We examined the long-term cost effectiveness of treating hypertensive patients aged 47 to 65 yrs with agents that promote regression of left ventricular hypertrophy (LVH). Beta-blockers, calcium channel blockers, and ACE inhibitors were compared with standard therapy. To estimate the effect of drug therapy on LVH regression, we pooled data from 25 studies. We estimated the effects of LVH regression on cardiovascular outcomes using Framingham data and the studies of Devereux. The estimated costs of differing treatment strategies included average costs of drug therapy and follow-up care, and direct costs of lost productivity owing to untoward outcomes. Patients were classified by initial left ventricular mass index (LVMI) as low (< 95 g/m2), moderate (95-125 g/m2) and high risk (> 125 g/m2). The data suggested that all three agents reduce LVMI in moderate- and high-risk groups. The respective average reductions in LVMI in high risk and moderate risk were 20% and 12% for ACE inhibitors, 18% and 8% for beta-blockers, and 5% and 6% for calcium channel blockers. We estimated the relative improvement in cardiovascular morbidity/mortality from LVH reversal required for overall cost savings with each drug class compared with a standard regimen. Patients treated with calcium channel blockers needed to realise at least 72% of the expected reduction in cardiovascular outcomes from LVH regression for this strategy to be less costly. For ACE inhibitors and beta-blockers, only 32% and 26%, respectively, of the expected reduction in poor outcomes from LVH reversal were required for these agents to be more cost effective. Long-term costs of treatment with ACE inhibitors versus beta-blockers were similar. A 25% higher treatment cost of ACE inhibitors was offset by a greater effectiveness in reversing LVH with attendant lower rates of cardiovascular complications.