[Use of homologous erythrocyte concentrates. Analysis of economical factors]

[Transfusion of erythrocyte concentrates. An evaluation in anesthesia and intensive care]

OBJECTIVE

To assess the relevance of perioperative packed red blood cell (PRBC) transfusion practice at the University Hospital of Nice, compared with information from the consensus conference on red blood cell transfusion, held by the French Society of Anaesthesia and Intensive Care (SFAR) and the National Agency for the Development of Medical Evaluation (ANDEM) in December 1993.

STUDY DESIGN

Retrospective case series analysis.

PATIENTS

The study included 240 medical files of surgical patients, transfused in 1994 with PRBC, obtained by drawing of lots following a methodology recommended by ANDEM.

METHOD

A reference list according to the statement of the consensus conference was designed for the various surgical specialities and the ICU in which PRBC had been transfused. It included the clinical and laboratory criteria which justified the transfusion, as well as the various categories of PRBC (phenotyped, cytomegalovirus negative, leukocyte-depleted, etc). Autotransfused PRBC were also considered. The data collected from the medical files of the 240 patients were compared with the reference list.

RESULTS

In 84.6% of patients (203/240), the PRBC transfusion had been decided with reference either only to a haematocrit level below 0.27 or a level between 0.27 and 0.30 associated with clinical evidence of bad tolerance of blood loss, according to the reference list. A lack of compliance with the reference list occurred in 15.4% of patients (37/240), who had been transfused without any reference to a biological criterion. Another non compliance existed in 50% of patients (12/24) transfused with phenotyped PRBC and in 35.3% (6/17) of those transfused with leucocyte-depleted PRBC. An autotransfusion with PRBC had been carried out in 30.4% of patients (75/240).

DISCUSSION

These deviations of transfusion practice from the consensus conference statement, which were more pronounced with phenotyped and leucocyte-depleted PRCB than conventional PRBC, resulted in the edition of a report, with an analysis of the causes of deviations and recommendations for all doctors of our institution prescribing blood transfusions. Another evaluation, extended also to the medical specialities of our hospital and including all blood derivates is planned for 1996.