Characteristics of Patients with Classic Congenital Adrenal Hyperplasia Missed on the Newborn Screen

Newborn screening for classic congenital adrenal hyperplasia (CAH) has been performed in all 50 states in the USA since 2009 [1] and was implemented even earlier in Indiana. This has allowed for earlier detection and treatment of patients with classic CAH and is particularly critical for boys who have a normal physical examination at birth. Despite cut-off values for 17-hydroxyprogesterone (17OHP) on the newborn screen (NBS) that have high sensitivity, it is known that some cases of classic CAH are missed [2]. In this study, our aim was to identify patients with classic CAH followed at our institution who were missed on the NBS in Indiana and determine if they had clinical characteristics that might have prevented a delayed diagnosis.

Medical records of children diagnosed with classic CAH due to 21-hydroxylase deficiency seen at Riley Hospital for Children in Indiana between January 2005 and December 2020 were reviewed. Variables analyzed included patient characteristics, birth history, visit information, and laboratory studies. Patients with non-classic CAH were excluded. Statistical analysis was performed using SPSS version 28. A p value of ≤0.05 was considered statistically significant. This study was approved by the Institutional Review Board at Indiana University Health.

A total of 64 patients with classic CAH were identified. Twelve patients (19%) were missed on the NBS and 52 patients were diagnosed on NBS. The mean age at presentation for the missed patients was 21.7 months (range 3–75 months) compared to 0.27 months (±0.31 months) in those who were detected by the NBS, p = 0.004. More than half of the patients who were missed on the NBS were girls (67%) compared to 48.1% of those detected by the NBS (p = 0.24). None of the patients in either group was exposed to antenatal steroids. One patient who was missed on the NBS had respiratory distress after birth and required a NICU stay but did not receive steroids. There was no statistically significant difference between the two groups with regard to sex, ethnicity, birth weight, mode of delivery, birth complications, or time of the NBS collection (Table 1). The results remained not statistically significant after controlling for birth weight (p = 0.63) and gestational age (p = 0.33).

A family history of CAH was present in 42% of patients who were missed on the NBS compared to 19% in those detected by the NBS (p = 0.10). Of the 12 patients missed, 8 (7/8 of the girls) presented with pubic hair or clitoromegaly, 5 of whom also had posterior labial fusion. One boy presented with pubic hair and testicular enlargement. Several had linear growth acceleration and an advanced bone age. The parents of 4 patients requested testing because of a family history of CAH. Of the entire cohort, 6 (50%) had confirmatory genetic testing and all variants identified have been reported to be associated with classic CAH except one. Patient #2, who had a positive family history, was found to have one variant that has been associated with classic CAH and one variant associated with the non-classic form. Based on her elevated renin level, her phenotype was felt to be consistent with classic CAH. Renin levels were elevated in 90% of the patients in whom they were measured, confirming some degree of salt wasting. Individual patient characteristics are seen in Table 2.

In Indiana, 17OHP levels on the NBS are measured using the DELFIA assay produced by PerkinElmer with different cut offs adjusting for birth weight. The 17OHP cut-off values on the NBS changed throughout the study period and a new platform was adopted for the test in 2012. Of our total cohort, 27 patients were born prior to the change in platform whereas 37 were born after this change. There was no difference in the number of patients missed on the NBS on the first compared to the second platform (18.5 vs. 18.9%). Cut-off values for 17OHP and testing methodologies used in Indiana throughout the study period are shown in Table 3. Mean 17OHP concentrations were not different in those who were missed (14,948 ng/dL vs. 16,701 ng/dL, p = 0.74). However, the mean testosterone level at diagnosis was lower in patients who were missed compared to those who were detected on the NBS (68.0 ng/dL vs. 196.16 ng/dL, p = 0.028). Mean values of androstenedione, DHEAS, and cortisol levels were not different between the two groups (Table 4). Individual hormone levels obtained in the missed patients are presented in Table 5.

The primary goal of the NBS in CAH is to identify affected boys before they experience life-threatening salt-wasting crises. In fact, it has been established that unrecognized CAH accounted for some apparent sudden infant deaths prior to the newborn screening era [3‒5]. Given this goal, 17OHP cut offs have been lowered over the years in an attempt to minimize false-negative results. Despite the success of newborn screening programs for CAH, challenges include that many states have adopted different assays for 17OHP that utilize disparate cut-off values for normal versus abnormal results [6, 7]. Factors associated with false-negative results include suboptimal collection time, immature adrenal function, exposure to pre- or postnatal steroids, and very mild forms of CAH [1].

In order to not miss patients with CAH, 13/50 states have adopted a second routine NBS that is generally obtained at 8–14 days of life [8]. A study in Texas found that 14% of patients diagnosed with classic CAH (of whom 55% were girls with the simple virilizing form) were detected on the second NBS rather than the first [9]. Another study in Colorado found that the first NBS missed nearly 30% of patients with classic CAH (27% of whom were girls) [10]. Both studies concluded that the patients detected on the second NBS likely had a milder form of CAH, as evidenced by a lower mean 17OHP levels on the second NBS in the Colorado study. This contrasts with our study where both groups had similar 17OHP levels at diagnosis, indicating that the missed patients did not necessarily have a milder form of CAH than those detected at birth. Another study from Minnesota found that the NBS missed 22% of patients with both simple virilizing and salt-wasting forms of CAH [2]. Interestingly, 9/15 patients that were missed were females and 3/9 of them were diagnosed late at 3 months, 3.4 years, and 6.5 years of age despite having ambiguous genitalia. A larger study comparing the NBS for CAH from 7 states (2 one-screen states and 5 two-screen states) showed that the first NBS had similar detection rates for salt-wasting CAH, but the second NBS detected 36% of all CAH cases of which only 6.5% had salt-wasting CAH [8]. A more recent study reviewed CAH cases from the Northwest Regional Newborn Screening Program over a 15-year period and showed that 25% of CAH cases were detected on the second NBS of which one-third were classified as salt-wasting CAH [11].

While the percentage of patients missed in our study (19%) are in line with these other reports, we found that more had evidence of salt-wasting CAH (75%), assuming that they would have been detected had a second screen been obtained. Renin levels were used as a reflection of the patients’ salt-wasting status as renin has historically served as an indicator of sodium homeostasis [12]. However, it is important to note that the designation of salt-wasting versus simple virilizing has fallen out of favor and is not considered clinically meaningful since all patients with classic CAH have some degree of mineralocorticoid deficiency [12, 13]. Although patient #2 had one genetic mutation for non-classic CAH, many studies have shown that genotype-phenotype correlations are inconsistent, and the clinical phenotype may be more severe than the genotype, especially in cases with genetic mutations for simple virilizing CAH [14, 15]. It is remarkable that 5 of our girls (#7, #8, #9, #11, and #12) who were missed on the NBS had evidence of significant prenatal hyperandrogenism that was not detected on physical examination for anywhere from 3-27 months. Another unusual finding in our cohort of missed patients is the presence of a family history of CAH in almost half, a finding which, to the best of our knowledge, has not been reported in previous studies. The higher testosterone levels in the patients detected as neonates are likely the result of the mini-puberty that occurs at birth [16]. We also compared testosterone levels in girls only which revealed lower values in those who were missed on the NBS compared to those detected at birth (45 ng/dL ± 23.9 vs. 296 ng/dL ± 230.4, p = 0.007). However, the 17OHP levels were not significantly different (20,995 ng/dL ± 14,265 vs. 21,189 ng/dL ± 27,523, p = 0.48).

Our study has several limitations primarily due to its retrospective nature, which may introduce missing or inaccurate data. For instance, some patients’ medical records were incomplete, and not all individuals underwent genetic testing. It is still possible that some of our patients who were missed had a milder form of CAH than those who were detected on the NBS despite the elevated renin levels and lack of difference in mean 17OHP levels. Nonetheless, our study’s findings are significant, as they underscore the importance of recognizing that some patients with classic CAH can be missed on the NBS. This highlights the wide clinical spectrum of CAH and the need for a high index of suspicion and ongoing vigilance to ensure early detection and appropriate management, potentially preventing adverse outcomes in affected individuals.

About a fifth of our patients with classic CAH were missed on the NBS. No specific features were identified that distinguished these children from those who were detected at birth. The reason for the delayed rise in 17OHP in these children is unclear. These results emphasize the importance of maintaining a high index of suspicion for CAH, particularly in the setting of clinical evidence of androgen exposure and positive family history even when the NBS is normal.

The study protocol was reviewed and approved by the Institutional Review Board at Indiana University Health (IRB: #13631) and was exempt from requiring written informed consent for the retrospective analysis of the data.

The authors have no conflicts of interest to declare.

Genetic testing of some of the participants in this study was funded by the CARES Foundation.

R.S. performed the data collection, statistics, and writing of the manuscript. T.D.N. reviewed the manuscript and suggested edits. E.A.E. supervised the writing of the full manuscript and suggested edits.

The data that support the findings of this study are not publicly available due to containing information that could compromise the privacy of research participants but are available from the corresponding author (R.S.) upon reasonable request.