1 Introduction

Rheumatoid arthritis (RA), a chronic autoimmune disease affecting approximately 1% of the population, primarily targets peripheral joints but is also associated with frequent multisystem, extra-articular manifestations.¹ In the axial skeleton, chronic inflammation promotes fibrovascular tissue proliferation, bone erosion, and ligamentous laxity, predisposing up to 86% of patients with RA to cervical spine pathologies such as atlantoaxial subluxation, cranial settling, and subaxial subluxation.^1–3 These conditions can progress to cervical instability or myelopathy, often requiring surgical intervention for stabilization and neurological protection.³

Studies show that patients with RA undergoing cervical fusion face a significantly elevated risk of postoperative infection, likely due to the immunosuppressive effects of corticosteroids and, to a lesser extent, disease-modifying antirheumatic drugs (DMARDs)0.^4–8 Sakuraba et al. found that higher ASA physical status, longer fusions, and high-dose corticosteroid use were significant risk factors for complications following cervical spine fusion in patients with RA.⁹ Additionally, the high rates of non-union and implant failure observed in this population may be attributed to the underlying prevalence of osteoporosis, which is estimated at 28% and influenced by factors such as older age, female sex, low BMI, prolonged disease duration, high disease activity, and glucocorticoid use.^10,11

Large-scale studies focused specifically on cervical fusion outcomes in patients with RA remain limited. Fields et al. used a national database to examine causes of 90-day readmissions in patients with RA after cervical spine fusion. However, complications that did not result in readmission, delayed complications beyond 90 days such as pseudoarthrosis or implant failure, or the impact of biologic and other immunosuppressive therapies were not captured.¹² Another study of elderly patients with RA over the age of 65 undergoing anterior cervical discectomy and fusion (ACDF) found significantly higher rates of major medical complications, infections, and revisions, along with longer hospital stays and greater associated costs compared to matched controls.⁴ Yet, because literature reports that the mean age of surgery for this population is around 50-60 years, these findings may not reflect the broader RA surgical population and potentially lack generalizability.^12,13

Despite the fact that rates of cervical spine fusion in patients with RA have generally been decreasing in recent years due to widespread adoption and improved efficacy of DMARDs, a clearer understanding of associated surgical risks in this vulnerable population is still essential.^14,15 This study aims to characterize the incidence and spectrum of medical and surgical complications, as well as patterns of healthcare utilization, among adults with RA undergoing single-level ACDF or posterior cervical fusion (PCF) compared with non-RA controls. Additionally, a subgroup analysis was performed to determine the impact of immunomodulatory therapy (IMT) on cervical fusion outcomes in the RA cohort.

2 Methods

3 Results

Baseline demographics did not differ significantly between the RA and non-RA cohorts, including age, sex distribution, and Elixhauser Comorbidity Index (Table 1). Compared to non-RA controls, patients with RA undergoing single-level ACDF experienced significantly more medical complications 90 days postoperatively, including higher rates of arrhythmia (15.1% vs 7.0%), MI (1.1% vs 0.5%), CVA (2.5% vs 1.0%), atelectasis (3.8% vs 1.8%), respiratory failure (3.0% vs 2.0%), pleural effusion (2.4% vs 1.1%), PE (0.9% vs 0.3%), urinary retention (2.9% vs 1.4%), UTI (15.4% vs 4.8%), renal failure (4.3% vs 2.0%), sepsis (1.7% vs 0.6%), and cervical radiculopathy (35.4% vs 16.8%) (all p < 0.001) (Table 2). Likewise, higher incidences of complications were found in multivariate analysis of patients with RA after single-level PCF, including arrhythmias (19.6% vs 11.1%, p < 0.001), CVA (3.0% vs 1.4%, p = 0.01258), UTI (17.8% vs 8.0%, p < 0.001), renal failure (8.1% vs 3.3%, p < 0.001), and cervical radiculopathy (29.4% vs 13.1%, p < 0.001) (Table 3). There were no significant differences in neurologic deficits after surgery.

Within two years post-operatively, patients with RA had significantly higher rates of pseudoarthrosis, with those undergoing ACDF (3.9% vs 2.2%, p < 0.001) experiencing a lower incidence than PCF (17.8% vs 7.4%, p < 0.001). After undergoing either ACDF or PCF, patients with RA had higher rates of ED visits and inpatient readmissions than patients without RA at 30 and 90 days postoperatively (p < 0.001 for all) (Tables 2 and 3). There were no significant differences in mortality at two years between groups.

In the RA subgroup analysis, multivariate regression found that patients with RA on IMT who underwent ACDF had higher rates of arrhythmia (19.0% vs 12.5%), CVA (3.1% vs 2.0%), atelectasis (4.6% vs 2.8%), respiratory failure (3.5% vs 2.3%), pleural effusion (3.0% vs 1.6%), PE (1.3% vs 0.6%), urinary retention (3.2% vs 2.2%), UTI (20.1% vs 11.8%), renal failure (5.1% vs 3.7%), sepsis (2.3% vs 1.0%), and cervical radiculopathy (44.4% vs 30.7%) when compared to patients with RA not on IMT (all p < 0.001) (Table 4). Further, these patients who received IMT experienced higher rates of pseudarthrosis (4.5% vs 3.2%, p < 0.001) two years post-ACDF (Table 4). In contrast, patients with RA on IMT who underwent PCF only experienced a significantly higher rate of postoperative arrhythmias relative to non-IMT controls (24.4% vs 17.2%, p < 0.001) (Table 5). There was not a statistically significant difference in rates of pseudoarthrosis post-PCF, despite there being a numerically higher incidence of this complication in the IMT group (19.6% vs 14.5%, p = 0.116) (Table 5). Lastly, patients with RA on IMT who underwent either kind of cervical fusion experienced significantly higher rates of ED visits and inpatient readmissions at 30 and 90 days postoperatively (Tables 4 and 5).

4 Discussion

Patients with autoimmune conditions are at an increased risk of medical and surgical complications following major operations.¹⁶ The findings of this investigation are consistent with existing evidence that patients with RA experience higher rates of postoperative complications after cervical spine fusion compared to those without the condition.^4,9,12,17,18

Evidence consistently reports that RA is characterized by excess morbidity and mortality from cardiovascular disease.^19–22 Still, there is some discrepancy in the literature regarding whether patients with RA experience higher rates of postoperative cardiovascular complications. While some studies have found that patients with RA may not have increased cardiovascular risk in the immediate postoperative period, other studies demonstrate elevated risks of MI, mortality, and venous thromboembolism (VTE)—particularly in those treated with biologic DMARDs—within weeks to months after major surgery.^23–25 The present investigation contributes to the latter body of evidence, finding that patients with RA who underwent ACDF or PCF were at increased risk of arrhythmia, MI, CVA, and PE at 90 days postoperatively, especially those treated with IMT. These findings may be related to the chronic systemic inflammation associated with RA, which is known to accelerate atherosclerosis and promote endothelial dysfunction. Key inflammatory mediators such as TNF-α and IL-6 drive vascular injury, destabilize atherosclerotic plaques, and disrupt lipid metabolism, while persistent immune activation and long-term glucocorticoid use further impair vascular health and promote microvascular dysfunction.^19–22 In addition, the present study also found that patients with RA experience higher rates of pulmonary complications in the perioperative period. Large population-based studies of total hip and knee arthroplasty have shown that patients with RA have increased odds of perioperative pulmonary complications, including a higher likelihood of requiring mechanical ventilation and a greater incidence of postoperative pneumonia.^26,27 The increased risk of pulmonary complications may be related to the high prevalence of RA-associated lung disease, particularly interstitial lung disease (ILD), airway disease, and pleural involvement.^28–30.

The prevalence of urinary tract and renal complications observed in this investigation is also consistent with the literature. Studies have demonstrated that individuals with RA are at increased risk for serious infections, including UTIs, due to both the underlying disease and the immunosuppressive therapies commonly used in management.³¹ It is plausible that immune system dysregulation, impairing mucosal defense mechanisms and pathogen clearance, may be partly responsible for these trends.³²

One salient finding in this study was the discrepancy in the incidence of pseudoarthrosis between patients with and without RA. Fusion relies on a coordinated inflammatory, reparative, and remodeling phase.³³ One contributing factor to this altered pathway involves pro-inflammatory cytokines such as TNF-α, IL-1, and IL-6, which prolong the inflammatory phase, stimulate bone resorption through osteoclast activation, and inhibit osteoblast function, leading to an imbalance in bone remodeling.^34–37 This cytokine-driven environment delays or prevents effective bone repair, potentially contributing to the increased risk of nonunion observed in patients with RA. Another contributory factor to nonunion is the underlying osteoporosis in many patients with RA.^10,11 Poor bone quality makes hardware fixation less secure and reduces the substrate available for bony fusion. Additionally, medications commonly used to treat RA may further impair the biological environment required for successful spinal fusion. Chronic corticosteroid use is associated with increased osteoporosis risk, and DMARDs may suppress immune function and wound healing, both of which can hinder solid fusion.^5–8.

Notably, pseudoarthrosis occurred less frequently in the ACDF cohort than in the PCF cohort. One possible explanation is that ACDF avoids disruption of the posterior paraspinal musculature and ligamentous structures, thereby preserving the posterior tension band and vascular supply that support bone healing and fusion.³⁸ However, because surgical approach selection is influenced by underlying pathology and surgeon preference, these groups are not directly comparable. The observed difference should be interpreted as an association rather than a causal effect of approach. Nonetheless, further studies are necessary to elucidate the exact mechanism for these observed discrepancies and to understand approach-specific risk profiles.

The present study also found that patients with RA experience higher healthcare utilization after cervical fusion. Multiple large cohort studies have demonstrated that RA is independently associated with increased 30- and 90-day readmission rates following major surgeries, including total knee and hip arthroplasty, even after adjusting for age, sex, and comorbidities.^26,39,40 Of note, although this analysis did not directly assess this variable, George et al. reported that the risk of readmission is further increased in patients with prolonged exposure to glucocorticoids and additional comorbidities.⁴¹ While no studies have specifically evaluated postoperative ED visits, the literature consistently shows increased ED utilization among individuals with RA, supporting the observed higher incidence of ED visits in patients with RA following cervical spine fusion.^42,43

Regarding the subgroup analysis, literature has found that IMT in patients with RA has a nuanced impact on surgical outcomes. There is evidence that conventional synthetic DMARDs are generally safe to continue perioperatively, with no significant increase in postoperative infection or complications.^44,45 Biologic DMARDs, including TNF inhibitors, carry a modestly increased risk of surgical site infection and VTE; however, the absolute risk remains low.^5,25 Further, large cohort studies show that the use of IMT does not increase 30-day hospital readmission risk after major non-spine surgeries.^5,41 The present study adds nuance to this growing body of literature, finding that patients with RA on IMT generally experienced significantly higher rates of medical and surgical adverse outcomes, including cardiovascular, pulmonary, infectious, and implant-related complications, as well as higher rates of postoperative ED visits and readmissions. These results suggest that cervical spine surgery may carry a distinct risk profile for patients with RA on IMT. Further studies and shared decision-making between clinicians and patients are needed to determine whether the benefits of continuing an IMT regimen outweigh the perioperative risks of cervical fusion surgery while on treatment.

Several limitations should be noted. First, coding errors or variations in coding practices are inherent limitations in database studies that rely on claims records. According to the 2023 Medicare Fee-for-Service Improper Payment Report, coding errors were present in 0.9% of claims.⁴⁶ As demonstrated in previous studies, patient records in the PearlDiver database may not accurately represent the demographics of the national population, thus limiting the generalizability of study findings.⁴⁷ In addition to age, gender, and ECI, other possibly confounding variables were not controlled for. Moreover, RA is a heterogeneous disease, and ICD or CPT codes may not provide sufficient granularity into disease severity and manifestations. For the IMT subgroup analysis, data on specific treatment duration, dosage, medication adherence, and treatment response could not be determined from the database. Future prospective studies are warranted to follow patients with RA after cervical fusion procedures longitudinally.

Despite these limitations, this study has several important strengths. Using a large, nationally representative claims database enabled robust evaluation of uncommon postoperative events and provided sufficient power to examine outcomes after ACDF and PCF separately. In addition, the inclusion of a medication-based subgroup analysis offers clinically relevant insight into the potential association between IMT exposure and perioperative risk in the RA population. Collectively, these findings have practical value for preoperative counseling, risk stratification, and complication surveillance. They also underscore the importance of coordinated perioperative planning between spine surgeons, rheumatologists, and perioperative medicine teams to balance disease control with surgical risk on a case-by-case basis.

5 Conclusion

Adults with RA undergoing single-level ACDF or PCF experienced higher rates of 90-day medical complications, two-year surgical complications, and post-operative healthcare utilization at 30 and 90 days compared with non-RA controls. Within the RA cohort, IMT exposure was associated with a higher burden of postoperative complications and healthcare utilization. These findings underscore the importance of multidisciplinary perioperative management and shared decision-making to optimize patient selection, counsel patients on expected risks, and tailor perioperative planning in this vulnerable population.

Informed consent (patient/guardian)

Not required since not a “case report” or “clinical images” article type.

Ethics, consent, and IRB approval

An Institutional Review Board (IRB) approval or Research Ethics Committee review was not required for this study, which utilizes a de-identified database for secondary data analysis.

Author contributions

Patrizia Manziano: Writing (Original Draft), Writing (Review & Editing), Visualization; Joyce En-Hua Wang: Conceptualization, Methodology, Software, Formal Analysis, Writing (Review & Editing); Madison D. Sroufe: Writing (Review & Editing); Wendy Novicoff: Formal Analysis; Xudong Li: Writing (Review & Editing); Stephen Lockey: Conceptualization, Writing (Review & Editing), Supervision.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.