In the present study in controlled acromegaly, joint complaints were reported frequently in previously-assessed peripheral joints (viz. hands, hips, knees). Additionally, shoulder complaints were common, whereas feet complaints were reported less frequently. Over 90% of patients showed radiographic signs of OA at any peripheral joint site, which occurred in a generalized manner in virtually all patients. Radiographic MTP1 OA was as prevalent as radiographic knee OA, whereas radiographic glenohumeral OA was similarly prevalent as hip OA. Higher pre-treatment IGF-1 levels, and current pharmacological treatment were identified as risk factors for radiographic glenohumeral OA, whereas no risk factors for MTP1 joint OA could be identified.
The high prevalence of acromegalic arthropathy - investigating hips, knees, hands, and spine joints – has been described extensively [2, 3, 6, 7, 13–15]. When investigating these previously-assessed joints in the present population, the mismatch between clinical symptoms and the presence of radiographic abnormalities was noted, which differed from the previously reported prevalence in some joints (e.g. hips) [61]. This difference might be due to the selected population of clearly well-controlled patients with years-long normal circulating GH, and IGF-1 levels at the time of the study visit.
Studies systematically investigating acromegalic arthropathy of other peripheral joints were lacking. Because of the (potential) invalidating nature of shoulder and forefoot OA, the present study investigated whether these joints were affected in acromegaly [20, 18, 26]. Moreover, both in acromegalic and primary OA patients, (radiographic) OA can be limited to one specific joint (viz. mono-articular), or generalized (viz. polyarticular) [59], which could only be investigated by assessing several peripheral joints simultaneously. Virtually all patients in our controlled acromegaly cohort had generalized radiographic OA, with solely two patients having localized MTP1 OA, underlining the differences between systemic causes of OA (i.e. GH excess), and biomechanical factors (e.g. strenuous use of one joint).
When assessing the shoulder joint in the present population of controlled acromegalics, prevalence of self-reported complaints varied from 12.2% (stiffness) to 38.8% (pain), being consistent with previous reports, although previous studies were mainly performed in active disease [16, 17]. Moreover, higher disability scores (using the validated DASH questionnaire as a measure for upper limb disability) were reported by our controlled acromegalics compared to the general population [62], although scores were lower than reported scores in patients with carpal tunnel syndrome [63], or rheumatoid arthritis [64]. Moreover, disability of the upper limb was negatively associated with HR-QoL in the present study, highlighting the invalidating nature of shoulder complaints.
For the first time, the characteristic radiographic features of acromegalic arthropathy were described in the glenohumeral joint [9–12, 65], with radiographic glenohumeral OA being present in 41.2% of controlled patients. This radiographic OA prevalence is higher than the prevalence of radiographic OA at other large joint sites [2, 3, 6, 7, 13–15], and higher than in the general population, although the latter varied greatly depending on the cohort and scoring methods across different studies [54, 66–68].
Although age, female sex, and BMI are established risk factors for radiographic glenohumeral OA in the general population [54–57], solely age was identified as a risk factor in the present population. Furthermore, higher pre-treatment IGF-1 levels, and current use of pharmacological treatment to achieve remission were associated with an increased risk of glenohumeral OA. These findings are in accordance with our previous studies, showing a significant relationship between acromegaly-specific risk factors, and the presence and progression of radiographic OA of the spine, hip, knee and hand joints [13, 69, 70], indicating the partial systemic nature of glenohumeral OA.
With respect to forefoot OA, we observed a comparable prevalence of self-reported pain or stiffness of the feet in controlled acromegaly, and the general population. To date, foot complaints have never been systematically assessed in patients with acromegaly, except for a few ultrasound studies focusing on tendinopathy [23, 71]. Complaints of the feet increase the demand of care, since almost 10% of all musculoskeletal consultations with general practitioners comprise foot (or ankle) pain [72], which have detrimental effects on daily functioning and QoL [25, 26].
Radiographic OA of the MTP1 joint was observed in over half of the controlled acromegaly patients, thereby being much more prevalent than radiographic OA at other peripheral joint sites in acromegalics [2, 3, 6, 7, 13–15]. By contrast, radiographic OA of the MTP2-5, and IP1 joint occurred infrequently. In the general population, radiographic OA of the MTP1 joint ranged from 5.0–42.0%, and 3.0–4.9% for the MTP2-5 joints, depending on age, sex, and country of origin [58]. Prevalence of radiographic OA of the MTP1 joint in Dutch individuals ranged from 31.4% for ages 55–59 years to 44.4% for ages > 80 years [68], which is lower than the prevalence observed in controlled acromegalics. In the present study, however, the exact clinical significance of the presence of foot complaints and radiographic OA cannot be determined in the absence of validated questionnaires. Nonetheless, the high prevalence of radiographic OA of the MTP1 joint combined with the reported clinical symptoms, as well as the association between foot deformities (e.g. hallux valgus) and radiographic MTP1 OA [58], the MTP1 joint is an important joint to evaluate in acromegaly patients.
Established risk factors for radiographic MTP1 OA in the general population are higher age, female sex and increased BMI [58, 59, 73], albeit these factors could not be identified in acromegalics. Moreover, measures of disease activity/severity were not detected as risk factors for MTP1 OA. Additional factors related to education, and occupation, e.g. lower educational attainment, and (a history of) physically demanding occupation (e.g. frequent stair climbing, professional dance) appear to contribute more to the risk of MTP1 clinical or radiographic OA in the general population, and primary OA [52, 74, 73, 75], and we therefore assume that the combination of biomechanical factors, including acromegaly-related physical changes (e.g. feet enlargement, joint misalignment), contribute most to the development of MTP1 OA in acromegaly. We advise to structurally assess the foot joints in acromegaly patients, since more stringent disease control is not likely to improve the foot complaints, whereas a multitude of other interventions are available.
To date, acromegalic arthropathy cannot be prevented, and its optimal management remains to be elucidated [15, 76]. We propose a diagnostic and therapeutic algorithm regarding the management of joint complications in acromegaly in Fig. 4, based on our extensive clinical expertise and research. All patients with acromegaly should be diagnosed, treated, and followed at a pituitary center of excellence (PTCOE) [77], of which the exact implementation is dependent on the (inter)national organization of health care. For acromegaly care, an exemplary PTCOE would harbor a multidisciplinary team (MDT) of an endocrinologist, neurosurgeon, rheumatologist, radiologist, orthopedic surgeon, physical therapist, and dietician. Stringent GH/IGF-1 hypersecretion control remains the cornerstone of acromegaly management, since smoldering disease might cause cartilage loss, and arthropathy progression [13, 70, 15]. Assessment of joint complaints should be performed regularly, preferably using validated questionnaires. In the case of joint symptoms, referral to MDT members might be necessary for adequate diagnostics, of which the exact route depends on the etiology of (osteo)arthritis and extensiveness of affected joints. In this respect, there should be a focus on etiologies for which effective treatment strategies are available, such as neuropathic pain [27], and inflammatory rheumatic disease [78]. Unfortunately, treatment strategies for acromegalic arthropathy are mostly symptomatic, and similar to treatment options for primary OA, including lifestyle advice, analgesics, physical therapy, intra-articular corticosteroid injections, or joint replacement therapy [79, 80]. Notably, none of these treatment strategies have been formally studied in acromegalic arthropathy to date. In case of persistent joint-related disability despite adequate biochemical disease control, and first-line treatment, we advise discussion in MDT meetings to evaluate an individualized, personalized strategy.
Several limitations of this study need to be addressed. First, although the patient population potentially was relatively small, this study describes a unique cohort of acromegaly patients in remission, and is the first study describing both clinical and radiographic OA in the shoulder and feet. Modified KL atlases based on the existing KL atlases of the knee and hand were used for the glenohumeral and MTP joints, similar to previous studies [48–51, 47, 52, 53]. No (modified) KL atlas was available for the acromioclavicular and sternoclavicular joints, and, therefore, as mentioned prior, solely the glenohumeral joint of the shoulder was assessed. Additionally, the unavailability of a widely-used KL atlas, as well as the unavailability of an OARSI atlas of the shoulder and foot, hampered the comparison of acromegalic patients with controls, since assessed radiographic glenohumeral and forefoot OA (age-matched) healthy controls were unavailable. Finally, validated questionnaires for feet complaints are unfortunately lacking, and therefore not included in this study.
In conclusion, for the first time, a high prevalence of both self-reported joint complaints and radiographic OA of newly assessed peripheral joint sites in patients with controlled acromegaly is reported. Whereas the direct effects of transient GH/IGF-1 excess are assumed to cause acromegalic arthropathy at most joint sites, resulting in a generalized (radiographic) OA pattern, primarily biomechanical factors play a role in the development of MTP1 joint OA. Available treatment options of acromegalic arthropathy remain symptomatic, and should be the focus of future studies, representing an important unmet need in the current care of acromegaly patients.