1 Introduction

Rheumatoid arthritis (RA) is a systemic autoimmune disease characterized by chronic inflammation that primarily affects the joints but can also produce extra-articular manifestations.¹ The exact etiology is unknown but is suspected to be multifactorial—involving both genetic and environmental interactions.^1,2 In the United States, the prevalence of the disease is estimated to range between 0.54 and 0.63 % of the population^3,4 and globally, the prevalence has been increasing since 1990.⁵ RA classically presents as a symmetrical polyarthritis that produces pain, swelling, and stiffness in the small joints of the hands and feet, initially.⁶ Without treatment, patients can develop erosion of the joints and deformity, leading to severe disability.^2,6 Diagnosis is made clinically and is supported by serological and radiographic studies; care must be taken to exclude other rheumatic pathologies.⁶ Treatment involves a combination of nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and disease-modifying antirheumatic drugs (DMARDs⁾.⁷ Though the pathological inflammatory state enabled by RA causes bone loss through a variety of mechanisms, the use of corticosteroids for treatment compounds this issue by independently decreasing bone mineral density.^6,8,9

After the joints of the hands and feet, the shoulder joint is the next most likely site of symptom onset of early RA, presenting in 18–21 % of patients.^10,11 As the disease progresses, 31–39 % of patients will eventually develop arthritic pain in the joint.^11,12 Shoulder arthroplasty is a common orthopaedic procedure used to treat intractable arthritic pain resulting from a range of etiologies.¹³ Annual incidence of the procedure continues to increase year-by-year, indicative of its utility in improving patients’ quality of life.^14,15 Despite shoulder arthroplasty being commonly implemented as a treatment option for patients with RA, post-operative outcomes data is not well-characterized in the literature, with the most robust studies and meta-analyses only pooling a few hundred patients at most.^16–18

The purpose of this study is to characterize the impact of RA on shoulder arthroplasty in a large commercial US population via a matched cohort analysis by focusing on the 90-day complications. We hypothesize that patients with RA will exhibit higher rates of postoperative complications after undergoing shoulder arthroplasty, as compared to their matched controls. The results of this study can inform future decision-making when counseling patients with RA undergoing total joint arthroplasty procedures.

2 Methods

3 Results

From 2010 to 2020, 33,282 patients underwent shoulder arthroplasty with 11,421 (34.3 %) having a preoperative diagnosis of RA. The matched cohorts of disease and control patients contained 11,142 patients and 22,140 patients, respectively.

Patients in the RA group had significantly higher rates of 90-day complications including infections (OR 4.4, 95 % CI 3.9–5.0, p < 0.001), ED visits (OR 3.0, 95 % CI 2.8–3.1, p < 0.001), cardiac arrythmia (OR 3.0, 95 % CI 1.2–4.0, p = 0.015), deep vein thrombosis (DVT) (OR 1.3, 95 % CI 1.0–1.7, p = 0.045), acute kidney injury (OR 2.3, 95 % CI 2.0–2.6, p < 0.001), wound disruption (OR 3.1, 95 % CI 2.4–4.0, p < 0.001), nerve injury (OR 2.1, 95 % CI 1.2–3.5, p = 0.009, pneumonia (OR 3.9, 95 % CI 3.5–4.4, p < 0.001), pulmonary embolism (OR 4.2, 95 % CI 3.3–5.4, p < 0.001), transfusion (OR 2.6, 95 % CI 2.2–3.1, p < 0.001), urinary tract infection (OR 4.5, 95 % CI 4.1–4.9, p < 0.001), sepsis (OR 3.3, 95 % CI 2.7–3.9, p < 0.001), periprosthetic infection (OR 3.1, 95 % CI 2.4–3.9, p < 0.001), and adhesive capsulitis (OR 9.0, 95 % CI 7.0–11.5, p < 0.001). Patients in the RA group were also significantly less likely to undergo subsequent manipulation under anesthesia (OR 0.7, 95 % CI 0.6–0.8, p < 0.001) (Table 1).

Patients in the RA group had no significant differences in revision arthroplasty rates compared to control patients (revision OR 0.9, 95 % CI 0.8–1.1, p < 0.05). Univariate modeling demonstrated no significant effect on revision (OR 1.1, 95 % CI 1.0–1.1, p < 0.05) or post-operative RCR (OR 1.0, 95 % CI 0.9–1.0, p < 0.05). In a multivariate analysis adjusted for age, gender, and CCI, RA was a significant predictor of all-cause complications within 90 days (OR-adj 3.4, 95 % CI 3.3–3.6, p < 0.001) (Table 2).

4 Discussion

While existing literature has investigated complications and outcomes of patients with RA after shoulder arthroplasty, our study is the first to leverage a commercial claims database that achieves patient cohort sizes on the scale of thousands. Our results align with the findings from Weinreb et al. which compared perioperative complications following shoulder arthroplasty in patients with and without rheumatoid arthritis.¹⁹ They identified blood transfusions rates (p < 0.001) and length of hospital stay (2.196 vs. 2.085 days; p < 0.001) to be significantly higher for RA patients than for healthy controls. Similarly, we also found that patients with RA were more likely to develop complications in the 90-day period after surgery, including: infection, ED visit, cardiac arrhythmia, DVT, acute kidney injury, hematoma, wound disruption, nerve injury, pneumonia, pulmonary embolism, transfusion, urinary tract infection, sepsis, periprosthetic infection, and adhesive capsulitis as compared to patients without RA (Table 1). These findings were further supported by the results of our multivariate modeling, which found that RA was a significant predictor of all-cause complications in the same 90-day period. Despite this, no significant differences were observed in rates of revision arthroplasty or RCR, suggesting that while RA increases perioperative risks, it does not appear to affect the long-term durability of the arthroplasty procedure itself. (Table 2).

A retrospective study by Sperling et al. investigated post-operative outcomes of total shoulder arthroplasty (TSA) versus hemiarthroplasty in 303 shoulders with RA that had at least 2 years of follow-up (mean was 11.6 years). In both surgical groups, a total of 27 shoulders experienced a post-operative complication (8.9 %) such as nerve injury/deficit (4.3 %), wound disruption (1.0 %), hematoma (0.3 %), and intraoperative fracture (3.3 %).¹⁸ There were a total of 28 shoulders that required revision surgery (9.2 %).¹⁸ However, it is important to note that this study did not compare post-operative arthroplasty complications between patients with and without RA, rather it compared the two surgical procedures in diseased patients to determine which operation produced fewer adverse effects.

Additionally, a systemic review performed by Cho et al. investigated reverse shoulder arthroplasty outcomes in patients with RA. Of the 123 shoulders they assessed, 27 (22 %) cases had one or more postoperative complication, 12 (44 %) of which had intraoperative or postoperative fractures and 9 (7 %) of which required one or more revision surgery. Other studies found surgical revision rates to range from 7 % to 14 % in patients with RA.^20,21 These findings are higher than our results, which identified 4.4 % (491) of cases that required revision arthroplasty and 6.7 % (752) of cases that received rotator cuff repair. Due to our population size being 36 to 92 times the size of the cohorts in the previous studies, this may explain why our complication rates were a bit lower and provide more a realistic perspective on postoperative outcomes.

Patients with rheumatoid arthritis have been identified to be at higher baseline risk for infectious diseases compared to the general population.^22,23 The use of immunosuppressive treatment in this autoimmune disease may further increase the risk of developing an infection or complication following surgical intervention.²⁴ In a retrospective study by Salt et al. diagnosis of RA (OR = 1.47; p = 0.031) was found to be a significant predictor of developing a postoperative infection following total joint arthroplasty (THA, TKA, or TSA).²⁵ In addition, individuals with postoperative joint infections were significantly more likely to be prescribed allopurinol (p = 0.002) and colchicine (p = 0.006), which can be utilized for their immunosuppressive effects respectively. However, it should be noted that no significant difference was found for the use of specific anti-rheumatic medications and TNA- α inhibitors.

Evaluating these findings through the lens of clinically significant outcomes (CSO) is critical. Minimal clinically important differences (MCID) represent the smallest changes in outcomes that patients perceive as meaningful, while substantial clinical benefit (SCB) denotes improvements that significantly enhance quality of life. Similarly, the patient acceptable symptom state (PASS) reflects the threshold of symptoms that patients find acceptable. While the results of this study demonstrate statistically significant differences, it remains uncertain whether these outcomes meet established thresholds for MCID, SCB, or PASS in RA patients undergoing shoulder arthroplasty. The observed increases in complications, such as infections or emergency department visits, may or may not translate into perceptible detriments to patient recovery or quality of life.^26–28

The large sample size inherent in our study likely amplified the detection of statistically significant differences, even when the clinical impact of these differences might be minimal. Future research should aim to establish MCID and SCB values tailored to RA patients undergoing shoulder arthroplasty. Additionally, incorporating PASS as a patient-centered metric would provide a more comprehensive understanding of the real-world implications of these findings on patient satisfaction and postoperative recovery.^1,5,18

RA's systemic inflammatory nature, compounded using immunosuppressive medications, likely contributes to the increased risks of infection, thromboembolic events, and other complications observed in this study. These findings underscore the need for careful perioperative planning and individualized risk mitigation strategies for RA patients. Optimizing medical therapy, ensuring adequate nutritional status, and employing aggressive infection prevention protocols may reduce the elevated complication rates in this vulnerable population.

There is conflicting evidence to determine the causative effect of immunosuppressants on postoperative complication rates. Some studies have determined that continued usage of these drugs may promote increased peri- and postoperative complication rates following joint arthroplasty.^29,30 Meanwhile, others have not found antirheumatic therapy or immunosuppressants to provide an independent effect on postoperative complications.^31,32 In one study that examined the continuation or discontinuation of the immunosuppressant medication Methotrexate for patients with RA within one year of elective orthopaedic surgery, and followed for postoperative outcomes, the continued use of Methotrexate treatment did not increase the risk of either infection or surgical complication.³³ Future work should seek to dispel these contradictory findings and offer a more definitive conclusion between taking immunosuppressant medications and the risk of developing postoperative complications following large joint surgery.

5 Limitations

This study has several limitations inherent to its design and data source. First, the use of the PearlDiver database, a national insurance claims database, introduces potential biases related to coding inaccuracies, missing data, and variability in billing practices. These limitations may affect the accuracy of patient identification and outcome assessment. Furthermore, the database lacks detailed clinical information, such as radiographic findings, disease severity, medication regimens, and functional scores, which could provide a more nuanced understanding of the impact of RA on postoperative outcomes.

Second, the analysis was restricted to a 90-day postoperative window, limiting the ability to assess long-term complications, functional recovery, and implant durability. While the large sample size enhances statistical power, it also increases the likelihood of detecting statistically significant differences that may not reflect clinically meaningful outcomes. The absence of patient-reported outcomes and measures of clinical significance, such as MCID, further limits the ability to determine the real-world impact of the observed complications on patient quality of life.

Lastly, the retrospective cohort design does not allow for control over confounding variables, despite the use of multivariate analysis to adjust for known factors. Variables such as adherence to postoperative care, surgical technique, and surgeon experience were not accounted for, which may have influenced outcomes. Additionally, the reliance on matching by age, gender, and comorbidities may not fully eliminate selection bias between the RA and control cohorts.

Future studies should aim to address these limitations by incorporating prospective designs, longer follow-up periods, and detailed clinical and patient-reported data. Such approaches would provide a more comprehensive evaluation of the relationship between RA and shoulder arthroplasty outcomes, enabling the development of targeted interventions to optimize care for this high-risk population.

6 Conclusion

Patients with rheumatoid arthritis undergoing shoulder arthroplasty face significantly higher risks of postoperative complications compared to matched controls, including infections, emergency department visits, and thromboembolic events. These findings underscore the systemic inflammatory nature of RA and its impact on surgical outcomes. Despite these elevated risks, revision arthroplasty and rotator cuff repair rates were comparable between RA and non-RA cohorts, indicating that RA does not compromise the long-term durability of shoulder arthroplasty. The results highlight the necessity of tailored perioperative strategies to mitigate complications, optimize patient outcomes, and ensure the effectiveness of surgical intervention in this vulnerable population. These insights contribute to the evidence base for managing RA patients undergoing shoulder arthroplasty and emphasize the importance of individualized, multidisciplinary care.

CRediT authorship contribution statement

Catherine Hand: Writing – review & editing. Camden Bohn: Writing – review & editing. Chase Gornbein: Conceptualization, Methodology. Daanish Khazi-Syed: Validation. Josh Chang: Data curation. Andrew Savoia: Investigation, Data curation. Brian Forsythe: Supervision, Project administration.

Guardian/Patients consent

The study utilized the PearlDiver Database, a national insurance claims database run through Humana Health Insurance (PearlDiver Inc, Fort Wayne, IN, USA). All data available in the database is anonymized and de-identified and was queried only using ICD-9, ICD-10, and Current Procedural Terminology (CPT) codes. Thus, no human subjects were directly involved nor any protected health information collected and no guardian/patient consent was necessary.

Ethical Statement for Solid State Ionics

Hereby, I Brian Forsythe consciously assure that for the manuscript “Outcomes and Complications of Shoulder Arthroplasty in Patients with Rheumatoid Arthritis: A Large Insurance Claims Matched Cohort Analysis.” the following is fulfilled:1)This material is the authors' own original work, which has not been previously published elsewhere.2)The paper is not currently being considered for publication elsewhere.3)The paper reflects the authors' own research and analysis in a truthful and complete manner.4)The paper properly credits the meaningful contributions of co-authors and co-researchers.5)The results are appropriately placed in the context of prior and existing research.6)All sources used are properly disclosed (correct citation). Literally copying of text must be indicated as such by using quotation marks and giving proper reference.7)All authors have been personally and actively involved in substantial work leading to the paper, and will take public responsibility for its content.The violation of the Ethical Statement rules may result in severe consequences.

To verify originality, your article may be checked by the originality detection software iThenticate. See also http://www.elsevier.com/editors/plagdetect.

I agree with the above statements and declare that this submission follows the policies of Solid State Ionics as outlined in the Guide for Authors and in the Ethical Statement.

Funding source declaration

This research was supported by internal department funds. No external funding was received for this project.