Left Ventricular Hypertrophy Regression—Losartan vs Atenolol

Abstract & Commentary

Synopsis: Losartan therapy of hypertensive patients with LVH resulted in greater LVH regression as compared to atenolol independent of blood pressure control and baseline ECG findings.

Source: Okin PM, et al. Circulation. 2003;108:684-690.

Since left ventricular hypertrophy (LVH) has been associated with preclinical atherosclerotic vascular disease and predicts cardiovascular morbidity and mortality, LVH regression has been the Holy Grail of hypertension therapy. Unfortunately, many previous studies have failed to provide clear information on how best to accomplish LVH regression. Meta analyses of available trials have suggested that agents that antagonize the rennin-angiotensin system may be best. Thus, the investigators in the LIFE trial undertook a sophisticated analysis of ECG-LVH regression in this large trial. Over 9000 hypertensive patients were randomized to losartan or atenolol therapy, treated to target blood pressure levels, and followed for up to 5 years. ECGs were done at baseline, 6 months, and then yearly for the duration of the study. Two criteria for ECG-LVH were used: Cornell method—amplitude of the R wave in lead AVL plus the S wave in lead V3 times the QRS duration; and Sokolow-Lyon method—the S wave amplitude in V1 plus the R wave in V5 or V6. Losartan reduced the Cornell LVH score by -200 vs -69 mm × ms on atenolol (P < .001) and the Sokolow-Lyon by -2.5 vs 0.7 mm (P < .001). These changes persisted over the follow-up period. There were no subgroup differences; age, gender, ethnicity, and diabetics all showed similar results. Also, the ECG results were adjusted for baseline ECG findings, blood pressure, and diuretic use. In addition, the prevalence of LVH by the Cornell method decreased 21% on losartan vs 13% on atenolol (P < .001) and by Sokolow-Lyon by 12 vs 8.5% (P < .001). Okin and colleagues concluded that losartan therapy of hypertensive patients with LVH resulted in greater LVH regression as compared to atenolol independent of blood pressure control and baseline ECG findings.

Comment by Michael H. Crawford, MD

The strength of this study is that it is large enough to detect a difference in LVH despite the use of 2 effective antihypertensive agents. It would seem to settle the issue between renin-angiotensin system blockers and beta-blockers, at least in a largely white population with LVH. Also, unlike previous studies, the results were adjusted to baseline ECG findings and blood pressure, as well as diuretic use.

These results are consistent with the HOPE trial, where it was found that ramipril as compared to placebo lowered the likelihood of developing LVH or its persistence and raised the likelihood of prevention or regression of LVH, even when adjusted for the larger change in blood pressure on ramipril. Also, these results are consistent with animal data showing that angiotensin II has a potent myocardial hypertrophy effect. What is not known is how losartan would compare to calcium channel blockers. Some studies have shown equivalent efficacy of calcium blockers in LVH regression and more potent antihypertensive effects.

There are some limitations to the study. The results are only applicable to those with moderate-to-severe hypertension and ECG evidence of LVH. Whether similar results would be seen in those with mild hypertension is doubtful. Also, the use of the ECG to diagnose LVH is problematic. In this study, patients were eligible if they met either the Cornell or the Sokolow-Lyon criteria, but the total number of patients who met LVH criteria by each method does not add up to the total number of patients in the study. Okin et al comment that after enrollment some patients no longer met criteria for LVH but were not excluded at that point. This raises the issue of whether some of the LVH regression observed was due to regression to the mean. In addition, the 2 ECG methods are quite different. The Sokolow-Lyon is very sensitive and the Cornell is much more specific. Combining them represents an interesting mix of patients. It would have been interesting to have echocardiograms in a subgroup. Despite these issues, this large trial strongly suggests that the greater reduction in cardiovascular end points on losartan as compared to atenolol observed in LIFE are probably partly due to the superior effect on LVH regression.

Dr. Crawford is Professor of Medicine, Associate Chief of Cardiology for Clinical Programs, University of California, San Francisco, Calif.