Object The long-term outcome of lumbar dynamic stabilization is uncertain. This study aimed to investigate the incidence, risk factors, and outcomes associated with screw loosening in a dynamic stabilization system. Methods The authors conducted a retrospective review of medical records, radiological studies, and clinical evaluations obtained in consecutive patients who underwent 1- or 2-level lumbar dynamic stabilization and were followed up for more than 24 months. Loosening of screws was determined on radiography and CT scanning. Radiographic and standardized clinical outcomes, including the visual analog scale (VAS) and Oswestry Disability Index (ODI) scores, were analyzed with a focus on cases in which screw loosening occurred. Results The authors analyzed 658 screws in 126 patients, including 54 women (42.9%) and 72 men (57.1%) (mean age 60.4 ± 11.8 years). During the mean clinical follow-up period of 37.0 ± 7.1 months, 31 screws (4.7%) in 25 patients (19.8%) were shown to have loosened. The mean age of patients with screw loosening was significantly higher than those without loosening (64.8 ± 8.8 vs 59.3 ± 12.2, respectively; p = 0.036). Patients with diabetes mellitus had a significantly higher rate of screw loosening compared with those without diabetes (36.0% vs 15.8%, respectively; p = 0.024). Diabetic patients with well-controlled serum glucose (HbA1c ≤ 8.0%) had a significantly lower chance of screw loosening than those without well-controlled serum glucose (28.6% vs 71.4%, respectively; p = 0.021). Of the 25 patients with screw loosening, 22 cases (88%) were identified within 6.6 months of surgery; 18 patients (72%) had the loosened screws in the inferior portion of the spinal construct, whereas 7 (28%) had screw loosening in the superior portion of the construct. The overall clinical outcomes at 3, 12, and 24 months, measured by VAS for back pain, VAS for leg pain, and ODI scores, were significantly improved after surgery compared with before surgery (all p < 0.05). There were no significant differences between the patients with and without screw loosening at all evaluation time points (all p > 0.05). All 25 patients with screw loosening were asymptomatic, and in 6 (24%) osseous integration was demonstrated on later follow-up. Also, there were 3 broken screws (2.38% in 126 patients or 0.46% in 658 screws). To date, none of these loosened or broken screws have required revision surgery. Conclusions Screw loosening in dynamic stabilization systems is not uncommon (4.7% screws in 19.8% patients). Patients of older age or those with diabetes have higher rates of screw loosening. Screw loosening can be asymptomatic and presents opportunity for osseous integration on later follow-up. Although adverse effects on clinical outcomes are rare, longer-term follow-up is required in cases in which screws become loose.

Object This study aimed to determine the age- and sex-specific incidence of cervical spondylotic myelopathy (CSM) and its associated risk of causing subsequent spinal cord injury (SCI). Methods Using the National Health Insurance Research Database (NHIRD), a 12-year nationwide database in Taiwan, this retrospective cohort study analyzed the incidence of hospitalization caused by CSM. All patients diagnosed with and admitted for CSM were identified during the study period. The CSM patients were divided into 2 groups, a control group and an operated group. An incidence density method was used to estimate age- and sex-specific incidence rates of CSM. The Kaplan-Meier method and Cox regression analyses were performed to compare the risk of SCI between the 2 groups. Results From 1998 to 2009, covering 349.5 million person-years, 14,140 patients were hospitalized for CSM. The overall incidence of CSM-related hospitalization was 4.04 per 100,000 person-years. Specifically, males and older persons had a higher incidence rate of CSM. During the follow-up of these patients for 13,461 person-years, a total of 166 patients were diagnosed with SCI. The incidence of SCI was higher in the control group than the operated group (13.9 vs 9.4 per 1000 person-years, respectively). During the follow-up, SCI was more likely to occur in CSM patients who were treated conservatively (crude HR 1.48, p = 0.023; adjusted HR 1.57, p = 0.011) than in those who underwent surgery for CSM. Conclusions In a national cohort of eastern Asia, the incidence of CSM-caused hospitalization was 4.04 per 100,000 person-years, with higher incidences observed in older and male patients. Subsequent SCI was more likely to develop in patients who received nonoperative management than in those who underwent surgery. Therefore, patients with CSM managed without surgery should be cautioned about SCI. However, further investigations are still required to clarify the risks and complications associated with surgery for CSM.

Object Dynamic stabilization systems are used to stabilize degenerative lumbar spondylosis. Loosening of the pedicle screws in such nonfusion implants is predictable. This retrospective study evaluated the incidence of screw loosening and its effect on clinical outcomes. Methods Charts, radiographic films, and medical records of 71 consecutive patients who underwent decompression using Dynesys dynamic stabilization for 1- or 2-level lumbar spondylosis were reviewed. Radiographic films were evaluated and compared to detect screw loosening. A visual analog scale (VAS) for back pain and the Oswestry Disability Index (ODI) were used for measuring clinical outcome. Statistical analysis was conducted using the chisquare test and Student t-test. Results The 71 patients in the study sample had a mean age of 59.2 ± 11.65 years (range 23–80 years), with slight female predominance (39 women, 32 men). The mean follow-up duration was 16.6 months (range 8–29 months). There were loose screws in 14 of 71 patients (19.7%), for a rate of 4.6% per screw (17 of 368 screws). Most screw loosening occurred in patients ≥ 55 years old (13 of 14 patients) although age and sex had no effect on screw loosening (p = 0.233 and 0.109, respectively). Both the loose screw and solid screw groups experienced significant improvement after the surgery in VAS and ODI scores. On the VAS, scores improved from 5.9 ± 2.99 to 2.1 ± 2.14 in the loose screw group (p = 0.003), and from 5.7 ± 3.45 to 2.9 ± 2.68 in the solid screw group (p < 0.001). For the ODI scale, scores improved from 43.5 ± 16.78% to 28.0 ± 18.18% (p = 0.006) in the loose screw group, and from 52.1 ± 20.92% to 24.6 ± 19.78% (p < 0.001) in the solid screw group. There were no significant differences between the 2 groups (p = 0.334 for VAS, p = 0.567 for ODI). Conclusions The preliminary study of this pedicle-based dynamic stabilization device for 1- and 2-level lumbar spondylosis shows radiographic evidence of screw loosening in 19.7% of patients and 4.6% of screws. Nonetheless, the loosening of screws has no adverse effect on clinical improvement.

Artificial intelligence (AI) based on convolutional neural networks (CNNs) has a great potential to enhance medical workflow and improve health care quality. Of particular interest is practical implementation of such AI-based software as a cloud-based tool aimed for telemedicine, the practice of providing medical care from a distance using electronic interfaces.Methods: In this study, we used a dataset of labeled 35,900 optical coherence tomography (OCT) images obtained from age-related macular degeneration (AMD) patients and used them to train three types of CNNs to perform AMD diagnosis.Results: Here, we present an AI- and cloud-based telemedicine interaction tool for diagnosis and proposed treatment of AMD. Through deep learning process based on the analysis of preprocessed optical coherence tomography (OCT) imaging data, our AI-based system achieved the same image discrimination rate as that of retinal specialists in our hospital. The AI platform's detection accuracy was generally higher than 90% and was significantly superior (p < 0.001) to that of medical students (69.4% and 68.9%) and equal (p = 0.99) to that of retinal specialists (92.73% and 91.90%). Furthermore, it provided appropriate treatment recommendations comparable to those of retinal specialists.Conclusions: We therefore developed a website for realistic cloud computing based on this AI platform, available at https://www.ym.edu.tw/~AI-OCT/. Patients can upload their OCT images to the website to verify whether they have AMD and require treatment. Using an AI-based cloud service represents a real solution for medical imaging diagnostics and telemedicine.

Object Heterotopic ossification (HO) after cervical total disc replacement (TDR) has been reported to impede artificial disc motion. In all previously reported cases of HO, assessment was based on plain radiographs. The authors hypothesized that CT scan is a more sensitive and accurate detector. The aims of this study were to assess the actual incidence of HO and its effect on outcome in a cohort of patients undergoing cervical TDR with the Bryan disc and to compare HO detection by means of plain radiographs and CT. Methods The authors retrospectively assessed data from medical records, radiological studies, and clinical evaluations of patients who underwent 1- or 2-level cervical TDR with the Bryan disc and were followed up for more than 12 months. The presence and grading of HO according to the McAfee classification were assessed by CT scan, and these findings were compared with findings on plain radiographs. Thirty-six patients (mean age 46.61 ± 7.24 years; range 29–60 years; 21 men and 15 women) who underwent Bryan TDR at 52 levels were included in the study. The mean duration of CT follow-up was 19.03 ± 4.64 months; the mean duration of clinical follow-up was 26.78 ± 7.20 months. Results On the basis of CT, HO was identified in 18 (50%) of 36 patients and 25 (48.1%) of 52 levels treated. Grade 1 HO was present in 9 of the levels treated (17.3%), Grade 2 in 13 levels (25.0%), Grade 3 in 2 levels (3.8%), and Grade 4 in 1 level (1.9%). Nineteen (76%) of the 25 affected levels were in patients who had undergone 2-level TDR. There was no significant association with patient sex or disc pathology. There was a tendency for HO development among older patients, but this finding was not statistically significant (mean age 48.8 ± 6.8 in patients with HO vs 44.4 ± 7.2 in those without HO, p = 0.065). Although HO was found in 25 levels, 96.2% of the treated levels (50 of 52) had segmental range of motion on dynamic (flexion and extension) radiographs. The concordance between HO grading by CT and radiography was high, with an intraclass correlation coefficient of 0.822 (lower limit of 95% CI: 0.710, p < 0.001). Patients who had HO had the same clinical success rate as those who did not (94.4% vs 94.4%, p = 1.00). The visual analog scale scores for neck and arm pain were significantly improved in both the HO and the non-HO group. Conclusions The rate of HO detected by CT scan in this cohort of patients undergoing cervical TDR with a Bryan disc was 48.1% per level treated and 50% per patient with minimal limitation of segmental motion (96.2% of levels remained mobile), but plain radiograph is an acceptable detection tool. Two-level surgery has a higher risk of HO, although development of HO does not affect clinical outcome.

No abstract

No abstract

Object Cervical arthroplasty is a valid option for patients with single-level symptomatic cervical disc diseases causing neural tissue compression, but postoperative heterotopic ossification (HO) can limit the mobility of an artificial disc. In the present study the authors used CT scanning to assess HO formation, and they investigated differences in radiological and clinical outcomes in patients with either a soft-disc herniation or spondylosis who underwent cervical arthroplasty. Methods Medical records, radiographs, and clinical evaluations of consecutive patients who underwent single-level cervical arthroplasty were reviewed. Arthroplasty was performed using the Bryan disc. The patients were divided into a soft-disc herniation group and a spondylosis group. Clinical outcomes were measured using the visual analog scale (VAS) for neck and arm pain and the Neck Disability Index (NDI), whereas HO grading was determined by studying CT scans. Radiological and clinical outcomes were analyzed, and the minimum follow-up duration was 24 months. Results Forty-seven consecutive patients underwent a single-level cervical arthroplasty. Forty patients (85.1%) had complete radiological evaluations and clinical follow-up of more than 2 years. Patients were divided into 1 of 2 groups: soft-disc herniation (16 cases) and the spondylosis group (24 cases). Their mean age was 45.51 ± 11.12 years. Sixteen patients (40%) were female. Patients in the soft-disc herniation group were younger than those in the spondylosis group, but the difference was not statistically significant (42.88 vs 47.26, p = 0.227). The mean follow-up duration was 38.83 ± 9.74 months. Sex, estimated blood loss, implant size, and perioperative NSAID prescription were not significantly different between the groups (p = 0.792, 0.267, 0.581, and 1.000, respectively). The soft-disc herniation group had significantly less HO formation than the spondylosis group (1 HO [6.25%] vs 14 Hos [58.33%], p = 0.001). Almost all artificial discs in both groups remained mobile (100% and 95.8%, p = 0.408). The clinical outcomes were not significantly different between the groups at all postoperative time points of evaluation, and clinical improvements were also similar. Conclusions Clinical outcomes of single-level cervical arthroplasty for soft-disc herniation and spondylosis were similar 3 years after surgery. There was a significantly higher rate of HO formation in patients with spondylosis than in those with a soft-disc herniation. The mobility of the artificial disc is maintained, but the long-term effects of HO and its higher frequency in spondylotic cases warrant further investigation.