Clinical outcomes of anchor-based versus transosseous arthroscopic rotator cuff repair: A retrospective cohort study with prospective follow-up

Introduction

Rotator cuff tears represent one of the most prevalent causes of shoulder pain and dysfunction, particularly in middle-aged and elderly populations, with a prevalence reaching up to 72% in cadaveric studies. ¹ Advances in suture-anchor technology have made arthroscopic repair the standard of care, providing advantages such as minimal soft tissue disruption and the capability to treat associated pathologies simultaneously. ² Nevertheless, despite advancements in implant and technique design, failure rates remain considerable – ranging from 12% to 40% – and anchor-related complications, including pullout, osteolysis, knot impingement, and challenges during revision surgery, continue to present significant hurdles. ¹ These issues have prompted renewed interest in all-arthroscopic, anchorless transosseous (TO) repair, which aims to replicate the biomechanically favourable TO fixation of open surgery while preserving the minimally invasive nature of modern arthroscopy. ³

Recent studies suggest that the TO technique is a safe and effective option. At mid-term follow-up, patients showed clear improvements in function and range of motion (ROM), while the complication rates were similar to those seen with standard anchored repairs. ⁴ Randomized and comparative studies, including the recent trial by Mesriga et al., have shown no significant differences in pain, functional scores (American Shoulder and Elbow Surgeons (ASES), OSS, and Constant), or tendon-healing integrity between arthroscopic TO and single-row anchor repairs, while the TO technique offers substantially lower implant costs.^5,6 An additional key factor in today's healthcare environment is cost-effectiveness. Multiple studies have shown that TO repairs can reduce implant costs by as much as 80% per case, without adding to operative time, thereby providing a clear economic advantage. In one comparative study, the mean implant cost savings exceeded US$900 per case, without compromising surgical efficiency or short-term clinical outcomes. ⁷

Although current findings are promising, robust prospective comparative studies are still lacking, particularly those that evaluate both functional outcomes and imaging-confirmed tendon healing when comparing TO and anchor-based (AB) repairs over longer follow-up periods. The present study, therefore, aims to conduct a rigorous, prospective comparative evaluation of arthroscopic TO versus AB rotator cuff repair, assessing pain (VAS), shoulder function (Oxford Shoulder Score (OSS)), and ROM at baseline, 6 months, and 12 months postoperatively.

Materials and methods

This retrospective cohort study with prospective follow-up, conducted at a high-volume tertiary referral centre for orthopaedic surgery. The study was conducted between January 2023 and May 2025. For the final analysis, only patients who underwent surgery on or before May 2024 were included, ensuring a minimum follow-up of 12 months; those with shorter postoperative follow-up durations were excluded. The study aimed to compare functional and clinical outcomes following arthroscopic rotator cuff repair using either the TO technique or the AB technique. The study protocol was approved by the Institutional Review Board, and all participants provided informed written consent after being counselled in their preferred language about the surgical procedures, associated risks, potential benefits, and their right to withdraw at any point.

All adult patients aged 40 to 75 years with symptomatic, magnetic resonance imaging (MRI)-confirmed full-thickness rotator cuff tears of varying sizes (small <2 cm, medium 2 to 3 cm, or large >3 cm) were included. Tear size was measured on preoperative T2-weighted MRI as the greatest linear extent of the full-thickness tendon defect. The classification was based on the system described by DeOrio and Cofield, ⁸ which categorizes tears according to maximal diameter as small (<2 cm), medium (2 to 3 cm), or large (>3 cm). Patients were eligible only if they had not responded to at least 3 months of supervised conservative treatment, which included physiotherapy and pharmacological pain management. Only patients with a preoperative Goutallier classification of stage 0 to 2 were included. Patients with partial or irreparable tears, cuff tear arthropathy, previous surgeries on the affected shoulder, advanced glenohumeral arthritis or osteonecrosis, irreparable tears, subscapularis tendon injury, active infection, open or revision repairs, or less than one year of postoperative follow-up were excluded.

The retrospective component involved identifying eligible patients from surgical records within the study period and collecting baseline demographic and clinical details (age, sex, tear size, preoperative imaging, and functional scores) from hospital records.

The prospective component consisted of standardized postoperative follow-up assessments at 6 and 12 months, during which pain (VAS), function (OSS), and ROM were systematically evaluated using calibrated instruments and validated scoring tools.

The sample size in this study was determined pragmatically, reflecting the number of eligible patients who underwent surgical repair either with the TO bone tunnel or double-row AB technique during the defined study period at our institution. Although no formal power calculation was performed, Dawson et al. ⁹ reported that comparative studies assessing shoulder function with the OSS typically identify detectable differences of 4 to 5 points, with standard deviations ranging from 7 to 10. Considering these benchmarks and the feasibility of consistent follow-up, a total of 90 patients were included in the final analysis: 42 patients in the TO group and 48 patients in the AB group, all of whom had complete clinical and functional assessments at 12 months.

Participants were divided into two groups based on the surgical technique employed. Group allocation was non-randomized and determined by surgeon preference and anatomical feasibility. The TO group underwent arthroscopic rotator cuff repair using a specialized bone tunnel technique, while the AB group underwent repair with suture anchors following a double-row construct. In the TO group, surgery was performed with the patient in the lateral decubitus position. A posterior viewing portal was first created to inspect the glenohumeral joint, followed by the establishment of a superomedial accessory portal. A 2.9 mm awl was used to drill TO tunnels adjacent to the rotator cuff footprint. The ArthroCuff tunnelling device was inserted through this portal and oriented laterally to produce a low-angle cortical exit with a preserved bone bridge of 15 to 20 mm. High-strength, colour-coded sutures were passed through the tunnels and the rotator cuff tendon using a horizontal mattress configuration to enhance tendon–bone contact and maximize footprint coverage. Final arthroscopic inspection ensured adequate tension and fixation.

For AB repairs, patients were similarly positioned, and a posterior portal was used to perform diagnostic arthroscopy. Subacromial bursectomy was completed, and acromioplasty was carried out in cases with type III acromion morphology. The rotator cuff footprint was debrided to a bleeding bone bed. Depending on the tear configuration, one or two double-loaded suture anchors were inserted into the medial row of the greater tuberosity. After horizontal mattress suturing, lateral-row fixation was achieved with knotless anchors to complete the double-row construct. Following surgery, all patients underwent a standardized rehabilitation protocol. Immobilization was maintained in an abduction sling for six weeks. Pendulum and passive ROM exercises were initiated thereafter, followed by active-assisted and active exercises as tolerated. Muscle strengthening was introduced after eight weeks and tailored to individual recovery trajectories under the supervision of a dedicated physiotherapy team.

Functional outcomes were systematically assessed using the OSS to evaluate shoulder function, while pain intensity was measured using the visual analogue scale (VAS). These assessments were conducted at baseline, 6 months, and 12 months postoperatively to monitor recovery and treatment efficacy over time. ROM, including forward flexion, abduction, and external rotation, was measured using a standardized goniometer, while internal rotation was assessed using vertebral level reach in the standing position. Data were collected using a pretested semi-structured questionnaire developed by the study team. A pilot run of this tool was conducted among 10 patients, whose data were excluded from the final analysis. Based on the pretesting, necessary linguistic and structural adjustments were made to improve clarity and reliability. A single research coordinator trained in shoulder scoring systems and physical examination techniques was responsible for data collection and follow-up assessments to ensure interobserver consistency and eliminate observer bias. Goniometric instruments were calibrated weekly against a standard anatomical template to ensure reliable measurements. All data were entered and managed in Microsoft Excel with password protection and double-entry validation. Prior to analysis, data were cleaned, and any inconsistencies were verified against the source documents. Statistical analysis was performed using STATA version 14.2. Continuous variables were expressed as mean ± standard deviation and compared using the independent samples t-test. Within-group differences were analyzed using repeated measures analysis of variance, with a two-tailed p-value <0.05 considered statistically significant.

Results

A total of 90 patients were included, comprising 48 AB and 42 TO repairs. Baseline characteristics, including age, sex, dominance, comorbidities, and tear size distribution, were comparable between groups (all p > 0.05; Table 1). The mean age was 56.1 ± 8.1 years, with a slight male predominance (56%), and medium-sized tears were most common.

OPEN IN VIEWER

Table 1.

Baseline characteristics of participants by repair type (n = 90).

Variable	AB (n = 48)	TO (n = 42)	p-value*
Age (years), mean ± SD	56.7 ± 8.4	55.8 ± 7.7	0.63
Sex, n (%)
Male	27 (56.3)	24 (57.1)	0.94
Female	21 (43.7)	18 (42.9)
Dominant shoulder affected, n (%)	31 (64.6)	26 (61.9)	0.78
Tear size, n (%)
Small (<2 cm)	8 (16.7)	7 (16.7)	0.99
Medium (2–3 cm)	30 (62.5)	25 (59.5)	0.78
Large (>3 cm)	10 (20.8)	10 (23.8)	0.74
Comorbidities, n (%)
Diabetes mellitus	10 (20.8)	9 (21.4)	0.94
Hypertension	12 (25.0)	9 (21.4)	0.71
Symptom duration (months), median [IQR]	7 5–9	8 6–9	0.45

AB: anchor-based; TO: transosseous; IQR: interquartile range.

*Between-group comparisons used an independent-sample t-test for continuous variables and a χ² test for categorical variables.

Both techniques produced significant improvements in pain, function, and ROM over 12 months (p < 0.001 for all within-group comparisons). VAS pain scores decreased to near-zero levels at one year, and OSSs improved from poor (≈12) to excellent (≈32) in both groups. Final ROM parameters, including flexion, abduction, and external rotation, were nearly symmetric between techniques. No statistically significant differences were detected between A and T repairs at any time point (Table 2). Neither superficial infection nor re-operations nor structural failures were observed in either cohort.

OPEN IN VIEWER

Table 2.

Longitudinal clinical outcomes of TO versus AB repair (n = 90).

Outcome	Baseline AB	Baseline TO	6 months AB	6 months TO	12 months AB	12 months TO	p-value*
VAS (pain)	6.86 ± 0.64	7.05 ± 0.79	4.02 ± 0.73	3.83 ± 0.70	0.71 ± 0.46	0.71 ± 0.46	0.14/0.15/0.95†
OSS	12.21 ± 1.48	12.45 ± 1.48	22.29 ± 1.51	22.60 ± 1.56	32.35 ± 1.44	32.67 ± 1.59	0.36/0.27/0.23†
Flexion (°)	94.24 ± 8.04	95.36 ± 9.01	139.63 ± 5.89	140.31 ± 6.31	165.09 ± 3.05	165.05 ± 2.97	0.45/0.53/0.93†
Abduction (°)	84.73 ± 8.81	85.10 ± 8.72	138.78 ± 5.70	140.10 ± 5.76	164.77 ± 2.92	164.62 ± 2.78	0.82/0.20/0.76†
External rotation (°)	25.08 ± 2.89	25.21 ± 2.54	50.10 ± 3.06	49.83 ± 2.95	64.58 ± 3.10	64.79 ± 3.09	0.79/0.62/0.71†
Complications, n (%)	–	–	-	-	–	–	0.32
Re-operations, n (%)	–	–	–	–	–	–	–

AB: anchor-based; TO: transosseous; VAS: visual analogue scale; OSS: Oxford Shoulder Score.

^†

Three p-values correspond to baseline/6 months/12 months.

Discussion

This retro-prospective observational study evaluated and compared the clinical effectiveness of two arthroscopic techniques for rotator cuff repair AB and anchorless TO in a cohort of 90 patients. Assessments of pain (VAS), shoulder function (OSS), and ROM were performed preoperatively and at 6 and 12 months postoperatively. Both techniques produced substantial and statistically significant improvements in all outcome parameters over time. At 6 months, both groups exhibited marked reductions in pain and improvements in mobility, and by 12 months, most patients demonstrated near-complete functional recovery. Importantly, there were no statistically significant differences between the AB and TO groups in VAS, OSS, or ROM at any follow-up interval, indicating that both approaches are equally effective in achieving satisfactory clinical outcomes at one year.

Bone quality is a key determinant of repair integrity, especially in middle-aged and older adults. Osteoporosis and age-related metaphyseal bone loss weaken cancellous support at the greater tuberosity, increasing anchor pull-out and reducing fixation strength. Low proximal-humerus bone mineral density has also been linked to higher re-tear rates after AB repairs. TO fixation avoids these limitations by relying on a stable cortical bone bridge, providing more predictable load-sharing in compromised bone. This advantage is directly relevant to our cohort (mean age 56 years) and supports the use of TO repair in patients with osteopenia or early osteoporosis, where anchor fixation may be less reliable. ¹⁰

Comparison with existing literature

The results of the present study align with a growing body of evidence supporting the comparable clinical efficacy of arthroscopic TO and traditional AB repair techniques. Several recent studies have demonstrated that TO repair yields equivalent improvements in pain relief, functional recovery, and tendon integrity, while offering distinct biomechanical and biological advantages. ¹¹

A recent level-I randomized clinical trial by Mesriga et al. ⁶ confirmed the absence of significant differences in postoperative pain, functional outcomes, or tendon healing between arthroscopic single-row AB and TO repairs, reinforcing that the anchorless construct is a reliable alternative for routine clinical use. Similarly, Binder et al. ¹² found no significant differences in Constant scores, subjective shoulder value, or re-tear rates between TO and suture bridge techniques, with MRI evaluations confirming equivalent structural healing. These findings support the mechanical robustness of the TO construct, which achieves footprint compression and tendon stability comparable to that of AB configurations.

In a matched cohort study, Srikumaran et al. ¹³ reported no differences in ASES scores, ROM, or re-tear incidence between TO and TO-equivalent techniques. Notably, procedural duration was also similar, suggesting that TO repair does not compromise operative efficiency when performed by experienced arthroscopic surgeons. Their single-surgeon, matched-pair design enhances internal validity and lends further credibility to the equivalence of the two methods.

Conversely, Sundar et al. ¹⁴ observed superior early functional outcomes following TO repair compared to double-row anchor repair, with significantly higher ASES and UCLA scores at 3 and 6 months, although these differences converged by 12 months. The authors attributed this transient early advantage to the absence of implant-induced local irritation and better preservation of bone architecture, which may favour more natural tendon-to-bone integration during the early healing phase. Randelli et al. ¹⁵ similarly reported faster pain relief in the immediate postoperative period following TO repair relative to single-row anchor repair, an effect linked to reduced subacromial irritation from the absence of protruding anchors. Collectively, these findings indicate that while long-term outcomes between TO and AB repairs are equivalent, TO repair may confer early postoperative comfort and facilitate more rapid rehabilitation.

A particularly elegant bilateral comparative study by Castagna et al. ¹ further strengthens this evidence. Patients underwent TO repair on one shoulder and AB repair on the contralateral side, thereby minimizing inter-individual variability. The study found no significant differences in postoperative pain or function between the two shoulders, supporting true equivalence between the techniques. This bilateral model provides unique internal control and is particularly valuable for subjective parameters such as pain and perceived function.

Despite occasional discrepancies among published studies, these variations are likely attributable to differences in tear size, tissue quality, surgical expertise, and postoperative rehabilitation protocols. Technical nuances, including tunnel configuration, suture number, and footprint coverage, may also influence biomechanical integrity and healing potential. Moreover, heterogeneity in imaging modalities and follow-up durations affects the sensitivity of re-tear detection. Nevertheless, the underlying biological rationale remains consistent across the literature: TO repair achieves uniform tendon compression across the footprint while avoiding anchor-related complications, potentially enhancing biological healing and long-term shoulder comfort.

Economic considerations are critical in shoulder surgery, particularly in high-volume centres and resource-limited settings. Suture anchors are among the major cost drivers in arthroscopic rotator cuff repair, with each additional anchor increasing procedure-related charges by approximately US$2245. Comparative studies consistently show that anchorless TO techniques reduce implant expenditure without affecting operative time or outcomes: Seidl et al. reported mean implant costs of US$678 for TO versus US$1014 for AB repairs, while another analysis demonstrated average savings of US$946.91 per TO case. Given that our study showed equivalent clinical effectiveness between techniques, these reductions in implant costs position the TO approach as a practical, value-based alternative, particularly for public hospitals and low-resource environments where cost containment is essential.^7,16

Clinical implications

Our findings confirm that arthroscopic TO and AB repairs provide comparable improvements in pain, function, and shoulder mobility, allowing technique selection to be guided by surgeon preference, intraoperative anatomy, and institutional resources rather than differences in clinical efficacy. However, TO repair offers several practical advantages: it provides more reliable fixation in osteopenia or osteoporotic bone by engaging a stable cortical bridge rather than relying on compromised cancellous support^17,18; it avoids anchor-related complications such as knot impingement, cyst formation, osteolysis, and technical difficulties during revision surgery^15,17–20; and it substantially reduces implant-related expenditure, with published studies demonstrating mean savings of approximately US$700 to US$1000 per case compared with AB repairs.^7,16,21,22 These combined mechanical, biological, and economic benefits make TO repair an attractive, value-based option, particularly for elderly patients, high-volume surgical centres, and resource-limited health systems.

Conclusion

Arthroscopic TO rotator cuff repair appears to be a clinically safe and effective alternative to traditional AB methods, offering comparable outcomes in pain relief, shoulder function, and ROM at one year. To substantiate these findings, further large-scale, randomized controlled trials with extended follow-up are warranted. Future studies should also explore the impact of tear size, tendon quality, and bone density on outcomes, and incorporate standardized imaging-based assessments to better guide surgical decision-making and optimize patient selection for the TO approach.