BACKGROUND: Data on hemoglobin (Hb) threshold levels for "appropriate" packed red blood cell (PRBC) transfusions have not taken into account patient-specific variables such as sex and age. We sought to define differences in perioperative transfusion practices based on patient sex and age among patients undergoing complex gastrointestinal (GI) and cardiothoracic-vascular (CT-V) surgical procedures. METHODS: All patients undergoing any major GI or CT-V procedures between January 2010 and April 2014 at the Johns Hopkins Hospital were identified. Data on sex, age, as well as other clinicopathologic and procedures were collected and analyzed relative to transfusion practices (restrictive: transfusion at blood Hb < 7 vs liberal transfusion at Hb ≥ 7 g/dL). RESULTS: Among the 10,772 patients included in the study cohort, 4,689 (44.0%) were transfused with ≥ 1 PRBC. Median preoperative Hb was lower among women (12.3 vs 13.4 g/dL in men) and the aged (<65 years, 13.1 vs ≥ 65 years, 12.7 g/dL) patients (both P < .05). On adjusted analysis, male sex (odds ratio [OR], 1.13; 95% CI, 1.02-1.26; P = .03) and age ≥ 65 (OR, 1.77; 95% CI, 1.35-2.33; P < .001) were associated independently with an increased odds of receipt of ≥ 1 PRBC. Although sex did not seem to impact transfusion strategy, patient age did impact the relative trigger used by providers for a transfusion. Specifically, patients 65-74 years (OR, 2.87; 95% CI, 1.93-4.26) and those ≥ 74 years (OR, 3.42; 95% CI, 2.28-5.14) were at a much greater odds of being transfused liberally compared with patients <65 years old (both P < .05). The proportion of patients who had a potentially avoidable transfusion (ie, both trigger ≥ 7 and target ≥ 9 g/dL) was greater among aged patients (50%) compared with nonaged patients (41%; P < .001). Of note, a restrictive transfusion strategy did not increase the risk of overall morbidity among women (OR, 0.76; 95% CI, 0.59-0.99; P = .04) or aged (OR, 1.13; 95% CI, 0.87-1.47; P = .37) patients. CONCLUSION: Sex and age were associated with receipt of transfusion and, in the case of older patients, transfusion strategy. Given the lack of evidence to support a higher Hb level in older patients, emphasis should be placed on aligning transfusion practices with current evidence to employ a more restrictive transfusion strategy to decrease overuse of blood resources.
BACKGROUND: Data on hemoglobin (Hb) threshold levels for "appropriate" packed red blood cell (PRBC) transfusions have not taken into account patient-specific variables such as sex and age. We sought to define differences in perioperative transfusion practices based on patient sex and age among patients undergoing complex gastrointestinal (GI) and cardiothoracic-vascular (CT-V) surgical procedures. METHODS: All patients undergoing any major GI or CT-V procedures between January 2010 and April 2014 at the Johns Hopkins Hospital were identified. Data on sex, age, as well as other clinicopathologic and procedures were collected and analyzed relative to transfusion practices (restrictive: transfusion at blood Hb < 7 vs liberal transfusion at Hb ≥ 7 g/dL). RESULTS: Among the 10,772 patients included in the study cohort, 4,689 (44.0%) were transfused with ≥ 1 PRBC. Median preoperative Hb was lower among women (12.3 vs 13.4 g/dL in men) and the aged (<65 years, 13.1 vs ≥ 65 years, 12.7 g/dL) patients (both P < .05). On adjusted analysis, male sex (odds ratio [OR], 1.13; 95% CI, 1.02-1.26; P = .03) and age ≥ 65 (OR, 1.77; 95% CI, 1.35-2.33; P < .001) were associated independently with an increased odds of receipt of ≥ 1 PRBC. Although sex did not seem to impact transfusion strategy, patient age did impact the relative trigger used by providers for a transfusion. Specifically, patients 65-74 years (OR, 2.87; 95% CI, 1.93-4.26) and those ≥ 74 years (OR, 3.42; 95% CI, 2.28-5.14) were at a much greater odds of being transfused liberally compared with patients <65 years old (both P < .05). The proportion of patients who had a potentially avoidable transfusion (ie, both trigger ≥ 7 and target ≥ 9 g/dL) was greater among aged patients (50%) compared with nonaged patients (41%; P < .001). Of note, a restrictive transfusion strategy did not increase the risk of overall morbidity among women (OR, 0.76; 95% CI, 0.59-0.99; P = .04) or aged (OR, 1.13; 95% CI, 0.87-1.47; P = .37) patients. CONCLUSION: Sex and age were associated with receipt of transfusion and, in the case of older patients, transfusion strategy. Given the lack of evidence to support a higher Hb level in older patients, emphasis should be placed on aligning transfusion practices with current evidence to employ a more restrictive transfusion strategy to decrease overuse of blood resources.