RESUMO
BACKGROUND: Adrenal insufficiency (AI) has been extensively described in sepsis but not in acute hemorrhage. We sought to determine the incidence of hyperacute AI (HAI) immediately after hemorrhage and its association with mortality. METHODS: Patients with acute traumatic hemorrhagic shock presenting to the R Adams Cowley Shock Trauma Center prospectively had serum cortisol levels collected on admission. Inclusion criteria were hypotension and active hemorrhage. Clinicians were blinded to results, and no patient received steroids in the acute phase. The primary outcome measure was death from hemorrhage within 24 hours of admission. RESULTS: Fifty-nine patients were enrolled during an 8-month period. Mean admission cortisol level was 18.3 ± 8.9 µg/dL. Acute mortality rate from hemorrhage was 27%. Overall mortality rate was 37%. Severe HAI (serum cortisol level <10 µg/dL) was present in 10 patients (17%). Relative HAI (<25 µg/dL) was present in 51 patients (86%). Those who died of acute hemorrhage had significantly lower mean cortisol levels (11.4 ± 6.2 µg/dL vs. 20.9 ± 8.4 µg/dL, p < 0.001) as did patients who ultimately died in the hospital (12.8 ± 7.6 µg/dL vs. 21.6 ± 8.1 µg/dL, p < 0.001). In multivariate analysis, cortisol levels were associated with mortality from acute hemorrhage, with an odds ratio of 1.17 (95% confidence interval, 1.02-1.35). Adjusted receiver operating characteristic analysis indicated that serum cortisol has a 91% accuracy in differentiating survivors of acute hemorrhage from nonsurvivors. CONCLUSION: This study is the first to report that AI occurs immediately after acute injury during hemorrhagic shock and is strongly associated with mortality. HAI may be a marker of depth of shock but is potentially rapidly modifiable as opposed to other markers, such as lactate or base deficit. Further work is needed to determine whether steroid administration can change outcome in selected patients. LEVEL OF EVIDENCE: Prognostic/epidemiologic study, level III.
