Objective: To estimate whether continuous veno-venous hemodiafiltration (CVVHDF) is superior to intermittent hemodialysis (IHD) in terms of survival of adult patients with acute renal failure (ARF) admitted to the Intensive Care Unit.
Selection of studies: Controlled clinical trials (CCT) and systematic reviews comparing CWHDF and IHD for managing ARF in adult patients (age > 19 years). Observational and case series were excluded. SEARCH SOURCES: The basic syntax <<hemodiafiltration OR continuous hemodi* OR continuous dialysis) and acute renal failure and acute renal insuficiency>> was used to search Pub Med and Ovid System databases. A manual search was done by reviewing the references in the corresponding topic of UpToDate.
Analysis: Data were extracted by two author and their methodological quality was assessed according to the Cochrane Renal Group recommendations that include the procedure for assigning, blinding, intention to treat analysis, and follow-up. OUTCOMES VARIABLES: All data relating to mortality were extracted, specifying the time of collection, time and circumstances (mortality in the ICU or hospitalization). Values gathered are expressed as mortality rates in both the experimental group (CVVHDF) and the control group (IHD), indicating the absolute risk reduction (ARR) and its 95% confidence interval. OUTCOMES AGGREGATION: Studies meeting clinical and methodological homogeneity criteria were combined with the fix effect model by using the Review Manager tool from Cochrane Collaboration. Methodological heterogeneity was analyzed by using the chi-squared test for n-1 freedom degrees, with an alpha value of 0.05. A sensitivity analysis was done adjusting for methodological quality to confirm the results obtained.
Results: Seven clinical trials directly comparing the survival of severe ARF patients in a prospective, randomized, and controlled way were identifiec. Almost all published estudies have quality problems because of being too small to study survival rates, treatment allocation problems and high numbers of loss to follow-up, differences in initial severity levels, or to premature study closure. When combining the results, it was observed that mortality was 64% for IHD and 65% for CVVHDF, with a relative risk of 0.98 (95% CI 0.89-1.07), p = 0.65, with no statistically significant heterogeneity between studies included. When excluding from the analysis the most questionable study due to selection bias, high loss to follow-up (21%), and baseline differences in co-variables influencing the study outcomes, the results are not changed, the observed mortality was 67% for extra-renal intermittent depurative techniques versus 65% for continous ones, with a relative risk of 1.03 (95% CI 0.94-1.14), p = 0.54, again with no statistically significant heterogeneity between studies included.
Conclusion: CVVHDF does not offer any benefit as compared to IHD in terms of survival and according to available data from the literature. However, continuous techniques bring other potential benefits such as hemodynamic stability, better tolerability of ultrafiltration, and depuration of solutes, which merit a systematic review to estimate and quantify their magnitude, and which would allow for better defining their place in the therapeutic armamentarium available for this high-mortality condition.