Pre-HYpertension in the Very Elderly Trial (HYVET)
In 1963, the Veteran's Administration (VA) commenced the first double-blind, placebo-controlled, multicentre trial to determine the value, if any, of treating non-malignant hypertension. Reporting its results in 1967, from a group of 143 male hypertensive patients with diastolic blood pressures averaging between 115 and 129 mmHg, it had assigned them to either active (hydrochlorothiazide plus reserpine and hydralazine) treatment or placebo. With twenty-seven severe hypertension-related complications and four mortalities arising in the placebo arm, compared with two complications and no deaths in the active group, the authors had demonstrated the benefits of anti-hypertensive therapy in non-malignant hypertension . However, with a mean age of 51 years, a new question would emerge – was this benefit, seen with the treatment of severe diastolic hypertension, applicable to all age groups with all forms of hypertension?
In 1979 the Hypertension Detection and Follow-up Program (HDFP) published its results, having elected to recruit patients between the ages of 30 and 69 years. Comparing stepped care (antihypertensive therapy in specialist centres, increased stepwise to achieve and maintain reduction of BP to or below set goals) with referred care (treatment via their usual source), the investigators found a 17% reduction in 5 year all-cause mortality between the two arms (6.4 vs. 7.7 per 100; P < 0.01). However, the mean age of the patients was again 50.8 years . Sub-group analysis by age (n = 2376) did not demonstrate any significant difference in effect amongst the older cohort (aged 60–69 years). Stepped care was associated with a 5 year all-cause mortality of 12.7%, compared with 15.2% amongst those receiving referred care .
In a specific effort to determine the value of treating hypertension in older adults, the European Working Party on High Blood Pressure in the Elderly (EWPHE) recruited 840 men and women over the age of 60 years with both systolic and diastolic hypertension (systolic BP 150–239 mmHg, diastolic BP 90–119 mmHg). A double-blind, placebo-controlled trial was commenced with active treatment consisting of 25 mg hydrochlorothiazide and 50 mg triamterene, which could be doubled as required. With an average baseline blood pressure of 183/101 mmHg, a 21/10 mmHg difference in blood pressure was observed between the two trial arms at 5 years, in favour of active treatment. Active treatment was also associated with reduced cardiovascular mortality (−38%; P = 0.023). This comprised a 38% reduction (P = 0.036) in cardiac mortality and a non-significant decrease in cerebrovascular mortality (−32%; P = 0.16). However this reduction in cardiac mortality was associated with increased rates of non-fatal MI amongst those receiving active treatment. This contrasted with the data in terms of cerebrovascular disease. Whilst a non-significant decrease in cerebrovascular mortality was observed, non-terminal cerebrovascular events were significantly reduced (−52%; P = 0.026). Nonetheless, with an age range of 72 ± 8 years, the trialists had demonstrated that treatment of severe systolic and diastolic hypertension amongst older adults had significant cardiac mortality benefit .
In 1984 the Systolic Hypertension in the Elderly Program (SHEP) was commenced in the United States. Rather than extending the age boundary, this trial sought to evaluate whether treating older patients (≥60 years, mean age 71.6 years) with grade 2 isolated systolic hypertension (systolic blood pressure >160 mmHg; diastolic blood pressure <90 mmHg) would be beneficial, harmful or make no difference. Again a randomized, double-blind, placebo-controlled trial, the study cohort consisted of almost 5000 subjects, who were commenced on either a placebo or active therapy (initially utilizing the diuretic chlorthalidone, with atenolol 25 mg or reserpine 0.05 mg available as adjuncts). With a mean blood pressure of 170/76 mmHg in both trial arms at baseline, the 5 year average blood pressure was 143/68 mmHg amongst those who were actively treated. A decline in blood pressure was also observed amongst the placebo group, largely due to a ‘drop in’ effect, whereby patients assigned to placebo were started on open label active therapy by their personal physicians. Amongst this cohort the 5 year average blood pressure was 155/72 mmHg .
With regard to the primary endpoint of stroke (non-fatal or fatal), a 36% reduction was observed amongst those assigned active therapy (P = 0.0003), the absolute benefit of treatment estimated at 5 years being 30 events per 1000 participants. In addition to this, active therapy was associated with a reduction in risk for coronary heart disease (RR 0.75; 95% CI 0.60, 0.94), congestive heart failure (RR 0.36; 95% CI 0.60, 0.94) and cardiovascular disease (RR 0.68; 95% CI 0.58, 0.79); a non-statistically significant trend in favour of reduced all cause mortality was also observed. However, these benefits were associated with significant adverse effects, as described later .
A year after the results from SHEP were published, the Medical Research Council (MRC) published data from its placebo-controlled, single-blind trial of hypertension treatment in older adults. Having recruited 4396 UK-based subjects, aged between 65 and 74 years, the MRC working party randomized them to one of four trial arms; (i) a potassium sparing diuretic regimen (hydrocholorothiazide 25 mg or 50 mg plus amiloride 2.5 mg or 5 mg daily), (ii) matching placebo tablets, c) the β-adrenoreceptor antagonist, atenolol, 50 mg daily or d) a matched placebo. Inclusion and exclusion criteria meant that subjects had mean systolic pressures of 160–209 mmHg and mean diastolic pressures of <115 mmHg during an 8 week run in, whilst not taking anti-hypertensive treatment .
The primary endpoints were defined as all cause mortality, mortality and morbidity due to stroke and mortality and morbidity due to coronary heart disease. The secondary aims were to compare the effects of the two active drugs and to determine whether the response to treatment differed on the basis of gender. When compared with the placebo group, those receiving active therapy (in a combined analysis) exhibited a 25% (95% CI 0.03, 0.42; P = 0.04) reduction in stroke, a 19% (95% CI −0.02, 0.36; P = 0.08) reduction in coronary events and a 3% (95% CI −0.14, 0.18; P = not available) reduction in all cause mortality. Following adjustment for baseline characteristics, the diuretic group had significantly reduced risks of stroke (RR 0.31; 95% CI 0.03, 0.51; P = 0.04), coronary events (RR 0.44; 95% CI 0.21, 0.60; P = 0.0009) and all cardiovascular events (RR 0.35; 95% CI 0.17, 0.49; P = 0.0005). However, those receiving the β-adrenoceptor antagonist, atenolol, showed no reduction in these endpoints .
Meanwhile in Europe a similar trial to SHEP, the Systolic Hypertension in Europe (Syst-Eur) trial, was established. This enrolled nearly 5000 older men and women (≥60 years, mean age: 70.2 years) again with grade 2 isolated systolic hypertension (average sitting systolic blood pressure 160–219 mmHg, diastolic blood pressure <95 mmHg). A double-blind, placebo-controlled trial, Syst-Eur utilized a long acting dihydropyridine calcium antagonist, nitrendipine, as initial therapy (with enalapril and hydrochlorthiazide available as additional agents, if required). At cessation of follow up, the difference in systolic and diastolic blood pressures between the placebo group and those actively treated was 10.7 mmHg (95% CI 8.8, 12.5) and 4·7 mmHg (95% CI 3·7, 5·6) respectively. As with SHEP, the combined primary endpoint for Syst-Eur was fatal and non-fatal stroke. This occurred in 77 patients on placebo and 47 on active therapy (RR 0.42; 95% CI 0.60, 0.17; P < 0.003) and based on this interim analysis, the study was terminated prematurely. In addition to the significant reduction in stroke risk, a favourable reduction in all cause cardiac endpoints (fatal and non-fatal heart failure, fatal and non-fatal myocardial infarction, and sudden death) was observed (P = 0.03) .
In contrast to the adverse effect data from SHEP, which demonstrated increased self-reporting of impaired memory/concentration, the Syst-Eur trialists ran a sub-study which reported that antihypertensive treatment for older adults experiencing isolated systolic hypertension, was associated with a lower incidence of dementia . This sub-study demonstrated a 50% reduction in dementia rates from 7.7/1000 patient years to 3.7/1000 patient years, when using an intention to treat analysis . Although the relative risk reduction was 50%, the absolute risk was low with only 19 cases of dementia prevented per 5000 patient years of active treatment, a fact which has led many critics to argue that the data should be interpreted with caution [24, 25]. Moreover, whilst the authors attributed the observed effect to the use of dihydropyridine calcium antagonists, whether this is truly a drug/class specific effect is unclear [26, 27].
Given the lack of data for the treatment of hypertension amongst octogenarians, the HYpertension in the Very Elderly Trial (HYVET) was commissioned in 1994. An initial open label pilot recruited 1283 patients over the age of 80 years from 10 different European countries. Subjects with a sustained blood pressure of 160–210/90–109 mmHg (based on an average of four readings) were allocated to one of three treatment arms, a diuretic based regimen, an angiotensin converting enzyme inhibitor based regimen or no active treatment. Active treatment was associated with a reduction in cerebrovascular events with a relative hazard rate (RHR) of 0.47 (95% CI 0.24, 0.93). However, there was a non-significant rise in all cause (RHR 1.23; 95% CI 0.75, 2.01) and stroke related mortality (RHR 0.57; 95% CI 0.25, 1.32), such that for each stroke prevented, there was the potential for one excess death .
In contrast, the HYVET study was a double-blind placebo-controlled trial performed in 195 centres in 13 countries. All patients had a sustained systolic blood pressure of 160–210 mmHg, off treatment, during a 2 month run in period. Whilst the initial protocol also required a mean seated diastolic blood pressure of 90 to 109 mmHg, a protocol amendment 3 years into the trial, relaxed this to <110 mmHg, thus allowing for the inclusion of those with isolated systolic hypertension. Subjects were either assigned to the non-thiazide, sulphonamide diuretic, indapamide (sustained release, 1.5 mg) or a placebo. If required, investigators were able to add perindopril (2 mg or 4 mg) or a matching placebo to reach the target systolic blood pressure of <150 mmHg and the target diastolic blood pressure of <80 mmHg .
After the second interim analysis, the trial was stopped early. Active treatment decreased BP when compared with placebo (−15 mmHg/−6 mmHg) and this was associated with a non-significant reduction in the primary outcome measure, stroke (unadjusted hazard ratio 0.70; 95% CI 0.49, 1.01; P = 0.06). Cardiovascular morbidity and mortality were also non-significantly reduced (unadjusted hazard ratio 0.77; 95% CI 0.60, 1.01; P = 0.06), whilst a statistically significant reduction in congestive cardiac failure was observed (unadjusted hazard ratio 0.36; 95% CI 0.22, 0.58; P < 0.001). However, integral to the decision to stop early was an unexpected significant reduction in the incidence of all-cause mortality (unadjusted hazard ratio 0.79; 95% CI 0.65, 0.95; P = 0.02), with active treatment being well tolerated (see Adverse effects) .
These results provide definitive evidence that anti-hypertensive therapy in those with sustained hypertension, over the age of 80 years (mean participant age 83 years), is associated with benefit, prolonging life as opposed to increasing mortality. Whilst the reduction in fatal strokes in HYVET was greater than that observed in previous studies performed using younger subjects, the relative risk reduction in total mortality was consistent with the 95% CIs observed in other studies of hypertension in older adults, namely the Swedish Trial in Older Patients with Hypertension, the Systolic Hypertension in the Elderly Program and the Systolic Hypertension in Europe trial [21, 23, 30].
Nonetheless, the limitations of such a trial should not be overlooked. HYVET was reliant on patients from Eastern Europe and China, with potentially different pathophysiological mechanisms responsible for the mediation of hypertension in these populations as opposed to those from Western Europe . It also focused on subjects in relatively good physical and mental condition, in common with many other clinical trials in older people, in an effort to limit drop-outs, rendering the trial population less representative of the real life setting . Furthermore, the majority of patients included in the trial had systolic/diastolic hypertension (67.5%) as opposed to isolated systolic hypertension (32.5%), suggesting that the benefits observed may be a function of a particular sub-type of hypertension . In addition to this, the early evidence of mortality benefit resulted in a relatively short duration of follow up (median 1.8 years) . Thus, it remains unclear whether such benefits persist or diminish over a longer time course.
In 2010, a secondary analysis of the 2003 INternational VErapamil SR-Trandolapril STudy (INVEST) was published, focusing on the relationship between blood pressure and adverse outcomes in older adults. The 22 576 clinically stable patients with both hypertension and coronary artery disease were stratified by age in 10 year increments (≥80, 70–80, 60–70, <60 years). Patients were then randomized to either a calcium antagonist (verapamil SR) or non-calcium antagonist (atenolol) based treatment strategy (with the option for additional trandolapril, with or without hydrochlorothiazide, when necessary to achieve blood pressure goals). BP targets were defined as having been achieved when a mean of two sitting clinic blood pressures were <140/90 mmHg (or <135/85 mmHg when diabetes mellitus or renal impairment were present) .
At baseline, increasing age was associated with higher systolic blood pressure, lower diastolic blood pressure and wider pulse pressure (P < 0.001). As expected, treatment was associated with a reduction in all three of these measures and those over the age of 80 years experienced primary outcome measures most frequently (first occurrence of all-cause death, non-fatal myocardial infarction or non-fatal stroke). The adjusted hazard ratio for these outcomes demonstrated a J-shaped relationship with both systolic and diastolic blood pressures amongst all age groups (Figure 2), regardless of treatment strategy. With advancing age the hazard ratio nadir increased from a systolic blood pressure of 110 mmHg for patients under 60 years of age, to 140 mmHg for the very old. However, the hazard ratio nadir relative to diastolic pressures was only marginally lower for those ≥80 years (70 mmHg) when compared with their younger counterparts (75 mmHg) – data which might yet inform a future debate about target blood pressure for older adults .