Tobacco Documents Online

CHAPTER 3: NON-NEOPLASTIC BRONCHOPW'LM0I3ARY D'ISEASES'

\.

Contents Page Introduction.. . . . . . . .. . . . . . . .1 . . .1 . . . . . . .65. Smoking and Respiratory Morbidity! . . . . . . . . . . . . . . 66 Smoking, and Air Pollution. . . . . .1 . . . . . . ., . . . . . • 66. Smoking,and'0'ccupational Disease., . . . . . . . . . . . . . .71 Mill Workers - Byssinosis.. . . . . . . . « . . . . . .71 Firemen. . . . . . . . . . .. . . . . . i . . . . . . . . . 7'1 Smoking, and!Pulmonary Function,Tests. . . . . . . . . . . . .71. .? 1-Antitrypsin. . . . . . . .i . . . . . . . . . . . . . . 74 Autopsy and Pathophysiologic Studies . . . . . . . . . . . . .7'6 Autopsy Studies. . . .i . . « . .. . . . . . . . . . . . .76 Pathophysiologic Studies: ih:Humans. . . . . .i . . . . . 80 Pathophysiologic Studies in Animals .. . . . . . . . . .80, Summary of Recent Bronchopulmonary FindingS. . . . . . . . . 83 Bibliogzaphy . . . . . . . . . . . . . .i . . . . . . . ., . . . 84

List of Figures Fi'gure. 1.--Respi'ratory bronchioli'tis in smok,ers and control g'roups. . . . . . . . . . .1 ., . . . . . . . List of Tables Tab~~Ie~ 1.~--Leve~l's~ of sulfur~ dioxi'~de~ (~S~~02)~~ and total suspended particulates (TS~P)~ in four Utah~ communities, 1971, and in five Rocky Mountain communities, 19'70. . . . . . . . . . . . . .I . . . . . Table 2'.--Mean annuall levels of sulfur dioxide (802,) and' totah suspended particulates (TBP):in four areas. . . . . . . . . . . . . . . . . . . . . . . . Table 3'.--Age-adjusted~percentage~of cigarette smokers and nonsmokers in each race-sex group responding positively to exposure to.chemi'cals, fumes, sprays, and' dusts. . . . . . . ., . . . .. .. . . ., . . . . . . Tabile 4. -The T '1.-anti'trypsin levels.and'frequency of pr~otea~s~e~~ inhibitor~ (Pi)~ phenotypes in~ h~ealthy, populations. . ., . . . . ... . . . . . . . . . . . . Table 5--Means of the numerical values.g,iven lung sections at autopsy of male current smokers and non- smokers, standardized for age.. . . . . . . . . . . . . Table 61.--Means of the: numerical values gi'ven, lung sections at autopsy of female current smokers and nonsmokers,, standardized' for age . . . . . . . . . . . . . . . . . Table 7.--Nieans of the numerical values given lung, sections at autopsy of male former cigarette smokers, standardiaed1for age . . . . . . . . . . . . . Page' . 81 . 681 . . 69' . 72 . .75 . 7'7 78' . 79 64

C INTRQDUCT ION' Chronic non-neoplastic lung diseases are major causes of permanent and temporary disability in the United States. Chronic obstructive pulmonary disease (COPD) is the largest subgroup of these diseases and in this report refers to chronic bronchitis and/'or emphysema. Relationships between smoking, and non-neoplastic lung diseases have been reviewed in previous reports on the health consequences of smoking (HCS 1, 2, 3, A, 5, 6,, Z, 8). Cigarette smoking, is the most important cause of COPD. Cigarette smokers have higher death rates from chronic bronchitis and emphysema, more frequentlyreport symptoms of pulmonary disease, and have poorer performance on pulmonary functionitests thanido nonsmokers. These differences become even more marked as the number of cigarettes smoked increases. The relationship between cigarette smoking and COPD has been demonstrated in many different national and ethnic groups, and is more stri.king,in men than in women. Pipe and cigar smokers have higher morbidity and mortality rates from COPDD than do nonsmokers but are at lower risk than cigarette smokers. Cessation of cigarette smoking results~in improvedipu'lmonary function tests, decreased pulmonary symptoms, and'reduced COPD mortali'ty rates., In additionto an increased risk of CC1PD~„ cigarette smokers are more frequently subject to and require longer convalescence from other respiratory infections than nonsmokers. Also,, if they require surgery, they are:more likely to develop postoperative respiratory complications. The relative importance of air pollution in the development of COPD remains controversial,, but is clearly less significant under most circumstances than cigarette smoking. The combination of cigarette smoking and polluted air, however,,may prod'uce higher rates of COPD' than either factor alone. Several occupational exposure groxips incur an increased risk of COPD, and cigarette smoking adds~significantly to this risk. In particu2ar, exposure to cotton fiber and coal dust appears to act in concert with. cigarette smoking to promote the development of pulmonary disease. Autopsy studies have demonstrated a dose-related effect of cigarette smokinQon the severity. of macroscopic emphysemal.Increased goblet cell density, alveolar septal rupture, thickened bronchial epithelium,, and mucous gland hypertrophy are more commonly!found in the lungs of smokers than in those of nonsmokers. Many pathophys3'.ologilc mech~anismsbywhichismoking, may cause COPD,hav!e been proposed'.. Decreased overall pulmonary clearance, reduced'1ciliary motion,, and impaired alveolar macrophage functions have all been related'to cigaret~tesmoking and probably play a role in the development of COPD. The exact mechanisms whereby cigarette smoking contributes to the development o£ COPD, O howevzr,. remain only parti'a1Tyunderstood-. ~. ~ O Ilk 65

SMOKING AND RESPIRATORY MORBIDITY An increased prevalence of respiratory symptoms in smokers from early teens to those past the age of 801has been well established. Bewleyr„ et al. (BP' 33), in a study in Derbyshire County, EngTand, extended these findings to include younger children. In alquestionnaire study of 7,115' school- children ages 10 to 11-1/2 years,, he foundd that 6.9 percent of the boys and 2.61percent of the girls smoked more than one cigarette per day. The boys who smoked reported more morning, cough (21.5% to 6,.,1%) „ cough during, the ) day or night (48.0% to 20%), and cough of 3-months duration (18'.0Z to 4.1%) than their nonsmoki!ng,schoolmates. The percentages for the girls were similar although based on smaller numbers of smokers., As in manystud'ies of'this type, it was impossible to control for air pollution, social class, or smoking,habits of the parents; nevertheless, the results suggest that cigarette smoking,even in this young age group produces respiratory symptoms. Fridy, et al. (BP 171), in alsomewhat older population (average age 25 years), examined' the effect of'smoking on airway function during,mild' viral illness. They measured' closing volumes for 22 subjects (9' cigarette smokers - average age 29.1, and 13 nonsmokers - average age 251.7) before onset andlat weekly intervals from the beginning,of a mild respiratory illness until all symptoms had subsided. The closing vollumes for smokers prior to illness were higher than those for nonsmokers, but the difference was not statistically significant. In the tests done during, the illness, the smokers had a statistically significant increase in the closing volumes (from 37.0 to! 45.8 percent of their total lung capacity, while nonsmokers hadino change 32.7 and 31.7 percent). Smokers remained symptomatic more thanitwice as long as nonsmokers (35.7'and 16.5' days, respectively), and the mean d'urationn d of pulmonary function abnormalities in smokers was 29.7 days. Nonsmokers had no change in pulmonary function tests during,ilTness.. SMOKING AND' AIR P'OLLUTION The relationships among air'po1!lution, smoking, and'COPD remain .controversial. Reasons for this controversy include difficulties in controlling such variables as socioeconomi'c class, degree of crowding, ethnic differences,, and age distribution as well as determining the exact type and amount~of individual pollution exposure. Measuri'ngind'ivi'.d'ual pollution exposure even withina small area is d'ifficult since both amount and type camvary dramatically from street to street (e.g., proximity of a street to a heavily traveled expressway). In an effort to control as many of these variables as possible, two basicap~proachesin~ studyd'esign have been,tried., Thefi'rst approach is to find areas where pollution levels have been well measured and then to select study populations that are as similar as possible in areas with different pollution levels. Thus, effects on a population in a.low pollution area can be compared to those on a similar population in a high pollution area. The secon&approach is to select a population that is as uniform as possible, for example, twins„ and then measure individual responses to different pollution exposure. Both approaches have drawbacks as will be evident from the following studies. 66

C Using, the first approach, the Community Eealth and' Environmental Surveillance System of the Environmental Protection~Agency (BP 29, BP' 14)) has conducted surveys in areas with different types and levels oTpMution in four different parts of the United States (Chicago,, New York City, the Salt Lake Basin, and1the Rocky Mountain area). Within each part of' the country, the researchers i'dentified commun3!ties of similar socioeconomic status but different pollution levels. They then administeredia questionnaire through the school systems to determine the frequency of lower respiratory tract infecti'onlin thechildren and their famil'ies:.Theyreported an increased incidence of lower respiratory tract illness in childrenlin high pollution communities compared to children in low pollution communities. This d'iffereince was demonstrable only in chil'dren1whosefamilies hadi lived in~ the high pollution communities for more than 3 years. They also reported an i'ncreased'prevalence of chronic bronchitis in parents who lived in highipollution communi!tiess compared tolparents from low pollution communities. They calculatedithe excess risk of chronicbr~onchitis prod'ucedbyairpo9!lution, to be one-thirdlof that produced by smoking but to be additive with smoking. Several major problems in these surveys make it difficult to evaluate the results. The authors describe the areas as having different kinds of polluti'on. The Salt Lake Basin and Rocky Mountain areas were felt to be high in sulfur dioxide (S©2) and'low in total suspended particles (TiSP), while New York and Chicago were high in both these pollutants. As a result,, in the Salt LakeB~asiniand Rocky Mountain areas, communiti'.eswere separated iinto lowand high pollution communities only on the basis of their SO2 levels.M'any communities classified as low pollution communities on the basis of their S02 levels had higher levels of total suspended particles than the communities classifiedias hi.gh pollution communities by SO level (Table 1). In fact, the average total suspended particles level for tie low pollution communities in the Salt Lake Basin was higher than that for the hligh pollution communities (Table 2) in the Salt Lake Basin. These differences exemplify the difficulties of using,only one pollutant as a marker of, total pollution exposure. Additional problems with these studies were the differences in socioeconomic class measurements between low and!high pollution communities in some of'the regions. In the Rocky Mountain area, the percentage of fathers who completed high school varied from 91 percent in one of the low communities to 58 percent in one of the high~pollution communities. There were also major differences between high and low pollution communities in the percentage of families with more than one person per.room in the Salt Lake Basin(59'.6% to,51.2Y)'~ , Rocky: Mountain area(',87'.,0% ' to68.0%') , and. New York (85.0%' to 72.0%). Residential stability (percentage of' families li'ving, in the community for more than 3 years) was different in the high and low pollution communities in New York (58.0T' to 36.Q%)', and Chicago (56.(1y' to. 46.0%). The percentage of parents who currently smoke also differed'for high and low pollution communities, iniNewY'ork (53% to45%'forthe fathers and 47% to 37% for the mothers). These differences rai'se questions as to whether b7

TABll.E~ 1'i. -~ bevels~of sulfur~ dioxide~(S(12) , and~to~tal suspended particulates~(T'91')~ in four i'l'tah communities, 19 71', and' iir five Rocky Mountain communities, 1970, ~ Atea Community' Pollu tion Pollution levels imu8/im3 Classdfi cation S02 TSP Utah (Salt Lake Basin). Low 8' 78 Intermediate 1 15' 81 Intermediate 2' 22 45 High, 62 66 Rocky Mountain Area Low 1. 1'0' S0 Low 2 26' 6'8' Low 3 46' 110. E$igh, ll 109 43 '. High 2 186' 102 SourcerChapman. R.S,, et al. (BP2A).' 68

TABLE 2. - Mean annual' leuels of sulfur dioxide (SO2)'and total suspended pareiculates (T'SP) in four areas Pollution lev els i n µg/m~' Area SOZ2 TSP ' During Study Low High~ Decade Preceding Study Low High During Study Low High Decade Preceding$tudy. Low High Five Rocky Mountain ~Areas~ 10 275 110' 263 45 110 50 1'01l Salt Lake Basin 9 65 < 20' 144 78 66 82 62 New York. 2'3 63' <'30' 431 34104 4'0, 2'01 Chicago 57 106 109 2S0 111 1'5'1 121 1i65 N'OTE., - Area includes, highest- and lowest-pollurtedlcommurrities. , Sourca: French, I1G., et ali (EP I4) ,

the high and low poll'ution communities were really similar enough populations to justify the claim that differences in incidence of respiratory tract illness could be attributable to differences in air pollution. Increased prevalence of'COPD has also been d'emonstrated in areas off high pollution in the Netherlands (BP 119), yokkaichi,, Japan (BP 78), , and' Cracow, Poland (;BP'112). Again, however,,these studies were poorly controlled for socioeconomic status. Several recently published studies have used the second major method of investigating the relationship between smoki'ng,, air pollution, and'COPD', i.e., to select a uniform:population and then to measure individual differences to poll'ution exposure. Comstock, et al. (BP' 13), in an attemptt to control for occupational exposure and socioeconomic class,,studied threeseparate,unsform populationsoftel!ephone workers and used as a measureoft pollution the location of'the place of work and residence. The populations studied were telephone installers and repairmen in~B'altimore~,, Newl'ork City, Washington, D.C.,, and rural fTTestchester County in 1962 (survey 1')' and in 19'67' (survey 2)'„ and telephone installers and'repairmen in Tokyo in 19'67 (survey 3). They were unable to find any relation between pulmonary symptoms andidegree of urbanization of place of work or place of residence (ei:ther current or past). They were:, however, able to establish alstrong correlation between smoking habits and' pulmonary symptoms. Given the crude estimation of pollution exposure used in this study (all workers in each city were treated, as th:oughthey:received the same:exposure),, a small d:Ifference, inisymptomss due to air pollution could have been missed,, whereas the difference due to smoking could be detected both because it was larger and'because it was possible to determine individual exposure more exactl:y. Hrubec, et al. (BP' 12), in a study of twins from the U.S. Veterans Registry, were unable to show a difference in respiratory symptoms either between individuals with different exposure to air pollution or between members of twin pairs with different air pollutiomexposures. However,, they too used a.crude measure of air pollution exposure (by each aip code area), and so . could have missed a small difference due to air pollution despite being, able to relate respiratory symptoms to smoking,,, socioeconoanicstaflus,,; and alcohol intake. Colley, et al. (BP 2321), in a study of 3,899 persons (',20-year-olds born during the last week of March 1946 in the United Kingdom)', were also unablee to show a relation between COPD and air pollution. They used as their esti'mates of air pollution exposure the domestic'coal consumption inithe towns where the subjects lived. This method' of estimating air pollution exposure is subject to the same limitation cited'for the previous two studies-limited sensitivitytoismall risks d'ueto,air pollu~tion. W