In clinical work, ACCF is widely used in the treatment of cervical degenerative diseases, especially multi-level cervical spondylosis, with good clinical efficacy. Our study showed that postoperative JOA, VAS and NDI scores were significantly improved in both the non-subsidence and subsidence groups than before surgery. Common vertebral replacements in ACCF surgery include titanium mesh and 3D-printed artificial vertebral body, and 3D-printed artificial vertebral body was selected in this study. The 3D printed artificial vertebral body was made of titanium alloy, with the upper end as a curved dome-like top and the lower end as a slightly concave slope structure design, which was highly consistent with the anatomical morphology of the upper and lower endplates of the fusion segment and increased the contact area. In addition, the unique microporous network structure is conducive to the induction of osteoblasts and the promotion of fusion [19]. When ACCF surgery achieves satisfactory curative effect, vertebral replacement sedimentation is the most important complication. The 3D-printed artificial vertebral subsidence causes internal fixation failure and even clinical symptoms caused by spinal cord and nerve root compression again, which is the main cause of postoperative reoperation[20]. Therefore, it is particularly important to analyze the causes of the subsidence of the 3D-printed artificial vertebral body after operation and how to take measures to prevent. At present, there is no unified standard for the measurement and standard of 3D-printed artificial vertebral body subsidence. Chen's [16] study classified the subsidence of vertebral body substitutes into mild (1-3mm) and severe (≥ 3mm). According to the study of Van Jonbergen [17], the postoperative intervertebral height drop greater than 3 mm is considered as subsidence, and greater than 6mm is considered as severe subsidence. The above literature shows that most vertebral replacement subsidence occurs within 3 months after surgery. In our study, the standard was defined as subsidence when the height of the intervertebral space measured by lateral cervical radiographs dropped by more than 3 mm within 3 months after surgery. In this study, 19 times of settlement occurred, with a settlement rate of 28.8% and a settlement range of 3.0 ~ 8.1mm.
The subsidence of the 3D-printed artificial vertebra is caused by a variety of factors. Common risk factors include advanced age, osteoporosis, smoking, multiple segments, the height of intervertebral distraction, and end-plate injury during operation [21–22]. Our study showed a significant age difference between the two groups, but the age of the patient was not an independent risk factor. The patients in the subsidence group is much older than the non-subsidence group. It is inferred that the bone density of elderly patients is lower, and the time of bone grafting fusion is longer, which is prone to sink. The literature [23] shows that the height of intervertebral space extension is one of the factors for the subsidence of the vertebral body replacement at the fusion level. Our study shows that the height loss of the fusion front is more significant, but the multi-factor analysis shows that the height loss of the fusion front is not an independent risk factor for subsidence, and the clinical experience is to avoid excessive extension when dealing with the fusion level. Our study showed that diabetes and smoking were factors for the subsidence of the 3D-printed artificial vertebra, and multivariate analysis confirmed that smoking was an independent risk factor for 3D-PAVB subsidence. Although the specific mechanism is not clear, the literature reports that patients with diabetes are at a higher risk of osteoporosis, especially type 1 diabetes [24–25]. Although there is almost no relevant literature revealing the direct relationship between smoking and 3D-PAVB subsidence, it is clear that long-term smoking can lead to osteoporosis [26–27]. Osteoporosis is the most important risk factor for plant material subsidence during spinal fusion surgery. Osteoporosis can lead to loss of bone mass and destruction of bone microstructure. Severe osteoporosis can lead to loosening of internal fixation and failure of fusion after spinal surgery. Osteoporosis is commonly evaluated and measured by bone mineral density (BMD). There are many methods to evaluate BMD [28], including dual energy X-ray absorptiometry (DXA) and quantitative computed tomography (QCT), peripheral quantitative computed tomography (pQCT), quantitative ultrasound (QUS), etc. The currently accepted diagnostic criteria for osteoporosis are based on the results of DXA measurements.
DXA was mainly measured in the axial bone, including the lumbar spine and hip joint. Lumbar measurements include the vertebral body and its posterior adnexal structure, which is influenced by body size (e.g., obesity) and abdominal aortic calcification, and does not directly reflect cervical bone density [29–30]. In addition, DXA is not a routine test before cervical surgery, which requires additional costs. Therefore, lumbar DXA values are not suitable for evaluating cervical osteoporosis. According to the empirical formula of lumbar spine [31], density ρ = 1.122*HU + 47 and elastic modulus E = 1.92*ρ-170, Heinz unit was converted into density value to evaluate bone density. Lower HU values in the lumbar spine have been reported to be associated with pseudarthrosis and graft sinking, reflecting lower bone density. Currently, the measurement of vertebral HU using routine CT examination is increasingly used to evaluate cervical osteoporosis. According to the study of Wang et al[32], low preoperative CT HU value is related to the implant settlement of single-segment ACDF, and preoperative HU measurement may help to predict the postoperative outcome of ACDF. Lee et al[33] reported that cervical CT HU value was of great significance in predicting bone density and settlement after anterior cervical discectomy, and HU value could accurately reflect cervical bone density. According to the study of He Lei et al[34], CT value of cervical vertebra is positively correlated with T value of bone density examined by DXA, which is helpful for preoperative clinical assessment of bone quality of patients. Biomechanical analysis [35] showed that with the increase of the degree of osteoporosis, the maximum stress distribution at the interface between the upper and lower endplates and the inner plant material of the fusion segment significantly increased, thus increasing the probability of the prosthesis sinking. The biomechanical strength of the lower vertebral body replacement -- endplate interface was higher than that of the upper vertebral body replacement -- endplate interface in the same fusion segment, so the bone quality of the lower vertebral body had a greater impact on the subsidence of the 3D-printed artificial vertebral body. Our study showed that HU in the upper and lower vertebrae of the subsidence group was lower than that in the non-subsidence group, and logistic regression analysis showed that HU in the lower vertebrae was an independent risk factor for 3D-PAVB sedimentation, which was consistent with the results of the biomechanical study mentioned above. Therefore, routine examination of CT scan HU values before ACCF can be used as a feasibility predictor of 3D-PAVB sedimentation after ACCF, providing guidance for surgical planning.
Cervical CT + 3D reconstruction is a routine preoperative examination for cervical spine patients. Measurement of HU value based on cervical CT is a simple, rapid and non-invasive method for bone quality assessment without additional medical costs. Some studies have proposed different diagnostic thresholds for HU values compatible with osteoporosis [9]. Our study showed that preoperative HU measurements based on cervical CT were a predictor of 3D-PAVB subsidence after ACCF. According to the ROC curve, the patients that HU value of the vertebral body under 272 had a higher probability of 3D-PAVB subsidence after ACCF (sensitivity: 87.5%; Specificity was 79.3%). Before ACCF surgery, a detailed medical history should be asked, cervical CT examination should be improved, the risk factors of 3D-PAVB subsidence should be carefully screened and evaluated, and the surgical plan and vertebral body replacement materials should be reasonably selected. Patients with severe osteoporosis and low bone mineral density should be carefully selected for this procedure. For patients with osteoporosis or bone loss undergoing ACCF surgery, the wearing time of neck support should be extended appropriately after surgery, anti-osteoporosis treatment, and strict regular follow-up should be conducted to reduce the occurrence of 3D-PAVB subsidence.
This study has some limitations. First of all, this was a retrospective study with a small sample size and short follow-up time, and we only studied the relationship between early 3D-PAVB subsidence and functional changes in patients. Second, although this study identified smoking and lower vertebral HU as independent risk factors for subsidence, we did not perform DXA tests to demonstrate the correlation between smoking and BMD and BMD and CT values. Finally, cervical CT should be used to measure fusion level subsidence and further evaluate fusion, rather than relying solely on lateral radiographs. Therefore, large sample, prospective and long-term follow-up studies are needed to support the conclusions of this study.