Pulse high-volume haemofiltration for treatment of severe sepsis: effects on hemodynamics and survival

Ratanarat R. et al. Critical Care 2005; 9: R294-R302

Introduction

Severe sepsis remains the leading cause of death in the ICU. In spite of attempts to improve survival with innovative strategies based on our increasing knowledge of the complex pathophysiology of sepsis, mortality remains high, ranging from 30% to 50%. Continuous High-volume haemofiltration (HVHF) is a potential promising therapeutic approach, and appears to improve hemodynamics with decreased vasopressor requirements in human studies. However, continuous HVHF is an expensive and labour-intensive intervention. The current study evaluates a potential alternative namely pulse HVHF (PHVHF)

Study

This is a prospective interventional study in fifteen patients with severe sepsis or septic shock receiving continuous renal replacement therapy. All patients were treated according to the guidelines of the European Society of Intensive Care Medicine (Surviving Sepsis campaign). PHVHF was performed through a 14-Fr central venous haemodialysis catheter (flow rates 250-300 ml/min) using a UF rate of 85 ml/kg/h for 6 hours per day, followed by CVVH using a UF rate of 35 ml/kg/h for 18 hours. Changes in hemodynamics and the required dose of noradrenaline to maintain the mean arterial pressure above 70 mm Hg were measured during and after PHVHF. APACHE II, SAPS II and SOFA scores were calculated as well as the expected mortality rates.

Fifty PHVHF treatments were performed with a mean number of treatments per patient of 3.4 (1-9). No treatment was prematurely discontinued due to clotting or catheter problems. Observed hospital mortality was 46.7% with a standardized mortality ratio of 0.65 (0.48-0.87) calculated from the APACHE II score. The mean number of PHVHF treatments per patient was significantly higher in the survivors compared the non-survivors (4.8  ± 2.7 vs. 1.9  ± 0.7, P=0.02) The required dose of noradrenaline for the maintenance of the target MAP showed a small but significant decrease during PHVHF and this reduction was maintained in the 12 hours thereafter (p=0.001). This was accompanied by a slight increase in systolic blood pressure (124  ± 26 mm Hg vs. 133  ± 24 mm Hg, P=0.04) and heart rate 97.3  ± 25.53 vs. 100  ± 21.8, P=0.04) There were no changes in other hemodynamic parameters, body temperature, oxygenation, arterial pH or urine output during the observation period. Net fluid balance showed no change during PHVHF as compared to the days before initiation of PHVHF (positive fluid balance of 1374  ± 2618 ml/day vs. 1514  ± 2548 ml/day, P=0.9). Daily Kt/V was 1.92  ± 0.29.

Discussion

In this pilot study, the authors show that PHVHF is a feasible and effective (in terms of renal replacement therapy) therapeutic modality in patients with severe septic shock already requiring CRRT. As the number of included patients is small and the CRRT experience of this research group is extensive, safety issues remain unanswered. The potential hemodynamic benefits appear to be small. Due to the design of the study, the suggested beneficial effect on noradrenaline requirements can be confounded by time and the dosage of other vasoactive drugs (median number of concurrently administered drugs was 2). The low standardized mortality ratio is interesting but can be explained by many other factors. The authors are right in concluding that in terms of mortality, the only way to demonstrate the effect of PHVHF in this population is to conduct a prospective randomized, controlled study on a larger scale.

M. Moviat, fellow Intensive Care
J.G. van der Hoeven, Internist-intensivist