Cushing's Disease in Older Patients

Presentation and Outcome

Nidan Qiao; Brooke Swearingen; Nicholas A. Tritos


Clin Endocrinol. 2018;89(4):444-453. 

In This Article

Subjects and Methods

Patient Population

We reviewed the records of all patients with CD admitted to MGH for surgery between 2000 and 2017. The mean age overall was 41.3 years and the age delimiting the oldest 10% of this surgical population was 59.8 years. Accordingly, we used 60 years as a cutoff point for the definition of "older patients" in this study. We also included two control groups: control group I included randomly selected patients in the remaining CD cohort matched by surgery year in a 2:1 assignment ratio, and control group II included randomly selected patients more than 60 years old from a cohort of patients with nonfunctional pituitary adenoma (NFPA) who also underwent pituitary surgery in our institution during the same time period (in a 1:1 assignment ratio). The study was approved by the Partners/Massachusetts General Hospital Human Research Committee (Institutional Review Board) and was conducted under the ethical standards of the Declaration of Helsinki.

Criteria for Diagnosis of CD

The diagnosis of Cushing's syndrome was based on at least two of the following criteria: (a) elevated 24-hour urinary free cortisol (UFC) above the upper limit of the reference range; (b) elevated late-night salivary free cortisol (LNSC), and/or (c) lack of serum cortisol suppression after 1 mg dexamethasone in the absence of other exogenous glucocorticoid use. ACTH-dependent hypercortisolism of pituitary origin (ie, Cushing's disease) was confirmed by normal or elevated preoperative plasma ACTH concentrations and at least one of the following: (a) immunopathology confirming the diagnosis; (b) clinical and endocrine remission after pituitary surgery; (c) preoperative bilateral inferior petrosal sinus sampling (IPSS) predicting a pituitary source.[13,14]

Preoperative Data Collection

We identified baseline characteristics based upon retrospective record review, including age at diagnosis, date of pituitary surgery, gender, height, weight, symptoms and signs (weight gain, central adiposity, muscle wasting, skin abnormalities, alopecia, hirsutism, fatigue, multiple fractures, hypokalemia or incidental presentation) and comorbidities (hypertension, diabetes mellitus, cardiovascular disease, mood disorder and venous thromboembolism) in all three groups. Body mass index, hypertension, diabetes mellitus and hypokalemia were quantified. We adopted a standardized method to retrieve data (Table S1, found in the Supporting Information). Cushing symptoms were recorded if certain words appear in physicians' or surgeons' notes. "Hypopituitarism" was defined based on the absence of one or more pituitary hormones. "Apoplexy" was defined as severe, acute headache with evidence of intra-tumour haemorrhage on MRI. "Vision loss" was defined as decreased visual acuity or visual field defect. "Multiple fractures" was defined as history of more than one fractures. "Hypokalemia" was defined as presentation of low serum potassium level. Comorbidities (hypertension, diabetes mellitus, cardiovascular disease, venous thromboembolism and mood disorder) were recorded if the diagnosis were present in either a preoperation note or anaesthesia preprocedure evaluation.

In addition to preoperative cortisol and ACTH data, we recorded information on free thyroxine (T4), thyroid-stimulating hormone (TSH) and insulin-like growth factor 1 (IGF-1) levels in all patients, and testosterone levels in men. The methodology used for these assays has changed over time, so we reported endocrine data as "fold elevation" above the upper limit of normal in this study.

In patients with CD, we recorded tumour size (either no visible tumour, microadenoma or macroadenoma) and the presence of cavernous sinus invasion on magnetic resonance imaging (MRI). We also analysed data on central to peripheral ACTH ratios at baseline (positive if >2) and after corticotropin-releasing hormone (CRH) stimulation (positive if >3) during bilateral IPSS. Biochemical data, including preoperative ACTH, preoperative 24-hour UFC, and preoperative LNSC, were reported as fold times above the upper limit of the normal range.

Postoperative Data Collection

In those cases where tumour was found on frozen section, we performed a selective resection. If tumour was not identified intraoperatively, we usually did hemi-hypophysectomy. We recorded data from pathology reports documenting the presence of ACTH-immunostaining in pituitary adenomas and the Ki-67 proliferation index. Any surgery-related adverse events were also reviewed, which included transient or permanent diabetes insipidus (DI), syndrome of inappropriate antidiuretic hormone secretion (SIADH), cerebrospinal fluid (CSF) leakage or postoperative haemorrhage.

We used intermittent pneumatic compression devices beginning with the induction of anaesthesia and continuing until the patient was mobilized, usually the night of surgery. The average length of stay was 2 days. We did not routinely use prophylactic anticoagulation.

All patients received dexamethasone 0.5 mg daily beginning on postoperative day one, after obtaining a fasting cortisol level. Repeat testing of fasting morning cortisol on dexamethasone was usually obtained on days 1, 4, 7, 11 as an outpatient (morning cortisol is obtained 24 hour after the last dexamethasone dose). Postoperative serum cortisol, UFC and LNSC were classified as consistent with: (a) remission, which was defined as the presence of low early morning serum cortisol concentrations (<138 nmol/L), or low 24-hour UFC levels (<55 nmol/24 h), or (b) persistent disease, defined as either normal or elevated postoperative serum cortisol levels or elevated 24-hour UFC. Recurrence was defined as representation with an elevated 24-hour UFC, inability to suppress serum cortisol after low-dose dexamethasone testing, elevated serum cortisol concentration, or elevated LNSC with clinical symptoms after initial remission.[15–17] Data on adjunctive treatments, including radiation therapy to the sella, medical therapy and bilateral adrenalectomy were also recorded in patients with persistent/recurrent disease.

Statistical Analysis

Statistical significance of differences in continuous variables among the three groups was determined using ANOVA tests. Other continuous variables were analysed using nonparametric tests (Mann-Whitney test). Categorical variables were analysed using chi-square test or Fisher exact test as appropriate. A Kaplan-Meier analysis was performed to investigate the recurrence of CD among patients in remission. The comparisons yielding P < 0.05 were considered significant. Posthoc, we estimated that the power of this study was 0.90 in detecting the present difference in BMI between the study groups of the present size with an alpha of 0.05. Values are expressed as mean ± SD. All statistical analyses were completed by R software version 3.4.2.