Blood pressure-lowering effects of biofeedback treatment in hypertension: a meta-analysis of randomized controlled trials.

To examine the blood pressure-lowering effects of biofeedback treatment in patients with essential hypertension, a meta-analysis was conducted on studies published between 1966 and 2001. A total of 22 randomized controlled studies with 905 essential hypertensive patients were selected for review. Compared with clinical visits or self-monitoring of blood pressure (non-intervention controls), biofeedback intervention resulted in systolic and diastolic blood pressure reductions that were greater by 7.3 mmHg (for systole; 95% confidence interval: 2.6 to 12.0) and 5.8 mmHg (for diastole; 95% confidence interval: 2.9 to 8.6). Compared with sham or non-specific behavioral intervention controls, the net reductions in systolic and diastolic blood pressures by biofeedback intervention were 3.9 (95% confidence interval: -0.3 to 8.2) and 3.5 (-0.1 to 7.0) mmHg, respectively. The results of multiple regression analysis also indicated that biofeedback intervention decreased systolic and diastolic blood pressures more than non-intervention controls (p < 0.001), but not more than sham or non-specific behavioral intervention controls (p > 0.05), when controlling for the effects of initial blood pressures. When biofeedback intervention types were classified into two types, simple biofeedback and relaxation-assisted biofeedback, only the relaxation-assisted biofeedback significantly decreased both systolic and diastolic blood pressures (p < 0.05) compared with those in sham or non-specific behavioral intervention controls. The results suggested that biofeedback was more effective in reducing blood pressure in patients with essential hypertension than no intervention. However, the treatment was only found to be superior to sham or non-specific behavioral intervention when combined with other relaxation techniques. Further studies will be needed to determine whether biofeedback itself has an antihypertensive effect beyond the general relaxation response.