Glucocorticoids and acute lung injury.

OBJECTIVES

To describe hypothalamic-pituitary-adrenal (HPA) activation and the role of glucocorticoids in immune modulation during critical illness, and to review clinical trials of pharmacologic and "replacement"' doses of glucocorticoids in early and late acute respiratory distress syndrome (ARDS) and sepsis.

DATA EXTRACTION AND SYNTHESIS

Selected review of published literature (1963 to present), clinical trials, and meta-analyses.

DATA SUMMARY

HPA axis activation is an important component of the compensatory anti-inflammatory response to critical illness. Cortisol supports vascular tone and endothelial integrity, modulates a large number of proinflammatory cytokines, and suppresses phospholipase A(2), cyclo-oxygenase, and nitric oxide synthase. Cortisol has putative antifibrotic activities, including inhibition of fibroblast growth and collagen deposition and stimulation of T-cell and monocyte apoptosis. During critical illness, neurohumoral factors, cytokines, endothelin, and atrial natriuretic peptide all may participate in HPA axis activation, resulting in elevated plasma cortisol production and plasma concentrations. In general, cortisol concentrations correlate with severity of illness, and higher plasma concentrations are associated with a poorer outcome. Failure of adrenocorticotropic hormone to augment plasma cortisol appears to be a poor prognostic finding in vasopressor-dependent sepsis and may indicate "relative adrenal insufficiency." Replacement glucocorticoid/mineralocorticoid therapy over 7 days appears to be beneficial in such individuals. However, a meta-analysis of high-dose, short-course glucocorticoid treatment involving 1,297 patients with sepsis enrolled in nine trials showed a trend toward harm, and four trials in patients with, or at risk for, ARDS showed no benefit or a greater likelihood of progression to ARDS. In contrast, observational studies and one small randomized controlled trial suggest that lower pharmacologic doses of glucocorticoids given late (>1 wk) in the course of ARDS may be beneficial. The National Heart, Lung, and Blood Institute's ARDS Network currently is testing the use of methylprednisolone in late ARDS. This study was reviewed by an independent data safety monitoring board for safety and efficacy after enrolling 60 and 120 patients and is currently ongoing.

CONCLUSION

Current evidence indicates that short-duration, high-dose glucocorticoid therapy is not effective for early ARDS or severe sepsis. One small randomized, controlled trial suggests that moderate doses of glucocorticoids may be beneficial for patients with late ARDS; a much larger randomized controlled trial is ongoing. Some patients with pressor-dependent severe sepsis appear to have relative adrenal insufficiency and benefit from replacement glucocorticoid/mineralocorticoid therapy. The accuracy of the diagnostic criteria for, and the prevalence of, relative adrenal insufficiency in patients with acute lung injury/ARDS is unknown. It is also unclear whether such a response, if present, predisposes patients to ongoing lung inflammation and the development of late fibroproliferative ARDS, or if it is predictive of a beneficial response to steroids. Studies of HPA axis activation and the role of relative adrenal insufficiency on the outcome of patients with acute lung injury are needed.