- Split View
-
Views
-
Cite
Cite
C. Courtillot, S. Laugier Robiolle, F. Cohen Aubart, M. Leban, R. Renard-Penna, A. Drier, F. Charlotte, Z. Amoura, P. Touraine, J. Haroche, Endocrine Manifestations in a Monocentric Cohort of 64 Patients With Erdheim-Chester Disease, The Journal of Clinical Endocrinology & Metabolism, Volume 101, Issue 1, 1 January 2016, Pages 305–313, https://doi.org/10.1210/jc.2015-3357
- Share Icon Share
Abstract
Erdheim-Chester disease (ECD) is a rare non-Langerhans cell histiocytosis, characterized by infiltration of foamy histiocytes in multiple organs. Endocrine involvement has mostly been described in case reports.
We performed systematic endocrine evaluation in a large cohort of patients with ECD.
This was a single-center observational study conducted between October 2007 and May 2013.
The evaluation was conducted in Pitié-Salpêtrière Hospital (Paris, France), a tertiary care hospital.
Sixty-four consecutive patients with ECD (sex ratio, 3.6; mean age, 57.6 years [range, 20–80 years]). Thirty-six patients had follow-up assessments.
There were no interventions.
Clinical, biological, and morphological evaluations of pituitary, gonadal, adrenal, and thyroid functions, as well as metabolic evaluation, were performed.
Diabetes insipidus was found in 33.3% of patients, frequently as the first manifestation of ECD. Anterior pituitary dysfunction was found in 91.3% of patients with full anterior pituitary evaluation, including somatotropic deficiency (78.6%), hyperprolactinemia (44.1%), gonadotropic deficiency (22.2%), thyrotropic deficiency (9.5%), and corticotropic deficiency (3.1%). Thirty-five patients (54.7%) had ≥2 anterior pituitary dysfunctional axes, rising to 69.6% (16 of 23) when only patients with complete evaluations were considered. Two patients had panhypopituitarism. Infiltration of the pituitary and stalk was found with magnetic resonance imaging in 24.4% of patients. Testicular insufficiency was found in 53.1% of patients, with sonographic testicular infiltration in 29% of men, mostly bilateral. Computed tomography adrenal infiltration was found in 39.1% of patients, and 1 case of adrenal insufficiency was observed. No patient was free of endocrine hormonal or morphological involvement. Endocrine dysfunctions were most often permanent, and new deficits appeared during follow-up.
Endocrine involvement is very frequent in ECD and should be evaluated carefully at diagnosis and during follow-up.
Erdheim-Chester disease (ECD) is a rare form of non-Langerhans cell histiocytosis (LCH), first described by Erdheim and Chester in 1930 (1). It is characterized by tissue infiltration with foamy CD68+/CD1a− histiocytes, different from those in LCH. The incidence of ECD is unknown, but fewer than 650 cases have been reported in the literature. Diagnosis is most often made after the age of 40, and there is a male predominance (sex ratio, 3). It is a systemic disease with diverse manifestations: the clinical course mainly depends on the extent and distribution of the disease (2). Whereas bone pain is the most common symptom of the disease, half of all patients have extraskeletal manifestations, including central nervous system and cardiovascular involvement, retroperitoneal infiltration, interstitial lung disease, exophthalmos, and endocrine disorders (3). First-line treatment is based on interferon (IFN)-α therapy, but the recent discovery of a high prevalence of BRAF V600E mutations in ECD has led to the use of BRAF inhibitors, which represent a promising therapeutic alternative in patients with mutations (4, 5). ECD and LCH share common clinical manifestations, but many are different. Numerous cases of ECD associated with LCH have been reported (6, 7).
Endocrine manifestations have been described in ECD, but only in a few case reports and in a small retrospective series focused on adrenal involvement (4). The infiltration of the hypothalamo-pituitary axis leading to diabetes insipidus is the main known endocrine symptom of ECD and may be the first manifestation of the disease, sometimes long before diagnosis (8). Other endocrine disorders, such as hyperprolactinemia, gonadotropic deficiency, adrenal insufficiency, and low levels of IGF-I, have been reported in rare cases (4, 9–11).
To date, there has been no exhaustive investigation of the endocrine manifestations in ECD and no large cohort study. One of the most important international reference centers for ECD is located in the same hospital as our department. We thus decided to perform hormonal evaluations on all consecutive patients diagnosed with ECD, referred from the reference center to our department. Our aims were to determine the prevalence of endocrine manifestations in ECD and their evolution during follow-up. We present our 6-year experience in the first monocentric study of the largest cohort evaluating the endocrine features of ECD.
Subjects and Methods
Study design and setting
This study is a single-center observational study conducted in Pitié-Salpêtrière Hospital (Paris, France), a tertiary care center, between October 2007 and May 2013. For all consecutive patients with a diagnosis of ECD (based on histological proof and/or typical bone lesions) seen in the internal medicine department an endocrine evaluation in the endocrinology and reproductive medicine department was proposed. Those for whom the latter was not possible had an endocrine evaluation in the internal medicine department. During follow-up of patients in the internal medicine and/or endocrinology and reproductive medicine departments, endocrine reevaluations were proposed. Data were collected retrospectively.
Informed consent was obtained from all patients before the study began.
Clinical, biochemical, and morphological data
All of the endocrinological evaluations were performed during a hospitalization. The following data were retrieved: age at first symptoms and at diagnosis of ECD, known complications of ECD, specific treatments for ECD, and hormone replacement therapies. Patients were interrogated to determine the presence of polyuria-polydipsia syndrome, erectile dysfunction, and decreased libido. The following clinical data were obtained: body mass index (BMI), blood pressure, 24-hour evaluation of water intake and diuresis, thyroid examination, testicular volume, and mammary examination. The following radiological examinations were performed: pituitary magnetic resonance imaging (MRI), scrotal or pelvic ultrasound, thyroid ultrasound, and bone mineral density (BMD) assessment by dual-energy X-ray absorptiometry (DEXA). Abdominal computed tomography (CT) was part of the systematic evaluation of ECD in internal medicine, and results were collected when available.
In some male patients, testicular explorations were completed by testicular MRI, sperm count (using normal values from the World Health Organization, 2010 [12]), or testicular biopsy.
Patients were considered to have diabetes insipidus if they were already being treated with desmopressin or if they had a polyuria-polydipsia syndrome responding to desmopressin treatment.
Hormonal assays were performed to explore pituitary and peripheral gland functions. Gonadotropic and gonadal functions were evaluated by FSH, LH, and estradiol (E2) concentrations (in women) or testosterone and inhibin B concentrations (in men). Lactotrophic function was evaluated by the prolactin (PRL) concentration. Thyrotropic and thyroid functions were evaluated by TSH, free T4, and anti-thyroid peroxidase (TPO) and anti-thyroglobulin (ATG) antibody concentrations. Somatotropic function was evaluated by IGF-I and GH concentrations under stimulation by an insulin tolerance test (ITT) or glucagon, when possible. As stated in the Endocrine Society guidelines (13), GH deficiency (GHD) was defined by a peak GH concentration of <20 mIU/L after a stimulation test (severe GHD was defined as a peak GH concentration of <10 mIU/L, and partial GHD was defined as a peak GH concentration between 10 and 20 mIU/L).
According to literature reports (14, 15), corticotropin deficiency was defined by plasma cortisol of <30 ng/mL at 8:00 am, or a response to an ITT or 0.25 mg of Synacthen of <200 ng/mL for plasma cortisol and <13 ng/mL for salivary cortisol, associated with normal or low ACTH. Adrenal glucocorticoid insufficiency was defined by the same thresholds for cortisol as associated with elevated ACTH. Adrenal mineralocorticoid insufficiency was defined by elevated renin +/− low aldosterone.
Metabolic function was assessed by fasting blood glucose, hemoglobin A1c, and the lipid profile (total cholesterol, LDL cholesterol, high-density lipoprotein cholesterol, and triglycerides).
Bone metabolism was assessed by calcemia, phosphoremia, PTH(1–84) and 25-hydroxyvitamin D2 and D3 concentrations.
Laboratory assays
Blood tests were performed at 8:00 am, and hormonal assays were processed in the biochemical and hormonal department in Pitié-Salpêtrière Hospital. FSH, LH, E2, progesterone, total testosterone, SHBG, prolactin, ACTH, plasma cortisol, TSH, free T3, free T4, and PTH(1–84) were measured in serum by the Modular E170 automated chemiluminescent immunometric method (Roche Diagnostics France). Salivary cortisol was measured by the radioimmunometry method (Orion Diagnostica). Inhibin B was measured by ELISA (Immunotech, Beckman Coulter). GH and IGF-I were measured in serum by an automated chemiluminescent immunometric method (LIAISON; DiaSorin). Dehydroepiandrosterone sulfate was measured in serum by an automated chemiluminescent immunometric method (Beckman Coulter). Androstenedione was measured by extraction plus RIA (Beckman Coulter). Renin was measured in serum using a radioimmunometric procedure (Cisbio France), and aldosterone was measured using a radioimmunometric procedure (Siemens Healthcare Diagnostics). 25-Hydroxyvitamin D2 and D3 was measured by chemiluminescence (LIAISON XL; DiaSorin). TPO and ATG antibodies were measured by the Varelisa system (Thermo Fischer).
Statistical analysis
Statistical analyses were performed with StatView 5.0 (SAS Institute). Descriptive statistics used numbers and percentages for qualitative variables and means ± SD or medians and interquartile range for quantitative variables. Groups were compared using the χ2 test or the Fisher exact test for qualitative variables or a parametric test (ANOVA) for continuous variables. Values of P < .05 were considered significant.
Results
Patients
Between October 2007 and May 2013, 86 patients with ECD were seen in the department of internal medicine in Pitié-Salpêtrière Hospital. Of these, 22 did not have an endocrine evaluation and 64 (50 men/14 women) were hospitalized for endocrine evaluation, either in the endocrinology (n = 47) or internal medicine (n = 17) department. Most patients were French; 9 patients were from foreign countries. The characteristics of the patients at their first endocrine evaluation are summarized in Table 1.
. | Value . |
---|---|
Age at diagnosis, y | 54.2 ± 14.8 |
Age at first clinical signs of ECD, y | 49.6 ± 15.8 |
Time before diagnosis, y | 4.9 ± 6.5 |
Age at first endocrinological symptoms, y | 44.8 ± 16.1 |
Initial endocrinological manifestations | 14/61 (23) |
Diabetes insipidus | 12/14 (86.7) |
Gonadotropic insufficiency | 3/14 (21.4) |
Age at first endocrinological evaluation, y | 57.6 ± 13.4 |
Known endocrinological involvement before evaluation | 23/64 (35.9) |
Diabetes insipidus | 21/23 (91.3) |
At least 1 anterior pituitary deficit | 9/23 (39.1) |
ECD | 51/64 (79.7) |
ECD + LCH | 9/64 (14.1) |
ECD + RDD | 3/64 (4.7) |
ECD + LCH + RDD | 1/64 (1.5) |
Histological proof of ECD | 62/63 (98.4) |
BRAF status | |
V600E mutation | 32/64 |
WT | 17/64 |
NI | 9/64 |
ND | 6/64 |
Skeletal involvement | 63/64 (98.4) |
Cardiovascular involvement | 48/64 (75) |
Retroperitoneal involvement | 44/63 (68.9) |
Sinusal involvement | 30/64 (46.9) |
Dermatological involvement | 21/60 (35) |
Central nervous system involvement | 21/64 (32.8) |
Pulmonary involvement | 16/64 (25) |
. | Value . |
---|---|
Age at diagnosis, y | 54.2 ± 14.8 |
Age at first clinical signs of ECD, y | 49.6 ± 15.8 |
Time before diagnosis, y | 4.9 ± 6.5 |
Age at first endocrinological symptoms, y | 44.8 ± 16.1 |
Initial endocrinological manifestations | 14/61 (23) |
Diabetes insipidus | 12/14 (86.7) |
Gonadotropic insufficiency | 3/14 (21.4) |
Age at first endocrinological evaluation, y | 57.6 ± 13.4 |
Known endocrinological involvement before evaluation | 23/64 (35.9) |
Diabetes insipidus | 21/23 (91.3) |
At least 1 anterior pituitary deficit | 9/23 (39.1) |
ECD | 51/64 (79.7) |
ECD + LCH | 9/64 (14.1) |
ECD + RDD | 3/64 (4.7) |
ECD + LCH + RDD | 1/64 (1.5) |
Histological proof of ECD | 62/63 (98.4) |
BRAF status | |
V600E mutation | 32/64 |
WT | 17/64 |
NI | 9/64 |
ND | 6/64 |
Skeletal involvement | 63/64 (98.4) |
Cardiovascular involvement | 48/64 (75) |
Retroperitoneal involvement | 44/63 (68.9) |
Sinusal involvement | 30/64 (46.9) |
Dermatological involvement | 21/60 (35) |
Central nervous system involvement | 21/64 (32.8) |
Pulmonary involvement | 16/64 (25) |
Abbreviations: ND, Not determined; NI, noninformative (poor DNA quality); RDD, Rosai Dorfman disease; WT, wild type. Data are means ± SD or n (%).
. | Value . |
---|---|
Age at diagnosis, y | 54.2 ± 14.8 |
Age at first clinical signs of ECD, y | 49.6 ± 15.8 |
Time before diagnosis, y | 4.9 ± 6.5 |
Age at first endocrinological symptoms, y | 44.8 ± 16.1 |
Initial endocrinological manifestations | 14/61 (23) |
Diabetes insipidus | 12/14 (86.7) |
Gonadotropic insufficiency | 3/14 (21.4) |
Age at first endocrinological evaluation, y | 57.6 ± 13.4 |
Known endocrinological involvement before evaluation | 23/64 (35.9) |
Diabetes insipidus | 21/23 (91.3) |
At least 1 anterior pituitary deficit | 9/23 (39.1) |
ECD | 51/64 (79.7) |
ECD + LCH | 9/64 (14.1) |
ECD + RDD | 3/64 (4.7) |
ECD + LCH + RDD | 1/64 (1.5) |
Histological proof of ECD | 62/63 (98.4) |
BRAF status | |
V600E mutation | 32/64 |
WT | 17/64 |
NI | 9/64 |
ND | 6/64 |
Skeletal involvement | 63/64 (98.4) |
Cardiovascular involvement | 48/64 (75) |
Retroperitoneal involvement | 44/63 (68.9) |
Sinusal involvement | 30/64 (46.9) |
Dermatological involvement | 21/60 (35) |
Central nervous system involvement | 21/64 (32.8) |
Pulmonary involvement | 16/64 (25) |
. | Value . |
---|---|
Age at diagnosis, y | 54.2 ± 14.8 |
Age at first clinical signs of ECD, y | 49.6 ± 15.8 |
Time before diagnosis, y | 4.9 ± 6.5 |
Age at first endocrinological symptoms, y | 44.8 ± 16.1 |
Initial endocrinological manifestations | 14/61 (23) |
Diabetes insipidus | 12/14 (86.7) |
Gonadotropic insufficiency | 3/14 (21.4) |
Age at first endocrinological evaluation, y | 57.6 ± 13.4 |
Known endocrinological involvement before evaluation | 23/64 (35.9) |
Diabetes insipidus | 21/23 (91.3) |
At least 1 anterior pituitary deficit | 9/23 (39.1) |
ECD | 51/64 (79.7) |
ECD + LCH | 9/64 (14.1) |
ECD + RDD | 3/64 (4.7) |
ECD + LCH + RDD | 1/64 (1.5) |
Histological proof of ECD | 62/63 (98.4) |
BRAF status | |
V600E mutation | 32/64 |
WT | 17/64 |
NI | 9/64 |
ND | 6/64 |
Skeletal involvement | 63/64 (98.4) |
Cardiovascular involvement | 48/64 (75) |
Retroperitoneal involvement | 44/63 (68.9) |
Sinusal involvement | 30/64 (46.9) |
Dermatological involvement | 21/60 (35) |
Central nervous system involvement | 21/64 (32.8) |
Pulmonary involvement | 16/64 (25) |
Abbreviations: ND, Not determined; NI, noninformative (poor DNA quality); RDD, Rosai Dorfman disease; WT, wild type. Data are means ± SD or n (%).
Pituitary and peripheral gland functions
Forty-one patients were evaluated with pituitary MRI. The pituitary stalk was infiltrated in 12.2% of patients (5 of 41). The pituitary region was infiltrated in 14.6% of patients (6 of 41); only 1 of these also had enlargement of the pituitary stalk. Among the 10 patients with a visible infiltrative lesion of the pituitary region or stalk on MRI, 9 (90%) had pituitary deficits (100% of the patients with an enlarged stalk).
An absence of the physiological T1 hypersignal of the posterior pituitary was found in 60% of patients (24 of 40). Diabetes insipidus was found in 33.3% of patients (19 of 57) (Table 2), which was not correlated to the loss of the posterior pituitary bright spot (P = .06).
Hormonal Dysfunction . | % of Patients (n) . |
---|---|
GH deficiency | 78.6 (22/28) |
Testicular deficiency | 53.1 (26/49) |
Hyperprolactinemia | 44.1 (26/59) |
Diabetes insipidus | 33.3 (19/57) |
Gonadotropic deficiency | 22.2 (14/63) |
Thyrotropic deficiency | 9.5 (6/63) |
Thyroid deficiency | 9.5 (6/63) |
Corticotropic deficiency | 3.1 (2/64) |
Adrenal deficiency | 1.6 (1/64) |
Hormonal Dysfunction . | % of Patients (n) . |
---|---|
GH deficiency | 78.6 (22/28) |
Testicular deficiency | 53.1 (26/49) |
Hyperprolactinemia | 44.1 (26/59) |
Diabetes insipidus | 33.3 (19/57) |
Gonadotropic deficiency | 22.2 (14/63) |
Thyrotropic deficiency | 9.5 (6/63) |
Thyroid deficiency | 9.5 (6/63) |
Corticotropic deficiency | 3.1 (2/64) |
Adrenal deficiency | 1.6 (1/64) |
Hormonal Dysfunction . | % of Patients (n) . |
---|---|
GH deficiency | 78.6 (22/28) |
Testicular deficiency | 53.1 (26/49) |
Hyperprolactinemia | 44.1 (26/59) |
Diabetes insipidus | 33.3 (19/57) |
Gonadotropic deficiency | 22.2 (14/63) |
Thyrotropic deficiency | 9.5 (6/63) |
Thyroid deficiency | 9.5 (6/63) |
Corticotropic deficiency | 3.1 (2/64) |
Adrenal deficiency | 1.6 (1/64) |
Hormonal Dysfunction . | % of Patients (n) . |
---|---|
GH deficiency | 78.6 (22/28) |
Testicular deficiency | 53.1 (26/49) |
Hyperprolactinemia | 44.1 (26/59) |
Diabetes insipidus | 33.3 (19/57) |
Gonadotropic deficiency | 22.2 (14/63) |
Thyrotropic deficiency | 9.5 (6/63) |
Thyroid deficiency | 9.5 (6/63) |
Corticotropic deficiency | 3.1 (2/64) |
Adrenal deficiency | 1.6 (1/64) |
The mean age of the 50 male patients was 57.4 ± 12.8 years. Interrogation found decreased libido in 43.8% (12 of 32) and erectile dysfunction in 52.9% (18 of 34). The hormonal evaluation showed gonadotropic deficiency in 20.4% and testicular deficiency in 53.1% of men (Table 3). The mean testicular volume was 11.4 ± 4.8 mL (right) and 12.1 ± 5.2 mL (left) on ultrasonography (US), which showed testicular infiltration in 29.1% of patients, most often bilateral (Figure 1). Testicular infiltration was not significantly correlated with testicular deficiency (P = .23). A testicular volume of <15 mL was strongly correlated with the testicular deficiency (P = .006), but the correlation to the gonadotropic deficiency did not reach significance (P = .63). On the other hand, having a testicular volume of >15 mL was correlated with normal pituitary-testicular function (P = .0002).
. | Results . | n (%) . |
---|---|---|
Hormonal evaluation | Normal pituitary-testicular axis | 13/49 (26.5) |
Gonadotropic deficiency | 10/49 (20.4) | |
Testicular deficiency | 26/49 (53.1) | |
US | Testicular volume of <15 mL | 22/27 (81.5) |
Normal testicular structure | 22/31 (71) | |
Unilateral infiltration | 3/31 (9.7) | |
Bilateral infiltration | 6/31 (19.4) |
. | Results . | n (%) . |
---|---|---|
Hormonal evaluation | Normal pituitary-testicular axis | 13/49 (26.5) |
Gonadotropic deficiency | 10/49 (20.4) | |
Testicular deficiency | 26/49 (53.1) | |
US | Testicular volume of <15 mL | 22/27 (81.5) |
Normal testicular structure | 22/31 (71) | |
Unilateral infiltration | 3/31 (9.7) | |
Bilateral infiltration | 6/31 (19.4) |
. | Results . | n (%) . |
---|---|---|
Hormonal evaluation | Normal pituitary-testicular axis | 13/49 (26.5) |
Gonadotropic deficiency | 10/49 (20.4) | |
Testicular deficiency | 26/49 (53.1) | |
US | Testicular volume of <15 mL | 22/27 (81.5) |
Normal testicular structure | 22/31 (71) | |
Unilateral infiltration | 3/31 (9.7) | |
Bilateral infiltration | 6/31 (19.4) |
. | Results . | n (%) . |
---|---|---|
Hormonal evaluation | Normal pituitary-testicular axis | 13/49 (26.5) |
Gonadotropic deficiency | 10/49 (20.4) | |
Testicular deficiency | 26/49 (53.1) | |
US | Testicular volume of <15 mL | 22/27 (81.5) |
Normal testicular structure | 22/31 (71) | |
Unilateral infiltration | 3/31 (9.7) | |
Bilateral infiltration | 6/31 (19.4) |
Testicular MRI was performed in 2 patients with ultrasonographic bilateral testicular infiltration. MRI anomalies were found in 1 patient: bilateral pseudonodular, poorly limited lesions, with low vascularization, a low T2 signal, and a progressive and moderate pattern for uptake of contrast material.
Sperm counts were performed in 6 men (Supplemental Table 1), and the parameters were all normal in only 1 man. In the 5 other men, the anomalies ranged from mild oligospermia and/or asthenospermia to azoospermia. We found no correlation between gonadotropic/gonadal function, sperm count, and testicular US findings.
One man with bilateral testicular infiltration on US had a testicular biopsy. The results confirmed the presence of a non-Langerhans histiocyte infiltrate indicating ECD (Figure 2).
The mean age of the 14 women was 58.2 ± 16.6 years. Eleven women were menopausal (mean age, 65.6 years; range, 55–78 years). Among the 3 nonmenopausal women (mean age, 31.3 years; range, 20–41 years), 1 woman had regular menses and the other 2 women had oligomenorrhea. One of them had a normal pregnancy. The hormonal evaluation showed that the pituitary-gonadal axis was normal in 71.4% of women (8 of 11 of the menopausal women and 2 of 3 of the nonmenopausal women). There was a gonadotropic deficiency in 28.6% of women (3 of 11 of the menopausal women and 1 of 3 of the nonmenopausal women). We found no ovarian insufficiency in the 3 premenopausal women. Pelvic sonographies were performed in 3 women and showed no anomaly (in particular, no ovarian infiltration).
Hyperprolactinemia was found in 44.1% of patients (26 of 59) (Table 2), and the frequency was the same in men and women (47.8% and 30.8%, respectively). Elevation of PRL was very mild in all patients: the mean PRL in men was 21.7 ± 7 ng/mL (normal, <15.2 ng/mL) and in women was 22.8 ± 18.8 ng/mL (normal, <23.3). No patient had galactorrhea on clinical examination. Three men had bilateral gynecomastia; none of them had hyperprolactinemia, but 1 had gonadotropic deficiency, 1 had testicular insufficiency, and 1 had a normal pituitary-testicular axis.
A stimulation test on GH was performed in 28 patients. It showed GHD in 78.6% of patients (22 of 28) (Table 2): partial in 32.1% of patients (9 of 28) and severe in 46.4% of patients (13 of 28). The IGF-I level was evaluated in 60 patients and 4 of them had a level of <2 standard deviation score. These 4 patients did not have a stimulation test, and 3 of them had associated anterior pituitary deficits and could thus be considered as having GHD.
The pituitary-adrenal axis was normal in 70.3% of patients (45 of 64) and was not evaluated in 23.4% of patients (12 of 64) being treated with corticosteroids at the time. Corticotropic deficiency was found in 4 patients: 2 of them had no history of corticosteroid treatment and the other 2 were being treated with hydrocortisone since they had stopped corticosteroid treatment. We thus considered that 3.1% of patients (2 of 64) patients had corticotropin insufficiency associated with ECD (Table 2). Only 1 patient (1.6%) showed the features of adrenal glucocorticoid insufficiency, despite the frequency of adrenal infiltration on abdominal CT scans: 26.1% (6 of 23) had bilateral adrenal infiltration and 13% (3 of 23) had unilateral adrenal infiltration.
Most patients had normal thyroid function (81%, 51 of 63). Six patients had thyrotropic insufficiency (9.5%), and 6 patients had peripheral thyroid insufficiency (9.5%) (Table 2). Only 5 patients had thyroid US at the first evaluation, and we found nonsuspicious nodules in 2 patients and a goiter in 1 patient. No fine needle aspiration of the nodules was performed.
Thirty-seven patients were initially treated with IFN (mean duration, 22.8 months), and 3 of them (8.1%) developed autoimmune thyroiditis.
Among the 64 patients, 45 had at least 1 dysfunctional axis. In the other 17 patients, no hormonal anomaly was found, but anterior pituitary explorations were incomplete. Anterior pituitary evaluations were complete in 23 patients and showed at least 1 dysfunctional axis in 21 patients (91.3%). Thirty-five patients in the cohort (54.7%) were found to have 2 or more anterior pituitary dysfunctional axes, and the percentage rose to 69.6% (16 of 23) when only patients with complete evaluations were considered. Two patients had panhypopituitarism. There was no correlation between anterior and posterior pituitary involvement.
There was no difference between men and women concerning the frequencies of the deficits, besides the gonadal insufficiency, which was not found in women.
In the whole cohort, only 1 patient did not have any hormonal disturbances at the first evaluation. He was one of the youngest patients (born in 1980); however, he had infiltration of the pituitary on MRI, of both testicles on US, and of 1 adrenal gland on CT.
Metabolic and bone evaluations
We found that 53.2% of the patients in the cohort were overweight, with a BMI of >25 kg/m2. Half of the patients (50.8%) were hypertensive. Diabetes mellitus was found in 29.8% of the patients (17 of 57). Lipid profile evaluation showed that 37% of patients (20 of 54) had triglyceride levels of >1.8 mmol/L and 15% of patients (8 of 52) had LDL cholesterol of >4 mmol/L. Those findings were not different between patients receiving or not receiving corticosteroid therapy.
We found no anomaly in calcium metabolism, other than 7 patients with moderately low calcium levels and 68% of patients with low vitamin D levels.
Thirty-seven patients had BMD assessment by DEXA. BMD was normal in 12 patients (32.4%), showed osteopenia in 18 patients (48.7%), and showed osteoporosis in 7 patients (18.9%). There were no differences for the various risk factors for osteoporosis (female sex, menopause status, smoking, BMI of <20 kg/m2, previous or ongoing corticosteroid therapy, low vitamin D level, and hypogonadism) according to the mineral status found on DEXA.
Evolution
Follow-up data were available for 36 patients: 28 patients had 2 endocrinological evaluations, and 8 patients had 3 assessments. The mean time between evaluations was 13.7 months. Most of the pituitary deficits were permanent; in particular, we did not observe any regression in diabetes insipidus, corticotropic deficiency, or GHD. Follow-up showed that deficits not present at the first evaluation could appear progressively.
The morphologic findings in patients who had 2 or 3 consecutive pituitary MRI scans (n = 5 and 2, respectively), testicular ultrasonographies (n = 3 and 2, respectively), or abdominal CT scans (n = 20 for a second CT scan) were unchanged.
Discussion
We report here a series of 64 consecutive patients with ECD, who had endocrinological evaluations of pituitary, gonadal, adrenal, and thyroid function, as well as metabolic and bone status.
As previously reported, posterior pituitary dysfunction is a frequent endocrinological manifestation of ECD, but we showed that in many patients, ECD also affects the anterior pituitary gland. The hormonal functions of the peripheral glands, such as the gonads in men, are also frequently affected. In all, we demonstrated the extremely high prevalence of endocrine involvement in ECD, because we found no patient without any endocrine hormonal or morphological involvement.
Similar to findings in LCH, diabetes insipidus is often the first endocrine manifestation in ECD, clinically sudden and symptomatic, and once present, it is definitive, whatever the evolution of the histiocytosis (16, 17). The prevalence of diabetes insipidus in our study is similar to that found in a previous retrospective study on 59 patients with ECD, in which 17 patients (28.8%) were affected (8). In that study, diabetes insipidus was described as the most frequent endocrinological manifestation of ECD, but there was no systematic evaluation of the anterior pituitary function. The prevalence of diabetes insipidus in LCH is similar, around 30% (16, 17). The absence of the physiological bright spot of the posterior pituitary is frequent in patients with diabetes insipidus, whether it is idiopathic or secondary to ECD, LCH, or tumors. This MRI finding is not specific for the etiology, and it is not correlated with the presence of diabetes insipidus.
Anterior pituitary dysfunction has previously been described in ECD, usually in case reports, but systematic assessments have not been conducted previously. Our study shows that its prevalence is very high. The global prevalence of anterior pituitary dysfunction in LCH is only 20%, but very few studies reported patients with a complete evaluation, probably underestimating the prevalence of some deficits (16).
In our cohort, the order of frequency of anterior pituitary deficits is similar to that found in LCH or after radiotherapy, in which GHD is by far the most frequent deficit (up to 67% and 100%, respectively) (16–19). GHD was probably underestimated in our study, because a dynamic test on GH secretion was performed in less then half of the patients. The lower prevalence of gonadotropic deficiency in our cohort, compared with that of LCH (53%–58%) (16) or after radiotherapy (40%–50%) (19), could be explained by the latency of the appearance of this deficit. In LCH, it was found to appear 9 years after the diagnosis of the disease; as in our cohort it was evaluated around 3 years after diagnosis. The prevalence of corticotropic deficiency was probably underestimated in our study, because we considered that patients who had received corticosteroid treatment were deficient because of the treatment rather than ECD.
We showed in the present series, as described in patients with LCH, that pituitary deficits can appear progressively during follow-up and that they generally are definitive, evolving independently from the disease course (16, 20, 21). This finding argues in favor of regular and continuous assessment of pituitary function in patients with ECD.
In our cohort of patients with ECD, the pituitary hormonal disturbances were more frequent than the pituitary MRI anomalies. Hence, a normal pituitary MRI scan is not sufficient to rule out pituitary impairment. On the other hand, anterior pituitary hormone dysfunction is always found when MRI visualizes stalk infiltration. As described in LCH, the pituitary deficits in ECD are suspected to result from the infiltration of the hypothalamic-pituitary region. This infiltration, probably often not seen on MRI, has been demonstrated in an autopsy case (22), but in our cohort we have no histological proof of such infiltration. Unlike LCH in which anterior pituitary deficits are almost always associated with diabetes insipidus, we found no correlation between anterior and posterior pituitary deficits in patients with ECD (16).
Interestingly, our study showed the high prevalence of testicular involvement in men with ECD. In previous studies, 4 cases of histological proof of non-Langerhans cell infiltration of the testis and 1 case of Rosai Dorfman disease had been described (9, 23). No radiological description of any testicular anomaly in ECD had ever been published. Based on these findings, we decided to perform testicular sonography in men with ECD and found a high prevalence of testicular infiltration. A testicular tumor could be suspected for such morphological findings, but as for testicular rest tumors in congenital adrenal hyperplasia, their bilateralism and the context of the main disease argues against the diagnosis of a testicular tumor (24). Moreover, in 1 of our patients testicular infiltration by ECD was proven by a testicular biopsy. In regard to testicular function, it had not previously been assessed in patients with ECD. In our study, we found a high prevalence of testicular insufficiency. Some of these men did not have sonographic testicular lesions, which could suggest the existence of microinfiltrative testicular lesions, not visible with the current imaging techniques; however, we cannot rule out another cause of testicular deficiency. Conversely, some men with sonographic testicular lesions, even bilateral, did not have gonadal insufficiency, suggesting that the level of destruction of the testis by non-Langerhans cell infiltration is very difficult to evaluate. Half of the men who had sperm count assessment had significant alterations, including azoospermia. Although we cannot draw conclusions on such a small number of sperm counts, we found poor correlations between gonadotropic/gonadal function, sperm count, and testicular US findings, and we cannot rule out causes other than ECD for sperm count alterations.
Adrenal infiltration is a common finding in patients with ECD and is generally bilateral. In previous studies, the prevalence of adrenal infiltration was significantly lower, probably because of incomplete exploration. Histological infiltration of the adrenal glands by ECD has already been described, and 1 case (among 7 patients) of adrenal glucocorticoid insufficiency has been reported in the literature (4), as well as in our series, showing that it is a rare feature, even in patients with bilateral adrenal infiltration.
Thyroid infiltration has rarely been described in ECD or in LCH (9). No description of sonographic findings exists in the literature. Few patients in our study had thyroid US, but all of the nodules found had the typical features of benign nodules and no suspicious nodule or image of infiltration was found. This feature has to be more thoroughly studied in the future.
It is difficult to evaluate the exact influence of specific ECD treatments on endocrine consequences. Indeed, in the literature, patients have been receiving multiple treatments, and the endocrine impairments were never an endpoint. One study has shown that high doses of IFN-α had variable efficacy, depending on the organs and that a morphological amelioration was observed in the pituitary in 2 of 7 patients (6). In our follow-up study, we observed regression of a few pituitary deficits, but it is difficult to link that finding to the specific ECD treatments. In LCH, the influence of histiocytosis-specific treatment on prevention of pituitary deficits, such as diabetes insipidus, remains a matter of debate. However, it mostly seems that it cannot prevent the development of pituitary impairment or of any other endocrine lesions (21).
In regard to bone metabolism, we found in our cohort a lower prevalence of osteoporosis than what could have been expected with the multiple risk factors in this population (proinflammatory cytokines, anterior pituitary deficits and hormonal substitutions, and corticosteroid therapy). However, bone lesions in ECD are condensing and could thus lead to an overestimation of bone density and an underestimation of demineralization.
Another important finding in our study was the fact that patients with ECD are at high cardiovascular risk, on the one hand because of the cardiovascular localization of the disease and on the other hand because of the high prevalence of hypertension and diabetes mellitus, which was found to be 6 times more frequent than in the general population (25). Moreover, anterior pituitary insufficiency is associated with increased mortality in patients from cardiovascular and cerebrovascular causes (26). This should be taken into account in the care of patients with ECD, in whom it could worsen the prognosis. Careful supplementation of hormonal deficits has to be instituted and monitored by endocrine specialists. In addition, besides lowering vascular risk, hormonal substitution can also prevent osteoporosis and significantly improve quality of life.
This study reports the first evaluation of endocrine involvement in a large series of patients with ECD. We have shown that endocrine impairment is frequent, and we found no patient without endocrine hormonal or morphological involvements. Even if endocrine insufficiencies are not usually the main concern in these patients who are often very sick, ignorance of such deficits could worsen the prognosis and alter their quality of life further.
We present in Table 4 our recommendations for endocrine assessment in patients with ECD.
. | Clinical Evaluation . | Morphological Evaluation . | Biological Evaluation . |
---|---|---|---|
Pituitary | Search for signs of anterior pituitary deficits 24-hour diuresis and water intake | Pituitary MRI | FSH, LH, E2 (women)/testosterone (men) PRL TSH, FT4 IGF-I, GH under insulin tolerance test ACTH, cortisol under insulin tolerance test or after Synacthen test Natremia and urinary osmolarity |
Gonads | Evaluation of testicular volume and search for palpable testicular nodules | Gonadal sonography (and sperm cryopreservation in men with testicular infiltration) | FSH, LH E2 (women) Testosterone + inhibin B (men) |
Thyroid | Search for a goiter and nodules | Thyroid sonography if clinical anomalies | TSH, FT4 (and TPO + ATG antibodies in patients undergoing IFN therapy) |
Adrenal glands | Search for signs of adrenal deficiency | Abdominal or adrenal CT scan | ACTH, cortisol under ITT or after Synacthen test Renin, aldosterone |
Breast | Search for lumps (in women and men) | Mammography ± mammary sonography if presence of clinical lumps | |
Metabolism | Blood pressure Electrocardiogram | Fasting glycemia ± HbA1c TC, TG, HDL-c, LDL-c |
. | Clinical Evaluation . | Morphological Evaluation . | Biological Evaluation . |
---|---|---|---|
Pituitary | Search for signs of anterior pituitary deficits 24-hour diuresis and water intake | Pituitary MRI | FSH, LH, E2 (women)/testosterone (men) PRL TSH, FT4 IGF-I, GH under insulin tolerance test ACTH, cortisol under insulin tolerance test or after Synacthen test Natremia and urinary osmolarity |
Gonads | Evaluation of testicular volume and search for palpable testicular nodules | Gonadal sonography (and sperm cryopreservation in men with testicular infiltration) | FSH, LH E2 (women) Testosterone + inhibin B (men) |
Thyroid | Search for a goiter and nodules | Thyroid sonography if clinical anomalies | TSH, FT4 (and TPO + ATG antibodies in patients undergoing IFN therapy) |
Adrenal glands | Search for signs of adrenal deficiency | Abdominal or adrenal CT scan | ACTH, cortisol under ITT or after Synacthen test Renin, aldosterone |
Breast | Search for lumps (in women and men) | Mammography ± mammary sonography if presence of clinical lumps | |
Metabolism | Blood pressure Electrocardiogram | Fasting glycemia ± HbA1c TC, TG, HDL-c, LDL-c |
Abbreviations: CT, computed tomography; FT4, free T4; HbA1c, hemoglobin A1c; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; TC, total cholesterol; TG, triglycerides. We also recommend that patients be followed up regularly by an endocrinologist for monitoring of hormonal substitutions and detection of apparition of new deficits.
. | Clinical Evaluation . | Morphological Evaluation . | Biological Evaluation . |
---|---|---|---|
Pituitary | Search for signs of anterior pituitary deficits 24-hour diuresis and water intake | Pituitary MRI | FSH, LH, E2 (women)/testosterone (men) PRL TSH, FT4 IGF-I, GH under insulin tolerance test ACTH, cortisol under insulin tolerance test or after Synacthen test Natremia and urinary osmolarity |
Gonads | Evaluation of testicular volume and search for palpable testicular nodules | Gonadal sonography (and sperm cryopreservation in men with testicular infiltration) | FSH, LH E2 (women) Testosterone + inhibin B (men) |
Thyroid | Search for a goiter and nodules | Thyroid sonography if clinical anomalies | TSH, FT4 (and TPO + ATG antibodies in patients undergoing IFN therapy) |
Adrenal glands | Search for signs of adrenal deficiency | Abdominal or adrenal CT scan | ACTH, cortisol under ITT or after Synacthen test Renin, aldosterone |
Breast | Search for lumps (in women and men) | Mammography ± mammary sonography if presence of clinical lumps | |
Metabolism | Blood pressure Electrocardiogram | Fasting glycemia ± HbA1c TC, TG, HDL-c, LDL-c |
. | Clinical Evaluation . | Morphological Evaluation . | Biological Evaluation . |
---|---|---|---|
Pituitary | Search for signs of anterior pituitary deficits 24-hour diuresis and water intake | Pituitary MRI | FSH, LH, E2 (women)/testosterone (men) PRL TSH, FT4 IGF-I, GH under insulin tolerance test ACTH, cortisol under insulin tolerance test or after Synacthen test Natremia and urinary osmolarity |
Gonads | Evaluation of testicular volume and search for palpable testicular nodules | Gonadal sonography (and sperm cryopreservation in men with testicular infiltration) | FSH, LH E2 (women) Testosterone + inhibin B (men) |
Thyroid | Search for a goiter and nodules | Thyroid sonography if clinical anomalies | TSH, FT4 (and TPO + ATG antibodies in patients undergoing IFN therapy) |
Adrenal glands | Search for signs of adrenal deficiency | Abdominal or adrenal CT scan | ACTH, cortisol under ITT or after Synacthen test Renin, aldosterone |
Breast | Search for lumps (in women and men) | Mammography ± mammary sonography if presence of clinical lumps | |
Metabolism | Blood pressure Electrocardiogram | Fasting glycemia ± HbA1c TC, TG, HDL-c, LDL-c |
Abbreviations: CT, computed tomography; FT4, free T4; HbA1c, hemoglobin A1c; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; TC, total cholesterol; TG, triglycerides. We also recommend that patients be followed up regularly by an endocrinologist for monitoring of hormonal substitutions and detection of apparition of new deficits.
Acknowledgments
Disclosure Summary: The authors have nothing to disclose.
P.T. and J.H. contributed equally to the study.
Abbreviations
- ATG
anti-thyroglobulin
- BMD
bone mineral density
- BMI
body mass index
- CT
computed tomography
- DEXA
dual-energy X-ray absorptiometry
- E2
estradiol
- ECD
Erdheim-Chester disease
- GHD
GH deficiency
- IFN
interferon
- ITT
insulin tolerance test
- LCH
Langerhans cell histiocytosis
- MRI
magnetic resonance imaging
- TPO
thyroid peroxidase
- US
ultrasonography.
References