Clinical Outcomes of Arthroscopic Subscapularis Release for Obstetric Brachial Plexus Palsy with Internal Rotation Shoulder Conracture

INTRODUCTION

Difficult delivery can lead to a major neurological insult in the upper limb and it is called Obstetric brachial plexus palsy (OBPP). Most children (80-90%) recover within few months after birth with minor or no residual functional deficits but unfortunately, some of them didn't recover completely and develop secondary deformity because of muscular imbalance with weak abductors and external rotators.¹ The foremost affected roots are C5 - C6 and this can result in adduction and  internal rotation contracture in most of cases.^2-4

 A lot of surgical techniques were reported to increase shoulder range of abduction and external rotation. These surgical procedures differ according to the patient’s age and the degree of deformity. Contracture release, specific tendon transfers, and rotational osteotomies are examples of these techniques. Shortened structures should be released firstly to gain a good range of shoulder external rotation.^5,6

There's no consensus about the technique that gives the satisfactory results for shoulder function and mobility. Surgical procedures for OBPP are widely different with no clear evidence to favor one procedure over the other in addition to insufficient reports about arthroscopic subscapularis release in the literature.⁷

In this study, we aimed to assess the arthroscopic subscapularis release as a treatment option for internal rotation shoulder contracture associated with OBPP.

PATIENTS AND METHODS

Over seven years (between 2012 and 2018) we conducted this prospective study including 28 patients of OBPP with internal rotation shoulder contracture with < 10° of passive external rotation and before 6 years of age. Arthroscopic subscapularis release was done for all of them without any additional tendon transfer. The mean age was 3 years (2 – 5 years). There was an injury at the level of C5-C6 in 22 children, and at the level of C5-C6-C7 in 4 children, and a complete plexus injury in only 2 children. The mean time for follow-up was 27 months (24 - 36 months). Full clinical examination including ranges of motion were assessed and recorded in all patients with positive trumpet sign in all of them with inability to elevate the hand up to the mouth level (figure-1). The patients were evaluated preoperatively and postoperatively using the Mallet classification score system (table-1).⁸ 

Fig. 1: Preoperative active range of motion.

Inclusion criteria: This study included patients with shoulder internal rotation contractures, limitation in passive external rotation (less than 10 degrees), younger than 6 years before uncorrectable glenoid deformities, with no history previous shoulder surgery.

Surgical technique:

All patients were operated under general anaesthesia and we used the lateral decubitus position for all patients because we found a great difficulty to use the beach chair position in such small children. The upper extremity was draped in a classic pattern and positioned freely to re-evaluate the passive range of motion under anaesthesia before and after release. The assistant was holding the upper extremity in a position of 20° abduction and neutral rotation with gentle traction during the procedure with no need for a traction system. We used a small scope (2.7 mm/ 30° angle), 90° radiofrequency probe, and 2.5 mm shaver blades for this procedure. The posterior portal was made at the posterior soft spot about 1 cm below and medial to the posterolateral angle of the acromion. Shoulder abduction and lateral traction of the arm were helpful to enter the contracted joint with a wide bore needle (18G epidural needle) and inflating the joint with a few amount of saline to distend this small joint, then insert a blunt trocar through the posterior portal. Then we made the anterior portal using outside-in technique under direct visualization at the rotator interval just above the subscapularis tendon. We didn't use a 5.5-mm working cannula as in adult shoulder but we used 18G epidural needle as an out-flow cannula and introduced the radiofrequency electrode directly through the anterior portal to start the release of subscapularis tendon and anterior capsule until we see the muscle fibres (figure 2: a & b). In most of the cases (22 out of 28), this release was sufficient to gain more than 45°of intraoperative passive external rotation (figure 3: a & b). In few cases (6 out of 28), we would have to extend our release to include rotator interval, coracohumeral ligament, superior and inferior glenohumeral ligaments to attain over 45° of intraoperative passive external rotation.

After closure of skin incisions, shoulder spica cast was applied for all patients with maximum external rotation and 90° abduction. The cast was then removed after 4 weeks and physical therapy was started including gentle stretching and night bracing for another 4 weeks.All patients underwent regular postoperative follow-up at 4 weeks, 8 weeks, 3 months, 6 months, 1 year, and at the end of follow up 2 years postoperatively reporting the ranges of motion, shoulder function, and shoulder strength using the Mallet scoring system.

Fig. 2: Subscapularis tendon before (a) and after (b) release.

Fig. 3: Intraoperative passive external rotation before (a) and after release (b).

Statistical analysis:

Statistical analysis was carried out using the IBM SPSS Statistics 26 for Windows. Results were expressed as mean ± SD for quantitative variables. Confidence interval of 95% and p < 0.05 were considered statistically significant.

RESULTS

This prospective study was conducted over seven years (between 2012 and 2018), and included 28 patients who were presented with shoulder internal rotation contracture secondary to obstetric palsy before the age of 6 years with passive external rotation <10°. They were treated with shoulder arthroscopy to release the subscapularis tendon without any tendon transfer.   The average age was 3 years (2 – 5 years). Twenty-two children had an injury at C5-C6, four had an injury at C5-C6-C7 and two had a complete injury. The average follow-up was 27 months (24 - 36 months). Tables 2 and 3 summarize the demographic and clinical data of all patients. The mean passive ER was 70° (40° - 90°) and the mean active ER was 60° (20° - 90°) at the end of follow up (figure 4: a, b, c & d). The mean global Mallet score improved from 11.29 ± 1.06 SD (range from 10 to 13) preoperatively to 17.38 ± 2.14 SD (from 13 to 20) postoperatively. No significant difference was found in Mallet score between patients with partial plexus injury at the level of C5-C6 and patients with partial plexus injury at the level of C5-C6-C7 roots, while there was a significant difference between patients with partial and complete plexus injury with low Mallet score for patients with complete plexus injury. There have been no intraoperative or postoperative complications.

Table 1: Mallet score for shoulder function.⁸

.

Table 2: Demographic data of all patients.

Table 3: Clinical data of all patients.

Fig. 4: a, b, c & d- postoperative ROM