Post-Operative Cushing Syndrome Care

Justine Herndon, PA-C, and Irina Bancos, MD, on Post-Operative Cushing Syndrome Care

– Curative procedures led to widespread resolution or improvement of hyperglycemia

by Scott Harris , Contributing Writer, MedPage Today January 18, 2022

In a recent study, two-thirds of people with Cushing syndrome (CS) saw resolved or improved hyperglycemia after a curative procedure, with close post-operative monitoring an important component of the process.

Among 174 patients with CS included in the longitudinal cohort study (pituitary in 106, ectopic in 25, adrenal in 43), median baseline HbA1c was 6.9%. Of these, 41 patients were not on any therapy for hyperglycemia, 93 (52%) took oral medications, and 64 (37%) were on insulin.

At the end of the period following CS remission (median 10.5 months), 37 (21%) patients had resolution of hyperglycemia, 82 (47%) demonstrated improvement, and 55 (32%) had no change or worsened hyperglycemia. Also at the end of follow-up, HbA1c had fallen 0.84% (P<0.0001), with daily insulin dose decreasing by a mean of 30 units (P<0.0001).

Justine Herndon, PA-C, and Irina Bancos, MD, both endocrinology researchers with Mayo Clinic in Minnesota, served as co-authors of the report, which was published in the Journal of the Endocrine Society. Here they discuss the study and its findings with MedPage Today. The exchange has been edited for length and clarity.

What was the study’s main objective?

Herndon: As both a hospital diabetes provider and clinic pituitary/gonadal/adrenal provider, I often hear questions from colleagues about how to manage a patient’s diabetes post-operatively after cure from CS. While clinical experience has been helpful in guiding these discussions, the literature offered a paucity of data on diabetes/hyperglycemia specifically after surgery. There was also a lack of data on specific subgroups of CS, whether by sub-type or severity.

Therefore, we felt it was important to see what our past patient experiences showed in terms of changes in laboratory data, medications, and which patients were more likely to see improvement in their diabetes/hyperglycemia. The overall goal was to help clinicians provide appropriate patient education and care following a curative procedure.

In addition to its primary findings, the study also identified several factors associated with resolution or improvement of hyperglycemia. What were these factors?

Bancos: Both clinical and biochemical severity of CS, as well as Cushing subtype, were associated with improvement. We calculated severity based on symptoms and presence of comorbidities, and we calculated biochemical severity based on hormonal measurements. As clinical and biochemical scores were strongly correlated, we chose only one (biochemical) for multivariable analysis.

In the multivariable analysis of biochemical severity of Cushing, subtype of Cushing, and subtype of hyperglycemia, we found that patients with a severe biochemical severity score were 2.4 fold more likely to see improved hyperglycemia than people with a moderate or mild severity score (OR 2.4 (95% CI 1.1-4.9). We also found that patients with the nonadrenal CS subtype were 2.9 fold more likely to see improved hyperglycemia when compared to people with adrenal CS (OR of 2.9 (95% CI 1.3-6.4).

The type of hyperglycemia (diabetes versus prediabetes) was not found to be significant.

Did anything surprise you about the study results?

Herndon: I was surprised to see improvement in hyperglycemia in patients who were still on steroids, as you would expect the steroids to still have an impact. This shows how much a CS curative procedure truly leads to changes in the comorbidities that were a result of the underlying disease.

Also, I was surprised that the type of hyperglycemia was not a predictor of improvement after cure, although it was quite close. We also had a few patients whose hyperglycemia worsened, and we could not find a specific factor that predicted which patients did not improve.

What are the study’s implications for clinicians who treat people with CS?

Bancos: We think our study shows the clear need for closer follow-up — more frequently than the typical three-to-six months for diabetes. This can be accomplished through review of more than just HbA1c, such as reviewing blood glucose logbooks, asking about hypoglycemia symptoms, and so forth.

Patients with severe CS who are being treated with insulin or hypoglycemic medications are especially likely to decrease their medications to avoid hypoglycemia during postoperative period.

Read the study here.

Bancos reported advisory board participation and/or consulting with Strongbridge, Sparrow Pharmaceutics, Adrenas Therapeutics, and HRA Pharma outside the submitted work. Herndon did not disclose any relevant financial relationships with industry.

FDA Approval for Endogenous Cushing’s Syndrome Drug Recorlev

Ahead of its New Year’s Day decision deadline at the FDA, Xeris Biopharma has snagged an approval for Recorlev, a drug formerly known as levoketoconazole.

Based on results from phase 3 studies called SONICS and LOGICS, the FDA approved the drug for adults with Cushing’s syndrome. Xeris picked up Recorlev earlier this year in its acquisition of rare disease biotech Strongbridge Biopharma. It’s planning to launch in the first quarter of 2022.

Recorlev’s approval covers the treatment of endogenous hypercortisolemia in adults with Cushing’s syndrome who aren’t eligible for surgery or haven’t responded to surgery.

Endogenous Cushing’s disease is caused by a benign tumor in the pituitary gland that prompts the body to produce elevated levels of cortisol, which over time triggers a range of devastating physical and emotional symptoms for patients.


In the SONICS study, the drug significantly cut and normalized mean urinary free cortisol concentrations without a dose increase, according to the company. The LOGICS trial confirmed the drug’s efficacy and safety, Xeris says.

Cushion’s is a potentially fatal endocrine disease, and patients often experience years of symptoms before an accurate diagnosis, the company says. After a diagnosis, they’re presented with limited effective treatment options.

Following the approval, the company’s “experienced endocrinology-focused commercial organization can begin rapidly working to help address the needs of Cushing’s syndrome patients in the U.S. who are treated with prescription therapy,” Xeris CEO Paul R. Edick said in a statement.

Aside from its forthcoming Recorlev launch, Xeris markets Gvoke for severe hypoglycemia and Keveyis for primary periodic paralysis.

Back in October, the company partnered up with Merck to help reformulate some of the New Jersey pharma giant’s monoclonal antibody drugs.


Assessment of Vitamin D Metabolism in Patients with Cushing’s Disease

Endocrinology Research Centre, 117292 Moscow, Russia
Author to whom correspondence should be addressed.
Academic Editor: Spyridon N. Karras
Nutrients 202113(12), 4329;
Received: 12 November 2021 / Revised: 26 November 2021 / Accepted: 27 November 2021 / Published: 30 November 2021


In this study we aimed to assess vitamin D metabolism in patients with Cushing’s disease (CD) compared to healthy individuals in the setting of bolus cholecalciferol treatment. The study group included 30 adults with active CD and the control group included 30 apparently healthy adults with similar age, sex and BMI. All participants received a single dose (150,000 IU) of cholecalciferol aqueous solution orally. Laboratory assessments including serum vitamin D metabolites (25(OH)D3, 25(OH)D2, 1,25(OH)2D3, 3-epi-25(OH)D3 and 24,25(OH)2D3), free 25(OH)D, vitamin D-binding protein (DBP) and parathyroid hormone (PTH) as well as serum and urine biochemical parameters were performed before the intake and on Days 1, 3 and 7 after the administration. All data were analyzed with non-parametric statistics. Patients with CD had similar to healthy controls 25(OH)D3 levels (p > 0.05) and higher 25(OH)D3/24,25(OH)2D3 ratios (p < 0.05) throughout the study. They also had lower baseline free 25(OH)D levels (p < 0.05) despite similar DBP levels (p > 0.05) and lower albumin levels (p < 0.05); 24-h urinary free cortisol showed significant correlation with baseline 25(OH)D3/24,25(OH)2D3 ratio (r = 0.36, p < 0.05). The increase in 25(OH)D3 after cholecalciferol intake was similar in obese and non-obese states and lacked correlation with BMI (p > 0.05) among patients with CD, as opposed to the control group. Overall, patients with CD have a consistently higher 25(OH)D3/24,25(OH)2D3 ratio, which is indicative of a decrease in 24-hydroxylase activity. This altered activity of the principal vitamin D catabolism might influence the effectiveness of cholecalciferol treatment. The observed difference in baseline free 25(OH)D levels is not entirely clear and requires further study.

1. Introduction

Cushing’s disease (CD) is one of the disorders associated with endogenous hypercortisolism and is caused by adrenocorticotropic hormone (ACTH) hyperproduction originating from pituitary adenoma [1]. Skeletal fragility is a frequent complication of endogenous hypercortisolism, and fragility fractures may be the presenting clinical feature of disease. The prevalence of osteoporosis in endogenous hypercortisolism as assessed by dual-energy X-ray absorptiometry (DXA) or incidence of fragility fractures has been reported to be up to 50%. Osteoporosis in CD patients has a complex multifactorial pathogenesis, characterized by a low bone turnover and severe suppression of bone formation [2]. Exogenous glucocorticoids are used in the treatment of a wide range of diseases and it is estimated that 1–2% of the population is receiving long-term glucocorticoid therapy. As a consequence, glucocorticoid-induced osteoporosis is the most common secondary cause of osteoporosis [3].
Native vitamin D (in particular D3, or cholecalciferol) and its active metabolites (such as alfacalcidol) are universally considered as the essential components of the osteoporosis management [4,5]. The search for the optimal treatment of bone complications during chronic exposure to glucocorticoid excess provoked the investigation of vitamin D metabolism in this state. Early studies on this topic were focused predominantly on the general vitamin D status (assessed as 25(OH)D level) and on the levels of the active vitamin D metabolite (1,25(OH)2D). These studies showed inconsistent results, reporting that the chronic excess of glucocorticoids decreased [6,7,8,9], increased [10,11,12] or did not change [13,14,15] the levels of 25(OH)D or 1,25(OH)2D. A likely reason for such inconsistency might have been the high heterogeneity of the studied groups. Some of these studies were performed in humans [6,7,9,10,11,12,13,15] and some in animal models [8,14], and only several of them included subjects with specifically endogenous hypercortisolism [10,12,14,15]. Only two studies assessed both the levels of the active (1,25(OH)2D) and the inactive (24,25(OH)2D) vitamin D metabolites in endogenous hypercortisolism. One of them lacked control group and reported low-normal 24,25(OH)2D levels in patients with Cushing’s syndrome [10]. The second study by Corbee et al. reported similar circulating concentrations of 25(OH)D, 1,25(OH)2D and 24,25(OH)2D in studied groups of dogs regardless of either the presence of CD or hypophysectomy status [14].
Several experimental studies were performed to evaluate the impact of glucocorticoid excess on the enzymes involved in vitamin D metabolism. In mouse kidney glucocorticoid treatment increased 24-hydroxylase expression [16] and 24-hydroxylase activity [17]. An increased expression of 24-hydroxylase was also shown in rat osteoblastic and pig renal cell cultures treated with 1,25(OH)2D [18]. Dhawan and Christakos showed that 1,25(OH)2D-induced transcription of 24-hydroxylase was glucocorticoid receptor-dependent [19]. However, some works showed conflicting results. In particular, the steroid and xenobiotic receptor (SXR) which is activated by glucocorticoids [20], repressed 24-hydroxylase expression in human liver and intestine in work by Zhou et al. [21]. Lower 24-hydroxylase expression was observed in the brain and myocardium of glucocorticoid-treated rats [22] as well as in human osteosarcoma cells and human osteoblasts [23].
Nevertheless, based on experimental data, it has been suggested that the acceleration of 25(OH)D catabolism in the presence of glucocorticoid excess may predispose to vitamin D deficiency. Yet, relatively recent meta-analysis of the studies assessing 25(OH)D levels in chronic glucocorticoid users showed that serum 25(OH)D levels in these patients were suboptimal and lower than in healthy controls, but similar to steroid-naive disease controls [24].
Glucocorticoids also affect calcium and phosphorus homeostasis. In particular, they were shown to reduce gastrointestinal absorption by antagonizing vitamin D action (reducing the expression of genes for proteins involved in calcium transport—epithelial Ca channel TRPV6 and calcium-binding protein calbindin-D9K) [25]. Glucocorticoids increased fractional calcium excretion due to mineralocorticoid receptor-mediated action on epithelial sodium channels [26]. Hypercalciuria is highly prevalent in people with CD [27]. These effects might result in a negative calcium balance, although plasma ionized calcium was normal in people and dogs with hypercortisolism compared to control subjects [12,28]. Glucocorticoids also reduced tubular phosphate reabsorption by inhibiting tubular expression of the sodium gradient-dependent phosphate transporter, and induced phosphaturia [29], which was accompanied by phosphate lowering in humans [12].
Overall, current data on vitamin D status in hypercortisolism are conflicting and need clarification. In particular, clinical data on the state of vitamin D metabolism in the state of glucocorticoids excess are quite scarce. Studies were very heterogeneous in design, some lacked a control group, and the absolute majority of the studies were performed before the introduction of vitamin D measurement standardization [30]. Nevertheless, determining the optimal vitamin D treatment regimen in these high-risk patients is fairly relevant.
The aim of this study was to assess vitamin D metabolism in patients with CD compared to healthy individuals particularly in the setting of cholecalciferol treatment.

2. Materials and Methods

2.1. Study Population and Design

The study group included 30 adult patients with CD admitted for inpatient treatment at a tertiary pituitary center. Diagnosis of CD was established in accordance with the federal guidelines [31]. All patients were confirmed to be positive for endogenous hypercortisolism in at least two of the following tests: 24-h urine free cortisol (UFC) greater than the normal range for the assay and/or serum cortisol > 50 nmol/L after the 1-mg overnight dexamethasone suppression test and/or late-night salivary cortisol greater than 9.4 nmol/L). All patients also had morning ACTH ≥ 10 pg/mL and pituitary adenoma ≥ 6 mm identified by magnetic resonance imaging (MRI) or a positive for CD bilateral inferior petrosal sinus sampling (BIPSS). MRI was performed using a GE Optima MR450w 1.5T with Gadolinium (Boston, MA, USA). BIPSS was performed according to the standard procedure described elsewhere [32,33].
The control group included 30 apparently healthy adult individuals recruited from the staff and the faculty of the facility.
Inclusion criteria were age from 18 to 60 for both groups and the presence of the disease activity for the study group (defined as the presence of endogenous hypercortisolism at the time of participation in the study). Exclusion criteria for both groups were: vitamin D supplementation for 3 months prior to the study; severe obesity (body mass index (BMI) ≥ 35 kg/m2); pregnancy; the presence of granulomatous disease, malabsorption syndrome, liver failure; decreased GFR (less than 60 mL/min per 1.73 m2); severe hypercalcemia (total serum calcium > 3.0 mmol/L); allergic reactions to vitamin D medications; 25(OH)D level more than 60 ng/mL (determined by immunochemiluminescence analysis). All patients were recruited in the period from October 2019 to April 2021. The study protocol ( Identifier: NCT04844164) was approved by the Ethics Committee of Endocrinology Research Centre, Moscow, Russia on 10 April 2019 (abstract of record No. 6), all patients signed informed consent to participate in the study.
All participants received standard therapeutic dose (150,000 IU) of an aqueous solution of cholecalciferol (Aquadetrim®, Medana Pharma S.A., Sieradz, Poland) orally as a single dose [34]. Blood and urine samples were obtained before the intake as well as on days 1, 3 and 7 after administration; time points of sample collection were determined based on the authors’ previous work evaluating changes in 25(OH)D levels after a therapeutic dose of cholecalciferol [35]. The assessment included serum biochemical parameters (total calcium, albumin, phosphorus, creatinine, magnesium), parathyroid hormone (PTH), vitamin D-binding protein (DBP), vitamin D metabolites (25(OH)D3, 25(OH)D2, 1,25(OH)2D3, 3-epi-25(OH)D3 and 24,25(OH)2D3), free 25(OH)D and urine biochemical parameters (calcium- and phosphorus-creatinine ratios in spot urine).

2.2. Socio–Demographic and Anthropometric Data Collection

At the baseline visit, patients underwent a questionnaire aimed to assess their lifestyle: the presence of unhealthy habits, physical activity level, balanced diet (consumption of dairy products, meat, coffee, soft drinks), exposure to ultraviolet (UV) radiation (solarium and sunscreen usage, traveling south and the number of daytime walks in the sunny weather in the 3 months preceding study participation). Smoking status was classified as current smoker, former smoker and non-smoker; current and former smokers were collectively referred to as total smokers. A unit of alcohol was defined as a glass of wine, a bottle of beer or a shot of spirits, approximating 10–12 g ethanol. Serving of dairy products was defined as 100 g of cottage cheese, 200 mL of milk, 125 g of yogurt or 30 g of cheese. Patients’ weight was measured in light indoor clothing with a medical scale to the nearest 100 g, and their height with a wall-mounted stadiometer to the nearest centimeter. BMI was calculated as weight in kilograms divided by height in meters squared.

2.3. Laboratory Measurements

Morning ACTH (reference range 7–66 pg/mL), serum cortisol after a low-dose dexamethasone suppression test (cutoff value for suppression, 50 nmol/L [36]), late-night salivary cortisol (reference range 0.5–9.4 nmol/L [37]) were assayed by electrochemiluminescence assay using a Cobas 6000 Module e601 (Roche, Rotkreuz, Switzerland). The 24-h UFC (reference range 60–413 nmol/24 h) was measured by an immunochemiluminescence assay (extraction with diethyl ether) on a Vitros ECiQ (Ortho Clinical Diagnostics, Raritan, NJ, USA).
Total 25(OH)D levels (25(OH)D2 + 25(OH)D3; reference range 30–100 ng/mL) at the baseline visit were determined by the immunochemiluminescence analysis (Liaison, DiaSorin, Saluggia, Italy). PTH levels were evaluated by the electrochemiluminescence immunoassay (ELECSYS, Roche, Basel, Switzerland; reference range for this and subsequent laboratory parameters are given in the Results section for easier reading). Biochemical parameters of blood serum and urine were assessed by the ARCHITECT c8000 analyzer (Abbott, Chicago, IL, USA) using reagents from the same manufacturer according to the standard methods. Serum DBP and free 25(OH)D levels were measured by enzyme-linked immunosorbent assay (ELISA) using commercial kits. The assay used for free 25(OH)D levels assessment (DIAsource, ImmunoAssays S.A., Ottignies-Louvain-la-Neuve, Belgium) has <6.2% intra- and inter-assay coefficient of variation (CV) at levels 5.8–9.6 pg/mL. The assay used for DBP levels assessment (Assaypro, St Charles, MO, USA) has 6.2% average intra-assay CV and 9.9% average inter-assay CV.
The levels of vitamin D metabolites (25(OH)D3, 25(OH)D2, 1,25(OH)2D3, 3-epi-25(OH)D3 and 24,25(OH)2D3) in serum were determined by ultra-high performance liquid chromatography in combination with tandem mass spectrometry (UPLC-MS/MS) using an in-house developed method, described earlier [38]. With this technique, the laboratory participates in DEQAS quality assurance program (lab code 2388) and the results fall within the target range for the analysis of 25(OH)D and 1,25(OH)2D metabolites in human serum (Supporting Information, Figures S1 and S2). All UPLC-MS/MS measurements were made after the first successful completion (5/5 samples within the target range) of the DEQAS distributions for both analytes simultaneously. Each batch contained control samples (analytes in blank serum) with both high and low analyte concentrations. The samples were barcoded and randomized prior to the measurements to eliminate analyst-related errors.
Serum samples (3 aliquots) collected at each visit were either transferred directly to the laboratory for biochemical analyzes, total 25(OH)D and PTH measurement (1 aliquot) or were stored at −80 °C avoiding repeated freeze-thaw cycles for measurement of DBP, free 25(OH)D and vitamin D metabolites at a later date (2 aliquots).
Albumin-adjusted serum calcium levels were calculated using the formula [39]: total plasma calcium (mmol/L) = measured total plasma calcium (mmol/L) + 0.02 × (40 − measured plasma albumin (g/L)).
Baseline free 25(OH)D levels were also calculated using the formula introduced by Bikle et al. [40,41]. The affinity constant for 25(OH)D and albumin binding (Kalb) used for the calculation was equal 6 × 105 M−1, and affinity constant for 25(OH)D and DBP binding (KDBP) was equal 7 × 108 M−1.

Free 25(OH)D=total 25(OH)D1+Kalbalbumin+KDBPDBP

2.4. Statistical Analysis

Statistical analysis was performed using Statistica version 13.0 (StatSoft, Tulsa, OK, USA). All data were analyzed with non-parametric statistics and expressed as median [interquartile range] unless otherwise specified. Mann-Whitney U-test and Fisher’s exact two-tailed test were used for comparisons between two groups. Friedman ANOVA was performed to evaluate changes in indices throughout the study and pairwise comparisons using Wilcoxon test with adjustment for multiple comparisons (Bonferroni) were also made if the Friedman ANOVA was significant. Spearman rank correlation method was used to obtain correlation coefficients among indices. A p-value of less than 0.05 was considered statistically significant. When adjusting for multiple comparisons, a p-value greater than the significance threshold, but less than 0.05 was considered as a trend towards statistical significance.

3. Results

The groups were similar in terms of age, sex and BMI (p > 0.05). Both groups consisted predominantly of young and middle-aged women and the majority of patients were overweight or moderately obese (Table 1). Patients from the study group presented with lower screening levels of total 25(OH)D (p < 0.05).
Table 1. General characteristics of the patients at the baseline visits. For detailed description of the data format please refer to the Section 2.
The features of the underlying disease course in the study group are listed in Table 2. 15 patients (50%) had diabetes mellitus with an almost compensated state at the time of participation in the study, and 7 patients (23%) reported a history of low-energy fractures.
Table 2. Characteristics of the patients with Cushing’s disease (CD) in terms of the underlying disease.
The groups did not differ significantly in the reported smoking status, the level of daily physical activity, dietary habits and UV exposure (p > 0.05) and although there was a slight difference in alcohol consumption (p < 0.05), the absolute values were minor in both groups (Table 3).
Table 3. Questionnaire results.

3.1. Baseline Laboratory Evaluation

Detailed results of laboratory studies are presented in Table 4 and Table 5.
Table 4. Changes in the levels of the biochemical parameters and parathyroid hormone (PTH) during the study.
Table 5. Changes in the levels of free 25(OH)D, vitamin D-binding protein (DBP) and vitamin D metabolites during the study.
Patients with CD had several alterations in biochemical parameters, in particular, lower baseline serum creatinine and albumin levels, while magnesium levels were higher than in the control group (p < 0.05). They also had higher levels of urine phosphorus-creatinine ratio (p < 0.05). The rest of the studied biochemical parameters did not show significant difference between the groups (p > 0.05). 3 patients (10%) from the study group and 5 patients (17%) from the control group had secondary hyperparathyroidism, one patient with CD (3%) was diagnosed with mild primary hyperparathyroidism.
As for the assessment of vitamin D metabolism, unexpectedly the levels of 25(OH)D3 occurred to be equal in the groups (p > 0.05), with only two patients (7%) from the study group and one patient (3%) from the control group having sufficient vitamin D levels, according to the Endocrine Society and the Russian Association of Endocrinologists guidelines (≥30 ng/mL [34,42]). The levels of the active vitamin D metabolite—1,25(OH)2D3—were equal between the groups as well (p > 0.05), whereas the levels of 3-epi-25(OH)D3 and 24,25(OH)2D3 were lower in CD patients. Further calculation of 25(OH)D3/24,25(OH)2D3 and 25(OH)D3/1,25(OH)2D3 ratios corresponded to the observed levels of metabolites: 25(OH)D3/24,25(OH)2D3 ratio was higher in the study group (p < 0.05) assuming lower 24-hydroxylase activity and 25(OH)D3/1,25(OH)2D3 ratio was equal between the groups (p > 0.05).
Levels of free 25(OH)D were lower in CD patients (p < 0.05) and the levels of DBP did not differ between the groups (p > 0.05). Although calculated free 25(OH)D showed prominent positive correlation with the measured free 25(OH)D in both groups (r = 0.63 in the study group, r = 0.87 in the control group, p < 0.05), the association appeared to be weaker in the study group. In the control group, DBP levels correlated with both measured and calculated 25(OH)D levels (r = −0.48, p < 0.05 and r = −0.69, p < 0.05 respectively), while in patients with CD there was no association with measured free 25(OH)D levels (r = 0.04, p > 0.05 and r = −0.50, p < 0.05 respectively).
Correlation with 24-h UFC in CD patients was observed for serum albumin level (r = −0.37, p < 0.05) and urine calcium-creatinine ratio (r = 0.51, p < 0.05) among assessed biochemical parameters, and only with 25(OH)D3/24,25(OH)2D3 ratio among the parameters of vitamin D metabolism (r = 0.36, p < 0.05).

3.2. Laboratory Evaluation after the Intake of Cholecalciferol

All patients from the study group and 28 patients (93%) from the control group completed the study.
The observed baseline differences in biochemical parameters mostly preserved during the follow-up. In the study group there was an increase in serum phosphorus levels by Day 1 (p = 0.006) and a tendency to an increase in the urine phosphorus-creatinine ratio by Day 7 (p = 0.02). Patients from the control group showed a clinically insignificant increase in serum creatinine levels by Day 1 (p = 0.002) and a non-significant trend towards an increase in serum total and albumin-adjusted calcium (p = 0.01 for both measurements). No change in PTH levels was observed in patients with CD during the follow-up (p > 0.05), while in the control group there was a tendency for PTH to decrease by Day 3 (p = 0.02). There were no new cases of hypercalcemia in both groups during the follow-up. One patient from the study group and one patient from the control group had persistently increased urine calcium-creatinine ratio throughout the study. Four patients from the study group (13%) and none from the control group developed hypercalciuria during the follow-up, however these patients had no clinical manifestations during the observation period.
By Day 7, 25 patients (83%) from the study group and 22 patients (79%) reached sufficient 25(OH)D3 levels (≥30 ng/mL). Levels of 25(OH)D3 continued to increase by Day 3 in both groups (p < 0.001), after which tended to decrease in the study group (p = 0.01) and remained stable in the control group (p = 0.65). The increase in 25(OH)D3 after cholecalciferol intake was equal between the groups (18.5 [15.9; 22.5] ng/mL in the study group vs. 16.6 [13.1; 19.8] ng/mL in the control group, p > 0.05). In the presence of obesity, Δ25(OH)D3 was higher in the CD patients than in the control group (18.3 [14.2; 23.0] vs. 12.1 [10.0; 13.1] ng/mL, p < 0.05), while in non-obese patients no difference was observed (p > 0.05).
Obese and non-obese patients with CD had equal Δ25(OH)D3 (18.3 [14.2; 23.0] vs. 19.6 [16.0; 21.5] ng/mL, p > 0.05), while in the control group it was significantly lower in obese patients (12.1 [10.0; 13.1] vs. 18.3 [15.3; 21.4] ng/mL, p < 0.05). BMI showed significant correlation with Δ25(OH)D3 only in the control group (r = −0.47, p < 0.05), while in CD patients there was no such association (r = −0.06, p > 0.05) (Figure 1).
Figure 1. Relationship between Δ25(OH)D3 and BMI in groups.
1,25(OH)2D3 levels increased in CD patients by Day 1 and were stable during the follow-up in the control group. The rest of the studied parameters of vitamin D metabolism changed in a similar way between groups: 3-epi-25(OH)D3 levels increased until the Day 3, after which they decreased by the Day 7; 24,25(OH)2D3 levels showed more graduate elevation throughout the follow-up. In both groups 25(OH)D3/24,25(OH)2D3 ratios increased by Day 1, after which they decreased by Day 7, and 25(OH)D3/1,25(OH)2D3 ratios increased by Day 1, after which they remained stable. DBP levels didn’t change and free 25(OH)D levels showed an increase in both groups during the follow-up. The levels of 25(OH)D2 did not exceed 0.5 ng/mL in all examined individuals throughout the study. Among assessed parameters of vitamin D metabolism, higher 25(OH)D3/24,25(OH)2D3 ratios in the study group was the only difference between the groups which remained significant throughout the observation period (p < 0.05) (Figure 2).
Figure 2. Dynamic evaluation of 25(OH)D3/24,25(OH)2D3 ratios in groups.

4. Discussion

The main goal of our study was to evaluate the 25(OH)D3 levels and its response to the therapeutic dose of cholecalciferol in patients with CD as compared to healthy individuals. We observed no difference in baseline 25(OH)D3 assessed by UPLC-MS/MS between groups. Similar to our data were obtained in most studies conducted specifically in the state of endogenous hypercortisolism in humans [12,15] and dogs [14]. The study by Kugai et al. lacked control group and reported plasma levels of 25(OH)D corresponding to the vitamin D deficiency in most of the examined patients [10], while in our study only 2/3 of the patients with CD had 25(OH)D levels below 20 ng/mL. As for exogenous hypercortisolism, the meta-analysis aimed to explore serum 25(OH)D levels in glucocorticoid users showed lower levels than in healthy controls, but similar to steroid-naive disease controls, thus causing concern regarding the influence of the disease status on 25(OH)D levels [24]. Somewhat surprisingly, we obtained significantly discordant results in the study group when screening total 25(OH)D by ELISA and when measuring baseline 25(OH)D3 by UPLC-MS/MS, since the initial difference between the groups revealed by ELISA data with lower total 25(OH)D levels in the study group was not replicated by UPLC-MS/MS. It should be noted that our ELISA method did not participate in an external quality control program at the time of the study unlike UPLC-MS/MS; furthermore, a lower analytical performance was previously described for this technique with tendency for low specificity and lower measurement results [45].
When assessing other parameters of vitamin D metabolism, the most significant finding was the higher 25(OH)D3/24,25(OH)2D3 ratio in CD patients, both initially and during the observation after the intake of the cholecalciferol loading dose, indicating consistently reduced activity of 24-hydroxylase, the main enzyme of vitamin D catabolism. Earlier clinical and experimental studies also suggested altered activity of enzymes of vitamin D metabolism in hypercortisolism. However, these studies were heterogeneous and aimed predominantly at studying the activity of 1α-hydroxylase [7,8,10,11,12,14], which was not altered in patients with CD as compared to healthy individuals in our study. In the setting of the short-term glucocorticoid administration, Lindgren et al. showed transient increase in 24,25(OH)2D3 levels in rats [8], while in the study of Hahn et al. there was no change in 24,25(OH)2D3 levels [11]. Dogs with CD had similar 24,25(OH)2D3 levels before and after hypophysectomy as well as compared to control dogs [14]. The only study of considerably similar design by Kugai et al. reported low-normal 24,25(OH)2D3 in patients with Cushing’s syndrome [10], which is consistent with our result, as well as some experimental works indicative of suppression on CYP24A1 expression by glucocorticoids in human osteoblasts [23], liver and intestine [21] and in rat brain and myocardium [22]. However, in the present work, the activity of 24-hydroxylase in patients with hypercortisolism was for the first time evaluated by calculating the 25(OH)D3/24,25(OH)2D3 ratio, which has recently emerged as a new tool for vitamin D status assessment [46,47]. Given the correlation of this parameter with laboratory marker of the underlying disease activity (24-h UFC), a direct effect of cortisol overproduction on 24-hydroxylase activity might be assumed. Interestingly, it seems that the decreased activity of 24-hydroxylase observed in CD influenced the effectiveness of cholecalciferol treatment, decreasing the negative effect of obesity, as patients with CD had similar increase in 25(OH)D3 in obese and non-obese state and lacked correlation between Δ25(OH)D3 and BMI, as opposed to the control group. Moreover, the increase in 25(OH)D3 in obese patients from the control group was lower not only than in non-obese controls, but also than in obese patients with CD.
Another intriguing finding was lower levels of free 25(OH)D observed in patients with CD despite similar DBP levels and lower albumin levels, which, on the contrary, allows one to expect higher values of free 25(OH)D. Considering the weaker correlation between the measured and calculated free 25(OH)D in patients with CD, as well as the lack of correlation of the measured 25(OH)D with the main transport protein, an altered affinity of DBP might be suspected. One possible explanation is protein glycosylation as a consequence of diabetes mellitus, which was present in half of the patients [38,48,49]. After cholecalciferol intake, which was accompanied by an increase in free 25(OH)D, the differences between the groups were leveled; therefore, another suggested explanation might be competitive binding to the ligand. Since actin binds DBP with high affinity [50] and considering catabolic action of glucocorticoids on muscle tissue [51], actin is a presumable competing ligand candidate. Although this is mostly speculative, as far as the authors are aware, the present work was the first to assess free vitamin D in the glucocorticoid excess, so the described findings require verification of reproducibility and further evaluation.
The obtained discrepancies in the biochemical parameters characterizing calcium and phosphorus metabolism were generally consistent with the data of early studies discussed in the introduction [12,25,26,27,28,29], except for similar to controls serum phosphorus levels and lower prevalence of hypercalciuria. An interesting observation was the complete absence of the PTH decrease in patients with CD after receiving a loading dose of cholecalciferol. The mechanism of this phenomenon is not entirely clear, we tend to agree with the earlier hypothesis that this may be an adaptation to chronic urinary calcium loss [52].
Our research is distinguished by a number of important strengths: a prospective design, substantial sample of patients with CD, accounting for social and behavioral factors affecting vitamin levels D, comprehensive spectrum of vitamin D metabolism parameters investigated and participation in an external quality control program for vitamin D metabolites measurement.
Nevertheless, the study also had several limitations: the amount of dietary vitamin D and phosphorus, as well as possible differences in DBP affinity to vitamin D metabolites due to genetic isoforms of DBP [53] or other possible involved parameters (e.g., fibroblast growth factor-23) were not taken into account. A few patients from both groups received therapy with possible impact on vitamin D and calcium metabolism within 3 months preceding the participation in the study (spironolactone, diuretics, proton pump inhibitors, oral contraceptives, antifungal treatment, antidepressants, barbiturates, antiepileptic drugs). The groups had a trend for differences in sex and BMI (p = 0.07 for both parameters). Also, the study lacked a study group of patients with remission of CD to test the hypotheses put forward, however, this is a promising direction for further research.

5. Conclusions

We report that patients with endogenous ACTH-dependent hypercortisolism of pituitary origin have a consistently higher 25(OH)D3/24,25(OH)2D3 ratio than healthy controls, which is indicative of a decrease in 24-hydroxylase activity. This altered activity of the principal vitamin D catabolism might influence the effectiveness of cholecalciferol treatment. There is also a lack of clarity regarding the lower levels of free 25(OH)D observed in patients with CD, which require further study. To test the proposed hypotheses and to develop specialized clinical guidelines for these patients, longer-term randomized clinical trials are needed.

Supplementary Materials

The following are available online at, Method validation against DEQAS, Figure S1: Comparison between DEQAS data for 25(OH)D scheme and our lab results, Figure S2: Comparison between DEQAS data for 1,25(OH)2D scheme and our lab results.

Author Contributions

Conceptualization, L.R., E.P., A.P. and A.Z.; methodology, V.B., Z.B., L.R. and G.M.; formal analysis, A.P.; investigation, A.P., V.B., E.P., L.D. and A.Z.; data curation, A.P. and V.B.; writing—original draft preparation, A.P.; writing—review and editing, V.B., E.P., A.Z., Z.B., L.R.; visualization, V.B.; supervision, L.D., L.R., G.M. and N.M.; project administration, L.R. and N.M.; funding acquisition, L.R. and N.M. All authors have read and agreed to the published version of the manuscript.


This research was funded by the Russian Science Foundation, grant number 19-15-00243.

Institutional Review Board Statement

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Endocrinology Research Centre, Moscow, Russia on 10 April 2019 (abstract of record No. 6).

Informed Consent Statement

Written informed consent was obtained from all individual participants included in the study.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.


We express our deep gratitude to our colleagues: Natalya M. Malysheva, Vitaliy A. Ioutsi, Larisa V. Nikankina for the help with the laboratory research.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.


  1. Nishioka, H.; Yamada, S. Cushing’s disease. J. Clin. Med. 20198, 1951. [Google Scholar] [CrossRef]
  2. Mazziotti, G.; Frara, S.; Giustina, A. Pituitary Diseases and Bone. Endocr. Rev. 201839, 440–488. [Google Scholar] [CrossRef] [PubMed]
  3. Compston, J. Glucocorticoid-induced osteoporosis: An update. Endocrine 201861, 7–16. [Google Scholar] [CrossRef] [PubMed]
  4. Buckley, L.; Guyatt, G.; Fink, H.A.; Cannon, M.; Grossman, J.; Hansen, K.E.; Humphrey, M.B.; Lane, N.E.; Magrey, M.; Miller, M. 2017 American College of Rheumatology Guideline for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Care Res. 201769, 1095–1110. [Google Scholar] [CrossRef]
  5. Belaya, Z.; Rozhinskaya, L.y.; Grebennikova, T.; Kanis, J.; Pigarova, E.; Rodionova, S.; Toroptsova, N.; Nikitinskaya, O.; Skripnikova, I.; Drapkina, O.; et al. Summary of the Draft Federal Clinical Guidelines for Osteoporosis. Osteoporos. Bone Dis. 202023, 4–21. [Google Scholar] [CrossRef]
  6. Klein, R.G.; Arnaud, S.B.; Gallagher, J.C.; Deluca, H.F.; Riggs, B.L. Intestinal calcium absorption in exogenous hypercortisolism. J. Clin. Investig. 197760, 253–259. [Google Scholar] [CrossRef] [PubMed]
  7. Seeman, E.; Kumar, R.; Hunder, G.G.; Scott, M.; Iii, H.H.; Riggs, B.L. Production, degradation, and circulating levels of 1,25-dihydroxyvitamin D in health and in chronic glucocorticoid excess. J. Clin. Investig. 198066, 664–669. [Google Scholar] [CrossRef]
  8. Lindgren, J.U.; Merchant, C.R.; DeLuca, H.F. Effect of 1,25-dihydroxyvitamin D3 on osteopenia induced by prednisolone in adult rats. Calcif. Tissue Int. 198234, 253–257. [Google Scholar] [CrossRef]
  9. Chaiamnuay, S.; Chailurkit, L.O.; Narongroeknawin, P.; Asavatanabodee, P.; Laohajaroensombat, S.; Chaiamnuay, P. Current daily glucocorticoid use and serum creatinine levels are associated with lower 25(OH) vitamin D levels in thai patients with systemic lupus erythematosus. J. Clin. Rheumatol. 201319, 121–125. [Google Scholar] [CrossRef]
  10. Kugai, N.; Koide, Y.; Yamashita, K.; Shimauchi, T.; Nagata, N.; Takatani, O. Impaired mineral metabolism in Cushing’s syndrome: Parathyroid function, vitamin D metabolites and osteopenia. Endocrinol. Jpn. 198633, 345–352. [Google Scholar] [CrossRef]
  11. Hahn, T.J.; Halstead, L.R.; Baran, D.T. Effects of short term glucocorticoid administration on intestinal calcium absorption and circulating vitamin D metabolite concentrations in man. J. Clin. Endocrinol. Metab. 198152, 111–115. [Google Scholar] [CrossRef]
  12. Findling, J.W.; Adams, N.D.; Lemann, J.; Gray, R.W.; Thomas, C.J.; Tyrrell, J.B. Vitamin D metabolites and parathyroid hormone in Cushing’s syndrome: Relationship to calcium and phosphorus homeostasis. J. Clin. Endocrinol. Metab. 198254, 1039–1044. [Google Scholar] [CrossRef] [PubMed]
  13. Slovik, D.M.; Neer, R.M.; Ohman, J.L.; Lowell, F.C.; Clark, M.B.; Segre, G.V.; Potts, J.T., Jr. Parathyroid hormone and 25-hydroxyvitamin D levels in glucocorticoid-treated patients. Clin. Endocrinol. 198012, 243–248. [Google Scholar] [CrossRef]
  14. Corbee, R.J.; Tryfonidou, M.A.; Meij, B.P.; Kooistra, H.S.; Hazewinkel, H.A.W. Vitamin D status before and after hypophysectomy in dogs with pituitary-dependent hypercortisolism. Domest. Anim. Endocrinol. 201242, 43–49. [Google Scholar] [CrossRef] [PubMed]
  15. Aloia, J.F.; Roginsky, M.; Ellis, K.; Shukla, K.; Cohn, S. Skeletal metabolism and body composition in Cushing’s syndrome. J. Clin. Endocrinol. Metab. 197439, 981–985. [Google Scholar] [CrossRef]
  16. Van Cromphaut, S.J.; Stockmans, I.; Torrekens, S.; Van Herck, E.; Carmeliet, G.; Bouillon, R. Duodenal calcium absorption in dexamethasone-treated mice: Functional and molecular aspects. Arch. Biochem. Biophys. 2007460, 300–305. [Google Scholar] [CrossRef] [PubMed]
  17. Akeno, N.; Matsunuma, A.; Maeda, T.; Kawane, T.; Horiuchi, N. Regulation of vitamin D-1-hydroxylase and -24-hydroxylase expression by dexamethasone in mouse kidney. J. Endocrinol. 2000164, 339–348. [Google Scholar] [CrossRef] [PubMed]
  18. Kurahashi, I.; Matsunuma, A.; Kawane, T.; Abe, M.; Horiuchi, N. Dexamethasone enhances vitamin D-24-hydroxylase expression in osteoblastic (UMR-106) and renal (LLC-PK 1) cells treated with 1a, 25-dihydroxyvitamin D3. Endocrine 200217, 109–118. [Google Scholar] [CrossRef]
  19. Dhawan, P.; Christakos, S. Novel regulation of 25-hydroxyvitamin D3 24-hydroxylase (24(OH)ase) transcription by glucocorticoids: Cooperative effects of the glucocorticoid receptor, C/EBPb, and the vitamin D receptor in 24(OH)ase transcription. J. Cell. Biochem. 2010110, 1314–1323. [Google Scholar] [CrossRef] [PubMed]
  20. Luo, G.; Cunningham, M.; Kim, S.; Burn, T.; Lin, J.; Sinz, M.; Hamilton, G.; Rizzo, C.; Jolley, S.; Gilbert, D.; et al. CYP3A4 induction by drugs: Correlation between a pregnane X receptor reporter gene assay and CYP3A4 expression in human hepatocytes. Drug Metab. Dispos. 200230, 795–804. [Google Scholar] [CrossRef]
  21. Zhou, C.; Assem, M.; Tay, J.C.; Watkins, P.B.; Blumberg, B.; Schuetz, E.G.; Thummel, K.E. Steroid and xenobiotic receptor and vitamin D receptor crosstalk mediates CYP24 expression and drug-induced osteomalacia. J. Clin. Investig. 2006116, 1703–1712. [Google Scholar] [CrossRef] [PubMed]
  22. Jiang, P.; Xue, Y.; Li, H.; Liu, Y. Dysregulation of vitamin D metabolism in the brain and myocardium of rats following prolonged exposure to dexamethasone. Psychopharmacology 2014231, 3445–3451. [Google Scholar] [CrossRef] [PubMed]
  23. Zayny, A.; Almokhtar, M.; Wikvall, K.; Ljunggren, Ö.; Ubhayasekera, K.; Bergquist, J.; Kibar, P.; Norlin, M. Effects of glucocorticoids on vitamin D3-metabolizing 24-hydroxylase (CYP24A1) in Saos-2 cells and primary human osteoblasts. Mol. Cell. Endocrinol. 2019496, 110525. [Google Scholar] [CrossRef] [PubMed]
  24. Davidson, Z.E.; Walker, K.Z.; Truby, H. Do glucocorticosteroids alter vitamin D status? A systematic review with meta-analyses of observational studies. J. Clin. Endocrinol. Metab. 201497, 738–744. [Google Scholar] [CrossRef] [PubMed]
  25. Huybers, S.; Naber, T.H.J.; Bindels, R.J.M.; Hoenderop, J.G.J. Prednisolone-induced Ca2+ malabsorption is caused by diminished expression of the epithelial Ca2+ channel TRPV6. Am. J. Physiol.-Gastrointest. Liver Physiol. 2007292, 92–97. [Google Scholar] [CrossRef] [PubMed]
  26. Ferrari, P.; Bianchetti, M.G.; Sansonnens, A.; Frey, F.J. Modulation of renal calcium handling by 11β-hydroxysteroid dehydrogenase type 2. J. Am. Soc. Nephrol. 200213, 2540–2546. [Google Scholar] [CrossRef]
  27. Faggiano, A.; Pivonello, R.; Melis, D.; Filippella, M.; Di Somma, C.; Petretta, M.; Lombardi, G.; Colao, A. Nephrolithiasis in Cushing’s disease: Prevalence, etiopathogenesis, and modification after disease cure. J. Clin. Endocrinol. Metab. 200388, 2076–2080. [Google Scholar] [CrossRef] [PubMed]
  28. Ramsey, I.K.; Tebb, A.; Harris, E.; Evans, H.; Herrtage, M.E. Hyperparathyroidism in dogs with hyperadrenocorticism. J. Small Anim. Pract. 200546, 531–536. [Google Scholar] [CrossRef]
  29. Freiberg, J.M.; Kinsella, J.; Sacktor, B. Glucocorticoids increase the Na+-H+ exchange and decrease the Na+ gradient-dependent phosphate-uptake systems in renal brush border membrane vesicles. Proc. Natl. Acad. Sci. USA 198279, 4932–4936. [Google Scholar] [CrossRef] [PubMed]
  30. Sempos, C.T.; Heijboer, A.C.; Bikle, D.D.; Bollerslev, J.; Bouillon, R.; Brannon, P.M.; DeLuca, H.F.; Jones, G.; Munns, C.F.; Bilezikian, J.P.; et al. Vitamin D assays and the definition of hypovitaminosis D: Results from the First International Conference on Controversies in Vitamin D. Br. J. Clin. Pharmacol. 201884, 2194–2207. [Google Scholar] [CrossRef]
  31. Melnichenko, G.A.; Dedov, I.I.; Belaya, Z.E.; Rozhinskaya, L.Y.; Vagapova, G.R.; Volkova, N.I.; Grigor’ev, A.Y.; Grineva, E.N.; Marova, E.I.; Mkrtumayn, A.M.; et al. Cushing’s disease: The clinical features, diagnostics, differential diagnostics, and methods of treatment. Probl. Endocrinol. 201561, 55–77. [Google Scholar] [CrossRef]
  32. Machado, M.C.; De Sa, S.V.; Domenice, S.; Fragoso, M.C.B.V.; Puglia, P.; Pereira, M.A.A.; De Mendonça, B.B.; Salgado, L.R. The role of desmopressin in bilateral and simultaneous inferior petrosal sinus sampling for differential diagnosis of ACTH-dependent Cushing’s syndrome. Clin. Endocrinol. 200766, 136–142. [Google Scholar] [CrossRef]
  33. Findling, J.W.; Kehoe, M.E.; Raff, H. Identification of patients with Cushing’s disease with negative pituitary adrenocorticotropin gradients during inferior petrosal sinus sampling: Prolactin as an index of pituitary venous effluent. J. Clin. Endocrinol. Metab. 200489, 6005–6009. [Google Scholar] [CrossRef] [PubMed]
  34. Pigarova, E.A.; Rozhinskaya, L.Y.; Belaya, J.E.; Dzeranova, L.K.; Karonova, T.L.; Ilyin, A.V.; Melnichenko, G.A.; Dedov, I.I. Russian Association of Endocrinologists recommendations for diagnosis, treatment and prevention of vitamin D deficiency in adults. Probl. Endocrinol. 201662, 60–84. [Google Scholar] [CrossRef]
  35. Petrushkina, A.A.; Pigarova, E.A.; Tarasova, T.S.; Rozhinskaya, L.Y. Efficacy and safety of high-dose oral vitamin D supplementation: A pilot study. Osteoporos. Int. 201627, 512–513. [Google Scholar] [CrossRef]
  36. Pivonello, R.; De Martino, M.C.; De Leo, M.; Lombardi, G.; Colao, A. Cushing’s Syndrome. Endocrinol. Metab. Clin. N. Am. 200837, 135–149. [Google Scholar] [CrossRef]
  37. Belaya, Z.E.; Iljin, A.V.; Melnichenko, G.A.; Rozhinskaya, L.Y.; Dragunova, N.V.; Dzeranova, L.K.; Butrova, S.A.; Troshina, E.A.; Dedov, I.I. Diagnostic performance of late-night salivary cortisol measured by automated electrochemiluminescence immunoassay in obese and overweight patients referred to exclude Cushing’s syndrome. Endocrine 201241, 494–500. [Google Scholar] [CrossRef]
  38. Povaliaeva, A.; Pigarova, E.; Zhukov, A.; Bogdanov, V.; Dzeranova, L.; Mel’nikova, O.; Pekareva, E.; Malysheva, N.; Ioutsi, V.; Nikankina, L.; et al. Evaluation of vitamin D metabolism in patients with type 1 diabetes mellitus in the setting of cholecalciferol treatment. Nutrients 202012, 3873. [Google Scholar] [CrossRef] [PubMed]
  39. Thode, J.; Juul-Jørgensen, B.; Bhatia, H.M.; Kjaerulf-Nielsen, M.; Bartels, P.D.; Fogh-Andersen, N.; Siggaard-Andersen, O. Comparison of serum total calcium, albumin-corrected total calcium, and ionized calcium in 1213 patients with suspected calcium disorders. Scand. J. Clin. Lab. Investig. 198949, 217–223. [Google Scholar] [CrossRef]
  40. Bikle, D.D.; Siiteri, P.K.; Ryzen, E.; Haddad, J.G.; Gee, E. Serum protein binding of 1, 25-Dihydroxyvitamin D: A reevaluation by direct measurement of free metabolite levels. J. Clin. Endocrinol. Metab. 198561, 969–975. [Google Scholar] [CrossRef]
  41. Bikle, D.D.; Gee, E.; Halloran, B.; Kowalski, M.A.N.N.; Ryzen, E.; Haddad, J.G. Assessment of the Free Fraction of 25-Hydroxyvitamin D in Serum and Its Regulation by Albumin and the Vitamin D-Binding Protein. J. Clin. Endocrinol. Metab. 198663, 954–959. [Google Scholar] [CrossRef]
  42. Holick, M.F.; Binkley, N.C.; Bischoff-Ferrari, H.A.; Gordon, C.M.; Hanley, D.A.; Heaney, R.P.; Murad, M.H.; Weaver, C.M. Evaluation, treatment, and prevention of vitamin D deficiency: An Endocrine Society clinical practice guideline. J. Clin. Endocrinol. Metab. 201196, 1911–1930. [Google Scholar] [CrossRef] [PubMed]
  43. Dirks, N.F.; Martens, F.; Vanderschueren, D.; Billen, J.; Pauwels, S.; Ackermans, M.T.; Endert, E.; den Heijer, M.; Blankenstein, M.A.; Heijboer, A.C. Determination of human reference values for serum total 1,25-dihydroxyvitamin D using an extensively validated 2D ID-UPLC–MS/MS method. J. Steroid Biochem. Mol. Biol. 2016164, 127–133. [Google Scholar] [CrossRef] [PubMed]
  44. Tang, J.C.Y.; Nicholls, H.; Piec, I.; Washbourne, C.J.; Dutton, J.J.; Jackson, S.; Greeves, J.; Fraser, W.D. Reference intervals for serum 24,25-dihydroxyvitamin D and the ratio with 25-hydroxyvitamin D established using a newly developed LC–MS/MS method. J. Nutr. Biochem. 201746, 21–29. [Google Scholar] [CrossRef]
  45. Máčová, L.; Bičíková, M. Vitamin D: Current challenges between the laboratory and clinical practice. Nutrients 202113, 1758. [Google Scholar] [CrossRef] [PubMed]
  46. Ginsberg, C.; Hoofnagle, A.N.; Katz, R.; Hughes-Austin, J.; Miller, L.M.; Becker, J.O.; Kritchevsky, S.B.; Shlipak, M.G.; Sarnak, M.J.; Ix, J.H. The Vitamin D Metabolite Ratio Is Associated With Changes in Bone Density and Fracture Risk in Older Adults. J. Bone Miner. Res. 2021, 1–8. [Google Scholar] [CrossRef]
  47. Cavalier, E.; Huyghebaert, L.; Rousselle, O.; Bekaert, A.C.; Kovacs, S.; Vranken, L.; Peeters, S.; Le Goff, C.; Ladang, A. Simultaneous measurement of 25(OH)-vitamin D and 24,25(OH)2-vitamin D to define cut-offs for CYP24A1 mutation and vitamin D deficiency in a population of 1200 young subjects. Clin. Chem. Lab. Med. 202058, 197–201. [Google Scholar] [CrossRef]
  48. Rondeau, P.; Bourdon, E. The glycation of albumin: Structural and functional impacts. Biochimie 201193, 645–658. [Google Scholar] [CrossRef]
  49. Soudahome, A.G.; Catan, A.; Giraud, P.; Kouao, S.A.; Guerin-Dubourg, A.; Debussche, X.; Le Moullec, N.; Bourdon, E.; Bravo, S.B.; Paradela-Dobarro, B.; et al. Glycation of human serum albumin impairs binding to the glucagon-like peptide-1 analogue liraglutide. J. Biol. Chem. 2018293, 4778–4791. [Google Scholar] [CrossRef]
  50. McLeod, J.F.; Kowalski, M.A.; Haddad, J.G. Interactions among serum vitamin D binding protein, monomeric actin, profilin, and profilactin. J. Biol. Chem. 1989264, 1260–1267. [Google Scholar] [CrossRef]
  51. Gupta, Y.; Gupta, A. Glucocorticoid-induced myopathy: Pathophysiology, diagnosis, and treatment. Indian J. Endocrinol. Metab. 201317, 913. [Google Scholar] [CrossRef] [PubMed]
  52. Smets, P.; Meyer, E.; Maddens, B.; Daminet, S. Cushing’s syndrome, glucocorticoids and the kidney. Gen. Comp. Endocrinol. 2010169, 1–10. [Google Scholar] [CrossRef] [PubMed]
  53. Arnaud, J.; Constans, J. Affinity differences for vitamin D metabolites associated with the genetic isoforms of the human serum carrier protein (DBP). Hum. Genet. 199392, 183–188. [Google Scholar] [CrossRef] [PubMed]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Desmopressin Stimulation Test in a Pregnant Patient with Cushing’s Disease rights and content
Under a Creative Commons license
open access


Due to the physiologic rise of ACTH during pregnancy, unstimulated ACTH levels may not be an accurate marker to differentiate between adrenal and ACTH-independent Cushing’s syndrome.

The desmopressin stimulation test can be done during pregnancy to investigate the etiology of Cushing’s syndrome.

Non-gadolinium enhanced pituitary imaging may not detect pituitary adenoma, which is the most common cause of Cushing’s disease. Contrast-enhanced pituitary magnetic resonance imaging should be considered in pregnant women with ACTH-dependent Cushing’s syndrome.

Due to increase maternal and fetal morbidities in active Cushing’s syndrome, prompt diagnosis and appropriate treatment are essential. The treatment of choice is transsphenoidal surgery during the second trimester, preferably at a high-volume pituitary center.

There were significantly lower rates of fetal complications in women with active Cushing’s syndrome than a cured disease, including low birth weight.



The hypothalamic-pituitary-adrenal axis stimulation during pregnancy complicates the investigation of Cushing’s syndrome. Our objective is to present a pregnant patient with Cushing syndrome caused by pituitary tumor in which the desmopressin stimulation test helped in the diagnosis and led to appropriate management.

Case report

A 27-year-old woman with 9-week gestation presented with proximal myopathy for 2 months. She had high blood pressure, wide abdominal purplish striae, and proximal myopathy. Her past medical history revealed hypertension and dysglycemia for 1 year. The 8 AM cortisol was 32.4 μg/dL (5-18), late-night salivary cortisol at 11 PM was 0.7 μg/dL (<0.4), and the mean 24-hour urinary free cortisol was 237.6 μg/day (21.0-143.0). The mean ACTH concentrations at 8 AM were 44.0 pg/mL (0-46.0). Non-gadolinium enhanced pituitary magnetic resonance imaging (MRI) reported no obvious lesion. The desmopressin stimulation test showed a 70% increase in ACTH levels from baseline after desmopressin administration. The pituitary MRI with gadolinium showed an 8x8x7-mm pituitary adenoma. Transsphenoidal surgery with tumor removal was done, which showed ACTH-positive tumor cells. After the surgery, the patient carried on the pregnancy uneventfully.


During pregnancy, the ACTH level may not be an accurate marker to help in the differential diagnosis of Cushing’s syndrome. Moreover, non-gadolinium pituitary imaging may not detect small pituitary lesions.


In the present Case, the desmopressin stimulation test suggested the diagnosis of Cushing’s disease, which subsequently led to successful treatment. This suggested that the desmopressin test may serve as a useful test to diagnose Cushing’s disease in pregnant individuals.


Cushing’s disease
Cushing’s syndrome
desmopressin stimulation test


Pregnancy rarely occurs during the course of Cushing’s syndrome (CS).1,2 Given the increase in maternal and fetal morbidities in women with active CS, early diagnosis and treatment of CS are essential.2

The diagnosis of CS using the usual diagnostic tests is challenging due to stimulation of the hypothalamic-pituitary-adrenal axis during pregnancy. The physiologic rise of ACTH from the 7th week of pregnancy also complicates the investigation for the etiology of CS.1 The concern of gadolinium use during pregnancy can affect the sensitivity in detecting small pituitary lesions in ACTH-dependent CS if using non-gadolinium pituitary imaging. Desmopressin is a vasopressin analog selective for V2 receptors. The desmopressin stimulation test has been proposed as a useful procedure for the differential diagnosis of CS.3 Desmopressin stimulates the increase in ACTH and cortisol in patients with CS caused by pituitary tumor or Cushing’s disease (CD) but not in the majority of normal, obese subjects and patients with adrenal CS or ectopic ACTH syndrome.3,4 However, there were limited data on the desmopressin stimulation test during pregnancy.

Here we present the 27-year-old woman with CS in which the desmopressin stimulation test helped in the diagnosis of CD and led to successful treatment.

Case presentation

A 27-year-old woman with 9-week gestation was referred from the orthopedic department to evaluate CS. She presented with proximal myopathy for 2 months. On physical examination, she had Cushingoid appearance, wide purplish striae, bruising, and proximal muscle weakness. Her blood pressure was 160/100 mmHg, and her body mass index was 32.2 kg/m2. Her past medical history revealed that she had hypertension, dyslipidemia, and impaired fasting glucose for 1 year without taking any medication. She also gained 20 kg in the past 2 years. The 8 AM cortisol (chemiluminescent immunometric assay, Immulite/Siemens) was 32.4 μg/dL (normal , 5.0-18.0), late-night salivary cortisol at 11 PM (electrochemiluminescence immunoassay, Roche Cobas) was 0.7 μg/dL (normal, <0.4), and the mean 24-hour urinary free cortisol (UFC) (radioimmunoassay, Immulite/Siemens) was 237.6 μg/day (normal, 21.0-143.0). ACTH concentrations at 8 AM (chemiluminescent immunometric assay, Immulite/Siemens) were 48.4 and 39.6 pg/mL (normal, 0-46.0) (Table 1). At 12 weeks of gestation, non-gadolinium enhanced pituitary magnetic resonance imaging (MRI) reported a mild bulging contour of the right lateral aspect of the pituitary gland without an obvious abnormal lesion (Figure 2A). The desmopressin stimulation test was then carried out at 14 weeks of gestation. Serial blood samples for ACTH and cortisol were obtained basally (at 8 AM) and at 15, 30, 45, and 60 minutes after the intravenous administration of 10 μg of desmopressin. The results were shown in Table 2. Compared with baseline, ACTH levels increased from 34.7 to 58.9 pg/mL (70%) at 15 minutes after desmopressin administration (a ≥35% increase in ACTH levels was considered an indication of CD in non-pregnant individuals)3 (Figure 1). The pituitary MRI with gadolinium revealed an 8x8x7-mm circumscribed lesion with heterogeneous iso- to hyperintensity on T2W in the right inferolateral aspect of the anterior pituitary lobe. The lesion had a delayed enhancement compared to normal pituitary tissue (Figure 2B). Non-contrast MRI adrenal glands showed bilateral normal adrenal glands without mass or nodule. Other abdominal organs were unremarkable. Regarding comorbidities, she had hypertension and gestational diabetes mellitus (GDM). The HbA1c level was 5.7% (39 mmol/mol). Using a two-step strategy, GDM was diagnosed at 12 weeks of gestation. Hypertension and GDM were controlled with 750 mg of methyldopa and 50 units of insulin per day, respectively.

Table 1. Laboratory investigations of the present Case

Variable At 9 weeks of gestation
8 AM cortisol, μg/dL (5.0-18.0) 32.4
Salivary cortisol (11 PM , <0.4 μg/dL) 0.7
UFC (21.0-143.0 μg/day) 183.5 and 291.6
ACTH, pg/mL (8 AM, 0-46.0) 48.4 and 39.6
DHEAS (8 AM, 35.0-430.0 μg/dL) 378.0
PAC (upright position, 8 AM), ng/dL 5.2
PRA (upright position, 8 AM), ng/mL/hr 2.1
Potassium, mmol/L 3.6

UFC, urinary free cortisol; ACTH, adrenocorticotrophic hormone; DHEAS, dehydroepiandrosterone sulphate; PAC, plasma aldosterone concentration; PRA, plasma renin activity.

Figure 2Pituitary imaging of the present Case. (A) A non-gadolinium MRI of the pituitary gland at 12 weeks of gestation showing a mild bulging contour of the right lateral aspect of the pituitary gland without an obvious abnormal lesion (B) An MRI of the pituitary gland with gadolinium at 14 weeks of gestation showing an 8x8x7-mm circumscribed lesion with heterogeneous iso- to hyperintensity on T2W in the right inferolateral aspect of the anterior pituitary lobe. The lesion had a delayed enhancement compared to normal pituitary tissue.

Table 2. Desmopressin stimulation test results performing at 14 weeks of gestation

Time 0 min 15 min 30 min 45 min 60 min
ACTH (pg/mL) 34.7 58.9 57.4 49.9 38.2
Cortisol (μg/dL) 30.6 30.2 29.7 29.6 31.0

ACTH, adrenocorticotrophic hormone

Figure 1. Percentage of ACTH increase after desmopressin administration (time 0 min).

Transsphenoidal surgery with tumor removal was performed at 18 weeks of gestation. Pathological findings showed a 1.3×1.0x0.3 cm of tissue with segments of the pituitary gland and tumor. There were monomorphous round nuclei, stippled chromatin, indistinct nucleoli, and pale eosinophilic cytoplasm cells. These cells were reactive with ACTH and showed loss of reticulin framework, unlike the normal pituitary gland. The next day after the surgery, her 8 AM cortisol was 6.0 μg/dL. Hydrocortisone supplement was started and continued throughout pregnancy. Antihypertensives were discontinued, and the insulin dosages decreased to less than 20 units per day. At 38 weeks of gestation, she gave birth to a 2300-gm male newborn (small for gestational age). Dysglycemia and hypertension resolved after the delivery. One year after the first child’s delivery, the patient had a spontaneous pregnancy without GDM or hypertension. The 8 AM cortisol was 3.9 μg/dL, and hydrocortisone replacement was continued. The patient successfully delivered a term 3300-gm male infant without fetal or maternal complications. Two years after the first transsphenoidal surgery, a 1-μg cosyntropin stimulation test was performed, the basal cortisol was 11.7 μg/dL, and the peak serum cortisol was 23.8 μg/dL. Steroid replacement was withdrawn.


Herein we present a 27-year-old woman who was evaluated during her first pregnancy for clinical and laboratory features suggestive of CD. Her morning serum and late-night salivary cortisol concentrations were elevated in addition to non-suppressed ACTH, but a definitive diagnosis was not obtained by a non-gadolinium pituitary MRI. The diagnosis of CD was suggested, however, by the results of a desmopressin stimulation test. The pituitary MRI with gadolinium was proceeded and revealed a pituitary lesion greater than 6 mm.

The prevalence of pregnancy is low due to reduced fertility in CS. To date, there have been less than 300 pregnant patients with CS reported in the literature.2 In pregnancy, the most frequent etiology of CS is adrenal CS (60%), followed by ACTH-producing pituitary adenomas or CD (35%), and very rarely ectopic ACTH (<5%).1 In contrast, CD is the most common cause of CS in non-pregnant people (approximately 70 percent). The clinical diagnosis of CS during pregnancy may be missed due to overlapping features between pregnancy and CS. However, wide purplish cutaneous striae and proximal myopathy are signs with high discrimination index when CS is suspected.5 These signs are not present in normal pregnancy.

In this present Case, CS was diagnosed with apparent clinical features of CS in addition to an elevated UFC and late-night salivary cortisol. The patient denied taking any supplements and her 8 AM cortisol was not suppressed and therefore did not suggest an etiology of exogenous steroid use. Pregnant women without CS may have elevated UFC and late-night salivary cortisol due to increased total and free plasma cortisol from the first trimester until the end of pregnancy.6 This results from an elevated concentration of cortisol transport protein and the increase in placental ACTH and CRH. According to the current guideline, UFC is the recommended test when CS is suspected during pregnancy.5 Since UFC increases during the second trimester, it may not be a reliable marker after the first trimester of pregnancy unless the level is clearly increased (up to 2- to 3-fold the upper limit of normal values).1 Late-night salivary cortisol is also one of the useful tests to diagnose CS during pregnancy because the circadian rhythm of cortisol is preserved in normal pregnancy. Furthermore, it is not influenced by the changes in the binding proteins.7 However, the previous study has shown that late-night salivary cortisol increased progressively throughout pregnancy. When compared with non-pregnant women, median values of late-night salivary cortisol in pregnant women were 1.1, 1.4, and 2.1 times higher in the first, second, and third trimesters respectively. The cutoff values for late-night salivary cortisol on each gestational trimester were: first trimester 0.255 μg/dL, second trimester 0.260 μg/dL, and third trimester 0.285 μg/dL. The respective sensitivities and specificities in each trimester were: first trimester 92 and 100%, second trimester 84 and 98%, and third trimester 80 and 93%.8

Given the non-suppressed ACTH levels after the 7th week of gestation, we were not able to summarize whether the etiology was adrenal CS or ACTH-dependent CS which could be either CD or ectopic ACTH syndrome. In non-pregnant individuals, ACTH suppression usually identifies adrenal CS. However, in pregnancy, ACTH levels were non-suppressed in half of those with adrenal CS due to continued stimulation of maternal hypothalamic-pituitary-adrenal axis by placental CRH.1 Therefore, using the ACTH thresholds in general populations can lead to misdiagnosis when investigating the etiology of CS in pregnant individuals. The hypothalamic-pituitary-adrenal axis response to exogenous glucocorticoids is blunted in pregnant women. Following an overnight dexamethasone administration, pregnant women without CS may have non-suppressed plasma cortisol and UFC.6 In non-pregnant individuals with CS, the high-dose dexamethasone suppression test identify CD with a sensitivity of 82% and a specificity of 50%.4 During pregnancy, the high-dose dexamethasone suppression test failed to identify almost half of the patients with CD.1 Inferior petrosal sinus sampling is usually avoided due to the risk of excessive radiation exposure. Since the non-gadolinium MRI also showed no obvious pituitary lesion in the present Case, in addition to the limitation of the high-dose dexamethasone suppression test and inferior petrosal sinus sampling in pregnancy, we used desmopressin stimulation to help in the investigation of CD since desmopressin can stimulate an ACTH response in a considerable proportion of patients with CD but not in most patients with adrenal CS or ectopic ACTH syndrome.3,4

Desmopressin has been assigned to pregnancy category B by the US Food and Drug Administration (FDA). In the most recent guideline update on the diagnosis and management of CD, the desmopressin stimulation test can be used to differentiate ectopic CS and CD in patients with normal or high ACTH and have no adenoma or equivocal results of pituitary MRI. However, the guideline did not mention the use of this test in pregnant individuals.9 The literature regarding the use of desmopressin stimulation tests in pregnancy is limited. We were able to identify one study in a pregnant patient with active CS, who was surgically confirmed as CD, in which the desmopressin stimulation test was performed at 10 weeks of gestation and after the delivery. Compared with age-matched healthy non-pregnant women, there were different responses of cortisol and ACTH after desmopressin administration in a pregnant patient with active CS.10 The ACTH peaks after the administration of desmopressin were higher in the pregnant patient. CRH stimulation test was also performed in the pregnant patient with CD. Desmopressin stimulated ACTH values during pregnancy and after the delivery were not significantly different, while the CRH stimulated ACTH values were significantly higher when the test was performed after the delivery. The authors did not mention optimal cutoff values for these diagnostic tests.10 In non-pregnant individuals, the ACTH increase of more than 35% at 15 minutes after the desmopressin administration gave the sensitivity of 84% and the specificity of 43% in the diagnosis of CD.3 Another recent study in ACTH-dependent CS showed that the threshold increase in the ACTH level after desmopressin stimulation of 45% identified CD with a sensitivity of 91% and a specificity of 75%.4 Using the non-pregnant cutoff values for the desmopressin stimulation test, the diagnosis of CD was made in our patient who was later surgically confirmed as CD.

Pituitary microadenomas were the cause of CD in almost 90% of non-pregnant individuals.11 In pregnant women with CD, pituitary microadenomas were also reported to be more common than macroadenomas.1,12 Almost 40% of pituitary microadenomas in CD were invisible or poorly visible in non-contrast MRI, in which contrast-enhanced MRI detected them.13 In the Case series from Lindsay et al., the non-contrast MRI could not correctly identify pituitary adenomas in 38% of pregnant patients with available data.1 The same case series reported a pregnant patient having normal pituitary MRI and was later surgically confirmed as having CD from a 3×3 adenoma with positive staining for ACTH. In the present case, a mild bulging contour of the pituitary gland, although without an obvious abnormal lesion, in addition to desmopressin test results, suggested the need for contrast-enhanced pituitary MRI. Gadolinium contrast is FDA pregnancy category C since it is water-soluble and can cross the placenta into the fetus and amniotic fluid.14 However, since a non-gadolinium MRI may not detect pituitary microadenoma even in patients with normal imaging results,1,15 we suggested physicians consider pituitary MRI with gadolinium as initial imaging in pregnant patients with clinical suspicion of CD.

Prompt diagnosis and treatment of CS are essential due to a higher rate of fetal loss in active CS patients without treatment than those who received either medical or surgical treatment. There are significantly lower rates of various fetal complications, including low birth weight, in women with active CS than in cured CS.2 Although medical and surgical treatment were not compared as prognostic factors for complications, experts recommend transsphenoidal surgery in the second trimester as the treatment of choice for CD in pregnancy.1,15 Medical treatment should be the second choice when surgery cannot be carried out or late diagnosis is made.


In the present Case, the results from the desmopressin stimulation test and the pituitary MRI with gadolinium suggested the diagnosis of CD, which subsequently led to successful treatment. This suggested that the desmopressin test may serve as a useful test to diagnose CD even in the context of pregnancy.

Conflicts of Interest

None of the authors have any potential conflicts of interest associated with this research.


Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.


The authors would like to thank you all the colleagues in the Division of Endocrinology and Metabolism, Department of Medicine, Faculty of medicine, Chulalongkorn University for all the support.

Cushing Disease Treated Successfully with Metyrapone During Pregnancy rights and content
Under a Creative Commons license
open access


Cushing’s Disease (CD) in pregnancy is rare, but poses many risks to the mother and fetus

Although surgery is still considered first line, this CASE highlights the successful use of metyrapone throughout pregnancy to manage CD in patients where surgery is considered high risk or low likelihood of cure

The dose of metyrapone can be titrated to a goal urinary free cortisol of < 150 ug/24 hours given the known rise in cortisol during gestation

Though no fetal adverse events have been reported, metyrapone does cross the placenta and long-term effects are unknown.



Cushing Disease (CD) in pregnancy is a rare, but serious, disease that adversely impacts maternal and fetal outcomes. As the sole use of metyrapone in the management of CD has been rarely reported, we describe our experience using it to treat a pregnant patient with CD.

Case Report

34-year-old woman with hypertension who was diagnosed with adrenocorticotropic hormone-dependent CD based on a urinary free cortisol (UFC) of 290 μg/24hr (reference 6-42μg/dL) and abnormal dexamethasone suppression test (cortisol 12.4 μg/dL) before becoming pregnant. She conceived naturally 12 weeks post-transsphenoidal surgery, and was subsequently found to have persistent disease with UFC 768μg/dL. Surgery was deemed high risk given the proximity of the tumor to the right carotid artery and high likelihood of residual disease. Instead, she was managed with metyrapone throughout her pregnancy and titrated to goal UFC of <150μg/24hr due to the known physiologic rise in cortisol during gestation. The patient had diet-controlled gestational diabetes, and well-controlled hypertension. She gave birth at 37 weeks gestation to a healthy baby boy, without adrenal insufficiency in the baby or mother.


This CASE highlights the successful use of metyrapone throughout pregnancy to manage CD in patients where surgery is considered high risk or low likelihood of cure. While metyrapone is effective, close surveillance is required for worsening hypertension, hypokalemia, and potential adrenal insufficiency. Though no fetal adverse events have been reported, this medication crosses the placenta and long-term effects are unknown.


We describe a CASE of CD during pregnancy that was successfully treated with metyrapone.

Key words

Cushing disease


Cushing disease (CD) is caused by endogenous overproduction of glucocorticoids due to hypersecretion of adrenocorticotropic hormone (ACTH) by a pituitary adenoma. CD in pregnancy is very rare, and when it occurs, it is considered a high-risk pregnancy with many potential adverse outcomes for both the mother and fetus.1 Infertility is common in CD due to cortisol and androgen excess leading to hypogonadotropic hypogonadism.1 Due to the rarity of CD in pregnancy, there is little guidance in terms of treatment for this patient population. Similar to non-pregnant patients, the first-line treatment is transsphenoidal pituitary adenoma resection, with medical therapy as a second-line treatment option. This report presents a CASE that highlights the use of metyrapone, a steroidogenesis inhibitor, as a sole therapy in cases where surgery is deemed to be high risk and unlikely curative due to location of the tumor.


A 34-year-old woman with a past medical history of hypertension and infertility for six years presented to endocrinology for evaluation. Aside from difficulty conceiving, her only complaints were nausea and easy bruising. On exam she did not have clinical features of CD –abdominal violaceous striae, moon facies or a dorsocervical fat pad were absent. Her laboratory results revealed an elevated prolactin level (50-60ng/mL, reference range 1.4-24), an elevated ACTH level (61 pg/mL, reference range 0-46), and low FSH and LH levels (1.7mIU/mL and 1.76mIU/mL, respectively). Further testing demonstrated an elevated urinary free cortisol level (UFC) (290μg/24 hour, reference range 6-42) and her cortisol failed to suppress on a 1mg dexamethasone suppression test (cortisol 12.4μg/dL). Magnetic resonance imaging (MRI) of the pituitary with and without contrast showed a T2 hyperintense, hypoenhancing lesion within the right side of the sella touching the right cavernous internal carotid artery measuring 8x8x9 mm consistent with a pituitary adenoma (Figure 1).

Figure 1. Caption: T1 weighted post gadolinium coronal image of the pituitary gland with a small hypoenhancing lesion within the right side of the sella.

After the presumed diagnosis of CD was made, she was referred to neurosurgery for transsphenoidal resection of the adenoma, which she underwent a few months later. Intra-operatively, a white friable tumor was found, and otherwise the surgery was uneventful. Three months later, however, she was found to have a persistent 8x8x9mm hypoenhancing lesion extending laterally over the right cavernous carotid artery on MRI. The mass approximated but did not contact the right intracranial optic nerve. The pathology from resected tissue was consistent with normal pituitary tissue with staining for growth hormone (80%), ACTH (30%), prolactin (40%), follicle stimulating hormone (5%), luteinizing hormone (40%) and thyroid stimulating hormone (15%), proving the surgery to have been unsuccessful.

Twelve weeks post-operatively, the patient discovered she was pregnant. At 12 weeks gestation, her UFC was 768μg/24h and two midnight salivary cortisol levels were elevated at 0.175 and 0.625μg/dL (reference <0.010-0.090). She was experiencing easy bruising and taking labetalol 400 mg twice daily for hypertension. She had gained 10 pounds by 12 weeks gestation.

A second transsphenoidal surgery during pregnancy was deemed high risk, with a high likelihood of residual disease due to the proximity of the tumor to the right carotid artery. The decision was made to treat the patient medically with metyrapone which was started at 250 mg twice per day at 12 weeks gestation and was eventually uptitrated based on UFC levels every 3-4 weeks (goal of <150μg /24h) to 1000 mg three times per day by the time of delivery with an eventual UFC level of 120μg/24h (Figure 2) . Morning ACTH and serum cortisol levels were monitored for potential adrenal insufficiency.

Figure 2. Caption: This figure depicts the patient’s 24 hour urinary cortisol levels over time as well as the titration of metyrapone dosage in mg/day.

Her hypertension was well controlled throughout pregnancy on labetalol with the addition of nifedipine XL 30mg daily in the second trimester. She remained normokalemic with potassium ranging from 3.8-4.1mEq/L. She was diagnosed with gestational diabetes at 24 weeks by an abnormal two-step oral glucose tolerance test, which was diet-controlled. The patient was induced at 37 weeks gestation due to cervical insufficiency with cerclage in place, and was given stress dose steroids along with metyrapone. She delivered a healthy baby boy vaginally without complications. His Apgar scores were 9 and 9 and he weighed 6 pounds and 5 ounces. At the time of delivery and one week later, the baby’s cortisol levels were normal (6 μg/dL, normal 4-20), without evidence of adrenal insufficiency.

The patient’s metyrapone dose was reduced to 500mg three times a day after pregnancy and her 2 month postpartum 24 hour UFC was 42μg/24hr. The patient stopped the metyrapone on her own four months later and her UFC was found to be elevated at 272ug/24hr (normal 6-42μg/24hr). An MRI one year postpartum revealed a 10x10x9 mm adenoma in the right sella with some suprasellar extension without compression of the optic chiasm, but with abutment of the right carotid artery. Due to the persistently elevated cortisol, large size of the tumor, and potential for cure, especially if followed by radiation therapy, a second transsphenoidal surgery was recommended. However, due to the COVID-19 pandemic the patient underwent a delayed surgery 1.5 years postpartum. The pathology was consistent with a pituitary adenoma that stained strongly and diffusely for ACTH and synaptophysin, only. Her postoperative day 2 cortisol was 1.1μg/dL (reference range 6.7-22.6) and hydrocortisone 20mg in the morning and 10mg in the afternoon was started. She remains on hydrocortisone replacement and went on to conceive again, one month after her second surgery.


We describe a patient with pre-existing CD who became pregnant and was managed successfully with metyrapone throughout her pregnancy.

Although CD is rare in pregnancy, it can occur, and poses risks to both the mother and fetus.1,2 Potential maternal complications include hypertension, preeclampsia, diabetes, fractures and more uncommonly, cardiac failure, psychiatric disorders, infection and maternal death.1,2 There is also increased fetal morbidity including prematurity, intrauterine growth retardation and less commonly CD can lead to stillbirth, spontaneous abortion, intrauterine death and hypoadrenalism.1,2

It is, therefore, imperative that these patients receive prompt care to control cortisol levels. The treatment of CD in pregnancy is challenging as there are no large research trials studying the efficacy and safety of medications in CD during pregnancy. Pituitary surgery is first-line recommendation and should be done late in the first trimester or in the second trimester to prevent spontaneous pregnancy loss.3 In this CASE, however, it was felt that a second surgery would be high-risk given the proximity of the tumor to the right carotid artery and possibly not curative, and thus surgery was not a feasible option. She was therefore successfully managed with medical therapy with metyrapone alone throughout her pregnancy.

Metyrapone use in pregnancy has been previously reported in the literature and has been shown to be effective in reducing cortisol levels.4,5,6 Although not approved for use in pregnancy, this steroidogenesis inhibitor is the most commonly used medication to treat Cushing’s syndrome in pregnant women.3,5 Due to metyrapone’s inhibition of 11-beta-hydroxylase, there is a buildup of steroidogenesis precursors such as 11-deoxycorticosterone, which can worsen hypertension, increase frequency of preeclampsia, and cause hypokalemia.3 Metyrapone also leads to elevation of adrenal androgens, which in conjunction with accumulation of 11-deoxycorticosterone, can cause hirsutism and virilization. 8

Though the use of Cabergoline has been reported in cases with Cushing disease during pregnancy, no long term safety data is available regarding it effects on pregnancy as well as the fetus. Moreover, studies assessing the effect of cabergoline in persistent or recurrent CD show a response rate of 20-30% only in cases with mild hypercortisolism. 9

There is no consensus on how to medically treat patients with CD during pregnancy. We chose a goal UFC of <150μg/24 hours because of the physiological rise of cortisol to two to three times the upper limit of normal during pregnancy.3,7 During pregnancy, there is an increase in corticotropin-releasing hormone from the placenta, which is identical in structure to the hypothalamic form.7 This leads to increased levels of ACTH which stimulates the maternal adrenal glands to become slightly hypertrophic and accounts for the rise in serum cortisol levels in pregnancy.7 Corticosteroid-binding globulin also increases in pregnancy, along with serum free cortisol, leading to urinary free cortisol increasing to 3-fold the normal range.7 We therefore aimed to keep our patient’s urinary free cortisol approximately 3 times the upper limit of normal on our assay, to maintain normal cortisol levels for pregnancy.

Close surveillance of patients is required for worsening hypertension, hypokalemia, and potential adrenal insufficiency.3 Although no fetal adverse events from metyrapone have been reported, the medication does cross the placenta, leading to the potential for fetal adrenal insufficiency, and long-term effects are unknown.3


This CASE demonstrates the successful use of metyrapone alone to treat CD throughout pregnancy resulting in the birth of a healthy baby without adrenal insufficiency. These cases are particularly challenging given the lack of FDA-approved therapies and the lack of consensus on directing titration of medications and the duration of therapy.

Uncited reference



Clinical Relevance: Cushing’s Disease (CD) in pregnancy is a rare, but serious, disease that has potential adverse effects on maternal and fetal health. Surgery is considered first line therapy, and there is little consensus on medical treatment of CD in pregnancy. This CASE demonstrates the successful use and titration of metyrapone throughout pregnancy.


%d bloggers like this: