Abstract
BACKGROUND: Relative adrenal insufficiency (RAI) and hyponatremia are both common in decompensated cirrhosis. The literature is scarce on the prevalence and clinical correlates of RAI in patients with acute decompensation (AD) of cirrhosis and hyponatremia. METHODS: Patients with a diagnosis of AD and serum sodium <135 mmol/l were included in this cohort study. Adrenal axis was assessed using the adrenocorticotrophic hormone (ACTH) stimulation test. Those with baseline cortisol <35 μg/dl and rise in cortisol<9 μg/dl were diagnosed to have RAI. ACTH stimulation was done using intravenous cosyntropin in 74 patients and subcutaneous long-acting porcine ACTH (AP) in 42 patients due to availability issues. Proportion of patients with RAI was the primary outcome. Secondary outcomes included in-hospital mortality, recurrent hospital admissions, 90-day mortality and its predictors. RESULTS: Of 116 patients who were enrolled (mean age: 46.5 ± 10.9 yrs; male: 89.7%, alcohol etiology: 78.5%; Child-Turcotte-Pugh score: 81%), RAI was diagnosed in 69 (59.5%) cases. The baseline cortisol, peak cortisol and proportion of patients with RAI were not different between those who underwent CTH stimulation using cosyntropin or AP. The 90-day mortality was significantly higher in the RAI as compared to non-RAI group [47 (68.1%) vs 21 (46.7%), P = 0.01]. In-hospital mortality and recurrent hospital admissions were not different between the groups. On multivariate logistic regression analysis, the model for end-stage liver disease (MELD) score and presence of RAI independently predicted 90-day mortality. On stratified survival analysis, RAI predicted 90-day mortality only in MELD<25. CONCLUSION: RAI portends poor short-term outcomes and independently predicts 90-day mortality in patients with AD and hyponatremia. Subcutaneous AP is a feasible alternative for diagnosing RAI in this population. IRB NUMBER: AIIMS/IEC/23/49.