Relationship between Hematological Parameters and Incidence of Ischemic Heart Diseases among Japanese White-Collar Male Workers

Transcription

1 J Occup Health 2001; 43: Journal of Occupational Health Relationship between Hematological Parameters and Incidence of Ischemic Heart Diseases among Japanese White-Collar Male Workers Toshio KOBAYASHI 1, Yuji MIYOSHI 2, Kazue YAMAOKA 3 and Eiji YANO 3 1 Department of Hygiene and Preventive Medicine, Fukushima Medical University School of Medicine, 2 Health Insurance Society, Meiji Life Insurance Company and 3 Department of Hygiene and Public Health, Teikyo University School of Medicine Abstract: Relationship between Hematological Parameters and Incidence of Ischemic Heart Diseases among Japanese White-Collar Male Workers: Toshio KOBAYASHI, et al. Department of Hygiene and Preventive Medicine, Fukushima Medical University School of Medicine The determination whether high hematocrit (Hct) and hemoglobin (Hb) are risk factors for ischemic heart diseases (IHD) remains controversial in Japan. To examine whether hematological parameters are independent risk factors for IHD, we conducted a case control study among 3,924 Japanese male white-collar workers aged yr. IHD cases were defined as men with a first episode of acute myocardial infarction or angina pectoris during the investigation period (from April 1995 to April 2000). Five controls without a history of IHD were assigned for each case, matching for age and smoking status. Twenty-two IHD cases were developed during the study period and 110 controls were assigned. Hematological parameters, obtained from periodic health examinations undergone before the onset of the disease, were analyzed with the conditional logistic regression model. Multivariate adjusted odds ratios (OR) with 95% confidence interval (95%CI) were estimated, controlling for Body Mass Index, total cholesterol, alcohol use, hypertension and fasting blood sugar. Crude OR (95%CI) for Hct, Hb, red blood cell count (RBC) and Mean corpuscular volume (MCV) were 1.21 ( ), 2.04 ( ), 1.09 ( ) and 1.09 ( ), respectively. The multivariate adjusted OR (95%CI) for Hct, Hb, RBC and MCV were 1.28 ( ), 2.75 ( ), 1.08 ( ) and 1.09 ( ), respectively. These results suggest that Hct and Hb are independent risk Received Sept 25, 2000; Accepted Jan 5, 2001 Correspondence to: T. Kobayashi, Department of Hygiene and Preventive Medicine, Fukushima Medical University School of Medicine, Hikarigaoka 1, Fukushima , Japan factors for ischemic heart diseases among Japanese middle-aged male white-collar workers. (J Occup Health 2001; 43: 85 89) Key words: Hematocrit, Hemoglobin, Ischemic heart disease, Cardiovascular risk factor, Japanese male worker The periodic health examination of workers in Japan traditionally includes hematological parameters as an essential component. First instituted because of the poor nutritional status of Japanese workers, hematological parameters were also expected to detect diseases caused by hazardous substances such as lead 1). But the general nutritional status of Japanese workers has improved significantly, and the National Nutritional Survey 2) revealed a higher intake of fat and protein, suggesting that malnutrition is no longer a prevalent problem for current Japanese workers. Additionally, specific examinations for hazardous substances such as lead have been developed 3) and introduced into routine examinations for workers handling toxic substances, so that there is little rationale for continuing to include hematological parameters in the periodic health examinations of Japanese workers. The systematic 4, 5) reviews of Canadian and US clinical preventive services also concluded that there is insufficient evidence to recommend for or against routine testing for anemia in asymptomatic persons, with the exception of pregnant women and infants. Simply following the reports from North America, it seems also reasonable to exclude routine screening for anemia in Japan. However, in general, once a system is abolished it is not easy to reintroduce, even if it is found to have some other benefit. Although hematological parameters may not be useful for the detection and treatment of anemia among workers, they can still be useful in predicting increased

2 86 J Occup Health, Vol. 43, 2001 cardiovascular risk. In fact, increased hematocrit (Hct) and hemoglobin (Hb) are considered to be possible risk factors for ischemic heart diseases (IHD) 6 12), but several studies have not supported this 13 17). The conflicting results may be explained by differences in ethnicity 13 16, 18) and sex 17). Therefore, it is worthwhile to further examine the validity of the hematological parameters as independent risk factors for IHD in Japan, where the relationships between IHD and hematological parameters have rarely been examined. If an association between high Hct or Hb and the risk of IHD exists, it may be explained either as the result of an increase in other well-known risk factors such as high cholesterol, hypertension and obesity, or as an independent effect of hematological parameters per se. The aim of the present study was to examine, by a matched case-control study among male white-collar workers, the relationship between hematological parameters and the incidence of acute myocardial infarction or angina pectoris with adjustment for other cardiovascular risk factors. Methods Subjects and study design The study subjects were 3,924 male white-collar workers aged yr working in a life insurance company in Tokyo. The cases were subjects with a first episode of acute myocardial infarction (AMI) or angina pectoris (AP) during the study period (from April 1995 to April 2000). They were assigned according to the medical certificates which were submitted to the health insurance society of this company. Diagnoses of AMI and AP were made when electrocardiographic changes and/or coronary angiographic findings were matched with typical clinical symptoms (chest discomfort and/or pain). Five male controls without a past history of IHD or abnormal ECG finding were randomly selected for each case from a pool of workers who participated in periodic health examinations in the study period. Controls were matched for age and smoking status (non smoker, 1 9 cigarettes per day, cigarettes per day, 20 cigarettes per day), because smoking is strongly correlated with polycythemia through increased blood carboxyhemoglobin levels 19, 20). The diagnoses were confirmed and agreed upon by two physicians. Subjects with recurrence of IHD, obstructive or restrictive pulmonary diseases and those with other anemic diseases were excluded from the case and control group. Hematological parameters Subjects underwent a periodic health examination at least once during the study period. The data related to hematological parameters of Hct, Hb, number of red blood cells (RBC) and mean corpuscular volume (MCV) were obtained from the last periodic health examination before the onset of IHD. Blood was collected from the antecubital vein. Hematological parameters were measured in an automatic hematological analyzer (NE- 8000, Toa Medical Co.), which measures Hct, Hb and RBC by the erythrocyte pulse height detection method, oxyhemoglobin method and sheath flow DC detection method, respectively. MCV was computed from the Hct and RBC values. The maximum coefficients of variation of each parameter were 1.5%, 1.0%, 1.5% and 1.5% for Hct, Hb, RBC and MCV, respectively. Other cardiovascular risk factors Among cardiovascular risk factors, Body Mass Index (BMI), smoking status, alcohol use, systolic and diastolic blood pressure (SBP and DBP), antihypertensive medication, total cholesterol (T-chol) and fasting blood sugar (FBS) were recorded. BMI and T-chol were treated as continuous variables. Alcohol use was classified into three categorical groups (never drink, 0 to 25 g per day, 25 g per day) and blood pressure was classified as normal blood pressure (SBP<140 mmhg and DBP<90 mmhg and no antihypertensive drug use) vs. hypertension (SBP 140 mmhg or DBP 90 mmhg or antihypertensive drug use). Since the distribution of FBS was skewed, FBS was classified into four categorical variables (<90, 90 to 96.5, 96.5 to 109, and 110 mg/dl) according to the quartile. Statistical analysis Crude and multivariate adjusted odds ratio (OR) and 95% confidence intervals (95%CI) based on maximum likelihood estimates were calculated with a conditional logistic regression model (SAS PHREG Procedure). BMI, T-chol, alcohol use, hypertension and FBS were included in the multivariate adjusted model. The stepwise variable selection method (inclusion and exclusion criteria: 10%) was used. All P values are two-tailed. Results Twenty-two IHD cases (14 for AMI and 8 for AP) were diagnosed during the investigation period and 110 controls were assigned. Mean ages of the cases and the controls were both 50.8 yr, with a range of yr. The proportion of current smoking both in the cases and the controls was 86.4%. The mean values for the hematological parameters and other cardiovascular risk factors in the cases and the controls are shown in Table 1. The hematological values for the cases were higher than those for the controls. The other established risk factors for IHD were higher in the cases than in the controls. Table 1 shows the crude odds ratio (OR) and 95%CI for the hematological parameters and cardiovascular risk factors. The relationship between IHD and hematological parameters was significant as evidenced by OR for Hct (1.21, 95%CI ) and

3 Toshio KOBAYASHI, et al.: Hematological Parameters and Ischemic Heart Diseases 87 Table 1. Mean values for the hematological parameters and cardiovascular risk factors in the IHD cases and controls with their unadjusted odds ratio (OR) and 95% confidence interval (95%CI). Cases (n=22) Controls (n=110) Mean (SD) Mean (SD) OR* 95%CI Hematocrit (%) 48.6 (3.7) 46.6 (3.5) Hemoglobin (g/dl) 15.9 (1.0) 15.3 (1.1) Red blood cell count ( 10 4 /mm 3 ) (42.1) (42.7) Mean corpuscular volume (µm 3 ) 98.8 (4.2) 97.3 (4.7) Systolic blood pressure (mmhg) (24.2) (15.3) Diastolic blood pressure (mmhg) 87.2 (12.8) 79.7 (10.3) Total cholesterol (mg/dl) (31.0) (34.8) Body mass index (kg/m 2 ) 25.3 (4.2) 23.8 (2.8) Hypertension** 54.5% (n=12) 26.4% (n=29) Alcohol use*** never drink 31.8% (n=7) 11.8% (n=13) to 25g per day 18.2% (n=4) 36.4% (n=40) g per day 50.0% (n=11) 51.8% (n=57) Fasting blood sugar*** <90 mg/dl 13.6% (n=3) 21.8% (n=24) to 96.5 mg/dl 18.2% (n=4) 30.9% (n=34) to 109 mg/dl 31.8% (n=7) 30.0% (n=33) mg/dl 36.4% (n=8) 17.3% (n=19) *Unit values for continuous variables are 1 for hematocrit, hemoglobin, mean corpuscular volume, body mass index, and 10 for red blood cell count, blood pressures, total cholesterol. **For hypertension (diagnosed when subjects met at least one of the following: SBP 140 mmhg; DBP 90 mmhg; antihypertensive drug use), percentage (%) of subjects with hypertension and its number (n) are shown. ***For alcohol use and fasting blood sugar, percentage (%) of subjects and its number (n) in each category are shown. Table 2. Multivariate-adjusted* OR in relation to ischemic heart diseases OR** 95%CI Hematocrit (%) Hemoglobin (g/dl) Red blood cell count ( 10 4 /mm 3 ) Mean corpuscular volume (µm 3 ) *Adjusted for Body Mass Index (BMI), total cholesterol (Tchol), alcohol use, hypertension and fasting blood sugar (FBS) by the stepwise method (inclusion and exclusion criteria: 10%). Ht: adjusted for FBS, Hb: adjusted for FBS and alcohol use, RBC: adjusted for FBS, hypertension and BMI, MCV: adjusted for FBS, hypertension and BMI. **Unit values 1 for hematocrit, hemoglobin, and mean corpuscular volume, and 10 for red blood cell count. Hb (2.04, 95%CI ), but not significant for RBC (1.09, 95%CI ) or MCV (1.09, 95%CI ). Among other cardiovascular risk factors, SBP (OR 1.46, 95%CI ), DBP (OR 1.83, 95%CI ), T-chol (OR 1.18, 95%CI ), BMI (OR 1.20, 95%CI ), hypertension (OR 2.99, 95%CI ) and FBS 110 mg/dl (OR 2.93, 95%CI ) showed significantly positive relationships with IHD. Multivariate-adjusted OR for hematological parameters after adjusting for BMI, T-chol, hypertension, alcohol use and FBS are shown in Table 2. Even after adjustment, Hct (1.28, 95%CI ) and Hb (2.75, 95%CI ) were still significantly related to IHD. Discussion In order to examine the relationship between hematological parameters and the onset of IHD, a matched case-control study was conducted on Japanese male white-collar workers. In the present study, positive relationships between hematological parameters such as Hct, Hb and the onset of IHD were observed. The results of the multivariate analysis suggest that these hematological parameters may act as independent risk factors for IHD in Japanese male white-collar workers aged yr. In this study measurements of hematological parameters were carried out as part of the last periodic health examination before the occurrence of IHD so that the results of the health examination were thus not biased by the presence of disease itself. Enhanced reliability in the diagnosis of IHD was maintained by the use of medical records judged by two physicians. Although many studies have been conducted on the

4 88 J Occup Health, Vol. 43, 2001 relationship between Hct or Hb and the occurrence of IHD 6 12), it seems to remain controversial 13 17). A significant relationship between high Hct and morbidity of coronary heart diseases after 8 years follow-up was found among urban, but not rural men in Puerto Rico 13). In the Framingham study 17), high Hct was recognized as an independent risk factor for occurrence and death from coronary heart disease after a 34-yr follow-up in women, but not in men, aged yr. In the 10-yr follow up study of males of Japanese ancestry aged yr in Honolulu 14, 15), no independent contribution of Hct to IHD risk was found. On the other hand, in the occupational health care setting in the Netherlands 11), a case-control study similar to the present study was conducted and likewise revealed that high Hct or Hb was independently associated with a high incidence of myocardial infarction in men aged yr. As for the Japanese population, studies on the relationship between Hct or Hb and IHD are very limited. As far as we know, only one populationbased study was conducted more than 15 yr ago 18), and no relationship was observed. The average Hct of the participants was 39.8%, which was far below the current levels for Japanese. In the present study, cases and controls combined had an average Hct of 46.9%, and also the average Hct of all the male workers who received health examinations in 1997 (n=3,405) was 46.8%. In the study of males of Japanese ancestry in Honolulu 14, 15), in which Hct was not identified as an independent risk factor for IHD, the mean Hct value in the cohort was 44.7%, which was lower than that in our study. Recent research 8) has indicated that there might be a threshold effect in the relationship between Hct and coronary events. The risk of major coronary events was significantly increased at an Hct level exceeding 46.0% compared to those with Hct less than 46.0%, even after adjustments for other risk factors 8). Among the subjects in the present study, Hct exceeded 46.0% in 19 out of 22 cases (86.4%) compared to 63 out of 110 controls (57.3%). In a report from the Netherlands 11), in which Hct was shown to be an independent IHD risk factor, the mean Hct value among the cases was also greater than 46.0%. Several hypotheses concerning the mechanisms for the association of IHD with high Hct or Hb value exist. High Hct value is associated with an increase in blood viscosity, which may lead to vascular occlusion 6, 21, 22). It has also been reported that the Hct level and the severity of the coronary artery stenosis are positively associated in IHD patients 22). Another report indicated that cerebral blood flow was inversely related to RBC or Hct 23). In addition, recent studies have suggested that high iron intake or high stored body iron might be related to an increased risk of coronary heart disease 24, 25). Apart from the high viscosity associated with high Hct, a high iron level per se may play a role in the pathogenesis of IHD through metabolic effects of iron such as free radical production 26, 27). There is much evidence to support the relationship between Hct or Hb and other well-known cardiovascular risk factors such as smoking, hypertension and high T- chol 28 33). For example, Gaisböck 34) found a positive relation between stress polycythemia and hypertension. Our recent report on Japanese male office workers also showed a positive relationship between Hct and blood pressure 35). Hypoxia caused by increased carboxyhemoglobin in the circulating blood of smokers is believed to induce compensatory polycythemia 19, 20). Even after matching for smoking and age in the present study, both crude and multivariate adjusted OR for Hct and Hb were significant. Although the effects of smoking might not be completely adjusted by matching, these results suggest that increased hematological parameters are independent risk factors for IHD. In Japan, the hematological test is a mandatory part of the workplace periodic health examination. Although the prevalence of anemia has decreased, the above-normal range values in the hematological test can be utilized to identify subjects at high-risk for IHD. When hematological examination remains on the list of mandatory health check-up items, the target population is to be selected and specified. Because the prevalent anemia group among workers used to be young females whereas those with high risk for cardiovascular diseases are middle and older aged males, the hematological examination should be performed in older workers. References 1) Baker EL Jr, Landrigan PJ, Barbour AG, Cox DH, Folland DS, Ligo RN, Throckmorton J. Occupational lead poisoning in the United States: clinical and biochemical findings related to blood lead levels. Br J Ind Med 1979; 36: ) Ministry of Health and Welfare, Summary of the National Surveillance Report of Nutrition Status in Eiyougaku Zasshi 1998; 56: (in Japanese). 3) WHO Study Group, Early detection of health impairment in occupational exposure to health hazards. WHO Technical Report Series 1975; ) Canadian Task Force on the Periodic health examination, The periodic health examination. Can Med Assoc J 1979; 121: ) Report of the U.S. Preventive Services Task Force, Guide to Clinical Preventive Services ) Elwood PC, Waters WE, Benjamin IT, Sweetnam PM. Mortality and anemia in women. Lancet. 1974; 1: ) Murphy JF, O Riordan J, Newcombe RG, Coles EC, Pearson JF. Relation of hemoglobin levels in first and second trimesters to outcome of pregnancy. Lancet 1986; 1: ) Wannamethee G, Shaper AG, Whincup PH. Ischaemic heart disease: association with haematocrit in the

5 Toshio KOBAYASHI, et al.: Hematological Parameters and Ischemic Heart Diseases 89 British Regional Heart Study. J Epidemiol Community Health 1994; 48: ) Erikssen G, Thaulow E, Sandvik L, Stormorken H, Erikssen J. Haematocrit: a predictor of cardiovascular mortality? J Intern Med 1993; 234: ) Goubali A, Voukiklaris G, Kritsikis S, Viliotou F, Stamatis D. Relation of hematocrit values to coronary heart disease, arterial hypertension, and respiratory impairment in occupational and population groups of the Athens area. Angiology 1995; 46: ) Knottnerus JA, Swaen GM, Slangen JJ, Volovics A, Durinck J. Haematologic parameters as risk factors for cardiac infarction, in an occupational health care setting. J Clin Epidemiol 1988; 41: ) Wannamethee G, Shaper AG, Macfarlane PW, Walker M. Risk factors for sudden cardiac death in middleaged British men. Circulation 1995; 91: ) Sorlie PD, Garcia-Palmieri MR, Costas R Jr, Havlik RJ. Hematocrit and risk of coronary heart disease: the Puerto Rico Health Program. Am Heart J 1981; 101: ) Carter C, McGee D, Reed D, Yano K, Stemmermann G. Hematocrit and the risk of coronary heart disease: the Honolulu Heart Program. Am Heart J 1983; 105: ) Yano K, Reed DM, McGee DL. Ten-year incidence of coronary heart disease in the Honolulu Heart Program. Relationship to biologic and lifestyle characteristics. Am J Epidemiol 1984; 119: ) Abu-Zeid HA, Chapman JM. The relation between hemoglobin level and the risk for ischemic heart disease: a prospective study. J Chronic Dis 1976; 29: ) Gagnon DR, Zhang TJ, Brand FN, Kannel WB. Hematocrit and the risk of cardiovascular disease-the Framingham study: a 34-year follow-up. Am Heart J 1994; 127: ) Szatrowski TP, Peterson AV Jr, Shimizu Y, et al. Serum cholesterol, other risk factors, and cardiovascular disease in a Japanese cohort. J Chronic Dis 1984; 37: ) Sagone AL Jr, Balcerzak SP. Smoking as a cause of erythrocytosis. Ann Intern Med 1975; 82: ) Smith JR, Landaw SA. Smokers polycythemia. N Engl J Med 1978; 298: ) Burch GE, DePasquale NP. Hematocrit, viscosity and coronary blood flow. Dis Chest 1965; 48: ) Lowe GD, Drummond MM, Lorimer AR, et al. Relation between extent of coronary artery disease and blood viscosity. Br Med J 1980; 280: ) Thomas DJ, du Boulay GH, Marshall J, et al. Cerebral blood-flow in polycythaemia. Lancet 1977; 2: ) Salonen JT, Nyyssonen K, Korpela H, Tuomilehto J, Seppanen R, Salonen R. High stored iron levels are associated with excess risk of myocardial infarction in eastern Finnish men. Circulation 1992; 86: ) Tzonou A, Lagiou P, Trichopoulou A, Tsoutsos V, Trichopoulos D. Dietary iron and coronary heart disease risk: a study from Greece. Am J Epidemiol 1998; 147: ) Bernier M, Hearse DJ, Manning AS. Reperfusioninduced arrhythmias and oxygen-derived free radicals. Studies with anti-free radical interventions and a free radical-generating system in the isolated perfused rat heart. Circ Res 1986; 58: ) Farber NE, Vercellotti GM, Jacob HS, Pieper GM, Gross GJ. Evidence for a role of iron-catalyzed oxidants in functional and metabolic stunning in the canine heart. Circ Res 1988; 63: ) Dintenfass L. Elevation of blood viscosity, aggregation of red cells, haematocrit values and fibrinogen levels with cigarette smokers. Med J Aust 1975; 1: ) Shimakawa T, Bild DE. Relationship between hemoglobin and cardiovascular risk factors in young adults. J Clin Epidemiol 1993; 46: ) Wannamethee G, Shaper AG. Haematocrit: relationships with blood lipids, blood pressure and other cardiovascular risk factors. Thromb Haemost 1994; 72: ) Elwood PC, Sweetnam P, Welsby E, Mahler R, Moore F. Association between circulating haemoglobin level, serum-cholesterol, and blood-pressure. Lancet 1970; 1: ) Bottiger LE, Carlson LA. Relation between serum cholesterol and triglyceride concentration and haemoglobin values in non-anaemic healthy persons. Br Med J 1972; 3: ) Smith S, Julius S, Jamerson K, Amerena J, Schork N. Hematocrit levels and physiologic factors in relationship to cardiovascular risk in Tecumseh, Michigan. J Hypertens 1994; 12: ) Gaisböck F. Die Bedeutung der Blutdruck messung fur die Praxis. Dtsch Arch Klin Med 1905; 83: ) Nishikido N, Kobayashi T, Kashiwazaki H. Hematocrit correlates with blood pressure in young male office workers. Ind Health 1999; 37: