Tobacco Documents Online

.: >•- # E.C. / ~.._.. (Diseases of the Chest) For P.M. (CDT). Release ?une 24, 1970 IN r~ t'~,r F'r

t 3;ffectB of Cigmrette S:aol:ing Upon. Doga `~~:::r .ri nent; Moxtali.ty; 1'indinrgn in Lung Parenchyma. by. E. Cuyler Sircr;nond, Sc.D., Dr. Med. (=scar Auerbach, M. D. David Kirwan, B. S., and Lmixcnce Garfinl;el, M. A. In ntimerous epidemiologic and histologic studies, it has been found . , . , that the ill of cigarette smoking increrse Vith the nursber of ciga-- . 1-6) rettes smoked per" day. 1lia points to a do, e-roopons.e relation5h3p. The dose of tar 2nd nicotine dewive;.ed in the ccncLp fr= a cig3rette can be reduced in any of several ways, one way being the use of a filter. Thus it _pight be suppoae.d that if tar and nicotine are =inly responsible for the i12 effects, then scro~~;ing cigarettes equippcd with an efficient filter may be less harmful than sWoRing the aame nt=ber of identical cigprettes not so equippcd. i~]`v j.Ts:sC:..''. .^.Xj':.Timer,t V'm-. conducted to test- tt1'-F !1'yPuLiico~: _...F .,., .i c i- t• i`t;1 i ii ~. vnether cigarette smoking i=ould evcutually lead: to lung cancer in dogs. ,In a-prelivin,~ experirment, ten beagle dogs smoked ci:garettes .: ..:.=-t_~ ...: . - 7,8) daily by means of a tube in: crted i_n a tracY,eostcmn. Five of the dogs died within the fir:rt 415 daya of vnol:ing, one deQth being duie to br.onchopncu.r:onfa and the other four to pulmonary ~nfa.rction duc to eraboli. In three of thc.^,e cases the enbol3: apparently ca.^e from a thrc :bus in the right auri_cular appen- dage. xhe remaining ff_%,e dogs were sacrifi.ced on dayn 421 to 423 -sftei- having . : . . ' swokeeet 3,700 to 4,100 non-filt:er cigarettes of a br,-.jid x-ehich Was reported to be high in tar and nicotine. At autopsy, all of the cacrifiGed dr,g, were ,. ' 100509i'761

2. focuid to have cmphyser~a togetY~er r~ith heurt cnlFZ~ge3ent; ~^.nd' it in doubtful whether they would have survived for p:ny more moiiths. They also had raoder- ately sdvanced histologic changes in the bronchial epithcli=; and three of them had lesions (dyskeratosis) resenbling cwrcinc-ma-in-situ. In addition, all five dogs showed hy~.~.^..rpla'sia ualls of the bronchial tubes.. beings. However, at relatively higr, d.aily duFage (up to 12 non-filter cigarettes with distended goblet ce;.ls a glands in the The preliminary experiment indicated that beagle dogs are suitable experimental animals for teating various t,-pcs of cigarettes in respect to ~ < •. . -. ~.: - - • • . severaT of the effects Vnich cigarette smol:ing is knout; to produce in htm;an a day), most of the animals are likely to die too soon for the investigation of Iong-tern effects. Therefore, in the pre; ent experi;mcnt, . we started with a 1ov daily dosage and very gradually increased it up'to 9-cigarettes a day. In. the Ili~'heSL exposure group in l:Tli~~' f:oiiiu wou ...•+ reached until the 858th day of cc,o kinp. atream smoke vhcn.a cigarette is uicchznicull.y ;,Taoked under stundard test Cig:1rctt-eg i ested Our desire was to deterr-aine the biologicul effectivenesa of a filter which re:soves exactly half of the tar e-nd: half of the nicotine from the main conditions. Several brands of filr.cr-typt: cige.rettc:.> were tc:•ted For us by Dr. Jacob Fitelson (a cl3emist with ].ong eYpcrience in this ficeld); but none .of the-,n precisely met the desired spe-cifi~~~a'ti.o:ls. The nearest was a popular brand of American cigarettes cold in cruch-rroof pac1:F9'e3. Each cigarette was 80 =: long with a 17 =. t:hit•e filtcr:.. - In testi'tig and later t:hcn givcn to dogs, each cigarette was sVLokc.'d until the first. 49 i; n. of the cigarette

filter attached, the t:,mol:e contained an aver4t,;e of 17.8 mg. of total tar (dry was consumed, (i.e. until the glowing tip reached the 49 n~. mark) . With the '3. wei:&ht) and 1.17 mg. of nicotine. With the filter removed, the smoke contained an_average of 34.8 mg. of tar and 1.85 mg. of nicotine. ';r-us ehe. filter - xeduced the tar by 497. but reduced the nicotine by only 37°io. We purchased 480,000 of these cigaretteswhich ca.me pc.cked in Heil-sealed boxes, each box containing 60 cartons with 10 pacl:ageg of cigarettes per carton. The boxes = were stored_ir a temperature-controlled, humidified room until used. P.xperi*^ental l~nitn~ ?. s -2linety-3even pedigreed male beagles Kere purchased from breeders who stated that the dogs were in good health. They were kept under.observa- tion in our Iaboratory, for several wceks to make sure that they were in good _ health. - TYacheostony was then performed on each dog. The tracheostoma was . - - . ~ . • . .,kept open by a hollo~.~ tube made of Teflon. In smoking, the tube was c`iangc:a Teflon tube equipped with a. socket -for coupling to tubing leading to a to a z. - . . _ . . cigarett.e holder. - . At the start of the experiment, 'the dogs ranged in age from 1.7 ?_0 cages. They were fed each day following thF: afternoon smoking set;rion, experiment -they were housed in five air ctnditio:led va.rds, each ward 1taving .from 20 pounds to 39 pounds, the mean being 28.3 pourtid3. Tnroughout the years to 3.3 -;eara, their mean age being 2.7 years. . They ranged in veight the ~e~l consisting of 400 grLms of cc.rr:7crci<:l dog food. The ni.nety-aeven dogs were a.ss)Vried to crzF;cs by compater.-bencratcd'

4. kept as controls. There was one or more of these dogs in each of the five randcxn numbers_. Eight of the doga were then sel.ected by rartdom ntic:bera and .I : wards. The ci€,ht control dogs, hereafter r?ferrcd to as. group C, never sr~oi:ed cigarettes; but their tracheostomas were I:ept open or the duration of i.he - experiment. During seven c,onaecutive days the other.89 dogs (i.e. all except -the eight group C control dogs), were given unlit cigarettes in order to let them become £artiLinr with the 'situation, We smoking of lit cigarettes began on May 2, 1967 %4hich we will hereaf ter cal:, "day -4.1". The first 56 conse- .period. On days 1-1411 each~ dog smoked one filter-tip cigarette in the morninfi; cutive days were used to train the doL-,s to smoke and to habituate them to tobacco smoke. Only filter-tip cigarettes were administered during this 5'wvr.ov days 22-35 each smoked lll- in the afternoon; on one in the morning and two in the afternoon; and on days 36-56 each smoked two in the morning and two in the afternoon. 'Plius each dog (except three that were ].ost) smoked a total of 154 cigarettes during the first 56 days. Each dog Wmoked cigarettes in the morning and in the afternoon of each day, seven days av'c.ek, from then on. At the beginning, sraol:e from the cigarettes was delivercd in puffs under forced draft. Gradually, the dogs became used; to smoking cnd "inhaled . the smol:e voluntarily by drrving on the cigarette. In between each five puffs (or more often when necessary to prevent anolia) the smoking tube was clampeci off allo:ring the dog to breathe fresh «ir through its motsth or nostr.i~:s. . ~ ". .

Whenever a ciog -showed signs of" d~tr~.` .. ,~z longer. period of air inhalal ion was:permitted. After a while the dc4-.:=appeared to enjoy, the experience as indicate& by tail-wagging and jumping into thi. sinoking box voluntarily. Forccd inhalation was discontinued and the dogs smoked voluntarily from t~en Three of the dogs were lo'~: -"z3'uring the 56 day training period: .Stzortly after smoking was`, 'started, the' eyes of the dogs were ex- . amined by Ur. Ha :old Spalter (Opthalmologist) and he discovered that one of them had a brain tumor. This dog ua5`°'withdrzwn frozr, the study and will not . . . sr..; : be mentioned hereafter. One dog died on the 32nd day of smoking after having smoked 61 filter-tip cigarettes. Autopsy revealed the cause of death to -he pulmonary infarction with multiple emboi i in Lite iukLr;ZY hrcint-}r,n- pneumonia on the 52nd d'ay after havinig smoked 13'4 filter-tiP cigarettes. On day f57, the rcmaining 86 smoking dogs were divided into four groups (F, L, li and h) . As described below, 12 dogs were put in group F, [2 cigarettes for the duration of the experiment. Group L, H and h dogs were group L, 24 in group It and 38 in group h. Group F dogs smoke& f i l t er-• t i p gradually transferred to non-fiLter cigarettes (i.e. the sz,:ne brand of citi;a- . .A Since tl,ere was reason to suspect that effective dosage (ir, contrast * to ahso2ute, do:.ai;c) 'in influenced by the site of animals, we attempted to obtain dogs all of close to the same si'-.: Unfortunatel.y, due to shortage of suphly, - . . _ ,. . _ . . _ _>.;

c~ 6. ti -the dogs delivcrcd to us va.ried in weiEht from 20 to 39 pounds. For this reason, we net. ut;ide the 38 heaviest dogs vhich will hercafter be referred to as group h. These dogs, which ranged in weight from slightly over 29 pounds to 39 po.unds,- were used in a sapRrr.te experiment, the prcliminary findings of which are reported . . . .. the averar;e being 25.0 pounds. By stratifiecl randoni selection they were . into gt'oups F, L and H(F for filter•-tip, L for ='lighter smol:er." , .. non-filter cigarettes and H for "heavier smo~cer" of each group in remaining 48 dog& ranged in weight from 20 pounds to 29 pounds, each of the five frurds. On any given non-filter cigarettes). t<ou 1 d be some dogs of day., the same technicians hzndled the sr:~oking of all the dogs in a-particular ward. 057 throus,h L140 may be referred to as a transitiom period. Group F dogs were kept on filter-tip cigarettes Vrile group L, 11 and-h dogs were gradually chanSed fro.~, filter-tip to non-filter cigarettes. On days #141- throu8h #210, group F dogs snloi;ed' 6 filter-tip ci8arettes, group L dogs smvl:ed 3 non-filter cigarettes, and group IT and Ii dogs smol.ed 6 non-filter The daily nwnber of cigarettes 'Has gradually increased there- We ai.r~ied to expose the group L dogs to half as much tar and nicotine after up to 9 cigarettes per day for group p,.H and Ii dogs. As the group 11 and h~ dogs during the courpc of the experiment. The'ref-ore, in order to co:spen:sate for the fact t:lrrt all the dogs stnoS:cd the sanie ntm,bet- af cigarettes during the first 56 days, ths group 1. dogs clcre 8iven fe~r,:r ::.than half as many cigarettes as the'H and li dogs during, the remainder of the .

7. e>:peri.ment: The surviving g,roup L dot;s smoked five cigarettes per day :_dura.nS 49 day:5,but smoked less than this nunber on all other days. During ,the last year before being sacrificed, they smoke.d' just four cigarettes a It should be noted that during the course of the experiment three . of the dogs became acutely ill (fever and other symptorrs) . One was of group- H and 'two were of group h. They were taken off of smoking for a, few days, treated with .:ntibiotics and when they recc-vered, put back on smoking. of these three dogs surv'_ved for many mon'_hs after the eFisodes of illness. Table 1 provides some information on each of the four groups of :smoking dogs and the control group. For ease of comparing groups, data are given on the total exposure during 875 days for those dogs that survived for. t'•tliti> lon6 v- iZ~ .. LC 1 ~..r-..7 f~ar anrl rnYal nico:.lne consmried are based upon the assumption that the dogs received the sanie quantity from each cigarette as was found when the cigarettes were smokcd by a smoking machine. -. As previously described, dogs of group.: L, 11 and h smoked filter-tip cigarettes at the start of the exp:.riment but smoked non-filter tip cigarettes during most of the experiment. This was taken into account in computing total tar and nicotine consumed in 875 dayn. . convenient measure of size. For this reason, table 1 shows: the grams of tar The size of the animal should be•consi:dered in estimating effective dosage; and body •eight of the animalt; at the start of the experiment is a and nicotinis consumed in 875 days divided by the mean starting weight of each group of anicials. By this index., group h had lower effective exposure to tar t _-.. , a;._ . .. . . . . .

I .and nicotine than group H. labeled Group F had 1.06 ti-mes as much exposure to :nicotine as group 1~. - Sacrifice of Dogs - - I- - _ ~ . - _ ---- for-a 150 lb. man" is base& ~upon the constant,'relative dosage nd H than man} Lhi"s experiment smoked no more cigarettes per day on the average t The lan was to sacrifice all of the remaining group F, L, a plan figures are_intended.only ' v .dl7pJ 4./ actical reasons sacxificin~; ~*as deferred for 16 days during which' but for pL ~ term ef fects) lialf of the group H dogs c'ere dead by the end of day V85v , s had di i,~1 f of the do~;s ini any onc of these four groupe l 7 •. t.~:~'"O}{C)~f'(~ VPYV lCnZ (Dogs of group h were to continue smo~:iub ,.~c orU4~ -. ~• - ' 1005091768 Thee a;icrificing of all remaining group F, L and I~ dogs was started on on day PS7Cand completed on day ~jS4'~• 7.1ie set of figures The exposure of sroup, L was half that of group H.. "equivalc ,t ntm=ber of ci,;arettes per asstr.n'ptio:~ that, holdins absolute dosage is inversely proportional to body waight. These to show that, in relation to size, the dogs in Sacrifi.cinp, was conducted in This length of *time Was requi.:'ed to zr out thoroughr post-rnortc,n cr.~ir.ations. stratifi:ed random or.der, of the . time. the 4tratif j-caLi.oit being such that the doSs of e<~c i three group; would be kept a'li\Ie for al=pro>timw.tely the same lcn,n,ttl^. of Until s.:crificed, each, F dof'l smolccd, 9 fil:tcr-tip ci.garc.ttes a d,ay, each L dog smoked 4 non-filter cigt;rettes, a day and eacl~r H' dog, smoked S Group C doga were sacrificed between days to tar and 1.29 times as much exposure 4

One cf the F dobs c' ~d on day.V879 which was prior to the day on -uhich he otherwise would have been sacrifi.ce6. The principal cause of death : in this case was pulmonary- infarction; and thrombi were found in veins of For simplic.`-t~I -n presentation, we will hereafter include - - ,. - . . . _ :. .,. ~_ ..- . : . . . : this dog with those• that werc-sz;crificcd. , Hons of ~the remaining group h dogs .of the-n died b:_ween days #876 a..d 2,,=906. All of them continued to smoke r -non-filter cigarettes. Later findings in this group will be reported after Hpne of the 8 control dogs died; two (16.77) of the 12 group F dogs died; two (1b.7%) =of the 12 group L dogs died; 12 (50i) of the 24 group 11 dogs died; and 12 (31.67o) of the 38 group It -dogs died. Thus the death rates of the five groups of dogs were in aligrwient with the dosage o; tar and nicotine relative to their rnean body weight at start of the experir_-lent.. (See table 1) . Table 2 shows the principal cause of death . of each dog that died, they have all died. F,-...,, riay f57 (the CIay vP(-n rb` smoking doos were divided into four groups) th-rough day i875. dog of group It which died on day P237. : TAble 3 shovs a stmtnary of the principal causes of death in identify each dog by group and day of deatl7. ' For example, dog 11287 weans the the day of death, and the nimiber of cigarettes smoked. We will hcreaf:ter.

group of dogs as ascertained at autopsy. Some of the dogs apparently died f a combination of causes; and all had at least slight indications of On this table, each dog is clas;ified according to what appRared ' to.be the most severe morphologic r:anirestation of cisease. We cannot rule out the possibility that the acute toxic effects of cigarette smoke may have precipitated the death of some of the dog's including some of those with morphologic manifestations of disease-ufiich appeared to be sufficient to account for -de..th., Furtheimore, we cannot ruLe out the possibility that the effects of car'',on monoxide in cigarette smc'cc may have contributed to the death of some of the dogs. Two deaths (H517 and 11563) were due principally to pulmonary fibro- sis and e,.physema, and eight were due to cor pulmonale. - All ten of these -f 14 ..i 1, i'F,n firot nrriirrnfj ?ft-or smoking. In those that died of cor pulmonale, great enlargement of the right -heart was a more pro;~inent feature than was fibrosis and enp5ysema. the degree of pulmonary Nine of the 28 .:eaths were due to pulmonary ;_:-ifarction. A thrombus was found in tha right auricular appendage of most of these dogs and sc:ve_al also had a thrcmbus in the left auricular appendage. One had a thrombus in -the vena cava and several had thrombi in other veins. absceas probat ly contributed to death. AL 1 four had moderately advanc :d Four deaths (11198, 11203, 11259 and 1~I287) were due principally to .bronchopneumonia. In one of these pulr:lonn;.y i.nfa.r.ction and in one pul.r;ionary

11. ->Ar=autopsy, food particles were found in the larynx of one dog ;;_V:3-471 and in both the larynx and trachea of another dog (IF380). The. lutigs ° of dog 7380! were collapsed. One of tliese dogs died sud: en:y while smol:ing a cigarette and the other died shortly after smoking. Asphyxiation resulting from aspiration of food appeared to be the major cause of death; but reduction : =Vf:_~Ugrsi 4upply or acute toxic effects may have been a contributing cause. The principal cause of death of three of the dogs is.uncertai~.n . . . . . .-:. . . : . . . . . ; . . : . : . - (H135, h289 and 11858). Dog H135 had a sligf,t to modezate r:i£fuse fibrc3sis emphysema in all lobes of the lungs. Dog h289 died shortly after the morning smoking session. lie had pulmor_ary fibrosis and emphysema in all lobes of the lungs and organized bronchopneumonia in the left diaphragmatic but these morphologic cha::ges did not appear to be sufficient to cause death. Dog 118:i8 died~ suddenly after smokiag his fifth cigarette in the _°:----mvz n in g s e s s i on . He had pulmonary fibrosis and emphysema in all lobes of the lung of a degree probably insufficient to account for death; and alarge number -..- of granulomata with brown pig:nent and fat with reaction around these areas in Jobes of the lungs. Since morphologic -findings at autopsy did not. appear to be sufficient to cause death of these three dogs, it may be that death was ;.aiiu*,-:principally to acute toxic- effects. ~40SQ9~:"~`71 : As shown in table 2, non-iiivasive. bronchiolo-alveolar tumors were' found in the lungs of 12 of the 28 dogs which died and invasive bronchi:olo- ` alveolar tumor. were found in the lungs of four of them (dogs h626, hG49, 9

Food Intake and F!eizht In F;eneral, the appetite of the smoking dogs as well as the ;-on- scrmewhat in ,eight from one weighing to'thz next. A few lost weight and ' The dogs were weighed periodically. Most of the dogs -flucttiated =•regained_it and a few ga'-ned and later Lo.-t weight. ?':owever, the different xhat is to say, they usually ate all of the 400 grams of food provided each day, although occasionally some of the dogs .failed to do so. .&roups of dogs did not vary significantly in_this respect. Two dogs of group F and two group L dogs died before day #876. Their -weigfits shortly before. death were approxiyaately the same as at start. a. r. ~. 11 a' -~r L..a .7....4..T.tr FT..,.n.. avi ic bzvuy ii-uub~ lllGll. U'iCa~ VcAVLa:- ..ca~ .~.1,., . .~ .- . o~- -- --c~ -, --- 10050991772 Findi.n ,s' in T6unR Pai1cncl«^na and Pleura Autopsy was performed and the lungs removed as soon as possible had lost weight and seven were at close to starting weight. Of 12 group h.- - dogs %fnich died, 8 weighed less just before death _ than at start; one had '.gained weight and three had'the same weight as at start. Of 40 dogs WhUc;I were sacrificed after day "875, 21 weighed more than at start, 17 weighed about the same as at start and two (one a group L:and. one a group 11 dog) t•.*ei~:&hed less than at start. into the tr.ac_,ea after which time was allowed for'adequ<<te fixation of the after death. Tte lungs were inflated by instillation of gxavity-fed fornialin tissue. The lungs of beagle dogs have seven lobes, four on the right side

13. ..apical, cardiac, intermediate and diap:hragmatic) and three on the left side (i.e. apical, cardiac and diaphragnatic). Olie specimen of tissue was cut froas-each of the seven lobes extending from the pleurQ irn4ard about two i r re arcd from each of these seven p p centimeters. ~ microscopic sect on as Altogether, there were 490 microscopic sections, seven from each of the-70~dogs th~:t died or were sacrificed (including the two dogs that died before ddy #57). The 490 sli.des were put in random oraer by computer- geherated random numbers, and, while in this order. labeled with consecutive the reader nuuibers, O:h.). had no clue as to.tl-he identity of the dog from which a section was On ex,^iination, the degree of each of the following factors '•as estimated• 1) fibrosis, 2) emp,rysema (rupturing or destruction of alveolar ta) 3) pad-like attachments to a].veolar sept.~a, 4) thickness of kalls of sep , - ,rt.n Has used to c arterioles, and 5) thickness of the pleura. k grading s)s record .the de:;ree of each type of histolog ic ch-1rnge. in each instance ,~tGrade 0 " means "no rhange" or "nor.r:~al" . For example, Grade 0 f ibros i_s cr,eans no

- .-z_ . .. . . . indication • of fibrosis. The walls of arterioles are normally thin and the .."normal" thickr.ess of these structures. G;:ade 0 for pad-like attachment:s Co pleura is normally fairly thin. 'Grades above "0" indicate greater than alveolar septa means that no such attach:nents were'found. 1'anphysenia was classified in four grades, 0 through IiI; but ;;rade III was only found in tbree sections (group H dogs). Pad-like attach-nents. Were classified in three grades; thic}oless of walls of arterioles in t:zree The various grades of each type of histologic change mentioned ' four gra3es. It may be that when more mon ths , greater changes will be than any so far observed. If so, higher grades will. be utilized in classifying the changes. ; In.uany instances,• all seven sections from the same dog were. graded the sz,_,ne in respect to a particular type of finding; and~there were no in- :: stances in which the seven sections fro-..c a dog varied •greatly in this respect. For example, there was nc. Instarice in which grade 0 c.:,physema was found i some of the aeven sections and~ grade II e-nphysema in other sections f-:om the same dog. Because of this 10QS09177-4 Sections fro:a~ the two dog: tThi:ch had c;,ioked only filter-tip cigarette and died befcre day V-1'57 showed a elight cegree of fibrosis, a very slight degree of emphysema, no pad-lil:e attacli*.r:enta to alvcolar &epta, normal many con;iEtency, in table8 4 and 5 the findings are shown in terms of all the sections-fro:n specified groups of dogs rather than being sho:an for each individual do?,.

thickness of the walls of ^,r.terioles and no greater thickness of pleura than yas found in uaany sections from the non-smoking dogs. . Sections from dogs F380 and F854 were not much different than those from the two.dogs ment:ioned above. Sections from dogs L347 and L812 were again about the same except that pad-like attachments (grade I and a few grade II) were found in all of shows findings in sections from the 5 group 11 and 6 group It dogs which eied between days f 57 and f-499; the 6 gr,,up H and 6 group h'dobs which~ died between days u500 and P875 and the 12 group H dogs which were between days i7576 and 16899. In both group. H and group h dogs, -there was a progression in the degree of-each of the five types of histologic change recorded. That is, in both H and h dogs, greater changes were found n sections from dogs dying between days ~:5O0 and x875 than in iihose dying between days V'57 and 0499, and the greatest changes were found in sections from~ group H dogs which were sacrificed betveen days #876 and f899. the 12 group H(non-filter) dogs which were sacrificed after day #875. As pre- Table 5 shows che findings in the 8 group C(non-smoking control . dogs),' the 10 group F (f iltcr-tip) dogs, the 10 group L (non-f ilter) dogs and .. . . ~ (non-fi_1t:er) as the group F(fi.lter-tip) dogs. However, the group F dogs had viously described, the group L doi,s stnoke<:t only about half as many cigarettes about 6% more exposure to tar and about 29% more exposure to nicotine than

/0( . .-'~_ . , . . number of cigarettes ac the group F dof;s. Far greater histologic changes were found in sections of lung parenchyma fror the group 11 (non-filter) than in sections fro-•n the group L_ (non-filter) at~d group F(fil.ter-tip) dogs. Except in respect to tile wall.s' of=arterioles, greater changea were found in sections f::olm group L than in sections from group F. Except as noted below, sections from group C (controls) and_89,3% shoued only a slight degree of fibrosis (grade Z). All of the sections from group F, L and if showed at least grade I fibrosis; and most were of higher-grudc. Grade IV or V fibrosis was found in 0% of the sectiona frow , 6LVUt+ v, .. • Of the 56 sections from group C, 10.7% showed no fibrosis (grade 0) S "7 °14~r r.rrs o•-~~~» F. 12.9% from group L and 91.7%" from group H, - (See A comparison may be made by consi-d!ering individual dogs rather than akl :the sections from each group of dogs.' Of the J.& dogs of group I', 8 had no -sections of grade 'IV ^r higher while two dogs each had just two sections of =grade IV (the other Bect"ions being gxttles Ix or XIZ) . Of the 12 dogs of these dogs all seven ce"ctionc were of grade IV or V. In otiier words, every gr_oup•f[; all had a majority of the sections grade IV or V and for six of one of the 12 group It dogs had a greater degree Of f.ibl:osis than any one of the ten dogs of group F. This di.fferencc-. is s,tatistically significant

'diffcrences notcd above, is statistically significant by both tests (p .( 0.05). significant by- both the t test (p ..-~ 0.0001) and the Wilcoxon rank sum test. =Tie• difference between group F and group L, though smaller than the other None of the 56 sections from group C showed any evidence of emphysema. Grad:e:xl:or III emphysema was found in 070 of the sections froa group C; 12.97. from. Proup F. 24.37. from firoup L and 98.87o from Aroup H.. Considering individual dogs,.the :iifferences between group F and group-1I, and between group L and group H are statistically significant (p<0,0001). 7ti'e=difference between group F and group L is not statistically significant the.sections from group C, but nonc of these were above grade I. Grade II ,padtJ:ike attachments were Found in 0% of sections from group C, 22.97o froin group F, 4E.67; from group i: and 97.6°l from group H. Pad-like attachments to alveolUr septa were found in over half of 17. (See table 5 and figure Considering individual dogs, the 2iifferences between group F and 11 and between group L and 11 arc statistically significant. (p < 0.0001). The difference betvecn group 1:' and 1, is statisti,ca,l1y sign ific,~1t (p G 0.05) . . Thiclcness of the Walls of Arteriol.es walls of arterioles were of rormal thickness in all the sections Thickneas of grade I or 11 was found in 0`l, of sections from 2,97; from group L and 41.77. from group H. :

18. table 5 and figure 8). The differences bctwecn group F and li and between Group L and H are statisticai ly significant (p lz, 0.0001). There was no difference between group F and L.. Thickness of rleura . The thicl-,ncss~of the pleura was evaluated as grade 0 in 30.4% and -grade I in 64.57, of the sections from group C. 2'liickness of grade II or III was found in J% of sectio-B from group C, 7 0% from group F, 21.4% froagroup L and 90.57o from-gi oup H. (See table 5- and .Ci gure 9). The differences betueen group F and It and between Group L'and It are statisticall-y sign.ificant (p ( 0.0001). The difference bet4een group F and L is not statistically ~.ignificant (p > 0.05). The mortality of the dogs in relation to ci.barette smoking pnra.llels : the finding in- epi_demiological atudies that death rates are far higher in cigarette nWokers than in non-smol:ers anu increa:se with amount of cigarette 1-4) u~coking. 10Q+.~?Q9177s J?uUnonaxy c~rl~r~yse~a and fibrosis uad cor puln,onale are rare ca:ises of death in dog.:y but of 62 he'avy smol:crs of non-filter cigarettes (groups It and h), 10 (16%) died of these di.ser:scs within 875 days. This paralie2s the finding in epidenaiologica1 studies that ci.garette cmoking is accomp;1nicd by a great inc7:Gasn in dcath ri:tes fro-.n theae disezsea. .I1i..tolcgic fi.ndings in the 1Llnf; pazencl7yna of the smol:ing do,s were , of the srune, typcc as h ia tologic f i:ndings in the lung parench}nna of h"„n z. ,. .._...r,. . . .... ..

19. -1 5) beings with a history of ci e smoking. Both in hu:nan beings and in "dogs, emphysema (rupturing :atzuction of alveolar septa), fibrosis, the occurrence. of pad-like attachments to aLvcol.ar septa,., and thickening of the' wal].s of arterioles seldom occur, or occur to only a slight degree, in the lungs of non-smokers (except non,.-smokcrc with occupational eYposure to certain 'dusts and vapors). Cigaret~ t:~~oking is acco-npanied by these histopathologic' -:: changes; the changes •beccr.n:°;°pi1ogressively more advanced so long as smoking is . smokers. this er.perintent, one group of dogs smoked- filter-tip cigarettes -and other groups of dogs smol:ed the same brand of cigarettes but with the fil:ter removed. The smoke from~ the filter-tip cigarettes contained 49% less lctX iut 'R% l:as PLrOLi1iG FY^-'~ rifiarettes Without the filters. G~- . Considering dogs i'nich smoked the same number of cigarettes per day, _ the death rate wds lower among those" that smoked fiZtex-tip cigarettes than among those 'which smoked non-filter cigarettes. Sruoking filter-tip cigarettes produced less extensive da-mage to the lung parenchyma (i.e. emphysem,a, fi:b1•oGis, etc.) than smoking thc same ntmrber of non-filter cigarettes for the s,we length of time. Ho;rever, smoking filter-tip cigarettes for a period of £:75 to 899 days prod«ced si&nificant damage to the lung parenclx}..~~a. Since the degree of such dUnage progresEes caith duration of smoking, the evidence indicates that lunE; parench}^x,-cl d'ainage -ndvances Ie.is rapidly tAth the smot.ing, of filter-tip cigarettes thzr. with the

20. filter-tip cigarettes for a longer length of time would have eventually resu ;,- i.e. years-in the case of men or months in the case of dobs) but in two . : - . ~ . . . . - . . , -Jactors 3_rieluding duration of smoking, the nirmber of cigarettes sraoked per day and the tar and nicotine content of the *srnoke. Now, let us suppose that the sane amount of tar and nicotine is received in the. same length of time different ways: 1) by smokino N high-tar, i,idh-nicotina cigarettes per day and 2) by smoking twice this ntanber of cigaxettes per day, but each cibarztte delivering just half as much tar and nicotine. In other vords, all exposure factors are the same except for the concentration. of tar and nicotine in the afinka.' Which of these two types of exposure would produce the greater The eegree of exposure to tar and nicotine depends upon sever<<I In planning this experiment we hypothes{ zed that under these con- ditions greater biological damage would be produced by smoke with a high con- centration of tar and nicotine than by smoke with half the concentration of -centrwtion. 'Tie princi.pal phy3iological cffecis of nicotine depend upon the of the body a.nd thereby le~F.d to'cffcct:ively hi.gtler uos.f;e tha.n a lower con - ' our ainls was to test this hypothesis. It was based upon the theory that high concentration would tend to overpower the, defense mechaniumtt tar and nicotine (total dosage being the 3ame over a period of many mcsnt`is or concentratioa of nicotine in the blood at: a moment of tirne. Since nicotine itz removed from the blood very rapidly, a,. hx4h concentration in the b)-ocfd can be

21. .Achicved onl~y_ by rapid delivery of nicotine into the- blood. It would seein that nicot-ine would be delivere& into the blood more rapidly from smoke With .~ . . . . a large amount of nicotine than by the same volum, e of smoke with half that amount of nicotine. Most of the tar and nicotine ih cigarette smoke is con- - tained in particles.. Particles are rcmovcd frcr,n the alveoli by alveolar phagocytes and are reraoved from the bronchial tubes and trachea in bronchial secretions propelled by ciliary action. Presumably, there is a Limit to the rapidity with which the removal can take p? ace even if the removal mechanisms are undamaged. Thus it scems likely that a higher concentration of particles occur on lung surfacds if particulate matter is delivered rapidly (concentrated smoke) than if delivered less rapidly (less concentrated smoke). In 'ac'.ditiorn, cigarette smoke retards ciliary acti_on and, in sufficient concentratiorl, can ((`ortnin of the c5.11astaticagents are in the g2sc4-s_:_ In order to test this hypothesis, we attempted to find filter:-tip cigarettes such that the filter produced a 50% reduction in both tar*r:nd nico- than in the particles contained in cigarette smoke). tine. We could not find Cucch cigarettes. The filters in the cigarettes used effected a 491, reduction in tar but only z. 37% reduction in nicotine. During a period of 875 days, group F dogs which smoked filter-tip cigarettes received 67. more tar and 29% more nzcotine than group I. dogs which smoked non-filter 4 cigarettes. (See table 1). Neverthel.ess, histopathologic chUnges in the luns' parencz)^la c~,c;-e sorae:+hat less extensive in group F dogs (tilter-tip) than in group I. du~s (tlon-f ilter) . 'f'ni.s finding ;s 'consistent with the hypotl esis. .,_ . -Findin~s" in relation to the occurrence of tumors will be descri:bed ~:-in the second of this series of papers.

/a7 Xn -the above di.9cussi_on, we have co:5paxed the filter-tip cigarettes used in this experi,ment with the non-filter cigaret-tca used inthis exPeria,ent. Our non-filter test cigarettes dr.livcred 34.8-mg. of tar and 1.85 . ~ . to _the filter-tip if any. Therefore, some co-nparisons are in order. Cigarettes now on the market vary widely both. as to the "tobacco end" and as "-tn8: of nicotine. In 1955, 'about 8170 of the cigaret_tes sold in the United States 10) did .not have a filter. At that datie, all of the non-filter and mosr of the filter-type cigarettes deliv-~rcd more than 34.'8 mg. of tar and more than 1.85 11) . mg. of nicotine.- delivered 45 mg. . . . - _ ~ _ The most popular cigarette on the American market in 1955 11,12) of tar and 3.3 mg. of nicotine; However, as of October 1969, there was no brand of American 'cigarettes sold in the United States khich~ delivered over 32 mg. of tar and there was no brand which deLivered over 2.5 . . - - - -- --~ -1'~l t'v nicotine, we estimate that less than 170 of the ci~_garettes sold in the United States in 1969 delivered over 1.85 mg. of nicotine while about 14% delivered . 14) mg.- of nicotine. Based upon sales data aau iov 13) over 1.7 mg. of nicotir.e. Our fi.lter-tip test cigarettes delivered 17.8 mg. of tar and 1.17 ms;. of.nicot-ine. In 1969, many brands of filtcr-tip cigarettes deli:vere3 18'mv. or-leas tar and also delivered 1.2 mg. or less nicotine; but these brands . 1 . - ' - . coma:anded only aUout. 171 of the total cigarette market and a`bout 22% of the filter•tip cigarette rnzrket in the United States. Several braiids delivered 12 mg. or less tar and alao 0.8 izg. or less nicotine; but in mi-d-1961) they . conznanded only about 17, of the filter-tip c`.garctte market.

23. I i ared w L but not and nicotinc.cccip uxd be conaidered relatl.vely :.o~~ in tar and nicotine, cigarettcs wo . .. . : • , h civzrettes sold but c=pAred, wi th those eo 1909• our filter-tip tes, . ,.our non-filter tcat cig4retteo ~'ould be conaidcre 1969 they c'ould be consi~,ercd high in tar ld in with cigarettea sol:d in the United Stwtcs in 1955, C t,parcd ow in tar and nicotine; d ~ as lotia as sott~e brnnde nc~~ on the rarket.

/O9 I. References Death rates by cause'. JlJ-kA 166:1159-1172; 1294-1.308, 1958. months of fcrllowup of 187,783 men. Part I. Total mortality. Part II. 1) Id=nond, E. C. and FIo:-n, D. Stiol,ing and dcath~ rates ~- report on forty-foui 2) H'aminond, E. C. Smoking in relation to the death rates of one million men and women. hat. Cancer Inst. Monograph 19:127-204, 1966. 3) Y.Rhn,1I. A. The Dorn study of smoking and mortality a.-nong U. S. veterans. 4) Doll, R. and Hill,-A. B. Mortality in relation to sesokind. Ten year's 0 . fibrosis and thickening of walls of smalj.' arteries and arterioles. New Eng. ' and: age in rc-lztion to Falrzonary ch4nges : rupture of alveolar septums, 5) Auerbach, 0. , Stout, A. P., Ha=ond, E. C. and Garfi.il:el, L. Smolcinp habits J. 1fed, 269:1049-1654, 1963. 6) Auerbach, 0., Stout, A. P., IL..ond, E. C. and Glrfinhel, L. Changes in bronchial epitlieliL^n in relation to cigarette smoking and in relation to Monograph 19: 1-'125 ; 1966. Report on eight and one half years of observation. Zat. Cancer Inst. - observation of British doctors. Brit. I`.Ed. J. 1:1399-1410, 1460-1467, 1964. . lung cancer. i~e:r Eng. J. Med. 265:253-267, 1961. ~ ~ - • r. n ~4J ~,.,.,. ~ ( f KuE'.1 U1t:21', v.', ta<wu~tvuv , ". . , -_------- '-F-Histologic changes in bronchial tubes 2055-2066, 1967. r, , r..,,-f;nS nl . T.. and. St-out, A- P. of cigarette-smoking (ioys. Cat,ee1 20: ) Auerbach, 0., Iiarr~,~ond, E. C., Kirc;an:, D:, and Garfinkel, L. Emphysema pro- duccd in dogs by cigarette scaol.ing,. J1sWiA. 199:241-246, 1967. ) W}-i ider, E. L. and Hoffc,U::n, D. Tobacco and Tobacco Smoke. Studies in Ex- perimental Carcinogenesis, pagcs 250-257, Aca:demic 3'ress, New York, 1967. . Situation, TS-1.03, March 1963; TS-127, ~'~arch 1969; and TS-129, September 1969. - 10) Econcnic R.e.car.cla Service, U. S. Departuent of Agriculture. The Tobacco 11) ConsLnmers Union. Cigarette tests for tars and nicotine. Consumer Reports 26: . 203-207, 3.0,61. 12) t+'ooten, 11. Cii;al=ette sales turn up cgairi in Dece:r.ber, 1955. Federal Tra,lc Co-Nrmi: sion: R:epor.t on comparison of tests of tar and nicotine content of -.i_ga:rettcfs. Dacet nbcr 1969. . .A . _ .. . . . . , _ 14) Maxwell, J. C., Jr'.. Cigarettes 169: The cancer ecare erodes sales - but not 13) very much. Jiarkcti.ng/Ccr.mnunications 297:22-38, November 1969.'

9_7 r r r- r- r r- r- r- r- r r r r- r- . Tabl'c 1. Data on pedigreed male beaE~e dog.s of groups F, L, 11, h and C. \ No No No Filter Filter Filter Filter Control Group F Group L Croup 1( Group h Group C o. of DoPS on Day457* 12'- 12 24 38' - 8 Deaths in Da s -157 th g #875 Y rou h 2 2 12 .12 0 7.' Dead . 16.77. 16.7'G 50.0% 31.67. - o,e at Start (Day 11) Mean Age (Years) 2.7 2.7 2.6 2.7 2.8 Youngest Dog (Years) 2.1 2.3 1.8' 1.7 2.4 Oldest Dog (Years)- 3.1 2.9 3.3 3.4 - 3.3 ai^,ht at :.tart (Day 41) Mcan kei 3ht (lbs.) . : 25.0 25.1 25.01 31.9 30.7 LiEhtestDog (lbs.) 22.0 20.0 20.5 29.3 22.0 lleaviest Dog (lbs.) 28.0 28.3 29.0 39.0 36.3 ^rettes ner Dog in 875 Dags 6,143 3,103 6,129 6,129 None Mean No. of Cigarettes per Day , 7.02 3.54 7.00 7.00 - Equivalent Na. of Cigarettes 42.1 21.2 42.0 32.9 - per Day for 150 lb. Man NO NO NO c rt ~rro~rk - a---- FILTER FILT'F-B: FILTr.R FILTF,R: - - MZ. of Tar per Cigarette 17.8 mg. 34.8 mg. 34!.8mg. 34.8 mg. -- Me. of Nicotine per Cigarette 1.17 ag. 1.85 mg. 1.85 ag. 1.85 mg. - ota1 Dos.a~te in, 875 Days Grams iar per pog, . 109.3 g. 103.5 g. 207.8 g. 207.8 g. - Grans Nicotine per Dog 7.19 g. 5.56 g. 11.12 g. 11.12 g. >osare in 375 Days Relative to StartinL_L•'eiht . Grams Tar/lbs. Weight 4.37 4.12 8.31 6.51, - Grams Nicotine/lbs. k'eight 0.29 0.22 0.44 0.35 - The smoking dogs were divided in :o groups F, L, H and h on Day V57. " Do&r of Froups L, H and h sciol:ed of the experiment, but smoked'no ,ote: Certain of the figures above filter-tip cigarettec during a training period at the start n-fi1'ter ci-garettes thcreafter. apply only to dol;J that aurvivcd 875 days or longer. sooSosa.7ss

C9:7. F_ IT. c r V- V- r F_ r~ rr Tabic 2. Mortality experience of ciFarette snok4nfi,doga and non-c:nking dbga fro.;i dry G57 • • through 1S75 by day, of death nnd cause of death. Dr.y G1 was the c]ay When the dogs started smoking. Filter Tip Group F No Filter Group L No Filter Group H No Filter Group h Control Group C Non- Invas- inv, Day of Death No, of - Cigar- ettes - Deatha (x) Principal cause of death ive T!ucnor (T) iv Tum, (M) 59 160 x Pulmonary infa:ction 86 276 x Pulmonary infarction 135 518 x Uncertain T 198 888 x IIronchopneu*ionin 203 924 X Dronchopneir.nonin 244 1219' x Pulmonary Infarction 259 ,1343' . x Rronchopneumonia & Pul.Inf. T 271 1440' x Pulmonary Infarction 287 1560 x Eronchopneunonia v/Abscecs 289 1596 x Uncertain 347 1055 x Aopiration of foo& T 380 1 2162 x ,Aapiration of food' 416 7:340 x Pulmonary infarciiuu 464 :346 x Pulmonary Iafaretion 517 :3062 x Pu1.Emphycema & fibroeis 563 3404 x Pul.Emphysema & fibrosis T 606 3796 x Cor Pulmonale T 626 3928 i x Cor Pulmonaie T H 649 414-1 x ror Pulmonale T M' 716 4689 x . .. Cor Pulmonale & Pu1.Inf. 153 5030 x t ; Cor Pulmonalle T H 760 5083 x Cor Pulrionale, T c-- 762 5110 x {:or Pulhnoncle 767 5148 x Pulmonary Infarct 6 Cor Piil 1794 15400 Cbr Pulmonale •n 8 12 2847 x Puim.onnry Infarction 'r 81514 5950 X Pulmonary Infarction 9'S8 - 5470 x Uncertnin T -_ - ~-- Nb- Dead 2 2 12 112 No. A1ive on Day 875 LO i0 12 26 8 1005091786 • Total 12 . 12 I I 24 38 8 .. . . . . . - .. .. . .. i .~~ .•~ ~2~ - -zvw-•~.-V- -i.t,-` ~ ..-~. Z i ~.~~.-v./~ .-_•-._ _ ~ ,• G. ~,,.~~~sE ll~w :~.. .:, ry~ ' -• . . .. vw..~..-

Summary of principal'cause of death, days ~57 through fi875 in dogs of groups F, L, 11, h and C. Princi.pal Cause of Death Pu1: Kdiphysema & Fibroeis Cor-P~ilmonale (pu1~. emphysema &:fibrosis with right heart enlargement) Pulmo:iary Infarction Bronchopneumonia aaner,,tion of Food Uncertain Numbcr-of Deaths No. Surviving 875 days Tota;. Dogs Filter , No No I'G iv o Tip 'Filter IF'ilter ,Filter Group IGroup Group roup ,F h Control Group C Total 1 2' 5 3 1 2 2 1 2 12 12 28 10 10 12 26 8 66 12 1!2 2ti 38 8 94 Note:-Some of these dogs apparently died of a combination of causes (e.g. pulmonary emphysema, pulmonary fibrosis, pulmonary infarction and bronchopneunonia).. In this table each death is classified according to what appeared to have been the most severe condition.

//3 I Percentage distribution of oections by grr.de of e°.ch of ive types o histologic ehnnEcs in the lung parcnchyma. Group It and h dogs which died in diffcrent periods of tin:e. - No Ho Filter Filter Group It - Group h Crade of Condition Da~ys Davs Days Days Davs ~57- 4500- Y876- - P57-. P500- 4499 #875 #899 0499 #875 Fibrosis . _ .x x Z %G O: - - - - 7x_ • IL= 23.8 - . . 19.1 4.8 tZI 66.7 47.6 8.3 71.4 47.6 IV; Y . . . 9.5 52.4 91.7 9•5 47.6 T6ta1: _ 100.0 100.0 100,0 100.0 100.0 EsDbysema . - . 0_ " - - 2.4 - I= 76.2 7.1 1.2 61.9 14.3 II; ZII=_ 23.8 92.9 98.8 35.7 85,7 100.0 100.0 10100 100.0 100.0 P--ci-1iYe Ati^ehr.rents 0 I: - 52.4 16.7 2.4 50.0 14.3 II= 47.6 83.3 97.6 50.0 86,7 Tbtal 100.0 100.0 100.0 100.0 100.0 Thtcfness of k'A11s of Arterioles 0. 100.0 -64.3 58.3 97.6 85.7 35.7 35.7 2.4 11.9 II= = - - 6.0 - 2.4 Tota1= 100.0 100.0 100.0 100.0 1100.0 Tnickness of Pleura 0 - 2.4 - 4.8 2.4 76.2 26.2 . 9.5 " 64.3 31.0 . 11; 11 23.8 71.4 90.5 30.9 66.7 Totall - 100.0 " 100.01 100.0 100.0 100.0 2.v :ber of Dogs 6 6 12 6 6 Nuriuer of Scctiors 42 42 84: . 42 - 42

pcrcentage dirtri.bution of r..ectiona by grc~de of ench of I i the luns parcnchy;ns. five typea of hiatologic chnn, c r- c+cri£icec3 after L t~nd li do$B uhich were a P Group C , , . 875 dnyn of mnokin8. . ' . • Cotatrol Filtcr Grade of Condition .Group Tip Group 0 100. 0 - 87e1 74.3 1.2 I ; - _ 12,~~ 24-.3 98:8 '- • . ~ ~ 100.~ ,lptgl 1.4 21.4 64.3 12.9 ~~•se~rs _ 1.4 ppd-li1:e Att~ctr~enta 0 Fm 1ticYr.e;a of r`r" 0 I Yx., Yl:z Tots1. 44.6 - 55.4 77.1 22.9 100.0 30.4 69.6 97.1 2.9 97.1 2.9 2.4 47.6 100.0 58.3 35.7 6.0 2.9 1.4 87.1 77.2 9.5 10.0 21.4 50.5 l_oo,a 10o.0 100.0 10 10 12 ~,_ .. 84 7(~ .. ,: 70 51.4 48.6 . 100.0 //Y

n r) f ..... .y • C ~ ; '~aUP Gi~OP L• UP f" . --~, CRaUP C CP4 '~f~ ROL5 ~.R •-Y lP NO FlLTER ., ~ ~,O ~r ILT -~r~ qs -tAFiY `.. r=. ' 6~ 6~. • b.~r k..r k... ..p ..j , .nw' I G~.: F T~• S),.,.d ~,d ti . ~ , . . ~ ~ . . . . - ~ . . ~ a ~rJ

pE: . R~ I Sv~.~ ~Ir~l~~r~ GRADe:o,~ '~d r~~~ ~'~'` ' ~3y~,,, ~ 1 ~C~~~~ (;smu~ ~ "OlJP L: . U.IKuUp !! ~'~ ' ~ i~ i GOWI R04- M, ,4NY , `„ ,.,,, .,,,ft ,m„ f.d C VIM R

~ 'PEr.C."ENi OF &E-C1 IO~S W8 i H GiR.~~~E lE PAD - LEK"E G PZ CU P L:, G R OUP F''+• BL© FILTER. NO FIL_~ ER (1/2 AS M111%NY. ""r Cfo* R G'Ebr--j.,

10050917i93 ;. ; A b~. ~ OR -~; T-~IC~~i~'~5;~ PE~:~E~'1T OF Si~CT6Oi~S !''$-~ ~~ S : : OF WALLS ()F' AR i -N4aLc-

r'~- ; r-- r--.' , r"~ .~r"n. (4~?.Z 1 ~.1~.~ ~ i ~r.; ~ ,'..,; • p,n,,, ;,,,~ ' n„~' ~ _ ~ • ~ S•d 3f1i .~G_~ ~a~nn,~~ .'"3 S'~~; .~.~"~.1,• ~ , yr.~..; ~ Q ~ • ~ ~~ :~ e'~'~~'~ :~3 ~ a~.~~~ ~.~. 1 tYA'ON : U_--3 ~c~41, -~~~:~.~. i -..