Clinical and radiographic outcomes of three-level anterior cervical discectomy and fusion using allograft cellular bone matrix

Introduction

Anterior cervical discectomy and fusion (ACDF) is one of the most commonly performed surgical techniques for the management of cervical radiculopathy, myelopathy, and degenerative disease (1). First introduced in 1958, ACDF has since evolved into a highly reliable operation and has become widely adopted in the United States, with more than 120,000 cases performed annually (2). Over time, the emphasis of the procedure has shifted towards optimizing fusion rates, reducing complications, and improving the long-term durability of the fusion construct (3). These factors have become increasingly important, particularly as the number of fused levels continues to increase. Although single-level ACDF achieves excellent fusion rates, often exceeding 90–95% in modern series, multilevel fusions are associated with markedly higher rates of pseudoarthrosis, dysphagia, and adjacent segment disease (ASD) (4). Three-level ACDF procedures, in particular, have a reported pseudoarthrosis rate of as high as 20–30%, depending on the graft choice and surgical technique (4). Given the inherent challenges with maintaining a high fusion success rate in 3-level ACDF, there is increasing interest within the field of cervical spine surgery to ascertain the optimal biological adjunct for 3-level constructs (5).

Selection of an appropriate bone graft is essential for achieving a successful arthrodesis (6). While iliac crest autograft has traditionally served as the gold standard source for graft harvest, it is limited by significant, well-documented morbidity (7). The biological advantages of an iliac crest graft center around its complement of osteoconductive, osteoinductive, and osteogenic properties, all of which make it ideal for stimulating new bone formation (8). However, there are several common downsides to an iliac crest harvest, including donor site pain, hematoma, infection, sensory disturbances, increased operative time, and greater length of hospital stay (9). When considered for multilevel ACDF procedures, the benefits do not always outweigh the associated morbidity. Other alternatives such as structural allograft and demineralized bone matrix (DBM) have become increasingly popular to avoid donor site morbidity (10). While these graft materials offer some osteoconductive and osteoinductive potential, they lack intrinsic osteogenicity, a factor that grows in importance as fusion constructs increase in complexity (11).

Recombinant human bone morphogenetic protein-2 (rhBMP-2) is a powerful biologic agent with osteoinductive capabilities that emerged in the early 2000s as a reliable tool for promoting fusion (12). It was validated in several investigations that reported high rates of fusion during different spinal procedures, but its widespread enthusiasm for use in cervical fusion later decreased as growing evidence highlighted significant dose-related complications such as soft tissue swelling, severe dysphagia, and even airway obstruction (13). Modern practices involve using ultra-low doses of rhBMP-2 and have demonstrated reduced complication rates; however, its usage in the cervical spine is still considered off-label (14). Many surgeons remain reluctant to use it routinely, especially in anterior approaches with multilevel constructs. Identifying a biologic adjunct with osteogenic potential similar to that of autograft or rhBMP-2 but without the risk profile is, therefore, of particular interest.

One such emerging alternative is allograft cellular bone matrix (ACBM), which is commercially available as products such as Osteocel^® (NuVasive, Inc., USA). These grafts consist of cryopreserved cadaveric mesenchymal stem cells (MSCs) and osteoprogenitor cells contained within a cancellous matrix as well as a demineralized cortical component (15). In this way, they can theoretically provide the full triad of osteoconduction, osteoinduction, and osteogenesis (16). ACBM has already been used successfully in a wide variety of orthopedic applications, including trauma, revision arthroplasty, and lumbar fusion procedures (17). Several studies within the spine surgery literature have reported promising early fusion rates. ACBM has demonstrated fusion rates exceeding 85% in cohorts undergoing 1–2 level ACDF, with clinical outcomes that are comparable to those using traditional grafts (18).

To our knowledge, virtually no literature exists that evaluates ACBM specifically in the setting of 3-level ACDF procedures. Multilevel constructs place increased biological demands on the grafts in the form of greater mechanical stress, increased motion across multiple adjacent segments, and diminished vascularity at terminal levels (particularly C6–C7) (19). This unique fusion environment positions 3-level ACDF procedures to be significantly more challenging to achieve osteogenesis with ACBM than with single-level or two-level procedures (20). Pseudoarthrosis risk is also particularly elevated in multilevel fusion, and so, it would be of great clinical utility to determine how ACBM performs in this higher-risk setting. Other complications including dysphagia and ASD are also at increased risk in multilevel fusion, but as of yet, there is no strong relationship reported in the literature that analyzes the association between graft selection and complication profile.

From a broader health perspective, biological graft selection is becoming increasingly influenced by cost. While ACBM is more expensive than many traditional grafts, the potential to reduce pseudoarthrosis and reoperation rates may make it more economical in the long-term. A study that evaluates such outcomes in 3-level ACDF would provide significant data to help guide future decisions about optimizing graft selection. The purpose of this study was to describe the clinical and radiographic outcomes of patients undergoing 3-level ACDF using ACBM to help contextualize this data. We present this article in accordance with the STROBE reporting checklist (available at https://jss.amegroups.com/article/view/10.21037/jss-2026-1-0028/rc).

Methods

A retrospective cohort review was performed at a single academic institution of all patients who underwent 3-level ACDF with ACBM augmentation from 2012 to 2020. Patients were included if they had a primary 3-level fusion for symptomatic cervical stenosis, were at least 18 years of age, and had a minimum follow-up of 24 months with complete clinical and radiographic data. Patients were excluded if they had undergone prior cervical spine procedures or if the indications for surgery were trauma, neoplasm, or infection. Medical records including history and physical examinations, operative reports, discharge summaries, office notes, and radiographic studies were thoroughly reviewed. Baseline demographic data that was collected included age, sex, body mass index (BMI), fused levels, smoking status, and history of diabetes mellitus. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was reviewed by the Institutional Review Board of St. Joseph’s University Medical Center, and the requirement for formal approval and informed consent was waived due to the retrospective design and the use of deidentified data.

Surgical technique

All procedures were performed by fellowship-trained orthopedic spine surgeons using a standard modified Smith-Robinson anterior cervical approach. Following decompression and discectomy, polyetheretherketone (PEEK) interbody cages were implanted at each level. Each cage was tightly packed with Osteocel^® ACBM. Anterior cervical plate fixation was used in all cases according to standardized institutional technique. No iliac crest autograft, DBM, or rhBMP-2 was used in any patient. ACBM served as the sole biologic graft material across all three fused levels.

Outcome measures

Patients were evaluated at routine postoperative intervals: 2 weeks, 6 weeks, 3 months, 6 months, and annually after that. Patient-reported functional outcomes were assessed using the Neck Disability Index (NDI) and Visual Analog Scale (VAS) scores for both neck and arm pain. Outcome measures were recorded preoperatively and at each postoperative visit, with the 24-month values used for formal statistical comparison.

Complications

All perioperative and postoperative complications were recorded and categorized as clinical, wound-related, hardware-related, or radiographic. Dysphagia was assessed at follow-up visits according to the Bazaz scale, which grades difficulty swallowing solids or liquids as none, mild, moderate, or severe. Resolution of dysphagia over time was determined from clinical documentation. Wound-related complications were classified as either superficial or deep infections and noted according to whether they required medical management or operative intervention. Hardware-related complications consisted of implant failure, instrumentation fracture, or implant prominence requiring revision or removal.

Radiographic evaluation

Standard postoperative cervical spine radiographs (anteroposterior and lateral views) were obtained at each follow-up visit. Radiographic fusion was evaluated on serial imaging and was defined as solid bridging bone with osseous incorporation of the graft across the operative levels, without evidence of implant failure or gross motion on sequential studies.

When fusion status was uncertain on standard radiographs, additional imaging was obtained to further assess for pseudoarthrosis. Flexion-extension radiographs were typically performed in cases of persistent clinical concern (such as ongoing neck or arm symptoms) and/or equivocal findings on serial static imaging. Computed tomography (CT) imaging was obtained when fusion remained indeterminate after radiographic follow-up, including dynamic studies, when performed, or when there was increased concern for nonunion based on symptoms or construct-related findings. On CT, fusion was defined as the presence of continuous bridging bone across the interbody space, whereas nonunion was defined as absence of bridging bone at one or more operative levels. All radiographic assessments were initially performed by the treating spine surgeon and then subsequently reviewed by additional members of the surgical team to improve reliability.

Pseudoarthrosis was defined as radiographic evidence of nonunion at one or more operative levels. The specific levels affected were documented for each case. Patients with radiographic pseudoarthrosis were then further categorized based on the presence or absence of associated clinical symptoms and whether revision surgery was required.

ASD was defined as the onset of new clinical symptoms attributable to adjacent-level degeneration on imaging, managed either nonoperatively or operatively. Only symptomatic adjacent segment pathology cases were counted as ASD. Radiographic degeneration at the adjacent level in the absence of new symptoms was not considered for the purposes of this study.

Statistical analysis

All data were de-identified and tabulated for analysis. Descriptive statistics were generated for demographic variables, complication rates, and radiographic outcomes. Continuous variables were reported as means with standard deviations, and categorical variables were reported as frequencies and percentages.

Changes in patient-reported outcome measures (NDI, VAS-neck, and VAS-arm) from the preoperative baseline to the 24-month follow-up were analyzed using paired Student’s t-tests. All statistical analyses were performed using GraphPad Prism (GraphPad Software, San Diego, CA). A P<0.05 was considered statistically significant.

Results

There were 108 patients who met the inclusion criteria for this study. Five patients were lost to follow-up before 24 months and were excluded from the final analysis. Thus, there were a total of 103 patients that were included in the final cohort. The mean follow-up duration was 45.5±15.5 months. The mean age of patients at the time of surgery was 57.8±7.9 years, and the mean BMI was 30.2±5.1 kg/m². The sex distribution was roughly equivalent, with 49.5% male patients. Seventeen patients (16.5%) had a documented history of diabetes, and 19 patients (18.4%) were active smokers. The most commonly treated operative segment was C4–C7, accounting for 79.6% of all cases, followed by C3–C6 at 20.4%. The baseline demographic data is summarized in Table 1.

Complications

There were no perioperative complications that resulted in a prolonged length of hospital stay greater than 48 hours. The most frequently observed postoperative complication (Table 2) was dysphagia, which occurred in 24 patients (23.3%). The majority of dysphagia cases were classified as mild and resolved without intervention by the 6-month follow-up visit. Two patients with moderate dysphagia were treated successfully with a short course of oral corticosteroids. One patient underwent removal of anterior instrumentation due to persistent moderate dysphagia associated with prominent hardware. No patients experienced severe dysphagia or airway compromise.

Two patients (1.9%) experienced wound-related complications, both of which were superficial infections treated successfully with a course of oral antibiotics. There were no deep infections or cases requiring surgical debridement.

Hardware-related complications occurred in three patients (2.9%), all of which resulted in subsequent removal of implants without further complication. These cases included a fractured screw, implant failure, and the aforementioned case of hardware prominence associated with persistent dysphagia.

There were no cardiac, neurologic, or respiratory complications observed during the study period.

Radiographic outcomes

Advanced imaging was obtained selectively to clarify fusion status. Flexion-extension radiographs were obtained in 43 patients (41.7%) and CT imaging was obtained in 17 patients (16.5%). The majority of CT studies were obtained in patients with equivocal radiographs or persistent clinical symptoms. By 24 months postoperatively, five patients demonstrated radiographic evidence of pseudoarthrosis. Three cases occurred at the C6–C7 level while two were at C5–C6. Four of the five patients remained asymptomatic and were managed nonoperatively. One patient developed clinical symptoms and required a revision posterior cervical fusion from C4 to C7.

ASD occurred in six patients (5.8%) during the follow-up period. Of these cases, three required posterior cervical fusion, two underwent decompression via posterior foraminotomy, and one was managed nonoperatively with physical therapy.

Functional outcomes

Significant improvements were observed in all patient-reported outcome measures at final follow-up (Table 3). The mean NDI score improved from 26.4±8.2 preoperatively to 11.2±4.5 (P<0.001). Mean VAS-neck scores improved from 7.7±2.9 (preoperative) to 2.4±1.6 (P<0.001), while the mean VAS-arm pain scores went from 5.2±3.4 preoperatively to 1.4±1.1 (P<0.001) (Table 4). These improvements were sustained throughout the available follow-up period as well.

Discussion

Pseudoarthrosis is a well-recognized contributor to poor outcomes after ACDF, so achieving reliable fusion, particularly in multilevel procedures, is a key surgical objective (4,21). Although traditional autologous bone grafts are associated with high fusion rates, their use is limited by donor site morbidity, prompting increased reliance on biologic substitutes like rhBMP-2 and DBM, each of which has its own unique limitations (21-23). ACBM, a Food and Drug Administration (FDA)-approved allograft containing osteoconductive, osteoinductive, and osteogenic components, has demonstrated high fusion rates in 1–2 level ACDF (18,22), but its efficacy in 3-level cohorts is, as of yet, unvalidated. Unlike purely structural allograft, cellular bone matrices are designed to incorporate a cancellous scaffold with demineralized components and viable osteogenic cells (22). This combination theoretically improves the likelihood of consistent graft incorporations by promoting early cellular activity in settings with biologically challenging fusion environments (18,22). Multilevel ACDF represents a biologically demanding environment, with prior literature consistently demonstrating higher rates of pseudoarthrosis, dysphagia, and reoperation as the number of fused levels increases (4,21,23).

Three-level ACDF specifically represents a meaningful clinical threshold rather than simply a broader extension multilevel surgery. While fusion rates for one- and two-level procedures are generally high (4), the addition of a third level has been consistently associated with a sharper increase in pseudoarthrosis and revision risk (21,24). In practice, three levels also tend to represent the upper extent of standalone anterior reconstruction, as increasing construct length beyond three levels is associated with greater complexity and much higher nonunion risk, often prompting consideration of posterior or circumferential techniques (23). As a result, three-level constructs combine increased mechanical demand with greater biologic stress. The added fusion interfaces, longer construct length, and higher stresses (particularly at the caudal level) place greater reliance on graft performance (19,25). In this setting, a cellular bone matrix that provides viable osteogenic cells alongside osteoconductive and osteoinductive support may be especially relevant in achieving reliable fusion compared with shorter constructs, where the biologic demand is less substantial (22).

Within this context, the present study evaluated clinical and radiographic outcomes of ACBM augmentation in 3-level ACDF procedures. The findings demonstrated significant improvements in patient-reported outcomes with a relatively low incidence of pseudoarthrosis. Among 103 patients followed for an average of almost four years, the overall rate of pseudoarthrosis was under 5%. Dysphagia was the most common postoperative complaint, but all cases were resolved by six months. ASD occurred in just over 5% of patients. Together, these results provide descriptive, real-world data regarding the performance of ACBM in a population that is traditionally at an elevated risk for fusion-related complications.

The favorable improvements in NDI and VAS scores support the clinical effectiveness of three-level ACDF augmented with ACBM, suggesting meaningful symptom relief and functional recovery. Functional improvement is a key indicator of success following ACDF, particularly in patients with combined axial neck pain, radiculopathy, and myelopathy. These improvements were sustained throughout available follow-up and are consistent with outcomes reported for multilevel ACDF using other grafts. While these findings do not establish superiority of ACBM over alternative graft options, they suggest that ACBM can be used successfully in complex multilevel constructs without compromising patient-reported outcomes.

Achieving reliable fusion in 3-level ACDF remains a major concern, as prior literature consistently demonstrates that pseudoarthrosis rates increase substantially with construct length, particularly as construct length increases (21). Previous data on 3-level ACDF using autograft or structural allograft have reported widely varying rates of pseudoarthrosis, with some series reporting rates exceeding 40%, reflecting differences in fixation techniques, patient-related risk factors, and graft selection (21,24). In the present analysis, pseudoarthrosis was identified in 4.9% of patients, with the majority of cases remaining asymptomatic and only one patient requiring revision surgery. Notably, observed instances of pseudoarthrosis most commonly involved the caudal segment of the construct, a pattern that has been described previously and is thought to reflect higher mechanical demands at distal levels (21,25). When compared with previous reports, the observed rate of pseudoarthrosis falls within the lower range of previously reported values, but the findings should still be interpreted with some caution. Although fusion was evaluated with serial plain-film radiographs, CT imaging was used only when radiographic findings were unclear rather than as part of a standardized follow-up protocol. This approach minimized unnecessary testing and reduced radiation exposure, but it may have also led to underdetection of asymptomatic pseudoarthrosis. However, because the pseudoarthroses identified were largely clinically silent and did not require revision, any missed cases were unlikely to have altered the overall clinical conclusions of the study.

Postoperative dysphagia was the most frequently observed complication, occurring in approximately one-quarter of patients. This finding aligns with prior literature demonstrating higher rates of dysphagia following multilevel cervical procedures compared with single-level surgery (23). The majority of dysphagia cases in this study were mild and resolved by six months postoperatively, with no cases of severe dysphagia, airway compromise, or prolonged hospitalization. Although biologic adjuncts such as rhBMP-2 have been associated with dysphagia following ACDF (23), the symptom is generally regarded as multifactorial, influenced by elements like surgical exposure, retraction time, plate prominence, and construct length rather than graft material alone. The transient nature of dysphagia observed in most cases, coupled with the low rate of required intervention, supports an acceptable morbidity profile for 3-level ACDF. As dysphagia in this cohort was assessed retrospectively, a causal relationship between ACBM use and dysphagia is difficult to establish. Nonetheless, both the incidence and clinical course observed here are comparable to patterns reported across existing multilevel ACDF literature.

ASD is another important long-term consideration following cervical fusion. In this study, ASD occurred in approximately 6% of patients, with the majority requiring subsequent reoperation. Reported ASD rates after ACDF vary widely in the literature, largely due to differences in definitions that include radiographic degeneration alone as opposed to symptomatic disease (23). By excluding asymptomatic radiographic changes, this study focuses on clinically meaningful ASD, which may better reflect outcomes relevant to patient counseling and treatment decisions. As with dysphagia, there are other biomechanical factors such as plate-to-disc distance and sagittal alignment that contribute to the development of ASD. Although establishing a direct link between ACBM use and ASD rates is challenging, the ASD rates observed in this cohort are comparable to those reported in other systematic reviews of ACDF outcomes (4,23).

Altogether, the observed clinical improvements, fusion outcomes, and complication profile suggest that ACBM can be used as an adjunct to 3-level ACDF with acceptable results. Although this study is not positioned to establish superiority over traditional graft choices, it does provide real-world data that serves as evidence of feasibility for a relatively large cohort of patients undergoing 3-level ACDF. Given the ongoing concerns regarding donor site morbidity associated with autograft and the off-label risks of rhBMP-2 usage (21,23), these findings provide a framework for validating ACBM augmentation in multilevel ACDF. Despite the fact that cellular bone matrices are associated with higher upfront material costs compared to traditional allograft options, the potential to reduce pseudoarthrosis or revision rates may offset these costs in select patient populations (22,26).

There are several important limitations of this study that should be acknowledged. The retrospective design introduces the potential for selection bias, and the absence of a control group limits direct comparison with alternative graft materials. Fusion assessment was also not standardized with advanced imaging for all patients. Because CT and/or dynamic imaging were obtained selectively, most often when static radiographs were equivocal or when clinical concern for nonunion was present, determination of fusion status may have been subject to detection bias, and asymptomatic nonunion may be underrepresented in the reported pseudoarthrosis rate. Radiographic evaluation performed by the treating surgeon rather than independent reviewers may also have led to the underestimation of asymptomatic nonunion. Additionally, the complication profiles were identified retrospectively through chart review, which may subject the data to underreporting or variability in documentation. The single-institution nature of this study also limits the generalizability of the findings to external populations. Despite these limitations, this study represents one of the largest reported cohorts evaluating ACBM specifically in 3-level ACDF. The findings support the feasibility of ACBM as an adjunct in multilevel ACDF and provide a foundation for further comparative and cost-effectiveness investigations.

Future studies should prioritize designs that incorporate standardized fusion assessment with consistent follow-up imaging protocols. Prospective cohorts comparing ACBM with commonly used alternatives would better clarify whether there are any clinically meaningful differences. This would also allow for a more methodical stratification based on known fusion risk factors to help identify which patients would benefit the most from ACBM compared to other graft materials. Ultimately, a formal cost-benefit analysis looking at the projected downstream cost savings would define the true value proposition of incorporating ACBM use in the standard of care.

Conclusions

This study provides functional and radiographic data describing the use of ACBM in 3-level ACDF, a setting in which reliable fusion remains challenging. The findings suggest that ACBM can be used as a biologic adjunct in multilevel cervical fusion without compromising clinical outcomes. While this study does not establish superiority compared to other grafts, the results contribute to real-world feasibility that may inform graft selection and guide future comparative and value-based studies.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jss.amegroups.com/article/view/10.21037/jss-2026-1-0028/rc

Data Sharing Statement: Available at https://jss.amegroups.com/article/view/10.21037/jss-2026-1-0028/dss

Peer Review File: Available at https://jss.amegroups.com/article/view/10.21037/jss-2026-1-0028/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jss.amegroups.com/article/view/10.21037/jss-2026-1-0028/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was reviewed by the Institutional Review Board of St. Joseph’s University Medical Center, and the requirement for formal approval and informed consent was waived due to the retrospective design and the use of deidentified data.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Martin D, Schroeder R, Toups C, et al. Comparing ACDF Outcomes by Cervical Spine Level: A Single Center Retrospective Cohort Study. Int J Spine Surg 2024;18:645-52. [Crossref] [PubMed]
2. Arif H, Razzouk J, Bohen D, et al. Analysis of reasons for medical malpractice litigation due to anterior cervical discectomy and fusion. World Neurosurg X 2024;23:100371. [Crossref] [PubMed]
3. Chong E, Pelletier MH, Mobbs RJ, et al. The design evolution of interbody cages in anterior cervical discectomy and fusion: a systematic review. BMC Musculoskelet Disord 2015;16:99. [Crossref] [PubMed]
4. Laratta JL, Reddy HP, Bratcher KR, et al. Outcomes and revision rates following multilevel anterior cervical discectomy and fusion. J Spine Surg 2018;4:496-500. [Crossref] [PubMed]
5. Cabrera JP, Agarwal N, Mesregah MK, et al. Analysis of Complications in Multilevel Anterior Cervical Discectomy and Fusion Using Osteobiologics Other than Bone Morphogenetic Protein: A Systematic Review. Global Spine J 2024;14:86S-93S. [Crossref] [PubMed]
6. Prost M, Windolf J, Konieczny MR. Bovine-derived xenograft is a viable bone graft substitute in multilevel, instrumented, spinal fusion. Orthop Rev (Pavia) 2022;14:37576. [Crossref] [PubMed]
7. Holmes C, Elder BD, Ishida W, et al. Comparing the efficacy of syngeneic iliac and femoral allografts with iliac crest autograft in a rat model of lumbar spinal fusion. J Orthop Surg Res 2020;15:410. [Crossref] [PubMed]
8. Osteogenic protein-1 for long bone nonunion: an evidence-based analysis. Ont Health Technol Assess Ser 2005;5:1-57.
9. Katz MS, Ooms M, Heitzer M, et al. Postoperative Morbidity and Complications in Elderly Patients after Harvesting of Iliac Crest Bone Grafts. Medicina (Kaunas) 2021;57:759. [Crossref] [PubMed]
10. Manawar S, Myrick E, Awad P, et al. Use of allograft bone matrix in clinical orthopedics. Regen Med 2024;19:247-56. [Crossref] [PubMed]
11. Yu D, Shen W, Dai J, et al. Treatment of large bone defects in load-bearing bone: traditional and novel bone grafts. J Zhejiang Univ Sci B 2025;26:421-47. [Crossref] [PubMed]
12. Roh JS, Yeung CA, Field JS, et al. Allogeneic morphogenetic protein vs. recombinant human bone morphogenetic protein-2 in lumbar interbody fusion procedures: a radiographic and economic analysis. J Orthop Surg Res 2013;8:49. [Crossref] [PubMed]
13. Vaidya R, Carp J, Sethi A, et al. Complications of anterior cervical discectomy and fusion using recombinant human bone morphogenetic protein-2. Eur Spine J 2007;16:1257-65. [Crossref] [PubMed]
14. Khajavi K, Shen A. Safety and efficacy of bioabsorbable cervical spacers and low-dose rhBMP-2 in multi-level ACDF. Int J Spine Surg 2014;8:9. [Crossref] [PubMed]
15. Neman J, Duenas V, Kowolik C, et al. Lineage mapping and characterization of the native progenitor population in cellular allograft. Spine J 2013;13:162-74. [Crossref] [PubMed]
16. Malagón-Escandón A, Hautefeuille M, Jimenez-Díaz E, et al. Three-Dimensional Porous Scaffolds Derived from Bovine Cancellous Bone Matrix Promote Osteoinduction, Osteoconduction, and Osteogenesis. Polymers (Basel) 2021;13:4390. [Crossref] [PubMed]
17. Tohmeh AG, Watson B, Tohmeh M, et al. Allograft cellular bone matrix in extreme lateral interbody fusion: preliminary radiographic and clinical outcomes. ScientificWorldJournal 2012;2012:263637. [Crossref] [PubMed]
18. Goldman SN, Paschal GK, Mani K, et al. Efficacy of an allograft cellular bone matrix as an alternative to autograft in anterior cervical discectomy and fusion: radiological results & safety. J Spine Surg 2024;10:372-85. [Crossref] [PubMed]
19. Kong Q, Li F, Yan C, et al. Biomechanical Comparison of Anterior Cervical Corpectomy Decompression and Fusion, Anterior Cervical Discectomy and Fusion, and Anterior Controllable Antedisplacement and Fusion in the Surgical Treatment of Multilevel Cervical Spondylotic Myelopathy: A Finite Element Analysis. Orthop Surg 2024;16:687-99. [Crossref] [PubMed]
20. Strenge KB, Heller JE, Williams DM, et al. Three-Level Anterior Cervical Discectomy and Fusion With or Without an Investigational Posterior Stabilization System Assessed Through 24 Months: A Multicenter Randomized Controlled Trial. Spine (Phila Pa 1976) 2025;50:1161-70. [Crossref] [PubMed]
21. Epstein NE. A Review of Complication Rates for Anterior Cervical Diskectomy and Fusion (ACDF). Surg Neurol Int 2019;10:100. [Crossref] [PubMed]
22. Diaz RR, Savardekar AR, Brougham JR, et al. Investigating the efficacy of allograft cellular bone matrix for spinal fusion: a systematic review of the literature. Neurosurg Focus 2021;50:E11. [Crossref] [PubMed]
23. Yee TJ, Swong K, Park P. Complications of anterior cervical spine surgery: a systematic review of the literature. J Spine Surg 2020;6:302-22. [Crossref] [PubMed]
24. Wewel JT, Kasliwal MK, Adogwa O, et al. Fusion rate following three- and four-level ACDF using allograft and segmental instrumentation: A radiographic study. J Clin Neurosci 2019;62:142-6. [Crossref] [PubMed]
25. Mun HW, Lee JJ, Shin HC, et al. Fusion Sequence and Risk Factors for Pseudarthrosis in Multilevel Anterior Cervical Discectomy and Fusion. Neurosurgery 2025;97:1153-61. [Crossref] [PubMed]
26. Demetriades AK, Mavrovounis G, Deml MC, et al. What Is the Evidence Surrounding the Cost-Effectiveness of Osteobiologic Use in ACDF Surgery? A Systematic Review of the Literature. Global Spine J 2024;14:163S-172S. [Crossref] [PubMed]