Original Article
Secondary Biosynthetic Defects in Women with Late-Onset Congenital Adrenal Hyperplasia
N Engl J Med 1990; 323:855-863September 27, 1990
- Abstract
- Abstract
Background and Methods.
Late-onset (nonclassic) congenital adrenal hyperplasia is a cause of hirsutism, menstrual disorders, and infertility, but its frequency and the patterns of abnormalities in adrenal hormone secretion are not well understood. We investigated the frequency and ethnic distribution of nonclassic congenital adrenal hyperplasia due to deficiencies of 3β-hydroxy
5-steroid dehydrogenase, 21-hydroxylase, or 11β-hydroxylase among 170 Israeli Jewish women with these clinical problems. All enzyme defects were identified by comparing the patients' hormonal responses to a 0.25-mg intravenous bolus dose of α1–24-ACTH with those of 26 age-matched normal women.Results.
Twenty women (12 percent) had 3β-hydroxy
5-steroid dehydrogenase deficiency, 18 (10 percent) 21-hydroxylase deficiency (14 homozygous), and 14 (8 percent) 11β-hydroxylase deficiency. All the homozygous women with 21-hydroxylase deficiency also had evidence of a partial deficiency in 11β-hydroxylase activity. Similarly, most of the women with 11β-hydroxylase deficiency also had evidence of a deficiency in 3β-hydroxy-5-steroid dehydrogenase. Among the 118 women with no adrenal biosynthetic defect, 38 had high plasma androgen concentrations, and 80 had normal concentrations.Conclusions.
About one third of Israeli Jewish women with hirsutism, menstrual disorders, or unexplained infertility had nonclassic congenital adrenal hyperplasia. Secondary adrenal biosynthetic defects were frequent in these women and were probably caused by intraadrenal androgen excess rather than by dual inherited enzymatic deficiencies. (N Engl J Med 1990; 323: 855–63.)
Media in This Article
Figure 1
Plasma Steroid Concentrations in Women with Nonclassic Congenital Adrenal Hyperplasia Due to 3β-Hydroxy-5-steroid Dehydrogenase Deficiency at Base Line, 60 Minutes after Intravenous Bolus Injection of 0.25 mg of α1–24-ACTH, and after a 48-Hour Course of Dexamethasone (Dex) (0.5 mg Four Times Daily).Figure 2
Plasma Steroid Concentrations in Women with Nonclassic Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency at Base Line, 60 Minutes after Intravenous Bolus Injection of 0.25 mg of α1–24-ACTH, and after a 48-Hour Course of Dexamethasone (Dex) (0.5 mg Four Times Daily).Article Activity
- Article
NONCLASSIC (late-onset) congenital adrenal hyperplasia is a frequent and relatively mild disorder of cortisol biosynthesis, characterized by a spectrum of clinical manifestations of postnatal androgen excess. These characteristics distinguish it from the classic, severe forms of congenital adrenal hyperplasia.1 The nonclassic form is usually caused by a deficiency of steroid 21-hydroxylase (steroid 21-monooxygenase)1 or 3β-hydroxy-
5-steroid dehydrogenase,1 , 2 whereas steroid 11β-hydroxylase (steroid 11β-monooxygenase)1 deficiency is considered extremely rare.3 4 5 The clinical manifestations of these three deficiencies are similar, and the exact diagnosis can be established only by hormonal measurements. Steroid 21-hydroxylase deficiency is characterized by high plasma 17-hydroxyprogesterone concentrations after stimulation with ACTH.1 The biochemical diagnosis of the other two defects is more complicated. The diagnosis of 3β-hydroxy-5-steroid dehydrogenase deficiency can be based on increased plasma concentrations of 17-hydroxypregnenolone and dehydroepiandrosterone (DHEA) after ACTH stimulation, a pattern that may be also encountered in 21-hydroxylase deficiency.2 Nonetheless, the two defects differ: 3β-hydroxy-5-steroid dehydrogenase deficiency is associated with an abnormally high ratio of 5-steroid to 4-steroid metabolites (the ratio of 17-hydroxypregnenolone to 17-hydroxyprogesterone), whereas the opposite is found in 21-hydroxylase deficiency. Criteria for the diagnosis of 11β-hydroxylase deficiency have not been established. Data on plasma 11-deoxycortisol levels in response to ACTH stimulation are not available, and this metabolite has not been measured systematically in patients suspected of having nonclassic congenital adrenal hyperplasia.HLA genotyping also may be used for the diagnosis of nonclassic adrenal hyperplasia due to 21-hydroxylase deficiency.1 The ethnic origin of the patient can be important as well. Recent studies have demonstrated that this disorder is extremely common, occurring in approximately 0.3 percent of the general white population, 1.6 percent of Yugoslavs, 1.9 percent of Hispanics, and 3.7 percent of Jews of Eastern European (Ashkenazi) origin.6
In this study we determined the frequency and ethnic distribution of nonclassic congenital adrenal hyperplasia due to deficiencies of 3β-hydroxy-
5-steroid dehydrogenase, 21-hydroxylase, and 11β-hydroxylase among Israeli Jewish women with hirsutism, menstrual disorders, unexplained sterility, or some combination of these symptoms. We also determined whether there were secondary enzymatic defects caused by the increased adrenal androgens that accumulate as a result of the primary enzymatic defect.Methods
Subjects
The control group for this study consisted of 26 normal women, from 18 to 41 years of age (mean, 26.9), with regular menses (every 26 to 32 days) and no signs of virilization. None had a family history of endocrinologic disorders, and none were taking any medications (including oral contraceptives).
The patient population consisted of 170 women referred to our institution between 1985 and 1988 for investigation of hirsutism, menstrual disorders, or unexplained sterility. Some of these women also had acne or were obese. The 117 women who were hirsute had long-standing, slowly progressive excessive hair growth on the face, chest, abdomen, buttocks, or limbs.7 The severity of hirsutism, scored according to the method of Ferriman and Gallwey, ranged between 5 and 25.8 The 38 women with acne had inflammatory papular or pustular lesions of the pilosebaceous unit (acne vulgaris) on the face, chest, or back.7 The 87 women with menstrual disorders had had very irregular menses or secondary amenorrhea for at least six months. Thirty-three women were obese, weighing 120 percent or more of their ideal body weights.9 In 31 women the only clinical problem was unexplained infertility. All patients underwent general endocrine evaluation, gynecologic examination, and pelvic ultrasonography, laparoscopy, or both when indicated. None of the women had had ambiguous genitalia at birth, and none had clitoromegaly or fusion of the labia on physical examination. Patients with other disorders that can cause these clinical problems, such as thyrotoxicosis, Cushing's syndrome, and virilizing adrenal or ovarian tumors, were excluded from the study.
The relevant clinical information for the 170 women is shown in Table 1Table 1
Clinical Characteristics and Ethnic Origins of 170 Women with Hirsutism, Menstrual Disorders, or Unexplained Infertility., according to the different defects encountered. The women who had no evidence of an adrenal biosynthetic defect were divided into two groups according to whether their plasma androstenedione or testosterone levels (or levels of both) were high (presumably because of production by the ovary) or normal.Of the women studied, 156 underwent evaluation for polycystic ovaries. The 46 women designated as having the polycystic ovary syndrome had one or more of the following: chronic anovulation, a ratio of luteinizing hormone to follicle-stimulating hormone in plasma of more than 3, or definite ovarian cystic abnormalities identified by ultrasonography, laparoscopy, or biopsy, whether their plasma androgen levels were elevated or normal.10
Protocols
We screened for nonclassic congenital adrenal hyperplasia and identified the different enzyme deficiencies by determining the hormonal responses to an intravenous bolus dose of 0.25 mg (25 U) of synthetic α1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24-ACTH (Cortrosyn) and to oral administration of dexamethasone (0.5 mg four times daily for 48 hours) during the follicular phase of the cycle in the women who had regular menstrual cycles. Blood samples were drawn between 8 and 9 a.m. basally, 60 minutes after the administration of ACTH, and after the 48-hour course of dexamethasone. Written informed consent was obtained from all patients or their parents or guardians.
Biochemical Criteria for Enzyme Deficiency
A diagnosis of 3β-hydroxy-
5-steroid dehydrogenase deficiency was made if the plasma concentrations of 17-hydroxypregnenolone, DHEA, and DHEA sulfate and the ratios of 17-hydroxypregnenolone to 17-hydroxyprogesterone and of DHEA to androstenedione were more than 2 SD above the normal mean after stimulation with ACTH.1 , 2 An abnormal value for the latter ratio, however, was not considered essential for diagnosis.A diagnosis of homozygous nonclassic 21-hydroxylase deficiency was made if the plasma concentration of 17-hydroxyprogesterone was 45.4 nmol per liter or higher after ACTH stimulation. Additional abnormalities not essential for but supportive of this diagnosis were elevated plasma concentrations of 21-deoxycortisol, androstenedione, and testosterone and a ratio of 17-hydroxyprogesterone to 11-deoxycortisol more than 2 SD above the normal mean. A diagnosis of heterozygous nonclassic 21-hydroxylase deficiency was made on the basis of the genetic pattern (in the case of obligate carriers), HLA genotype, or both and a plasma 17-hydroxyprogesterone concentration more than 2 SD above the normal mean (15.1 nmol per liter) but less than 45.4 nmol per liter after ACTH stimulation. These ranges of plasma 17-hydroxyprogesterone values were defined in previous studies of homozygous and heterozygous persons with nonclassic 21-hydroxylase deficiency.11
A diagnosis of 11β-hydroxylase deficiency was made if the plasma concentration of 11-deoxycortisol and the ratio of 11-deoxycortisol to cortisol were both more than 2 SD above the normal mean after ACTH stimulation. Additional abnormalities not essential for but supportive of the diagnosis were elevated plasma concentrations of 11-deoxycorticosterone, androstenedione, and testosterone.
The diagnosis of a secondary enzyme deficiency in patients with 21-hydroxylase or 11β-hydroxylase deficiency was made according to our criteria. For example, the patients with 21-hydroxylase deficiency all had an abnormal 17-hydroxyprogesterone pattern (and, when determined, linkage to HLA), and their secondary defect was 11β-hydroxylase deficiency, as defined earlier. On the other hand, none of the patients with 11β-hydroxylase deficiency had any evidence of 21-hydroxylase deficiency, but most had 3β-hydroxy-
5-steroid dehydrogenase deficiency, according to our definition, as a secondary defect. Thus, only women with no evidence of deficiencies of 21-hydroxylase and 11β-hydroxylase had the biochemical abnormalities of 3β-hydroxy-5-steroid dehydrogenase deficiency as a primary defect.Although plasma concentrations of DHEA or DHEA sulfate (or both), androstenedione, and testosterone were measured in all women before and after the administration of ACTH, we used the basal values for androstenedione and testosterone to subdivide the group of patients with no enzyme deficiencies into those with values more than 2 SD above the normal mean and those with normal androgen production.
Definition of Ethnic Origin
The Jewish communities have been classified into several large ethnic groups according to historic, geographic, and cultural criteria.12 Ashkenazi Jews are those originating in Eastern Europe. Non-Ashkenazi Jews comprise two broad ethnic groups: Sephardic (Spanish) and Asian Jews. Ashkenazi Jews compose approximately 82 percent of world Jewry,12 but only about 38 percent of the current Israeli population.13
The genealogic data were collected in personal interviews of the patients and their relatives. Ethnic heritage was determined according to the origin of the parents. When one or both parents or their ancestors had been born in Israel, ethnic heritage was determined according to the origin of the first ancestor who could be traced.
Techniques and Statistical Analysis
We used sensitive and specific radioimmunologic methods described previously to measure the concentrations of Cortisol,14 17-hydroxyprogesterone,15 17-hydroxypregnenolone,16 11-deoxycortisol,17 21-deoxycortisol,15 11-deoxycorticosterone,14 DHEA and DHEA sulfate,16 androstenedione,18 and testosterone.18 HLA typing of peripheral-blood lymphocytes was performed with the fluorochromatic cytotoxicity test and the standard two-stage microcytotoxicity test of the National Institutes of Health.19 The hormonal data were analyzed with Student's unpaired t-test, and the ethnic data were analyzed by the z-proportion test or the binomial-distribution test.20 The results are reported as means ±SD, and the normal values in the figures as means ±2 SD.
Results
Normal Women and Women with No Enzyme Deficiency
The plasma steroid concentrations of the 26 normal women before and in response to the administration of ACTH and dexamethasone and the precursor-to-product ratios are shown in Table 2Table 2
Plasma Androgen Concentrations in 26 Normal Women and 118 Women with Hirsutism, Menstrual Disorders, or Unexplained Infertility Who Did Not Have an Adrenal Biosynthetic Defect.* and Figures 1Figure 1Plasma Steroid Concentrations in Women with Nonclassic Congenital Adrenal Hyperplasia Due to 3β-Hydroxy-5-steroid Dehydrogenase Deficiency at Base Line, 60 Minutes after Intravenous Bolus Injection of 0.25 mg of α1–24-ACTH, and after a 48-Hour Course of Dexamethasone (Dex) (0.5 mg Four Times Daily). through 4Figure 4
Plasma Steroid Precursor-to-Product Ratios 60 Minutes after ACTH Stimulation in Women with Nonclassic Congenital Adrenal Hyperplasia Due to Deficiency of 3β-Hydroxy-5-steroid Dehydrogenase, 21-Hydroxylase, or 11β-Hydroxylase and in 118 Women with No Adrenal Biosynthetic Defect..Of the 170 patients, 118 had no biochemical evidence of an adrenal biosynthetic defect. All had normal basal plasma concentrations of cortisol, 17-hydroxypregnenolone, 17-hydroxyprogesterone, 11-deoxycortisol, 21-deoxycortisol, and 11-deoxycorticosterone and normal responses to stimulation with ACTH and dexamethasone (data not shown). Their precursor-to-product ratios, shown in Figure 4, also were normal. Among these 118 women, 38 had elevated plasma concentrations of androstenedione, testosterone, or both, basally and in response to ACTH (P<0.001), and 80 had normal levels. Dexamethasone reduced the androstenedione and testosterone levels in the women with normal androgen levels to the same extent as it did in the normal women, but suppression was incomplete in the group with high androgen levels (Table 2).
Although some women in all groups had the polycystic ovary syndrome, the incidence (and that of the more severe clinical forms) was higher (45 percent) in the group with high androgen levels and nearly as high (39 percent) in the group with 3β-hydroxy-
5-steroid dehydrogenase deficiency.Women with Partial 3β-Hydroxy-
5-steroid Dehydrogenase DeficiencyTwenty women (12 percent) had partial 3β-hydroxy-
5-steroid dehydrogenase deficiency. Nineteen of them had increased plasma 17-hydroxypregnenolone concentrations in response to stimulation with ACTH, and all 20 had an increased ratio of 17-hydroxypregnenolone to 17-hydroxyprogesterone (mean [±SD], 7.3±3.2, vs. 3.1±1.2 in normal women; P<0.0005), increased plasma DHEA concentrations, and an increased ratio of DHEA to androstenedione (mean, 7.6±2.3, vs. 3.8±1.2 in normal women; P<0.001) (Fig. 1 and 4). Their cortisol response to ACTH was normal, and DHEA sulfate levels were increased in only 14 of the 20 patients. Their androstenedione and testosterone concentrations were normal, moderately increased, or greatly increased, and their testosterone levels did not increase further in response to stimulation with ACTH.Two women in this group had increased plasma 11-deoxycortisol concentrations; however, their ratios of 11-deoxycortisol to cortisol were normal. Another two had slightly elevated ratios, but their 11-deoxycortisol response to ACTH stimulation was normal. Therefore, none of these four women fulfilled the biochemical criteria for 11β-hydroxylase deficiency.
In these women, all glucocorticoid and androgen levels (and levels of their intermediate metabolites) decreased after dexamethasone administration. Similar decreases also occurred in the women with deficiencies of 21-hydroxylase and 11β-hydroxylase.
Women with Partial 21-Hydroxylase Deficiency
Fourteen women (8 percent) had increases in their plasma 17-hydroxyprogesterone concentrations in response to ACTH (range, 51.1 to 183.7 nmol per liter) that were more than 2 SD above the mean response in known carriers,11 whereas they had normal cortisol responses. These women were considered homozygous for 21-hydroxylase deficiency. Further evidence in support of the diagnosis was provided by their increased 21-deoxycortisol concentrations (P<0.0005) (Fig. 2Figure 2
Plasma Steroid Concentrations in Women with Nonclassic Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency at Base Line, 60 Minutes after Intravenous Bolus Injection of 0.25 mg of α1–24-ACTH, and after a 48-Hour Course of Dexamethasone (Dex) (0.5 mg Four Times Daily).). In addition, all 14 women had increased 11-deoxycortisol concentrations and increased ratios of 11-deoxycortisol to cortisol (mean [±SD], 25.7±10.2×10–3, vs. 4.5±1.7×10–3 in normal women; P<0.0005) after ACTH stimulation, indicative of partial 11β-hydroxylase deficiency (Fig. 2 and 4). Although the ratios of 17-hydroxyprogesterone to 11-deoxycortisol were also significantly increased (mean, 5.9±3.1, vs. 1.6±1.3 in normal women; P<0.0005), four women had ratios within 2 SD of the mean normal values because they had high 11-deoxycortisol levels. Therefore, this ratio was less useful than the response of 17-hydroxyprogesterone to ACTH stimulation for the diagnosis of 21-hydroxylase deficiency. Plasma 17-hydroxypregnenolone concentrations were markedly elevated after ACTH stimulation in seven of the eight women in whom this steroid was measured; however, the ratios of 17-hydroxypregnenolone to 17-hydroxyprogesterone were significantly lower than normal (0.76±0.38; P<0.0005), as a result of the marked increase in 17-hydroxyprogesterone concentrations.Four women (2 percent) were heterozygous for 21-hydroxylase deficiency. They had peak 17-hydroxyprogesterone responses to ACTH stimulation that ranged from 19.1 to 34.8 nmol per liter, whereas their cortisol responses were normal. Their 11-deoxycortisol concentrations were in the normal range, and the ratio of 11-deoxycortisol to cortisol was significantly elevated only in two.
The women who were homozygous for 21-hydroxylase deficiency and those who were heterozygous had androstenedione, testosterone, and DHEA sulfate concentrations that ranged from normal to very high both basally and after ACTH stimulation.
Five patients and all available family members assumed to carry the allele for 21-hydroxylase deficiency underwent HLA typing (Table 3Table 3
HLA Genotypes in Five Patients with Nonclassic Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency and Their Families.*). Linkage to HLA was demonstrated in all the patients. None of these families were consanguineous, and four family members had haplotypes identical to those of the probands. These four homozygous family members (two men and two women), whose status was previously unknown, were asymptomatic but had significantly increased 17-hydroxyprogesterone concentrations in response to ACTH stimulation. Of the nine family members who were homozygous, seven had HLA antigen B14 and six had HLA-DR1.Women with Partial 11β-Hydroxylase Deficiency
Fourteen women (8 percent) had partial 11β-hydroxylase deficiency. All had increased plasma 11-deoxycortisol concentrations and high ratios of 11-deoxycortisol to cortisol (mean [±SD], 14.8±4.1× 10–3, vs. 4.5±1.7×10–3 in normal women; P<0.0005) after ACTH stimulation (Fig. 3Figure 3
Plasma Steroid Concentrations in Women with Nonclassic Congenital Adrenal Hyperplasia Due to 11β-Hydroxylase Deficiency at Base Line, 60 Minutes after Intravenous Bolus Injection of 0.25 mg of α1–24-ACTH, and after a 48-Hour Course of Dexamethasone (Dex) (0.5 mg Four Times Daily). and 4). Their cortisol concentrations were normal. 11-Deoxycorticosterone concentrations, measured in five patients, were either normal or high. 17-Hydroxypregnenolone concentrations were markedly elevated basally and after ACTH stimulation in 8 of the 10 women in this group in whom this steroid was measured, and 9 also had increased ratios of 17-hydroxypregnenolone to 17-hydroxyprogesterone (mean, 7.1±5.2, vs. 3.1 ±1.2 in normal women; P<0.05). These biochemical measurements indicated the presence of a partial deficiency of 3β-hydroxy-5-steroid dehydrogenase. The concentrations of androgens were either normal or moderately increased.Analysis of Ethnic Data
The ethnic distribution of the entire study group was not different from that of 770,000 age-matched women within the Israeli Jewish population13; however, there was an increased frequency of non-Ashkenazi Jews among the women with 3β-hydroxy-
5-steroid dehydrogenase deficiency (90 percent, P<0.0001) (Table 1). Except for two patients whose parents and father, respectively, were of Ashkenazi origin, the parents of all other patients in this group were non-Ashkenazi Jews. Among the 18 with 3β-hydroxy-5-steroid dehydrogenase deficiency who were non-Ashkenazi Jews, 11 were Asian Jews and 7 were Spanish Jews. On the other hand, the patients with 21-hydroxylase deficiency were mostly Ashkenazi Jews (79 percent, P<0.005). The ethnic distribution of the women with partial 11β-hydroxylase deficiency was similar to that of the population as a whole. Non-Ashkenazi Jews also constituted a higher proportion of the women with high androgen levels and no evidence of a biosynthetic defect (P<0.001).Discussion
We found that women with partial 21-hydroxylase or 11β-hydroxylase deficiency often have a second adrenal enzymatic defect. All 14 women who were homozygous for 21-hydroxylase deficiency also had partial 11β-hydroxylase deficiency. The exact nature of these combined defects, first described by Gandy et al. in 196021 and later by others,22 23 24 25 is unknown, but several possibilities have been considered.24 , 25 On the basis of the recent discovery that the genes encoding 21-hydroxylase and 11β-hydroxylase are located on different chromosomes (6 and 8, respectively),1 it is extremely unlikely that a single woman, much less all the women with 21-hydroxylase deficiency studied, would have the mutant genes for both defects. The hypothesis of a genetically abnormal or aberrant 11β-hydroxylase, with increased affinity for an alternative substrate (17-hydroxyprogesterone), inducing abnormal steroidogenesis24 seems very unlikely too. Such a disorder would presumably represent an allelic variant of 11β-hydroxylase deficiency and should not be linked to the HLA system.26 Yet we and others have shown that such linkage occurs in these patients in a previous study25 and in this study, in which we also found a significant association of HLA antigens B14 and DR1. This association is known to occur in patients with nonclassic 21-hydroxylase deficiency.1 , 6 Partial deficiency of both enzymes can best be explained by the proposition that one defect is congenital (21-hydroxylase) and the other acquired (11β-hydroxylase) as a result of the inhibitory effect of increased androgens on 11β-hydroxylation. This phenomenon has been well documented in animals, both in vitro27 and in vivo.28 29 30 The levels of androgens critical for enzyme inhibition would be determined by the intraadrenal concentrations achieved31 as a result of the primary defect.1 This would explain why increased androgen levels of adrenal but not ovarian origin would inhibit 11β-hydroxylation, even though the peripheral concentrations may be similar in both conditions.32 Furthermore, the degree of inhibition achieved differs for each androgen, androstenedione and testosterone being more potent than DHEA or its sulfate in inhibiting the enzyme.27
The presence of 3β-hydroxy-
5-steroid dehydrogenase deficiency among many of the women with 11β-hydroxylase deficiency may be explained by a similar mechanism. In human adrenal microsomes, 3β-hydroxy-5-steroid dehydrogenase is consistently inhibited by a variety of steroids, in particular by androstenedione, DHEA, progesterone, and 17-hydroxyprogesterone.33 Therefore, in patients with 11β-hydroxylase deficiency who produce many of these substrates in large quantities, the in vivo 3β-hydroxy-5-steroid dehydrogenase activity would be determined not only by the amount of enzyme available, but also by the relative concentrations of the various competing substrates and inhibitory ligands.33 If increased adrenal concentrations of androgens and other steroids are capable of inhibiting 11β-hydroxylase and 3β-hydroxy-5-steroid dehydrogenase, a second defect would also be expected to occur in women with partial 3β-hydroxy-5-steroid dehydrogenase deficiency. As we have mentioned, DHEA, which is the only adrenal androgen produced in excess in 3β-hydroxy-5-steroid dehydrogenase deficiency, is much less potent than androstenedione or testosterone in inhibiting 11β-hydroxylation.27 (The elevated concentrations of androstenedione and testosterone found in this defect probably derive from enhanced DHEA conversion in peripheral tissues.2) This may explain the absence of a secondary defect in patients with such a deficiency.The true incidence of adrenal biosynthetic defects in women with hyperandrogenism is still not well established. Steroid 21-hydroxylase deficiency has been diagnosed by screening such women with an ACTH test. In individual studies during the past decade,2 , 23 , 34 35 36 37 38 39 40 the frequency of the defect ranged from 1 percent to 30 percent. Cumulatively, however, there were 113 cases among 1291 patients — an incidence of 9 percent. Partial 3β-hydroxy-
5-steroid dehydrogenase deficiency has also been described increasingly in recent years.2 , 39 , 41 The frequency of this defect among women with hyperandrogenism has ranged from 12 to 38 percent, the cumulative value being 17 percent (53 cases among 316 women).In our study, the frequency of deficiencies of 3β-hydroxy-
5-steroid dehydrogenase (12 percent) and 21-hydroxylase (8 percent) matched the cumulative values in these other studies. Unexpectedly, we found 11β-hydroxylase deficiency to be as frequent as 21-hydroxylase deficiency. With two exceptions39 , 42 the search for nonclassic adrenal hyperplasia in women with hyperandrogenism has always focused on the presence of deficiencies of 21-hydroxylase, 3β-hydroxy-5-steroid dehydrogenase, or both. Nonetheless, the frequency of 11β-hydroxylase deficiency in this study is similar to that (6.5 percent) in earlier studies39 , 42 in which this defect was specifically sought. It should be pointed out that it is more difficult to identify a defect in 11β-hydroxylase activity than in 21-hydroxylase activity, because the differences in plasma steroid values between normal and affected persons are smaller than the differences between normal subjects and patients with 21-hydroxylase deficiency. Whereas plasma 11-deoxycortisol concentrations increased 2.5-fold to 5-fold after ACTH stimulation in patients with 11β-hydroxylase deficiency, plasma 17-hydroxyprogesterone concentrations rose 10-fold to 45-fold in patients with 21-hydroxylase deficiency. Therefore, the diagnosis was confirmed only when both the plasma levels and the precursor-to-product ratio were significantly elevated.The ethnic origin of the women with 11β-hydroxylase deficiency was similar to that of the general population of women in Israel.13 (This is not true for classic 11β-hydroxylase deficiency, which is very common among Jews of North African origin.43) In contrast, we found a higher frequency of non-Ashkenazi Jews among the women with 3β-hydroxy-
5-steroid dehydrogenase deficiency and a higher frequency of Ashkenazi Jews among the women with 21-hydroxylase deficiency, as compared with the general population. The latter observation has been reported by Speiser et al.,6 although it is not possible to compare the two studies, since their population consisted almost entirely of Ashkenazi Jews (95 percent),12 whereas ours was a random population of women with hyperandrogenism from different ethnic backgrounds.The identification of patients with nonclassic congenital adrenal hyperplasia not only is of academic interest but also has practical, therapeutic implications. Since these patients' basal hormone levels are often within the normal range, an ACTH stimulation test is essential for accurate diagnosis. The administration of dexamethasone can help to distinguish these defects from other causes of hyperandrogenism (e.g., adrenal tumors), although it may also inhibit ovarian production of androgen. It must be emphasized, however, that a search for the various biosynthetic defects must be done in parallel. Finding increased plasma concentrations of 11-deoxycortisol and a high ratio of 11-deoxycortisol to cortisol does not distinguish patients with 11β-hydroxylase deficiency from those with 21-hydroxylase deficiency, if plasma 17-hydroxyprogesterone is not measured at the same time. On the other hand, increased plasma 17-hydroxypregnenolone and DHEA concentrations can be found in all three defects. Their differentiation therefore requires the additional measurement of 17-hydroxyprogesterone and 11-deoxycortisol and calculation of the appropriate ratios.
We conclude that women with nonclassic congenital adrenal hyperplasia have secondary adrenal biosynthetic defects that are due to intraadrenal androgen excess, rather than to dual inherited deficiencies or production of genetically abnormal 11β-hydroxylase. It appears that deficiencies of 3β-hydroxy-
5-steroid dehydrogenase, 21-hydroxylase, and 11β-hydroxylase are not mutually exclusive, so that patients may have more than one such deficiency. Now that the genes encoding these enzymes have been cloned,1 it may be possible for molecular genetic studies to elucidate the exact nature of the primary disorder.Address reprint requests to Dr. Rösler at the Department of Endocrinology and Metabolism, Hadassah—Hebrew University Medical Center, P.O. Box 12000, il-91120 Jerusalem, Israel.
Presented in part at the 71st annual meeting of the Endocrine Society, Seattle, June 21 to 24, 1989.
We are indebted to Professor Chaim Brautbar from the Department of Immunohematology, Hadassah—Hebrew University Medical Center, for HLA determinations; to Dr. Einat Goberman for advice and assistance with the statistical analysis; and to Mrs. Nina Weshler for excellent technical assistance.
Source Information
From the Departments of Obstetrics and Gynecology (T.E.-G., A.H., Z.P., A.M.) and Endocrinology and Metabolism (A.R.), Hadassah—Hebrew University Medical Center, Jerusalem, and the Department of Pharmacology, University of Heidelberg, Heidelberg, West Germany (P.V.).
- References
References
1
New MI, White PC, Pang S, Dupont B, Speiser PW. The adrenal hyperplasias. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic basis of inherited disease. 6th ed. Vol. 2. New York: McGraw-Hill, 1989:1881–917.
2
Pang
CrossRef | Web of Science | Medline3
Gabrilove
Full Text | Web of Science | Medline4
Cathelineau
CrossRef | Web of Science | Medline5
Guthrie
CrossRef | Web of Science | Medline6
Speiser
Web of Science | Medline7
Rittmaster
Web of Science | Medline8
Ferriman
CrossRef | Web of Science | Medline9
Speroff L, Glass RH, Kase NG, eds. Clinical gynecologic endocrinology and infertility. 4th ed. Baltimore: Williams & Wilkins, 1989:445–59.
10
Idem. Clinical gynecologic endocrinology and infertility. 4th ed. Baltimore: Williams & Wilkins, 1989:216–32.
11
New
CrossRef | Web of Science | Medline12
Goodman RM. Genetic disorders among the Jewish people. Baltimore: Johns Hopkins University Press, 1979:1–26.
13
Statistical abstracts of Israel. No. 38. Jerusalem: Hamakor Press, Ltd., Central Bureau of Statistics, 1987:58–9, 68–9.
14
Rösler
CrossRef | Web of Science | Medline15
Milewicz
CrossRef | Medline16
Abraham
CrossRef | Web of Science | Medline17
Schumert
CrossRef | Web of Science | Medline18
Rösler
CrossRef | Medline19
Bodmer W, Tripp M, Bodmer J. Application of the fluorochromatic cytotoxicity test assay to leukocyte typing. In: Curtoni ES, Mattiuz PL, Tosi RM, eds. Histocompatibility testing 1967. Copenhagen: Munksgaard, 1967:341–50.
20
Cochran WG. Sampling techniques. 3rd ed. New York: John Wiley, 1977:50–68.
21
Gandy
CrossRef | Web of Science | Medline22
Newmark
Medline23
Blankstein
CrossRef | Web of Science | Medline24
Finkelstein
CrossRef | Medline25
Hurwitz
CrossRef | Web of Science | Medline26
Brautbar
Web of Science | Medline27
Sharma
Web of Science28
Fragachan
CrossRef | Web of Science | Medline29
Colby
CrossRef | Web of Science | Medline30
McCall
CrossRef | Web of Science | Medline31
Dickerman
CrossRef | Web of Science | Medline32
Parker
Web of Science | Medline33
Byrne
CrossRef | Web of Science | Medline34
Chrousos
Web of Science | Medline35
Carmina
Web of Science | Medline36
Chetkowski
CrossRef | Web of Science | Medline37
Kuttenn
Full Text | Web of Science | Medline38
Benjamin
Web of Science | Medline39
Lucky
CrossRef | Web of Science | Medline40
Azziz
CrossRef | Web of Science | Medline41
Lobo
CrossRef | Web of Science | Medline42
Bouallouche
Web of Science | Medline43
Rösler A, Leiberman E. Enzymatic defects of steroidogenesis: 11β-hydroxylase deficiency congenital adrenal hyperplasia. In: New MI, Levine LS, eds. Adrenal diseases in childhood. Vol. 13 of Pediatric and adolescent endocrinology. Basel: S. Karger, 1984:47–71.
- Citing Articles (40)
Citing Articles
1
V. Deckwart, K. Diedrich, N. Bündgen, D.A. Beyer. (2011) Bedeutung der Late-onset-Form des adrenogenitalen Syndroms in der Kinderwunschbehandlung. Gynäkologische Endokrinologie 9:2, 93-96
CrossRef2
SelmaFeldman Witchel, Ricardo Azziz. (2010) NonClassic Congenital Adrenal Hyperplasia. International Journal of Pediatric Endocrinology 2010:1, 625105
CrossRef3
2009. Adrenal Gland. , 277-316.
CrossRef4
Holger S. Willenberg, Maryam Bahlo, Matthias Schott, Tina Wertenbruch, Joachim Feldkamp, Werner A. Scherbaum. (2008) Helpful diagnostic markers of steroidogenesis for defining hyperandrogenemia in hirsute women. Steroids 73:1, 41-46
CrossRef5
Manuel Luque-Ramírez, José Luis San Millán, Héctor F. Escobar-Morreale. (2006) Genomic variants in polycystic ovary syndrome. Clinica Chimica Acta 366:1-2, 14-26
CrossRef6
2005. Hypertension (High Blood Pressure). .
CrossRef7
2005. Androgens. .
CrossRef8
2005. Androgens. .
CrossRef9
2005. Anticoagulants. .
CrossRef10
John R.D Stalvey. (2002) Inhibition of 3β-hydroxysteroid dehydrogenase-isomerase in mouse adrenal cells: a direct effect of testosterone. Steroids 67:8, 721-731
CrossRef11
Songya Pang. (2001) CONGENITAL ADRENAL HYPERPLASIA OWING TO 3β-HYDROXYSTEROID DEHYDROGENASE DEFICIENCY. Endocrinology & Metabolism Clinics of North America 30:1, 81-99
CrossRef12
C MORAN, R AZZIZ. (2001) The Role Of The Adrenal Cortex In Polycystic Ovary Syndrome. Obstetrics and Gynecology Clinics of North America 28:1, 63-75
CrossRef13
Jean Fiet, Frank Giton, Ahmed Boudi, Philippe Boudou, Hany Soliman, Jean-Marie Villette, Hervé Galons. (2001) Plasma 17-OH pregnenolone: comparison of a time-resolved fluoroimmunoassay using a new tracer 17-OH pregnenolone-3-oxyacetyl-biotine with a radioimmunoassay using 125I 17-OH pregnenolone-3-hemisuccinate-histamine. Steroids 66:2, 81-86
CrossRef14
Josefina Romaguera, Carlos Moran, Teresa P Diaz-Montes, Gene A Hines, Rosa I Cruz, Ricardo Azziz. (2000) Prevalence of 21-hydroxylase–deficient nonclassic adrenal hyperplasia and insulin resistance among hirsute women from Puerto Rico. Fertility and Sterility 74:1, 59-62
CrossRef15
Melanie Blackless, Anthony Charuvastra, Amanda Derryck, Anne Fausto-Sterling, Karl Lauzanne, Ellen Lee. (2000) How sexually dimorphic are we? Review and synthesis. American Journal of Human Biology 12:2, 151-166
CrossRef16
Ricardo Azziz, Luis A Hincapie, Eric S Knochenhauer, Didier Dewailly, Liesl Fox, Larry R Boots. (1999) Screening for 21-hydroxylase–deficient nonclassic adrenal hyperplasia among hyperandrogenic women: a prospective study. Fertility and Sterility 72:5, 915-925
CrossRef17
Héctor F Escobar-Morreale, José L San Millán, Rhonda R Smith, José Sancho, Selma F Witchel. (1999) The presence of the 21-hydroxylase deficiency carrier status in hirsute women: phenotype-genotype correlations. Fertility and Sterility 72:4, 629-638
CrossRef18
Alan R Jacobs, Phyllis B Edelheit, Anton E Coleman, Andrew G Herzog. (1999) Late-onset congenital adrenal hyperplasia: a treatable cause of anxiety. Biological Psychiatry 46:6, 856-859
CrossRef19
Carlos Moran, H.Downing Potter, Rosario Reyna, Larry R. Boots, Ricardo Azziz. (1999) Prevalence of 3β-hydroxysteroid dehydrogenase–deficient nonclassic adrenal hyperplasia in hyperandrogenic women with adrenal androgen excess. American Journal of Obstetrics and Gynecology 181:3, 596-600
CrossRef20
Robert L. Rosenfield. (1999) OVARIAN AND ADRENAL FUNCTION IN POLYCYSTIC OVARY SYNDROME. Endocrinology & Metabolism Clinics of North America 28:2, 265-293
CrossRef21
Peter R. Garner. (1998) Congenital adrenal hyperplasia in pregnancy. Seminars in Perinatology 22:6, 446-456
CrossRef22
Rossella Valentino, Antonio Pasquale Tommaselli, Silvia Savastano, Maurizio Dorato, Riccardo Rossi, Gaetano Lombardi. (1997) Different dysregulations in adrenal steroid biosynthesis as a prevalent cause of hyperandrogenism in women from southern Italy. Fertility and Sterility 68:2, 236-241
CrossRef23
H.F. Escobar-Morreale, J. Serrano-Gotarredona, R. Garcia-Robles, J. Sancho, C. Varela. (1997) Mild adrenal and ovarian steroidogenic abnormalities in hirsute women without hyperandrogenemia: Does idiopathic hirsutism exist?. Metabolism 46:8, 902-907
CrossRef24
Talia Eldar-Geva, Ehud J. Margalioth, Baruch Brooks, Nurit Algur, Michael Gal, Edith Zylber-Haran, Ilya Bar, Yoram Z. Diamant. (1997) Elevated serum progesterone levels during pituitary suppression may signify adrenal hyperandrogenism. Fertility and Sterility 67:5, 959-961
CrossRef25
Stefan A. Wudy, Janos Homoki, Walter M. Teller. (1994) Successful prenatal treatment of congenital adrenal hyperplasia due to 21-hydroxylase deficiency. European Journal of Pediatrics 153:8, 556-559
CrossRef26
Aarno Hautanen, Matti Mänttäri, Vesa Manninen, Leena Tenkanen, Jussi K. Huttunen, M. Heikki Frick, Herman Adlercreutz. (1994) Adrenal androgens and testosterone as coronary risk factors in the Helsinki Heart Study. Atherosclerosis 105:2, 191-200
CrossRef27
D.M. Cathro,, S.G. Golombek,. (1994) Non-Classical 3ß-Hydroxysteroid Dehydrogenase Deficiency in Children in Central Iowa. Difficulties in Differentiating this Entity from Cases of Precocious Adrenarche Without an Adrenal Enzyme Defect. Journal of Pediatric Endocrinology and Metabolism 7:1, 19-32
CrossRef28
William Jeffcoate. (1993) The treatment of women with hirsutism. Clinical Endocrinology 39:2, 143-150
CrossRef29
MICHEL PUGEAT, MARIE HÉLÉNE NICOLAS, JEAN CHARLES CRAVES, CHRISTINE ALVARADO-DUBOST, SYLVIE FIMBEL, HENRI DÉCHAUD, HERVÉ LEJEUNE. (1993) Androgens in Polycystic Ovarian Syndrome. Annals of the New York Academy of Sciences 687:1 Intraovarian, 124-135
CrossRef30
Phyllis W Speiser, Perrin C White, Maria I New. (1993) Congenital adrenal hyperplasia. Reproductive Medicine Review 2:01, 1
CrossRef31
Stephen F.W. Cavanah, Robert F. Dons. (1993) Partial 3β-hydroxysteroid dehydrogenase deficiency presenting as new-onset gynecomastia in a eugonadal adult male. Metabolism 42:1, 65-68
CrossRef32
Fernand Labrie, Jacques Simard, Van Luu-The, Alain Bélanger, Georges Pelletier. (1992) Structure, function and tissue-specific gene expression of 3β-hydroxysteroid dehydrogenase/5-ene-4-ene isomerase enzymes in classical and peripheral intracrine steroidogenic tissues. The Journal of Steroid Biochemistry and Molecular Biology 43:8, 805-826
CrossRef33
M. A. ARNAOUT. (1992) Late-onset congenital adrenal hyperplasia in women with hirsutism. European Journal of Clinical Investigation 22:10, 651-658
CrossRef34
Ehrmann, David A., Rosenfield, Robert L., Barnes, Randall B., Brigell, Deborah F., Sheikh, Zubie, . (1992) Detection of Functional Ovarian Hyperandrogenism in Women with Androgen Excess. New England Journal of Medicine 327:3, 157-162
Full Text35
Elaine I. Turner, Malcolm J. Watson, Leslie A. Perry, Michael C. White. (1992) Investigation of adrenal function in women with oligomenorrhoea and hirsutism (clinical PCOS) from the north-east of England using an adrenal stimulation test. Clinical Endocrinology 36:4, 389-397
CrossRef36
F. Labrie, J. Simard, V. Luu-The, G. Pelletier, A. Bélanger, Y. Lachance, H.F. Zhao, C. Labrie, N. Breton, Y. de Launoit, M. Dumont, E. Dupont, E. Rhéaume, C. Martel, J. Couët, C. Trudel. (1992) Structure and tissue-specific expression of 3ß-hydroxysteroid dehydrogenase/5-ene-4-ene isomerase genes in human and rat classical and peripheral steroidogenic tissues. The Journal of Steroid Biochemistry and Molecular Biology 41:3-8, 421-435
CrossRef37
Irene N. Sills, Robert Rapaport, Kathryn A. Skuza, Mary N. B. Horlick. (1992) 46,XX pure gonadal dysgenesis with growth hormone deficiency and impaired 3β-hydroxysteroid dehydrogenase activity. American Journal of Medical Genetics 42:1, 100-103
CrossRef38
YVES LACHANCE, VAN LUU-THE, HUGUES VERREAULT, MARTINE DUMONT, ERIC RHÉAUME, GILLES LEBLANC, FERNAND LABRIE. (1991) Structure of the Human Type II 3β-Hydroxysteroid Dehydrogenase/Δ 5 -Δ 4 Isomerase (3β-HSD) Gene: Adrenal and Gonadal Specificity. DNA and Cell Biology 10:10, 701-711
CrossRef39
JoeLeigh Simpson, Sara Mccluskey, J.Malcolm Pearce, J.Hubert Lacey. (1991) Polycystic ovaries. The Lancet 337:8732, 53
CrossRef40
Ehrmann, David A., Rosenfield, Robert L., . (1990) Hirsutism — Beyond the Steroidogenic Block. New England Journal of Medicine 323:13, 909-911
Full Text
