Uric Acid: A Risk Factor

Transcription

1 Uric Acid: A Risk Factor for Coronary Heart Disease? VICTORIA W. PERSKY, M.D., ALAN R. DYER, PH.D., ELIZABETH IDRIS-SOVEN, M.A., JEREMIAH STAMLER, M.D., RICHARD B. SHEKELLE, PH.D., JAMES A. SCHOENBERGER, M.D., DAVID M. BERKSON, M.D., AND HOWARD A. LINDBERG, M.D. SUMMARY The association between serum uric acid and the prevalence of ECG abnormalities was analyzed for 24,997 employed men and women, white and black, age years, from the Chicago Heart Association (CHA) Detection Project in Industry. In addition, the relationships between uric acid and 5-year mortality from all causes, from cardiovascular diseases (CVD), and from coronary heart disease (CHD) were analyzed for 7804 white men and women age years from this study and 967 white men age years from the Chicago People's Gas Company Study. For men, the associations between uric acid and the prevalence of ECG abnormalities and with mortality appear to be secondary to associations between uric acid and other risk factors. For women, however, the associations could not be explained by other risk factors. A POSSIBLE ASSOCIATION of serum uric acid with risk of coronary heart disease (CHD) has been noted in several studies.' 7 Other reports have either not confirmed these findings8-10 or noted that the association disappears when controlling for relative weight or use of diuretics."'8 The importance of determining if uric acid is an independent risk factor for CHD derives in part from the hyperuricemic effects of diuretics and their widespread use in the treatment of hypertension."' An independent contribution of uric acid to CHD risk would necessitate re-evaluation of the benefit/risk ratio of the use of diuretics in hypertension, especially of the common less severe grades. In this study we analyzed the association between serum uric acid and the prevalence of ECG abnormalities for 24,997 men and women, white and black, age years, from the Chicago Heart Association (CHA) Detection Project in Industry. In addition, the relationships of baseline serum uric acid and mortality from all causes, from cardiovascular diseases (CVD) and from CHD are assessed for 7804 white men and women originally age years from this same study, and for 967 white males age years in 1962 from the Chicago People's Gas Company study. Materials and Methods From the fall of 1967 until early 1973, the Chicago From the Department of Community Health and Preventive Medicine, Northwestern University Medical School, Rush- Presbyterian St. Luke's Medical Center, and the Chicago Heart Association, Chicago, Illinois. Supported by the American Heart Association, Chicago Heart Association, the Illinois Regional Medical Program, and the NHLBI, NIH, USPHS. This research was done while Dr. Dyer was an Established Investigator of the American Heart Association. Address for reprints: Victoria Persky, M.D., Morton Building, Room 1-693, 303 East Chicago Avenue, Chicago, Illinois Received July 20, 1978; revision accepted December 7, Circulation 59, No. 5, Heart Association Detection Project in Industry screened 39,665 young adult and middle-aged men and women, both black and white, employed in 85 firms in the Chicago area. The volunteer rate for this screening effort was 55%. In all facilities, all employees were encouraged to participate regardless of the type of job or shift worked. At each site of employment, standardized tests were administered by one of two specially trained four-person teams of nurses and technicians. They included a selfadministered questionnaire to collect demographic data and information on chronic diseases; measurements of height and weight; a single casual supine blood pressure reading (standard mercury sphygmomanometer, fifth phase diastolic pressure); venipuncture for determination of serum cholesterol, uric acid and plasma glucose 1 hour after a 50 g oral load; and an ECG. Relative weight was computed as the ratio of observed body weight to desirable weight for height and sex, based on 1959 actuarial tables from the Metropolitan Life Insurance Company.20' 21 Methods of chemical analyses have been described previously.22 Heart rate was determined from the ECG. Electrocardiographic abnormalities were divided into major and minor by criteria of the Hypertension Detection and Follow-up Program (HDFP), which are a slight modification of the criteria of the National Cooperative Pooling Project.23 A detailed description of criteria for major ECG abnormalities is given in the Appendix. Of the group screened, 37,604 were age years. The group of 24,997 men and women with complete baseline data on all relevant variables and not hypertensitives on drug treatment are examined in this report for possible associations at baseline between serum uric acid and electrocardiographic abnormalities. For this study, 7804 white men and women age years, free of definite CHD at baseline and followed for an average of 5 years, were examined for possible associations between serum uric acid and mortality from all causes, CVD and CHD. Cause of death was determined from the death certificate and classified according to the eighth revision of the 969

2 970 CIRCULATION VOL 59, No 5, MAY 1979 TABLE 1. Correlations Between Serum Uric Acid and Other Baseline Variables by Age, Sex and Race-Chicago Heart Association Detection Project in Industry Correlation coefficients No. Age Heart Relative cigarettes/ Plasma Group (years) No. Age SBP DBP Cholesterol rate weight day glucose White men White men White men White women White women White women Black men Black men Black men Black women Black women Black women Abbreviations: SBP = systolic blood pressure; DBP = diastolic blood pressure. ICDA. CVD included codes and ; epidemiology and etiology of the adult CVD in males CHD, codes , 25 age years on January 1, Of the 1594 male The Chicago People's Gas Company study is a employees in this age range on that date, 1465 (91.9%) longitudinal investigation of the natural history, underwent a complete physical examination in TABLE 2. Baseline Serum Uric Acid Level and Percent with Electrocardiographic Abnormalities, Major and All, by Age-Sex-Race, Chicago Heart Association Detection Project in Industry Men White men Serum Age years Age years Age years uric Abnormality Abnormality Abnormality acid Major Any Major Any Major Any (mg/dl) No. (%) (%) No. (%) (%) No. (%) (%) < > All x2 NA t 12.1 Women White women Serum Age years Age years Age years uric Abnormality Abnormality Abnormality acid Major Any Major Any Major Any (mg/dl) No. (%) (%) No. (%) (%0) No. (%) (%) < > All x t 20.50t *Numbers of black men years were too small for this analysis. tp <0.05. Abbreviation: NA = not appropriate because there were less than five cases in at least one uric acid group.

3 URIC ACID AND CHD/Persky et al. 971 Serum uric acid was not measured in this study until 1962, when 1106 men from the original cohort of 1465 were reexamined. Details of this examination have been published elsewhere.26 Of these men, 967 were white, age years and free of definite CHD in Data on treatment for hypertension were not available on these men, who have been followed for 13 years. The associations between serum uric acid and mortality from all causes, CVD, CHD and stroke were examined. Cause of death was determined primarily from the death certificate, with classification by a combination of the seventh and eighth revisions of the ICDA. Autopsy and hospital reports were also used when available. The data are analyzed using both univariate and multivariate techniques. The univariate techniques include cross-classification with chi-square analysis when appropriate, and t tests of the difference in uric acid means between those with and without ECG abnormalities and between those who survived and those who died. The multivariate analysis was based on multiple linear regression. In all instances the statistical software package developed by the Statistical Package for the Social Sciences was used.27 Results Serum Uric Acid and Prevalence of Electrocardiographic Abnormalities: Chicago Heart Association Detection Project in Industry The correlations of serum uric acid with other baseline variables for each age-sex-race group of the CHA cohort are shown in table 1. Uric acid correlates most strongly with relative weight. Correlations with diastolic and systolic blood pressure are somewhat lower. Associations are less strong with heart rate, age, glucose and cholesterol. The prevalence rates of electrocardiographic abnormalities for both major and any abnormalities, by uric acid level for each age-sex-race group, are shown in table 2. In several of the groups, the abnormality rate is significantly higher for the group with the highest uric acid, suggesting a threshold effect. Based on the chi-square test, the differences reach statistical significance (p < 0.05) for major abnormalities in white men age years, for all abnormalities in black women age years, and for major and all abnormalities in white women years. The mean serum uric acid of those persons with and TABLE 2. (Continued) Men Black men* Age years Age years Abnormality Abnormality Major Any Major Any No. (%) (%) No. (%) (%) NA NA 3.25 Women Black women Age years Age years Age years Abnormality Abnormality Abnormality Major Any Major Any Major Any No. (%) (%) No. (%) (%) No. (%) (%) NA NA NA 13.05t NA NA

4 972 CIRCULATION VOL 59, No 5, MAY 1979 TABLE 3. Mean Uric Acid in Persons with and without Baseline Electrocardiographic Abnormalities by Age- Sex-Race, Chicago Heart Association Detection Project in Industry Any abnormality Number Mean uric acid t Adjusted Group Age ECG abnormalities ECG abnormalities Adjusted for all Race/Sex (years) None Any None Any tt for age variables White men White men White men * 2.52* 1.27 White women White women t 3.44t 3.35t White women t 3.67t 2.21* Black men Black men Black men Black women Black women t 3.16t 2.43* Black women * Major abnormality ti Number Mean uric acid t Adjusted Group Age ECG abnormalities ECG abnormalities Adjusted for all Race/Sex (years) None Major None Major tt for age variables White men White men White men * 2.88* 1.81 White women White women * 2.72* 3.08* White women t 3.45t 2.32* Black men ** Black men Black men Black women * Black women Black women *p <0.05 tp < It values for the difference in mean uric acid between those with and without abnormalities. t values of serum uric acid coefficients in multiple linear regression controlling for age. t values of uric acid coefficients in multiple linear regression controlling for age, heart rate, relative weight, systolic blood pressure, plasma glucose, serum cholesterol, and number of cigarettes smoked per day. **Too few numbers for multivariate analysis. without ECG abnormalities, along with a two-sample t test of significance of the difference between the means within each age-sex-race group, are shown in table 3. The uric acid means of white men age years and of white women age years and years with any ECG abnormality are significantly higher than the means of men and women without abnormalities. Similarly, the uric acid means of white men age years and of white women age years and years with major ECG abnormalities are significantly higher than the uric acid means of those without major ECG abnormalities. The uric acid means of black women age years and years with any abnormalities are significantly higher than the uric acid means of black women without abnormalities. There is no significant difference, however, in uric acid means of black women with and without major abnormalities. The associations between uric acid and any ECG abnormalities remained significant after adjusting for age in a bivariate linear regression analysis in white men age years, white women age years and years, and black women age years. The significant difference in black women age years noted on univariate analysis was no longer significant, although it was nearly so. Uric acid differences remained significant only in white women age years and years and black women age years after adjustment for differences in baseline heart rate, relative weight, systolic blood pressure, number of cigarettes smoked per day, plasma glucose, serum cholesterol, and age using multiple linear regression. The differences in white men were no longer significant after adjustment for these other variables. The associations between uric acid and major ECG abnormalities were significant adjusting -for age in white men age years, white women age years, and white women age years. These differences remained significant adjusting for other variables only in white women age 25-44

5 URIC ACID AND CHD/Persky et al. 973 TABLE 4. Mortality by Baseline Serum Uric Acid, White Men and Women Age Years, Chicago Heart Association Detection Project in Industry, Followed an Average of 5 Years Men Serum Including hypertensives on treatment Excluding hypertensives on treatment uric All-cause death CVD death CHD death All-cause death CVD death CHD death acid Total Rate Rate Rate Total Rate Rate Rate (mg/dl) no. No. (per 1000) No. (per 1000) No. (per 1000) no. No. (per 1000) No. (per 1000) No. (per 1000) < > All Women Serum Including hypertensives on treatment Excluding hypertensives on treatment uric All-cause death CVD death CHD death All-cause death CVD death CHD death acid Total Rate Rate Rate Total Rate Rate Rate (mg/dl) no. No. (per 1000) No. (per 1000) No. (per 1000) no. No. (per 1000) No. (per 1000) No. (per 1000) < > All Abbreviations: CVD = cardiovascular disease; CHD = coronary heart disease. years and years. Uric acid differences in black women age years became significant after adjustment for all variables with major abnormalities as the end point. Baseline Serum Uric Acid and 5-year Mortality: Chicago Heart Association Detection Project in Industry Five-year mortality rates by level of serum uric acid at baseline from all causes, CVD and CHD for white men age years and for white women age years are shown in table 4. The numbers of total and cause-specific deaths in the other age-sex-race groups are too small for a meaningful examination of the questions considered here, e.g., there were no CHD deaths in any of the groups of black women, in black men younger than 45 years or in white men and women younger than 25 years. In addition, none of the other groups had more than six CHD deaths. There is a tendency in white men age years for all-cause mortality to be higher in the highest uric acid group. This tendency persists when hypertensives on treatment are excluded. Correspondingly, CVD and CHD death rates are also higher in men in the highest uric acid groups. This tendency does not persist as strongly when hypertensives on treatment are excluded. In white women age years the highest death rate from all causes occurred in the group with uric acid Cause-specific mortality data in white women age years are shown for completeness. The numbers, however, are too small for meaningful comparison. Mean serum uric acid levels by sex and cause of death are shown in table 5. T tests of significance are computed for the differences in means between those who died and those who were still alive at the time of analysis. T tests, adjusted for age differences, and for differences in age, serum cholesterol, plasma glucose, relative weight, heart rate, systolic blood pressure and number of cigarettes smoked per day are also given. Mean uric acid levels of men who died from all causes, CHD and CVD are significantly higher than mean uric acid values of men who survived. These differences persist for death from all causes, but not for death from CVD or CHD when controlled for differences in the other variables. These differences do not persist if hypertensives on treatment are excluded.

6 974 CI RCULATION VOL 59, No 5, MAY 1979 TABLE 5. Mean Serum Uric Acid by Vital Status and Cause of Death: White Men and Women A ge Years Followed an Average of 5 Years, Chicago Heart Association Detection Project in Industry Men Including hypertensives on treatment Excluding hypertensives on treatment t Adjusted t Adjusted Mean Adjusted for all Mean Adjusted for all Cause of death Number uric acid tt for age variables Number uric acid tt for age variables Alive All-cause death * 2.34* 1.98* CVD death * 2.53* CHD death * 2.46* Women Includinig hypertensives on treatment Excluding hypertensives on treatment ti ~~~~~~~~~~~~~~~~t1 t Adjusted t Adjusted Mean Adjusted for all Mean Adjusted for all Cause of death Number uric acid tt for age variables Number uric acid tt for age variables Alive All-cause death t 3.05* 2.61* t 3.26t 2.92* CVD death ** ** ** ** CHD death ** ** ** ** *p <0.05. tp < tt values for the difference in mean uric acid between those who are still alive and those who have died. t values of the uric acid coefficients in multiple linear regression controlling for age. t values of the uric acid coefficients in multiple linear regression controlling for age, heart rate, relative weight, systolic blood pressure, serum cholesterol, plasma glucose, and number of cigarettes smoked per day. **Number of deaths too small for the multivariate analysis to be appropriate. Abbreviations: CVD = cardiovascular disease; CHD = coronary heart disease. The mean serum uric acid of women who died from any cause is significantly higher than the mean serum uric acid of women who survived. This difference persists when adjusted for all other variables and when hypertensives on treatment are excluded from the analysis. For these women, the uric acids of those who died from CVD or CHD tended to be higher than the uric acids of those who survived, but given the small numbers, the differences are not statistically significant. Baseline Serum Uric Acid and 13-Year Mortality: Chicago People's Gas Company Study Thirteen-year mortality rates by baseline serum uric acid of white males age years from the Chicago People's Gas Company Study are shown in table 6. There are no consistent trends in either total or causespecific mortality, although the all-cause, CVD and CHD death rates are higher in those with uric acid. 8 mg/dl. Mean serum uric acid by vital status and cause of death, with t tests of significance of the difference between the means of those who died and those who survived 13 years, are shown in table 7. Also noted are the t tests of significance for uric acid differences adjusted for age, and for age, pulse, serum cholesterol, relative weight, systolic blood pressure and number of cigarettes smoked per day. There are no significant differences at the 0.05 level in a two-tailed test, although death from all causes and stroke approach significance (p < 0.10) controlling for other variables. Discussion The possible association of elevated serum uric acid with coronary heart disease has been controversial. In 1951 Gertler et al. reported higher mean serum uric acid levels in young men with CHD compared with healthy controls.' Several other case-control studies with small numbers of persons confirmed these findings.' '7 Larger epidemiologic studies, however, have yielded conflicting results. An Israeli prevalence study found no association of uric acid with CHD on univariate analysis.9 However, an Australian prevalence study did note an association of uric acid with CHD in men, but not in women.4 This association persisted when other variables were controlled, but diuretic usage was not mentioned. Similarly, incidence data from Honolulu showed significant associations of baseline uric acid with CHD in 2 years of follow-up, which persisted on multivariate analysis.6 This study also failed to control for diuretic use. In contrast, in several other studies, apparent associations of uric acid with CHD were secondary to associations with other risk factors. In the Framingham study, significant associations in women were noted in some age groups between baseline uric acid and 16-year incidence of CVD." Except for an in-

7 URIC ACID AND CHD/Persky et al. 975 TABLE 6. Thirteen-Year Mortality by Baseline Serum Uric Acid: White Men Age Years, Free of Coronary Heart Disease in 1962, Chicago People's Gas Company Study All causes death CVD death CHD death Stroke death Uric acid Rate Rate Rate Rate (mg/dl) No. No. (per 1000) No. (per 1000) No. (per 1000) No. (per 1000) < > All Abbreviations: CVD cardiovascular disease; CHD = coronary heart disease. creased incidence of myocardial infarction in men age years with elevated uric acid, no associations with uric acid were noted in men. None of the associations persisted in multivariate analyses.12 In the Tecumseh Study, no association was found between uric acid and CHD prevalence with age and weight controlled.'3 In the Evans County study, a significant association was noted between uric acid and CHD prevalence for white women and black men. These associations were not significant with blood pressure and cardiovascular medication controlled." Four recent studies have found that the apparent association of uric acid with CHD may be secondary to use of diuretics. Prevalence data from the Honolulu study excluding men on diuretic therapy showed no correlation between uric acid and CHD on multivariate analysis.'8 Isomaki in Finland found higher uric acid levels in patients hospitalized with CHD than in healthy rural persons; this difference was not present when men on diuretic medication were excluded.15 Similarly, Bengtsson and Tibblin from Goteborg noted that the higher uric acid levels in women age years with "coronary" ECGs were secondary to diuretic usage.'6 Finally, the Coronary Drug Project Research Group, studying men age years with prior evidence of myocardial infarction, noted that uric acid was a predictor of long-term mortality only if diuretic usage was not controlled.'7 In this study we have tried to determine if uric acid is an independent risk factor for the atherosclerotic diseases, i.e., after control for other risk factors including diuretic usage. The results in men are consistent with most,"1'8 but not all,4 6 other epidemiological studies. The data from the Chicago People's Gas Company study presented here indicate no statistically significant associations between baseline serum uric acid and 13-year mortality from all causes, CVD or CHD, although the association with all causes and stroke are nearly so. Prevalence data from the CHA Detection Project in Industry showed no association between uric acid and electrocardiographic abnormalities in black men. Data from the CHA study in white men age years did show an association between uric acid and prevalence of electrocardiographic abnormalities, and between uric acid and death from all causes, CVD, and CHD. However, these latter associations did not persist after control for use of diuretics or for other baseline variables. TABLE 7. Mean Serum Uric Acid by Cause of Death: White Men Years Followed 13 Years, Chicago People's Gas Company Study t Vital status and Mean serum tj Adjusted for cause of death Number uric acid tt Adjusted for age all variables Alive All causes * 1.67* CVD CHD Stroke * 1.84* 1.65* *0.05 < p < tt values for the difference in mean uric acid between those who are still alive and those who have died. tt values of the uric acid coefficients in multiple linear regression controlling for age. t values of the uric acid coefficients in multiple linear regression controlling for age, pulse, cholesterol, systolic blood pressure, relative weight and number of cigarettes smoked per day. Abbreviations: CVD = cardiovascular disease; CHD = coronary heart disease.

8 976 CIRCULATION VOL 59, No 5, MAY 1979 There is thus little evidence in these Chicago studies to support the hypothesis that in white men age years uric acid is an independent risk factor for coronary artery disease. Prevalence data in white and black women and mortality data in white women from the CHA Detection Project in Industry show an association of uric acid with prevalence of ECG abnormalities and with all-cause mortality. This possible association of -uric acid with all-cause mortality in white women is unexplained. It may be secondary to low-order associations with CHD, CVD or cancer, which do not reach statistical significance because of small numbers. It does not appear, in contrast to previous literature,19 to be secondary to diuretic usage or to associations with other measured variables such as relative weight, glucose, cholesterol, blood pressure, age, heart rate or smoking. Findings from this study do not support the hypothesis that uric acid is an independent risk factor for CHD in white men years. Associations of uric acid with prevalence of electrocardiographic abnormalities and with incidence of CHD in white men years appear to be secondary to associations with other variables such as diuretic usage, blood pressure and relative weight. In white women the positive associations between uric acid and ECG abnormalities, and in white women years the positive association between uric acid and death from all causes, suggest that uric acid may be an independent risk factor for CHD in women. These associations did not appear to be secondary to other measured variables and are unexplained. Further investigations are needed to clarify these findings. Acknowledgments It is a pleasure to pay tribute to the officers, executive leadership, medical department and research staff of the People's Gas Company for their continuous support and cooperation, to the many Chicago companies and organizations that gave valuable cooperation in this effort, to those involved in the Chicago Heart Association Detection Project in Industry, to the Heart Disease Detection Committee of the Chicago Heart Association, and to the leadership and staffs of the Chicago Health Department and the Chicago Health Research Foundation. The biochemical analyses for the studies were done in the research laboratories of the Chicago Health Department and the Chicago Health Research Foundation, under the direction of Morton B. Epstein, Ph.D. and Howard Adler, Ph.D.; we also acknowledge the fine contribution of the two chief technicians, Dana King and Ika Tomaschewsky. It is a pleasure to express appreciation to Ron Hoeksema and Dan Garside, who assisted in programming, and to Margie Shores, who typed the manuscript. Appendix Hypertension Detection and Follow-up Program (HDFP): Criteria for Major ECG Abnormalities Standardized by the Minnesota Code 1. Q-QS ST Depression T Inversion Complete atrioventricular (AV) block AV block Wolff- Parkinson-White Artifical pacemaker Complete left bundle branch block Complete right bundle branch block (RBBB) Interventricular block Left anterior hemiblock and intermittent RBBB or complete RBBB Frequent ectopic ventricular beats 8.12, 8.13, Ventricular arrhythmia Atrial fibrillation or atrial flutter Supraventricular tachycardia Sinoatrial arrest or block AV dissociation ST elevation 9.2 References 1. Gertler MM, Garn SM, Levine SA: Serum uric acid in relation to age and physique in health and in coronary heart disease. Ann Intern Med 34: 1421, Kohn PM, Prozan GB: Hyperuricemia-relationship to hypercholesterolemia and acute myocardial infarction. JAMA 170: 1909, Benedik TG: Correlations of serum uric acid and lipid concentrations in normal, gouty, and atherosclerotic men. Ann Intern Med 66: 851, Welborn TA, Cumpston GN, Cullen KJ, Curnow DH, McCall MG, Stenhouse NS: The prevalence of coronary heart disease and associated factors in an Australian rural community. Am J Epidemiol 89: 521, Jacobs D: Hyperuricemia and myocardial infarction. S Afr Med J 46: 367, Kagan A, Gordon T, Rhoads G, Schiffman JC: Some factors related to coronary heart disease incidence in Honolulu Japanese men: the Honolulu Heart Study. Int J Epidemiol 4: 271, Shoshkes M: Systolic hypertension, hyperuricemia, and hyperglycemia as risk factors in cardiovascular disease. J Med Soc NJ 73: 219, Higgins ITT, Higgins MW, Lochshin MD, Canale N: Coronary disease in mining communities in Marion County, West Virginia. J Chron Dis 22: 165, Medalie JH, Kahn HA, Neufeld HN, Riss E, Goldbourt U: Five-year myocardial infarction incidence - II. Association of single variables to age and birthplace. J Chron Dis 26: 329, Allard C: Serum uric acid: not a discriminator of coronary heart disease in men and women. Can Med Assoc J 109: 986, Shurtleff D: Some characteristics related to the incidence of cardiovascular disease and death: Framingham Study, 16-year followup. Section 26. In The Framingham Study - an Epidemiological Investigation of Cardiovascular Diseases, edited by Kannel WB, Gordon T. U.S. Department of Health, Education and Welfare, Washington, D.C., Gordon T, Sorlie P, Kannel WB: Coronary heart disease, atherothrombotic brain infarction, intermittent claudication - a multivariate analysis of some factors related to their incidence: Framingham Study, 16-year followup. Section 27. In The Framingham Study - an Epidemiological Investigation of Cardiovascular Diseases, edited by Kannel WB, Gordon T. U.S. Department of Health, Education and Welfare, Washington, D.C., Myers AP, Epstein FH, Dodge HJ, Mikkelsen WM: The relationship of serum uric acid to risk factors in coronary heart disease. Am J Med 45: 520, Klein R, Klein BE, Coroni JD, Maready J, Cassel JC, Tyroler HA: Serum uric acid: its relationship to coronary heart disease risk factors and cardiovascular disease, Evans County, Georgia. Arch Intern Med 132: 401, Isomaki H: Hyperuricemia in Northern Finland, an epidemiologic study of serum uric acid in rural, urban, and hospital populations. Ann Clin Res 1 (suppl 1): 1, Bengtsson C, Tibblin E: Serum uric acid levels in women. Acta Med Scand 196: 93, 1974

9 STATIC-DYNAMIC VS DYNAMIC EFFORT/DeBusk et al The Coronary Drug Project Research Group: Serum uric acid: its association with other risk factors and with mortality in coronary heart disease. J Chron Dis 29: 557, Yano K, Rhoads GC, Kagan AK: Epidemiology of serum uric acid among 8,000 Japanese-American men in Hawaii. J Chron Dis 30: 171, Veterans Administration Cooperative Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension III. Influence of age, diastolic blood pressure, and prior cardiovascular disease; further analysis of side effects. Circulation 45: 991, Build and Blood Pressure Study, 1959, Vol I, Chicago, Society of Actuaries, New weight standards for men and women. Statist Bull Metrop Life Insur Co 40: 1, Stamler J, Rhomberg P, Schoenberger JA, Shekelle RB, Dyer A, Shekelle S, Stamler R, Wannamaker J: Multivariate analysis of the relationship of seven variables to blood pressure: findings of the Chicago Heart Association Detection Project in Industry, J Chron Dis 28: 527, The Pooling Project Research Group: Relationship of blood pressure, serum cholesterol, smoking habit, relative weight, and ECG abnormalities to incidence of major coronary events: Final report of the Pooling Project. J Chron Dis 31: 201, US Department of Health, Education, and Welfare, National Center for Health Statistics: Eighth Revision International Classification of Diseases, adapted for Use in the United States (ICDA), PHS Publ No 1693, Washington, DC, Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death. Geneva, World Health Organization, Stamler J, Lindberg HA, Berkson DM, Shaffer A, Miller W, Poindexter A: Prevalence and incidence of coronary heart disease in strata of the labor force of a Chicago industrial corporation. J Chron Dis 11: 405, Nie NH, Hull CH, Jenkins JG, Steinbreimer K, Bent DH: Statistical Package for the Social Sciences, 2nd ed. New York, McGraw-Hill Book Company, 1975 Comparison of Cardiovascular Responses to Static-Dynamic Effort and Dynamic Effort Alone in Patients with Chronic Ischemic Heart Disease ROBERT DEBUSK, M.D., WILLIAM PITTS, B.S., WILLIAM HASKELL, PH.D., AND NANCY HOUSTON, R.N. SUMMARY Thirty men, mean age 55 years, known to have treadmill-induced ischemic ST-segment depression, performed static and dynamic effort, i.e., forearm lifting and treadmill exercise, separately and combined. Static effort was sustained at 20%, 25% or 30% of maximal forearm lifting capacity. Two symptomlimited treadmill tests, one with and one without added static effort, were performed on each of two visits. Compared with dynamic effort alone, combined static-dynamic effort decreased treadmill work load and increased heart rate, systolic blood pressure and rate-pressure product at the onset of ischemic ST-segment depression or angina pectoris: 7.1 ± 0.4 vs 8.0 ± 0.5 (SEM) multiples of resting oxygen consumption (mets), estimated; 141 ± 3 vs 134 ± 3 beats/min; 170 ± 4 vs. 162 ± 4 mm Hg and 239 ± 8 vs 218 ± 9 (p < 0.001). The prevalence of angina pectoris was significantly less with combined static-dynamic effort than with dynamic effort alone. Static effort causes a resetting of the threshold at which ischemic abnormalities appear during dynamic effort. COMBINED STATIC-DYNAMIC EFFORT is often considered hazardous to patients with chronic ischemic heart disease." 2 This belief is primarily based on a study of normal persons in whom systolic blood pressure and, by inference, myocardial oxygen consumption increased disproportionately during combined static-dynamic effort compared with submaximal dynamic effort alone.' When these two modes of effort have been directly compared in patients with chronic ischemic heart disease, the ac- From the Division of Cardiology, Department of Medicine, Stanford University School of Medicine, Palo Alto, California. Supported in part by grant HL from the NHLBI. Address for reprints: Robert F. DeBusk, M.D., Director, Stanford Cardiac Rehabilitation Program, 730 Welch Road, Palo Alto, California Received July 20, 1978; revision accepted November 10, Circulation 59, No. 5, tual prevalence of ischemic abnormalities has been found to be lower with combined static-dynamic effort than with dynamic effort alone.3' 4 These studies have reported the heart rate-systolic blood pressure product, or "double product," only during peak staticdynamic effort3 1 4or during a fixed submaximal work load,3 but not at the onset of ischemic ST-segment depression or angina pectoris. Consequently, we do not know whether the double product at which ischemic abnormalities appear with dynamic effort is the same as that at which ischemic abnormalities appear with combined static-dynamic effort. Moreover, a systematic difference between the double product at the onset of ischemic abnormalities during these two modes of effort may imply a significant difference in mechanisms underlying myocardial ischemia. We had three aims in this study: 1) to devise a simple, standardized and clinically useful method for