 |
|
Viewpoint Indian Pediatrics 2002; 39:1011-1016 |
||
|
Adrenal Insufficiency of Critical Illness |
||
|
Cortisol is the major glucocorticoid produced by the adrenal cortex. It is required for normal function in all cells in the body. Deficiency of cortisol is associated with increased morbidity and mortality during critical illness. We had known about increased requirements of corticosteroids during stress of illness in children with previously diagnosed adrenal insufficiency, and have routinely increased replacement doses in these cases. The practice is based on the observation that plasma cortisol and ACTH concentration are increased in stressful conditions such as fever, trauma, surgery and pain. However, the observations that some critically ill patients may have low plasma cortisol and/or ACTH concentration led to a resurgence of interest in adrenal insufficiency in intensive care setting(1). Several studies and reviews in the recent past have focussed on adrenal insufficiency in critically ill adults(2,3). However, literature on the subject in critically ill infants(4) and children is scanty(5). In this review, I have focussed on the current issues in adrenal insufficiency of critical illness that are applicable to pediatric patients. I have sought to answer the following:
Very little data is available on incidence of adrenal insufficiency in critically ill children. Studies in adult patients suggest that overall incidence of adrenal insufficiency in high risk critically ill patients such as those with sepsis, hypotension, shock and post-operative state is high and increases with severity of illness(3). Most studies in critically ill adults report an incidence of 25-40%(3). The incidence in critically ill children is close to it; 52% in septic shock patients(5) and 23% in a recent study from Canada(6). Theoretically adrenal insufficiency in critically ill patients could be due to direct damage to adrenal glands because of infection, inflammation, exhausted HPA axis, exhausted adrenal cortex with no cortisol reserve, impaired capacity to produce cortisol(7) and possibly an immature HPA axis in children. There is evidence that each of these mechanisms may be operating. Systemic inflammatory states such as sepsis are associated with both primary and secondary adrenal insufficiency(8) that is reversible with treatment of inflammation. It is believed to result from HPA suppression by cytokines and other inflammatory mediators(9). The immune and endocrine system share common mediators. TNF-alpha is potent inducer of ACTH secretion(10), but it impairs CRH stimulated ACTH release(11). TNF-alpha inhibits adrenal function(7, 11) through direct effects on the adrenal gland and has been shown to reduce adrenal cortisol synthesis by inhibiting the stimulatory actions of ACTH and angiotensin II on adrenal cells(7). A number of clinical studies have reported inappropriately low ACTH levels in patients with sepsis and critical illness(8). A decreased response to corticotropin has been reported in patients with septic shock compared with response in the patients with recovery, despite similar basal cortisol levels(9,12). Critical illness may also impair cortisol action at cellular level. In patients with sepsis, number and affinity of cellular glucocorticoid receptors is decreased(13). Endotoxins cause a rapid decline in glucose cortisol binding capacity in liver and decrease the maximum binding capacity of the receptor(14,15). It appears that tissues of some patients with septic shock have an impaired ability to metabolize cortisol(16). Whether these effects are secondary to tissue injury or etiologic in causing poor outcome is not clear. Adrenal insufficiency affects all organs of body, therefore most of the signs and symptoms are non-specific. It is clinically not easy to differentiate between a critically ill child with adrenal insufficiency and one with normal adrenal function. However, some patients are more prone to acute adrenal insufficiency such as those with coagulopathy, thromboembolic disease, traumatic shock, septis etc. It is perhaps due to direct damage to adrenal cortex. These patients will have signs of acute adrenal failure. All very ill patients with sudden deterioration should therefore be screened for adrenal insufficiency. More difficult is to recognize relative adrenal insufficiency. These patients may have hypotension that is refractory to fluids and requires vasopressors(5,17) and responds to cortico-steroids(18). Adrenal insufficiency should be suspected in patients with increasing requirement of catecholamines, unexplained hyponatremia, hyperkalemia, hypoglycemia or eosinophilia(19,20). Adrenal insufficiency in critically ill children could be either primary (because of primary adrenal dysfunction) or secondary because of hypothalamic or pituitary dysfunction. It is likely that dysfunction of HPA axis i.e. secondary adrenal insufficiency may not be as serious a problem as the primary adrenal insufficiency. Three patients in the series by Menon and Clarson(6) probably had secondary adrenal insufficiency. They had a serum cortisol less than 7 µg/dL and ACTH £ 17 pg/mL, and the cortisol increased on ACTH stimulation test. All of them survived. On the other hand, one patient had primary adrenal insufficiency; had low normal cortisol and failed to respond to ACTH stimulation. This patient died of refractory shock and needed fluid and inotropes. However, patients with adrenal insufficiency (defined as inadequate cortisol response to ACTH) in the series by Hetherill et al.(15) had average daily Therapeutic Intervention Scale Score (TISS) and multiorgan failure (MOF) scores similar to those with normal adrenal functions. The mortality was also similar in two groups. Nonetheless, patients with adrenal insufficiency required adrenaline or nor-adrenaline more often and in a greater vasopressor doses(5). Also lower basal plasma cortisol concentration were found in non-survivors compared with survivors of severe sepsis in critically ill adults(9,21). However, there was no difference in basal cortisol or increment or peak concentration in response to ACTH stimulation in survivor and non-survivors(5,9). What is the real indicator of Adrenal insufficiency? Is it low cortisol which shows what is functionally available or a low response to ACTH stimulation which shows low cortisol reserve? Perhaps the best marker would be low free (unbound) cortisol that is physiologically active and homeostatically regulated. In critically ill patients there is a decrease in cortisol binding, which may increase cortisol bioavailability to tissues. However, there is no clinically useful test to assess cellular actions (end organ effects) of cortisol during illness. Moreover, no consensus exist as to what constitutes the lower limit of cortisol (should it be <10 µg/dL) and what should be interpreted as an appropriate cortisol response to ACTH stimulation? Cortisol values <7 µg/dL, <10 µg/dL, <20 µg/dL all have been proposed as cut off for defining adrenal insufficiency. A combination of both as used by Menon and Clarson(6) may be a better option as it takes in consideration "the present" and "what is in reserve for tomorrow". A word of caution is needed about interpretation of ACTH stimulation test. Best stimuli for testing adrenal function are endogenous stresses; pain, surgery, tissue injury, and other diseases stimulate cortisol release. Endogenous stresses such as hypotension, hypoxemia, and hypoglycemia are superior stimuli for cortisol secretion compared with ACTH testing. These endogenous stresses test the integrity of the entire brain HPA axis. Exogenous ACTH bypasses the brain-hypothalamic-pituitary axis and tests the integrity of the adrenal gland directly. Failure of the adrenal gland to secrete adequate cortisol in response to ACTH is considered diagnostic of adrenal insufficiency with a high positive predictive power. However, a normal response to ACTH does not rule out adrenal insufficiency(3). Patients with partial adrenal insufficiency and acute adrenal insufficiency can have normal responses to exogenous ACTH(22,23). It can take 10-14 days for the adrenal gland to atrophy enough so that responses are subnormal to ACTH.
ACTH stimulation test can aid in the evaluation of adrenal function. The traditional test uses 250 µg of synthetic corticotropin intravenously. A subnormal response (indicative of adrenal insufficiency) is a cortisol level less than 25 to 20 µg/dL. A normal response to 250 µg corticotropin (high dose) does not rule out adrenal insufficiency as the stimulus used is over one hundred fold greater than normal stress-induced ACTH levels(3,22,23). Such high supraphysiologic dose can override adrenal resistance to ACTH and result in a normal cortisol response in patients with acute secondary adrenal insufficiency potentially serious conse-quence(23,24). The rise in serum cortisol following ACTH is an indictor of adrenal reserve. Some patients may have an appropriately high serum cortisol concentration but be unable to respond further after corticotropin injection (no reserve). Some authors therefore suggest that the increase in cortisol after administration of corticotropin should not be used as a criterion for the diagnosis of adrenal insufficiency(3). Many clinicians use the more physiologic low-dose (1 to 2 µg) corticotropin stimulation test, which better approximates ACTH levels found in severe stress (25,26). Most critically ill patients with stress-related cortisol levels <15 µg/dL the cortisol levels do not increase higher than 20 µg/dL. It appears that low dose corticotropin stimulation test is more sensitive than the high-dose test; most patients with adrenal insufficiency diagnosed by the low-dose corticotropin test improve clinically with hydrocortisone treatment(3). Low plasma cortisol levels in the face of high corticotropoin levels (i.e. > 100 pg/mL or 22 pmol/L) suggest primary adrenal insufficiency. Normal or low corticotropin levels in the presence of low cortisol levels suggest secondary adrenal insufficiency. It is not necessary to obtain cortisol levels at a specific time of the day because most critically ill patients lose the diurnal variation in their cortisol levels. The level should be obtained during the stress such as hypo-tension, hypoxemia, hypoglycemia(3). Once adrenal insufficiency is suggested by serum cortisol levels, it is essential to evaluate the cause of the adrenal insufficiency and consider hydrocortisone therapy. Patients with cortisol levels less than 10 µg/dL or failure to increase cortisol levels above 25 µg/dL on corticcotropin (1 to 2 µg) stimulation test indicates adrenal insufficiency that should be treated with hydrocortisone. Beneficial effect of treatment with glucocorticoids in stress doses (200-300 µg/day) has recently been reported in adults. It resulted in reduced mortality, improved shock reversal, and decreased requirement of vasopressor support(20,27,28). Catecholamines dependent hyperdynamic shock responds well to glucocorticoids(18). Evidence for benefit from supplemental glucocorticoids treatment in critically ill children is awaited. However, it would be reasonable to treat such patients with hydrocortisone stress doses (5 mg/kg/day) based on adult studies till such time that relevant data becomes available. To end, diagnosis and treatment of adrenal insufficiency in critically ill children presents a significant challenge. There is sufficient literature support to suggest that adrenal insufficiency should be looked for in all critically ill children and appropriate treatment considered if serum cortisol is low and further does not increase after ACTH (1 µg) stimulation test. It is time that trials to evaluate role of replacement therapy and follow-up studies of adrenal functions after recovery from critical illness in children are undertaken. Contributors: SS conceived the paper, did the literature search, wrote the draft and final manuscript of the paper. He stands as the guarantor to the paper. Funding: None Competing interests: None stated.
| ||
|
References | ||
|
