PGA and PLLA Arthroscopic Bankart Repairs: Tied Score A 7-Year Prospective, Randomized, Clinical, and Radiographic Study After Arthroscopic Bankart Reconstruction Using 2 Different Types of Absorbable Tack

Background: Polygluconate-B polymer (PGA) implants absorb quickly; and although they may not cause longterm joint or bone damage, they may cause synovitis and may not last long enough to effect healing. Poly-Llactic acid polymer (PLLA) implants absorb much more slowly but may cause delayed reactions or changes years after reconstruction. Objective: To compare the clinical and radiographic results of PGA and PLLA implants for arthroscopic Bankart repairs. Design: Prospective, randomized controlled trial. Participants/Methods: 40 patients with recurrent anterior shoulder instability underwent arthroscopic Bankart repair, half with a Suretac (Smith-Nephew) PGA implant, and half with the Bionx Bankart tack (Linvatec) PLLA implant. Patients underwent clinical and radiographic assessment preoperatively to include a C-reactive protein level in the blood. Both groups underwent similar postoperative rehabilitation and guidelines for return to sports. Results: 35 of the 40 patients were available for follow-up at a mean of 80 months (range, 69-96). The groups were very similar demographically for age, sex, number of dislocations, and follow-up. The mean time from initial dislocation to repair was 34 months (range, 8-262 months). Both groups had very similar clinical exam and Rowe and Constant scores. At final follow-up, patients had very similar external rotation. Both groups had approximately 10° less on the operated side than on their contralateral side. Both groups had similar Isobex strength and similar Constant and Rowe scores. Neither group had elevated C-reactive protein blood levels at day 1, day 4, or 6 weeks postoperatively. In the PGA group, there were 3 failures, 2 dislocations, and 1 subluxation. In the PLLA group, there were 2 failures, 1 dislocation and 1 subluxation. This did not reach statistical significance. The failure rate for this cohort was 14% at 7 years. The previously reported rate for this group at 2 years was 5%. Seven patients had early complications (3 in PLLA, 4 in PGA) of pain, low-grade fever, and grinding. One PGA patient had development of a reflex sympathetic dystrophy. Two PLLA patients had a severe restriction in motion, with one developing significant radiographic degenerative changes. Radiographically, the PLLA group had a significantly greater number of visible drill holes still visible at 7 years compared with the PGA group. Both groups showed an increase of degenerative changes by radiography at 7 years. Conclusions: PLLA and PGA arthroscopic Bankart reconstructions had similar clinical results 7 years after surgery. Reviewer's Comments: This article shows no clear advantage of PLLA implants over PGA implants. The redislocation or failure rate was 14% at 7 years: very acceptable. Both implants demonstrated a similar rate of development of early synovitis. This article provides a good standard to compare newer arthroscopic fixation devices for Bankart repairs. (Reviewer-John H. Wilckens, MD).