Adrenal Dysfunction Does Not Predict Increased Rates of Portal Hypertension-Related Decompensation in Outpatients With Decompensated Cirrhosis

Introduction: Adrenal dysfunction (AD), defined as inadequate cortisol production in response to physiologic demand, is common in cirrhosis and associated with increased mortality in critically ill patients. In contrast, clinical outcomes in non-critically ill patients with adrenal dysfunction are unknown. We designed a prospective study to evaluate various clinical outcomes in outpatients with AD. Methods: Adults with cirrhosis and Child-Pugh Score B or C were recruited from outpatient clinics. Key exclusion criteria were alcohol use within 3 months or medications known to affect the hypothalamic-pituitary-adrenal axis. Participants were administered a standard-dose (250µg) Cosyntropin stimulation test and AD was defined as an increase in total cortisol level of < 9 µg/dL at 60 minutes. Patients were followed for 6 months or until death or transplant (transplant-free survival). Portal hypertension (pHTN) related hospitalizations were defined as admissions for volume overload, hepatic encephalopathy, or bleeding varices. pHTN-related decompensation was defined as development of SBP; ascites requiring diuretic initiation, uptitration, new paracentesis, or TIPS; variceal bleed; or hepatic encephalopathy requiring medication titration. Multivariate logistic regression was performed to assess predictors for decompensation. Results: Analysis of 59 patients revealed 18 (31%) meeting criteria for AD. Baseline characteristics including age, sex, comorbidity index, and MELD were similar (Table 1). Outcomes at 6 months were notable for no differences in transplant, mortality, or pHTN-related hospitalizations. Patients with AD were not more likely to have decompensations related to pHTN at 6 months compared to patients without AD (83.3% vs 68.3%, P=0.38). On multivariable logistic regression, the presence of AD was not associated with pHTN-related decompensation (OR 3.95, 95% CI 0.94-21.9, P = 0.08), even when accounting for age, MELD, and co-morbidity index. Conclusion: In this prospective study of outpatients with decompensated cirrhosis, the presence of AD was not associated with developing pHTN-related decompensations or hospitalizations. Transplant-free survival was independent of adrenal functionality. These results suggest that the presence of adrenal dysfunction may be independent of progressive liver dysfunction and its prognostic implications may differ based on the clinical setting. Table 1. - Analysis of Patients with and without Adrenal Dysfunction No Adrenal Dysfunction Adrenal Dysfunction Selected Demographics n=41 n=18 P-value Age (years) 57.8 (11.6) 60.5 (9.3) 0.39 Female (%) 22 (53.7) 9 (50.0) 1.00 MELD Score 12.8 (3.8) 13.2 (3.4) 0.71 CCI 2.2 (1.4) 2.1 (1.3) 0.68 Cortisol at 0 minutes (µg/dL) 7.1 (3.7) 9.5 (4.3) 0.03* Cortisol at 60 minutes (µg/dL) 19.4 (4.2) 17.0 (3.9) 0.04* ∆Cortisol (µg/dL) 12.3 (2.8) 7.5 (1.2) < 0.001* Outcomes Transplant (%) 2 (4.9) 2 (11.1) 0.75 Mortality (%) 1 (2.4) 0 (0) 1.00 pHTN Hospitalizations (%) 8 (19.5) 3 (16.7) 1.00 Non-pHTN Hospitalizations (%) 8 (19.5) 3 (16.7) 1.00 Decompensation (%) 28 (68.3) 15 (83.3) 0.38 Multivariate Analysis for pHTN-related Decompensation Independent Variable OR 95% CI P-value Age 0.85 0.74 – 0.95 0.01* MELD Score 1.05 0.87 – 1.26 0.62 CCI 3.27 1.29 – 11.0 0.03* Adrenal Dysfunction 3.95 0.94 – 21.9 0.08 AD: Adrenal dysfunction; pHTN: portal hypertension; CCI: Charlson Comorbidity Index; OR: odds ratio; CI: confidence interval; Continuous variables presented as means (SD). Categorical variables presented as counts (percentage); *: statistically significant with P < 0.05