UW Shoulder and Elbow Academy: Managing the stiff (frozen) shoulder

Effects of comorbidities on the outcomes of manipulation under anesthesia for primary stiff shoulder

These authors sought to assess how comorbidities influence the recovery speed and clinical outcomes after manipulation under anesthesia (MUA) in 281 primary stiff shoulders. They divided patients into the control (n . 203), diabetes mellitus (DM)(n . 32), hyperlipidemia (n . 26), and thyroid disorder (n . 20) groups. The mean HbA1c (normal range: 4.5%-5.6%) of the DM group was 7.3%. Among the 32 diabetic patients, 24 (75%) had HbA1c ≥ 6.5%.

MUA was considered if all four criteria were met:

(1) limited passive ROM compared with the opposite normal shoulder (passive glenohumeral motion is 20 or less of forward flexion with holding down the scapula by the surgeon, and movements are made mainly using the scapulothoracic motion);

(2) pain and stiffness persisted for at least 3 months and did not respond to sufficient conservative treatment;

(3) stiff shoulder is in the frozen stage; and

(4) normal shoulder x-rays.

MUA was performed as an outpatient procedure under interscalene regional anesthesia with the patient in the supine position. With one hand stabilizing the scapula, the range of glenohumeral movement was assessed and recorded. To avoid iatrogenic injuries such as humeral fracture, the surgeon performed manipulation by holding the patient’s arm between the shoulder and elbow with the other hand to form a short lever arm. The shoulder was manipulated sequentially through a range of forward flexion, abduction, external rotation, cross-body adduction, and internal rotation. The procedure was followed by immediate physical therapy.

A successful MUA was defined as achieving the ROM for the passive forward flexion and external rotation at the side within 15 degrees and internal rotation to the posterior within 3 spinal levels compared with the normal contralateral side.

Significant improvements in range of motion (ROM) and clinical scores at 3 months after MUA were observed in all groups.

Significant differences in ROM among the 4 groups were also observed during follow-up. The DM group had significantly lower ROM values, even at 3 months after MUA, compared with the control group. The ROM recovery speed after MUA was slowest in the DM group, followed by the thyroid disorder, hyperlipidemia, and control groups.

Most (90.6%) of the DM group experienced late recovery. The proportion of nonsuccessful MUA was higher in the DM and thyroid disorder groups than that in the control and hyperlipidemia groups.

One case of spiral humeral fracture occurred after MUA while rotating the arm externally during this study. No other complications occurred.

During follow-up, there were no statistically significant differences among groups regarding the visual analog scale, University of California at Los Angeles shoulder, and Constant scores.

Comment: We have found manipulation under anesthesia to be useful for refractory shoulder stiffness in patients without glenohumeral arthritis or osteopenia. We use a short (5 minute) general anesthetic coupled with complete muscle paralysis achieved with succinylcholine. This has been a safe and effective procedure for patients with refractory idiopathic frozen shoulder and for shoulders with refractory post operative stiffness.

In the photo below, one anesthesiologist (upper left) is providing oxygen ventilation while another (upper right) is administering the intravenous succinylcholine. To avoid the risk of humeral fracture (such as the one reported in this paper) or cuff injury, we manipulate in flexion, cross body adduction, and abduction, but not in rotation.