Comparative effectiveness of fourth-line anti-hypertensive agents in resistant hypertension: A systematic review and meta-analysis.

Sinnott, S.J. ; Tomlinson, L.A. ; Root, A.A. ; Mathur, R. ; Mansfield, K.E. ; Smeeth, L. ; Douglas, I.J. ;
Comparative effectiveness of fourth-line anti-hypertensive agents in resistant hypertension: A systematic review and meta-analysis.
Eur J Prev Cardiol, 2016;

We assessed the effectiveness of fourth-line mineralocorticoid receptor antagonists in comparison with other fourth-line anti-hypertensive agents in resistant hypertension.

We systematically searched Medline, EMBASE and the Cochrane library from database inception until January 2016. We included randomised and non-randomised studies that compared mineralocorticoid receptor antagonists with other fourth-line anti-hypertensive agents in patients with resistant hypertension. The outcome was change in systolic blood pressure, measured in the office, at home or by ambulatory blood pressure monitoring. Secondary outcomes were changes in serum potassium and occurrence of hyperkalaemia. We used random effects models and assessed statistical heterogeneity using the I(2) test and corresponding 95% confidence intervals. From 2,506 records, 5 studies met our inclusion criteria with 755 included patients. Two studies were randomised and three were non-randomised. Comparative fourth-line agents included bisoprolol, doxazosin, furosemide and additional blockade of the renin angiotensin-aldosterone system. Using data from randomised studies, mineralocorticoid receptor antagonists reduced blood pressure by 7.4 mmHg (95%CI 3.2 – 11.6) more than the active comparator. When limited to non-randomised studies, mineralocorticoid receptor antagonists reduced blood pressure by 11.9 mmHg (95% CI 9.3 – 14.4) more than the active comparator.

On the basis of this meta-analysis, mineralocorticoid receptor antagonists reduce blood pressure more effectively than other fourth-line agents in resistant hypertension. Effectiveness stratified by ethnicity and comorbidities, in addition to information on clinical outcomes such as myocardial infarction and stroke, now needs to be determined.