Reductions in albuminuria and in electrocardiographic left ventricular hypertrophy independently improve prognosis in hypertension:: the LIFE study

Background In the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study, reduced urine albumin/creatinine ratio (UACR) as well as regression of left ventricular hypertrophy have been associated with lower incidence of cardiovascular events. We wanted to investigate whether these prognostic improvements were independent. Methods In 6679 hypertensive patients included in the LIFE study, we measured UACR, left ventricular hypertrophy by electrocardiography, serum cholesterol, plasma glucose and blood pressure after 2 weeks of placebo treatment and again after 1 year of anti-hypertensive treatment with either an atenolol- or a losartan-based regimen. During this first year of treatment, 77 patients encountered a non-fatal stroke or myocardial infarction and were excluded to avoid bias. During the next 3-4 years, 610 composite endpoints [cardiovascular death (n = 228), fatal or non-fatal myocardial infarction or stroke] were recorded. Results In Cox regression analyses, the composite endpoint was after adjustment for treatment allocation predicted by baseline logLIACR [hazard ratio (HR) = 1.16 per 10-fold increase, P < 0.05], 1-year logUACR (HR = 1.29 per 10-fold increase), baseline Sokolow-Lyon voltage (HR = 1.01 per mm, both P < 0.001) and 1-year Cornell product (HR = 1.01 per 100 mm x ms, P < 0.01). Cardiovascular death was predicted by 1-year logLIACR (HR = 1.59, P < 0.001), baseline Sokolow-Lyon voltage (HR = 1.01, P = 0.06) and 1-year Cornell product (HR = 1.02, P < 0.001). Both were predicted independent of age, Framingham risk score, current smoking, history of cardiovascular disease and diabetes. Gender, serum cholesterol, plasma glucose and blood pressure did not enter the models. Conclusions Baseline UACR and Sokolow-Lyon voltage, as well as in-treatment UACR and Cornell product added to the risk prediction independent of traditional risk factors, indicating that albuminuria and left ventricular hypertrophy reflect different aspects of cardiovascular damage and are modifiable cardiovascular risk factors.