CRD summary

This review concluded that intravenous aprotinin and tranexamic acid use significantly reduced the risk of allogeneic erythrocyte transfusion in orthopaedic surgery but that there was a lack of evidence regarding safety. The review methods were reliable (although a more thorough exploration of the quality of the evidence was warranted) and the author's conclusions appear appropriate.

Authors' objectives

To evaluate whether the use of intravenous antifibrinolytics in orthopaedic surgery reduces the need for perioperative allogeneic erythrocyte transfusion in adults, or increases the risk of venous thromboembolism.

Searching

MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials were searched to July 2005. Search terms were reported. Meeting abstracts and references of studies, reviews and meta-analyses were also searched. No language restrictions were applied.

Study selection

Randomised controlled trials (RCTs), open or using single or double-blinding, comparing an antifibrinolytic agent with control groups (placebo, no treatment or another antifibrinolytic agent) to reduce perioperative allogeneic blood transfusion in orthopaedic surgery patients were eligible for inclusion. Studies including an additional method for reducing allogeneic blood transfusion were eligible if the same method was used in both the intervention and control groups. Only studies of orthopaedic surgery were eligible. Studies in paediatric population or dose-ranging studies were excluded.

Eligible outcomes were efficacy and/or safety recorded by: receipt of at least one unit of allogeneic erythrocyte according to a transfusion protocol; receipt of any kind of erythrocyte transfusion; number of allogeneic blood units transfused; total perioperative and postoperative blood loss; venous thromboembolism confirmed by an objective test; or arterial thrombosis (heart attack, stroke or limb ischaemia).

The included trials evaluated mostly the antifibrinolytic agents aprotinin or tranexamic acid, with only a few evaluating ε-aminocaproic acid or two different agents compared with placebo. Most studies used low-molecular weight heparin for deep vein thrombosis prophylaxis. Doses and transfusion protocols varied between studies and the most common procedures were total hip or total knee arthroplasty.

Studies were selected by two reviewers independently with disagreements referred to a third reviewer.

Assessment of study quality

The authors stated that only studies with a transfusion protocol defining when a transfusion of allogeneic and/or autologous erythrocytes was necessary would be included in the meta-analyses to limit evaluation bias. They also assessed the level of blinding, commented on the size of the trials and how many open studies reported appropriate methods of randomisation.

The authors do not state that they assessed validity.

Data extraction

For binary outcomes, the proportions of patients in each group were extracted and used to calculate either the odds ratio (OR), risk ratio (RR) or risk difference (RD) and 95% confidence interval (CI) for each trial. For continuous outcomes the standardised mean differences (SMD) and 95% CI were calculated from the mean and standard deviation (SD) of each group in each trial. If a trial had more than one active treatment group then these groups were combined.

Data were extracted by two authors independently with disagreements referred to a third reviewer. Authors were contacted for missing data or clarification when necessary.

Methods of synthesis

Studies were pooled using meta-analysis on an intention-to-treat basis. Heterogeneity was assessed using χ² tests with p≤0.10 used to indicate statistically significant heterogeneity. For binary outcomes, the OR was presented as it showed the least heterogeneity. For continuous outcomes, trials reporting median and range had the mean and SD estimated; studies with obviously skewed outcomes were excluded. Random-effects models were planned in the presence of unexplained heterogeneity but only fixed-effect model results were presented for binary outcomes. Separate analyses were performed for each drug.

Subgroup and sensitivity analyses were performed to explore reasons for heterogeneity. Subgroups were: double-blind versus open studies; primary hip or knee surgery versus other major orthopaedic surgery; type of anaesthesia; dose regimen; and number of bolus doses (single versus continuous infusion or repeated doses). Sensitivity analyses removed each study one at a time and imputed missing data for any per-protocol results. Publication bias was assessed using funnel plots.

Results of the review

43 RCTs (n=2,523) were included in the review; 31 trials were double-blind and five trials were open-label but reported an appropriate randomisation method. Sample sizes ranged from 17 to 391, with the majority of studies including less than 100 patients.

Aprotinin (23 RCTs, n=1,268):

Fifteen trials reported using a transfusion protocol but did not comment on whether these were followed. The trials were mostly small, with only two trials including more than 100 patients.

Aprotinin was associated with a statistically significant reduction in the odds of requiring allogeneic erythrocyte transfusion compared with placebo (OR 0.43, 95% CI: 0.28, 0.64, p<0.01,10 trials). Results were similar for the double-blind studies but there was no evidence of a difference for the open studies. There was no evidence of heterogeneity. There were no influential studies or evidence of publication bias.

Other outcomes showed significant reductions in any kind of erythrocyte transfusion (OR 0.49, 95% CI: 0.33, 0.73, p<0.01); blood loss (SMD 0.79, 95% CI: 0.64, 0.94, p<0.01) and no evidence of an increased risk of venous thromboembolism.

Tranexamic acid (20 RCTs, n=1,084):

Fourteen trials were double-blind, three open trials reported appropriate randomisation and three reported allocation concealment. The transfusion protocol was reported in 15 trials. Only one trials included more than 100 patients.

Tranexamic acid was associated with a statistically significant reduction in the odds of requiring allogeneic erythrocyte transfusion compared with placebo (OR 0.17, 95% CI: 0.11, 0.24, p<0.01,15 trials). There was no evidence of heterogeneity. Results were similar for the double-blind and open trials.

Other subgroup analyses showed that tranexamic acid was significantly more effective in knee than hip arthroplasty, higher doses (≥30mg/kg) and when it was given as repeated bolus or bolus plus continuous infusion. There were no influential studies, or evidence of publication bias.

Other outcomes showed significant reductions in any kind of erythrocyte transfusion (OR 0.17, 95% CI: 0.12, 0.25, p<0.01) but no differences for blood loss or risk of venous thromboembolism.

ε-aminocaproic acid (four RCTs, n=171):

Three trials had a predefined transfusion protocol.

There was no evidence of a difference between ε-aminocaproic acid and placebo for allogeneic erythrocyte transfusion, venous thromboembolism or blood loss.

Details of other adverse events were reported in the review.

Authors' conclusions

The use of aprotinin and tranexamic acid significantly reduces the risk of allogeneic erythrocyte transfusion in orthopaedic surgery. There was a lack of evidence regarding safety, so no definite conclusions about the clinical risk/benefit of antifibrinolytics can be derived from this review.

CRD commentary

This review had clearly stated objectives, study inclusion and exclusion criteria. The search made efforts to locate unpublished studies and no studies were excluded on the basis of language, which reduced the risk of publication and language bias. Study screening and data extraction were performed by two reviewers independently which minimised the risk of error and bias in the process. The methods of meta-analysis were appropriate but only studies meeting a specific criterion for having a transfusion protocol were included in the analyses. As other studies were included in the systematic review, it might have been helpful to readers to report the results of these studies (or make them available elsewhere). The main limitation to this review was the lack of a quality assessment; some aspects of study quality were reported in the text but not any overall indication of the quality of the evidence. However, a number of subgroup analyses were performed, one of which assessed a quality aspect (blinding). Overall, this appears to be a reliable review with appropriate conclusions, particularly in relation to the safety of antifibrinolytics.

Implications of the review for practice and research

Practice: The authors did not state any recommendations for practice.

Research: The authors stated that large prospective trials are needed to define the optimal regimen and assess the safety and cost-effectiveness of antifibrinolytics before they can be recommended for use in orthopaedic surgery.

Funding

None.

Bibliographic details

Zufferey P, Merquiol F, Laporte S, Decousus H, Mismetti P, Auboyer C, Samama C M, Molliex S. Do antifibrinolytics reduce allogeneic blood transfusion in orthopedic surgery? Anesthesiology 2006; 105(5): 1034-1046. [PubMed: 17065899]

Original Paper URL

http://journals.lww.com/anesthesiology/toc/2006/11000

Indexing Status

Subject indexing assigned by NLM

MeSH

Adult; Aminocaproic Acid /therapeutic use; Antifibrinolytic Agents /therapeutic use; Aprotinin /therapeutic use; Blood Transfusion; Erythrocyte Transfusion; Humans; Orthopedic Procedures; Postoperative Hemorrhage /prevention & control; Randomized Controlled Trials as Topic; Tranexamic Acid /therapeutic use

AccessionNumber

12006007778

Database entry date

10/06/2009

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.