Research suggests that women experience greater cardiovascular ischemic effects from stress than men. Visceral adiposity is an endocrine tissue that differs by sex and interacts with stress hormones. We hypothesized that urinary cortisol would be associated with increased cardiovascular events and change in coronary artery calcium score (CAC) in women, and these relationships would vary by central obesity. In the Multi-Ethnic Study of Atherosclerosis Stress Ancillary study, cortisol was quantified by 12-h overnight urine collection. Central obesity was estimated by waist-hip ratio (WHR). Multivariable Cox models estimated the relationship between cortisol and cardiovascular events and assessed for moderation by WHR. The relationship between cortisol and change in CAC Agatston score was assessed by Tobit regression models. 918 patients were analyzed with median follow up of 11 years. There was no association between urinary cortisol and cardiovascular events in the cohort. However, in individuals with below median WHR, higher urinary cortisol levels (upper tertile) were associated with higher cardiovascular event rates in the full cohort, women, and men, but not in groups with above median WHR. There was significant moderation by WHR in women, but not men, whereby the association between elevated cortisol and increased cardiovascular events diminished as WHR increased. Urinary cortisol was associated with increased change in CAC in women (P = 0.003) but not men, without moderation by WHR. Our study highlights associations between cortisol and subclinical atherosclerosis in women, and moderation of the relationship between cortisol and cardiovascular events by central obesity in both genders.

Background

Sepsis is a common cause of death in patients with pulmonary arterial hypertension (PAH). Treatment requires careful fluid management and hemodynamic support. This study compares patients with or without PAH presenting with sepsis with a focus on initial fluid resuscitation.

Methods

This retrospective analysis compared adults with and without PAH admitted for sepsis at two academic hospitals between 2013 and 2022. Prior PAH diagnosis was verified by review of right heart catheterization data and sepsis present on admission was verified by chart review. Demographics, vital signs, laboratory values, imaging results, treatment approaches, and all-cause mortality data were obtained. Controls were propensity score weighted by age, sex, and Charlson Comorbidity index. Logistic regression models controlling for age and Charlson comorbidity indices were used to examine factors associated with survival.

Results

Thirty patients admitted for sepsis with pre-existing PAH were compared to 96 matched controls. Controls received significantly more fluids at 24 h compared to PAH patients (median 0 mL v. 1216 mL, p < 0.001), while PAH patients were more likely to receive vasoactive medications (23.3% vs. 8.3%, p = 0.037). At 30 days, 7 PAH patients (23.3%) and 13 control patients (13.5%) had died (p = 0.376). PAH patients that received more fluids had decreased mortality (OR 0.31, 95% CI 0.11-0.92, p = 0.03) and patients who received fluids had shorter mean time to antibiotics (2.3 h v. 6.5 h, p = 0.04), although decreased time to antibiotics was not associated with mortality. Patients who received no fluids more often had previously identified right ventricular systolic dysfunction (62.5% v. 28.6%, p = 0.136).

Conclusion

Patients with PAH and sepsis have high mortality and receive different treatments than controls, with more reliance on vasopressors and less on fluid resuscitation. PAH patients who received less fluids had higher mortality and those who received no fluids had a longer time to receiving antibiotics, indicating a potential delay in recognizing sepsis. Timely recognition of sepsis and dynamic decision-making around fluid resuscitation remains critical in this high-risk population.