Our study has limitations. We could not perform a
multivariate meta-regression analysis because data were
missing for important covariables, such as ethnicity (17
trials) and hypertensive status (7 trials). We did, however,
conduct subgroup analyses based on these important
covariables. Aerobic exercise had a slightly greater
effect on blood pressure in hypertensive participants
than in normotensive participants. The reduction was
greater for systolic blood pressure in black persons and
for diastolic blood pressure in Asian persons, but only a
relatively small number of trials were conducted in persons
of these ethnic backgrounds. In addition, all but
one trial included in our meta-analysis used an open
design. However, many trials masked observers who obtained
blood pressure measures. We identified a potential
publication bias by funnel-plot method. Several
large trials reported a moderate reduction in blood pressure.
However, blood pressure was not the primary outcome
of interest in most of these trials (23, 26, 30).
Furthermore, in a formal test using the “trim and fill”
method, reductions in systolic and diastolic blood pressure
remained unchanged.
The underlying mechanism or mechanisms responsible
for an exercise-induced reduction in blood pressure
remain unclear. Recent evidence shows that insulin resistance
and hyperinsulinemia may contribute to the
pathogenesis of hypertension (56, 57). Clinical trials
have shown that aerobic exercise reduces insulin resistance
and insulin levels in hypertensive patients (58,
59). Brett and associates (60) have reported that change
in blood pressure during exercise is strongly associated
with reduction in serum concentrations of to