Comorbidity-Specific Risk of Revision Following Single-Level Degenerative Spinal Fusion (2010-2023): US Cohort Study Among Over 399K Patients.

Abstract

Study design: Retrospective cohort.

Objective: To evaluate the associations of 28 preoperative comorbidities with all-cause and etiology-specific revisions (pseudarthrosis, mechanical failure, adjacent segment disease [ASD], infection) following single-level degenerative spinal fusion.

Summary of background data: Revision following spinal fusion for degenerative spinal disease remains frequent and costly. While technical factors are well studied, the extent to which preoperative comorbidities differentially predispose to specific revision etiologies is unclear.

Materials and methods: Using the PearlDiver Mariner170 claims database (2010-2023), we identified adults undergoing single-level degenerative spinal fusion and followed them for 2 years for reoperation. All-cause and etiology-specific revisions were defined by CPT/ICD codes. Preoperative comorbidities were analyzed using multivariable cox proportional hazards models adjusted for age, sex, and all comorbidities. Bonferroni correction set significance at P≤0.00036.

Results: Among 399,186 patients (mean age 59.7±13.0; 56.7% female; mean CCI 1.4±1.2), anaemia (HR 1.33), osteoporosis (HR 1.31), substance-related disorders (HR 1.25), and depression (HR 1.19) conferred the greatest risk of all-cause revision. Type 2 diabetes (HR 1.07) and essential hypertension (HR 1.14) had smaller but significant effects. Etiology-specific analyses showed that osteoporosis, anaemia, depression, and substance-related disorders were consistently associated with increased risk across pseudarthrosis, mechanical failure, and ASD. Mechanical failure was additionally associated with hypertension (HR 1.20) and diabetes (HR 1.28), which were not significant in the other revision subtypes. In contrast, ASD revisions were uniquely associated with osteoarthritis (HR 1.26). No comorbidity was significantly associated with infection-related revision. All hazard ratios were significant at P<0.001 unless otherwise specified.

Conclusion: In this national cohort of patients, osteoporosis, anaemia, depression, and substance-related disorders emerged as significant predictors of non-infectious revision within two years. Hypertension and type 2 diabetes conferred more modest risks for mechanical failure. These findings support the integration of comorbidity-informed risk stratification into preoperative planning and highlight the potential optimization to minimize revisions and enhance clinical outcomes.