CRD summary

This review concluded that the use of hydroxyethyl starch 130/0.38-0.45 in patients with sepsis increased use of renal replacement therapy, red blood cell transfusions and serious adverse events compared to human albumin or crystalloid. The authors' conclusions reflect the results of the review and appear likely to be reliable.

Authors' objectives

To assess the effects of fluid therapy with hydroxyethyl starch 130/0.38-0.45 versus crystalloid or albumin on mortality, kidney injury, bleeding and serious adverse events.

Searching

Nine databases and trials registers, including MEDLINE, EMBASE, CINAHL and The Cochrane Library were searched from 1995 to September 2012. The search strategy was available in an online appendix. References of identified studies and systematic reviews were checked and authors and manufacturers were contacted. There were no restrictions on language or publication status.

Study selection

Randomised controlled trials (RCTs) that compared hydroxyethyl starch 130/0.38-0.45 with crystalloid or albumin in patients with sepsis were eligible for inclusion. There were no restrictions on the indication for fluid therapy or the predefined outcomes of the trial. Trials in which sepsis patients were a subgroup were included only if the randomisation had been stratified for sepsis or if more than 500 patients had sepsis. Cross-over trials were excluded. The primary outcomes of the review were mortality and number of patients receiving renal replacement therapy at follow-up. Secondary outcomes related to blood transfusion requirements, bleeding, renal replacement therapy at any time, acute kidney injury and adverse events. Quasi-randomised and observational studies in which at least 500 patients received hydroxyethyl starch 130/0.38-0.45 were also included for the evaluation of serious adverse events only.

All included studies took place in intensive care units. Patients had a diagnosis of sepsis, severe sepsis (organ failure), septic shock or sepsis and tissue hypoperfusion. The indication for treatment was resuscitation in all except two trials where it was to administer a fixed dose or to maintain pulmonary capillary wedge pressure. Intervention periods ranged from 24 hours to a maximum intensive care unit stay of 90 days. Where reported the intervention was given as a 6% solution; total mean or median doses ranged from 2.1 to 6.4 litres. Two trials compared hydroxyethyl starch 130/0.38-0.45 with 20% human albumin, others used crystalloid.

Two reviewers evaluated the papers for inclusion; disagreements were resolved through consultation with a third reviewer.

Assessment of study quality

The trials were appraised using the Cochrane Collaboration risk of bias assessment tool which assessed sequence generation (randomisation), allocation concealment, blinding, incomplete outcome data and selective outcome reporting. Baseline imbalances, financial interests and academic bias were also assessed.

Two independent reviewers carried out the assessment.

Data extraction

Data on trial and patient characteristics such as indication for therapy and disease severity were extracted with outcome data to permit the calculation of relative risks (RR) or risk differences (RD) with 95% confidence intervals (CI). Authors were contacted for data on outcomes which were not reported. Non-English reports were translated.

Two reviewers independently extracted the data using a standardised form.

Methods of synthesis

Pooled relative risks with 95% confidence intervals were calculated for dichotomous data and pooled risk differences with 95% confidence intervals for continuous data. Heterogeneity was assessed using Ι²; where Ι² was greater than zero, random-effects analyses were undertaken in addition to fixed-effect ones. A pre-specified subgroup analysis which stratified the trials according to risk of bias was conducted; a post-hoc analysis stratified by length of follow-up. Sensitivity analyses were used to explore the impact of excluding the smallest and largest trials and trials funded by industry. In addition to standard meta-analysis, trial sequential analysis was also carried out.

Results of the review

Nine RCTs with 3,456 patients were included in the review. Four RCTs were considered at low risk of bias, the other five were at high risk of bias in at least one domain.

Mortality: There was no statistically significant difference between groups treated with hydroxyethyl starch 130/0.38-0.45 and comparison groups (RR 1.04, 95% CI 0.89 to 1.22; Ι²=37%; eight trials). Sensitivity, subgroup and trial sequential analyses did not change the overall result.

Renal replacement therapy at end of follow-up: Only three events (two in the intervention groups, one in the control group) were reported for this outcome (five trials); the data were not pooled.

Other efficacy outcomes: Patients treated with hydroxyethyl starch 130/0.38-0.45 had a statistically significantly higher chance of needing renal replacement therapy at any point during follow-up (RR 1.36, 95% CI 1.08 to 1.72; Ι²=0; five trials) and of having a red blood cell transfusion (RR 1.29, 95% CI 1.13 to 1.48; Ι²=0%; three trials). Other transfusion related outcomes did not differ significantly between the groups but favoured controls. The risk of acute kidney injury was also non-significantly higher in the intervention groups.

Serious adverse events: The risk of a serious adverse event was statistically significantly higher in the intervention groups (RR 1.30, 95% CI 1.02 to 1.67; Ι²=0%; three trials).

Results of trial sequential analyses were also reported for secondary outcomes.

Authors' conclusions

Conventional meta-analysis found that the use of hydroxyethyl starch 130/0.38-0.45 in patients with sepsis increased the use of renal replacement therapy and red blood cell transfusions and resulted in more serious adverse events compared to albumin or crystalloid. It appeared unlikely that hydroxyethyl starch 130/0.38-0.45 provided overall clinical benefit for patients with sepsis.

CRD commentary

This review assessed a clear question with specific and reproducible inclusion criteria. The search was thorough and measures were taken to reduce the chances of relevant studies being missed. The authors reported using methods designed to reduce reviewer error and bias in all stages of the review process. An appropriate tool was used to assess quality of the included studies and the results of the assessment were used to inform the synthesis. The meta-analysis was suitable and measures were taken to assess potential sources of heterogeneity.

The authors' conclusions reflect the results of the review and appear likely to be reliable.

Implications of the review for practice and research

The authors stated that if use of hydroxyethyl starch 130/0.38-0.45 in the surgical setting continues, despite the safety issues in patients with sepsis that were raised in this review, then adequately powered surgical trials were urgently required to ensure patient safety.

Funding

Not reported.

Bibliographic details

Haase N, Perner A, Inkeri Hennings L, Siegemund M, Lauridsen B, Wetterslev M, Wetterslev J. Hydroxyethyl starch 130/0.38-0.45 versus crystalloid or albumin in patients with sepsis: systematic review with meta-analysis and trial sequential analysis BMJ 2013; 346: f839. [PMC free article: PMC3573769] [PubMed: 23418281]

Original Paper URL

http://www.bmj.com/content/346/bmj.f839.abstract

Indexing Status

Subject indexing assigned by NLM

MeSH

Fluid Therapy /methods; Humans; Hydroxyethyl Starch Derivatives /therapeutic use; Isotonic Solutions /therapeutic use; Sepsis /therapy; Serum Albumin /therapeutic use

AccessionNumber

12013010136

Database entry date

20/02/2013

Record Status

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.