Is Plasma Renin Activity a Biomarker for the Prediction of Renal and Cardiovascular Outcomes in Treated Hypertensive Patients? Observations From the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT)

Study Questions:

What is the relationship between cardiovascular (CV) and renal outcomes and all-cause mortality with baseline measurements of plasma renin activity (PRA) among hypertensive adults randomized in the ASCOT trial?

Methods:

In the United Kingdom and Ireland, ASCOT included 9,098 hypertensive adults randomized to either calcium channel blocker (CCB)- or beta-blocker (BB)-based treatment. Four thousand eight hundred and fifty-three patients with total cholesterol ≤6.5 mmol/L (250 mg/L) were further randomized to atorvastatin or placebo. Over 5.5 years, there were 399 CV events (fatal coronary heart disease [CHD], nonfatal myocardial infarction [MI], coronary revascularization, and fatal and nonfatal stroke), 96 cases of new-onset renal impairment, and 220 deaths. Cases were age, sex, and ethnicity-matched with 1,525 controls. Conditional logistic regression models were used to evaluate the association between CV events, renal impairment, all-cause mortality, and PRA.

Results:

For those on antihypertensive (AHT) treatment at the baseline (91.5%), PRA was influenced by prior drug treatment. The median (interquartile range; ng/ml/h) levels were 1.04 (0.52, 1.3) for BBs, 1.30 (0.78, 2.72) for CCBs, 1.56 (0.91, 3.50) for diuretics, and 2.33 (1.30, 5.57) for angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers. Odds ratios (ORs) and 95% confidence intervals (CIs) for CV and other events were estimated for 1-standard deviation increase in log-transformed PRA levels and by categorizing PRA into quartiles with the lowest as the referent category. Baseline PRA did not predict CV events in models adjusted for baseline characteristics (OR, 0.92; CI, 0.81-1.06; p = 0.25) and for pre-randomized AHT treatment (OR, 0.91; CI, 0.79-1.04; p = 0.17) and was not associated with all-cause mortality (OR, 1.12; CI, 0.92-1.37; p = 0.25) and OR, 1.06 (CI, 0.91-1.24; p = 0.46) in the fully adjusted model. Baseline levels of PRA were positively, but nonsignificantly associated with the development of renal impairment in models adjusted for baseline characteristics (OR, 1.39; CI, 0.97-1.97; p = 0.07), and also for pre-randomized AHT treatment (OR, 1.35; CI, 0.95-1.94; p = 0.10). Quartile analyses, however, demonstrated a significant positive association of higher levels of PRA with the development of impaired renal function (p = 0.03 and 0.05 in adjusted models, respectively) compared with the lowest quartile.

Conclusions:

The authors concluded that their analyses do not support the use of PRA to predict future CV events or all-cause mortality in treated hypertensive patients without diagnosed CHD.

Perspective:

The current study reports that in a large subgroup of hypertensive subjects with no pre-existing CV disease, PRA at the baseline was not a predictor of the subsequent development of future CV events, either MI, stroke, or heart failure. However, in models adjusted for associated baseline characteristics, there was a positive association of PRA with the subsequent development of renal impairment, which achieved statistical significance in trend analyses. However, given the profound effects that prior AHT drugs have on PRA levels and the uncertainty with which such treatments may confound the use of PRA as a potential biomarker for CV and renal outcomes, the association of PRA with renal impairment needs further prospective study.

Clinical Topics: Dyslipidemia, Heart Failure and Cardiomyopathies, Prevention, Lipid Metabolism, Nonstatins, Novel Agents, Statins, Acute Heart Failure, Heart Failure and Cardiac Biomarkers, Hypertension

Keywords: Uncertainty, Odds Ratio, Angiotensin Receptor Antagonists, Myocardial Infarction, Stroke, Great Britain, Renin, Diuretics, Coronary Disease, Heptanoic Acids, Ireland, Pyrroles, Cholesterol, Renal Insufficiency, Biological Markers, Heart Failure, Cardiovascular Diseases, Confidence Intervals, Hypertension, Logistic Models


< Back to Listings