Abstract
Context
Successful first-line treatment of Cushing’s syndrome by resection of the underlying tumor is usually followed by adrenal insufficiency.
Purpose
The aims of this study were to determine the recovery rate and time to recovery of adrenal function after treatment for different forms of endogenous Cushing’s syndrome and to identify factors associated with recovery.
Methods
In this retrospective study of 174 consecutive patients with Cushing’s syndrome, the recovery rate and time to recovery of adrenal function after surgery were assessed.
Results
The 1-year, 2-year and 5-year recovery rates of patients with Cushing’s disease were 37.8, 70.1 and 81.1%, respectively. For patients with adrenal Cushing’s syndrome, the 1-year, 2-year and 5-year recovery rates were higher: 49.3, 86.9 and 91.3%, respectively. Median time to recovery for patients with Cushing’s disease and adrenal Cushing’s syndrome was 13.9 and 12.1 months, respectively. The median time to recovery of adrenal function in patients with Cushing’s disease with and without recurrence was 9.9 versus 14.4 months, respectively. Higher age was associated with a lower probability of recovery of adrenal function: HR 0.83 per decade of age (95% CI 0.70–0.98).
Conclusion
The recovery rate of adrenal function after successful surgery as first-line treatment in patients with Cushing’s syndrome is high. However, it may take several months to years before recovery of adrenal function occurs. In case of early recovery of adrenal function, clinicians should be aware of a possible recurrence of Cushing’s disease.
Introduction
Cushing’s syndrome (CS) is characterized by chronic exposure to an excess of glucocorticosteroids (1). Endogenous hypercortisolism is a rare disorder with an estimated incidence of 0.2–5 patients per million per year (1). CS can cause severe, disabling signs and symptoms and is associated with significantly increased morbidity and mortality. In approximately 70% cases, endogenous CS is caused by an ACTH-producing pituitary adenoma, also known as Cushing’s disease (CD). In 15–25% cases, an ACTH-independent form of CS is caused by a unilateral adrenal adenoma, adrenal carcinoma or bilateral micro- or macronodular hyperplasia (adrenal CS). An ACTH-producing ectopic tumor is a rare cause of CS. First-line treatment of CS is surgical removal of the pituitary, adrenal or ectopic tumor (1, 2).
Successful first-line treatment by resection of the underlying tumor is usually followed by adrenal insufficiency (AI) due to suppression of the hypothalamic–pituitary–adrenal axis after prolonged exposure to high concentrations of cortisol (3, 4, 5). Theoretically, one would expect that the hypothalamic–pituitary–adrenal axis recovers over time and that the substitution of glucocorticosteroids can slowly be reduced and stopped as long as there is no irreversible damage to the remaining adrenal or pituitary tissue. However, in clinical practice, AI is not always transient. In a subset of patients, this is caused by permanent AI due to perioperative damage to the pituitary gland or irreversible atrophy of the contralateral adrenal gland. In other cases, tapering the dosage of glucocorticosteroids is not possible because this causes worsening of symptoms. Despite the glucocorticoid replacement therapy, patients often experience symptoms resembling AI, such as fatigue, myalgia, arthralgia, depression, anxiety and decreased quality of life, also known as glucocorticoid withdrawal syndrome (GWS) (6). GWS is caused by dependence on supraphysiologic glucocorticoid concentrations after chronic exposure to high concentrations of glucocorticoids, which can complicate and delay the withdrawal of exogenous steroids. As a result, patients and physicians often struggle with a dilemma: on the one hand, lowering the cortisol substitution is necessary to enable functional recovery of the hypothalamic–pituitary–adrenal axis. On the other hand, lowering the substitution therapy often causes worsening of symptoms. In clinical practice, it is not always possible to completely taper the substitution of steroids due to GWS, even in spite of intensive guidance and support by the treating physician, specialized nurse and other healthcare professionals. Moreover, in patients remaining on glucocorticoid replacement, it is not always clear whether the failure to recover from AI is caused by the irreversible damage of the remaining pituitary or adrenal tissue or the failure to overcome the GWS. The time after which adrenal function recovers and substitution therapy can be tapered off varies largely between patients but may take several years (7).
A recent survey among patients with CS highlighted the need of patients for better information about the difficult post-surgical course (8). However, scientific data about this post-operative period, particularly regarding the recovery rate and time to recovery from AI are scarce (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). Because of the rarity of CS, most studies are hampered by a limited number of patients. The reported recovery rates of adrenal function after first-line treatment for CS vary widely, between 37 and 93% for CD (9, 10, 11, 12, 13) and between 38 and 93% for overt adrenal CS (10, 12, 14, 15, 16, 17).
The reported duration to recovery of the hypothalamic–pituitary–adrenal axis after CD and adrenal CS also varies widely, between 13 and 25 months after CD (9, 10, 11, 13, 19) and between 11 and 30 months in overt adrenal CS (10, 14, 15, 16, 18, 20).
Factors which influence the recovery rate and the duration to recovery of adrenal function are not entirely clear. A few studies reported a lower chance of recovery and a longer duration to recovery of adrenal function in patients who are younger, have more severe hypercortisolism, and longer duration of symptoms before diagnosis, whereas other studies could not confirm these findings (10, 13, 21). By contrast, other studies reported a higher chance of recovery in younger patients (21). Identification of these factors may help provide patients with more information about the expected post-surgical course.
Therefore, the aims of the present study were to assess the recovery rate and time to recovery of adrenal function after successful first-line treatment in the different subtypes of CS in a large series of consecutive patients treated at a tertiary referral center and to identify factors associated with recovery.
Methods
Patients
The medical records of adult and pediatric patients treated for CS at Radboud University Medical Center, Nijmegen, between 1968 and 2022 were examined retrospectively. This is a tertiary referral hospital where practically all cases of CS from the large surrounding geographic area are managed. All patients with CD, adrenal CS and ectopic CS who were in remission and developed AI after first-line surgical treatment were included. Exclusion criteria were bilateral adrenalectomy as first-line treatment, adrenocortical carcinoma, radiotherapy of the pituitary gland before surgery, pituitary carcinoma and the therapeutic use of corticosteroids for conditions other than AI. Data were collected on age, sex, body mass index (BMI), duration of CS symptoms, comorbidities, the use of medication, biochemical results at diagnosis and during follow-up, preoperative imaging, surgical treatment and histology.
The study was assessed by the Committee for Research with Humans, Arnhem/Nijmegen Region and the need for written approval by individual patients was waived since this study did not fall within the remit of the Medical Research Involving Human Subjects Act (WMO). The study has been reviewed by the ethics committee on the basis of the Dutch Code of conduct for health research, the Dutch Code of conduct for responsible use, the Dutch Personal Data Protection Act and the Medical Treatment Agreement Act. The ethics committee has passed a positive judgment on the study. The procedures were conducted according to the principles of the Declaration of Helsinki.
Diagnostics and definitions
Patients were diagnosed with CS according to the guidelines available at the time, i.e., the presence of signs and symptoms of hypercortisolism in combination with confirmatory biochemical tests, including the 1 mg dexamethasone suppression test (DST), 24-h urine free cortisol (UFC), late-night salivary cortisol concentrations and/or hair cortisol. The cutoff value for adequate cortisol suppression after the DST was <50 nmol/L (22). For UFC, the times upper limit of normal was calculated because several assays with different reference values were used over time.
First-line treatment consisted of pituitary surgery in patients with CD and unilateral adrenalectomy in patients with ACS. In patients with bilateral macronodular hyperplasia, adrenalectomy of the largest adrenal was performed after carefully outweighing the risks and benefits of surgery together with the patient, taking into account factors such as age, severity of symptoms, comorbidities associated with hypercortisolism (e.g., diabetes mellitus type 2, cardiovascular disease, osteoporosis) and the severity of the hypercortisolism (2).
Peri- and postoperatively, all patients received glucocorticoid stress dosing, which was tapered off within a few days after surgery. Adrenal function was initially evaluated with a postoperative morning fasting cortisol concentration, measured at least 24 h after the last dose of hydrocortisone or cortisone acetate, within 7 days after surgery. If the postoperative morning fasting cortisol was <200 nmol/L, the patient was considered to have AI and glucocorticoid replacement therapy was continued. The starting dose was usually hydrocortisone 30 mg once daily (or an equivalent dose of cortisone acetate in the early years). For children, the dose was weight-based. Afterwards, the dose was slowly tapered off according to the symptoms/well-being of the patient and fasting cortisol values. During follow-up, the dose was usually divided into two or three doses a day.
Remission of CS after treatment was defined as either a morning cortisol of ≤50 nmol/L, adequate cortisol suppression after DST or a late-night salivary cortisol concentration within the reference range. Duration of AI was defined as the time between surgery and discontinuation of glucocorticoid replacement therapy. Complete recovery of adrenal function was assessed by spontaneous fasting cortisol concentration, an insulin tolerance test or a 250 μg ACTH stimulation test after discontinuation of glucocorticoid replacement therapy. In cases where fasting morning cortisol ≥520 nmol/L, adrenal function was considered as completely recovered. For the dynamic tests, assay-dependent cutoff values were used according to the guidelines available at the time. The dynamic tests were not performed routinely in all patients until 1999. In patients for whom no dynamic tests (results) were available, complete recovery of AI was defined as complete discontinuation of replacement therapy. Recurrence of CS was defined as the presence of signs and symptoms of hypercortisolism in combination with confirmatory biochemical tests, including the 1 mg DST, 24-h UFC, late-night salivary cortisol concentrations and/or hair cortisol.
Statistical analysis
Continuous data were expressed as mean ± SD or median + interquartile range (IQR), and categorical data were presented as frequency (n) and percentage (%). We produced Kaplan–Meier curves to determine the unadjusted probability of recovery of adrenal function over time. Patients that tapered off and completely stopped the glucocorticoid replacement therapy were assigned in the survival analyses as having an event (=recovery of adrenal function). The date of the last follow-up visit was assigned in the survival analyses as the last date and patients that were lost to follow-up or developed a recurrence before stopping the glucocorticoid replacement therapy were censored. In order to identify factors associated with recovery of adrenal function, we compared Kaplan Meier curves between several subgroups of patients: CD versus adrenal CS versus ectopic CS, age (at diagnosis) groups of ≤35 versus 36–55 versus ≥56 years old, patients with or without postoperative pituitary deficiencies, patients with or without recurrence of CS during follow-up, patients with or without preoperative medical treatment (PMT), patients operated before versus after 2010 and patients with a low versus slightly higher post-operative morning cortisol (<100 nmol/L versus 100–200 nmol/L), measured within 7 days after surgery. The Kaplan–Meier curves of the subgroups were compared using the two-sided log-rank test. The P-value ≤0.05 was considered statistically significant. The Kaplan–Meier curves provided the 1-year, 2-year and 5-year recovery rates and the median time to recovery of the adrenal gland. We used Cox proportional hazards models to calculate hazard ratios (HRs) with a 95% confidence interval (CI) of the probability of recovery of adrenal function over time in order to identify factors associated with recovery of adrenal function (univariate analyses). Cox proportional hazards models with multivariate analyses were performed to calculate the adjusted HRs with 95% CI. The model of multivariate analysis for the whole group included the variables: etiology of CS, age, sex, BMI, duration of symptoms before diagnosis, UFC and postoperative cortisol 0.10–0.20 versus <0.10 mcmol/L. The model of multivariate analysis for the patients with CD only included the variables: etiology of CD, age, sex, BMI, duration of symptoms before diagnosis, UFC, post-operative cortisol 0.10–0.20 versus <0.10 mcmol/L, PMT, hormonal deficiencies of the anterior pituitary gland other than AI and micro/macroadenoma. A 95% CI not including 1 was considered statistically significant.
All statistical analyses were performed using STATA version 11 (StataCorp, USA).
Results
In total, 174 patients were included in the analysis. The assessment of eligibility, the number of patients excluded from this study and the reasons for exclusion are shown in Fig. 1. The baseline characteristics are described in Table 1. The median follow-up was 6.8 years (IQR: 2.2–12.6). In 69.6% (94/135) of all patients who discontinued their glucocorticoid replacement therapy, the recovery of adrenal function was confirmed with a dynamic test or a morning cortisol concentration ≥520 nmol/L.
Flowchart showing the assessment for eligibility, the number of patients excluded from the study and the reasons for exclusion.
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
Baseline characteristics.
| Variable | All patients | CD | Adrenal CS |
|---|---|---|---|
| Participants (n) | 174 | 135 | 35 |
| Female (%) | 135/174 (77.6%) | 102/135 (75.6%) | 32/35 (91.4%) |
| Median age at diagnosis (y) | 44 (35–55) | 43 (32–55) | 47 (36–54) |
| Median BMI at diagnosis (kg/m2) | 28.3 (24.7–32.4) | 28.6 (24.7–32.9) | 28.0 (26.0–31.8) |
| Median duration of symptoms before diagnosis of CS (years) | 3.0 (1.0–5.6) | 3.0 (1.0–6.0) | 3.5 (1.5–5.6) |
| Median times upper limit of normal UFC at diagnosis | 3.7 (1.9–5.8) | 3.9 (2.0–6.4) | 2.4 (1.4–4.1) |
| Median cortisol after DST (nmol/L) | 480 (320–630) | 460 (290–620) | 550 (330–710) |
| Median salivary cortisol at diagnosis (nmol/L) | 8.6 (5.4–15.4) | 10.1 (5.9–18.0) | 6.1 (4.0–8.9) |
| Median follow up (years) | 6.8 (2.2–12.6) | 8.4 (3.0–13.5) | 2.2 (1.2–4.7) |
| Preoperative medical therapy* (n) | 120/174 (69%) | 106/135 (78.5%) | 10/35 (28.6%) |
| Pituitary microadenoma/macroadenoma/no adenoma detected on MRI scan (n) | - | 64/27/28** | - |
| Bilateral disease (n) | - | - | 7/35 (20.0%) |
CD, Cushing’s disease; CS, Cushing’s syndrome; BMI, body mass index; UFC, 24-h urine free cortisol; DST, 1 mg dexamethasone suppression test. Continuous data are summarized as median and interquartile ranges. Categorical data are presented as frequencies and percentages.
Cortisol-lowering medication, either metyrapone or ketoconazole.
Missing data on MRI in 16 patients.
Recovery rates and recovery times of adrenal function
The probability of recovery of AI for CD, adrenal CS and ectopic CS are depicted in Fig. 2. The 1-year, 2-year and 5-year recovery rates of adrenal function for the entire cohort were 40.1, 73.4 and 83.3%, respectively. The median time to recovery of adrenal function was 13.9 months. The 1-year, 2-year and 5-year recovery rates of patients with CD were 37.8, 70.1 and 81.1%, respectively. The median recovery time was 13.9 months for patients with CD. For patients with adrenal CS, the 1-year, 2-year and 5-year recovery rates were higher: 49.3, 86.9 and 91.3%, respectively (two-sided log-rank test: P = 0.14). The median recovery time for patients with adrenal CS was 12.1 months. Seven out of the 35 patients with adrenal Cushing had bilateral disease. The median time to recovery in patients with bilateral disease was 17.5 versus 11.0 months in patients with unilateral disease.
Cumulative probability of recovery of adrenal function in CD (n = 135), adrenal CS (n = 35) and ectopic Cushing (n = 4).
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
Of the 15 evaluated patients with ectopic CS, only four patients underwent successful resection of the ectopic tumor and were included in our study. All four patients had a neuroendocrine tumor of the lung and recovered from AI. The time to recovery of adrenal function was known in three patients: 5.7, 7.9 and 14.5 months.
Factors associated with recovery of adrenal function
Age at diagnosis
Figure 3 shows the Kaplan–Meier curves of three different age groups (group 1: 0–35 years old, group 2: 36–55 years old and group 3: 56–100 years old). The 1-year recovery rates of patients aged between 0–35, 36–55 and 56–100 years old were 54.6, 37.2 and 31.4%, respectively. The 2-year recovery rates were 79.3, 72.6 and 68.4%, respectively and the 5-years recovery rates were 89.6, 83.8 and 75.1%, respectively. The median times to recovery of adrenal function of patients aged between 0–35, 36–55 and 56–100 years old were 11.2, 13.4 and 17.6 months, respectively. The probability of recovery of AI was higher in young patients (0–35 years old) (two-sided log-rank test: P = 0.05).
Cumulative probability of recovery of adrenal function by age groups.
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
Recurrence after primary treatment
In total, 17.8% patients with CD (24/135) had developed a recurrence during follow-up. Figure 4 shows the Kaplan–Meier curves with the probability of recovery of AI of the groups with and without recurrence during follow-up in patients with CD. The probability of recovery of AI was higher in patients with a recurrence (two-sided log-rank test: P-value = 0.02). In patients with a recurrence, the 1-, 2- and 5-year recovery rates of AI were 60.9, 78.3 and 87.0%, respectively. In patients without a recurrence, the 1-, 2- and 5-years recovery rates of AI were 32.6, 68.3 and 79.7%, respectively. The median time to recovery of adrenal function in patients with CD with and without recurrence was 9.9 versus 14.4 months, respectively.
Cumulative probability of recovery of adrenal function by recurrence during follow-up in patients with CD.
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
There was only one patient with adrenal CS with a recurrence. This was a patient with bilateral macronodular hyperplasia. During the first surgery, the largest adrenal was removed. However, 3 years later, the contralateral adrenal was also removed because of the recurrence of CS.
Hypopituitarism after pituitary surgery
In patients with CD, we performed a sub-analysis based on the presence of anterior pituitary deficiencies after pituitary surgery for CD, besides AI. Antidiuretic hormone (ADH) deficiency was not included in this analysis. As expected after pituitary surgery and in line with the literature, temporary ADH deficiency occurred in a substantial part of the patients after surgery (23). Therefore, only central hypothyroidism, hypogonadotropic hypogonadism and growth hormone deficiency were taken into account (Fig. 5). The probability of recovery of AI was lower in patients with one or more pituitary deficiencies versus patients with intact pituitary function after surgery (two-sided log-rank test: P-value = 0.05). In patients with anterior pituitary deficiencies, the 1-, 2- and 5-years recovery rates of AI were 35.6, 60.4 and 67.6%, respectively. In patients without anterior pituitary deficiencies, the 1-, 2- and 5-years recovery rates of AI were 39.2, 76.0 and 89.1%, respectively. The median time to recovery of adrenal function in patients with CD with and without anterior pituitary deficiencies was 15.9 versus 13.4 months, respectively. Figure 6 shows the Kaplan–Meier curves by the number of hormonal deficiencies of the anterior pituitary gland, other than AI. Although statistical significance was not reached, there is a trend showing that the more postoperative hormonal deficiencies present, the lower the probability of recovery of AI is (two-sided log-rank test: P-value = 0.15).
Kaplan–Meier curve by the presence/absence of hormonal deficiencies of the anterior pituitary gland (other than AI) after surgery in patients with CD.
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
Kaplan–Meier curve by the number of hormonal deficiencies of the anterior pituitary gland after surgery in patients with CD.
Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0612
Preoperative cortisol-lowering medical therapy, year of surgery and fasting cortisol concentration at the initial postoperative evaluation
Sub-analyses regarding patients who received PMT versus patients without PMT did not show any difference in the probability of recovery. In patients without PMT, the 1-, 2- and 5-years recovery rates of AI were 42.6, 77.6 and 85.2%, respectively. In patients with PMT, the 1-, 2- and 5-years recovery rates of AI were 41.6, 73.7 and 82.3%, respectively. The median time to recovery of adrenal function in patients without PMT and with PMT was 14.1 versus 13.5 months, respectively.
Sub-analyses regarding patients operated on before versus after 2010, regarding the results of the 1 mg dexamethasone suppression test at diagnosis and regarding patients with a low versus slightly higher postoperative morning cortisol within 7 days after surgery (<100 versus 100–200 nmol/L) also did not show any difference in the probability of recovery of adrenal function.
Table 2 shows HRs of univariate and multivariate Cox regression analyses. Adrenal CS and ectopic CS were associated with a higher probability of recovery of AI in comparison with patients with CS. Higher age was associated with a lower probability of recovery of AI.
Uni- and multivariate Cox regression analyses.
| Variable | Univariate Cox regression | Multivariate Cox regression | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Etiology of CS (CD/adrenal CS/ectopic) | 1.44 | 1.00–2.08 | 0.05 | 1.76 | 1.11–2.80 | 0.02 |
| Etiology of CS (CD/adrenal CS) | 1.42 | 0.91–2.22 | 0.12 | |||
| Age (decades) | 0.81 | 0.71–0.93 | 0.002 | 0.83 | 0.70–0.98 | 0.03 |
| Sex (male/female) | 1.02 | 0.68–1.55 | 0.92 | 0.74 | 0.46–1.20 | 0.22 |
| BMI (kg/m2) | 1.00 | 0.97–1.02 | 0.81 | 1.01 | 0.97–1.05 | 0.61 |
| Duration of symptoms before diagnosis (years) | 0.95 | 0.90–1.01 | 0.08 | 0.95 | 0.89–1.01 | 0.09 |
| UFC (ULN) | 1.03 | 0.99–1.07 | 0.15 | 1.01 | 0.97–1.05 | 0.57 |
| Post-operative cortisol 0.10–0.20 versus <0.10 mcmol/L | 0.92 | 0.56–1.50 | 0.73 | 1.16 | 0.58–2.30 | 0.67 |
| In patients with CD only | ||||||
| PMT (no/yes) | 1.23 | 0.75–2.02 | 0.40 | 1.26 | 0.53–3.02 | 0.60 |
| Hormonal deficiencies of the anterior pituitary gland, other than AI (no/yes) | 0.65 | 0.43–0.99 | 0.05 | 0.67 | 0.40–1.11 | 0.12 |
| Micro/macroadenoma | 1.13 | 0.70–1.83 | 0.62 | 1.42 | 0.79–2.52 | 0.24 |
PMT, preoperative medical treatment; HR, hazard ratio; CI, confidence interval; CS, Cushing’s syndrome; CD, Cushing’s disease; BMI, body mass index; UFC (ULN), times upper limit 24-h urine free cortisol; AI, adrenal insufficiency. The model of multivariate analysis for the whole group included the variables: etiology of CD, age, sex, BMI, duration of symptoms before diagnosis, UFC and postoperative cortisol 0.10–0.20 versus <0.10 mcmol/L. The model of multivariate analysis for the patients with CD only included the variables: etiology of CD, age, sex, BMI, duration of symptoms before diagnosis, UFC, postoperative cortisol 0.10–0.20 versus <0.10 mcmol/L, preoperative medical treatment, hormonal deficiencies of the anterior pituitary gland other than AI and micro/macroadenoma.
Discussion
In this study, we investigated the recovery rate of adrenal function and time to recovery after first-line treatment in patients with CS. The main finding is that the recovery rates of adrenal function are high. However, it may take several months to years before recovery of adrenal function occurs.
Patients with adrenal CS had higher recovery rates than patients with CD. This can be explained by the fact that the cortisol excess is generally less severe in adrenal CS and the fact that one adrenal gland remains completely intact after unilateral adrenalectomy. By contrast, patients who undergo pituitary surgery are at risk of developing new pituitary hormone deficiencies, including corticotrope deficiency, due to permanent structural damage to the pituitary gland. Our finding that patients with additional pituitary deficiencies after surgery for CD had lower recovery rates of adrenal function supports this hypothesis.
The recovery rates of adrenal function in CD, as well as in adrenal CS, are higher than what was reported in some previous studies (10, 11, 12), but are similar to other reports (13, 15, 17, 18). As shown in Table 3, it is difficult to compare previous studies because they all differ in design, study population and inclusion and exclusion criteria. For example, Berr et al. and Klose et al. used a different cutoff value of postoperative cortisol (<100 nmol/L) than we did (<200 nmol/L) to define initial AI shortly after surgery. However, only 25 patients in our cohort had a postoperative morning cortisol between 100 and 200 nmol/L and sub-analysis of patients with a morning cortisol <100 nmol/L versus patients with a morning cortisol between 100 and 200 nmol/L did not show any difference in recovery rate or time. Another difference between studies is the strategy for tapering off and stopping glucocorticoids in the postoperative period. In our study, patients started with 30 mg hydrocortisone per day after surgery. One might expect that a higher dose of hydrocortisone leads to a longer time to recovery of adrenal function. However, there are no data or evidence-based guidelines regarding the best strategy for tapering off and stopping glucocorticoids in the postoperative period.
Overview of previous studies regarding recovery of adrenal function after surgery in patients with CS.
| Author | n, etiology | Recovery rate AI | Time to recovery, years | Follow up years | Definition of AI/remission | Substitution therapy (start doses) | Recurrence rate (CD) |
|---|---|---|---|---|---|---|---|
| Alexandraki, 2013 (8) | 131 CD | 49/81 (60.5%) during follow up | Median 1.5 years | Minimum 6 years, mean 15.9 ± 6 years | Postoperative cortisol ≤50 nmol/L | Prednisolone 5 + 2 mg or HC 20 mg in divided doses | 22.7% (microadenoma) 33.3% macroadenoma |
| Berr, 2015 (9) | 5-year: | Median: | Mean 8.2 years | Morning cortisol ≤100 nmol/L | HC 40–50 mg/day | ||
| 54 CD | CD: 58% | CD: 1.4 years | CD: 7.0 years | ||||
| 26 ACS | ACS: 38% | ACS: 2.5 years | ACS: 8.5 years | ||||
| 11 ECS | ECS: 82% | ECS: 0.6 years | ECS: 13.5 years | ||||
| Serban, 2019 (12) | 61 CD | 5-year: | Median 1.6 years | Minimum 3 years, median 6 years | Morning cortisol <3 μg/dL or cortisol after 250 μg synacthen test <18 μg/dL | Cortisone acetate 25 mg, divided in 2–3 doses | 16.4% |
| Persistent remission: 55.8% | 2.1 years | ||||||
| Recurrence: 100% | 1.0 years | ||||||
| Ciric 2012 (10) | 86 CD | 59.3% during follow up | Mean 1.1 years | Minimum 0.5 years, mean 5.7 years | Drop in immediate postoperative cortisol, range <0.5–5.3 µg/dL and symptoms | No specific unified algorithm | 9.7% |
| Klose, 2004 (11) | 2-year: | Median: | Post-operative cortisol <100 nmol/L and/or UFC <50 nmoL/24h | Hydrocortisone 20–30 mg/day | |||
| 18 CD | CD: 67% | CD: 2 years | CD: 22.2% | ||||
| 14 ACS | ACS: 79% | ACS: 2 years | ACS: 0% | ||||
| Prete, 2017 (18) | Median: | Minimum 2 years | Postoperative morning serum cortisol <5 μg/dL/138 nmol/L | Hydrocortisone 20–30 mg/day in divided in 2–3 doses | Patients with recurrence were excluded | ||
| 15 CD | CD: 1.3 years | CD: median 5.8 years | |||||
| 31 ACS | ACS: 0.8 years | ACS: Median 4.0 years | |||||
| 14 overt ACS | Overt ACS: 1.5 years | ||||||
| 17 subclinical ACS | Subclinical ACS: 0.5 years | ||||||
| Hurtado, 2018 (14) | 81 ACS | 87.8% during follow up | Median ACS: 0.4 years | Median ACS: 1.2 years | Postoperative morning (day 1) serum cortisol <10 μg/dL/276 nmol/L or hemodynamic instability or received perioperative GC due to anticipated AI after unilateral adrenalectomy | Prednisone or hydrocortisone, median hydrocortisone-equivalent dose 40 mg/day | |
| 27 severe CS | Severe: 1.0 years | Severe: 1.0 years | |||||
| 24 moderate CS | Moderate: 0.2 years | Moderate: 1.0 years | |||||
| 30 MACE | MACE: 0.2 years | MACE: 1.5 years | |||||
| Dalmazi, 2014 review on adrenal function after adrenalectomy for subclinical CS, 28 studies (17) | ACS: 376 overt ACS 141 subclinical ACS | Overt ACS: 93.4% subclinical ACS: 97.9% | Mean overt ACS: 0.9 years subclinical ACS 0.5 years |
CD, Cushing’s disease; ACS, adrenal Cushing’s syndrome; ECS, ectopic Cushing’s syndrome; AI, adrenal insufficiency; Subclin: subclinical; MACE, mild autonomous cortisol excess.
One might also hypothesize that the studies reporting high recurrence rates are related to higher recovery rates in CD patients. In our study, the recurrence rate was 17.8%, which is in line with previous studies (9, 13, 24). The establishment of recovery of adrenal function in patients with a recurrence later on is a difficult matter: despite the exclusion of patients with immediate obvious persistent disease in our study, recovery of glucocorticoid secretion in patients who developed a recurrence later on could be an early manifestation of recurrence instead of true recovery of physiological adrenal function. A striking finding in this study, in line with the aforementioned hypothesis, was the considerably higher 1-year recovery rate and the shorter time to recovery of patients with a recurrence in comparison to patients without a recurrence. Recovery of adrenal function is more rapid in patients with recurrences (13, 25). These findings imply that in case of an early recovery of adrenal function, clinicians should be aware of a possible recurrence of CD.
Another difference between studies is the inclusion or exclusion of patients with mild autonomous cortisol secretion (MACS), formerly known as subclinical CS. Previous studies have shown that patients with subclinical CS have a higher probability of recovery and a shorter duration of AI (14, 15, 16, 18). In our study, only two patients were diagnosed with subclinical CS (in this study characterized as inadequate suppression after DST in combination with values of UFC within the reference range) and therefore subgroup analysis was not possible.
In the present study, a rather high number of patients received PMT in comparison to other studies. In our institution, it was common practice to start PMT 3 months before pituitary surgery in patients with CD with the aim to improve hemostasis and other Cushing-related comorbidities, although the benefit of PMT has not yet been well established by randomized controlled trials. At the liberty of the treating physician, the dose of ketoconazole or metyrapone was titrated with the aim to normalize the 24-h UFC excretion. The doses needed to achieve normal 24-h UFC and the time to normalization of 24-h UFC varied between patients.
One could hypothesize that lowering cortisol levels during the weeks to months before surgery may result in a faster recovery of adrenal function. However, this was not the case in this study.
Overall, the present study shows a high recovery rate of adrenal function after treatment for CS. The time until recovery is partly dependent on the strategy and success of tapering off of glucocorticoids replacement and therefore may be very long because of GWS. These are meaningful findings. Tapering glucocorticoid substitution in parallel with the recovery of cortisol secretion after surgery for CS is often a challenging and lengthy trajectory for both patients and physicians. The lack of standardization of the follow-up and of the tapering protocols, the need for constant shared decision-making and personalized support for patients, particularly of those who are also confronted with severe associated comorbidities and unpredictable withdrawal symptoms, may discourage patients and physicians from proceeding in this endeavor. Given the rarity of the disease, knowledge on this topic is scarce. Previous, mainly smaller studies reported a wide range of recovery rates of adrenal function after first-line treatment for CS (varying between 37 and 93% for CD, and for overt adrenal CS between 38 and 93%) (10, 11, 12, 13, 15, 17, 18). The rather low percentages of recovery of adrenal function in some of these previous studies could discourage patients and physicians to persevere the attempt to taper off hydrocortisone. Our findings in a large cohort of patients with CS, including a sizable subgroup of patients with CD, allow us to deepen the multivariate analysis to uncover factors that are associated with a better chance of recovery. The data indicate that in this real-life setting, despite the long time to achieve recovery, the recovery rates are high and while this occurs for most of the patients within 1–2 years after treatment, recovery is still possible even after a longer follow-up. Moreover, this study showed that the recovery rate is higher in patients with adrenal CS versus CD, in younger patients and in patients with CD with preserved pituitary function after pituitary surgery. These findings are very important for clinical practice. They highlight the importance of continuing to taper off the glucocorticoids, if necessary slowly and steadily, in the years after surgery. They also help us better inform the patients beforehand and to improve the management and the expectations of both patients and physicians to motivate them to persevere in tapering of the glucocorticosteroids while considering the factors such as those identified to influence the chance of recovery during their personalized counseling and guidance of the patients in this often very difficult and lengthy period.
In our institution, it is common practice to counsel and provide guidance intensively to patients in this difficult period, both by the treating physician and a specialized nurse, as we consider this coordinated guidance of utmost importance. Moreover, all patients are provided with contact details so that they can reach to us for advice 24 h a day, either by phone or by secure email throughout this process. When indicated, patients are referred to other healthcare professionals such as psychologists, physical therapists, social workers and other specialists.
One important strength of our study is the large size of our single-center cohort, considering the rarity of the disease. This has also allowed us to do subgroup analyses and assess factors associated with recovery from postoperative AI. The limitations include the retrospective character of this study and the fact that patients were included over a long period of time (1968 to 2022) during which diagnostic tools and management protocols for CS have somewhat changed over this period of time. We have tried to mitigate the limitations that are inevitable with a retrospective study by being thorough and extensive in the quality and amount of data that we were able to collect. In addition to that, the diagnostic assessment and the treatment of the patients followed very strict and uniform protocols in conformity with the internationally recognized clinical guidelines available at the time. On the other hand, the fact that this represents a real-life study renders the results more relatable for clinical practitioners and strengthens its impact.
We collected data from medical records regarding the duration of CS-related signs and symptoms before diagnosis, as mentioned by the patient during history taking. We are well aware that these data are rather subjective and dependent on the accuracy of the recollection of the patient. However, this is the only way to assess the duration of symptoms before diagnosis. In our opinion, these data still could be very valuable.
In conclusion, our study shows that the large majority of patients with CS recover their adrenal function after first-line surgical treatment, even though the time to recovery may take several months to years. Informing patients beforehand and providing support, encouragement and guidance in this process is therefore paramount. Herewith, one could consider factors such as the age of the patient, the etiology of CS and the presence of additional pituitary deficiencies after pituitary surgery. In case of an early recovery of adrenal function, clinicians should be aware of a possible recurrence of CD. Future studies should establish the optimal postoperative management for CS to improve the chance for success of recovery of adrenal function.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
References
- 1↑
Lacroix A , Feelders RA , Stratakis CA , et al. Cushing's syndrome. Lancet 2015 386 913–927. (https://doi.org/10.1016/s0140-6736(14)61375-1)
- 2↑
Fassnacht M , Tsagarakis S , Terzolo M , et al. European Society of Endocrinology clinical practice guidelines on the management of adrenal incidentalomas, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 2023 189 G1–G42. (https://doi.org/10.1093/ejendo/lvad066)
- 3↑
Balasko A , Zibar Tomsic K , Kastelan D , et al. Hypothalamic–pituitary–adrenal axis recovery after treatment of Cushing's syndrome. J Neuroendocrinol 2022 34 e13172. (https://doi.org/10.1111/jne.13172)
- 4↑
Hermus AR , Pieters GF , Pesman GJ , et al. Coexistence of hypothalamic and pituitary failure after successful pituitary surgery in Cushing's disease? J Endocrinol Investig 1987 10 365–369. (https://doi.org/10.1007/bf03348149)
- 5↑
Fitzgerald PA , Aron DC , Findling JW , et al. Cushing's disease: transient secondary adrenal insufficiency after selective removal of pituitary microadenomas; evidence for a pituitary origin. J Clin Endocrinol Metab 1982 54 413–422. (https://doi.org/10.1210/jcem-54-2-413)
- 6↑
He X , Findling JW & Auchus RJ . Glucocorticoid withdrawal syndrome following treatment of endogenous cushing syndrome. Pituitary 2022 25 393–403. (https://doi.org/10.1007/s11102-022-01218-y)
- 7↑
Balasko A , Zibar Tomsic K , Kastelan D , et al. Hypothalamic–pituitary–adrenal axis recovery after treatment of Cushing's syndrome. J Neuroendocrinol 2022 34 e13172. (https://doi.org/10.1111/jne.13172)
- 8↑
Acree R , Miller CM , Abel BS , et al. Patient and provider perspectives on postsurgical recovery of cushing syndrome. J Endocr Soc 2021 5 bvab109. (https://doi.org/10.1210/jendso/bvab109)
- 9↑
Alexandraki KI , Kaltsas GA , Isidori AM , et al. Long-term remission and recurrence rates in Cushing's disease: predictive factors in a single-centre study. Eur J Endocrinol 2013 168 639–648. (https://doi.org/10.1530/eje-12-0921)
- 10↑
Berr CM , Di Dalmazi G , Osswald A , et al. Time to recovery of adrenal function after curative surgery for Cushing's syndrome depends on etiology. J Clin Endocrinol Metab 2015 100 1300–1308. (https://doi.org/10.1210/jc.2014-3632)
- 11↑
Ciric I , Zhao JC , Du H , et al. Transsphenoidal surgery for Cushing disease: experience with 136 patients. Neurosurgery 2012 70 70–81. (https://doi.org/10.1227/neu.0b013e31822dda2c)
- 12↑
Klose M , Jorgensen K & Kristensen LO . Characteristics of recovery of adrenocortical function after treatment for Cushing's syndrome due to pituitary or adrenal adenomas. Clin Endocrinol 2004 61 394–399. (https://doi.org/10.1111/j.1365-2265.2004.02111.x)
- 13↑
Serban AL , Sala E , Carosi G , et al. Recovery of adrenal function after pituitary surgery in patients with cushing disease: persistent remission or recurrence? Neuroendocrinology 2019 108 211–218. (https://doi.org/10.1159/000496846)
- 14↑
Alesina PF , Hommeltenberg S , Meier B , et al. Posterior retroperitoneoscopic adrenalectomy for clinical and subclinical Cushing's syndrome. World J Surg 2010 34 1391–1397. (https://doi.org/10.1007/s00268-010-0453-0)
- 15↑
Hurtado MD , Cortes T , Natt N , et al. Extensive clinical experience: hypothalamic–pituitary–adrenal axis recovery after adrenalectomy for corticotropin-independent cortisol excess. Clin Endocrinol 2018 89 721–733. (https://doi.org/10.1111/cen.13803)
- 16↑
Kim HK , Yoon JH , Jeong YA , et al. The recovery of hypothalamic–pituitary–adrenal Axis is rapid in subclinical cushing syndrome. Endocrinol Metab 2016 31 592–597. (https://doi.org/10.3803/enm.2016.31.4.592)
- 17↑
Morelli V , Minelli L , Eller-Vainicher C , et al. Predictability of hypoadrenalism occurrence and duration after adrenalectomy for ACTH-independent hypercortisolism. J Endocrinol Investig 2018 41 485–493. (https://doi.org/10.1007/s40618-017-0788-6)
- 18↑
Di Dalmazi G , Berr CM , Fassnacht M , et al. Adrenal function after adrenalectomy for subclinical hypercortisolism and Cushing's syndrome: a systematic review of the literature. J Clin Endocrinol Metab 2014 99 2637–2645. (https://doi.org/10.1210/jc.2014-1401)
- 19↑
Prete A , Paragliola RM , Bottiglieri F , et al. Factors predicting the duration of adrenal insufficiency in patients successfully treated for Cushing disease and nonmalignant primary adrenal Cushing syndrome. Endocrine 2017 55 969–980. (https://doi.org/10.1007/s12020-016-1007-5)
- 20↑
Doherty GM , Nieman LK , Cutler GB Jr , et al. Time to recovery of the hypothalamic–pituitary–adrenal axis after curative resection of adrenal tumors in patients with Cushing's syndrome. Surgery 1990 108 1085–1090.
- 21↑
Flitsch J , Ludecke DK , Knappe UJ , et al. Correlates of long-term hypocortisolism after transsphenoidal microsurgery for Cushing's disease. Exp Clin Endocrinol Diabetes 1999 107 183–189. (https://doi.org/10.1055/s-0029-1212095)
- 22↑
Nieman LK , Biller BM , Findling JW , et al. The diagnosis of Cushing's syndrome: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2008 93 1526–1540. (https://doi.org/10.1210/jc.2008-0125)
- 23↑
Brooks EK & Inder WJ . Disorders of salt and water balance after pituitary surgery. J Clin Endocrinol Metab 2022 108 198–208. (https://doi.org/10.1210/clinem/dgac622)
- 24↑
Atkinson AB , Kennedy A , Wiggam MI , et al. Long-term remission rates after pituitary surgery for Cushing's disease: the need for long-term surveillance. Clin Endocrinol 2005 63 549–559. (https://doi.org/10.1111/j.1365-2265.2005.02380.x)
- 25↑
Lonser RR , Nieman L & Oldfield EH . Cushing's disease: pathobiology, diagnosis, and management. J Neurosurg 2017 126 404–417. (https://doi.org/10.3171/2016.1.jns152119)
