Predictive Value of Preoperative Nutritional Risk Index for Screw Loosening After Lumbar Interbody Fusion in Elderly Patients With Lumbar Spine Diseases.

Abstract

Objective: Pedicle screw loosening is one of the common complications in elderly patients undergoing transforaminal lumbar interbody fusion (TLIF) for lumbar spine disease. Malnutrition, prevalent among elderly patients, has been shown to be associated with increased complications. The Geriatric Nutritional Risk Index (GNRI) serves as a simple indicator of nutritional status. However, the relationship between malnutrition, particularly GNRI, and pedicle screw loosening has not been adequately investigated. This study aims to investigate the relationship between GNRI and pedicle screw loosening following TLIF to guide the perioperative nutritional management of patients and prevent postoperative complications.

Methods: A retrospective review was conducted on clinical data from patients who underwent single-level TLIF between 2014 and 2022. Data collection encompassed patient demographics, preoperative laboratory parameters, surgery-related data, perioperative radiographic data, and patient-reported outcomes were comprehensively documented. All patients were followed up for a minimum of 12 months. The relationship between GNRI and pedicle screw loosening was evaluated by univariate and multivariate Cox regression analysis, restricted cubic spline (RCS) analysis, receiver operating characteristic (ROC) analysis, and Kaplan-Meier survival analysis.

Results: A total of 426 patients were included in the study. The rate of pedicle screw loosening rate was 16.4% at a minimum follow-up of 12 months. Patients with pedicle screw loosening exhibited significantly lower GNRI (89.0 ± 8.0 vs. 99.2 ± 9.3, p < 0.001) and volumetric bone mineral density measured by quantitative computed tomography (QCT-vBMD) (84.2 [interquartile range (IQR) 79.6-92.2] vs. 104.0 [IQR 88.2-126.0] mg/cm³, p < 0.001) compared with those in the non-loosening group. Multivariate Cox regression analysis identified sex (hazard ratio [HR] 1.433, 95% confidence interval [CI] 0.714-2.876, p = 0.027), age (HR 1.062, 95% CI 1.014-1.113, p = 0.012), GNRI (HR 0.841, 95% CI 0.711-0.994, p = 0.043), and QCT-vBMD (HR 0.982, 95% CI 0.967-0.997, p = 0.019) as independent risk factors for screw loosening. RCS analysis showed that GNRI was negatively correlated with screw loosening (p < 0.0001). The area under the curve (AUC) for the GNRI in predicting pedicle screw loosening was 0.794, with a cut-off value of 95.590 (sensitivity, 85.7%; specificity 65.2%). Kaplan-Meier survival analysis identified that the lower-level GNRI group exhibited a higher cumulative incidence of screw loosening (log-rank test, p < 0.0001).

Conclusion: The GNRI was an independent risk factor for postoperative screw loosening in elderly patients undergoing TLIF for lumbar spine disease. Preoperative GNRI may potentially serve as a valuable tool in predicting postoperative screw loosening in elderly patients undergoing TLIF.