Polyethylene glycol vs. sodium phosphate for bowel preparation: a treatment arm meta-analysis of randomized controlled trials.

Background: Results of meta-analyses of randomized trials comparing PEG and NaP are inconsistent and have not included trials comparing either or both preps to less traditional ones.

Aim: To perform a meta-analysis by treatment arm.

Methods: Using MEDLINE and EMBASE, we identified English-language trials published from 1990 to 2008 that included PEG and/or NaP, and aggregated them by treatment arm into: 4 liter (L) PEG; 2 L PEG; split-dose PEG; two 45 ml doses of NaP +/- adjunctive medication; and NaP tablets. We compared prep quality and the proportion completing the prep.

Results: Among 71 trials (patient N = 10,201), excellent prep quality was present in 34% (CI, 26-41%) for 4 L PEG alone; 39% (CI, 26-51%) for 2 L PEG; 37% (CI, 28-46%) for split-dose PEG; 42% (CI, 33-51%) for NaP solution; 44% (CI, 38-51%) for NaP with adjunctive meds; and 58% (CI, 49-67%) for NaP tablets. Patients receiving NaP were more likely to complete the prep (97% [CI, 96-98%] vs. 90% [CI, 87-92%] for 4L PEG alone); however, completion rates for 2L PEG (98%) and split dose PEG (95%) were similar to NaP.

Conclusions: NaP tablets resulted in better prep quality and higher completion rates compared to other regimens. In comparisons limited by sample size, split dose PEG was not statistically different from NaP solution for completion rate or prep quality.