Abstract
Introduction:
Irreparable rotator cuff tear (RCT) presents a difficult treatment challenge for the orthopaedic surgeon. Many treatment strategies with varying degrees of success have been performed over the years. One of the suggested surgical treatment options is the use of a biodegradable sub-acromial balloon spacer.
Methods:
A retrospective study of patients treated with sub-acromial balloon spacer between the years 2011 and 2016 was conducted. Mean follow-up time was 14.4 months. Patient charts were reviewed to evaluate the early clinical results and complications of sub-acromial spacer for irreparable RCTs.
Results:
The study cohort included 24 shoulders in 22 patients. The average postoperative Disability of the Arm, Shoulder and Hand score was 62.4. The average preoperative University of California at Los-Angeles Shoulder score was 10.9 and improved to 15.9 (p = 0.001). Forty-six per cent of patients were satisfied with their clinical postoperative outcome. We found moderate–strong positive correlation (r = 0.64) between preoperative range of motion (ROM) and general satisfaction. None of the postoperative radiographs showed an improvement regarding the proximal migration of the humeral head. In total, four (16.7%) patients experienced postoperative complications, and two (8.3%) patients required an additional surgery as a consequence of a postoperative complication.
Conclusion:
Our results show unsatisfactory improvement in patients with irreparable RCT treated with the sub-acromial balloon spacer. Careful patient selection with attention to preoperative ROM should be considered.
Level of Evidence:
Therapeutic level IV.
Introduction
Rotator cuff tears (RCTs) are a common cause of pain in adult population and often produce lasting symptoms such as pain and limitations in normal activities. 1 The terms ‘massive’ and ‘irreparable’ RCTs are incorrectly used interchangeably, as not all massive tears are irreparable tears. 2 Massive RCTs are classified based on their size; Cofield et al. 3 were the first to define a massive tear as one measuring >5 cm in diameter. Gerber et al. 4 defined a massive tear by the number of tendons involved (i.e., two or more). While the repair of massive tears can be technically challenging, many are repairable. 5 In contrast, irreparable RCT is a functional definition; Rockwood et al. 6 described the irreparable tear as the inability to achieve a direct repair of the native tendon to the insertion on the tuberosities, despite conventional techniques of mobilization and soft tissue release. Signs of irreparability include static superior migration of the humeral head, a narrowed or absent acromiohumeral interval and fatty infiltration affecting more than 50% of the rotator cuff musculature. 2
While the success of surgical treatment of partial, small and medium sized RCTs is well established, 7,8 massive and irreparable RCTs (IRCT) have been considered as a surgical challenge for decades. 2 Surgical procedures such as debridement with sub-acromial decompression, 9 muscle–tendon slide, 10 tuberoplasty or reverse decompression 11 and reconstruction with use of rotator cuff allografts and/or synthetic grafts 12 are a few of previously reported treatment options for irreparable RCT treatment. Recently, Mihata et al. 13 described the technique of arthroscopic superior capsular reconstruction for the treatment of irreparable RCT. Tendon transfer surgery was suggested for younger patients without arthritis. 14 –16 In the last decade, reverse shoulder arthroplasty has emerged as a treatment option in the setting of rotator cuff insufficient shoulder with or without glenohumeral arthritis. 17
A new surgical technique based on utilizing a biodegradable sub-acromial spacer has been reported by several authors. The spacer was described in 2012 by Savarese and Romeo 18 and later by Gervasi et al. 19 and Senekovic et al. 20 The surgical technique includes implanting a biodegradable spacer between the acromion and the humeral head, with the intention of restoring normal shoulder biomechanics by pushing down and recentring the humeral head against the glenoid cavity, relieving pain and increasing range of motion (ROM) and improving shoulder function by increasing the ROM before impingement. Furthermore, the use of the balloon spacer as a protective measure following rotator cuff repair was reported by Szöllösy et al. 21
However, all previous reported studies share a common significant limitation. Conflict of interest was reported in all studies; therefore, results should be interpreted cautiously.
To date, no clinical study, except those with proclaimed conflict of interests, has examined the postoperative outcomes of surgical treatment with sub-acromial balloon spacer.
The purpose of this study was to report the results of a retrospective study of the clinical and radiographic outcomes of biodegradable sub-acromial balloon spacer for IRCT.
It was hypothesized that patients following sub-acromial balloon spacer insertion would have postoperative clinical improvement of shoulder function, pain reduction and high satisfaction.
Materials and methods
Between May 2011 and February 2016, 26 shoulders in 24 consecutive patients with symptomatic irreparable RCTs were treated with arthroscopic sub-acromial balloon spacer.
Two patients in the study group were lost to follow-up 3 months after surgery and were excluded from the study.
The remaining 24 shoulders (22 patients) had an average follow-up of 14.4 months (SD = 15, range 4–56). The average age of the study group was 70.7 (SD = 7.9, range 57–83, median 71) years, and 12 (55%) patients were men. Sixty-two per cent of shoulders were right sided.
All patients had preoperative shoulder pain and ROM limitation adversely affecting their activities of daily living. Patients were treated with surgery only after failure of conservative treatment that included activity modification, non-steroidal anti-inflammatory medications, intra-articular injection of corticosteroids and physical therapy.
Surgical technique
Surgery was performed under general anaesthesia. Patients were placed in a ‘beach chair’ position. Standard arthroscopic approach to the shoulder was used.
First, debridement and bursectomy of the sub-acromial space were performed. Second, a tenotomy of the long head of biceps was performed. Finally, the biodegradable balloon spacer was introduced through the lateral port and placed in the sub-acromial space. The spacer was inflated to the maximal volume, approximately 15 cc of saline.
All procedures were performed in a single medical centre study by senior shoulder surgeons.
The study was approved by the institutional review board of Sheba Medical Center (approval no. 15-2892-15-SMC). Exclusion criteria from surgery were (1) inflammatory arthropathy and (2) glenohumeral osteoarthritis.
Clinical outcome was assessed with preoperative and postoperative University of California at Los-Angeles Shoulder score (UCLA shoulder score), 22 including general satisfaction that was determined as a yes/no question. Also, subjective postoperative evaluation of shoulder function was performed using the abbreviated Disability of the Arm, Shoulder and Hand questionnaire (quick-DASH). 23 The quick-DASH in presented on a scale from 0 being the best to 100 being the worst.
For each patient, the preoperative and postoperative forward elevation (FE), internal rotation (IR) and external rotation (ER) ROM were recorded. FE was measured in degrees of sagittal elevation from a neutral position of the arm lying next to the torso. IR was measured as the most cephalad level that the thumb could reach on the spine. 24 ER was measured in degrees from a neutral position of the forearm pointing directly forward with the elbow lying next to the torso.
Preoperative and the 3 months postoperative follow-up radiographs were analyzed for the difference of proximal humeral migration measured by the upward migration index (UMI) as described by van de Sande and Rozing. 25 The distance was measured in millimetres between the centre of the humeral head and the undersurface of the acromion divided by the radius of the humeral head. A UMI of >1.35 indicates that there is no migration; a UMI of 1.25 to 1.35 intermediate migration and a UMI <1.25 severe proximal migration. 26
Early and late postoperative complications were recorded for all the patients who underwent the arthroscopic sub-acromial balloon insertion.
A paired t-test was used to compare preoperative and last follow-up data for each parameter for each patient. Statistical significance was set at p < 0.05. Point biserial correlation coefficient test was used to calculate the correlation between general satisfaction and other parameters in patient-reported outcomes. Statistical analysis was performed with the SPSS software package, version 20.0 (SPSS, Chicago, IL, USA).
Results
Clinical outcome
The average preoperative UCLA shoulder score was 10.9 (range 6.0–16.0; SD 3.24) of a maximum score of 35. A score of 27 or more was considered as good/excellent, and a score of <27 was considered as fair/poor. 9
The average postoperative UCLA shoulder score was 15.9 (range 7.0–28.0, SD 6.87). The difference was statistically significant (p = 0.001).
In 11 of the 24 (46%) shoulders, patients reported that they were satisfied with the treatment outcome. The satisfaction parameter was included in the UCLA shoulder score and added 5 points to the total score.
Excluding patient satisfaction, the average postoperative UCLA shoulder score was 13.8 (range 6.0–23.0, SD 4.82); the improvement in the score remained significant (p = 0.004).
When analyzing the different UCLA domains, only pain had been significantly improved from 2.0 to 3.35 on average (on a scale of 10, p = 0.002).
Function, active forward flexion and strength had not been improved significantly (p = 0.09, p = 0.17 and p = 0.30, respectively). Results are summarized in Table 1.
Preoperative and postoperative UCLA score.
ULCA: University of California at Los-Angeles Shoulder.
aStatistically significant.
The average postoperative total Quick-DASH score was 62.4 (range 4.5–95.5; SD = 28),
The median score was 72.7. The median Quick-Dash score for ‘heavy housework’ (e.g., wash walls and wash floors), ‘wash your back’, ‘recreational activities’ and limitations at work was the maximum possible score of 5, which translated into ‘unable’.
Range of motion
As mentioned above, FE has improved insignificantly (p = 0.17) from 90.0° preoperatively (range 20–160°, SD 26.9°) to 106.5° postoperatively (range 20–170°, SD 29.0°). The active IR and ER ROM improvement after surgery was statistically insignificant. The average IR changed from L5 vertebra (range Thigh-T7) preoperatively to L4 vertebra postoperatively (range Thigh-T5) (p = 0.37). The average ER changed from 34.1° preoperatively (range 7–60°, SD 11.75°) to 37.5° postoperatively (range 7–60°, SD 9.22°) (p = 0.48).
Correlations
We examined the correlation between satisfaction and different parameters of function. The calculated correlation between the total Quick-DASH score and general satisfaction was r = −0.61 (negative correlation—the lower the quick-DASH, the higher the satisfaction), which indicates a moderate–strong 27 correlation between the two parameters.
When isolating the parameter from the Quick-Dash score, it correlated to satisfaction by r = −0.71, which indicates strong correlation.
Examining the different parameters of the Quick-DASH score, ‘Shoulder pain’, ‘Do heavy household chores’ and ‘Open a tight or new jar’ had the strongest correlation with satisfaction (r = −0.71, r = −0.63 and r = −0.7, respectively). The rest of the Quick-DASH parameters had weak correlation to satisfaction when isolated. The correlation coefficients between the Quick-DASH score and satisfaction are summarized in Table 2.
Correlation between quick-DASH score and satisfaction.
DASH: Disability of the Arm, Shoulder and Hand.
aStrong correlation.
bModerate–strong correlation.
The calculated correlation coefficient between preoperative ROM and general satisfaction scores was r = 0.64, which indicates positive moderate–strong correlation between the two parameters. Postoperative ROM correlated similarly with general satisfaction at r = 0.61.
The calculated correlation coefficient between preoperative UCLA shoulder score and general satisfaction was r = 0.45 (weak correlation). Nonetheless, we found strong correlation between postoperative UCLA score and general satisfaction (p = 0.71).
Radiological outcome
The preoperative UMI 25 was 1.15 (range 1.00–1.33, SD 0.104). The postoperative UMI was 1.16 (range 1.00–1.33, SD 0.092). The change in proximal migration was statistically insignificant (p = 0.7).
Complications
In total, four (16.7%) patients experienced postoperative complications, of which, three (12.5%) had early complications; one of the patients (4.2%) had balloon migration 4 weeks postoperatively. The balloon remnant was bulging subcutaneously anterior to the humeral head and was surgically removed under local anaesthesia. One of the patients (4.2%) had a 6-month transient neural damage and exhibited forearm dysaesthesia in the distribution of the lateral cutaneous nerve of forearm. One of the patients (4.2%) experienced superficial wound infection which resolved with superficial wound treatment and PO antibiotics. Per os (PO) One (4.2%) patient had late deep wound infection and was treated with balloon removal 2 months after the first surgery. Cultures were negative, possibly because of prior treatment with antibiotics. The patient was treated with IV for 1 week and with PO antibiotics for two more weeks.
In total, two (8.3%) patients in the study group underwent a second surgery following the insertion of the balloon spacer.
Discussion
Clinical outcome in this study, as evaluated by the UCLA shoulder score, showed an improvement from 10.9 to 15.9 (p = 0.001). While this is a statistically significant result, it is a small improvement reflecting pain reduction rather than functional improvement.
UCLA shoulder score has been used to evaluate shoulder function and satisfaction for almost 3 decades. Our results indicate a statistically significant improvement in total UCLA shoulder score from 10.9 to 15.9 (p = 0.001). Nonetheless, in comparison with the literature, the improvement found in our study regarding UCLA shoulder score is discouraging.
To date, the only clinical results of the balloon spacer are presented in the article by Senekovic et al., 20 a one-arm prospective study of 20 patients, which was part of the product Conformité Européene marking application. The authors reported a 3-year postoperative follow-up period and showed significant improvement in total Constant score from 33.41 at baseline to 65.42 at the final follow-up (p < 0.0001).
Other authors reported on their results with different surgical treatments for IRCT. In a study by Burkhart et al. in 1994, 28 partial repair was performed in 14 patients. The average score on the UCLA shoulder score improved from a preoperative value of 9.8 to a postoperative value of 27.6. The author concluded that all but one patient was very satisfied with their result. In 1997, Gartsman 29 published his results of arthroscopic sub-acromial decompression. In a group of 33 patients, he performed acromioplasty and resection of the coracoacromial ligament and found significant improvement in all parameters of the UCLA score excluding FE. The total score improved from 11.5 preoperatively to 21.0 postoperatively. Fenlin et al. 11 followed 20 patients for an average of 27 months who underwent open tuberoplasty. Overall UCLA score increased from 9.3 to 27.7. Lo and Burkhart 5 in 2004 treated nine irreparable RCTs using single- and double-interval slides. UCLA score increased from 10.0 preoperatively to 28.3 postoperatively. All patients showed some improvement in active motion, strength or function. Moore et al. 30 in 2006 questioned the use of tendon allograft, due to a high failure rate on magnetic resonance imaging and results equivocal to a simple debridement. Still their cohort which consisted of 32 patients improved in the mean UCLA score from 12.1 to 26.1. In a recent long-term follow-up after tuberoplasty or reverse arthroscopic decompression,(31) Park et al. 31 (31)(8,31)(31)(31)(31) (31)followed up 16 patients for 8 years in average and found that the mean UCLA score improved from 10.3 to 27.2. The results are summarized in Table 3.(31)
Comparison of preoperative and postoperative UCLA shoulder scores.
ULCA: University of California at Los-Angeles Shoulder.
Reviewing the results in the literature from similar sized cohorts reveals that though the average preoperative results are similar to our study (10.8 vs. 10.9), the average postoperative results have improved considerably more compared to our cohort (25.5 vs. 15.9).
As mentioned above, the average postoperative Quick DASH scores were 62.4 (of 100 being worst). There is no established cut-off for Quick-DASH score, still, our Quick-DASH results are inferior to other results in the literature. In a study by Debeer and De Smet, 32 26 shoulders with an irreparable posterosuperior RCT were assessed preoperatively and after latissimus dorsi transfer. The mean postoperative Quick-DASH score was 31.7. In a prospective, randomized study by Berth et al. 33 involving 42 patients treated with either arthroscopic partial repair or arthroscopic debridement, the postoperative DASH score had no significant difference between the two study groups (23.8 in the partial repair group and 35.3 in the debridement group), but both scores have significantly improved from the preoperative score (64.6 and 69.5).
Our data showed that the correlation between the functional score, represented by the total quick-DASH score, and general satisfaction was moderately strong. This presumably supports the assumption that good functionally of the arm is a key to satisfaction from medical treatment in orthopaedics and in general. 34,35
Our data suggested that the component ‘pain’ had a strong correlation with satisfaction, stronger than any other parameter. Pain had been found to have strong correlation with function. 36 Pain had been distinguished as a key factor of satisfaction by Hamilton et al. 37 in study of 4709 patients following total joint replacement. Hamilton et al. concluded that achieving satisfactory pain relief is one of the three factors that broadly determine the patient’s overall satisfaction, along with meeting preoperative expectations, and a satisfactory hospital experience. Other major determinants of satisfaction after surgery are preoperative expectations 34,35 and emotional health. 38
Our data suggest that patients with better preoperative and postoperative ROM were more satisfied than the others. This phenomenon has been addressed in the literature before, 34 as the total functional outcome and not the increment in function may provide better satisfaction with surgical results. Furthermore, the postoperative UCLA shoulder score (after deducting the points for satisfaction), highly correlated with satisfaction, supports this concept.
The finding that better preoperative ROM correlates with higher rates of postoperative satisfaction may contribute to a concept that patients whose main functional impairment is pain and have somewhat preserved active ROM may benefit from the pain-relieving effect of the balloon spacer. These patients can regain strength and improve functionally with physiotherapy during the immediate and near postoperative period, in which the balloon still has its pain-relieving effect. On the contrary, in patients whose limited preoperative active ROM is their main functional limitation may not benefit from the balloon spacer, which does not address ROM directly. Our data also showed that pain improved significantly postoperatively, whereas ROM did not. Nevertheless, pain and limited active ROM are many times entwined together, making patient selection in this matter challenging.
In the 3-month follow-up radiographs, we found insignificant change in terms of proximal humeral migration. The original papers that describe the technique states that the spacer degrades within 12 months and indicate that these months are the appropriate time for shoulder rehabilitation. However, they agree that it is unclear how long the spacer remains inflated. 18 Based on the 3-month radiograph follow-up, the spacer effect is shorter, and the shoulder returns to preoperative anatomical relations before the complete degradation of the balloon spacer. This may possibly indicate that the window of opportunity for rehabilitation is narrower than the aforementioned 12 months.
A few limitations of this study should be recognized. First, this is a retrospective study, with a small cohort of patients lacking a control group. Second, the average follow-up period was 14 months, so for part of the cases only short-term follow-up data were available. Taking into account that spacer effect might diminish over time due to its biodegradable properties, the results might not capture accurately the efficacy of this treatment.
Conclusion
In this study, we investigated the clinical outcome in patients with irreparable RCT treated with a sub-acromial balloon spacer. Other than an early substantial pain relief, our results show unsatisfactory clinical improvement and merely 46% of general satisfaction. Our data suggest that patients with better preoperative ROM may be better candidates for this procedure. The lack of improvement in proximal humeral migration in the 3-month follow-up radiographs leads us to assume that the spacer effect is time limited and that early rehabilitation may be preferable. Our results show that the balloon spacer should be used with caution. Further research can shed light on who may be the right patient for this procedure and whether it provides a sustained benefit over sub-acromial debridement alone.
Footnotes
Authors’ note
All authors, their immediate family, and any research foundation with which they are affiliated did not receive any financial payments or other benefits from any commercial entity related to the subject of this article.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
