Pedicle Hounsfield Unit Threshold for Predicting Screw Malposition İn Idiopathic and Degenerative Scoliosis Surgery

Abstract

Study Design

Retrospective comparative study.

Objective

To investigate the relationship between pedicle Hounsfield Unit (HU) values and transpedicular screw malposition in scoliosis surgery and to determine a predictive HU threshold for malposition risk.

Methods

Patients who underwent free transpedicular screw fixation for idiopathic or degenerative scoliosis between 2011 and 2024 and had both preoperative and early postoperative spinal computed tomography (CT) scans available were retrospectively reviewed. Screw malposition was identified on postoperative CT, and pedicle HU values and the widths of malpositioned screws were measured on the corresponding preoperative CT images and compared with contralateral and control pedicles. Receiver operating characteristic (ROC) analysis was used to determine a predictive HU threshold for screw malposition.

Results

The study included 121 patients. Mean HU values were significantly higher in malpositioned pedicles compared to controls in both the idiopathic (692 vs 299.5; P < 0.001) and degenerative scoliosis (343.5 vs 250; P < 0.001) groups. ROC analysis determined the HU cut-off value to be 479 and above to predict screw malposition (AUC = 0.796; 95% CI: 0.717-0.876; sensitivity 73.6%, specificity 79.6%). The location of screw malposition according to the convex and concave sides, and the location above and below the apex did not show any significant differences.

Conclusion

Higher pedicle HU values are significantly associated with screw malposition in scoliosis surgery. A HU threshold of 479 may help identify pedicles at increased risk of malposition during free screw placement in both idiopathic and degenerative scoliosis. Preoperative pedicle HU assessment may contribute to surgical planning and risk reduction strategies.

Keywords

scoliosis surgery transpedicular screw hounsfield unit screw malposition sclerotic pedicle structure

Introduction

Transpedicular screw fixation has been widely used for spinal stabilization and deformity correction since its introduction by Boucher in 1959.^1-6 Despite advances in technique, screw malposition remains a significant concern in deformity surgery, with reported rates ranging from 8% to 22%.^7-9 In complex spinal deformities, this rate can range from 5% to 41%.^10-13 Although navigation systems can consistently improve screw placement accuracy compared with freehand techniques,^10-13 they are not universally available and can increase cost and operative time^.10,14,15 Also, access to navigation systems is limited in many parts of the world. Therefore, freehand techniques are widely used in many settings. Identifying pedicles at increased risk of malposition during freehand screw placement before surgery can allow for the implementation of predictive strategies (e.g., altered entry points, careful trajectory adjustment, or selective use of aids) and potentially reduce neurovascular complications.^8-14

Hounsfield Units (HU), derived from routine CT scans, provide a simple and reproducible surrogate for local bone mineral density and trabecular quality.^15-24 HU has been associated with clinically relevant endpoints such as osteoporosis screening, vertebral degeneration, and implant-related complications. The relationship with implants has been particularly important in the literature through studies predicting implant subsidence or loosening.^16-31 Previous studies have primarily evaluated HU as a predictor of postoperative stability (e.g., screw loosening) rather than initial technical accuracy at the time of screw placement.^16-31 Asymmetric loading in scoliosis may lead to regional differences in trabecular structure,^17,24,25 but it remains unclear whether these differences increase the risk of malposition during drilling/tapping.

Scoliosis surgery involves correcting the deformity by placing pedicle screws.^24,27 Screw malposition remains a significant challenge in scoliosis surgery, as in all procedures involving transpedicular screw placement.^8,9,25,30,31 The causes of malposition have been extensively studied. Previous reports have suggested that screw location relative to the apex of the scoliosis^21,23 and higher Cobb angles^21-23 may increase malposition risk. In our clinical practice, we have observed that pedicle bone density—reflected as pedicle stiffness perceived by the surgeon—is highly influential in screw malposition, particularly in free-hand scoliosis surgeries. This observation prompted us to design the present study. The relationship between bone density and HU values has been well documented in prior studies,^28-35 which further motivated us to evaluate pedicle HU values alongside other variables.

From a technical perspective, very dense (sclerotic) pedicles may resist smooth advancement of the pedicle probe along the ideal cancellous path, increasing the risk of slight stripping or reorientation during freehand insertion, even if the intended starting point and gross trajectory are correct. In this context, HU, reflecting local pedicle bone density, can serve as a practical preoperative marker to identify “difficult” pedicles where maintaining the ideal trajectory is more difficult. Therefore, the primary objective of this study was to investigate the relationship between pedicle HU values and transpedicular screw malposition in scoliosis surgery and to determine a predictive HU threshold for malposition risk using the Youden index and ROC analysis. Secondary objectives were to investigate whether this relationship varied according to scoliosis etiology (idiopathic or degenerative), curve anatomy (convex/concave; supra-apex/infra-apex/at-apex), medial or lateral fracture pattern, and pedicle thickness. We hypothesized that higher pedicle HU values might be associated with an increased risk of screw malposition during freehand insertion, and that a predictive risk value could be derived to guide preoperative planning.

Design and Methods

The study was approved by the Ethics Committee of the Faculty of Medicine of xxx University (Selcuk University, Konya, Turkey, 2024/543). Informed consent was not obtained due to the retrospective nature of the study. This retrospective study included patients with degenerative and idiopathic scoliosis who underwent surgery at our clinic between 2011 and 2024. Patients were eligible if they had both preoperative and postoperative spinal CT scans and if screw malposition was detected on early postoperative follow-up CT scans.

Patient and Data Selection

Screw malposition was defined as a ≥2 mm cortical tear in the medial or lateral direction on postoperative CT images or any screw trajectory impinging on adjacent neural or vascular structures. All scans were independently evaluated by 2 spine surgeons, and discrepancies were resolved by consensus. Patients diagnosed with idiopathic or degenerative scoliosis who underwent thoracic, thoracolumbar, or lumbar transpedicular screw placement were included in the study. All surgeries were performed by a single, experienced surgeon using a freehand technique. The study cohort consisted of pediatric patients with idiopathic scoliosis and elderly patients with degenerative scoliosis. For each case, the thickness and HU values of the pedicles with screw malposition were compared with the contralateral pedicle of the same vertebra without malposition and with all control pedicles in the study. Pedicles with screw malposition were analyzed in relation to patient sex, mean age, scoliosis etiology, location relative to the curve apex, location within the scoliotic curve (convex or concave side), direction of malposition relative to the midline (medial or lateral), and pedicle thickness and HU values. Measurements of pedicle thickness and HU values were initially performed by a single neurosurgeon with more than 10 years of experience. The same measurements were then independently repeated by a second neurosurgeon with more than 5 years of experience, and no clinically significant discrepancies were identified. Therefore, the original measurements were retained as the final data set for analysis. Because only a single final set of measurements was included in the statistical analysis, intraclass correlation coefficient (ICC) values could not be calculated.

Measurement Methods

The pedicle of the malpositioned screw was evaluated on preoperative CT images. Pedicle thickness and HU values were measured on axial CT slices (Figure 1).

Figure 1.

(A) Pedicle HU measurement method, (B) Pedicle thickness measurement method (mm)

Evaluators were provided with a detailed measurement guide defining anatomical landmarks, slice selection criteria, and ROI placement for HU and diameter measurements. The prepared measurement guide was as follows:

1. In spinal computed tomography, the pedicle measurement area should be determined from the axial image corresponding to the screw projection of the desired pedicle in sagittal sections.

2. The sclerotic structure of the bone cortex should not be included in the measurements (especially to ensure that the HU measurement represents the spongy bone, care should be taken to measure the pedicle from inside the spongy bone).

Assessment Criteria:

• Position of the malpositioned screw relative to the apex (above, at, or below the apex)

• Comparison of pedicle thickness and HU values of malpositioned screws with control pedicles

• Differences in these parameters between degenerative and idiopathic scoliosis groups

Statistical Analysis

All statistical analyses were performed using SPSS version 22 (SPSS Inc., Chicago, IL, USA). Data normality was assessed using the Shapiro–Wilk test. Comparisons of categorical variables were conducted using Pearson’s chi-square test or Fisher’s exact test, as appropriate. Continuous variables that were not normally distributed were compared between groups using the Mann–Whitney U test. A P value of <0.05 was considered statistically significant.

A power analysis based on the observed AUC of 0.796 demonstrated that a minimum of 23 patients would be required to achieve 80% power at α = 0.05; therefore, our cohort of 121 patients was adequately powered for ROC analysis.

Given the exploratory nature of subgroup analyses, no formal adjustment for multiple comparisons was applied. Subgroup findings are interpreted with caution; these analyses were performed to assess the consistency and potential generalizability of the HU threshold (≥479) across anatomical contexts.

Inter-measurement reliability was assessed using the Intraclass Correlation Coefficient (ICC), calculated using a two-way random effects model. The absolute agreement approach was adopted in the analysis, and single-measure ICC values were reported along with 95% confidence intervals (CI). ICC values were interpreted as: <0.50 low, 0.50-0.75 moderate, 0.75-0.90 good, and >0.90 excellent reliability. Interobserver measurement agreement was evaluated using Bland–Altman analysis. In this analysis, the mean difference (bias) and 95% agreement limits were calculated. The presence of proportional bias was examined by evaluating the relationship between the differences and the magnitude of the measurements in the Bland–Altman plots.

Results

A total of 121 patients were included in our study. The mean age was 45.31 ± 23.04 (range, 9-78). The mean age in the degenerative group was 63.30 (range, 40-74), while the mean age in the idiopathic group was 18.24 (range, 9-24). Of the patients included in the study, 53.7% (n: 65) were female and 46.2% (n: 56) were male. The gender distribution in the degenerative group was 48.6% (n: 35) male and 51.4% (n: 37) female, while in the idiopathic group, this ratio was 42.9% (n: 21) male and 57.1% (n: 28) female. No significant gender differences were found in demographic characteristics based on etiology in our study (Table 1). Of the patients participating in the study, the mean HU value of the pedicles in the degenerative group was 343.5 (122.5-513.25), while the mean HU value of the pedicles in the idiopathic group was 692 (499-855.5). The mean Cobb angle values were 29.5 (25.75-32) degrees in the degenerative group, while this value was 42 (36-47) degrees in the idiopathic group. The mean Cobb angle values in the idiopathic group and the HU values of the pedicles in the idiopathic group were statistically significant and higher compared to the degenerative group (Mann-Whitney-U, <0.001) (Table 1). While the mean pedicle thickness value in the degenerative group was 5.92 (4.8-7) mm, these values were determined as 4.8 (3.85-6.11) mm in the idiopathic group; The thickness of the pedicles in the degenerative group was statistically significant and higher compared to the idiopathic group (Table 1).

Table 1.

Demographic and Clinical Characteristics According to Etiology

Variable	Degenerative n (%)/Median(IQR)	Idiopathic n (%)/Median(IQR)	P value
Sex
Male	35 (48.6)	21 (42.9)
Female	37 (51.4)	28 (57.1)	0.580
HU value	343.5 (122.5-513.25)	692 (499-855.5)	<0.001
Pedicle width (mm)	5.92 (4.8-7)	4.8 (3.85-6.11)	0.006
Cobb angle (°)	29.5 (25.75-32)	42 (36-47)	<0.001

Note. Values are presented as n (%) or median (IQR). P values were calculated using Chi-square test for categorical variables (sex) and Mann–Whitney U test for continuous variables (HU value, pedicle width, Cobb angle). Statistically significant data are indicated in bold.

The regional distribution of scoliosis in the patients included in the study was 62.8% (n: 76) thoracolumbar, 36.3% (n: 44) lumbar and 0.8% (n: 1) thoracic, while in the degenerative group this distribution was 38.8% (n: 28) thoracolumbar, 61.1% (n: 44) lumbar region. In the idiopathic group, 97.9% (n: 48) thoracolumbar and 2.04% (n: 1) thoracic region.

The distribution of malpositioned screws in the degenerative and idiopathic groups was as follows; in the degenerative group, there was 48.6% (n: 35) convex, 51.3% (n: 37) concave screw malposition, while in the idiopathic group, there was 53.06% (n: 26) convex, 46.9% (n: 23) concave screw malposition. This distribution according to the apex of scoliosis was as follows; in the degenerative group, there were 40.2% (n: 29) malpositioned screws above the apex, 47.2% (n: 34) below the apex, 12.5% (n: 9) at the apex, while in the idiopathic group, this rate was as follows; 53.06% (n: 26) malpositioned screws above the apex, 42.8% (n: 21) below the apex, and 4.08% (n: 2) at the apex. No statistically significant difference was detected in these distributions.

When the characteristics of malpositioned screws were compared according to medial and lateral displacement, it was found that idiopathic screws were more malpositioned laterally than those with degenerative etiology (Pearson chi-square, 0.019). Those in the thoracic region were more malpositioned laterally than those in other regions (Fisher Freeman Halton test, 0.05) (Table 2). The distribution of malpositioned screws in the degenerative and idiopathic groups was not found to be statistically significant in terms of concave or convexity, apex, subapex, or supraapex (Table 2).

Table 2.

Comparison of Characteristics of Medial and Lateral Screw Malpositions

Variable	Medial n (%)	Lateral n (%)	P value
Sex
Male	21 (37.5)	35 (62.5)
Female	21 (32.3)	44 (67.7)	0.550
Etiology
Degenerative	31 (43.1)	41 (56.9)
Idiopathic	11 (22.4)	38 (77.6)	0.019
Side
Left	25 (37.9)	41 (62.1)
Right	17 (30.9)	38 (60.1)	0.423
Location relative to apex
At apex	5 (41.7)	7 (58.3)
Below apex	20 (37.7)	33 (62.3)
Above apex	17 (30.4)	39 (69.6)	0.625
Curve side
Concave	23 (39.7)	25 (60.3)
Convex	19 (30.2)	44 (69.8)	0.273
Region
Thoracolumbar	20 (26.0)	57 (74.0)
Lumbar	21 (48.1)	22 (51.2)
Thoracic	1 (100)	0 (0)	0.005

Note. Values are presented as n (%). P values were calculated using Pearson’s chi-square test for most comparisons; the Region comparison used the Fisher–Freeman–Halton test. Statistically significant data are indicated in bold.

When the pedicles of all patients were evaluated regardless of etiology, the mean HU value of the pedicles with malpositioned screws was 468 (221.5-748.5), the mean HU value of the pedicles without malpositioned screws was 279.5 (204.75-375), and the higher HU value of the pedicles with malpositioned screws was found to be statistically significant (Table 3). When grouped according to etiology, the mean HU value of the pedicles with malpositioned screws in the degenerative group was 343.5 (122.5-513.25), while this value was 250 (190.25-339.5) in the control pedicles with non-malpositioned screws. The HU values of the pedicles with malpositioned screws in the degenerative group were also statistically significantly higher compared to the control group. When the idiopathic group was evaluated within itself, the mean HU value of the pedicles with malpositioned screws was 692 (499-855.5), while the mean HU value of the pedicles without malpositioned screws was 299.5 (226.75-408). The higher HU values of the pedicles with malpositioned screws in the idiopathic group compared to the control group were found to be statistically significant (Wilcoxon, <0.001) (Table 3). Across all pedicles pooled, the mean HU was 399.6.

Table 3.

Comparison of Pedicle HU Values and Widths According to Etiology

Variable	Degenerative median (IQR)	Idiopathic median (IQR)	Total median (IQR)
HU value
Malpositioned pedicle HU value	343.5 (122.5-513.25)	692 (499-855.5)	468 (221.5-748.5)
Non-malpositioned pedicle HU value	250 (190.25-339.5)	299.5 (226.75-408)	279.5 (204.75-375)
P value	<0.001	<0.001	<0.001
Pedicle width (mm)
Malpositioned pedicle width	5.99 (4.8-7.5)	4.8 (3.85-6.11)	5.4 (4.38-6.55)
Non-malpositioned pedicle width	6.0 (5.02-7.0)	4.7 (3.77-5.68)	5.66 (4.5-6.9)
P value	0.606	0.909	0.031

Wilcoxon signed-rank test was used.

Note. Values are presented as median (IQR). Within-group comparisons between malpositioned and non-malpositioned pedicles were performed using the Wilcoxon signed-rank test (reported P values). Statistically significant data are indicated in bold.

Because the mean HU values of pedicles with malpositioned screws were statistically significantly higher when evaluated both across all patients and within groups, a predictive HU value was investigated. In this study of all pedicles, the malpositioned pedicle HU value was calculated as AUC: 0.796 (95% confidence interval: 0.717-0.876, P < 0.001) according to ROC analysis, and when analyzed with the Youden index, the predictive HU value for pedicle malpositioned screws was calculated as 479 (79.6% specificity, 73.6% sensitivity) (Figure 2).

Figure 2.

Roc curve of pedicle HU value of malpositioned screw

Our measurement technique was evaluated in independent data for Intraclass Correlation Coefficient (ICC) assessment. Measurements were performed by 2 independent neurosurgeons with 10 years of experience and repeated after 2 weeks. For the first evaluator, the intra-observer ICC value for the HU value measurement was found to be 0.992 (95% CI: 0.986-0.996), and for the second evaluator, the intra-observer ICC value was 0.997 (95% CI: 0.995-0.998). The inter-observer ICC value for the first HU value was found to be 0.982 (95% CI: 0.969-0.990). The inter-observer ICC value for the second HU value was found to be 0.992 (95% CI: 0.985-0.995). For the first evaluator, the intra-observer ICC value for pedicle thickness (mm) was found to be 0.968 (95% CI: 0.945-0.982), and for the second evaluator, the inter-observer ICC value was 0.992 (95% CI: 0.985-0.995). The intra-observer ICC value was determined to be 0.930 (95% CI: 0.880-0.960). For the first pedicle thickness (mm) measurement, the inter-observer ICC value was found to be 0.896 (95% CI: 0.824-0.940). For the second pedicle thickness (mm) measurement, the inter-observer ICC value was found to be 0.940 (95% CI: 0.897-0.965). The mean deviation between HU value measurements was found to be 9.34 (95% confidence interval: 137.01-118.33). The mean deviation between pedicle thickness (mm) measurements was found to be 0.16 (95% confidence interval: 2.06-1.74) (Figures 3-4). The ICC ranges of our study are between good and excellent.

Figure 3.

Bland–Altman plot for pedicle thickness (mm) measurement

Figure 4.

Bland–Altman plot for HU measurement

Discussion

In this retrospective study, which included patients with idiopathic and degenerative scoliosis who were operated on by a single surgeon and whose screw malposition was detected on routine postoperative spinal CT scans, our primary finding was that higher pedicle HU values were associated with an increased risk of pedicle screw malposition during freehand scoliosis surgery, with a predictive threshold of ≥479 (AUC 0.796) obtained from ROC analysis. This threshold remained consistent across anatomic and etiologic subgroups, supporting its potential utility as a practical preoperative marker for identifying high-risk pedicles. Clinically, preoperative identification of pedicles with a HU ≥479 can help surgeons anticipate technical difficulties during free screw placement. This threshold can guide adjustments in entry point selection, trajectory planning, guiding technique, or selective use of fluoroscopy/navigation in high-risk pedicles. Therefore, HU-based risk estimation has the potential to improve intraoperative safety, reduce malposition rates, and contribute to personalized deformity surgery planning. In practical application, basing pedicle screw placement decisions on pedicles on preoperative spinal CT scans is an idea inspired by the work of Gertzbein and Robbins, who pioneered this study and demonstrated the validity and reproducibility of pedicle assessment on spinal CT.³⁶

When the pedicle characteristics of the malpositioned screws were evaluated together with their thickness, location relative to the scoliosis (e.g., convex vs concave side; supra-apical vs infra-apical vs apical), and scoliosis characteristics (idiopathic, degenerative, Cobb angles, etc.), no significant differences were detected, making the obtained information clinically applicable to other patients. When pedicle thickness was evaluated in all patients, pedicle thicknesses were significantly less in the idiopathic group. However, when analyzed within each group, there was no statistically significant difference in thickness between the malpositioned and control pedicles (Table 3). When all pedicles, including controls, were evaluated, idiopathic scoliosis patients showed significantly higher HU values but significantly lower pedicle thicknesses. This finding, consistent with previous literature,^24,25,28 may be attributed to the younger age of idiopathic scoliosis patients and the absence of significant bone mineral density loss. In degenerative scoliosis, age-related bone loss and degenerative trabecular changes appear to reduce HU values.^{27,30,32,34,35} Although the mean Cobb angle was higher in the idiopathic group, no significant difference in Cobb angle was observed between patients with malpositioned screws and controls. Contrary to some reports,^25,27 increasing the Cobb angle was not associated with a higher risk of malposition in our series. Similarly, screw malposition was not significantly associated with being located at, above, or below the curve apex, likely reflecting the effect of all procedures being performed by a single experienced surgeon.

One notable finding was the predominance of screw malposition in the thoracic region and a tendency toward lateral breaches. A majority of patients had thoracolumbar scoliosis (63.6%). All idiopathic scoliosis patients (100%, n = 49) underwent surgery involving the thoracolumbar region, compared with 38.8% (n = 28) of degenerative scoliosis patients. In the idiopathic group, the higher frequency of malpositioned screws in the thoracic region was considered related to differences in stabilization site distribution between groups. The lateral bias in the idiopathic group may be explained by the surgeon’s deliberate avoidance of medial penetration in the thoracic region. In a previously published study by the same surgeon, no postoperative complications were observed with lateral pedicle wall breaches, whereas medial breaches resulted in neurological complications in 44.4% of cases (radicular pain and motor weakness).²⁶ This tendency to avoid the medial wall may have contributed to the increased lateral malposition rate.

Previous scoliosis studies have shown that one pedicle of a given vertebral body may exhibit denser cancellous bone, while the contralateral pedicle may exhibit softer or more normal trabecular structure.^27-32 This is primarily explained by Wolff’s law, which states that bone tissue remodels in response to mechanical loading and that changes in bone mass distribution occur to adapt to the load without microdamage or fracture.^23-38 In our study, we did not attempt to validate our findings with Wolff’s law. Conversely, if Wolff’s law were the dominant mechanism, we would expect increased pedicle density on the convex side and, consequently, a higher concentration of malposition on the convex side.^37-39 However, neither the convex-concave comparison nor the distribution based on the apex (above/beside/below) showed significant differences in our cohort. In the entire sample, HU values were significantly higher in malpositioned pedicles compared to non-malpositioned controls (Table 3). These observations suggest that Wolff’s law alone is insufficient to explain this phenomenon. Instead, we propose that localized sclerotic pedicles (HU ≥479) may impede the smooth advancement of the pedicle probe and make it difficult to reach the vertebral body; even if the initial trajectory is correct, the increased resistance may cause slight stripping or reorientation during free screw insertion, predisposing to malposition. Subgroup comparisons (apex location, concavity/convexity, pedicle thickness) were intentionally conducted to test whether the HU cutoff value (≥479) retains its validity in different anatomical contexts. The absence of significant subgroup effects supports the robustness and potential generalizability of this cutoff value for routine application.

The HU values obtained in our study and our clinical experience should not be compared to the better implant retention or lower risk of loosening found in previous studies with higher HU.^30-35 The final point here is initial technical accuracy: in very dense (sclerotic) pedicles, advancement of the probe along the ideal cancellous trajectory may be more difficult and may predispose to slight stripping or redirection despite an accurate entry point; these mechanisms are more relevant to freehand techniques and can be mitigated by navigation.^10-14 Therefore, we interpret the HU threshold as a warning sign of intraoperative difficulty, not as a marker of poor fixation quality.

Importantly, most HU loosening studies measure HU from the vertebral body and assess postoperative stability (looseness)^15-35; However, this study targeted an intraoperative endpoint (initial malposition) and measured HU specifically within the pedicles, that is, in the actual transpedicular corridor used during screw placement. The numerical HU values obtained in this article and their clinical interpretations are not directly comparable across these literatures.^15-35 Therefore, our findings complement, rather than contradict, previous studies focusing on loosening by highlighting a different mechanism and measurement domain related to freehand trajectory control.

Conclusion

This is an original study evaluating the relationship between pedicle characteristics and screw malposition during transpedicular screw placement in scoliosis surgery. In the literature, HU values are generally examined for the identification of osteoporotic vertebrae or the malposition of fusion instruments such as intervertebral cages. However, a specific threshold value for using pedicle HU to predict the risk of malposition during transpedicular screw placement has not been previously reported in the literature.

Pedicle HU ≥479 predicts increased malposition risk during freehand scoliosis surgery (AUC 0.796) and appears applicable across subgroups. Surgeons should consider added trajectory safeguards when HU is high; external validation is warranted.

Limitations

This study has limitations. Its retrospective design is a primary limitation. While the fact that all surgeries were performed by a single experienced surgeon ensures procedural consistency, it may limit the generalizability of the results to less experienced operators or different surgical techniques.

Footnotes

ORCID iD

Derya Karaoglu Gundogdu

Ethical Considerations

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (Ethics Committee of the Charité University Hospital in Berlin, Germany; EA4/046/16) and with the Declaration of Helsinki (1964) and its later amendments or comparable ethical standards.

Author Contributions

Concept and design: DKG, EK, FH, MS. Data collection: DKG, MS, SUA, FMÖ, EYO, MSŞ, NR. Data interpretation: DKG, EK, FH, HK. Statistical analysis: DKG. Writing of the manuscript: DKG, NR. Critical revision of the manuscript for important intellectual content: all authors.

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.

References

Skinner

Maybee

Transfeldt

Venter

Chalmers

. Experimental pullout testing and comparison of variables in transpedicular screw fixation. A biomechanical study. Spine. 1990;15:195-201. doi:10.1097/00007632-199003000-00007

Boucher

. A method of spinal fusion. J Bone Joint Surg Br. 1959;41:248-259. doi:10.1302/0301-620X.41B2.248

Myers

Belmont

PJJ

Richardson

Harper

Nightingale

. The role of imaging and in situ biomechanical testing in assessing pedicle screw pull-out strength. Spine. 1996;21:1962-1968. doi:10.1097/00007632-199609010-00004

Ebraheim

Darwich

Yeasting

. Anatomic relations between the lumbar pedicle and the adjacent neural structures. Spine. 1997;22:2338-2341. doi:10.1097/00007632-199710150-00003

Ashman

Galpin

Corin

Johnston

. Biomechanical analysis of pedicle screw instrumentation systems in a corpectomy model. Spine. 1989;14:1398-1405. doi:10.1097/00007632-198912000-00019

Yadla

Maltenfort

Ratliff

Harrop

. Adult scoliosis surgery outcomes: a systematic review. Neurosurg Focus. 2010;3:28. doi:10.3171/2009.12.FOCUS09254

Kosmopoulos

Schizas

. Pedicle screw placement accuracy: a meta-analysis. Spine. 2007;32:111-120. doi:10.1097/01.brs.0000254048.79024.8b

Halvorson

Kelley

Thomas

Whitecloud

3rd Cook

. Effects of bone mineral density on pedicle screw fixation. Spine. 1994;19:2415-2420. doi:10.1097/00007632-199411000-00008

Löffler

Sollmann

Burian

, et al. Opportunistic osteoporosis screening reveals low bone density in patients with screw loosening after lumbar semi-rigid instrumentation: a case-control study. Front Endocrinol. 2021;11:552719. doi:10.3389/fendo.2020.552719

10.

Gelalis

Paschos

Pakos

, et al. Accuracy of pedicle screw placement: a systematic review of prospective in vivo studies comparing freehand, fluoroscopy guidance and navigation techniques. Eur Spine J. 2012;21(2):247-255. doi:10.1007/s00586-011-2011-3

11.

Kostas

Zimmermann

Pneumaticos

Schizas

Bouillon

. Incidence and risk factors for the misplacement of pedicle screws in scoliosis constructs: a systematic analysis of 1662 screws. Eur Spine J. 2017;26(4):1009-1017. doi:10.1007/s00586-016-4833-4

12.

Kotil

Akçetin

Sucessu

. Computer tomography assessment of pedicle screw placement in the thoracic spine: is it safe to place the screw without navigational guidance? J Spinal Disord Tech. 2011;24(1):1-8. doi:10.1097/BSD.0b013e3181dc9b10

13.

Verma

Krishan

Haendlmayer

Mohsen

. Functional outcome of computer-assisted spinal pedicle screw placement: a systematic review and meta-analysis of 23 studies including 5,992 pedicle screws. Eur Spine J. 2010;19(3):370-375. doi:10.1007/s00586-009-1258-4

14.

Lonstein

Denis

Perra

Pinto

Smith

Winter

. Complications associated with pedicle screws. J Bone Joint Surg Am. 1999;81(11):1519-1528. doi:10.2106/00004623-199911000-00003

15.

Choi

Kim

Lim

. Diagnostic efficacy of hounsfield units in spine CT for the assessment of real bone mineral density of degenerative spine: correlation study between T-Scores determined by DEXA scan and hounsfield units from CT. Acta Neurochir. 2016;158:1421-1427. doi:10.1007/s00701-016-2821-5

16.

Zou

Jiang

Zhou

, et al. Prevalence of osteoporosis in patients undergoing lumbar fusion for lumbar degenerative diseases: a combination of DXA and hounsfield units. Spine. 2020;45:406-410. doi:10.1097/BRS.0000000000003284

17.

Wang

Zou

Sun

Wang

Ding

. Hounsfield unit for assessing vertebral bone quality and asymmetrical vertebral degeneration in degenerative lumbar scoliosis. Spine. 2020;45:1559-1566. doi:10.1097/BRS.0000000000003639

18.

Amin

MFM

Zakaria

WMW

Yahya

. Correlation between hounsfield unit derived FromHead, thorax, abdomen, spine and pelvis CT and T-Scores from DXA. Skelet Radiol. 2021;50:2525-2535. doi:10.1007/s00256-021-03801-z

19.

Colantonio

Saxena

Vanier

Rodkey

Tintle

Wagner

. Cervical spine computed tomography hounsfield units accurately predict low bone mineral density of the femoral neck. Clin Spine Surg. 2020;33:58-62. doi:10.1097/BSD.0000000000000879

20.

Schreiber

Anderson

Rosas

Buchholz

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

21.

Liang

Zhao

Zhou

. Hounsfield unit for assessing bone mineral density distribution within cervical vertebrae and its correlation with the intervertebral disc degeneration. Front Endocrinol. 2022;13:899705. doi:10.3389/fendo.2022.899705

22.

Zhou

Wang

Zhou

Wang

. Hounsfield unit value on CT as a predictor of cage subsidence following stand-alone oblique lumbar interbody fusion. BMC Muscoskelet Disord. 2021;22:23. doi:10.1186/s12891-020-03887-z

23.

Ryoo

Chai

Hong

Choi

Yoo

Chae

. CT hounsfield unit and histogram analysis for differentiation of recent versus remote vertebral compression fractures. Br J Radiol. 2021;94:20210941. doi:10.1259/bjr.20210941

24.

Jin

Liu

. Reliability of hounsfield unit for assessing asymmetrical vertebral bone mass in adult degenerative scoliosis. Int J Gen Med. 2022;15:5869-5877. doi:10.2147/IJGM.S368718

25.

Cheng

Liu

, et al. Hounsfield unit for assessing asymmetrical loss of vertebral bone mineral density and its correlation with curve severity in adolescent idiopathic scoliosis. Front Surg. 2022;9:842740. doi:10.3389/fsurg.2022.842740

26.

Koktekir

Ceylan

Tatarli

Karabagli

Recber

Akdemir

. Accuracy of fluoroscopically-assisted pedicle screw placement: analysis of 1,218 screws in 198 patients. Spine J. 2014;14:1702-1708. doi:10.1016/j.spinee.2014.03.044

27.

Jin

Liu

. Reliability of hounsfield unit for assessing asymmetrical vertebral bone mass in adult degenerative scoliosis. Int J Gen Med. 2022;15:5869-5877. doi:10.2147/IJGM.S368718

28.

Inagaki

Tanaka

Udaka

Akiyama

Matsuoka

Saito

. Distribution of hounsfield unit values in the pelvic bones: a comparison between young men and women with traumatic fractures and older men and women with fragility fractures: a retrospective cohort study. BMC Muscoskelet Disord. 2022;23:305. doi:10.1186/s12891-022-05263-3

29.

Mertens

Hecht

Bauknecht

Vajkoczy

. The use of intraoperative CT hounsfield unit values for the assessment of bone quality in patients undergoing lumbar interbody fusion. Glob Spine J. 2023;13(8):2218-2227. doi:10.1177/21925682221078239

30.

Zhang

Xie

Song

. The preoperative hounsfield unit value at the position of the future screw insertion is a better predictor of screw loosening than other methods. Eur Radiol. 2023;33:1526-1536. doi:10.1007/s00330-022-09157-9

31.

Ishikawa

Katsumi

Mizouchi

Sato

Yamazaki

. Importance of computed tomography hounsfield units in predicting S1 screw loosening after Lumbosacral fusion. J Clin Neurosci. 2023;113:1-6. doi:10.1016/j.jocn.2023.04.019

32.

Lee

. Value of computed tomography hounsfield units in predicting pedicle screw loosening in the thoracic spine. Sci Rep. 2022;12:18279. doi:10.1038/s41598-022-23142-8

33.

Sun

Jiang

, et al. A study of screw placement to obtain the optimal pull-out resistance of lumbar pedicle screws—analysis of hounsfield units measurements based on computed tomography. BMC Muscoskelet Disord. 2022;124:2-11. doi:10.1186/s12891-022-05074-6

34.

Zhao

Wang

, 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

35.

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;32:716-721. doi:10.3171/2019.11.SPINE19868

36.

Gertzbein

Robbins

. Accuracy of pedicular screw placement in vivo. Spine. 1990;15(1):11-14. doi:10.1097/00007632-199001000-00004

37.

Frost

. Bone’s mechanostat: a 2003 update. Anat Rec Part A. 2003;275(2):1081-1101. doi:10.1002/ar.a.10119

38.

Turner

. Three rules for bone adaptation to mechanical stimuli. Bone. 1998;23(5):399-407. doi:10.1016/S8756-3282(98)00118-5

39.

Barak

Lieberman

Hublin

. A wolff in sheep’s clothing: trabecular bone adaptation in response to changes in joint loading orientation. Bone. 2011;49(6):1141-1151. doi:10.1016/j.bone.2011.08.020