Elsevier

The Lancet

Volume 354, Issue 9187, 16 October 1999, Pages 1337-1341
The Lancet

Articles
Haemodialysis-membrane biocompatibility and mortality of patients with dialysis-dependent acute renal failure: a prospective randomised multicentre trial

https://doi.org/10.1016/S0140-6736(99)01213-1Get rights and content

Summary

Background

There is controversy as to whether haemodialysismembrane biocompatibility (ie, the potential to activate complement and neutrophils) influences mortality of patients with acute renal failure. We did a prospective randomised multicentre trial in patients with dialysis-dependent acute renal failure treated with two different types of low-flux membrane.

Methods

180 patients with acute renal failure were randomly assigned bioincompatible Cuprophan (n=90) or polymethylmethacrylate (n=90) membranes. The main outcome was survival 14 days after the end of therapy (treatment success). Odds ratios for survival were calculated and the two groups were compared by Fisher's exact test. Analyses were based on patients treated according to protocol (76 Cuprophan, 84 polymethyl methacrylate).

Findings

At the start of dialysis, the groups did not differ significantly in age, sex, severity of illness (as calculated by APACHE II scores), prevalence of oliguria, or biochemical measures of acute renal failure. 44 patients (58% [95% CI 46–69]) assigned Cuprophan membranes and 50 patients (60% [48–70]) assigned polymethyl-methacrylate membranes survived. The odds ratio for treatment failure on Cuprophan compared with polymethyl-methacrylate membranes was 1·07 (0·54–2·11; p=0·87). No difference between Cuprophan and polymethyl-methacrylate membranes was detected when the analysis was adjusted for age and APACHE II score. 18 patients in the Cuprophan group and 20 in the polymethyl-methacrylate group had clinical complications of therapy (mainly hypotension).

Interpretation

There were no differences in outcome for patients with dialysis-dependent acute renal failure between those treated with Cuprophan membranes and those treated with polymethyl-methacrylate membranes.

Introduction

Despite various improvements in diagnosis, monitoring, and treatment, acute renal failure is associated with high mortality,13 particularly in patients requiring haemodialysis.4, 5, 6 Although septicaemia and multiorgan failure are generally believed to be the main determinants of outcome in patients with acute renal failure,7 clinical studies have shown that dialysis modality (ie, choice of dialysis membrane) may influence clinical course and mortality. Hakim and colleagues8 prospectively studied 72 patients with acute renal failure requiring haemodialysis, and suggested that the biocompatibility of dialysis membranes had a substantial impact on the recovery of renal function, and, in patients who were not oliguric at the start of dialysis, on mortality. Schiffl and colleagues9 studied 52 patients with postoperative acute renal failure, and reported significantly higher mortality and delayed renal recovery in patients treated with Cuprophan (bioincompatible) low-flux haemodialysers than in those treated with high-flux polyacrilonitrile membranes. These findings, if true, would necessitate a revision of current dialysis for acute renal failure and have repercussions on treatment strategies for chronic haemodialysis in which patients are repeatedly exposed to dialysis membranes for long periods. Both studies have, however, set off a debate among dialysis specialists because of suspected methodological flaws, statistical insufficiencies, and bias.10, 11, 12, 13, 14 Moreover, several similar studies have been unable to confirm these results,15, 16, 17 leaving this important question unresolved.

We therefore did a prospective, randomised, open-label multicentre comparison of the effects on the mortality of patients with dialysis-dependent acute renal failure of two dialysis membranes with different biocompatibility properties in terms of the potential to activate complement and neutrophils.

Section snippets

Patients

We enrolled all patients over 18 years of age who had acute renal failure that necessitated dialysis and who were admitted to any of the study centres (Berlin, Germany; Skopje, Republic of Macedonia; Gdansk, Poland; Göttingen, Germany; and Antwerp, Belgium). Criteria for dialysis were as follows: clinical symptoms of uraemia (ie, uraemic pericarditis, nausea, vomiting, or neurological abnormalities); oliguria and fluid overload; laboratory findings (facultative), including blood urea nitrogen

Design

The study was approved by the institutional review boards of the participating centres. Informed consent was obtained from all patients or their next of kin. Once the nephrologist had decided that the patient needed haemodialysis, he or she assigned treatment (by external telephone randomisation) with one of the membranes. To assign treatment, the nephrologist called a telephone number at the Department of Biometry and Medical Documentation, University of Ulm, Germany, and was connected to a

Statistical analysis

The study was monitored by the University of Ulm, Germany. All data were collected and collated at the Biometrical Department in Ulm, where data were double-typed and compared by computer. Statistical analysis was done by SAS (version 6·12) and StatXact (version 3).

We calculated that with 75 patients in each group the study would establish 95% CIs for population mortality no larger than observed mortality plus or minus 12%. To compare proportions between the two groups, odds ratios were used as

Results

Between June 12, 1996 and Oct 30, 1997, 180 patients with acute renal failure requiring haemodialysis were assigned treatment (90 Cuprophan and 90 polymethyl methacrylate). 104 patients were recruited in Berlin, 32 in Skopje, 22 in Gdansk, 17 in Göttingen, and five in Antwerp. 20 patients were excluded from the analysis: nine were not eligible for the study (three did not have acute renal failure, four had cancer, one had chronic renal disease, and one had received continuous venovenous

Discussion

Because of the high mortality associated with acute renal failure, every advance in the management and treatment of these patients must be thoroughly investigated. Some studies have suggested that use of a high-complementactivating Cuprophan membrane may be associated with increased morbidity and mortality,89 which is surprising and alarming since this membrane has been used successfully for decades to treat patients with acute and chronic renal failure. If true, the hypothesis that membrane

References (21)

There are more references available in the full text version of this article.

Cited by (0)

Members listed at end of paper

View full text