Outcomes of restrictive versus liberal transfusion strategies in older adults from nine randomised controlled trials: a systematic review and meta-analysis

Summary Background Guidelines for patient blood management recommend restrictive transfusion practice for most adult patients. These guidelines are supported by evidence from randomised controlled trials (RCTs); however, one of the patient groups not explicitly examined in these studies is the geriatric population. We examined RCTs relevant to transfusion outcomes in older patients. Our aim was to determine whether special guidelines are warranted for geriatric patients, recognising the different pathophysiological characteristics of this group. Methods For this systematic review and meta-analysis, we searched PubMed, Scopus, and the Cochrane Library databases from their inception to May 5, 2017, for evidence relating to transfusion outcomes in adults aged 65 years and older. This criterion was widened to include RCTs where a substantial proportion of the study population was older than 65 years. We also included study populations of all clinical settings, and did not limit the search by date, language, or study type. For articles not in English, only available translations of the abstracts were reviewed. Studies were excluded if they did not specify age. Observational studies and duplicate patient and outcome data from studies that generated multiple publications were also excluded. We screened bibliographies of retrieved articles for additional publications. We analysed data extracted from published RCTs comparing restrictive and liberal transfusion strategies in older adults. We generated fixed effects risk ratios (RR) for pooled study data using the Mantel-Haenszel method. Primary outcomes were 30-day and 90-day mortality events for patients enrolled in restrictive and liberal transfusion study groups. We included intention-to-treat outcome data in the meta-analysis when available, otherwise we used per-protocol outcome data. Findings 686 articles were identified by the search, and a further 37 by the snowball approach. Of these articles, 13 eligible papers described findings from nine RCTs (five trials investigating orthopaedic surgery, three cardiac surgery, and one oncology surgery; including 5780 patients). The risk of 30-day mortality was higher in older patients who followed a restrictive transfusion strategy than in those who followed a liberal transfusion strategy (risk ratio [RR] 1·36, 95% CI 1·05–1·74; p=0·017). The risk of 90-day mortality was also higher in those who followed a restrictive transfusion strategy than in those who followed a liberal transfusion strategy (RR 1·45, 95% CI 1·05–1·98; p=0·022). Interpretation Liberal transfusion strategies might produce better outcomes in geriatric patients than restrictive transfusion strategies. This outcome contradicts current restrictive transfusion approaches. Population ageing will challenge resources globally, and this finding has implications for blood supply and demand, and optimal care of older adults. Further research is needed to formulate evidence-based transfusion practice across clinical specialties specific to the geriatric population, and to examine resource effects. Funding Australia's National Blood Authority.