Relative adrenal insufficiency in patients with septic shock; a close look to practice patterns

Impact of underlying chronic adrenal insufficiency on clinical course of hospitalized patients with adrenal crisis: A nationwide cohort study

BACKGROUND

Chronic adrenal insufficiency (AI) is an established risk factor for adrenal crisis (AC). However, the proportion of patients with newly diagnosed chronic AI during admission for AC is unclear.

METHODS

This retrospective cohort study used a Japanese claims database involving 7.39 million patients at 145 acute care hospitals between 2003 and 2014. Study patients with AC met these criteria: 1) newly coded in claims as AI; 2) glucocorticoid therapy administered; 3) admission; and 4) age ≥ 18 years. We investigated the prevalence of underlying chronic AI and assessed in-hospital mortality. Additionally, we explored risk factors for in-hospital mortality through multivariate analysis using a Cox proportional hazards model.

RESULTS

Among 504 patients with AC, chronic AI was diagnosed before and during admission in 73 (14.5%) and 86 (17.1%) patients, respectively. In-hospital mortality rates were 1.4% and 5.8%, respectively, lower than that of the total population (14.1%). Significant risk factors for increased mortality were: age (hazard ratio [HR] 1.45/10 years; 95% confidence interval [CI] 1.17-1.78), requiring mechanical ventilation (HR 3.81; 95% CI 1.88-7.72), vasopressor administration (HR 2.05; 95% CI 1.16-3.64), sepsis (HR 3.79; 95% CI 1.57-9.14), AI-related symptoms (HR 2.00; 95% CI 1.02-3.93), and liver disease (HR 3.24; 95% CI 1.10-9.58).

CONCLUSIONS

Relative to patients without chronic AI, those diagnosed before admission tended to survive to discharge; however, the difference with those diagnosed during admission was not significant. Hospital admission due to nonspecific AI-related symptoms was associated with an increased risk of in-hospital mortality.

The utility of the corticotropin test to diagnose adrenal insufficiency in critical illness: an update

OBJECTIVE

One of the most common dynamic testing procedures for assessment of adrenocortical function is the standard corticotropin or the cosyntropin test. The aim of this review was to examine the evidence base underlying the corticotropin test in the management of the critically ill patient.

DATA SYNTHESIS

The principle behind the corticotropin test is the demonstration of an inappropriately low cortisol production in response to exogenous ACTH, a situation analogous to physiological stress. The corticotropin test was originally described in nonstressed subjects, and its applicability and interpretation in the setting of critical illness continues to generate controversy. Attempting to determine the prevalence of an abnormal corticotropin test in critical illness is complicated by the use of different end-points and different populations. Moreover, the test result is also influenced by the assay used for measurement of plasma cortisol. Trials assessing the relationship between corticotropin response and severity of stress and organ dysfunction have produced divergent results, which may reflect differences in the methodology and the association being measured. Moreover, controversy exists with respect to the methodology and the interpretation with respect to the following variables: dose of corticotropin, end-points for assessment of total or free cortisol, effect of plasma cortisol variability, adrenal blood flow and its equivalence with other tests of adrenocortical function.

CONCLUSIONS

The corticotropin test is used widely in the evaluation of adrenocortical function in the endocrine clinics. Its role in the critically ill patient is less well established. Several confounding variables exist and to have a 'one-size-fits-all' approach with a single end-point in the face of several methodological and pathophysiological confounders may be flawed and may result in the institution of inappropriate therapy. The current evidence does not support the use of the corticotrophin test in critical illness to assess adrenocortical function and guiding steroid therapy in critical illness.

Stress response in critical illness

Sepsis brings about neuroendocrine dysfunction in children that differs significantly from that of adults and can thus be difficult to interpret and manage. Aggressive treatment of sepsis with appropriate and judicious use of antibiotics remains a top priority. Strict glycemic control in children has been associated with significant risk of hypoglycemia, which may independently contribute to morbidity and mortality. Timely initiation of hydrocortisone in persistently hypotensive children with fluid-refractory, catecholamine-resistant shock is controversial, but its use in children with suspected or proven adrenal insufficiency is suggested. Fluid and electrolyte abnormalities must be corrected. Treatment of thyroid dysfunction has been shown to be beneficial in certain specific populations but cannot be extrapolated to all septic patients with the current available data.

Critical illness-related corticosteroid insufficiency in cancer patients

PURPOSE

Critically ill cancer patients with sepsis represent a high-risk sub-group for the development of critical illness-related corticosteroid insufficiency (CIRCI); however, the incidence of CIRCI in this population is unknown. The purpose of this study was to determine the incidence of CIRCI in cancer patients with severe sepsis or septic shock.

METHODS

A single-center, retrospective, observational study was conducted in a 52-bed medical-surgical intensive care unit of a National Cancer Institute-recognized academic oncology institution. Eighty-six consecutive patients with a diagnosis of severe sepsis or septic shock who received a high-dose 250-μg cosyntropin stimulation test were included. CIRCI was identified by a maximum delta serum cortisol of 9 μg/dL or less post cosyntropin.

RESULTS

Overall, 59% (95% CI, 48-70%) of cancer patients with severe sepsis or septic shock were determined to have CIRCI. When compared to patients without CIRCI, patients with CIRCI had higher baseline serum cortisol (median, 26.3 versus 14.7 μg/dL; p = 0.002) and lower delta cortisol levels (median, 3.1 versus 12.5 μg/dL; p < 0.001). Mortality did not differ between the two groups. An inverse relationship was identified between baseline serum cortisol and maximum delta cortisol (maximum delta cortisol = -0.27 × baseline cortisol + 14.30; R (2) = 0.208, p < 0.001).

CONCLUSIONS

The incidence of CIRCI in cancer patients with severe sepsis or septic shock appears high. Further large-scale prospective trials are needed to confirm these findings.

Adrenocortical insufficiency in horses and foals

The adrenal cortices produce various steroid hormones that play vital roles in several physiologic processes. Although permanent adrenocortical insufficiency is rare in all species, emerging evidence in both human and equine medicine suggests that transient reversible adrenocortical dysfunction resulting in cortisol insufficiency frequently develops during critical illness. This syndrome is termed relative adrenal insufficiency (RAI) or critical illness-related corticosteroid insufficiency (CIRCI) and can contribute substantially to morbidity and mortality associated with the primary disease. This review discusses the mechanisms, diagnosis, and clinical consequences of adrenocortical insufficiency, with particular focus on the current understanding of RAI/CIRCI in horses and foals.

Practice variability in the assessment and treatment of critical illness-related corticosteroid insufficiency

PURPOSE

Little is known about how published evidence regarding use of steroids in septic shock has been incorporated into clinical practice.

MATERIALS AND METHODS

All patients admitted to an intensive care unit at a single, tertiary-care, academic medical center from November 1, 2004, through February 28, 2005, were screened using the hospital's computerized pharmacy database to determine if they had received at least 1 qualifying medication: cosyntropin, fludrocortisone, hydrocortisone, or dopamine, epinephrine, or norepinephrine as a vasopressor infusion.

RESULTS

Only 58% (95% confidence interval, 47%-69%) of the 81 patients who met criteria for vasopressor-dependent septic shock (VDSS) were evaluated for critical illness-related corticosteroid insufficiency. Forty-seven percent of the 81 patients who met the criteria for VDSS and 49% of the 47 patients who did not meet the criteria for VDSS were treated with corticosteroids. Nearly all (85%; 95% confidence interval, 72%-94%) patients who did not meet the criteria for VDSS received an adrenocorticotropic hormone test.

CONCLUSIONS

Treatment and evaluation of critical illness-related corticosteroid insufficiency in critically ill patients at our institution are inconsistent. Many patients with VDSS do not receive either treatment or evaluation for critical illness-related corticosteroid insufficiency, and patients who do not meet the current criteria are being evaluated and/or treated for critical illness-related corticosteroid insufficiency.

Baseline plasma cortisol and ACTH concentrations and response to low-dose ACTH stimulation testing in ill foals

OBJECTIVE

To evaluate baseline plasma cortisol and ACTH concentrations and responses to low-dose ACTH stimulation testing in ill foals.

DESIGN

Cross-sectional study.

ANIMALS

58 ill foals.

PROCEDURES

Baseline cortisol and ACTH concentrations and cortisol concentrations after administration of a low dose of cosyntropin were determined within 6 hours after admission. Foals were assigned to 4 groups on the basis of age (<or=24 hours vs 1 to 56 days) and presence of septicemia (yes vs no). Values were compared among groups and with values previously reported for healthy foals.

RESULTS

Plasma cortisol concentrations 30 and 60 minutes after cosyntropin administration in foals<or=24 hours old were significantly higher than corresponding cortisol concentrations in older foals. In all 4 groups, plasma cortisol concentration 30 minutes after cosyntropin administration was significantly higher than baseline cortisol concentration or concentration 60 minutes after cosyntropin administration. No differences in baseline cortisol or ACTH concentration or in the ACTH-to-cortisol ratio were detected between groups or when ill foals were compared with healthy foals. A small number of ill foals had low baseline cortisol and ACTH concentrations or low responses to cosyntropin administration, compared with healthy foals.

CONCLUSIONS AND CLINICAL RELEVANCE

Results indicated that most ill foals in the present study population had adequate responses to cosyntropin administration. However, a small subset of ill foals appeared to have dysfunction of the hypothalamic-pituitary-adrenal axis.

Neuroendocrine dysfunction in pediatric critical illness

BACKGROUND

The extent of neuroendocrine dysfunction (NED) has not been well defined in critically ill children and likely varies significantly from that in adults. We sought to define the prevalence of neuroendocrine dysfunction in a group of children in a multidisciplinary pediatric intensive care unit and determine the relationship of neuroendocrine dysfunction with severity of illness and presence of sepsis.

METHODS

Prospective observational study in a pediatric intensive care unit at a referral childrens hospital. Blood samples were evaluated within 12 hrs of admission for serum cortisol, thyroid stimulating hormone, total triiodothyronine (T3), reverse triiodothyroine (rT3), free thyroxine, and arginine vasopressin. Pediatric risk of mortality, pediatric logistic organ dysfunction scores, and length of stay were calculated.

RESULTS

Seventy-three children were enrolled over a 13-month period. Median patient age was 72 months (range, 3-228 months). Overall prevalence of absolute adrenal insufficiency ranged from 7% to 58% based on cortisol cutoff chosen. Presence of absolute adrenal insufficiency, low T3 syndrome (LT3S), or vasopressin insufficiency did not differ between septic or nonseptic patients. NED did not correlate with pediatric logistic organ dysfunction, Pediatric Risk of Mortality Score III, length of stay, or mortality. Prevalence of multiple NED was 62% (28 of 45 children), where 62% had 2 neurohormonal deficiencies and 24% had 3 neurohormonal deficiencies.

CONCLUSION

NED is common in both septic and nonseptic critically ill children in a single pediatric intensive care unit. Larger scale studies are necessary to determine whether presence of NED, or specific combinations of neurohormonal dysfunction, is important in predicting outcomes or benefit of early hormonal replacement therapies in critically ill children.

Adrenal exhaustion in septic patients with vasopressor dependency

PURPOSE

The use of low-dose corticosteroids for patients with septic shock who remain vasopressor dependent after adequate fluid resuscitation is recommended, but there is lack of agreement on how to diagnose relative adrenal insufficiency (RAI) and when to start steroid supplementation among these patients. This case series reports changes in cortisol concentrations during the course of vasopressor-dependent septic shock.

METHODS

Observational study was performed at a university hospital medical intensive care unit. Consecutive adult patients with vasopressor-dependent septic shock admitted to the medical intensive care unit were studied. Clinical data, cortisol concentrations, and dose of vasopressor agents at different times during the course of septic shock were recorded and reported as mean +/- SD.

RESULTS

Thirteen patients were included. Mean age was 59 +/- 15 years. Mean basal nonstimulated cortisol level was 41.7 +/- 30.9 microg/dL (within 24 hours of intensive care unit admission in all but 2 patients). Steroids were initiated in 8 patients and then discontinued after cortisol values were obtained and RAI was ruled out. Because of inability to discontinue vasopressor support, cortisol testing was repeated after 6.2 +/- 4.8 days of initial assessment. Repeated concentrations were 10.0 +/- 6.3 mug/dL (P < .001). Steroids were then reinitiated, and resolution of vasopressor dependence was achieved 1.5 +/- 1.4 days later.

CONCLUSIONS

Adrenal function in the critically ill is a dynamic process, and an appropriate initial adrenal response does not preclude later development of RAI.

Low-dose steroid therapy does not affect hemodynamic response in septic shock patients

PURPOSE

Several studies showed that low-dose steroid therapy (LDST) in patients with septic shock leads to a significantly shorter duration of shock and a decreased mortality. However, these results have been criticized. Our purpose was to evaluate the effects of LDST on time to shock reversal and mortality in septic shock.

MATERIALS AND METHODS

We retrospectively studied 203 patients with septic shock admitted to the intensive care unit of our tertiary hospital. A short corticotropin test was performed in all patients within 72 hours of septic shock onset. We performed a propensity score analysis through a logistic regression model with baseline relevant characteristics, and evaluated the influence of LDST on time to shock reversal and inhospital mortality.

RESULTS

One hundred twenty-four patients were treated with LDST (steroid group) and 79 without LDST (control group). Patients treated with steroids presented higher Simplified Acute Physiology Score II and maximum Sepsis-Related Organ Failure Assessment scores. Both groups presented similar baseline and stimulated cortisol values. The hazard ratio of remaining on shock adjusted by severity of illness, inadequate antibiotic, and propensity score was 1.15 (95% confidence interval 0.71-1.86) for patients treated with steroids. Inhospital mortality was 62% in the steroid group and 52% in the control group (P = .84). Logistic regression analysis with propensity score neither showed differences between steroid and control group in the inhospital mortality. Predictors of inhospital mortality were age, maximum Sepsis-Related Organ Failure Assessment score, and inadequate antibiotics.

CONCLUSION

In our study, treatment with low-dose steroid therapy was not associated to a reduction in time to shock reversal or mortality.