Spine Journal - 2026-11-10 - Journal Article
Preoperative Geriatric Nutritional Risk Index Complements Vertebral Hounsfield Units for Osteoporosis-Related Risk Stratification After Lumbar Fusion Surgery.
Kinoshita Y, Tamai K, Taniwaki H, Namikawa T, Matsumura A, Kato M, Hori Y, Hoshino M, Iwamae M, Takahashi S, Suzuki A, Toyoda H, Okamura Y, Kobayashi Y, Uematsu M, Sasaki R, Suzuki M, Tsujino M, Wakasa J, Nakamura H, Terai H
Topics
Key Takeaway
Among lumbar fusion patients with low vertebral HU (≤100), a low preoperative GNRI (<98) was associated with a 6.22-fold increased odds of overall osteoporosis-related mechanical complications.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This retrospective multicenter cohort examined whether preoperative GNRI adds risk stratification beyond vertebral HU values for predicting osteoporosis-related mechanical complications (vertebral fracture, cage subsidence, screw loosening) after short-segment lumbar fusion for stenosis. In multivariable analysis, low HU (≤100) predicted vertebral fractures, screw loosening, and cage subsidence, while low GNRI (<98) independently predicted screw loosening. In the low-HU subgroup specifically, low GNRI conferred OR 6.22 for overall complications and OR 5.85 for screw loosening.
Key Limitation
The low-GNRI subgroup comprised only 33 patients, making the OR estimates for subgroup analyses statistically unstable and potentially unreliable for clinical decision thresholds.
Original Abstract
BACKGROUND CONTEXT
Bone health optimization is increasingly emphasized in spinal surgery because poor bone health is associated with postoperative mechanical complications. Computed tomography-derived vertebral Hounsfield unit (HU) values primarily reflect local vertebral bone mineralization, which is a component of overall bone quality. In contrast, the Geriatric Nutritional Risk Index (GNRI) may reflect systemic nutritional and physiological vulnerabilities. We hypothesized that combining HU and the GNRI would provide complementary risk information about mechanical complications after lumbar fusion surgery.
PURPOSE
To determine whether the preoperative GNRI provides additional risk stratification beyond vertebral HU values in predicting postoperative osteoporosis-related complications after lumbar fusion surgery for lumbar spinal stenosis.
STUDY DESIGN/SETTING
Retrospective multicenter cohort study.
PATIENT SAMPLE
A total of 150 patients who underwent short-segment lumbar fusion surgery for lumbar spinal stenosis and completed a minimum 5-year follow-up were included.
OUTCOME MEASURES
Postoperative osteoporosis-related complications included incident vertebral fractures, cage subsidence, or screw loosening. Individual complications were also analyzed. Longitudinal clinical outcomes included Japanese Orthopaedic Association and visual analog scale scores for low back and leg pain.
METHODS
Preoperative nutritional status was assessed using the GNRI, with low GNRI defined as <98. Vertebral bone quality was assessed using mean L1-L4 HU values on computed tomography, with low HU defined as ≤100. Multivariable logistic regression analysis adjusted for age, sex, and osteoporosis medication was performed, including subgroup analyses of patients with low HU. This study received no external funding, and the authors report no relevant financial relationships or conflicts of interest.
RESULTS
Thirty-three patients had low GNRI scores. The low GNRI group had lower HU values and a higher incidence of postoperative osteoporosis-related complications than the high GNRI group. In multivariable analyses, a low GNRI was associated with screw loosening, whereas low HU was associated with incident vertebral fractures, screw loosening, and cage subsidence. Among patients with low HU, a low GNRI was associated with an increased risk of overall complications (odds ratio, 6.22; p = 0.018) and screw loosening (odds ratio, 5.85; p = 0.009).
CONCLUSIONS
Preoperative GNRI complements vertebral HU values for risk stratification after lumbar fusion surgery. In patients with a low HU, a low GNRI identified a subgroup with a higher risk of osteoporosis-related complications, particularly screw loosening. The combined assessment of HU and the GNRI may support risk-adapted perioperative bone health optimization.