The emergency department is an environment with a potential risk for diagnostic errors during trauma care, particularly for fractures. Convolutional neural network (CNN) deep learning methods are now widely used in medicine because they improve diagnostic accuracy, decrease misinterpretation, and improve efficiency. In this study, we investigated whether automatic localization and classification using CNN could be applied to pelvic, rib, and spine fractures. We also examined whether this fracture detection algorithm could help physicians in fracture diagnosis. A total of 7664 whole-body CT axial slices (chest, abdomen, pelvis) from 200 patients were used. Sensitivity, precision, and F1-score were calculated to evaluate the performance of the CNN model. For the grouped mean values for pelvic, spine, or rib fractures, the sensitivity was 0.786, precision was 0.648, and F1-score was 0.711. Moreover, with CNN model assistance, surgeons showed improved sensitivity for detecting fractures and the time of reading and interpreting CT scans was reduced, especially for less experienced orthopedic surgeons. Application of the CNN model may lead to reductions in missed fractures from whole-body CT images and to faster workflows and improved patient care through efficient diagnosis in polytrauma patients.

BackgroundComponent design is one of the contributory factors affecting the postoperative flexion angle. The purpose of this study was to compare short-term outcomes of flat surface and medial pivot designs in posterior cruciate-retaining (CR) total knee arthroplasty (TKA).MethodsA retrospective, case-control, and observational cohort study consisted of matched-pairs of the flat surface design (Hi-Tech Knee II) and the medial pivot design (FINE Knee) in CR-TKA with a two-year follow-up period.ResultsHi-Tech Knee II and FINE knee groups each included 7 males and 38 females. Surgical time was significantly shorter in the FINE Knee group than in the Hi-Tech Knee II group (104.8 min versus 154.9 min, p = 0.001). Estimated total blood loss was significantly lower in the FINE Knee group than in the Hi-Tech Knee II group (654 ml versus 1158 ml, p = 0.001). The postoperative flexion angle was significantly better in the FINE Knee group than in the Hi-Tech Knee II group (119.3 degrees versus 112.5 degrees), and was positively correlated with the preoperative flexion angle. Postoperative Knee Society scores were significantly better in the FINE Knee group than in the Hi-Tech Knee II group (93.0 points versus 85.0 points, p = 0.001), especially for postoperative pain relief (46.0 points versus 39.0 points out of 50, p = 0.001). Complications were not observed in either group over a two-year follow-up period.ConclusionThe short-term outcome of the medial pivot design used in CR-TKA was more favorable than the flat surface design, especially for surgical time, estimated total blood loss, postoperative flexion angle, and knee pain.Electronic supplementary materialThe online version of this article (10.1186/s12891-018-2138-z) contains supplementary material, which is available to authorized users.

OBJECTIVE It is unclear whether anterior cervical decompression and fusion (ADF) or laminoplasty (LMP) results in better outcomes for patients with K-line–positive (+) cervical ossification of the posterior longitudinal ligament (OPLL). The purpose of the study is to compare surgical outcomes and complications of ADF versus LMP in patients with K-line (+) OPLL. METHODS The study included 478 patients enrolled in the Japanese Multicenter Research Organization for Ossification of the Spinal Ligament and who underwent surgical treatment for cervical OPLL. The patients who underwent anterior-posterior combined surgery or posterior decompression with instrumented fusion were excluded. The patients with a follow-up period of fewer than 2 years were also excluded, leaving 198 patients with K-line (+) OPLL. Propensity score matching was performed on 198 patients with K-line (+) OPLL who underwent ADF (44 patients) or LMP (154 patients), resulting in 39 pairs of patients based on the following predictors for surgical outcomes: age, preoperative Japanese Orthopaedic Association (JOA) score, C2–7 angle, and the occupying ratio of OPLL. Clinical outcomes were assessed 1 and 2 years after surgery using the recovery rate of the JOA score. Complications and reoperation rates were also investigated. RESULTS The mean recovery rate of the JOA score 1 year after surgery was 55.3% for patients who underwent ADF and 42.3% (p = 0.06) for patients who underwent LMP. Two years after surgery, the recovery rate was 53.4% for those who underwent ADF and 38.7% for LMP (p = 0.07). Although both surgical procedures yielded good results, the mean recovery rate of JOA scores tended to be higher in the ADF group. The incidence of surgical complications, however, was higher following ADF (33%) than LMP (15%; p = 0.06). The reoperation rate was also higher in the ADF group (15%) than in the LMP group (0%; p = 0.01). CONCLUSIONS Clinical outcomes were good for both ADF and LMP, indicating that ADF and LMP are appropriate procedures for patients with K-line (+) OPLL. Clinical outcomes of ADF 1 and 2 years after surgery tended to be better than LMP, but the analysis did not detect any significant difference in clinical outcomes between the groups. Conversely, patients who underwent ADF had a higher incidence of surgery-related complications. When considering indications for ADF or LMP, benefits and risks of the surgical procedures should be carefully weighed.

No abstract

The number of elderly patients with spinal cord injury without radiographic abnormalities (SCIWORA) has been increasing in recent years and common of most cervical spinal cord injuries. Basic research has shown the effectiveness of early decompression after spinal cord injury on the spinal cord without stenosis; no studies have reported the efficacy of decompression in models with spinal cord compressive lesions. The purpose of this study was to evaluate the effects of decompression surgery after acute spinal cord injury in rats with chronic spinal cord compressive lesions, mimicking SCIWORA. A water-absorbent polymer sheet (Aquaprene DX, Sanyo Chemical Industries) was inserted dorsally into the 4–5th cervical sublaminar space in 8-week-old Sprague Dawley rats to create a rat model with a chronic spinal compressive lesion. At the age of 16 weeks, 30 mildly myelopathic or asymptomatic rats with a Basso, Beattie, and Bresnahan score (BBB score) of 19 or higher were subjected to spinal cord compression injuries. The rats were divided into three groups: an immediate decompression group (decompress immediately after injury), a sub-acute decompression group (decompress 1 week after injury), and a non-decompression group. Behavioral and histological evaluations were performed 4 weeks after the injury. At 20 weeks of age, the BBB score and FLS (Forelimb Locomotor Scale) of both the immediate and the sub-acute decompression groups were significantly higher than those of the non-decompression group. There was no significant difference between the immediate decompression group and the sub-acute decompression group. TUNEL (transferase-mediated dUTP nick end labeling) staining showed significantly fewer positive cells in both decompression groups compared to the non-decompression group. LFB (Luxol fast blue) staining showed significantly more demyelination, and GAP-43 (growth associated protein-43) staining tended to show fewer positive cells in the non-decompression group. Decompression surgery in the acute or sub-acute phase of injury is effective after mild spinal cord injury in rats with chronic compressive lesions. There was no significant difference between the immediate decompression and sub-acute decompression groups.