Abstract: Improvement in survival in patients with advanced cancer is accompanied by an increased probability of bone metastasis and related pathologic fractures (especially in the proximal femur). The few systems proposed and used to diagnose impending fractures owing to metastasis and to ultimately prevent future fractures have practical limitations; thus, novel screening tools are essential. A CT scan of the abdomen and pelvis is a standard modality for staging and follow-up in patients with cancer, and radiologic assessments of the proximal femur are possible with CT-based digitally reconstructed radiographs. Deep-learning models, such as convolutional neural networks (CNNs), may be able to predict pathologic fractures from digitally reconstructed radiographs, but to our knowledge, they have not been tested for this application. Questions/purposes (1) How accurate is a CNN model for predicting a pathologic fracture in a proximal femur with metastasis using digitally reconstructed radiographs of the abdomen and pelvis CT images in patients with advanced cancer? (2) Do CNN models perform better than clinicians with varying backgrounds and experience levels in predicting a pathologic fracture on abdomen and pelvis CT images without any knowledge of the patients’ histories, except for metastasis in the proximal femur? Methods A total of 392 patients received radiation treatment of the proximal femur at three hospitals from January 2011 to December 2021. The patients had 2945 CT scans of the abdomen and pelvis for systemic evaluation and follow-up in relation to their primary cancer. In 33% of the CT scans (974), it was impossible to identify whether a pathologic fracture developed within 3 months after each CT image was acquired, and these were excluded. Finally, 1971 cases with a mean age of 59 ± 12 years were included in this study. Pathologic fractures developed within 3 months after CT in 3% (60 of 1971) of cases. A total of 47% (936 of 1971) were women. Sixty cases had an established pathologic fracture within 3 months after each CT scan, and another group of 1911 cases had no established pathologic fracture within 3 months after CT scan. The mean age of the cases in the former and latter groups was 64 ± 11 years and 59 ± 12 years, respectively, and 32% (19 of 60) and 53% (1016 of 1911) of cases, respectively, were female. Digitally reconstructed radiographs were generated with perspective projections of three-dimensional CT volumes onto two-dimensional planes. Then, 1557 images from one hospital were used for a training set. To verify that the deep-learning models could consistently operate even in hospitals with a different medical environment, 414 images from other hospitals were used for external validation. The number of images in the groups with and without a pathologic fracture within 3 months after each CT scan increased from 1911 to 22,932 and from 60 to 720, respectively, using data augmentation methods that are known to be an effective way to boost the performance of deep-learning models. Three CNNs (VGG16, ResNet50, and DenseNet121) were fine-tuned using digitally reconstructed radiographs. For performance measures, the area under the receiver operating characteristic curve, accuracy, sensitivity, specificity, precision, and F1 score were determined. The area under the receiver operating characteristic curve was used to evaluate three CNN models mainly, and the optimal accuracy, sensitivity, and specificity were calculated using the Youden J statistic. Accuracy refers to the proportion of fractures in the groups with and without a pathologic fracture within 3 months after each CT scan that were accurately predicted by the CNN model. Sensitivity and specificity represent the proportion of accurately predicted fractures among those with and without a pathologic fracture within 3 months after each CT scan, respectively. Precision is a measure of how few false-positives the model produces. The F1 score is a harmonic mean of sensitivity and precision, which have a tradeoff relationship. Gradient-weighted class activation mapping images were created to check whether the CNN model correctly focused on potential pathologic fracture regions. The CNN model with the best performance was compared with the performance of clinicians. Results DenseNet121 showed the best performance in identifying pathologic fractures; the area under the receiver operating characteristic curve for DenseNet121 was larger than those for VGG16 (0.77 ± 0.07 [95% CI 0.75 to 0.79] versus 0.71 ± 0.08 [95% CI 0.69 to 0.73] ; p = 0.001) and ResNet50 (0.77 ± 0.07 [95% CI 0.75 to 0.79] versus 0.72 ± 0.09 [95% CI 0.69 to 0.74] ; p = 0.001). Specifically, DenseNet121 scored the highest in sensitivity (0.22 ± 0.07 [95% CI 0.20 to 0.24]), precision (0.72 ± 0.19 [95% CI 0.67 to 0.77] ), and F1 score (0.34 ± 0.10 [95% CI 0.31 to 0.37]), and it focused accurately on the region with the expected pathologic fracture. Further, DenseNet121 was less likely than clinicians to mispredict cases in which there was no pathologic fracture than cases in which there was a fracture; the performance of DenseNet121 was better than clinician performance in terms of specificity (0.98 ± 0.01 [95% CI 0.98 to 0.99] versus 0.86 ± 0.09 [95% CI 0.81 to 0.91]; p = 0.01), precision (0.72 ± 0.19 [95% CI 0.67 to 0.77] versus 0.11 ± 0.10 [95% CI 0.05 to 0.17]; p = 0.0001), and F1 score (0.34 ± 0.10 [95% CI 0.31 to 0.37] versus 0.17 ± 0.15 [95% CI 0.08 to 0.26]; p = 0.0001). Conclusion CNN models may be able to accurately predict impending pathologic fractures from digitally reconstructed radiographs of the abdomen and pelvis CT images that clinicians may not anticipate; this can assist medical, radiation, and orthopaedic oncologists clinically. To achieve better performance, ensemble-learning models using knowledge of the patients’ histories should be developed and validated. The code for our model is publicly available online at https://github.com/taehoonko/CNN_path_fx_prediction. Level of Evidence Level III, diagnostic study.

Abstract: The purpose of the present study was to evaluate the incidence rate (IR) of suicide in elderly patients with hip fracture on the basis of a nationwide cohort and to analyze the change in the hazard ratio for suicide after hip fracture over time in comparison with a control group. Methods: Patients with hip fracture and their matched controls were selected from the National Health Insurance Service-Senior cohort (NHIS-Senior) of the Republic of Korea. The NHIS-Senior consists of 558,147 people selected by a 10% simple random-sampling method from a total of 5.5 million subjects ≥60 years of age in 2002. Risk-set matching (1:2) on the propensity score was performed with use of a nearest neighbor matching algorithm with a maximum caliper of 0.1 for the hazard components. The IR of suicide and 95% confidence interval (CI) were calculated on the basis of a generalized linear model with a Poisson distribution. The effect size was presented as a hazard ratio (HR) with use of the Cox proportional-hazard model with a robust variance estimator that accounts for clustering within matched pairs. Results: A total of 11,477 patients with hip fracture and 22,954 matched controls were included. The mean duration of follow-up was 4.59 years, generating 158,139 person-years. During follow-up, a total of 170 suicides were identified. Comparisons at up to 180 days and 365 days showed that patients with hip fracture were at higher risk for suicide than matched controls (p = 0.009 and 0.004, respectively; stratified log-rank test). During the first 180 days of follow-up, 14 suicides were identified in patients with hip fracture during 11,152 person-years (IR, 266.1 per 100,000 person-years; 95% CI, 157.6 to 449.4). Patients with hip fracture were 2.97 times more likely to kill themselves than their matched controls during the same period (HR = 2.97; 95% CI, 1.32 to 6.69). Conclusions: Hip fracture in elderly patients increased suicide risk within a year. A new approach to psychiatric evaluation and management is needed in elderly patients with hip fracture. Level of Evidence: Therapeutic Level III . See Instructions for Authors for a complete description of levels of evidence.

Abstract: Based on its mechanism, the use of balloon guide catheters (BGCs) may be beneficial during endovascular treatment, regardless of the type of mechanical recanalization modality used—stent retriever thrombectomy or thrombaspiration. We evaluated whether the use of BGCs can be beneficial regardless of the first-line mechanical endovascular modality used. Methods— We retrospectively reviewed consecutive acute stroke patients who underwent stent retriever thrombectomy or thrombaspiration from the prospectively maintained registries of 17 stroke centers nationwide. Patients were assigned to the BGC or non-BGC group based on the use of BGCs during procedures. Endovascular and clinical outcomes were compared between the BGC and non-BGC groups. To adjust the influence of the type of first-line endovascular modality on successful recanalization and favorable outcome, multivariable analyses were also performed. Results— This study included a total of 955 patients. Stent retriever thrombectomy was used as the first-line modality in 526 patients (55.1%) and thrombaspiration in 429 (44.9%). BGC was used in 516 patients (54.0%; 61.2% of stent retriever thrombectomy patients; 45.2% of thrombaspiration patients). The successful recanalization rate was significantly higher in the BGC group compared with the non-BGC group (86.8% versus 74.7%, respectively; P 〈 0.001). Furthermore, the first-pass recanalization rate was more frequent (37.0% versus 14.1%; P 〈 0.001), and the number of device passes was fewer in the BGC group (2.5±1.9 versus 3.3±2.1; P 〈 0.001). The procedural time was also shorter in the BGC group (54.3±27.4 versus 67.6±38.2; P 〈 0.001). The use of BGC was an independent factor for successful recanalization (odds ratio, 2.18; 95% CI, 1.54–3.10; P 〈 0.001) irrespective of the type of first-line endovascular modality used. The use of BGC was also an independent factor for a favorable outcome (odds ratio, 1.40; 95% CI, 1.02–1.92; P =0.038) irrespective of the type of first-line endovascular modality used. Conclusions— Regardless of the first-line mechanical endovascular modality used, the use of BGC in endovascular treatment was beneficial in terms of both recanalization success and functional outcome.

Abstract: Chronic hepatitis B (CHB) remains a major worldwide public health concern. Besifovir dipivoxil maleate (BSV) is a new promising treatment for CHB. However, long-term efficacy and safety have not yet been evaluated. Therefore, the goal of the study is to determine the antiviral efficacy and safety of BSV treatment over a 144-week duration (BSV-BSV) in comparison with those of a sequential treatment with tenofovir disoproxil fumarate (TDF) followed by a 96-week duration BSV administration (TDF-BSV). METHODS: After 48 weeks of a double-blind comparison between BSV and TDF treatments, patients continued the open-label BSV study. We evaluated antiviral efficacy and drug safety up to 144 weeks for BSV-BSV and TDF-BSV groups. The primary endpoint was a virological response (hepatitis B virus DNA 〈 69 IU/mL). RESULTS: Among the 197 patients enrolled, 170 and 158 patients entered the second-year and third-year open-label phase extensional study, respectively, whereas 153 patients completed the 144-week follow-up. The virological response rate over the 144-week period was 87.7% and 92.1% in BSV-BSV and TDF-BSV groups, respectively ( P = 0.36). The rates of ALT normalization and HBeAg seroconversion were similar between the groups. No drug-resistant mutations to BSV were noted. Bone mineral density and renal function were well preserved in the BSV-BSV group and were significantly improved after switching therapy in TDF-BSV patients. DISCUSSION: This extensional study of a phase 3 trial (NCT01937806) suggests that BSV treatment is efficacious and safe for long-term use in treatment-naïve and TDF-experienced patients with CHB.

Abstract: Cervicogenic headache has been known to originate from the convergence of the upper 3 cervical and trigeminal afferents. The administration of conservative treatments, interventional procedures, and more recently, pulsed radiofrequency, has been used to relieve cervicogenic headache. In this study, the authors evaluated the clinical efficacy and safety of pulsed radiofrequency targeting the mid-cervical medial branches. Materials and Methods: From September 2012 until December 2017, 395 patients were diagnosed with cervicogenic headache based on the third edition of the International Classification of Headache Disorders. The authors treated them conservatively at first, and those patients with nonresolution of pain were treated with mid-cervical medial branches block applied from C3 to C5 twice. Subsequently, if any patient continued to experience persistent pain, the authors classified them as having intractable cervicogenic headache and performed pulsed radiofrequency treatment targeting the mid-cervical medial branches from C3 to C5 bilaterally. The authors analyzed their demographics and used a Visual Analogue Scale to assess their pain for 12 months. Results: Fifty-seven patients were enrolled in this study. The mean age was 49.8 years, and the mean duration of symptoms was 47.7 months. The mean Visual Analogue Scale score was 6.21 before pulsed radiofrequency treatment, and it improved to 1.54 immediately after the procedure without the symptoms recurring for a minimum of 12 months. There were no severe complications, such as vascular or nerve injuries. Conclusions: In patients with intractable cervicogenic headache, pulsed radiofrequency treatment targeting the mid-cervical medial branches resulted in a satisfactory, long-lasting outcome without serious complications.

Abstract: Although cilostazol has shown less hemorrhagic events than aspirin, only marginal difference was observed in hemorrhagic stroke events among patients at high risk for cerebral hemorrhage. To identify patients who would most benefit from cilostazol, this study analyzed interactions between treatment and subgroups of the PICASSO trial (Prevention of Cardiovascular Events in Asian Ischemic Stroke Patients With High Risk of Cerebral Hemorrhage). Methods— Ischemic stroke patients with a previous intracerebral hemorrhage or multiple microbleeds were randomized to treatment with cilostazol or aspirin and followed up for a mean 1.8 years. Efficacy, defined as the composite of any stroke, myocardial infarction, and vascular death, and safety, defined as the incidence of hemorrhagic stroke, were analyzed in the 2 groups. Interactions between treatment and age, sex, presence of hypertension and diabetes mellitus, index of high-risk cerebral hemorrhage, and white matter lesion burden were analyzed for primary and key secondary outcomes. Changes in vital signs and laboratory results were compared in the 2 groups. Results— Among all 1534 patients enrolled, a significant interaction between treatment group and index of high risk for cerebral hemorrhage on hemorrhagic stroke ( P for interaction, 0.03) was observed. Hemorrhagic stroke was less frequent in the cilostazol than in the aspirin group in patients with multiple microbleeds (1 versus 13 events; hazard ratio, 0.08 [95% CI, 0.01–0.61]; P =0.01). A marginal interaction between treatment group and white matter change on any stroke ( P for interaction, 0.08) was observed. Cilostazol reduced any stroke significantly in patients with mild (5 versus 16 events; hazard ratio, 0.36 [95% CI, 0.13–0.97]; P =0.04)-to-moderate (16 versus 32 events; hazard ratio, 0.50 [95% CI, 0.29–0.92]; P =0.03) white matter changes. Heart rate and HDL (high-density lipoprotein) cholesterol level were significantly higher in the cilostazol group than in the aspirin group at follow-up. Conclusions— Cilostazol may be more beneficial for ischemic stroke patients with multiple cerebral microbleeds and before white matter changes are extensive. Registration— URL: https://www.clinicaltrials.gov . Unique identifier: NCT01013532.