Document R207XXEdZM3dOaBBLOK27Yw67

The New England Journal of Medicine Volume 273 Copyright, 1965, by die Man^ckuioti* Medico] fioclaty DECEMBER 16, 1965 Number 25 DISTRIBUTION OF POLONIUM?10 IN PULMONARY TISSUES OF CIGARETTE SMOKERS* John B. Little, M.D.,f Edward P, Radford, Jr., H. Louis McCombs, Vilma R. Hunt, B.D.S.fl and BOSTON IGARETTE smoke contains trace amounts of G an alpha-particle-emitting radioactive element, polonium1110 (Po910),1 a naturally occurring daugh ter isotope of radium898. In establishing whether this source of radiation exposure may be involved in the initiation of bronchial cancer in smokers, an im portant step is to show that pulmonary tissues of smokers, particularly certain regions of the bron chial epithelium, contain more of this element than those of nonsmokers. Because polonium has the property of binding strongly to solids, it is a reason able hypothesis that a major part of the inhaled PoTM may remain attached to smoke particles as they are excreted. On this basis analysis of pul monary tissues for Poao may also offer' some infor mation concerning the dynamics of smoke absorption and excretion in human lungs. Present knowledge of these turnover processes is largely bated on ani mal experiments. The present investigation was undertaken, there fore, to measure PoU0 concentrations in various pul monary tissues of smokers and nonsmokers. From such measurements estimates may be obtained of the radiation dose delivered to the lung and bron^ chial epithelium by this element. Equally important, ] however, these measurements offer an opportunity to evaluate the dynamics of pulmonary turnover of r smoke itself, MFvom Ujo Dcfrak*racr\i cd PKydoWy, Harvard School of X^Uic HbUH, and the Department of PatboWy. Harvard Medical ScbooL Supported l>y a contract (AT (3(M]*5l70). with the United State* Atomic Energy Comcrnaion, by grant (4T1-EH-27-03) from the of Stata Servian United State* Public Hodtli Soiwo, zrtd by lordiuljonftl grant from the Rockofdler Poundetien. tAssvUnl proioaror o( radiobielogr> Harvard School of Public Health,' corwwltanl in radiology, MauachuKU* General Hoiplul- tProfcssor of environmental Health, proftuor of phyilology and dl tCt<kr, Kettarlng Laboratory, Unlvwifty ot Clnolnnail OoUoge of Medi co (formerly, aiiociate profcaor of physiology, Hnrvaru School of Pablie Health). fiReatorch a&iodate In piihology, Harvard School of Public Health 1 formerly, fcllmv in oilhology, Harvard Medical School, aod rcarch auodaio, Wcit floxbtiry Vote ram AdmtnblraUtm Hospital)* fiRereareh auodato la physiology, Harvard School of Public Health. Materials This investigation is based on luDg specimens from 40 different subjects, of whom 36 have been studied in detail. Of the latter, 31 were whole lungs ob tained at autopsy, and 5 were surgical specimens from patients who underwent elective lobectomies or pneumonectomies. Among these patients 25 were currently cigarette smokers, 2 currently pipe smokers, and 1 formerly a cigarette smoker, as well as 8 who had never smoked. The clinical data on each subject, including smok ing histories, pertinent autopsy findings and causes of death are presented in Table I. Histories were obtained from hospital records and the attending physician and in most cases by direct contact with the next of kin. All current smokers had smoked continuously for periods of at least twenty years, up until a maximum of ten days before death or sur gery. The nonsmoking interval indicated in Table 1 represents the time elapsed between the moment cigarette smoking was stopped and death or sur gical intervention. Autopsy Specimens Methods The average elapsed time between death and au topsy was ten hours. With the exception of Cases 1-13 whole lungs were obtained directly from the autopsy table and brought immediately to our labo ratory. In Cases 1-13 an additional period up to twenty-four hours elapsed after autopsy; during this time the lungs were kept under refrigeration at 4C. Lungs either were dissected immediately after they were obtained or were quick frozen and subsequently dissected after careful thawing within ten days. In most cases several tissue samples were obtained from various parts of the bronchial tree for hiBtologic examination. These samples were taken either just 1344 THE NEW ENGLAND JOURNAL OF MEDICINE Dec' 16, 1965 Table 1, Clinical Data on Cast* Sluditd in Dalail. Case No. Sex iM 2M 3M 4M 6M 7M fi U 9M 10 M 12 M 13 U 17 u 18 M 19 M 21 F 22 U 23 -M 24 U 25 U 30 M 31 M 32 M 33 38 M 37 M 5M 11 F 14 u 16 26 F 28 F 34 p F 15 U 20 M. 29 U Ace ST. 63 42 70 55 57 60 41 76 56 52 48 66 51 56 75 52 55 76 56 56 42 70 57 40 55 72 IS 67 67 so 64 56 49 78 74 66 $MOKOtO Hiatohy TAOKAOCA UP OtQAQEWEa/ YU, SM0XUJ WQMAMOlCfNo ittmVAi mVtj*T. DOATM 54 1% 1-2 i'A > 54 2-9 1-2 /.-l 2 1% 1/4-2 i-154 1 2 1/4-2 1-2 154 1 >2 114 1-2 154-2 54 114 >2 >30 25 >20 35 >20 50 20 >20 >30 30 >20 >30 30 25 50 25 25 SO >20 30 20 >50 30 27 >30 10 day* 4 day* 7 days 5 deyi 3 days 10 days 4 day* 2 day* 2 days 4 days 4 hr. 3 dyu 2 day* 3 days 10 day* 2 days 0 6-12 hr. 7 days 1 2 days 7 days 7 days 1 day 1 day 12 hi*. (Kousmakcr) (Nocfiftoicr) (Nonfmotar) (NonmioVcr) (Nowmoktr) (Noranwkcr) (Nornmolttr) (N ojuuiokcv) (Pipe smoker until 2 day* bafor* death) (Stopped cigarettes 18 yr. before death) (Pipe wnokc e until 1 day before death) Treayulnt of IWNM ArTBH AUTOPSY Comment ft Cause o# Deatk Pjjh. )ung* Fresh lung* Fresh lung*' Fresh lung* Fresh lung* Fresh lung* Fresh lung* Freih king* Fresh, lung* Freih. king* Freih lung* Frpila hvng) Freih lung) Fresh lung) Freih lung) Erejh lung) Fresh. king) Fresh, long) Fresh, long) Fresh lung) Freih hmgf FrU lung) Freih Umg) Freih lung) Fresh lung) Fresh, lung* Fresh lung* Freih lung) Fresh long) Freih lung) Freih lung) Fresh lung) Freah lung) Freih, lung) Fresh lung) Freih lung) Smoke Inhalation, pulmonary emboli & edema Alcoholic pntiont; skull fracture & subdural hematoma, Pulmonary emphysema, with coronary heart disease Status aelhmjEluaui Alcoholic patient found dead in bed; chronic pincrtt(i|j. Bronchopneumonia, with, abscess Ruptured enroUd-bifureation. aneurysm Patient dead on arrival; coronary heart disease. Alcoholic pationt; FriedWndcr'i pneumonia. Sub&rachaold hemorrhage Acuta myocardial infarct Cirduomi of right lung Acute myocardial Infarct Acute myocardlei Infarct Carcinoma of lung >-- advanced Cfrrhoth; heart failure, with acute pulmonary edema. Acute myocardial iofqrct Acuta septicemia Subdural hemorrhage Alcoholic patient; acute myocardial infarct. ? Dissecting aneurysm; death during eleatiye aurgery. Carcinoma of lung, with metateaics Sui^icat specimen.* right long after pneumonectomy; ar- cmomi. Surgical specimen: left upper lobe after lobectomy; carci noma. Surgical specimen: left upper lobe after pneumonectomy; carcinoma. Acute pAocrenlitis; coronary hoari disease; pulmonary edema- Loolovoia Ccrtbrovaaculnr accident Cerebrovascular accident; bronchopneumonia. Acute intextiaal ohstruaiton Meiartaiic cardoama from breast Surgical specimen: right middle lobe, after lobectomy; bronchiectasis. Siufcical specimen? left lung after pnonnvonectomy; until' ferentiated carcinoma. Acute myocardial infarct Acute myocardial inforot Acute myocardial infarct DutecLion delayed 2-24 hr. after autopry. fFroien inunedl&toly -- diucctcd Utcr'(je text). tDusmeied Jmcncdiateiy. before dissection or before the lung was frozen. Systematic microscopical and histocheraical examina tions were performed on multiple epithelial sections from 14 lungs; these findings will be described in detail in a future publication. The left lung was used uniformly in these studies, except in a few cases in which it was specifically diseased and the right lung only was available for analysis. With the ex ception of Cases i-13, 17 and 28 analyses were per formed on lungs in which the bronchial tree had not been opened at autopsy. Surgical Specimens Whole lungs or single Jobes were obtained during elective surgery in Cases 32-36. To minimize post mortem changes, surgical specimens were dissected within twenty minutes after removal. The bronchial tree had not been opened, and samples were dis sected for Po810 analysis before routine histologic sections were taken. In 2 of these specimens (Cases 33 and 34) relatively little normal bronchial tissue was available. Dissection of Specimens Peribronchial lymph nodes were removed from the region of the lower-lobe bronchus and, if necessary, the hilar area. Whenever possible, about 1 gm. oi lymphoid tissue was obtained. Samples of lung parenchyma weighing about 5 gm. were dissected from areas as free as possible of bronchi. Paren chymal samples contained no bconchi larger titan 1 mm. in exterior diameter and were taken from the periphery of the lower lobe and, in 16 cases, from the region of the Jower-lobe primary segmental bi furcations. In 10 cases peripheral parenchymal sam ples from both the upper and the lower lobes were obtained. To study in greater detail the gross paren TEW 289 Vol. 273 No. 25 POLONIUM"4 IN SMOKERS--LITTLE ET AL. 13 chymal distribution of Po410, 4 to 7 contiguous specific for polonium isotopes, but, of the natural parenchymal samples Wei's dissected from 4 addition isotopes, only Po213 has a long enough half-life to al lungs, following the course of a bronchus from be counted one or two hours after plating. Radio the perihilar region out to the lung periphery. activity was measured in gas-flow proportional Samples of bronchia] epithelium were obtained counters, with background rates of 0,2 to 0.6 counts whenever possible from the main-stem bronchus, per hour, and efficiencies of 51 per cent for polonium lower-lobe bronchus and lower-lobe segmental bron alpha particles. Counters with backgrounds of 0.2 chi, as well as from primary and secondary seg to 0.3 counts per hour were used for most epithelial mental bifurcations in the upper and Iowa- lobes. samples; these samples were counted for periods of When the bronchial tree had been opened at autopsy twenty-four or forty-eight hours in different count areas were chosen that had been minimally trauma ers for a total elapsed time of three to seven days, tized and where the mucous sheet appeared intact until a minimum of 15 to 20 counts above back grossly. The epithelium, along with overlying mucus, ground had been obtained. For samples with min was gently scraped off with a dull scalpel and imally detectable activity the count rate ringed from analyzed separately from the underlying wall and 35 to 65 per cent above background, and the stan submucosa. Separate determinations were also per dard deviation of a particular measurement was formed on samples of superficial mucus obtained by roughly 50 per cent When the total net count was careful lifting off of areas of the raucous sheet with less than 15, and the count rate less than 35 per curved forceps, Samples of bronchial epithelium were cent above background, the activity of the sample obtained from areas averaging 1 or 2 square centi was considered not significantly different from zero. meters except in bifurcations, where they were as Determination of background count rates, as well as small as 0.1 to 0.2 square centimeter. In very small the counting of reagent blanks, was performed at samples epithelial tissue was usually dissected from regular intervals; these showed remarkably little vari the underlying bronchial wall and cartilage with ation over long periods. iris scissors, and the mucosa and submucosa were analyzed together, Since epithelial samples contained varying amounts Lung Parenchyma Results of superficial mucus or submucosa the Poao con Polonium410 concentrations found in lung paren centrations in bronchial epithelium were calcu chyma, peribronchial lymph nodes and bronchial lated from the surface areas of the samples. In each epithelium are listed for each subject in Table 2. lung the concentrations in mucus and in specimens These are expressed as picocuries (10"lJ curies, or of bronchia] wall and submucosa were separately 2.2 disintegrations per minute) of Po410 per gram determined; these concentrations were in general of wet tissue. The average concentration in periph similar to those found in lung parenchyma, but much eral parenchyma of current cigarette smokers was lower than in epithelium. The weight of the epithe 0.0074 picocurie per gram {range, 0.002 to 0.023) lium alone in a given sample was calculated from as compared with 0.0016 (range, 0.001 to 0.002) for its surface area on the basis of an average epithelial nonsmokers. These results are shown graphically in thickness of 40 microns -- a value derived from the Figure l, in which the difference between smokers study of histologic sections in the present series and and nonsmokers is clear. Parenchymal concentrations from the figures reported by Altshuler et aL4 This in the 2 patients currently smoking pipes were sim calculated epithelial weight was subtracted from ilar to those of nonsmokers (0.001 and 0.0015 pico the measured weight of the entire sample, yielding curies per gram), the weight component due to superficial mucus or In 12 cigarette smokers samples of more cen submucosa or both. The Po*10 content of this latter trally located parenchyma were also analyzed (Table component was then calculated from the concentra 2). Po410 content in these samples was greater than tions measured in mucus and submucosa obtained in peripheral samples in 9 of the 12 subjects and elsewhere in the same lung. By subtraction of this less in only l; the average concentration in central value from the total activity measured in the sam specimens was roughly twice that in peripheral sam ple the Po410 content of the epithelium alone was ples from the same group. In 10 smokers' lungs in obtained. which peripheral parenchyma from both upper and lower lobes was analyzed the upper-lobe parenchyma Polonium'14 Analyses showed greater activity in 9 of 10 cases though the The details of the radiochemical analysis and differences were not large (average concentration counting of radioactivity due to Po410 have previ in the upper-lobe peripheral parenchyma was 0.010 ously been described.3 Briefly, tissue samples were picocurie per gram as compared with 0.008 pico digested for an hour in hot concentrated hydrochloric curie in lower-lobe parenchyma from these 10 pa acid, after which the solution was diluted to 100 tients) . In the 4 additional lungs in which contigu nil. with 0.5-N hydrochloric acid and put into plat ous parenchymal specimens were taken along the ing cells maintained at 90 to 95C. This technic is course of a bronchus from the hilar region to the 1346 THE NEW ENGLAND JOURNAL OF MEDICINE Dec. IS, 1355 Tabli 2. Polonium"* Conctnlrationi in Pulmonary Titian. sa No. 1 2 3 4 6 7 a 9 10 12 13 17 18 19 21 22 23 24 25 30 31 32 33 30 37 5 II 14 10 20 20 34 35 15 20 29 CotrtBpraAfION IN PasbnOHYAJA PZtlTPKBHAC OeNTBAl/f picocurtrif l iii, oj rtnl lUltii 0.013 0.013 0.002 0.004 0.039 0.004 0.0(0 0.033 0.097 0.005 0.035 0.003 ' 0.004 0X07 0.003 0.004 0X09 0,002 0.01S 0,007 0.002 0.002 0X06 0.017 0.O2S 0.0015 0.002 -- ' 0.0015 0.002 0.001 0.0015 0.0015 0.00IS 0.0013 0.001 Iticocurigs/ itin. of wel lift H0 0.025 0.005 0.007 -- -- -- -- 0.012 -- 0.0J0 -- 0.019 0.007 -- -- 0.009 0.0025 -- 0.005 o.ooc 0.00(25 -- -- -- 0.001 -- -- -- 0.004 0.0015 -- -- -- -- 0.001 ConchnuuTLON IN Peal- D.ONCUrAL Lvupk Noocs I'icocurict/ got. ol lutl tiuua 0X07 0.007 0X13 0.006 0.0(8 0.010 0.000 0.010 0.014 0.020 0.003 0.013 0.010 0.015 0X00 0.020 0X15 <0.002 0.306 0.010 0.005 0X03 -- 0.034 0.010 0.007 0.017 0X09 ' 0,004 <0.0015 0.0035 -- 0.0023 0.007 0.006 0.002 CONCf,NT*AnON IT* UstfJtfONlAL EbWHBl.MfM* ItAlN-BTBM WtOWOBTVJ plooouricj/ gin. of Wet fiu\u -- 0J <0.2 0.2 -- 0.2 0.6 -- -- -- -- -- <0.2 0.2 0.1 <0.2 0J 1.7 -- -- <0.2 _ -- -- 0.L __ <0.5 .-- <03 0 -2 <0.6 _ <0.6 03 -- 03 H.ONCKua picoeoritt/ tn.'Of val lUJBt 0.7 <0.2 0.1 <0.2 <0.2 0.2 0,1 0.1 0.3 0.3 <0.2 0.2 <(M 0.5 0.3 0.6 0.4 <0.3 <0.2 0.2 <0.2 0.1 -- <1.0 -- <0.2 <0.2 <01 <0.5 <0.2 0.6 -- <03! 0,9 0.3 0.3 suuumi imOMOtflfft pieocuriu/ Sro. / ant lutuc -- <0.3 <0.2 <0.2 0.4 0.2 <0.2 0.2 0.1 <0.2 <0.3 0.1 1.0 <0.5 0.3 1.1 2.6 0.4 <0.2 <0.4 0.2 <0.4 2.5 <1.2 -- <0.3 <0.4 -- <0.4 <1.5 <0.6 <0,6 <0.2 <0.2 <0.5 <0.2 UBrr*v*piaeewittr niPtrROATrON) f>UocurteJ/ gin . of wtl liSiHi 1.0 1-3 0.5 0.6 0.8 <0.0 <4.0 0.8 <5,0 <1.0 3.8 2.0 2.7 4.2 -- 3.0 3.1 <1.0 5.0 2.1 1.4 (.4 7.8 1.5 -- -- <2.0 <0.7 < 1.0 <1.5 <1.5 <1.3 LBFT-LOIVU. KOUiWr M'PUIeCATlClNt piooridij IW, of wtj lUiHi 0.3 <0.8 2.9 0.5 0.G 1.8 <1.5 3.8 2.9 13.9 3.5 2.7 13.0 6.7 4.9 7.5 4.9 <1.0 5.0 -- <2.5 0.7 -- -- 12.8 <0.2 4.5 <0.3 <1.0 1.6 6.8 -- 0.0 7.2 <1.5 6.2 1.5 1J> 2.7 "Cftlculaitjd from rerfsee urea (*ce text). fS'Brctu^rymft. from region oi primary aegrotat*! bifurcedoitaV&hii givdii for ringic bifurcation in fl*ch lobe vrith highlit coHCcntralinn, periphery, the results varied slightly, depending on the lobe studied. In the lower lobes Pono concen trations were greatest centrally, in the region of the lobar bronchi and major segmental bifurcations, and gradually decreased to a minimum in the most peripheral samples. In the right middle lobe (1 case) and the lingular segment of the left upper lobe (3 cases), concentrations reached a maximum near the tertiary segmental bifurcations, in 2 of the 4 cases being lower more centrally (though still higher than in the most peripheral samples). We have attempted to correlate concentrations in the lung parenchyma with smoking history. There was no correlation, with total cigarettes smoked ex pressed as pack years (packages per day times num ber of years smoked), or with the age of the indir vidual at death. As shown in Figure 2, however, a trend toward higher Po" levels is suggested in the persons whose daily cigarette consumption was high er. Correlation with daily cigarette consumption, but not with total cigarettes smoked, is not un expected, considering the rapid clearance time for particulates by the lung, as well as the relatively short half-life of the polonium isotope. Such rela tions are difficult to establish, however, owing to the inaccuracies inherent in the next of kin's estimation of cigarette consumption and the lack of specific data in hospital records. When parenchymal Per10 was studied as a function of time (one to ten days) since cessation of smoking a significant trend toward higher- levels was evident only in those who smoked up until twenty-four hours or less before death or surgery, but the scatter was great. The parenchymal Pos, concentration in the patient who had not smoked cigarettes for sixteen years (Case 20) was very low (0,0015 picocurie per gram). Lymph Nodes The concentrations of Po" found in peribron chial lymph nodes of smokers and nonsmokers arc TEW 291 Vol. 273 No. 25 POLONIUM5* IK SMOKERS -- LITTLE ET AL. 1M7 PoUMmlM1" COWMNtIUtlOK ftUtctiriei/srti. .035 .030' .023- .020 .015- i * .010- ooa- # Ul MM MM V** SMOKCTS NON-SMOttB* PtmPHSRlL UJNfl paiunchtma t SMOKfftg 0K-S*CW*S AtRISaONCKAL ltofh wars Fwurs 1. Po" Concentrations in Peripheral Lung Paren chyma and Peribronchial Lymph Nodes from 25 Patients Currently Smoking Cigarettes and from 8 Nonsmokers. Each point represents aoerage findings in a single patient. No results were available in the lymph nodes from 2 lungs and in the parenchyma from 1 other lung. Poiomim CoNCZNTUTWH picDturter/gin. .025- .020- .0(5- .010- .005- >< BETWEEN 2 OR [ and Z MORE PACKAGES 20AY SMOKED Fionas 2. Po'" Concentrations in Peripheral Lung Paren chyma as a Function of Daily Cigarette Consumption in 25 Patients Currently Smoking Who Smoked until Ten Days or Less before Death or Surgical Intervention. also shown in Figure 1. Again, considerable indi vidual variation was evident. Though the average level in nonsmokers (0,0063 picocurie per gram) was somewhat below that found in those currently smok ing (0.011 picocurie per gram) the difference was far less marked than in. the lung parenchyma. There was no correlation with total cigarettes smoked, or number smdked per day, and little if any relation existed between Poai concentrations in lung paren chyma and peribronchial lymph nodes in the same subject (for example, Gases 12, 22, 25 and 37). Bronchial Epithelium Though the Poil# concentration in bronchial wall was similar to that present in lung parenchyma, it was generally about two orders of magnitude greater in bronchial epithelium than in parenchyma or lymph nodes. The individual- results in epithelium from 5 different sites are shown in Table 2. The concen trations indicated for bifurcations of the upper and lower lobes, where several similar specimens were frequently analyzed in each lung, were those found in the single bifurcation in each region containing the greatest measured activity. Since parenchymal and lymph-node samples weighed 1 to 5 gm, good ac curacy could be achieved in measurement of Po" activity tit these specimens. Epithelial samples usu ally weighed less than 25 mg. and contained much less total activity. The accuracy of each determina tion was therefore considerably reduced, particularly in small bifurcations, When no activity was ob served, Po" was recorded in Table 2 as being less than the concentration that we should have been able to detect on the basis of the epithelial weight and total counting time of the sample. Measurable Po110 was also present in superficial mucus from all smokeis; the concentrations ranged from 0.002 to 0.044 picocurie per gram of mucus, with the exception of Case 36, in which the concen tration was much higher (0.28 picocurie per gram). The latter result may have been due to an error in analysis. The results in mucus are probably the least representative of equilibrium conditions during life, owing to the rapid clearance characteristics of superficial mucus and the abnormally large quanti ties of mucus undoubtedly present in many of these patients at death. When histologic sections were obtained at the time of dissection in the lungs in which dissection was delayed for two to twenty-four hours after autopsy they showed varying degrees of post-mor tem changes, especially .loss of areas of bronchial epithelium. We shall consider in detail, therefore, the results from the 14 smokers' lungs that were analyzed or frozen immediately. In these cases his tologic sections revealed minimal post-mortem changes and generally intact bronchial epithelium. The anatomic locations of epithelial samples, as well as the averages and ranges of Poai<1 concen trations found in epithelium from this group of 14 smokers, are shown graphically in Figure 3. With the exception of Gases 33 and 36 lobar and segmental 1348 THE NEW ENGLAND JOURNAL OP MEDICINE Dec. 16, |9g5 A. MAIN-STEM BRONCHUS IT8. LOBAR BRONCHJS C. BASAL SEGMENTAL BRONCHUS segmental bifurcation UPPER LOBE E. SEGWENTAL bifurcation LOWER LOBE (<1.0 -13.0) 6- i 5' ry S i i 0 ABODE Figure 3. Average Po'K Concent-rations in Bronchial Epi thelium from Various Regions of the Bronchial Tree of It Cigarette Smokers. The figures in parentheses are ranges. For the purpose of these averages, samples in which no activity could be mea sured (see Table 2) were considered as having xero activityIn bifurcations the averages are of concentrations found in the single bifurcation in each lobe with the highest measured activity. bronchi analyzed were all from the lower lobes. Po814 activity was observed in epithelium from 5 of the 8 main-stem and 7 of the 12 lobar bronchi ana lyzed. With 1 exception the concentrations in these regions ranged from 0.1 to 0.6 picocurie per gram of epithelium. Activity was found in segmental bronchi from 8 of 13 lungs; epithelial concentrations were 1.0 picocurie per grain or greater in 4 of these bronchi, hut were similar to those in lobar and main-stem bronchi in the other 4, The levels were considerably higher, however, in segmental bifurcations, Po814 was -found in epithelium from upper-lobe bifurcations in 11 of the 12 lungs and in lower-lobe bifurcations in 9 of the 11 lungs in which they were analyzed. In 9 lungs in which bifurcations from both upper and lower lobes were analyzed 7 showed activity in both regions, In all 14 lungs the average Po114 concen tration in segmental bifurcations from both lobes was 4.5 picocuries per gram. Fo814 in epithelium from individual bifurcations, however, exceeded 10 picocuries per gram in 2 lungs, and ranged from 4.5 to 10 picocuries per gram in 7 of the remaining 12 specimens. It is evident from Table 2 that considerable vari ation in epithelial Po814 concentration was present among smokers. Irr addition, however, considerable variation was found among different bifurcations in the same lung and lobe. In the II additional lungs from cigarette smokers in which delayed dissections were performed (Cases I to 13) Po3*4 activity was found in segmental bi furcations from 10 of the 11 specimens, but the concentrations were generally lower than in the lungs that were frozen or dissected immediately after au topsy. We believe that even in the latter cases polonium may have been lost from local "hot spots" because of the delay between death and autopsy. No con-elation existed between number of ciga rettes smoked, or time since the last cigarette, and the Po810 levels in bronchial epithelium. Similarly, there was no correlation between parenchymal and epithelial concentrations. In the 2 pipe smokers epithelial Po*10 content was similar to that of ciga rette smokers though the parenchymal concentra tions were very low. Significant, though low, levels of activity were found-in the lobar bronchus and 1 bifurcation of the single past smoker (Gasc 20). The difference between these results and those in bronchial epithelium from the lungs of the 8 non. smokers is not as clear- as the results in the lung paren chyma. Though in -no nonsmolcer was activity found in segmental bronchi or upper-lobe bifurca tions, and in only 1 case each was there activity in main-stem or lobar bronchi, significant Po110 was found in epithelium from lower-lobe bifurcations in 4 of the 8 nonsmokers. In 2 of these the concen trations were similar to those found in many smok ers, but in the remaining 2, only low Po310 levels were observed in I of 3 bifurcations analyzed in each case. Obviously, cigarette smoking may not ba the only environmental source of Po514 for bron chial epithelium. DiSCUSSroN It is evident that human lung tissues contain measurable amounts of the radioactive isotope polo nium414. In general, this element may enter the body in two ways. In the first, "unsupported" Po414, or polonium -not present with its long-lived parent, lead110, may be taken into the body directly. Be cause the half-life of Po*10 is only a hundred ami thirty-eight days, exposure to the isotope -would have to be fairly continuous for a steady-state concentra tion to be reached in tissues. On the basis of pre liminary measurements of lead'14 we believe that most of the Po214 in smokers' lungs conies from such | unsupported Po810 present in cigarette smoke. Sec ondly, when lead310 is present, Po114 may arise in the body from a decay of this element* Lead114 is present in many foodstuffs, as well as in the air in low concentrations.-1 Its major route of entry is probably the gastrointestinal tract, from either in gested food or swallowed mucus from the lungs. Wc *Po" la V.t tail >n - (erica of ndtouKvo elamcuu dcm-cU 1IV ' rdlum,M. 1/ PoS|* 11 .present(with it long-lived pAroni* lead"4 (physical half-life, twenty ycirl), It will bo In equilibrium when (lie rate f Po1* decay 1 o Cubic lead equ*b the rate a/ production of new /roro ilia decay of lead114, Polonium"0, thora/orc, may be present lor long perLodi dciplte IU rcUlfvoly short half-liU (a hundred find (lurt nlffbt days! if il b "wpjiorlcd" In this ,vmy by Jonjt-livcd kod**- Fat eyainplc, iho Pc4W In cigarettes depends on the presence of IcAd'11 Iveceute of the long rime between harvesting of tobacco nnd coniouM1- tlon of the cigarette. It ti chiefly the however, that >j volntiltwj* when (bo cigarette burns and Is carried off In lhe tmofcc and lolialw- I TEW 293 Vol. 273 No. 23 POLONIUM1" IN SMOKERS -- LITTLE BT AL. 13+9 believe that the Po416 activity in soft tissues other /than the lung may arise principally from this source; levels in liver, spleen and kidneys, for ex ample, are similar in smokers and nonsmokers.6 Like wise, PoSJO in nonsmokers' lungs may also arise from dead416. Though some pulmonary Po110 comes from normal :environmental sources, such as Ingested or inhaled lead444, the present studies indicate that cigarette smoke is a significant additional source of this isotope in the lungs. fpolonium as a Tracer tor Clearanoe of Smoke Particles ' jn Human Lungs b If we assume that Po440 in cigarette smoke vevmsrins attached to the smoke particles, we may in|fer some of the characteristics of particle deposition <and cleai-ance in human lungs from the pattern of polonium distribution in pulmonary tissues of smok ers. The relatively low Po416 concentration in lung parenchyma of cigarette smokers, for example, sug gests that tire majority of inhaled smoke particles is rapidly cleared from the lung. The average excess Po"6 concentration in lung parenchyma of cig arette smokers as compared to nonsmokers was 0.010 'picocurie per gram, on the basis of our findings in central and peripheral samples. Tills excess wotdd represent about 10 picocuries of Pouo in both lungs :at any given time, or the amount deposited from smoking of about 7 packages of cigarettes.1'4 Turn over- of polonium in the longs is therefore a rela tively rapid process. As discussed below, Po116 clear ance appears to occur primarily by way of the mucous sheet. On the basis of our measurements in mucus from persons with no lung disease who smoked until less than forty-eight hours before death (excluding Case 36), and on the assumption that each lung contains less than 10 gnr. of superficial mucus, the total Poal in the mucus of both lungs at any given time would be less than 0.4 picocuric -- a quantity that would be deposited from the smoking of 6 cigarettes. Though the findings rele vant to superficial mucus in these specimens prob ably do not duplicate in vivo equilibrium conditions our results are consistent with the rapid mucus transit and clearance times calculated for human beings by Albert and Arnett-4 . The distribution of polonium activity in the lung parenchyma of cigarette smokers suggests that either deposition or clearance of smoke is not uniform. Be cause deposition of cigarette smoke depends On dif fusion it should be relatively uniform within the lung. The lower polonium content in peripheral parenchyma, therefore, probably reflects a more rapid clearance of smoke from peripheral lung tissue, into the bronchial tree, than from more central regions. Unexpectedly, the lymphatics apparently do not have an important role in clearance of smoke particles, ot at least in clearance of the isotope itself. This con clusion is based on the observations that peribron chial lymph nodes in smokers contained relatively little Po*14 and that the concentrations in smokers were little different from those found in persons who had never smoked. Finally, the observation that blood of cigarette smokers' contains only about 0.0016 picocurie per gram of Po410 as compared with 0.0007 picocuric per gram in nonsmokers,4 despite a low urinary clearance, indicates that only a very small fraction of inhaled Po81* in cigarette smoke is absorbed into the bloodstream. In summary, then, clearance of the majority of inhaled cigarette-smoke particles appears to be rapid and to occur pri marily by way of the bronchi. The epithelial measurements indicate that Po810 in cigarette smoke can he absorbed in measurable quan tities into the bronchial epithelium. As previously pointed out,1'6 high local concentrations within the epithelium would be by far the most important factor contributing to-a significant radiation dose to bron chial tissues. It is logical that Po210 deposition from the mucous sheet should occur in epithelium from bi furcations, because at these points the flow must split and move laterally to the wall of the single bron chus.16 In the region where the splitting takes place, a dead spot or eddy frequently occurs, and material contained in the mucus may more readily penetrate to the underlying epithelium. It is of interest that Auerbach et al.11 found the highest incidence of premalignant changes in the bronchial epithelium to occur at bifurcations. The wide variation iu Po416 concentrations in segmental bifurcations is not un expected, considering the normal anatomic and phys iologic variability and the varying degrees of pul monary disease and post-mortem changes present in many of these lungs. The observation that Po*16 was also present in some epithelial samples from nonsmokers indicates that cigarette smoke is not the only environmental source of this isotope in bronchial epithelium. In general, however, the levels were much lower in epithelium from nonsmokers than in those currently smoking cigarettes; no activity was found in any epithelial sample from 4 of the 8 nonsmokers. The Po110 activity measured in bronchial epithelium of nonsmokers may arise from lead416, and specific physiologic factors relevant to each subject may de termine the accumulation of localised concentrations of Pou0 or lead410 from sources other than cigarette smoke. 'Radiation Dose from Po,u In Bronchial Epithelium Although the epithelial concentrations measured may not represent equilibrium conditions during life, we may use them as a base line in calculating the radiation dose from Po410 to local areas of bron chial epithelium. Owing to post-mortem changes, primarily some loss of epithelium, one would ex 1350 THE NEW ENGLAND JOURNAL OF MEDICINE Dec. 16, I9gj pect that any errors in such a calculation would lead to a calculated dose lower than that actually re ceived. If an equilibrium concentration of 10 picocuries per gram is distributed in a volume of branchial epithelium (levels of 7.5-14 picocuries per gram were measured in bifurcations from 7 subjects in this series), and 80 per cent of the energy from the alpha radiation (5.3 Mev., with a range in tissue of about 40 microns) is absorbed in this volume, the total radiation dose to the tissue in twenty-five years will be about 20 rad, or 200 rem, on the basis of a relative biologic-effectiveness factor for alpha particles of 10. (Normal background-radiation ex posure to the bronchial epithelium during this period would be in the order of 5 rem.J) This calculation is based on a homogeneous distribution of Po210 ra dioactivity within the small epithelial volume and must therefore be considered a minimum dose. It is possible that localized "hot spots" were present even within the small tissue samples that we have measured; in this case local radiation doses would be considerably liighcr. It should be pointed out, how ever, that the relative importance of localized in radiation m the production of bronchial cancel', as compared to irradiation of a large volume of tissue, is not yet well understood. Certainly, on the basis of our findings in lung parenchyma and superficial mu cus of smokers, the radiation dose from Po2' to the lung or bronchial epithelium as a whole would be very small as compared to normal background ra diation. The best evidence available tor the importance of alpha-emitting isotopes in the production of bron chial cancer in man comes from recent studies of underground mine workers exposed to moderately elevated radon concentrations in the air.12,13 The latest report of the continuing evaluation of urani um miners in the Colorado Plateau12 has permitted an estimation of the dose-response relation between exposure to alpha radiation and the mortality rate for bronchial cancer, Direct comparison, of mortal ity rates between the general smoking population and these miner's is not possible because the age . distribution and smoking histories are not given. The best comparison is within the miner group itself. The radiation exposures have been given in "work ing-level months." Altshuler et al.* and Jacobi12 have calculated in detail the radiation dose to the' basal-ce]) layer at various sites of the bronchial epithelium arising from inspired radon daughters. Al though their estimates depend on many assumptions they indicate that the highest radiation dose to local ized areas of epithelium would be about 1 rad of alpha-radiation exposure per working-level month. This highest dose would apply only to small areas of the segmental bronchi where the epithelium is thin. From the data of Wagoner et al.5S the incidence of bronchial cancer in the lowest exposure group (leas than 120 working-level months or leas than 120 rad of added exposure) is found to be only half tba( observed in the second exposure group, who received local cumulative exposures of about 240 rad. It clear-, therefore, that relatively small increments in alpha-radiation exposure to a localized region 0( bronchial epithelium can significantly affect the inti, dence of cancer at that site. A direct application of this dose-response relation to the doses arising from Po210 hr the epithelium o( smokers is not possible, however, for three niain reasons: too much uncertainty about the degree of linearity of the dose-response curve at low dosage levels remains; there are differences in the lateni period and types of tumors found in mine workers as compared to smokers, and longer-term follow-up in the miners may reveal a higher incidence of bron. chial cancer; and, finally, in comparing nonsmokers with smokers, we are assessing1 differences not only in exposure to alpha radiation but also to all other components of cigarette smoke. It is unlikely that alpha radiation is the sole factor responsible far- bronchial tumors in smokers. Other agents in cigarette smoke may well contribute signifi cantly as cocarcmogens, and the effect of a small radiation dose may be considerably magnified by the action of these agents. In this connection, the high incidence of bronchial cancel- in Newfoundland fluorspar- miners is of interest.13 We estimate that this incidence has been about 1 per cent per year ior men having a cumulative radiation exposure of ap proximately 1000 working-level months. If we as sume that the age distribution and smoking histories are comparable, the incidence of bronchia! cancer ap peal's to be at least five times as high among the Newfoundland fluorspar miners as among the Col or-ado uranium miners for similar radiation expo sures. Although radon and its daughters in the air are common to both, the dusts in the mines from the two areas are quite different; in the Newfound land mineB considerable fluorspar dust, but no ar senic or uranium compounds, was found in the air. The greater- incidence in these miners than in the Colorado miners suggests that an additional factor or cocarcinogen is present, and the possibility that fluorspar itself is the cocarcinogen is under investi gation.14 Because of the uncertainty associated with dose estimates to bronchial stem cells in both miners and cigarette smokers it is premature to assert that Po"# is or is not likely to be the major factor in induc tion of bronchial cancer in smokers. We should like to emphasize, however, the point that it is not particularly relevant to compare these localized radi ation doses from alpha emitters to external x-ray or gamma-ray exposure to lungs. Since it is highly probable that the relative biologic effectiveness of al pha particles in producing long-term effects on cell populations is very great, at least ten to twenty times greater than external gamma radiation give1' TEW 295 Vol. 2 73 No. 23 POLONIUM1" IN SMOKERS--LITTLE ET AL. 1351 acutely or chronically,111 small radiation doses from alpha particles will be considerably more effective than a comparison with the sparsely ionizing elec tromagnetic radiations would suggest Furthermore, it is unlikely that there is a threshold dose below which no effect would be produced by alpha radi ation; on this basis any dose, no matter how small, would have a certain probability for tumor induc tion. Finally, recent preliminary studies in animals indicate that alpha radiation may be a much more potent carcinogen in the production of certain skin cancers than more sparsely ionizing radiation (in this case, protons) ,w As for the distribution of Po810 within the lung, the high levels found in segmental bifurcations are in regions where bronchial carcinomas frequently arise. The relatively low concentrations in lung pa renchyma indicate that significant localization does not occur In the alveoli, and, indeed, parenchymal tumors are relatively uncommon in cigarette smok ers. Lung cancer in males is repented to occur about twice as frequently in left-upper-lobe segmental and lobar bronchi as in similar areas of the left lower lobe.18,10 Auerbach et al.,u however, found the inci dence of premalignant changes to be only very slightly higher in the upper-lobe bronchi of cigarette smok ers. These findings do not correlate with the Po'u` levels that we have observed though the difference in PoJ1 concentrations in bifurcations of the upper and lower lobes is not as significant as it appeals because of the wide individual variation. The values given in Table 2 are for the single segmental bifur cation in each lobe with the highest Po" concen tration; in most cases only 1 or 2 upper-lobe bifurca tions were analyzed, whereas levels in 2 or more lower-lobe bifurcations were frequently measured. It has been shown in animals that ionizing radiation may potentiate the effects of directly applied com ponents of cigarette smoke in the experimental pro duction of malignant skin tumors.80 We conclude, on the basis of the available evidence, that radiation from Po!I may be an important factor in the ini tiation of bronchial cancer in cigarette smokers. Summary and Conclusions The alpha-emitting radioactive isotope polonium810 present in cigarette smoke was found in higher con centrations in lung parenchyma, peribronchial lymph nodes and bronchial epithelium of 25 persons cur rently smoking cigarettes than in those of 8 nansmokers. From its distribution within the lung, cer tain characteristics of the pulmonary deposition and clearance of inhaled cigarette smoke may be de termined. The clearance of the majority of smoke particles appears to be rapid and to occur primar ily by way of the bronchi, but absorption of Pon0 into bronchial epithelium does occur. By far the highest local concentrations of PoJL0 were found in bronchial epithelium from segmental bifurcations, and with continual exposure to the isotope, as ex perienced by cigarette smokers, the cumulative local radiation dose to these small regions of bronchial tissue may bp quite high. Od the basis of these re sults, we believe that Po110 may be an important factor in the initiation of bronchial carcinoma in man. We are Indebted to Dr. Richard H. Overholt for provid ing the surgical spedmesu, to Dr. Jonas Hallgrirruson and Dr. Benjamin Caslleraan, of the Massachusetts Oeneral Hos pital, Or. Gustave .1. Darnmin, of tho Peter Bent Brigham Hospital, and Dr. Harold J. While, of the Wwc Roxbury Veterans Administration Hospital, for providing the autopsy specimens, to Mr. Clement Nelson and Mr. Shaun Flaherty for technical assistance with the ntdiochcroicol analyses and to Mr. Edmond J. Baratla, of the Northeastern Radiological Health Laboratory, Winchester, for help with preliminary Po" analyses. References I. Radford, E. P., Jr., snd Hunt, V. R. Polcnium-2l0: volatile redfodomciU la cigarettes. Science 142>247-249, 19G4, 2- Alultfilcr, B., Najson, N-> and Xuichncr, M. Estimation of lacy (Lnua dose from Inhalation of r*dor and douehtrrt- Health Phytici lOi1137-1161/ 1964. 3. Radford, B. P,, Jr., Honl, V. JL, nrvd Shtrry, D. Analytu or teeth nod bones for aIpbn-emitting olomentf. Radiation Rouorch 19:298.315, 1963. 4. Holttmao, IL B. c9Pb (RaD) in inhabitant of Caribbean bland. Health Phyi'm 11:477-480, J9G5. 3, Little, J. 6., Radiord, E. P., Jr.. McCombs, H. L.. Hunt, V. R., And NeJsoit, O. Polonium-210 in long* and soft tissues of cigarette smokers, Radiation Reiedfdh 22:209, 1964. G. Polydorovi, M. Attempt to determine retention of tobacco smoko by means 0/ mcmUrono ftJtprt, In British Occupational Hygiene SocJoty. Inhaled Particlti and Vapom P/acsedingt 0/ an inter- notional lsiuporitim orgatriced hy the Societyt Oxford, 79 Mnr.- I Apr. I960. Edited by C. N. Davgci, 495 pp, London: PtiTjamoo. 190. Pp. 142147. 7. Albert. R. \L, And Arne It, L. C. Clearance cl radioactive dujt from lung. Arek. htdiiti. Health 12:99406, 1955. 6. Little, J. 8., McCandy, R, B., and Rodford, E. P., Jr. Ur- niibliined data. 9. Radford, B. P,, Jr,, Hunt, V. R-, ond Liltlo, J. B. Polonlum-210 in cignreHo jmoken. ScUnu UG;8?, 1964. 10. Hilding. A. 0. Phagaeyiofu, mucous now, and ciliary action. ArtU. Enoironmeniol Health 6:61-73, 1963, if. Auerbacli. O,, Stout, A. P>, Hammond, E. G., end Garfinkd. L. Change* in bronchial apilbcllum In relation to cigarette smoking ajvd in relation to lung cantor, bfno &ng. J. Med. 265:255-267, 1961. 12. Wagoner, J. K., Archer, V, B,, Lundin, F. E., HoJfldey, 33. A., nod Lloyd. J, W. Radiation at cause of lung cancer Among uranium minora. 7Uw En*. J. Med. 273:101-108, 1966. 13. do VlUlors, A. T., and Windulv, J. P. Lung cancer in fluorspar mining community. L Radiation, dust, and mortality txpe/iencs. Brit. J. ludiut. Med. 21:94409, 1964. 14. Jacobi) W. Dora to human respiratory LracL by inhalation of short'-bved ,nRn- and *wRn-dccfty producls. Health Phyticr 10: 1163-1175, 1964. 15. do Vi LIi en. A. Personal communication, 16. Roport pi RBt commlttco to l^loruAtloAAl Comral|om on Radiological Protection and on Radiological Units and Measure- nonti. Health Phytict 9:357-304, 1963. 17. Llpnlncolt, S. W,. Jctjnph, J, B.. Baker, C.' P., and Poudtcc, J. JH. S. Cofopartion of efteots 0/ 10 Mcv profoni and 40 Mcv alpha pftrifctex on Induction of malignant neoplasms In mouse jkua. rroo. Am. A. Cartoer R etcarch. 1965. Id- Hormau, > D. L.( nod Odttendeo, M, DUiribution ol primary lung carcinoniAf m relation to tlmn ai determined by hutoenemicnl icdfvrUcjuca. ). Hat. Cancer 1ml. 271J227-1271, 1961. 19. Ocisanor, A-, Ofllivner, A., Jr., H^Doubler, 0., nnd Blalock, J. Bronchogontu carckxoma. 2>/;. of Chttl 3711-12, 19C0. 20. Bock, F. G.. and Moore, O. E. Carcinogenic activity of cigarette- smoke condensate- 1. Effect 0/ trauma, and remote v-lvradleiion. J. Hal. Cancer Inti. 22i40Mll, 1959.