Blood pressure and its associations in 554 children and young people with congenital adrenal hyperplasia

Abstract

Background: Congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (21OHD) affects approximately 1 in 15 000 individuals. We leveraged the power of multicentre registry data to assess the trend and predictors of blood pressure (BP) within children and young persons with 21OHD to inform monitoring strategies.

Method: Data from the International CAH Registry in patients younger than 20 years was compared to normative values. Values of BP were modeled to create reference curves, multiple change point analysis applied to quantify the difference with normative data. Covariate adjustment was informed by a directed acyclic graph, prior to joint outcome regression modeling to accurately assess predictors of BP.

Results: A total of 6436 visits within 554 patients (52.5% females) showed BP-Standard deviation scores (SDS) were higher at younger ages. Patients under five years had systolic BP-SDS of 1.6 (Q1:0.6-Q3:2.7) decreasing to 1.0 (Q1:0.2-Q3:1.8) over 5 years, equating to 31.0% over the 95th centile decreasing to 15.0%. Higher doses of fludrocortisone were associated with a small increase in systolic BP equivalent to 1.2 mmHg with every 100 µg extra fludrocortisone. Renin of 100µU/mL was associated with 4.6 mmHg lower systolic BP than a renin of 1µU/mL, higher 17OH-progesterone and androstenedione also predicted lower systolic and diastolic BP (P < .05).

Conclusion: Higher BP in children with 21OHD is common and particularly pronounced at a younger age, but may not be attributable to excessive mineralocorticoid replacement. There is a need to improve our understanding of the determinants of this raised BP as well as its long-term effects.

Keywords: Bayesian analysis; blood pressure; congenital adrenal hyperplasia; glucocorticoids; statistical modeling.