Renal replacement therapy in the intensive care unit

Indian Journal of Critical Care Medicine has a great review article titled Renal replacement therapy in the intensive care unit by Jose Chacko. The article covers all the big hitters: IHD, CVVHD, CVVH, SCUF, and SLEDD. The most interesting part to me:

Continuous Vs Intermittent therapies - what's the evidence?

The physiological superiority of CRRT over conventional IHD is unquestioned. Does this result in improved clinical outcomes? Many trials that compare these modalities involve patients with different severity of illness at baseline; crossover from one arm to the other was also allowed in many studies, making it difficult to interpret the results. Besides, the majority of trials do not include patients with significant haemodynamic instability - precisely the subgroup of patients who are likely to have a survival advantage with continuous therapies. Kellum et al did a meta-analysis of 13 clinical trials involving 1400 patients. Out of this, only 3 were randomised. On unadjusted analysis, there was no difference survival between CRRT and IHD. However, when adjusted for study quality or baseline severity of illness, or both, CRRT was associated with improved survival. Under no conditions, either of inclusion criteria or adjustment method did CRRT fare worse. Mehta et al, in their study, randomised 166 patients to receive either IHD or CRRT. CRRT was associated with increased ICU and hospital mortality. However, patients in the CRRT group had a higher baseline severity of illness by APACHE II and III scores, more hepatic failure and more organ failures. There was strong bias in favour of IHD by all these counts, thus obviating meaningful analysis. In a more recent trial, Vinsonneau et al, for the haemodiafe group compared alternate day IHD with CRRT and reported no survival benefit at 60 days. However, the trial did not standardise the time of initiation of therapy or the dose of delivered dialysis. The actual CRRT dose delivered was only 25 ml/kg/hr, significantly less than the optimal 35 ml/kg/hr that has been associated with improved outcomes.

Chacko J. Renal replacement therapy in the intensive care unit. Indian J Crit Care Med 2008;12:174-80