Complications following thoracolumbar spinal surgery for adult spinal deformity in the setting of rheumatoid arthritis with minimum 2-year follow-up surveillance Authors: Neil V. Shah, George A. Beyer, Samuel Gedailovich, Juhayer S. Alam, Adam J. Wolfert, Benjamin Krasnyanskiy, Douglas A. Hollern, Rohan Desai, Peter G. Passias, Carl B. Paulino, Virginie Lafage, Vincent Challier, Bassel G. Diebo Published: 2026 ABSTRACT ABSTRACTStudy Design:Retrospective cohort study.Objectives:Given its high and increasing prevalence in the United States, rheumatoid arthritis (RA) has been a growing area of focus for spine surgeons. The literature regarding the long-term outcomes of RA patients undergoing adult spinal deformity (ASD) surgery is limited. Thus, the purpose of this study was to compare the postoperative complications and revisions through 2 years postoperatively between patients with and without RA undergoing ASD surgery.Methods:A retrospective analysis of the New York Statewide Planning and Research Cooperative System database was performed. Patients undergoing spinal fusion for ASD were selected for and divided into two cohorts based on RA status. Univariate analysis compared demographics, complications, length of stay, and revisions between cohorts. Multivariate logistic regression models, adjusted for age, gender, number of levels fused, and Deyo score.Results:A total of 6020 patients were identified, of which 238 (3.9%) patients had RA. Patients with RA had a higher rate of total complications (61.3% RA+ vs. 44.0% RA−, P < 0.001), surgical complications (24.4% RA+ vs. 15.6% RA−, P < 0.001), medical complications (55.9% RA+ vs. 37.2% RA−, P < 0.001), and revision (26.9% vs. 19.1, P = 0.003). RA was identified as an independent predictor of both overall complications (OR: 1.7 [95% confidence interval: 1.2–2.3]; P = 0.001) and revision (odds ratio: 1.6 [CI: 1.05–2.5]; P = 0.03).Conclusion:This study found RA to be an independent risk factor for increased complications and subsequent revision surgery after fusion surgery for ASD. Surgeons can use this data when counselling patients with RA regarding their operative risk. INTRODUCTION Adult spinal deformity (ASD) is a chronic and degenerative condition of the thoracic and thoracolumbar regions of the spine.[12] Clinical presentation among patients varies, ranging from low-back pain to neurogenic claudication, radiculopathy, and gait disturbance.[3] Overall, ASD encompasses a wide spectrum of disease, including de novo scoliosis, progressive adolescent idiopathic scoliosis, iatrogenic flatback, and hyperkyphosis.[134] While medical management may be considered for ASD, surgery is regarded as the superior option in terms of cost efficiency and effectiveness.[5678] The type and extent of surgical intervention is unique for each individual, and meticulous consideration regarding patient comorbidities is paramount. Spinal fusion is one option that has been increasing in prevalence as treatment for degenerative spinal conditions such as scoliosis, disc degeneration, and spondylolisthesis in the United States.[9] Furthermore, research has demonstrated that chronic medical conditions, such as osteoporosis and obesity, markedly impact the management in ASD and are generally associated with poorer postoperative outcomes.[3101112] Given its high and increasing prevalence in the United States, rheumatoid arthritis (RA) has been a growing area of focus for spine surgeons.[1314] This condition classically involves dysfunction of the immune system, increasing infection risk and impairing wound healing.[1516] The medications most frequently utilized to treat RA, including corticosteroids, disease-modifying anti-rheumatic drugs (DMARDs) and biologics, are effective in delaying the progression of RA yet can yield complications in the peri- and postoperative periods.[1718] Sagittal parameters measured through spine radiographs have become increasingly significant when evaluating patients with ASD and determining severity/treatment options. Specifically, pelvic incidence-lumbar lordosis mismatch and large sagittal vertical axis (SVA) have been associated with worse patient reported outcome measures.[12] Higher RA disease activity has been correlated with larger SVA and worse lower back pain.[19] This highlights a need for increased consideration of surgical management in this cohort. Considering the complex pathophysiology of RA along with the increasing prevalence of both RA and ASD in the United States, improving our understanding of risks associated with ASD surgery in patients with RA is critical to being able to mitigate postoperative complications. In addition, the literature regarding the long-term outcomes of patients with RA undergoing ASD surgery is limited to small case series or limited follow-up data. Thus, this study sought to objectively describe and compare the postoperative complications and revisions through 2 years postoperatively between patients with and without RA following surgery for ASD. METHODS Data source This study was deemed exempt by the Downstate Institutional Review Board (IRB) due to the de-identified nature of the data. The Statewide Planning and Research Cooperative System (SPARCS) IRB study number is 1847009-2. A retrospective analysis of the New York (NY) SPARCS database was performed using data from January 1, 2009 to December 31, 2011, ensuring a minimum of 2-year follow-up surveillance for all patients. The SPARCS dataset is an entity of the NY State (NYS) Department of Health, maintained with additional funding from the United States federal government.[20] This database comprises the most robust all-payer data available for NYS, with each patient receiving a unique tracking identification number. Patient selection Using International Classification of Diseases, Ninth Revision, Clinical Modification codes, adult patients (≥18 years) who were admitted for the following principal diagnoses of ASD, defined as idiopathic scoliosis (737.30, 737.32), or degenerative disc disease (737.10, 737.20, 722.52, 722.51, 724.02, 721.3, 738.4, 722.10, 756.12, 722.73, 721.42, 724.01, 721.2, 722.72, 721.41, 722.11, 724.03, 756.11) who underwent 2–3 (81.62), 4–8 (81.63), and more than or equal to 9-level spinal fusion (81.64) were identified. A minimum follow-up surveillance period of 2 years was deemed to be the threshold for inclusion. Patients were then stratified based on the presence or absence of a diagnosis of RA (code 714.0). Statistical analysis Univariate analysis compared demographics, complication rates, length of stay, and subsequent revision between patients with and without RA. The Chi-square analysis was used to compare the categorical variables and two-tailed independent t-tests were used to compare the continuous variables. Multivariate binary stepwise logistic regression was used to evaluate RA as an independent predictor of these outcomes, with age, gender number of levels fused, and Charlson/Deyo score adjusted for as covariates. All statistical analyses were performed with SPSS version 24 (IBM Corporation, Armonk, NY, USA). P < 0.05 was deemed to be statistically significance. RESULTS Demographics 6020 patients were included, 238 (3.95%) of whom had a diagnosis of RA and 5,782 (96.05%) who did not. In terms of demographics, patients with RA were older (64.8 years vs. 58.8 years, P < 0.001), and more likely to be female (81.1% vs. 52.6%, P < 0.001). Patients with RA also had a higher comorbidity burden, evidenced by a higher Charlson/Deyo score (1.45 vs. 0.65, P < 0.001). Meanwhile, there was no statistically significant difference in race distribution between the two cohorts [Table 1]. Table 1Demographic comparisons between RA+ and RA- patientsRA+RA- P Age (years)64.858.8P<0.001Charlson/Deyo Index1.450.65P<0.001Sex Male18.9%47.4%P<0.001 Female81.1%52.6%Race White87.8%80%P=1.0 Black4.6%7.3% Hispanic3.0%5.2% Asian or Pacific Islander0.4%1.2% Native American0%0.2% Other4.2%6.2%Insurance Medicare60.9%39.8%P<0.001 Medicaid2.9%6.0% Private Insurance28.6%39.5% Self-pay0%0.6% No Charge0%0% Other7.6%14.2%Fused Spinal Levels 2-30%0%P=0.446 4-890.8%89% >99.2%11.0%Total Charges Across Visits (USD)$276,230$191,994P<0.001Length of stay (days)7.356.46P=0.024 Demographic comparisons between RA+ and RA- patients Hospital related parameters Patients with RA had longer average lengths of hospital stay (7.35 days vs. 6.46 days, P = 0.024) and higher total hospital charges ($276,230 vs. $191,994, P < 0.001) than patients without RA over the follow-up surveillance period [Table 2]. Table 2Complication rates by rheumatoid arthritis status (rheumatoid arthritis + or rheumatoid arthritis -)ComplicationRA-RA+ P Nervous1.70%1.30%0.622DVT3.90%6.60%0.032PE1.50%1.30%0.810Hemorrhage2.40%6.30%<0.001Wound Disruption2.40%6.30%<0.001Device5.60%5.50%0.935Infection0.20%0.80%0.015Puncture0.70%0.00%0.182Hematoma3.40%4.20%0.520Irrigation/Debridement4.90%14.70%<0.001Any Medical Complications37.2%55.9%<0.001Any Surgical Complications13.6%24.4%<0.001Any Complications44.0%61.3%<0.001 Complication rates by rheumatoid arthritis status (rheumatoid arthritis + or rheumatoid arthritis -) Postoperative outcomes Patients with RA experienced a significantly higher rate of overall complications (61.3% vs. 44.0%, P < 0.001), surgical complications (24.4% vs. 15.6%, P < 0.001), and medical complications (55.9% vs. 37.2%, P < 0.001). Patients with RA experienced a higher rate of specific complications including need for irrigation/debridement (14.7% vs. 4.9%, P < 0.001), deep-vein thrombosis (6.6% vs. 3.9%, P = 0.032), hemorrhage (6.3% vs. 2.4%, P < 0.001), wound disruption (6.3% vs. 2.4%, P < 0.001) infection (0.80% vs. 0.20%, P = 0.015) as well an increased rate of revision surgery (26.9% vs. 19.1, P = 0.003) [Table 3]. Table 3Rate of total complications and revisions by rheumatoid arthritis status (rheumatoid arthritis + or rheumatoid arthritis -)ComplicationRA+RA- P Total Complications61.3%44.0%<0.001Revisions26.9%19.1%0.003 Rate of total complications and revisions by rheumatoid arthritis status (rheumatoid arthritis + or rheumatoid arthritis -) Predictors of adverse outcomes On multivariate analysis, RA was found to be independently associated with both overall complications (odds ratio [OR]: 1.7 95% confidence interval [CI]: 1.2–2.3]; P = 0.001), and revision spinal surgery (OR: 1.6 95% CI: 1.05–2.5]; P = 0.03). DISCUSSION Operative treatment, in particular spinal fusion, has become a mainstay in the treatment of ASD.[123] While effective, reducing complication rates has been a growing area of focus.[2122] Despite recent developments in the literature correlating medical comorbidities with poorer outcomes, the relationship between RA and thoracolumbar surgery has not been fully elucidated. This study demonstrates the elevated complications rates and an increased frequency of revision surgeries in RA patients undergoing spinal fusion procedures compared to those who do not have the disease. RA is a chronic inflammatory condition that affects nearly 1% of adults worldwide.[2324] Dysregulation of the immune system facilitates synovial tissue destruction, commonly resulting in swelling, erythema, arthralgia, and decreased range of motion.[2325] The MCP and PIP joints of the hands, the wrists and MTP joints of the feet are the most common joints affected.[232526] However, RA often targets the spine as well.[27] While cervical degeneration is a highly characteristic and prevalent feature of this disease, thoracolumbar involvement has been documented as well.[2829] When surgical intervention is warranted, such involvement may dictate the approach, technical difficulty, and overall success of the procedure.[22293031] The relationship between RA and presence of postoperative complications is multifaceted. This is due to the etiology of the disease, as well as the medications designed to slow its progression. It has been well documented that the dysfunctional immune system associated with RA results in an increased likelihood of infection.[1532] Through the same mechanism, wound healing is typically impaired.[163334] Both factors can help explain the increased rate of complications among patients with RA reported in this study. These findings differ from those expressed by Crawford et al.[35] who reported that there was no difference in outcomes when comparing patients with and without RA who underwent a corrective procedure for ASD. However, their study was limited in power, only including a total of 38 patients. This study is more generalizable and overwhelmingly consistent with other reports in the literature that have focused on RA in the setting of noncervical spinal surgery.[29363738] Bernstein et al.[37] queried the National Inpatient Sample database and reported patients with RA carried more comorbidities and experienced increased complications when compared to patients without RA following noncervical spine fusion surgery. Other smaller studies have also reported an increased risk of complications among patients with RA undergoing spinal fusion. Kang et al.[12] reported on 174 patients who underwent lumbar spinal fusion and found patients with RA experienced a higher rate of overall complications including symptomatic nonunion, implant failure, adjacent segment disease, and infection. In addition, Mesfin et al.[36] studied the outcomes of patients with RA undergoing long spinal fusions for scoliosis and reported an increase in complications. Our work finds similar conclusions and builds on the literature by including a 2-year follow-up period to study the additional complications with delayed presentations and identifying RA as an independent predictor for increased complications and revision surgery in the setting of other covariates. From a surgical standpoint, anti-rheumatic drugs have proven to be a double-edged sword. Three common used classes of medication include corticosteroids, DMARDS, and biologics.[172325] All have at least some evidence of increasing infection risk and diminishing wound healing, thus likely contributing to the findings in this study.[17182325] Further research is warranted on the risks of treated and untreated RA in the setting of spine surgery. Epidemiological studies have demonstrated a strong association between osteoporosis and RA with a diagnosis of osteoporosis being twice as likely to be found in those with RA compared to the general population. The relationship between the two is complex, but the factors such as an increase in circulating cytokines, decreased physical activity, and pharmacological side effects have been implicated.[39] Furthermore, poor bone quality associated with osteoporosis may lead to vertebral collapse and hardware erosion, negatively affecting the outcomes and necessitating subsequent revision surgery.[12] There were several limitations to the current study. This was a retrospective study that relied on ICD-9 coded data from the SPARCS database and thus was subject to inherent variability in the coding and inclusion of different diagnoses and procedures. In addition, RA and ASD symptom severity and treatment status were not specified in the dataset and could not be included in our analysis. While there were ICD-9 codes corresponding with certain fusion lengths (2–3, 4–8, and ≥9), the exact length or specific vertebrae fused could not be elucidated through the dataset. Despite these limitations, this study drew from a large statewide database with robust follow-up data, thus maximizing the ability of this data to be generalized to the larger ASD population. CONCLUSION Surgical intervention is regarded as a proven and reliable treatment for ASD. While the effects of various comorbid conditions on outcomes have been investigated, there is a paucity of literature pertaining to RA and its relationship to adverse outcomes after spinal fusion. This study demonstrates a clear association between RA and increased risks of both total complications and subsequent revision surgery. These data underscore the importance of discussing a patient’s comorbidities and medication with them and the possible risks they pose to help guide the screening, preoperative optimization and risk stratification in an effort to minimize the adverse outcomes.[40] Conflicts of interest There are no conflicts of interest.