Congenital adrenal hyperplasia in Denmark: a nationwide study of neonatal screening, clinical presentation and mental health

Congenital adrenal hyperplasia (CAH) is a group of diseases caused by rare enzymatic deficiencies in the cortex of the adrenal gland. In Denmark, the prevalence of CAH is 12 per 100,000 newborns, and its average incidence is 1.55 per 1,000,000. In 95% of cases, CAH is due to 21-hydroxylase deficiency (21-OHD), caused by pathogenic variants in the CYP21A2 gene. CAH symptoms are caused by impaired synthesis of aldosterone, cortisol and a concomitantly increased androgen production. Since 2009, CAH has been included in the neonatal screening program, aiming to detect the most severe salt-wasting (SW) phenotype of CAH. SW CAH and simple virilising (SV) CAH are defined as classic CAH. Non-classic (NC) CAH represents the milder CAH phenotype.

The three retrospective studies of this thesis were based on epidemiological data from health registries, screening data, genetic tests and clinical data from medical records. We identified a national Danish cohort of 462 patients with CAH, distributed as follows: SW: 222, SV: 59, NC: 142, rare forms: 24 and unknown phenotype: 15, diagnosed between 1943 and 2018.

Study I evaluated the efficacy of the neonatal screening programme. Its sensitivity was 96% and its specificity was 100%. The PPV was 55% overall (61% in full-term- and 33% in premature newborns). Comparing the pre-screening period with the screening period, the incidence of SW CAH did not differ (5 per 100,000 newborns). However, after 2009, the femaleto-male ratio for SW CAH changed from 2.5 to 0.76, indicating missing male cases before 2009. The frequency of salt-wasting crises decreased (from 29% to 10%), and significantly more newborns were asymptomatic at the time of diagnosis (from 0% to 20%). Neonatal screening aims to identify the most severe forms of CAH, and today there is no national screening for milder forms of CAH. Therefore, CAH suspicion must rest on clinical manifestations, family investigation and/or molecular genetic testing leading to the diagnosis.

In Study II, the correlation between genotype and phenotype. Genotypes were grouped as Null, A, B and C, corresponding to an increasing 21-OH activity, and the expected phenotype; SW, SW, SV and NC, respectively. In patients diagnosed at age ≤18 years (n=379), the correlation between genotype and the expected phenotype was expressed as PPV: Null: 100%, A:95%, B:51% and C: 76%. In a subgroup of patients diagnosed with CAH between 1999 and 2018, ≤ 18 years of age (n=159), genotype groups were correlated with clinical data. We found that every female in genotype group Null and A was prenatally virilised, but prenatal virilisation was surprisingly also seen in milder genotype groups. This implicated that genotypes cannot reject possible virilisation in CAH females. Virilisation, both prenatally and postnatally, was found among 53% of females, but only prenatally virilised females underwent feminising surgery, corresponding to 85% of prenatally virilised females. Feminising surgery was performed at a median age of 0.90 (IQR 0.5-1.6) years. Signs of pseudo precocious puberty at the time of diagnosis were the most frequent symptom of CAH (39%). Bone age was significantly more advanced in males than in females (P=0.0009) at the time of diagnosis, emphasising a delayed diagnosis of males. Furthermore, the female-tomale ratio was 1.19 in SW, 1.85 in SV and 2.68 in NC CAH, thus underlining the non-diagnosis of CAH males. The virilisation among CAH females and the delayed diagnosis of the milder CAH phenotypes may have other long-term impacts on morbidity.

Study III, an epidemiological register-based study of psychiatric morbidity among patients with CAH compared with 100 randomly age- and sex-matched background population controls. We took advantage of the extensive Danish health registries holding data on hospital in- and out patients’ admissions and corresponding diagnosis codes, suicide, data on medicinal prescriptions related to psychiatric disorders. We found that approximately half of the patients with CAH had a minimum of one registration of a psychiatric diagnosis or related medication. A significantly increased risk of being registered with any psychiatric disorders was found in both CAH males (incidence rate ratio (IRR) 3.20 (95% confidence interval (CI) 1.58-6.50)) and females (IRR): 2.48 (95% CI 1.57-3.91)). The significantly increased risk also applied to medication prescription related to psychiatric disorders in both males (HR: 1.80 (95% CI 1.36-2.38)) and females (HR 1.70 95% CI 1.40-2.06)). The risk of psychiatric disorders due to psychoactive substance use was significantly increased in both CAH males and females, and this applied specifically to patients with SW CAH. Suicidal behaviour among CAH males was significantly increased (IRR: 10.66 (95% CI 3.74-30.42).

In conclusion, in Denmark, we have an efficient neonatal screening for severe forms of CAH, whereas milder forms suffer from late diagnosis and even non-diagnosis, especially in males. In severe genotypes, an almost complete correlation with the SW phenotype was found, while the intermediate and milder genotypes predict the phenotype with less confidence. CAH females with prenatal virilisation underwent feminising surgery in 85% of cases. The prevalence of psychiatric disorders among CAH patients is significant, warranting more attention to psychological and psychiatric support and care in patients with CAH.