Tobacco Documents Online

I I I I I I I I I I I I INDOOR AND OUTDOOR AIR POLLUTION AND LUNG CANCER Du Ying-Xiu*, Huang Lan-fang**, Feng Zhen-zhi** and Feng 7ian-wei* * Department of Hygiene, Guangzhou Medical College, Guangzhou, China ** The Municipal Health & Antiepidemic Station of Guangzhou, Guangzhou, China Introduction In China, the urban lung cancer mortality rate is the highest of all cancers. In 16 Chinese cities, the mortality rate for all types of cancers rose from 100/100,000 in 1982 to 125/100,000 in 1986 (b = 0.0117, P < 0.05), with lung cancer not only accounting for 25 % of the total cancer deaths but also increasing at the fastest rate: from 25/100,000 in 1982 to 32/100,000 in 1988 (b = 0.0151, P < 0.01). The city of Guangzhou now has the third highest lung cancer death in China; only Chongqing and Shanghai have higher rates. In 1989, the five leading cancer-related standardized mortality rates (SMRs) in Guangzhou were: lung cancer 39.79/100,000, liver cancer 24.12/100,OOQ stomach cancer 9.67/100,000, nasopharyngeal cancer 6.07/100,000 and esophageal cancer 5.00/100,000. Notably, the lung cancer death rate was higher than the combined SMR for liver, stomach and nasopharyngeal cancers. Air pollution, smoking and certain occupational exposures are considered to be the three most important risk factors for lung cancer. However, their significance may vary due to locality or sex differences. In order to examine the potential risk factors for lung cancer in Guangzhou, we have undertaken an analysis of the relevant data collected during the last 20 years. This paper primarily discusses the relationship between indoor/outdoor air pollution and lung cancer. The association of atmospheric air pollution and lung cancer has long been noted. Stocks reported that the lung cancer incidence in various areas of Great Britain was related to the local atmospheric deposit index, smoke index, and population density.(1) A close relationship was found between lung cancer and 3, 4 benzo(a)pyrene, beryllium, molybdenum, vanadium and arsenic in the atmosphere(2). Blot and Fraumeni analyzed the distribution of lung cancer deaths in the United States and found higher lung cancer incidence in males in locations near paper manufacturing, chemical engineering, petroleum, and vehicle manufacturing plants, and concluded that lung cancer was associated with both atmospheric pollution and occupational exposures(3). Xu and Blot performed a case-control study in the industrial city of Shenyang and found that, in addition to smoking, lung cancer was related to indoor/outdoor air pollution, because a higher incidence of lung cancer was found in men and women living near refining facilities for long periods of time(4). The traditional coal-burning "kangs" appeared to be an important contributing factor to indoor air pollution. Moreover, they thought that smoking and indoor/outdoor air pollution were associated with squamous cell carcinoma and small cell carcinoma. In their 1987 case-control study in Shanghai, Gao etal.(5) found that indoor air pollution and the use of rapeseed oil for cooking had an important significance in female lung cancer. Our current study combines the atmospheric pollution data for the last 20 years, the indoor air pollution data from a 2-year study, and Guangzhou's lung cancer mortality records for the last 10 years in a retrospective case-control study in order to examine the relationship between indoor/outdoor air pollution and lung cancer.

I Materials and Methods 1. Retrospective survey of lung cancer deaths. Guangzhou has a population of approximately 2 million, who live in four city districts and who are served by 63 local police stations. Each local police station keeps complete and systematic records of the residents, which include population characteristics such as the number, sex, age, occupation, street address of the residents, and the annual registration of births and deaths, (including time and cause of death). We have conducted retrospective surveys of all lung cancer deaths between 1980-1988 by interviewing the relatives of the decedents using standardized questionnaires. Surveys were conducted once each year (to cover lung cancer deaths that occurred in the previous year) by trained medical staff who also examined the hospital records prior to death. All data were entered into computers to establish a database. The data fields are as follows: name, sex, city, date of birth, date of death, prior native city, proximity to sources of pollution, education, marital status, religion, date of initial diagnosis of lung cancer, where diagnosed, level of diagnosis, method of diagnosis, cell type, tumor site, lung involved, metastasis, site of metastasis, medical history, smoking history, smoking members of household, family status, kitchen facilities, diet history, occupational history, family tumor history, relative giving the information, hospital record and cooperation of the relative providing the information. 2. Atmospheric pollution. The geographic area of Guangzhou is approximately 55 square kilometers, divided into the four districts of Liwan, Yuexiu, Dongshan and Haizhu. From 1972 to 1977, the Guangzhou Health and Antiepidemic Station established 30 atmospheric pollution sampling stations and tested the atmospheric content of S02 and sedimented dust (SD). SO2 (µg/m3) was analyzed using the hydrochloric-rosaniline method from seasonal samples, collected during 5 consecutive days, 3 samples a day, once every month. The samples were collected in glass containers, 15 cm diameter and 30 cm height, placed on roof-tops. The naturally sedimented dust was measured by weight in ton/km2/month. As part of the WHO's global monitoring plan, a reference station and four sampling stations were established in Liwan, Yuexiu, Dongshan, Haighu to study the total suspended particulates (TSP) and SO2 in 1981-1990. SO2 and TSP, continuously collected from the first to the 15th day of each month, were analyzed according to WHO's criteria.(6) The databases from these two tests were quite large. The following formula was used to calculate the air pollution index of various localities of Guangzhou. 1 n Ci API = E where: n i=1 Si O 00 n - air pollutants tested ~ Ci - concentration of air pollutants 00 Si - acceptable level of pollutant (according to government guidelines) W W I I I I I I , I I I I I I I I

I I I I In December, 1984, the Guangzhou Municipal Science & Technology Committee organized an aero-remote sensing test for the city of Guangzhou. Atmospheric pollution was assessed by observing the effects of pollution on plants by the vegetation ecoline on air infrared color film and sychromonitoring. Levels of air pollution were classified as clean, light, medium, heavy or severe. 3. Indoor air pollution. I I I I I I I In order to examine indoor air pollution caused by coal fumes, a two-year continuous systematic study of indoor air pollution was conducted in 1984-1985. Five households from each of the four districts of Guangzhou, 20 in total, were randomly selected. The daily and seasonal variations of SO2 and NOx were tested 7 times daily with samples taken every two hours from 7 a.m. to 7 p.m. for 5 consecutive days, once during each season, i.e. four times in a year. Since the 1980s, residents of Guangzhou have been in the process of gradually switching over from coal to gas. In order to compare the indoor air pollution associated with either fuel, 5 coal-user households and 5 gas-user households were again selected from the four districts in 1986. The indoor air contents of 502, NOx, CO, TSP, SD, B(a)P, radon and thoron were tested. Moreover, the urine B(a)P levels of housewives in both coal and gas households were also tested. The housewives were all non-smokers with at least one year's experience in managing the households. S02 was analyzed by the hydrochloric-rosaniline method, NOx by the diphenylamine hydrochloride colorimetric method, CO by gas chromatography, TSP by weighing 4-hour continuous samples from ambient air collected using large flow samplers. SD was weighed after 15 days' exposure of flat dishes of 15 cm diameter placed 1.7 meters above ground. B(a)P was analyzed by paper chromatography fluorospectrophotometer and urine B(a)P content by subjecting polyurethane foam absorbed B(a)P (from urine volume above 2000 m]) to fluorimetric analysis. Radon and thoron samples were taken by DK-60 particulate samplers and examined by a FJ-13 a radiometer. 4. Case-control study. In 1985, there were 806 cancer deaths (531 males, 275 females) in Guangzhou. We matched the cases of lung cancer deaths with controls of noncancer deaths in the same year and in the same residential areas for sex, age (± 2 years), and obtained 659 pairs (82% of total number of cases); there were 143 male pairs and 216 female pairs. By using the Mantel-Haenszel method, we calculated the relative risks of smoking and exposure to coal dust at the 95% confidence level. Results 1. Distribution of lung cancer deaths by districts. ' Between 1980 and 1988, a total of 6,812 lung cancer deaths were registered at the 63 local police . stations of Guangzhou. After excluding nonprimary lung cancers and less than 10-year residencies, 5546 cases were obtained (3760 males, 1786 females; ratio 2.1:1). Based on police station records, the M ' regional distribution of lung cancer deaths was calculated and is shown in Figure 1. The highest lung 0 , 00 cancer rate (37.94/100,000) was in Liwan, and in descending order, Yuexiu (35.99/100,000), Haizhu -3 V O t 3- W  IV O I

! 1 (31.50/100,000), and Dongshan (30.79/100,000), suggesting a history of pronounced difference in lung cancer death rates by districts in Guangzhou. 2. Atmospheric pollution. Between 1972 and 1979, data obtained at 30 sampling stations were used to calculate the atmospheric pollution index (API). It was found that Liwan had the highest API (2.49), followed by Yuexiu (1.68), Haizhu (1.64) and Dongshan (1.17). For comparison, the "control" region had only an API of 0.57. Between 1982 and 1990, four sampling stations were established in order to provide 24-hour around the clock monitoring. Samples were obtained on 180 days each of the nine years and then analyzed. The results showed that Liwan had the most severe API (0.898), followed by Yuexiu (0.721), and lastly Dougshan (0.470). For comparison, the control district only showed an air pollution index of 0.246. The severity of atmospheric pollution was also studied in 1984 with an aero-remote sensing system which further substantiated the atmospheric pollution for Liwan to be severe/heavy, Yuexiu to be heavy/medium and Dongshan and Haizhu to be medium/light, and control to be clean. (Table 1) Table 1. Comparison of Atmospheric Pollution and Lung Cancer Death Rate in Four Districts of Guangzhou Liwan Yuexiu Dongshan Haizhu Control AP Index (1972-79) 2.49 1.68 1.17 1.64 0.57 AP Index (1982-90) 0.898 0.721 0.470 -- 0.246 Degree of pollution by aero severe/ heavy/ medium/ medium/ clean remote sensing (1984) heavy medium light light Lung Cancer death rate (1976-87) 37.94 35.99 30.79 31.50 -- Results of these surveys are also shown in Figures 2-4 which show clearly that atmospheric pollution for the past 20 years was the most severe in the Liwan district, followed by the Yuexiu district. This is highly correlated with the higher lung cancer death rates for these two districts. 3. Indoor air pollution. Indoor and outdoor air pollution studies carried out over a two-year period (1984-1985) showed that indoor air pollution was more severe than outdoor air pollution (Table 2). This probably reflected the fact that during this period of time most of the factories in Guangzhou were located outside the city and automobiles were rather scarce. 4 I I I I I I I I ' I I I I I I I I

I I I I , I I I I I I I I I I I I I Table 2. Comparison of Average Levels of SO2, NOX, TSP and B(a)P Between Indoor and Outdoor Air in Guangzhou (19841985) B(a)P SO (µ g/m3) NO (µ g/m3) TSP (µ g/m3) (µ g/10om3) Indoor 190 t 80 70 t 30 , 210 t 70 1.30 t 0.98 Outdoor 80 t 20 40 f 10 200 t 30 0.50 f 0.26 Three peaks of indoor SO2 and NOX were found, (7 a.m., 11 a.m. and 7 p.m.) which were also substantially higher in winter/spring months when doors/windows were generally closed, than in the summer or fall (Figure 5), suggesting that the primary source of indoor air pollution originated from cooking. Residents of Guangzhou depended on wood in the 50s, on coal during the 60s, and progressively switched to propane and gas in the 80s. In 1991, about 40% of the families in Guangzhou used gas. Thus, indoor air pollution was probably most pronounced from 1960 to 1990. Table 3 compares the indoor air pollutant levels in houses using gas as compared with those using coal. Not only is indoor air pollution higher in coal users' homes, housewives and family members of households using coal also have higher concentrations of benzo(a)pyrene in the urine. -5- I

I Table 3. A Comparison of the Concentrations of Some Air Pollutants and Urine-B(a)P Between the Coal-Burning and Gas-Burning Kitchens Briquette coal- butning kitchen Liquefied petroleum gas- burning kitchen Coal/gas SO2 (µ g/m3) 279 58 4.81 NOX (µ g/m3) 76 63 1.21 CO (µ g/m3) 9420 2340 4.03 TSP (µ g/m3) 332 188 1.77 SD (µ g/m3) 12 5 2.40 BaP (µ g/100 m3) 11.9 2.2 5.41 Radon (Bq/m3) 18.6 16.6 1.12 Thoron (Bq/m3) 42.5 28.3 1.50 Urine-BaP (ng/1) 4.0 2.8 1.43 11 4. Case-control study. The results of this study are shown in Table 4 which shows that cigarette smoking is the most significant risk factor for lung cancer in males and is less important for lung cancer in females. Indoor air pollution is the most significant risk factor in females and has no obvious association with lung cancer in males. -6- I I I I I I I I I I I I I I I I I ,

I I I I ' I I 1 I I , I I I ' I Table 4. A Case-Control Study Among 659 Lung Cancer Deaths RR (95% CL) P-Value Male Smoking 3.53 (2.44 - 5.11) < 0.001 Coal fumes exposure 0.89 > 0.05 Female Smoking 1.93 (1.30 - 2.87) < 0.01 Coal fumes exposure 2.21 (1.16 - 4.21) < 0.01 Discussion The incidence of lung cancer is known to be higher in industrialized nations compared to developing countries. For a given country, lung cancer is more prevalent in highly industrialized zones, compared to agricultural regions and is also significantly higher in the city than in the country. These observations underscore the importance of atmospheric pollution as a significant factor for the induction of lung cancer. However, because the development of lung cancer is known to be associated with a multitude of risk factors, and with an extremely long latency, and because atmospheric pollutants are known to be complex and have dynamic and variable interactions with humans, studies that intend to explore the inter-relationship between atmospheric pollution and lung cancer require data generated by long-term observations. Moreover, the existence of lung carcinogens in the atmospheric pollution should also be clarified. The severity of atmospheric pollution is influenced by many factors, such as: population density, the nature and sophistication of industries in that locale, source and type of energy and number of automobiles, city planning and number of trees, geographic location, weather conditions, etc. At the present time in China, the primary source of atmospheric pollution comes from coal fumes generated by the use of coal for cooking and heating in the home. In northern China where coal is used for heating, the smoke is commonly vented to the outside by a chimney. This results in severe atmospheric pollution. In southern China, by contrast, chimneys are usually not available to vent the coal smoke. Instead, smoke from burning coal remains indoors, giving rise to severe indoor air pollution. We have investigated the relationship between atmospheric pollution in the last 20 years and the number of lung cancer deaths in the last 10 years. Our studies show that the districts of Guangzhou with the more severe atmospheric pollution are the same ones with the higher rate of lung cancer deaths, thus confirming the close relationship between atmospheric pollution and lung cancer deaths. Our case-control study further shows that in the case of females, the majority of whom are lifetime nonsmokers, indoor air pollution actually is a more important and significant risk factor than active smoking. N Lb ~ y 00 _7_ ~ A J

I Both cigarette smoking and coal-burning are known to produce benzo(a)pyrene. While most of the existing studies of the effects of benzo(a)pyrene are based on the use of animal cells and animal models, our recent studies show that benzo(a)pyrene is effectively biotransformed by microsomes prepared from human lung tissues. The metabolites, when incubated with human fetal tracheal epithelial cells, induce an increase in unscheduled DNA synthesis, promote the formation of micronuclei, and introduce point mutation in codon 12 of the H-ras oncogene. Taken as a whole, these results provide direct evidence for the ability of benzo(a)pyrene to induce human lung cancer. Additional studies are needed to show whether other carcinogens exist in atmospheric pollutants. Many other studies in China have demonstrated a close relationship between lung cancer and atmospheric and indoor air pollutants. Liang et al.(7) showed that although both coal and wood smoke contain potential carcinogens, coal smoke is regarded to have a higher carcinogenic potential than wood smoke. Wang et al.(8) showed that a high coal consumption index, indoor smog pollution and low ceiling height in the living quarters were major risk factors for lung cancer. Guan et al.(9) reported that extracts of air particles collected from Beijing, Taiyuan and Xuanwei all had potential carcinogens. Moreover, the smaller size particles were found to have a stronger potential for carcinogenicity. Ye et al.(10) showed that air particles collected in five Chinese cities (Beijing, Taiyuan, Wuhan, Shenyang and Xuanwei) were mutagenic based on the Ames and the SCE tests, with the highest lung cancer incidence area, Xuanwei, showing the most severe air pollution. Wang et al.(11) demonstrated that cooking oil fumes were a common risk factor for lung cancer and that winter heating by coal stoves was a risk factor for squamous cell carcinoma of the lung. In conclusion, while the relationship between atmospheric pollution and lung cancer has been established, some observations on this relationship remain to be explained. For example, while indoor air pollution due to coal burning has generally been regarded as a risk factor for lung cancer in nonsmoking females, coal has been in use in China for thousands of years, the question arises why the incidence of lung cancer has been rising only in the last 20 to 30 years. Moreover, atmospheric pollutants are thought to give rise primarily to squamous cell carcinoma; this appears to be inconsistent with the prevailing adenocarcinoma cell type commonly observed in nonsmoking females. The answers to these seemingly conflicting observations must, therefore, await additional studies in the future. -8- I I I I I I I I I I I I I

I I i I I I , I I I I References Stocks, P. "Cancer and bronchitis mortality in relation to atmospheric deposit and smoke," Br. Med. J. 1: 74-79, 1959. 2. Stocks, P. "The relationship between atmospheric pollution in urban and rural localities and mortality from cancer, bronchitis and pneumonia with particular reference to 34 benzopyrene, beryllium, molybdenum, vanadium and arsenic," Br. J. Cancer 14(3): 397-418, 1960. 3. Blot, W.J. and Fraumeni, J.F. "Geographic patterns of lung cancer: industrial correlation," Am. J. Epidemiol. 103(6): 539-550, 1976. 4. Xu, Z.Y.; Blot, W.J.; Zhao-yi, Xu et al. "Smoking, air pollution, and the high rates of lung cancer in Shanyang," China J. of the National Cancer Institute 81(23): 1800-1806, 1989. 5. Gao, Y.T. et al. "A case-control smdy of female lung cancer in Shanghai," Second Symposium on Lung Cancer Research, Guangzhou, p. 7, 1987. 6. WHO. "Sulfur oxides and suspended particulate matter," Environmental Health Criteria 8, Geneva, 1979. 7. Liang, C.K. et al., "Kuming mice skin tumor initiating activity of extracts of inhalable particles in indoor air," Chinese J. Prev. Med. 21: 316-318, 1987. 8. Wang, F.L. et al. "Analysis of risk factors for female lung adenocarcinomas in Harbin," Chinese J. Prev. Med. 23: 270-273, 1989. 9. Guan, N.Y. et al., "A study of carcinogenicity of extracts from different size particles in air," Chinese J. Prev. Med. 24: 9-12, 1990. 10. Yu, S.Y. et al., "Study on mutagenicity of size fractionated air particles," Chinese J. Prev. Med. 25(2): 70-74, 1991. 11. Wang, G.X. et al. "Multi-variate analysis of causal factors including cooking oil fume and others in matched case control study of lung cancer," Chinese J. Prev. Med. 26(2): 89-91, 1992. O 00 ..a -4 ' 00 W ~ ! 9 I

.. .. m .. .rm a. .m .m m .m w w .= m .. ..m m .. _ 2081783327 Figure 1. DISTRIBUTION OF LUNG CANCER DEATH RATE (pcr 100,000) IN GUANGZIfOU (1976-1983) 111111 :ra~, '[~ ~~~_ Llunn Yucatu Unuptslulu ~ " )~ . - II y.: I Ilfl)Z1111\ I.uug Cnncer DeoCh RnCe ( per 100,000 ) 0 - 21) - 30 - 60 - 50 - FM ® MEN win ®. lf. 1-.~r I?