TABLE 1.
Results
| REFERENCE | CHARACTERISTICS | MEASURES | FINDINGS | STRENGTHS AND LIMITATIONS |
|---|---|---|---|---|
| Kelly4 | Sample size: 209; age: 39.7 (mean); gender: 47 M, 162 F; psychiatric interview of patient | Hamilton Rating Scale for Depression, Present State Examination | In total, 25 patients experienced psychiatric symptoms, mostly depressive, prior to any signs of CS manifesting. Overall, 120 patients were experiencing or had experienced psychiatric symptoms, mostly depressive, at diagnosis of CS. Diagnoses include anxiety in 26 patients, psychotic illness in 16 patients, mania or hypomania in 6 patients, and confusion in 3 patients. | Strengths: large sample size, four institutions involved; limitation: retrospective study |
| Haskett8 | Sample size: 30; age: 36.7±10.7 (mean±SD); gender: 6 M, 24 F; psychiatric interview of patient and relatives | SADS-L | In total, 25 patients experienced an episode of affective disorder during the course of their CS. Twenty patients experienced depression, 8 patients experienced mania or hypomania, and 2 patients experienced psychotic depression. There was 1 suicide attempt. Schizophrenic illness was not detected. Patients often tried to minimize or hide psychiatric symptoms. | Strength: long term follow-up, multiple sources of information |
| Hudson et al9 | Sample size: 16; age: 44.4±14.1 (mean±SD); gender: 3 M, 16 F; psychiatric interview of patient | Semi-structured diagnostic interview to determine lifetime psychiatric history according to DSM-III criteria | In total, 13 out of 16 patients with CS were diagnosed with major affective disorder in their lifetime according to DSM-III. Relative to patients with major depression, the rate of familial major affective disorder among these patients was found to be significantly lower. | Limitation: sample drawn from a single institution, use of family history method rather than family interview method for ascertaining psychiatric illness in relatives |
| Bolland et al10 | Sample size: 253; age: 39±15 (mean±SD); gender: 81 M, 192 F; retrospective survey | Clinical evaluation | After treatment of CS, the prevalence of mild (not requiring medication) psychiatric symptoms decreased. However, 10 to 20 percent of patients continued to experience psychiatric symptoms at final follow-up. | Strength: comparison group of rheumatoid arthritis patients |
| Sonino et al11 | Sample size: 66; age: 39.2±13.0 (mean±SD); gender: 12 M, 48 F; psychiatric interview of patient | Semi-structured interview based on Paykel’s clinical interview for depression | Overall, 27 percent of patients experienced depression prodromal to CS and 62 percent of patients experienced major depressive disorder during CS. About 70 percent of patients made a full recovery from depression, though there was no significant change in others, and even worsening of depression in two patients. | Limitations: use of semi-structured rather than fully structured psychiatric interviews, use of family history method rather than family interview method for ascertaining psychiatric illness in relatives, small sample size |
| Starkman et al12 | Sample size: 23; age range: 19 to 60; gender: 5 M, 18 F; psychiatric interview of patient | Semi-structured psychiatric interview; Hamilton Rating Scale for Depression | Prior to treatment of CS, 18 of 23 patients reported depressed mood. Six patients reported suicidal thoughts and, of these, 2 had attempted suicide after diagnosis of CS. After treatment of CS, 9 of the 18 patients with depressed mood prior reported normal mood; 4 patients reported improvement in depressed mood. Improvement in mood was described as fewer days feeling depressed, shorter duration of depressed episodes, and a less “deep” or “all-encompassing” quality of depression; patients stopped experiencing depression apart from an external trigger. The mean modified Hamilton depression score was 10.8±5.8 at the initial visit but 5.5±4.8 at the last visit, which was a statistically significant difference. | Strengths: large sample size, nationwide survey between 1960 and 2005, included four institutions; limitation: retrospective review |
| Dorn et al13 | Sample size: 33, age: 19 to 30; gender: 3 M, 14 F; psychiatric interviews of patients and questionnaires completed by patients | SADS-L, ADDS, Hamilton Rating Scale for Depression, POMS, SCL-90-R, STAI, STAS | In total, 67 percent of patients with CS reported psychiatric symptoms during illness. Of those, 50 percent reported major depression. Additionally, 52 percent of all patients enrolled in the study presented with atypical depression, the most common psychiatric manifestation. Of these patients, 47 percent also reported a comorbid psychiatric disorder. | Strength: comparison with Graves’ disease patients; limitation: sample drawn from a single institution |
| Jeffcoate et al14 | Sample size: 40; age: 43 (mean); gender: 12 M, 28 F; psychiatric interview of patient | Standardized psychiatric interview before and after treatment | Among the 40 patients enrolled, depression was the most common psychiatric symptom. In characterizing depression, 5 patients reported severe, 4 reported moderate, and 13 reported mild symptoms. Four patients experienced psychiatric symptoms unrelated to depression. Fourteen patients did not experience any psychiatric symptoms. After metyrapone treatment for hypercortisolism, 6 of the 13 patients who reported mild depression experienced improvement. | Strength: extensive qualitative description of psychiatric symptoms; limitation: small sample size |
| Loosen et al15 | Sample size: 20; age: 39.0±11.2 (mean±SD); gender: 1 M, 19 F; psychiatric interview of patient | Structured Clinical Interview for DSM-III, RDC, Family History RDC, Hamilton Rating Scale for Depression, Beck Depression Inventory | Generalized anxiety disorder was present in 79 percent of this cohort, the most common psychiatric diagnosis, while 68 percent of patients experienced major depressive disorder, 53 percent experienced panic disorder, and 63 percent experienced a combination of these 3 diagnoses. Psychiatric symptoms typically manifested at or after the occurrence of physical symptoms of CS, although the onset of panic disorder was associated with chronic stages of CS. These data reveal a syndrome of anxious depression in patients with CS. | Strength: used multiple questionnaires, patient group compared to matched controls; limitation: sample drawn from single institution |
| Langenecker et al16 | Sample size: 21; age: 34.4±14.9 (mean±SD); gender 4 M, 17 F; task administered to patient during fMRI | FEPT, fMRI | Patients with CS had more difficulty distinguishing facial expressions and had less activation in areas of the brain responsible for emotional processing. Hypercortisolism is associated with changes in brain structures related to emotional processing, perception, and regulation. | Strength: patients captured over a 10-year period; limitation: 65 percent formally interviewed |
| Bas-Hoogendam et al17 | Sample size: 21; age: 45.0±7.9 (mean ±SD); gender: 4 M, 17 F; questionnaires completed by patient, MRI scan | MADRS, Inventory of Depression Symptomatology, Beck Anxiety Inventory, Fear Questionnaire, Apathy Scale, Irritability Scale, Cognitive Failures Questionnaire, CS Severity Index, fMRI Faces task | Patients with CS experienced more depressive and anxious symptoms relative to healthy controls. Patients also scored higher on the Apathy Scale and noted higher levels of cognitive failure. Patients had decreased brain activation while processing facial emotions when compared with controls, even in patients who had achieved long-term remission of CS. | Strength: multiple measures used to assess symptom severity; limitation: small sample size |
| Alcalar et al18 | Sample size: 40; age: 39.6±10.6 (mean ±SD); gender: 9 M, 31 F; questionnaires completed by patient | Beck Depression Inventory, SF-36, MBSRQ | Patients reported a lower quality of life related to general health and body image in comparison with healthy controls. There were no differences in depression scores between the two groups. However, the mean Beck Depression Inventory depression score was higher in patients without remission than in patients with remission and healthy controls. On the SF-36, patients without remission scored lower on physical functioning, bodily pain, and general health when compared with patients with remission and healthy controls. Overall, patients with CS scored lower on quality of life measures, had worse body image perception, and had more severe depression compared to healthy controls, especially if surgery was not curative of illness. | Strength: comparison with matched healthy control group; limitation: small sample size |
| Heald et al19 | Sample size: 15; age: 18 to 75; gender: 5 M, 10 F; QOL questionnaires completed by patients, one questionnaire completed by informant | Psychological rating scales: HADS-UK, WHOQOL-BREF, GHQ-28, FACT, SAS1 and SAS2 | Patients with CS scored significantly worse on psychological well-being and psychosocial adjustment questionnaires compared to patients with all other types of pituitary tumors; scores for GHQ, HADS depression, and HADS anxiety were significantly higher. Treated patients with CS continued to perceive themselves as being more depressed, anxious, fatigued, and having poorer physical health and environmental and social adjustment relative to patients with treated other pituitary tumors. | Strength: multiple measures used to assess symptom severity; limitation: small sample size |
| Starkman et al20 | Sample size: 35; age range: 19 to 59; gender: 7 M, 38 F; semi-structured psychiatric interview of patient | Hamilton Depression Scale, Research Diagnostic Criteria | Regarding the severity of psychiatric symptoms, 34 percent of patients with CS in this cohort were considered mild, 26 percent were moderate, 29 percent were severe, and 11 percent were very severe. There was a statistically significant relationship between overall neuropsychiatric disability and levels of cortisol and ACTH. | Strength: multiple measures used to assess symptom severity; limitation: sample drawn from single institution |
| Castinetti et al21 | Sample size: 20; age range: 20 to 63; gender: 7 M, 13 F; retrospective study | Clinical evaluation every 15 to 30 days after initiation of mifepristone treatment | Five out of 20 patients in this cohort experienced psychiatric symptoms. Of these, four patients showed improvement of symptoms within the first week of mifepristone treatment for hypercortisolism. Of four patients who exhibited signs of psychosis, symptoms subsided in three patients after a week of treatment. | Strength: comparison with patients with other types of pituitary tumor; multiple questionnaires administered; limitation: response rate of 62 percent |
| Kelly et al22 | Sample size: 26; age range: 18 to 70; gender: 7 M, 19 F | Hamilton Rating Scale for Depression | Depressive symptoms rated at three and 12 months after treatment were significantly improved relative to before treatment. At three months, 20 of 26 patients experienced improvement in depression. At 12 months, 25 patients experienced improvement. | Strength: studied association between psychiatric symptoms and hormone levels; limitation: relies on subjective reports by patient |
CS: Cushing’s syndrome; SADS-L: Schedule for Affective Disorders and Schizophrenia, lifetime version; DSM: Diagnostic and Statistical Manual of Mental Disorders; fMNI: functional magnetic resonance imaging; SF-36: Short Form-36; SD: standard deviation; ADDS: Atypical Depression Diagnostic Scale; POMS: Profile of Mood States; SCL-90-R: Symptom Checklist 90-revised; STAI: State Trait Anxiety Inventory; STAS: State Trait Anger Scale; RDC: Research Diagnostic Criteria; FEPT: Facial Emotion Perception Task; MADRS: Montgomery-Asberg Depression Rating Scale; MBSRQ: Multidimensional Body-Self Relations Questionnaire; HADS-UK: Hospital Anxiety and Depression Scale – UK version; WHOQOL-BREF: World Health Organization Quality of Life - Brief version; GHQ-28: General Health Questionnaire-28; FACT: Functional Assessment of Cancer Therapy; SAS1: Social Adjustment Scale-modified-completed by patient; SAS2: Social Adjustment Scale-modified-completed by another person who knows the patient well.