Cardiovascular Risk in Special Populations II: Hypertension

Racial differences in blood pressure (BP) were reported as early as 1932 when a 7-mm Hg higher systolic BP was recorded in 6000 black workmen compared with 8000 white workmen in New Orleans.1 Hypertension is of course an important contributor to cardiovascular (CV) events, yet, as of recent surveys, less than one third of the US hypertensive population has adequate BP control despite the availability of a large variety of antihypertensive medications.2 In fact, hypertension prevalence is increasing, by 3.7% from 1988–1991 to 1999–2000 based on National Health and Nutrition Examination Surveys (NHANES).2 Although almost 70% of persons were aware of their hypertension in the more recent survey and almost 60% were treated (an increase of 6% from the earlier survey), hypertension was controlled in only 31% (an increase of slightly more than 6%). In terms of special populations, women, the elderly (older than 60 years), and Mexican Americans had lower rates of control than men, younger persons, and non-Hispanic whites. A more recent sampling comparison using expanded sample sizes of NHANES (1998–1994 vs 1999–2002) provided further information in regard to race-specific data on hypertension comparing blacks with whites.3 The prevalence of hypertension was highest in blacks, increasing from 36% in the earlier NHANES to 41% in the later one (and from 24% to 29% among whites)3 (Table I). However, awareness of their hypertension status was higher in blacks (78% vs 70%), with an increase of 4% for blacks since the earlier survey without a change in whites. Treatment increased in blacks from 58% to 68%. Although BP control in blacks increased from 23% to 32% since the earlier survey, it remained below whites in both surveys, with 35% of whites having BP control in the latter survey (Table II). In the 1999–2004 NHANES, the prevalence of hypertension increased with age, from 7.3% in persons 18 to 39 years, to 66.3% in persons 60 years and older.4 The BP control rate (to <140/90 mm Hg) increased from 29.2% in 1999–2000 to 36.8% in 2003–2004. This increase was significant in sexes, blacks, and Mexican Americans. In the Multi-Ethnic Study of Atherosclerosis (MESA) involving 6814 adults without clinical heart disease, the prevalence of hypertension or self-reported treatment of hypertension was significantly higher in blacks than whites (60% vs 38%)5 (Figure 1). After adjustment for age, body mass index (BMI), presence of diabetes, and smoking, black race as well as Chinese ethnicity were associated with higher prevalence of hypertension compared with white race, and socioeconomic factors were significantly associated with treated but uncontrolled hypertension in blacks compared with other racial groups. From the Multi-Ethnic Study of Atherosclerosis (MESA). Treatment and control of hypertension (HTN) among hypertensive MESA participants by race and ethnicity. HTN was defined as self-reported treatment with 1 of 6 classes of antihypertensive medications (thiazide diuretics, β-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin-2 receptor blockers, and α-blockers) or systolic blood pressure (BP) ≥140 mm Hg or diastolic BP ≥90 mm Hg. Treated and controlled HTN: whites>Hispanics (P<.01). Treated and uncontrolled HTN: blacks>whites (P<.0001); Hispanics>whites (P<.001); and Chinese>whites (P<.01). Unaware of HTN: whites>blacks (P<.0001). From Kramer et al5 with permission from The American Journal of Hypertension ©2004. In a study of adherence to guidelines, in this case the Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI) guidelines (1997), a review of medical records of 15,768 visits to 12 internal medicine clinics in 2001–2002 revealed a greater physician adherence among blacks (84%) and Hispanics (83%) than whites (78%) (P<.001).6 On the other hand, BP was controlled most among whites (39%) vs 35% for blacks and 33% for Hispanics (P<.001). However, in all groups, intensification of therapy led to better control of BP, although Hispanics were less likely than blacks and non-Hispanic whites to have BP therapy intensified. In different regions of the country there appears to be a lack of substantial geographic differences in hypertension awareness comparing the “Stroke Belt” region (southeastern United States) with the country as a whole in racial assessment.7 However, there appears to be a trend toward better treatment and control in the Stroke Belt region. The prevalence of hypertension is as high as 65% in persons older than 60 years. Increasing BMI, an epidemic in this country, was independently associated with increased rates of hypertension. Further analysis of the 2 NHANES indicated that only 2.0% to 3.6% of the increase in hypertension prevalence was due to increase in BMI.3, 4 The increase in the diagnosis of hypertension may have also been due to the lower BP indication for antihypertensive therapy in diabetics (130/85 mm Hg vs 140/90 mm Hg) and indeed exclusion of diabetics from the analysis demonstrated that the increase in prevalence was no longer significant. The latter NHANES indicated that almost 75% of all diabetics with hypertension did not have BPs controlled to <130/85 mm Hg.8 This underscores the health problem of uncontrolled hypertension in patients with diabetes in the United States, especially since blacks have a higher prevalence of diabetes than whites. Hypertension is not predominantly a condition of the elderly. Based on NHANES 1999–2002 findings, 63% of black adults and 45% of white adults with hypertension are younger than 60 years.3 However, treatment rates are significantly higher in blacks than whites (68% vs 60%), predominantly based on the higher treatment rates of black women than white women. Nonetheless, BP control remains lower in blacks than whites. Although NHANES provide important cross-sectional assessments of the awareness, prevalence, and treatment response of hypertension in the general population, they omit persons in nursing homes and other institutions, include only Mexican Americans among the Hispanic population, and restrict analysis to participants who have BP measurements taken during a specific visit and not hypertensives who do not have a BP taken during the visit.9 In addition, participants who have been told they have hypertension but have BPs below 140/90 mm Hg and were not taking medications were not counted as hypertensive.2 Presumably, a higher prevalence would be found if these factors were taken into account. Data are sparse in assessment of the factors underlying hypertension in younger and middle-aged women, especially in East Asian women. The Study of Women’s Health Across the Nation (SWAN) has evaluated ethnic variation in hypertension among premenopausal and postmenopausal women.10 This study, being conducted in community-based groups of women, evaluated 3292 women aged 42 to 52 years at baseline in various locations across the United States. Substantial BP variation was seen among ethnic groups as follows: black, 38%; Hispanic, 28%; white, 15%; Chinese, 13%; and Japanese, 11%. After multivariable adjustment for metabolic syndrome variables, lipids, fasting glucose levels, waist-hip ratio, and BMI, hypertension prevalence was 2 to 3 times higher in Hispanics and blacks compared with whites and East Asians. Among hypertensive participants, although the prevalence of antihypertensive treatment in all hypertensive participants was highest among blacks (59%) and whites (55%), control of BP to goal was highest in whites (43%) and Japanese (39%) and especially low in Hispanic women (11%). Among treated hypertensive women, goal BP (<140/<90 mm Hg) varied between 72% and 80% in whites, Chinese, and Japanese, but was only 51% in blacks and 30% in Hispanics. Different covariates were especially associated with hypertension among ethic groups of women. This included smoking among black women, BMI among Chinese women, and waist-hip ratio among Japanese women. Possible causes of differences in hypertension control among ethnic groups are cost of medications and poor awareness in patients about risks of hypertension for CV disease. As is well-known, the prevalence of hypertension increases with age. In non-Hispanic white men, for example, 7% had stage 2 hypertension (160–179/100–109 mm Hg) by JNC V criteria at age 50 to 59 years vs 13% at age 70 to 79 years. However, using Framingham risk criteria, the residual lifetime risk for hypertension in 55-year-olds was 90%, primarily due to systolic hypertension with normal diastolic BP.11 On the basis of NHANES III and IV findings (1988–1994 and 1999–2004), evaluation of participants 60 years and older showed a significant increase in the prevalence of hypertension in all elderly groups between surveys.12 For example, at ages 60 to 69 years, from 49% to 60%; at ages 70 to 79 years, from 62% to 72% years, and 80 years and older from 69% to 77%. For men it increased from 54% to 61% and for women, from 60% to 72%. In non-Hispanic whites the prevalence increased from 56% to 66% and in blacks from 71% to 82%. In those with strong risk factors, conditions in which presumably extra efforts might be added to reduce BP elevation, hypertension increased from 67% to 78% in diabetics and from 64% to 73% in those with established CV disease. Hypertension control was less likely in Mexican Americans than non-Hispanic whites as was the case in elderly patients 70 years and older compared with those 60 to 69 years. Since the surveys included only noninstitutionalized participants it is likely that the prevalence of hypertension is higher in individual groups. Reasons for lack of BP control in the elderly include less frequent BP monitoring and lack of awareness by physicians as well as patients of the definition of a healthy BP level. Another consideration is the reticence to treat the older elderly with pharmacologic therapy if the BP is in the hypertensive range.13 As Goodwin points out, the consequences of hypertension and its treatment appear to differ in very old men and women compared with younger populations.14 To support this assessment of the very old (now considered 80 years and older), a very curious finding in several European population–based studies indicated that hypertension was associated with lower 5-year mortality than normotension.15-17 Even the Framingham study demonstrated an inverse relationship between BP and mortality in persons 75 to 84 years of age.14, 18 Several explanations for these seemingly paradoxical findings are (1) increased “physiologic vigor” with higher BPs and (2) better organ perfusion with higher BPs in the very old. Along with this is the evidence for increased mortality with treatment of hypertension in the very old. As previously indicated, physician and patient attitudes can affect the degree of control of hypertension in individual patients. Although there may be some differences in interracial responses to some classes of antihypertensive agents, the aggressiveness of treatment intervention and willingness of the patient to adhere to therapeutic recommendations also have to be taken into consideration. Valuable information on interracial responses to antihypertensive interventions has been obtained from assessment of clinical trial results. This section will evaluate specific hypertension clinical trial results in relation to racial groups. Black hypertensives have been found to respond better to calcium channel blockers (CCBs) and diuretics than angiotensin-converting enzyme (ACE) inhibitors and β-blockers.19, 20 In a systematic evaluation review of 30 studies of antihypertensive drug therapy in black patients through March 2004,21 the results can be summarized as follows: (1) most of the antihypertensive drugs were more effective than placebo in reducing systolic and diastolic BP; (2) β-blockers did not differ significantly from placebo in reduction of systolic BP; (3) ACE inhibitors did not differ significantly from placebo in decreasing diastolic BP; however, only 23% of actively treated patients in this group of studies reached target BP goals; and (4) CCBs were the only drug type effective for all subgroups of black participants. It was concluded that although drugs differed in their efficacy of reducing BP, there was no significant difference in morbidity and mortality outcomes if BP reduction goals were achieved. The efficacy of agents affecting the renin-angiotensin system has generally been less pronounced in blacks than whites. For example, in patients with left ventricular (LV) dysfunction, enalapril has been shown to significantly reduce the risk of hospitalization for white patients with heart failure but not blacks.22 This is especially important in LV dysfunction patients with hypertension. However, the selective aldosterone blocker eplerenone appears to be superior in antihypertensive effects in blacks compared with the angiotensin receptor blocker (ARB) losartan.23 Eplerenone as well as losartan significantly reduced urinary albumin/creatinine ratios in blacks and in whites. A consensus statement has been published on management of high BP in blacks that uses the results of population-based studies and clinical trials to guide antihypertensive therapy.24 Most of the recommendations for risk assessment apply to all special groups. A salient feature of risk assessment in blacks emphasizes evidence for renal disease that is 3 to 4 times more likely to lead to end-stage renal disease in blacks compared with whites. Evaluation should include evidence for albuminuria with special effort to get BP targets to <130/80 mm Hg with nondiabetic renal disease and albuminuria of >1 g/d. Summaries of the use of antihypertensive agents and management issues in blacks are found in Table III and Table IV. In terms of lifestyle modification, sodium restriction appears to lead to more pronounced BP-lowering in blacks than in other groups25 (Figure 2). The greater salt sensitivity in blacks may also be associated with lower potassium intake than in whites, so that potassium intakes should also be emphasized in the absence of renal failure. From a study of 412 participants randomly assigned to eat a typical US diet or the Dietary Approaches to Stop Hypertension (DASH) diet. The effect on systolic blood pressure of dietary sodium intake during the control diet (panel A) and the DASH diet (panel B), according to subgroup. The error bars represent the 95% confidence intervals of the blood pressure changes for each subgroup. Hypertension was defined as an average blood pressure of 140 to 159 mm Hg or an average diastolic blood pressure of 90 to 95 mm Hg during the 3 screening visits. The “other” category of race and ethnic group is composed primarily of non-Hispanic whites. *P≤.05; †P≤.01; and ‡P≤.001 indicating significant differences between levels of sodium intake. From Sachs et al25 with permission from the Massachusetts Medical Society ©2001. The evaluation of pharmacologic intervention points out that all classes of antihypertensive agents will lower BP in blacks, but that the BP-lowering efficacy of thiazide diuretics and dihydropyridine CCBs were particularly effective, in comparison with ACE inhibitors, ARBs, or β-blockers. However, the majority of patients need ≥2 antihypertensive agents to bring BP to target goals and thus combinations of agents may mitigate these differences in relative efficacy. It was pointed out that with blacks as well as other groups, low doses of 2 agents yield additional decreases in BP of 8 to 11/4 to 6 mm Hg compared with the highest dose of monotherapy.26 The importance of ARBs was noted in the African-American Study of Kidney Disease and Hypertension (AASK) trial in which the ACE inhibitor ramipril reduced clinical events by 46% compared with the dihydropyridine CCB amlodipine and reduced the decline in kidney function to a significantly greater extent than amlodipine and metoprolol. ARBs have been especially useful in blacks when combined with thiazide diuretics, especially in type II diabetic nephropathy. ACE inhibitors are useful in LV dysfunction but blacks appear to be at increased risk for side effects from these agents such as cough and angioedema.27, 28 In the 2003–2004 NHANES, Mexican Americans had a similar age-adjusted prevalence of hypertension compared with non-Hispanic whites, approximately 28%. Although there was improvement in hypertension awareness, treatment and control in non-Hispanic whites and blacks between NHANES II (1976–1980) and NHANES III (1988–1991), no changes were found between a Hispanic Health and Nutrition Survey (1982–1984) and NHANES III in Mexican Americans.29 Although these disparities between non-Hispanics and Mexican Americans persisted through the 2002 survey, the survey in 2003–2004 demonstrated an improvement of awareness (from 57% to 64% now) and treatment (from 37% to 48% now); however, still lower than non-Hispanics. BP control was also found to be lowest in Mexican Americans (27%) compared with blacks and non-Hispanic whites. In the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), in which Hispanic whites comprised 16% of the total group, an interesting finding was that this racial group, although having less BP control at enrollment, had higher control at 4 years into the study (72%) than non-Hispanic whites (67%) and non-Hispanic blacks (59%) (Table V). This finding suggested that Hispanic ethnicity did not necessarily presage inferior BP control when there was equal access to medical care and medication cost was not an issue under the auspices of a clinical trial.29 In the International Verapamil SR/Trandolapril (INVEST) trial as well, the Hispanic cohort had better BP control and lower risk of adverse CV outcomes than the non-Hispanic cohort.30 The patients were randomized to sustained-release verapamil (with trandolapril) or atenolol (with hydrochlorothiazide). The increased risk for diabetes in Hispanics, due possibly to the higher prevalence of overweight and obesity compared with other groups suggests that, since diuretics and β-blockers may be more likely to produce this adverse outcome, these agents should be used cautiously in overweight Hispanic patients. CCBs and ACE inhibitors on the other hand would either have a neutral effect or actually increase insulin sensitivity. Certain explanations for the findings of NHANES in relation to Hispanic whites compared with non-Hispanic whites include: (1) less health insurance and less preventive services, and (2) lower use of antihypertensive treatment.31, 32 The importance of dietary salt restriction in BP control was studied in a group of salt-sensitive black, Hispanic, and non-Hispanic white patients with hypertension in conjunction with antihypertensive agents.33 In all racial groups, the absolute BP achieved by all racial groups was consistently lowered when salt restriction was combined with an ACE inhibitor or a CCB. In all racial groups, the impact of a high- vs a low-salt diet with the ACE inhibitor showed an enhanced effect on BP reduction compared with the CCB. Some conclusions about the impact of hypertension on various ethnic groups include the following: the incidence of hypertension, mortality from hypertensive disease, stroke, and hypertensive renal disease is higher in blacks than in other groups. Some Hispanic Americans appear to have a lower risk for hypertension but a greater risk for risk factors such as diabetes and hyperlipidemia. A similar association of mortality and income is found in blacks and non-black Hispanics. In response to antihypertensive therapy, blacks respond less favorably to ACE inhibitors and β-blockers and more favorably to diuretics and CCBs. There appears to be no such hierarchy in response in non-black Hispanics. However, ethnicity cannot predict the initial response in an individual to any class of antihypertensive agent. Primarily comparing blacks and Hispanics with non-Hispanic whites, several trials showed different clinical outcomes associated with antihypertensive drug therapy.70 For example, ALLHAT found a greater magnitude of benefit of diuretics for blacks vs non-blacks and in the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial, ARBs were less effective than β-blockers. Several trials exclusively in Asians have demonstrated CCBs to be effective in preventing CV outcomes.34 Whatever the ethnic minority group, hypertension is not under adequate control in the United States at present. Since older persons are a rapidly growing component of the American population, on top of an increased prevalence of diabetes and obesity, the potential impact of hypertension on the population becomes even more significant. Although inroads have been made within all racial groups on awareness and treatment of hypertension, efforts at prevention take on increasing importance for in CV morbidity and For hypertension this efforts to control salt and increase This can only be within a that the at risk to lifestyle preventive may better than for BP to increase with a pharmacologic It also