Correspondence University of Pittsburgh, School of Medicine, Department of Critical Care Medicine, 608 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA

Email kellumja@ccm.upmc.edu

Original article

Saudan P et al. (2006) Adding a dialysis dose to continuous hemofiltration increases survival in patients with acute renal failure. Kidney Int 70: 1312–1317   PubMed

Synopsis

Background

Continuous venovenous hemofiltration (CVVH) removes solutes from the blood by convection; continuous venovenous hemodiafiltration (CVVHDF) uses both convection and diffusion to remove solutes.

Objective

To establish whether increasing the dialysis dose, by using CVVHDF rather than CVVH, improves survival of patients with acute renal failure (ARF) in intensive care.

Design

The medical and surgical intensive care units of the Geneva University Hospitals provided the patients for this prospective, randomized trial. Individuals with urine output <200 ml/12 h despite fluid replacement and intravenous diuretics, and/or with blood urea nitrogen >30 mmol/l (>84 mg/dl) and urine output <1,500 ml/12 h, were eligible to enter the study. Patients with prerenal or postrenal failure, end-stage renal disease or suspected glomerular disease were among those excluded. Analysis was by intention to treat.

Intervention

Participants were randomized to receive CVVH or CVVHDF, via a dual-lumen central venous catheter, for 48 h. If possible, patients were then observed for 24 h; renal replacement therapy (RRT) was administered for another 48 h if blood urea nitrogen increased and/or oliguria persisted. Ultrafiltration flow rates, calculated as the rate required to clear the patient's estimated urea distribution volume in 24 h, were between 1 and 2.5 l/h. For CVVHDF, the dialysate flow rate was 1–1.5 l/h. Severe hypotension or irreversible lactic acidosis prompted cessation of RRT.

Outcome measures

The primary end points were 28-day and 90-day survival.

Results

Of the 371 patients with ARF who were undergoing continuous RRT at the participating centers during the period October 2000 to December 2003, 206 underwent randomization (102 to CVVH and 104 to CVVHDF). Thirteen patients in each group did not complete treatment but were included in the analysis. Initial plasma creatinine level was significantly higher in the CVVHDF patients than in the CVVH patients (P <0.03); other baseline clinical and demographic characteristics did not differ significantly between the two groups (mean age 62 years vs 65 years, and mean APACHE II score 24 vs 26, respectively). The mean prescribed ultrafiltration dose was 25 ml/kg/h in the CVVH group and 24 ml/kg/h in the CVVHDF group, and the patients on CVVHDF were prescribed a mean dialysis dose of 18 ml/kg/h. The CVVHDF patients had significantly higher mean urea and creatinine reduction ratios 48 h after the initiation of continuous RRT than did the CVVH patients (50% vs 40%, P <0.009, and 46% vs 38%, P <0.014, respectively). Survival rates at 28 days and 90 days were higher with CVVHDF than with CVVH (59% vs 39%, P = 0.03, and 59% vs 34%, P = 0.0005, respectively). Cox proportional hazard modeling revealed that use of CVVHDF versus CVVH was an independent predictor of survival at 90 days (adjusted hazard ratio for death 0.59, 95% CI 0.40–0.87; P = 0.008). There were no significant differences in the incidences of major bleeding or filter clotting between the groups.

Conclusion

Adding a dialysis dose to CVVH might improve survival of patients with ARF in intensive care.

Commentary

The study by Saudan and colleagues comes at an interesting time. In 2002, an expert panel concluded that there was insufficient evidence to recommend a dose of RRT for ARF beyond what is recommended for chronic disease.¹ At that time the number of studies was limited and results were conflicting.^{2, 3, 4} Shortly afterward two multicenter trials were devised, one in the US and the other in Australia. These efforts ultimately became the ATN Study and the RENAL trial, respectively, both of which are currently underway.⁵ Pre-trial surveys of prospective study sites in both trials indicated that a specific dose of RRT had not been adopted in clinical practice. In short, there was equipoise.

The principle of equipoise demands that participants in biomedical research should be randomized only if there is an equal likelihood that either treatment would be of most benefit. In reality, however, there is almost never an exactly equal chance. Instead, there is insufficient evidence that one therapy is better than another, combined with an expectation that a new or different therapy will prove superior. Although research is conducted under this principle, clinical practice never is. Even if there is substantial uncertainty about the best treatment, clinicians never just 'flip a coin'. The standard for choosing a therapy is to use the one for which the best evidence for efficacy is available. This evidence might well be insufficient to establish one therapy as superior, but it is nonetheless the basis for clinical decision-making.

The question for us now is whether this latest study should change clinical practice. Should we wait until the ongoing studies are complete, as suggested by one observer?⁵ Should we adopt augmented dosing in the meantime? Or should we even consider halting ongoing trials? The four existing studies have all had different designs and compared different strategies. Like previous trials, the Saudan et al. study is underpowered; furthermore, it confounds the effects of dose and technique by adding dialysis to filtration. Nevertheless, pooled results from all four studies indicate a very large effect on survival in favor of augmented dosing, with an odds ratio of 1.95 (95% CI 1.48–2.58, P <0.001; see Supplementary Figure 1 online). The results are identical regardless of whether a fixed or random effects model is used, and the study results are homogeneous (Q statistic 1.73; P = 0.63).

Although these data may still not be definitive, the best evidence to date supports the use of at least 35 ml/kg/h for CVVH, CVVHDF, or daily hemodialysis. Lower doses of RRT might still be effective and indeed could be safer, but their use should now be limited to research protocols that actively test this hypothesis. When the ATN and RENAL studies were designed, augmented dosing was an experimental therapy with possible benefits, but also possible harm. Given the evidence that has accumulated since, lower doses, previously thought of as standard, are now the experimental therapy. The exact method with which to achieve an augmented dose, and indeed what the correct dose should be, remain matters of clinical judgment. Nevertheless, our response to the evidence now available should be either to enroll patients in one of the ongoing studies or to adopt a higher dose. We may still have scientific equipoise, but our clinical recommendations outside research should change.

Acknowledgments

The synopsis was written by Chloë Harman, Associate Editor, Nature Clinical Practice.

References

1. Kellum JA et al. (2002) The first international consensus conference on continuous renal replacement therapy. Kidney Int 62: 1855–1863 | PubMed |
2. Ronco C et al. (2000) Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: a prospective randomised trial. Lancet 356: 26–30 | Article | PubMed | ISI | ChemPort |
3. Schiffl H et al. (2002) Daily hemodialysis and the outcome of acute renal failure. N Engl J Med 346: 305–310 | Article | PubMed | ISI |
4. Bouman CS et al. (2002) Effects of early high-volume continuous venovenous hemofiltration on survival and recovery of renal function in intensive care patients with acute renal failure: a prospective, randomized trial. Crit Care Med 30: 2205–2211 | Article | PubMed |
5. Bellomo R (2006) Do we know the optimal dose for renal replacement therapy in the intensive care unit? Kidney Int 70: 1202–1204 | Article | PubMed | ChemPort |

Supplementary information

Supplementary Figure 1 (ppt 24 KB)

Forest plot pooling trials of RRT dose

Contact the journal about this article

Subject areas under which this article appears: Acute renal failure | Dialysis (hemodialysis, peritoneal dialysis, continuous renal replacement)