Persisting Muscle Dysfunction in Cushing's Syndrome Despite Biochemical Remission

Frederick Vogel; Leah T. Braun; German Rubinstein; Stephanie Zopp; Heike Künzel; Finn Strasding; Adriana Albani; Anna Riester; Ralf Schmidmaier; Martin Bidlingmaier; Marcus Quinkler; Timo Deutschbein; Felix Beuschlein; Martin Reincke


J Clin Endocrinol Metab. 2020;105(12) 

In This Article


This is the first prospective longitudinal study investigating the long-term outcome of CS-associated myopathy. The main finding of our study is that muscle strength remains impaired even after years in remission, and independent of glucocorticoid replacement status. Another interesting finding is that at 6 months follow-up, grip strength and CRT performance show opposite effects; whereas grip strength has worsened, CRT performance has improved. Because of a correlation between CRT performance and BMI at baseline (P = 0.007), it is likely that an improvement in CRT after achieving remission is due to the considerable reduction in BMI. However, grip strength remained impaired and showed no improvement during follow-up, while the temporary worsening of grip strength during the steroid withdrawal phase seems to be reversible. Despite biochemical remission, myopathy can thus be regarded as another persistent clinical symptom, as it is known for cognitive function, mental fatigue, and cardiovascular risk.[10,11,25] Why patients with CS in remission showed a temporary worsening in grip strength 6 months after surgery remains unclear in term of pathophysiology. That muscle function and thus myopathy is affected by glucocorticoid withdrawal due to a relative glucocorticoid resistance or inflammatory processes is quite conceivable.[26,27] Also, high-dose glucocorticoid replacement for the prevention of a steroid withdrawal syndrome directly after surgery could account for the transient aggravation. Grip strength measurements performed by the JAMAR hydraulic hand dynamometer depends on muscle function but also requires an intact function of joints and connective tissue. It is possible that muscle and joint pain, which can as well be provoked by glucocorticoid withdrawal, affect strength measurements.

As predictors for myopathy outcome and muscle function in remission of CS, we identified age, WHR, and HbA1c. Age correlated inversely with relative changes of grip strength and CRT in relation to baseline (Supplement Figure 1; all supplementary material and figures are located in a digital research materials repository[17]). Moreover, a high WHR and HbA1c at baseline were associated with a poorer outcome in muscle function (Table 2 and Table 3), while BMI and body fat percentage estimated with bioimpedance measurements were not. These results suggest that particularly an unfavorable fat distribution with elevated WHR and diabetic metabolic state contributes to adverse outcome of myopathy. These findings are concordant with our previous analysis of patients with florid CS and matched obese controls, where we found that patients suffering from both hypercortisolism and impaired fasting glucose had the strongest impairment in grip strength and CRT performance, implicating impaired glucose metabolism as an important driver in the pathomechanism of myopathy. According to our data analysis we assume that impaired glucose metabolism affects both muscle function during hypercortisolism and the long-term outcome of CS-associated myopathy. The link between impaired muscle function and diabetes has already been described, but the underlying pathomechanisms remain largely unclear.[28] Furthermore, other factors such as growth hormone and sex hormone suppression are proposed to play a crucial role in influencing CS-associated myopathy in the long run.[29,30]

Whether the necessity of a long-term glucocorticoid replacement influences muscle strength or myopathy outcome remains controversial. Geer et al found an inverse correlation between duration of glucocorticoid replacement and skeletal muscle mass in patients with CS in remission, suggesting continued muscle loss resulting from glucocorticoid replacement therapy.[13] In our subgroup analysis of 29 patients followed for more than 3 years in remission, we could not observe any differences in muscle function or estimated muscle mass by bioimpedance measurements between patients with or without glucocorticoid replacement 3 years after surgical cure. Consistent with previous findings on the body composition of patients with CS in remission,[12] estimated body fat percentage decreased, whereas estimated muscle mass remained at a constant level. These results indicate that weight loss and reduction of BMI after successful treatment is most likely due to a loss of body fat than due to continuing muscle loss after hypercortisolism. However, clinically manifest—even though subclinical—hypercortisolism was associated with muscle waste and the severity of hypercortisolism was correlated with lower muscle mass,[31,32] the remission of CS exhibited no relevant gain in muscle function or muscle mass.[12,13] The severity of cortisol excess in patients with adrenal CS was also reported to be related to visceral fat area, visceral-to-subcutaneous fat ratio and skeletal muscle radiodensity 1 year after adrenalectomy.[33] In a recently published cross-sectional study with 17 patients, Roerink et al reported a reduced aerobic exercise capacity after years in remission. Due to nonsignificant differences in muscle fiber cross-sectional area and capillarization, impaired aerobic capacity was explained by a persisting cardiac dysfunction, as seen in patients with florid CS.[15] In terms of pathology, muscle biopsy in glucocorticoid-induced myopathy demonstrated severe histochemical type-2 muscle fiber atrophy.[34] Whether these alterations during hypercortisolism are reversible and whether intermuscular fat infiltration plays a critical role in its pathophysiology remains largely unknown.[35] Likewise, if muscle atrophy leads to an abnormal glucose metabolism or if a diabetic metabolic state promotes atrophy and/or fat infiltration is uncertain. We have recently shown that hand grip strength seems to be more susceptible to hypercortisolism in a glucocorticoid receptor A3669G wild type than in A3669G minor allele carriers, explaining the partially interindividual differences of glucocorticoid-induced myopathy in patients with endogenous CS.[36] Given the fact that corticosteroid-induced myopathy also affects 60% of patients receiving corticosteroid treatment,[3] and muscle function showed a strong correlation with quality of life, especially in the long-term period, there is great importance to clarify pathophysiologic basics and to find treatment options for patients with glucocorticoid-induced myopathy.

Strengths and Limitations of the Study

This is a prospective study with a reasonable number of patients having adequate statistically power. A limitation of the study is the decreasing number of patients with time in remission caused by the continuing longitudinal inclusion of patients with florid CS between 2012 and 2018. Given that the statistical power of this study was calculated based on grip strength and CRT changes, it cannot be ruled out that the statistical power for other results and clinical parameters is insufficient. Some results might therefore change after all 88 patients have completed the 4-year follow-up. Moreover, a further limitation could be a selection or observer bias, as the study was performed in 3 sites of the German Cushing's Registry.