Systematic Review of All-Arthroscopic Versus Mini-Open Repair of Rotator Cuff Tears: A Meta-Analysis

The objective of this study was to compare outcomes in patients with rotator cuff tears undergoing all-arthroscopic versus mini-open rotator cuff repair. A systematic review and meta-analysis of outcomes of all-arthroscopic repair versus mini-open repair in patients with rotator cuff repair was conducted. Studies meeting the inclusion criteria were screened and included from systematic literature search for electronic databases including Medline, Embase, Cochrane CENTRAL, and CINAHL library was conducted from 1969 and 2015. A total of 18 comparative studies including 4 randomized clinical trials (RCTs) were included. Pooled results indicate that there was no difference in the functional outcomes, range of motion, visual analog scale (VAS) score, and short-form 36 (SF-36) subscales. However, Constant-Murley functional score was found to be significantly better in patients with mini-open repair. However, the results of the review should be interpreted with caution due to small size and small number of studies contributing to analysis in some of the outcomes. All-arthroscopic and mini-open repair surgical techniques for the management of rotator cuff repair are associated with similar outcomes and can be used interchangeably based on the patient and rotator tear characteristics.

functional scores (University of California Los Angeles [UCLA] shoulder score, American Shoulder and Elbow Surgeons [ASES] shoulder outcome score, Constant-Murley scores), range of motion (abduction, forward flexion, external rotation); pain visual analog scale (VAS) score, and complications (retear, adhesive capsulitis). No restriction was employed to study design, with both retrospective and prospective cohort studies were included, except for case studies.
Data collection. Bibliographic details and abstracts of all citations were retrieved through database searches.
A team of independent reviewers specialized in evidence-based medicine determined the eligibility of each publication. Citations were initially screened on the basis of title/abstract supplied with each citation by applying the defined set of eligibility criteria described above. Duplicates of citations (due to overlap in the coverage of the databases) were excluded at this stage. Full text copies were ordered for studies that potentially met the eligibility criteria. The eligibility criteria were then applied to the full-text publications, with each publication being reviewed by an independent two review process.
Data was extracted from the full-text articles of included studies using a specifically designed data extraction grid. Only one dataset per study was compiled from all publications related to that study in order to avoid duplication of data. Outcome data from eligible studies were extracted from the latest time point in all trials.  Study quality. Study quality was assessed using Newcastle-Ottawa Quality Assessment Scale for cohort studies and Cochrane Collaboration's tool for assessing risk of bias for randomized controlled trials. Statistical analysis. Results will be expressed as mean differences for continuous outcomes (standardized vs. weighted to be determined by available data); and the appropriate ratio/difference for dichotomous outcomes as determined by available data. For pooled analyses and forest plot generation, we used Comprehensive Meta-Analysis software. To test the robustness of our results, we will perform sensitivity analyses to be determined by the available data. Random effects models will be used, as will appropriate tests for heterogeneity.

Results
Identification of relevant studies. Literature search yielded 1799 studies, of which 87 potentially relevant full-text articles were identified for detailed evaluation. Following detailed screening, 18 studies evaluating arthroscopy and mini-open repair for rotator-cuff repair were included in the review (Fig. 1). The included evidence was based on comparative studies assessing clinical outcomes or providing sub-group data on outcomes of interest in patients with rotator-cuff tear.
The list of studies included in the review along with study characteristics is presented in Table 1. Of the 18 included studies, 7 studies were conducted in the USA, three in South Korea, two in Germany, and one each in Patient demographics. The summary of patient demographics for all included studies is presented in Table 1. Preoperative patient characteristics did not show any significant difference between these two groups with respect to the number of patients, gender and age.
Outcome Measurements. The results of standardized mean difference (SMD) and 95% confidence interval (CI) for each comparison were shown in Table 2. Data at the study endpoint was pooled directly without stratifying for the study period due to considerable variability in the time period of follow-up in the included studies.
Analyzable data was only reported in limited studies, which might contribute bias to our final results.         there was considerable heterogeneity in the results of Constant score and sensitivity analysis was conducted after removing Zwaal et al. 9 , and the results were still found to be significant (SMD = 0.477 95% CI 0.039, 0.915);    (Figs 2-6). Several studies, in which only mean value was reported, were not included in the analysis. However, authors in these studies gave similar results as our outcome.   A heterogeneity (p < 0.05) was found in the VAS analysis, and sensitivity analysis was performed by excluding the study by Zwaal et al. 9 . The reason might be attributed to arthroscopic development as discussed earlier (Figs 13-15).  (Figs 16 and 17).

Discussion
To our knowledge, this is most up-to-date systematic review including 18 studies, including both randomized and observational studies comparing arthroscopic repair with mini-open rotator cuff repair. Earlier conducted systematic reviews focus on specific study designs, RCTs 14 or observational study design 15 . While some other reviews include limited publications, 5 studies 16 and 12 studies 17 .
The results of our review are consistent with the previously conducted systematic reviews [14][15][16][17] , concluding that the two techniques (mini-open rotator cuff repair and arthroscopic repair) have similar outcomes and can be considered as alternative treatment options. However, the result of our meta-analysis show that the Constant-Murley score (CMS) was significantly better in the mini-open repair group compared to all arthroscopic repair. CMS is a 100-points scale composed of four subscales: pain (15 points), activities of daily living (20 points), strength (25 points) and range of motion: forward elevation, external rotation, abduction and internal rotation of the shoulder (40 points). On a 100-points scale, higher score is related to higher quality of the function 18 .  Tear size is an important factor for achieving satisfactory results, with more patients with large or massive cuff tears obtaining unsatisfactory response outcomes 19 . Zhang et al. noted that patients treated with arthroscopic group displayed better shoulder strength but a significantly higher retearing rate as compared to mini-open group at 24-month follow-up 12 . For full-thickness tears, retearing rates were 74% for the arthroscopic group and 35% for the mini-open group (p < 0.05). For partial-thickness tears, no significant difference was detected 12 . Kim et al. conclude that surgical outcomes depend upon the size of the tear, rather than the method of repair 6 . The operative time for arthroscopic repair was also significantly longer than that for mini-open repair 4 .
In a study by Verma et al., there was no difference in the outcome measure for VAS (pain) and ASES score between the intact and failed repair group, indicating that excellent symptomatic relief can be achieved regardless of tendon healing. However, significant differences existed between intact and failed repairs in the restoration of forward flexion, showing an adequate repair remains vital, if strength is to be restored 10 .
Surgical technique had an impact on return to work, with an open procedure (66% patients) being advantageous compared to arthroscopic repair (45.3%) and mini-open repair (41.6%) (p = 0.004). However, there was no significant difference in the time away from work between the groups, even if it was slightly longer for open procedures 20 .
Warner et al. tested two hypotheses in a retrospective study evaluating 21 patients with full-thickness rotator cuff tears 21 . First, that there was no difference in clinical outcome and patient satisfaction between single tendon tears repaired through mini-open repair (MOR) or arthroscopic repair (ASR) technique and second, that stiffness would be less and recovery would be faster with ASR. However, the results of the study support the first hypothesis but not the second hypothesis 21 . In a study by Chung et al. evaluating postoperative stiffness in 288 patients with full-thickness rotator cuff tears, patients who underwent mini-open repair had more stiffness compared to all-arthroscopic group at the final follow-up (p = 0.02) 22 . However, there was no significant difference postoperative stiffness, pain scores, and range of motion in the two groups, in an RCT conducted by Cho et al. 13 .
Severud et al. noted that no patients in the arthroscopic group developed fibrous ankylosis, whereas 4 patients in the mini-open group developed the condition (14%), defined as failure to achieve greater than 120° forward flexion by 12 weeks postoperatively. The lower incidence of fibrous ankylosis favors the all-arthroscopic technique. A trend for better early motion was also noted in the all-arthroscopic group 19 .
Kose et al. reported preference of mini-open repair due to its low cost and high patient satisfaction, while also providing similar results to arthroscopic surgery 7 .
No statistically significant improvement was observed at six months in SF-36 general health, role-emotional, and mental health, in a retrospective study of 65 patients treated with arthroscopic rotator cuff repair and 63 treated with mini-open rotator cuff repair 4 . Similarly, in a case-control study design to report on 52 patients treated with either technique, the SF-36 was not significantly different postoperatively between the two groups 23 . However, in a retrospective study conducted by Osti et al., evaluating the two techniques in 64 patients with rotator cuff tears less than 3 cm, postoperative assessment showed a statistically significant improvement in the self-administered SF-36 scores from the preoperative values at 6 months 8 . The differences could be due to patient selection in individual studies, further, Osti et al. compared only rotator cuff tears with similar size and similar fixation (suture anchor) 8 .
Limitations. The review includes both RCTs and retrospective studies, with more number of studies having retrospective study design. However, this may be due to the lack of RCT studies conducted in this area, as an unbiased methodology was used for study selection and inclusion irrespective of the study design. There were also differences in time to follow up postoperatively, with studies ranging from 6 months 4,5,13 to 50.6 months 23 . Further, we did not investigate the impact of tear size on the outcomes, with population consisting of patients with partial-thickness rotator cuff tears less than 3 cm and full-thickness rotator cuff tears larger than 3 cm. Both single-row and double-row fixation techniques have been widely used for rotator cuff tears. Differentiation based on the use of fixation techniques was not investigated in this review, due to limitation of evidence reporting the impact of the techniques. However, in a meta-analysis it has been found that double-row fixation technique is associated with increase in post-operative rotator cuff integrity and improved clinical outcomes, especially in patients with tears larger than 3 cm [24][25][26] . Also, arthroscopic procedures were performed during the transition from mini-open to all-arthroscopic techniques; consequently, this occurred early in the learning curve in majority of the studies.

Conclusion
In conclusion, arthroscopy repair and mini-open repair are associated with similar clinical outcomes. The choice of the operating technique depends upon the tear size and surgeon's preference. Future research should focus on tear patterns, size, degree of delamination, mobility, and outcomes from surgical repair.