Eight RCTs (1,315 participants randomised;1,286 participants analysed) were included in the review.
For all three outcomes, the heterogeneity P-values were statistically significant: these ranged from 0.048 to less than 0.005, indicating greater-than-expected variation among the trial results and implying that aggregating the quantitative data may be problematic.
Failure to complete preparation (favours PEG).
For all studies (8 studies; n=1,286), the RR was 0.23 (95% confidence interval, CI: 0.18, 0.28) and the NNT was 7 (95% CI: 5, 12). When excluding any outliers (7 studies; n=1,214), the RR was 0.20 (95% CI: 0.16, 0.24) and the NNT was 7 (95% CI: 5, 9). The best estimate (5 trials; n=842) RR was 0.25 (95% CI: 0.20, 0.30) and the NNT was 7 (95% CI: 5, 10).
Adequate preparation (favours NaP).
For all studies (8 studies; n=1,286), the RR was 1.19 (95% CI: 1.09, 1.32) and the NNT was 9 (95% CI: 5, 625). When excluding any outliers (5 studies; n=685), the RR was 1.02 (95% CI: 0.99, 1.05); the NNT was not calculated. The best estimate (3 studies; n=313) RR was 1.06 (95% CI: 0.95, 1.19); the NNT was not calculated.
Excellent preparation (favours NaP).
For all studies (7 studies; n=1,068), the RR was 1.28 (95% CI: 1.11, 1.48) and the NNT was 10 (95% CI: 5, 53). When excluding any outliers (5 trials; n=644), the RR was 1.24 (95% CI: 1.02, 1.52) and the NNT was 11 (95% CI: 7, 26). The best estimate (3 trials; n=272) RR was 1.72 (95% CI: 1.16, 2.53) and the NNT was 5 (95% CI: 4, 7).
Patient tolerance and side-effects.
There were no consistent, clinically important differences in symptom side-effects. However, because of variation in the extent, detail and method of measuring the side-effects, a clinically meaningful direct quantitative comparison was not possible. In one study, orthostasis was more common among patients using PEG. There were also consistent, transient increases in serum phosphorous and sodium, and in two studies a transient decrease in calcium, although no clinical adverse side-effects accompanied these changes.