A Single, Post-ACTH Cortisol Measurement to Screen for Adrenal Insufficiency in the Hospitalized Patient.
J Hosp Med. 2018 Feb 07;:E1-E5
Authors: Odom DC, Gronowski AM, Odom E, Clutter W, Thoelke M
Abstract
BACKGROUND: Cosyntropin stimulation testing (CST) is used to screen patients for adrenal insufficiency (AI). Traditionally, CST includes baseline cortisol concentration, the administration of cosyntropin, and cortisol concentration at 30 and 60 minutes poststimulation. There is debate surrounding the utility of testing and cut-off points for concentrations at each time point.
OBJECTIVE: To determine if a single cortisol measurement at 30 or 60 minutes could replace the traditional approach.
DESIGN: looked retrospectively at inpatients who underwent standard, high-dose CST (n = 702) and evaluated the number of patients who would screen positive for AI by using a single time point (30 or 60 minutes) compared with the traditional CST.
SETTING: A tertiary-care, academic medical center.
PATIENTS: Hospital inpatients present between January 2012 and September 2013.
RESULTS: Of tests, 84.3% were normal, which was defined as at least 1 cortisol concentration of 18 mcg/dL or higher at any time after stimulation. The average 60-minute concentration was higher than the average 30-minute concentration (P < .001). A single 60-minute concentration is 100% concordant with the full CST in the intensive care unit (ICU) subgroup and 99.6% concordant in floor patients. A single 30-minute concentration is significantly less concordant, 91.9% and 86.9%, in the ICU and floor subgroups, respectively.
CONCLUSIONS: Overall, a single 60-minute cortisol concentration to screen for AI was 99.7% concordant with the traditional CST, and the positive percent agreement was 98%. Fewer false-positive screens would occur with a single 60-minute cortisol concentration compared with a single 30-minute concentration (P < .001). High-dose CST screening may safely be interpreted with single 60-minute poststimulation cortisol serum concentrations.
PMID: 29444197 [PubMed - as supplied by publisher]
Well written