A total of 24 RCTs met the inclusion criteria, of which one trial had two treatment groups that were compared with an identical placebo control group. Therefore, there were 25 treatment strata overall, with a total of 1,404 participants.
The meta-analysis showed a non-statistically significant effect of fibre supplementation on systolic BP of -1.13 mmHg (95% confidence interval, CI: -2.49, 0.23) and a statistically significant effect on diastolic BP of -1.26 mmHg (95% CI: -2.04, -0.48). Univariate subgroup analyses showed that systolic BP reduction was statistically significantly greater in those over 40 years of age, compared with those 40 years or younger (P=0.001), with overall effects of -3.08 mmHg (95% CI: -4.61, -1.56) and 0.35 mmHg (95% CI:-1.00, 1.69), respectively. It was also shown in univariate subgroup analyses that BP reduction was statistically significantly greater in hypertensive populations than normotensive populations. The overall effects on systolic BP for hypertensive and normotensive populations were -4.53 mmHg (95% CI: -6.69, -2.38) and -0.23 mmHg (95% CI: -1.43, 0.98), respectively (P<0.001), while the overall changes in diastolic BP for the same subgroups were -2.37 mmHg (95% CI: -3.56, -1.19) and -0.68 mmHg (95% CI: -1.44, 0.09), respectively (P=0.02). No other subgroup or sensitivity analyses showed any statistically significant effect of fibre supplementation on BP.
The funnel plot showed that there was a possibility that more small trials showing large reductions in BP were identified and included in the systematic review. Executing the trim-and-fill method revealed that one trial might have been missing from the review. When incorporating the results from this theoretical trial, the overall effect of dietary fibre on systolic BP was attenuated to -0.94 mmHg (95% CI: -2.34, 0.46)