No Short-term Clinical Benefit to Bovine Collagen Implant Augmentation in Primary Rotator Cuff Repair: A Matched Retrospective Study.

Background: Bovine bioinductive collagen implants (herein, "bovine collagen implant") can be used to augment rotator cuff repair. Concern exists that these bovine collagen implants may not yield clinical benefits and may actually increase postoperative stiffness and the need for reoperation.

Questions/purposes: Among patients who underwent primary rotator cuff repair with or without a bovine collagen implant, we asked: (1) Did the proportion of patients undergoing reoperation for postoperative stiffness and inflammation differ between the bovine collagen implant and control groups? (2) Did short-term patient-reported outcomes differ between the two groups? (3) Did the proportion of patients receiving postoperative methylprednisolone prescriptions and corticosteroid injections differ between the two groups?

Methods: We performed a retrospective, matched, comparative study of patients 18 years and older with minimum 2-year follow-up who underwent primary arthroscopic repair of partial or full-thickness rotator cuff tears diagnosed by MRI. All procedures were performed by one surgeonbetween February 2016 and December 2021. During the period in question, this surgeon broadly offered the bovine collagen implant to all patients who underwent rotator cuff repair and who (1) consented to xenograft use and (2) had surgery at a facility where the bovine collagen implant was available. The bovine collagen implant was used in rotator cuff tears of all sizes per the manufacturer's instructions. A total of 312 patients were considered for this study (243 control, 69 implant). Minimum 2-year clinical follow-up data were available for 83% (201 of 243) of patients in the control group and 90% (62 of 69) of patients in the bovine collagen implant group. After we applied the exclusion criteria, 163 control and 47 implant group patients remained and were eligible for matching. Propensity score matching was conducted to balance cohorts by age, gender, race (Black, White, other), ethnicity (Hispanic, non-Hispanic), health insurance status, Area Deprivation Index, BMI, American Society of Anesthesiologists physical status classification, diabetes, smoking, rotator cuff tear size, concomitant surgical procedures, preoperative American Shoulder and Elbow Surgeons (ASES) score, Subjective Shoulder Value (SSV), VAS score for pain, and shoulder ROM. We included 141 patients (47 in the implant group and 94 in the control group) after matching. Patients were categorized according to whether they received the bovine collagen implant. Before matching, the control cohort was older (mean ± SD 57 ± 10 years versus 52 ± 11 years; p = 0.004), more likely to be White (58% versus 23%; p < 0.001), with a smaller proportion of concomitant distal clavicle excisions (43% versus 21%; p = 0.003), and a smaller proportion of "other" concomitant procedures (17% versus 6%; p = 0.011) compared with the implant cohort. After matching, the cohorts were well matched in all demographic variables. The primary study outcome was reoperation for inflammation and stiffness, defined as a failure of nonoperative treatment for a minimum of 9 months, including physical therapy, NSAIDs, at least one course of oral methylprednisolone, and at least one cortisone injection (reoperations for traumatic retears were excluded). Secondary outcomes were patient-reported outcomes (SSV, ASES score, and VAS score for pain), receipt of methylprednisolone prescriptions, and receipt of corticosteroid injections. Chi-square, Fisher exact tests, and independent-samples t-tests were used to assess relationships between treatment group and study outcomes.

Results: A greater proportion of patients in the bovine collagen implant group (9% [4 of 47]) underwent reoperation for inflammation and stiffness than in the control group (0% [0 of 94; p = 0.01]). At minimum 2-year follow-up, the cohorts did not differ by ASES score (mean ± SD 81 ± 24 implant versus 85 ±19 control; p = 0.24), SSV (79 ± 24 implant versus 85 ± 18 control; p = 0.30), or VAS score for pain (2.0 ± 2.9 implant versus 1.5 ± 2.3 control; p = 0.11). The cohorts did not differ in the proportion who received postoperative corticosteroid injections (15% implant versus 11% control; p = 0.46) or methylprednisolone prescriptions (49% implant versus 37% control; p = 0.18).

Conclusion: At minimum 2-year follow-up, patients undergoing primary arthroscopic rotator cuff repair with bovine collagen implant augmentation had a greater proportion of reoperation due to inflammation and stiffness compared with patients who did not receive the implant. Furthermore, the implant offered no benefit in patient-reported outcomes or need for postoperative corticosteroid injections or methylprednisolone prescriptions. Because of the lack of clinical benefit and potential increase in postoperative complications, we recommend against the use of these bovine collagen implants unless high-quality randomized controlled trials are able to demonstrate their clinical effectiveness, cost-effectiveness, and overall safety.

Level of evidence: Level III, therapeutic study.