Preoperative Predictors for Intraoperative Blood Transfusion in Crossmatched Adult Patients Undergoing General, Gynaecologic and Orthopaedic Surgeries.

Abstract

Background: Transfusion of blood intraoperatively in general, gynaecological, orthopaedic surgeries can be lifesaving. At KNH, grouping and crossmatching for blood compatibility of donor blood is carried out for many of these patients as part of preoperative preparation. However, no standardized criteria have been applied for this blood ordering. This has resulted in high crossmatch to transfusion ratios, contributing to wastage of blood products. Development of hospital-based prediction tools and protocols for intraoperative blood transfusion would guide the efficiency of preoperative blood ordering.

• Study objective: To determine preoperative predictors for intraoperative blood transfusion in crossmatched adult patients undergoing general, gynaecologic and orthopedic surgeries at Kenyatta National Hospital.

• Study design and site: This study was a descriptive cross-sectional study, carried out at the Kenyatta National Hospital’s main theatres and trauma theatres.

I. Participants and Methodology: It involved 226 crossmatched adult patients undergoing general, gynaecological and orthopaedic surgeries over a period of 3 months. Inclusion criteria- Crossmatched adult patients and patients who gave written informed consent undergoing general, gynaecological and orthopaedic surgery at KNH. Exclusion criteria- blood transfused to patients during transfer into theater/ blood that accompanied patient to theater, patient with documented blood disorder, patients who underwent multiple surgeries, patients scheduled for hemostatic surgery, and revision surgeries.

• Data management: Pre- and intraoperative data collected in the wards and A & E, was cross-checked, cleaned, categorized, and entered using the statistical analysis software package, SPSS version 22. Measures of central tendency such as mean and standard deviation was used to describe variables with normal distribution while skewed distributions was described in terms of medians and interquartile ranges. Descriptive

statistics such as frequencies and percentages were used to describe demographic characteristics like age and sex.

• Expected results: This required identification of patients who have been crossmatched and scheduled for surgery. Preoperative data were collected based on patient and operative factors and correlated to intraoperative transfusion. I expect to observe an association between preoperative factors and intraoperative transfusion among patients who had been grouped and crossmatched.