Atkinson, Elizabeth J.

Abstract Disclosure: R. Sandooja: None. A. Kittithaworn: None. P. Dogra: None. D. Gruppen: None. E. Atkinson: None. S. Achenbach: None. K. Yu: None. M. Connelly: None. V. Simha: None. R.P. Dullaart: None. I. Bancos: None. INTRODUCTION: Based on the spectrum of cortisol excess, adrenal adenomas are classified as adenomas with Cushing syndrome (CS), adenomas with mild autonomous cortisol secretion (MACS), and non-functioning adenomas (NFA). Patients with adrenal adenomas demonstrate a high prevalence of cardiovascular morbidity and mortality proportional to the degree of hypercortisolism. Evidence on whether dyslipidemia is a contributing cardiovascular risk factor in patients with NFA and MACS is discrepant. We aimed to characterize (apo) lipoprotein abnormalities across the spectrum of cortisol excess in patients with adenomas as compared to referent subjects. METHODS: Study Design: Single center cross sectional study between 01/2015 - 05/2021.Patients: Adults with NFA, MACS, or CS were consecutively enrolled. MACS was diagnosed based on post-dexamethasone cortisol >1.8 mcg/dL. Referent subjects were recruited from local population of patients with available abdominal imaging and no adrenal disorders. All subjects provided fasting ethylenediaminetetraacetic acid (EDTA) venous samples. Measurements: Nuclear magnetic resonance (NMR) spectroscopy measurements included total cholesterol (TC), low-density lipoprotein (LDL) and LDL particle concentrations, high-density lipoprotein (HDL) subclasses, apolipoprotein B (apo B), triglycerides (TG) and TG-rich lipoprotein concentrations. RESULTS: Participants included 101 patients with CS (median age 44 years, IQR 31-58, 85% women), 158 patients with MACS (median age 62 years, IQR 51-70, 62% women), 166 patients with NFA (median age 60 years, IQR 53--66, 63% women) and 136 referent subjects (median age 68 years, IQR 57-77, 52% women). Patients had higher BMI (32 vs 27 kg/m2), higher prevalence of hypertension (72% vs 46%) and diabetes mellitus (27% vs 9%) as compared to reference subjects. Statin therapy was used in 32% of patients and referent subjects. After adjusting for age, sex, BMI, smoking, diabetes, hypertension, and statin therapy, as compared to referent subjects, we found inconsistent increase in TC, LDL cholesterol and TG across CS, MACS and NFA and no consistent differences in LDL size, HDL variables, TG and TG-rich lipoproteins. However, when compared to referent subjects, LDL particle concentrations- CS(OR 1.78, 95% CI 1.33-2.39), MACS (OR 1.31, 95% CI 1.03-1.67), and NFA (OR 1.43, 95% CI 1.12-1.82) and Apo B levels were higher in all patient categories: CS (OR 2.26, 95% CI 1.57-3.24), MACS (OR 1.36 95% CI 1.00-1.85 ), and NFA (OR 1.54 95% CI 1.13-2.09). These differences were particularly evident in patients not on statin therapy. CONCLUSIONS: We found that LDL particle and Apo B concentrations were elevated in all patient groups, including those with MACS and NFA. In comparison to conventional lipid measurements, LDL particle and Apo B may serve as a better biomarker of the future cardiovascular risk in patients with adrenal adenomas. Presentation: Friday, June 16, 2023

Abstract Disclosure: J. Saini: None. R. Nathani: None. C. Zhang: None. M. Suresh: None. K. Thangamuthu: None. S. Singh: None. V. Fell: None. E.J. Atkinson: None. S. Achenbach: None. N.H. Stricker: None. I. Bancos: None. Background: Patients with mild autonomous cortisol secretion (MACS) have impaired cognition compared to patients with non-functioning adrenal adenomas. However, the impact of adrenalectomy on cognitive function in patients with MACS is unclear. Objective: To determine the impact of adrenalectomy on cognitive function in patients with MACS. Methods: Single-center prospective longitudinal study of adults with MACS. MACS was defined as serum cortisol >1.8 mcg/dL after the 1 mg dexamethasone suppression test (DST) in patients with adrenal adenomas without any overt features of hypercortisolism. Patients underwent cognitive evaluation using the standardized National Institute of Health (NIH) Toolbox Cognition Battery (7 tests of attention, episodic and working memory, language, executive function, cognitive flexibility, and processing speed) prior to adrenalectomy and at follow-up at least 10 months post-surgery. Fully corrected T-scores for age, sex, education, and race were compared using a paired t-test before and after adrenalectomy. A T-score of 50 is average performance (SD of 10). Total cognition is composite of fluid cognition (executive function, episodic memory, cognitive flexibility, working memory and processing speed) and crystallized cognition (language and vocabulary). Results: A total of 27 patients with MACS (median age 59.7 years, IQR 51.4-66.9, 17, 63% women) underwent cognitive evaluation prior to and at a median of 1.5 years (IQR, 1.1- 2.0) after adrenalectomy. Before surgery, median post-DST cortisol was 3.8 mcg/dL (IQR, 2.7-6.4) mcg/dL. Of the 27 patients, 21 (78%) patients developed post-operative adrenal insufficiency with 6 (21%) patients still requiring still requiring glucocorticoid replacement at the time of the last follow-up. At baseline, the mean total composite T-score was 50.7 (9.3), with fluid composite T-score of 49.9 (9.4) and crystallized composite T-score of 51.3 (8.7). Following adrenalectomy, all three composite T-scores improved compared to baseline. The crystallized composite T-score (composed of language and vocabulary) increased by a mean of 3.7 (5.5), P=0.001, the fluid composite T-score (composed of episodic, executive, working memory, cognitive flexibility, and processing speed) increased by a mean of 3.1 (8.3), P=0.068, and the total composite T-score improved by a mean of 3.9 (6.2), P=0.004. Conclusion: We showed that adrenalectomy in patients with MACS is associated with improvement in cognition. However, as test results may improve with practice (i.e., improvement due to taking the test the second time), to determine whether the improvements observed represent a reliable change in cognition, the next step is to evaluate for cognition score changes in referent subjects. Presentation: Friday, June 16, 2023

Abstract Disclosure: C.D. Zhang: None. S. Singh: None. J. Saini: None. M. Suresh: None. K. Thangamuthu: None. R. Nathani: None. A.L. Ebbehøj: None. V. Fell: None. S.J. Achenbach: None. E.J. Atkinson: None. S. Khosla: None. I. Bancos: None. Background: Patients with mild autonomous cortisol secretion (MACS) have higher rates of fractures compared to patients with non-functioning adrenal nodules, but bone mass is typically only modestly reduced. We hypothesized that chronic exposure to mild levels of cortisol excess may lead to a deterioration in bone quality. Objective: To characterize the effect of MACS on bone metabolism and bone microarchitecture, as measured by high-resolution peripheral quantitative computed tomography (HR-pQCT). Methods: In this cross-sectional single-center study, patients with MACS were matched by age, sex, and gonadal status to referent subjects without adrenal nodules. MACS was defined as serum cortisol concentration >1.8 mcg/dL after the 1 mg overnight dexamethasone suppression test (1-mg DST), in the absence of a classic Cushingoid phenotype. Those with other known secondary causes of osteoporosis and/or on pharmacologic treatment for low bone mass were excluded. All study participants underwent a bone health interview, blood draw for bone turnover markers, dual-energy x-ray absorptiometry (DXA) scan with trabecular bone score (TBS), and HR-pQCT of the non-dominant radius and tibia. Participants with a recent fracture within 12 months were excluded from the analysis of bone turnover markers. The Kruskal-Wallis test was used to compare measurements between groups. Results: A total of 82 patients with MACS and 82 referent subjects were prospectively enrolled between 1/2019 and 7/2022. The median age of study participants at the time of enrolment was 57 years (interquartile range [IQR], 47-65), and 57 (70%) were women. In the MACS cohort, the median cortisol following 1-mg DST was 3.2 mcg/dL (IQR, 2.3-6.5) with median ACTH of 8.7 pg/mL (IQR, 6.0-14.0) and median DHEA-S of 42.0 mcg/dL (IQR 23.0-68.0). While no differences in bone mineral density (BMD) were observed at any site (lumbar spine, total hip, femoral neck, or distal radius) between the two groups, patients with MACS had decreased TBS (1.374 vs 1.467, P <0.001) and lower osteocalcin concentrations (16.9 vs. 19.7 ng/mL, P = 0.005) than referent subjects without adrenal nodules. On HR-pQCT at the tibia, patients with MACS had significantly reduced trabecular volumetric BMD (151.6 vs 167.4 mg hydroxyapatite/cm3, P = 0.002), trabecular bone volume/tissue volume ratio (0.219 vs 0.237, P = 0.011), and number of trabeculae (1.428 vs 1.506 1/mm, P = 0.045), while trabecular separation (0.707 vs 0.656, P = 0.023) was increased compared to referent subjects. Conclusions: Patients with MACS have altered bone metabolism and microstructural deterioration that is not evident on routine DXA BMD testing. These changes may mediate their increased fracture risk and could serve as biomarkers of cortisol-induced changes in the bone. Presentation: Thursday, June 15, 2023

Abstract Disclosure: J. Saini: None. R. Nathani: None. S. Singh: None. V. Fell: None. E. Atkinson: None. S. Achenbach: None. S. Khosla: None. I. Bancos: None. Background: Previous studies reported that both cortisol secretion and bone metabolism biomarkers follow a circadian pattern of secretion. Patients with mild autonomous cortisol secretion (MACS) demonstrate an increased risk of fractures. As most patients with MACS do not have increased cortisol production, we hypothesized that abnormal circadian cortisol secretion is associated with impaired bone metabolism. Objective: To compare the circadian secretory pattern of bone metabolism biomarkers in relation to cortisol circadian secretion in patients with MACS versus referent subjects. Methods: We conducted a pilot cross-sectional study of patients with MACS and referent subjects. MACS was diagnosed in patients with adrenal adenomas based on an abnormal 1 mg dexamethasone suppression test (cortisol >1.8 mcg/dL). Referent subjects with available abdominal imaging within 5 years and adrenal disorder were recruited from the community. Participants were admitted to the clinical research unit for blood sampling every 2 hours over 24h period. Measurements included total and free cortisol, calcium, parathyroid hormone (PTH), osteocalcin, amino-terminal propeptide type 1 collagen (PINP), and carboxy-terminal telopeptide of type 1 collagen (CTX). Subjects with disorders or medication affecting bone metabolism were excluded. Time periods of blood draw were categorized into the morning (6 to 12 hours), afternoon (14-18 hours) and night (20-2 hours). Delta (morning measurements/afternoon measurements) was calculated. Results: Subjects included 12 patients with MACS (median age 57, IQR 48-67, 83% women) and 10 referent subjects (median age 57, IQR 50-63, 80% women). The 24-hour area under the curve for total (10342 vs 9090 mcg/dL, p=0.468) and free cortisol (437 vs 329 mcg/dL, p=0.086) was no different between patients with MACS and referent subjects. However, the mean night total (6.7 vs 4.3 mcg/dL, p=0.056) and free cortisol (0.2 vs 0.1 mcg/dL, p=0.035) was higher in MACS vs referent subjects. The area under the curve for calcium, phosphorus, CTX, osteocalcin and PINP were similar between patients and referent subjects, however, the area under the curve for PTH was higher in patients vs referent subjects (mean 64179 vs 42602 pg/mL, p=0.021). Both delta total cortisol (mean 8.6 vs 5.5, p=0.01) and delta PTH (mean -1.5 vs 4.2, p=0.051) was higher in referent subjects vs patients with MACS. Conclusion: When compared to referent subjects, patients with MACS have similar 24-hour cortisol production but demonstrate an increase in night-time cortisol secretion and a blunting of diurnal cortisol (lower morning/afternoon cortisol). Patients with MACS have higher PTH production overall, as well as an abnormal diurnal PTH secretion. Noted abnormalities in PTH may contribute to the MACS-associated impairment in bone health. Presentation: Thursday, June 15, 2023

Abstract Disclosure: J. Saini: None. R. Nathani: None. S. Singh: None. K. Thangamuthu: None. M. Suresh: None. C. Zhang: None. A.L. Ebbehøj: None. M. Thomas: None. V. Fell: None. E. Atkinson: None. S. Achenbach: None. N. Stricker: None. I. Bancos: None. Introduction: While patients with overt Cushing syndrome are reported to have impaired cognition, the impact of milder, long-standing hypercortisolism on cognition is incompletely characterized. Objective: To determine the impact of MACS on cognition as compared to age, sex, and education-matched referent subjects. Methods: We conducted a single-center cross-sectional study of adults with MACS and referent subjects. MACS was defined as serum cortisol>1.8 mcg/dL after 1 mg dexamethasone suppression test (DST) in patients with adrenal incidentalomas. Referent subjects were included if they had abdominal imaging and had no adrenal disorders. Exclusion criteria were malignancy and exogenous glucocorticoid use within 1 year. All participants underwent a cognitive evaluation using the NIH Toolbox Cognition Battery between 1/2019 to 7/2022. NIH Toolbox assesses 8 cognitive domains using 7 tests and yields 3 composite scores: fluid cognition (executive function, episodic memory, cognitive flexibility, working memory, and processing speed), crystallized cognition (language and vocabulary), and total composite cognition (composite of all). Results are given in fully corrected T scores (adjusted for age, sex, race, and education) with an average of 50 and a standard deviation (SD) of 10. A subgroup analysis was performed based on DST cortisol. The Kruskal-Wallis test was used to compare scores between groups. Results: A total of 94 patients with MACS (median age 57.2 years, IQR 47.8-66.5, 70% women) and 94 referent subjects (median age 58 years, IQR 49.1-66.4, 70% women) were included. The median post-DST cortisol in patients with MACS was 3.3 mcg/dL (IQR, 2.4-6.2). The mean (SD) total composite T-score was lower in patients with MACS vs referent subjects (51 (9.1) vs 56.3 (9.3), P<0.001). Patients with MACS demonstrated lower mean fluid composite T-score when compared to referent subjects (49.5 (10.4) vs 56.9 (9.7), P<0.001). However, mean crystallized composite T-scores were similar in patients with MACS and referent subjects (52.3 (8.1) vs 53.9 (9.6), P=0.342). A subgroup analysis of patients with post-DST cortisol <3 mcg/dL (n=40) vs patients with post-DST cortisol >3 mcg/dL (n=54) did not show any differences in the total, fluid, or crystalized composite scores. Conclusion: Patients with MACS demonstrate lower cognitive performances than a referent group suggesting MACS has a subtle impact on cognitive functioning. The pattern of cognitive impairment in patients with MACS is similar to that seen with normal cognitive aging: impaired fluid cognitions and unaffected crystallized cognition). Cognitive performance did not differ based on the degree of MACS suggesting that in the absence of overt features of hypercortisolism, the degree of MACS is less important than the presence of MACS. Presentation: Thursday, June 15, 2023

Abstract Disclosure: L. Rahimi: None. A. Kittithaworn: None. R. Gregg Garcia: None. J. Saini: None. P. Dogra: None. E. Atkinson: None. S. Achenbach: None. A. Kattah: None. I. Bancos: None. Objective: Patients with mild autonomous cortisol secretion (MACS) have been reported to have a higher prevalence of chronic kidney disease (CKD), possibly due to the direct effect of hypercortisolism on kidney function and indirect effect of MACS-associated cardiovascular comorbidities, such as diabetes, hypertension, obesity, and dyslipidemia. Limited studies suggest improvement of kidney function following adrenalectomy in patients with MACS. In this study, we aimed to assess the factors associated and the degree and timing of change in kidney function post adrenalectomy in patients with MACS or nonfunctioning adrenal adenoma (NFA). Methods: We conducted a retrospective single-center study and included adult patients with MACS or NFA diagnosed based on a post-dexamethasone cortisol cutoff of 1.8 mcg/dL, who underwent adrenalectomy between 1999-2022. The change in estimated glomerular filtration rate (eGFR) before and up to 3.5 years after adrenalectomy was defined as the primary outcome. Demographics and comorbidities were included in univariable and multivariable analyses. Results: We included 155 (76.0%) patients with MACS and 49 (24.0%) patients with NFA treated with adrenalectomy. The proportion of women was high in both MACS and NFA (71.6% vs 69.4%, p=0.77), however, patients with MACS were older (median age 59.2 vs 54.6 years in NFA, p=0.05). The prevalence of hypertension was higher in MACS vs NFA (71.0% vs 53.1%, p=0.02), while the prevalence of diabetes (25.2% vs 28.6%, p=0.89), dyslipidemia (47.7% vs 46.9%, p=0.92), smoking (74.2% vs 68.9%, p=0.49) and obesity (median BMI 30.4 vs 31.8 kg/m2, p=0.35) was similar between the two groups. Prior to adrenalectomy, patients with MACS had lower eGFR (median 84.4 vs 89.2 ml/min/1.73m2, p=0.09) and a higher prevalence of CKD stage ≥ 3 (16.1% vs 8.2%, p= 0.16). In patients with MACS, after an initial decline of eGFR within 6 months post adrenalectomy (mean decrease 5.3 ml/min/1.73m2), eGFR increased by a mean of 4.9 points at 18-30 months, and by a mean of 8.1 points at 30-42 months post-adrenalectomy (p<0.01). Patients with NFA followed a similar eGFR change trend (p=NS). When comparing post-adrenalectomy delta eGFR in patients with MACS vs NFA, no significant differences were found. After adjusting for age, sex, pre-adrenalectomy eGFR, patient group, hypertension, antihypertensive therapy, and diabetes, there were still significant changes in eGFR over time. In a multivariable analysis, younger age (est -031, p<0.001) and lower pre-adrenalectomy eGFR (est-0.27, p=0.001) were associated with kidney function improvement post-adrenalectomy. Conclusion: Patients with MACS and NFA demonstrate improvement of eGFR over the 3.5 years of follow up post-adrenalectomy. The factors that are associated with improvement of eGFR are lower eGFR before surgery, younger age at time of adrenalectomy and longer follow up post adrenalectomy. Presentation: Friday, June 16, 2023

Abstract Objective Adrenal adenomas are commonly encountered in clinical practice. To date, population-based data on their impact on cognition, mental health, and sleep are lacking. We aimed to study possible associations between adrenal adenomas and dementia, psychiatric or sleep disorders. Design Population-based cohort study, Olmsted County, MN, 1995-2017. Methods Patients with adrenal adenoma and absent overt hormone excess were age- and sex-matched 1:1 to a referent person without adrenal adenoma. Outcomes were baseline and incident diagnoses of dementia, psychiatric or sleep disorders, assessed using ICD codes. Results Of 1004 patients with adrenal adenomas, 582 (58%) were women, and median age at diagnosis was 63 years. At baseline, and after adjusting for age, sex, education, BMI, and tobacco use, patients with adenoma had higher odds of depression (adjusted odds ratio, aOR: 1.3, 95% CI, 1.1-1.6), anxiety (aOR: 1.4, 95% CI, 1.1-1.8), and substance abuse (aOR: 2.4, 95% CI, 1.7-3.4) compared to referents. During a median follow-up of 6.8 years, and after adjusting for age, sex, socioeconomic status, BMI, tobacco, and substance abuse, patients demonstrated a higher risk of psychiatric and sleep disorders [adjusted hazard ratio (95% CI)]: depression [1.7 (1.3-2.2)], anxiety [1.4, CI (1.1-1.7)], insomnia [1.4 (1.0-1.9)], sleep-related breathing disorders [1.5 (1.1-1.9)], hypersomnias [2.1 (1.0-4.2)], parasomnias [2.1 (1.0-4.2)], and sleep-related movement disorders [1.5 (1.0-2.1)], but not dementia. Conclusions Patients with adenomas demonstrate a higher incidence of psychiatric and sleep disorders, possibly due to the underlying subtle increase in cortisol secretion.

Abstract Objective Frailty, characterized by multi-system decline, increases vulnerability to adverse health outcomes and can be measured using Frailty Index (FI). We aimed to assess the prevalence of frailty in patients with adrenal disorders (based on hormonal sub-type) and examine association between FI and performance-based measures of physical function. Design Multi-centre, cross-sectional study (March 2019-August 2022). Methods Adult patients with adrenal disorders (non-functioning adrenal adenomas [NFA], mild autonomous cortisol secretion [MACS], Cushing syndrome [CS], primary aldosteronism [PA]) and referent subjects without adrenal disorders completed a questionnaire encompassing 47 health variables (comorbidities, symptoms, daily living activities). FI was calculated as the average score of all variables and frailty defined as FI ≥ 0.25. Physical function was assessed with hand grip, timed up-and-go test, chair rising test, 6-minute walk test, and gait speed. Results Compared to referent subjects (n = 89), patients with adrenal disorders (n = 520) showed increased age, sex, and body mass index-adjusted prevalence of frailty (CS [odds ratio—OR 19.2, 95% confidence interval—CI 6.7-70], MACS [OR 12.5, 95% CI 4.8-42.9], PA [OR 8.4, 95% CI 2.9-30.4], NFA [OR 4.5, 95% CI 1.7-15.9]). Prevalence of frailty was similar to referent subjects when post-dexamethasone cortisol was <28 nmol/L and was higher when post-dexamethasone cortisol was 28-50 nmol/L (OR 4.6, 95% CI 1.7-16.5). FI correlated with all measures of physical function (P < .001). Conclusions Whilst frailty prevalence was highest in patients with adrenocortical hormone excess, even patients with NFA demonstrated an increased prevalence compared to the referent population. Future longitudinal studies are needed to evaluate the impact of various management strategies on frailty.

Abstract Objective Glucocorticoid withdrawal syndrome (GWS) is a scarcely studied phenomenon that complicates the recovery following surgical remission of hypercortisolism. We aimed to characterize the presence and trajectory of glucocorticoid withdrawal symptoms in the postoperative period and to determine presurgical predictors of GWS severity. Design Longitudinal observational study. Methods Glucocorticoid withdrawal symptoms were prospectively evaluated weekly for the first 12 weeks following surgical remission of hypercortisolism. Quality of life (CushingQoL and Short-Form-36) and muscle function (hand grip strength and sit-to-stand test) were assessed at the baseline and at 12 weeks after surgery. Results Prevalent symptoms were myalgias and arthralgias (50%), fatigue (45%), weakness (34%), sleep disturbance (29%), and mood changes (19%). Most symptoms persisted, while myalgias, arthralgias, and weakness worsened during weeks 5-12 postoperatively. At 12 weeks after surgery, normative hand grip strength was weaker than at baseline (mean Z-score delta −0.37, P = .009), while normative sit-to-stand test performance improved (mean Z-score delta 0.50, P = .013). Short-Form-36 Physical Component Summary score worsened (mean delta −2.6, P = .015), but CushingQoL score improved (mean delta 7.8, P < .001) at 12 weeks compared to baseline. Cushing syndrome (CS) clinical severity was predictive of postoperative GWS symptomology. Conclusion Glucocorticoid withdrawal symptoms are prevalent and persistent following surgical remission of hypercortisolism with baseline CS clinical severity predictive of postoperative GWS symptom burden. Differential changes observed in muscle function and quality of life in the early postoperative period may reflect the competing influences of GWS and recovery from hypercortisolism.