The pattern of blood loss in adolescent idiopathic scoliosis

Background context
Previous studies have shown that modern intraoperative blood-saving techniques dramatically reduce the allogeneic transfusion requirements in surgery for adolescent idiopathic scoliosis (AIS). No studies have looked at the pattern of postoperative hemoglobin (Hb) in AIS patients undergoing corrective spinal surgery and correlated this with the timing of allogeneic transfusion.
Purpose
To describe the pattern of perioperative blood loss in instrumented surgery for AIS. We look at the recommendations regarding an ideal preoperative Hb, the need for preoperative cross-matching, and the timing of postoperative Hb analysis.

Study design
This was a retrospective case series. Surgeries were performed by one of four substantive pediatric spinal surgeons within a single regional center over a 3-year period.

Patient sample
A consecutive series of 86 patients who underwent posterior instrumented fusion for AIS were included: 10 males and 76 females. Mean age was 14 years (range 10–17 years). All patients had posterior instrumented fusion using various blood-saving techniques (eg, cell-saver). All patients were cross-matched preoperatively, and our transfusion trigger value (TTV) was 7 g/dL.

Outcome measures
Hemoglobin level was the outcome measure. Hemoglobin readings were obtained preoperatively, within 2 hours of surgery, and daily up to 5 days after surgery. This physiologic measure was assessed using routine blood sampling techniques and standardized laboratory processing.

Methods
Patient predictor variables (demographic and surgical) were assessed for association with Hb levels in a hierarchical model, with repeated Hb readings at the lower level being clustered within an individual patient at the upper level of the structure. The variation of Hb levels within individuals was compared with mean levels in different individuals via the variance partition coefficient of the model structure.

Results
No patients required intraoperative allogeneic transfusion. Only four patients (4.65%) received allogeneic transfusion, all within 2 days of surgery. A clinically important drop in Hb occurred within the first 2 postoperative days, rising thereafter. The average postoperative drop in Hb was 4.1 g/dL. Young males had lower postoperative Hb values. Neither the preoperative curve magnitude (Cobb angle of major curve) nor the number of vertebrae/levels fused significantly affected the blood loss.

Conclusions
We recommend setting a minimum preoperative Hb value that is 5 g/dL higher than your TTV. Because no patients required an intraoperative transfusion when using modern blood-saving techniques, preoperative cross-matching is unnecessary and potentially wasteful of blood reserves. Hemoglobin analysis beyond the second postoperative day is unnecessary unless clinically indicated