行政院國家科學委員會專題研究計畫 期中進度報告
食入砷與血管疾病劑量效應關係之長期追蹤研究(1/3)
計畫類別： 個別型計畫 計畫編號： NSC91-2320-B-002-075- 執行期間： 91 年 08 月 01 日至 92 年 07 月 31 日 執行單位： 國立臺灣大學公共衛生學院流行病學研究所 計畫主持人： 陳建仁 報告類型： 精簡報告 報告附件： 出席國際會議研究心得報告及發表論文 處理方式： 本計畫可公開查詢中 華 民 國 92 年 6 月 3 日
Full title: Ingested Arsenic is a Dominant Risk Factor for
Cardiovascular Mortality: a Ten-year Follow-up Study
First author: Wang
Short title: Ingested Arsenic and Cardiovascular Mortality
Search Codes: [8], [135]
Word counts: 4485
Chih-Hao Wang, MD; Chi-Ling Chen, PhD; San-Lin You, PhD; Lin-I Hsu, PhD;
Hung-Yi Chiou, PhD; Yu-Mei Hsueh, PhD, Shu-Yuan Chen, PhD; Meei-Maan Wu,
PhD, Chuhsing Kate Hsiao, PhD; Chien-Jen Chen, ScD; for the Blackfoot Disease
Study Group
From the Graduate Institute of Epidemiology, National Taiwan University, Taipei
(C.-H.W., C.-L.C., S.-L.Y., L.-I.H., C.K.H., C.-J.C.); School of Public Health, Taipei
Medical University, Taipei (H.-I.C.,Y.-M.H.); Division of Biostatistics, National
Health Research Institutes, Taipei (S.-Y.C.); and Institute of Biomedical Sciences,
Academia Sinica (M.-M.W.). Address reprint requests to Dr. Chien-Jen Chen at
University, 1 Jen-Ai Road Section 1, Room 1547, Taipei 100, Taiwan, or at
ABSTRACT
Background － Ingested arsenic has been documented to increase the risk of
atherothrombotic diseases. This prospective study aimed to compare cardiovascular
mortality and risk factors in arsenic-exposed and unexposed cohorts.
Methods and Results－ During the period of 1988-1992, we enrolled 1563 subjects
from the southwestern area of endemic arseniasis in Taiwan and 23942 subjects from
seven non-endemic areas in Taiwan and Penghu archipelago. Traditional risk factors
and ingested arsenic exposure were evaluated through health examination and
structured questionnaire interview. Deaths from ischemic heart disease and stroke
were ascertained through data linkage with national death certification programs.
Cox's proportional hazards regression analyses were used to estimate
multivariate-adjusted relative risks of cardiovascular mortality. There were 17,666
and 210,885 person-years under observation for arsenic-exposed and unexposed
cohorts, respectively. Exposed cohort had a significantly higher cardiovascular
mortality than unexposed cohort, showing the relative risks of 2.4 for ischemic heart
disease and 2.2 for stroke mortality after adjustment for age (1-year increment),
gender, smoking and alcohol consumption (both p values <0.01). There was a
significant dose-response relationship between the cumulative cardiovascular
risk factors such as current smoking, diabetes mellitus and hyperlipidemia were
significantly associated with mortality only in the unexposed cohort and lost their
significance in exposed cohort.
Conclusions － Ingested arsenic was a dominant risk factor and significantly
increased mortality from ischemic heart disease and stroke. Traditional risk factors
may be dominated or attenuated by arsenic.
Condensed abstract: This prospective study enrolled 1563 subjects from
arseniasis-endemic area in southwestern Taiwan and 23942 subjects from
non-endemic areas in Taiwan and Penghu archipelago. Exposed cohort had a
significantly higher cardiovascular mortality than unexposed cohort, showing the
multivariate-adjusted relative risks of 2.8 for ischemic heart disease and 2.2 for stroke
mortality (both p values <0.01). There was a significant dose-response relationship
between the cardiovascular mortality and the arsenic level in drinking water (P
<0.0001 trend test). Ingested arsenic is a dominant risk factor for cardiovascular
Arsenic is a ubiquitous metalloid in the crust of the earth. Human exposure to
inorganic arsenic is mainly through ingestion of drinking water contaminated with
naturally occurring arsenic.1 Chronic arsenic poisoning is becoming an emerging
epidemic in Asia. Over 100 million people are exposed to underground water with
high concentration of arsenic.2 The magnitude of this arsenic calamity was projected
to be the largest in history of environmental disaster that will be more serious than
those at Chernobyl, Ukraine in 1986 and Bhopal, India in 1984.3 In the United States,
over 350,000 people are estimated to be exposed to water contaminated with arsenic > 50 g/L and over 2.5 million to water with arsenic > 25 g/L.4
Long-term exposure to ingested arsenic has been documented to induce various
cancers and atherothrombotic diseases.5,6 The maximum contamination level for
arsenic in drinking water has recently been lowered from 50 to 10 g/L by the US
Environmental Protection Agency.7 However, there remains argument on the scientific
basis of this new regulatory standard and its adequacy for protection of public health.
Ingested inorganic arsenic has been documented to cause blackfoot disease, a
unique endemic peripheral vascular disease in southwestern Taiwan.8 Clinically, the
disease begins with coldness and numbness of lower extremities, progresses over
amputation of distal parts of affected extremities. Pathologically, the disease is
compatible with thromboangiitis obliterans (30 %) and arteriosclerosis obliterans (70
%) with a fundamental change of an unduly developed severe systemic
arteriosclerosis.8,9 The pleiotropism of inorganic arsenic-induced health effects may
be characterized by its associations with hypertension, diabetes mellitus, carotid
atherosclerosis, ischemic heart disease (IHD), stroke and various cancers in a
dose-response relationship and its adverse health effects on gastrointestinal,
pulmonary, hematological and immune system.2, 5, 6, 10-16 However, the biological
gradient between ingested arsenic and the cardiovascular mortality has never been
reported by a large-scale long-term follow-up study.
This follow-up study aimed to compare the mortality from IHD and stroke as
well as their risk factors among residents in arsenic-exposed and unexposed areas.
The dose-response relationship between arsenic level in drinking water and
cardiovascular mortality was also examined.
Methods
Study Cohorts in Arsenic-exposed and Unexposed Areas
recruited. The arsenic-exposed area included Homei, Fuhsin, and Hsinming villages
in Putai Township located on the southwestern coast of Taiwan. Residents started
using arsenic-contaminated artesian well water since early 1910s.14 To recruit a cohort
of residents for prospective study, we selected as eligible only those who lived > 5
days a week in the township. A total of 2258 residents were registered, but only 1563
of them were qualified. A public water supply system using surface water was
implemented in this area since early 1960s, but its coverage remained low until early
1970s. Artesian well water was not used for drinking and cooking after the
mid-1970s.
To establish an unexposed cohort for long-term follow up study, Sanchi,
Chutung, Putze and Kaoshu were selected randomly from Taiwan, and Paisha, Huhsi
and Makung from Penghu archipelago. Geographically, Sanchi and Chutung are
located on northern Taiwan, Putze and Kaoshu on southern Taiwan, and Paisha,
Huhsi and Makung on Penghu archipelago positioned between Taiwan and Mainland
China in the middle of Taiwan Strait. A total of 23,942 residents in the seven areas
were qualified for recruitment. The residents in unexposed townships used tap water
and shallow well water for cooking and drinking. The seven non-endemc townships
are compatible with the endemic area in many aspects. Most residents in these areas
average socioeconomic status, life style, dietary pattern and average index of
accessibility to health care facility are comparable in both cohorts.
Questionnaire Interview and Arsenic Exposure
Based on a structured questionnaire, well-trained public health nurses conducted
the standardized personal interview of participants. Information obtained from the
interview including socioeconomic and demographic characteristics, alcohol intake,
cigarette smoking, physical activities, dietary pattern, residential history, water
consumption, and history of hypertension, diabetes.
The estimation of ingested arsenic exposure of study subjects in the exposed
cohort has been described previously.14-17 In brief, the arsenic level in water of
artesian wells of southwestern exposed area was obtained from previous studies
conducted in early 1960s.18, 19 As residents in this area shared few wells in the same
village, the median arsenic level in water of shared wells was used as the exposure
level of arsenic for residents in this village. The average arsenic exposure level in
drinking water of each study subject was derived by the formula (Ci×Di)/ (Di);
where Ci was the median arsenic level in water of shared wells of a village in which
the subject inhabited, and Di was the duration of drinking artesian well water in the
area ranged from 700 to 930 g/L in 1960s.10
Residents in unexposed areas used tap
water and/or shallow well water for cooking and drinking, and the arsenic level in drinking water in these areas have been measured to be <10 g/L.19, 20
Laboratory Examinations
Fasting blood samples were collected from study subjects participating the
health examination and serum level of total cholesterol, and triglycerides were tested.
Glucose tolerance test was also performed. Diabetes mellitus was defined as 1) a fasting serum glucose level 140 mg/dL, 2) a two-hour glucose level 200 mg/dL, or 3) a history of diabetes mellitus treated with oral hypoglycemic agents or insulin.
Anthropometric characteristics including height, weight, and systolic and diastolic
blood pressures were measured according to a standard protocol. The average of
three blood pressure measurements with a mercury sphygmomanometer was used to
define the status of hypertension. Hypertension was defined as 1) an average systolic blood pressure 160 mmHg, 2) an average diastolic blood pressure 95 mmHg, or 3) a history of hypertension treated with anti-hypertensive agents. Hyperlipidemia was defined as 1) a serum total cholesterol level 240 mg/dL, or 2) a serum triglycerides level 200 mg/dL.
Ascertainment of Causes of Death
Causes of death of study subjects were ascertained through data linkage with
national death certification profiles. It is mandatory to register any event of birth,
education, marriage, employment, and death in Taiwan. The registration information
of each household is double-checked annually by household registration officers.
Almost all (99 %) death certificates are confirmed and issued by clinical doctors in
private practice or in hospitals in Taiwan. The death certification profiles in Taiwan
have been considered complete, updated and accurate. Causes of deaths were
classified according to the codes of the International Classifications of Diseases, 9th
Edition. The codes were 410-414 for IHD and 430-438 for stroke. Cumulative
cardiovascular mortality included IHD and stroke death.
Statistical Analysis
The t or χ2 test was used to compare the demographic characteristics between
arsenic-exposed and unexposed cohorts. To assess the association between traditional
risk factors and IHD and stroke mortality, Cox's proportional hazards regression
analyses were performed to estimate the age-gender-adjusted and
multivariate-adjusted relative risks and 95% confidence interval for arsenic-exposed
drinking water and relative risk of cumulative cardiovascular mortality was examined
by a trend test.
RESULTS
Cohort Characteristics
The characteristics of arsenic-exposed and unexposed cohorts were shown in
table 1. The exposed cohort had a significantly higher prevalence of alcohol
consumption, hypertension, diabetes mellitus and hyperlipidemia and than the
unexposed cohort. The unexposed cohort had a significantly higher prevalence of
current smoking status and male gender (both P<0.001). Age and BMI were
marginally higher in exposed cohort.
Age-gender- adjusted risks of mortality from IHD and stroke
Up to December 31, 2000, there were 210,885 and 17,666 person-years under
observation, for unexposed and exposed cohort respectively.
Table 2 showed the age-gender-adjusted relative risks of mortality from IHD and
stroke in arsenic-exposed and unexposed cohorts. Current smoking, diabetes mellitus,
from IHD and stroke for unexposed cohort after adjustment for age and gender (all p
values < 0.05). However, the associations between mortality from IHD and ischemic
stroke and these traditional risk factors (male gender, current smoking, diabetes
mellitus, hypertension, hyperlipidemia, and BMI) were not statistically significant for
exposed cohort after adjustment for age and gender (all p values > 0.05).
Multivariate-adjusted risks of mortality from IHD and stroke
Table 3 shows multivariate-adjusted relative risks of mortality from IHD and stroke
in two cohorts. Hypertension was significantly associated with the mortality from
IHD and stroke in both exposed and unexposed cohorts (all p values < 0.05, except
IHD mortality in exposed cohort). The association between IHD and stroke mortality
and current smoking, and diabetes mellitus were significant (marginally significant
for hyperlipidemia) only in the unexposed cohort but were still insignificant in the
exposed cohort. Figure 1 showed the cardiovascular (IHD and stroke) mortality rates
per 100,000 person-years for age groups of < 50, 50-54, 55-59 and 60+ years
respectively, were 23.8, 104.8, 132.2 and 144.6, for unexposed cohort; and 58.7,
306.3, 315.9 and 434.5 for exposed cohort. Exposed cohort had a significantly higher
2.4 for IHD and 2.2 for stroke mortality after adjustment for age (1-year increment),
gender, cigarette smoking and alcohol consumption (both p values <0.01). Figure 2
showed relative risk of cumulative cardiovascular (IHD and stroke) mortality after
adjustment for age, gender, current smoking and alcohol smoking in exposed and
unexposed cohorts categorized by arsenic level in drinking water. The
multivariate-adjusted relative risks of cumulative cardiovascular mortality were 2.30
(0.84-6.29), 3.83 (1.64-8.94), and 2.29 (1.45-3.62) for groups <50, 50-699, and > 700
ppb respectively, compared with group <10 ppb. A significant dose response between
average arsenic level and relative risk of cumulative cardiovascular mortality after 10
years follow up (p <0.0001 by trend test). Cancer death was an important competing
cause of death. The multivariate-adjusted relative risks of all cancer death were 2.59
(1.41-4.77), 2.23 (1.09-4.57), and 2.77 (2.08-3.67) respectively for groups <50,
50-600, and > 700 ppb respectively, compared with group <10 ppb (p <0.0001 by
trend test).
Previous findings on the increased risk of atherothrombotic diseases including
peripheral vascular disease, ischemic heart disease and stroke were mostly reported
by ecological correlation studies or cross-sectional surveys. They were limited by
ecological fallacy, a small number of affected patients or difficulty in delineation of
causal temporality. This study followed a large number of residents in
arsenic-exposed and unexposed areas in Taiwan, and found a significant
dose-response relationship between ingested arsenic and mortality from ischemic
heart disease and stroke. This is the first long-term follow-up study demonstrated a
biological gradient between ingested arsenic and mortality from IHD and stroke in
exposed and unexposed cohorts. As arsenic may induce various lethal cancers as
competing causes of death, the mortality from ischemic heart disease and stroke by
ingested arsenic level might be underestimated.
Exposed cohort had a significantly higher prevalence of hypertension, diabetes
mellitus, and hyperlipidemia than unexposed cohort. Diabetes mellitus,
hyperlipidemia and current smoking were found to play an insignificant role in
cardiovascular mortality for the exposed cohort. The hazardous effect of these risk
factors might be attenuated or dominated by ingested arsenic, a dominant
environmental risk factor. A dose response relationship between ingested arsenic and
identified a dose response relationship between ingested arsenic and hyperlipidemia
(unpublished data). Ingested arsenic may in fact be a causal factor in hypertension,
diabetes mellitus and hyperlipidemia. Therefore relative risk of IHD and stroke
mortality adjusted by these three factors would represent overcontrol. The drop
phenomenon in the relative risk of cumulative cardiovascular mortality (in study
group >700 ppb) was most likely due to competing cause of lethal cancers such as
lung, liver, kidney and bladder cancers and may lead to underestimation of
cardiovascular mortality.
Pathology of BFD has been extensively studied in 51 patients from southwestern
exposed area.9 The universal finding is the systemic arteriosclerosis involving large,
medium and small arteries. Severe atherosclerosis of coronary and other
medium-sized arteries, and repeated and extensive myocardial infarction were found
in a young female BFD patient.In an autopsy study at Antofagasta of Chile, infants
exposed to ingested arsenic had systemic arterial intimal thickening in small and
medium arteries involving the heart, gastrointestinal tract, liver, skin and pancrease.21
Cytotoxic effect of arsenic results from its damage on the mitochondrial
respiratory function, specifically on nicotinamide adenine dinucleotide (NAD)-linked
enzymes, such as pyruvate dehydrogenase.22-24 Arsenic also alters mitochondrial
oxidative phosphorylation and decrease in cellular production of ATP may result in
the induction of reactive oxygen species and induction of stress proteins.23, 25
Disturbance of mitochondrial respiratory function, generation of oxidative stress, and
alterations in the mitochondrial structure may lead to cellular injury, necrosis and/or
apoptosis.22-25
Inorganic arsenic is methylated to monomethylarsonic acid and dimethylarsinic
acid in humans using S-adenosylmethionine as the main source of methyl group.26
The process of arsenic methylation may thus lead to the elevation of blood
homocysteine level, which has been known to be associated with increased risk of
cardiovascular disease.27, 28
Arsenic may induce various cancers and vascular diseases including
angiosarcoma and atherosclerotic plaques.29 These findings suggest that somatic
mutation and cell proliferation may be involved in the dual effects of arsenic on
carcinogenesis and atherosclerosis.29 NADH oxidase activation and superoxide
production were reported to be involved in arsenic-induced oxidative DNA damage in
human vascular smooth muscle cells.30 Sodium arsenite can induce the increase in
mRNA transcripts of growth factors, including granulocyte macrophage-colony
stimulating factor, transforming growth factor, tumor necrosis factor peroxynitrite generation and cyclooxygenase-2 protein expression31-34
arsenic may induce atherothrombosis through its roles in the induction of oxidative
stress, homocysteinemia, chromosomal abnormalities, gene amplification,
inflammatory reaction, increased expression of many growth factors, cytokines and
chemokines, and augmented aggregation of platelets 29-36
References
1. World Health Organization. Environmental health Criteria 18: Arsenic. Geneva,
Switzerland: World Health Organization. 1981:43-102.
2. Chen CJ, Hsu LI, Tseng CH, et al. Emergent epidemics of arseniasis in Asia, in
Chappell WR, Abernathy CO, Calderon RL, ed. Arsenic Exposure and Health
Effects. Elsevier, Amsterdam, 1999;113-121.
3. Smith AH, Lingas EO and Rahman M. Contamination of drinking-water by
arsenic in Bangladesh: a public health emergency. Bull World Health Org. 2000;
78(9):1093-1103.
4. Nordstrom DK. Worldwide occurrences of arsenic in ground water. Science.
2002; 296:2143-2145.
5. Chen CJ, Wu MM, Lee SS, et al. Atherogenicity and carcinogenicity of
high-arsenic artesian well water: multiple risk factors and related malignant
neoplasms of blackfoot disease. Arteriosclerosis. 1988;8:452-460.
6. Chen CJ, Kuo TL, Wu MM. Arsenic and cancers. Lancet 1988;20:414-415.
7. Smith AH, Lopipero PA, Bates MN, Steinmaus CM. Arsenic epidemiology and
drinking water standards. Science. 2002;296:2145-2146.
disease with arsenic. Environ Health Prospect. 1977;19:109-119.
9. Yeh S, How SW. A pathological study on the blackfoot disease in Taiwan.
Reports Inst Pathol Natl Taiwan Univ. 1963;14:25-73.
10. Chen CJ, Hsueh YM, Lai MS, et al. Increased prevalence of hypertension and
long-term arsenic exposure. Hypertension. 1995; 25:53-60.
11. Lai MS, Hsueh YM, Chen CJ, et al. Ingested inorganic arsenic and prevalence of
diabetes mellitus. Am J Epidemiol.1994; 139:484-492.
12. Tseng CH, Tai TY, Chong CK, et al. Long-term arsenic exposure and incidence
Of non-insulin-dependent diabetes mellitus: a cohort study in
arseniasis-hyperendemic villages in Taiwan. Envron Health Pospect.
2000;108:847-851
13. Tseng CH, Chong CK, Chen CJ, at al. Dose-response relationship between
peripheral disease and ingested inorganic arsenic among residents in blackfoot
disease endemic villages in Taiwan. Atherosclerosis.1996;120:125-133.
14. Chen CJ, Chiou HY, Chiang MH, et al. Dose-response relation between ischemic
heart disease and mortality and long-term arsenic exposure. Aterioscler Throm
Vasc Biol. 1996;16:504-510.
15. Chiou HY, Huang WI, Su CL, et al. Dose-response relationship between
1997;28:1717-1723.
16. Wang CH, Jeng JS, Yip PK, et al. Biological gradient between long-term arsenic
exposure and carotid atherosclerosis. Circulation. 2002;105:1804-1809.
17. Chiou HY, Chiou ST, Hsu YH, et al. Incidence of transitional cell carcinoma and
arsenic in drinking water: A follow-up study of 8,102 residents in an
arseniasis-endemic area in northeastern Taiwan. Am J Epidemiol. 2001;
153:411-418.
18. Kuo TL. Arsenic content of artesian well water in endemic area of chronic
arsenic poisoning. Rep Inst Pathol Nat Taiwan Univ. 1964;20:7-13.
19. Lo MC, Hsen YC, Lin BK. Arsenic content of underground water in Taiwan:
Second report. Taichung, Taiwan: Taiwan Provincial Institute of Environmental
Sanitation; 1977:1-17.
20. Taiwan Tap Water Corporation. Annual Report. Taipei: Taiwan Tap Water
Corporation; 2000.
21. Rosenberg HG. Systemic arterial disease and chronic arsenicism in infants. Arch
Pathol.1974;97:360-365.
22. Packer L. Metabolic and structural states of mitochondria, II: regulation by
phosphate. J Biol Chem.1961;236:214-220.
structure and function: Morphometric and biochemical evaluation of in vivo
perturbation by arsenate. Lab Invest.1979;41:313-320.
24. Chen B., Burt PL, Goering BA, et al. In vivo 32-P nuclear magnetic resonance
studies of arsenite induced changes in hepatic phosphate levels. Biochem
Biophys Res Commun. 1986;139:228-23.
25. Chin, KV, Tanaka S, Darlington I, et al. Heat shock and arsenite increase
expression of the multi-drug resistance (MDR1) gene in human renal carcinoma
cells. J Biol Chem. 1990;265:221-226.
26. Zhao CQ, Young MR, Diwan BA, et al. Association of arsenic-induced
malignant transformation with DNA hypomethylation and aberrant gene
expression. Proc Natl Acad Sci USA. 1997;94:10907-10912.
27. Clark R, Daly L, Robinson K, et al. Hyperhomocysteinemia: An independent
risk factor for vascular disease. N Eng J Med. 1991;324: 1149-55.
28. Graham IM, Daly LE, Refsum HM, et al. Plasma homocysteine as a risk factor
for vascular disease: The European Concerted Action Project. J Am Med Assoc.
1997;277:1775-1781
29. Hansen ES. Shared risk factors for cancer and atherosclerosis: A review of the
epidemiological evidence. Mutat Res. 1990;239:163-179.
arsenic-induced oxidative DDA damage in human vascular smooth cells. Circ
Res. 2000;86:514-519.
31. Germolec, DR, Spalding J, Boorman DA, et al. Arsenic can mediate skin
neoplasia by chronic stimulation of keratinocyte-derived growth factors. Mutat
Res. 1997;386:209-218.
32. Germolec, DR, Spalding J, Yu HS, et al. Arsenic enhancement of skin neoplasia
by chronic stimulation of growth factors. Am J Pathol. 1998; 153:1775-1785.
33. Kitchen KT. Recent advances in arsenic carcinogenesis: Modes of action, animal
model systems, and methylated arsenic metabolites. Toxicol Appl Pharmacol.
2001;171:249-261
34. Bunderson M, Coffin JD, Beall HD. Arsenic induces peroxynitrite generation
and cyclooxygenase-2 protein expression in aortic endothelial cells: possible role
in atherosclesosis. Toxicol Appl Pharmacol 2002;184:11-18.
35. Lee TC, Tanaka N., Lamb PW, et al. Induction of gene amplification by arsenic.
Science. 1988;241:79-81.
36. Lee MY, Bae ON, Chung SM, et al. Enhancement of platelet aggregation and
thrombus formation by arsenic in drinking water: a contributing factor to
TABLE 1. Characteristics of Two Cohorts By Traditional Risk Factors Unexposed cohort (n=23942) Arsenic-exposed cohort (n=1563)
Variable Number Mean ± SD, or %
Number Mean ± SD, or %
P value Age (in years) 23942 47.3 ± 10.0 1563 48.8 ± 11.0 <0.001
Male 12025 50.2 705 45.1 Female 11917 49.8 858 54.9 <0.001 Smoking status No 16972 71.1 1164 77.6 Past 829 3.5 49 3.1 Current 6033 25.3 350 22.4 0.022 Current Alcohol consumption
status
No 21321 89.4 1356 86.8
Yes 2542 10.7 206 13.2 0.002
Diabetes mellitus status
No 23252 97.5 1287 82.3 Yes 603 2.5 276 17.7 <0.0001 Hypertension status No 22414 94.0 1225 78.4 Yes 1437 6.0 338 21.6 <0.0001 Hyperlipidemia No 18571 78.1 562 48.5 Yes 5224 21.9 696 51.1 <0.0001 Body Mass Index, kg/m2
23865 24.0 ± 3.4 1128 24.2 ± 3.3 <0.001
TABLE 2.Age-Gender-Adjusted Relative Risks of Mortality from IHD and Stroke in Two Cohorts
Unexposed Cohort Exposed Cohort Traditional Risk Factors IHD
RR (95% CI) Stroke RR (95% CI) IHD RR (95% CI) Stroke RR (95% CI) Age* < 50 1.00 1.00 1.00 1.00 50-54.9 8.20 (3.38 - 19.9) 2.96 (1.42 - 6.15) 3.28 (0.66 - 16.2) 16.4 (1.92 - 141) 55-59.9 9.63 (4.05 - 22.9) 5.27 (2.79 – 9.98) 7.87 (1.88 - 33.0) 19.0 (2.13 - 170) > 60 17.8 (7.80 - 40.1) 13.7 (7.76 – 24.1) 11.2 (2.96 - 42.3) 63.4 (8.37 - 481) P for trend < 0.0001 < 0.0001 < 0.0001 < 0.0001 Gender¶ Female vs. Male 0.34 (0.20 - 0.60) 0.65 (0.43 - 0.99) 0.46 (0.18 - 1.18) 0.64 (0.29 - 1.40) Smoking status Current vs. Never 2.28 (1.22 - 4.23) 2.04 (1.17 - 3.57) 1.20 (0.37 - 3.92) 2.05 (0.63 - 6.66) Past vs. Never 1.93 (0.73 – 5.05) 1.80 (0.75 - 4.34) 0.80 (0.09 – 6.93) 1.64 (0.30 –9.00)
Alcohol consumption status
Yes vs. No 1.51 (0.84 - 2.69) 2.30 (1.40 - 3.78) 0.75 (0.20 - 2.74) 2.14 (0.76 - 6.01)
Diabetes mellitus status
Yes vs. No 3.11 (1.54 - 6.29) 3.05 (1.66 - 5.60) 1.95 (0.76 - 5.00) 1.46 (0.63 - 3.38) Hypertension status Yes vs. No 3.35 (1.98 - 5.67) 3.37 (2.17 - 5.25) 2.36 (0.95 - 5.85) 2.31 (1.05 - 5.08) Hyperlipidemia Yes vs. No 2.12 (1.33 - 3.40) 1.66 (1.10 - 2.50) 1.36 (0.44 - 4.17) 1.16 (0.40 - 3.40)
Body mass index
< 23 1.00 1.00 1.00 1.00
> 23 – 28 1.33 (0.76 - 2.32) 1.16 (0.74 - 1.82) 0.90 (0.29 - 2.86) 0.83 (0.28 - 2.51) > 28 3.13 (1.65 - 5.93) 1.66 (0.93 - 2.99) 0.54 (0.06 - 4.69) 0.52 (0.06 - 4.34)
P for trend 0.001 0.111 0.617 0.538
TABLE 3.Multivariate-Adjusted Relative Risks of Mortality from IHD and Stroke in Two Cohorts
Unexposed Cohort Exposed Cohort Traditional Risk Factors IHD RR (95% CI) Stroke RR (95% CI) IHD RR (95% CI) Stroke RR (95% CI)
Age in 1-year increment 1.12 (1.08 -1.16) 1.11 (1.08 - 1.15) 1.08 (1.00 - 1.15) 1.16 (1.08 - 1.24) Gender Female vs. Male 0.57 (0.28 - 1.18) 1.21 (0.66 - 2.19) 0.49 (0.11 – 2.13) 0.50 (0.12 - 2.15) Smoking status Current vs. Never 2.34 (1.24 – 4.39) 2.01 (1.12 – 3.60) 0.99 (0.18 – 5.44) 1.97 (0.43 – 8.97) Past vs. Never 1.75 (0.67 – 4.62) 1.66 (0.68 – 4.07) -- 1.54 (0.14 – 16.7)
Alcohol consumption status
Yes vs. No 1.24 (0.69 –2.24) 2.05 (1.23 - 3.41) 0.88 (0.14 – 5.34) 0.67 (0.15 – 2.96)
Diabetes mellitus status
Yes vs. No 2.32 (1.12 - 4.81) 2.39 (1.28 - 4.49) 2.81 (0.80 – 9.83) 1.39 (0.47 - 4.11) Hypertension status Yes vs. No 2.55 (1.45 - 4.47) 2.91 (1.81 - 4.70) 1.95 (0.56 –6.83) 3.73 (1.18 - 11.8) Hyperlipidemia Yes vs. No 1.76 (1.08 - 2.86) 1.48 (0.97 - 2.25) 1.28 (0.37 - 4.51) 1.10 (0.36 - 3.38)
Body mass index
< 23 1.00 1.00 1.00 1.00
> 23 – 28 1.12 (0.63 - 2.00) 0.99 (0.63 - 1.57) 0.77 (0.21 – 2.81) 0.67 (0.22 - 2.08) > 28 2.15 (1.09 - 4.23) 1.16 (0.63 - 2.13) 0.44 (0.05 – 4.21) 0.36 (0.04 - 3.09)
P for trend 0.04 0.70 0.47 0.30
Legend:
Table 1. Characteristics of both cohorts were shown by traditional risk factors.
Table 2. Relative risks of mortality from IHD and stroke respectively for various risk
factors were shown after adjustment for age and gender. Traditional risk factors were
dominated or attenuated by ingested arsenic in exposed cohort and became
insignificant.*gender-adjusted;¶ age-adjusted;
Table 3. In multi-variate analysis, relative risk of mortality from IHD and stroke
respectively in both cohorts were shown. The traditional risk factors were still
dominated or attenuated by arsenic exposure, such as current smoking, diabetes
mellitus and hyperlipidemia in exposed cohort.
Figure 1. Cardiovascular mortality rate (including IHD and stroke) in
arsenic-exposed and unexposed cohorts by age during follow-up.
Figure 2. Relative risk of cumulative cardiovascular (including IHD and stroke)
mortality after adjustment for age (1-year increment), gender, current smoking and
alcohol smoking in arsenic-exposed and unexposed cohorts categorized by arsenic
