The Losartan Intervention for Endpoint Reduction (LIFE) Trial—Have Angiotensin‐Receptor Blockers Come of Age?

Abstract

The Losartan Intervention for Endpoint Reduction trial is one of several end‐point trials that are now available with angiotensin‐receptor blockers. This trial compared two regimens—losartan‐based therapy to atenolol‐based therapy—in 9193 hypertensive patients with electrocardiographic evidence of left ventricular hypertrophy. In the instance of each of these therapeutic groups, hydrochlorothiazide add‐on therapy was permitted as per protocol. Although blood pressures were comparably reduced in both the losartan and the atenolol‐based treatment groups, stroke rate was notably less in the losartan‐treatment group. The 1195 patient diabetic cohort in this trial also experienced a substantial reduction in total and cardiovascular mortality favoring losartan. An additional finding in this trial was that new‐onset diabetes developed 25% less frequently in the losartan‐treated group. The results of this trial are both interesting and relevant to what is an expanding use of angiotensin‐receptor blockers in the hypertensive population.

Seven angiotensin‐receptor blockers (ARBs) are currently available in the United States. Each of these several drugs reduces blood pressure (BP) comparably to what has been observed with other drug classes such as angiotensin‐converting enzyme (ACE) inhibitors, calcium channel blockers, and β blockers. To date, the major advantage of these drugs has resided in the virtual absence of side effects when they are used. It has long been recognized, though, that a favorable side‐effect profile can serve as the marketing mantra of a drug class for only a limited period of time. Ultimately, hard end point trials are needed to prove/disprove the merit of a drug class. The findings from one such trial, the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) study, were recently published and offer important insights into the workings of this drug class. ¹ , ²

AIMS

The primary aim of the LIFE study was to determine if selective AT₁‐receptor antagonism with losartan would be more effective than β blockade with atenolol in reducing cardiovascular morbidity and death in patients with essential hypertension and definitive electrocardiographic signs of left ventricular hypertrophy (LVH).

METHODS

The LIFE study was a double‐masked, randomized, parallel‐group trial of a losartan‐based treatment program compared to an atenolol‐based regimen in hypertensive patients with LVH. Patients aged 55–80 years were enrolled from June, 1995 to May 2, 1997 with follow‐up for at least 4 years. In the final analyses, 9193 patients were available of which 1195 (13%) were diabetic (586 losartan, 609 atenolol). Treatment groups were matched with respect to demographic characteristics, severity of hypertension, and prevalence of coexisting cardiovascular conditions, Framingham risk score, and electrocardiographic (ECG)‐LVH criteria. In the diabetics, however, there were significantly more patients with atrial fibrillation at baseline in the atenolol group. The BP goal in this study was a target BP of less than 140/90 mm Hg with pre‐study BP values having to fall in the 160–200 and 95–115 mm Hg ranges for systolic and diastolic BP, respectively.

In both randomized groups, a 50‐mg dose of losartan or atenolol was the starting dose. Thereafter, hydrochlorothiazide (HCTZ) (12.5–25.0 mg/day), together with titration of the starting drug to 100 mg, as well as other agents, was permitted—with the exception of β blockers, ACE inhibitors, or ARBs—if BP remained persistently high during follow‐up. An identical protocol was followed in the diabetic cohort. In the diabetic cohort, only 124 (21%) losartan‐ and 127 (21%) atenolol‐treated patients went untreated with antidiabetic drugs or insulin during the trial. A primary composite end point of cardiovascular morbidity and mortality (cardiovascular death, stroke, or myocardial infarction) was utilized in this study. Both treatment groups were well balanced for treatment with lipid‐lowering drugs and aspirin throughout the study. All end points in the LIFE study were assessed by an intention‐to‐treat analysis.

RESULTS

In the main trial, BP fell by 30.2±18.5/16.6±10.1 and 29.1±19.2/16.8±10.1 mm Hg in the losartan and atenolol groups respectively. At the end of the follow‐up, sitting systolic BP was reduced by 1‐mm Hg more with losartan than with atenolol. Mean Bps at the last visit were 144.1±17.1/81.3±9.6 and 145.4±16.4/80.9±9.6 mm Hg respectively, in the losartan and atenolol groups. Heart rate decreased more in patients assigned to atenolol than losartan (−7.7±12.8 and −1.8±12.0 beats/min, respectively, (p<0.0001).

There was no significant difference between the losartan and atenolol treatment groups in adjusted relative risk of cardiovascular mortality, which was one of the primary endpoint components. For the overall primary endpoint, there was a 13% overall relative risk reduction for those on losartan, mostly the result of a 24.9% difference in stroke between the two groups. There was no significant difference between the losartan and atenolol treatment groups in adjusted relative risk for the primary end point component of fatal and nonfatal myocardial infarction. LVH was reduced more in the losartan‐based group than in the atenolol group. ³

In the diabetic subset the findings were somewhat different. Losartan‐treated patients demonstrated a 39% reduction in total mortality (p=0.002), a 37% reduction in cardiovascular mortality (p=0.028), and a 24% reduction in reaching the primary composite end point. There was no statistically significant difference, however, on strokes or myocardial infarctions in the diabetic cohort. New‐onset diabetes, defined according to 1985 World Health Organization criteria, was 25% less frequent with losartan. ¹ , ² At follow‐up, 84% and 80% of the losartan and atenolol groups, respectively, remained on study drug. In patients who stayed on study drugs until the completion of the study, mean doses of losartan and atenolol were 82±24 and 79±26 mg, respectively. Only about 10% of all patients in each treatment group were still receiving the 50‐mg starting dose of losartan or atenolol.

DISCUSSION

The LIFE study has provided an additional basis for the use of ARB‐based therapy in the treatment of hypertension. An important consideration in this study is the fact that per protocol most patients required more than one‐drug for BP control and most times the maximum allowable number of titrations were required; thus, in this study it was the regimen per se and not losartan alone that provided the benefit. The LIFE study was designed to mimic “real world” conditions where a combination of drugs and nonpharmacologic measures is normally required to provide satisfactory control of BP. In this regard, only 10% of patients in each study group reached goal BP with monotherapy—the remainder did so with the addition of HCTZ. Importantly, LIFE confirmed, like the Hypertension Optimal Treatment (HOT) ⁴ and the African American Study of Kidney Disease and Hypertension (AASK) trials ⁵ before it, that it is possible to achieve current BP targets in community‐based populations with up‐titrated drug doses and combination treatments.

The end point findings from the LIFE study are of interest—in particular, a 25% reduction in stroke rate with losartan compared to the atenolol group. ¹ , ² , ⁶ That losartan could have a significant effect on reducing stroke rate over and above BP reduction extends the line of reasoning that emerged from the Heart Outcomes Prevention Evaluation (HOPE) trial, wherein a group treated with an ACE inhibitor (ramipril)‐based regimen appeared to provide stroke protection in a BP‐independent manner. ⁷

How compounds that interact with the reninangiotensin axis protect against stroke is unclear at this time. The substantial reduction in stroke rate in the losartan‐treated arm of the LIFE study does not answer this question since this was not a mechanistic study and a priori there was not a specific reason to believe that losartan would decrease stroke rate more than other medications. However, it is known that LVH (both on ECG and echocardiography) is a BP‐independent predictor of cerebrovascular events, ⁸ and left ventricular mass regressed more so with losartan in the LIFE trial. This was sufficient to explain about one third of the difference in stroke rate. In addition, the selection of atenolol, and for that matter a β blocker, as the control therapy could explain some of the difference between the treatment groups. In this regard, central pressure determinants are increasingly viewed as a truer reflection of cardiovascular risk than is the case with peripheral BP readings. Thus, a plausible, although untested explanation, for the findings in the LIFE trial is the differing effects of β blockers and ARBs on central pulse pressure and blood vessel stiffness despite similar peripheral Bps. Although these parameters were not determined in the LIFE trial it might be expected that angiotensin‐receptor blockade more favorably impacted these measures than did β blockade, ⁹ and therein reduced central pressures to a greater degree. ¹⁰ In this regard, Schiffrin et al. ¹¹ have also previously shown that treatment with losartan provided greater reduction in small arterial wall thickness than did atenolol for equal peripheral BP control.

Yet, these findings remain puzzling and to a degree internally inconsistent. For example, in a higher‐risk group—the diabetics in the LIFE trial—losartan did not afford any additional benefit as regards stroke rate. Likewise, the losartan treatment group in the Reduction of Endpoints in Non‐Insulin‐Dependent Diabetes Mellitus With the Angiotensin II Antagonist Losartan (RENAAL) trial did not realize any reduction in stroke rate compared with other antihypertensives. ¹² Future studies, such as the Valsartan Antihypertensive Long‐term Use Evaluation (VALUE) trial, will hopefully further clarify the question of whether or not specific stroke protection benefits exist for the ARB class beyond BP reduction.

In addition, in the LIFE trial the lower rate of new‐onset diabetes (difference of 25%) with losartan confirms findings from prior studies with the ACE inhibitors ramipril and captopril. ¹³ , ¹⁴ This finding could have related to a differential effect on insulin resistance—the precursor condition for overt diabetes—favoring losartan, although prior studies with losartan have not shown it to improve insulin resistance. ¹⁵ , ¹⁶ Likewise, the same was the case in the LIFE trial; therein, there was not a specific positive benefit of losartan on insulin resistance. It was more so a negative effect with β blockade atenolol, since in those receiving atenolol, insulin sensitivity fell throughout the study. ¹⁷ It is also possible that in a multi‐year study such as this that the deleterious effects of aging might have affected insulin and glucose metabolism in these patients. Of interest is the observation that body weight decreased similarly in both treatment groups (84.3 to 82.5 kg and 84.3 to 82.6 kg for losartan and atenolol, respectively). ¹⁸ Typically, body weight increases with long‐term β blocker therapy. ¹⁹

Unresolved Questions on the LIFE Trial.

Aspirin Therapy.

As of yet, the LIFE study has not reported whether there was an aspirin and losartan interaction. This has been an area of considerable interest in the congestive heart failure arena and should also be a consideration in the LIFE study as relates to the stroke finding. Of note, patients in the HOPE trial, who had previously been treated with aspirin, tended toward a reduced stroke benefit from ramipril compared to patients who had not been treated with aspirin. ⁷

Did Race or Geography Influence the LIFE Results?.

One limitation of the LIFE trial was that the study population was predominantly white and drawn from western countries. Only 19% of the overall study population came from the United States. In addition, the overall study findings have yet to be formally reported on either a regional or a country basis. This is pertinent since observations from the LIFE study suggest that the stroke benefit of losartan was not found in black subjects. In fact, just the reverse was observed with atenolol‐treated blacks having a lower rate of stroke. In the LIFE trial, only 533/9193 (6%) of enrolled subjects were black, of which 270 were randomized to losartan. In the black hypertensives taking losartan, 46 patients (17%) experienced a heart attack, stroke, or cardiovascular death, compared to 29 patients (11%) in the atenolol group. This occurred despite the finding that BP was reduced similarly for both blacks and nonblacks with both treatments, and that black patients in the losartan‐treated group regressed left ventricular mass in a fashion comparable to what was observed with atenolol. ²⁰

This finding is difficult to interpret for several reasons. First, the geographic location of blacks has yet to be reported and the criteria used to define ethnicity were not commented on per se. Second, the analysis of data according to race was just one of several such data cuts and the only one so positive. Like all such post‐hoc analyses in large clinical trials there is little statistical power to the observation and at best it can only be viewed as hypothesis generating. Finally, this racial difference occurred in the face of multidrug therapy, including a high rate of HCTZ use. This observation is also in no way consistent with the existing base of information supporting diuretic therapy in primary or secondary stroke prevention. ²¹ , ²² Black hypertensives should not be denied ARB therapy on the basis of these findings.

Did the Effect of Losartan on LVH Influence Other Clinical End Points?.

Both drugs caused regression of LVH throughout the 5‐year study, although the effect of losartan was significantly greater than that of atenolol. While there were no overall differences between the two study groups in cardiac events, there was a significant reduction in heart failure hospitalizations for diabetic patients treated with losartan. Because LVH‐dependent diastolic dysfunction is a common cause of heart failure in the elderly, it is possible that sustained treatment with an ARB could be beneficial in protecting against this serious outcome.

Did the Effect of Losartan on Uric Acid Contribute to the Difference in Stroke Rate?.

In the LIFE trial there was a differential effect on serum uric acid between the two treatment groups with the study‐related rise in uric acid being less in the losartan treatment group. Serum uric acid rose from 328 to 348 µmol/L in the losartan group compared to a rise from 329 to 376 µmol/L in the atenolol group. Losartan has one unique feature, which distinguishes it from other ARBs, which is that it is uricosuric. ²³ This is a useful feature in the diuretic‐treated patient, since diuretics cause a dosedependent form of hyperuricemia. The rise in uric acid in this study was likely secondary to two processes: first, the influence of HCTZ on uric acid handling and second the gradual decline in renal function that occurred during the 4 years of the study (86 to 97 mmol/L (losartan) vs. 85 to 96 mmol/L (atenolol). It seems unlikely and difficult to understand how this small difference in uric acid could be responsible for the protection from stroke. The relationship between uric acid and stroke is complex and poorly explored mechanistically. Moreover, in patients with acute ischemic stroke, there is a 12% increase in the odds of a good clinical outcome for each mg/dL increase of serum uric acid. This finding reinforces the relevance of oxidative damage in ischemic stroke since uric acid is an efficient antioxidant. ²⁴

Do the Findings of the LIFE Study Suggest a Class Effect for ARBs?.

This is a complicated question and one asked whenever the results of outcomes studies become available with one member of a drug class. Acceptance of the concept of a “class‐effect” does not require adopting an all‐or‐none approach since there is considerable middle ground for this concept. First, the more specific the therapeutic target, such as the angiotensin receptor for the ARBs, the more difficult it is to differentiate one compound from another in a drug class. In addition, if the ARB dose being given in a clinical trial falls within conventional dosing boundaries it becomes easier to substitute one for the other. This was, in fact, the case in the LIFE trial, wherein the losartan dose ranged from 50–100 mg/day. Yet the ARBs are structurally and physicochemically distinct, with differing potency, pharmacokinetics, and tissuepenetration characteristics. ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ The latter property may be of particular importance if relevant central nervous system penetration occurs for any of these drugs. Thus, although substitution of one ARB for the other seems a simple proposition one should be ever mindful of the fact that evidence‐based medicine derives from specific drugs having been used under specific clinical trial circumstances. Departure from the proven treatment paradigm that generated a positive result should only occur cautiously.

CONCLUSION

The LIFE study provides an additional basis for the use of ARBs in the treatment of hypertension. Additional information will be forthcoming from this trial. The early finding of a decreased stroke rate in losartan‐treated patients holds considerable promise. In addition, the finding of a lesser incidence of new‐onset diabetes lends some additional support for the use of drugs known to interrupt activity in the renin‐angiotensin axis.