Factors affecting healing of rotator cuff repairs: microfracture of the greater tuberosity
Article information
Abstract
Background
This study aimed to investigate the impact of microfractures generated within the footprint of the greater tuberosity (GT) on postoperative cuff healing following arthroscopic rotator cuff repair (ARCR).
Methods
A retrospective analysis was conducted on patients who underwent ARCR for full-thickness rotator cuff tear (FTRCT) between April 2020 and October 2023 at our institution. A total of 73 patients was categorized into two groups based on the presence of microfractures: a microfracture group (group M, n=33) and a non-microfracture group (group N, n=40). Six months post-surgery, magnetic resonance imaging was performed to assess cuff healing and retear rates between the two groups. Furthermore, patients were stratified into retear and healing groups based on cuff integrity to analyze the factors influencing retear.
Results
There was no significant difference in retear rates between groups M and N (18.2% vs. 10.0%, P=0.332). Among demographic factors, age showed a significant difference between the retear and healing groups (67.4±8.5 vs. 61.6±6.1, P=0.044). ML tear size (3.1±1.7 vs. 2.0±1.1, P=0.015), AP tear size (2.4±1.2 vs. 1.6±1.0, P=0.332), FI of the supraspinatus (2.3±1.3 vs. 1.4±1.0, P=0.029), and FI of the infraspinatus (1.6±1.3 vs. 0.9±0.8, P=0.015) exhibited significant differences between the retear and healing groups.
Conclusions
ARCR with concurrent microfracture of the GT footprint did not significantly impact cuff healing in patients with FTRCT. However, older age and larger ML tear size were associated with an increased risk of retear.
Level of evidence
III.
INTRODUCTION
Rotator cuff tears are a primary etiology of shoulder pain [1]. In cases where conservative treatment fails to ameliorate symptoms, surgical intervention is imperative for patients afflicted with full-thickness rotator cuff tears (FTRCTs) to mend the disrupted rotator cuff tendon [1]. Depending on the extent of the tear, open rotator cuff repair may be warranted; however, contemporary advancements in arthroscopic techniques have rendered arthroscopic repair the predominant approach [2,3].
Despite progressive strides in surgical methodologies, the reported success rate of rotator cuff tear repair is approximately 80%. For extensive and massive tears, this rate is as low as 66% [4,5]. Furthermore, longitudinal studies have linked retears with hastened progression of arthritis, diminished functional outcomes, and earlier onset of pain compared to healed tears, precipitating various endeavors to mitigate retear incidence [4]. Collagen or platelet-rich plasma injections [6-8] in microfracture of the humeral greater tuberosity footprint during arthroscopic repair is an antiquated approach aimed at curtailing retear frequency [9-11]. Microfracture of the bone has been postulated to instigate the release of growth factors and mesenchymal stem cells from bone perforations, fostering the regeneration of tendon-to-bone connective tissue [10].
Previous studies have suggested that the adjunctive use of microfractures can reduce retear incidence post-rotator cuff repair from 45.2% to 22.2% [12]. Nevertheless, contradictory findings exist, with some studies reporting no discernible impact of microfracture on retear rates or postoperative functional outcomes based on Constant, University of California at Los Angeles, and visual analog scale scores [13,14]. Hence, the efficacy of this procedure remains undetermined. Consequently, we designed the present study to ascertain whether creation of microfractures in the footprint of the glenohumeral head concurrent with arthroscopic rotator cuff repair (ARCR) notably decreases the incidence of retears.
METHODS
This study received approval from the Institutional Review Board of Kyung Hee University Hospital (No. 2024-02-026) at Gangdong, Seoul, and informed consent was obtained from all participants. Between April 2020 and October 2023, a cohort of 183 patients who underwent arthroscopic repair for rotator cuff tears underwent investigation, with subsequent retrospective analysis conducted on 73 patients who had completed a minimum 1-year follow-up period. Inclusion criterion was diagnosis of a FTRCT exhibiting no symptomatic alleviation following a minimum 3-month course of conservative treatment. Exclusion criteria were lack of magnetic resonance imaging (MRI) 1 year post-surgery or partial repairs insufficient to fully encompass the footprint.
Patient Information
Age, sex, dominant hand, body mass index, bone mineral density (BMD), smoking status, alcohol consumption, occupational activity, shoulder stiffness, diabetes mellitus, and hyperlipidemia were evaluated preoperatively. BMD was assessed using dual-energy x-ray absorptiometry, which partitioned the humeral head into three equally wide columns and quantified the BMD of the humeral greater tuberosity within a 1-cm2 region of interest centered on the outermost column [15]. Occupational activity was categorized as high for strenuous manual labor, medium for manual labor with reduced physical demands (e.g., catering, homemakers, and sales personnel), and low for predominantly sedentary work or unemployment [16]. Patients with diabetes mellitus exhibiting a glycosylated hemoglobin (HbA1c) level exceeding 7.0 were classified with poorly controlled diabetes [17].
Imaging Evaluation and Intraoperative Factors
A preoperative MRI was conducted to quantify the retraction distance (cm) of the ruptured tendon both mediolaterally and anteriorly for tear size determination. Evaluation of fatty degeneration in the rotator cuff muscles slated for suture repair was performed following the Goutallier classification for the subscapularis, supraspinatus, infraspinatus, and teres minor muscles, at the juncture of the scapular pole and scapular body in a Y-shaped configuration on preoperative MRI [18]. Critical shoulder angle and acromiohumeral interval were gauged on true anteroposterior (AP) X-rays. Critical shoulder angle was measured as the angle created by the line connecting the superior and inferior edges of the glenoid and the line extending from the inferior edge of the glenoid to the outermost border of the acromion [19]. Acromiohumeral interval was measured as the shortest distance from the inferior cortex of the acromion to the humeral head [20].
In instances of intraoperative biceps brachii long head rupture and medial subluxation, either tenotomy or tenodesis was executed based on the age of the patient [21]. The degree of subscapularis (SBS) tear was classified according to the Yoo and Rhee Subscapularis Tendon Tear Classification into non-tear, type ≤2A with tendon detachment <50% of the first facet, and type >2A with tendon detachment >50% of the first facet [22]. Rotator cuff repair was performed using the single-row repair method or suture bridge method [21].
Confirmation of the presence or absence of rotator cuff tears was performed on MRI during the outpatient follow-up at 1-year post-surgery. Retears were classified using the Sugaya classification as type IV (discontinuity of the rotator cuff on one or two images in both the sagittal and coronal planes of the MRI) or V (discontinuity of the rotator cuff on three or more images in both the sagittal and coronal planes of the MRI) [23].
Surgical Technique
All surgeries were conducted employing a standardized protocol by a senior author under general anesthesia, with the patient in a beach chair position. Procedures included subacromial decompression and anterior release of the coracoacromial ligament. Microfracture of the greater tuberosity footprint of the humerus was achieved using a perforating awl before suturing the rotator cuff. Microfractures were precisely placed in the greater tuberosity footprint of the humerus using a perforating awl, ensuring a minimum distance of 3–4 mm from the anchor entry point. Each perforation was 2–4 mm deep and spaced 3–4 mm from its neighbor (Fig. 1) [24]. Following microfracture, tendon repair was conducted utilizing either a single-row repair or suture bridge technique depending on the size and morphology of the rupture.
Arthroscopic view from the posterosuperior portal at the right shoulder with the patient in the beach chair position. (A) Supraspinatus cuff tear, showing the footprint after debridement. (B) Microfractures were performed every 3–4 mm of distance to a depth of 2–4 mm on the exposed footprint. (C) Bleeding and fat droplets from the subchondral bone through the microfracture were visible.
Postoperative Rehabilitation
Postoperative rehabilitation was consistent for all patients. Following surgery, patients wore an abduction brace for six weeks and avoided active shoulder exercises. On the afternoon of surgery, patients commenced active flexion exercises for the finger and wrist joints, alongside passive flexion exercises for the elbow joint, starting with finger-pinching exercises. Passive anterior elevation range of motion exercises for the shoulder joint were initiated at 6 weeks postoperatively, followed by passive external rotation range of motion exercises at eight weeks postoperatively.
Statistical Analysis
Statistical analyses of the study data utilized the independent t-test, chi-square test, or Fisher's exact test to compare patient and radiologic factors between the microfracture and retear groups. Pearson's chi-square and Fisher's exact tests were employed to assess the integrity of repaired rotator cuffs. Multivariate logistic regression analysis was conducted using variables that exhibited significance in the univariate analysis and to explore risk factors for rotator cuff tears. Statistical significance was determined when the p-value was less than 0.05, with an associated 95% CI. Analysis was performed using IBM SPSS Statistics version 24.0 (IBM Corp.). An a priori analysis performed using G*Power (version 3.1.9.7) indicated that the required total sample size was 58 patients; thus, the sample size in this study (n=73) was sufficient to yield a power >0.80.
RESULTS
Patient Information and Imaging Assessment in Group M vs. Group N
In groups M and N, mean age was 63.5±6.9 and 61.5±6.6 years, respectively (P=0.202). There were no significant differences in demographic data between groups M and N. Preoperative rotator cuff tear sizes were 2.4±1.3 cm and 2.0±1.3 cm for size mediolateral (ML), measurements and 1.8±1.0 cm and 1.6±1.1 cm for AP measurements in groups M and N, respectively. These differences were not significant (P=0.228 and P=0.484, respectively). Additionally, no significant differences were observed in other radiologic parameters, such as fatty infiltration (FI), biceps surgery, or subscapularis muscle tears, between the groups (Table 1). Mean time from surgery to follow-up MRI in groups M and N was 13.0±1.4 and 12.4±2.6 months, respectively (P=0.263). Retears occurred in 6 of 33 cases (18.2%) in group M and 4 of 40 cases (10%) in group N, which was not a significant difference (P=0.332) (Table 2).
Patient demographic and preoperative radiologic factors in group M vs. group N
Postoperative RCT integrity in group M vs. group N
Factors Affecting Retear vs. Healing
Mean age of patients was 67.4±8.5 and 61.6±6.1 years in the retear and healing groups, respectively, which was a significant difference (P=0.044). Furthermore, preoperative rotator cuff tear size significantly varied between the retear and healing groups, with ML sizes of 3.1±1.7 cm and 2.0±1.1 cm, respectively (P=0.015), and AP sizes of 2.4±1.2 cm and 1.6±1.0 cm (P=0.040). A significant difference in FI existed between the retear and healing groups, with values of 2.3±1.3 and 1.4±1.0 in the SS region (P=0.029) and 1.6±1.3 and 0.9±0.8 in the IS region, respectively (P=0.015). No significant differences were observed in other radiologic or intraoperative factors between the retear and healing groups (Table 3).
Patient demographics and postoperative radiologic factors in retear vs. healing groups
Independent Risk Factors for Retear
When subjected to multivariate logistic regression analysis, patient age (odds ratio [OR], 1.153; 95% CI, 1.026–1.295; P=0.016) and size of the ML rotator cuff tear (OR, 1.988; 95% CI, 1.103–3.582; P=0.022) were significant risk factors for retear (Table 4).
Independent risk factors for retear
DISCUSSION
This study assessed the impact of microfractures on tendon healing following ARCR in patients with FTRCTs. No significant discrepancy in retear rates between the microfracture and no-microfracture groups was observed. Among patient factors, advanced age and ML tear size were correlated with elevated risk of retear.
Research into measures to mitigate retear occurrences post-rotator cuff repair is ongoing. Intraoperative injections of atelocollagen and hyaluronic acid into the periphery of the greater tuberosity of the humerus to foster rotator cuff tear healing, platelet-rich plasma injections [8-10], and microfracture procedures at the footprint of the greater tuberosity of the humerus [7,11,12] are some approaches currently being evaluated. Microfractures are hypothesized to stimulate the release of growth factors and mesenchymal stem cells from the perforated bone area, facilitating the regeneration of tendon-to-bone connective tissue [12]. This technique is relatively easy to execute and adds minimal cost and time to the surgery.
In a study comprising 123 patients diagnosed with FTRCTs who underwent MRI follow-up over 2 years, Anil et al. observed retears in 6 of 44 patients (13.6%) who underwent microfracture and 27 of 79 patients (34.1%) who did not, suggesting an association between microfracture and reduced incidence of rotator cuff retears [25]. Ruiz Ibán et al. [26] reported similar findings in their investigation involving 69 patients with rotator cuff tears, wherein retears were documented in 19.4% of patients who received microfracture treatment and 42.4% of those who did not, supporting the premise that microfractures contribute to a diminished rate of rotator cuff retears. Conversely, Osti et al. [24] based on evaluation of 57 patients with FTRCTs followed for over 2 years reported conflicting findings. Although microfracture initially provided pain relief and functional enhancement, it failed to demonstrate significant functional or imaging benefits at the 2-year mark. Toro et al. [27], in a study involving 123 rotator cuff tear patients with MRI follow-up at 6 months postoperatively, reported retear rates of 4.8% and 11.4% in patients with and without microfractures, respectively, with no notable difference between the two groups. Hence, there remains no consensus regarding the efficacy of concurrent microfractures during rotator cuff repair based on the existing literature. In the current investigation, we focused on patients with FTRCTs; MRI follow-up at approximately 1 year post-repair revealed retear rates of 18.2% in patients with microfractures and 10% in those without microfractures, which was not a significant difference.
Risk factors for retear after rotator cuff repair are diverse. Prior research has identified patient-related factors including BMD, age, smoking, and diabetes mellitus as being correlated with postoperative cuff healing [17,28-30]. Radiological factors such as the supraspinatus occupation ratio, infraspinatus FI, initial tear size, and supraspinatus FI have also been linked to postoperative cuff healing [31,32]. In the current study, age and ML tear size were identified as retear risk factors, consistent with previous findings. However, smoking, osteoporosis, diabetes mellitus, supraspinatus, and infraspinatus FI had no impact on the retear rate, despite being recognized risk factors. Further investigations encompassing a broader array of patient and radiological risk factors in a larger cohort are warranted.
Certain limitations of this study should be noted. First, this was a retrospective study that excluded patients who did not undergo postoperative MRI and those who were not followed for at least 1 year after surgery, potentially introducing selection bias. Second, the sample size was small, with only 73 cases. As previously mentioned, a prospective, randomized controlled study with a larger patient cohort is necessary. Despite these limitations, we found no significant difference in retear rates according to the presence or absence of microfractures in patients with FTRCTs who underwent ARCR. We performed a post hoc power analysis of the retear rate and showed significantly higher retear rate in the microfracture group more than non-microfracture group, with an effect size of 0.768, power of 0.723, and alpha error of 0.05. Previous studies have reported that microfracture reduces the retear rate. While the effectiveness of this procedure has not been confirmed and there is no consensus, the present study differs from previous studies because the retear rate was higher with microfracture than without. If microfracturing is ineffective and increases the retear rate, it should be avoided, especially as it adds time to the surgery, suggesting the need for further research.
CONCLUSIONS
In patients with FTRCT, simultaneous microfracture procedures targeting the greater tuberosity footprint, when performed alongside ARCR, did not have a significant impact on cuff healing outcomes. However, advanced age and increased ML tear dimensions exhibited a positive correlation with heightened susceptibility to retear.
Notes
Author contributions
Conceptualization: MSK. Data curation: MSK. Methodology: MSK. Project administration: MSK. Visualization: GKS. Writing – original draft: GKS. Writing – review & editing: MSK.
Conflict of interest
None.
Funding
None.
Data availability
Contact the corresponding author for data availability.
Acknowledgments
None.