Thirty-three studies were included (n=8,933): nine RCTs (n=1,635); 18 non-randomised studies (n=6,436); and six retrospective studies (n=861).
The authors stated that in all studies severity scores of patients were similar between treatment groups. None of the studies were blinded. Three reported intention-to-treat analysis.
Overall studies, CRRT was associated with a statistically significant reduction in survival compared to intermittent haemodialysis (OR 0.77, 95% CI 0.62 to 0.95; 33 studies). For RCTs, there was no statistically significant difference in mortality between CRRT and intermittent haemodialysis (OR 0.89, 95% CI 0.63 to 1.24, p=0.95; nine studies). Random-effects models were used for both these analyses.
For all studies published after 2000 and for the seven RCTs published after 2000, CRRT was associated with a statistically significant reduction in survival compared to intermittent haemodialysis (OR for all studies was 0.61, 95% CI 0.50 to 0.74; six studies and for RCTs OR was 0.72, 95% CI 0.58 to 0.90; seven studies; fixed-effect model).
For all studies, CRRT was associated with a statistically significant reduction in dialysis dependence (OR 0.44, 95% CI 0.34 to 0.58; nine studies). For RCTs (four studies), there was no statistically significant difference in dialysis dependence.