Nine RCTs (1,190 patients, sample size range 38 to 287) were included in the review. For quality, four trials scored 5 points on the Jadad scale, three trials scored 3, and two trials scored 2. Allocation concealment was adequate in seven trials. Blinding was conducted in four trials. Intention-to-treat analysis was conducted in five trials. Follow-up ranged from one month to two years. Loss to follow-up was high in three trials.
Compared with glucose for peritoneal dialysis long-dwell exchange, icodextrin was associated with a significantly higher ultrafiltration efficiency ratio (MD 6.84, 95% CI 4.43 to 9.25; two trials; Ι²=0%), higher peritoneal clearances of creatinine (MD 0.51, 95% CI 0.35 to 0.67; five trials; Ι²=0%) and higher peritoneal clearances of urea nitrogen (MD 0.43, 95% CI 0.26 to 0.61; four trials; Ι²=0%). There was no significant difference between groups for preservation of residual renal function (four trials; Ι²=34%).
Icodextrin was associated with significant increases in net ultrafiltration compared to glucose concentrations of 1.5%, 2.5% and 4.25%; it was also superior to glucose in high-average/high and low-average transporters, but not low transporters (results reported in the paper).
Negative net ultrafiltration was significantly lower in the icodextrin group (RR 0.13, 95% CI 0.07 to 0.25; four trials; Ι²=18.2%).
The only statistically significant safety outcome was the increased occurrence of rash in the icodextrin group (RR 2.41, 95% CI 1.42 to 4.10; four trials; Ι²=0%).
There was insufficient information to explore the influence of diabetic status.
Sensitivity analysis did not alter the main findings.
There was no evidence of publication bias.