| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Endocrinology (V.T.-F., L.M.R.-N., I.T.N.V., C.E.K.), Department of Medicine, Universidade Federal de Sao Paulo, 04039-032 Sao Paulo, Brazil; Pediatric Endocrinology Service (V.T.-F.), Hospital Infantil Darcy Vargas 05614-040 Sao Paulo, Brazil; Division of Pediatric Endocrinology (S.H.V.L.-M.), Department of Pediatrics, Universidade de Campinas, 13083-100 Campinas, Sao Paulo, Brazil; and Division of Pediatric Endocrinology (H.K.), Instituto da Criança, Universidade de Sao Paulo, 05403-900 Sao Paulo, Brazil
Address all correspondence and requests for reprints to: Claudio E. Kater, M.D., Steroids Laboratory, Division of Endocrinology, Universidade Federal de Sao Paulo, Rua Pedro de Toledo, 781-13 andar, SP 04039-032 Sao Paulo, Brazil. E-mail: kater{at}endocrino.epm.br.
| Abstract |
|---|
|
|
|---|
Materials and Methods: In 133 patients (89 females/44 males; 10 d-20.9 yr) with alleged classic 21OHD and five (three females/two males; 7.321 yr) with documented 11ßOHD, we measured serum 21-deoxycortisol (21DF), 17-hydroxyprogesterone (17OHP), and 11-deoxycortisol (S), 48 h after glucocorticoid withdrawal. We also studied 20 sex- and age-matched control subjects. Serum steroid levels were determined by RIA after HPLC purification.
Objectives: The objectives of this study were to: 1) quantify 21DF in patients with congenital adrenal hyperplasia, 2) correlate hormonal with clinical data, and 3) identify possible misdiagnosed patients with 11ßOHD among those with 21OHD.
Results: In 21OHD, 17OHP (217100,472 ng/dl) and 21DF (<3914,105 ng/dl) were mostly elevated and positively correlated (r = 0.7202; P < 0.001). Except for higher 17OHP in pubertal patients, 17OHP and 21DF values were similar according to sex, disease severity, or prevailing glucocorticoid dose. One additional patient with 11ßOHD was detected (1%) and also one with apparent combined 11ß- and 21OHD. S levels were elevated in 11ßOHD and normal but significantly higher in 21OHD than in controls.
Conclusion: To recognize patients with 21- and/or 11ßOHD, we recommend evaluation of 17OHP or 21DF and S. Also, 21DF may be useful to follow up pubertal patients with 21OHD. Because 1% of patients with alleged 21OHD may have 11ßOHD, its frequency seems underestimated, as per our experience in a Brazilian population.
| Introduction |
|---|
|
|
|---|
Because serum 17-hydroxyprogesterone (17OHP) is distinctly elevated in 21OHD, it stands as the classical steroid marker for this disease. Deficiency in the next step for F biosynthesis, 11ß-hydroxylase deficiency (11ßOHD), also allows 17OHP to accumulate, but in concert with the immediate precursors 11-deoxycortisol (S) and 11-deoxycorticosterone. The resultant clinical picture in 11ßOHD is similar to that of 21OHD, except for the variable presence of hypertension and hypokalemia due to 11-deoxycorticosterone excess. Although uncommon, probably accounting for less than 5% of all cases (3), the prevalence of 11ßOHD may go underestimated. Because biochemical confirmation of clinically suspected CAH is routinely investigated by checking serum 17OHP alone, a subset of patients whose primary diagnosis would be 11ßOHD may be mistakenly included among the larger population due to 21OHD, whenever S is not measured.
Due to the marked buildup of 17OHP in 21OHD, this precursor steroid can proceed directly to 11-hydroxylation, producing distinct elevations of 21-deoxycortisol (21DF), an additional marker of the disease (4, 5). Although greatly activated in 21OHD, this alternate pathway is negligible in normal subjects and virtually nonfunctional in 11ßOHD, supporting its use to discriminate between both conditions. However, serum or plasma assays for 21DF are not commercially available, and only a few research centers test for 21DF in diagnosing CAH (4, 5, 6, 7, 8). We have recently developed an adapted RIA for 21DF preceded by steroid separation by HPLC (9, 10).
In the present study, we evaluated 138 patients with CAH: 133 with a presumptive diagnosis of classic 21OHD (15 untreated), based exclusively on previously elevated levels of 17OHP and androgens, five with documented 11ßOHD, and 20 control subjects, measuring morning serum 21DF, 17OHP, and S to: 1) quantify serum levels of 21DF in treated and untreated patients with CAH and normal subjects, 2) correlate hormonal with relevant clinical parameters in the patients, and 3) identify possible misdiagnosed patients with 11ßOHD among the vast group with alleged 21OHD.
| Subjects and Methods |
|---|
|
|
|---|
A total of 138 patients with CAH were studied and were seen at four different endocrine pediatric services in the State of Sao Paulo, Brazil; 118 had been followed up with a presumed diagnosis of classic 21OHD, and 15 were either recently diagnosed and untreated or off medications for more than 3 months; another five had the 11ßOHD form of CAH. Diagnosis of 21OHD was established on the basis of previously elevated serum levels of 17OHP and androgens; 89 were female, and 44 were male, age ranging from 10 d to 20.9 yr (median of 8.4 yr); 55 (41%) were simple virilizers (SVs), and 78 (59%) were salt wasters (SWs), as defined by increased plasma renin activity and/or electrolyte abnormalities and/or the presence of clinical manifestations of sodium loss. Diagnosis in all five 11ßOHD patients (three females/two males, 7.321 yr) was established by elevated serum levels of S in addition to androgens and confirmed genetically in all (11).
Before the study period, 118 patients with 21OHD and all five with 11ßOHD were on regular replacement therapy with glucocorticoid (GC); 21OHD patients also received mineralocorticoid. For those 31 receiving GC other than hydrocortisone (HC), we calculated for statistical purposes the dose equivalencies as follows: 15 mg HC = 1 mg prednisone or prednisolone = 0.2 mg dexamethasone (12, 13). Fludrocortisone was given in a standard dose of 0.050.1 mg/d.
Twenty control subjects (eight females/12 males; 0.315.2 yr, at their expected sexual development stages) were also studied: four were normal infants, and in 16, investigation for short stature or other unrelated medical condition had unremarkable results.
All patients, parents or legal guardians signed a written consent form after having the purposes of the protocol, which had been previously approved by the Ethics Committees of all centers involved in the study, thoroughly explained to them.
Study protocol
This is a cross-sectional study in which patients with CAH on regular treatment were oriented to discontinue their medications 48 h before blood collection to permit partial reemergence of the abnormal serum steroid profile. A single blood sample for steroid determinations was drawn between 0730 and 0830 h in all 30-min recumbent and 10-h fasting patients with alleged 21OHD and 11ßOHD and control subjects. All samples were centrifuged at room temperature within 2 h after collection, and serum was separated and stored at 20 C until processing.
Steroid assays
Serum concentrations of 17OHP (14), 21DF (9, 10), and S (15) were determined by slight modifications of previously reported routine in-house RIAs, with assay sensitivities of 11, 39, and 24 ng/dl, respectively. Before RIA, all samples were submitted to ether extraction and separation by HPLC, as previously reported (10). Serum steroid concentrations were calculated by RIACALC software (Wallac Oy, Turku, Finland), and the results were corrected for the initial volume and subsequent dilutions.
Anthropometry
Bone age was evaluated according to the Greulich and Pyle Atlas (16). Z scores for height and weight were calculated from the reference database for the Brazilian population (17), and body mass index was calculated by the Quetelet method (18). Pubertal development was ascertained according to Marshall and Tanners method (19, 20): girls were distributed according to breast development (B1B5) and boys according to development of the genitalia (G1G5). B1 and G1 patients are referred to herein as prepubertal and all other categories (B2B5 and G2G5) as pubertal.
Statistical analysis
GraphPad Prism version 2.0 program (GraphPad Software Inc., San Diego, CA) was used for calculations. The Kolmogorov-Smirnov test was used to verify whether serum steroid values had normal distributions. The Spearman test was used to tentatively correlate serum levels of 17OHP and 21DF with clinical parameters in patients and controls. The nonparametric ANOVA of Kruskall-Wallis, complemented by the Mann-Whitney test, was used to compare patients and controls separated according to chronological age (CA) and prevailing GC doses. For statistical purposes, all steroid values below the sensitivity limit for the particular assay were arbitrarily considered equal to the detection level divided by the square root of 2 (therefore defining an average number for all undetectable values).
| Results |
|---|
|
|
|---|
|
In the remaining 131 patients with confirmed 21OHD (either untreated or previously treated), basal serum levels of 17OHP and 21DF were elevated and widely distributed around median values: 8260 (217100,472) and 810 (2814,105) ng/dl, respectively (Table 1
). All 131 pairs of 17OHP and 21DF values were positively and significantly correlated (r = 0.7202; P < 0.001). This correlation was maintained when patients were subgrouped by sex (female, r = 0.7289; male, r = 0.7046), disease severity (SW, r = 0.7483; SV, r = 0.6515), or pubertal status (prepubertal, r = 0.8611; pubertal, r = 0.5808). In this group of 21OHD patients, serum S levels were peculiarly in the upper normal range or slightly above it and significantly higher than those obtained in control subjects (71 vs. 32 ng/dl; P < 0.001).
When 21OHD patients were subgrouped according to sex or disease severity, no significant differences in 17OHP or 21DF levels were observed between female and male and SW and SV. However, when pubertal patients were compared with prepubertal patients, significantly elevated levels of 17OHP were observed in the former (10,397 vs. 5231 ng/dl; P < 0.005), a fact not observed with 21DF (810 vs. 779 ng/dl; not significant) (Fig. 1
).
|
|
Although 17OHP levels were normal to moderately elevated (50466 ng/dl) in all six patients with 11ßOHD (Table 1
), 21DF levels ranged from undetectable to normal (<3968 ng/dl). Serum S levels in three of six patients were characteristically elevated (3537, 6088, and 6357 ng/dl), in contrast to the normal values seen in the other three patients (31, 34, and 64 ng/dl); at diagnosis, however, their levels were unquestionably elevated (4510, 8755, and 6027 ng/dl, respectively).
In control subjects, the normal serum levels of 17OHP (Table 1
) exhibited a significant correlation with CA (r = 0.7171; P < 0.0001). Although there were no differences according to sex, pubertal controls (n = 9) had significantly higher levels of 17OHP than prepubertal (n = 11) (119 vs. 67 ng/dl; P < 0.005). Serum levels of 21DF ranged from less than 39 to 72 ng/dl.
| Discussion |
|---|
|
|
|---|
The clinical features of 21OHD and 11ßOHD are identical, except for the inconsistent presence of hypertension and hypokalemia in the latter (22). However, a substantial number of patients with confirmed 11ßOHD had never had their blood pressure or electrolytes checked at diagnosis, and even if so, up to 35% of them may have been normotensive and normokalemic (23, 24). A precise pretreatment diagnosis is essential for proper replacement therapy and genetic counseling. It is also crucial to alert patients with 11ßOHD of their higher risk for cardiovascular morbidity (23, 24, 25, 26). Diagnostic confirmation must then rely on the evaluation of a serum steroid profile, identifying hormonal markers for each condition.
In the present study, a distinct adrenal steroid panel was evaluated in a large series of patients seen at four different pediatric endocrinology centers, where they had been followed with a presumptive diagnosis of classic 21OHD; none had 21DF or compound S ever measured. The short-term (48 h) discontinuation of replacement therapy in these patients was planned to lessen the possible GC influence upon the hypothalamic-pituitary-adrenal (HPA) axis, allowing partial restoration of the pretreatment steroid pattern while maintaining the study on stringent ethical grounds.
The serum concentrations of 21DF, 17OHP, and S obtained in the conditions of the present study (Table 1
) were similar to those previously reported in CAH (4.68.27). In patients with 21OHD, basal values of 17OHP and 21DF vary from normal to extremely elevated, ranging across a spectrum of three orders of magnitude, and both were significantly correlated, confirming their common origin in the adrenal cortex (4, 28).
There were no differences for both steroids regarding sex. However, expected differences due to disease severity (SW vs. SV) were not observed, possibly because of the presumed attenuation of the HPA axis and adrenal suppression, as a result of long-term therapy. In this setting, 48 h off medications proved insufficient to produce resurgence of the original hormonal pattern in a significant number of patients. This could also be supported by the evidence of a gradual decrease in serum levels of 17OHP and 21DF along with the GC dose being used. Untreated patients and those receiving small doses of GC still had conceivably higher levels of both steroids, which were close to normality when optimal replacement doses (1020 mg/m2·d) were given. Conversely, the higher levels of 17OHP (but not 21DF) observed when patients were on the highest and usually suppressible doses (2030+ mg/m2·d) were paradoxical. Doses up to that level are not usually necessary to treat patients with CAH and may cause cushingoid effects (29), and because they were prescribed, a question of compliance should then be raised (30). Also, patients in this group were older (median age, 12.5 yr) and almost 80% pubertal, indicating an additional GC-independent gonadal contribution for the circulating levels of 17OHP.
By the same token, the initially elevated levels of 17OHP and 21DF in patients within the first 2 yr of life and treatment are probably a remnant of the still recent untreated period, whereas similar elevated levels of 17OHP but not 21DF in the older group reflect participation of the extraadrenal source. This disproportionate elevation of 17OHP levels seems to reflect a combination of reduced adherence to treatment in this period of life and a substantial 17OHP contribution from the gonads, especially the ovaries (31), which may be exquisitely sensitive to the combined stimulation of ACTH and LH in the adolescence. Also, because androgen levels increase briskly in this scenario, it may be responsible for the inhibition of 11ß-hydroxylase in the adrenal cell mitochondria, acting as pseudosubstrates and reducing its efficiency. This phenomenon is also observed with androgen-producing adrenocortical carcinomas (32). In the control group, 17OHP levels were also significantly higher in pubertal than in prepubertal children, reinforcing the gonadal contribution to the circulating levels (31, 33). Similar observations were confirmed by others in normal pubertal children (34) and in adult women during the luteal phase of the menstrual cycle (35).
Because 11ß-hydroxylation is confined to the adrenal cortex, in contrast to 21-hydroxylase, 21DF does not follow the same secretory pattern of 17OHP; its levels do not change with either puberty or the menstrual cycle (7, 8, 34). Thus, in the pubertal patient with 21OHD, 21DF may be useful for therapeutic decisions.
Considering that 17OHP could also be moderately elevated in 11ßOHD, additional laboratory clues must be employed to differentiate both entities. Compound S is the key steroid for that matter because it is characteristically elevated in 11ßOHD but not in 21OHD. Three of six patients with 11ßOHD in this study (Table 1
) had typically elevated S levels, together with moderately elevated 17OHP; however, in the other three patients, both S and 17OHP levels were normal, suggesting adrenal suppression upon treatment. This could be related to continued therapy, despite the recommended 48-h pause or, alternatively, to an exquisite sensitivity of the HPA axis to replacement doses of GC in these patients. Nevertheless, all 11ßOHD patients had their diagnosis confirmed by elevated pretreatment S levels and by molecular genetics (11), including the misdiagnosed case of this series.
Two patients, in whom a diagnosis of 11ßOHD was initially considered because of typically elevated levels of S, were identified in our series of alleged 21OHD. The first was a clear-cut case, whose 17OHP and S, levels were 410 and 3537 ng/dl, respectively, in the face of undetectable 21DF. This 2-yr-old patient was normotensive and normokalemic at diagnosis but later became hypertensive (blood pressure, 140/95 mm Hg). The second, however, was a salt-wasting female newborn in whom all three steroids were extremely elevated (Table 1
), suggesting an apparent combined 21OHD and 11ßOHD but whose genotyping confirmed the diagnosis of 21OHD only. Although uncommon and challenging in the light of current biochemical and molecular evidence, patients with such a steroid pattern have been occasionally reported in the literature (6, 36, 37, 38, 39).
Because 21DF is an 11ßOH derivative of 17OHP, its serum levels are parallel and proportionally elevated in patients with 21OHD but decreased or undetectable in those with 11ßOHD. Also, because the possibility of misdiagnosing 11ßOHD among patients with 21OHD is at the level of 1%, based on the present large series and within the study conditions, we recommend a careful steroid evaluation based on the association of serum 17OHP and/or 21DF with compound S, which also allows for detection of cases of apparent combined 21OHD and 11ßOHD. Furthermore, because of its exclusive adrenal source (as opposed to 17OHP), 21DF may be useful, if not preferential, for the evaluation of pubertal patients with 21OHD.
| Acknowledgments |
|---|
| Footnotes |
|---|
Members of the Brazilian CAH Multicenter Study Group are: Gil Guerra-Junior, Maria Tereza Matias Baptista, Nuvarte Setian, Durval Damiani, Margaret de Castro, and Tânia Bachega.
First Published Online March 21, 2006
Abbreviations: CA, Chronological age; CAH, congenital adrenal hyperplasia; 21DF, 21-deoxycortisol; F, cortisol; GC, glucocorticoid; HC, hydrocortisone; HPA, hypothalamic-pituitary-adrenal; 11ßOHD, 11ß-hydroxylase deficiency; 21OHD, 21-hydroxylase deficiency; 17OHP, 17-hydroxyprogesterone; S, 11-deoxycortisol; SV, simple virilizer; SW, salt waster.
Received August 23, 2005.
Accepted March 10, 2006.
| References |
|---|
|
|
|---|
-hydroxyprogesterone: methodological considerations]. Arq Bras Endocrinol Metab 24:2430 (Portuguese)Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |