Relative adrenal insufficiency among trauma patients in a community hospital

Diagnosis and Management of Adrenal Insufficiency in Hospitalized Patients

BACKGROUND

Plasma cortisol is commonly obtained in hospitalized hypotensive patients, and adrenocorticotropic hormone (ACTH) challenge is typically conducted to further workup hypocortisolemia. It is important to recognize that relative adrenal insufficiency (AI) is the most common cause of low cortisol levels and failed ACTH challenge in ill patients. Both cortisol and synthetic ACTH challenge assays are unreliable in critically ill patients. In clinical practice, corticosteroid therapy in septic shock patients results in immediate hemodynamic benefits with less vasopressor and ventilator dependence.

AREAS OF UNCERTAINTY

There is no consensus about the diagnostic criteria of relative AI, appropriate cortisol level, and the dose used for synthetic ACTH challenge in patients with septic shock. There is controversy about the mortality benefits of supplemental steroid therapy and about the use of adjunctive fludrocortisone.

DATA SOURCES

PubMed search of randomized control trials and meta-analyses.

THERAPEUTIC OPINION

Despite all the controversies, hospital physicians frequently use steroids in patients with septic shock with hypocortisolemia. Hydrocortisone should be the choice of steroid for most relative AI patients, and fludrocortisone can be added on a case-by-case basis in refractory shock. Most of the adverse effects induced by a short course of steroids are easily managed in the inpatient setting.

Sugar or salt? The relative roles of the glucocorticoid and mineralocorticoid axes in traumatic shock

BACKGROUND

Glucocorticoid deficiency (GD) has been proposed as a key contributor to shock states, but the presence and role of acute mineralocorticoid deficiency may be of equal or greater significance. We sought to analyze the incidence and degree of acute mineralocorticoid deficiency and GD in an animal model of severe hemorrhage and shock.

METHODS

Fifty-seven swine underwent 35% volume-controlled hemorrhage followed by aortic cross-clamping for 50 minutes to induce truncal ischemia-reperfusion. Protocol-guided resuscitation was performed. Laboratory analysis included cortisol, aldosterone, and plasma renin activity. The aldosterone-to-renin ratio (ARR) was calculated at each time point, and changes were correlated to markers of perfusion.

RESULTS

Mean baseline cortisol levels were 5.8 μg/dL. Following hemorrhage, there was a significant increase in mean cortisol to 9.2 μg/dL (p < 0.001). After 1 hour of reperfusion, there was no change in mean cortisol levels (9.8 μg/dL, p = 0.12). Mean baseline aldosterone was 13.3 pg/mL. Aldosterone levels before cross-clamp removal increased significantly to 115.1 pg/mL (p < 0.001) and then rapidly declined to 49.2 pg/mL (p < 0.001) after 1 hour of reperfusion. Conversely, baseline plasma renin activity was 0.75 ng/mL per hour and increased significantly before cross-clamp removal (1.8) and at 1 hour (8.9, both p < 0.001). The ARR at baseline was 96.1 and increased to 113.5 (p = 0.68) before cross-clamp removal but significantly declined following 1 hour of reperfusion to 7.6 (p < 0.001). Overall, this represented a 93% reduction in mean ARR following reperfusion. The degree of aldosterone deficiency correlated with degree of systemic shock as measured by arterial base deficit (r = 0.47, p = 0.04), while cortisol showed no correlation.

CONCLUSION

Hemorrhagic shock with ischemia-reperfusion injury resulted in only modest impact on the glucocorticoid axis, but major dysfunction of the mineralocorticoid axis and severe hyperreninemic hypoaldosteronism. The degree of aldosterone deficiency may provide prognostic information or offer potential targets for pharmacologic intervention.

LEVEL OF EVIDENCE

Diagnostic study, level III.

Adrenal response after trauma is affected by time after trauma and sedative/analgesic drugs

BACKGROUND

The adrenal response in critically ill patients, including trauma victims, has been debated over the last decade. The aim of this study was to assess the early adrenal response after trauma.

METHODS

Prospective, observational study of 50 trauma patients admitted to a level-1-trauma centre. Serum and saliva cortisol were followed from the accident site up to five days after trauma. Corticosteroid binding globulin (CBG), dehydroepiandrosterone (DHEA) and sulphated dehydroepiandrosterone (DHEAS) were obtained twice during the first five days after trauma. The effect of time and associations between cortisol levels and; severity of trauma, infusion of sedative/analgesic drugs, cardiovascular dysfunction and other adrenocorticotropic hormone (ACTH) dependent hormones (DHEA/DHEAS) were studied.

RESULTS

There was a significant decrease over time in serum cortisol both during the initial 24 h, and from the 2nd to the 5th morning after trauma. A significant decrease over time was also observed in calculated free cortisol, DHEA, and DHEAS. No significant association was found between an injury severity score ≥ 16 (severe injury) and a low (< 200 nmol/L) serum cortisol at any time during the study period. The odds for a serum cortisol < 200 nmol/L was eight times higher in patients with continuous infusion of sedative/analgesic drugs compared to patients with no continuous infusion of sedative/analgesic drugs.

CONCLUSION

Total serum cortisol, calculated free cortisol, DHEA and DHEAS decreased significantly over time after trauma. Continuous infusion of sedative/analgesic drugs was independently associated with serum cortisol < 200 nmol/L.

Mineralocorticoid deficiency in hemorrhagic shock

BACKGROUND

In the critically ill, mineralocorticoid deficiency (MD) is associated with greater disease severity, the development of acute renal insufficiency, and increased mortality. We hypothesized that severely injured trauma patients presenting with hemorrhagic shock would demonstrate a high degree of MD. We also hypothesized that MD in these patients would be associated with increased length of stay, hypotension, fluid requirements, and acute kidney injury (AKI).

MATERIALS AND METHODS

Thirty-two trauma patients in hemorrhagic shock on admission to the trauma bay (SBP <90 mm Hg × 2) were enrolled. Blood samples were obtained on ICU admission and 8, 16, 24, and 48 hours later. Plasma aldosterone (PA) and renin (PR) were assayed by radioimmunoassay. MD was defined as a ratio of PA/PR ≤2. Demographic data, injury severity score, ICU and hospital length of stay, fluid requirements, mean arterial pressure, serum sodium, hypotension, and risk for AKI were compared for patients with and without MD.

RESULTS

At ICU admission, 48% of patients met criteria for MD. Patients with MD were significantly more likely to experience hypotension (MAP ≤60 mm Hg) during the study period. MD patients required significantly more units of blood in 48 h than non-MD patients (13 [7-22] versus 5 [2-7], P = 0.015) and had increased crystalloid requirements (18L [14-23] versus 9L [6-10], P < 0.001). MD patients were at higher risk for AKI according to RIFLE and AKIN criteria.

CONCLUSIONS

MD is a common entity in trauma patients presenting in hemorrhagic shock. Patients with MD required a more aggressive resuscitative effort, were more likely to experience hypotension, and had a higher risk of AKI than non-MD patients. Future studies are needed to fully understand the impact of MD following trauma and the potential role for hormonal replacement therapy.

Acute Adrenal Insufficiency May Affect Outcome in the Trauma Patient

Acute adrenal insufficiency in the trauma patient is underrecognized and the impact poorly understood. Our hypothesis was that the identification and treatment of acute adrenal insufficiency reduces mortality in trauma patients. Institutional Review Board approval for the retrospective review of a prospective database from a Level 1 trauma center for 2002 to 2004 was obtained. The study population included patients receiving a cosyntropin stimulation test (250 μg) and/or random Cortisol level based on our practice management guideline and an intensive care unit stay longer than 24 hours. Demographic, acuity, and outcome data were collected. The nonresponders had baseline Cortisol levels less than 20 μg/dL or poststimulation rise less than 9 μg/dL. Independent t tests and χ2 statistics were used. One hundred thirty-seven patients had cosyntropin stimulation tests performed. Eighty-two (60%) patients were nonresponders of which 66 were treated with hydrocortisone and 16 went untreated as a result of the discretion of the attending physician. The 55 (40%) responders showed no statistical differences in outcome variables whether or not they received hydrocortisone. The untreated adrenal-insufficient patients had significantly higher mortality, longer hospital length of stay, intensive care unit days, and ventilator-free days. Conclusions were: 1) treatment of acute adrenal insufficiency reduces mortality by almost 50 per cent in the trauma patient; and 2) acute adrenal insufficiency recognized by low random Cortisol levels or nonresponse to a stimulation tests should be considered for treatment.

Time-course changes in nuclear translocation of hepatic glucocorticoid receptor in rats after burn trauma and its pathophysiological significance

Nuclear translocation is a determining step for the glucocorticoid receptor (GR) to exert its functions in response to traumatic conditions. This study was designed to observe the nuclear translocation changes of hepatic GR in severely burned rats during early postburn stage, and to explore the effects of high-dose dexamethasone on GR nuclear translocation. Rats with 35% total body surface area full-thickness burn injury, parallelized with a sham-burn group, were killed at consecutive time points to examine the changes in plasma corticosterone and expression of hepatic GR at both whole-cell and nuclear levels. The effects of high-dose dexamethasone on GR nuclear translocation and suppression of proinflammatory cytokine overproduction were subsequently analyzed. In burned rats, plasma corticosterone increased remarkably soon after burn injury. On the contrary, the hepatic GR levels showed an initial phase of decrease as measured in both whole-cell and nucleus by Western blot, followed by a rapid elevation in the nucleus but a slow recovery at whole-cell level. By comparing the changes of GR in both whole-cell and nuclear levels, we found that GR nuclear translocation was relatively enhanced in the early postburn period. High-dose dexamethasone administered at 1 or 48 h postburn did not further elevate GR nuclear translocation, neither did it restrain the increased release of proinflammatory cytokines such as TNF-alpha and IL-1 beta. These studies suggest that although the whole-cell level of hepatic GR is decreased, GR nuclear translocation is relatively enhanced at early postburn stage. High-dose exogenous glucocorticoids may not promote more nuclear translocation of GR to reinforce its functions.