Radiographic severity of knee osteoarthritis in adult spinal deformity: the effect on rod fracture after long spinal fusion in deformity correction.

Abstract

Background context: Instrumentation failure, notably rod fracture (RF), may occur even after ideal spinal deformity correction in adult spinal deformity (ASD). As RF is the major reason behind a revision surgery, various risk factors of RF are reported in literature, including pedicle subtraction osteotomy (PSO) technique itself. However, whether hip and knee joint diseases serve as another risk factor for RF remains unexplored.

Purpose: To investigate the effects of lower-extremity joint diseases on RF in ASD patients who underwent deformity correction through long-segment fixation with PSO.

Study design: Retrospective study.

Patient sample: About 96 consecutive ASD patients (mean age, 71.2 years) who underwent deformity correction through PSO and long-segment fixation from T10 to S1 between 2008 and 2019 were included. Those with radiographic data less than 2 years were excluded.

Outcome measures: RF was confirmed on radiograph, computed tomography (CT) and bone scan or SPECT images. Coronal and sagittal spinopelvic parameters, lower-extremity osteoarthritis (OA) parameters and clinical outcomes (Oswestry Disability Index [ODI] and Visual Analog Scale [VAS]) were analyzed.

Methods: Radiographic measurements were performed at preoperative, postoperative, and last follow-up visits. Patients were divided into 2 groups: non-RF (n=60) and RF (n=36). Radiographic factors were assessed, correlations between parameters were examined, and multivariate logistic regression analysis was performed to evaluate risk factors for RF.

Results: Structural and functional leg length discrepancies (LLDs), pelvic obliquity (PO), and the differences of mechanical axis (MAD) between both extremities had significantly differed between the groups (p<.05). The between-knee differences in the extent of OA, clinically significant MAD and PO, and the incidences of both structural and functional LLD were significantly greater in the RF group than in the non-RF group (p<.05). On correlation analysis, PO was not correlated with coronal radiographic parameters but was correlated with structural and functional LLDs and MAD (p<.05). Logistic analysis revealed that the difference in OA grade between the knee joints, functional LLD, and preoperative lower extremity surgery were significantly associated with RF (p<.05).

Conclusion: ASD in the elderly often presents alongside degenerative changes in the lower-extremities, and even with ideal spinal deformity correction, RF may still occur if preexisting joint pathologies are not resolved. In the deformity correction of ASD, it is essential that we not only pursue the restoration of spinopelvic harmony, but also apply perioperative measures for lower-extremity degenerative joint diseases.