Adrenal Insufficiency and Response to Stress Dose Hydrocortisone in Patients With Cirrhosis and Vasopressor Dependency Using Cirrhosis-Specific Cortisol Thresholds

Background: Cortisol thresholds defining adrenal insufficiency (AI) in some cirrhosis-specific studies differ from those recommended by the SCCM/ESICM (Society of Critical Care Medicine/European Society of Intensive Care Medicine) guidelines, which may influence treatment decisions. Objective: To determine if stress-dose hydrocortisone (HC) improves outcomes in vasopressor-dependent patients meeting cirrhosis-specific criteria for AI. Methods: In this retrospective study, AI was defined using criteria from 2 studies in critically ill cirrhosis patients showing mortality reduction with HC (random cortisol <20 µg/dL, or if a standard-dose cosyntropin test was performed, baseline cortisol <15 µg/dL or delta cortisol <9 µg/dL if baseline = 15-34 µg/dL). Use of HC was at the discretion of the intensivist. The primary endpoint was days of vasopressor therapy. Secondary endpoints included hospital mortality and newly acquired infections. Sixty-four patients were evaluated; 40 patients received HC and 24 did not. Results: Mean random cortisol was significantly lower in the HC group (9.8 ± 3.2 vs 12.0 ± 3.7 µg/dL, P = 0.04). Delta cortisol also tended to be lower in the HC group (8.2 ± 4.4 vs 11.3 ± 6.4 µg/dL, P = 0.25). Patients in the HC group exhibited significantly fewer median days of vasopressor therapy (4.0 [2.0-7.0] vs 7.0 [4.2-10.8], P = 0.006), lower mortality (22.5% vs 50%, P = 0.02), and a similar incidence of newly acquired infections. Conclusion and Relevance: The use of HC in patients meeting cirrhosis-specific criteria for AI resulted in significantly shorter duration of vasopressor therapy, lower mortality, and no increased risk of infection. Use of traditional AI definitions may exclude patients with cirrhosis that could benefit from HC therapy.