A pre-operative scoring model to estimate the risk of blood transfusion over an ovarian cancer debulking surgery (BLOODS score): a Memorial Sloan Kettering Cancer Center Team Ovary study

Objectives To develop a pre-operative tool to estimate the risk of peri-operative packed red blood cell transfusion in primary debulking surgery.Methods We retrospectively reviewed an institutional database to identify patients who underwent primary debulking surgery for ovarian cancer at a single center between January 1, 2001 and May 31, 2019. Receiver operating characteristic curve and area under the receiver operating characteristic curve (AUC) were calculated. Five-fold cross-validation was applied to the multivariate model. Significant variables were assigned a 'BLOODS' (BLood transfusion Over an Ovarian cancer Debulking Surgery) score of +1 if present. A total BLOODS score was calculated for each patient, and the odds of receiving a transfusion was determined for each score.Results Overall, 1566 patients met eligibility criteria; 800 (51%) underwent a peri-operative blood transfusion. Odds ratios (OR) were statistically significant for American Society of Anesthesiologists scores of 3 and 4 (OR 1.34, 95% confidence interval (95% CI) 1.09 to 1.63), pre-operative levels of cancer antigen 125 (CA125) (OR 2.43, 95% CI 1.98 to 2.99), platelets (OR 1.59, 95% CI 1.45 to 1.74), obesity (OR 0.76, 95% CI 0.60 to 0.96), presence of carcinomatosis (OR 2.45, 95% CI 1.93 to 3.11), bulky upper abdominal disease (OR 2.86, 95% CI 2.32 to 3.54), pre-operative serum albumin level (OR 0.31, 95% CI 0.24 to 0.40), and pre-operative hemoglobin level (OR 0.56, 95% CI 0.51 to 0.61). The corrected AUC was 0.748 (95% CI 0.693 to 0.804). BLOODS scores of 0 and 5 corresponded to 11% and 73% odds, respectively, of receiving a peri-operative blood transfusion.Conclusions We developed a universal pre-operative scoring system, the BLOODS score, to help identify patients with ovarian cancer who would benefit from surgical planning and blood-saving techniques. The BLOODS score was directly proportional to the American Society of Anesthesiologists score, presence of upper abdominal disease, carcinomatosis, CA125 level, and platelets level. We believe this model can help physicians with surgical planning and can benefit patient outcomes.