Linda G. Martin,     Pullman, Washington

Pathophysiology and Causes

The pathogenesis of CIRCI in dogs and cats is unknown but it is most likely multifactorial, involving complex interactions between endocrine and immune systems. Possible mechanisms for the development of CIRCI in humans and animals include the following:

The ABCB1 gene mutation that results in lack of P-glycoprotein (Pgp) at the blood-brain barrier may also be a contributing factor in some dog breeds (e.g., collies, Shetland sheepdogs, Australian shepherds, Old English sheepdogs, English shepherds, German shepherds, long-haired whippets, silken windhounds). Pgp restricts the entry of cortisol into the brain, limiting cortisol’s feedback inhibition of CRH and ACTH. In ABCB1 mutant dogs, Pgp is not present, allowing greater concentrations of cortisol within the brain. There is greater feedback inhibition of the HPA axis and, ultimately, inhibition of sufficient cortisol secretion, potentially leading to the inability to respond appropriately to critical illness and stress.

Clinical Signs

Clinical signs of CIRCI can be vague and nonspecific, such as depression, weakness, fever, vomiting, diarrhea, and abdominal pain. In addition, clinical signs that are secondary to the underlying disease process responsible for CIRCI (e.g., septic shock, hepatic disease, trauma) can mask the clinical features of CIRCI. The most common clinical abnormality associated with CIRCI in critically ill human patients is hypotension refractory to fluid resuscitation, requiring vasopressor therapy. Hyponatremia and hyperkalemia, abnormalities consistent with aldosterone deficiency, are uncommon in humans with CIRCI and, to date, have not been reported in canine or feline critically ill patients with insufficient adrenal or pituitary function. Laboratory assessment of critically ill human patients with CIRCI may demonstrate eosinophilia or hypoglycemia or both, but these abnormalities are not consistently found in all humans with CIRCI. Eosinophilia and hypoglycemia have not been reported in critically ill veterinary patients with CIRCI.

Diagnosis

CIRCI should be considered as a differential diagnosis in all critically ill patients requiring vasopressor support. At the present time, there is no consensus regarding the identification of patients with CIRCI in human or veterinary medicine, and normal reference ranges do not exist for basal and ACTH-stimulated cortisol concentrations in critically ill dogs and cats.

Various tests have been recommended for diagnosing CIRCI, including random serum or plasma (total) basal cortisol concentration, serum free cortisol concentration, ACTH-stimulated cortisol concentration, delta cortisol concentration (the difference when subtracting basal from ACTH-stimulated cortisol concentration), the cortisol-to-endogenous ACTH ratio, and combinations of these methods. The optimal way to identify critically ill veterinary patients with CIRCI has yet to be determined.

Evaluation of adrenal function in veterinary patients typically involves administration of an ACTH stimulation test. The most commonly used protocol for ACTH stimulation testing in dogs involves intravenous administration of 5 µg/kg of cosyntropin, up to a maximum of 250 µg. In cats, intravenous administration of 125 µg/cat of cosyntropin is commonly used. Serum or plasma is obtained for cortisol analysis before and 60 minutes after ACTH administration for both dogs and cats. The standard doses of cosyntropin (5 µg/kg in dogs and 125 µg/cat) currently used are greater than the doses necessary to produce maximal adrenocortical stimulation in healthy small animals. Doses of 0.5 µg/kg in healthy dogs (Martin et al, 2007) and 5 µg/kg in healthy cats (DeClue et al, 2011) have been shown to induce maximal adrenocortical cortisol secretion. The use of higher doses is considered supraphysiologic and may hinder the identification of dogs and cats with CIRCI. Low-dose (0.5 µg/kg IV) ACTH stimulation testing has been compared with standard-dose (5 µg/kg IV) testing in critically ill dogs (Martin et al, 2010). Every critically ill dog that was identified to have insufficient adrenal function (i.e., ACTH-stimulated serum cortisol concentration below the reference range or <5% greater than the basal cortisol concentration) by the standard-dose ACTH stimulation test was also identified by the low-dose test. Additional dogs with adrenal insufficiency were identified by the low-dose ACTH stimulation test but not by the standard-dose test. ACTH administered at a dose of 0.5 µg/kg IV appears to be at least as accurate in determining adrenal function in critically ill dogs as the standard dose. The low-dose ACTH stimulation test may be a more sensitive diagnostic test in detecting patients with insufficient adrenal gland function than the standard-dose test.

Assays that measure cortisol concentration typically measure total hormone concentration (i.e., serum free or unbound cortisol plus a protein-bound fraction). However, the serum free cortisol fraction is thought to be responsible for the physiologic function of the hormone. Serum free cortisol concentrations may be a more precise predictor of adrenal gland function.

The relationship between free and total cortisol varies with serum protein concentration. In critically ill human patients, cortisol-binding globulin and albumin concentrations can decrease by approximately 50% because of catabolism at the inflammatory sites and inhibition of hepatic synthesis via cytokine induction. Serum total cortisol concentration may be falsely low in hypoproteinemic patients, resulting in overestimation of CIRCI. Serum free cortisol concentration is less likely to be altered in states of hypoproteinemia. Consequently, serum total cortisol concentrations may not accurately represent the biologic activity of serum free cortisol during critical illness. Several human studies suggest that serum free cortisol concentrations are a more accurate measure of circulating glucocorticoid activity than total cortisol concentrations. At this time, canine and feline studies are sparse, and the ability to measure serum free cortisol concentration is not widely available. However, serum free and total cortisol concentrations were compared in a group of 35 critically ill dogs having one of the following diseases: sepsis, severe trauma, or gastric dilation-volvulus (Martin et al, 2010). Fewer critically ill dogs with adrenal insufficiency (i.e., an ACTH-stimulated serum cortisol concentration below the reference range or <5% greater than the basal cortisol concentration) were identified by serum free cortisol concentration than by serum total cortisol concentration. However, basal and ACTH-stimulated serum total cortisol concentrations were not lower in hypoproteinemic dogs compared with normoproteinemic dogs. The significance of this finding is unknown, and further investigation is warranted in veterinary patients.

The delta cortisol concentration has been advocated as a method to identify critically ill patients with CIRCI in both human and veterinary medicine. A study in human patients with septic shock found that basal cortisol concentrations of 34 µg/dl (938 nmol/L) or less combined with delta cortisol concentrations of 9 µg/dl (250 nmol/L) or more in response to an IV 250 µg/person ACTH stimulation test were associated with a favorable prognosis. In addition, basal cortisol concentrations greater than 34 µg/dl combined with delta cortisol concentrations less than 9 µg/dl were associated with a poor prognosis. Because this protocol was successful in predicting outcome, a delta cortisol concentration less than 9 µg/dl is frequently used as the diagnostic criterion for CIRCI in critically ill human patients.

Veterinary studies have also assessed delta cortisol concentration as a criterion for diagnosing CIRCI in critically ill patients. One study found that septic dogs with delta cortisol concentrations of 3 µg/dl (83 nmol/L) or less after an IM 250 µg/dog ACTH stimulation test were more likely to have systemic hypotension and decreased survival (Burkitt et al, 2007). In addition, another study investigating acutely ill dogs (i.e., dogs with sepsis, severe trauma, or gastric dilation-volvulus) found that dogs with delta cortisol concentrations of 3 µg/dl or less after an IV 5 µg/kg ACTH stimulation test were more likely to require vasopressor therapy as part of their treatment plan (Martin et al, 2008). Sensitivity of delta cortisol concentrations of 3 µg/dl or less in the diagnosis of critically ill veterinary patients with CIRCI has yet to be determined.

A further confounding factor in interpreting pituitary-adrenal function tests of any kind is that parameters can change over time. Test results obtained on a single day may not reflect the findings on previous or future days. The relationship between abnormal parameters can also change over time. For example, dogs that died from parvoviral diarrhea had a lower delta cortisol concentration than dogs that survived on day 1 of hospitalization but not on day 3 (Schoeman and Herrtage, 2008).

Based on the current veterinary literature, three scenarios may indicate the presence of CIRCI in critically ill dogs (especially in the presence of refractory hypotension): (1) dogs with a normal or an elevated basal cortisol concentration and an ACTH-stimulated cortisol concentration less than the normal reference range, (2) dogs with a normal or an elevated basal cortisol concentration and an ACTH-stimulated cortisol concentration that is less than 5% greater than the basal cortisol concentration (flatline response), or (3) dogs with a delta cortisol concentration of 3 µg/dl or less (≤83 nmol/L). Based on a few clinical studies and case reports, CIRCI also appears to occur in cats. However, there is no consensus regarding the diagnostic criteria in cats at this time.

At the present time, it is recommended that ACTH stimulation testing not be used to identify human patients with septic shock who should receive supplemental corticosteroid therapy. This recommendation is based on the lack of compelling evidence demonstrating that a patient’s response to ACTH administration predicts the benefit from corticosteroid therapy. Similarly, no studies in veterinary medicine have investigated the usefulness of ACTH stimulation (or other diagnostic) testing for identifying patients that would benefit from supplemental corticosteroid therapy. ACTH stimulation testing may still prove useful in veterinary patients if only to gain more information on the syndrome of CIRCI in animals and to help identify animals that should receive supplemental corticosteroid treatment. The results of ACTH stimulation testing also can be used to decide if corticosteroid therapy should be stopped or continued (see next section on Treatment).