Abstract
Study Design
Retrospective cohort study.
Objectives
To determine whether preoperative cervical bone quality assessed by computed tomography Hounsfield units (CT HU) predicts pseudarthrosis, cage subsidence, or loss of cervical lordosis after anterior cervical discectomy and fusion (ACDF) and to identify perioperative predictors.
Methods
Consecutive patients who underwent 1-3-level ACDF with ≥5-year follow-up were reviewed. Outcomes included neck/arm pain visual analog scale scores, Neck Disability Index, and C2-C7 lordosis and segmental height. Bone quality was assessed by preoperative CT HU. Subgroup comparisons, multivariable regression, and receiver operating characteristic (ROC) analyses evaluated predictors.
Results
Among 253 patients, subsidence, pseudarthrosis, and loss of lordosis occurred in 15 (5.9%), 26 (10.3%), and 41 (16.2%), respectively. Index HU and clinical outcomes did not differ between subgroups defined by subsidence or pseudarthrosis. HU ≤ 231.5 was associated with a higher incidence of loss of lordosis. On regression, higher BMI and greater segmental lengthening independently predicted subsidence, while multilevel surgery predicted pseudarthrosis. Greater preoperative C2-C7 lordosis predicted loss of lordosis; very low HU (≤231.5) was significant in univariable analysis but not after multivariable adjustment. On ROC analysis, Index HU did not predict subsidence (AUC: 0.511), pseudarthrosis (AUC: 0.414), or loss of lordosis (AUC: 0.551). In contrast, segmental lengthening predicted subsidence (AUC 0.695) with threshold around 3 mm (3.12 mm).
Conclusion
Preoperative cervical CT HU did not predict pseudarthrosis or subsidence over ≥5 years after ACDF; low HU showed an unadjusted association with loss of lordosis. Bone quality alone may be insufficient to justify avoiding ACDF. Limiting segmental lengthening to ∼3 mm may mitigate subsidence risk.
Keywords: anterior cervical discectomy and fusion, osteoporosis, bone quality, computed tomography Hounsfield unit, cage subsidence, pseudarthrosis, complication
Introduction
Anterior cervical discectomy and fusion (ACDF) is a commonly performed surgical procedure for degenerative cervical radiculopathy and myelopathy. Pseudarthrosis and cage subsidence are common complications following ACDF. These complications may result in suboptimal clinical outcomes and frequently necessitate revision surgeries.1,2
Low bone mineral density (BMD) is a well-recognized risk factor for implant failure and subsidence in lumbar spine fusion.3-6 However, BMD measured by dual-energy X-ray absorptiometry (DXA) has limitations because degenerative changes, such as osteophytes, can artificially increase the measured values. 7 Computed tomography–derived Hounsfield units (CT HU) measured directly within the vertebral body have emerged as a practical surrogate marker of bone quality8-11 and are increasingly used to predict postoperative complications.12-14
However, the extent to which poor bone quality predisposes to complications such as pseudarthrosis, cage subsidence, and loss of lordosis after cervical surgery remains unclear. Several recent studies have reported that osteoporosis or osteopenia do not significantly predict pseudarthrosis or subsidence following ACDF.15-17 Cadaveric studies have demonstrated substantial biomechanical differences between cervical and lumbar vertebrae, particularly in trabecular architecture and distribution.18,19 Therefore, the relationship between bone quality and complications after ACDF may differ from that observed in lumbar spine surgery.
This study aimed (1) to determine whether poor cervical bone quality, assessed by CT HU, is associated with increased risk of pseudarthrosis, subsidence, or loss of lordosis following ACDF; and (2) to identify perioperative factors associated with these complications.
Methods
Study Design and Population
This retrospective study was approved by the Institutional Review Board of the participating institution (2025-07-009), and informed consent was waived because of the study’s retrospective design. We reviewed 1231 patients (age range, 27-82 years) who underwent one- to three-level ACDF performed by a single surgeon at a single center between February 2006 and January 2020. Inclusion criteria were: (1) a diagnosis of degenerative cervical disease (eg, degenerative disc disease, cervical spondylosis) causing radiculopathy, myelopathy, or myeloradiculopathy; (2) receipt of one- to three-level plated ACDF; (3) availability of preoperative CT suitable for HU measurement; and (4) a minimum follow-up of 5 years. Exclusion criteria were: (1) prior cervical surgery; (2) concomitant procedures such as posterior fusion, vertebral body sliding osteotomy, or corpectomy; (3) ACDF using a cage alone (without plate fixation); (4) surgery for tumor, trauma, or infection. Based on these criteria, 253 patients were included in the final analysis (Figure 1).
Figure 1.
Flow diagram of patient selection for the study
Surgical Technique
One surgeon performed all operations at a single center. The patient was positioned supine, and a standard left-sided Smith–Robinson approach was used to remove the intervertebral disc and osteophytes. Minimal decortication of the superior and inferior endplates was performed to minimize the risk of cage subsidence. Interbody cages were inserted at each level and packed with local autograft harvested during osteophytectomy and endplate preparation. When arm pain caused by foraminal stenosis was present, an additional uncoforaminotomy was performed. All operated levels were stabilized with an anterior cervical plate and variable-angle screw constructs to promote postoperative graft compression and gradual settling. The cage was positioned 2-3 mm posterior to the anterior vertebral body margin. A cervical collar was worn postoperatively for 6-12 weeks.20-22
Variables
Patient-reported outcomes included neck pain visual analog scale (VAS), arm pain VAS, and the Neck Disability Index (NDI) scores, which were assessed preoperatively, at 2 years postoperatively, and at the final follow-up (≥5 years). Neck and arm pain were measured using a 10-point VAS.
Radiographic outcomes were evaluated across three domains at predefined time points. The literature employs varying definitions of cage subsidence, often based on reductions in intervertebral disc height.23,24 Previous studies have demonstrated that changes in fused segmental body height correlate strongly with true subsidence 25 ; therefore, subsidence was defined as a >3-mm reduction in segmental height at the fused level(s), consistent with previous studies.26-29 Segmental height was measured on 150%-magnified neutral lateral radiographs obtained preoperatively, within 1 week postoperatively, and at the final follow-up. 16 Immediate postoperative segmental lengthening (increase from preoperative to immediate postoperative values) was prespecified as a surrogate of intraoperative distraction, and cage subsidence as the decrease from immediate postoperative to final follow-up16,30 (Table 1) (Figure 2).
Table 1.
Variables Definitions and Assessment Timings
| Variables | Definition | Timing |
|---|---|---|
| Bone quality (index HU) | CT-measured mean vertebral HU at the index level | Preoperative |
| Segmental lengthening | Immediate postoperative increase of segmental height (mm) | Preoperative–Immediate postoperative |
| Subsidence | ≥3 mm loss of segmental height | Immediate postoperative–Final |
| Pseudarthrosis | Non-union confirmed by ISM or bridging bone | Final (5-year) |
| Loss of lordosis | ≥5° loss of C2-C7 lordosis | Preoperative–Final |
| Revision | Reoperation due to pseudarthrosis or mechanical failure | ≥ Final follow-up |
HU, Hounsfield unit; CT, computed tomography; ISM, interspinous motion.
Figure 2.
Radiographic measurement of segmental height, segmental lengthening, and subsidence.
A 66-year-old woman underwent 2-level ACDF (C5-6 and C6-7) (A) Preoperative neutral lateral radiograph showing the segmental height (SH, white dashed arrow), measured at 30.3 mm. (B) Immediate postoperative neutral radiograph; SH 34.5 mm, indicating segmental lengthening 4.2 mm. (C) Final follow-up neutral radiograph; SH 30.9 mm, indicating subsidence 3.6 mm
Pseudarthrosis was defined using two complementary criteria, consistent with prior validated methods20,30-34: (1) interspinous motion (ISM) >1 mm on 150%-magnified dynamic lateral radiographs (flexion–extension), or (2) absence of bridging bone across the graft connecting adjacent endplates or outside the graft on CT. ISM and CT bridging were evaluated at 2 years and at the final follow-up, and union status or pseudarthrosis was recorded at each time point (Table 1) (Figure 3).
Figure 3.
Evaluation of pseudarthrosis using CT and dynamic radiographs.
(A and B) 5-year postoperative CT (sagittal and coronal) showing absence of a bridging bone across the graft (black arrows), consistent with nonunion. (C and D) Final flexion–extension lateral radiographs showing the C6-7 interspinous distance (white dashed arrows): 28.6 mm in flexion and 20.9 mm in extension, yielding an interspinous motion of 7.7 mm, supporting pseudarthrosis
Cervical alignment was assessed as C2-C7 lordosis on neutral lateral radiographs obtained preoperatively, within 1 week postoperatively, and at the final follow-up. Loss of lordosis was defined as a decrease of ≥5° at the final follow-up relative to the preoperative value35,36 (Table 1) (Figure 2).
Cervical bone quality (Index HU) was derived from preoperative noncontrast CT by measuring HU on mid-sagittal vertebral body reformats using small trabecular regions of interest (ROIs) that excluded the cortical rim, endplate sclerosis, and osteophytes. HU values were obtained from each operative-level vertebral body, and the Index HU was defined as the mean of these values. For multilevel surgeries, the Index HU represented the average HU across all vertebral bodies included in the fusion construct 25 (Table 1) (Figure 4).
Figure 4.
Measurement of preoperative bone quality using Hounsfield units (HU).
Representative mid-sagittal noncontrast CT image showing trabecular regions of interest (ROIs; blue circles) placed within the vertebral body while excluding the cortex, endplate sclerosis, and osteophytes. HU values were obtained from each operative-level vertebral body, and the Index HU was calculated as the mean HU across operative levels (representative measurement output shown)
Two fellowship-trained attending spine surgeons independently performed all radiographic measurements to assess interobserver reliability, with intraclass correlation coefficients (ICCs) ranging from 0.842 to 0.953.
Statistical Analysis
The normality of continuous variables was assessed using the Shapiro–Wilk test. Subgroup analyses were conducted based on the presence or absence of subsidence, pseudarthrosis, and loss of lordosis at the final follow-up, using independent t-tests for continuous variables and chi-square tests for categorical variables. Intergroup analyses also used published CT HU thresholds for cervical osteoporosis—231.5 in a Korean cohort and 326.5 in a U.S. cohort.37,38
We conducted univariable and multivariable regression analyses to identify predictors of complications (subsidence, pseudarthrosis, and loss of lordosis). Candidate covariates included demographic variables and relevant preoperative and intraoperative factors. Variables with a screening P-value of ≤.100 in univariable analyses were entered into multivariable models.
Receiver operating characteristic (ROC) curves were constructed to evaluate the ability of the preoperative Index HU to predict pseudarthrosis, cage subsidence, and loss of lordosis. For each outcome, the area under the curve (AUC) and 95% confidence intervals (CI) were calculated. The optimal cutoff was determined by maximizing the Youden index (sensitivity + specificity − 1), and the corresponding sensitivity and specificity were reported. An identical ROC analysis was performed for postoperative segmental lengthening. All analyses were performed using IBM SPSS Statistics, version 29.0 (IBM Corp., Armonk, NY, USA), with statistical significance set at P < .05.
Results
Baseline Patient Characteristics
Among 253 patients, the distribution of operated levels was 1-level in 113 patients (44.7%), 2-level in 98 (38.7%), and 3-level in 42 (16.6%). The mean Index HU was 324.67 ± 77.07. At the final follow-up, subsidence occurred in 15 patients (5.9%). Pseudarthrosis was identified in 79 patients (31.2%) at 2 years; 67.1% (53/79) subsequently achieved delayed union, resulting in a final pseudarthrosis rate of 26 patients (10.3%) (Figure 5). Loss of lordosis was observed in 41 patients (16.2%). Subgroup analysis by subsidence status showed no significant difference in Index HU between groups (subsidence: 328.85 ± 70.76 vs no subsidence: 324.40 ± 77.58; P = .829) or in other demographic factors. Similarly, when stratified by pseudarthrosis status at the final follow-up, Index HU values did not differ significantly (pseudarthrosis: 307.53 ± 87.77 vs union: 323.63 ± 75.72; P = .232). When stratified by loss of lordosis, the mean Index HU was comparable between groups (P = .269). Notably, the proportion of patients with HU ≤ 231.5 was higher in the loss subgroup than in the no loss of lordosis subgroup (19.5% vs 8.5%; P = .047). The pseudarthrosis subgroup had a higher body mass index (BMI) (P = .047) and a greater proportion of smokers (P = .036). The distribution of underlying pathology differed between groups (P = .035), and multilevel procedures were significantly more frequent in the pseudarthrosis subgroup than in the union subgroup (P < .001) (Table 2, Supplemental Table 1).
Figure 5.
Change in union status from 2 years to ≥5 years after ACDF.
At 2 years, the union rate was 68.8% and the pseudarthrosis rate was 31.2%. By ≥5 years, 67.1% of cases with pseudarthrosis at 2 years had converted to union (delayed union = 20.9% overall), with persistent pseudarthrosis in 10.3% of patients. The dotted connector indicates the transition of the 2-year pseudarthrosis subgroup
Table 2.
Baseline Characteristics of Included Patients and Subgroup Comparisons by Subsidence and Pseudarthrosis at Final Follow-Up
| Characteristics | Included patients (n = 253) | Subsidence subgroup | Pseudarthrosis subgroup | ||||
|---|---|---|---|---|---|---|---|
| Subsidence (n = 15) | No subsidence (n = 238) | P value | Pseudarthrosis (n = 26) | Union (n = 227) | P value | ||
| Age (year) | 57.23 ± 11.20 | 61.53 ± 11.02 | 56.95 ± 11.18 | .125 | 57.85 ± 9.05 | 57.15 ± 11.43 | .722 |
| Sex (Female) | 120 (47.4%) | 6 (40.0%) | 114 (47.9%) | .552 | 10 (38.5%) | 110 (48.5%) | .334 |
| BMI (kg/m2) | 24.48 ± 3.22 | 26.14 ± 5.14 | 24.37 ± 3.05 | .105 | 25.66 ± 3.72 | 24.34 ± 3.14 | .047* |
| DM | 42 (16.6%) | 3 (20.0%) | 39 (16.4%) | .721 | 4 (15.4%) | 38 (16.7%) | .560 |
| HTN | 73 (28.9%) | 5 (33.3%) | 68 (28.6%) | .770 | 11 (42.3%) | 62 (27.3%) | .110 |
| Smoking | 98 (38.7%) | 7 (46.7%) | 91 (38.2%) | .516 | 15 (57.7%) | 83 (36.6%) | .036* |
| Index HU | 324.67 ± 77.07 | 328.85 ± 70.76 | 324.40 ± 77.58 | .829 | 307.53 ± 87.77 | 323.63 ± 75.72 | .232 |
| Pathology | .605 | .035* | |||||
| Myelopathy | 118 (46.6%) | 6 (40.0%) | 112 (47.1%) | 7 (26.9%) | 111 (48.9%) | ||
| Radiculopathy | 77 (30.4%) | 5 (33.3%) | 72 (30.3%) | 10 (38.5%) | 67 (29.5%) | ||
| Myeloradiculopathy | 58 (22.9%) | 4 (26.7%) | 54 (22.7%) | 9 (34.6%) | 49 (21.6%) | ||
| Number of levels operated | .774 | <.001* | |||||
| 1–level | 113 (44.7%) | 7 (46.7%) | 106 (44.5%) | 4 (15.4%) | 109 (48.8%) | ||
| 2–level | 98 (38.7%) | 6 (40.0%) | 92 (38.7%) | 13 (50.0%) | 85 (37.4%) | ||
| 3–level | 42 (16.6%) | 2 (13.3%) | 40 (16.8%) | 9 (34.6%) | 33 (14.5%) | ||
| Follow-up period (month) | 89.25 ± 22.70 | 98.13 ± 35.26 | 88.69 ± 21.66 | .322 | 90.12 ± 25.73 | 89.15 ± 22.39 | .838 |
BMI, body mass index; DM, diabetes mellitus; HTN, hypertension; Index HU, CT-measured mean vertebral Hounsfield Unit at index level.
*P < 0.050.
Outcome Comparison Based on Complications
Clinical and radiographic outcomes were compared between subgroups stratified by each complication. Stratification by subsidence revealed no preoperative or final follow-up differences in neck pain VAS or NDI, although the subsidence group showed higher preoperative arm pain VAS (P = .006). Radiographically, C2-C7 lordosis did not differ between groups at either time point, and the proportion of patients with loss of lordosis was comparable (P = .780). Segmental height was similar between subgroups at baseline and the final follow-up; however, postoperative segmental lengthening was significantly greater in the subsidence group (2.81 ± 1.64 vs 1.59 ± 1.68 mm; P = .007). The subsidence group also showed higher rates of pseudarthrosis at final follow-up (40.0% vs 8.4%; P = .002) and revision surgery (20.0% vs 5.0%; P = .050) (Table 3).
Table 3.
Clinical and Radiologic Outcomes at Final Follow-Up After ACDF; Comparison Based on Subsidence and Pseudarthrosis
| Variables | Subsidence subgroup | Pseudarthrosis subgroup | |||||
|---|---|---|---|---|---|---|---|
| Subsidence (n = 15) | No subsidence (n = 238) | P value | Pseudarthrosis (n = 26) | Union (n = 227) | P value | ||
| Clinical outcomes | |||||||
| Neck pain VAS | Preoperartive | 4.27 ± 3.53 | 3.96 ± 2.73 | .371 | 4.65 ± 2.80 | 3.89 ± 2.77 | .187 |
| Final | 2.36 ± 2.58 | 2.43 ± 2.17 | .461 | 2.56 ± 2.41 | 2.41 ± 2.18 | .794 | |
| Arm pain VAS | Preoperartive | 6.80 ± 2.14 | 5.02 ± 2.70 | .006* | 6.35 ± 2.21 | 4.98 ± 2.72 | .014* |
| Final | 3.64 ± 2.29 | 3.45 ± 2.64 | .410 | 3.22 ± 2.34 | 3.49 ± 2.65 | .685 | |
| NDI | Preoperative | 18.60 ± 12.20 | 14.96 ± 8.28 | .096 | 17.59 ± 9.69 | 14.91 ± 8.40 | .219 |
| Final | 10.45 ± 9.09 | 8.34 ± 6.56 | .157 | 9.56 ± 8.10 | 8.35 ± 6.58 | .473 | |
| Radiologic outcomes | |||||||
| C2-C7 lordosis | Preoperartive | 5.65 ± 14.55 | 5.91 ± 11.19 | .466 | 2.54 ± 12.40 | 6.28 ± 11.23 | .113 |
| Final | 9.85 ± 12.70 | 9.97 ± 9.70 | .482 | 7.94 ± 12.53 | 10.20 ± 9.52 | .270 | |
| Loss of lordosis | 1 (6.7%) | 40 (16.8%) | .501 | 4 (16.0%) | 37 (16.2%) | 1.000 | |
| Segmental height | Preoperartive | 33.27 ± 2.36 | 32.33 ± 2.81 | .103 | 33.14 ± 2.20 | 32.30 ± 2.84 | .152 |
| Final | 32.54 ± 2.35 | 33.15 ± 2.88 | .209 | 32.97 ± 2.49 | 33.13 ± 2.89 | .786 | |
| Segmental lengthening | 2.81 ± 1.64 | 1.59 ± 1.68 | .007* | 1.59 ± 1.80 | 1.67 ± 1.69 | .826 | |
| Subsidence | Final | NA | NA | NA | 6 (23.1%) | 9 (4.0%) | .002* |
| Pseudarthrosis | Final | 6 (40.0%) | 20 (8.4%) | .002* | NA | NA | NA |
| Revision | After final | 3 (20.0%) | 12 (5.0%) | .050* | 4 (5.9%) | 11 (4.8%) | .055 |
VAS, visual analog scale; NDI, neck disability index.
*P < .050.
Stratification by pseudarthrosis revealed no between-group differences in neck pain VAS or NDI, except for higher preoperative arm pain VAS in the pseudarthrosis subgroup (P = .014). Radiographically, there were no significant between-group differences preoperatively or at the final follow-up. Subsidence occurred more frequently in the pseudarthrosis subgroup than in the union subgroup (23.1% vs 4.0%; P =.002), while revision surgery rates did not differ (P = .055) (Table 3).
When stratified by loss of lordosis, no significant between-group differences were observed in preoperative or final patient-reported outcomes (VAS and NDI). Radiographically, the loss subgroup had greater preoperative C2-C7 lordosis (16.01 ± 8.98 vs 3.94 ± 10.76; P < .001), whereas the final C2-C7 lordosis was comparable between groups (P = .061). Postoperative segmental lengthening was modestly smaller in the loss subgroup (1.14 ± 1.61 vs 1.76 ± 1.70; P = .032). Revision surgery was numerically more frequent in the loss subgroup, but the difference was not significant (12.2% vs 4.7%; P = .075) (Supplemental Table 2).
Comparison of Postoperative Complications by HU Cutoff Subgroups
Previously proposed cervical CT HU thresholds for poor bone quality (231.5 and 326.5) were used to compare complication rates across subgroups. For the 231.5 cutoff, 26 patients had HU ≤ 231.5 and 227 had HU > 231.5; by the 326.5 cutoff, 128 patients had HU ≤ 326.5 and 125 had HU > 326.5. Stratification by the 231.5 cutoff showed no significant between-group differences in subsidence, pseudarthrosis, or revision surgery rates. Loss of lordosis was more frequent in patients with HU ≤ 231.5 (30.8% vs 14.5%; odds ratio [OR]: 2.613; 95% CI: 1.051-6.497; P = .039). Stratification by the 326.5 cutoff revealed no significant between-group differences in complication rates, including loss of lordosis (Table 4).
Table 4.
Postoperative Complications According to HU Cutoffs of 231.5 and 326.5
| Characteristics | HU cutoff 231.5 | HU cutoff 326.5 | ||||
|---|---|---|---|---|---|---|
| HU ≤ 231.5 (n = 26) | HU > 231.5 (n = 227) | P value | HU ≤ 326.5 (n = 128) | HU > 326.5 (n = 125) | P value | |
| Subsidence | 1 (3.8%) | 14 (6.2%) | 1.000 | 8 (6.3%) | 7 (5.6%) | .827 |
| Pseudarthrosis | 5 (19.2%) | 20 (8.8%) | .154 | 15 (11.7%) | 10 (8.0%) | .322 |
| Loss of lordosis | 8 (30.8%) | 33 (14.5%) | .039* | 22 (17.2%) | 19 (15.2%) | .796 |
| Revision | 2 (7.7%) | 13 (5.7%) | .657 | 7 (5.5%) | 8 (6.4%) | .754 |
HU, Hounsfield Unit.
*P < .050.
Regression and ROC Analysis for Predicting Complications
Univariable and multivariable regression analyses were performed to identify predictors of complications. Higher BMI (OR: 1.194; 95% CI: 1.031-1.383; P = .018) and greater segmental lengthening (OR: 1.721; 95% CI: 1.193-2.483; P = .004) independently predicted subsidence. Multilevel surgery significantly predicted pseudarthrosis (OR: 2.585; 95% CI: 1.476-4.527; P < .001). Greater preoperative C2-C7 lordosis predicted loss of lordosis (OR: 1.122; 95% CI: 1.077-1.169; P < .001). Although very low HU (≤231.5) was significantly associated with loss of lordosis in univariable analysis (OR: 2.613; 95% CI: 1.051-6.497; P = .039), this association was not significant after multivariable adjustment (OR: 1.294; 95% CI: 0.411-4.074; P = .660). Moreover, the Index HU and HU below cutoff were not associated with subsidence or pseudarthrosis (Tables 5-7).
Table 6.
Multivariate Regression Analysis for the Predictors of Pseudarthrosis
| Predictors | Odds ratio | 95% CI | P-value |
|---|---|---|---|
| Univariate analysis | |||
| Age | 1.006 | 0.970-1.043 | .765 |
| Sex | 0.665 | 0.289-1.527 | .336 |
| Pathology | 1.694 | 1.029-2.788 | .038* |
| Operation level | 2.585 | 1.476-4.527 | <.001* |
| BMI | 1.126 | 1.000-1.269 | .050* |
| DM | 0.904 | 0.295-2.774 | .860 |
| HTN | 1.952 | 0.850-4.480 | .115 |
| Smoking | 2.366 | 1.038-5.391 | .040* |
| Preoperartive C2-C7 lordosis | 0.971 | 0.937-1.007 | .113 |
| Preoperartive SVA | 0.989 | 0.948-1.032 | .616 |
| Segmental lengthening | 0.975 | 0.769-1.236 | .834 |
| CT HU at index level | 0.997 | 0.991-1.002 | .231 |
| HU ≤ 326.5 | 1.643 | 0.715-3.774 | .242 |
| HU ≤ 231.5 | 2.336 | 0.798-6.834 | .122 |
| Multivariate analysis | |||
| Pathology | 1.540 | 0.933-2.541 | .091 |
| Operation level | 2.585 | 1.476-4.527 | <.001* |
| BMI | 1.128 | 0.993-1.281 | .063 |
| Smoking | 1.884 | 0.797-4.453 | .149 |
VAS, visual analog scale; NDI, neck disability index; CI, confidence interval; BMI, body mass index; DM, diabetes mellitus; HTN, hypertension; SVA, sagittal vertical axis.
*P < .050.
Table 5.
Multivariate Regression Analysis for the Predictors of Subsidence
| Predictors | Odds ratio | 95% CI | P-value |
|---|---|---|---|
| Univariate analysis | |||
| Age | 1.039 | 0.989-1.091 | .127 |
| Sex | 0.725 | 0.250-2.101 | .554 |
| Pathology | 1.184 | 0.625-2.243 | .605 |
| Operation level | 0.899 | 0.434-1.860 | .773 |
| BMI | 1.163 | 1.005-1.344 | .042* |
| DM | 1.276 | 0.344-4.732 | .716 |
| HTN | 1.250 | 0.412-3.792 | .694 |
| Smoking | 1.413 | 0.496-4.029 | .517 |
| Preoperartive C2-C7 lordosis | 0.998 | 0.953-1.045 | .931 |
| Preoperartive SVA | 0.988 | 0.935-1.044 | .672 |
| Segmental lengthening | 1.629 | 1.141-2.327 | .007* |
| CT HU at index level | 1.001 | 0.994-1.008 | .828 |
| HU ≤ 326.5 | 1.124 | 0.395-3.198 | .827 |
| HU ≤ 231.5 | 0.609 | 0.077-4.826 | .638 |
| Multivariate analysis | |||
| BMI | 1.194 | 1.031-1.383 | .018* |
| Segmental lengthening | 1.721 | 1.193-2.483 | .004* |
VAS, visual analog scale; NDI, neck disability index; BMI, body mass index; DM, diabetes mellitus; HTN, hypertension; SVA, sagittal vertical axis.
*P < .050.
Table 7.
Multivariate Regression Analysis for the Predictors of Loss of Lordosis
| Predictors | Odds ratio | 95% CI | P-value |
|---|---|---|---|
| Univariate analysis | |||
| Age | 1.002 | 0.973-1.033 | .882 |
| Sex | 0.750 | 0.381-1.475 | .404 |
| Pathology | 1.130 | 0.748-1.708 | .562 |
| Operation level | 0.687 | 0.422-1.120 | .132 |
| BMI | 1.043 | 0.942-1.155 | .413 |
| DM | 1.269 | 0.540-2.985 | .585 |
| HTN | 1.345 | 0.660-2.740 | .415 |
| Smoking | 1.636 | 0.835-3.207 | .152 |
| Preoperartive C2-C7 lordosis | 1.121 | 1.076-1.166 | <.001* |
| Preoperartive SVA | 0.982 | 0.947-1.017 | .304 |
| Segmental lengthening | 0.812 | 0.669-0.985 | .035* |
| CT HU at index level | 0.997 | 0.993-1.002 | .268 |
| HU ≤ 326.5 | 1.158 | 0.592-2.263 | .668 |
| HU ≤ 231.5 | 2.613 | 1.051-6.497 | .039* |
| Multivariate analysis | |||
| Preoperartive C2-C7 lordosis | 1.122 | 1.077-1.169 | <.001* |
| Segmental lengthening | 0.800 | 0.630-1.017 | .068 |
| HU ≤ 231.5 | 1.294 | 0.411-4.074 | .660 |
VAS, visual analog scale; NDI, neck disability index; CI, confidence interval; BMI, body mass index; DM, diabetes mellitus; HTN, hypertension; SVA, sagittal vertical axis.
*P < .050.
ROC analysis showed that preoperative Index HU did not predict subsidence (AUC: 0.511; 95% CI: 0.372-0.649; P = .882), pseudarthrosis (AUC: 0.414; 95% CI: 0.298-0.530; P = .147) or loss of lordosis (AUC: 0.551; 95% CI: 0.447-0.655; P = .338). In contrast, segmental lengthening significantly predicted subsidence (AUC: 0.695; 95% CI: 0.550-0.840; P = .008), with an optimal cutoff of 3.12 mm (Figure 6).
Figure 6.
Receiver operating characteristic curve analysis for postoperative segmental lengthening.
Receiver operating characteristic curve analysis for segmental lengthening, identifying a cutoff value of 3.12 mm for predicting cage subsidence at final follow-up. The area under the curve (AUC) was 0.695 (95% confidence interval, 0.550–0.840; P = .008). At the optimal cutoff determined by the Youden index, sensitivity was 48.5% and specificity was 85.5%
Discussion
This study assessed whether complications after ACDF—pseudarthrosis, subsidence, and loss of lordosis—were predicted by preoperative bone quality measured by CT HU over at least 5 years of follow-up. Cervical bone quality did not predict pseudarthrosis or subsidence, but lower HU values were associated with postoperative loss of cervical lordosis. Immediate postoperative segmental lengthening—used as a surrogate for intraoperative distraction—was associated with subsidence; a threshold of approximately 3 mm may guide operative planning. Overall, these findings highlight that differences in cervical and lumbar biomechanics warrant region-specific interpretation of bone quality.
Although lumbar fusion studies consistently linked osteoporosis-related bone quality to adverse outcomes and implant complications,4-6 recent cervical studies report no significant association between bone quality and pseudarthrosis or subsidence after ACDF.15-17 These findings support this site-specific distinction. Cadaveric studies by Schröder et al showed that cervical vertebrae have thicker, denser trabeculae and less age-related trabecular loss than lumbar counterparts.18,19 MRI-based vertebral bone quality studies similarly reveal distinct bone-quality profiles between the cervical and lumbar regions. 39 These microarchitectural features likely increase load-bearing capacity per unit area, suggesting greater cervical spine resilience to postoperative mechanical failure, even in osteoporotic bone.
Mechanically, ACDF involves a short moment arm between the plate and the fusion interface (cage–endplate), and the plate acts as a tension band limiting micromotion at the interbody surface.40-42 In contrast, posterior lumbar constructs have longer distances between pedicle screw–rod anchors and the fusion segment, transmitting greater bending moments to the anterior column under similar loads.43,44 These structural differences may explain why reduced bone quality does not directly lead to higher pseudarthrosis or subsidence rates following ACDF (Figure 7).
Figure 7.
Moment-arm contrast between ACDF and posterior lumbar constructs.
(A) In ACDF, the plate–graft moment arm (d) is short, and the plate functions as a tension band that stabilizes the graft–endplate interface by limiting micromotion. (B) In posterior lumbar constructs, the screw–rod–to-graft moment arm (d) is longer, generating greater bending moments on the anterior column under equivalent loads. Dashed boxes indicate the fused level, and double-headed arrows represent the moment arm (d)
In this cohort, pseudarthrosis, subsidence, and loss of lordosis had limited effects on clinical outcomes, including pain scales and NDI. Notably, 67.1% of pseudarthroses identified at 2 years achieved delayed union by final follow-up, consistent with previous ACDF reports.20,32 Although the subsidence group had a higher revision rate, interpretation is limited by the small subgroup size (n = 15) and the few revision events. Revision rates did not differ significantly when outcomes were stratified by pseudarthrosis or loss of lordosis. Overall, the anatomic advantages of the cervical spine and the stability provided by the short moment arm in ACDF may mitigate both the incidence and clinical impact of mechanical complications, even in patients with low bone quality.
CT HU-based bone quality did not meaningfully predict pseudarthrosis or subsidence, suggesting that bone quality alone may be insufficient as a rationale to avoid ACDF. In clinical practice, low bone quality identified by screening measures (eg, BMD) should not, by itself, justify forgoing an anterior approach (potential undertreatment) or planning overly extensive combined procedures (potential overtreatment). In additional analyses, very low HU (≤231.5) was associated with an increased risk of postoperative loss of lordosis. The prevalence of HU ≤ 231.5 was higher in the loss subgroup, and a significant association was observed in univariable regression; however, this association was not retained after multivariable adjustment, suggesting limited independent predictive value. Few studies have directly linked preoperative bone-quality metrics to postoperative loss of lordosis after ACDF. These findings indicate that bone quality is more closely associated with alignment maintenance than traditional mechanical complications. Because postoperative loss of lordosis is strongly influenced by baseline alignment and extent of alignment correction, bone quality may be most appropriately interpreted alongside alignment-related variables rather than as a stand-alone metric.
Excessive segmental lengthening was associated with increased subsidence risk, consistent with previous studies30,45; this study corroborates that association in a large cohort and identifies a threshold around 3 mm (3.12 mm). Limiting segmental lengthening relative to preoperative disc height to approximately 3 mm may serve as a pragmatic guardrail to mitigate subsidence risk. Because higher BMI also independently predicted subsidence in this cohort, this threshold may be particularly relevant for patients with high BMI. However, univariable analyses suggested a directional trade-off in which smaller segmental lengthening was associated with more frequent loss of lordosis, indicating that the 3-mm threshold is best viewed as a guide to avoid excessive distraction rather than a universal rule for all patients. Accordingly, flexible application tailored to individual alignment, target correction, and clinical necessity may be required, and further studies are needed to refine distraction strategies that balance these competing risks.
Limitations
This study has several limitations. First, although the sample size was large and the follow-up was long-term, the retrospective design could not eliminate selection bias. Second, because the study was conducted at a single center, its findings may have limited generalizability. Third, although DXA-derived BMD is a widely used index of bone quality, it is not routinely obtained in cervical surgery candidates; therefore, it was not included in the analyses. Fourth, adjacent segment degeneration, an important long-term complication, was not systematically assessed, precluding evaluation of its relationship with bone quality and postoperative alignment. Finally, the study design did not allow determination of the temporal sequence among pseudarthrosis, subsidence, and loss of lordosis, precluding causal inference; future prospective studies are needed to clarify these relationships.
Conclusion
In this 5-year longitudinal cohort, preoperative CT HU did not predict pseudarthrosis or subsidence after ACDF. Very low HU (≤231.5) was associated with postoperative loss of lordosis in univariable analysis, but did not retain independent predictive value after multivariable adjustment. Notably, these complications had limited impact on 5-year patient-reported outcomes. Immediate postoperative segmental lengthening (a surrogate of intraoperative distraction), and higher BMI independently increased cage subsidence risk, whereas multilevel surgery independently predicted pseudarthrosis. These findings indicate that bone quality in the cervical spine should be interpreted differently from that in the lumbar spine, particularly with respect to mechanical failure. Accordingly, bone quality alone should not be used as a primary rationale to avoid ACDF. A pragmatic guardrail of approximately 3 mm of segmental lengthening relative to preoperative disc height may help mitigate subsidence risk.
Supplemental Material
Supplemental Material for Does Bone Quality Predict Pseudarthrosis, Subsidence, or Loss of Lordosis After Anterior Cervical Discectomy and Fusion? Minimum 5-Year Cohort Study by San Kim, MD, Sehan Park, MD, PhD, Dong Ki Ahn, MD, PhD, Dong-Ho Lee, MD, PhD, Gumin Jeong, MD, Ki Chol Park, MD, Yong Kim, MD, and Changmin Choi, MD, PhD in Global Spine Journal
Appendix.
Abbreviations
- ACDF
anterior cervical discectomy and fusion
- BMD
bone mineral density
- HU
Hounsfield unit
- VAS
visual analog scale
- NDI
Neck Disability Index.
Author Contributions (CRediT): Conceptualization: San Kim, Dong-Ho Lee.
Methodology: San Kim, Dong Ki Ahn, Ki Chol Park.
Formal analysis: San Kim, Sehan Park.
Investigation: San Kim.
Data curation: San Kim, Sehan Park, Gumin Jeong, Ki Chol Park, Yong Kim, Changmin Choi.
Validation: Sehan Park, Dong-Ho Lee.
Visualization: Changmin Choi.
Project administration: Dong Ki Ahn, Dong-Ho Lee.
Funding acquisition: San Kim.
Resources: Gumin Jeong.
Software: Yong Kim.
Supervision: Dong Ki Ahn, Dong-Ho Lee.
Writing – original draft: San Kim.
Writing – review & editing: Sehan Park, Dong-Ho Lee.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a research grant from Jeju National University Hospital in 2025 (grant number 2025-22).
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Supplemental Material: Supplemental material for this article is available online.
ORCID iDs
San Kim https://orcid.org/0000-0002-4725-9082
Sehan Park https://orcid.org/0000-0001-8959-8579
Dong-Ho Lee https://orcid.org/0000-0003-3704-6355
Gumin Jeong https://orcid.org/0000-0002-8279-3125
Ethical Considerations
This retrospective study was approved by the Institutional Review Board of the participating institution (IRB approval number: 2025-07-009).
Consent to Participate
Informed consent was waived because of the study’s retrospective design.
References
- 1.Lee YS, Kim YB, Park SW. Risk factors for postoperative subsidence of single-level anterior cervical discectomy and fusion: the significance of the preoperative cervical alignment. Spine. 2014;39(16):1280-1287. doi: 10.1097/brs.0000000000000400 [DOI] [PubMed] [Google Scholar]
- 2.Brenke C, Dostal M, Scharf J, Weiß C, Schmieder K, Barth M. Influence of cervical bone mineral density on cage subsidence in patients following stand-alone anterior cervical discectomy and fusion. Eur Spine J. 2015;24(12):2832-2840. doi: 10.1007/s00586-014-3725-9 [DOI] [PubMed] [Google Scholar]
- 3.Chin DK, Park JY, Yoon YS, et al. Prevalence of osteoporosis in patients requiring spine surgery: incidence and significance of osteoporosis in spine disease. Osteoporos Int. 2007;18(9):1219-1224. doi: 10.1007/s00198-007-0370-8 [DOI] [PubMed] [Google Scholar]
- 4.Keaveny TM, Adams AL, Fischer H, et al. Increased risks of vertebral fracture and reoperation in primary spinal fusion patients who test positive for osteoporosis by biomechanical computed tomography analysis. Spine J. 2023;23(3):412-424. doi: 10.1016/j.spinee.2022.10.018 [DOI] [PubMed] [Google Scholar]
- 5.Levy HA, Magera CA, Messer CJ, et al. Lumbar endplate Hounsfield units enhance transforaminal lumbar interbody fusion subsidence prediction compared to trabecular Hounsfield units alone. Glob Spine J. 2025;16:21925682251356986. doi: 10.1177/21925682251356986 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Löffler MT, Sollmann N, Burian E, 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 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rand T, Seidl G, Kainberger F, et al. Impact of spinal degenerative changes on the evaluation of bone mineral density with dual energy X-ray absorptiometry (DXA). Calcif Tissue Int. 1997;60(5):430-433. doi: 10.1007/s002239900258 [DOI] [PubMed] [Google Scholar]
- 8.Choi MK, Kim SM, Lim JK. 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(7):1421-1427. doi: 10.1007/s00701-016-2821-5 [DOI] [PubMed] [Google Scholar]
- 9.Ahmad A, Crawford CH, 3rd, Glassman SD, Dimar JR, 2nd, Gum JL, Carreon LY. Correlation between bone density measurements on CT or MRI versus DEXA scan: a systematic review. N Am Spine Soc J. 2023;14:100204. doi: 10.1016/j.xnsj.2023.100204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Nagata K, Glassman SD, Dimar JR, 2nd, et al. Comparison of bone mineral density in children and adolescents on CT versus DEXA scan. Spine. 2024;49(19):E322-e326. doi: 10.1097/brs.0000000000004877 [DOI] [PubMed] [Google Scholar]
- 11.Zaidi Q, Danisa OA, Cheng W. Measurement techniques and utility of Hounsfield unit values for assessment of bone quality prior to spinal instrumentation: a review of current literature. Spine. 2019;44(4):E239-E244. doi: 10.1097/brs.0000000000002813 [DOI] [PubMed] [Google Scholar]
- 12.Egu C, Najjar E, Komaitis S, et al. Evaluating the prognostic role of computed tomography Hounsfield units in anticipating spinal outcomes post-instrumentation: a systematic review and meta-analysis. Eur Spine J. 2025;34(4):1420-1432. doi: 10.1007/s00586-025-08737-9 [DOI] [PubMed] [Google Scholar]
- 13.Chanbour H, Steinle AM, Chen JW, et al. The importance of Hounsfield units in adult spinal deformity surgery: finding an optimal threshold to minimize the risk of mechanical complications. J Spine Surg. 2023;9(2):149-158. doi: 10.21037/jss-22-102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Duan PG, Mummaneni PV, Rivera J, et al. The association between lower Hounsfield units of the upper instrumented vertebra and proximal junctional kyphosis in adult spinal deformity surgery with a minimum 2-year follow-up. Neurosurg Focus. 2020;49(2):E7. doi: 10.3171/2020.5.Focus20192 [DOI] [PubMed] [Google Scholar]
- 15.Baumann AN, Trager RJ, Gong DC, Anaspure OS, Strony JT, Aleem I. Osteoporosis is not associated with reoperation or pseudarthrosis after anterior cervical discectomy and fusion through 4-years’ follow-up: a retrospective cohort study of US academic health centers. Spine J. 2025;25(2):290-298. doi: 10.1016/j.spinee.2024.09.031 [DOI] [PubMed] [Google Scholar]
- 16.Hong CG, Nam WD, Lee YJ, Lee DH. Effect of osteoporosis on clinical and radiological outcomes following one-level anterior cervical discectomy and fusion. Asian Spine J. 2024;18(2):182-189. doi: 10.31616/asj.2023.0207 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ahmed W, Duey A, Rajjoub R, et al. Role of preoperative Hounsfield units in predicting subsidence after anterior cervical discectomy and fusion in the United States: a retrospective analysis including osteopenia diagnosis. Asian Spine J. 2025;19(4):609-618. doi: 10.31616/asj.2024.0214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schröder G, Reichel M, Spiegel S, et al. Breaking strength and bone microarchitecture in osteoporosis: a biomechanical approximation based on load tests in 104 human vertebrae from the cervical, thoracic, and lumbar spines of 13 body donors. J Orthop Surg Res. 2022;17(1):228. doi: 10.1186/s13018-022-03105-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Schröder G, Baginski AM, Schulze M, et al. Regional variations in the intra- and intervertebral trabecular microarchitecture of the osteoporotic axial skeleton. Anat Sci Int. 2023;98(4):566-579. doi: 10.1007/s12565-023-00726-6 [DOI] [PubMed] [Google Scholar]
- 20.Lee D-H, Cho JH, Hwang CJ, et al. What is the fate of pseudarthrosis detected 1 year after anterior cervical discectomy and fusion? Spine. 2018;43(1):E23-E28. doi: 10.1097/brs.0000000000002077 [DOI] [PubMed] [Google Scholar]
- 21.Park S, Lee DH, Lee CS, Hwang CJ, Yang JJ, Cho JH. Anterior decompression and fusion for the treatment of cervical myelopathy caused by ossification of the posterior longitudinal ligament: a narrative review. Asian Spine J. 2023;17(3):582-594. doi: 10.31616/asj.2022.0003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lee D-H, Lee HR, Riew KD. An algorithmic roadmap for the surgical management of degenerative cervical myelopathy: a narrative review. Asian Spine J. 2024;18(2):274-286. doi: 10.31616/asj.2023.0413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wu W-J, Jiang L-S, Liang Y, Dai L-Y. Cage subsidence does not, but cervical lordosis improvement does affect the long-term results of anterior cervical fusion with stand-alone cage for degenerative cervical disc disease: a retrospective study. Eur Spine J. 2012;21(7):1374-1382. doi: 10.1007/s00586-011-2131-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bartels RH, Donk RD, Feuth T. Subsidence of stand-alone cervical carbon fiber cages. Neurosurgery. 2006;58(3):502-508. doi: 10.1227/01.Neu.0000197258.30821.50 [DOI] [PubMed] [Google Scholar]
- 25.Lee HJ, You ST, Kim JH, Kim IS, Sung JH, Hong JT. Significance of cervical spine computed tomography Hounsfield units to predict bone mineral density and the subsidence after anterior cervical discectomy and fusion. Clin Spine Surg. 2021;34(8):E450-E457. doi: 10.1097/bsd.0000000000001218 [DOI] [PubMed] [Google Scholar]
- 26.Feng N, Li W, Yu X, et al. Cervical vertebra bone quality score predicts zero-profile anchored spacer interbody fusion cage subsidence after anterior cervical diskectomy and fusion: a retrospective study. Glob Spine J. 2024;15:21925682241280258. doi: 10.1177/21925682241280258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Noordhoek I, Koning MT, Jacobs WCH, Vleggeert-Lankamp CLA. Incidence and clinical relevance of cage subsidence in anterior cervical discectomy and fusion: a systematic review. Acta Neurochir. 2018;160(4):873-880. doi: 10.1007/s00701-018-3490-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wang M, Mummaneni PV, Xi Z, et al. Lower Hounsfield units on CT are associated with cage subsidence after anterior cervical discectomy and fusion. J Neurosurg: Spine SPI. 2020;33(4):425-432. doi: 10.3171/2020.3.SPINE2035 [DOI] [PubMed] [Google Scholar]
- 29.Safdar A, Headley B, Rommelman M, Haseeb A, Motiei-Langroudi R. The effect of interbody cage parameters on the rate of subsidence in single-level anterior cervical discectomy and fusion (ACDF): a retrospective analysis of 98 patients. Cureus. 2023;15(12):e50386. doi: 10.7759/cureus.50386 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lee Y-S, Kim Y-B, Park S-W. Risk factors for postoperative subsidence of single-level anterior cervical discectomy and fusion: the significance of the preoperative cervical alignment. Spine. 2014;39(16):1280-1287. doi: 10.1097/brs.0000000000000400 [DOI] [PubMed] [Google Scholar]
- 31.Riew KD, Yang JJ, Chang D-G, et al. What is the most accurate radiographic criterion to determine anterior cervical fusion? Spine J. 2019;19(3):469-475. doi: 10.1016/j.spinee.2018.07.003 [DOI] [PubMed] [Google Scholar]
- 32.Lee DH, Park S, Seok SY, et al. Fate of pseudarthrosis detected 2 years after anterior cervical discectomy and fusion: results of a minimum 5-year follow-up. Spine J. 2023;23(12):1790-1798. doi: 10.1016/j.spinee.2023.07.016 [DOI] [PubMed] [Google Scholar]
- 33.Chanbour H, Bendfeldt GA, Johnson GW, et al. Longer screws decrease the risk of radiographic pseudarthrosis following elective anterior cervical discectomy and fusion. Glob Spine J. 2023;15:21925682231214361. doi: 10.1177/21925682231214361 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Zuckerman SL, Devin CJ. Pseudarthrosis of the cervical spine. Clin Spine Surg. 2022;35(3):97-106. doi: 10.1097/bsd.0000000000001259 [DOI] [PubMed] [Google Scholar]
- 35.Sakai K, Yoshii T, Hirai T, et al. Cervical sagittal imbalance is a predictor of kyphotic deformity after laminoplasty in cervical spondylotic myelopathy patients without preoperative kyphotic alignment. Spine. 2016;41(4):299-305. doi: 10.1097/brs.0000000000001206 [DOI] [PubMed] [Google Scholar]
- 36.Li X-Y, Wang Y, Zhu W-G, Kong C, Lu S-B. Impact of cervical and global spine sagittal alignment on cervical curvature changes after posterior cervical laminoplasty. J Orthop Surg Res. 2022;17(1):521. doi: 10.1186/s13018-022-03421-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Han K, You ST, Lee HJ, Kim IS, Hong JT, Sung JH. Hounsfield unit measurement method and related factors that most appropriately reflect bone mineral density on cervical spine computed tomography. Skelet Radiol. 2022;51(10):1987-1993. doi: 10.1007/s00256-022-04050-4 [DOI] [PubMed] [Google Scholar]
- 38.Fluss R, Lo Bu R, De la Garza Ramos R, Murthy SG, Yassari R, Gelfand Y. Nonquantitative CT scan Hounsfield unit as a determinant of cervical spine bone density. J Neurosurg Spine. 2024;41(6):803-812. doi: 10.3171/2024.5.Spine24249 [DOI] [PubMed] [Google Scholar]
- 39.Razzouk J, Ramos O, Ouro-Rodrigues E, et al. Comparison of cervical, thoracic, and lumbar vertebral bone quality scores for increased utility of bone mineral density screening. Eur Spine J. 2023;32(1):20-26. doi: 10.1007/s00586-022-07484-5 [DOI] [PubMed] [Google Scholar]
- 40.Dvorak MF, Pitzen T, Zhu Q, Gordon JD, Fisher CG, Oxland TR. Anterior cervical plate fixation: a biomechanical study to evaluate the effects of plate design, endplate preparation, and bone mineral density. Spine. 2005;30(3):294-301. doi: 10.1097/01.brs.0000152154.57171.92 [DOI] [PubMed] [Google Scholar]
- 41.Fogel GR, Li Z, Liu W, Liao Z, Wu J, Zhou W. In vitro evaluation of stiffness and load sharing in a two-level corpectomy: comparison of static and dynamic cervical plates. Spine J. 2010;10(5):417-421. doi: 10.1016/j.spinee.2010.02.004 [DOI] [PubMed] [Google Scholar]
- 42.Brodke DS, Klimo P, Jr, Bachus KN, Braun JT, Dailey AT. Anterior cervical fixation: analysis of load-sharing and stability with use of static and dynamic plates. J Bone Joint Surg Am. 2006;88(7):1566-1573. doi: 10.2106/jbjs.E.00305 [DOI] [PubMed] [Google Scholar]
- 43.Chun DS, Baker KC, Hsu WK. Lumbar pseudarthrosis: a review of current diagnosis and treatment. Neurosurgical Focus FOC. 2015;39(4):E10. doi: 10.3171/2015.7.FOCUS15292 [DOI] [PubMed] [Google Scholar]
- 44.Levy HA, Pinter ZW, Reed R, et al. Transforaminal lumbar interbody fusion subsidence: computed tomography analysis of incidence, associated risk factors, and impact on outcomes. J Neurosurg Spine. 2024;41(4):463-472. doi: 10.3171/2024.4.SPINE231007 [DOI] [PubMed] [Google Scholar]
- 45.Yamagata T, Takami T, Uda T, et al. Outcomes of contemporary use of rectangular titanium stand-alone cages in anterior cervical discectomy and fusion: cage subsidence and cervical alignment. J Clin Neurosci. 2012;19(12):1673-1678. doi: 10.1016/j.jocn.2011.11.043 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Material for Does Bone Quality Predict Pseudarthrosis, Subsidence, or Loss of Lordosis After Anterior Cervical Discectomy and Fusion? Minimum 5-Year Cohort Study by San Kim, MD, Sehan Park, MD, PhD, Dong Ki Ahn, MD, PhD, Dong-Ho Lee, MD, PhD, Gumin Jeong, MD, Ki Chol Park, MD, Yong Kim, MD, and Changmin Choi, MD, PhD in Global Spine Journal