Reimbursement, efficiency, and safety in rotator cuff repair: a comparative study of arthroscopic and open approaches
Article information
Abstract
Background
Rotator cuff repair (RCR) is one of the most common orthopedic procedures in the United States. Although arthroscopic and open RCR yield similar long-term outcomes, differences in reimbursement and short-term safety are less understood. This study aimed to evaluate differences between RCR techniques in work relative value units (wRVUs), operative times, Medicare reimbursement amounts, and 30-day complication rates.
Methods
For data collection, the National Surgical Quality Improvement Program database was retrospectively analyzed (2006–2023). Patients undergoing arthroscopic or open RCR were identified by Current Procedural Terminology codes and matched 1:1 via propensity score matching according to demographics and comorbidities. Outcomes included operative time, wRVUs, Medicare reimbursement amount, and 30-day postoperative complications.
Results
A total of 18,988 matched patients (9,494 with arthroscopic RCR, 9,494 with open RCR) were included. Arthroscopic RCR cases had significantly longer operative times than open RCR cases (91.82±43.71 minutes vs. 80.93±41.02 minutes, P<0.001) but generated 17.7% higher wRVUs per minute (0.20±0.09 vs. 0.17±0.07, P<0.001) and 36.0% more revenue per case ($485.20±$0.00 vs. $356.77±$7.54, P<0.001). Arthroscopic RCR patients had significantly lower overall 30-day complication rates (1.1% vs. 1.5%, P=0.010), including fewer superficial (0.2% vs. 0.4%, P=0.023) and deep surgical site infections (0.0% vs. 0.2%, P=0.004).
Conclusions
Arthroscopic RCR is associated with lower short-term complication rates than open RCR, despite requiring longer operative times and more resources. This clinical trade-off suggests that the additional operative demands of arthroscopic surgery are justified by the associated improvements in patient safety, highlighting the value of reduced complications as a key advantage of the arthroscopic approach.
Level of evidence
III.
INTRODUCTION
Rotator cuff tears are a common cause of shoulder pain, especially among older adults [1]. Most cases are degenerative, atraumatic injuries that can be asymptomatic; however, when symptoms arise, they can cause reduced active range of motion; significant pain, especially with overhead activities; and functional limitations in activities of daily living [1]. The management of rotator cuff pathology is typically conservative at first, primarily comprising physical therapy and nonsteroidal anti-inflammatory drugs [2,3]. For patients who do not respond to nonoperative management, however, surgical repair is considered [4,5]. Approximately 250,000 rotator cuff repairs (RCRs) are performed each year in the United States, with the vast majority being conducted arthroscopically, according to recent estimates [6-8]. Studies have demonstrated that arthroscopic RCR carries lower rates of short-term postoperative complications and better short-term recovery of range of motion and pain compared to open RCR, although the approaches have similar long-term outcomes [3,9-11].
While prior studies have compared clinical outcomes between arthroscopic and open RCR, few to date have investigated differences in surgical efficiency, physician work effort, and reimbursement. Work relative value units (wRVUs) are a standardized metric used in the United States to measure physician labor for a particular service or procedure and are determined by a combination of factors, including complexity of the task, time, technical skill, and physician effort [12,13]. wRVUs directly impact physician compensation and practice economics, as they determine reimbursement under Medicare and private insurers [14].
The purpose of this study was to evaluate and compare operative times, wRVU generation, reimbursement rates, and short-term postoperative complication rates between arthroscopic and open RCR using a large cohort from a robust national database. Balancing surgical time, financial value, and clinical outcomes is critical for optimizing surgical decision-making, resource allocation, and clinical reimbursement structures. This study contributes a new analysis of the relative efficiency, safety, and cost of arthroscopic versus open RCR.
METHODS
This retrospective study sourced information from a publicly available national database without identifying patient details and thus did not require institutional review board approval and informed consent.
Data Source
The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was queried for cases recorded during 2006–2023. NSQIP is a multi-institution program in the United States that collects de-identified patient demographics, surgical case profiles, and 30-day postoperative outcomes from more than 700 hospitals. A data use agreement was submitted to facilitate access to this dataset. Procedures were selected using Current Procedural Terminology (CPT) codes.
Patient Selection
Patients who underwent either primary arthroscopic RCR (CPT code 29827) or open RCR (CPT codes 23410 and 23412) were selected from NSQIP. Patients with missing demographics and comorbidity data were excluded (n=17,133). In line with previous studies, cases with an operative time of either <30 minutes or >500 minutes (n=19,202) were also excluded, as such recorded operating times were likely data errors [15]. Demographic variables included age, sex, height, and weight; body mass index was calculated using the following formula: (weight in pounds)/(height in inches)2×703. The following comorbidities were also considered: smoking status, diabetes mellitus, American Society of Anesthesiologists classification, preoperative functional status, hypertension requiring medication, history of severe chronic obstructive pulmonary disease, ascites, heart failure in the 30 days before surgery, disseminated cancer, dialysis use, ventilator use, bleeding disorder, and preoperative packed red blood cell transfusion <72 hours before surgery.
Operative time and length of stay were included, as were 30-day postoperative outcomes of interest. Minor complications were defined as superficial surgical site infection, urinary tract infection, deep vein thrombosis, or bleeding requiring transfusion. Major complications included ventilator use for more than 48 hours, unplanned intubation, pneumonia, myocardial infarction, stroke, progressive renal insufficiency or acute renal failure requiring dialysis, pulmonary embolism, deep surgical site infection, wound dehiscence, sepsis, and septic shock.
wRVUs were obtained for each patient from the NSQIP variable “WORKRVU.” wRVUs per minute were calculated by dividing the “WORKRVU” variable by the operative time for each case. To determine the reimbursement rate, wRVUs were multiplied by the U.S. Centers for Medicare and Medicaid Services (CMS) conversion factor for January–December 2025 ($32.3465), yielding the dollar amount received per minute of work. Revenue per case was calculated by multiplying each case’s wRVU value by the CMS conversion factor, reflecting the total payment received for the completed procedure. To ensure contemporary relevance, all dollar values were calculated using the January–December 2025 CMS conversion factor ($32.3465), regardless of the year of surgery. Of note, this approach affects only the absolute payment values and does not alter the subsequent relative comparisons between the arthroscopic and open groups.
Statistical Analyses
Prior to matching, univariate statistics (two-sided independent-samples t-test, chi-square test, and Fisher exact test) were used to compare demographics and comorbidities between the arthroscopic RCR and open RCR groups. Subsequently, 1:1 nearest-neighbor propensity score matching without replacement was conducted using the MatchIt R package (R Foundation for Statistical Computing). The propensity score was estimated through logistic regression, controlling for baseline characteristics (age, sex, body mass index, smoking status, diabetes mellitus, American Society of Anesthesiologists classification, preoperative functional status, hypertension requiring medication, history of severe chronic obstructive pulmonary disease, ascites, heart failure in the 30 days before surgery, disseminated cancer, dialysis use, ventilator use, bleeding disorder, and preoperative packed red blood cell transfusion <72 hours before surgery). Following matching, univariate statistics (two-sided independent-samples t-test, chi-square test, and Fisher exact test) were calculated to identify significant differences in demographics, comorbidities, and 30-day postoperative outcomes between the matched cohorts. Results of comparative analyses of operative time and reimbursement metrics between arthroscopic and open RCR were expressed as percent differences. All analyses were performed using RStudio version 4.4.0 (R Foundation for Statistical Computing). Statistical significance was set at P<0.05.
RESULTS
Patient Demographics and Comorbidities
Between 2006 and 2023, 64,089 patients met the inclusion criteria, with 54,595 arthroscopic RCR procedures and 9,494 open RCR procedures performed for treatment of rotator cuff tears, respectively. Baseline demographics and comorbidities are presented in Table 1. Using 1:1 propensity score matching, 9,494 arthroscopic RCR cases were matched to 9,494 open RCR cases, creating a cohort of 18,988 patients. After matching, the two cohorts did not differ significantly in any baseline characteristic (Table 2).
Baseline characteristics of patients undergoing arthroscopic or open rotator cuff repair prior to propensity score matching
Baseline characteristics of patients undergoing arthroscopic or open rotator cuff repair after propensity score matching
Comparison of wRVUs between Arthroscopic and Open RCR
On average, arthroscopic RCR cases had longer operative times than open RCR cases (91.82±43.71 minutes vs. 80.93±41.02 minutes, P<0.001). However, the mean wRVU value for arthroscopic RCR was 36% higher compared to that for open RCR (15.00±0.00 vs. 11.03±0.23, P<0.001). When standardized per minute of operative time, arthroscopic RCR delivered 17.7% more wRVUs per minute than open RCR did (0.20±0.09 vs. 0.17±0.07, P<0.001). Likewise, the reimbursement rate and total revenue per case for arthroscopic RCR were 19.1% and 36.0% higher, respectively, than those for open RCR (reimbursement: $6.42±$2.80 vs. $5.39±$2.29; revenue per case: $485.20±$0.00 vs. $356.77±$7.54; both P<0.001) (Table 3).
wRVU comparison between arthroscopic and open rotator cuff repair after propensity score matching
Complication Rates
Thirty-day postoperative outcomes after arthroscopic and open RCR are shown in Table 4. Length of stay was statistically significantly shorter for the arthroscopic RCR group than for the open RCR group (0.15±1.63 days vs. 0.34±2.13 days, P<0.001), although the clinical significance of this difference is likely small. The incidence of any overall postoperative complication was lower (1.1% vs. 1.5%, P=0.010), as was that of any minor complication (0.6% vs. 1.0%, P=0.007). This was driven by a decreased rate of superficial surgical site infections (0.2% vs. 0.4%, P=0.023) and deep surgical site infections (0.0% vs. 0.2%, P=0.004) (Table 4).
Univariate analysis of postoperative outcomes of patients undergoing arthroscopic or open rotator cuff repair after propensity score matching
DISCUSSION
wRVUs serve as the basis of physician reimbursement in the United States, with a value assigned to every service that a physician can perform [16]. Over the past two decades, reimbursements for orthopedic surgeries have consistently decreased, with those for shoulder arthroscopies experiencing the greatest decline [17,18]. Given the high use of RCR in the United States, this study sought to compare reimbursement rates between open and arthroscopic RCR. The present study’s results indicate that arthroscopic RCR is reimbursed at a higher rate and displays lower complication rates compared to open RCR.
The primary result of this study was that arthroscopic RCR was reimbursed at a higher rate than open RCR. Although arthroscopic RCR cases had longer operative times than open RCR cases, both the total wRVUs and wRVUs per minute were higher on average for arthroscopic cases than for open cases. This finding is largely consistent with conclusions in the broader orthopedic literature. In an analysis of common procedures performed by sports medicine–trained orthopedic surgeons, Simcox et al. found that arthroscopic procedures broadly received greater reimbursement amounts than open procedures [19]. Among common shoulder and elbow procedures, Nayer et al. similarly found that arthroscopic procedures were more highly valued compared to other surgeries [20]. The present study’s results add to the existing literature by demonstrating that the current Medicare Fee Schedule provides higher reimbursement amounts for arthroscopic RCR than for open RCR. This observed financial differential is consistent with existing reports and suggests that the fee schedule acknowledges the increased complexity, time, and technical skill associated with the arthroscopic approach. Orthopedic surgeons without specialized training are often less familiar with arthroscopic RCR than with open RCR [21,22]. Arthroscopy is a demanding procedure, requiring surgeons to operate in a three-dimensional environment from a small two-dimensional camera image. In addition, the restricted range of movement associated with arthroscopy’s small incisions poses another challenge.
The present study reveals an important clinical trade-off: although arthroscopic RCR requires greater operative times and resources compared to open RCR, these investments are justified by its significantly lower short-term complication rates. This trend is largely driven by the decreased rates of deep and superficial surgical site infections observed following arthroscopic cases. Again, these findings are consistent with the literature, with several multi-center studies reporting higher infection rates in patients receiving open RCR compared to those receiving arthroscopic RCR [23-25]. The present study further demonstrated that there were otherwise no differences between RCR cohorts in the 30-day incidence of mortality, urinary tract infections, deep vein thrombosis, bleeding resulting in transfusion, pulmonary complications, myocardial infarction, stroke, sepsis, or renal failure. Ultimately, this indicates that the arthroscopic approach to RCR may offer clinical advantages without compromising patient safety in other domains. These findings support the continued shift toward applying minimally invasive techniques in shoulder surgery and may inform decision-making for both surgeons and healthcare systems aiming to implement value-based care [26].
Beyond the context of US-based reimbursement, the efficiency and safety differences observed between arthroscopic and open RCR may carry important implications for healthcare systems worldwide. In diverse global settings, where reimbursement structures vary, the demonstrated reduction in complication rates with arthroscopic RCR may translate to substantial clinical and economic benefits, including reduced postoperative morbidity, shorter hospital stays, and decreased need for reinterventions. These advantages may support the broader adoption of arthroscopic techniques, especially where surgical safety outcomes are prioritized. Moreover, the use of relative metrics, such as percent differences in complication rates and efficiency, provides a framework for evaluating procedural value across varying healthcare models, making these findings relevant for international surgeons and policymakers. Nevertheless, these implications may be speculative, given that the reimbursement modeling in this study was based solely on data from hospitals in the United States.
Despite these findings, the results of this study should be considered in the light of several important limitations. First, this study only assessed the incidence of complications in the first 30 days following RCR. As such, cohort-specific differences in complication rates over a longer period of time may have been omitted. The NSQIP database does not include functional outcomes, so differences in range of motion, pain, and other patient-reported outcome measures may exist between cohorts but could not be evaluated. However, it is unlikely there are actual differences in these characteristics, as prior studies have demonstrated no significant difference in long-term functional outcomes, retear rates, or complication rates between patients receiving arthroscopic RCR and those receiving open RCR [11]. This study’s use of the NSQIP database is also a limitation, as any retrospective analysis of this database is limited by the accuracy of its coding. Nonetheless, the use of the NSQIP database provides key advantages over single-center studies, including broader generalizability across diverse patient populations and practice settings. Furthermore, while the outcomes in this study were statistically significant, the absolute difference between groups was modest, and so any associated clinical impact may be limited. Finally, the NSQIP database does not capture important details such as rotator cuff tear size, chronicity, or number of tendons involved. Open RCR is often selected for more complex or larger tears, which may influence outcomes and resource usage. As a result, these variables can substantially influence resource usage and perioperative outcomes such as operative time and risk of complications, and their absence may lead to unmeasured confounding. Similarly, surgeon experience and expertise, both of which were not recorded in the NSQIP database, may have influenced the choice of technique and postoperative results.
CONCLUSIONS
This study demonstrates notable differences in reimbursement between RCR techniques, with arthroscopic RCR generating higher wRVUs and revenue than open RCR, despite having lower short-term complication rates. These findings suggest that the current Medicare Fee Schedule may appropriately compensate the arthroscopic approach for its increased complexity and clinical advantages. Future research should investigate whether these financial and clinical advantages persist across different tear characteristics, including size and chronicity, as well as assess their impact on long-term functional outcomes and healthcare costs.
Notes
Author contributions
Conceptualization: ARZ. Data curation: ARZ. Formal analysis: ARZ. Investigation: ARZ. Methodology: ARZ. Project administration: ARZ. Resources: ARZ. Software: ARZ. Supervision: US. Validation: ARZ. Visualization: ARZ. Writing – original draft: ARZ, EM, GQC, ARZ. Writing – review & editing: ARZ, US. All authors read and agreed to the published version of the manuscript.
Conflict of interest
Outside of this work, Umasuthan Srikumaran serves as a board or committee member for American Academy of Orthopaedic Surgeons, American Shoulder and Elbow Surgeons, and International Congress of Shoulder and Elbow Surgery; reports stock or stock options from ROM3, Sonogen, and Tigon Medical; is a paid consultant for, and receives intellectual property royalties from, Fx shoulder and Tigon Medical; receives other financial or material support from Arthrex, Depuy–a Johnson & Johnson Company, and Thieme; and has been a paid presenter or speaker for, and received research support from, Fx Shoulder. The remaining authors have no disclosures to report.
Funding
None.
Data availability
Contact the corresponding author for data availability.
Acknowledgments
None.