The effect of patient-managed stress dosing on electrolytes and blood pressure in acute illness in children with adrenal insufficiency

Background Adrenal crises (AC) are acute episodes of adrenal insufficiency (AI). Manifestations include hypotension and electrolyte disturbances. Glucocorticoid stress dosing (SD) can prevent AC progression, but its effect on physiological parameters has not been assessed in a 'real world setting'. Aims To assess the effect of prior self-managed glucocorticoid dose escalation on physiological markers in children with congenital adrenal hyperplasia (CAH) presenting to hospital for an acute illness. Methods An audit of records of all children with CAH presenting to paediatric referral hospital between 2000 and 2015. Potassium, sodium and glucose levels, and hypotension were compared between children who had and had not used SD. Results There were 321 attendances by patients with CAH and an acute illness during the study period. Any form of SD was used by 64.2% (n = 206); intramuscular (IM) hydrocortisone was used by 22.1% (n = 71) and oral only by 41.7% (n = 134). Use of SD (oral and/or IM) was associated with a significantly lower mean potassium level (4.02 +/- 0.71 vs. 4.27 +/- 0.79 mmol/l, P < .05). Linear regression analysis showed that age (beta: -0.04 years (95% CI -0.06, -0.02)), diarrhoea (beta: -0.41 (95% CI -0.06, -0.02)) and any form of stress dosing (oral, IM or both) (beta: -0.29 (95% CI -0.55, -0.04)) were each independently and significantly associated with potassium levels. SD was not significantly associated with sodium or glucose concentrations or with estimates of hypotension. Conclusion Patient-initiated SD resulted in a significant reduction in hyperkalaemia and lowered mean potassium levels in paediatric patients with CAH but did not alter significantly sodium and glucose concentrations or incidences of hypotension.