Adult height in patients with congenital adrenal hyperplasia: A systematic review and metaanalysis

Context: Treatment for patients with congenital adrenal hyperplasia (CAH) may affect the final height of these patients. Objective: Our objective was to determine the distribution of achieved height in patients with classic CAH diagnosed at infancy or early childhood and treated with glucocorticoids. Data Sources: We searched MEDLINE, EMBASE, Cochrane Library, ISI Web of Science, and Scopus through September 2008; the reference sections of included studies; and expert files. Study Selection: Eligible studies included patients diagnosed with CAH before age 5 and followed to final height. Data Extraction: Reviewers working in duplicate independently extracted data on study characteristics and outcomes and determined each study's risk of bias. Data Synthesis: The SD score (SDS) for final height and corrected height (defined as final height SDS - midparental height SDS) were estimated from each study and pooled using random-effects metaanalysis. The I2 statistic was used to assess inconsistency in results across studies. Results: We found 35 eligible studies, most of which were retrospective single-cohort studies. The final height SDS achieved by CAH patients was -1.38 (-1.56 to -1.20; I² - 90.2%), and the corrected height SDS was -1.03 (-1.20 to -0.86; I² = 63.1%). This was not significantly associated with age at diagnosis, gender, type and dose of steroid, and age of onset of puberty. Mineralocorticoid users had a better height outcome in comparison with the nonusers (P = 0.02). Conclusion: Evidence derived from observational studies suggests that the final height of CAH patients treated with glucocorticoids is lower than the populationnormand is lower than expected given parental height.