The authors concluded that, compared with standard dialysis, reducing the temperature of the dialysate lowers the incidence of intradialytic hypotension without reducing dialysis adequacy in patients on chronic dialysis. However, the effects on cold symptoms are unknown. Overall, this was a well-conducted review and the authors' conclusions are likely to be reliable.

To evaluate the effects of cool haemodialysis on intradialytic hypotension (IDH), thermal symptoms and clearance of small solutes.

The Cochrane CENTRAL Register, MEDLINE, EMBASE, CINAHL, four named renal journals and four named databases of ongoing trials were searched; the search terms were reported. In addition, the reference lists of reviews and full-text versions of included studies were screened. Published and unpublished studies were eligible but they had to be reported in English.

Study designs of evaluations included in the review

Blinded and non-blinded randomised controlled trials (RCTs) were eligible for inclusion. All of the included studies were of a crossover design. The duration of most of the included studies was two sessions; other studies evaluated interventions over 4 to 24 sessions.

Specific interventions included in the review

Studies that compared any method of reducing the dialysate temperature with standard dialysis were eligible for inclusion. The studies had to use a bicarbonate-based dialysate. Studies that used dialysis methods enhanced by convection were excluded. The included studies evaluated fixed empirical reductions in dialysis temperature and programmed cooling or isothermic dialysis using biofeedback temperature-control devices (BTM). Where reported, temperatures ranged from 34.0 to 35.7 degrees C in the intervention groups and from 36.5 to 38.5 degrees C in the control groups. Most of the included studies evaluated a standard duration of dialysis (approximately 4 hours); in two studies the duration was 1 hour.

Participants included in the review

Studies of adult patients (aged older than 18 years) who were stable or unstable on chronic haemodialysis for end-stage renal disease were eligible for inclusion. Where reported, the mean age of participants in the included studies ranged from 55 to 69 years.

Outcomes assessed in the review

The primary review outcomes were frequency of IDH, change in blood-pressure (BP) with dialysis, frequency and severity of cold sensation during dialysis, and adequacy of dialysis based on measures of urea clearance. The secondary review outcomes were changes in systemic haemodynamics during dialysis and changes in body temperature.

How were decisions on the relevance of primary studies made?

One reviewer screened titles and abstracts. Two reviewers independently selected studies from full-text articles and resolved any disagreements through discussion.

Two reviewers independently assessed the validity of the studies by considering the adequacy of allocation concealment (graded using the Cochrane scoring system), the Jadad criteria (randomisation, adequacy of blinding, and the number and handling of drop-outs), and potential confounders specific to studies of haemodialysis. Any disagreements were resolved through discussion.

The data were extracted using a pre-designed form, but the number of reviewers involved in the process was not reported.

How were the studies combined?

The studies were grouped by outcome. Studies reporting comparable data were pooled using a random-effects meta-analysis. Pooled rate ratios and 95% confidence intervals (CIs) were calculated for data reported as rates, while pooled weighted mean differences (WMDs) and 95% CIs were calculated for continuous data.

How were differences between studies investigated?

Statistical heterogeneity was assessed using the chi-squared and I-squared statistics.

A subgroup analysis was used to examine the effects of the method used to reduce dialysate temperature (fixed empirical reduction or BTM). The effects of cool dialysis on IDH and BP was also evaluated according to the intradialytic stability of patients; pooled IDH rates were calculated after excluding two studies that reported zero events in the intervention group.

Twenty-two crossover RCTs were included (n=408): 16 studies (n=235) evaluated fixed empirical temperature reductions and 6 studies (n=173) evaluated BTM.

One study reported adequate allocation concealment. Nineteen studies scored 2 out of a maximum of 5 on the Jadad scale for quality; the remaining three studies scored 1, 3 and 4, respectively. Though randomisation was generally adequate, most of the studies did not use blinding; only one was double-blinded. Most of the studies were short term and reported no drop-outs; the 4 studies that reported drop-outs only analysed data from patients who completed the study.

IDH.

IDH was 7.1 times less common (95% CI: 5.3, 8.9) with cool dialysate than with standard dialysis (based on 8 studies). Statistically significant heterogeneity was found (chi-squared p<0.00001, I-squared 99.3%). The reduction in IDH was found for both fixed reductions in temperature (IDH rate 9.5 times lower, 95% CI: 6.7, 12.4; based on 6 studies; significant heterogeneity was detected, p<0.00001) and BTM (IDH rate 2.0 times lower, 95% CI: 1.9, 2.1; based on 2 studies; no significant heterogeneity). Reductions in IDH rate were smaller, but remained significant, when the 2 studies that reported zero events in the intervention group were excluded.

BP.

Compared with standard dialysis, cool dialysis was associated with significantly higher post-dialysis mean arterial pressure (MAP) (WMD of MAP 11.3 mmHg, 95% CI: 7.7, 15.0; based on 10 studies) and a significantly smaller delta MAP (WMD 7.1 mmHg, 95% CI: 4.9, 9.4; based on 7 studies). The treatment difference was no longer statistically significant when only studies evaluating BTM were pooled.

Symptoms.

Reporting of symptoms during dialysis was generally poor. The 6 studies reporting symptoms presented no consistent ratings of severity and did not clearly report the frequency of symptoms.

Adequacy of dialysis.

There was no significant difference between cool and standard dialysis for dialysis adequacy measured using the Kt/V urea (based on 5 studies) or the urea reduction ratio (based on 2 studies).

Reducing the temperature of dialysate lowers the incidence of IDH without reducing the adequacy of dialysis. There was insufficient evidence to determine the effect of cool dialysis on cold symptoms. Suggestions for further research were reported.

The review question was clear in terms of the study design, participants, intervention and outcomes. Several relevant sources were searched and attempts were made to locate unpublished data, thus reducing the potential for publication bias. By limiting the included studies to those in English, the authors might have missed some relevant studies. Methods were used to minimise reviewer errors and bias in the validity assessment and parts of the study selection process, but it was unclear whether similar steps were taken in the data extraction. Validity was assessed using specified criteria and the results of the assessment were reported.

Adequate information about the included studies was presented. Significant statistical heterogeneity was shown for the meta-analysis examining IDH, but no comment was made about this in the text. The direction of the treatment effect appeared consistent across studies, and the influence of method of temperature reduction and other variables was examined. However, the demonstration of significant heterogeneity showed that treatment effects were not consistent across studies and suggested that a summary treatment effect might not have been appropriate. Overall, this was a well-conducted review and the authors' conclusions are likely to be reliable.

Practice: The authors stated that lowering the dialysate temperature using either a fixed temperature reduction or BTM reduces the incidence of IDH, but the effects on intolerable cold symptoms remain uncertain.

Research: The authors stated that no additional studies are required to evaluate the effects of reduced dialysate temperature on IDH. However, there is a need for further studies to define the optimal temperature and to evaluate long-term patient outcomes. There is also a need to compare empirical reductions in dialysate temperature with isothermic dialysis with BTM, and a need to develop a scoring system to rate the severity and frequency of cold symptoms.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.