Sex, Acute Kidney Injury, and Age: A Prospective Cohort Study

Abstract

Animal models of kidney disease suggest a protective role for female sex hormones but in humans, some authorities assert that female sex is a risk factor for acute kidney injury (AKI). To better understand the risk of AKI, we studied the strength of association between sex and AKI incidence in hormonally distinct age groups across the life span. Prospective cohort study. All patients hospitalized in the Montefiore Health System between 10/15/2015 and 1/1/2019, excluding those with kidney failure or obstetrics diagnoses. Male versus female sex. Acute kidney injury (AKI) occurring during hospitalization based on KDIGO definitions. Generalized Estimating Equation logistic regression adjusted for comorbidities, socio-demographic factors, and severity of illness. Analyses were stratified into 3 age categories, 6 months to ≤16 years, age >16 years - <55 years, and age ≥55 years. A total of 132,667 individuals were hospitalized a total of 235,629 times. The mean age was 55.2 (SD 23.8) years. The counts (%) of hospitalizations for women were 129,912 (55%). Hospitalization counts (%) among Black and Hispanic patients were 71,834 (30.5%) and 24,199 (10.3%), respectively. AKI occurred in 53,926 (22.9%) hospitalizations. In adjusted models, there was a significant interaction between age and sex (p<0.001). Boys and men had higher risk of AKI across all age groups, an association more pronounced in the age group >16 years to <55 years in which the OR for men was 1.7 (95% CI, 1.6-1.8). This age-based pattern remained consistent across prespecified types of hospitalizations. In a sensitivity analysis, women older than 55 years who received prescriptions for estrogen had lower odds of AKI than those without prescriptions. Residual confounding. The greatest relative risk of AKI for males occurred during ages >16 to <55 years. The lower risk among post-menopausal women receiving supplemental estrogen supports a protective role for female sex hormones.