Tobacco Documents Online

le wit eI-l-llll red e of~ tre- or !e rage PRIMARY BRONCHIOGENIC CARCINOMA An Analysis cf 190 Cases, 58 of Which Wer~ Successfully Treated by Pneumonectomy, with a Review ~f the Literature INCIDENCE " pKIMAKY bronchiogenic carcinoma has become one of the most frequent malignant neoplasia encountered in humans. Whereas it has been su~ested that the increase in the incidence of these neoplasia js relative and not absolute, it is our conviction fl~at bronchiogenic malignancies have increased and are actually increasing in frequency. On the other hand, there is undoubtedly also a relative increase as well, because the condition is now more frequenfly.smpected and its existence recognized and proved. Unquestionably, before the frequency .of primary bronchiogenic carcinoma was appreciated, there were many cases which were diagnosed as pulmonary abscess, unresolved paeumonia," and many other conditions. As late as 1912, Adler stated, "On one point, however, there is nearly, complete con- sensus of opinion and that is that primary malignant neo- plasms of the lung are among the rarest form of disease." At the present time it is well known that bronchiogenic carci- noma is extremely comnmn and is probably second in frequency to gastric maligTtancies. This is evidenced by ot~'r own exparience and by the experience of Brines and Kenning and Koletsky. That these fig-ures show the true status is e~idenced by the fact that the statistics are based on ~Utopsy e~tminati0ns and not upon clinical di%~nosis. According. to Johnson and Keinha~ the inddence of prima~ bronchiogenic carcingma as determined at autopsy increased from 0.54% in the period from 19~2 to I9~7 " From the Department of Surgery, Sdaool of Medicine, Tulane Vni~er~b/o anti the Oeb~er Clinic. New Oread, 1187 .. . .- • ". .... _

OCHSNER, DIXON, AND DEBAKEY ei to 0.92% in the period from 1937 to 1942. In the Charity pital at New Orleans the incidence of primary carcinoma in 1931 was 0.47% and that of gastric 2.3%," whereas in 1938 these respective incidences were 1.9% (Fig. 1). In this institution the incidence of bronchiogenic as determined at autopsy increased five times in seven Seyfarth ~und increases in the incidence of pulmonary to all carcinomas from 5.05% during 1900 to 1906 to 8.75c ing 1919 to 1923. During the first half of 1942 there was an crease of. 15.5%. Dorn found that between 1914 and 1930 the~ death rote from cancer of the bronchus in the United States im Geased 3.7 times as contrasted with an increase of only 20%..fo~ all forms of cancer combined. He also found that the death rate [or cancer. ing treated {urther est United St ,ut 13,000 ichus. Appro.', time each yea deaths of the respira cardnoma of million as co~ that bronchic and that this the stomach of bronchiog doubled and a primary b: ddence of ( Wegelin ob~, chiogenic ca cases studie~ the number was as follc from 1910 1924, 4.8; ff i-: 1935 to 19 Weller fou: c~rcinomz c incidence iz in 450 case: At the ~ case~ the! period .191 1900 tO 19 genic carc 1914 the i

BRONCHIOGENIC CAKCINOMA ~, 1941, Imonary ght-year 30 the tes in- ,% for.~ rate 1930 emales ent : for cancer. Five of the females and 15 of the males are be- ing treated for cancer of the lung." He further estimates that between 450,000 and 500,000 people in 7the United States are under medical care. for cancer, of which about 13,000 are being treated for primary cancer of the bron- claus. Approximately 8,000 new cases of pr.imary bronchiogenic carcinoma are diag-nosed and receive medical care for the first time each year. Harnett found that whereas the total number of cancer deaths in the British Empire had increased 22%, cancer of the respiratory tract had increased 120%; and the deaths from carcinoma of the lung in males from 1921 to 1930 was 21.1 per million as contrasted with 100.9 per million in 1937. Jaff6 states that bronchiogenic carcinoma represents 11.47% of all carcinomas and that this lesion is second in fr .eftuency onIy to carcLrmma of the stomach and the intestines. According to Dick, the incidence of bronchiogenic carcinoma in the Glasgow Royal Infirmary has doubled and [or the years 1934 to 1938 in 3.98% of all autopsies a primary bronchiogenic carcinoma was found, an admission in- cidence of 0.101 of bronchiogenic carcinoma to all admissions. Wegelin observed a progressive increase in the incidence of bron- chiogenic carcinoma as determined at autopsy. In a series of lit cases studied at the Bern Pathologic Institute, it was found that the number of bronchiogenic carcinomas in each. 1,000 autopsies was as follows: from 1900 to 1904, .9.5; from 1905 to 1909, 1.8; from 1910 to 1914, 1.6; fi:om 1915 to 1919, 7.5; from 19E0 to 1924, 4.8; from 1925 to 19.99, 7.8; from I930 to I934, If2; from 1935 to 1939, 142. At the University of Michigan Hospiuil, Weller found that .in. the first 1,000 autopsies the incidence of carcinoma of the bronchus',to all autopsies was. 0.1% and that this incidence increased to 0.5% in the second thousand, and to 0.8% in 450 cases of the third thousand. At the University of Leipzig, Assmann found that in autopsy cases the incidence of bronchiogenic malignanci.'es during the period 1919 to 1922-was double the inddemce during the p~,'od 1900 to 1906. Hoffanan observed that the inddence of bronchio- genic carcinoma among all deaths has increased. Whereas in 1914 the incidence was 0.6% 100,000 p0pi~lation, in 1924 it had increased to 1.6% and in 1928 to I~9%. According to Fdenczy-

OCHSNER, DIXON, AND DEBAKEY and Matolcsy, the incidence of bronchiogenic carcinoma creased from 0.54% in 1896 to 10.3% in 1925. Fischer in Germany an incidence of bronchiogenic carcinoma creased from one and one-half to three times. According leb and Ang-rist, the incidence of bronchiogenic carcinomas as determined at autopsy is 14.71%, and the of bronchiogenic carcinomas to all autopsies is 2.41%. Staderman found that of the fatal cases that came to auto incidence of bronchiogenic carcinoma to all cancers was as contrasted with a similar incidence of cancer of the - 21.8% and of cancer of the stomach of 14.1%. At the Wisconsin General Hospital, according to 5.9% of deaths from neoplasms coming to autopsy have a chiogenic carcinoma. ~ccording to Olin and Elliott the of-primary bronchiogenic carcinoma to all malignancies as mined clinically was 4.7%. Perrone and Levimon state that of all carcinomas coming to necyopsy are bronchiogenic. .~, Steiner, in a series of collected cases, found that primary chiogenic carcinoma occurred in 0.99% of 93,560 autopsies formed throughout the United States, representing 9.47% ca~;cinomas. He found that at the University of Chicago genic carcinoma was third in f:requenc3" of all carcinomas as mined at autopsy. Holzer'found that the incidence of genic carcinoma as determined at autopsy has increased. 1895 to 1904 the ratio of bronchiogenic carcinoma was 0.07 with an incidence of ~.04% of all carcinomas. From 1905 to these respective incidences increased to 0.19% and 2".36%, from 1915 to 1924 to 0.47 and 6.69%. Olson, in reporting .cases of carcinoma cff the bronchus from the Boston City ' pital, found that the incidence of these lesions to all ~ increased annually from 7.5% in 1929 to 19.1% in 19.34. larly, Matz found that the incidence of bronchiogenic c~rcinoma -~--- toall carcinomas as determined at autopsy in the Veterans' -~r~u Hospital ~as 6.4% for the years 1927 to 1931, 15.8%.for ,years-1932 to .1937, and 23.'~% during th'e year. 1937., L6izaga -. : : ported an.incldence of 5.18% of carcinoma of the bronchus all' carcinomas in .the autopsies at the:'Institute of Anatomy of the Faculty of Ciendes Medicu~ of Buenos

BRONCHIOGENIC CARCINOMA the period 1898 to 1917 and an incidence of 1.4.6% during the od from 1918 to 1937. He also found that the incidence of carcinoma to all carcinomas in autopsy series at Francisco Janvier Hospital was 29.62% during the period to 1931 and 4225,% from 1932 to 1937. The relatively Lter increase in the incidence of bronchiogenic carcinoma is nsrrated by Rasahn's collected cases. Whereas the incidence all carcinomas as determined at autopsy increased from 10.27% period 1910 to I919 to 12% in the period 1920 to 1928, the of bronchiogenic carcinoma to all autopsies increased ~rom 0.44% to 0.89% in these respective periods. Similar result~ ,,ere obtained in the Dresden City, Hospital. Reinhaxd reported of autopsy studies in the period from 1852 to 1856 ix~ which the incidence of bronchiogenic carcinoma to all autopsies Was .0.05~, and to all carcinomas x~a.s 0.92%. Wolf reporting from the mine institution found an incidence of bronchiogenic malig- nancy to all autopsies was 0.35%, and Ran found that from 1909 t6 1914 the incidence of the bronchiogenic lesions to all card- noma was 2.72% knd from 1914 to 1919 was 4.66%. Thus, the incidence of bronchiogenic carizinoma in the same institution as determined at autopsy increased five fold during a 67-year period. We previously showed that from a review of the mortality sta- tistics of all persons dying in the United States of bronchiogenic raalignancies there was not only an actual increase in the number of deaths from this cause but also an annual increase in the death rate per hundred thousand population. We found that bronchio- genic carcinoma is increasiflg more than other malignancies. Whereas in the period from 1920 to 1936, inclusive, the incidences of carcinoma of the stomach, duodenum, liver, g-allbladder, uterus, and skin showed little or no increase, the incidence of cardnoma of the Imag showed a progressive rise." . Of 234,490 admissions m the Chari~y Hospital in New Or- leans from the tint of Janrary 1940 to the first of August 1944, -- bronchioge~|c carcinoma was diagno~, d 175 times, 154 Of which Were histologic~IIy prov~ed and 21 of which were diagnosed clini- .-- cally and by x-my (Fig. 2)- : ...... : ':-.--- - --:" : " ....... , Sex..~Broncl~_iogenic carcinoma occur~ pred~minantIy in males " ... and alth0ugh~the~r~Iative ~cidence.. in females has incr ,ea# re-_ .-_ _ ....~i.-.~.ii_ -~-. . - " Hgl .... ':i~f:)-: ........ " - " "-. ~.-~

OCHSNER, DIXON, AND DEBAKEY cently, it can be stated that there is a definite sex malignant disease of the bronchus in males. In a previous we found that in a series of 8,575 collected cases in which was stated, there were 6,769 males (78.9%) and 1,806 (21%). Thus it is seen that approximately 3 out of every nomas of the lung affect the male. As it has been the ~eater incidence of carcinoma of the bronchus in men past has been due to a ~eater prevalence of smoking sex (Ferrari), there will tend to be an equalization sexes because of the increased smoking among women. A~ cated previously, however, the incidence will probably so ~eat in women as in men because there probably is a predisposition in the male sex. It has been su~ested by that the cases which occlar in women might be due to change occurring on the basis of an adenoma, as adenomas in about equal frequency in both sexes. In Adler's series, mately 70% of the bronchiogenic.carcinomas occurred in W~elin, however, in his series of bronchiogenic mali 96 occurring in males and 21 in females, a relationship tol. Farberov found that of 150 case~ studied at the State Roentgen Radium Oncologic Institute at Karkov, 121 in males and 9 in females. According to Harnett, in a 1,065 cases of bronchiogenic carcinoma there were 871 in and 192 in females, a ratio of 4.5 to 1. In t90 cases of genic carcinomas which we have observed 166 (87.4%) and 24 (12.6%) were women (Fig. 3). In 58 cases in which have done a pneumonectomy for bronchiogenic carcinoma~,:~ (8I%) were in men and 14" (18.9%) in women (Fig. 4). - ~lge Incidence.. Primary bronchiogenic carcinoma with- ~exceptions occlaxs in older persons. Brunn. found that 576 cases occurred between the ages of forty and sixty Weller found the greatest incidence in the sixth and. a~e~. In Olson's series, 73.8% occurred in the seventh decades..F'tscher, in a series of 1,888 collected cases, ~that one-third of the tumors occurred in the fifth third after the ag~ of sixty~ and only 13% under the --There were four tumors in children under $ repo in a b, aca age. A less we itirst decat :fourth de~ d ~ have studi lade 2 fifth decao ~5 (_98.0% s). seventy 8~; the yo~ ectomy pe $e seventh de cent were i nineteen o~ patients. i the age There ge~ted to igeni~ catch

BRONCHIOGENIC C.,-~RCINOMA ere hich of collected cases, found the most frequent occurrence be- the ages of fifty and seventy years. Simons, in a collected of over 5,000 cases, found that over four-fifths of the cases between the ages of forty and seventy. The youngest reported was by iMcAldowie. He described a cancer of the in a child five and one-half months old. The lesion was not by biopsy but was demonstrated at autopsy. Beardsley te- a case which was histologically proved and in which the first nodule was noted at the age of ten months. Weill-Hall4 a primary bronchiogenic carcinoma in a child one year age. A number of cases have been reported occurring in pa- :tients less than I0 years of age (Gould; Lereboullet et al.; L6izaga; Field and Quilliam). In a series of 4,307 collected cases we previously reported, the following results were found: decade, 0.16%; second decade, 0.7%; third decade. 2.9%; fourth decade, 10.7%; fifth decade, 25.4%; sixth decade, 34.1%; seventh decade, 20%; eighth decade, 5.4%; and ninth decade, .0.58%. In the 190 cases of bronchiogenic carcinoma which we • have studied the following age incidences were found: second dec- 2 (1.1%); -,hird decade 2 (1.1%); fourth decade 12 (6.3%); fifth decade 44 (23.2%); sixth decade 64 (33.4%); seventh decade 55 (28_9%); eighth decade 10 (5.3%); ninth decade 1 (0.5%) (Fig. 5). Eighty-five per cent occurred between the ages ,of forty and seventy years. The average age ~-as 54. The oldest patient was 83; the youngest was 19 years of age. In our 58 cases of pneumon- ectomy performed for ,bronchiogenic carkinoma, the ages were as follows: second decade 2 (3.4%); third decade 2 (3.4%); fourth decade 5 (8.6%); fifth decade 14 (24.1%); sixth decade 24 (41.4%); ~. seventh decade 10 (17.2%); eighth decade 1 (1.7%). Sixty-five per cent were in the fifth and sixth decades. The youngest patient was nineteen ariZ._the eldest, seventy-two (Fig. 6). It is obvious the age - of patients subjected to pneumonectomy is somewhat lower than the age of all patients with bronchiogenic carcinoma (Fig. 7). __. -- ETIdLO~Y - " " =~ There have been many-explanations which have beqm sag- gested to account for the increase in the incidence o[ bronchio- genic carcinoma. Wintemi~ Wkson,and McNamara; because Of

OCHSNER, DIXON, AND DEBAKEY ~' anaplasia of the bronchial mucosa in persons dying enza, suggested that this change might be a precancerous and that as a result of the changes occurring in persons from influenza in the epidemic o[ 1918 an increase in dence of bronchiOgenic carcinoma may be accounted for. lar suggestion was made by Askanazy, who [ound in 38 o|.

BRONCHIOGENIC CARCINOMA ent~ dying o[ influenza "the normal columnar epithelium of the is replaced by stratified squamous epithelium or that occurred a metaplasia (protoplasia) of the bronchial mu- Similarly, others .(Barton, Berblinger, l.~schke, Mittasch, ~Mo~e, Schmidtmann, Teutschlaender) have suggested a possible p between the influenza epidemic in 1918 and the in- Fr~. ~ Fro. 5--Grai~hic representation of age inddence according t¢~ decades in Fro. e~Graphi¢ repress/ration of age iaciaen~e~ accSrd!ng to a~a= -

OCHSNER, DIXON, AND D~.BAKEy ~ crease in bronchiogenic carcinoma. Bauer and M( the first to report cases of bronchiogenic carcinoma fluenza. Berblinger and more recendy Oberndorfer that influenza i~ among the principal causes of ~mcer of On the other hand, Fried, in a series of 47 cases carcinoma observed since 1918, had only one in which a history of influenza. He stated the belief that little if any role in causing pulmonary carcinoma. Kiku~ the! thes, don freq~ opin for of th our ~cto~ worl mali: theli ,ot~ ~ fro= bro~ geni, Usp, chio

at BRONCHIOGENIC CARCINOMA the incidences being 40% and 25%, respectively. This suggest that metaplasia of the bronchial mucosa following [uenza is no Factor in the increased incidence of carcinoma. 139 autopsies studied by Simpson in patients with pulmonary only 5 of the patients had influenza preceding the lung Hueper is also of the opinion that there is no significant p betnceen influenza'and bronchiogenic carcinoma. An apparently plausible theory concerning the cause o£ bron- carcinoma is inhalation of irritating +oases such as ha~t gas of combustion motors and gases emanating Prom tarred roads. Kawahata observed 21 cases of carcinoma of the lung in years among workmen employed in an illuminating g-as gen- erator and consequently exposed to dust and hot gases containing tar. Hampeln stated the belief that there is a definite relation to the increased production of smoke and dust in large cities in that these substances by constant inhalation produce a chronic irrita- tion of the bronchial and pulmonary epittxelium, increasing the f-requency o[ caxcinoma of the lung. Staehelin also stated the opinion that the inhalation of dust containing chemical sub- stances which possess a specific carcinogenic agent is responsible for pulmonary carcinoma. An increased incidence .of.carcinoma of the lung among open air workers exposed to road dusts was ob- served by Kermaway and Kennaway. This is not substantiated by our experience because in our 58 patients subjected to pneumon- ectomy, 32 (55.2%) had indoor occupations and 26 (44.8%) worked out of doors .(Fig. 8). Whereas it is possible to produce malignant lesions experimen,tally by the application of tar to epi- thelial sure;aces, it is difl~cult~to ima~ne how either the inhalation . 0f exhaust gases of combustion motors or the inhalation of gases from tarred roads can account for the increased incidence of bronchiog~.nic malignancies. Campbell demonstrated experimen- tally that exposure ot~ mice-to exhaust gases Prom combustion en~-:4nes had little effect on the incidence of primary brdnchio- genic carcinoma when compared with the controls. Davydofl~ and Uspensky observed a definite increase in the incidence of bron- chiogenic_.carcinoma in Russia in the-past ten years, although therewer~ few_.a, utomobiles and practically.no tar/ed roads in that country.:~ Similar observations have been made by Boyd in~ Can-.

OCHSNER, DIXON, AND D~BAKEY ada and by Husted and Biilmann in Denmark. Holmes state that the increased incidence of noma be~n in Great Britain before roads were tarred. and Franke state that the incidence of bronchiogenic increasing in the city, of Ri~, even though there is no the tarring of roads or in the number of automobiles. cIusions are drawn by Lehmann and Probst, who believe inhalation of gases fi-om tarred roads has little to do with creased incidence of bronchiogenic carcinoma. .~ FI~ 8--Graphic representation of o~:upation in authors' • 58 pneumonectomies. _ We have repeatedly emphasized the probable relationship tween smoking and the increased incidence of bronchiogenic cinoma. The chronic irritation resulting from the "- ~ , cigarette smoke over long periods of time is well known and chronic,bronchitis in smokers is so common that their cougtt~ consid~ed inconsequential and o[ no significance. Whereas longed chronic irritation o[th~, bronctiial mucosaas a . , :inhalation of smoke can in itself be a'factor in the- .. neoplastic disease just as any prolonged and continued .. can produce such a lesion, it is probabld ttmt smoking, ex. _e~. _= ;..- . .additional and more"active, influence. than ~ ..... :ii i:::-) alone, . ,¢_-x_ perimentalIy.. : it_ has .beeri~ demoustrated

hip ~ic don .~ don ~ BRONCHIOGENIC CARCINOMA a carcinogenic effect (Wacker and Schminke, Leitch, Phil- Lickint, Roffo, Lu-Fu-hua, Morpurgo, and Boehncke). is some contr6versy concerning which component of to- is responsible for the epithelial proliferation and the" effect. In addition to nicotine, there are other sub- in tobacco, such as pyridine bases, phenolic bodies, am- and certain hygroscopic agents, which are irritants. Stoe- • and ~Vacker produced epithelial proliferation in animals with Hamilton stated that pyridine •produces lesions on the skin similar to those observed in patients handling tarry sub- i~nces. McNally states that in addition to nicotine the tarry material separated from tobacco smoke has a significant irritating He demonstrated that from 6.5 to 11.5% o[ this residue could be absorbed or retained in the body. Hygroscopic agents which are added to cig-arette tobacco for the purpose o[ assuring satisfactory moisture content are usually irritating. Glycerin and diethylene glycol are commonly used. Of ~eat ~igTdficance are the investigations of Roffo, who demonstrated that tobacco does exert a carcinogenic effect and that the c~rcinogenic effect of tobacco tar varies considerably with different types o[ tobacco. He found that tar obtained from. black Kentucky' tobacco.when applied to the skin o[ rabbit~ produced epithelial tumors- as read- ily as coal tar. In both a tumor developed in 100%. On the basis o~ his clinical observations on 78,000 patients treated at the Uni- versity Institute for Experimental Medicine and for the Study of the Treatment of Cancer in Buenos Aires, Roffo cond'udes that tobacco is the most i~i..'portant factor in determining the localization of cancer. Hoffanan on the b~sis of his statistical analyses of the incidence of cancer states, "Stnoking habits unquestionably increase the Iiabili~ to cancer of the mouth, the throat, the esoph%mas, the larynx, and the lung. The_ change in the cancer death rate dur- ing recent years has not, however, been at all dispropor- tionate to the enormous increase in-.the dgarette smoking - ~.: habit which has replaced the older method of smoking, ~ unquestionably more injurious than Smoking of dgars.-- .. The in~reas~ o~ .caiacer.3f the lung observed in this and. " tent directly traceable .to the commdn/practiceo[ cigarette:-5~

OCHSNER, DIXON, AND-DEBAKEy smoking and inhalation of cigarette smoke. The factors unquestionably increase the danger of cancer velopm.ent." As early as 1923 Fahr stated that he believed the " bronchiogenic carcinoma was due to the increased inci cigarette smoking. Lickint also believes that digarette is an etiologic factor in the production of bronchiogenic noma and that many case~ can be prevented by the h-ore smoking, particularly in patients belonging to known to have a high incidence of cancer. Harnett found in a g-roup of 69 men with bronchiogenic carcinoma, were non-smokers, 26.1% were moderate smokers, 40.,5% excessive smokers; and in 29% it was not stated whether smoked or not. Anderson quotes Kramer as stating that a 5 cigar yields 200 rag. of tobacco tar, whereas an eight-hour behind an automobile resulted in the collection of only I ~ of a similar tar. M~Nally found that the tarry residue in cigarettes varies h:om 4.84% to 15.29% with an average of Recently Black suggested that commercial lead may be inciting factor "in bronchiogenic carcinoma and that the lead tobacco may bE the [actor, in the causation of the lesion in ers. Wallace and Ja "ckson surest that the small amount in tobacco may be responsible for the carcinogenic effect bronchial mucosa. In an attempt to determine the causal donship between inhalation of-exhaust gases, cigarette and bronchiogenic carcinoma we ga'aphed the incidence of bronchial lesion aga.inst the sale of cigarettes and the sale automobile tags in the United States (Fig. 9). As seen by g'raph there was no parallelism between the sale of automobile- licenses and the incidence of bron..chiogenlc carcinom,'~ but is a distinct parallelism between the sale of cigaxettes and the • incidence, of the bronchial lesions, . .,: . - Barnard compares the development of_ bronchiogeni~ noma with the development of. basal cell garcinomd of the Skin and believes that chronic irritation ofthe bronchiis the/esult of- repeated infection and other inqtations which are [actors in stimu- _ gating, the fftowth of cardr/oma~.. He mainl~ tha~-bronchial. " mucosais a tissue in which carcinoma.is likely "

BRONCHIOGENIC CARCINOMA and Dible on the basis of the silica content and the histologic of pneumonoconiosis in 70 patients with and 50 without carcinoma concluded that silicosis plays an impor- etiologic role. Klotz reported that of 50 patients with sill- coming tO autopsy 8% had bronchiogenic carcinoma whereas iin 4,500 neeropsies in which there was no silicosis, only 1.18% ishowed bronchiogenic carcinoma. Dick found that in a series of 1131 casesof bronchiogenic carcinoma, in 44 there was evidence of silicotic nodules, whereas in 11 others there was fibrosis ~ith suggestive silicosis although no "~ao~Iules could be demonstrated. According to Hamett, silicosis :plays a definite etiologic role in bronchiogenic carcinoma. He found that metal gTinders were • likely to develop.fibrosis of the lung from silica dust and that they developed bronchiogenic cardn0ma two and one-fourth times more frequently than other individualL - ,-- Other- specific and non:Specific pulmonary /rffecdons have been suggested as responsible etiologic factors in the production- of bronchiogenic carcinoma. Of these, tuberculosis is probably the most trequ_emly mentioned. According to ~.wing, the tubercle badllns, becatis~, it is an-irritating agent, is one of the most f re- - 1201

OCHSNER, DIXON, AND DEBAKEY quent etiologic factors in bronchiogenic carcinoma. Bah'on .Cherry suggest that tuberculosis plays an etiologic role in the duction of these neoplasia. Of 31 patients reported by had tuberculosis. Fried reported 13 cases in which and bronchiogenic carcinoma coexisted. On the other Rokitansky long ago emphasized the rare occurrence of losis and cancer in the same person at necropsy..These dons were subsequendy made by others (Albrecht; Kramer; L6izag~a and Vivoli; L~abarsch, and Reinhardt). and Monserrat report that whereas the incidence of F tuberculosis is high in the Philippines, bronchiogenic " is encountered in£requendy. In a series of 24~ cases of genic c~rcinoma, Kikuth found only 22 with tuberculosis. comparison of 886 persons with active tuberculosis and a ~oup without tuberculous involvement, Pearl found the dence of malig~aant tumors in.. the former to be 1.2% and in latter, 9.3~o. He concludes that the rarity of the two lesions the same person "is apparently due to a significant tween the two patholo~c phenomena which disappears when if the tuberculous process retrogres~ or heals, particularly by fibrotic route.'" Similar Observations have been made by Carlson and Bell. L6izaga and Vivoli found only one cas~ of of the lung in 2,400 autopsies on tuberculous patients. Simpson and Ziemssen have suggested that syphilis may an etiolo~c factor. Letulle observed evidence of syphilis in 11 cases of bronchiogenic carcinoma. The coexistence of and bronchiogenic carcinoma has been observed by Bonnamour, Brouardel, Martin and Colmt, Popper, Rouslacroix and There is, however, not su~icient evidence to suggest that syphi~ is a factor in the production of bronchiogenic carcinoma. Non.sl~ecific,-chronic inflammato~ lesions of the bronchi been suggested as possible etiologic factors. Frommel found chronic bronchitis, bronchiectasis, and emphysema were in ~9 of 41-cases of bronchiogenic ~-cinoma. Klotz observed case in which a malignancy developed in a bronchiectatic in a patient who had suffered from bronchiectasis for fifteer~, Simpson, L~scke, Fried a~id Hunt:have observed the .... _ of chronic lesi0 in ci~ br~ br b~ p1 p1 s~ pl 6'. d, r~ a t, t I t

)If, ,nch In BRONCHIOGENIC CARCINOMA cinoma. Relatively recently Stewart and Allison reported a case in which a microscopic focus of oat cell carcinoma was found in a bronchiectatic cavity of a pulmonary lobe which was removed for bronchiectasis. Blake reports an inoperable case of bronchiogenic carcinoma which developed at the site of a retained metallic for- eig-n body which had been aspirated 6 years previously. It has been suggested by Bonner, Goltz, and Siegmund that the meta- plastic changes consequent to the injury of the ePithelium in these chronic inflammatory lesions are responsible for the neo- plastic change. It has long been known that the inhalation of radioactive sub- stances is responsible for the development of bronchiogenic neo- plasia. This is demonstrated conclusively by a high incidence of carcinoma among the workers in the Schneeberg mines, first em- phasized by Anastein. Rostosk], Saupe, and SchmorI found that 62% of workers in the Schneeberg mines who were followed until death died of primary bronchiogenic carcinoma. A more recent report of studies on the Schneeberg miners conducted by the gov- ernment committee for the prevention of cancer states that these mali~nant tumors are considered to be due to radioactive emana- tions. D6hnert kept 48 mice in the Schneeberg mines ~or" periods rang-ing from two hundred to three hundred and seventy days, and of 26 submitted to microscopic examination, 7 were found to have neoplasms, 2 of which were in the lung. He concluded that these tumors were caused by the radium contained in the air. In addition to being radioactive, the dust contains a high con- tent of arsenic and cobalt (Schmorl and Uhlig). Similar observatiohs were, made by Pirchan and Sikl, who stud- ied the pitdablende mines bf Joachimsthal, which is across, the mountains ~om Schneeberg. The latter authors found that ra- dium emanation in the air of J~ichymo.v pits is as high as 50 mach6 units. Peller-fotmd that during the years 1929 to. 19~8, 89 Joa- chimsthal miners ch-ed, 6 of whom were ekamined postmortem. Of these, 47 (52.8%) died of cancer, 42. of whom had primary br~nchiogeni.c carcinoma,. Peller is of the opinion that. the high incidence of bronchioger~ic carcinoma among these-miners is due to the radi_'oactive factor in the mines: Although the presertcd of dust alone~migl_a_t be responsible for the development of bronchio-

OCHSNER, DIxoN, AND DEBAKEY genic carcinoma, the investigative work of Willis and would tend to disprove this, as would the fact that bronchi. carcinoma is not as ~equently found in other miners who. pneumonoconiosis as do those in the Schneeberg mines. and Brutsaert were able to produce tumor-like structures lun~ of guinea pigs exposed to silica dust, but there was dence of carcinoma developing. ~ Trauma has been blamed for the development of genic carcinoma but ~ually plays little or no role since the pulmonary lesion existed before the trauma was Isolated instances have been reported in which trauma ably played a role in thd development of the bronchiogenic (Aufrecht; Georgi; Gomez; Hand:ford; Hedinger; Luckow, Sch6ppler; Scott and Forman; and ~Vells and PATHOLOGY Carcinoma involving the lung is almost entirely a the bronchi, although rarely it may begin in the alveolL right side is involved more frequently than the left. In a of .4,732 cas~ which we previously collected from the there were 2,761 (58.3%) tumors involvihg the right lung 1,97I (41.6%) involving the le~t lung. In Fischer's series of cases of pulmonary carcinoma the right lung was involved in 53' the left in 45% and in 2% the lesion was bilateral. In the cases which Fischer reported ixt which the location was according to the bronchus involved, the findings were as right main stem bronchus, 142; left main stem bronchus, II bronchus of the right upper lobe, 148; bronchus of the left lobe~ 130; bronchus of the right lower lobe, 120; bronchus of left lower lobe,. 105; bronchus of the right middle lobe, 15. In the 190 cases which we have observed, .the right lung involved in.98 (51.6%); the left lung in 87 (45.8%), and in:-i (2.6%) both lungs were involved (Fig. I0).. Ix~ the 58 cases which we have done-a pneumonect0my [or p,rimar~ genic carcinoma, the right lung was involved in 34 (58.5%) the left lung in 24(41.4%) (Fig. 11). The involvement lobes 2 lobe lower in the neopla~ high to us Knox's high, heir 'were o~ dJJ~se, 80% o~ b: near the the perip bronchi~ ~-' Most hye of" mors axis with its

BRONCHIOGENIC CARCINOMA ~be 2 (3A%), right lower lobe 14 (24.1%), right upper and mid- lobes 2 (3.4%), right middle and lower lobes 2 (3.4%), left lobe 13 (17.2%), left lower lobe 7 (12.1%), and le[t upper lower lobes 4 (8.9%). Most bronchiogenic neoplasms are lo- ated in the region of the hilum. According to Boyd, 90% of these neoplasms are in this area. This incidence seems somewhat high to us since in our series it is not quite so high. In Frissel and Knox's series the incidence o[ hil~ carcinoma also was not so high. being only 49.7%;. 17.8% involved the parenchyma and were of/.he nodular variety; 6.5% were peripheral; 23.9% were disuse, and 2.1% were bilateralm'.fliary. According to Edwards, 80% o[ bronchiogenic carcinomas 6¢cur in the larger brondfi or near the origin o[ the secondar~ bronchi and only 20% occur in- the peripheral bronchioles. According to Betts~ in 62 cases of bronchiogenic carcinoma, 46 were located in the re,on o[ the. hilum and 16 peripherally, Most bronchio.genic carcinbmas probably be~--, in the basal layer o[ the bronchial mucosa. According to Bamard, these tu- mors arise most f~equenfly at @e junction o£ the bronchial branch ~th its parent stem and that early in the growth of the lesion the ~ ~ 1205 " " "

OCHSNER., DIXON, AND DEBAKEY mucosa overlying the neoplasm is intact, but shows color, particularly that of a whitish ~ay or whitish pink. have been many classifications of bronchiogenic gested and the one most frequently described has been ferentiation between squamous cell carcinoma, small differentiated cell carcinoma (oat cell carcinoma), and carcinoma. There has been litde uniformity of cerning this classification, however, and for this reason it most logical to us that the classification proposed by the most logical one. This classification is based on the ment of cells lining the bronchi and. explains adec histologic structure of all primary" bronchJogenic Normally the ceils lining the mucous membrane of the tree represent varying degrees of differentiation an& don of the ori~al endodermal cells. "Apparently the undifferentiated, end~lermal ancestor is capable of developing into ciliated cylindrical lium, goblet cells, cuboidal cells, arranged into acinar tubular structures producing a serous or mucous indifferent cells, lining the ducts of these glands, and another kind of cuboidaI or low cuboidaI ceils cilia which line partsof the terminal bronchioles. In tion to the variety of cells, just enumerated there are, neath the ciliated cylindrical-and goblet ceils, g number of bther epithelial cells whk:h, like the basal in the epidermi~, are lined up along the border toward tunica propria. They are the cells from which the layer o.f ciliated cylindrical and goblet cells is replenishecL~- These cells, which may be called 'reserve ceLts,' are the ent cells of the ciliated~ cylindrical, and goblet cells. In addition they. naturally also possess the qualitie~ .of ~ ancestor cells in that they may differentiate into any _ of epithelium that an endoderrfiaI ceilis capable of i Halpert is of theopinion that it is" from these- reserve cells bfonchiogenic carcinomas originate by a pi-ocess liferatiOn:. He, therefore, dassifie~ .brdnchiogenic r three types depending-apoh the embryo!ogic direction of • :-..; .~/ tende: genic c carcinc cell car AIt~ ered b~ to the ff that and me bronchi dence Neely, bronckic and fxeq is distan~ fast man in by determi~

BRONCHIOGENIC C--kRCINOMA 1) "reserve cell" carcinoma, (2) cylindric cell carcinoma, and (3) mous cell carcinoma. According to D'Aunoy, Pearson, and the "reserve cell" carcinomas consist o[ round, elongated, polygonal cells ~owing in solid masses and forming no par- ular structure. Characteristically, they have a palisade arrange- of the peripheral cells. The cylindric cell carcinomas are ed of cuboidal or columnar cells [orming tubular or acinar ures or are mounted on delicate connective tissue stalks in a 7papillary arrangement. The squamous cell carcinomas have a toward keratinization or pearl formation with central keratinization. In 135 cases o£ primary bronchiogenic carcinoma reviewed by Halpert at the Charity Hospital in New Orleans, 74 were squamous cell, 39 (30%) were reserve cells, and 22 (20%) were columnar cell carcinomas. In 56 cases o[ bronchio- genic carcinoma reported by Menne and .anderson, squamous cell carcinoma was present in 31, adenocarcinoma in 13, and reserve cell carcinoma in Although metastasis fi'om bronchiogenic carcinoma is consid- ered by many to occur relatively early, this is findoubtedly due to the Fact that a diag'nosis is not made until late. It is our opinion that g~nerally bronchiogenic carcinoma ~ows relatively slowly and metastasizes relatively late. Rogers, found that in 50 cases of bronchiogenic carcinoma studied at autopsy there were no metas- tases in 10%, whereas Moise, in 327 cases, found the same inci- dence, with no metastases..,W~elin observed similar results.. Neely, in a series of 80 cases; found no evidence of metast,xses in 33.8%. Unfortunately, until recently the relative incidence o[ bronchiogenic carcinoma has not been sufficiently appreciated, and frequently., many times the first manifestation o1~ a carcinoma is distant metastasis. In 50 autopsy cases studied by Rogers, the first mani/estation ot~ a bronchiogenic lesion ~as distant metastasis in 44%, and this same finding was true in 9% o~ 72 cases analTt.~l by King. In the latter ~oup of cases, the correct diagnosi~ Was. determin_e_d by biopsy o[ the cervical axillary lymph nodes~in 15%. Lind~.ko~.repo.rted that a correct diagnosis was made bv • 1207

OCHSNER, DIXON, AND DEBAKEY biopsy of the supraclavicular lymph nodes in 9 cases and axillary in 2 of 24 proved cases. Viacava and Pack re: of 334 fatal cases of bronchiogertic carcinoma seen at morial Hospital, 11 had an enlarged supraclavicular lyre as the. only sig'n on admission, 20 others had involvement nodes on admission, and 11 developed involvement apy. As we have previously emphasized, the modes of extension in primary, bronchiogenic carcinoma are as (1) by direct extension, (2) bronchial or intraluminal, plantation by aspiration biopsy or by operation, (4) nous, (5) lymphogenous. Metastases in primary carcinoma, as malignancies elsewhere, constitute the portant single prognostic factor. As might be ima~ned, the most important sites of involvement are the re~onal bronchial and mediastinal nodes. In Wegelin's series which were observed at autopsy were Iymphogenous metastases in 812%, the bronchial , mediastinal nodes were involved in 93 of 117 cases studied.: Koletsky's series of 88 cases, these nodes were involved in Olson observed the regional lymph nodes were involved in in his series of 67 cases. On the other hand~ Miller and their analysis of 808 collected cases, found that the re~onal nodes were involved in only 30%. In a series of 3,047 cases which we previously reported, the regional lymph were involved in 72_9% (~:ig. 12). This high figure is Confusing and does not indicate that three-fourths of genic carcinomas will have lymph node metastasis because in of these cases the observations were made at autopsy. Next to regional lymph nodes, the liver is most frequently involved- 808 collected cases reported by Miller and Jones, the liver involved in 30.7%, Jaff6 found the liver involved in 36% o[ • " cases, Koletsky in 40% of I00 cases, and Frissell and Knox i: 48.7% of 39 cases. In a previously collected series of 3,047 we founcf metastasis to the liver in 33.3%, Maher and found the liver involved in 35%: The pleura and,lungs next most frequent site of metastases..In our collected series, " volvement of the pleura occurred in 29.8%, and the l_un~ in 25_~o. -The frequency of involvement bf these is no direct inv the. rant roles probably

BRONCHIOGENIC CARCINOMA ~ns is not surprising because their proximity is likely to lead to direct invasion. In addition to this method of extension, how- ever, the lymphatic and aerogenous routes probably play impor- tant roles in conveying the neoplastic cells. The latter route ks probably insufficiently appreciated. r

OCHSNER, DIXON, AND DEBAKEY The method of metastasis, which Letulle and Jaquelin termed "Metastases a~riennes" and which we have referred bronchial embolism is probably a more important method generally considered. Letulle and Jaquelin in 1924 called tion to this method of dissemination and reported a case in the primary, lesion was on the right side and in which occurred on the opposite side. They believed that metasta volvement resulted from implantation of tumor cells by the intrabronchial route. Lumsden refers to this as spread. Koletsky, in a series of I00 ~ases, observed 8 in aer0genous transmission was present. In Moise's series there 4 cases with this type of extension. This method of ex~ension.~ probably responsible in. many instances for peripheral ment of the same lung or the opposite lung in those cases in the primary lesion ori~nates proximally. The incidence of involvement of distant viscera varies ing to reported cases and apparendy depends to a ~eat upon the thoroughness of e_xarninadon at the autopsy table. M:tler's serie~ of collected cases; the kidneys and bone~ were in frequency of involvement to the pleura and lung. servadons were made by Ferenczy and Matolcsy and by and Jones. On the other hand, D0squ~t and FrisselI and found the brain and the adi:enals to be the next most fre~ site of metastasis. In our previously collected ser~ex of cases incidences were as follows: bone, 21.~%; adrenals, 20.3%; neys, 17.5%,. and brain, 16.5% (Fig. I2). Involvement of distant viscera undoubtedly represents, in most instances, togenous spread. The frequency of metastasis to the central _. ous system is particularly interesting and is additional - of the h .ematogenous route Of metastasis ha primary carcinoma. The clinical sig-nificance of this fact has been " sized b~; Parker and more recently by Craig and his who state that they routinely obtain roemtgenogram~ of the • 7~- in all cases in .which a neoplasm o[ the brain i~ suspected2 -- . : The heart and pericardium are involved not infreq.uendy " undoubtedly. represent a. ~lirect exte~a~ion. In Our previ0usly. '~ lected ca~es, involvement of these structures . Involvement 0f the pericardium is certainly mucl~ more 1 C, •

BRONCHIOGENIC CARCINOMA than involvement of the heart. In 4 of ou~ pneumonectomy cases a subtotal pericardectomy was done in order to remove the in- volved portion of the pericardium. In 2 cases exploration revealed extension of the tumor from the lung into the wall of the auricle. Hetastases to numerous other structures have been observed, and in the collected series which we previously reported the incidences lon were as follows: pancreas, 7.3%; peritonetun, 4.8%; gastro- intestinal tract, 4.3%; skin, 3.6%; spleen, 3.5%; th);roid, 2.3%; tonsil, 1.8%, and tohgue, I~6% (Fig. 13). Of the more unusual. areas in which metastases tiave been observed are the nasal sep- turn, the tip of the nose, the urinary bladder, the eye, the gall- bladder, the ovary, the uterus, the testicle, the prostate, the seminal vesicle, and the'skeletal muscles. .- - From a. prognostic_ standpoint particularly, the-occurrence of metastasis in the skin and subcutaneous tissue is of interest (Fig. 14). Some authors have recorded_relatively high incidenceS. Grove. and Kramer found such metastases in 13-~o of 24 cases; Maxwell _and Nicholson in 8% o[. 100 cases; Arkin and Wagner in " _-~'-.~-'~.~'.- : " ~-_ - ~: ~ . " :- ~.-- .... - " ..~ -C~" :-.~- I~II.--. - .... - -_.. ...... -- - - - - 7 .C~':7 -.- "- ..... • ...... - -> -.-5~-. ..... -.".5~ ~. " .'.:" " -"--:- : -~ "" ":: ~ ~ " - ". " " "'~.~7:~ ".-.~

OCHSNER, DIXON, AND DEBAKEY 5.2% of 185 cases. In a review of this subject, Carache From the literature 80 (2.8~o) examples among 1,063 bronchiogenic carcinoma, and they added another case. and Nichokson stated that the nodules are small and ally less than 2 ¢rn. in size, and commonly distributed over back and front of the chest and shoulders and the anteriorl dominal wall surrounding the umbilicus. They emphasized significance of these nodules and stated that should always be considered as a possible primary source in with mems.tasis involving, the skin and subcutaneous tissu~ our series we have seen 4 cases o[ skin metastas~s, and it was sible by the removal o[ a nodule to confirm the diag'nosis, of course, precluded operability. - The lymph nodes represent the largest g-toup of In a previous series of collected cases we were abIe to 1R98 cases in which metastases to the lymph nodes were The incidences in this ~oup Of cases were as follows: bronchial, 69.7%; abdominal, 20.7%; cervical, 17.4%; peritoneal, 8.1%; fliac, 6.8%; axillary, 6.6%; 6.5%; supraclavicular, 4.2%; biliary, 3.1%; submax~ary, and femoro-inguinal, 22% (Fig.-15). It was amazing that abdominal nodes were involved so ffequendy, occurring 20.7%.. In Stein and Joslin's series of I00 cases, the toneal nodes were involved in 27%, the mesenteric in 26%, sup .radavicular in 9%, the cervical in 6%, and the axillary As shown by these figaxres, the incidence of metastases is This migant indicate that the sur~cal treatment of primary chiogenic malignancy is relatively hopeIess. It should be that these fig~ares are based on autopsy cases in which the obviously were advanced. The fact that in approximately o[ cases the metastases were limited ~x~ the regional lymph makes the prognosis as regards the surgical treatment

y BRONCHIOGENIC CARCINOMA _~4etasta~es ~@ pneumor~omies [] ~mber ~ ca~e~ t8.9%

OCHSNER, DIXON, AND DEBAKEy nant disease of the bronchus it is as important to remove regional lymph nodes together with the primary focus as it is i do an axillary dissection for malignant lesions of the breast. necessity of total pneumonectomy in cases of bronchiogenic noma has been repeatedly emphasized by us, because it is im tive to remove not only the regional lymph nodes, but also collecting lymph channels, which is frequently not_ possible lobectomies. This contention is based upon the fundamen tornlc investigadons of Rouvidre (Fig. 17). Because of the extensive metastases, particularly in cases are seen late, operation is either contraindicated or resection! not" possible in most cases which are operated upon. found in 155 cases of primary bronchiogenic carcinoma that eration was feasible in 52 (3$.6%), but that resection was s~le in only ~7 (17.4%). In a previous publication we that in a collected series of 139 cases, only 68 (49%) were operable, whereas in our series of 30 cases at that time, 19 were operable. In a series of 9_94 cases reported by Brock, cotomy was advocated in 40, refused in 4, and performed in • (16%); 18 (8%) were resected; 18 were inoperable at the time thoracotomy, making a total of 206 (9_9%) which were " Fetter, in a series of 31 proved cases of carcinoma of the lung the Philadelphia Naval Hospital, states that only 7 were consid- ered candidates for exploratory thoracotomy; one tel'reed o don, and the other 6 were found to be i~.operable. Overhoh re- ports 165 cases in which the diagnosis was verified during life 156 and was made at autopsy in 8. Of this .number, 71 (46%) were not even considered amenable for surgery; 85 (51%) were o~ered surgery, but 5 refused. Of the 80 who were operated upon, 32 (42.6%) had curative resection, 9 had palliative resection, and 39 were inoperable. This author states that 2 out of every 5 ca~s -.. - explored were found to be free of an extrapulmonary extension- " • -- At..the Mayo Clinic, as reported by Clagett and Brindley, explora- . ti0n .was, _thought to-be feasible in only 18% of the cases whi_ch • -i_ . -. were micr~, pically proved to be a bronchiogenic carcinoma and ~7-~ -. - in oliiy 8~ of the entire group was the diagnosis imde .... :~ Of the 90 cases which were operated upon, resection was i::-;:::---::.:- > :: :.. " :. -. :> ........ ~7 ~ - " " 7I ~" "- - " ~ " - -~ '" ecto per: riot but pne ing plo wa in ill

.ble BRONCHIOGENIC CARCINOMA ectomy and 2 a lobectomy." In our 190 cases in which a diagnosis of primary bronchiogenic carcinoma was made, operation was performed in 10(5 cases (55.8%), in 49 cases (25.8%) operation was not considered feasible, and in 35 (18.4%) operation was advised but refused. Of the 106 which were explored, 58 (54.7%) had pneumonectomies and 48 (45.3%) were found to be inoperable (Fig. 18). Of }39 cases which were definitely proved as hav- ing bronchiogenic carcinoma, operation was performed in 106 (76.2%), pneumonectomy was possible in 58 cases (41~7%), ex- ploration only was done in 't8 cases (34.5%), and no-operarion was performed in 33 (23.7%) (Fi~. 19 and 20). In the 58 cases in which pneumonectomy was possible metastasis was found in the mediastinal nodes in 33 cases (56.9%), they were suspected in an additional 11 cases (18~q%), and no metastase~ were found in I4 cases (24.I%) (Fig. 16). There are few conditions in which the onset is more insidious than in bronchiogenic carcinoma, and unfortunately there are no characteristic symptoms and signs of the lesion. This has doubtedly be~n responsible for the high incidence of incorrect diagnoses being made early and for the delay in suspecting the condition. In a large number of cases there is a. history of an antecedent respiratory tract in~ection which is usually diagT~osed as influenza or a pneumonlds from which the patient does not recover as he normally should. Frequently the symptoms are disregarded because the cough which is present is attributed to smoking. Hochberg and. Lederer state, "At present when cough seems to be habitual with ~e average patient, every patient with a cough cannot be looked'upon with suspicion as having plasm of the lung.'" In spite of this statement, cough is the most frequent sy~nptom of bronchiogenic carcinoma. According to Brunn. and Simons, it occurred in 65% and 72% of their respec*: " rive collected cases. In Brines and Kenning's series, cough was '. present.in 87%, ~hereas in FrisselI and Knox's series it '~ras' present in 91% and was the first.symptom in 23.8%. B~o~k has emphasized, the possibility Of bronchiogenic "malignancy being"

---

BRONCHIOGENIC CARCINOMA oma masked by an antecedent pulmonary lesion without the diagnosis being suspected. He states that delayed resolution or an unre- solved pneumonia and chronic lung sepsis should be suspected" as being caused by bronchiogenic .carcinoma, In 82% of Over- holt's cases, there was a complaint of chronic cough on first con- sulting a physician, and in 63% the condition was incorrectly diagnosed by the first physician consulted. The various condi- tions for which treatment was ~ven in the 63% were as follows: tubercul6sis, 40 cases; unxesolved pneumonia, I8 cases; lung ab- scess, I3 cases; bronchit£s, I1 cases; asthma, 5 cases; heart disease, 4 cases; pleurisy, 6 cases; metastatic .malignancy, 2, and misceI- laneous, 9. Staehelin round cough was present in all but 16% o[ his cases. In 448 cases reported by Tinney, it was present in 91% of Clagett and Brindley's 45 cases which were operable, and in 76% of their 45 cases which were inope~ble. Cough was pres- ent in 89.7% of our cases of pneumonectomy for bronchiogenic carcinoma (Fig. 21). The cough may be either dry or may be associated with e_xpectoradon- When it is associated with hemop- tysis, it is of ~eat ~alue. In 25.8% of Brines' and Kemain~'s series, symptoms o£ acute respiratory tract in£ection were the ini- tial symptoms, and hemoptysis was the chief complaint in 18.9%. It was present at ~me time during the course of the iIlness in 48.3%. Simonn reported this symptom in 40% of his collected series. Hem0ptysis was present in 62% of Clagett and Brindley's operable cases and in 4:7% of the inoperable ones. Hemoptysis was present in .our pneumonectomy series ha 53.4%: Thoracic discomfort may be the only manifestation of bronchio~nic carci- noma. It was present in ~60%, and was. the chief complaint in 44.8% o£ Brine~ and K~nings's Series. These respective inci- dences were 71.7% and 21.7% in Frissell and Knox's series. Simons found it present in 59.8% of his series. It was present in 70% of Overholt's_ 165 cases, and it was present in 31% .our ca~es. It was presentY in 58%-ot~ the operable, canes and"m 62% of the ino~erable cases reported by Clagett and Brindley. Dyspnea is an i .n~requent ~nit~estadon, andis occasionally, out of proportion to the pulmonary involvement. It was present ha 48.3% of ~;Ur case~.. In 2 it was the most prominent.clinical mani- festati~, but i~is usually a late manifestation. It usually indg

OCHSNER, DIXON, 'AND D~:BAKEY . Cc~~iw Inc~er~ of ~co~ed F~ 21-Graphic representation o£ comparative incidence of ~rmptom~ in 58 of author~" cases of c~xdnoma o[ the lung.

BRONCHIOGENIC CARCINOMA ,ares extensive involvement. It was present in 96 of Overholt's 165 cases and in 73 of Staehelin's 115 cases. In Clagett and s series it occurred in 42%. Occasionally a padent with a bronchiogenic carcinoma may no symptoms referable to the thorax. Hochberg and Lederer in 13 cases in which there were no thoracic symptoms found the .most fi-equent complaint was as follows: epigastric distress, 7 anorexia, 7 (5~.9%); nausea and ~'omidng, 6 (46.2%); 5 (~8.5%); loss of weight, 4 (~0.8%); constipation, 4 (50.8%); aphasia, 2 (15.4%). There were no symptoms referable to the respiratory tract in 5% of cases reported by Timaey,.the diagnosis-being made usually during routine roentgenoga-am of the thorax. In 4% of his cases, the only symptoms were those rderable to the central nervous system. We, too, have observed similar experiences in which there were no ,symptoms referable to the thorax. In 4 of our cases which were operated upon and which were resecmble, the original manifestadons were nausea and vomiting, the diag-nosis being first suspected upon fluoroscopy of the esophagus. We have recend~ observed a patient who con- suited the Clinic becahse he thought he had sinusitis. He had. no s)wnptoms referable to his chest but the routine thoracic film demonstrated_ the bronchiogertic lesion. Loss of weight may be an important manifestation. It was observed in 71% of Tinncy's cases and in 63.8% of our cases. The: physical findin~ in bronchiogenic carcinoma vary con- siderably according to location and the size of the tumor. In the small lesion which is not large enough to produce bronchial ob- struction there are few., if any, early physical finding. However, in th~ centrally locatdd tumor even though the lesion may be quite small but sufficiendy large to occlude the involved bronchus partially, th.e physical findings may be quite ma~,ked, because of the resulting atelect~is. As emphasized by Tinney, careful in- spection of th~ thorax in the recumbent position is of importan.ce, in order to elidt minor changesin physical findings. It is our belief that every respiratory disturbancein a patient past forty years o~ age which cannot be explained otherwise should, be i~-. vestigated in order to exclude a prima/y bronchiogenic carcinoma. An-acute respirator/tract infection which does not subside within - 1219

OCHSNER, DIXON, AND DEBAKEY a reasonable period of time in a patient past forty years of especially a man, should be considered as caused by a mali bronchiogenic lesion until it has been definitely disproved. occurrence of a pulmonary abscess in an older has not had an antecedent operation or an antecedent respiratory tract h'ffection must be considered a malig-nant until proved otherwise. Only by such constant v~ilance caa" early diagnosis of carcinoma of the bronchus be made. The clinical course o[ a bronchiogenic carcinoma may gress relatively slowly, although from the reported statistics len~-nh of time elapsing from the first onset o[ symptoms death is usually short. This is undoubtedly due to the fact it is ditficult to determine just when the symptoms b%,im cording to Koletsky, the average duration of symptoms from set to death is six months. Kihg found an average duration life of 9.3 months. Frissell and Knox had one patient with toms of less than one month's duration, 15 with symptoms of than three months' duration, 13 with symptoms of less than months' duration, 9 with symptoms of less than twelve duration, 5 with symptoms of less than fifteen months' and only 3 with symptoms lasting over fifteen months. and Kenning found that 4% 6f patients had s)anptoms less one month, 27.8% from .one to three months, 39.7% from to six months, 9.2% from six to nine months, and 7% ~or than a year. The average Iength 0f life in D'Aunoy, Pearson, Halpert's series of cases was five months. In Clagett and ley's series of 90 cases in which exploration was done, 40% of the 45 operable cases had symptoms from one to six months and 51' of .the inoperable.cases had symptoms for the same lengxh of time;) 33% of-the operable cases had symptoms from seven to months and 38% of the inoperable cases; 9,% of the operable cases had symptoms from'thirteen to. eighteen months and 4~ _of the inoperable cases; 9% of the operable cases had symptomS. .~.-.It is our. belief in contradistinction to many of these. ;_. that the growth of bronchiogenic .carcinoma is slow and that the~, statistics axe based upq.n histories which were taken by _. not;~onsider the possibili.ty of carcinoma of the .. Y~ ~ ," ; - - 22Z. 5"-': . ...-."l.:! "7 " ."7 "".. := ." :-. "; ".." .... -'..-:, :" -% - .-..'". -:7::;:.-- -" -f' - :;.: " - , ": '- ': " .... '.'" -.:_ _.~-:-:z:- ~7- ~'- 2%:--" ~

;rind- ff the time; ,velve ruble )tOIIlS BRONCHIOGENIC CARCINOMA and consequently made no serious effort to elicit the inaugural symptoms. Overholt found that in 125.cases in which the data ~-ere accurate, only 36% placed themselves under a doctor's care within a month of their s)wnptoms. Most of these did not consult a physician until three months after the onset and did not have an x-ray until an additional three months had elapsed. The diag- nosis was usually not established, until nine months had elapsed. Robertson found that only 34% of the cases of bronchiogenic carcinoma which he saw had been accurately diagnosed by the family physician before being sent in. D/.AGNOSIS The most important factor in the diagnosis of bronchiogenic carcinoma is a consideration of its possible occurrence. As men- tioned above, it must be suspected in every male patient forty years or older with cough, hemoptTsis, or thoracic discomfort. Sehrt, in 1904, reported only 3.3 ~e7o of antemortem dia~nose~ in a serie~ of 178 cases, indicating that until recently an incidence of correct diagnoses was small. The incidence of correct diagnoses ~a~ 36.8/~.o of Probst's series, 30~ro in Kikuth's, and _0/o of those of Cottin and his co-workers. Koletsky found that a correct diag- nosis was made in only 56% of his cases. In 9 ca~es a dia~osis of tuberculosis was made and in an additional 9 cases a diagnosis of abscess was erroueousIy made. Undoubtedly the high incidence of incorrect diag-nosis is due in part to ~e ability of bronchiogenic carcinoma to simulate other conditions. However, of even ~eater importance in this regard is the frequent lack of its comideradon. Roentgenography is of coiasiderable diag-nostic importance in bronchiogenic lesions. Th~ roentgenographic film may show a shadow produced by the tumor itself, a shadow produced by metastatic mediastinal nodes, or a shadow produced by an atelec- tatic lung. O~ the other hand, in small lesions a roentgeno~m-am may show nothing so that'a ndg-4.tive roentgenogram is of no value and does not in any way rule out. the possibility of a bronchiogettic tumor. Roentgenographic interpretation of c~.ntrally located le- sions is generally more difficult because of the corffusion with hilar shadows-prodiaced by other lesions and by normal structures. -A t221 ," • "

OCHSNER, DIXON, AND DEBAKEY This is of particular significance because most plasia are in the hilar region. Occasionally, a very small erally located tumor, insu~ficiemly large enough to ca.st a is associated with massive mediastinal metastases which considered the prima~ tumor. We have observed two o~ this type, one in which it was impossible e'cen after ~ the presence o~ the peripherally located tumor to find i-t roentgenographic plate in subsequent examinations. In trally located lesions which produce bronchial obstruction the characteristic atelectasis with shif-ting o~ the mediastix ward the affected side and elevation o~ the diaphra~m on tha X-ray examination is diagnostic in a high incidence og Andrns' series o~" 64 cases the correct condition was roentgenologicafly in 58 (90.6%). According to Moersch, a rect diagnosis was made in 73% of 448 proved cases. ports a correct diagnosis roentgenologically in 97.1%. phy (Chaoul and Grineder; Gravano and Malenchini) may be additional value in making a diagnosis of bronchiogenic noma. In our own series roentgenography made a correct nosis in 79.3% (Fig. 22). The roentgen diagnoses made in 58 cases in which a pneumonectomy was done f6r carcinoma although correct in over three-fourths of the varied from lung abscess to pneumonia (Fig. 23). Bronchog-raphy is also a valuable diagnostic aid. It is lady advantageous in those cases in which a mass does not cast shadow and in which the tumor is in a bronchus which is visibIe by bxonchoscopic examination. Farinas has the importance of bronchography and states that it is possible determine 'i[ the lesion is polypoid or infiltrated and the extension into the bronchial wall. Whereas we orig'inally tended that bronchography performed bythe passive technic desirable, we now believe that visualization ot the bronchial by mearis of a catheter introduced into the trachea is to visualize the bronchi satisfactorily. -. " • The best method of m.aking a positive diag-nosis of genic carcinoma is bronchoscopic .visualization of the tumor the obtaining-of a piec.e of tissue for biopsy. bronchoscopy is not performed often enough. It

a cast ~s as ized ~ ~le m ladve con-[ I tree BRONCHIOGENIC CARCINOMA .0.9 °~o of Kolet~ky's cases; in 50~ of Brines' and Kennings's cases, in 50% of King's cases. Theoretically, bronchoscopy should ire a positive diag-nosis in about 70% of cases, because this is about the incidence of hilar involvement. On the .other hand, in .[th°se cases in which the lesion is in the upper lobe bronchus, the [lesion may be beyond the vision of the bronchoscopi, t. In order 7to visualize the upper lobe bronchus, Gebauer has described a !bronchoscopic mirror. Frequently in a lesion involving the upper lobe in which bronchoscopy does not permit the visualization of the lesion because of the position of the bronchus, .the lesion can Ix made visible by the production of artificial pneumothorax, which permits the upper lobe to assume a lower position and come more in line with the trachea. Holinger and Hara recently reported that a positive diag-nosis was made bronchoscopically in 8 4~" According to Betts, a positive diag-nosis was made bron- choscopically in 46 of 6_'2 cases (74%). In 80 cases reported by Craver, from the Memorial Hospital, in which bronchoscopy made the final diagnosis, this method alone was responsible for the diao, mosis in 59. It ~-as supplemented by autopsy in 11 and supplemented by other methods in I0. According to White and his associates bronchoscopy was performed in 24 (43.%). of 56 cases. A positive biopsy was obtained in 19 (79%). By means of a bron- choscopy a Positive diag-nosis was made in 62% of Overholt's cases. In our ~8 pneumonectomies a bronchoscopy was done in 50 cases (86.2%) (Fig. 25). Of these 50 cases, no biopsy was done in 24 (48%) cases, whereas biopsy was done in 2fi (52~o) cases. Of the 2~ cases in which a biopsy x,~s done, the findin~ were Positive in 9.0 cases (7&9%) and were iaeg-ative in ~ (23.1%) cases (Fig. 24). The finding of malig~nant cells in expectorated material is of diaga~osdc importance and careful examination of the sputum should be made iri all suspected cases of bronchiogenic carcinoma. Dudgeon found carcinoma ceils in ~0% of cases in whom a posi- tive diag-n0sis of pulmonary, neoplasm was subsequendy proved. This method of diaomaosis has been used success~lly by Barrett and by Gamba and Lamberti. It has been of .value in a few of our cases.._In using this. method it is important, as. emphasized by Edw'~rds, that. the patholog-ht examining this ti~. ue should be • 122~ ,~ • -

OCHSNER, DIXON, AND DEBAKEY I~su.[ts o~ 50 l~onchoscopLc- ,, ' ,,.,-, 6xami.na.{:ior~ rticularl :lls in to th ted ma in lit Craver metho 58. It and b~- o* .1935 to 1 we [eel way of ar whom th, cases i~ v mg remo by DolIe, of diag-nc A fin: whom a don. tion was preopem sary in o genic detected operable tion in by explc mary les

•ors" 58 ;copic BRONCHIOGENIC CARCINOMA ~rticularly trained in the recognition of these cells. Similarly, a e~onstration according to 3Iandlebaum's technic of malig-nant in the pleural fluid in cases in which there has been exten- to the pleura is o~: diagnostic importance. Seecol~ demon- malignant cells in about 7_0% o£ cases and Goldman in 80%. This method is of little use early in the disease, because of the relatively late extension to the periphery ~ in peripherally loca.ted lesions. The importance ol~ the :thod lies principally in its prognostic value. Craver has been an ardent advocate of aspiration biopsy and method is undoubtedly a splendid one for use in making a diagnosis of bronchiogenic carcinoma. In 66 cases in which the -dia~osis was made in this way, it was made by this method alone in 58. It was supplemented by autopsy in two additional, cases and bv other methods in 6. This same author states that from 1935 to 1939, aspiration biopsy accounted for 51.6% ot~ all the cases in which histolo~c proof was obtained. On the other hand, we feel that this diagnostic method even though it is a splendid x~-av of arriving at a diagnosis should not be used in patients in whom there is chas~ce of resection, because we have observed two cases in which implant metastases occurred in the pleura follow- ing removal of the specimen. A similar case has been rel~rted by Dolley and Jones. Alexander is also opposed to this method of diagx~osis except in the moribund case. .-k final method of diagnosis which is justified in a patidnt in whom a diagnosis cannot be made otherwise is thoracic explora- tion. Whereas at one time we contended that thoracic explora- tion was not justified a'nless a positive diag-nosis could be made preoperatively, we now believe that thoracic exploration is neces- sary in order to confirm or rule out the possibility of a bronchio- genic neoplasm. Onlyin th/sway can many of the early cases be detected and can a resection be done ~,t a time when the lesion is operable. Thornton has .e~iphasized the imp0rtance.of, explora- tion in suspedted cases and reporu. 2 cases in which it was only by exploration that it was pbs/.ible to differentiate between a pri- mar,/lesion of the hmg and a metastatic one.

OCHSNER, DIXON, AND DEBAKEY TRKAT~tENT OF BRONCHIOGENIC CARCINOMA .-kS in carcinoma elsewhere, irradiation by x-ray and has been used in the treatment of bronchiogenic is the consensus at the present" time, however, that the rive treatment of this lesion ~ removal of the diseased Whereas occasionally a case may be benefited by irradiatior el'ally the results ~rom irradiation are of little or no value. be used as a palliative procedure anc~ is of value in ho in alleviating symptoms. In a series of patients created by don King found that the average duration of life of those was 15.4 months as compared with 9.3 months for the entire I Chandler and Potter found that the average duration of 59 patients treated by irbadiafion was eleven month,, for 61 untreated cases wa~ six months. Ormerod also slight increase in the duration of life in treated patients. in a series of 160 calm advised local radon therapy in 10. but 3 of these the radon therapy was suggested merely as a cebo. He stated that irradiation had little to do with life; in those cases in which the patient's life was apparendy longed it was probably because the carcinoma g'rew slowly. quoted the history of a patient.who lived eight years after his symptoms without any treatment. Faust similarly found that about 15% of 200 patients showed definite improvement ing irradiation. Baum and his associates report a case of cell sarcoma of the bronchus in which an apparent cure was rained by deep roentgen therapy, but in which death as a result of a rapidly pro~essive pulmonary tuberculosis plicated by a pulmonary g'anga'ene resulting from thrombosis the pul~onary artery. Bloch and Bog'ardus observed 46 which were not r.reated with x-ray of which 35 (75.2%) are in 9. the results are unknown, and 2 are still living. In a seri~ 6f 42 in which x-ray was used 36 (85.7%) are dead, in.4 results are unknown, and 2 are living. " craver repori~ i75 cases reporte~. at tl~e'Memorial 142 of which~w~retreated by .roentgen therapy, 14 by pack', 17 tin,reared, and only 2 were operated upon, but these were'operated up6n at other institutions. Of the 175 " : "_ .- ' ' '. .... ~ ~26 ~. dead. the ax er: i by a corr therapy. nine yea therapy fibrosis size of a embed& being he are e~-id two gTO~ roent~et; ~oup to twelV, WaS usec in a seri, survival months_ group slow-gro lobar

fife ]3RONCHIOGENIC cARCINOMA _o0 are living, one of these is one of the two patients operated ten months previously, _9 are of too recent observation to suitable for analysis. Of the 19 cases which were treated by _o survived over four and one-half years, one for three years, 'for two years, one for eighteen months, 5 for six months, and [or less than sLx months. Farberov and Baslow report 44 cases ,roved bronchiogenic carcinoma which were treated with x-ray, ida an average survival of eight months. In the first ~oup there 6 who lived 15 months to three years after the diagnosis was in the second ~oup of untreated cases, one lived over two and one whose disease was proved only at autopsy lived over i three years after onset of the disease. Hammond reports 26 cases [treated by x-ray, 15 by a combination of intrabronchial radium !therapy and external x-ray and 11 by x-ray alone. Of these, 99 are dead, the longest survival being 39 months, the shortest nine days, the average six months. The four who are still alive were treated by a combination of intrabronchial radium therapy and roentgen therapy. The danger of roentgen therapy is evidenced by a case reported by Jacobsen in which a man had deep x-ray therapy over nine years and whose death was undoubtedly the result of the therapy and not directly due to the neoplasm. A very, marked fibrosis of the lung had resulted, the lung being only. about the size of a man's fist. Carcinoma was found just outside the hilum embedded in a dense mass of fibrous tissue and was apparendy being held in check by the fibrous tissue. Palliative effects of x-ray are evidenced by the cases reported by Leddy and Moersch. Of two ~oups of cases of..125 each, one of which was treated with roentgen ray and the other not, it was found that in the former ~oup treated by x-ray there were 95 patients who lived from one to twelve years, whereas in the hitter ~oup in which no treatment was used no survival was longer t,han one year. Santy found that in a series of 53 cases in which no irradiation was used the ave.rage survival after onset of symptoms was 10.5 months whereas in 10 cases treated with x-ray the average survival of life was 11.9 months. On the other hand, one of .the cases in the irradiated ~oup was a malignant bronchial adenoma which is notoriously slow-growing. This patient lived.forty-eight months and died of lobar pneumonia, If this case were excluded, the average dura- l227

OCHSNER, DIXON, AND D~BAKEy tion of survival [or the treated cases would be 10.1months. son and Urban report the results obtained by means of ml volt roentgen therapy for bronchiogenic carcinoma. seven of these cases were treated, of which 25 complet course. Of these, 23 died. The averao~e length of life after: ing therapy was 7.3 months. The other 2 are still alive wi dence of the residual disease, 17 and 9 months respectively beginning treatment. Flood reports 70 cases ot~ primary genic carcinoma treated in the Radiolo~c Department Royal Cancer Hospital. Of these, 69 are dead, one is years and two months after the be~nning of the treatmen It is thus seen that irradiation may be of value. N1 others (Bonner; Fried; Frissell; Hunt; Kaplan; Kernan; Pancoast; Robet; Rodenbaugh; Rubenstone; Seifert; Staehelin; and Turtle) have observed that irradiation for noma of the lung is of little or no value. Overholt and found that patients subjected to irradiation lived only as long as untreated cases. Pohle reporting the results radiation in 41 cases in which the course of irradiation x~as pleted found that whereas one patient survived months, only 14% survived twelve months, an average of 8.6 months. Widmann believes, that irradiation is of palliative value. He compared a series of 167 cases which irradiated with a series of 419 which received no treatment. ot~ the latter control group lived as long as one year and 11-° l~ved less than sLx months. Of the 167 irradiated cases, 18 twelve months or longer from the time of diagnosis and 5 two to six years. Schinz reported from the Zfirich Radiolo~c stitute the following results in 79 cases treated with no of metastasis; 6~ were living at the end ot~ three months, 16~o were improved and able to work. At the end of six ~5~o were living and only 7.5~o were improved and able to at the end of the year only 10% were living and only 5% able to work. Of the 20 patients who had definite met ~o were living two months after treatment and only 6.2% able to work. Of this ~oup only 6.2~'o were alive three after treatment and only 2.5~o were im.prove.cl and able Schinz states that, of all malignant diseases, irradiation offer~ ~ - . -.: ~

ived - In~ only BRONCHIOGENIC CARCINOMA ~teast in primary bronchiogenic carcinoma. From these statistics it can be concluded that irradiation ol~ers practically no chance o[ cure and that it should be reserved Jor the inoperable case of bronchiogenic carcinoma or [or cases in which operation is defi- nitely contraindicated and in which it is to be used as a palliative procedure. • Surgical extirpation of the involved lung is the only curative ~reatment of bronchiogenic carcinoma. Whereas it has been sug- gested t_b, at simple resection of an involved lobe is all that is neces- may,, it is our belief that this procedure is wholly inadequate Bonner; Fried; Frissell and Knox; Hunt; Kaplan; Kernan; Max- well and Nicholson; Pancoast and Pendergxass; Robet; Roden- baugh; Rubenstone and Schwartz; Seifert; Simpson; Staehelin; ;Tuttle and Womack). Certainly, only by complete excision of !the entire involved lung can the primary focus be adequately rernoved and 'only in this ~ay can an adequate extirpation of the regional lymph nodes be accomplished..as mentioned pre- viously, the fact that the incidence of lymph node involvement is relatively high is sufficient indication that the re~onal lymph nodes should be extirpated together with the.primary lesion. AIo though total pneumonectomy may seem a radical procedure, par- dcularly in an elderly person suffering from a bronchiogenic new plasm it should be realized that there can or should be no com- promise when treating malig-nant disease and that complete, ex- tirpation is a requisite for cure. The production of an artificial pneumothorax as a preliminary prodedure is extremely important before performing a total pneu- monectomy. This should be done in stages, the amount o[ intra- pleurat pressure being increased gradually until the pressure is definitely on the positive side. ; Preoperative pneumothorax is of dia~ostic importance in determining the presence, extent, and location of adhesions, thus permitting the preoperative planning of the operative procedure. Another equally important advantage of a pneumothorax is g-radual compression of the pulmonary .capillary bed, which gives the heart time to compensate for the increased peripheral resistance in this area and mitigates the sud- den change that ,~ollows the cutting off o[ the blood to the in. .volved lung_at the time of li~tion o[ the pulmonary vessel. Thi~

OCHSNER, DIXON, AND DEBAKEY is particularly important in elderly persons whose cardiac is diminished and in whom malig~nant tumors are likely to If there is any infection distal to the tumor, preoperative scopic aspiration is of value and the administration o[ ch~ p~utic agents is advantageous. I[ there is an anemia, of whole blood should be done preoperatively. It is a rule service to administer vitamin C to these patients pre~ because many of them have a vitamin C deficiency as the associated infection. The re-establishment o[ a normal balance is equally as important. ~V'hereas older persons withstand an operative procedure as well as younger ones lieve that age itself is not necessarily a contraindication to monectomy and have performed the procedure on a seventy-two )'ears of age. It is desirable to have an infusion started before the is begun using a large needle, preferably a DeBakey trocar. Although usually saline solution is started at the ning, at least 1 ;500 cc. of whole blood should be available operating room. Generally tiffs amount is not neces~-y, bt casionally an injury, to one of the larger vesseh does occur, ing in the loss of a cousiderable amount of blood in a very period of time. The value of having this amount of blood able was demonstrated in one of our recent ca~es in which , in the arch of the aorta resulted [r0m the fi'eeing of nodes in this area. Although the opening in the aorta proximately 3 cm. in length, it was possible to administer blood through a DeBakey transfusion needle, which is used in these cases, to permit suture of the vessel. The systolic pressure dropped only to 80, and after closure o[ vessel came back immediately to the normal preoperative This patient recovered, returned hom~ and is doing nicely. • :-There are probably few conditions in surgery in trained anesthetist is more essential than in thoracic st~rger/i particularly in performing a pneumone~tomy.. We propane intratracheal itive ressure anesthesia, ..... pos ~ p .... ..

but d .hich~ it BRONCHIOGENIC CARCINOMA ~ces the least reaction of all the anesthetic agents whicla we have In pneumonectomy [or bronchiogenic carcinomas we prefer anterior incision popularized by Rienhoff, although we have the posterolateral incision ,as suggested by Crafoord in some A posterolateral incision is particularly undesirable in older ,ns whose cardiovascular system is already impaired because placing of the patient on his sound side further interferes with his cardiorespiratory function. It also has the distinct disadvan- that the mediastinum in which almost all of the dissection is :done g-ravitates away from the operative field. On the other hand, i if there are dense adhesions posteriorly and laterally there may be some diNcultv in freeing the adhesions by sharp dissection, and ~ ~t is in these instances that the posterolateral incision is particu- i larlv indicated. Although we prefer the anterior approach for re- ! moral of a lung for neoplastic disease, we almost invariably use a posterolateral approach in performing a lobectomy for inilamma- roD" disease, because in these cases there is l~ely to be more ad- hesions and dissecting within the lung is easier when approaches can be made from several sides. ~aen the anterior approach is used, the patient is placed in the supine position with the side to be operated upon s.lighfly ele- vated by placing sandba~ beneath the shoulder and the hip of the affected side. This is advantageous because it makes the pro- cedure somewhat easier for the operator and detkn_itely easier for the assistants who othenvise have considerable difficulty in visual- izing the mediastinum. The incision is made in the third inter- costal space extending from th.e midportion of the sternum to the midaxillary line. After divisi6n of the skin, Fascia, and pectoral muscles, the intercostal muscles and pleura are incised close to the upper border o~ the fourth rib. The incision is extended medially to the Iateml portion of the costal cartilages. The.third and fourth costal cartilages are divided subperichondrially in the parasternal portions and the medial portion of each costal carti- lage is removed by means of a rbngeur. By means of a hand inn-o- duced into the pleural cavity, it is possible to feel the pulsation of the internal mammary vessels posterior to the costal cartilages. A transfixion suture of crochet cotton (No. 20) is placed around 12~1

OCHSNER, DIXON, AND D~BAKEY these vessels in the second and fourth intercostal spaces, tively, securing these vessels. Following this, the intercostal i cles and pleura and the internal mammary vessels are the sternum. The thoracic wound is opened by third and fourth ribs by means ,o~ rib spreader. Any which are found are divided by sharp dissection. If the are very dense, making it impossible to determine where ends and where the pleura begins, it is best to remove the pleura with the lung. Early in the operation one should mine whether the case is operable or not. The presence o[ mediastinal nodes does not preclude operability nor does ment of the pericardium. Unless the tumor is densely the aorta, to the vena cava or to the trachea we believe tempted extirpation is justified because noddxtg else offers unfortunate individuals anything curative. In many which the case is appaxendy inoperable in the beginning, mobilization of the mediastinal structures, extirpation of the and its contained tumor is found possible. AYter fi'eeing hesions an incision is made in the mediastinal pleura poste/ior to the phrenic nel~'e. The incision is extended and below around the hilum. In lesions on the right vena azygos major is~ isolated, doubly li~t6d, and doubly fixed, using crochet cotton, No. 10, following which the divided between the ligatures. By careful bIunt dissection, the index finger and small pledgers of cotton on the end. forceps, the individual hilar structures are isolated, usually ladng the superior pulmonary vein. Each of these s doubly lig-ated and doubly transfixed with crochet cotton. fortunately the pulmonary artery and two pulmonary veins large and short vessels so that f'requendy the diameter of the sel is larger than its available length. Ochsner forceps are on the lung side of the vessel be[ore its division and the divided immediately adjacent to the clamp so ~hat a long portion of the vessel i~ Ieft on the cardiac side. In five paratively recent cases it Was ,neCessary to ligate and pulmonary" artery and two pulmonary veins " cause o~ extension o~ the lesion into the mediastinum. In these it was necessary 'to extirpate a large portion o~ the . - 1232 :.7 .

~S, rati adh, adah, here uld :e of zdh~ ~e ~ ,fiefs ," c~- _aing, he ng ded t side, ably he :t/on, the endi usually cructures : vehas of the i a five divide ,rdially, m fou~! ,f the BRONCHIOGENIC CARCINOMA ium in order to remove the involvement of that later. In one portion of the auricle was resected. None of these patients suf- any untoward symptoms. Followi.ng lig-ation and division the pulmonary vessels, the mediastinal 1)anph nodes are [reed the bronchus is isolated close to the carina. Two crushing ,s are applied to the bronchus just beyond the carina, and bronchus i.s divided by means of a scalpel between the clamps. posterior reflection of the pleura is then divided, and the is removedi The bronchus is closed by introducing a series crochet cotton ~No. 20) mattress sutures proximal to the clamp. iThe taro end sutures are tied before the clamp is removed, but :the others are not tied until the clamp is removed. Following re- ~mo~al of the clamp the mattress sutures are fled. The crushed end !bf the bronchus is then closed with a row of interrupted quilting cotton sutures..ks emphasized by Rienhoff, it is essential, we be- lieve, to cover the bronchial stump with pleura following closure of the bronchus. In fact, we believe that this is the most impor- tant maneuver in the prevention of an insufficiency of the stump. Whenever possible, the edge~ of the mediastinal pleura are ap- proximated, covering the bronchus and the hilar vessels. If there is insufficient amount of pleura to cover the bronchial stump satisfactorily, a pleural flap from the pericardium or from the mediastinum is swung into position to cover the bronchus. After careful pleuralizafion of the bronchial stump, the thoracic wall is closed without drainage using crochet No. 10 cotton sutures to approximate the adjacent ribs by placing the sutures around the fibs. These are tied after dae ribs areapproximated by means of retractors. The muscles and Fascia are approximated by means of interrupted quilting cotton sutures. The skin is closed with a con- tinuous quilting cotton suture. A compression bandage of ma- chinist's waste is applied. Immediately after the operation a bronchoscopy is done in order t6 aspirate any secretion that might be present in the tracheobronchial tree. This i~ of importance because it ~eatly facilitates the postoperative convalescence since it is not. necessary [or the patient to expectorate the material that might have ac- Cumulated in the bronchi. 1233

OCHSNER, DIXON, AND D~BAKEY Postoperati~ely, oxygen is administered during twenty-four hours, usually by means of a Lombard mask. patient is kept in the supine position until there is recovery fi-om the anesthesia and then placed on a back addition to the blood which is received in the operating the patient is ~ven a liter to a liter and one-half of 5% making a total quanti~" of about 3,000 cc. in the first 24 Generally on the morning of the first postoperative day, dent is able to eat without any difficulty and is able to full diet by noon 'of that day. He is usually out of bed third or fourth day and with few exceptions is able to hospital between the seventh and tenth days. We have never served any undesirable sequelae because of early these cases. In fact, it is "considered distinctly desirable probably responsible for the rapid convalescence. The " seem to improve more rapidly and certainly are better as their morale is concerned because of early ambulation. most of these patients are beyond middle age and because have maligxiant disease, they are particular candidates for development of intravenous clotting. In order to obviate in addition to early ambulation we apply compression from their toes to their ~oin, which by compressing the ficial veins will increase the flo~,' ~f blood into the deep We have never observed intravenous clotting in any of cases which we believe is due to these prophylactic measures. ~.~sun~s o~ ~,NrU.x~oNrc'ro~c As is generally appreciated, total pneumonectomy for malignancy o~ the bronchus is a formidable procedure and in the past carried a prohibitively high mortality rate. At present time, however, because of improvement in operative nic and the better unde~ding o~ preoperative and tire treatment, the results have become much better. We collected a total of 414 reported c~ses of primary carcinoma in which a total pneumonectomy was done. Of number 219 02.8%) are living and 19~ (47.1%) are dead 25); Evarts Graham recently reported a series of 81 cases of nectomy [o ,ined mortali~ deathS, a me which cura ~ths; 2 die ted 13 ca.' with one hos of ~e pat tasis, 2 Fxo. 25 collected ca: Fro. -°6 thors" 58 ca Clagett a ectomy ~ the first died in months known ~ pleura ~ the oth, heard. month~, month: to thir

~neu- BRONCHIOGENIC CARCINOMA :tomy for primary pulmonary neoplastic disease with a com- bined mortality rate of 30%. In the last 25 cases, there were only deaths, a mortality rate of 12%. Overholt, in a series of 32 cases which curative resection of the lung was done, had 6 operative lths; 2 died of carcinoma and 4 of other causes. Johnson re- wrted 13 caseg of pneumonectomy for bronchiogenic carcinoma with one hospital death and 2 subsequent deaths ~rom anesthesia; .~ of the patients have gone two years without any evidence of :metastasis, 2 are well after one year, and 6 are less t[han one year. Clagett and Brindley report 43 cases in which total pneumon- ectomy was done for bronchiogenic carcinoma; 7 (16%) died in the first twenty-four hours after operation; 7 additional patients died in the hospital alter qi~eration; one patient died three months after-pneumonectomy was performed; one patient was known to have had a recurrence of the malignant lesion in the pleura nine months after operation; 4 cases were not followed; the other 4 were alive and without known recurrence when last heardl Those which were living were a~ follows: one to six months, 36%; seven to twelve months, 24%; thirteen to eighteen months, 20%; nineteen to twenty4our months, 4%; twenty-five to thirty months, 12%; and forty-eight months, 4%. .. 1235

OCHSNER, DIXON. AND D~BAKEY We have performed a total of 58 pneumonectomies mary bronchiogenic carcinoma of which 28 (48.2%) are and 30 (51.7%) are dead (Fig. 26). Of the 30 patients who 16 (27.6%) of the total cases died within the hospital (24.1%) of the total cases died after leaving the incidence of complications was relatively low (13.8%): 6 of the entire g-roup developed empyema, 2 (3A%) had a 12 :ratic make our 5~ ha red Fro. 27 Fro. Fro. 27-Graphic representation of results in authors' ~ p mies since 194 I. Fro. 28-Grapkic representation of mortality in authors' 58 ectomies, bronchus. Although a mortality rate of 51.7% is extremely it must be remembered that these represent all of the cases, the ones in the earlier period when relatively little was ahout the technic. In the ~8 pneumonectomies performed 194I, 22 (61,1%) are living and 14 (38.9%) are dead (Fig, When one considers the hospital mortality in the earlier and the later one, the importance of improved technic is emphasized. In the 2'2. cases which were operated upon. 1941, the hospital mortality rate was 45.4%, whereas in the cases, operated upon since 1941, there were only 2 deaths, a pital mortality rate of 5.6% (Fig. 28). The deaths which Occurred since leaving the hospital are higher now than in the earlier period. Whereas in the 22 ca.~es Which were crated upon before 1941 there were 6 (27.2%) deaths after months died of sect 58 indicati, of cases The surgery [or bro, the hosl l~th th

:s for ;ire who ital. BRONCHIOGENIC CARCINOMA the hospital, of the 36 cases operated upon since 1941 there .ere 12 deaths (33.3%). This high mortality rate at first con- deration may appear undesirable. However, we have n° apology make for it; it simply indicates that we have not been too strict our selection of cases and have done resections in patients who ave had extension beyond the local confines. Although they ,rvived the operation without ahy difficulty and many lived for _'29 Fro. ~0 29--Graphic representation og inddeace o[ resectabtliry ia 106 of 30--Graphic representation of hospital deaths following pneumon- months and sometimes years in perfect comfort, they subsequently died of their malig~xant lesion. The fact that we were able to re- sect 58 (54.7%) of 106 cases operated upon indicates a liberal indication for resection (Fig': 29). In a previously collected series o[ cases we found that ther~ was a reses_.tability incidence of only CONCLUSIONS The pro~ess made in the technicaI development of thoracic surgery and its effect upon .the operation of total pneumonectomy for bronchiogenic carcinoma are iLlustrated by a comparison of the hospiud deaths in our own cases operated upon bdore 1941 with those collected from the literature since 1939 and our own I~7 - o

OCHSNER., DIXON, AND DzBAKEY cases since 1941 (Fig. 30). In the 22 cases operated upon 1941, there was a 45.5~o hospital mortality rate, and in cases collected from the literature and operated upon since this hospital mortality, rote was 31.4%, whereas in our 36 erated upon since 1941 there were only 2 deaths, a hospital tality rate of 5.G~o. These figur.es demonstrate, we believe, conclusively that total pneumonectomy for primary carcinoma can be done relatively safely and is associated low mortality rate even though careful selection of the cases: done. Although there may be some criticism of our that attempted extirpation of the lung is justified even in the of apparent inoperability, we are wil .Ling to defend this stand we believe that.the outlook in the bronchiogenic mall sions is so bad and the death so horri ,~wing that one is attempting to perform a resection even though it may be palliative. This contention is justified particularly when a tality rate of slightly over 5% can be obtained. desideratum in the treatment of primary bronchiogenic noma is the removal of the lung together x~ith the regional nodes at a time when the lesion is sdlI limited to the " This desideratum will be attained only when diag~noses are early enough so that the re~:tion can be performed when sion is still limited to the lung.

BRONCHIOGENIC CARCINOMA

OCHSNER, DIXON, AND DzBAKEY cLZ

927. 4:370, I940. ~ ~927. ( speziellea ringer, 1951, 1936.

OCHSNER, DIXON, AND DEBAKEY

---

OCHSNER, DIXON, AND DEBAKEY Rouvs~, H.: Anatomy of the Human Lymphatic System, Ann Arbor, Edwards Brothers, Inc. RUB~"~$TONE, A. I., ~ M. SCHW~: Med.Jour.and Rec., 126:719, 1927. Iax~.: Ar~.f.~werbepa~u.Gewerbehyg., 1:582, 1930. I~m.: Fo~.a.d.~b.d.RO~ens~en. ~3:549, 1936. S~co~, D. P.: Pr~.NewYork Pa~.~., ~4:3, 1924. S~T: ~t~ bv W. Egenolf, Z~.f.~ebsfo~, 31:396, i930. "- St~toy~ ~ J.: ~m~ ~no~ of ~e Lung, ~, The Y~ 1927.

1245