| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Internal Medicine and Endocrine & Metabolic Sciences, University of Perugia (S.L., F.C., G.C., G.A., P.B., F.S., A.F.), I-06126 Perugia, Italy; INSERM U-342, Hôpital Saint Vincent de Paul, Faculté Cochin, Université René Descartes (P.A., F.R.), Paris, France; and Department of Medicine, University of Washington (Å.L.), Seattle, Washington 98195-7710
Address all correspondence and requests for reprints to: Stefano Laureti, Department of Internal Medicine and Endocrine & Metabolic Sciences, University of Perugia, Via E. Dal Pozzo, 06126 Perugia, Italy. E-mail: laureti{at}dimisem.med.unipg.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adrenal insufficiency can be the consequence of the autoimmune destruction of adrenal cortical cells. Autoimmune adrenal insufficiency is associated with some human histocompatibility leukocyte antigen (HLA) haplotypes, namely HLA-DR3/DQ2 (5), and the appearance of circulating adrenal cell autoantibodies (ACA) (6, 7) that are considered a sensitive and specific immune marker (3, 8, 9, 10). However, in spite of their high diagnostic sensitivity, 10–30% of idiopathic patients cannot be accounted for by ACA (6, 7). The identification of a major adrenal autoantigen as steroid-21-hydroxylase (21OHAb) (11, 12) recently enabled the development of a sensitive radiobinding assay for 21OHAb, which was shown to have a high diagnostic sensitivity (13) and specificity for adrenal insufficiency (14). It has also been proposed that steroid-17-hydroxylase (17OH) and side-chain cleavage enzyme would be targets of autoantibodies in the autoimmune polyglandular syndrome type I (APS I) (15, 16). However, the sensitivity of 17OH or side-chain cleavage autoantibodies seems to be very low in patients with APS II or isolated autoimmune adrenal insufficiency (10, 17). Accordingly, 17OH and side-chain cleavage autoantibodies do not appear to be useful for a differential etiological diagnosis of primary adrenal insufficiency.
Other forms of adrenal insufficiency may result from targeting of the adrenal glands by chronic infection, such as tuberculosis (4) or fungal infection (18), adrenal hemorrhage (4, 19), cancer (20), or from several recently identified genetic disorders. These include congenital lipoid adrenal hyperplasia (21), adrenal hypoplasia congenita (22), triple A syndrome (23), and adrenoleukodystrophy (ALD) (24, 25, 26, 27). ALD is a hereditary peroxisomal disorder characterized by progressive demyelination within the central nervous system and adrenal insufficiency (24, 25), and is the only genetic disorder that causes primary adrenal insufficiency in adults (26, 28). Half of the phenotypic forms of ALD affects adults between 20 and 40 yr of age and is characterized by progressive spastic paraplegia. In many cases, clinical signs of adrenal insufficiency precede the appearance of neurological symptoms by several years or decades. In all cases, the demonstration of high plasma levels of very long chain fatty acids (VLCFA) allows an unambiguous diagnosis of affected males.
Because an effective protocol for the etiological diagnosis of primary adrenal insufficiency is needed to guarantee proper management of the patient, we retrospectively evaluated the diagnostic accuracy of a combination of immune (ACA, 21OHAb) and biochemical (VLCFA) markers in the total population of patients with primary adrenal insufficiency attending the Department of Internal Medicine and Endocrine and Metabolic Sciences in Perugia, Italy. In addition, we also studied a large group of French patients with adrenal insufficiency caused by ALD. Based on the results of our study, a simple diagnostic flowchart that allows an accurate etiological diagnosis of primary adrenal insufficiency was developed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A total of 56 patients with primary adrenal insufficiency (27
males and 29 females, median age 55 yr and range 15–84 yr) (Table 1
) attended the Department of Internal
Medicine and Endocrine and Metabolic Sciences in Perugia, Italy,
between 1992 and 1996. The patients had a disease duration ranging
between 0 and 49 yr. All the patients underwent routine analyses to
evaluate the function of adrenal cortex (such as plasma levels of
cortisol, aldosterone, renin activity, and ACTH), and all patients were
treated with cortisone acetate 25–50 mg/day. A total of 28 patients
was also treated with fludrocortisone 0.05–0.1 mg/day. Serum samples
were collected from all the 56 patients at the time of the first visit
and stored at -20 C until they were used to evaluate levels of ACA and
21OHAb, as well as of pancreatic islet autoantibodies [islet cell
antibodies (ICA), glutamic acid decarboxylase autoantibodies (GAD65Ab),
IA-2/ICA512 autoantibodies (ICA512Ab), and thyroid autoantibodies
(thyroid peroxidase autoantibodies, TPOAb)], because insulin-dependent
diabetes mellitus (IDDM) and thyroid diseases can frequently occur in
patients with autoimmune adrenal insufficiency. In addition, fasting
plasma samples from all the 56 patients were used to determine VLCFA
concentration. Adrenal imaging (X-ray and/or computerized tomography
scan) was performed in all patients found negative for adrenal
autoantibodies and in 15 patients with adrenal autoantibodies.
|
We also studied a population of 24 French male patients with adrenal insufficiency caused by ALD (duration of adrenal insufficiency at sampling: median 4 yr, range: 0.5–15 yr) (age at sampling: median 18 yr, range 3–40 yr). In these latter 24 patients, diagnosis of ALD was made at the Hôpital Saint-Vincent de Paul, Paris, France, between 1989 and 1995, according to plasmatic VLCFA concentrations. Serum samples from all the 24 ALD patients were used to evaluate levels of ACA, 21OHAb, ICA, GAD65Ab, ICA512Ab, and TPOAb.
Antibody testing
ACA. ACA were determined using an indirect immunofluorescence method on cryostatic sections of bovine adrenal glands (29, 30). Levels of ACA were expressed as end point dilution titer.
21OHAb. 21OHAb were determined in a radiobinding assay using in vitro translated recombinant human [35S]21OH (13). 21OHAb levels were expressed as a relative index (21OH index) based on the analysis of one positive and two negative standard sera included in each assay (13): the cut-off level for 21OH index was 0.06.
ICA. ICA were determined in an indirect immunofluorescence assay using cryostatic sections of human pancreas of blood group O. Levels of ICA were expressed in JDF units using the world reference standard serum sample (31).
GAD65Ab and ICA512Ab. GAD65Ab and IA-2/ICA512Ab were determined using radiobinding assays with in vitro translated recombinant human [35S]GAD65 or [35S]ICA512 as previously described in detail (32, 33). The complementary DNA of human ICA512bdc was a kind gift of Dr. George S. Eisenbarth, Barbara Davis Center for Childhood Diabetes, Denver, CO. GAD65Ab and ICA512Ab levels were expressed as relative indices (GAD65 index and ICA512 index) using one positive and two negative standard sera in each assay: the upper levels of normal were 0.03 and 0.04, respectively.
TPOAb. TPOAb were detected by RIA using a commercially available kit (Bio-Line, Brussels, Belgium). Results were expressed as international units per milliliter of bound TPO: the upper level of normal was 100 IU/mL.
VLCFA. Plasma VLCFA concentration was determined by capillary gas chromatography-mass spectrometry following procedures described previously (34). Upper levels of normal were: C26:0 = 0.6 µg/mL, C26/C22 ratio = 0.031, and C24/C22 ratio = 1.2.
Statistical analysis
Differences in frequencies of autoantibodies or of elevated
concentrations of VLCFA were tested with the 
-square method. Yates’
correction or the Fisher exact test were used when necessary. Because
the 21OHAb and ACA levels were not normally distributed among the
patients (Kolmogorov-Smirnov test), the relationship between 21OHAb
index or ACA levels and disease duration was analyzed, by linear
regression, after logarithmic transformation of antibody levels.
Correlation between 21OHAb and ACA levels was analyzed by Spearman’s
rank correlation test. A P value <0.05 was considered
significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A total of 35/56 (62%) patients had 21OHAb, representing 35/42
(83%) patients with a clinical diagnosis of idiopathic adrenal
insufficiency (P < 0.0001 vs.
posttuberculosis patients) (Table 2). Of
these 35 patients, 17 were also positive for ACA (P =
0.005 vs. posttuberculosis patients; P =
0.0001 frequency of 21OHAb vs. frequency of ACA). No
patients were found positive for ACA in the absence of 21OHAb. Plasma
concentrations of VLCFA higher than the upper levels of normal were
found in 4/56 (7%) (C26:0 = 0.52–1.04 µg/mL; C26/C22 =
0.033–0.1; C24/C22 = 1.34–1.88) patients, representing 4/42
(10%) patients with a clinical diagnosis of idiopathic adrenal
insufficiency, and, in particular, 4/19 (21%) males. Age of these
latter 4 patients ranged from 14–36 yr at the time of diagnosis of
primary adrenal insufficiency. In 2 of the four ALD cases, adrenal
insufficiency preceded the appearance of neurological clinical signs by
1–9 yr. In the remaining 2 ALD patients, no neurological clinical
signs have been detected at a 6- to 29-yr follow-up period after
diagnosis of adrenal insufficiency. Magnetic resonance imaging of the
central nervous system showed cerebral involvement and/or spinal
atrophy in all the 4 ALD patients. Similarly, electrophysiological
studies (such as brain stem auditory evoked responses, somatosensory
evoked potentials, and nerve conduction velocity) were abnormal in all
the 4 ALD patients.
|
).
|
A total of 39/42 (93%) patients with a clinically idiopathic adrenal insufficiency were positive for either 21OHAb or elevated levels of VLCFA. Thus, a definitive etiological diagnosis was possible in 53/56 (95%) total cases when radiological evaluation of adrenal glands was taken into consideration. In the remaining 3 patients, concomitant hypothyroidism was present in 2 with a duration of primary adrenal insufficiency of 22 and 32 yr, respectively. Hence, 34/35 (97%) idiopathic patients with less than 20 yr of disease duration were positive for either 21OHAb (31/35, 89%) or VLCFA (3/35, 8%). The last patient who could not be accounted for using a combination of 21OHAb and VLCFA was a 63-yr-old male with new onset primary adrenal insufficiency and periventricular demyelination lesions.
Organ-specific autoantibodies in patients with primary adrenal insufficiency
Of the 56 patients with primary adrenal insufficiency, 20 had hypothyroidism (disease duration: median 9, range 0–34 yr), 7 had IDDM (disease duration: median 10, range 5–25 yr), 2 had Graves’ disease (disease duration: range 0–1 yr), 7 had primary ovarian failure (disease duration: median 10, range 0–31 yr), 1 had vitiligo (disease duration: 21 yr), 1 had atrophic gastritis (disease duration: 32 yr), and 1 had pernicious anemia (disease duration: 1 yr). In 28 patients, no other concomitant diseases were present.
ICA were detected in 5/35 (14%) autoimmune and 1/14 (7%)
posttuberculosis patients (Table 3). Four
of the 6 ICA-positive subjects (66%) had clinical signs of IDDM.
Similarly, 4/7 (57%) IDDM patients were ICA-positive. GAD65Ab were
found in 8 patients, representing 7/35 (20%) cases with autoimmune and
1/14 (7%) cases with posttuberculosis disease (Table 3). In 6/8 (75%)
patients, GAD65Ab were associated with clinical signs of IDDM.
Similarly, 6/7 (86%) IDDM patients were found positive for GAD65Ab.
ICA512Ab were found in 4/35 (11%) patients with autoimmune adrenal
insufficiency (Table 3). IDDM was present in 3/4 (75%)
ICA512Ab-positive cases. Globally, 3/7 (43%) IDDM patients were
positive for ICA512Ab.
|
), representing 18/35
(51%) cases with autoimmune adrenal insufficiency. Clinical signs of
thyroid disease were present in 18/23 (78%) TPOAb-positive patients.
On the other hand, none of the two TPOAb-positive patients with
posttuberculosis adrenal insufficiency had clinical signs of thyroid
disease. Globally, 17/20 (85%) patients with clinical hypothyroidism
and 1/2 (50%) with hyperthyroidism were positive for TPOAb. A total of 23/35 (66%) idiopathic patients were positive for at least one islet or thyroid autoantibody type, which is significantly higher than the prevalence observed in posttuberculosis patients (2/14, 14%; P = 0.003).
In patients with autoimmune adrenal insufficiency, there was no significant difference between the age of APS II subjects (median 44 yr, range 15–75 yr) and the age of subjects with isolated adrenal insufficiency (median 42 yr, range 19–79 yr). Patients with TPOAb and/or thyroid disease (age: median 53 yr, range 19–75 yr) tended to be older than TPOAb-negative and thyroid disease-free subjects (age: median 34 yr, range 15–79 yr) even though the difference did not reach statistical significance. Prevalence or titer of thyroid or islet cell autoantibodies did not vary significantly with increasing age of patients.
Organ-specific autoantibodies in French ALD patients
None of 26 French ALD patients was found to be positive for any of the tested organ-specific autoantibodies (ACA, 21OHAb, ICA, GAD65Ab, ICA512Ab, or TPOAb).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our observation of a high diagnostic sensitivity of 21OHAb for autoimmune adrenal insufficiency supports the hypothesis that two thirds of patients are associated with autoimmune destruction of the adrenal cortical cells. In addition, in 74% of our clinically defined idiopathic patients, clinical signs and/or circulating autoantibodies typical of an organ-specific autoimmune disease were present. Only 48% of these patients were positive for ACA. On the other hand, 88% of them had 21OHAb and, conversely, 80% of 21OHAb-positive patients had a concomitant autoimmune disease and/or circulating organ-specific autoantibodies. These results indicate that the presence of 21OHAb is highly specific for autoimmune adrenal insufficiency. This conclusion is also supported by the absence of 21OHAb in posttuberculosis or ALD patients. On the other hand, the presence of thyroid or islet autoantibodies in 2 of the 14 posttuberculosis patients suggests that the presence of a concomitant endocrine autoimmune disease (or of organ-specific autoantibodies) is not sufficient for a definitive diagnosis of autoimmune adrenal insufficiency.
The presence of elevated VLCFA concentrations in 21% of our total population of consecutively diagnosed adult idiopathic males (representing 6% of total population and 13% of males with < 20 yr of disease duration) confirms the conclusions of previous reports showing the high incidence of ALD among patients with adrenal insufficiency (26, 28, 35, 36). Because our patient population was represented by adult subjects, patients with early-onset adrenal insufficiency, in whom ALD is highly prevalent (36), are underrepresented in our study. However, the overall frequency of X-linked ALD in our population of 19 adult idiopathic males is similar to that previously reported by Jorge et al. (36) in a group of males that included subjects with early-onset Addison’s disease. Accordingly, ALD must always be considered as a possible and not infrequent cause of primary adrenal insufficiency.
The observation that no patient was found simultaneously positive for
21OHAb and VLCFA demonstrates that these two parameters are specific
markers of two distinct causes of primary adrenal insufficiency.
Accordingly, based on the results of our study, we propose an original
flowchart for the etiological classification of primary adrenal
insufficiency (Fig. 2). Because the
majority of cases are caused by autoimmune destruction of
adrenocortical cells, and because of the very high diagnostic
sensitivity of 21OHAb, this immune marker can be used as first
screening to discriminate between autoimmune and nonautoimmune
patients. The 21OHAb assay is simple and low cost (approximately $10
U.S./sample) and therefore is the best screening assay for a
cost-effective algorithm for the etiological diagnosis of primary
adrenal insufficiency. In the presence of 21OHAb there is no need to
perform VLCFA assay or adrenal imaging. In fact, adrenal imaging has a
low diagnostic value in patients with autoimmune adrenal insufficiency
(4); this is also demonstrated by the absence of radiological
abnormalities of the adrenal glands in our patients with autoimmune
adrenal insufficiency. The availability of commercial assays for 21OHAb
determination makes this assay practical for clinical use. As a second
choice, the classical ACA assay can be used. In fact, although in our
study the prevalence of ACA was significantly lower than that of 21OHAb
in patients with more than 20 yr of disease duration, the frequency of
ACA was only slightly lower than that of 21OHAb in newly diagnosed
patients. ACA assay is commercially available, and the accuracy of the
commercial tests is similar to that obtained in the research
setting.
|
In our study, two idiopathic patients with long-term adrenal insufficiency (22 and 32 yr of disease duration) and hypothyroidism were negative for both adrenal autoantibodies (ACA and 21OHAb) and elevated VLCFA concentration. However, because the frequency and levels of 21OHAb (13) and ACA (38) decrease with increasing disease duration, we cannot rule out the possibility that an autoimmune process also caused adrenal insufficiency in these patients. On the other hand, it must be noted that, in our study, only 1/35 (3%) idiopathic patients with less than 20 yr of disease duration could not be accounted for by the combination of adrenal autoantibodies and VLCFA.
Nonautoimmune patients must be tested for ALD (especially if male), because early detection of central nervous system demyelination and/or medullary atrophy in ALD patients, and consequent institution of the appropriate therapy, may change the clinical course of the disease (24). In the presence of elevated levels of plasma VLCFA, magnetic resonance imaging of the central nervous system and neuroelectrophysiological tests should be carried out to discriminate between patients with and without neurological involvement. An additional important reason to detect ALD is genetic counseling. In fact, ALD is characterized by a wide phenotypic variation and less than 8% of de novo mutation (39, 40). The diagnosis of ALD may therefore lead to diagnosis of other cases in the family when genetic counseling is performed. Half of the cases of ALD involve children for whom there is no cure except when bone marrow transplantation at an early stage can be proposed (41, 42). It is therefore important to detect heterozygous women in ALD families to whom a prenatal diagnosis could be suggested.
Finally, adrenal imaging should be carried out in remaining patients to detect radiological signs of tumor, adrenal hemorrhage, or the presence of adrenal calcifications typical of chronic infection. In our study, the use of adrenal and lung imaging allowed a definitive diagnosis of posttuberculosis adrenal insufficiency in 25% of total cases and 22% of patients with short-medium disease duration.
In conclusion, the combined use of immune and biochemical markers allows a reliable etiological diagnosis in almost all patients with short-term idiopathic adrenal insufficiency, with only 2–3% of cases not accounted for. In these latter idiopathic cases, other rare conditions should be considered.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received April 27, 1998.
Revised May 28, 1998.
Accepted June 1, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase. Lancet. 339:770–773.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  I. Wallace, S. Cunningham, and J. Lindsay The diagnosis and investigation of adrenal insufficiency in adults Ann Clin Biochem, September 1, 2009; 46(5): 351 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Anglin, P. I. Rosebush, and M. F. Mazurek The Neuropsychiatric Profile of Addison's Disease: Revisiting a Forgotten Phenomenon J Neuropsychiatry Clin Neurosci, November 1, 2006; 18(4): 450 - 459. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Falorni, S. Laureti, A. De Bellis, R. Zanchetta, C. Tiberti, G. Arnaldi, V. Bini, P. Beck-Peccoz, A. Bizzarro, F. Dotta, et al. Italian Addison Network Study: Update of Diagnostic Criteria for the Etiological Classification of Primary Adrenal Insufficiency J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1598 - 1604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Betterle, C. Dal Pra, F. Mantero, and R. Zanchetta Autoimmune Adrenal Insufficiency and Autoimmune Polyendocrine Syndromes: Autoantibodies, Autoantigens, and Their Applicability in Diagnosis and Disease Prediction Endocr. Rev., June 1, 2002; 23(3): 327 - 364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Myhre, D. E. Undlien, K. Lovas, S. Uhlving, B. G. Nedrebo, K. J. Fougner, T. Trovik, J. I. Sorheim, and E. S. Husebye Autoimmune Adrenocortical Failure in Norway Autoantibodies and Human Leukocyte Antigen Class II Associations Related to Clinical Features J. Clin. Endocrinol. Metab., February 1, 2002; 87(2): 618 - 623. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Domenice, A. C. Latronico, V. N. Brito, I. J. P. Arnhold, F. Kok, and B. B. Mendonca Adrenocorticotropin-Dependent Precocious Puberty of Testicular Origin in a Boy with X-Linked Adrenal Hypoplasia Congenita Due to a Novel Mutation in the DAX1 Gene J. Clin. Endocrinol. Metab., September 1, 2001; 86(9): 4068 - 4071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ten, M. New, and N. Maclaren Addison's Disease 2001 J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 2909 - 2922. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Gambelunghe, A. Falorni, M. Ghaderi, S. Laureti, C. Tortoioli, F. Santeusanio, P. Brunetti, and C. B. Sanjeevi Microsatellite Polymorphism of the MHC Class I Chain-Related (MIC-A and MIC-B) Genes Marks the Risk for Autoimmune Addison's Disease J. Clin. Endocrinol. Metab., October 1, 1999; 84(10): 3701 - 3707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Laureti, L. Vecchi, F. Santeusanio, and A. Falorni Is the Prevalence of Addison's Disease Underestimated? J. Clin. Endocrinol. Metab., May 1, 1999; 84(5): 1762 - 1762. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
