Cushing’s disease (CD) is a rare disorder caused by an adrenocorticotropic hormone (ACTH)-secreting pituitary adenoma. Chronic exposure to hypercortisolism leads to significant morbidities, which may be only partially reversible after remission of the disease, as well as to impairment of the health-related quality of life (HRQoL) and an increase in mortality. Transsphenoidal surgery (TSS) is the treatment of choice, and recurrence rates vary widely, confirming the need for lifelong follow-up. This review summarises the studies performed on HRQoL, recurrence rates and morbidities in patients who have CD.
Cushing’s disease, quality of life, recurrence, morbidity
Isabel Huguet, Georgia Ntali, Ashley Grossman and Niki Karavitaki have no conflicts of interest to declare. No funding was received in the publication of this article
February 18, 2015
Isabel Huguet, Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill Hospital, Old Rd, Headington, Oxford, OX3 7LJ, UK.
Open Access: This article is published under the Creative Commons Attribution Noncommercial License, which permits any non-commercial use, distribution, adaptation and reproduction provided the original author(s) and source are given appropriate credit.
Cushing’s disease (CD) is a rare condition caused by an adrenocorticotropic hormone (ACTH)-secreting pituitary adenoma. Chronic hypercortisolism is associated with the development of several morbidities that impair health-related quality of life (HRQoL) and contribute to an increased mortality rate.1–5 Obesity and metabolic alterations, hypertension and cardio-/ cerebrovascular complications, neuropsychiatric, muscle/skeletal, hypercoagulability/thromboembolism and immune consequences remain the most challenging.
Despite successful treatment of CD, a number of adverse consequences may persist long after cure or may even be irreversible. Moreover, the remission/cure criteria of CD vary between different studies, making comparison of the results difficult.
This paper aims to review and summarise the studies performed on HRQoL, recurrence rates and morbidities during long-term follow-up in patients who have CD.
Materials and Methods
The available literature was evaluated to address questions on HRQoL, recurrence and morbidities in CD. The literature search was conducted in two stages: (1) identification, review and inclusion of all the most relevant articles published in PubMed having the keywords CD, remission, cure, HRQoL and morbidities and (2) additional hand research conducted on the basis of bibliographies of identified articles, with articles referring to paediatric population and case reports excluded and with papers referring to Cushing’s syndrome (CS) reviewed and included only if presenting data on CD.
Quality of LifeHRQoL was initially assessed in CD patients using such generic measures as the Short Form (SF)-366 and the SF-127 and measures of specific symptoms associated with the disease, including the Hospital Anxiety and Depression Scale (HADS).8 More recently, two disease-specific measures, the CushingQoL9 and the Tuebingen CD-2510,11 have been developed. Tables 1 and 2 present studies evaluating QoL using various questionnaires in CD patients having active disease or in remission.
Quality of life is significantly impaired not only in patients with active CD,9 but also in those in long-term remission,12,13 regardless of the presence of hormonal deficiencies9 or treatment strategies,14,15 and patients who have CS report more negative illness perceptions than do patients who have other acute or chronic conditions.16
Quality of Life Assessed by Generic Questionnaires
Lindholm et al. reported that patients in remission for more than 5 years after initial surgery scored significantly worse in all subscales of SF-36 except for mental health and bodily pain.17 van Aken et al. evaluated patients cured for a mean period of 13.6 years and showed that general perceived well-being was reduced compared with healthy controls for all subscales in SF-36 and the Nottingham Health Profile (NHP). Moreover, such patients scored worse in all subscales of fatigue (Multidimensional Fatigue Inventory [MFI]-20), anxiety and depression (HADS).18 In comparison with subjects having other pituitary adenomas, patients who had CD were the most severely affected in all parameters of the SF-36 questionnaire.8 Sonino et al. studied patients who had CS in remission (the majority of them had CD) for 1 to 3 years and found significantly higher scores in anxiety, depression, anger, hostility and psychotic symptoms in the Symptom Rating Test (SRT) questionnaire in comparison with healthy controls.19 When the Beck Depression Inventory (BDI), SF-36 and the multidimensional body-self relations questionnaire (MBSRQ) were used, patients who had CD demonstrated lower QoL, lower body image perception and higher levels of depression compared with healthy controls, particularly in cases of persistent disease.15
Quality of Life Assessed with Disease-specific Questionnaires
Since 2008, two disease-specific questionnaires have been developed: CushingQoL and Tuebingen CD-25. The CushingQoL questionnaire was evaluated by Webb et al. in a multicentre European study. Active CD was associated with worse scores, but the presence of hypopituitarism or prior pituitary radiotherapy did not determine differences in the scores.9 Similarly, Santos et al. found that active CD patients scored worse on the CushingQoL questionnaire than did cured subjects.20 Wagenmakers et al. found that CD patients in remission without hormonal deficiencies scored significantly better than those having hormone deficiencies but significantly worse thanthe control group on 50 % of the items of the questionnaires.14 The Tuebingen CD-25 also demonstrated significant differences in all subscales and the total score between active CD patients and healthy controls.10,11
The post-operative improvement in HRQoL could be predicted by the presence of pre-operative HRQoL impairment, andyounger patients were more likely to improve. Patients without post-operative pituitary deficiencies improved significantly in the cognition scale.21,22
QoL does not change after short-term biochemical remission induced by medical therapy but may improve after sustained control of the hypercortisolism.23,24
To summarise, most of the results on HRQoL in CD derive from generic questionnaires raising concerns about how appropriate these are for the reliable and accurate assessment of the HRQoL of patients with this rare condition. Interestingly, despite the use of the same type of questionnaire in some studies, the subscales mainly affected show variation among them, suggesting that either CD affects several dimensions in QoL in a heterogeneous way in different patient groups, or these questionnaires are not specific enough. The newly developed questionnaires focus on important disease-specific aspects of the QoL, and their sensitivity in detecting changes renders them a very promising and useful tool in clinical practice.
Transsphenoidal surgery is the treatment of choice in CD, with immediate post-operative remission rates ranging from 59 % to 94 % and recurrence rates from 3 % to 46 %, both depending upon the definition criteria, the duration of follow-up, the number of patients studied and the inclusion of macroadenomas (see Table 3).2,25–29 A small number of studies have used undetectable or very low post-operative serum cortisol levels as a strict criterion of remission, but most have defined effective remission as the resolution of clinical features and the reversal of hypercortisolism (in serum or urine), along with the recovery of cortisol suppressibility on dexamethasone administration or a normal cortisol circadian rhythm. Predictors of remission in CD include age at diagnosis, presence of hypertension or diabetes,4,5 response to desmopressin testing,30 identification of tumour at surgery and an adenoma histology positive for ACTH.31–33
Out of concern about recurrence of CD after initial remission, several parameters have been evaluated and are still matter of debate. Factors that have been associated with a low (but not zero) risk of CD recurrence include undetectable or low early morning serum levels of cortisol,33 low plasma levels of ACTH and prolonged requirement for glucocorticoid replacement after pituitary surgery. The effects of serum cortisol levels in the early post-operative period on predicting relapse have been assessed in various studies. No recurrences were found by Trainer et al.34 during a median follow-up of 40 months in patients with a postoperative serum cortisol of <50 nmol/l; similar results were reported after a median follow-up of 58 months by McCance et al.35 In contrast to these findings, several series have reported recurrence rates of 11.5 % and 15 % despite a serum cortisol <50 nmol/l post surgery.27,28 Another possible parameter is the length of adrenal insufficiency post-TSS. Thus it has been suggested that patients who have a shorter duration of adrenal insufficiency have a significantly higher risk of relapse.36
Accordingly, after successful treatment of CD, there is no accurate criterion that can ensure lifelong cure, and although the evidence shows a more optimal outcome in patients who achieve severe cortisol deficiency after TSS, lifelong follow-up is mandatory.
CD is associated with significant comorbidities – metabolic and vascular complications, osteoporosis, neuropsychiatric dysfunction and immunosuppression – that increase mortality and affect daily life. A number of these may persist long after cure or may even be permanent.37
Metabolic and Vascular Complications
It is well recognised that CD increases cardiovascular risk, with hypertension and obesity the most common associated risk factors. The cardiovascular complications are part of the metabolic syndrome, but hypertension related to endogenous hypercortisolism is not simply a component of the CS-related metabolic syndrome. The renin–angiotensin system, mineralocorticoid activity, the sympathetic nervous system and the vasoregulatory systems have been reported to be involved in the pathophysiology of hypertension, but the mechanisms are only partially understood.38
The adverse cardiovascular risk profile of patients who have CD39 is attributed to metabolic and vascular aberrations, as well as to changes in cardiac structure and function. Patients who have CD have increased leptin,40,41 resistin42 and pro-inflammatory agents, such as tumour necrosis factor-α and interleukin-6, C-reactive protein and low ghrelin levels.40 Moreover, they are characterised by a prothrombotic phenotype attributed to various abnormalities of coagulation and fibrinolysis: these include shortened activated partial thromboplastin time,43 increased factor VIII, von Willebrand factor, fibrinogen and plasminogen activator inhibitor-1,43 decreased fibrinolytic capacity44 and increased α2-antiplasmin.45 Endothelium-dependent flow-mediated vasodilatation is impaired, and several humoral markers of endothelial dysfunction (such as endothelin, homocysteine, vascular endothelial growth factor, osteoprotegerin and cell adhesion molecules) are elevated. Cardiac echocardiograms demonstrate left ventricular hypertrophy, concentric remodelling and diastolic and systolic dysfunction.
Persistent clinical abnormalities have been documented in terms of cardiovascular complications, showing that remission of hypercortisolaemia reduces, but does not completely eliminate them. Colao et al. reported that 27 % of patients having CD in remission for 5 years had persistently atherosclerotic plaques, compared with only 3 % of gender-, age- and body mass index–matched controls.45 Faggiano et al. also found persistence of the metabolic syndrome, vascular damage and atherosclerotic plaques after disease remission.46
Thus it is likely that disease remission does not entirely reverse cardiovascular morbidities affecting long-term survival and that lifelong follow-up is needed, with particular emphasis on cardiovascular risk factors.
Bone loss is attributed to decreased osteoblastic activity, increased osteoclastic bone resorption and impaired enteral calcium absorption. In the ERCUSYN study, osteopenia at the spine and hip was reported in 40 % and 46 % of patients who had CD, respectively, and osteoporosis at the spine and hip in 22 % and 12 %, respectively.47,48 Bone mineral density (BMD) does not completely recover following remission,49,50 though normalisation in some skeletal sites has been reported in the long-term.51
Glucocorticoid-induced vertebral fractures may develop even in the presence of normal or slightly low BMD. The risk of fractures at comparable BMD values with controls suggests that components of bone strength, not assessed by routine densitometric approaches, are also affected by glucocorticoids (including bone architecture, geometry and remodelling). Methods of assessment other than measurement of BMD are required, because despite the improvement of BMD after correction of hypercortisolism, the quality of the bone likely remains compromised.
Hypercortisolism induces reversible immunosuppression. During hypercortisolaemia, autoimmune disorders improve but may worsen during remission and new ones develop.52 There is a high risk of superficial fungal, opportunistic or bacterial infections.
Glucocorticoids are known to influence many functions of the central nervous system. Hypercortisolaemia is associated with depression, disrupted sleep and a wide range of cognitive impairments (derangement of memory – especially short-term – irritability and decreased concentration). High anxiety levels and low externalising behaviour are common emotional disorders.53 Smaller hippocampal volumes and generalised brain atrophy have been described.54 Functional magnetic resonance imaging studies in patients having CD have demonstrated emotion-processing difficulties and hyperactivity in the frontal and subcortical regions, similar to major depressive disorders. After remission, hippocampal volumes increase and emotional and cognitive functions improve,53–55 but profound structural alterations in the brain remain such as smaller volumes in the anterior cingulate cortex, a structure involved in cognitive–affective processes and behavioural adaptation.56
The new data on persistent changes in the central nervous system after cure of CD support the hypothesis that psychiatric symptoms and cognitive impairment could be related to structural changes, providing the basis for future research on the neurobiological background of psychological dysfunction in this complex condition.
CD is associated with significant adverse sequelae affecting longterm morbidity, mortality and quality of life. To some extent, the duration and severity of hypercortisolism determine the possibility of reversion of the morbidities, but a number of manifestations may persist long after cure – possibly permanently. Generic questionnaires and disease-specific measures have evaluated many aspects of the quality of life impaired, not only during the active state, but also when in remission.
Despite initial successful treatment, there is a risk of relapse, and post-operative hypocortisolism is the most significant predictor of remission. These long-term and persistent changes are challenging to our understanding of hypercortisolaemia but in a clinical context suggest that long-term follow-up is essential in all patients having CD, even those apparently cured.