Sex-Specific Associations Between Preoperative Hemoglobin and Outcomes After Major Noncardiac Surgery: A Retrospective Cohort Study

Abstract

BACKGROUND: Preoperative anemia is an established risk factor for morbidity and mortality after surgery. Men and women have different hemoglobin concentrations and are at different risks of postoperative complications. However, sex-stratified analysis on the association between preoperative hemoglobin and outcomes after noncardiac surgery has been limited in previous studies METHODS: This was a retrospective cohort study of adult patients undergoing elective major noncardiac surgery in a large academic hospital. The primary outcome was a collapsed composite of postoperative mortality or cardiovascular, renal, pulmonary, and infectious complications during hospitalization. Sex-specific univariable associations between preoperative hemoglobin and the composite outcome were visualized using moving-average and cubic-spline smoothing plots. Multivariable regression models adjusting for patient demographics, comorbidities, medication uses, laboratory tests, and anesthesia/surgery features were used to estimate confounder-adjusted associations. Restricted cubic spline and piecewise linear functions were used to assess the possible nonlinear relationships between preoperative hemoglobin and the outcomes. The interaction between patient sex and hemoglobin on outcomes was assessed using a likelihood-ratio test. RESULTS: We included 22,550 patients, with 6.7% (622 of 9268) of women and 9.7% (1293 of 13,282) of men developing the primary outcome. Lower preoperative hemoglobin was associated with a higher incidence of the primary composite outcome in both men and women. Nonlinearity for the association was not statistically significant in either women (P = .539) or men (P = .165). The multivariable-adjusted odds ratios per 1 g/dL increase in hemoglobin were 0.93 (95% confidence interval [CI], 0.87–0.98; P = .013) for women and 0.94 (95% CI, 0.90–0.97; P < .001) for men, with no interaction by sex (P interaction = .923). No hemoglobin thresholds were confirmed at which the associations with the primary outcome changed significantly. CONCLUSIONS: Low preoperative hemoglobin was associated with a higher risk of complications or mortality after elective noncardiac surgery in both men and women. No differences in the strength of associations between sexes were found. Further studies are needed to assess whether these associations are linear or there are sex-specific thresholds of preoperative hemoglobin concentrations below which postoperative risks begin to increase.