To Remplissage or not to Remplissage Part 2: Recurrent Instability After Primary Stabilization Surgery (RIPSS) Risk Tool Assesses Preoperative Failure Rates in On-track Shoulders Undergoing Primary Arthroscopic Anterior Stabilization with or without Remplissage Augmentation

Abstract

Objectives: Given the high rates of recurrent instability following the initial arthroscopic Bankart repair (ABR), it is crucial to provide recommendations for appropriate surgical interventions for patients at a higher risk of experiencing recurring instability. In Part 1 of the study, it was discovered that on-track Hill-Sachs lesions (HSL) with specific prognostic factors (i.e., younger age, increased shoulder laxity, lower DTD, and 2+ preoperative instability events) were associated with a greater risk of failure. Therefore, the objective of Part 2 was to develop a risk assessment tool that takes into account significant prognostic factors for recurrent shoulder instability following primary ABR and evaluate the role of remplissage augmentation given a patient’s risk profile. Methods: We retrospectively reviewed prospectively collected data of consecutive patients aged 14-40 who underwent either ABR (arthroscopic Bankart repair) or ABR+R (ABR with remplissage) procedures between 2013 and 2021 for anterior glenohumeral instability. Preoperative magnetic resonance imaging was used to determine the values of glenoid bone loss, Hills-Sachs Interval (HSI), glenoid track (GT), and distance-to-dislocation (DTD). On-track lesions have a distance-to-dislocation (DTD) value greater than zero. In addition, within the category of on-track lesions, there is a subset known as “near-track” lesions, with a DTD ranging from 0 to 10 mm. Capsuloligamentous laxity scores were categorized according to hyperlaxity status, defined as external rotation greater than 85 degrees and/or grade 2+ posterior and inferior load-and-shift on examination under anesthesia. Recurrent shoulder instability was defined as recurrent dislocation and/or subjective subluxation postoperatively. Patients were excluded if the indexed surgery was a revision procedure, < 2-year follow-up, or glenoid bone loss (GBL) >20%. A final multivariate survival analysis was constructed using categorical prognostic factors (i.e., patient age, “near-track” status, hyperlaxity, and 2+ preoperative instability episodes) identified in Part 1, while adjusting for GBL. Although contact athlete status was not identified as a significant predictor of recurrent shoulder instability in Part 1, the final model stratified by contact athlete status given its established importance in prior literature. Multivariate hazard ratio estimates were utilized to create a risk assessment tool and correlated with patient-specific risk estimates via post-estimation analysis. Results: One-hundred-and-fifty-five patients were included for analysis (ABR: 116 | ABR+R: 39) with an average age of 21.6 ± 6.2 years and an average follow-up of 5.1 ± 2.0 years (range: 2.0 – 8.7 yrs). Patients with near-track lesions had a three-fold higher risk of recurrent instability (Hazard Ratio [HR]: 2.5, p = 0.03) compared to on-track HSL with DTD > 10mm. Individuals with evidence of hyperlaxity exhibited a five-fold increased risk (HR: 5.4, p=0.03) relative to patients without hyperlaxity. Younger patient groups had twice the risk of recurrent shoulder instability (HR: 2.2, p=0.003) compared to older adjacent groups. Patients with two or more preoperative recurrent instability episodes faced a four-times greater risk (HR: 4.2, p=0.004) for recurrent shoulder instability compared to patients who experienced a single instability event preoperatively. Lastly, patients who underwent primary ABR only had almost 10-times greater risk of recurrent shoulder instability (HR: 9.5, p=0.002) than those who underwent primary ABR+R. The Recurrent Instability after Primary Shoulder Stabilization (RIPSS) risk assessment score considers patient age, near-track status, hyperlaxity, preoperative instability episodes, and surgical technique (ABR vs. ABR+R) as important prognostic indicators, Table 2. The RIPSS score was created with risk stratifying subgroups: Low-risk (0 – 2), Moderate-risk (3 – 6), High-risk (7 – 9), and Extreme-risk (10+). Figure 1 illustrates a strong correlation between patient RIPSS scores with patient-specific hazard ratio estimates derived from the multivariate Cox regression modeling computed in “To Remplissage or not to Remplissage Part 1”. According to RIPSS score subgroups, recurrent shoulder instability rates range from 4.3% among low-risk groups to 54.6% among extreme-risk groups, Figure 1. Conclusions: Patients with a low-risk RIPSS score without ABR+R may not benefit from remplissage augmentation. Their risk for recurrent instability would remain low despite the addition of remplissage. However, patients with a high- or extreme-risk RIPSS score prior to considering ARB+R would likely benefit from remplissage supplementation. The RIPSS score may be an effective risk assessment tool that may help identify patients who may benefit from a primary Bankart repair with remplissage augmentation. The risk assessment tool developed may optimize surgical treatment and allow physicians to determine patient risk scores with and without remplissage augmentation.