ABSTRACT
Background: Whereas animal models of sepsis demonstrate survival benefits for the pro-estrus state, human observational studies have failed to demonstrate a consistent survival advantage among female patients. Estrogen biosynthesis differs substantially in primate and non-primate animals, and estrogens have diverse immunologic actions. Estrogen concentrations are elevated in response to critical illness and injury (regardless of sex), and elevated concentrations of serum estradiol are associated with a higher mortality rate. Our objective was to determine the predictive ability and test characteristics of the serum estradiol concentration at 48 h in critically ill patients.
Methods: A prospective cohort study of surgical and trauma adult intensive care unit patients at two academic tertiary-care centers. Sex hormones (estradiol, progesterone, testosterone, prolactin, and dehydroepiandrosterone) and cytokines were assayed at 48 h, and the 28-day all-cause mortality rate was assessed.
Results: There was no difference in mortality rates between the sexes (survivors being male in 75.2% of cases vs. 76.0% in non-survivors; p = 0.43). The serum estradiol concentration was significantly elevated in non-survivors regardless of sex (median 18.7 pg/mL [interquartile range {IRQ} 9.99–43.6] in survivors and 40.7 pg/mL [IQR 9.99–94.8] in non-survivors; p < 0.001). The area under the receiver-operating characteristic (ROC) curve for serum estradiol was 0.64 (95% confidence interval [CI] 0.55, 0.72). The parameter with the largest ROC curve was the Acute Physiology and Chronic Health Evaluation (APACHE) II score (0.75; 95% CI 0.68, 0.82). A serum estradiol cut-point of 50 pg/mL was 48% sensitive and 80% specific in predicting death and classified the outcome of 76% of patients correctly.
Conclusions: Serum estradiol concentration is a valuable prognostic tool and potential contributor to adverse outcomes of critically ill or injured surgical patients.
