Different Effects of Antihypertensive Therapies Based on Losartan or Atenolol on Ultrasound and Biochemical Markers of Myocardial Fibrosis

Abstract

Background— In hypertensive left ventricular hypertrophy (LVH), myocardial texture is altered by a disproportionate increase in fibrosis, but there is insufficient clinical evidence whether antihypertensive therapy or individual agents can induce regression of myocardial fibrosis. Methods and Results— We compared the effects of an angiotensin II receptor antagonist with a β-blocker on myocardial collagen volume (assessed by echoreflectivity and serum collagen markers) in 219 hypertensive patients with echocardiographically documented LVH. Patients were allocated randomly to receive losartan 50 to 100 mg/d (n=111) or atenolol 50 to 100 mg/d (n=99) with or without hydrochlorothiazide 12.5 to 25 mg/d for 36 weeks. Echoreflectivity analysis was conducted on ultrasound tracings of the midapex septum with specifically designed and validated software. A color histogram of reflecting echoes was obtained, and its spread (broadband [BB], previously shown to correlate directly with collagen volume fraction on endomyocardial biopsies) was used as the primary outcome measure. Mean color scale and serum markers of collagen synthesis (PIP, PIIIP) or degradation (CITP) were secondary outcome variables. Echoreflectivity analysis proved feasible in 106 patients (losartan 52, atenolol 54). Losartan reduced BB over 36 weeks (from 114.5 to 104.3 color levels, P <0.02), whereas atenolol treatment was associated with an increase in BB (from 109.0 to 113.6 color levels, P =NS), the difference between treatments being −12.8 color levels (95% CI −23.6 to −2.0, P =0.02). Secondary end points (mean color scale and collagen markers) also changed in the direction of decreased collagen in patients receiving losartan, but differences between groups were not statistically significant. Conclusions— In hypertensive patients with LVH, losartan decreases myocardial collagen content, whereas atenolol does not. The difference between the 2 treatments is statistically significant.