Abstract

Background:

Continuous renal replacement therapy (CRRT) is the preferred mode of dialysis in critically ill hemodynamically unstable patients and in addition removes the inflammatory cytokines that accumulate in sepsis. Sustained low-efficiency dialysis (SLED) is a hybrid modality of intermittent dialysis with the advantage of metabolic control, hemodynamic stability, at a reduced cost. We aimed to compare CRRT versus SLED in patients with cirrhosis and sepsis-associated AKI(SA-AKI) in preventing the incidence of intradialytic hypotension (IDH).

Methods:

Prospective single-center open-label randomized controlled trial wherein critically ill patients with cirrhosis (CICs) with SA-AKI and norepinephrine dose <0.10 ug/kg/min) were randomized to CRRT versus SLED. Pre and post (8 hours) rotational thromboelastometry (ROTEM,n=62) and inflammatory cytokines was performed in 20 patients (10-CRRT, 10-SLED)

Results:

A total of 62 patients were randomized to SLED (n=31) vs. CRRT (n=31). The baseline characteristics were comparable. The mean age was 45.7± 10.8 years, 95% males, 81% alcohol-related, with arterial lactate 2.3±1.9 umol/L ,serum creatinine 3.1 ± 2.1 mg/dl, MELD 53.4 ±9.7 and SOFA score 12.6±2.9. The indication for dialysis was metabolic acidosis in 90%. The norepinephrine (NE) dose was comparable at baseline [(0.06 ±103) vs (0.7 ±10.3) ug/kg/min; p=0.13). On intention-to-treat analysis, the development of IDH was lower in CRRT [ 11(35.5%) vs. 22(71%);p=0.01). There was no difference in 28-day mortality [(12(38.7%) vs.14(45.2%); p=0.19], the recovery of AKI at day 14 [11 (35.5%) vs. 7 (22.6%); p=0.40] but reduction in SOFA score by 2 points was higher in CRRT[13 (41.9%)vs.4(12.9%); p=0.02] with a significant lower NE dose at day 2 [(0.9±0.16) vs. (0.17±0.11) ug/kg/min ;p=0.03) vs. SLED respectively. A significant reduction in arterial lactate from baseline [(2.1±1.9 to 1.9±1.1; p=0.047) was noted in CRRT vs. SLED (2.0±1.9 to 2.2±1.8; p=0.33)]. The days of mechanical ventilation and ICU stay were not different. There were 7(22.5%) patients with protocol violation in SLED group who required CRRT due to IDH. The median duration of CRRT was 48 (IQR 15-116 hours) while the mean number of SLED sessions were 3±2.1. A significant derangement in coagulation by ROTEM was noted more with SLED. Inflammatory cytokines (serum ferritin, c-reactive protein) showed significant reduction with CRRT vs. SLED while levels of interleukin-6 increased after CRRT. (Figure)

Conclusion:

CRRT compared to SLED is a better modality which lowers the incidence of IDH achieves better lactate clearance and reduction in SOFA scores in cirrhosis patients with SA-AKI but does not confer any survival benefit. CRRT is possibly more effective in clearing inflammatory cytokines and causes lesser impairment in coagulation compared to SLED. Large controlled trials are required for exploring the observed benefits of CRRT in CICs.