ENDOCRINE SOCIETY
OXFORD
Clinical Research Article
Epidemiology and Comorbidity of Adrenal Cushing Syndrome: A Nationwide Cohort Study
Chang Ho Ahn,1,2 Jung Hee Kim,1,3 Man Young Park,4 and Sang Wan Kim 1,5
1Department of Internal Medicine, Seoul National University College of Medicine, Seoul, 03080, Republic of Korea; 2Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam, 13620, Republic of Korea; 3Department of Internal Medicine, Seoul National University Hospital, Seoul, 03080, Republic of Korea; 4Korea Institute of Oriental Medicine, Daejeon, 34054, Republic of Korea; 5Department of Internal Medicine, Seoul Metropolitan Government Boramae Medical Center, Seoul, 07061, Republic of Korea
ORCID number: 0000-0003-1932-0234 (J. H. Kim).
Abbreviations: ACTH, adrenocorticotropic hormone; CI, confidence interval; CRH, corticotropin-releasing hormone;CS, Cushing syndrome;CT, computed tomography; HPA, hypothalamic-pituitary-adrenal;KNHANES, Korean National Health and Nutrition Examination Survey; NHIS, National Health Insurance Services;OR, odds ratio.
Received: 15 June 2020; Editorial Decision: 11 October 2020; First Published Online: 19 October 2020; Corrected and Typeset: 11 November 2020.
Abstract
Context: Adrenal Cushing syndrome (CS) is a major subtype of CS and has a high surgical cure rate. However, only a few studies have investigated the epidemiology and long-term outcomes of adrenal CS.
Objective: We aimed to investigate the nationwide epidemiology, long-term prognosis, and postoperative glucocorticoid replacement therapies of adrenal CS in Korea.
Design: Retrospective cohort study.
Setting: A nationwide claim database.
Patients: Adrenal CS patients who were defined as having undergone adrenalectomy, a diagnosis code of CS, and not having pituitary gland surgery.
Main Outcome Measures: Crude incidence and age-standardized incidence rates, long- term mortality, comorbidities diagnosed preoperatively or developed postoperatively, and the pattern of postoperative glucocorticoid replacement therapy.
Results: From 2002 to 2017, there were a total of 1199 new adrenal CS patients, including 72 patients with adrenocortical carcinoma (malignant adrenal CS), in Korea. The crude and age-standardized incidence rates were 1.51 and 1.27 per million person-years, respectively. The overall standardized mortality ratio was 3.0 (95% confidence interval [CI], 2.4-3.7) for benign adrenal CS and 13.1 (95% CI, 7.6-18.6) for malignant adrenal CS. Adrenal CS patients had a high risk of having coronary artery disease, stroke, metabolic diseases, and depression. A similar proportion of patients were diagnosed with these comorbidities both preoperatively and postoperatively, suggesting a significant residual risk even after adrenalectomy. The median time of postoperative glucocorticoid
replacement therapy was 10.1 months, and the major types of glucocorticoids used were prednisolone (66.6%) and hydrocortisone (22.4%).
Conclusions: Adrenal CS is associated with multiple comorbidities even after treatment, which necessitates meticulous postoperative care.
Key Words: adrenal Cushing syndrome, epidemiology, comorbidity, postoperative glucocorticoid replacement
Cushing syndrome (CS) is a syndrome characterized by glucocorticoid excess and leads to various complications and mortality. Endogenous CS cases are categorized as adrenocorticotropic hormone (ACTH)-dependent CS and ACTH-independent CS. ACTH-dependent CS can be caused by ACTH-producing pituitary adenoma (Cushing disease) or ectopic ACTH or corticotropin-releasing hor- mone (CRH) secretion (ectopic CS). ACTH-independent CS is almost exclusively caused by excess glucocorticoid secretion from the adrenal gland (adrenal CS). The most common subtype of endogenous CS is Cushing disease, which accounts for 70% of endogenous CS cases in Western countries (1). Adrenal CS accounts for 20% to 25% of en- dogenous CS cases, and the remaining cases are ectopic CS (1). The underlying pathologies of adrenal CS include cortisol-producing adrenal adenoma, adrenal cortical car- cinoma, and bilateral hyperplasia of the adrenal gland.
Although adrenal CS is the major subtype of endogenous CS, there are only a limited number of studies that have comprehensively studied the epidemiology and long-term prognosis of adrenal CS. The long-term survival of adrenal CS patients varied across previous studies. The standard- ized mortality ratio of adrenal CS ranged from 1.7 to 7.5 (2-6). However, these studies included only small numbers of patients (ranging from 16 to 84 patients). The long-term prognosis of adrenal CS needs to be determined with a sufficient number of patients and a sufficient follow-up duration.
CS is also associated with various complications, including metabolic diseases, cardiovascular diseases, osteoporosis, thrombotic diseases, and mental illnesses. The resolution of hypercortisolemia can improve the metabolic profile and halt the progression of these comorbidities. However, the risk of several complications remains elevated, even after the remission of CS (2, 7). Thus, the long-term follow-up data of various complications after the surgical treatment of adrenal CS can be useful for determining the optimal follow-up strategy for adrenal CS.
After the treatment of CS, patients experience postoperative adrenal insufficiency resulting from chronic suppression of the hypothalamic-pituitary-adrenal (HPA) axis. Glucocorticoid replacement therapy is required until the HPA axis recovers. A previous study reported that the recovery time varied between patients with CS of different
etiologies (8). In this study, adrenal CS had a longer re- covery time than did Cushing disease and ectopic CS. Furthermore, approximately 40% of adrenal CS patients still exhibited adrenal insufficiency over the mean follow-up period of 8.5 years (8). However, other previous studies on adrenal CS reported recovery times of 12 months (9) and 15 months (10), but these studies included only small num- bers of patients.
Herein, we aimed to investigate the national epidemi- ology, comorbidities, postoperative glucocorticoid replace- ment therapies, and long-term prognosis of adrenal CS in Korea using nationwide data based on the National Health Insurance Service.
Methods
Data sources
We used the National Health Insurance Services (NHIS) database in Korea. The insurance system was established by the Korean government and covered approximately 97.2% of Korean residents in all age groups. The NHIS database contains data on all forms of healthcare services reim- bursed by the insurance system, including demographics, diagnoses, prescriptions, diagnostic or surgical procedures, and medical costs associated with the claims. The diagnoses were classified by the 10th version of the International Classification of Diseases (ICD-10). The data used for this study comprised all NHIS databases from 2002 to 2018. The incidence of adrenal CS was investigated from 2002 to 2017, and the comorbidities and survival rate were ana- lyzed from 2002 to 2018 for at least 1 year of follow-up data.
Operational definition of adrenal CS
Adrenal CS was defined as any patients fulfilling all 3 fol- lowing criteria: (i) a procedural code for unilateral or bi- lateral adrenalectomy (P4571, P4572), (ii) ICD-10 code of CS (E240, E248, E249, E270), and (iii) the absence of pro- cedural codes for pituitary gland surgery (S4633, S4743). The number of patients fulfilling the above criteria was 1230. Patients with a diagnostic code of C741 (malig- nant neoplasm of adrenal gland medulla) were excluded (n = 21). Given that ectopic or pituitary CS patients can be
treated with bilateral adrenalectomy, patients with a diag- nostic code of ectopic CS (E243) or pituitary CS (E240) and underwent bilateral adrenalectomy (P4572) were also excluded (n = 10). As a result, a total of 1199 patients were included in the final analysis. Among all adrenal CS pa- tients, those with the ICD-10 code of the neoplasm of the adrenal gland (C740, C749) were classified as malignant adrenal CS patients, and the remaining patients were clas- sified as benign adrenal CS patients.
Validation of the definition of adrenal CS
To validate the operational definition of adrenal CS, we checked the false-positive rate and false-negative rate of the operational definition using our institution’s database. To estimate the false-positive rate, we identified patients fulfilling the above definition of adrenal CS at our institu- tion from January 2010 to September 2019. A total of 88 patients fulfilled the criteria. The medical records of these patients were reviewed, and 86 (97.7%) patients were con- firmed as having true adrenal CS. The remaining 2 patients were those with Cushing disease who underwent bilat- eral adrenalectomy due to uncontrolled hypercortisolemia despite having undergone pituitary-directed treatments. To estimate the false-negative rate, we applied the above definition to our institution’s cohort of 61 proven adrenal CS patients who were prospectively enrolled in a registry of adrenal tumors. Among these patients, 58 (95.1%) ful- filled the criteria. The remaining 3 patients did not have either the diagnosis code of CS or a procedural code for adrenalectomy. From these analyses, the false-positive and false-negative rates of the adrenal CS definition were esti- mated to be 2.3% and 4.9%, respectively.
Comorbidities of adrenal CS
Comorbidities were considered to exist if the following diagnostic codes for at least 2 principal or secondary diagnosis were present. The comorbidities examined were depressive disorder (F32, F33, F34), psychotic dis- order (F20-F29), osteoporosis (M80, M81, M82), fracture (M843, M844, S12, S22, S32, S42, S52,S62,S72,S82, S92, T02), tuberculosis (A15-A19), other bacterial infection (A20-A28, A30-A49, A50-A64), coronary artery disease (I20-I25), ischemic stroke (I63, I64, I693, I694, G45), hem- orrhagic stroke (I60-I62, I690, I692), pulmonary embolism (I26), diabetes mellitus (E11, E12, E13, E14), dyslipidemia (E78), and hypertension (I10-I13, I15). If comorbidities existed before adrenalectomy, the comorbidities were classified as “preoperative comorbidities.” Based on the time of the first appearance of each diagnosis code, each
comorbidity was classified as being present at more than 1 year before adrenalectomy (>1 year preoperatively) and within 1 year before adrenalectomy (within 1 year pre- operatively). For preoperative comorbidities, the propor- tion of patients who were preoperatively diagnosed with each comorbidity among all patients is shown. To assess the risk of having preoperative comorbidities in adrenal CS pa- tients, we compared the prevalence of each comorbidity in adrenal CS patients with that in a control population. The control population comprised age- and sex-matched (1:2) individuals to the adrenal CS patients selected from the 2015 Korean National Health and Nutrition Examination Survey (KNHANES) database. The KNHANES is a na- tionally representative cross-sectional survey that included approximately 10 000 individuals (11). In the KNHANES database, information on depressive disorder, osteopor- osis, tuberculosis, coronary artery disease, stroke, diabetes mellitus, dyslipidemia, and hypertension were available, but information on fracture, bacterial infection, psych- otic disorder, aortic disease, and pulmonary embolism were not available. The odds ratios (ORs) for having these comorbidities were calculated.
Ifcomorbidities werenewly diagnosedafteradrenalectomy, they were classified as “postoperative comorbidities.” For postoperative comorbidities, we calculated the proportion of patients who were diagnosed with each comorbidity postoperatively among those without the comorbidity at the time of surgery. To explore the time interval to the diagnosis of comorbidities from adrenalectomy, we also as- sessed the results according to the time of the first diagnosis code of each comorbidity as follows: within 3 months after adrenalectomy (within 3 months postoperatively), within 3 months to 1 year after adrenalectomy (3 months to 1 year postoperatively), and more than 1 year after adrenalectomy (>1 year postoperatively). The median time to diagnosis from adrenalectomy was analyzed for each comorbidity.
Analysis of postoperative glucocorticoid replacement therapy
We examined the type of glucocorticoid used for postoperative replacement therapy and the duration of glucocorticoid replacement after adrenalectomy. Considering that the type of oral glucocorticoid can be modified for each patient during postoperative replacement therapy, the type of oral glucocorticoid used for the longest time after adrenalectomy by each patient was recorded as the representative type. The date of the last glucocorticoid replacement therapy was determined as the last day of any oral glucocorticoid prescription, with more than a 1-year glucocorticoid-off period.
Statistical analysis
Subjects’ demographic characteristics were summarized as the means ± standard deviations for interval variables and numbers (%) for categorical variables. The median time and interquartile range are presented for the follow-up and time to diagnosis of each comorbidity. The crude inci- dence rate and age-standardized incidence rate of adrenal CS were calculated. The WHO world standard population (12) was used to calculate the age-standardized incidence rate. The Kaplan-Meier survival curves of malignant ad- renal CS (adrenal CS due to adrenocortical carcinoma) and benign adrenal CS (adrenal CS due to noncarcinoma diseases) were compared by the log-rank test. The Cox proportional hazards model was used to evaluate the im- pact of age, sex, comorbidities of CS, and having bilateral adrenalectomy on mortality. A P value of <0.05 was con- sidered statistically significant. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute, Cary, NC, USA), and R, version 3.5.0 (R Foundation for Statistical Computing, Vienna, Austria).
Ethical considerations
The Institutional Review Board of Seoul National University Hospital (IRB No. 1811-007-982 and 1801-010-911) ap- proved the study protocol. Informed consent was waived since personal information was anonymously encrypted in the NHIS and KNHANES databases.
Results
Incidence and prevalence of adrenal CS in Korea
From 2002 to 2017, 1199 patients were newly diag- nosed with adrenal CS in Korea. The mean age was 44.9 ± 13.4 years, with the peak age between 36 and 55 years (Table 1 and Fig. 1A). Patients were predominantly female, with a male:female ratio of 1:3.2. The prevalence of adrenal CS was 23.4 per million people. The overall crude
incidence rate of adrenal CS was 1.51 per million person- years, and the age-standardized incidence rate was 1.27 per million person-years (Fig. 1B). Among the total adrenal CS patients, 72 (6.0%) had malignant adrenal CS. The mean age of the patients with malignant adrenal CS was similar to that of all the patients with adrenal CS (47.8 ± 16.3 years, Table 1). The peak age was 56 to 60 years, which was slightly higher than that of all the patients with adrenal CS. There was also a female predominance in the patients with malignant adrenal CS (male:female ratio of 1:1.8).
Mortality of adrenal CS
During the follow-up period, which had a median duration of 9.7 (4.7-12.5) years, 96 patients (74 benign adrenal CS and 22 malignant adrenal CS patients) died. The 5-year sur- vival rates of benign adrenal CS and malignant adrenal CS were 96.3% (95% confidence interval [CI]: 95.2-97.5%) and 68.6% (95% CI, 55.8-79.2%), respectively (Fig. 2). The overall standardized mortality ratio was 3.0 (95% CI, 2.4- 3.7) for benign adrenal CS and 13.0 (95% CI, 7.6-18.6) for malignant adrenal CS. The most mortality (21/22) occurred during the first 5 years after adrenalectomy in the malig- nant adrenal CS patients, while mortality occurred steadily for 10 years after adrenalectomy in the benign adrenal CS patients (Fig. 2). In survival analysis, older age (HR 1.06; 95% CI, 1.04-1.08), bilateral adrenalectomy (2.92, 1.25- 6.81), stroke (3.56, 1.95-6.52), and diabetes (1.95, 1.21- 3.15) were independently associated with higher mortality in addition to the malignant adrenal CS (Table 2). These factors were still independent risk factors of mortality in patients with benign adrenal CS.
Comorbidities of adrenal CS
We investigated the preoperative and postoperative comorbidities of adrenal CS, which were categorized on the basis of whether the comorbidity was diagnosed before or after the treatment of adrenal CS. The most common
| Total | Benign adrenal CS | Malignant adrenal CS | Age and sex-matched control population | |
|---|---|---|---|---|
| Total N | 1199 | 1127 | 72 | 2404 |
| Age, years | 44.9 ± 13.4 | 44.8 ±13.3 | 47.8 ± 16.3 | 45.1 + 13.6 |
| Sex | 23.5 ± 3.6 | |||
| Male | 287 (23.9%) | 261 (23.2) | 26 (36.1%) | |
| Female | 912 (76.1%) | 866 (76.8) | 46 (63.9%) | 580 (24.1%) |
| Surgical treatment | 1824 (75.9%) | |||
| Unilateral adrenalectomy | 1161 (96.8%) | 1092 (96.9%) | 69 (95.8%) | NA |
| Bilateral adrenalectomy | 38 (3.2%) | 35 (3.1%) | 3 (4.2%) | NA |
The data are shown as the mean ± SD or n (%). Abbreviations: CS, Cushing syndrome; NA, not applicable.
A
Number of adrenal CS patients
200
150
100
50
0
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
61-65
66-70
71-75
76-80
Age group
B
Incidence per million person-years
5
Crude incidence rate
Age-standardized incidence rate
7
0
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Overall
Year
preoperative comorbidities were hypertension (67.6%), dyslipidemia (56.3%), and diabetes (37.9%), in order of prevalence (Table 3). All preoperative comorbidities were diagnosed mostly within 1 year before adrenalectomy. We compared the risk of preoperative comorbidities in ad- renal CS patients with that in the age- and sex-matched KNHANES control population. Adrenal CS patients had a clearly increased risk of having coronary artery dis- ease (OR 17.5; 95% CI, 11.8-26.0), stroke (OR 14.4; 95% CI, 8.9-23.1), dyslipidemia (OR 10.6; 95% CI, 8.9- 12.6), diabetes (OR 9.6; 95% CI, 7.8-11.8), hypertension (OR 9.2; 95% CI, 7.9-10.8), osteoporosis (OR 5.0; 95% CI, 3.9-6.3), and depression (OR 4.6; 95% CI, 3.6-6.0) (Fig. 3A). Various comorbidities also commonly occurred postoperatively (Table 3). The median time to diagnosis of the postoperative comorbidities after adrenalectomy ranged from 0 to 4.2 years. Hypertension, dyslipidemia, diabetes, coronary artery disease, and ischemic and hem- orrhagic stroke were more commonly noted as preopera- tive comorbidities than as postoperative comorbidities
(Fig. 3B). However, only hypertension was more com- monly noted as a preoperative comorbidity by more than 2-fold, whereas other comorbidities were noted as a postoperative comorbidity after adrenalectomy in a con- siderable proportion of patients. Osteoporosis, bacterial infection, and pulmonary embolism were more commonly noted as postoperative comorbidities. These findings sug- gest that adrenal CS patients still have a significant residual risk for these comorbidities, even after adrenalectomy. All comorbidities were diagnosed within 1 year before or after adrenalectomy in more than half of the patients, suggesting that this perioperative 2-year period is a high-risk period for developing various comorbidities.
Postoperative glucocorticoid replacement therapy
Among 1161 patients who had unilateral adrenalectomy, 973 (83.8%) patients were prescribed postoperative oral glucocorticoids. The most common type of glucocorticoid
1.00
- Benign —. Malignant
Survival probability
0.75
p < 0.0001
0.50
0.25
0.00
0
1
2
3
4
5
6
7
8
9
10
Number at risk (number of events)
Benign
1127 (0)
1109 (18
1039 (27)
978 (31)
923 (34)
836 (39)
762 (43)
715 (47)
659 (54)
606 (59)
530 (60)
Malignant
72 (0)
67 (5)
58 (11)
48 (17)
43 (19)
40 (21)
37 (21)
34 (21)
32 (21)
30 (22)
28 (22)
0
1
2
3
4
5
6
7
8
9
10
Follow-up duration (year)
| All adrenal CSª | Benign adrenal CSb | |||||||
|---|---|---|---|---|---|---|---|---|
| Unadjusted HR | P value | Adjusted HRd | P value | Unadjusted HRC | P value | Adjusted HRd | P value | |
| Female sex | 0.69 (0.43-1.10) | 0.116 | 1.07 (0.66-1.74) | 0.795 | 0.78 (0.44-1.38) | 0.395 | 1.06 (0.59-1.90) | 0.854 |
| Age (year) | 1.07 (1.05-1.09) | <0.001 | 1.06 (1.04-1.08) | <0.001 | 1.07 (1.05-1.1) | <0.001 | 1.06 (1.04-1.09) | <0.001 |
| Bilateral adrenalectomy | 2.53 (1.10-5.80) | 0.029 | 2.92 (1.25-6.81) | 0.013 | 3.69 (1.59-8.59) | 0.002 | 3.97 (1.66-9.50) | 0.002 |
| Coronary artery disease | 1.15 (0.57-2.30) | 0.694 | 0.52 (0.25-1.07) | 0.075 | 1.25 (0.57-2.74) | 0.585 | 0.59 (0.26-1.34) | 0.206 |
| Stroke | 4.41 (2.52-7.72) | <0.001 | 3.56 (1.95-6.52) | <0.001 | 5.22 (2.83-9.65) | <0.001 | 3.26 (1.67-6.33) | 0.001 |
| Diabetes | 2.48 (1.60-3.82) | <0.001 | 1.95 (1.21-3.15) | 0.006 | 3.27 (1.95-5.48) | <0.001 | 2.10 (1.02-3.66) | 0.009 |
| Hypertension | 1.93 (1.15-3.26) | 0.013 | 1.05 (0.59-1.85) | 0.880 | 2.14 (1.14-4.03) | 0.018 | 0.88 (0.45-1.73) | 0.714 |
| Fracture | 1.58 (0.82-3.07) | 0.173 | 1.43 (0.72-2.85) | 0.311 | 1.93 (0.95-3.91) | 0.070 | 1.31 (0.63-2.72) | 0.470 |
| Depression | 1.33 (0.67-2.67) | 0.415 | 0.99 (0.48-2.03) | 0.970 | 1.85 (0.91-3.75) | 0.090 | 1.12 (0.54-2.35) | 0.757 |
| Malignant adrenal CS | 6.72 (4.12-10.96) | <0.001 | 8.36 (4.94-14.13) | <0.001 | NA | NA | NA | NA |
Abbreviations: CS, Cushing syndrome; HR, hazard ratio; NA, not applicable.
ªA Cox proportional hazards model was applied for all 1199 adrenal CS patients.
“A Cox proportional hazards model was applied for 1127 benign adrenal CS patients.
‘Unadjusted hazard ratio was calculated for each factor in univariate Cox proportional hazards model.
dAdjusted hazard ratio was calculated in multivariate Cox proportional hazards model including all factors.
used was prednisolone (66.6%). Hydrocortisone (22.4%), methylprednisolone (5.3%), and dexametha- sone (4.9%) were also used (Fig. 4A). The median dur- ation of postoperative glucocorticoid replacement was 10.1 months (interquartile range: 4.7-20.2) (Fig. 4B). Among various types of glucocorticoid, the median
duration of postoperative glucocorticoid replacement was 7.5 (3.6-18.8) months for hydrocortisone and 11.4 (5.5- 21.0) months for prednisolone or methylprednisolone (P = 0.02). Postoperative glucocorticoid replacement con- tinued until 1, 2, and 5 years after adrenalectomy in 47.0%, 23.6%, and 10.8% of the patients, respectively.
| Preoperative comorbiditiesa | Postoperative comorbiditiesb | |||||
|---|---|---|---|---|---|---|
| >1 y preoperatively | Within 1 y preoperatively | Within 3 mos postoperatively | 3 mos to 1 y postoperatively | >1 y postoperatively | Median time to diagnosis postoperatively ℃ (y) | |
| Hypertension | 220 (18.3%) | 590 (49.2%) | 64 (5.3%) | 5 (0.4%) | 20 (1.7%) | 0.0 (0.0-0.8) |
| Dyslipidemia | 151 (12.6%) | 524 (43.7%) | 70 (5.8%) | 30 (2.5%) | 161 (13.4%) | 1.7 (0.3-5.5) |
| Diabetes | 123 (10.3%) | 332 (27.7%) | 98 (8.2%) | 19 (1.6%) | 93 (7.8%) | 0.6 (0.1-3.8) |
| Osteoporosis | 99 (8.3%) | 137 (11.4%) | 103 (8.6%) | 31 (2.6%) | 117 (9.8%) | 0.9 (0.2-3.0) |
| Fracture | 51 (4.3%) | 178 (14.8%) | 25 (2.1%) | 17 (1.4%) | 118 (9.8%) | 3.0 (1.0-6.2) |
| Coronary artery di- sease | 45 (3.8%) | 166 (13.8%) | 25 (2.1%) | 5 (0.4%) | 73 (6.1%) | 2.5 (0.7-6.0) |
| Depression | 44 (3.7%) | 145 (12.1%) | 48 (4.0%) | 24 (2.0%) | 86 (7.2%) | 1.3 (0.4-4.6) |
| Bacterial infection | 18 (1.5%) | 84 (7.0%) | 14 (1.2%) | 10 (0.8%) | 83 (6.9%) | 4.2 (1.5-6.5) |
| Ischemic stroke | 14 (1.2%) | 87 (7.3%) | 10 (0.8%) | 8 (0.7%) | 43 (3.6%) | 3.3 (0.9-5.4) |
| Hemorrhagic stroke | 10 (0.8%) | 18 (1.5%) | 2 (0.2%) | 2 (0.2%) | 10 (0.8%) | 2.8 (1.0-3.6) |
| Tuberculosis | 6 (0.5%) | 18 (1.5%) | 6 (0.5%) | 2 (0.2%) | 9 (0.8%) | 1.6 (0.2-4.1) |
| Psychotic disorder | 3 (0.3%) | 9 (0.8%) | 2 (0.2%) | 1 (0.1%) | 7 (0.6%) | 3.2 (0.9-5.4) |
| Aortic disease | 2 (0.2%) | 4 (0.3%) | 2 (0.2%) | 0 (0.0%) | 3 (0.3%) | 1.3 (0.5-5.0) |
| Pulmonary embolism | 1 (0.1%) | 3 (0.3%) | 6 (0.5%) | 0 (0.0%) | 3 (0.3%) | 0.1 (0.0-3.5) |
ªThe data are expressed as n (percentage among total adrenal CS patients).
“The data are expressed as n (percentage among adrenal CS patients without each comorbidity at the time of surgery).
“The median time from adrenalectomy to the diagnosis of each comorbidity was analyzed for postoperative comorbidities. Median time (interquartile range).
Discussion
Current nationwide analysis based on the NHIS claim database demonstrated that the prevalence of adrenal CS was 23.4 per million people, and the age-standardized incidence rate was 1.27 per million person-years in Korea. This is the first epidemiologic study of adrenal CS in Korea. Adrenal CS is the second most common subtype of endogenous CS. In the nationwide analysis in Denmark that was performed on the basis of the na- tional registry database, the incidence of adrenal CS with adrenal adenoma and adrenocortical carcinoma was 0.6 per million person-years and 0.2 per million person- years, respectively (3). In New Zealand, according to the medical records of 4 main endocrinology centers, the incidence of endogenous CS was estimated to be 1.8 per million person-years, and 18.2% of these cases were caused by adrenal adenoma (2). Thus, the incidence was calculated to be 0.3 per million person-years for benign adrenal CS in New Zealand. Another study in western Sweden reported the incidence of adrenal CS to be 0.7 per million person-years for benign CS and 0.2 per million person-years for malignant adrenal CS (13). Our study showed a higher incidence of adrenal CS in Korea than in Western countries. The Korean guideline for adrenal incidentaloma is similar to other guidelines (14, 15). The 1 mg overnight dexamethasone suppression test is re- commended as a screening test for adrenal incidentaloma (16). Additionally, late-night salivary cortisol, midnight
serum cortisol, and 24-hour urinary free cortisol can be used for the diagnosis of endogenous CS. Although the salivary cortisol test is not available in Korea, a combin- ation of other available tests could accurately diagnose CS patients in most cases.
The inconsistency between the incidence of adrenal CS in Korean and Western countries might be due to the dif- ferences in databases used; the national claim database was used in our study, and the national registry database was used in other studies. However, the operational definition of adrenal CS had a balanced rate of false positives and false negatives in our validation analysis. The operational definition of adrenal CS may have a minimal influence, if any. Thus, it is possible that the incidence of adrenal CS is higher in Korea than in Western countries. Although it was conducted almost 20 years ago, a previous Korean na- tional survey reported that the etiology of endogenous CS was Cushing disease for 43.3% of cases and adrenal CS for 41.7% of cases (17). A single-center study in Korea also re- ported that adrenal CS (62.0%, n = 57) was more prevalent than Cushing disease (38.0%, n = 35) (18). Adrenal CS can have a higher incidence in Korea than in Western countries and comprises a higher proportion of endogenous CS cases. Another explanation of the higher incidence of adrenal CS in Korea is that the use of computed tomography (CT) has increased over the recent decades (19). This increased use of CT might have led to an increased detection rate of adrenal adenomas and, eventually, cortisol-producing
A
Hypertension
9.2 (7.9-10.8)
Dyslipidemia
10.6 (8.9-12.6)
Diabetes
9.6 (7.8-11.8)
Osteoporosis
5.0 (3.9-6.3)
Coronary artery disease-
17.5 (11.8-26.0)
Stroke
14.4 (8.9-23.1)
Depression
4.6 (3.6-6.0)
Tuberculosis
0.6 (4.0-1.0)
1.0
5
10
15
20
25
30
B
Odds ratio
Proportion of patients diagnosed with each comorbidities (%)
80
67.6
Preoperative
Postoperative
60
56.3
49.8
40
37.9
22.9
28.2
19.7
26.1
19.1
20
16.5
17.6
10.4
15.8
15.6
8.5
9.8
8.4
5.6
2.3
1.2
2.0
1.4
1.0
0.8
0.5
0.4
0.3
0.8
0
Hypertension
Dyslipidemia
Diabetes
Osteoporosis
Fracture
Coronary artery disease
Depression
Bacterial infection
Ischemic stroke
Hemorrhagic stroke
Tuberculosis
Psychotic disorder
Aortic disease
Pulmonary embolism
adenomas. However, these patients who presented with adrenal incidentalomas would have subclinical CS rather than overt CS. It is necessary to investigate the nationwide data of other CS subtypes in future studies.
Benign adrenal CS patients had a more than 3-fold higher mortality rate than did the standard population. Although the number of deaths was small, mortality oc- curred steadily for 10 years after adrenalectomy in the benign adrenal CS patients. Due to limited access to data, we were unable to present the cause of death for adrenal CS patients. However, we demonstrated that a risk of car- diovascular disease remained even after adrenalectomy in adrenal CS patients, which could contribute to the higher mortality in adrenal CS patients. Clayton et al also showed that the increased cardiovascular mortality of CS patients, even after biochemical remission, can increase the overall
mortality of benign adrenal CS patients (20). Most deaths occurred during the first 5 years after surgery in the malig- nant adrenal CS group. This finding is probably attributed to the prognosis of adrenocortical carcinoma (21).
Various comorbidities were prevalent at the presenta- tion of adrenal CS. The most prevalent comorbidities were hypertension, dyslipidemia, and diabetes. Atherosclerotic cardiovascular diseases were also highly prevalent, and these patients had a 17-fold and 14-fold increased risk of coronary artery disease and stroke compared with the control population. Cardiovascular disease is the leading cause of death in CS patients (22). CS patients have multiple cardiovascular risk factors. Moreover, hypercortisolemia itself can directly cause myocardial hypertrophy (23) and vascular damage (24). The car- diovascular mortality remained elevated, even after the
A
B
Dexamethasone
Others 0.7%
Proportion of glucocorticoid replacement therapy
1.00
4.9%
Median 10.1 months (interquartile range: 4.7 - 20.2)
Methylprednisolone 5.3%
0.75
0.50
Hydrocortisone 22.4%
0.25
0.00
Prednisolone 66.6%
0
12
24
36
48
60
72
84
96
108
120
Number at risk (number of events)
973 (30) 424 (484) 193 (691) 114 (755) 82 (780)
70 (790)
) 55 (802) 43
(810)
38 (813)
27 (817)
17 (820)
0
12
24
36
48
60
72
84
96
108
120
Follow-up duration (month)
biochemical remission of CS (20). The cardiovascular risk factors, including a high body mass index, hyperglycemia, and dyslipidemia, also remained elevated at 5 years after the remission of Cushing disease (7). In our study, for all the comorbidities, considerable proportions of patients were diagnosed with postoperative comorbidities after adrenalectomy, suggesting a significant residual risk for these comorbidities, even after the treatment of adrenal CS. Interestingly, pulmonary embolism was mostly diagnosed within 3 months after adrenalectomy (46.2% of all pul- monary embolism cases were diagnosed within 3 months after adrenalectomy). A previous study also reported that the risk of venous thromboembolism was remarkably high (hazard ratio of 59.9) in the 3-month period after the sur- gical treatment of CS (25). High thrombotic diathesis oc- curs in CS patients (22). Thus, the perioperative period is the period in which CS patients are at the highest risk for thromboembolism. Special consideration should be taken for the early detection and proper management of these comorbidities during the follow-up of adrenal CS patients after adrenalectomy.
After the curative resection of adrenal CS, the remaining adrenal gland becomes atrophied and requires gluco- corticoid replacement therapy until the HPA axis is re- covered. It has been reported that the recovery time of the HPA axis differs according to the etiology of CS (8). The recovery time was longest for adrenal CS cases, followed by Cushing disease and ectopic CS cases, with median re- covery times of 2.5, 1.4, and 0.6 years, respectively (8). However, a more recent study reported a shorter recovery time of 11.4 months for severe CS and 2.8 months for moderate CS, which were defined based on the number of
clinical features of CS presented in the patient (26). In our study, the median duration of postoperative glucocorticoid replacement was 10.1 months. However, approximately 10% of patients received glucocorticoid replacement therapy, even after 5 years. The more worrisome finding was that the main type of glucocorticoid replacement used was prednisolone, not hydrocortisone. Prednisolone has a longer action duration and higher glucocorticoid re- ceptor binding affinity, which results in more suppression of CRH secretion from the hypothalamus and ACTH se- cretion from the pituitary gland. Among the patients who underwent unilateral adrenalectomy, 188 (16.2%) patients did not receive postoperative glucocorticoid replacement therapy. This may be attributed to bilateral macronodular adrenal hyperplasia or subclinical CS in these patients. Furthermore, since we only included postoperative gluco- corticoid that was prescribed regularly, those patients who took postoperative glucocorticoid as needed were not in- cluded in the analysis on postoperative glucocorticoid re- placement. However, considering the extremely high risk of adrenal insufficiency that exists after the treatment of CS (27), these patients could suffer from adrenal insuffi- ciency after adrenalectomy. More attention is warranted for optimal glucocorticoid replacement therapy after adrenalectomy in adrenal CS patients.
The strength of our study is that the NHIS database has nationally representative data that covers nearly 97% of the whole population. Our study investigated long-term prognosis and comorbidities without any pa- tients being lost to follow-up. The treatment of adrenal CS, when the diagnosis is definitive, is surgical resec- tion of adrenal pathology for all adrenal CS patients.
Thus, the operational definition of adrenal CS was val- idated in our institution and had low false-positive and negative rates.
Our study also has several limitations. First, we could not retrieve data on the cause of death. Second, we cat- egorized adrenal CS cases as only either benign or ma- lignant. Among the benign adrenal CS cases, bilateral macronodular or micronodular adrenal hyperplasia was not distinguishable from adrenal adenoma, be- cause it was not indicated by a diagnostic code. Bilateral macronodular adrenal hyperplasia may have different postoperative course from adrenal adenoma. Third, be- cause our study was based on a claim database, the re- sults of biochemical, imaging, and pathologic exams could not be obtained. Thus, patients with subclinical CS and overt CS could not be differentiated in our study. In addition, we could not precisely evaluate the recovery of the HPA axis using biochemical tests. Last, the defin- ition of comorbidities was solely based on the diagnosis codes, which could have been imprecise in some cases. We could not differentiate newly developed postoperative comorbidities from preexisting comorbidities that were only diagnosed postoperatively.
In summary, the prevalence of adrenal CS was 23.4 per million people, and the age-standardized incidence rate of adrenal CS was 1.27 per million person-years in Korea. Approximately 6.0% of adrenal CS cases were caused by adrenocortical carcinoma. Benign adrenal CS patients had a 3-fold higher mortality than did the standard population. Various comorbidities, including metabolic diseases, cardiovascular diseases, osteopor- osis, and mental illness, were already prevalent at the presentation of adrenal CS, and a residual risk remained after the treatment of adrenal CS. The early detection and management of these comorbidities are necessary to optimally follow-up adrenal CS patients. More attention should also be paid to have optimal postoperative gluco- corticoid replacement therapies based on the physiologic regimens.
Acknowledgments
We would like to thank the Committee of Research of the Korean Endocrine Society. This study was supported by the National Health Insurance Service (NHIS-2019-4-007).
Financial Support: None.
Author Contributions: Conception or design of the study: JH Kim, SW Kim
Acquisition, analysis, or interpretation of the data: CH Ahn, MY Park
Drafting or revision of the manuscript: CH Ahn, JH Kim
Final approval of the manuscript: CH Ahn, MY Park, JH Kim, and SW Kim
Additional Information
Correspondence and Reprint Requests: Jung Hee Kim, MD, PHD, Department of Internal Medicine, Seoul National University Hos- pital, Seoul National University College of Medicine, 101 Dae-hak ro, Jongno gu, Seoul, 03080, Korea. E-mail: jhkxingfu@gmail.com
Disclosure Summary: The authors have nothing to disclose.
Data Availability: This study used National Health Information Database (NHIS-2019-4-007) made by National Health Insurance Ser- vice (NHIS). Restrictions apply to all the availability of data generated or analyzed during this study to preserve patient confidentiality. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.
References
1. Newell-Price J, Bertagna X, Grossman AB, Nieman LK. Cushing’s syndrome. Lancet. 2006;367(9522):1605-1617.
2. Bolland MJ, Holdaway IM, Berkeley JE, et al. Mortality and morbidity in Cushing’s syndrome in New Zealand. Clin Endocrinol (Oxf). 2011;75(4):436-442.
3. Lindholm J, Juul S, Jørgensen JO, et al. Incidence and late prog- nosis of Cushing’s syndrome: a population-based study. J Clin Endocrinol Metab. 2001;86(1):117-123.
4. Ntali G, Asimakopoulou A, Siamatras T, et al. Mortality in Cushing’s syndrome: systematic analysis of a large series with prolonged follow-up. Eur J Endocrinol. 2013;169(5):715-723.
5. Pikkarainen L, Sane T, Reunanen A. The survival and well-being of patients treated for Cushing’s syndrome. J Intern Med. 1999;245(5):463-468.
6. Yaneva M, Kalinov K, Zacharieva S. Mortality in Cushing’s syn- drome: data from 386 patients from a single tertiary referral center. Eur J Endocrinol. 2013;169(5):621-627.
7. Colao A, Pivonello R, Spiezia S, et al. Persistence of in- creased cardiovascular risk in patients with Cushing’s disease after five years of successful cure. J Clin Endocrinol Metab. 1999;84(8):2664-2672.
8. 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(4):1300-1308.
9. Välimäki M, Pelkonen R, Porkka L, Sivula A, Kahri A. Long- term results of adrenal surgery in patients with Cushing’s syn- drome due to adrenocortical adenoma. Clin Endocrinol (Oxf). 1984;20(2):229-236.
10. Doherty GM, Nieman LK, Cutler GB Jr, Chrousos GP, Norton JA. Time to recovery of the hypothalamic-pituitary- adrenal axis after curative resection of adrenal tumors in patients with Cushing’s syndrome. Surgery. 1990;108(6):1085-1090.
11. Kweon S, Kim Y, Jang MJ, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol. 2014;43(1):69-77.
12. Ahmad OB, Boschi-Pinto C, Lopez AD, Murray CJ, Lozano R, Inoue M. Age standardization of rates: a new WHO standard. Geneva: World Health Organization, 2001. https://www.who. int/healthinfo/paper31.pdf. Accessed April 15, 2020.
13. Wengander S, Trimpou P, Papakokkinou E, Ragnarsson O. The incidence of endogenous Cushing’s syndrome in the modern era. Clin Endocrinol (Oxf). 2019;91(2):263-270.
14. Fassnacht M, Arlt W, Bancos I, et al. Management of ad- renal incidentalomas: European Society of Endocrinology Clinical Practice Guideline in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol. 2016;175(2):G1-G34.
15. Nieman LK. Approach to the patient with an adrenal incidentaloma. J Clin Endocrinol Metab. 2010;95(9):4106-4113.
16. Lee JM, Kim MK, Ko SH, et al .; Korean Endocrine Society, Committee for Clinical Practice Guidelines. Clinical guidelines for the management of adrenal incidentaloma. Endocrinol Metab (Seoul). 2017;32(2):200-218.
17. The Survey Committee for Endocrine Disease in Korea. The incidence and clinical characteristics of Cushing’s syndrome in Korea. J Korean Soc Endocrinol. 2000;15(1):31-45.
18. Hong AR, Kim JH, Hong ES, et al. Limited diagnostic utility of plasma adrenocorticotropic hormone for differentiation between adrenal Cushing syndrome and Cushing disease. Endocrinol Metab (Seoul). 2015;30(3):297-304.
19. Oh HY, Kim EY, Cho J, et al. Trends of CT use in the adult emergency department in a tertiary academic hospital of Korea during 2001-2010. Korean J Radiol. 2012;13(5):536-540.
20. Clayton RN, Jones PW, Reulen RC, et al. Mortality in patients with Cushing’s disease more than 10 years after remission: a multicentre, multinational, retrospective cohort study. Lancet Diabetes Endocrinol. 2016;4(7):569-576.
21. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2):282-326.
22. Pivonello R, Isidori AM, De Martino MC, Newell-Price J, Biller BM, Colao A. Complications of Cushing’s syn- drome: state of the art. Lancet Diabetes Endocrinol. 2016;4(7):611-629.
23. Muiesan ML, Lupia M, Salvetti M, et al. Left ventricular struc- tural and functional characteristics in Cushing’s syndrome. J Am Coll Cardiol. 2003;41(12):2275-2279.
24. Albiger N, Testa RM, Almoto B, et al. Patients with Cushing’s syn- drome have increased intimal media thickness at different vascular levels: comparison with a population matched for similar cardio- vascular risk factors. Horm Metab Res. 2006;38(6):405-410.
25. Dekkers OM, Horváth-Puhó E, Jørgensen JO, et al. Multisystem morbidity and mortality in Cushing’s syndrome: a cohort study. J Clin Endocrinol Metab. 2013;98(6):2277-2284.
26. Hurtado MD, Cortes T, Natt N, Young WF Jr, Bancos I. Extensive clinical experience: hypothalamic-pituitary-adrenal axis recovery after adrenalectomy for corticotropin-independent cortisol excess. Clin Endocrinol (Oxf). 2018;89(6):721-733.
27. Di Dalmazi G, Berr CM, Fassnacht M, Beuschlein F, Reincke M. Adrenal function after adrenalectomy for sub- clinical hypercortisolism and Cushing’s syndrome: a sys- tematic review of the literature. J Clin Endocrinol Metab. 2014;99(8):2637-2645.
Downloaded from https://academic.oup.com/jcem/article/106/3/e1362/5930839 by National Library of Medicine user on 03 April 2026