Pulse Pressure and Effects of Losartan or Atenolol in Patients With Hypertension and Left Ventricular Hypertrophy
In the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study, the primary composite end point of cardiovascular death, stroke, and myocardial infarction was reduced by losartan versus atenolol in patients with hypertension and left ventricular hypertrophy. The objective of this post hoc analysis was to determine the influence of pulse pressure on outcome. Patients were divided into quartiles of baseline pulse pressure. Cox regression, including baseline Framingham risk score as a covariate, was used to compare risk in the quartiles. In the atenolol group, there were significantly higher risks in the highest versus lowest quartile for the composite end point 28% (confidence interval [CI], 2% to 62%; P=0.035), stroke 84% (CI, 32% to 157%; P<0.001), and total mortality 41% (CI, 7% to 84%; P=0.013). Risk for myocardial infarction was 44% higher (CI, −5% to 120%; P=0.089). The risks in the losartan group also increased with increasing quartile, but were lower than in the atenolol group, and differences between the highest and lowest quartiles were not significant: composite end point 12% (CI, −13% to 44%; P>0.2), stroke −5% (CI, −34% to 37%; P>0.2), myocardial infarction 30% (CI, −13% to 94%; P>0.2), and total mortality 32% (CI, −1% to 76%; P=0.062). In patients with hypertension and left ventricular hypertrophy in the LIFE study, there were significantly higher risks, adjusted for the Framingham risk score, for the primary composite end point, stroke, and total mortality in the highest versus lowest quartile of pulse pressure with atenolol-based treatment. The risks in the losartan group also increased with increasing pulse pressure quartile, but were lower than those in the atenolol group, and were not significant.
Pulse pressure (PP) is a powerful predictor of cardiovascular risk,1–7 particularly in older patients,5–7 and in several studies was a better predictor than systolic blood pressure (SBP) and diastolic blood pressure (DBP).3–7 The prognostic importance of PP increases with age.5–7
The present analysis concerns the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study, which showed 13% reduction in the primary composite end point (cardiovascular death, stroke, and myocardial infarction) and 25% reduction in stroke in patients with hypertension and left ventricular hypertrophy treated with losartan-based compared with atenolol-based therapy.8 Arterial blood pressure was similarly reduced in the 2 treatment groups throughout the study and, therefore, differences in achieved blood pressure did not explain the differences in outcomes between treatments.
In a prespecified subanalysis of LIFE of 1326 patients with isolated systolic hypertension (baseline SBP 160 to 200 mm Hg and DBP <90 mm Hg), and therefore elevated PP, patients treated with losartan had 25% reduction in the primary composite end point and 40% reduction in stroke compared with patients treated with atenolol.9 SBP, DBP, and mean blood pressure were similarly reduced in both groups throughout the study, and changes in blood pressures could not explain differences in outcomes between treatments.
In the present analysis, we examined the impact of baseline PP and change in PP on the benefits of losartan versus atenolol in the overall LIFE population.
The LIFE study8 was a double-blind, randomized, parallel group study. The main outcome8 and the study design, organization, clinical measures, end point definitions, exclusion criteria, reasons for choice of comparative agents, statistical power calculations, and baseline characteristics have been published.8–11 Patients gave written informed consent, and the study protocol was approved by relevant ethics committees. An independent data and safety board monitored the study.
Patients and Treatment
The LIFE study randomized 9193 patients aged 55 to 80 (mean, 66.9) years with hypertension (either treated or untreated) and signs of left ventricular hypertrophy on ECG (according to Cornell voltage duration or Sokolow-Lyon voltage criteria) to losartan-based versus atenolol-based treatment after 1 to 2 weeks of placebo if they had sitting blood pressure of 160 to 200/95 to 115 (mean, 174.4/97.8) mm Hg. We followed-up patients for at least 4 years with regular visits. We recorded sitting blood pressure 24 (range, 22 to 26) hours after study drug dose and permitted uptitration of study drugs with the addition of hydrochlorothiazide and other agents (excluding β-blockers, angiotensin-converting enzyme inhibitors, or angiotensin-II receptor antagonists) if blood pressure was >140/90 mm Hg during follow-up. PP was calculated as SBP −DBP.
The primary end point was a composite of the first occurrence of cardiovascular death, stroke, or myocardial infarction. Other prespecified outcomes were total mortality, angina pectoris or heart failure requiring hospitalization, resuscitated cardiac arrest, and new-onset diabetes. The present analysis examined the primary composite end point and its components, hospitalization for heart failure, and total mortality. End points were adjudicated by an independent end point committee, which was blinded to therapy.
For the present analysis, patients were divided into the actual quartiles of baseline PP as follows: <67, 67 to 76, 77 to 86, and ≥87 mm Hg. We assessed all end points by intention-to-treat analysis. We assessed the difference between quartiles with respect to end points with hazard ratios adjusted for the Framingham risk score12 derived from a Cox proportional hazards model. The model contained 4 terms, which were 3 indicator variables representing the second, third, and fourth quartiles of baseline PP, and a term representing the Framingham risk score. The hazard ratios for the second, third, and fourth quartiles represent risks relative to the first quartile. We calculated the risk as 100 × (1−hazard ratio). For each end point, an unadjusted analysis was performed excluding the Framingham risk score to corroborate the adjusted analysis. The atenolol and losartan treatment groups were analyzed separately. However, a global test of treatment effect was performed across the 4 quartiles simultaneously to determine whether there was a treatment effect in at least one quartile of baseline PP at a 0.10 level of significance. A model with blood pressure throughout the trial as a time-varying covariate adjusted for treatment effect was analyzed as well. To determine whether the treatment effect varied with increasing achieved reduction in blood pressure, the interaction of achieved reduction in SBP, DBP, and PP with treatment was investigated. Achieved reduction in blood pressure was defined as the change from baseline to the last blood pressure before an end point. Baseline characteristics were tested across PP quartiles with ANOVA for continuous variables and the χ2 test for categorical variables. For all tests, the level of 2-sided significance was 5%, except for interaction tests, which used 10%.
The treatment groups were well-matched with regard to baseline characteristics.8 There were differences in baseline characteristics between the groups defined by PP quartiles (Table 1). Patients in the lowest quartile of baseline PP (<67 mm Hg, n=2291) compared with those in the highest PP quartile (≥87 mm Hg, n=2390) were younger (mean±standard deviation age 63.9±6.7 versus 69.9±6.4 years) and fewer had diabetes (9% versus 17%). More patients were female in the highest PP quartile than in the lowest quartile of PP. Slightly more current/former smokers were in the lowest PP quartile. Mean SBP increased and DBP decreased with increasing PP. Framingham risk score increased from the lowest to the highest PP quartile. Mean total cholesterol was modestly higher and serum uric acid was lower in the third and fourth quartiles of baseline PP, whereas high-density lipoprotein cholesterol did not show a difference among quartiles. There was no difference among groups defined by PP in serum creatinine level, but the mean urinary albumin/creatinine ratio increased progressively from the first to the fourth quartile of PP.
Similar to previously reported data in the overall population,8 slightly more patients remained on blinded losartan than atenolol across quartiles of PP, without differences in dosage or use of additional therapy at end point or end of follow-up (Table 2).
Blood Pressure Control
Blood pressure results during the LIFE study have been published.8 SBP at end of follow-up or at last visit before a primary end point occurred decreased by (mean±standard deviation) 30.2±18.5 and 29.1±19.2 mm Hg in the losartan and atenolol groups, respectively, and DBP decreased by 16.6±10.1 and 16.8±10.1 mm Hg, respectively.
The blood pressure reduction from baseline to year 5 in the fourth and first baseline PP quartile were similar for SBP in the losartan and the atenolol groups (−38.6 and −20.6 versus −40.2 and −18.1 mm Hg), as was the case for DBP (−15.8 and −18.4 versus −16.4 and −17.8 mm Hg) (Table 3). DBP and mean arterial pressures (data not shown) were reduced similarly by losartan-based or atenolol-based treatment.
Cardiovascular Outcomes by Baseline PP Quartile and Treatment
The global test of treatment effect across baseline PP quartiles established that there was a treatment effect in at least one quartile of baseline PP for the primary composite end point (borderline P=0.101) and stroke (P=0.001).
Risk of the primary composite end point adjusted for Framingham risk score increased from the first to the fourth PP quartile in losartan-treated patients by 12% (confidence interval [CI], −13% to 44%; P>0.2) and in atenolol-treated patients by 28% (CI, 2% to 62%; P=0.035) (Table 4). In the unadjusted analysis, the increased risk was significant in losartan-treated (48%) (CI, 16% to 90%; P=0.002) and atenolol-treated (66%) (CI, 32% to 108%; P<0.001) patients. Kaplan–Meier curves for the primary composite end point comparing the lowest to the highest baseline PP by treatment group are shown in the Figure.
Risk of cardiovascular death increased from the first to the fourth PP quartile (Table 4) in losartan-treated and atenolol-treated patients. The risks of cardiovascular death adjusted for the Framingham risk score were not significant, possibly because of the relatively low number of events (Table 4).
Risk of stroke did not increase significantly from the first to the fourth PP quartile (Table 4) in losartan-treated patients, but increased by 84% (CI, 32% to 157%; P<0.001) in atenolol-treated patients (Table 4). Kaplan–Meier curves for stroke that compare the lowest to the highest baseline PP by treatment group are shown in Figureb.
Risk of definite myocardial infarction was increased from the first to the fourth PP quartile by 30% (CI, −13% to 94%; P>0.2) in the losartan group and 44% (CI, −5% to 120%; P=0.089) in the atenolol group. Risk of hospitalization for heart failure also increased from the first to the fourth PP quartile (Table 4) in losartan-treated and atenolol-treated patients. The risks of myocardial infarction or hospitalization for heart failure adjusted for the Framingham risk score were not significant, possibly because of the relatively low number of events (Table 4). In the unadjusted analysis, the increased risk was not significant only for hospitalization for heart failure in losartan-treated patients.
The risk of total mortality adjusted for the Framingham risk score increased from the first to the fourth PP quartile (Table 4) in losartan-treated patients by 32% (CI, −1% to 76%; P=0.062) and in atenolol-treated patients by 41% (CI, 7% to 84%; P=0.013).
The interaction tests of achieved reduction in blood pressure and treatment indicated significant interaction for cardiovascular death and DBP (interaction P<0.001), mortality and DBP (interaction P=0.022), and nonsignificant interaction for cardiovascular death and PP (interaction P=0.065).
We showed that the benefit of losartan-based treatment compared with atenolol-based treatment on outcomes in the LIFE study was most marked in the highest compared with the lowest quartile of baseline PP. These differences were for the primary composite end point, stroke, myocardial infarction, and total mortality. In atenolol-treated patients, the hazard ratio adjusted for the Framingham risk score significantly increased from the lowest to the highest PP quartile for the primary composite end point, stroke, myocardial infarction, and total mortality. In losartan-treated patients, a similar comparison of baseline PP quartiles showed smaller, nonsignificant increments in hazard ratio, particularly for stroke, with increasing PP. This analysis extends previous information provided by the initial LIFE study publication8 and the substudy on isolated systolic hypertension,9 demonstrating increasing benefit from losartan-based treatment compared with atenolol-based treatment from the lowest to the highest PP quartile. Such differences in outcomes depending on baseline PPs when comparing 2 antihypertensive drugs has not been previously described.
In the Study on Cognition and Prognosis in the Elderly (SCOPE),13 which included 4964 elderly patients with hypertension randomized to candesartan or placebo, reduction of all stroke was not significant (23.6%, P=0.056); however, in the substudy on isolated systolic hypertension in 1518 patients, stroke reduction was 42% (P=0.049).14 Although not directly comparable to the LIFE study, the results from SCOPE also indicate particular protection against stroke by an angiotensin-receptor blocker in patients with high PP. In the Valsartan Antihypertensive Long-term Use Evaluation study (VALUE), 15 245 hypertensive patients at high cardiovascular risk were treated with valsartan or amlodipine plus a diuretic and add-on drugs.15 Blood pressure reduction was significantly larger in the amlodipine group: 4.0/2.1 mm Hg after 6 months and ≈2.1/1.6 mm Hg later. The differences by treatment on primary end point and stroke were not significant and were thought to be explained by differences in achieved blood pressure. No subanalysis on impact of PP in VALUE has been published.
The strong predictive power of PP for cardiovascular risk in older patients shown in earlier studies4–7 was shown also in this analysis of the LIFE study. However, the superiority of losartan-based treatment over atenolol-based treatment in LIFE was not confined to patients at high risk. In a substudy of LIFE with 6886 patients without clinically evident vascular disease (that is, with lower cardiovascular risk), losartan was superior to atenolol even in that subpopulation.16
Because differences in achieved blood pressure (neither SBP, DBP, nor PP) did not appear to explain differences in outcomes between atenolol-based and losartan-based treatment, blood pressure–independent explanations should be explored. Increased PP is associated with left ventricular hypertrophy,17 with increased intimal and medial thickness of the carotid artery,18 with stiffening of the arterial wall,18 and with shortening of telomeres.19 Increased PP is also associated with reduced nitric oxide production by the endothelium,20 which may cause endothelial dysfunction. Consequently, elevated PP may be viewed as a marker of advanced vascular aging and disease,19 and a contributor to disease progression.7 Losartan was shown in the LIFE study to decrease left ventricular hypertrophy more effectively than atenolol.8 Further, losartan, antagonizing potentially vasculotoxic effects of angiotensin II,21 causes regression of arterial wall thickness in hypertension22 and improves endothelial function,23 all effects that atenolol does not have. Recently, it was suggested that angiotensin type 1 receptor-blockers may offer cerebroprotection via activation of angiotensin type 2 receptors.24 This hypothesis has not been proven. The superiority of cardiovascular protection in patients at high risk with the highest PP may relate to beneficial effects of losartan. These beneficial properties may translate into more effective prevention of cardiovascular morbidity and mortality particularly evident in, but not confined to, patients with high PP.
The majority of patients received additional thiazide treatment, similarly in the losartan-based (58.9%) and the atenolol-based (58.0%) treatment groups. Because combined losartan–thiazide treatment is synergistic, combined atenolol–thiazide may have unfavorable metabolic consequences (eg, decreased glucose tolerance), possibly explaining part of the difference in outcomes reported here.
Some limitations of the study need to be mentioned. The study population was mainly white and was from western countries. Measurement of blood pressure was performed by sphygmomanometry in the sitting position, which is considered to provide less accurate information than measurement of 24-hour ambulatory blood pressure.6 This was a post hoc analysis. Differences in key risk predictors at baseline for PP quartiles were taken into consideration by adjusting results for Framingham risk score; we did not adjust for severity of left ventricular hypertrophy.
We conclude that in patients with hypertension and left ventricular hypertrophy in the LIFE study, there were significantly higher risks, adjusted for the Framingham risk score, for the primary composite end point (cardiovascular death, stroke, and myocardial infarction), stroke, and total mortality in the highest versus lowest quartile of baseline PP with atenolol-based treatment. The risks in the losartan group also increased with increasing baseline PP quartile, but were lower than those in the atenolol group, and were not significant. Achieved reductions in SBP, DBP, and PP could not explain the differences in cardiovascular outcomes between treatment groups in the LIFE study. Although treatment with atenolol may no longer be a wise choice in patients with hypertension and high PP,8,25 losartan-based treatment appears to offer powerful cardiovascular protection to these patients.
In this subanalysis of the LIFE study involving 9193 patients with hypertension and left ventricular hypertrophy, we show the prognostic importance of baseline PP in terms of cardiovascular outcomes, particularly stroke. Further, we show superior protection against stroke in patients treated with losartan when compared with atenolol, most pronounced in the highest quartile of PP (>87 mm Hg). This difference may be related to specific effects of angiotensin type 1 receptor antagonism. Because the majority of patients in both treatment groups also received a diuretic, a further explanation may be that combined losartan–thiazide offers metabolic advantages (eg, better glucose tolerance) compared with combined atenolol–thiazide. Reduction of SBP, DBP, and PP were similar in both treatment groups and did not explain the difference in outcomes. Treatment with losartan may offer particular benefits beyond blood BP reduction in hypertensive patients with high cardiovascular risk and high PP.
We thank Sigrid Helle Berg for her dedicated work with the LIFE Study. The trial was supported by an unrestricted grant from Merck & Co, Inc.
- Received September 16, 2004.
- Revision received October 6, 2004.
- Accepted February 16, 2005.
Benetos A, Rudnichi A, Safar M, Guize L. Pulse pressure and cardiovascular mortality in normotensive and hypertensive subjects. Hypertension. 1998; 32: 560–564.
Domanski MJ, Davis BR, Pfeffer MA, Kastantin M, Mitchell GF. Isolated systolic hypertension: prognostic information provided by pulse pressure. Hypertension. 1999; 34: 375–380.
Millar JA, Lever AF. Implications of pulse pressure as a predictor of cardiac risk in patients with hypertension. Hypertension. 2000; 36: 907–911.
Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease? The Framingham Heart Study. Circulation. 1999; 100: 354–360.
Khattar RS, Swales JD, Dore C, Senior R, Lahiri A. Effect of aging on the prognostic significance of ambulatory systolic, diastolic, and pulse pressure in essential hypertension. Circulation. 2001; 104: 783–789.
Dahlöf B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002; 359: 995–1003.
Kjeldsen S, Dahlöf B, Devereux RB, Julius S, Aurup P, Edelman J, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Snapinn S, Wedel H. Effects of losartan on cardiovascular morbidity and mortality in patients with isolated systolic hypertension and left ventricular hypertrophy: a Losartan Intervention for Endpoint Reduction (LIFE) substudy. JAMA. 2002; 288: 1491–1498.
Dahlöf B, Devereux R, de Faire U, Fyhrquist F, Hedner T, Ibsen H, Julius S, Kjeldsen S, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H. The Losartan Intervention For Endpoint reduction (LIFE) in Hypertension study: rationale, design, and methods. The LIFE Study Group. Am J Hypertens. 1997; 10: 705–713.
Dahlöf B, Devereux RB, Julius S, Kjeldsen SE, Beevers G, de Faire U, Fyhrquist F, Hedner T, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H. Characteristics of 9194 patients with left ventricular hypertrophy: the LIFE Study. Losartan Intervention For Endpoint Reduction in Hypertension. Hypertension. 1998; 32: 989–997.
Kannel WB, McGee DL. Diabetes and cardiovascular risk factors: the Framingham study. Circulation. 1979; 59: 8–13.
Lithell H, Hansson L, Skoog I, Elmfeldt D, Hofman A, Olofsson B, Trenkwalder P, Zanchetti A. The Study on Cognition and Prognosis in the Elderly (SCOPE): principal results of a randomized double-blind intervention trial. J Hypertens. 2004; 21: 875–886.
Papademetriou V, Farsang C, Elmfeldt D, Hofman A, Lithell H, Olofsson B, Skoog I, Trenkwalder P, Zanchetti A. Stroke prevention with the angiotensin II type 1-receptor blocker candesartan in elderly patients with isolated systolic hypertension. The Study on Cognition and Prognosis in the Elderly (SCOPE). J Am Coll Cardiol. 2004; 44: 1175–1180.
Julius S, Kjeldsen SE, Weber M, Brunner HR, Ekman S, Hansson L, Hua T, Laragh J, McInnes GT, Mitchell L, Plat F, Schork A, Smith B, Zanchetti A. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomised trial. Lancet. 2004; 363: 2022–2031.
Devereux RB, Dahlöf B, Kjeldsen SE, Julius S, Aurup P, Beevers G, Edelman JM, de Faire U, Fyhrquist F, Helle Berg S, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Snapinn S, Wedel H. Effects of losartan or atenolol in hypertensive patients without clinically evident vascular disease: a substudy of the LIFE randomized trial. Ann Intern Med. 2003; 139: 169–177.
Gardin JM, Arnold A, Gottdiener JS, Wong ND, Fried LP, Klopfenstein HS, O’Leary DH, Tracy R, Kronmal R. Left ventricular mass in the elderly. The Cardiovascular Health Study. Hypertension. 1997; 29: 1095–1103.
Franklin SS, Sutton-Tyrrell K, Belle SH, Weber MA, Kuller LH. The importance of pulsatile components of hypertension in predicting carotid stenosis in older adults. J Hypertens. 1997; 10: 1143–1150.
Aviv A. Hypothesis: pulse pressure and human longevity. Hypertension. 2001; 37: 1060–1066.
Dzau VJ. Theodore Cooper Lecture: tissue angiotensin and pathobiology of vascular disease: a unifying hypothesis. Hypertension. 2001; 37: 1047–1052.
Schiffrin EL, Park JB, Intengan HD, Touyz RM. Correction of arterial structure and endothelial dysfunction in human essential hypertension by the angiotensin receptor antagonist losartan. Circulation. 2000; 101: 1653–1659.
Fournier A, Messerli FH, Achard JM, Fernandez L. Cerebroprotection mediated by angiotensin II. A hypothesis supported by recent randomized trials. J Am Coll Cardiol. 2004; 43: 1323–1327.