Microalbuminuria (MA) is associated with increased cardiovascular risk in adult hypertensive patients, but no study has specifically examined the effects of MA lowering on regression of left ventricular hypertrophy (LVH) among pediatric patients with hypertension. Fifty-five patients with essential hypertension, 11-19 years old, were prospectively studied. All patients received concomitant therapy of hydrochlorothiazide and angiotensin-converting enzyme inhibitor. Five patients also required angiotensin receptor blocker to achieve the blood pressure goal. Baseline and 12-month follow-up measures of left ventricular mass index (LVMI), determined by echocardiography and urine microalbumin/creatinine ratio (MA/Cr), were collected. MA was defined as MA/Cr >30 μg/mg. LVH was defined as LVMI >38.6 g/m2.7. The primary end points were reductions in MA and LVMI of 25% or more. Weight (r = 0.83), body surface area (r = 0.85), body mass index (BMI) (r = 0.86), systolic blood pressure (SBP) (r = 0.57), diastolic blood pressure (DBF) (r = 0.49), mean arterial pressure (r = 0.53) and MA (r= 0.87) were all univariate correlates of LVMI. In a multiple regression analysis, MA, BMI and SBP were significant correlates of LVMI. MA alone explained 76% of the variance of LVMI, whereas BMI and SBP explained only 1.6 and 0.4% of the variance, respectively. MA was the most significant correlate of follow-up LVMI after BMI and SBP were included in the overall multiple regression models. Thus, MA is a strong predictor of LVH in hypertensive children and adolescents. MA lowering halts the progression of LVH or induces its regression.