Comparison of the Predictive Roles of CT- and MRI-Based Pedicle Regional Osteoporosis Status Measurements for Pedicle Screw Loosening After Posterior Lumbar Interbody Fusion

Abstract

Study Design

Retrospective Study.

Objectives

To compare the efficacy of pedicle Hounsfield unit (HU) values and pedicle bone quality (PBQ) scores in predicting pedicle screw loosening (PSL) after posterior lumbar interbody fusion (PLIF) in elderly patients and to identify the most discriminative bone mineral density (BMD) assessment indicator.

Methods

This retrospective analysis included 275 consecutive patients who underwent PLIF. L₁ and pedicle HU values were obtained from CT scans, whereas vertebral bone quality (VBQ) and PBQ scores were obtained from MRI. Logistic regression analysis determined factors associated with PSL. Receiver-operating characteristic curve analysis was conducted to assess the predictive value of pedicle HU and PBQ score for PSL and to additionally compare their predictive value with that of L₁ HU value and VBQ score.

Results

The PSL rate was 20.36% (56/275). The loosening group demonstrated a higher postoperative low-back pain visual analog scale score (P < 0.05), lower L₁ (P < 0.001) and pedicle HU values (P < 0.001) and higher VBQ (P < 0.001) and PBQ scores (P < 0.001) than the non-loosening group. The logistic regression analysis identified L₁ (OR = 0.98, 95% CI = 0.97-0.99, P < 0.001) and pedicle HU values (OR = 0.99, 95% CI = 0.98-0.99, P < 0.001) and VBQ (OR = 2.81, 95% CI = 1.43-5.52, P = 0.003) and PBQ scores (OR = 3.85, 95% CI = 2.03-7.32, P < 0.001) as independent predictors of PSL. The area under the curve for L₁ and pedicle HU values and VBQ and PBQ scores were 0.751, 0.766, 0.684, and 0.702, respectively. The optimal pedicle HU cut-off for predicting PSL was 106.32 (sensitivity: 78.49%; specificity: 75.00%).

Conclusions

Pedicle HU values exhibited a relatively higher predictive performance for PSL compared with the PBQ score and served as the most discriminative BMD indicator in patients who underwent PLIF. Measuring the pedicle HU value preoperatively help surgeons to select a more appropriate surgical plan and is expected to improve the patient outcomes.

Keywords

hounsfield unit value vertebral bone quality score pedicle bone quality score screw loosening CT MRI

Introduction

Pedicle screw loosening is one of the most common complications after posterior lumbar interbody fusion (PLIF),¹ potentially causing low-back pain, functional impairment, and even requiring revision surgery.^1,2 Further, PSL may precipitate further complications, such as screw breakage, nonunion, pseudarthrosis, and kyphosis,^3,4 which substantially decrease patient quality of life and increase the economic burden. Accordingly, accurate preoperative prediction and prevention of PSL are essential for optimizing surgical outcomes.

Studies have revealed that low bone mineral density (BMD) is a major risk factor for PSL.^5-7 The incidence of PSL ranges from 1% to 15% in patients with no osteoporosis but may reach up to 60% in those with osteoporosis.^1,8,9 Biomechanical in vitro experiments have demonstrated that specimens with normal bone density require significantly higher screw pull-out forces than osteoporotic specimens.^5,10 Furthermore, additional research indicates that pedicle bone density exerts a greater impact on screw stability than whole vertebral body density, with approximately 80% of the caudocephalad stiffness and 60% of the reduction in screw pullout strength attributable to the pedicle.¹¹

Dual-energy X-ray absorptiometry (DXA) is the current gold standard for bone density assessment.¹² However, its clinical application is limited by BMD overestimation in patients with degenerative lumbar disease^13,14 and by low clinical application rates.¹⁵ The vertebral body Hounsfield unit (HU) value obtained from computed tomography (CT) serves as a reliable BMD indicator,^16,17 and the vertebral bone quality (VBQ) score derived from magnetic resonance imaging (MRI) has appeared as a novel tool for bone quality assessment.¹⁸ Both appear to be viable alternatives to DXA and have been closely associated with PSL.^19-23

The pedicle HU values and pedicle bone quality (PBQ) score have been developed for preoperative BMD assessment based on the vertebral body HU value and VBQ score, and further studies have determined both as independent predictors of PSL.^24-28 However, it remains unclear whether pedicle HU or PBQ score has superior predictive ability for PSL.¹¹ The selection of BMD assessment indicators that are more closely associated with PSL is critical for individualized screw selection, fixation strategy decisions, and the development of tailored anti-osteoporotic treatment plans.^1,29-32 Therefore, this study is the first to systematically compare the predictive efficacy of the pedicle HU value and PBQ score for PSL after PLIF in the same patient cohort. It further investigates BMD assessment indicators measured at the pedicle vs the conventional vertebral body regions, aiming to determine the more valuable PSL predictor to improve preoperative BMD evaluation, guide clinical decision-making, and ultimately reduce the risk of PSL.

Methods

Patients

We retrospectively reviewed data of consecutive patients who underwent surgery for degenerative lumbar disease at our institution from January 2016 to October 2023. Our Institutional Ethics Committee approved this study (approval number: 2024-KE-346). Informed consent was waived due to the retrospective study design. The inclusion criteria were (1) PLIF with pedicle screw fixation for degenerative lumbar disease; (2) lumbar CT and MRI performed within 1 month preoperatively; and (3) minimum follow-up of 12 months. The exclusion criteria were (1) prior lumbar surgery; (2) congenital spinal deformity, ankylosing spondylitis, lumbar infection, spinal tumor, or lumbar trauma; (3) metabolic bone disease or long-term use of medications affecting BMD (e.g., corticosteroids); (4) fixation levels of > 4; and (5) revision surgery within 12 months postoperatively for non-PSL reasons, such as intraoperative neural injury, postoperative infection or hematoma, or adjacent-segment disc degeneration. Postoperatively, all patients received long-term antiosteoporosis therapy and attend regular outpatient follow-up visits. This study ultimately included 275 patients. PSL was assessed at 12 months postoperatively. Two senior spine surgeons independently evaluated PSL on the radiographs (N.F. with 10 years of experience and P.D. with 18 years of experience). In cases of disagreement, the final decision was made by a spine expert (L.Z. with 21 years of experience). Figure 1 illustrates the flowchart of patient selection.

Figure 1.

Patient Inclusion and Exclusion Flow Chart

Data Collection and Outcome Assessment

The patient demographic data collected included age, sex, and body mass index (BMI). Radiological data included fixation levels, the lowest instrumented vertebra (LIV) at S₁, L₁ HU value, pedicle HU value, VBQ score, and PBQ score. Screw loosening was defined as the presence of a radiolucent zone ≥ 1 mm around the pedicle screw on radiography.^1,33 Clinical outcomes were assessed using the visual analog scale (VAS) and the Oswestry Disability Index score at 12 months postoperatively.

HU Value Measurements

The L₁ HU value was measured following the method of Schreiber et al.³⁴ An elliptical region of interest (ROI) was placed at 3 axial levels on CT scans: below the superior endplate, at the vertebral midsection, and above the inferior endplate. This encompasses as much trabecular bone as possible while avoiding cortical bone and heterogeneous areas such as the posterior venous plexus, bone islands, and endplate sclerosis. The mean of the 3 measurements was documented as the L₁ HU value. The pedicle HU value was measured following the method of Xu et al.³⁵ the ROI was placed at the narrowest portion of the bilateral L₁ pedicles, and the average of the 2 measurements was recorded (Figure 2).

Figure 2.

ROIs Were Placed at 3 Axial levels (Below the Superior Endplate, Vertebral Midsection, and Above the Inferior Endplate) of L1 for L1 HU Value Measurement (A). ROIs were Placed at the Narrowest point of the Bilateral Pedicles on the Midsagittal Slice of the L1 CT scan for Pedicle HU value Measurement (B)

VBQ and PBQ Score Calculation

All MRI scans were performed on the same scanner. The VBQ score was calculated utilizing lumbar T1-weighted MRI without contrast. Following the method described by Ehresman et al.¹⁸ ROIs were placed in the medullary region of L₁–L₄ at 3 mm from the cortical margin and the cerebrospinal fluid (CSF) space at the L₃ level on the mid-sagittal plane. The mean signal intensity (SI) of each vertebral ROI and the CSF mean SI were recorded. The VBQ score was then calculated as the ratio of the average vertebral SI to the CSF SI. The PBQ score was measured following the VBQ method. ROIs were placed in L₁–L₄ regions 3 mm from the pedicle cortex to obtain the pedicle SI on T1-weighted sagittal MRI images. The PBQ score was then calculated as the average pedicle SI divided by the CSF SI at the L₃ level. CSF from the L₂ or L₄ level was used instead if the CSF at the L₃ level was unavailable (Figure 3).

Figure 3.

VBQ (A) and PBQ Scores (B) were Assessed on Preoperative Lumbar MRI and Standardized Based on the L3 Cerebrospinal Fluid (CSF) Signal Intensity

Supplemental File 1 presents detailed CT and MRI parameters, along with a description of the surgical procedure. Two experienced spine surgeons (T.W. with 5 years of experience and A.W. with 5 years of experience) independently measured the vertebral body and pedicle HU values, as well as the VBQ and PBQ scores using RadiAnt DICOM Viewer version 2025.1 (Medixant, Poznań, Poland), and the average of their measurements was recorded as the final result. The anonymised DICOMs were available. Both surgeons were blinded to the patient’s clinical outcomes and also blinded to each other’s first set of ROIs. After 2 weeks later, they repeated the measurements, and the intra- and inter-reader correlation coefficients (ICC) were calculated. The ICC values less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater than 0.90 were defined as indicating poor, moderate, good, and excellent reliability, respectively.³⁶ A third spine surgeon independently assessed screw loosening.

Statistical Analysis

Statistical analysis was performed using SPSS Statistics version 27.0 (IBM Analytics, New York, USA), and figures were generated using GraphPad Prism version 10.3.1 (GraphPad Software, San Diego, California, USA). The continuous variables were tested for normality with the Shapiro–Wilk test. Data with a normal distribution were analyzed using Student’s t-test, whereas nonnormally distributed data were analyzed using the Mann–Whitney U test. Categorical variables were analyzed using the chi-square test or Fisher’s exact test. Potential confounders (Table 1) were entered into multivariate logistic regression model. Receiver-operating characteristic (ROC) curves adjusted by the confounders via logistic regression residuals were constructed, the goodness of fit of the final adjusted model was evaluated using the Hosmer–Lemeshow test. The area under the curve (AUC) was utilized to assess the predictive value of L₁ and pedicle HU values and VBQ score and PBQ scores for PSL. The AUC difference was compared using the DeLong test, and the optimal cut-off value was determined using the Youden index. Using the McNemar test to evaluate the clinical applicability of the simplified threshold. Pearson’s correlation analysis was conducted to assess the associations between variables. A P value of < 0.05 indicated statistical significance.

Table 1.

Comparison of Patient Characteristics Between the Loosening and Non-loosening Groups

	Non-loosening (n = 219)	Loosening (n = 56)	P value^a
Age, years	63.07 ± 9.58	65.02 ± 8.93	0.171
Sex			0.984
Male	102 (46.58%)	26 (46.43%)
Female	117 (53.42%)	30 (53.57%)
BMI, kg/m²	26.55 ± 3.94	26.91 ± 3.78	0.541
Average fusion length, levels	2.42 ± 0.61	2.39 ± 0.49	0.797
LIV at S₁	61 (27.85%)	29 (51.79%)	<0.001
L₁ HU value	147.48 ± 49.67	102.71 ± 41.77	<0.001
Pedicle HU value	277.47 ± 84.45	196.39 ± 87.23	<0.001
VBQ score	2.99 ± 0.52	3.27 ± 0.40	<0.001
PBQ score	2.85 ± 0.62	3.25 ± 0.40	<0.001

BMI, body mass index; LIV, lowest instrumented vertebrae; HU, Hounsfield unit; VBQ, vertebral bone quality; PBQ, pedicle bone quality.

^aBold italics indicate statistical significance.

Results

This study included 275 patients (128 males and 147 females). The mean age was 63.47 ± 9.47 years, the mean BMI was 26.62 ± 3.91 kg/m², and the mean fusion length was 2.41 ± 0.59 levels. The screw loosening rate was 20.36% (56/275), and 32.73% of patients had the LIV at S₁. The mean L₁ and pedicle HU values and mean VBQ and PBQ scores were 138.36 ± 51.38, 260.96 ± 90.95, 3.05 ± 0.51, and 2.91 ± 0.60, respectively. Postoperative low‐back pain VAS score was significantly higher in the loosening group, whereas other clinical outcomes did not significantly differ between the groups (Supplemental File 2). The ICCs for L1 and pedicle HU values, as well as VBQ and PBQ scores, all exceeded 0.80, indicating good intra- and inter-reader reliability. (Supplemental File 3).

Table 1 compares patient characteristics between the loosening and non-loosening groups. Age, sex, BMI, and fusion levels did not significantly differ between the groups (P > 0.05). Notably, the loosening group had significantly lower L₁ and pedicle HU values and higher LIV proportions at S₁, VBQ scores, and PBQ scores compared with the non-loosening group (P < 0.001).

Age, sex, BMI, average fusion length, and LIV at S1 were selected as potential confounders and included into separate multivariate logistic regression models due to significant linear correlations among L₁ HU, pedicle HU, VBQ score, and PBQ score (Supplemental File 4). Results revealed that L₁ HU (odds ratio [OR] = 0.98, 95% confidence interval [CI] = 0.97-0.99, P < 0.001), pedicle HU (OR = 0.99, 95% CI = 0.98-0.99, P < 0.001), VBQ score (OR = 2.81, 95% CI = 1.43-5.52, P = 0.003), and PBQ score (OR = 3.85, 95% CI = 2.03-7.32, P < 0.001) were significantly associated with PSL (Table 2). Supplemental File 5 shows the full logistic regression results. Notably, LIV at S₁ remained significantly associated with PSL across the models (P < 0.05). The P-values of Hosmer-Lemeshow goodness of fit test were all greater than 0.05, indicating adequate model fit (L₁ HU value: P = 0.585, pedicle HU value: P = 0.523, VBQ score: P = 0.210, and PBQ score: P = 0.306).

Table 2.

Logistic Regression Analysis of the Influencing Factors for Screw Loosening After Lumbar Spine Surgery

	OR (95% CI)	P value^a
L₁ HU value	0.98 (0.97-0.99)	<0.001
Pedicle HU value	0.99 (0.98-0.99)	<0.001
VBQ score	2.81 (1.43-5.52)	0.003
PBQ score	3.85 (2.03-7.32)	<0.001

HU, Hounsfield unit; VBQ, vertebral bone quality; PBQ, pedicle bone quality; OR, odds ratio.

^aBold italics indicate statistical significance.

ROC curves were constructed to assess the predictive value of each bone quality indicator for PSL (Figure 4), and the Youden index was employed to determine the optimal cut-off values (Table 3). The AUC for L₁ and pedicle HU values and VBQ and PBQ scores were 0.751 (optimal cut-off: 106.32; sensitivity: 73.11%; specificity: 58.93%), 0.766 (optimal cut-off: 237.78; sensitivity: 78.49%; specificity: 75.00%), 0.684 (optimal cut-off: 3.18; sensitivity: 66.07%; specificity: 78.08%), and 0.702 (optimal cut-off: 3.09; sensitivity: 67.86%; specificity: 69.68%), respectively. Using 110 as the simplified optimal cut-off value for pedicle HU, the McNemar test showed no statistically significant difference compared to 106.32 (P = 0.125). The DeLong test revealed that only the difference between the AUC of the pedicle HU value and the VBQ score reached statistical significance (P < 0.05) (Supplemental File 6).

Figure 4.

ROC Curve Analysis of L1 and Pedicle HU Values and VBQ and PBQ Scores

Table 3.

AUC and Optimal Cut-Off Value of Predictors in Screw Loosening

Variable	AUC	Cut-off value	Sensitivity	Specificity
L₁ HU value	0.751	106.32	73.11%	58.93%
Pedicle HU value	0.766	237.78	78.49%	75.00%
VBQ score	0.684	3.18	66.07%	78.08%
PBQ score	0.702	3.09	67.86%	69.68%

HU, Hounsfield unit; VBQ, vertebral bone quality; PBQ, pedicle bone quality; AUC, area under the curve.

Discussion

To the best of our knowledge, this study is the first to systematically compare the predictive performance of the pedicle HU value and PBQ score for PSL after PLIF. We revealed that pedicle and L₁ HU values and VBQ and PBQ scores were all independently associated with PSL after PLIF, and the pedicle HU value was the most discriminative BMD indicator.

PSL is a common complication after PLIF,¹ frequently causing restricted mobility and persistent low-back pain, with severe cases requiring revision surgery.^1-3 In our study, the loosening group demonstrated a higher postoperative back VAS score (P < 0.05), consistent with the findings of Zou et al²¹ Loosened screws may undergo micromotion during physical activity, causing repeated irritation of the surrounding bone and compromised spinal stability, which may ultimately result in low-back discomfort. These findings indicate that PSL is closely associated with poor clinical outcomes.²² Therefore, the preoperative identification of high-risk patients and the implementation of individualized preventive strategies are crucial to improving surgical success and clinical outcomes. Previous studies have determined common risk factors for PSL, including advanced age, osteoporosis, sagittal imbalance, multilevel fixation, and LIV at S₁.^6,23,37,38 In our study, the screw loosening rate was 20.36% (56/275), consistent with previous reports.^1,9,27 We revealed LIV at S₁ as an independent risk factor for PSL, aligning with previous findings,^22,23 possibly due to the shorter, wider anatomy of the S₁ pedicle and the greater shear and torsional stresses borne by S₁ as the stress conduction center between the spine and pelvis.³⁹ However, neither age nor fusion length appeared as a significant risk factor in our study. The former may indicate that bone loss, rather than age per se, drives screw loosening, whereas the latter may be caused by the limited statistical power from our sample size.

BMD is one of the major risk factors for PSL after lumbar instrumentation.^5,6,40 DXA remains the gold standard for BMD assessment; however, DXA measurements in patients with degenerative lumbar changes are frequently spuriously increased due to osteophytes,^14,41-43 making DXA alone insufficient to identify all high-risk individuals preoperatively. Recently, the CT-based HU value and MRI-based VBQ score have appeared as potential alternatives.⁴³ Joseph et al revealed that the HU value effectively differentiates healthy, osteopenic, and osteoporotic populations,³⁴ and the VBQ score correlates moderately with DXA T-scores¹⁸ while exhibiting good intra- and inter-reader reliability.⁴⁴ In our study, both the L₁ HU value and VBQ score were independent PSL predictors, consistent with prior reports.^21-23,27 Ye et al revealed that the VBQ score outperformed the HU value in predicting PSL among patients with lumbar spinal stenosis (P_AUC = 0.003). Similarly, Li et al. revealed that the VBQ score was superior to the HU value.²³ Contrasting these findings, our results indicated a lower AUC for the VBQ score compared with the L₁ HU value (0.684 vs 0.751). We attribute this discrepancy to the following reasons. First, there are fundamental differences in the working principles of CT and MRI. The CT-based HU value was obtained by quantifying the degree of attenuation of X-rays passing through the trabecular bone during CT imaging. This is a direct positive correlation. Selectively delineating and measuring the ROI can prevent the influence of the cortical bone and osteophytes, thereby obtaining more accurate BMD measurement results.³⁴ The MRI-based VBQ score indirectly assesses bone quality via vertebral fat infiltration^45,46; however, the fat content is susceptible to disc degeneration and inflammatory factors,^47,48 and the errors are not fully mitigated by ROI selection. Wang et al. further demonstrated that vertebral marrow fat infiltration tends to plateau in individuals aged ≥ 80 years despite ongoing bone loss.⁴⁹ Second, our method for measuring vertebral HU values differs from that used by Li et al. Following the method described by Schreiber et al.³⁴ we measured the HU values at 3 levels of the L₁ vertebral body—below the superior endplate, at the mid-vertebral body, and above the inferior endplate—and used the average value as the final result. Fan et al. also reported that HU measurements at L₁ have the highest inter-observer consistency (ICC = 0.909), compared to T₁₂ and L₂ (ICC = 0.895 and 0.887, respectively), making L₁ the most representative vertebral level.⁵⁰ Li et al defined the ROI only at the mid-vertebral body and averaged HU values across L₁–L₄, which may introduce greater variability and result in a lower AUC for HU values. Finally, the patients in our study were generally older than those in the study by Liu et al. Given that advanced age is associated with more severe disc degeneration, the resulting variability in VBQ scoring may have further contributed to the lower AUC observed for VBQ in our study. However, the difference between the HU value and VBQ score was not statistically significant (P_AUC = 0.077), indicating that the HU value demonstrates greater reliability, but it remains intrinsically correlated with the VBQ score.

Screw stability depends on the structural characteristics of the pedicle,^11,51,52 and pull-out strength correlates much more strongly with the pedicle region than with the vertebral body BMD,⁵³ and approximately 60% of the screw pull-out strength is attributable to the pedicle.¹¹ Consistent with previous studies, we revealed that the AUC for the pedicle HU value in predicting PSL was slightly greater than that for the vertebral body HU value,^27,54 and that PBQ’s AUC slightly surpassed VBQ’s.²⁸ Using the same measurement machine, BMD assessment at the pedicle, which is the site most directly related to screw stability, provides a relatively higher predictive accuracy. To the best of our knowledge, no previous research has directly compared the predictive capacities of BMD indicators at the pedicle region. Our study revealed that the pedicle HU value not only outperformed PBQ in predicting PSL (0.766 vs 0.702) but also exhibited slightly higher sensitivity and specificity. Interestingly, we observed that the PBQ score’s AUC was slightly lower than that of the L₁ HU value (0.702 vs 0.751), indicating that differences in measurement tools cannot be fully compensated by the choice of target region. These results emphasize that prioritizing CT as the tool for BMD evaluation—rather than focusing solely on the measurement site—may be a more valuable strategy in assessing the risk of PSL. Moreover, the difference between the AUC of the pedicle HU value and the VBQ score was statistically significant (P_AUC = 0.047), which we attribute to the “superposition of differences” in the measurement tool and region. When there is a discrepancy between the pedicle HU value and the VBQ score in predicting the risk of PSL, surgeons should prioritize the pedicle HU value to formulate a more appropriate surgical strategy.

In addition to measuring BMD at the vertebral body and pedicle, some studies have explored measuring the three-dimensional (3D) HU value along the pedicle screw trajectory as a predictive tool for PSL.⁵⁴ However, this method requires specialized 3D image-analysis software and is not readily applicable in routine clinical practice. In summary, L₁ and pedicle HU values and VBQ and PBQ scores all serve as independent PSL predictors after lumbar instrumentation in patients with degenerative lumbar disease, with the pedicle HU value showing greater predictive accuracy than the PBQ score, making it a more reliable and clinically valuable indicator. We recommend that surgeons utilize already-available preoperative CT scans to obtain pedicle HU values as a complementary tool to DXA for preoperative BMD assessment, in order to streamline patient selection for augmentation strategies. According to the results of the McNemar test, the threshold may be set at 110 (originally 106.32) for clinical convenience. We recommend adding teriparatide to conventional antiosteoporosis therapy for patients at high risk of screw loosening with pedicle HU values below 110. Ohtori et al. revealed that teriparatide can significantly reduce PSL.⁵⁵ Surgically, adopting novel materials and techniques that strengthen the screw–bone interface stability, such as expandable screws,^1,30-32 hydroxyapatite-coating of pedicle screws,⁵⁶ or cement-augmented screws,⁵⁷ may improve patient outcomes.

Limitation

This study has several limitations. First, this was a single-center retrospective study, and patients with spinal deformity, tumors, infections, and metabolic bone diseases were excluded, which may limit the external validity and generalizability of the findings. Second, the duration and adherence of postoperative anti-osteoporotic treatment were not systematically recorded or compared, potentially introducing bias. Future multicenter prospective studies with larger sample sizes and standardized protocols for monitoring medication adherence and treatment duration are warranted to further validate the effectiveness and applicability of our findings. Third, data on patients’ DXA T scores were lacking, which prevented us from comparing the predictive ability of pedicle HU values with that of DXA for PSL. In future studies, we plan to incorporate patient DXA data to benchmark against standard BMD tools of the pedicle and vertebral body, and to compare their predictive performance. Fourth, the manual selection of ROIs for HU value and VBQ score measurements is prone to subjectivity. However, previous research has demonstrated that HU value and VBQ score measurements have good intra- and inter-reader reliability.^44,58 Future development of a deep learning model capable of automated delineation of ROIs on CT/MRI scans for BMD assessment is anticipated to effectively resolve this issue. Fifth, constructing separate logistic regression models for each predictor may limit the ability to assess their relative contributions. Finally, we only analyzed some of the factors associated with PSL and other well-established influencing factors, such as pedicle screw size, insertion technique, postoperative sagittal alignment, and spinopelvic parameters, were not included in the statistical analysis due to limitations in sample size.^4,59-62 Future studies are warranted to incorporate these surgical factors and integrate various BMD indicators, with the ultimate goal of developing a precise predictive model for clinical applications.

Conclusion

The pedicle HU value more accurately predicts PSL than the PBQ score and is the most discriminative indicator of BMD in patients undergoing PLIF for lumbar degeneration. Preoperative assessment of pedicle HU values facilitates the identification of patients at high risk of PSL and may help surgeons in optimizing surgical strategies to improve the patient’s clinical outcomes.

Supplemental Material

Supplemental Material - Comparison of the Predictive Roles of CT- and MRI-Based Pedicle Regional Osteoporosis Status Measurements for Pedicle Screw Loosening After Posterior Lumbar Interbody Fusion

Supplemental Material for Comparison of the Predictive Roles of CT- and MRI-Based Pedicle Regional Osteoporosis Status Measurements for Pedicle Screw Loosening After Posterior Lumbar Interbody Fusion by Peng Du, MM, Minghui Liang, MM, Ruiyuan Chen, MM, Tianyi Wang, MD, Ning Fan, MD, Shuo Yuan, MD, Aobo Wang, MD, Ziqian Ma, MD, Yu Xi, MM, Lei Zang, MD in Global Spine Journal

Footnotes

ORCID iDs

Ruiyuan Chen

Tianyi Wang

Aobo Wang

Yu Xi

Lei Zang

Ethical Considerations

The research conducted has been performed in accordance with the Declaration of Helsinki. Approval for the study was obtained from the ethics committees of the Beijing Chaoyang Hospital (2024-KE-346).

Author Contributions

PD and ML and RC contributed equally to this work. PD and ML and RC contributed equally to this work. Conception and design: LZ. Acquisition of data: PD and ML and RC. Analysis and interpretation of data: PD and ML and RC. Drafting the article: PD and ML and RC. Critically revising the article: LZ, TW, NF, SY, and AW. Reviewed submitted version of manuscript: ZM and YX.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. *

Supplemental Material

Supplemental material for this article is available online.

References

Galbusera

Volkheimer

Reitmaier

Berger-Roscher

Kienle

Wilke

. Pedicle screw loosening: a clinically relevant complication? Eur Spine J. 2015;24(5):1005-1016. doi:10.1007/s00586-015-3768-6

J-C

Huang

W-C

Tsai

H-W

, et al. Pedicle screw loosening in dynamic stabilization: incidence, risk, and outcome in 126 patients. Neurosurg Focus. 2011;31(4):E9. doi:10.3171/2011.7.Focus11125

Rometsch

Spruit

Zigler

, et al. Screw-related complications after instrumentation of the osteoporotic spine: a systematic literature review with Meta-analysis. Glob Spine J. 2020;10(1):69-88. doi:10.1177/2192568218818164

Yuan

Zhang

Zeng

Chen

. Incidence, risk, and outcome of pedicle screw loosening in degenerative lumbar scoliosis patients undergoing long-segment fusion. Glob Spine J. 2023;13(4):1064-1071. doi:10.1177/21925682211017477

Varghese

Saravana Kumar

Krishnan

. Effect of various factors on pull out strength of pedicle screw in normal and osteoporotic cancellous bone models. Med Eng Phys. 2017;40:28-38. doi:10.1016/j.medengphy.2016.11.012

Okuyama

Abe

Suzuki

Tamura

Chiba

Sato

. Influence of bone mineral density on pedicle screw fixation: a study of pedicle screw fixation augmenting posterior lumbar interbody fusion in elderly patients. Spine J. 2001;1(6):402-407. doi:10.1016/s1529-9430(01)00078-x

Bokov

Bulkin

Aleynik

Kutlaeva

Mlyavykh

. Pedicle screws loosening in patients with degenerative diseases of the lumbar spine: potential risk factors and relative contribution. Glob Spine J. 2018;9(1):55-61. doi:10.1177/2192568218772302

Saman

Meier

Sander

Kelm

Marzi

Laurer

. Reduced loosening rate and loss of correction following posterior stabilization with or without PMMA augmentation of pedicle screws in vertebral fractures in the elderly. Eur J Trauma Emerg Surg. 2013;39(5):455-460. doi:10.1007/s00068-013-0310-6

Uehara

Takahashi

Ikegami

, et al. Pedicle screw loosening after posterior spinal fusion for adolescent idiopathic scoliosis in upper and lower instrumented vertebrae having major perforation. Spine. 2017;42(24):1895-1900. doi:10.1097/brs.0000000000002305

10.

Zhang

Tan

Chou

. Investigation of fixation screw pull-out strength on human spine. J Biomech. 2004;37(4):479-485. doi:10.1016/j.jbiomech.2003.09.005

11.

Hirano

Hasegawa

Takahashi

, et al. Structural characteristics of the pedicle and its role in screw stability. Spine. 1997;22(21):2504-2509. doi:10.1097/00007632-199711010-00007

12.

Link

. Osteoporosis imaging: state of the art and advanced imaging. Radiology. 2012;263(1):3-17. doi:10.1148/radiol.2633201203

13.

Muraki

Yamamoto

Ishibashi

, et al. Impact of degenerative spinal diseases on bone mineral density of the lumbar spine in elderly women. Osteoporos Int. 2004;15(9):724-728. doi:10.1007/s00198-004-1600-y

14.

Westerveld

Verlaan

J-J

Lam

MGEH

, et al. The influence of diffuse idiopathic skeletal hyperostosis on bone mineral density measurements of the spine. Rheumatology. 2009;48(9):1133-1136. doi:10.1093/rheumatology/kep177

15.

Neuner

Binkley

Sparapani

Laud

Nattinger

. Bone density testing in older women and its association with patient age. J Am Geriatr Soc. 2006;54(3):485-489. doi:10.1111/j.1532-5415.2005.00622.x

16.

Pan

Shi

Wang

, et al. Automatic opportunistic osteoporosis screening using low-dose chest computed tomography scans obtained for lung cancer screening. Eur Radiol. 2020;30(7):4107-4116. doi:10.1007/s00330-020-06679-y

17.

Zou

Deng

. The use of CT Hounsfield unit values to identify the undiagnosed spinal osteoporosis in patients with lumbar degenerative diseases. Eur Spine J. 2019;28(8):1758-1766. doi:10.1007/s00586-018-5776-9

18.

Ehresman

Pennington

Schilling

, et al. Novel MRI-based score for assessment of bone density in operative spine patients. Spine J. 2020;20(4):556-562. doi:10.1016/j.spinee.2019.10.018

19.

Bredow

Boese

Werner

CML

, et al. Predictive validity of preoperative CT scans and the risk of pedicle screw loosening in spinal surgery. Arch Orthop Trauma Surg. 2016;136(8):1063-1067. doi:10.1007/s00402-016-2487-8

20.

Zou

Sun

Zhou

Zhong

. Hounsfield units value is a better predictor of pedicle screw loosening than the T-score of DXA in patients with lumbar degenerative diseases. Eur Spine J. 2020;29(5):1105-1111. doi:10.1007/s00586-020-06386-8

21.

Zou

Muheremu

Sun

Zhong

Jiang

. Computed tomography Hounsfield unit-based prediction of pedicle screw loosening after surgery for degenerative lumbar spine disease. J Neurosurg Spine. 2020;32(5):716-721. doi:10.3171/2019.11.Spine19868

22.

Y-H

Chou

J-H

Yeh

Y-C

, et al. The MRI-based vertebral bone quality score is a predictor of pedicle screw loosening following instrumented posterior lumbar fusion. Sci Rep. 2025;15(1):1696. doi:10.1038/s41598-025-85625-8

23.

Zhu

Hua

, et al. Vertebral bone quality score as a predictor of pedicle screw loosening following surgery for degenerative lumbar disease. Spine. 2023;48(23):1635-1641. doi:10.1097/brs.0000000000004577

24.

Zhan

Wei

Guo

, et al. The predictive value of endplate morphology and pedicle screw bone quality score on screw loosening after single-level lumbar spinal fusion surgery. J Orthop Surg Res. 2024;19(1):898. doi:10.1186/s13018-024-05367-7

25.

Yang

Wang

Song

. Novel MRI-based pedicle bone quality score independently predicts pedicle screw loosening after degenerative lumbar fusion surgery. Orthop Surg. 2024;16(10):2372-2379. doi:10.1111/os.14146

26.

Chen

Lei

, et al. Prediction of pedicle screw loosening using an MRI-based vertebral bone quality score in patients with lumbar degenerative disease. World Neurosurg. 2023;171:e760-e767. doi:10.1016/j.wneu.2022.12.098

27.

Zou

, et al. Hounsfield units of the vertebral body and pedicle as predictors of pedicle screw loosening after degenerative lumbar spine surgery. Neurosurg Focus. 2020;49(2):E10. doi:10.3171/2020.5.Focus20249

28.

Yang

Wang

Song

. MRI-based pedicle bone quality score: correlation to quantitative computed tomography bone mineral density and its role in quantitative assessment of osteoporosis. Spine J. 2024;24(10):1825-1832. doi:10.1016/j.spinee.2024.06.007

29.

Ohtori

Inoue

Orita

, et al. Comparison of teriparatide and bisphosphonate treatment to reduce pedicle screw loosening after lumbar spinal fusion surgery in postmenopausal women with osteoporosis from a bone quality perspective. Spine. 2013;38(8):E487-E492. doi:10.1097/BRS.0b013e31828826dd

30.

Z-X

Gong

F-T

Liu

, et al. A comparative study on screw loosening in osteoporotic lumbar spine fusion between expandable and conventional pedicle screws. Arch Orthop Trauma Surg. 2012;132(4):471-476. doi:10.1007/s00402-011-1439-6

31.

Cook

Barbera

Rubi

Salkeld

Whitecloud

. Lumbosacral fixation using expandable pedicle screws. an alternative in reoperation and osteoporosis. Spine J. 2001;1(2):109-114. doi:10.1016/s1529-9430(01)00020-1

32.

Frankel

Jones

Wang

. Segmental polymethylmethacrylate-augmented pedicle screw fixation in patients with bone softening caused by osteoporosis and metastatic tumor involvement: a clinical evaluation. Neurosurgery. 2007;61(3):531-537. doi:10.1227/01.Neu.0000290899.15567.68

33.

Sandén

Olerud

Petrén-Mallmin

Johansson

Larsson

. The significance of radiolucent zones surrounding pedicle screws. Definition of screw loosening in spinal instrumentation. J Bone Joint Surg Br. 2004;86(3):457-461. doi:10.1302/0301-620x.86b3.14323

34.

Schreiber

Anderson

Rosas

Buchholz

. Hounsfield units for assessing bone mineral density and strength: a tool for osteoporosis management. J Bone Joint Surg Am. 2011;93(11):1057-1063. doi:10.2106/jbjs.J.00160

35.

Zou

36.

Koo

. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155-163. doi:10.1016/j.jcm.2016.02.012

37.

Bokov

Bulkin

Aleynik

Kutlaeva

Mlyavykh

. Pedicle screws loosening in patients with degenerative diseases of the lumbar spine: potential risk factors and relative contribution. Glob Spine J. 2019;9(1):55-61. doi:10.1177/2192568218772302

38.

Marie-Hardy

Pascal-Moussellard

Barnaba

Bonaccorsi

Scemama

. Screw loosening in posterior spine fusion: prevalence and risk factors. Glob Spine J. 2020;10(5):598-602. doi:10.1177/2192568219864341

39.

Bydon

Fredrickson

Garza-Ramos

, et al. Sacral fractures. Neurosurg Focus. 2014;37(1):E12. doi:10.3171/2014.5.Focus1474

40.

Amirouche

Solitro

Magnan

. Stability and spine pedicle screws fixation strength-A comparative study of bone density and insertion angle. Spine Deform. 2016;4(4):261-267. doi:10.1016/j.jspd.2015.12.008

41.

Muraki

Yamamoto

Ishibashi

, et al. Impact of degenerative spinal diseases on bone mineral density of the lumbar spine in elderly women. Osteoporos Int. 2004;15(9):724-728. doi:10.1007/s00198-004-1600-y

42.

Bolotin

. DXA in vivo BMD methodology: an erroneous and misleading research and clinical gauge of bone mineral status, bone fragility, and bone remodelling. Bone. 2007;41(1):138-154. doi:10.1016/j.bone.2007.02.022

43.

Deshpande

Hadi

Lillard

, et al. Alternatives to DEXA for the assessment of bone density: a systematic review of the literature and future recommendations. J Neurosurg Spine. 2023;38(4):436-445. doi:10.3171/2022.11.Spine22875

44.

Mierke

Ramos

Macneille

, et al. Intra- and inter-observer reliability of the novel vertebral bone quality score. Eur Spine J. 2022;31(4):843-850. doi:10.1007/s00586-021-07096-5

45.

Zha

Wang

, et al. Quantitative evaluation of vertebral microvascular permeability and fat fraction in alloxan-induced diabetic rabbits. Radiology. 2018;287(1):128-136. doi:10.1148/radiol.2017170760

46.

Scheller

Rosen

. What's the matter with MAT? Marrow adipose tissue, metabolism, and skeletal health. Ann N Y Acad Sci. 2014;1311:14-30. doi:10.1111/nyas.12327

47.

Lee

Song

Baek

, et al. Interleukin-1β induces angiogenesis and innervation in human intervertebral disc degeneration. J Orthop Res. 2011;29(2):265-269. doi:10.1002/jor.21210

48.

Krug

Joseph

Han

, et al. Associations between vertebral body fat fraction and intervertebral disc biochemical composition as assessed by quantitative MRI. J Magn Reson Imaging. 2019;50(4):1219-1226. doi:10.1002/jmri.26675

49.

Wang

Liu

, et al. Simplified S1 vertebral bone quality score in the assessment of patients with vertebral fragility fractures. World Neurosurg. 2024;185:e1004-e1012. doi:10.1016/j.wneu.2024.03.011

50.

Fan

Wang

, et al. Vertebral CT Hounsfield units in postmenopausal women with osteoporotic vertebral compression fracture: identification and validation of reference intervals. Eur Spine J. 2025;34(5):1663-1672. doi:10.1007/s00586-025-08828-7

51.

Weinstein

Rydevik

Rauschning

. Anatomic and technical considerations of pedicle screw fixation. Clin Orthop Relat Res 1992;(284):34-46. https://www.ncbi.nlm.nih.gov/pubmed/1395312.

52.

Defino

Vendrame

. Role of cortical and cancellous bone of the vertebral pedicle in implant fixation. Eur Spine J. 2001;10(4):325-333. doi:10.1007/s005860000232

53.

Wichmann

Booz

Wesarg

, et al. Quantitative dual-energy CT for phantomless evaluation of cancellous bone mineral density of the vertebral pedicle: correlation with pedicle screw pull-out strength. Eur Radiol. 2014;25(6):1714-1720. doi:10.1007/s00330-014-3529-7

54.

Zhao

Wang

Y-J

Wang

R-G

, et al. Three-dimensional Hounsfield units measurement of pedicle screw trajectory for predicating screw loosening in lumbar fusion surgery. Clin Interv Aging. 2023;18:485-493. doi:10.2147/cia.S389059

55.

Ohtori

Inoue

Orita

56.

Hasegawa

Inufusa

Imai

Mikawa

Lim

. Hydroxyapatite-coating of pedicle screws improves resistance against pull-out force in the osteoporotic canine lumbar spine model: a pilot study. Spine J. 2005;5(3):239-243. doi:10.1016/j.spinee.2004.11.010

57.

Amendola

Gasbarrini

Fosco

, et al. Fenestrated pedicle screws for cement-augmented purchase in patients with bone softening: a review of 21 cases. J Orthop Traumatol. 2011;12(4):193-199. doi:10.1007/s10195-011-0164-9

58.

Pompe

Jong

, et al. Inter-observer and inter-examination variability of manual vertebral bone attenuation measurements on computed tomography. Eur Radiol. 2016;26(9):3046-3053. doi:10.1007/s00330-015-4145-x

59.

Kim

J-B

Park

S-W

Lee

Y-S

Nam

Park

Kim

. The effects of spinopelvic parameters and paraspinal muscle degeneration on S1 screw loosening. J Korean Neurosurg Soc. 2015;58(4):357-362. doi:10.3340/jkns.2015.58.4.357

60.

Bokov

Pavlova

Bulkin

Aleynik

Mlyavykh

. Potential contribution of pedicle screw design to loosening rate in patients with degenerative diseases of the lumbar spine: an observational study. World J Orthop. 2021;12(5):310-319. doi:10.5312/wjo.v12.i5.310

61.

Yang

J-X

Luo

Liu

J-H

Wang

. Incomplete insertion of pedicle screws triggers a higher biomechanical risk of screw loosening: mechanical tests and corresponding numerical simulations. Front Bioeng Biotechnol. 2023;11:1282512. doi:10.3389/fbioe.2023.1282512

62.

Peduzzi

Concato

Kemper

Holford

Feinstein

. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373-1379. doi:10.1016/s0895-4356(96)00236-3

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB