Abstract
Background:
Body mass index (BMI) has been commonly studied in relation to hip and knee procedures, but the effect on shoulder instability procedures is less understood. This study aims to analyze BMI's relationship to clinical outcomes and recurrence rates after shoulder instability surgery.
Hypothesis:
Shoulder instability surgery in patients with elevated BMI will produce similar postoperative functional outcomes and recurrence rates to those with normal BMI.
Study Design:
Cohort study; Level of evidence, 3.
Methods:
This is a single-institution, retrospective study of patients who underwent operative treatment for shoulder instability and had 6 months of clinical follow-up between 2009 and 2023. Primary procedures included Bankart repair, Latarjet, distal tibial allograft, Hill-Sachs bone grafting, and remplissage. Patients were stratified into 3 cohorts by BMI: underweight or normal weight (U/NW; BMI, ≤25 kg/m2), overweight (OW; BMI, >25 to ≤30 kg/m2), and obese (BMI, >30 kg/m2). Data collected from electronic medical records included patient demographics, clinical outcomes, and functional markers.
Results:
A total of 509 patients (391 male, 118 female) underwent surgery for recurrent shoulder instability. Of the total, 210 patients were normal or underweight (23 kg/m2 [range, 21-24 kg/m2]), 184 patients were overweight (27 kg/m2 [range, 26-28 kg/m2]), and 115 patients were obese (34 kg/m2 [range, 32-38 kg/m2]). Age statistically differed across BMI cohorts at time of surgery (U/NW vs OW vs obese, 21 years [range, 18-27 years]; 24 years [range, 20-33 years]; 28 years [range, 20-35 years]; P < .001). Obese individuals had a greater depression rate, 36%, compared with 19% and 21% for U/NW and OW, respectively (P = .001). Preoperatively, the 3 cohorts statistically differed in terms of external rotation, forward elevation, and internal rotation range of motion. Obese patients had significantly lower postoperative internal rotation (T12) compared with the other cohorts (T10) (P < .001). Obese patients had an overall instability recurrence rate of 27%, compared with 16% in U/NW and 13% in OW patients (P = .33).
Conclusion:
Our cohort demonstrated no significant association between obesity and outcomes of shoulder instability surgery. While the range of motion differences between the groups were statistically significant, these differences were not clinically significant and may be attributed to body habitus. Our findings suggest that shoulder instability surgery is a safe and effective procedure regardless of BMI.
Shoulder instability is a common problem surgeons face, as the prevalence of anterior instability in the US population is estimated to be 0.08 per 1000 person-years. 4 With that said, shoulder instability remains a complex clinical and surgical challenge. Despite advancements in operative techniques, the variability in outcomes and recurrence rates emphasizes the multifactorial nature of this condition. 14 Among patient-specific factors, body mass index (BMI) has emerged as a critical yet underexplored determinant of surgical success; especially considering the growing prevalence of obesity, with >2 in 5 adults (42.4%) in the United States currently being obese, up from 30.5% in 2000. 11 Elevated BMI is associated with altered joint biomechanics, increased periarticular stress, and systemic inflammation, all of which adversely affect surgical stabilization outcomes and postoperative recovery. 8 Postoperative complications after orthopaedic procedures, such as humeral fracture repair, have been well-documented as being increased in the obese population.8,15 Interestingly, this relationship is not consistent within hand surgery, as obese and nonobese patients had similar rates of postoperative complications, indicating this trend is not consistent across all orthopaedic subspecialties. With that said, the precise relationship between BMI and the outcomes of shoulder stabilization surgery, including the risk of recurrent instability, has not been comprehensively outlined.
This study seeks to fill this gap by examining the relationship between BMI and both surgical outcomes and recurrence rates in shoulder stabilization. Through the assessment of preoperative and postoperative clinical metrics, complication rates, and instability survival of stabilization procedures, this study seeks to understand the influence of BMI on the trajectory of recovery and recurrence. We hypothesize that shoulder instability surgery will be an effective procedure allowing surgical intervention to be a viable option across all BMI cohorts.
Methods
Patient Population
This study was approved through the institutional review board at our home institution. Patients underwent surgery for shoulder instability at a single institution between 2009 and 2023. Patients were identified through an internal search using the following International Classification of Diseases, 9th and 10th Revision, codes: 719.2 and 718.3 (indicating pathologic and recurrent dislocation of the shoulder, respectively) and Current Procedural Terminology codes 23455 (open capsulorrhaphy with labral repair (eg, Bankart procedure), 29806 (arthroscopic capsulorrhaphy with labral repair), 23462 (Latarjet procedure), and 29807 (arthroscopic labral repair). Procedures were performed by 7 fellowship-trained surgeons (including R.C.R., G.L.J., J.Y.B., and G.L.C.). Inclusion criteria were patients who underwent surgery for instability, had ≥6 months of in-office clinical follow-up, were alive at the time of data collection, and had accessible medical records. Patients were stratified into 3 cohorts by BMI: underweight or normal weight (U/NW; BMI ≤25 kg/m2), overweight (OW; BMI >25 to ≤30 kg/m2), and obese (BMI >30 kg/m2). BMI was tabulated at time of index surgery. An initial screening of 1233 patients who underwent shoulder instability surgery was performed, with 509 patients meeting eligibility (Figure 1).

Inclusion criteria flowchart. BMI body mass index; OW, overweight; U/NW, underweight or normal weight.
Data Collection
A retrospective chart review was completed on the eligible patients to obtain demographic information. Provider notes were used to obtain preoperative and postoperative functional measurements, postoperative complications, and documented episodes of instability. Preoperative measurements were taken as close to surgery as possible, while postoperative measurements were taken at the date of final follow-up. Operative reports were used to quantify procedure type, glenoid bone loss, surgical approach (open vs arthroscopic), and bone graft. Anesthesia reports provided American Society of Anesthesiologists (ASA) scores and Charlson Comorbidity Index (CCI) scores.
Patient-Reported Outcomes
Patients were called between July 2024 and November 2024 to obtain patient-reported outcomes (PROs) using a volunteer phone questionnaire. The goal was to obtain long-term follow-up by assessing the patients’ current perspectives on their shoulder. The duration of long-term follow-up was determined by the date of the index procedure. After 3 separate failed attempts to gain contact, patients were deemed unable to be reached. PROs were obtained using the visual analog scale (VAS) for pain on a scale of 0 to 10, Single Assessment Numeric Evaluation (SANE), American Shoulder and Elbow Surgeons (ASES) Standardized Shoulder Assessment Form, and the Shoulder Instability–Return to Sport after Injury (SIRSI) 6 scale.
Statistical Analysis
This study utilized various statistical methods to analyze the data. The distribution of all continuous variables was assessed using the Shapiro-Wilk test for normality (Appendix Table A1). The Kruskal-Wallis test was used for comparison across the 3 study arms. Following a significant result, pairwise post hoc comparison completed using Dunn tests with Bonferroni correction was completed (Appendix Table A2). Ordinal variables were analyzed using the Wilcoxon signed-rank test. Kaplan-Meier survival curves were used to estimate instability survival probability up to 5 years, with a 95% CI to assess failure rates. Differences in survival distributions were evaluated using log-rank tests. A significance threshold of P < .001 was set for all statistical tests. Continuous variables were presented as median (IQR, 25th-75th). Ordinal and categorical data variables are presented with means. Statistical analyses were performed using Stata/SE Version 17.0 (StataCorp).
Results
Patient Population
A total of 509 patients (391 male, 118 female) who underwent surgery to address recurrent shoulder instability were included in this analysis. Overall, 210 patients were normal or underweight (23 kg/m2 [range, 21-24 kg/m2]), 184 patients were overweight (27 kg/m2 [range, 26-28 kg/m2]), and 115 patients were obese (34 kg/m2 [range, 32-38 kg/m2]) (Table 1). Age statistically differed across BMI cohorts at time of surgery (U/NW vs OW vs obese, 21 [range, 18-27] years vs 24 [range, 20-33] years vs 28 [range, 20-35] years; P < .001). U/NW individuals had a mean follow-up of 13 (range, 9-45) months, OW individuals had a mean follow-up of 19 (range, 8-61) months, and obese individuals had a mean follow-up of 17 (7-55) months (P = .11). There was no statistical difference in CCI scores between the 3 subgroups, but there was a difference in ASA scores (U/NW vs OW vs obese, 1 [range, 1-2] vs 1 [range, 1-2] vs 2 [range, 2-2]; P < .001). Obese individuals had a greater depression rate of 36%, compared with 19% and 21% for U/NW and OW, respectively (P = .001). There was no significant difference in the percentage of glenoid bone loss observed between the 3 groups (U/NW vs OW vs obese, 13 [range, 5-20] vs 13 [range, 9-20] vs 14 [range, 6-23]; P = .62).
Demographic Data Across BMI Cohorts a
Data is reported as mean and range unless otherwise indicated. Bold P values indicate statistical significance. ASA, American Society of Anesthesiologists; BMI, body mass index; F, female; M, male.
The study cohort underwent a large variety of surgical procedures to address recurrent instability (Table 2), including arthroscopic Bankart and labral repairs, open Bankart, open Latarjet, and distal tibial autograft. A few patients underwent open reduction and internal fixation because of glenoid fracture. Several patients underwent multiple procedures at the same time.
Procedure Types for Recurrent Instability a
Data are presented as n. Some patients were classified into multiple arthroscopic labral repair categories based on location and extent of lesion. BMI, body mass index; DTA, distal tibial autograft; ORIF, open reduction and internal fixation; SLAP, superior labrum anterior-posterior.
Clinical Evaluation
All patients were evaluated for preoperative and postoperative range of motion (ROM) and strength measurements. There was a statistical difference in preoperative forward elevation (FE) between the 3 cohorts (U/NW vs OW vs obese, 170° [range, 160°-170°] vs 170° [range, 160°-170°] vs 160° [range, 150°-170°]; P = .001). Furthermore, there was a difference in preoperative external rotation (ER) between the U/NW patients (70° [range, 60°-70°]), OW patients (60° [range, 50°-70°]), and obese patients (60° [range, 50°-70°]) (P < .001) (Table 3). In both preoperative measurements, the obese cohort had significantly less ROM compared with the U/NW cohort (Appendix Table A2). There was a significant difference in preoperative internal rotation (IR) between the 3 cohorts (U/NW vs OW vs obese, T10 vs T11 vs T11; P = .001). Postoperatively, there was no difference between the 3 cohorts in ER and FE. Obese patients had significantly lower postoperative IR (T12) compared with U/NW (T10) and OW (T10) (P < .001). ROM improvement was quantified by calculating the difference in preoperative and postoperative ROM. There was no difference in ROM improvement within ER, IR, and FE across the cohorts. On average, all patients had 5/5 strength in pre- and postoperative FE, ER, and IR. There was no statistically significant difference in VAS pain, SANE, ASES, and SIRSI scores (Table 4).
Preoperative and Postoperative Range of Motion and Strength Measurements a
Data are presented as mean and range unless otherwise indicated. ROM for FE and ER is reported in degrees. ROM for IR is reported as highest vertebral level patient thumb reached behind back. Bold P values indicate statistical significance. BMI, body mass index; ER, external rotation ; FE, forward elevation; IR, internal rotation; ROM, range of motion.
Patient-Reported Outcomes a
Data are presented as mean and range. ASES, American Shoulder and Elbow Surgeons; BMI, body mass index; SANE, Single Assessment Numeric Evaluation; SIRSI, Shoulder Instability−Return to Sport after Injury; VAS, visual analog scale.
Instability Survival and Postoperative Complications
U/NW individuals had an overall instability recurrence rate of 16%, with recurrence-free rates projected at 80.3% and 64.2% at 2 and 5 years, respectively (Table 5). OW individuals had an overall instability recurrence rate of 13%. They had recurrence-free rates of 84.3% and 74.2% at 2 and 5 years. Obese patients had an overall instability recurrence rate of 27%. At 2 and 5 years, the recurrence-free rate for obese patients was 81.0% and 59.9%. There was no statistical significance in the overall instability recurrence-free rates between the 3 cohorts (P = .33) (Figure 2).
Instability Survival Rates Across Different Time Points a
Data are presented as percentages. Number at risk denotes patients still at risk for recurrence (eg, event-free at time point and not censored/lost to follow-up). BMI, body mass index.

Instability recurrence-free survival Kaplan-Meier curve. OW, overweight; U/NW, underweight or normal weight.
Discussion
Our findings suggest that although BMI influenced preoperative measures, such as ER, IR, and FE, it was not significantly associated with recurrence rates or postoperative ROM improvements after shoulder stabilization surgery. Notably, obese patients demonstrated reduced postoperative IR compared with their U/NW and OW counterparts, but no significant differences were observed in other postoperative motion parameters. This suggests that the mechanical challenges posed by obesity may disproportionately affect specific aspects of shoulder movement. Additionally, all cohorts demonstrated a broad spectrum of ROM improvement outcomes, spanning negative changes, no improvement, and positive gains. As a result of recurrent instability, patients may trade mobility for stability through surgical intervention. Thus, achieving optimal joint stability within this cohort may involve a reduction in ROM to restore proper biomechanics.
PROs, including VAS, SANE, ASES, and SIRSI scores, were consistent across all BMI groups, contrary to similar postoperative trends seen in hip arthroplasty. 13 Despite these similar functional outcomes, obese individuals exhibited higher ASA scores and an increased prevalence of depression.
Our study explored a sample cohort in which 36.1% of the population was OW and 22.6% of the population was obese. While these figures are slightly lower than the reported prevalence of OW and obesity in the general US population—30.7% and 42.4%, 11 respectively—they align more closely with the demographic typically undergoing shoulder instability surgery. 4 These procedures are most common in younger populations, for whom the obesity rate is reported at 35.5% among individuals aged 20 to 39. 3 The slight underrepresentation of obesity in our sample is likely attributable to the cohort's younger mean age, as BMI generally increases with age. 3 Nonetheless, we believe our study population is a reasonable representation of those undergoing surgical intervention for shoulder instability.
The absence of significant differences in recurrence rates among BMI groups suggests that modern shoulder stabilization techniques can effectively address the biomechanical challenges associated with obesity. These findings reinforce the importance of surgical precision and individualized perioperative care, which appear capable of offsetting the potentially detrimental effects of elevated BMI. It is important to note that the number of patients available for long-term follow-up decreased over time; this loss to follow-up may have influenced the reported results. However, the higher systemic health burden among obese patients highlights the need for a multidisciplinary approach to care, including preoperative optimization and postoperative support tailored to this group. These results also support the inclusion of patients with higher BMI in surgical pathways, provided they meet other eligibility criteria. While BMI alone should not be a determinant for surgical candidacy, additional interventions, such as weight reduction and personalized physical therapy programs, may further enhance the outcomes for these patients.
Limitations
It is important to acknowledge that a minimum of 6 months of clinical follow-up was a limitation to this study, as recurrent instability has been reported to occur around 2 years after instability surgery on average. 9 Several other limitations in this study must also be acknowledged. As a single-institution study conducted in a specific regional context, the findings may not be easily generalized to broader populations. Variations in surgeon experience, institutional protocols, and patient demographics could all influence the results. Furthermore, a wide range of procedures were used to address instability; these procedures may have a different influence on postoperative outcomes and recurrent instability. 5 We were unable to gather preoperative PROs, as they were not previously recorded. There is also a possibility of bias in PROs and follow-up, as patients with less favorable outcomes are more likely to return to clinic than those with promising outcomes. Additionally, regional differences in BMI trends, particularly with the highest obesity rates observed in the Midwest and South, 2 should be considered. While it is possible that these findings may not be broadly applicable outside the Midwest, the safety and efficacy of shoulder instability surgery suggest that the conclusions may be relevant nationwide. The age distribution of our sample population also warrants mention, as it does not fully represent the general population. However, we believe this demographic reflects the typical cohort undergoing shoulder instability procedures, which are often sports-related injuries.
BMI, while a convenient and widely used metric, has inherent limitations in assessing patient-specific risks. It does not differentiate between fat mass and lean body mass, 12 nor does it reflect variations in body composition that could affect joint stability and surgical recovery. As such, reliance on BMI alone may oversimplify the risk stratification process. Incorporating more sophisticated measures, such as fat distribution (waist-to-hip ratio 7 ) or muscle mass evaluations, could provide a deeper understanding of the role body composition plays in surgical success.
Building on this study, future research should explore the use of advanced imaging and body composition tools, such as dual-energy X-ray absorptiometry, to refine our understanding of the effect of adiposity and muscle mass on surgical outcomes. 10 Additionally, preoperative identification of shoulder instability pathology presents a significant challenge for surgeons, as physical examination and MRI have proven to be unreliable in accurately diagnosing Bankart lesions in obese patients. 1 Consequently, the observation that obese individuals exhibit reduced preoperative IR and ER may aid surgeons in more effectively diagnosing shoulder instability. It is important to note that without baseline ROM measurements before injury, the reduced ROM is not sufficient for diagnostic criteria.
Moreover, predictive models that integrate BMI with other clinical variables, such as glenoid bone loss and psychological health, could enhance preoperative planning and patient counseling. Longitudinal studies are also warranted to explore the durability of stabilization outcomes across a wider range of BMI values and over extended follow-up periods.
Conclusion
Shoulder instability surgery was a safe and effective procedure regardless of BMI classification. Although some differences were found in preoperative and postoperative ROM measurements, instability recurrence rates did not differ across BMI cohorts. These results correspond to a younger population that commonly undergoes surgery for shoulder instability, so special attention should be made for an older individual. Future directions to address the limitations of this study should focus around a multi-institutional, nationwide approach with an improved metric to quantify body composition. Overall, individuals with higher BMI should be included in surgical pathways for shoulder instability.
Footnotes
Appendix
Post Hoc Pairwise Comparisons of Statistically Significant Continuous Variables a
| Variable | Group 1 | Group 2 | P |
|---|---|---|---|
| Age at surgery, y | |||
| U/NW | OW |
|
|
| U/NW | Obese |
|
|
| OW | Obese | .15 | |
| BMI, kg/m2 | |||
| U/NW | OW |
|
|
| U/NW | Obese |
|
|
| OW | Obese |
|
|
| ASA | |||
| U/NW | OW | .95 | |
| U/NW | Obese |
|
|
| OW | Obese |
|
|
| Preoperative ER | |||
| U/NW | OW |
|
|
| U/NW | Obese |
|
|
| OW | Obese |
|
|
| Preoperative FE | |||
| U/NW | OW | .18 | |
| U/NW | Obese |
|
|
| OW | Obese |
|
Bold P values indicate statistical significance. ASA, American Society of Anesthesiologists; BMI, body mass index; ER, external rotation; FE, forward elevation; OW, overweight; U/NW, underweight or normal weight.
Final revision submitted September 30, 2025; accepted October 31, 2025.
One or more of the authors has declared the following potential conflict of interest or source of funding: R.C.R. has received nonconsulting fees from Arthrex. G.L.J. has received honoraria from the Musculoskeletal Transplant Foundation. J.Y.B. has received consulting fees from Stryker Corp and Arthrex. G.L.C. has received research funding from Smith & Nephew. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from The Ohio State University Institutional Review Board (Approval #2022H0150).
