External validation of the SORG machine learning algorithms for predicting 90-day and 1-year survival of patients with lung cancer-derived spine metastases: a recent bi-center cohort from China

BACKGROUND CONTEXTThe survival prediction of lung cancer-derived spinal metastases is often underestimated by several scores. The SORG machine learning (ML) algorithm is considered a promising tool to predict the risk of 90-day and 1-year mortality in patients with spinal metastases, but not been externally validated for lung cancer.PURPOSEThis study aimed to externally validate the SORG ML algorithms on lung cancer-derived spinal metastases patients from two large-volume, tertiary medical centers between 2018 and 2021.STUDY DESIGN/SETTINGRetrospective, cohort study.PATIENT SAMPLEPatients aged 18 years or older at two tertiary medical centers in China are treated surgically for spinal metastasis.OUTCOME MEASURESMortality within 90 days of surgery, mortality within 1 year of surgery.METHODSThe baseline characteristics were compared between the development cohort and our validation cohort. Discrimination (receiver operating curve), calibration (calibration plot, intercept, and slope), the overall performance (Brier score), and decision curve analysis was used to assess the overall performance of the SORG ML algorithms.RESULTSThis study included 150 patients with lung cancer-derived spinal metastases from two medical centers in China. Ninety-day and 1-year mortality rates were 12.9% (19/147) and 51.3% (60/117), respectively. Lung Cancer with targeted therapies had the lowest Hazard Ratio (HR=0.490), showing an optimal protecting factor. The AUC of the SORG ML algorithm for 90-day mortality prediction in lung cancer-derived spinal metastases is 0.714. While the AUC for 1-year mortality prediction is 0.832 (95CI%, 0.758–0.906). The algorithm for 1-year mortality was well-calibrated with an intercept of 0.13 and a calibration slope of 1.00. However, the 90-day mortality prediction was underestimated with an intercept of 0.60 and a slope of 0.37. The SORG ML algorithms for 1-year mortality showed a greater net benefit than the “treats all or no patients” strategies.CONCLUSIONSIn the latest cohort of lung cancer-derived spinal metastases in China, the SORG algorithms for predicting 1-year mortality performed well on external validation. However, 90-day mortality was underestimated. The algorithm should be further validated by single primary tumor-derived metastasis treated with the latest comprehensive treatment in diverse populations.     

Updated external validation of the SORG machine learning algorithms for prediction of ninety-day and one-year mortality after surgery for spinal metastasis

BACKGROUND CONTEXTSurgical decompression and stabilization in the setting of spinal metastasis is performed to relieve pain and preserve functional status. These potential benefits must be weighed against the risks of perioperative morbidity and mortality. Accurate prediction of a patient's postoperative survival is a crucial component of patient counseling.PURPOSETo externally validate the SORG machine learning algorithms for prediction of 90-day and 1-year mortality after surgery for spinal metastasis.STUDY DESIGN/SETTINGRetrospective, cohort studyPATIENT SAMPLEPatients 18 years or older at a tertiary care medical center treated surgically for spinal metastasisOUTCOME MEASURESMortality within 90 days of surgery, mortality within 1 year of surgeryMETHODSThis is a retrospective cohort study of 298 adult patients at a tertiary care medical center treated surgically for spinal metastasis between 2004 and 2020. Baseline characteristics of the validation cohort were compared to the derivation cohort for the SORG algorithms. The following metrics were used to assess the performance of the algorithms: discrimination, calibration, overall model performance, and decision curve analysis.RESULTSSixty-one patients died within 90 days of surgery and 133 died within 1 year of surgery. The validation cohort differed significantly from the derivation cohort. The SORG algorithms for 90-day mortality and 1-year mortality performed excellently with respect to discrimination; the algorithm for 1-year mortality was well-calibrated. At both postoperative time points, the SORG algorithms showed greater net benefit than the default strategies of changing management for no patients or for all patients.CONCLUSIONSWith an independent, contemporary, and geographically distinct population, we report successful external validation of SORG algorithms for preoperative risk prediction of 90-day and 1-year mortality after surgery for spinal metastasis. By providing accurate prediction of intermediate and long-term mortality risk, these externally validated algorithms may inform shared decision-making with patients in determining management of spinal metastatic disease.     

External validation of a predictive algorithm for in-hospital and 90-day mortality after spinal epidural abscess

BACKGROUND CONTEXTMortality in patients with spinal epidural abscess (SEA) remains high. Accurate prediction of patient-specific prognosis in SEA can improve patient counseling as well as guide management decisions. There are no externally validated studies predicting short-term mortality in patients with SEA.PURPOSEThe purpose of this study was to externally validate the Skeletal Oncology Research Group (SORG) stochastic gradient boosting algorithm for prediction of in-hospital and 90-day postdischarge mortality in SEA.STUDY DESIGN/SETTINGRetrospective, case-control study at a tertiary care academic medical center from 2003 to 2021.PATIENT SAMPLEAdult patients admitted for radiologically confirmed diagnosis of SEA who did not initiate treatment at an outside institution.OUTCOME MEASURESIn-hospital and 90-day postdischarge mortality.METHODSWe tested the SORG stochastic gradient boosting algorithm on an independent validation cohort. We assessed its performance with discrimination, calibration, decision curve analysis, and overall performance.RESULTSA total of 212 patients met inclusion criteria, with a short-term mortality rate of 10.4%. The area under the receiver operating characteristic curve (AUROC) of the SORG algorithm when tested on the full validation cohort was 0.82, the calibration intercept was -0.08, the calibration slope was 0.96, and the Brier score was 0.09.CONCLUSIONSWith a contemporaneous and geographically distinct independent cohort, we report successful external validation of a machine learning algorithm for prediction of in-hospital and 90-day postdischarge mortality in SEA.     

Development of a machine learning algorithm for predicting in-hospital and 1-year mortality after traumatic spinal cord injury

BACKGROUND CONTEXTCurrent prognostic tools such as the Injury Severity Score (ISS) that predict mortality following trauma do not adequately consider the unique characteristics of traumatic spinal cord injury (tSCI).PURPOSEOur aim was to develop and validate a prognostic tool that can predict mortality following tSCI.STUDY DESIGNRetrospective review of a prospective cohort study.PATIENT SAMPLEData was collected from 1245 persons with acute tSCI who were enrolled in the Rick Hansen Spinal Cord Injury Registry between 2004 and 2016.OUTCOME MEASURESIn-hospital and 1-year mortality following tSCI.METHODSMachine learning techniques were used on patient-level data (n=849) to develop the Spinal Cord Injury Risk Score (SCIRS) that can predict mortality based on age, neurological level and completeness of injury, AOSpine classification of spinal column injury morphology, and Abbreviated Injury Scale scores. Validation of the SCIRS was performed by testing its accuracy in an independent validation cohort (n=396) and comparing its performance to the ISS, a measure which is used to predict mortality following general trauma.RESULTSFor 1-year mortality prediction, the values for the Area Under the Receiver Operating Characteristic Curve (AUC) for the development cohort were 0.84 (standard deviation=0.029) for the SCIRS and 0.55 (0.041) for the ISS. For the validation cohort, AUC values were 0.86 (0.051) for the SCIRS and 0.71 (0.074) for the ISS. For in-hospital mortality, AUC values for the development cohort were 0.87 (0.028) and 0.60 (0.050) for the SCIRS and ISS, respectively. For the validation cohort, AUC values were 0.85 (0.054) for the SCIRS and 0.70 (0.079) for the ISS.CONCLUSIONSThe SCIRS can predict in-hospital and 1-year mortality following tSCI more accurately than the ISS. The SCIRS can be used in research to reduce bias in estimating parameters and can help adjust for coefficients during model development. Further validation using larger sample sizes and independent datasets is needed to assess its reliability and to evaluate using it as an assessment tool to guide clinical decision-making and discussions with patients and families.     

30-Day, 90-day and 1-year mortality after emergency colonic surgery

Purpose

Emergency surgery is an independent risk factor in colonic surgery resulting in high 30-day mortality. The primary aim of this study was to report 30-day, 90-day and 1-year mortality rates after emergency colonic surgery, and to report factors associated with 30-day, 90-day and 1-year mortality. Second, the aim was to report 30-day postoperative complications and their relation to in-hospital mortality.

Methods

All patients undergoing acute colonic surgery in the period from May 2009 to April 2013 at Copenhagen University Hospital Herlev, Denmark, were identified. Perioperative data was collected from medical journals.

Results

30-day, 90-day and 1-year mortality was 21, 30 and 41%, respectively. Age >70 years, Performance status ≥3 and resection with stoma were independent factors associated with 30-day mortality. Age >70 years, Performance status ≥3, resection with stoma and malignant disease were independent risk factors associated with 90-day mortality. Age >70 years, Performance status ≥3, resection with stoma and malignant disease were independent factors associated with 1-year mortality. Overall, 30-day complication rate was 63%, with cardiopulmonary complications leading to most postoperative deaths.

Conclusion

Mortality and complication rates after emergency colonic surgery are high and associated with patient related risk factors that cannot be modified, but also treatment related outcomes that are modifiable. An increased focus on medical and other preventive measures should be explored in the future.

     

External validation of a predictive model of survival after cytoreductive nephrectomy for metastatic renal cell carcinoma

Introduction

Recent trials have emphasized the importance of a precise patient selection for cytoreductive nephrectomy (CN). In 2013, a nomogram was developed for pre- and postoperative prediction of the probability of death (PoD) after CN in patients with metastatic renal cell carcinoma. To date, the single-institutional nomogram which included mostly patients from the cytokine era has not been externally validated. Our objective is to validate the predictive model in contemporary patients in the targeted therapy era.

Methods

Multi-institutional European and North American data from patients who underwent CN between 2006 and 2013 were used for external validation. Variables evaluated included preoperative serum albumin and lactate dehydrogenase levels, intraoperative blood transfusions (yes/no) and postoperative pathologic stage (primary tumour and nodes). In addition, patient characteristics and MSKCC risk factors were collected. Using the original calibration indices and quantiles of the distribution of predictions, Kaplan–Meier estimates and calibration plots of observed versus predicted PoD were calculated. For the preoperative model a decision curve analysis (DCA) was performed.

Results

Of 1108 patients [median OS of 27 months (95% CI 24.6–29.4)], 536 and 469 patients had full data for the validation of the pre- and postoperative models, respectively. The AUC for the pre- and postoperative model was 0.68 (95% CI 0.62–0.74) and 0.73 (95% CI 0.68–0.78), respectively. In the DCA the preoperative model performs well within threshold survival probabilities of 20–50%. Most important limitation was the retrospective collection of this external validation dataset.

Conclusions

In this external validation, the pre- and postoperative nomograms predicting PoD following CN were well calibrated. Although performance of the preoperative nomogram was lower than in the internal validation, it retains the ability to predict early death after CN.

     

Development and internal validation of machine learning algorithms for end-stage renal disease risk prediction model of people with type 2 diabetes mellitus and diabetic kidney disease

AimsDiabetic kidney disease (DKD) is the most common cause of end-stage renal disease (ESRD) and is associated with increased morbidity and mortality in patients with diabetes. Identification of risk factors involved in the progression of DKD to ESRD is expected to result in early detection and appropriate intervention and improve prognosis. Therefore, this study aimed to establish a risk prediction model for ESRD resulting from DKD in patients with type 2 diabetes mellitus (T2DM).MethodsBetween January 2008 and July 2019, a total of 390 Chinese patients with T2DM and DKD confirmed by percutaneous renal biopsy were enrolled and followed up for at least 1 year. Four machine learning algorithms (gradient boosting machine, support vector machine, logistic regression, and random forest (RF)) were used to identify the critical clinical and pathological features and to build a risk prediction model for ESRD.ResultsThere were 158 renal outcome events (ESRD) (40.51%) during the 3-year median follow up. The RF algorithm showed the best performance at predicting progression to ESRD, showing the highest AUC (0.90) and ACC (82.65%). The RF algorithm identified five major factors: Cystatin-C, serum albumin (sAlb), hemoglobin (Hb), 24-hour urine urinary total protein, and estimated glomerular filtration rate. A nomogram according to the aforementioned five predictive factors was constructed to predict the incidence of ESRD.ConclusionMachine learning algorithms can efficiently predict the incident ESRD in DKD participants. Compared with the previous models, the importance of sAlb and Hb were highlighted in the current model.

Highlights

• What is already known? Identification of risk factors for the progression of DKD to ESRD is expected to improve the prognosis by early detection and appropriate intervention.
• What this study has found? Machine learning algorithms were used to construct a risk prediction model of ESRD in patients with T2DM and DKD. The major predictive factors were found to be CysC, sAlb, Hb, eGFR, and UTP.
• What are the implications of the study? In contrast with the treatment of participants with early-phase T2DM with or without mild kidney damage, major emphasis should be placed on indicators of kidney function, nutrition, anemia, and proteinuria for participants with T2DM and advanced DKD to delay ESRD, rather than age, sex, and control of hypertension and glycemia.

     

External validation of a 5-year survival prediction model after elective abdominal aortic aneurysm repair

Objective

The benefit of prophylactic repair of abdominal aortic aneurysms (AAAs) is based on the risk of rupture exceeding the risk of death from other comorbidities. The purpose of this study was to validate a 5-year survival prediction model for patients undergoing elective repair of asymptomatic AAA <6.5 cm to assist in optimal selection of patients.

机器学习算法和COX列线图在肝细胞癌术后生存预测中的应用价值

目的探讨机器学习算法和COX列线图在肝细胞癌术后生存预测中的应用价值。方法采用回顾性描述性研究方法。收集2012年1月至2017年1月中国医学科学院北京协和医学院肿瘤医院收治的375例肝细胞癌行根治性肝切除术患者的临床病理资料;男304例,女71例;中位年龄为57岁,年龄范围为21~79岁。375例患者通过计算机产生随机数方法以8∶2比例分为训练集300例和验证集75例,应用逻辑回归、支持向量机、决策树、随机森林、人工神经网络机器学习算法构建肝细胞癌患者术后生存的预测模型,筛选性能最优的机器学习算法预测模型;构建肝细胞癌患者术后生存预测的COX列线图预测模型;比较最优机器学习算法预测模型和COX列线图预测模型预测肝细胞癌患者术后生存的性能。观察指标:(1)训练集与验证集患者临床病理资料分析。(2)训练集与验证集患者随访及生存情况。(3)机器学习算法预测模型构建及验证。(4)COX列线图预测模型构建及验证。(5)随机森林机器学习算法预测模型与COX列线图预测模型预测性能评价。采用门诊或电话方式进行随访,了解患者生存情况。随访时间截至2019年12月或患者死亡。正态分布的计量资料以±s表示,组间比较采用配对t检验。偏态分布的计量资料以M(P25,P75)或M(范围)表示,组间比较采用Mann-Whitney U检验。计数资料以绝对数表示,当Tmin≥5,N≥40时,组间比较采用χ2检验;当1≤Tmin≤5,N≥40时,采用校正χ2检验;当Tmin<1或N<40时,采用Fisher确切概率法。采用Kaplan-Meier法计算生存率和绘制生存曲线。采用COX比例风险模型进行单因素分析,将P<0.2的变量纳入Lasso回归分析,根据λ值筛选影响预后的变量,最后将变量纳入COX比例风险模型进行多因素分析。结果(1)训练集与验证集患者临床病理资料分析:训练集和验证集患者微血管侵犯(无、有),肝硬化(无、有)分别为292、8例,105、195例和69、6例,37、38例,两组患者比较,差异均有统计学意义(χ2=4.749,5.239,P<0.05)。(2)训练集与验证集患者随访及生存情况:训练集与验证集患者均获得随访。训练集300例患者随访时间为1.1~85.5个月,中位随访时间为50.3个月。验证集75例患者随访时间为1.0~85.7个月,中位随访时间为46.7个月。375例肝细胞癌患者术后1、3年总体生存率分别为91.7%、79.5%。训练集和验证集患者术后1、3年总体生存率分别为92.0%、79.7%和90.7%、81.9%。两组患者术后生存情况比较,差异无统计学意义(χ2=0.113,P>0.05)。(3)机器学习算法预测模型构建及验证。①筛选最优机器学习算法预测模型:根据变量对预测肝细胞癌术后3年生存的信息增益度,应用逻辑回归、支持向量机、决策树、随机森林和人工神经网络5种机器学习算法对肝细胞癌临床病理因素进行变量综合排名。筛选主要预测因素为乙型肝炎e抗原(HBeAg)、手术方式、肿瘤最大直径、围术期输血、肝被膜侵犯、肝脏Ⅳ段侵犯。将预测因素前3、6、9、12、15、18、21、24、27、29个变量依次引入5种机器学习算法。其结果显示:当引入9个变量时,逻辑回归、支持向量机、决策树、随机森林机器学习算法预测模型受试者工作特征曲线的曲线下面积(AUC)趋于稳定。当引入变量>12个时,人工神经网络机器学习算法预测模型AUC波动明显,逻辑回归、支持向量机机器学习算法预测模型AUC稳定性可继续改善,而随机森林机器学习算法预测模型AUC接近0.990,说明随机森林机器学习算法预测模型为最优机器学习算法预测模型。②随机森林机器学习算法预测模型优化和验证:将预测因素29个变量依次引入随机森林机器学习算法预测模型中,构建训练集最佳随机森林机器学习算法预测模型。其结果显示:当引入变量=10个时,网格搜索法示最佳决策树结点个数=4,最佳决策树数目=1000;当引入变量≥10个时,随机森林机器学习算法预测模型AUC稳定在0.990左右。其中当引入变量=10个时,随机森林机器学习算法预测模型预测训练集术后3年总体生存AUC为0.992,灵敏度为0.629,特异度为0.996,预测验证集术后3年总体生存AUC为0.723,灵敏度为0.177,特异度为0.948。(4)COX列线图预测模型构建及验证。①训练集患者术后生存因素分析。单因素分析结果显示:HBeAg、甲胎蛋白、围术期输血、肿瘤最大直径、肝被膜侵犯、肿瘤分化程度是影响肝细胞癌患者术后生存的相关因素(风险比=1.958,1.878,2.170,1.188,2.052,0.222,95%可信区间为1.185~3.235,1.147~3.076,1.389~3.393,1.092~1.291,1.240~3.395,0.070~0.703,P<0.05)。将P<0.2的临床病理因素纳入Lasso回归分析,其结果显示:性别,HBeAg,甲胎蛋白,手术方式,围术期输血,肿瘤最大直径,肿瘤位置在肝脏Ⅴ段和肝脏Ⅷ段,肝被膜侵犯,肿瘤分化程度(高分化、中高分化、中分化、中低分化)是影响肝细胞癌患者术后生存的相关因素。进一步将上述临床病理因素纳入多因素COX回归分析,其结果显示:HBeAg、手术方式、肿瘤最大直径是肝细胞癌患者术后生存的独立影响因素(风险比=1.770,8.799,1.142,95%可信区间为1.049~2.987,1.203~64.342,1.051~1.242,P<0.05)。②COX列线图预测模型的构建和验证:将训练集COX多因素分析结果中P≤0.1的临床病理因素引入Rstudio软件及其rms软件包,构建训练集COX列线图预测模型。COX列线图预测模型预测术后总体生存的C-index为0.723(se=0.028),预测训练集术后3年总体生存AUC为0.760,预测验证集术后3年总体生存AUC为0.795。训练集校准图验证显示COX列线图预测模型对术后生存有较好预测效果。COX列线图回归函数=0.62706×HBeAg(正常=0,异常=1)+0.13434×肿瘤最大直径(cm)+2.10758×手术方式(腹腔镜=0,开腹手术=1)+0.54558×围术期输血(无输血=0,输血=1)-1.42133×高分化(非高分化=0,高分化=1)。计算所有患者COX列线图风险评分,应用Xtile软件寻找COX列线图风险评分最佳阈值,风险评分≥2.9分为高危组,风险评分<2.9分为低危组。Kaplan-Meier总体生存曲线结果显示:训练集低危组和高危组患者术后总体生存比较,差异有统计学意义(χ2=33.065,P<0.05)。验证集低危组和高危组患者术后总体生存比较,差异有统计学意义(χ2=6.585,P<0.05)。进一步采用决策曲线分析结果显示:联合HBeAg、手术方式、围术期输血、肿瘤最大直径和肿瘤分化程度因素的COX列线图预测模型预测性能优于单一因素的预测性能。(5)随机森林机器学习算法预测模型和COX列线图预测模型预测性能评价:通过对2种模型中共同含有的重要变量(肿瘤最大直径)进行分析,并将2种模型通过预测误差曲线进行比较,观察2种模型的预测差异。其结果显示:肿瘤最大直径为2.2 cm时,随机森林机器学习算法和COX列线图预测模型预测患者术后3年生存率分别为77.17%和74.77%(χ2=0.182,P>0.05);肿瘤最大直径为6.3 cm时,随机森林机器学习算法和COX列线图预测模型预测患者术后3年生存率分别为57.51%和61.65%(χ2=0.394,P>0.05);肿瘤最大直径为14.2 cm时,随机森林机器学习算法和COX列线图预测模型预测患者术后3年生存率分别为51.03%和27.52%(χ2=12.762,P<0.05)。随着肿瘤最大直径增加,2种模型预测患者生存率差异增大。验证集中,随机森林机器学习算法预测模型预测患者术后3年总体生存AUC为0.723,COX列线图预测模型预测患者术后3年总体生存AUC为0.795,两者比较,差异有统计学意义(t=3.353,P<0.05)。采用Bootstrap交叉验证结果显示:随机森林机器学习算法预测模型和COX列线图预测模型预测3年生存的整合Brier得分分别为0.139、0.134,COX列线图预测模型预测误差低于随机森林机器学习算法预测模型。结论与机器学习算法预测模型比较,COX列线图预测模型预测肝细胞癌术后3年生存性能更佳,且其变量少,易于临床使用。     

Development and internal validation of a clinical prediction model for 90-day mortality after lung resection: the RESECT-90 score

 Open in a separate windowOBJECTIVESThe ability to accurately estimate the risk of peri-operative mortality after lung resection is important. There are concerns about the performance and validity of existing models developed for this purpose, especially when predicting mortality within 90 days of surgery. The aim of this study was therefore to develop a clinical prediction model for mortality within 90 days of undergoing lung resection.METHODSA retrospective database of patients undergoing lung resection in two UK centres between 2012 and 2018 was used to develop a multivariable logistic risk prediction model, with bootstrap sampling used for internal validation. Apparent and adjusted measures of discrimination (area under receiving operator characteristic curve) and calibration (calibration-in-the-large and calibration slope) were assessed as measures of model performance.RESULTSData were available for 6600 lung resections for model development. Predictors included in the final model were age, sex, performance status, percentage predicted diffusion capacity of the lung for carbon monoxide, anaemia, serum creatinine, pre-operative arrhythmia, right-sided resection, number of resected bronchopulmonary segments, open approach and malignant diagnosis. Good model performance was demonstrated, with adjusted area under receiving operator characteristic curve, calibration-in-the-large and calibration slope values (95% confidence intervals) of 0.741 (0.700, 0.782), 0.006 (−0.143, 0.156) and 0.870 (0.679, 1.060), respectively.CONCLUSIONSThe RESECT-90 model demonstrates good statistical performance for the prediction of 90-day mortality after lung resection. A project to facilitate large-scale external validation of the model to ensure that the model retains accuracy and is transferable to other centres in different geographical locations is currently underway.     

Prognostic factors for long-term survival in patients with thoracic metastatic disease: a 10-year experience.

OBJECTIVE: To compare survival results after resection in patients with thoracic parenchymal metastatic disease versus non-parenchymal metastatic disease and to identify prognostic factors for survival. METHODS: From 1990 to 2002, we retrospectively studied 134 procedures performed on 93 patients (3-84 years old). There were 73 patients with parenchymal resection and 20 patients with non-parenchymal resection. Tumor histology was epithelial in 62 patients, sarcoma in 21 patients, and teratomas and melanoma in 6 and 4 patients, respectively. Sixty-five patients underwent a metastasectomy once, whereas 28 had their metastatic disease repeatedly resected. RESULTS: Follow-up was 100% complete with a mean time of 43 months (range 1-169). In-hospital mortality was 2.2% (3/134 procedures) and major morbidity 5.5%. Median survival was 39 months (95% CI: 21-56 months). Overall, the actuarial survival at 1, 3, and 5 years were 84%, 55%, and 44%, respectively. For the entire group, by univariate analysis, among the 13 predictor variables selected, only the number of metastases (Hazard Ratio (HR)=3.4 [95% CI: 1.9-6.1]) and completeness of resection (HR=2.3 [95% CI: 1.3-4.2]) were found to be significant for death whereas repeated metastasectomy was found to be a significant predictor for survival (HR=0.25 [95% CI: 0.12-0.55]). In the group of parenchymal metastatic disease, a size greater than 3cm was a predictor for death (HR=2 [95% CI: 1.1-3.7]). In the subgroup of patients with colorectal metastasis, bilateral disease was also found to be a significant predictor for death (HR=3.6, [95% CI: 1.2-11.1]). CONCLUSION: This study supports our current aggressive approach to metastatic thoracic disease. Indeed, patient's survival is improved while a low mortality and morbidity is achieved. The most beneficial impact on long-term survival is correlated to the completeness of the surgery whereas the increasing number and size of the metastasis inversely correlate with survival.     

Utility of machine learning algorithms in degenerative cervical and lumbar spine disease: a systematic review

Machine learning is a rapidly evolving field that offers physicians an innovative and comprehensive mechanism to examine various aspects of patient data. Cervical and lumbar degenerative spine disorders are commonly age-related disease processes that can utilize machine learning to improve patient outcomes with careful patient selection and intervention. The aim of this study is to examine the current applications of machine learning in cervical and lumbar degenerative spine disease. A systematic review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A search of PubMed, Embase, Medline, and Cochrane was conducted through May 31st, 2020, using the following terms: “artificial intelligence” OR “machine learning” AND “neurosurgery” AND “spine.” Studies were included if original research on machine learning was utilized in patient care for degenerative spine disease, including radiographic machine learning applications. Studies focusing on robotic applications in neurosurgery, navigation, or stereotactic radiosurgery were excluded. The literature search identified 296 papers, with 35 articles meeting inclusion criteria. There were nine studies involving cervical degenerative spine disease and 26 studies on lumbar degenerative spine disease. The majority of studies for both cervical and lumbar spines utilized machine learning for the prediction of postoperative outcomes, with 5 (55.6%) and 15 (61.5%) studies, respectively. Machine learning applications focusing on degenerative lumbar spine greatly outnumber the current volume of cervical spine studies. The current research in lumbar spine also demonstrates more advanced clinical applications of radiographic, diagnostic, and predictive machine learning models.

     

Complications in the management of metastatic spinal disease

Metastatic spine disease accounts for 10% to 30% of new cancer diagnoses annually. The most frequent presentation is axial spinal pain. No treatment has been proven to increase the life expectancy of patients with spinal metastasis. The goals of therapy are pain control and functional preservation. The most important prognostic indicator for spinal metastases is the initial functional score. Treatment is multidisciplinary, and virtually all treatment is palliative. Management is guided by three key issues; neurologic compromise, spinal instability, and individual patient factors. Site-directed radiation, with or without chemotherapy is the most commonly used treatment modality for those patients presenting with spinal pain, causative by tumours which are not impinging on neural elements. Operative intervention has, until recently been advocated for establishing a tissue diagnosis, mechanical stabilization and for reduction of tumor burden but not for a curative approach. It is treatment of choice patients with diseaseadvancement despite radiotherapy and in those with known radiotherapy-resistant tumors. Vertebral resection and anterior stabilization with methacrylate or hardware (e.g., cages) has been advocated.Surgical decompression and stabilization, however, along with radiotherapy, may provide the most promising treatment. It stabilizes the metastatic deposited areaand allows ambulation with pain relief. In general, patients who are nonambulatory at diagnosis do poorly, as do patients in whom more than one vertebra is involved. Surgical intervention is indicated in patients with radiation-resistant tumors, spinal instability, spinal compression with bone or disk fragments, progressive neurologic deterioration, previous radiation exposure, and uncertain diagnosis that requires tissue diagnosis. The main goal in the management of spinal metastatic deposits is always palliative rather than curative, with the primary aim being pain relief and improved mobility. This however, does not come without complications, regardless of the surgical intervention technique used. These complication range from the general surgical complications of bleeding, infection, damage to surrounding structures and post operative DT/PE to spinal specific complications of persistent neurologic deficit and paralysis.     

Development and validation of machine learning models for the prediction of blunt cerebrovascular injury in children

BackgroundBlunt cerebrovascular injury (BCVI) is a rare finding in trauma patients. The previously validated BCVI (Denver and Memphis) prediction model in adult patients was shown to be inadequate as a screening option in injured children. We sought to improve the detection of BCVI by developing a prediction model specific to the pediatric population.MethodsThe National Trauma Databank (NTDB) was queried from 2007 to 2015. Test and training datasets of the total number of patients (885,100) with complete ICD data were used to build a random forest model predicting BCVI. All ICD features not used to define BCVI (2268) were included within the random forest model, a machine learning method. A random forest model of 1000 decision trees trying 7 variables at each node was applied to training data (50% of the dataset, 442,600 patients) and validated with test data in the remaining 50% of the dataset. In addition, Denver and Memphis model variables were re-validated and compared to our new model.ResultsA total of 885,100 pediatric patients were identified in the NTDB to have experienced blunt pediatric trauma, with 1,998 (0.2%) having a diagnosis of BCVI. Skull fractures (OR 1.004, 95% CI 1.003–1.004), extremity fractures (OR 1.001, 95% 1.0006–1.002), and vertebral injuries (OR 1.004, 95% CI 1.003–1.004) were associated with increased risk for BCVI. The BCVI prediction model identified 94.4% of BCVI patients and 76.1% of non-BCVI patients within the NTDB. This study identified ICD9/ICD10 codes with strong association to BCVI. The Denver and Memphis criteria were re-applied to NTDB data to compare validity and only correctly identified 13.4% of total BCVI patients and 99.1% of non BCVI patients.ConclusionThe prediction model developed in this study is able to better identify pediatric patients who should be screened with further imaging to identify BCVI.Level of evidenceRetrospective diagnostic study–level III evidence.     

脊柱转移性肿瘤的手术疗效及生存危险因素分析

目的 探究脊柱转移肿瘤手术对改善疼痛及神经功能的临床疗效,并对可能影响脊柱转移瘤病人生存的危险因素进行分析。方法 选取我院2010年6月至2018年12月期间接受脊柱外科手术（后路脊柱肿瘤刮除椎弓根螺钉减压内固定术、椎体成形术、射频消融术）的45例脊柱转移性肿瘤病人进行回顾性分析。术前应用Tomita评分、改良Tokuhashi评分对病人进行术前评估。术后应用疼痛视觉模拟量表（visual analogue scale, VAS）及Frankel分级对病人疼痛的改善及神经功能恢复情况进行评估。对转移性脊柱肿瘤病人的生存时间进行单因素分析,纳入的变量包括性别、BMI、手术年龄是否大于60岁、原发肿瘤的恶性程度、原发肿瘤手术治疗、肿瘤的位置、转移瘤的数目、是否存在病理性骨折、术前Frankel分级、术中出血量、术前ECOG评分。根据单因素分析的结果进一步进行Cox生存分析。结果 45例均接受了脊柱转移瘤手术并获得随访,随访时间为2~80个月,中位随访时间为9个月。与术前相比,术后的生活质量有明显的改善,疼痛VAS评分明显下降（P＜0.001）,神经功能Frankel等级明显改善（P＜0.001）。转移性脊柱肿瘤病人术后1、2年生存率分别为（54±8）%、（46±10）%,而是否存在病理性骨折（HR=2.5,P=0.043）是影响此研究预后的独立危险因素。结论 外科手术治疗可以明显改善病人的疼痛水平、生活质量与神经功能状态,而术前是否存在病理性骨折是影响转移性脊柱肿瘤病人生存预后的主要因素。     

External validation of a commercially available deep learning algorithm for fracture detection in children

PurposeThe purpose of this study was to conduct an external validation of a fracture assessment deep learning algorithm (Rayvolve®) using digital radiographs from a real-life cohort of children presenting routinely to the emergency room.Materials and methodsThis retrospective study was conducted on 2634 radiography sets (5865 images) from 2549 children (1459 boys, 1090 girls; mean age, 8.5 ± 4.5 [SD] years; age range: 0–17 years) referred by the pediatric emergency room for trauma. For each set was recorded whether one or more fractures were found, the number of fractures, and their location found by the senior radiologists and the algorithm. Using the senior radiologist diagnosis as the standard of reference, the diagnostic performance of deep learning algorithm (Rayvolve®) was calculated via three different approaches: a detection approach (presence/absence of a fracture as a binary variable), an enumeration approach (exact number of fractures detected) and a localization approach (focusing on whether the detected fractures were correctly localized). Subgroup analyses were performed according to the presence of a cast or not, age category (0–4 vs. 5–18 years) and anatomical region.ResultsRegarding detection approach, the deep learning algorithm yielded 95.7% sensitivity (95% CI: 94.0–96.9), 91.2% specificity (95% CI: 89.8–92.5) and 92.6% accuracy (95% CI: 91.5–93.6). Regarding enumeration and localization approaches, the deep learning algorithm yielded 94.1% sensitivity (95% CI: 92.1–95.6), 88.8% specificity (95% CI: 87.3–90.2) and 90.4% accuracy (95% CI: 89.2–91.5) for both approaches. Regarding age-related subgroup analyses, the deep learning algorithm yielded greater sensitivity and negative predictive value in the 5–18-years age group than in the 0–4-years age group for the detection approach (P < 0.001 and P = 0.002) and for the enumeration and localization approaches (P = 0.012 and P = 0.028). The high negative predictive value was robust, persisting in all of the subgroup analyses, except for patients with casts (P = 0.001 for the detection approach and P < 0.001 for the enumeration and localization approaches).ConclusionThe Rayvolve® deep learning algorithm is very reliable for detecting fractures in children, especially in those older than 4 years and without cast.     

Patient survival after hip arthroplasty for metastatic disease of the hip

BACKGROUND: The hip joint is a common location for metastatic disease. Actual as well as impending fractures at this site are frequently due to mechanical instability after tumor invasion and are usually treated surgically with hip arthroplasty. The objective of this study was to analyze survival and influences on survival after hip arthroplasty for metastatic hip disease. METHODS: Two hundred and ninety-nine patients who had undergone a total of 306 hemiarthroplasty or total hip arthroplasty procedures for treatment of a pathologic or an impending pathologic hip fracture between 1969 and 1996 at our institution were included in this study. Data that had been acquired prospectively within the total joint registry of our institution were reviewed retrospectively. RESULTS: The median duration of survival after the arthroplasty was 8.6 months. The duration of survival was significantly associated with the site of the fracture, location of the primary tumor, and time from the diagnosis of the primary tumor to the surgery for the fracture (p < or = 0.05). The time from the diagnosis to the arthroplasty was a significant independent predictor of survival. CONCLUSIONS: Patients undergoing hip arthroplasty for metastatic disease have a limited life expectancy, with only 40% (120) of the 299 patients in our series still alive at one year after the surgery. By identifying prognostic factors regarding life expectancy, this study provides surgeons and oncologists with information with which to weigh risks and benefits of hip arthroplasty for individual patients preoperatively.     

Liver resection for metastatic disease prolongs survival in renal cell carcinoma: 12-year results from a retrospective comparative analysis

Abstract  

The value of surgical resection of renal cell carcinoma (RCC) liver metastases still remains unclear.