Abstract Disclosure: A.J. Han: None. S.J. Achenbach: None. B.J. Atkinson: None. I. Bancos: None. Background: Diagnosis of adrenal insufficiency (AI) is often delayed due to its vague and non-specific symptomatology. As classic symptoms of AI such as fatigue and nausea are prevalent in the general population, identifying clinical features that are more specific to AI may help clinicians determine which patients require additional testing. Methods: A single-center retrospective study was performed of patients who underwent outpatient Cosyntropin stimulation testing (CST) between 2005-2018. Patients with oral glucocorticoid or estrogen use within 2 months of CST, and patients without available Current Visit Information (CVI) form at the time of CST were excluded. Presence of symptoms was obtained from CVI, and laboratory findings were electronically extracted. Adrenal insufficiency was defined as post-CST peak cortisol <18 mcg/dL. Morbidity was assessed using the Elixhauser index. Variables for multivariable analysis were defined a priori and included fatigue, dizziness, muscle pain, joint pain, weight loss >10 pounds, abdominal pain, nausea/vomiting, and hyponatremia. Results: Among 5347 patients who underwent CST, 376 (7.0%) of patients were found to have AI. There were no sex differences between patients with and without AI, however patients with AI were slightly older (median age 48.9 vs 47.1 years, p=0.026), and had a higher median BMI (26.5 vs 25.6 kg/m2, p=0.002). Prevalence of comorbidities was similar between the two groups, with a median Elixhauser index of 1 (IQR 1-3), however, patients with AI were more likely to report difficulties with activities of daily living (69% vs 57%, p<0.001). Hyponatremia was diagnosed in 4% of patients with AI and 3% of those without AI, p=0.208. Although patients in our cohort reported a high prevalence of fatigue (79%), dizziness (50%), joint pain (50%), muscle pain (48%), nausea (40%), abdominal pain (39%), and weight loss in the past 6 months (29%), there were no statistically significant differences in reported symptoms between patients with and without AI with the exception of joint pain (59% vs 49%, p<0.001) and muscle pain (54% vs 47%, p=0.009). Based on multivariable analysis, only joint pain (OR 1.51, 95% CI 1.14-1.99) and weight loss >10 pounds in the past 6 months (OR 1.31, 95% CI 1.02-1.68) were associated with higher odds of AI diagnosis. Interestingly, patients who reported fatigue had lower odds of AI diagnosis on CST (OR 0.66, 95% CI 0.49-0.88). Conclusions: Commonly associated symptoms of AI such as fatigue, dizziness, nausea, and abdominal pain were poor predictors of AI diagnosis based on CST. Consequently, the presence of these symptoms alone may not warrant additional testing. Notably, clinical features such as myalgia, arthralgia, and weight loss >10 pounds in the past 6 months were found to be more predictive of AI. Presentation: 6/1/2024
Read full abstract