Blood Transfusion Incidence, Risk Factors, and Associated Complications in Surgical Treatment of Hip Dysplasia

Abstract

Background: Perioperative bleeding requiring blood transfusion is a known complication of hip dysplasia (HD) surgery. Here we examine rates of, risk factors for, and postoperative complications associated with transfusion during HD surgery.

Methods: The National Surgical Quality Improvement Program (NSQIP) Pediatric database was queried for patients treated by an orthopaedist from 2012 to 2013. HD cases were categorized by Current Procedural Terminology codes into femoral osteotomies, acetabular osteotomies, combined femoral/acetabular osteotomies, and open reductions. Patients were grouped by comorbidities: neuromuscular (NM) disease (eg, cerebral palsy) group, non-NM with other comorbidity (Other) group, and no known comorbidity (NL) group. Patients were stratified by weight-normalized transfusion volume. Multivariate regression analysis of transfusion association with procedures, demographics, comorbidities, preoperative laboratory values, and 30-day complications was performed.

Results: A total of 1184 HD cases were included. Transfusion rates for the NL, Other, and NM groups, respectively, were 44/451 (9.8%), 61/216 (28.2%), and 161/517 (31.1%). Transfusion volumes (mean±SD) for the NL, Other, and NM groups, respectively, were 8.4±5.4, 13.9±8.8, and 15.5±10.0 mL/kg (P<0.001). Combined osteotomies had the highest transfusion rates in the NM and Other groups (35.7% and 45.8%, respectively), whereas acetabular osteotomies had the highest rate in the NL group (15.8%). Open reductions had the lowest transfusion rate (all groups). Longer operations were independently associated with transfusion (all groups, per hour increase, OR>1.5, P<0.001). Independent patient risk factors included preoperative hematocrit <31% (NM group, OR=18.42, P=0.013), female sex (NL group, OR=3.55, P=0.008), developmental delay (NM group, OR=2.37, P=0.004), pulmonary comorbidity (NM group, OR=1.73, P=0.032), and older age (NL group, per year increase: OR=1.29, P<0.001). In all groups, transfusion was associated with longer hospitalization (P<0.001). We observed a volume-dependent increase in overall complication rate within the Other group for transfusion volumes >15 mL/kg (25.0% vs. 5.4% for <15 mL/kg, P=0.048).

Conclusions: We identified several risk factors for transfusion in HD surgery. The incidence of transfusion in HD surgery and its association with adverse outcomes warrants development of appropriate patient management guidelines.

Level of evidence: Level III-prognostic.

Figure 1

(A) Absolute transfusion volume (mean ± standard deviation) for each study group. (B) Body weight-normalized transfusion volume (mean ± standard deviation) for each study group. Statistical tests were performed using one-way ANOVA. NM = neuromuscular, NL = otherwise normal, Other = other non-NM comorbidity, NS = not significant.