Title: Reproducibility of the Corticotropin Stimulation Test in Septic Shock
Abstract: To the Editor: We read with interest the article by Loisa et al. (1) regarding the poor reproducibility of the corticotropin (ACTH) stimulation test in septic shock. They concluded that a single ACTH test may not adequately diagnose adrenal insufficiency (AI) in septic shock patients. First, we would like to report our experience with repeated ACTH stimulations in septic shock; second, we question the methodology and, in particular, the conclusion of the cited article. We studied 15 septic shock patients (8 men and 7 women), having a median age of 72 yr. Patients underwent two small-dose (1 mcg) ACTH stimulation tests within 48 h. Blood was drawn before the tests to measure baseline cortisol levels and again at 30 min after the tests to determine stimulated cortisol. APACHE II scores at the times of the two tests were 26 ± 4 and 26 ± 6, respectively. There was a positive correlation between baseline cortisol levels obtained during the two tests (r = 0.65, P = 0.01) (Fig. 1, top). Similarly, there was a correlation between stimulated cortisol levels from the two tests (r = 0.76, P = 0.001) (Fig. 1, bottom). In contrast, there was no correlation between delta cortisol—the difference between the two levels (r = 0.27, P = 0.34) (Fig. 2).Figure 1.: Correlations between baseline cortisol (top) and stimulated cortisol (bottom) levels on tests No. 1 and No. 2. Dotted lines represent the thresholds of 15 and 18 mcg/dL, respectively, which separate patients with normal adrenal function from those with adrenal insufficiency.Figure 2.: Correlation between delta cortisol on tests No. 1 and No. 2. Dotted lines represent the threshold of 9 mcg/dL, which separates patients with normal adrenal function from those with adrenal insufficiency.In agreement with the results of Loisa et al., our findings suggest that in patients with septic shock, the ACTH stimulation test is not reproducible if delta cortisol is taken into consideration. On the other hand, our data point toward a good correlation between stimulated cortisol concentrations obtained in two consecutive ACTH tests. Why did Loisa et al. use only the delta cortisol responses to determine the reproducibility of the ACTH test? Were there any correlations between stimulated cortisol concentrations on Day 1 and 2 in their septic shock patients? Furthermore, the P value obtained (P = 0.08) suggests that by studying more patients, this relationship might become statistically significant. To summarize, we feel that the study of Loisa et al. does not give a definite answer regarding the reproducibility of the ACTH stimulation test in patients with septic shock. A larger cohort along with a more complex analysis might clarify this important issue. Ioanna Dimopoulou, MD Maria Zervou, MD Evangelia Douka, MD Callirhoe Augustatou, MD Apostolos Armaganidis, MD 2nd Department of Critical Care Medicine Attikon University Hospital Athens University—Medical School [email protected] Stylianos Tsagarakis, MD Department of Endocrinology Athens' Polyclinic Athens, Greece