Adrenal Hyperplasia, Congenital, Due To 21-Hydroxylase Deficiency

A number sign (#) is used with this entry because this form of congenital adrenal hyperplasia is caused by homozygous or compound heterozygous mutation in the CYP21A2 gene (613815), encoding steroid 21-hydroxylase, on chromosome 6p21.

Description

Congenital adrenal hyperplasia (CAH) results from a deficiency in one or another of the enzymes of cortisol biosynthesis. In about 95% of cases, 21-hydroxylation is impaired in the zona fasciculata of the adrenal cortex so that 17-hydroxyprogesterone (17-OHP) is not converted to 11-deoxycortisol. Because of defective cortisol synthesis, ACTH levels increase, resulting in overproduction and accumulation of cortisol precursors, particularly 17-OHP, proximal to the block. This causes excessive production of androgens, resulting in virilization.

Slominski et al. (1996) presented evidence that the CYP21A2, CYP11A1 (118485), CYP17 (609300), and ACTHR (202200) genes are expressed in skin (see 202200). The authors suggested that expression of these genes may play a role in skin physiology and pathology and that cutaneous proopiomelanocortin activity may be autoregulated by a feedback mechanism involving glucocorticoids synthesized locally.

Clinical Features

There are 4 recognized clinical forms of congenital adrenal hyperplasia, the majority of cases being associated with 21-hydroxylase deficiency: salt-wasting (SW), simple virilizing (SV), nonclassic (NC) late-onset (also called attenuated and acquired), and cryptic. All 4 forms are closely linked to HLA and represent the effects of various combinations of alleles.

In female newborns, the external genitalia are masculinized; gonads and internal genitalia are normal. Postnatally, untreated males as well as females may manifest rapid growth, penile or clitoral enlargement, precocious adrenarche, and ultimately early epiphyseal closure and short stature. A mild form of late-onset adrenal hyperplasia due to 21-hydroxylase deficiency can occur in adults and has hirsutism as the only manifestation in the most attenuated form.

All types of adrenal hyperplasia were reviewed exhaustively by Bongiovanni and Root (1963). Prader et al. (1962) reported an enormous interlocking Swiss kindred. (See precocious puberty (176400) for a simulating condition.)

Galal et al. (1968) concluded that the 2 clinical forms of 21-hydroxylase deficiency (with and without salt loss) correlate with the extent of the defect in the cortisol pathway. Some had suggested the existence of 2 different 21-hydroxylating systems, one specific for progesterone and concerned with aldosterone synthesis and the other specific for 17-alpha-hydroxyprogesterone involved in cortisol synthesis. However, Orta-Flores et al. (1976) presented evidence that there is only one 21-hydroxylation system with 2 active sites: one active on progesterone only and a second active on either substrate indiscriminately. The authors suggested that both sites are defective in the salt-losing variety and only the second in the non-salt-losing form.

Presentation with gynecomastia and bilateral testicular masses was reported by Kadair et al. (1977) in a case of 21-hydroxylase deficiency. Others have reported bilateral testicular tumors. Lewis et al. (1968) found that intelligence is increased in the adrenogenital syndrome, a remarkable and possibly significant feature from the point of view of selection and gene frequency. However, McGuire and Omenn (1975) presented data indicating that patients with congenital adrenal hyperplasia do not have higher IQs than expected from the family background. Wenzel et al. (1978) found similar results.

Blankstein et al. (1980) reported a possible allelic form of 21-hydroxylase deficiency in 2 sisters, aged 28 and 30 years, who had primary infertility and mild hirsutism but normal puberty, regular menses, and normal female sexual characteristics. Two sibs were normal. The affected sibs were HLA-identical; their healthy sibs were of different HLA type.

Levine et al. (1980) studied serum androgen and 17-hydroxyprogesterone levels as well as HLA genotypes in 124 families of patients with classic 21-hydroxylase deficiency. In 8 kindreds, 16 pubertal or postpubertal persons of either sex were found to have biochemical evidence of 21-hydroxylase deficiency without clinical symptoms of excess virilism, amenorrhea, or infertility. They designated the disorder 'cryptic 21-hydroxylase deficiency.' Within each generation, the family members with the cryptic form were HLA identical. They suggested that these persons were compound heterozygotes for the classic gene and a cryptogenic gene. Of 42 pediatric patients with 21-hydroxylase deficiency (from 36 families) treated in Milwaukee between 1965 and 1981, 4 developed a malignant tumor: sarcoma or astrocytoma (Duck, 1981).

Kuttenn et al. (1985) found that 21-hydroxylase deficiency was the basis of hirsutism in 24 of 400 women (6%). The diagnosis was made by a high plasma level of 17-hydroxyprogesterone and its marked increase after ACTH stimulation. From genotyping of the 24 families, a high correlation with HLA-B14 and Aw33 was found. Nine HLA-identical sibs showed similar biologic profiles but had no hirsutism; skin sensitivity to androgens may be important in determining clinical expression of the disorder. (It was previously known that unusual sensitivity to androgens can lead to hirsutism despite normal plasma levels of androgen (Kuttenn et al., 1977).) The patients were not distinguishable from women with idiopathic hirsutism or polycystic ovarian disease (184700), either clinically or in plasma androgen levels.

Knochenhauer et al. (1997) hypothesized that heterozygosity for CYP21 mutations in women increases their risk of developing clinically evident hyperandrogenism, and that this risk is related to the severity of the mutation of CYP21 and/or the 17-hydroxyprogesterone (17-OHP) response to ACTH stimulation. To test these hypotheses, they studied 38 obligate carriers for 21-hydroxylase deficiency (i.e., mothers of children with CAH1 or nonclassic CAH), comparing them to 27 controls. Their data indicate that heterozygosity for CYP21 mutations does not appear to increase the risk of clinically evident hyperandrogenism, although carrying the defect was associated with higher mean and free T levels. Finally, due to the low frequency of androgen excess in their heterozygote population, they were unable to correlate the severity of CYP21 mutations and/or 17-OHP responses to ACTH stimulation with the presence of the phenotype.

Sinnott et al. (1989) presented analyses of families that showed profound discordance between the clinical features of sibs with 21-hydroxylase deficiency who appeared to be HLA identical, both in terms of serologically defined HLA polymorphism and in gene organization at the 21-hydroxylase and C4 loci (C4A, 120810; C4B, 120820). For example, in 1 family a boy had the simple virilizing form while his 2 younger sisters, who were both HLA-identical to their brother, had additional salt-wasting features. In 1 family they made the unusual observation of HLA-Bw47-bearing haplotypes that appeared to carry a functional 21-hydroxylase gene.

Jaresch et al. (1992) found a high frequency of asymptomatic adrenal tumors in association with homozygosity (82%) and heterozygosity (45%) for 21-hydroxylase deficiency. Jaresch et al. (1992) suggested that CAH should always be ruled out in the case of incidentally detected adrenal masses. Since CAH is a relatively frequent disorder and adrenal carcinoma belongs to the rarest malignant tumors, they concluded that malignant transformation of these tumors is unlikely.

Ravichandran et al. (1996) pointed out that both homozygous and heterozygous patients with congenital adrenal hyperplasia have an increased cross-sectional area of their adrenal glands as well as an increased prevalence of adrenal incidentalomas, i.e., adrenal tumors discovered incidentally in the course of imaging studies performed for unrelated reasons (Jaresch et al., 1992). The prevalence of adrenal tumors may be more than 70% in nonclassic CAH and 'unmasked heterozygotes.' Ravichandran et al. (1996) presented 2 patients, female pseudohermaphrodites with the simple virilizing form of CAH and 21-hydroxylase deficiency, who functioned successfully as married phenotypic males. Both came to medical attention in their sixth decade by virtue of massive adrenal incidentalomas encountered in the evaluation of recurrent urinary tract infections. Each had a 46,XX karyotype, no palpable testes, and markedly elevated baseline levels of 17-hydroxyprogesterone. Both responded appropriately to dexamethasone suppression. Histologic and autopsy examination of the first patient's tumor and computed tomographic characteristics of the second patient revealed benign adenoma and mild lipoma, respectively. Ravichandran et al. (1996) concluded that these observations extended and confirmed previous recommendations that CAH be included in the differential diagnosis of adrenal incidentaloma and that baseline 17-hydroxyprogesterone levels be obtained, with ACTH stimulation if necessary, to diagnose the presence of nonclassic CAH.

Beuschlein et al. (1998) noted that 21-hydroxylase deficiency had been implicated in the pathogenesis of adrenocortical tumors. They investigated the mutation spectrum of the CYP21B gene and the mRNA expression of P450c21 in 6 aldosterone-producing adenomas, 7 cortisol-producing adenomas, 2 nonfunctional incidentally detected adenomas, and 4 adrenal carcinomas. The 10 exons, intron 2, intron 7, all other exon/intron junctions, and 380 bp of the promoter region of CYP21B were sequenced. In samples from 2 patients (1 with a cortisol-producing adenoma and 1 with an androgen-secreting adrenocortical carcinoma), they detected the heterozygous germline mutation val281 to leu in exon 7 (V281L; 613815.0002). A somatic, heterozygous microdeletion was found in exon 3 of 1 aldosterone-producing adenoma. The P450c21 gene expression correlated with the clinical phenotype of the tumor, with low P450c21 mRNA expression in nonfunctional adenomas (18.8%, 1.5%) compared with high P450c21 expression in aldosterone- and cortisol-producing adenomas (84 +/- 8% and 101 +/- 4%, respectively, vs normal adrenals, 100 +/- 10%). They concluded that the pathophysiologic significance of this finding in the presence of 1 normal CYP21B gene seems to be low, suggesting that 21-hydroxylase deficiency is not a major predisposing factor for adrenal tumor formation.

Stikkelbroeck et al. (2001) investigated the prevalence of testicular tumors in 17 adolescent and adult male patients with CAH aged 16 to 40 years. In 16 of 17 patients, one or more testicular tumors ranging in maximal length from 0.2 to 4.0 cm were found on ultrasonography. In 6 patients, the testicular tumors were palpable. Undertreatment, defined as the presence of a salivary androstenedione level above the upper reference morning level, was found in 5 of 17 patients at the time of investigation. The other 12 patients were treated adequately or even overtreated at the time of investigation. Nevertheless, 11 of these 12 patients showed testicular tumors on ultrasonography. Tumor size was significantly larger in patients who were heterozygous or homozygous for deletion or conversion of the CYP21 gene than in patients who did not have this genotype. Impairment of Leydig cell function as manifested by decreased plasma levels of testosterone was found in 6 of 17 patients. Semen analysis in 11 patients revealed azoospermia in 3 patients and poor semen quality in 4 patients. The authors concluded that, when carefully sought for, testicular adrenal rest tumors are frequently present in adolescent and adult males with CAH and are often accompanied by impaired spermatogenesis and Leydig cell failure.

In a follow-up study of 52 males with congenital virilizing adrenal hyperplasia seen at Johns Hopkins between 1950 and 1978, 51 had 21-hydroxylase deficiency and 1 had 11-hydroxylase deficiency (Urban et al., 1978).

Because little is known about the relation between endogenous TSH and cortisol secretion under physiologic or slightly disturbed conditions, Ghizzoni et al. (1997) evaluated the pulsatility, circadian rhythmicity, and 24-hour secretory patterns of cortisol and TSH in 8 prepubertal children with nonclassic CAH and 8 age-matched short normal children. In both groups, TSH and cortisol were secreted in a pulsatile and circadian fashion, with a clear nocturnal TSH surge. Although no difference in mean 24-hour TSH levels was observed between the 2 groups, daytime TSH levels were lower in the nonclassic CAH group than in controls (P less than 0.05). Cross-correlation analysis showed that TSH and cortisol were negatively correlated, possibly reflecting a negative glucocorticoid effect on TSH under physiologic conditions. The authors concluded that the hypothalamic-pituitary-adrenal axis has a primarily negative influence on endogenous TSH secretion and that even mild disturbances in cortisol biosynthesis can be associated with slight alterations in TSH secretion.

Meyer-Bahlburg (1999) noted that women with classic CAH have relatively low fertility rates. The author stated that the largest clinic population was studied by Mulaikal et al. (1987), who studied 80 women with classic 21-hydroxylase deficiency who were evenly split into the SV and SW forms. Half of the women were not heterosexually active. Those who were heterosexually active nevertheless appeared to have low fertility. Among the 25 SV women who reported both adequate vaginal reconstruction and heterosexual activity, the fertility rate was 60%. Among the 15 SW women with both adequate introitus and heterosexual activity, the fertility rate was only 7%; a single pregnancy was reported and that ended in an elective termination. Meyer-Bahlburg (1999) reviewed the various physical and behavioral factors that could account for the observed low rates of child bearing.

Merke et al. (2000) studied a group of patients with congenital adrenal hyperplasia in whom plasma epinephrine and metanephrine concentrations and urinary epinephrine excretion were approximately 50% lower in those who had been hospitalized for adrenal crises than in those who had not. In 3 patients with congenital adrenal hyperplasia who had undergone bilateral adrenalectomy, the formation of the adrenal medulla was incomplete, and electron-microscopic studies revealed a depletion of secretory vesicles in chromaffin cells. Thus, the authors concluded that congenital adrenal hyperplasia compromises both the development and the functioning of the adrenomedullary system.

Green-Golan et al. (2007) compared 6 adolescents with classic CAH with 7 age-, sex-, and body mass index group-matched controls to assess hormonal, metabolic, and cardiovascular response to prolonged moderate-intensity exercise comparable to brisk walking. The CAH patients showed defective glycemic control and altered metabolic and hormonal responses.

Studies had shown that girls with CAH, a syndrome resulting in overproduction of adrenal androgens from early fetal life, are behaviorally masculinized. Nordenstrom et al. (2002) studied play with toys in a structured play situation and correlated the results with disease severity, assessed by CYP21 genotyping, and age at diagnosis. Girls with CAH played more with masculine toys than did controls when playing alone. In addition, the authors demonstrated a dose-response relationship between disease severity (i.e. degree of fetal androgen exposure) and degree of masculinization of behavior. They concluded that prenatal androgen exposure has a direct organizational effect on the human brain to determine certain aspects of sex-typed behavior.

Hormones of the hypothalamic-pituitary-adrenal axis and sex hormones interact with extrahypothalamic regulatory centers of the brain, including the amygdala and hippocampus. The amygdala is important in the processing of emotion and generation of fear, whereas the hippocampus plays an important role in memory. Chronic hypercortisolemia is associated with hippocampal damage, while glucocorticoids and corticotropin-releasing factor play a major role in the regulation of amygdala function. Merke et al. (2003) performed MRI of the brain on 27 children with classic CAH and 47 sex- and age-matched controls. Volumes of the cerebrum, ventricles, temporal lobe, amygdala, and hippocampus were quantified. Females with CAH did not have brains with male-specific characteristics. In contrast, a significant decrease in amygdala volume was observed in both males and females with CAH (males, P = 0.01; females, P = 0.002). Iatrogenic effects on the hippocampus due to glucocorticoid therapy were not observed in children with CAH. The authors concluded that prenatal glucocorticoid deficiency with resulting alterations in hypothalamic-pituitary-adrenal axis regulation, sex steroid excess, or some combination of these preferentially affect the growth and development of the amygdala, a structure with major functional implications that warrant further exploration.

Berenbaum and Bailey (2003) studied gender identity in girls with CAH in relation to characteristics of the disease and treatment, particularly genital appearance and surgery. Gender identity in girls with CAH was not related to degree of genital virilization or age at which genital reconstructive surgery was done. The authors concluded that moderate androgen excess early in development appears to produce a small increase in the risk of atypical gender identity, but this risk cannot be predicted from genital virilization.

Gidlof et al. (2007) found that female patients with severe CYP21 deficiency had longer gestational age than did patients with a milder form of the disease, indicating that androgen excess, increased 17-hydroxyprogesterone levels, or cortisol deficiency, or a combination of these factors, may be of importance for prolongation of pregnancy. The same correlation was not seen for male patients. The authors concluded that steroid hormones may affect the prolongation of pregnancy or onset of labor or both.

Moran et al. (2006) studied the frequency of CAH and nonclassic CAH (NCAH) infants born to mothers with 21-OH-deficient NCAH. The outcome of 203 pregnancies among 101 women with 21-OH-deficient NCAH was reviewed. The risk of a mother with 21-OH-deficient NCAH giving birth to a child affected with CAH was found to be 2.5%; at least 14.8% of children born to these mothers had NCAH.

Other Features

Winqvist et al. (1992) demonstrated that 21-hydroxylase, which is prominent in the zona glomerulosa of the adrenal cortex, is a major autoantigen in idiopathic Addison disease (240200). This is another example of the way in which genetic disease can be mimicked by the development of autoantibodies against the gene product that is genetically deficient in the inherited disorder. Other examples are hemophilia A (306700), dystrophic epidermolysis bullosa (131750), hereditary angioedema (106100), and perhaps congenital myasthenia gravis (254210). All of these hereditary disorders appear to have an acquired mimic which is an autoimmune disorder.

Inheritance

Congenital adrenal hyperplasia-1 is an autosomal recessive disorder.

Spiro et al. (1999) reported the first case of maternal uniparental disomy for chromosome 6 ascertained through congenital adrenal hyperplasia, which arose because of reduction to homozygosity (or hemizygosity) of an autosomal recessive mutation. The mother was heterozygous for the I172N mutation (613815.0001); the father had no detectable mutations. DNA microsatellite analysis with polymorphic markers spanning the entire chromosome 6 indicated inheritance of a single maternal allele and absence of paternal alleles in the proband. The patient was born with intrauterine growth retardation, followed by catch-up growth.

Mapping

Patients with 21-hydroxylase deficiency also show genetic linkage disequilibrium with complement allotypes. Different forms of 21-hydroxylase deficiency are associated with characteristic HLA haplotypes. Holler et al. (1985) studied HLA types and plasma 17-hydroxyprogesterone levels after ACTH stimulation in 134 German families of patients with salt-wasting (SW), simple virilizing (SV), or nonclassic (NC) late-onset CAH. Hormone evidence for CAH was found in 6 otherwise healthy relatives who, therefore, were thought to be NC cryptic cases. The SW form was strongly associated with HLA Bw47, whereas the SV form was associated with B5(w51). The almost complete association of the NC form with HLA B14 was confirmed. These alleles, especially Bw47 and B14, are components of normally rare haplotypes. Thus, all or almost all persons in the general population with 1 of these haplotypes will be heterozygotes.

Dupont et al. (1977) demonstrated close linkage of 21-hydroxylase deficiency and the HLA complex (lod score = 3.394 at a recombination fraction of 0.00). One patient had inherited a maternal recombinant between HLA-A and HLA-B. Studies in this family indicated that the abnormal gene is close to the HLA-B locus. Both the salt-losing and non-salt-losing forms of 21-hydroxylase deficiency show linkage to HLA, suggesting allelism. Murtaza et al. (1978) identified possible genetic compounds.

Levine et al. (1978) obtained a lod score of 9.5 with a 0.00 recombination fraction. In a study of 48 patients, 48 sibs and their parents, all patients were HLA-different from their unaffected sibs. When 2 or more children were affected in a sibship they were always HLA-B identical. In 34 unrelated patients no selective increase of a particular haplotype was observed, thus excluding association or linkage disequilibrium. Klouda et al. (1980) found a lod score of almost 9.0 for the linkage of HLA-B and 21-hydroxylase deficiency at a recombination fraction of 0.03. They pointed to the association of an excess of HLA-Bw47 with a deficiency of HLA-B8 persons. The workers concluded that the 21-hydroxylase locus 'lies outside the HLA system and is closely linked to the HLA-DR locus.' Fleischnick et al. (1983) demonstrated that extended MHC haplotypes are markers for different mutations causing 21-hydroxylase deficiency, just as the extended restriction nonalpha-globin haplotypes are markers for different beta-thalassemia mutations. In studying 29 families, more than 20% were found to have a very rare extended haplotype (taking into consideration complement loci and glyoxalase I as well). Furthermore, 3 other haplotypes were each found twice in unrelated patients concordant for their disease phenotype and ethnic background. Previously, striking linkage disequilibrium was noted; e.g., in Sheffield, England, the frequency of Bw47 was 27.3% in the patient population and 0.4% in the general European population. They commented on the fact that Klouda et al. (1980) as well as at least 1 other group placed CAH between D/DR and GLO1, whereas others place it between HLA-A and HLA-D/DR (Pucholt et al., 1980; Bias et al., 1981). Sobel et al. (1980) pointed out that heterozygotes can be detected by the linkage principle. They also reported the first instance of presumed recombination between AH3 and the HLA-B locus.

Presumably because of linkage disequilibrium, the common severe form of 21-hydroxylase deficiency is positively associated with Bw47 and negatively with B8, while the late-onset type is positively associated with B14 (reviewed by Petersen et al., 1982). HLA haplotyping was used to confirm the genetic compound nature of the cryptic form of 21-hydroxylase deficiency (Zachmann and Prader, 1979; Levine et al. (1980, 1981)).

Patients with the HLA-Bw47 antigen invariably show simultaneous deficiencies of 21-hydroxylase activity and the C4A (Rodgers) form of C4. The HLA-Bw47(w4) antigen is very similar serologically and otherwise to the more common antigen HLA-B13(w4). Therefore, it was proposed that a deletion or rearrangement simultaneously affected the B13 gene and the closely linked 21-hydroxylase locus. White et al. (1984) used a plasmid with bovine adrenal cDNA insert encoding part of the cytochrome P450 polypeptide to examine this hypothesis. The hybridization patterns of normal DNA and that from 21-hydroxylase-deficient persons were compared. One band from both EcoRI and TaqI digests was absent in DNA from a patient homozygous for HLA-Bw47. Of 6 unrelated patients homozygous for CAH and heterozygous for HLA-Bw47, 5 had a relative intensity of this band consistent with heterozygosity and one had complete absence. The deletion segregated with HLA-Bw47 in a large pedigree with 21-hydroxylase deficiency and HLA-Bw47. These authors referred to the structural gene for P450(C21). Close linkage of said gene and that for C4A is indicated by the occurrence of the null allele at that locus. Apparently only one of the two 21-hydroxylase genes is mutant. Several alternative explanations might be considered. The second gene may in fact be mutated also. The second gene may be regulated by the renin-angiotensin system and be involved in aldosterone synthesis in the zona glomerulosa. The second gene may be a pseudogene or may be expressed only at certain times in ontogeny or in other organs (the kidney and liver also contain 21-hydroxylase activity). In an addendum, White et al. (1984) stated that reexamination of the C4 allotypes associated with HLA-Bw47 led to the conclusion that the data are consistent with the P450(C21) gene being near the C4B (Chido) gene and both of those genes being deleted in the case of the HLA-Bw47 haplotype.

Speiser et al. (1985) studied the frequency of 21-hydroxylase deficiencies in several ethnic groups and showed that the gene for the nonclassic form is in linkage disequilibrium with HLA-B14. The classic form shows linkage disequilibrium with HLA-Bw47;DR7.

By analysis of data collected on 157 families ascertained through a proband with the classic form of 21-hydroxylase deficiency, Aston et al. (1988) could not arrive at a definitive conclusion as to whether the gene is closer to HLA-B or to HLA-DR. They pointed out the limitations of present methods of estimating genetic distance when recombination frequencies are of the order of 0.005.

Diagnosis

Merkatz et al. (1969) could not diagnose the disorder early in pregnancy by amniocentesis and hormone assay of the amniotic fluid.

Levine et al. (1980) expressed the opinion that experience is still so limited with HLA typing of amniotic cells and with hormonal measurements of amniotic fluid that both approaches to prenatal diagnosis should be used. Gueux et al. (1988) found significant elevations of both 21-deoxycortisol and 17-hydroxyprogesterone in the amniotic fluids of affected pregnancies, as determined by HLA typing and linkage analysis to HLA probes. Hughes et al. (1987) determined the concentration of 17-OH-progesterone in the amniotic fluid collected from 55 pregnant women who had previously had a child with 21-hydroxylase deficiency. In 8 pregnancies the levels were raised. These parents elected to terminate in 4 cases, and examination of the fetus confirmed the diagnosis of CAH. In each case, the affected sib had been a salt-loser. The remaining 4 affected pregnancies proceeded to term, and each infant had salt-losing 21-hydroxylase deficiency. All 47 infants predicted to be unaffected were normal at birth; however, an increased plasma concentration of 17-OH-progesterone was documented in a male non-salt-loser at 3 months of age. Hughes et al. (1987) concluded that prenatal diagnosis of congenital adrenal hyperplasia by amniotic fluid steroid analysis is reliable only for the salt-losing form. They published a photograph of the external genitalia of a female fetus with 21-hydroxylase deficiency which showed clitoromegaly and fusion of the labia.

Wudy et al. (1999) used routine stable isotope dilution/gas chromatography-mass spectrometry to profile 17-hydroxyprogesterone, androstenedione, testosterone, dehydroepiandrosterone, androstanediol, and 5-alpha-dihydrotestosterone in amniotic fluids of midgestation in 77 normal fetuses and 38 untreated or dexamethasone-treated fetuses at risk for 21-hydroxylase deficiency. Dexamethasone was suspended 5 to 7 days before amniocentesis. Regarding prenatal diagnosis of 21-hydroxylase deficiency, 17-hydroxyprogesterone and androstenedione presented the diagnostically most valuable steroids and were of equal diagnostic potential. They permitted successful diagnosis in 36 of 37 (97%) fetuses at risk; 12 were untreated and unaffected, 13 were treated and unaffected, 4 were untreated and affected (3 salt wasters and 1 simple virilizer), and 8 were treated and affected (5 salt wasters and 3 simple virilizers). In the latter group, 1 simple virilizer revealed normal steroid concentrations. The authors proposed that isotope dilution/gas chromatography-mass spectrometry, providing the highest specificity in steroid analysis, be routinely used in clinical steroid analysis whenever maximal reliability is requested.

Definitive neonatal diagnosis of CAH is frequently complicated by normal 17-hydroxyprogesterone levels in 21-hydroxylase-deficient patients, residual maternal steroids, and other interfering substances in blood. In an effort to improve the diagnosis, Caulfield et al. (2002) developed a gas chromatography/mass spectrometry method for simultaneous measurement of 15 urinary steroid metabolites as early as the first day of life. Random urine samples from 31 neonatal 21-hydroxylase-deficient patients and 59 age-matched normal newborns were used in the development. Furthermore, the authors developed 11 precursor/product ratios that diagnosed and clearly differentiated the 4 enzymatic deficiencies that cause CAH. The throughput for one bench-top gas chromatography/mass spectrometry instrument was 20 samples per day. The authors concluded that this method afforded an accurate, rapid, noninvasive means for the differential diagnosis of CAH in the newborn period without the need for invasive testing and ACTH stimulation.

New et al. (1983) published nomograms relating baseline and ACTH-stimulated levels of adrenal hormones. These nomograms distinguished the milder symptomatic and asymptomatic nonclassic forms of 21-hydroxylase deficiency (termed late-onset and cryptic forms, respectively), as well as heterozygotes for all of the forms, from normal subjects.

The cutoff level for ACTH-stimulated 17OHP for the diagnosis of the nonclassic form of 21-hydroxylase deficiency (21OHD), established before molecular studies, is based on the mean +2 SD of 17OHP levels of obligate heterozygotes. However, carriers of CYP21 mutations present variable ACTH-stimulated 17OHP levels, ranging from normal values up to 30 nmol/liter. Bachega et al. (2002) sought to determine if ACTH-stimulated 17OHP levels in obligate carriers for 21OHD would be correlated with the impairment of the enzyme activity caused by these mutations, which would affect the 17OHP cutoff level for the diagnosis of the nonclassical form. Fifty-nine parents of patients with the classical and nonclassical forms of 21OHD had their DNA screened for the mutations found in the index case and were divided into 3 mutation groups according to the impairment of enzyme activity (A equal to 0%, B equal to 3%, and C greater than 20%). Blood samples were collected at baseline condition and 60 minutes after ACTH (250 microg intravenously) to measure 17OHP levels. The levels among groups A, B, and C were compared using the Kruskall Wallis test. ACTH-stimulated 17OHP levels identified 39% of the carriers (9 in group A, 2 in group B, and 12 in group C). The mean +/- SD basal 17OHP levels in groups A, B, and C were: 2.94 +/- 1.89, 1.77 +/- 0.81, and 3.90 +/- 2.43 nmol/liter, respectively (P greater than 0.05) and for ACTH-stimulated levels were 12.6 +/- 7.2, 13.2 +/- 12.9, and 16.8 +/- 7.8 nmol/liter, respectively (P greater than 0.05). Two carriers presented ACTH-stimulated 17OHP levels between 30 and 45 nmol/liter and their entire CYP21 sequencing revealed only 1 mutation in heterozygous state, indicating that the cutoff level might overestimate the diagnosis of the nonclassical form. The authors concluded that the variable ACTH-stimulated 17OHP levels in carriers are not related to CYP21 gene mutations with different impairment of enzyme activity.

Mornet et al. (1986) demonstrated that one can use linkage of HLA-DNA probes in chorion villus samples in the first trimester diagnosis. They also used determination of 17-hydroxyprogesterone in the first trimester amniotic fluid in the diagnosis.

Reindollar et al. (1988) described the use of a RFLP of the 21-hydroxylase gene for prenatal diagnosis.

Lee et al. (1996) developed primers for differential PCR-amplification of the CYP21 gene and the nonfunctional CYP21P gene. Using the amplification created restriction site (ACRS) approach for direct mutation detection, a secondary PCR was then performed using a panel of primers specific for 11 mutations associated with CAH. Subsequent restriction analysis allowed not only the detection but also the determination of the zygosity of the mutations analyzed. In the analysis of 20 independent chromosomes in 11 families of CAH patients in Taiwan, Lee et al. (1996) detected 4 CYP21 mutation types besides deletion. In 5 different alleles, the CYP21P pseudogene contained some polymorphisms that the authors believed to be associated with the CYP21 gene. This finding suggested that gene conversion events are occurring in both CYP21P and CYP21. The combined differential PCR-ACRS protocol was described as simple, direct, and applicable to prenatal diagnosis of CAH using chorionic villi or amniotic cells.

During the course of genetic analysis of CYP21 mutations in CAH families, Day et al. (1996) noticed a number of relatives genotyped as nucleotide 656G (613815.0006) homozygotes who showed no clinical signs of disease. They proposed that the putative asymptomatic nucleotide 656G/G individuals are incorrectly typed due to a dropout of 1 haplotype during PCR amplification of CYP21. They recommended that for prenatal diagnosis, microsatellite typing be used as a supplement to CYP21 genotyping in order to resolve ambiguities at nucleotide 656.

Lako et al. (1999) reported the development of a linkage analysis approach using novel, highly informative microsatellite markers from the class III HLA region to allow highly accurate prenatal diagnosis in all families where samples are available from an affected child.

To evaluate genotyping as a diagnostic complement to neonatal screening for CAH, Nordenstrom et al. (1999) analyzed DNA from 91 children who had been diagnosed with CAH between 1986 and 1997 for mutations in the CYP21 gene. Screening levels of 17-hydroxyprogesterone (17OHP) were compared in patients representing different genotypes. Genotyping was done by allele-specific PCR, the patients were divided into 4 groups by the severity of their mutations, and neonatal screening results were compared between these groups as well as with 141 values representing false positive samples. The screening levels of 17OHP were significantly different in the 5 groups of samples. Values above 500 nmol/L were clearly associated with the most severe genotypes, whereas conclusions concerning disease severity could not be drawn from individual samples representing lower levels. The authors concluded that genotyping is a valuable diagnostic tool and a good complement to neonatal screening, especially in confirming or discarding the diagnosis in cases with slightly elevated 17OHP levels.

Koppens et al. (2002) noted that duplication of the CYP21A2 gene complicates mutation analysis. They recommended that whenever CYP21A2 mutation analysis is performed in an individual who is not a known carrier of the deficiency, the overall structure of the CYP21/C4 region (the RCCX area) should be determined by haplotyping to avoid erroneous assignment of carrier status.

To improve the specificity of newborn screening for CAH, Minutti et al. (2004) developed a method using liquid chromatography-tandem mass spectrometry to measure 17-hydroxyprogesterone, androstenedione, and cortisol simultaneously in blood spots. The authors recommended the assay as a second-tier test of blood spots with positive results for CAH screening by conventional methods.

Homma et al. (2004) studied the diagnostic value of the metabolite of 21-deoxycortisol, also known as pregnanetriolone (Ptl), and the metabolite of 17OHP, or pregnanetriol (PT), in identifying 21OHD in term and preterm neonates with elevated blood 17OHP on the newborn screening. They found spot urine Ptl to be a highly specific marker of 21OHD with a cutoff value of 0.1 mg/g creatinine, yielding an unambiguous separation between 21OHD and non-21OHD in term and preterm neonates. They recommended that spot urine Ptl measurement by gas chromatography/mass spectrometry in selected ion monitoring (GC/MS-SIM) be routinely performed in neonates with elevated blood 17OHP detected by newborn screening, if the diagnosis of 21OHD is uncertain.

Van der Kamp et al. (2005) determined that gestational age rather than birth weight provides a better basis for cutoff levels of 17OHP in newborn blood screening tests for CAH.

Janzen et al. (2007) reported a second-tier liquid chromatography-tandem mass spectrometry procedure that could be used to reduce false-positive results of standard 21-CAH newborn screening.

Clinical Management

Jones (1978) found cases of mild 'adult' adrenal hyperplasia manifest by oligomenorrhea and treated like the usual form with adrenocorticosteroids.

Cutfield et al. (1983) described 2 male cousins with partial 21-hydroxylase deficiency presenting with bilateral testicular masses and infertility. In both cases, the testicular masses, consisting of adrenocorticotropic hormone-dependent pluripotential interstitial cells, were thought to play a major etiologic role in infertility. Nighttime (11 p.m.), low-dose dexamethasone therapy led to disappearance of the masses and restoration of fertility. Hydrocortisone, 10 mg 3 times daily, had failed to accomplish this reversal.

Cutler and Laue (1990) investigated the use of a new form of therapy which would combine hydrocortisone in strictly physiologic dosage with antiandrogen and aromatase-inhibitor therapy. By blocking androgen and the conversion of testosterone to estrogen, they hoped to achieve normal growth. The proposal, which remained to be tested, was suggested by the success of a similar program in the treatment of familial male precocious puberty (176410).

A multicentric study of prenatal treatment of 21-hydroxylase deficiency with dexamethasone administered by mouth to the mother was undertaken in France (Forest et al., 1989). Wudy et al. (1994) reported successful treatment of a single case with dexamethasone (0.5 mg, tid, p.o.) starting from the beginning of the eighth week of gestation.

Cornean et al. (1998) studied 22 prepubertal children with 21-hydroxylase deficiency whose steroid therapy was considered to be optimal in terms of linear growth and skeletal maturity. The authors reported a significant increase in body mass index (BMI) as a consequence of increased body fat. This was consistent with an early 'rebound' of BMI, which is associated with obesity in later childhood and an increased risk of long-term health problems related to adult obesity.

In children, CAH is often treated with cortisone acetate and fludrocortisone. Certain patients with CAH require very high substitution doses of cortisone acetate, and a few do not respond to cortisone acetate at all. Nordenstrom et al. (1999) reported a patient with 21-hydroxylase deficiency in whom elevated pregnanetriol levels in urine were not suppressed during treatment with cortisone acetate (65 mg per m2-day). The patient's lack of response to treatment with cortisone acetate was caused by a low conversion of cortisone to cortisol, assumed to be secondary to low 11-beta-hydroxysteroid dehydrogenase activity. These results supported the use of hydrocortisone, rather than cortisone acetate, for substitution therapy in adrenal insufficiency.

Travitz and Metzger (1999) discussed prenatal treatment of classic 21-OH forms of congenital adrenal hyperplasia. Dexamethasone (DEX) is a potent glucocorticoid that inhibits the adrenal cortex through feedback on the hypothalamus and pituitary. It is used for prenatal treatment because, compared with other glucocorticoids, it crosses the placenta more efficiently (approximately 50% reaches the fetal side), has a longer half-life (approximately 4 to 6 hours), and has a greater suppressive effect on ACTH. The arguments for and against prenatal DEX therapy were reviewed.

Lo et al. (1999) reported the pregnancy outcomes and serial measurements of maternal serum steroid levels in 4 women with classic 21OH deficiency, 3 of whom were female pseudohermaphrodites with the salt-losing form. These glucocorticoid-treated women gave birth to 4 healthy female newborns with normal female external genitalia, none of whom were affected with 21OH deficiency. In 3 women, circulating androgen levels increased during gestation, but remained within the normal range for pregnancy during glucocorticoid therapy. In the fourth patient, androgen levels were strikingly elevated during gestation despite increasing the dose of oral prednisone from 5 to 15 mg/day (2 divided doses). The authors concluded that despite the high maternal serum concentration of androgens, placental aromatase activity was sufficient to prevent masculinization of the external genitalia of the female fetus and quite likely the fetal brain.

Laue et al. (1996) reported better control of linear growth, weight gain, and bone maturation in a short-term crossover study of a 4-drug treatment regimen containing an antiandrogen (flutamide), an inhibitor of androgen-to-estrogen conversion (testolactone), reduced hydrocortisone dose, and fludrocortisone, compared to the effects of a control regimen of hydrocortisone and fludrocortisone. Merke et al. (2000) reported the results of a subsequent long-term randomized parallel study comparing these 2 treatment regimens. Twenty-eight children completed 2 years of follow-up. During 2 years of therapy, compared to children receiving hydrocortisone and fludrocortisone treatment, children receiving flutamide, testolactone, reduced hydrocortisone dose, and fludrocortisone had significantly higher plasma 17-hydroxyprogesterone, androstenedione, dehydroepiandrosterone, dehydroepiandrosterone sulfate, and testosterone levels. Despite elevated androgen levels, children receiving the new treatment regimen had normal linear growth rate and bone maturation. No significant adverse effects were observed after 2 years. The authors concluded that the regimen of flutamide, testolactone, reduced hydrocortisone dose, and fludrocortisone provides effective control of CAH with reduced risk of glucocorticoid excess.

In CAH due to 21-hydroxylase deficiency, treatment with glucocorticoid and mineralocorticoid substitution is not always satisfactory. Suboptimal control is often observed in pubertal patients, despite adequate replacement doses and adherence to treatment. Charmandari et al. (2001) investigated whether the pubertal process is associated with alterations in cortisol pharmacokinetics resulting in a loss of control of the hypothalamic-pituitary-adrenal axis. They found that the serum total cortisol clearance curve was monoexponential. Mean clearance was significantly higher in the pubertal group compared with the prepubertal and postpubertal groups. The mean volume of distribution was also significantly higher in the pubertal than in the prepubertal patients but not in the postpubertal patients. In addition, the half-life of free cortisol was significantly shorter in females compared with males. Charmandari et al. (2001) concluded that puberty is associated with alteration in cortisol pharmacokinetics resulting in increased clearance and volume of distribution with no change in half-life. They also concluded that these alterations probably reflect changes in the endocrine milieu at puberty and may have implications for therapy of CAH and other conditions requiring cortisol substitution in the adolescent years.

In a multicenter retrospective chart review of 54 patients with salt-wasting 21-OHD CAH who were diagnosed in the first 6 months of life and had reached adult height, Muirhead et al. (2002) found that adult height was negatively correlated with androstenedione in infancy (p = 0.03) and childhood (p less than 0.01) and with testosterone in childhood (p = 0.01). They recommended that androgen levels be used in conjunction with growth velocity measurements to optimize glucocorticoid dosing in persons with 21-OHD CAH.

Bonfig et al. (2007) studied final height outcome and influences of steroid treatment in 125 patients (77 females) with CAH. They concluded that patients with CAH are able to achieve adequate FH with conventional therapy. Total pubertal growth is significantly decreased, and treatment with prednisone results in decreased FH.

The Joint LWPES/ESPE CAH Working Group (2002) published a consensus to address the best practice, management guidelines, and innovative therapies for CAH caused by 21-hydroxylase deficiency, including guidelines for neonatal diagnosis and treatment, clinical evaluation in term and premature neonates, newborn screening for CAH, prenatal diagnosis, and treatment and management in adolescence and adulthood.

Creighton et al. (2003) objected to the surgical management guidelines of the consensus statement of the Joint LWPES/ESPE CAH Working Group (2002) and stated that the only consensus attainable at that time would be that of a dedicated multidisciplinary team addressing an individual case including the full participation of the affected family.

Van Wyk and Ritzen (2003) summarized follow-up studies in 18 patients who underwent bilateral