Preoperative hemoglobin thresholds for survival equity in women and men

Abstract

Anemia affects humans throughout life, and is linked to higher morbidity and mortality. Unclear is whether hemoglobin values are equivalent between women and men. This study evaluates the association of preoperative hemoglobin levels with in-hospital mortality and estimates thresholds for survival equity between men and women. All adult patients undergoing surgery between 2010 and 2019 from 14 German hospitals were included in the study. Thresholds for survival equity were determined with generalized additive models. In total, 842,130 patients with a median in-hospital follow-up time of 7 days were analyzed. During follow-up 20,370 deaths occurred. Preoperative hemoglobin stratified in-hospital mortality (log-rank test p < 0.001) and was associated with mortality independently of demographic risk, surgical risk and health status. For each 1 g/dL reduction in preoperative hemoglobin, the odds of mortality increased by a factor of 1.22 (95% CI 1.21-1.23, p < 0.001). A preoperative hemoglobin threshold of 10.5 g/dL reflected equivalent risk for both male and female patients. Hemoglobin levels below 10.5 g/dL had higher risk of mortality for women than for men. The findings from this study aid evidence-based thresholds, inform anemia management and promote equitable care, thus enhancing patient outcomes.

Keywords: anemia; equitable health care; hemoglobin; perioperative management; perioperative medicine.

Figure 1

Study population: patients included in this study from the German Patient Blood Management Network.

Figure 2

Preoperative hemoglobin levels: (A) distribution of preoperative hemoglobin in female (n = 385,774) and male patients (n = 456,356), (B) smoothed conditional means (solid line) with the 5th percentile (dashed line) for a healthy subpopulation (Charlson comorbidity score <3, minor surgery). Measures are given continuously for age in female (n = 234,430) and male patients (n = 274,762).

Figure 3

Preoperative hemoglobin stratifies in-hospital mortality: survival curves were derived from the Kaplan–Meier estimator stratified by preoperative hemoglobin. Decreasing hemoglobin levels were associated with increasing mortality risk (log-rank p < 0.001).

Figure 4

Hemoglobin thresholds for survival equity between women and men: in order to investigate comparable hemoglobin levels linked to mortality in both men and women, interactions of hemoglobin thresholds and sex were analyzed using generalized additive models. Survival equity was defined as the hemoglobin thresholds that show the least difference based on sex. The smooth line indicates similar hemoglobin levels for both genders. The color scale illustrates the associated risk (odds ratio) with these thresholds.