication with the bronchus, but bronchial remnants are seen within the tumor. In some instances the sarcomatous tissue does not completely destroy the septa, so that the alveolar struc- ture in some places at least remains distinctly visible. The air-vesicles are then filled with a mass of polymorphous cells which, according to the individual bias of the observer, may pass either for epitheUal cells or for deformed sarcoma (round) cells or for endothelial cells. The dispute concerning endothelium will be touched upon later. For the present it may be said that some authors consider the endothelium to play a considerable role in the histology of sarcoma, and Burkhardt,^ after extensive researches, thinks that sarcoma and endothelioma are not to be separated from each other, inasmuch as every sarcoma, besides the proliferating cells of the connective tissue, contains a greater or less proportion of endothelia of the lymph spaces as well as adventitia cells. All sarcoma are, therefore, according to him, more or less endothelioma, and only according as the connective tissue cells or the endothelia react stronger do the various types stand out. This is, of course, a very extreme point of view
^ Sarkome und Endotheliome nach ihrem path.-anatom. und klin. Ver- halten, Bnins Beitr. z. klin. Chir. 36, 1902.
PATHOLOGY 49
and will have to be discussed later when endothelioma is touched upon. The microscopic picture often speaks for this theory, as it presents distinct alveolar structure with much enlarged septa consisting of spindle cells and alveoles filled with polymorphous cells. It is this type of tumor that probably comes under the head of what Virchow termed carcinoma sarcomatodes.^ The case of Weichselbaum ^ seems to be a true adeno-sarcoma. Is it not possible that this kind of tumor resembles those produced experimentally by Ehrlich and his school, in which the stroma of a carcinoma was ultimately converted into genuine spindle or round celled sarcoma?
Carcinoma. The epithelium found in the lungs (lungs being taken in the broader sense and including the bronchi) consists of cylindrical epithelium, cihated as well as not cihated. The ciliated cells form the hning of the mucous membrane of the larger bronchial tubes. As with continued dichotomous division the branches of the bronchial tree be- come smaller, so the high ciliated cells become lower, the cilia gradually disappear, and the very smallest bronchioles are simply lined by a small, low, cuboid epithelium without cilia. The bronchial epithelium in the minutest bronchioles is by gradual transformation changed into the respiratory and alveolar epithelimn. In the adult this consists of fiat, squamous cells resembling endothelium. They line the septa and the pulmonary alveoli. The endothelium itself, those cells which form the inner coating of the lymph vessels and spaces, must be presently considered somewhat more in detail, as it is still a subject of dispute. Cyhndrical epithelium is also found in the bronchial mucous glands. This has no cilia and differs in no way from the ordinary cylindrical cell as found in glands.
Considering only the very limited group of cells that contribute to the structure and formation of the carcinoma of the lung, it is often surprisingly difficult to distinguish the kind of epithelial cells that make up the tumor, and its
^ Bohme, M., Primares Sarco-Carcinom der Pleura, Virchows Archiv., Vol. 81, 1880, p. 181. 2 Table III, No. 94.
5
50 PRIMARY MALIGNANT GROWTHS OF THE LUNG
structural peculiarities, and to understand the histogenesis. The enormous plasticity of the epithelium, the influence which territorial hmitations, intense proliferation, pressure upon each other, and various other intra- and extra-cellular changes bring to bear upon the cells, — all these features conspicuously increase the difficulties. It may really appear at times as if there were no specific kinds of epithe- lium, but that the epithelial cell, according to merely extrinsic conditions, might assume any form, cylindrical cells being transformed into pavement cells, pavement cells into horny pearls, etc. One is frequently at a loss to decide whether, in the section before him, the cells are of epithelial or connective tissue origin, whether it is a carci- noma or a sarcoma. Frankel, in the discussion of Simmond's paper, ^ states emphatically that great difiiculty is often experienced in distinguishing between carcinoma and sarcoma, owing, on the one hand, to the alveolar structure of the lung simulating carcinoma, and on the other hand to the almost limitless proliferation and change of form of the epithelia suggesting sarcoma. A good example of this is shown in Plate 3. Here the cells are so crowded, the prolif- eration is so rapid, that it would be impossible at the spot photographed to make any other diagnosis than that of a small round-celled sarcoma. No one would easily believe that these cells are mere transformations of epithelial cells and that the tumor is a true carcinoma. Plate 4 shows the same section with a higher power. One sees a great variety of polymorphous cells, some of which resemble epithelial, others sarcoma cells. In one spot a mitosis is plainly to be seen. Plate 5 is a section of the same tumor from another place, photographed with a moderate magnification, which plainly demonstrates the alveolar structure, the typical stroma, and in several places undoubted epithelial cells. There can be no hesitancy in calling this tumor a carci- noma. Plate 6 is a section from the kidney of the same patient, photographed with high power and showing most
^ liber die Histologie des prim. Lungenkrebses, Miin. Med. Woch,, 1896, p. 189.
PATHOLOGY 51
beautifully a few undoubted epithelial cells just after their entrance into Bowman's Capsule. This picture may serve to remove all possible doubt as to the true natiire of the tumor.
The various well-known types of carcinoma are all repre- sented. The carcinoma simplex. Plate 7 is a good illustra- tion of this. The alveolar structure is very plain, the alveoles varying in size, lined with cuboid or cylindrical cells and filled with polymorphous cells jumbled together, compressed out of shape and partly degenerated (horny, mucoid, colloid, fatty degeneration, etc., are frequently met with). The stroma is usually rich in cells and here and there a lymph space filled with epithelial cells is seen. It is very interest- ing to note in the picture a tolerably large alveole projecting its epithelial material directly into a lymph vessel. Plate 8 shows the typical glandular carcinoma without any distinc- tive features, and consisting mostly of flat and cuboidal epithelial cells with very little stroma. In this section there is nothing to suggest the origin of the tumor from the lung. Plate 9 shows the same form of carcinoma with smaller and more plexiform alveolar structure, more voluminous and firmer interstitial tissue, and a very plain demonstration of the infiltration of lymph vessels and spaces from the alveolar contents. In Plate 10 is shown a good example of a can- croid with the characteristic horny epithehal pearls. The basilar lining of cuboid cells is in this section not very plain.
The cylindrical celled carcinoma. Plate 11. The cells are not ciliated. The alveolar structure is evident, the alveoles varying in size. The larger ones are about the size of a moderately large bronchus, and it is obvious that they are formed by the confluence of a number of smaller alveoles. The contents of these larger alveolar spaces, sometimes sug- gesting small cavities, consist of cellular and mucous detritus and scattered epithelial cells in various stages of degenera- tion. The stroma between the alveoles generally consists of rather soft connective tissue containing moderately abun- dant connective tissue cells. This form of carcinoma, occur-
52 PRIMARY MALIGNANT GROWTHS OF THE LUNG
ring as it does quite frequently, is considered by many pathologists to be the typical, if not the only form, in which carcinoma occurs in the lungs. It is demonstrable that this type of tumor develops from the cells of the bronchial mucous glands. That this is so was first shown by Langhans,^ whose views were widely accepted. ^ In Plate 12 there is seen very clearly to the right of the pic- ture a dilated bronchus with mucoid detritus in its interior and a partially detached epithelial lining. In the middle of the picture are shown the bronchial epithelial glands, the majority of them unchanged, others just at the beginning of carcinomatous proliferation. Toward the left are some alveoles lined with cylindrical cells and the transition from proliferating bronchial mucous glands to carcinomatous alveoles is clearly perceptible. Plate 13 illustrates similar conditions. The bronchial cartilage is in parts destroyed and there are similar carcinomatous degenerations as in the preceding figure. Some of the alveoles, evidently originat- ing from degenerated bronchial mucous glands, contain carci- nomatous epithelium, not typically glandular, but exhibiting the usual character of pavement epithelium.
Carcinoma may also develop from the surface epithelium of the bronchi. It is still a matter of some dispute what kind of cells are characteristic of this form of carcinoma. It is thought by competent authorities that the surface epithelium of the bronchi develops a carcinoma of alveolar structure with polymorphous and polyedric cells that are, in the great majority of cases flat, but sometimes varying numbers of cylindrical cells are mingled with them. Such forms of carcinoma are exemplified by Plates 8 and 9. It was contended by some^ that the carcinoma just described might develop from the bronchial mucous membrane, but might also take its origin from the flat epithelium of the pulmonary alveoles. This contention caused considerable
1 Virch. Arch., Vol. 53, 1871, p. 470.
2 Chiari, Table I, No. 51; Ebstein, Table I, No. 75; Stilling, Table I, No. 310, and others.
3 Ehrich, Table I, No. 77, and others.
PATHOLOGY 53
discord among the few pathologists who studied the subject. A number of these without hesitation considered every pul- monary carcinoma, where they found fiat polyedral epithe- lium, as necessarily derived from the alveolar cells. A little closer study showed the untenable character of these theories. It is unnecessary to enter into all the details of the discussion. Some considered the flat epithelium in pulmonary carcinoma extremely rare, others considered it very frequent. Frohhch,^ for instance, found it twelve times among sixteen cases. According to the statistics of Watsuji,2 32.2% of all pulmonary carcinomata are of the pavement cell variety. There is, however, no evidence that these carcinomata develop from the pulmonary alveoles. On the contrary there is considerable evidence against the supposition. It is now held that carcinoma starting from the pulmonary alveoles is extremely rare, and some go so far as to deny its existence altogether. Marchand and his pupils ^ succeeded in demonstrating beyond doubt a tumor starting from the alveolar respiratory epitheUum. The tumor in question would hardly be recognized as tumor by the naked eye, but rather suggested the opaque and some- what translucent tissues as they occur in chronic broncho- pneumonia, and the structure as shown by the microscope was a great siuprise. It was found that the tumor was made up of cylindrical cells with more or less of a papillary arrangement. As the respiratory epithelium in the embryo is of the cylindrical type, the occiu-rence of cylindrical cells in these growths is not surprising. The tumor is probably congenital. Plate 14 shows a section of this sort of tumor, in which remnants of alveolar structure, with somewhat irregular but nevertheless recognizable high cylindrical cells, can still be traced. There are perfectly clear patches showing papillary arrangement.
Neglecting in this place all further detail, it may be briefly stated that it is at present the common consensus of opinion, and probably justly so, that the great majority of
1 Table I, No. 88. * Zeitschr. f. Krebsforsch., Vol. I, p. 445.
* Ejretschmer, loc. cit.
54 PRIMARY MALIGNANT GROWTHS OF THE LUNG
primary carcinomata of the lungs develop from the bronchi, and that a cancer of the lung is, taken strictly, a bronchial carcinoma; that, on the other hand, a carcinoma starting from lung tissue itself occurs, but is extremely rare, and is built up, not of flat, but of cylindrical epithelium.
CHAPTER VI
PATHOLOGY (Continued)
\ NY attempt to work out the histogenesis of lung tumors "^~*- leads at once to troublesome questions concerning epithelium, metaplasia, and other fundamental problems about which there exist great differences of opinion in the pathological world. It may be said at once that it is gen- erally impossible to determine the histogenesis of a fully developed lung tumor and it rarely or never happens that we meet with a tumor so small that its very beginnings can be clearly seen. Even the close study of the growing edges of the tumor will give no satisfaction, and any certainty with regard to the histogenetic origin of the majority of lung tumors must, for the present at least, be given up as hope- less. Turning to epithelium, it is at this moment practically impossible to say what "epitheUum" really means and what its relations are to other kinds of cells, especially to endo- thehum. The literature on the subject of endothelium and its relation to tumors, as well as to acute and chronic inflam- mations in adult tissue and its embryonal history, is really enormous, and no attempt at even a sketch can be made here. The work of Borst^ in his large treatise on tumors, and his several other separate publications, ^ and the critical compilations of Monckeberg,' go deeply into the question of endothelioma, while Volkmann,^ and before him Kolaczek,^ have done fundamental work in the study of these tumors. Leaving this mass of literature to those specially interested, it is important to arrive, at the very beginning, at some un-
^ Lehre von den Geschwiilsten, Wiesbaden, 1902. 2 Das Verhalten der Endothelien, Wurzburg, 1897, and others. ' Lubarsch, Ergebnisse, 10 Jahrg., Wiesbaden, 1906. * Deut. Z'tschrift f. Chir., Vol. XLI, 1895. 6 Deut. Z'tschrift f. Chir., Vols. IX and XIII, 1878 and 1880.
55
56 PRIMARY MALIGNANT GROWTHS OF THE LUNG
derstanding of the nature of epithelial cells. It is generally accepted that epithelium assumes various forms differing in morphological structure and in physiological function. The forms recognized by all are: (1) cylindrical epithelium, which is differentiated into several species : (a) endowed with cilia upon which certain physiological motor functions depend, and (6) without cilia, dispersed in a single layer or in several strata, serving as an inner coating to numerous hollow organs, and lastly, (c) glandular cylindrical epithelium, to which are allotted duties of secretion and excretion; (2) fiat, squamous, or pavement epithelium, arranged either in single layers or; in numerous strata and modified in its morpho- logical structure according to the physiological function which it is called upon to perform. The lining of numerous internal organs consists of this type of epithelium. The epidermis which protects the surface of the entire common integument is in the main built up of such cells, specially differentiated as to their structure and chemical constitu- tion (kerato-hyalin, intra-cellular structure, and protoplas- matic bridges). No further detailed description of epithelial cells is necessary. Until very recently it was accepted as a fact that the three germinal layers were the dominant factors in the histogenesis of all the tissues and organs in intra- as well as extra-uterine life. All the epithelium that was needed for the viscera of the chest and abdomen was supposed to be furnished by the entoderm. The epithelium of the common integument and of several other organs closely connected with the outer surface is referred to the ectoderm. There is besides this a certain class of flat cells bearing nearly all the hallmarks of genuine flat epithelial cells, which are universally found in the body as a lining of the great lymphatic cavities (pleura, peritoneum, etc.). The inner coat of the arteries and veins and the perivas- cular lymph spaces, as well as all lymph spaces throughout the body, are lined with this peculiar epithelium. Its origin is said to be from the mesoderm, the mesoderm being the third germinal layer, from which the fibrous and connective tissue, the bones, cartilages, elastic fibres, etc., — aptly
PATHOLOGY (Continued) 57
called by the Germans ''Stiitzgewebe," — are said to origi- nate. These cells just mentioned as coming from the meso- derm could not be classified as genuine epithelium and were therefore called by His endothelium. They showed, on the one hand, close connection with the connective tissue cells, with which, indeed, they have much in common, espe- cially the property of forming fibro-plastic cells. There are many tumors that are supposed to be developed from the endothelium and are therefore named endothelioma. These are usually non-malignant, but there are also malig- nant forms of endothelioma. Borst and his followers have also not infrequently found endothelioma as a primary malignant neoplasm in the lung. The writer himself^ was at one time convinced of the occurrence of primary malig- nant endothelioma in the lungs, but has since been forced to change his opinion.
At the present writing opinions as to the embryonal development of the so-called endothelium are extremely perplexing. The doctrine that the endothelium, as well as the connective, osseous, and other specific elements, are derived from the mesoderm, is becoming more and more discredited. Hertwig^ derives the mesoderm from the primary entoderm, and according to him, at a very early stage independent mesenchym germinal cells emigrate and proliferate in the spaces between the ento- and ecto- derm, and thus form the basis for the development of the connective tissue substances and blood. Schultze,^ on the other hand, derives the mesoderm from the ectoderm, and according to him nearly all the cells of the mesoderm possess considerable mobiUty of their own, so that they wander through all the organs developed from either of the germinal layers. It will be seen by these two quotations how unsatis- factory as yet the embryonal history of endothelium is. It will also be seen that embryology is tending more and more
^ I. Adler, Remarks on Primaxy Endothelioma of the Lung, Pleura, etc., Journal of Medical Research, VI, 1901.
* O. Hertwig, Lehrbuch d. Entwicklungsgeschichte, 1896.
* O. Schultze, Grundriss der Entwicklungsgeschichte, Leipzig, 1896.
58 PRIMARY MALIGNANT GROWTHS OF THE LUNG
toward giving up the mesoderm as a primary germinal layer and is depending more and more upon the ento- and ecto- derm, with only secondary and varying assistance from a secondary mesoderm. It is impossible to go further into details. Let it suffice to say that at present there is little doubt, though the various workers on this subject have not arrived at a uniform opinion as to what cells should be classed as endothelium and what as epithelium, that there is a form of cell which may rightly be called endothelium, which occu- pies a unique position in so far that it lines the banks of seas and streams of fluid, where it is not only acting as a mere mechanical agent, but has certain other physiological properties which will be touched upon presently.
Suppose the endothelium to be derived from the meso- derm and to be an integral part of the connective tissue system, it follows, and rather absurdly, that a tumor pos- sessing alveolar structure and cells, not to be distinguished from the true epithelial (carcinomatous) cells, — a neoplasm, in short, that acts altogether like a carcinoma, — must be classed among the malignant connective tissue tumors; in other words, must be called a sarcoma. Thus Remak, Thiersch, Billroth, and Waldeyer classed as sarcoma all tumors that develop in localities where normally no epithe- lium is found. This may in part be responsible for such designations as adeno-sarcoma, alveolar carcinoma, lympho- sarcoma, etc. Koster^ does not employ the term ''endo- thelioma," but assumes that all carcinomata take origin from the lymph vessels. Of late the opinion is gaining ground that the intimate structure of the tumor is not dependent upon certain phases of embryological develop- ment nor upon the morphological relations of the three germinal layers. It is held that whatever tumor possesses carcinomatous structure and behaves clinically as a carci- noma is a carcinoma, no matter whether its component epithelial constituents be derived from the mesoderm, the entoderm, or the ectoderm. In other words, it is said that, .while the germinal layers are of utmost importance
1 Die Entwicklung der Carcinome und Sarcome, Wiirzburg, 1869.
PATHOLOGY (Continued) 59
as regards differentiation, topography, and ultimate devel- opment and function of the tissues, their influence to a great extent ceases when the organism is complete and the foetus is fully developed. Extra-uterine pathology should not be tyrannized over by embryology.^ Klaatsch^ also points out that the concept of a mesoderm is gradually disappearing and that the ectoderm is of paramount importance. He shows, moreover, the necessity of being guided in one's judgment more by the physiological requirements and functions than by the merely morpho- logical and embryological point of view. He demonstrates convincingly that the morphological character of cells may be changed to a considerable extent, consequent upon changes in the surrounding tissues, especially when gaps in the con- tinuity of the tissues are formed. He is totally opposed to a classification of tumors in their relations to the three germinal layers. It is to be noted that both functionally and physiologically the endothelium appears closely related to typical epithelium.
It is not necessary to go into all the finer distinctions between endothelium and epithelium. It is best, in the opinion of the writer, to agree with Borst that there are tumors undoubtedly taking origin from endothelium, and as the endothelium occupies a peculiar position, on the one hand appropriating to itself some of the functions of epithelium,^ on the other hand being intimately associated with connective tissue, even forming fibro-plastic cells, it is best to call these tumors by the special name of endothe- Uomata. That there are malignant endotheliomata, we cannot doubt, such perhaps as the much discussed primary cancer of the pleura, concerning which there is still no unity of opinion and a lack of clear and sharp definition. This is
^ Marchand, Uber die Beziehungen der path. Anatomie zur Entwicklungs- geschichte, besonders der Keimblattlehre, Verhand. Deut. Path. Ges., II, 1900, pp. 38 ff.
2 t)ber den jetzigen Stand der Keimblattfrage mit Rucksicht auf die Patho- logie, Miinch. Med. Woch., 1899, N. 6, p. 169.
^ Haidenhain, Verhand. des X. internat. Congresses, Berl. 1891, Vol. II; also Archiv. f. Physiol, v. Pfltiger, Vol. 49, 1891, and Vol. 56, 1894; also Orlow, Recklinghausen, Adler and Meltzer, Meltzer, and others.
60 PRIMARY MALIGNANT GROWTHS OF THE LUNG
shown by the various names, as for instance ''lymphangitis carcinomatodes " ^ or "lymphangitis prolif erans. " ^ As to the lung, however, the writer has not as yet been so fortunate as to be able to diagnosticate an endothehoma of the lung, though Borst and his pupils and others^ have published a number of cases.
If one beheves, as does the writer, that these malignant tumors, carcinoma and others, grow not peripherically, but centrally, out of themselves, as it were,'* then the mere fact of the lymph spaces and lymph vessels at the periphery of the growth being filled with endotheUal cells
1 Schottelius, Table I, No. 289.
2 A. Frankel, tlber primaren Endothelkrebs der Pleura, Berl. Klin. Woch., 1892, 21 and 22. In this connection it might be well to mention the case of Bostrom (Das Endothelcarcinom, Diss. Erlangen, 1876). It concerns a female twenty-eight years of age who had complained of no lung symptoms whatever, but who suffered mainly from the stomach, and the diagnosis of ulcer of the stomach was made. She died suddenly from profuse gastric hgemorrhage. At autopsy the ulcer of the stomach was found and carefully examined, by as high an authority as Zenker, and no trace of anything that could be taken for car- cinoma was detected. Nevertheless, besides about half a litre of bloody sermn in both pleural cavities without any adhesions of the lungs, there was extensive carcinomatous lymphangitis on the pleura of both sides and carcinomatous infiltration of the bronchial, tracheal, and retroperitoneal glands. Cases of carcinoma of the stomach with extensive carcinomatous lymphangitis cover- ing the lungs have been frequently reported (Hilliarie, I'Union m4d., 1874, Nos. 53, 54, and 55; Frantzel, Charite-Annalen, 1878, III, 306; Debove, Gas. Hebd., 1879, N. 43, p. 688). But in these cases there was usually a con- spicuous primary carcinomatous nodule to be found in the stomach. In this case of Bostrom's we have a practically certain assurance that there was no carcinoma in the stomach. By means of very careful examination, the bronchial mucous glands, the bronchial and alveolar surface epitheUum could be positively excluded, and the author, after most painstaking study, by means of serial sections of both pleura, comes to the conclusion that the pleural affec- tion has nothing whatever to do with the gastric ulcer, but is an independent carcinoma of the endotheliima of the pleural lymph vessels.
3 Wack, Ein seltener Fall von primarem Endotheliom der Lunge, Diss. Wurzburg, 1898; Klemm, "Cber ein primares Endotheliom der Lunge, Diss. Miinchen, 1905; Bostrom, Endothelcarcinom der Lunge, Diss. Erlangen, 1876; Cahen, Diss. Wurzburg, 1896; Neelsen, Deut. Arch. Klin. Med., Vol. 31, p. 375.
* Borrman (Die Entstehung und das Wachstum des Hautcarcinoms, Z. f. Krebsforsch., II, 1904) is an enthusiastic adherent of imi-central or possibly multi-central growth of carcinoma. He calls attention justly to the fact that nobody has ever yet seen the conversion of a normal epithelial cell into a can- cerous epithelial cell, and as his material consisted of carcinoma of the skin in its very earliest stages of development, his findings possess considerable weight.
PATHOLOGY (Continued) 61
means nothing as to histogenesis, while on the other hand it will never be possible to study a tumor at a stage early enough to show a possible development of the endothelium into maUgnant cells. Thus the diagnosis of primary endo- thelioma of the lungs is at present not possible, and it is preferable to call these tumors, not endothelioma, or sar- coma, on purely theoretical grounds, but carcinoma, if they are built and act like one, and sarcoma imder similar conditions.
There are many microscopic pictures which are adduced as characteristic of endothelioma, especially those show- ing ramifications simulating a network of deep interlacing meshes, strongly suggesting a system of lymphatics, more or less completely filled with fiat, endothelial-like cells. Plate 15, taken from the same tumor as Plate 9, shows this rami- fication. Neither Plate 9 nor Plate 15 can possibly be taken for an endothelioma, as other parts of the same tumor show typical carcinoma. In the same way Plate 16 shows very prettily the injection of the lymph vessels and lymph spaces with carcinomatous material, but it is from the same tumor from which Plate 7 is taken, in which was shown the mechanical injection of cells from a large typical carci- nomatous alveolus into a lymph vessel, and it is not possi- ble to prove, with any kind of magnification, that lymph endothelium was converted into carcinomatous cells.
CHAPTER VII
PATHOLOGY (Continued)
THE aphorism of Bard/ "Omnis cellula e cellula ejusdem generis," has been mentioned. If each kind of epithe- lium be considered a specific genus, then, according to him, cyhndrical epitheHum should produce only cylindrical epi- thelium; cuboid, or fiat, or horny, should always and under all conditions produce a similar kind of epithelium. It soon became evident, however, that histology did not completely bear out the theory of the strict and hmited production of cells of a certain character and structure from cells of identically the same character and structure. A long, and at this writing still unsettled, discussion has taken place concerning these questions, which are summarized under the title of '^ Metaplasia." It is necessary to touch briefly on some of the problems of metaplasia in order to obtain a proper notion of certain changes in structure and char- acter of the cells that occur here and there, perhaps not infrequently, in lung tumors.
Virchow, as is well known, assigned a very great role to metaplasia in pathology, which meant for him something entirely different from what is understood to-day by the term. He attributed, especially to the connective tissue cells, all sorts of possible metaplastic changes, deriving osseous tissue therefrom as well as the epithelial cells of carcinoma. It is useless to enumerate the multitude of pathologists who have devoted time and no slight labor to this question of metaplasia. Opinions differ as to whether such a process actually exists, and, if it does exist, what the meaning of the process is. Ribbert defines metaplasia as a sort of regression, the cells losing their speci- ficity and attaining a simpler structure, or in other words
1 Loc. cit. 62
PATHOLOGY (Continued) 63
returning to some lower state of differentiation through which, in the regular course of development, they had already passed, and this without regaining new properties. Hansemann speaks of histological accommodation and of anaplasia as being a lower grade of differentiation along embryological lines, to which the metaplastic cells return. It is a mooted point whether this metaplasia of the cells proceeds under the laws of strict embryonal development and is ruled by the theory of the three germinal layers. If this hypothesis were true, then the metaplastic alterations to which, say, an entodermal epithelial cell is subjected would result only in such types of cell as normally originated from the entoderm.
On the other hand, it is maintained that metaplasia is entirely independent of embryonal influences and that the alterations in the character of the cell are produced by mechanical and physical conditions and in a great measure by causes as yet unknown. Finally, there is a theory entertained by many that the so-called metaplasia of cells and tissues, especially when occurring in tumors, is the outcome of congenitally displaced germinal remnants.^ It is not necessary to go into further details on this point. For further reference to these questions in regard to tumors see Lubarsch.2 Most important, and throwing light also on the metaplasia in tumors, is the work of Schridde.^ Speak- ing only for lung tumors, and indifferent to what may take place in other tumors or organs with reference to metaplasia, it is to be noted that only such cells can justly be considered as metaplastic cells that reproduce not only the superficial character of the cells, such as localization, general appear- ance, etc., but the cell must exhibit the intimate and charac- teristic structure of the type of cells which is supposed to be represented. Thus, an ordinary flat epithelium can by no
1 Ernst, Table I, No. 82.
* Lubarsch, Die Metaplasiefrage und ihre Bedeutung fur die Geschwulst- lehre, Arbeiten aus der path. Anatom. Abteilung des Kgl. Hyg. Institut in Posen, 1901, N. 305 ff.
' Schridde, Die Entwicklungsgeschichte des menschlichen Speiserohren- epithels und ihre Bedeutung fur die Metaplasielehre, 1907; Die Ortsfremden Epithelgewebe des Menschen, Jena, 1909.
64 PRIMARY MALIGNANT GROWTHS OF THE LUNG
means be considered as an epidermal cell unless it shows the pecuhar structm-e, the fibres, and protoplasmatic bridges of the latter. A high cuboid or a laterally compressed flat cell is not converted into a cylindrical cell unless it shows at least some of the typical characteristics of the latter, — the nucleus at the base, the colloid, mucoid, or other secretion, etc. It is reasonable to assume, and seems to be the result of common experience, that the nearer the epithelia are related to each other, the more readily they will interchange in form and structure.^ The transforma- tions of one sort of epithelium into another, usually of cylindrical or cuboid epithehum into squamous epithelium, as has been frequently found in many kinds of inflamma- tory processes, in granulations, in pneumonias, ^ in the gall bladder,^ in the urinary bladder, in the uterus, in the pan- creas,^ and other organs, are well known. They are usually the results of acute or chronic inflammations. It would indeed be strange if similar metaplasia of the epithelium were not also found in the bronchi and in the lungs. Under purely physiological conditions and under perfectly normal development, certain epithelial changes in the bronchi are regularly found. The largest and larger bronchial tubes are lined with ciUated cylindrical epithehum. In the smaller orders of the bronchial tubes these cylindrical cells lose their ciha. In still smaller orders the cells become cuboid, and finally, and without break in the continuity, the very small- est bronchioles and the pulmonary alveoles are lined with flat epitheUal cells. Metaplastic changes in the epithelium under pathological conditions are shown by the work of Kitamura,^ who finds in almost every grade of catarrhal
1 Let it be understood that even in the question of metaplasia, the speci- ficity of cells as postulated by Bard is still maintained to a certain extent. Metaplasia can take place only among cells embryologically closely related.
2 Conf. the work of Friedlander, tjber Epithelwucherimg und Krebs, Strass- burg, 1877, 57 S. mit 2 Tafeln.
» Dietz, Virch., Arch., Vol. 164, p. 381.
* Lewisohn, Zwei Seltene Carcinomfalle zugleich ein Beitrag zur Meta- plasiefrage, Z'tschrift f. Krebsforsch., Ill, 1905, p. 528.
^ Kitamura, Uber secundare Veranderungen der Bronchien und einige Bemerkungen uber die Frage der Metaplasie., Virch. Arch. 190, 1907, p. 160.
PATHOLOGY (Continued) 65
inflammations of the severer types, and especially in tuber- culosis, the transformation of single layers of cyhndrical ciliated cells into cuboid or polygonal cells. He does not consider this a true metaplasia, but simply a change in form, a "histological accommodation" in the sense of Hansemann.^ On the other hand he finds genuine stratified epidermal epi- thelium with typical keratohyalin in the uppermost strata. This occurs in the large bronchi that are in open communica- tion with tubercular cavities. Later, islets of this epidermal epithelium are found. There are many other metaplasias throughout the bronchial system, such as chalky degenera- tions and the formation of bone in the bronchial wall, etc. These metaplasias seem to occur very frequently as phe- nomena secondary to tuberculosis. In this connection, too, there is the work of McKenzie.^ His conclusion, after very careful study of four cases in very young children, — the oldest only two years old, — is that real genuine metaplasia exists. Not only chronic inflammatory processes, as Sim- monds beheves, but also acute inflammations in the lungs may lead to metaplasia. The existence of such islets of pavement epithelium in the lungs after acute inflammation may have some connection with the development of pavement celled cancer in the lungs. The assumption of dislocated germinal cells is not needed to explain the development of pavement epithelium cancer in the lungs.
Eichholz,^ in his very excellent experimental researches concerning the conversion of the epidermis into mucous mem- brane, and conversely, is inclined to think that metaplasia is not to be excluded with certainty, but on the whole it does not seem likely to him. In most of the cases where true epidermis was formed it could be demonstrated that it was due to a proliferation of the epidermis from without.
1 Loc. cit.
2 Ivy McKenzie, Epithelmetaplasie bei Bronchopneumonie, Virch. Arch. 190, p. 351. (Note, by the author. — We know of many cases of conversion of cyhndrical into pavement epithehum; we know of none as yet of pavement into cyhndrical epithehum.)
' Eichholz, Experimentelle Untersuchungen iiber Epithelmetaplasie, Lan- genbecks Arch. f. klin. Chir., Vol. 65, p. 959. 6
66 PRIMARY MALIGNANT GROWTHS OF THE LUNG
Cylindrical epithelium, according to him, is able to produce epidermis. If, however, epidermis occurs in tissue of cylin- drical epithelium, it is to be explained either through the proUferation of the epidermal epithelium from without or by the assumption of a dislocated embryonal germ.
It is, therefore, not difficult to explain the occurrence of true cancroid, to use the old name, — that is to say, of nodules consisting of typical epidermal cells with the charac- teristic structure and the formation of cancer pearls. It appears natural, too, according to the views of Kitamura, that these cases generally occur in connection with tubercu- losis, as in the cases of Friedlander,^ Perrone,^ Gougerot,' and a number of others. The tumor either came from with- out and penetrated through the wall, and thus projected into the tubercular cavity,* or developed directly from the wall of the cavity. In the case of Ernst ^ the cancroid took its origin from the wall of the main bronchus of the right upper lobe. As from this location no epidermal tissue could normally be expected, Ernst attributed his tumor to develop- ment from a germinal remnant. In view of this widespread instability in the types and forms of the epithelial cells and the apparent lawlessness with which these transformations from cylindrical to cuboid and from flat to cylindrical, from ciliated to non-ciliated, recur, one is tempted to share with John Marshall ® the belief in a complete anarchy as the essence of cancerous proliferation. This anarchy Mar- shall is inclined to attribute to the lack of nerve influence, no nerves having as yet been demonstrated in any malignant tumor, with the exception of a very few perivascular nerve fibrils. According to this view there would be no meaning in metaplasia and no reversion to embryonal types or conditions. The process would simply be anarchy, which might be subdivided into anarchimorphic, anarchibolic,
1 Friedlander, Table I, No. 87.
2 Perrone, Table I, No. 257.
3 Gougerot, Table I, No. 98.
* Perrone.
" Ernst, Table I, No.82.
* Marshall, The Morton Lecture on Cancer and Cancerous Disease, Lancet, II, 1889, pp. 1045 ff.
PATHOLOGY (Continued) 67
anarchisynthetic forms. Beneke^ does not agree with this view. According to him the nervous system can only regulate the forces contained in the cell, and he suggests a disturbed equilibrium in the relations and proportions of the cell function as a causal factor. In the writer's opinion all these facts and theories lead necessarily to the conviction that epithelium is a highly plastic material, designed to accommodate itself in manifold ways to the demands which local, physiological, and pathological conditions require. The changes thus produced, however, can only take place among the specific epithelial cells, whether derived from entoderm, ectoderm, or mesoderm. The divisions into squamous, epidermal, cylindrical, ciliated, and epithelial depend upon more or less functional and often unstable qualities and are employed more for the sake of con- venience than as a description of the character of the cells. The numerous studies with reference to the ques- tion of metaplasia 2 do not appear to give much enlighten- ment as to tumors, but seem to corroborate the opinion here upheld. The theory of persisting and abnormally dispersed germinal centres and remnants, while it cannot be dis- proven, is not necessary for the explanation of the so-called metaplastic transformations.^
1 Beneke, Neuere Arbeiten zur Lehre vom Carcinom, Schmidts Jahrbiicher, 1892, pp. 73 £f.
2 Kawamura, Beitrage zur Frage der Epithelmetaplasie, Virch. Arch., Vol. 203, No. 3, 1911.
' Fixtterer, Uber Epithelmetaplasie, Lubarsch-Ostertag, Ergebnisse, IX, 2, p. 706. Simmonds, Munch. Med. Woch., 1898, p. 189. Watsuji, Zeitschr. f. Krebsforschung, Vol. 1, No. 5, 1904.
CHAPTER VIII CLINICAL
UNTIL very recently it was the conunon consensus of medical opinion that the diagnosis of primary carci- noma or sarcoma of the lung, if it could be made at all, was one of a more or less high degree of probability, but never of certainty and precision. Within the last few years, how- ever, decided advances have been made in our diagnostic methods, rendering it possible to diagnosticate a timaor of the lung with nearly as much certainty as the present status of our diagnostics permits a cancer diagnosis for any other internal organ of the body. Stokes's remark, speaking of the diagnosis of primary cancer of the lung, that "though none of the physical signs of this disease are, separately considered, peculiar to it, yet the combinations and modes of succession are not seen in any other affection of the lung,"^ has been true for nearly a hundred years and has been a source of stimulation and hope to many. The clinician's ambition to-day is not, at the conclusion of long and anxious obser- vation, to make a diagnosis of lung tumor that is merely probable. His object should be to diagnosticate the tumor at the earliest possible stage of its development, and with such accuracy as is needed for the basis of surgical treat- ment. This, however, is by no means an easy task.
Note. — It will be necessary to refer frequently to the writings of Stokes (Table III, No. 78), Hughes (Table I, No. 121), Graves (Table III. No. 30), Frankel (Table I, No. 85), Passler (Table I, No. 241), Leopold (Table I, No. 174), and Lenhartz (Table II, No. 46), and to that most recent and excellent pubHcation of Wolff (Die Lehre von der Krebskrankheit, Vol. II, Jena, 1911). In making this general statement of indebtedness, the writer hopes to be ex- cused from special references to these authors where such reference is deemed unnecessary.
1 Diseases of the Chest, New Sydenham Society, London, 1882, pp. 420 and 421.
68
CLINICAL 69
In many cases the diagnosis is impossible because there are no symptoms pointing to the lungs and the tumor is an unexpected discovery on the autopsy table. To illustrate this, some cases may be singled out, — that reported by Colomiatti'^ and that of Bernouilli.^ The latter was a case of a female fifty-one years of age, without chnical history except that she died of peritonitis after operation for um- bilical hernia. Autopsy was held the day after. A small round celled sarcoma of the size of a walnut was lodged in the right upper lobe and evidently had not caused any symptoms. There were no metastases, not even of a single gland.
In some cases there are symptoms, but none pointing toward disease of the lungs, and therefore the observer is misled. The patient of Beveridge,^ it is true, had a shght cough and some pressure over the chest, but not sufficient to interfere with his work. He worked until death, which came suddenly from haemorrhage of the lungs. Kliiber ^ reports an apparently healthy woman, dying suddenly from a bum, without any lung symptoms. In the case reported by Walshe,^ there was no cough, nothing pointing to the lungs, but the symptoms were exclusively psychic. Davy's patient^ was healthy until he acquired jaundice and pain in abdomen; physical examination of lungs was negative, no symptoms pointing to lungs, no cough, no pain. Degen^ reports a patient healthy and strong; sudden death from haemorrhage of lungs; no other cUnical symptoms. The much cited case of McAldowie ^ is that of a child five and a half months old, — no dyspnoea, no cough, percussion clear over both lungs.
It is obvious that tumors such as the malignant neoplasms of the lungs, varying so widely in type and localization, entering into so many unstable relations with other organs of the chest and, through metastases, with almost every
1 Table II, No. 14.
2 tlber primare Lungensarkomatose, Diss. Miinchen, 1907.
» Table I, No. 38. « Table I, No. 56.
* Table I, No. 145. ^ Table I, No. 59.
<• Table I, No. 329. s Table III, No. 53.
70 PRIMARY MALIGNANT GROWTHS OF THE LUNG
organ in the body, cannot be expected to present a perma- nent and characteristic set of symptoms. One is reminded of Graves/ who, reporting a case of maUgnant disease of the lungs, probably sarcoma, gives a minute analysis of the cUnical symptoms and shows how both he and Stokes were misled. He candidly confesses that he should have made the proper diagnosis during hfe, but adds, in his characteristic manner, "I became quite tired of the difficulty of attempt- ing to explain the phenomena observed and gave up all further attempts at diagnosis." It may be said in a general way that the possibility of a clean-cut diagnosis depends largely upon the anatomical localization of the tumor and upon the degree of development which the disease has reached when the patient is presented. It is not probable that the actual beginning of the blastomic development will ever be perceived, since it is necessary that the tumor attain a certain size before it can be recognized. Again, in the last stages, the clinical picture may be so complicated, nearly every organ of the body participating in the morbid process and causing symptoms which almost completely mask the pulmonary lesions, that the difficulties are greatly augmented and a diagnosis rendered practically impossible.
There are, however, certain symptoms which are common to all malignant neoplasms and some which are more or less peculiar to malignant neoplasms of the lungs, to which brief attention must be given.
I. Pain. This is frequently not a real, acute pain, but rather a sense of discomfort and pressure in the chest. According to Schmidt ^ the pulmonary parenchyma is prob- ably insensible to pain, therefore the acute or chronic genuine stabbing pain is brought about when the pleura participates in the inflammatory processes which are apt to accompany the progress of the disease. Taking into account the well-known relations between the two folds of the pleura and the nerves, — the brachial plexus, intercostal nerves, phrenic nerve, — and the diaphragm, it is clear that
1 Table III, No. 30.
* Die Schmerzphenomene bei inneren Krankheiten, etc., Wien, 1906.
CLINICAL 71
the pain produced in one place may be referred to localities quite distant from the point of origin. The pain in the shoulder and around the clavicle, the neuralgias of the arm, the intercostal pains along the chest and in the abdomen and diaphragm, which so often occur both in carcinoma and in sarcoma, are thus easily explained, and it is understood that where there is no pain the pleura has evidently not been involved. Schmidt also points out that a large area of dulness, without spontaneous or pressure pain, excludes any inflammatory process of either fold of the pleura and suggests the possibility of a neoplasm. Figures represent- ing an approximate estimate of the occurrence of pain in malignant lung tumors can be obtained from Tables I and II. In Table I pain is not mentioned in 206 cases out of 374. This, of course, does not mean that pain was not present, but merely that any reference to pain was omitted. The probability therefore is that the cases in which pain was a fea- ture are much more numerous than would appear from the Table. In eighteen cases it is distinctly stated that there was no pain during the entire course of the disease, while pain is mentioned as present in one hundred and fifty cases. In Table II, dealing with sarcoma, pain is given as a symptom at some time during the disease in fifty-two cases, in two cases only is it distinctly stated that there was no pain whatever, in six cases there is no clinical history, and pain is not mentioned in the history of thirty-four cases.
The possible irradiations along various nerve tracts are illustrated by the case of Demange,^ in which the pain was constantly referred to the healthy side. In two cases the pain was mostly abdominal, while in the case of Harris ^ the pain was referred to both sides of the chest. If one could draw deductions from these figures, it would seem that sarcoma causes more pain than carcinoma. This result, however, is probably illusory and caused by the imperfect statistics.
11. Cough. This complication is one that would natu- rally be expected in any malady of the lungs, and therefore
1 Table II, No. 17. ^ Table II, No. 33.
72 PRIMARY MALIGNANT GROWTHS OF THE LUNG
in tumors of the lung. Indeed, cough is probably the most common of all symptoms appertaining to lung tumors, and there are but few cases in which it is not a factor. A rather insignificant, but fairly constant, irritating cough, mostly without expectoration, may be the earliest symptom of tumor. Where this cough exists and nothing abnormal is found in the chest, the upper air-passages, oesophagus, etc., the possibility of the presence of a lung tumor should, in the writer's opinion, suggest itself. A case observed by the writer, which does not appear among the material collected, may serve to illustrate this rather important point. It con- cerned a lady of some sixty-odd years, fairly healthy, and so far as known, without any hereditary strain of malignancy. She began to cough this same short, hacking cough, without pain, without expectoration. Both lungs on close examina- tion gave no indication of anything abnormal and nothing abnormal could be detected anywhere, except a trifling pharyngitis. Very gradually some loss of flesh and strength became apparent, and after several months a very small area of dulness at the right hilus, together with some fairly loud cornage, could be made out. The dulness gradually extended. For some time previous a tumor had been sus- pected, principally from the cornage, and the diagnosis was corroborated when the dulness and cornage were also found at the apex. There was never much expectoration, and no blood. The emaciation and weakness increased, the area of dulness on the right lung extended over the entire lower and middle lobes, with diminished voice and breathing, secondary plainly palpable nodules appeared, especially in the hver, accompanied by jaundice, and death from exhaus- tion took place in about a year from the beginning of the cough. No autopsy could be obtained, but there is httle room for doubt that this was a genuine case of carcinoma of the lung.
Besides this slight hacking cough, accompanied by little or no distress, all varieties of cough, up to the most violent, explosive, and harassing forms, are reported. The cough may, as just mentioned, be an early symptom of the disease;
CLINICAL 73
on the other hand there may be no cough until shortly before the fatal end. As bronchitis is one of the ordinary features of the case, the fairly loose cough, accompanied by large and small mucoid rales, is present in the majority of cases. If bronchiectatic cavities, or cavities of other origin, are present, there will probably be attacks of coughing of an explosive character, discharging large quantities of muco- purulent or purely purulent expectoration, often mixed with blood. When the cavities are sufficiently refilled or com- munication with the bronchus is again restored, these spells are apt to recur. The distressing, rasping, but usually dry cough that is caused by compression or irritation of the larger bronchi and the trachea is often noted. At times this cough is accompanied by considerable stridor. Schwalbe ^ claims that carcinoma produces very little stridor, if any at all, but that it occurs in its greatest intensity and most frequently in sarcoma, and his explanation of this is that sarcoma gives rise to earlier and more extensive involvement of the mediastinal organs than carcinoma, thereby exerting more pressure on the trachea and nerves. This does not, perhaps, quite correspond with the actual facts, and it can be seen from the material collected here that carcinoma also can, and frequently does, involve all the mediastinal organs. There is, furthermore, the hoarseness, also the well-known laryngeal cough, both of which usually occur in late stages of the disease, when either one or both superior larjmgeal recurrent nerves have become involved and paralyzed. In Table I cough in its various forms is mentioned in 174 cases, while in 191 cases it is not mentioned. In nine cases it is distinctly stated that there was no cough. In Table II cough is mentioned as a symptom forty-six times; five cases had no cough, and thirty-nine passed without any mention of it.
III. Sputum. Much more important than the cough, — in fact, one of the principal signs to be depended upon for the diagnosis of malignant lung tumors, — is the character of the sputum. This, however, can only be satisfactory as the result of close study. It is necessary to bear in mind that
1 Deut. Med. Woch., 1891, No. 45.
74 PRIMARY MALIGNANT GROWTHS OF THE LUNG
a single examination of the sputum will rarely give reliable results. The ordinary routine examination of the expecto- ration, such as is the common practice, which consists in a search for tubercle bacilh or elastic fibres, and at best a few cells, is entirely insufficient when so delicate a diagnosis as that of primary lung tumor is the object. It is necessary to examine the sputa systematically and thoroughly, both morphologically and bacteriologically, and under certain conditions even chemically, as frequently as possible, until the diagnosis is assured. In Table I there are 143 instances out of 374 in which no mention is made of the sputum. It is, therefore, not ascertainable whether in these cases there was any expectoration or what its character may have been if present. In thirty-six cases it is clearly stated that there was no expectoration. Stokes^ was the first to speak of a pecuharly homogeneous and tenacious sputum, the color of which he compared to black currant jelly and which is spoken of by others as resembUng raspberry jelly or prune juice. The latter designation is particularly used in American textbooks. Stokes considered this sputum as pathognomonic of lung tumor, especially of carcinoma, and many textbooks still spread this behef. It has been shown, however, that this peculiar sputum is per se not pathognomonic for malignant tumors of the lung. It occurs in other diseases, and even in primary carcinoma of the lungs it is not constant and is recorded in but few cases. Looking over Table I, it is foimd that the currant, rasp- berry, and prune juice sputa have been placed on record in only six out of the 374 cases. This may not absolutely coincide with the actual facts, but it is reasonable to suppose that where there is a clinical history given, so characteristic a symptom would be mentioned. In Table II only two cases are recorded out of a total of ninety. But though this kind of sputum cannot be considered pathognomonic, it should, in the writer's opinion, if associated with other symptoms that all point toward tumor of the lung, be considered corroborative of the diagnosis. The processes ultimately
^ Loc. cit.
CLINICAL 75
at work in the production of this peculiar type of sputum are entirely unknown up to date. It seems certain that the peculiar color is not merely due to the presence of blood; there must be other conditions involved. Perhaps it is not unreasonable to suspect that some specific kind of haemolysis, caused, it may be, by some toxic product of the tumor, formed only under certain conditions (perhaps oleic acid — conf. Faust 0 is responsible. The subject has been insufficiently studied and is well worth further research.
Bloody expectoration is associated with most cases of lung tumors at some period of their development. The sputum, either mucoid or mucopurulent, as the case may be, may be intimately mixed with the blood, or the latter may appear in the form of haemoptysis, varying in profuseness. It has been claimed ^ that haemoptysis is uncommon in lung tumors. According to the writer's own experience and his study of the hterature of the subject, which is to a great measure collected in the Tables, this statement cannot be verified. It seems, on the contrary, that haemoptysis is of rather frequent occurrence. A number of cases are reported in which the very first symptom was a profuse haemoptysis, others where haemoptysis occurred frequently in the course of the sickness, and in quite a number of cases, sev- eral of them under the writer's own observation, death was caused by very profuse haemorrhage. The mere bloody sputum, too, may appear as one of the very first symptoms, though it sometimes requires all the skill of a trained cross-examiner to elicit the fact that there has at one time been some slight bloody expectoration. On the other hand, blood may appear at a later stage, or even at the very last stage, and sometimes, again, be constantly present throughout the course of the disease. The records in Table I show about one hundred cases in which the sputum was bloody, not counting the currant, raspberry, and prune juice sputa mentioned before, and not counting
^ 'Ober chronische Olsaurevergiftung, Archiv. f . exp. Path, und Phar. Festschrift f. Schmiedeberg, p. 171.
2 West, Table I, No. 326. Also Hampeln, €ber den Auswurf bei Lungen- carcinom, Z'tschrift f. klin. Med., Vol. 32, 1897, p. 246.
76 PRIMARY MALIGNANT GROWTHS OF THE LUNG
sixteen cases of profuse haemoptysis. In sixty-five of these one hundred cases pure blood seems to have been expecto- rated, representing, as it were, small hsemoptyses. The others were various kinds of sputa, — mucoid, mucopuru- lent, purely purulent, etc., — all of them mixed more or less with blood. In three cases tubercle bacilli were found in the bloody expectoration. In thirteen cases the sputa were entirely free from blood. In forty-five cases the expectoration was ordinarily without blood, and character- istic merely of the condition of the bronchi and the lungs, without reference to tumor. Greenish expectoration is mentioned twice, and one case is reported of olive-green sputum.^ Just what kind of sputa these are cannot be ascertained, as there was no detailed examination recorded. They are probably not characteristic. In Table II sputum is not mentioned in thirty-one cases, in eight cases no expectoration took place, in ten others there was not even a cough, while twenty-five were bloody, three with profuse hsemoptyses. In twelve cases haemoptysis is the main charac- teristic of the sputum. Green sputum is noted five times, and it is believed that Bell ^ was the first to mention it as occurring in sarcoma. There are no means of judging of its character or its relation to sarcoma. In Janssen's case^ the sputum was not merely green, but grass-green, and he believes this to be characteristic of sarcoma of the lung. Traube ^ finds grass-green sputa associated with pneumonia or bronchitis, accompanied by jaundice, — the so-called ''bilious pneumonia," — and also in chronic pneumonia without icterus. He claims that the varying colors of these sputa are due to the red blood cells and the hsematin going through the same cycle of discolorations as an ordinary haemorrhage into the skin, the last being green and repre- senting, according to Traube, the last stages of oxidation of the haematin. He does not mention tumor. That grass-green sputum cannot be characteristic of sar-
1 Elliott, Table III, No. 24.
» Table II, No. 3.
3 Table II, No. 39.
* Gesammelte Beitrage f. Path. u. Phys., Vol. II, 1871, p. 699.
CLINICAL 77
coma of the lungs may be deduced from the fact that it does not appear in the majority of cases, while sputum, mentioned as merely green, is seen in carcinoma, as well as in other diseases of the lungs and bronchi. Moreover, grass-green sputum is said to occur rather frequently in cases of chronic pneumonia and of pulmonary abscess. Here, also, further study is imperative, not only to determine the diagnostic value, but also the conditions under which such peculiar sputa are produced. Perhaps there is some special conjunc- tion of circumstances in cases of sarcoma of the lung which, while not occurring very frequently, produces when present this peculiarly characteristic sputum. The writer feels that in a case of suspected sarcoma of the lungs the grass-green sputiun of Janssen would be strong corroborative evidence. It seems at first glance almost self-evident that sputa from a malignant growth of lungs and bronchi must necessarily contain tumor elements, and that thus the diagnosis of such tumors could easily be made certain beyond doubt. Some reflection will show, however, that this is not so simple as it seems, and must in fact be a rather rare occurrence. There are first to be considered the quantities of various kinds of epithelial cells that can normally be present in the mouth and air-passages; the cylindrical cells, ciliated and without cilia, that come from the bronchi, the nose, etc., the possible admixture of cells from the oesophagus, etc., all of which would prevent the direct recognition of tumor cells. It is, therefore, always hazardous to suspect lung tumor merely from the presence of scattered epithelial or round cells. On the other hand, if the cells in question occur in unusually large quantities and more or less constantly, or if cells which normally are not found in the expectoration are constantly present, the suspicion of tumor is permissible, provided the clinical symptoms correspond. The tumor elements are not apt to be expectorated unless there is open communica- tion with a bronchus and the tumor itself has softened and is in a state of incipient disintegration. Tumor cells, also, that are expectorated under such circumstances are as a rule in such a state of degeneration that their character as
78 PRIMARY MALIGNANT GROWTHS OF THE LUNG
derivatives of a neoplasm can only be recognized if some remnants of their blastomic structm-e and organization remain. This, of com"se, would make the diagnosis abso- lutely certain, especially as secondary lung tumors seldom cause marked symptoms, and never such as are peculiar to primary growths. Some cases in point are on record. It has even happened that a portion of necrosed lung tissue has been expectorated before any other symptoms of pulmonary disease were apparent, as in the case of Claisse.^ In the case of Ehrich,2 villous and bloody masses containing can- cerous material were expectorated. Pearson^ records a case in which pieces of necrosed lung tissue were coughed up, accompanied by tubercle bacilli, and the tumor was diagnos- ticated by him as ''encephaloid." A similar case was that of Turnbull and Worthington,^ in which a lump the size of a walnut, of alveolar structure and containing cylindrical and cuboidal ceUs, was expectorated. Still another, was the case recorded by Peacock,^ in which masses were expectorated consisting of spindle and round cells. There are a number of other cases which can be found by reference to the Tables, most of which are doubtful, however, because they lack the all-important microscopic examination. Most of the cases in which the expectoration is recorded of larger or smaller portions of tumor, which are degenerated but nevertheless distinctly recognizable as either carcinoma or sarcoma, belong as a rule to late stages, and while they clinch the diagnosis they do so at a time when all hope of beneficial therapeutic interference is practically gone. It is quite natural therefore that anxious search is made for elements whose appearance in the sputum, while characteristic of lung tumors, is not delayed until the later stages of development. Hampeln ^ found certain cells in the expectoration from cases of carcinoma of the lungs which, according to him, if only
^ Table I, No. 52. In the discussion of this case, Troisier reports a case of primary cancer of the lung in which the diagnosis was confirmed by tumor particles in the sputum. Menetrier also reports similar cases.
2 Table I, No. 78. ^ Table III, No. 59.
3 Table I, No. 249. e Loc. cit. « Table I, No. 321.
CLINICAL 79
seen but a single time, assure the diagnosis of carcinoma. He says, '^ Polymorphic, polygonal cells that are entirely free from pigmentation are seen in the sputum where there is carcinoma of the lungs, and in no other case but carcinoma. In all other cases, if there are epithelial cells at all in the sputa, they are principally round or oval cells, pavement or ciliated cells, highly pigmented." These cells do not seem to have gained favor in the eyes of diagnosticians. The writer is not aware that Hampeln's views have been corrob- orated by others, and he himseK has never seen the cells in question. He must confess, however, that his examinations with reference to them have not been sufficient to warrant a definite conclusion. Lenhartz^ finds large spherical cells filled with a multitude of fatty granules and associated with abundance of epithelial cells that are strangely deformed and possess club-like or tail-like projections. He is of opinion that these fatty or granular cells are pathognomonic of pul- monary carcinoma. Tuberculosis may be present without changing anything in the character and diagnostic value of these cells. In Table I the granular fatty cells are found in the sputum seven times. The writer is inclined to agree with Lenhartz that these cells are strictly pathognomonic, at least of carcinoma of the lung, there being as yet insuf- ficient experience as to sarcoma. Since the writer's attention was drawn to these cells he has found them in every case of primary carcinoma that has come under his observation (about twelve cases), and a very long and close study of sputa from all manner of other lung diseases tends to show that they occur in carcinoma alone. The technique of examina- tion is very simple, inasmuch as no staining is required, and a spread of sputum, not too thin, perhaps in a little glycerine and water, or perhaps without any addition, if examined carefully with a moderate magnification, will not fail to show these ''Kornchenzellen" if they are present. The cells can sometimes be obtained, also, by puncture of the pleura or the tumor. 2 It is to be remembered that the
1 Miinch. Med. Woch., 1898, No. 1, p. 28.
2 Muser, Table I, No. 209.
80 PRIMARY MALIGNANT GROWTHS OF THE LUNG
conditions under which these cells are formed are still un- known. Lenhartz believes that they are produced by fatty degeneration of the large epithelial cells of the tumor. This, however, is merely hypothesis. Their appearance in the sputum, — for what reason is not known, — is, moreover, very inconstant and irregular. It may be necessary to hunt for them for days in succession before they are found; it may be, on the other hand, that the first examination will show them. They may occur in great profusion, or again only scattered singly here and there through the smear. But it is the writer's conviction that when found they are pathog- nomonic of pulmonary carcinoma, and furthermore that a daily, systematic examination of the sputum is necessary and that one should not be discouraged if the cells are not found at once.
IV. That respiratory difficulties constitute one of the most frequent symptoms in lung tumors is obvious. An insignificant shortness of breath on slight exertion is fre- quently reported as the first symptom. This may be present long before percussion and auscultation give evidence of any lesion in the lungs. The difficulty in breathing is often so slight that only a rigid inquiry will elicit the fact of its existence. Its gradual increase may be the first thing to alarm the patient and cause him to submit to a medi- cal examination. Beginning with this slightest form of dyspnoea, all transitions up to the severest orthopncea occur. Among the material here collected, numerous examples will be found of death from suffocation. No physician who has ever seen the intolerable and hopeless suffering of those unfortunates who are doomed to the awful death by suffocation accompanied by intensest orthopnoea extending over weeks, sometimes even months, will ever forget it. Fortunately, it is not always continuous, but is apt to come in spells. Nevertheless, it is one of the most cruel tortures to which man can be subjected and before which the physi- cian has stood powerless. Not only is he unable to cure, but even to relieve, as morphine loses its virtue and surgery is helpless. Complete closure of a bronchus does not cause
CLINICAL 81
these worst forms of suffocation, but at most only a very moderate degree of dyspnoea following exertion. The in- tensest forms are brought about mainly by compression or obstruction of the trachea. The tumor may grow up from below through a main bronchus into the trachea and thus obstruct it, or, as is perhaps more frequently the case, in- volvement of the mediastinal glands may form large masses pressing upon the trachea from without so as to produce almost entire closure. Though a most frequent symptom, dyspnoea does not necessarily complicate lung tumors. In Table I there is a record of twenty cases in which no dyspnoea of any kind was found throughout the disease. There are 189 cases where dyspnoea is not mentioned. In 165 instances dyspnoea was present, and this number includes all the differ- ent forms of respiratory disturbance, from the slightest incip- ient dyspnoea to the most terrific orthopnoea. In Table II appear two cases in which it is recorded that no dyspnoea was present, fifty-two cases in which dyspnoea is recorded as present at some stage of the disease, leaving thirty-six cases in which no mention is made of this symptom.
V. Cachexia, the usual companion of malignancy, is also a very frequent accompaniment of lung tumors. Its incidence, however, is extremely irregular. There are cases on record, as the Tables show, in which loss of flesh and weight are apparently among the earliest symptoms, cer- tainly before anything abnormal could be detected on the lungs. ^ In other cases there is no apparent loss in flesh and weight throughout the course of the disease. In one of the writer's own cases,^ though there were profuse haemorrhages and the disease lasted about four years, the man kept stout and florid and apparently without any loss of strength until his death, which was caused by suffocation from a profuse and sudden haemorrhage. A positive gain in weight during the progress of the disease has been observed by v. Fetzer'
1 Rottman, Table I, No. 277.
2 Table I, No. 3.
' Bronchuscarcinom, Correspondenzblatt Wiirtemberg artzlicher Landes- verein, Feb. 25, 1905. 7
82 PRIMARY MALIGNANT GROWTHS OF THE LUNG
and also by Rothman.^ Le Sourd ^ reports a distinct ten- dency to obesity throughout the disease. Notwithstanding all that, a great number of cases are recorded in which death ensued from exhaustion.
VI. There is still considerable diversity of opinion as to fever in carcinoma and sarcoma of the lungs. Kast^ and Ebstein and others recognize a somewhat typical intermit- tent, but usually not very high, fever in the course of the growth of sarcoma. DaroUes ^ is of opinion that there is no fever in uncomplicated cases of carcinoma of the lungs. On the other hand Hampeln ^ finds an intermittent fever similar to the malarial type in cases of occult visceral carcinoma. The same is maintained by Kast^ and a number of others, who also find fever of an intermittent character, especially in cases of cancer of the stomach. Without going into the details of this subject for carcinoma in general, but consid- ering only the carcinoma of the lungs, it appears, looking over the list of cases, that such as seem to be uncomplicated have, as a rule, no rise of temperature of any significance. That fever in an absolutely uncomplicated case of cancer of the lungs is possible, cannot be denied, in view of the modern researches on auto-intoxications and metabolic disturbances caused by the carcinoma itself. In the case of cancer of the lungs, however, it is hardly possible to determine whether the tumor is uncomplicated or not, and in the overwhelming majority of cases it will probably be sufficiently complicated by bronchitis, inflammatory conditions of the lung tissue, bronchiectatic dilatations, etc., to account for whatever temperatures may occur.
VII. Difference in pulse in the two radials has fre- quently been reported. This is easily explained by the tumor pressing upon one or the other of the subclavian arteries.
1 Table I, No. 275. » Table I, No. 179. ' Jahrbuch der Hamburger Staatsanstalten, 1889, I.
* Du cancer pleuro-pulmonaire au point de vue clinique. These, Paris, 1877. »Z't8chrift f. klin. Med., 1884, Vol. 8, p. 221; and 1888, Vol. 14, p. 566,
Zur Symptomatologie okkulter visceraler Karzinome.
• Loc. cit.
CLINICAL 83
Japha^ reports a distinct bradycardia in one of his cases, but no cause for it is mentioned. So far as one can see from the clinical and post-mortem notes, it does not seem to have any connection with the lung tumor.
VIII. The blood count has not thus far been of much assistance in the diagnosis of lung tumors. There are but a few cases in which the blood count is reported, — in all less than a dozen, — and even these lose greatly in value inasmuch as it does not appear from the records how the haemoglobin was estimated and how often and under what varying conditions the blood count was done. One almost involuntarily gets the impression that the blood count was done only once, while it is obvious that it should be repeated at stated intervals. Here also is a fruitful field for further investigation.
Of the few blood counts that are on record, it may be well to mention, first, that of Kappis.^ He finds cancer cells with mitosis in the sputum. The blood he reports as follows: Hb., 120; red cells, 6,200,000; white cells, 50,560-40,700; eosinophiles, 33-39^-12%; polynu- clears, 56.9%. The pleuritic effusion was a sanguinolent serum which contained no eosinophiles. In this case the blood count appears to have been taken repeatedly, but is thus far inexplicable in that there is nothing in the history as given by the author to explain the enormous leucocytosis, the accompanying polycythsemia, and the very high per- centage of eosinophiles, the polynuclears, at the same time, being rather low. The autopsy also throws no light upon this curious condition. The author remarks in his descrip- tion of the microscopical structure that enormous heaps of eosinophiles were found in places free from tumor. It is best in this case to indulge in no hypotheses as to the possible cause of this blood picture and its contradictions.
Another imperfect blood count is given by Naun^: Hb., 40; leucocytes, 15,000. It is to be regretted that the number of erythrocytes is not stated, because without knowing the number of red cells one is left in doubt whether this is a
1 Table I, No. 136. » Table I, No. 139. ' Table I, No. 224.
84 PRIMARY MALIGNANT GROWTHS OF THE LUNG
mere haemoglobin anaemia with a moderate leucocytosis, or whether the red cells also are diminished. A complete blood count, including differential, and repeated several times during the course of the disease, should in the future be considered an essential requirement. In a similar way Musser^ records merely increased leucocytosis, without fur- ther details, in both his cases. In two of the writer's own cases, 2 where the advantages of hospital observation could be had, the blood count was taken repeatedly with the average, in Case No. 2, of: Hb., 65; red cells, 4,500,000; leuco- cytes, 15,000. This corresponds very nearly with the blood count given by Cohen and Kirkbride^: Hb., 60; red cells, 4,400,000; leucocytes, 18,000; differential count of leucocytes not stated. In Case No. 4 the blood count was as follows: Hb., 62; red cells, 3,980,000; leucocytes, 14,300; differen- tial fairly normal. In this case, besides the haemoglobin anaemia, there is a distinct reduction in the number of red cells, but no deformation or other alterations in them.
The case of Ebstein^ is very similar to this latter case: Hb., 62; red cells, 3,492,000; but the leucocytes are unusu- ally high, there being 32,000 (differential not stated). It is impossible at present, there being so few blood counts avail- able, to come to any definite conclusion. The leucocytosis is easily accounted for by the inflaromatory and often puru- lent processes going on in the lungs. Whether there is a real disproportion between the number of red cells and the percentage of haemoglobin, thus pointing perhaps to some haemolytic process, or whether in the majority of cases there is only the usual anaemia, both of red cells and of haemo- globin, to be expected in any case of increasing malignancy, — especially if there is considerable loss of blood, — is a problem that awaits further study. In the case of Cohen and Kirkbride the disproportion between 4,400,000 red cells and only sixty haemoglobin is very striking. The blood counts given by Faust ^ show some resemblance to the
1 Table I, Nos. 222 and 223. * Table I, No. 76.
2 Table I, Nos. 2 and 4. ^ Loc. cit.
3 Table II, No. 13.
CLINICAL 85
blood counts mentioned here, inasmuch as his rabbits showed a continuous decrease in the haemoglobin with a comparative increase in the red cells and a tendency to some leucocytosis. The interesting coincidence is certainly worthy of note.
Miiller^ has among his cases no case of lung tumor. As a result of his careful blood counts nothing character- istic is shown. The haemoglobin has a tendency to go down steadily, as also the number of red cells, and there is a tendency to leucocytosis and to an increase of the polynuclear cells, but nothing characteristic of the blood in lung tumors is shown.
IX. Incidentally, there should be mentioned two cases in which diabetes was a complication of the disease, as in the cases of Kratz^ and Liibbe.^ There is no evidence, so far as can be seen, that the diabetes stands in any relation to the lung tumor.
X. The clubbed fingers which are sometimes reported have, it is obvious, no specific relation to malignant growths. They are not different from the clubbed fingers that we see in other chronic diseases, especially of the lungs, and more particularly where pus is present.
1 Oswald Miiller, tjber den Blutbefund bei Krebskranken, Diss. Berlin, 1909.
2 Table I, No. 151. » Table I, No. 187.
CHAPTER IX
CLINICAL {.Continued)
WHEN one is compelled to face the almost infinite variety of pathological lesions and compUcations that are associated with most of the primary malignant neoplasms of the Imigs, the clinical pictm^es and their symptomatology appear to present an almost hopeless con- fusion. A larger experience and comparative study will show that there is, after all, a certain monotony of essen- tial sjTuptoms, around which the varying complications and lesions are grouped. It is possible in this way to arrange the entire clinical material at our disposal into certain groups which, with their subdivisions, supply a fairly well-classified arrangement of the clinical phenomena. A certain number of tumors, as has been shown above, are apt to withdraw themselves from diagnosis by causing no symptoms whatso- ever, and others in which a diagnosis is not likely because symptoms caused by metastatic deposits^ completely domi- nate the chnical picture and successfully mask the pulmonary disease. For the great majority of tumors which do produce symptoms, the remark of Stokes, that ''the faciUty of diagnosis mainly depends on the anatomical disposition of the disease," is still true.
According to Passler,^ the clinical pictures accom- panying pulmonary mahgnant neoplasms can be aptly arranged in two main groups. The first group contains
1 There is much difference of opinion among authors as to the frequency of metastases in maUgnant tumors of the lung, some claiming that secondary deposits are very rare in carcinoma and correspondingly numerous in sarcoma, others expressing directly opposite opinions. By consulting Appendices C and D the reader will obtain a fair idea of the occurrence of metastases in the various organs both in carcinoma and in sarcoma and he will find very little difference between carcinoma and sarcoma in this respect.
2 Loc. cit.
86
CLINICAL (Continued) 87
the cases in which the symptoms referable to diseases of the lungs and bronchi largely predominate. These tumors, mostly carcinoma, nearly always take their origin from the bronchial ramifications from the second order downwards to the smaller and smallest bronchioles, and as a rule do not directly implicate the hilus. The second group embraces to a large extent the tumors of the root of the lung. This group may be accompanied by intense and agonizing symptoms on the part of the respiratory organs: lungs, bronchi, etc.; but these are usually of a secondary nature, though they may dominate the clinical picture. The typical symptoms of this variety of lung tumor are largely mechanical and composed mainly of such symp- toms as result from pressure on or compression of the tho- racic organs, especially of the mediastinum, and from the overcrowding of the intrathoracic spaces. The elementary symptoms mentioned above are common to both groups.
The classification of Marfan,^ identical in principle with that of Passler, is perhaps a little more convenient, and is adopted here. It reads as follows:
I. The acute or galloping form of pleuro-pulmonic cancer. II. Chronic pleuro-pulmonic cancer.
1. Broncho-pulmonary type, being the classical type of carcinoma
of the lungs.
2. Type suggesting tiunor of the mediastinum.
3. Pleuritic type.
(a) Pleuritic type of the pleuro-pulmonary tumor without effusion.
The first main division, the acute or galloping miliary car- cinoma of the lungs, runs an extremely rapid course, accom- panied by cough, dyspnoea, and asphyxia; death usually in a month or thereabouts. The clinical picture in many respects resembles that of acute miliary tuberculosis, and at autopsy both lungs and pleura are found studded with miliary nodules which, however, on microscopic examination, are found to be cancerous. This form is extremely rare and only a very few scattered cases have been reported. The case of Elisberg2 may possibly come under this heading. In
* Quoted from Chauvain, loc. cit. * Table I, No. 80.
88 PRIMARY MALIGNANT GROWTHS OF THE LUNG
this case the primary tumor was in the bronchus. It is generally denied that this form of carcinosis ever occurs as a primary pulmonary lesion. This statement, however, cannot be supported by absolute proof. Granted that it does occur as a primary lesion, it seems that at present there are no means of obtaining a correct diagnosis during hfe.
II. The chronic pletjro-pulmonary cancer. This is the ordinary chronic form of cancer of the lung, in which the lungs, bronchi, and pleura are mainly affected by the tumor. The subdivisions which have been mentioned are, it is necessary to insist, merely for the convenience of the clinician and do not represent strictly defined and firmly established independent syndromes. With the progressive development and extension of the blastomic lesion, accom- panied by a varying degree of destruction of the lung and the secondary effects of the tumor on its environment, the symptoms must necessarily vary, and the so-called subor- dinate groups may merge one into the other. It may often be observed that several or all of the various types here mentioned are exemplified in the course of a single case.
1. Pulmonary cancer. The classical type of cancer of the lung. This represents the ordinary bronchial carci- noma which, as shown above, is by far the most frequent form of the disease. The dominant symptoms are referable mainly to the lungs and bronchi. The earlier stages usually suggest merely a chronic bronchitis.
It is commonly said that in the very earliest stages of the development of the tumor, percussion will fail to show any appreciable difference from the normal. This may, in the main, be true. It is, however, the writer's deep conviction that, even in very early stages, percussion may be found significantly altered, if a sufficiently dehcate technique be adopted.
It cannot fall within the scope of this study to enter in detail into a discussion as to the relative values of the vari- ous methods of percussion or into the manifold theories that have been put forward in this most important chap-
CLINICAL (Continued) 89
ter of diagnostics. But it is the writer's opinion that the ordinary loud, resounding, finger to finger or hammer to finger or plessimetre percussion cannot be made to give proper results in these earher stages. The writer has employed for years the " Schwellenwerthperkussion " and orthopercussion as elaborated by Goldscheider, Plesch, and Curschmann, in combination with the auscultatory percussion according to Ewald and the friction method of Bianchi. The results, checked by the orthodiascope, have as a rule been most satisfactory. These methods, if carried out with the dehcacy of touch and hearing which they require, may be expected to lead to the detection of compara- tively slight pathologic lesions where other methods of per- cussion will fail. It is understood that percussion must vary according to the different stages of development and the various complications that may occur in the course of malignant disease of the lungs.
There are cases on record, as for instance that of Rottman/ where it is reported that physical signs on the lungs were negative, although a large tumor was found. This is only one of many similar examples reported. In early stages a dull percussion note is found at one apex or the other, or, which is much more difficult to find, at the hilus posteriorly. The anterior aspect of the upper chest is more frequently the seat of dulness than the posterior, but the dulness at the hilus, of course, can only be heard near the spine. This dulness may gradually increase from a shght change in the percussion note to absolute flat- ness. The flatness and boardlike resistance to the per- cussing finger are very often due, not to the tumor itself, but to the atelectasis caused by the tumor. Woillez 2 desig- nated as characteristic of lung tumor what he called the 'Hympanisme thoracique," which consists of a tympanitic, immediately preceding the full, percussion note. This has not turned out to be a pathognomonic sign and is wellnigh forgotten.
1 Table I, No. 277.
2 Dictionn. de Diagnost. m6d., Paris, 1870, 2d Ed.
90 PRIMARY MALIGNANT GROWTHS OF THE LUNG
Characteristic of these earher stages is, further, the fact that with dull or flat percussion, auscultation shows diminished respiration. Where pleuritic effusion or pleu- ritic adhesions and thickenings can be excluded, which is comparatively easy for the upper anterior portions of the chest, this sign of increasing dulness with diminishing voice and breathing sounds is extremely suggestive, and while not absolutely pathognomonic of tiunor, should make the presence of tumor highly probable. The mechanism of the sign, — increasing dulness with diminishing voice and breathing without pleuritic effusion, — is of course given in the more or less complete obstruction of a bronchus, by which means those portions of the lung not affected by tumor are in a more or less complete state of atelectasis. Most interesting in this connection is the case reported by Korner.i In this case there was flattening of the right chest, absolute flatness of percussion, and entire absence of respiratory and vocal sounds, — in a word uncomphcated and complete obstruction of the right main bronchus, a diag- nosis that was confirmed by autopsy. The area of dull per- cussion note in these cases is usually sharply defined, as distinguished from tuberculosis and pneumonic conditions, where the delimitation is more diffused, the abnormal per- cussion merging gradually into the normal. The configura- tion of the area of dulness or flatness is, however, usually quite irregular, according to the topographical disposition of the tumor, its depth, its extension, and its surrounding reactive processes.
As the tumor grows and degenerations of various kinds make their appearance, as breaking-down and irregular excavations in the tumor come about, — and it has been stated above that this happens much more frequently than most authors concede, — the percussion note and ausculta- tory signs must necessarily change in character and become variable to a considerable extent. Tympanitic percussion note, amphoric breathing, metallic rales will show the presence of a cavity, and when a case has reached this stage
1 Table I, No. 147.
CLINICAL (Continued) 91
one is apt to pardon the clinician who does not hesitate to diagnosticate tuberculosis. Besides more or less profuse haemorrhages, it is not unusual to find at this stage irregular fever of considerable intensity and night sweats. The fever may resemble the hectic type. Notice is to be taken, also, of the bronchiectatic dilatations which occur so often and to so great an extent, as a consequence of obstructed bronchi. Here percussion as well as auscultation offers frequently interesting changes. If the bronchus is completely closed for a long time, the bronchiectatic cavity naturally fills with secretion, — pus, mucus, blood, and so on, — possibly continually dilating, and the percussion note over this will be dulness to flatness, and auscultation will hear neither voice nor breathing. Suddenly, as it were, the bronchus is reopened by ulceration and degeneration of the obstructing tumor, there is a free discharge of the bronchiectatic con- tents, and in the place where formerly there was abso- lute flatness, we have now the tympanitic note and the auscultatory symptoms pointing to a cavity.
It is obvious that these signs can only occiu" in very late stages of the disease. The process may be varied in different ways and it may be taken as characteristic of these later ulcerative stages when such sudden changes in auscultation and percussion appear. As a good illus- tration of these conditions may be mentioned the case of Amal.^ In this case there was total absence of breath- ing, but normal percussion over the entire right lower lobe. There were all the other symptoms of a malignant growth in the lungs. Very suddenly, and only a few days before death, the respiratory murmur was again distinctly heard over the right lower lobe, — in other words, the tumor, partly compressing, partly proliferating into the right main bronchus of the lower lobe and completely filling it and preventing the passage of air, had ulcerated away to a great extent and thus again permitted communication with the air. It has frequently been said that percussion over a neoplasm of the lung offers a greater resistance to the finger
» Table I, No. 13.
92 PRIMARY MALIGNANT GROWTHS OF THE LUNG
than is normal. This sign, however, depends on so many varying factors, as the closeness of the tumor to the chest wall, the condition of the lungj etc., that it is not constant and not characteristic, though when present a welcome corroboration.
Another sign of great diagnostic value is the auscultatory symptom, to which Behier ^ gave the name of ''cornage." This is a sound very similar to that obtained from the trachea when partially compressed. It is pathognomonic of bronchial obstruction and might be considered, especially when heard about the root of the lungs, and better still when accompanied by some dulness, as an almost certain sign of tumor. It must be remembered, however (and for that reason the word ''almost" is inserted), that certain other conditions which may result in bronchial obstruction must be excluded. This should not be difficult, for probably all the processes which may result in bronchial obstruction, and thus in an audible cornage, are acute. Thus it is not unusual to find the sign in acute, severe bronchitis or in an influenza pneumonia, or even in chi'onic bronchitis when a bronchus happens to be obstructed by masses of viscous and tenacious mucus. But in all these cases the obstruction is temporary and disappears as a rule in twenty-four hours. But in tumor the cornage is practically constant and will remain so until the bronchus is completely obstructed, or will disappear after a comparatively long time when the bronchus, through ulceration, becomes again freely perme- able to air. Cornage may be a very early symptom.
2. The mediastinal type of lung tumor. A bronchial cancer, — and it is indifferent of what order the bronchus may be, whether large or small, — has two main preformed routes of extension at its disposal. The easiest and most natural, and the one that is in the majority of cases primarily resorted to, is along the bronchial ramifications and the peribronchial tissues into the interior of the lung. This holds good also for those sarcomata that originate in the minute peribronchial glands or in the peribronchial connec- 1 Gaz. de Hop., AprU, 1867.
CLINICAL (Continued) 93
tive tissue. In the later stages the bronchial wall is apt to be broken down and penetrated by the tumor, and thus the bronchial and then the mediastinal lymph nodes become involved and are occasionally enormously enlarged. The mediastinal lymph nodes, possibly both anterior and pos- terior, now take part, the mediastinum is filled with tumor masses, the pericardium may be covered or even penetrated by the neoplasm, pericarditis develops, secondary growths in the heart appear, the large vessels, both aorta and cavse, the pulmonary arteries and veins are surrounded and either compressed or penetrated by the tumor. It should be mentioned that the aorta, while often much compressed, so far as the writer's knowledge goes, never takes part in the tumor proliferation and is never penetrated by it. As a con- sequence of all this crowding of the mediastinal organs, the superficial veins of the chest are dilated, sometimes to a huge extent, and cedcema, varying from cedoema of a single arm, or the face, to a general oedoema of the entire body, arises. One or the other, sometimes both, of the laryngeal recurrent nerves are involved, the trachea, large bronchi, oesophagus, are compressed, obstructed, and even penetrated by the tumor. The participation of the oesophagus causes the dysphagia so frequently reported. And thus all the symp- toms of an intrathoracic growth, or more especially of primary mediastinal tumor, are evolved. Sarcoma, origi- nating at the hilus of either lung, differs from this group of symptoms in so far as the direction of the growth is less towards the lung and tends to advance more rapidly and at an earlier stage of the disease toward the mediastinum. It is this mediastinal type of tumor that usually causes the dreadful attacks of asphyxia and orthopnoea mentioned above.
3. The pleuritic type. In cases belonging to this type, the symptoms referable to the pleura predominate. So far as tumors of the lungs and bronchi are concerned, this form corresponds to a rather late stage of the disease. In primary mahgnant disease of the pleura, however, which is beyond the scope of this monograph, this form usually marks
94 PRIIMARY MALIGNANT GROWTHS OF THE LUNG
the beginning of the lesion. The symptoms in the main are those of acute, sub-acute, or chronic pleurisy. There is stabbing pain in the chest, radiating to the shoulders or in other directions, and all the signs of a persistent pleuritic effusion, which too often tend to mask more or less com- pletely the symptoms of pulmonary disease. We have the absolute flatness on percussion, the total absence of voice and breathing on auscultation, very often the obliteration of the intercostal spaces, frequently the bulging of these same spaces.
In nearly every case of lung tumor, the pleura partici- pates to a certain extent in the morbid process, sometimes with sometimes without effusion; according to Herrmann ^ in fifty per cent of the cases. In this pleuritic type, how- ever, effusion more or less profuse is always present and is hkely to recur after tapping of the chest, so that these tappings must be repeated again and again, at longer or shorter intervals. In ordinary pleurisy the aspiration of the effusion affords prompt reUef of the harassing symptoms. Even in the pleurisy associated with extensive tuberculosis, this rehef can be recognized. It is characteristic of the type of tumor under discussion here, — though it applies also to primary carcinoma of the pleura, — that relief after removal of the pleuritic effusion either does not follow at all, or lasts but a very short time. As a rule there is no abatement of the cough, dyspnoea, expectoration, and general distress, but there may be intense pain caused by the wrenching of the diseased tissues. Some exceptions to this fairly general rule are on record, such as the case of Unverricht,2 where, after one or two aspirations of sanguin- olent fluid, all symptoms seemed to disappear, the patient felt entirely well and gained in weight, until secondary tumors made their appearance in the skin where the aspirat- ing needle had penetrated. Hampeln^ also reports a case
1 Deut. Archiv. f . klin. Med., Vol. 63, 1899, p. 583.
^ Beitrage zur klin. Geschichte der krebsigen Pleuraerglisse, Z'tschrift f. klin. Med., Vol. IV, 1882, pp. 79 ff. 3 Table I, No. 101.
CLINICAL (Continued) 95
in which the pleuritic effusion was absorbed without tapping and without recurrence. These cases, however, are rare exceptions.
The fluid recovered by the first few tappings may be clear yellow serum, but sooner or later it is certain to become bloody. It is well known that bloody pleural effusion occurs in other diseases, especially in tuberculosis, and is in itself, therefore, not pathognomonic of malignant tumor of the lungs or pleura. It is said, however, that the change from initial clear serum to bloody effusion is charac- teristic of neoplasms of the lung. It is uncertain whether this is correct or not. It is reported, on the other hand, very often that a thick, chocolate-hke fluid is recovered in the later tappings. This, according to the writer's opinion, is certainly pathognomonic for malignant disease in the pleural cavities. Adipose and chylous effusions into the pleura are reported, but are found very rarely in malignant neoplasm of the lung, — certainly much less frequently than in the disease of the peritoneum. The same holds good for empyema. In the case of Walch^ it was evidently a pneumococcic affection and had no direct relation with the carcinoma. Nothing characteristic has as yet been found by the bacteriological examination of the pleuritic effusions.
The results of the cytological examinations have been a subject of much discussion, with no positive conclusions. Ehrhch^ has called attention to the diagnostic importance of the presence of organically connected cell-groups in the effusion. Frankel has called attention to large vacuolized cells, sometimes attaining gigantic dimensions. These are probably tumor elements and this is assured if they are found to contain glycogen, but they probably belong to primary diseases of the pleura. It is therefore not very difficult to diagnose the presence of malignant tumor in the chest from the study of the cells in the effusion, if such can be found. It is, however, almost impossible, under the condi-
* Cancer du poumon gauche, pleur^sie purulente pneumocoques, Soc. anat. de Paris, 1893, VII, Ser. 5.
» P. Ehrlich, Charit6-Annaleii, 1880, Jahrg. VII, p. 226.
96 PRIMARY MALIGNANT GROWTHS OF THE LUNG
tions given, to distinguish an endothelial from an epithelial cell, and therefore a primary endothelioma of the plem*a from a carcinoma of the Imigs, and it is wise not to depend for diagnosis on the cytology of the pleural exudate alone. This rule should hold, even though exceptions are possible, as in the case of HeUendall,^ who found in the bloody effu- sion in the chest white particles consisting of heaps of round cells, sufi&ciently characteristic to warrant the diagnosis of sarcoma of the lung, — a diagnosis which was confirmed by autopsy. Kronig,^ on making a probatory puncture, penetrated the tumor with the needle and found attached thereto white particles which microscopic examination showed to be lympho-sarcoma, and he was thus enabled to obtain an absolutely certain diagnosis during life. He devised a method based on this, by which in every doubtful case the attempt was to be made to remove particles of tumor by aspiration. There are serious objections to this method. It is not only very uncertain in its results, as the needle does not always return with tumor particles, but usually only with a little blood, but there is actual danger of causing a haemorrhage.
It may be taken as a trustworthy sign of malignancy if a paralysis of the recurrent laryngeal is observed on the side of the pleuritic effusion. It has been stated above that as a rule there is no relief after removing the effu- sion in cancerous pleuritic effusions. It may also be said that, after removal of the fluid, the various phenomena of percussion and auscultation, which until then had been masked, will appear in unmistakable distinctness, and thus greatly assist in the diagnosis. The dislocated heart * which, on removal of the pleuritic effusion, will make no attempt to return to its normal place, — other symp- toms being favorable, — suggests tumor. The retraction of the affected side of the thorax, accompanied by increased dulness and impaired or entirely abolished respiratory motions, when caused by a thickening of the pleura, some- times to an enormous degree, is not at all characteristic of 1 Table II, No. 35. 2 Table II, No. 42.
CLINICAL (Continued) 97
malignant growth in the lungs after the stage of effusion is over, but is well known to occur in other forms of pleurisy, especially in tuberculosis.
(a) The pleuritic type without effusion. This is most typical and applies almost exclusively to those large mas- sive sarcomata or lympho-sarcomata that are apt to fill the greater part of the chest. It marks, of course, a late stage of the disease. There are all the signs of a pleuritic effu- sion, often increased circumference of the side of the chest involved, displacement of the heart, etc. There may also be present, but not necessarily so, the ordinary general symptoms of maUgnant growth of the lung, — the cough, dyspnoea, fever, sweats, haemoptysis, cachexia, etc. The exploring needle fails to discover any fluid. On the con- trary it seems to penetrate into a more or less solid mass extending to such depths as to preclude any possibility of its being merely an abnormally thickened pleura. Par- ticles of tumor may be brought away by the needle. It is characteristic of this type that, while there is complete absence of respiratory murmur or vocal fremitus, there is a very loud propagation of the heart sounds, so that if the tumor occupies, for instance, the right chest, the heart sounds can be heard very distinctly over the whole of the right chest, both in front and in back.^ This sign alone is sufficient to assure the diagnosis of a solid intrathoracic mass. Consequently in most of these cases there is dilata- tion of the superficial veins of the chest and possibly of those of the abdomen, more or less intense dyspnoea, paralysis of one or both recurrent laryngeal nerves, direct or indirect affection of the heart itself, the large vessels, etc.
A few words should be said concerning some morbid processes which are found in the train of pulmonary tumors. Pneumonias, both acute and chronic, are among the most frequent accompaniments of lung tumors. In a number of cases the pneumonia is recorded as the first symptom. The patients state that they were taken acutely ill with chill, high fever, cough, rusty sputum, from which they recovered,
1 Withauer, Table I, No. 342. Budd, Table III, No. 13. 8
98 PRIMARY MALIGNANT GROWTHS OF THE LUNG
but that from then on they were never quite well. These acute pneumonias may be pneumococcic pneumonias or pro- duced by other well-known bacteria. The chronic form, if not of the cheesy tubercular character, is principally of the indurative type. These pneumonias may lead to symptoms which mask the signs of the tumor, or at least are most perplexing. Sometimes, though rarely, they are followed by a genuine empyema. Atelectasis ^ has been mentioned above and is the natural consequence of the blocking by tumor of larger or smaller bronchi, resulting in the collapse of the entire territory which the bronchus supphes with air, as well as its splenification, if no change occurs in the bronchus. There will be moderate dulness on percussion, though sometimes, — particularly if the area is small, — the percussion note will remain fairly normal. But vocal fremitus and breathing sounds are completely abolished. It is on account of these secondary processes that the extent of the dull area does not coincide with the actual size of the tumor. The tumor, as the X-rays have shown,^ may be larger than the dull percussion would lead one to expect. On the other hand these secondary processes give a dull percussion note of their own, which, merging into that caused by the tumor, is apt to give an exaggerated idea of the tumor's size.
Another complication which requires mention, though abeady hinted at above, is gangrene. It is easily conceiv- able, in fact it is almost self-evident, that a proliferating tumor in the lung, rapidly destroying lung tissue and pene- trating into blood vessels, can at any time envelop and, by compression, obstruct an artery of some size, or, by breaking through the arterial wall, close an artery completely, and by either of these means cause total ischsemia, followed by gangrene. According to the size of the artery involved, the gangrenous territory will be larger or smaller, occasion- ally occupying the greater part of a lobe. When a case is first seen in this condition, the diagnosis is intensely diffi- cult, — wellnigh impossible, — as even those signs in the
^ Korner, loc. cit. ' Leo, loc. cit.
CLINICAL (Continued) 99
sputum which we have found to be pathognomonic are apt to be lacking. Under these conditions, too, the X-rays will not give any useful information, and it is only by most careful study of the history and the progress of the disease that a probable diagnosis can be arrived at. On the other hand, if the gangrene appears, after previous examination and observation of the patient have settled the diagnosis of tumor, or at least have caused tumor to be suspected, the gangrene will rank only as a complication. It may be casually added that there may be interesting involvements of the sympathetic which will in no wise interfere with the cardinal symptoms and the diagnosis, but which are of interest as again demonstrating the manifold complications that are constantly arising.^
It was not very long ago that A. Frankel ^ wrote that the X-rays were of little service in the diagnosis of lung tumors. Since then the X-rays have become a most remarkable and efficient aid to diagnosis in general, and there exists the well-founded hope of their increasing efficiency as further improvements in the apparatus and advances in technique are made. They have also proved, as is well known, a powerful therapeutic agent in many diseases, but not as yet for treatment of lung tumors. The hope may reasonably be entertained that with the systematic and proper appli- cation of the X-rays to the exploration of the chest, the diagnosis of lung tumor may be assured when no other means will give equally certain results. Leo^ diagnosticated an osteosarcoma of the lungs, secondary to a sarcoma of the right knee, during life, with certainty and much topograph- ical detail by means of the X-rays, which also showed a much greater extent of the tumor than could be ascertained by percussion and auscultation. It may also be possible, per- haps, to obtain this diagnosis at a time when the tumor is as yet very small and causing but little subjective distiu-bance.
If this happy result is ever to be reahzed, it will be neces-
^ Kronig, loc. cit. ^ Loc. cit.
' Nachweis eines Osteosarkoms der Lunge durch Rontgenstrahlen, Berl. Klin. Woch., Vol. XXXV, 1898, No. 16, p. 349.
100 PRIMARY MALIGNANT GROWTHS OF THE LUNG
sary to examine the chest with the Rontgen rays even where there are no symptoms pointing to any disease in the chest. It has been the writer's practice for a great many years, as an essential part of the routine examination in every case that presents itself at his office, no matter what the patient's complaint, to subject the chest to a thorough exploration with the Rontgen rays. We prefer the examination with the orthodiascope (de la Campe) and a very large (12'''xl6") fluorescent screen. Thus one is enabled at a single glance to observe heart, lungs, in fact, taking advantage of various positions, nearly all the thoracic contents during action. It is particularly useful, also, for watching the respiratory mobiUty of the lungs and diaphragm. It has repeatedly been noted that in lung tumor the mobihty of the lung is markedly diminished or entirely abolished. In cases of medi- astinal tumor the respiratory mobility of the lung remains unchanged or is increased, and Jacobson ^ has found this valuable in distinguishing between the two types of tumor. With good light, good apparatus, and some experience, com- paratively minute lesions in the lungs can be discovered. Any abnormality that is thus brought to notice can be per- manently fixed for further reference by the photographic plate, approximately accmrate measurements can be ob- tained, and thus the gradual enlargement of the tumor verified and its blastomic nature determined. The shadow of a carcinoma or sarcoma just starting from the hilus and gradually extending toward one of the pulmonary lobes is a very striking picture when seen with the Rontgen rays, and often suggests the tumor diagnosis when the observer, though other characteristic symptoms were present, would have been led astray. The interpretation is more difficult when the shadow extends over the upper lobe of either side, as this is the favorite localization of tuberculous processes. Sometimes the sharp hnear delimitation at the base of the shadow makes for tumor rather than tuberculosis. It speaks for tumor, also, if the affection is confined to one
* Primare Lungen vmd Mediastinal Tumoren, Festschr. f. Lazarus, Berlin, 1889.
CLINICAL (Continued) 101
upper lobe, for as these pictures are seen only after the dis- ease has progressed to a certain extent, the upper lobes of both lungs, if the process were tuberculous, would probably have been affected. The shadow remaining unilateral speaks for tumor. The absence of tubercle bacilli in the bloody sputum, with the increasing shadow on one lobe only, also suggests tumor. But where tuberculosis is associated with advancing carcinoma or sarcoma of the lung, the Rontgen rays are of Uttle value, and if a differential diag- nosis is possible, it must be attempted by other means. It is beyond the scope of this study to enter into further details concerning the X-rays. The reader is referred to the well-known books of Holzknecht,^ Grodel,^ Grunmach,^ and Amsperger.* The details, however, as to the value of the X-rays in malignant lung tumors may be studied by the reader in the cases recorded by Otten ^ and Muser,^ from the Eppendorf Krankenhaus, Hamburg, under the direction of Lenhartz.
Another recent aid to diagnosis is the bronchoscope, that has been so successfully employed in various affections of the trachea and the larger bronchi. It has also done service in establishing beyond doubt the presence of a bronchial neoplasm. 7 Karrenstein^ reports the case of a male forty- eight years of age, in which the tumor, taking origin from the large bronchus immediately below the first division of the right main bronchus, was made distinctly visi- ble by the bronchoscope, the tumor having been suspected. H. von Schrotter ^ reports a case of a male f orty-foiu- years of age where the bronchoscope showed very plainly
^ Mitteil. aus Laboratorium fiir radiologische Diagnostik und Therapie, Jena, 1907.
2 Rontgendiagnostik in der inn. Med., Miinch., 1909.
2 tiber die diagnostische und ther. Bedeutung der X-Strahlen f. d. inn. Med. u. Chir., Deut. Med. Woch., 1899, No. 37.
* Die Rontgenuntersuchung der Brustorgane, Leipzig, 1909.
6 Table I, No. 228.
6 Table I, No. 205.
^ Killian, Zur diagnostischen Verwertung der oberen Bronchoskopie bei Lungencarcinom, Berl. Klin. Wochenschr., 1900, p. 437.
8 Table I, No. 141.
« Table I, No. 325.
102 PRIMARY MALIGNANT GROWTHS OF THE LUNG
a prominent tumor in the right bronchus from which a piece was exsected for microscopic examination, which showed cancerous epitheUa with glycogen reaction, and thereby settled the diagnosis.
It is always unwise to endeavor to prophesy as to future possibilities, at least within the domain of biology and pathology. It cannot be denied that the field of bron- choscopy may be greatly extended by improvements in appa- ratus and in technique. It is, however, the writer's opinion that its usefulness in the diagnostics of lung tumor, at this writing at least, is limited. It appears at present that from the nature of things, bronchoscopy can make visible only such tumors as have involved the upper bronchi. Of what occurs in the bronchi of lower orders and in the depths of the lung, the bronchoscope leaves us in utter ignorance. Moreover, there are undoubtedly many cases that come under observation, late in the course of the disease, where the dyspnoea, brain involvements, and other concomitant symptoms are of such gravity, and menace life to such a degree, that even the boldest would hesitate to introduce a bronchoscope, though there remained but little doubt that the instrument could make visible the involvement of the upper bronchi. In such cases the diagnosis should be made by other means, — especially as even the exact recognition of the tumor by the bronchoscope would be of little avail to the patient.
In concluding the clinical part of the subject, it is still necessary to mention a few points which may be helpful in differentiating lung tumors from other diseases closely resembling them in symptomatology, and for which they might easily be mistaken. First and foremost, of course, is the question — tuberculosis or tumor? This question can be easily answered at autopsy, but it is not quite so simple in the living person. Some points in the differential diag- nosis have already been brought out. The small tumors, particularly cancroids, described as growing from the walls of a tuberculous cavity, will probably never be diag- nosticated, unless pathognomonic cells in the sputum direct
CLINICAL (Continued) 103
attention to the possible existence of tumor in the respiratory system. At any rate it is always advisable to remember the exhortation of Gerhardt, — always to suspect tumor in persons of advanced age where tuberculosis is not