Insuficiencia poscorticosteroides: búsqueda de un valor umbral de cortisol a las 8 horas, estudio prospectivo (SYNASTOPRO)

Synthetic glucocorticoids are widely used in medicine for their anti-inflammatory and immunomodulating effects. There are 1-2% of the population treated with synthetic glucocorticoids; among them, around 0.5% underwent prolonged treatment. Glucocorticoids act to reduce inflammation, to modulate immunity but also, to regulate physiological functions such as glucose or bone homeostasis. This is why, despite great efficiency, glucocorticoids are associated with numerous side effects. One of the most frequent is glucocorticoid-induced adrenal insufficiency (GI-AI), which is caused by exogenous glucocorticoid negative feedback on the pituitary and the hypothalamus that leads to adrenal insufficiency. Within 10 weeks after glucocorticoid treatment withdrawal, more than 40% of the patients displayed biological glucocorticoid insufficiency. It is important to diagnose GI-AI to prevent life-threatening adrenal crisis. Diagnosis of GI-AI is biological. In pediatrics, the Low Dose Short Synacthen Test (LD-SST) is usually used. LD-SST consists of injection of a low dose of Synacthen (1µg) to stimulate the adrenal glands and evaluate their capacity to secrete a cortisol peak superior to 500 nmol/L. Plasma cortisol elevation is measured at T20, T30, and T60 and compared to an unstimulated value at T0. As LD-SST requires an outpatient hospital and multiple blood samples, less than 5% of the patients underwent adrenal testing after prolonged glucocorticoid treatment.

In a retrospective study, the investigators showed that morning plasmatic cortisol value can be used to predict LD-SST outcomes, facilitating GI-AI diagnosis. Based on 91 pediatric patients who underwent LD-SST after prolonged glucocorticoid therapy, the investigators showed that a morning plasma cortisol value superior to 317 nmol/L could rule out the diagnosis of GI-AI with a sensitivity of 95% whereas a value inferior to 144 nmol/L is associated with GI-AI with a specificity of 94%. However, our study had several limits, such as selection bias and population representativity, and those thresholds needed to be confirmed with a larger prospective study.

The main objective of our study is to confirm thresholds of morning plasma cortisol value to diagnose glucocorticoid-induced adrenal insufficiency in a larger population, more representative of the variety of glucocorticoid treatment in pediatrics. Secondary objectives are the evaluation of salivary cortisol values either to replace plasma cortisol value in response to Synacthen or as a single morning cortisol value. The tertiary objective is to determine GI-AI prevalence in children treated with glucocorticoids for more than 3 weeks and to evaluate clinical predictive factors.

It is a prospective bicentric study. The investigator will include patients aged from 6 months to 18 years old who underwent glucocorticoid treatment for at least 3 weeks at a dose superior to or equal to 3 mg/m2/day. As asthma is the most common pediatric pathology to require prolonged glucocorticoid treatment, patients treated with inhaled glucocorticoids for more than 12 months, or more than 6 months if associated with topic glucocorticoids or if the dose is considered high (GINA 2022) will also be included.

Four weeks after treatment discontinuation or tapering, an LD-SST will be performed to diagnose GI-AI. It will be associated with salivary cortisol measurement at T0 and T30. The investigators will evaluate the association between T0 plasma or salivary cortisol values and GI-AI (plasma cortisol < 500 nmol/L after LD-SST). Thresholds of morning plasma or salivary cortisol associated with the presence or absence of GI-AI will be determined using a receiver operating characteristic (ROC) curve. The risk factors of GI-AI will be assessed through linear regression.

The investigators believe that using morning plasma or salivary cortisol values to screen patients after glucocorticoid withdrawal should improve systematic GI-AI screening and, thus, prevention of adrenal crisis.