Old Wine in New Bottles: Continuous Versus Intermittent Renal Replacement Therapy in the ICU*
In this issue of Critical Care Medicine, Bonnassieux et al (3) report the results of another retrospective cohort study, using data from a large database derived from 291 French ICUs. Patients treated for AKI from 2010 until 2014 were grouped according to the initial modality of treatment: CRRT or intermittent hemodialysis (IHD). Renal recovery was defined as more than 3 days without RRT before discharge. In this observational study, IHD was associated with a lower likelihood of recovery after hospital discharge, with an odds ratio of 0.91 (95% CI, 0.88–0.92), suggesting that CRRT does offer a small but significant therapeutic benefit in terms of renal recovery.
However, this suggestion can be contested. First, it is a matter of debate whether not needing RRT for more than 3 days before discharge covers renal recovery in the long run. Second, the investigators extracted only the initial RRT modality from the database (IHD or CRRT), and details on blood and dialysate flow rates, dialysis dose, and fluid balance are missing. It is therefore impossible to determine whether the two groups were comparable in terms of these variables. This may be important, since during IHD using low blood and dialysate flow rates (prolonged IRRT [PIRRT] or slow low efficiency dialysis [SLED]), hemodynamic stability has been demonstrated to be as well preserved as during CRRT (4, 5).
In recent years, the use of PIRRT has gained popularity because of its greater efficiency in small solute removal combined with lower costs in comparison with CRRT (6). In addition, PIRRT reduces the need of anticoagulation and the treatment can be performed during the night, allowing the patient to be mobilized for physical therapy as well as for diagnostic and therapeutic procedures during the day (7). Recently, a systematic review and meta-analysis comparing the effect of SLED and CRRT on renal recovery after AKI in the ICU was published (8). Eighteen studies were included, four RCTs and 14 observational studies. Renal recovery was defined as not requiring RRT. Meta-analysis results indicated no significant difference between SLED and CRRT in terms of renal recovery and time to renal recovery in surviving patients.