原发性自发性气胸复发预测模型的构建与验证

目的 分析原发性自发性气胸(primary spontaneous pneumothorax, PSP)复发的危险因素,并建立预测模型。方法 2010年1月～2021年1月我院明确诊断为PSP的病人803例,并随机将70%的病人纳入建模组(562例),30%的病人纳入验证组(241例)。使用R 4.2.1软件进行单因素及多因素Cox回归分析,并建立Nomogram预测模型。绘制受试者工作特征曲线,计算曲线下面积(AUC)以评估模型区分度,并绘制校准曲线以评估模型校准度。结果 总体复发率为22.67%(182/803),多因素分析显示,年龄、吸烟指数、病变严重程度评分和治疗手段是PSP复发的独立危险因素,Nomogram预测模型的AUC=71.7%(95%CI 64.1～79.2),预测效能较高。结论 构建的PSP复发预测模型可协助临床医生个体化评估病人复发风险。     

基于术后短期PSA变化的前列腺癌根治术后生化复发预测模型及其验证

目的:探讨根治术后4周PSA水平对前列腺癌患者术后生化复发的预测能力,并与经典的Kattan列线图进行比较。方法:纳入PC-follow数据库中2004年10月—2018年8月期间1485例行前列腺癌根治术的患者,经筛选后313例纳入训练集用于建立模型(训练组),90例纳入验证集用于验证模型(验证组)。利用COX比例风险回归模型确定与生化复发相关的独立风险因素并纳入最终预测模型,绘制列线图便于临床应用。利用R软件计算模型的一致性指数(C-index),绘制模型预测值与实际值的校准曲线对模型进行评价。内部验证采用bootstrap方法重抽样,外部验证基于验证组数据。最后通过计算综合判别改善指数(integrated discrimination improvement, IDI)比较新模型和传统Kattan列线图的预测效能。结果:基于术后4周PSA水平和Gleason评分的新模型能较好地评估前列腺癌根治术后的生化复发风险(训练组C-index=0.819,验证组C-index=0.751),且预测值同实际发生情况具有较好的一致性。相较于传统Kattan列线图,新模型的IDI在训练组和验证组中分别为10.68%(P0.001)和12.38%(P=0.016)。结论:利用患者术后4周PSA水平和术后Gleason评分能够很好地预测患者生化复发概率,且较经典Kattan列线图更加方便准确,能够为个体化随访策略的制定提供依据。     

HBV相关肝细胞癌微血管侵犯术前预测模型的构建与应用

目的构建HBV相关肝细胞癌(HCC)微血管侵犯(MVI)的术前预测模型,并验证其预测效能。方法回顾分析2017年1月至2021年1月海南省肿瘤医院收治的812例HBV相关HCC患者资料,根据手术顺序分为模型组589例和验证组223例,单因素和多因素Logistic回归分析模型组发生MVI的独立影响因素,建立术前预测模型,采用受试者工作特征(ROC)曲线评估预测模型预测HBV相关HCC患者发生MVI的效能,并在验证组中进行独立验证。结果多因素Logistic回归分析显示年龄、中性粒细胞/淋巴细胞比值、甲胎蛋白异质体L3百分比、异常凝血酶原、肿瘤最大直径是HV相关HCC患者发生MVI的独立影响因素(OR分别为0.944、1.215、1.059、2.815、1.201),根据Logistic回归分析结果构建预测模型,ROC曲线分析结果显示,在模型组预测模型预测MVI的曲线下面积(AUC)为0.801,截断值为0.15时,灵敏度为75.2%,特异度为82.4%。在验证组预测模型预测MVI的AUC为0.824,灵敏度75.8%,特异度82.6%,预测低危MVI的AUC为0.788,灵敏度73.8%,特异度80.4%,预测高危MVI的AUC为0.858,灵敏度77.6%,特异度83.7%。结论基于Logistic回归分析构建的术前预测模型对预测HBV相关HCC患者是否发生MVI,特别是高危MVI具有较高的临床价值,可为HBV相关HCC患者的术前治疗方案,手术规划提供参考。     

基于CT影像组学胰十二指肠切除术后胰瘘风险预测模型及其价值

目的 探讨基于CT影像组学模型术前预测胰十二指肠切除术后胰瘘（POPF）的应用价值。方法 回顾性分析106例接受胰十二指肠切除术患者的临床及腹部CT资料,其中POPF（+）组36例,POPF（–）组70例。采用ITAK-SNAP软件勾画CT图像感兴趣区域（ROI）,Python程序的radiomics包进行影像组学特征提取,使用最小绝对收缩和选择算子（least absolute shrinkage and selection operator,LASSO）回归进一步筛选特征、建立影像组学评分（Rad-score）,构建影像组学预测模型。然后将临床特征、Rad-score纳入多因素Logistic回归分析,筛选出POPF发生的独立危险因素,构建临床预测模型以及联合影像学组学特征的混合模型。最后采用受试者工作特征曲线（ROC）评估不同模型的预测效能。结果 共筛选出7个非零影像组学特征并建立了Rad-score。BMI、胰管扩张及Rad-score均是发生POPF的独立危险因素。影像组学预测模型、临床特征预测模型及混合预测模型预测POPF的曲线下面积（AUC）为分别为0.72、0.69、0.80;Delong检验表明临床特征预测模型与混合预测模型间的差异具有统计学意义。结论 基于CT影像组学模型在术前辅助预测胰十二指肠切除术POPF方面具有一定的价值,联合临床指标能够提高模型的预测效能。     

胃癌病人淋巴结转移nomogram图预测模型构建及评价

目的 探讨胃癌淋巴结转移的影响因素并构建nomogram图预测模型预测淋巴结转移概率。 方法 回顾性分析2013-01-01-2020-01-01上海交通大学医学院附属新华医院收治的879例未发生远隔转移及腹膜转移且经手术治疗的胃癌病人临床资料，利用单因素logistic回归、lasso回归筛选等统计学方法筛选变量，建立nomogram图并用ROC曲线进行评价模型。结果 CA242、肿瘤病灶大小、肿瘤浸润深度以及肿瘤分化程度被纳入构建nomogram图预测模型，内部验证评价模型AUC值为0.802（95%CI 0.766~0.838），外部验证评价AUC值0.791（95%CI 0.699~0.883）。 结论 构建胃癌淋巴结转移nomogram图预测模型，纳入变量在术前易于获得，模型具有较好的预测效果和较低的泛化误差，或可为临床医生在胃癌术前分期及治疗决策选择上提供参考。     

急诊留观患者病情变化风险预测模型的构建

目的 构建急诊留观患者病情变化风险预测模型,并检验其预测效能.方法 回顾性收集急诊留观室收治的568例患者资料,采用随机数字表法抽取400例作为建模组,168例作为验证组.建模组将是否发生病情变化作为因变量构建Logistic回归模型,采用Hosmer-Lemeshow判断模型的拟合优度,根据β系数建立风险预测评分系统.验证组采用ROC曲线下面积检验评分的预测效能,对预测结局与实际临床结局采用一致性检验.结果 纳入预测模型的因子有年龄(OR=2.192)、校正改良早期预警评分(OR =3.081)、低蛋白血症(OR=26.712)、高乳酸血症(OR=13.929)、呼吸兴奋剂(OR=14.415)和抗心律失常药物(OR=4.488),模型Hosmer-Lemeshow检验P=0.220.预测评分系统ROC曲线下面积为0.985,灵敏度0.972,特异度0.919,识别患者病情变化的最佳截断值为9分,预测结局与实际临床结局的Kappa值为0.935.结论 构建的病情变化风险预测模型预测效能较好,可为急诊留观室筛查潜在危重患者提供评估工具.     

基于SEER数据库的壶腹周围癌预后预测模型的构建

目的 分析影响壶腹周围癌(periampullary carcinoma)预后的危险因素，建立壶腹周围癌预后预测模型。方法 回顾性分析美国SEER数据库中壶腹周围癌病人临床数据，根据设定标准最终纳入1 775例，按7∶3的比例分为建模组(1 242例)和验证组(533例)。在建模组中，通过Cox比例风险回归模型筛选影响壶腹周围癌预后的危险因素，在回归分析结果的基础上构建预后预测模型并绘制Nomogram图。分别在建模组和验证组中对模型的预测效能进行验证。结果 Cox回归模型多因素分析结果显示年龄、T分期、N分期、肿瘤病理分级、肿瘤病理类型、是否手术为壶腹周围癌的危险因素，将上述6个变量纳入预测模型，绘制Nomogram图，进行1年、3年、5年生存率预测。在建模组和验证组中C指数分别为0.704 7(95%CI:0.685 4,0.724 1)和0.700 1(95%CI:0.668 9,0.731 4)。建模组中1年、3年、5年的ROC曲线的AUC值分别为0.766、0.756、0.757,验证组中1年、3年、5年的ROC曲线的AUC值分别为0.736、0.733和0.742。校正曲线...     

绝经后女性中医症候群骨质疏松风险预测工具构建

目的 构建中医证候的绝经后女性的骨质疏松(osteoporosis,OP)风险预测工具。方法 问卷调查,收集研究对象一般资料、病史和中医症状,并进行骨密度检测。Logistic回归方法构建模型,并通过与常用预测工具OSTA指数比较及外部验证,分析其预测效能。结果 年龄、体重、驼背和精神不振是OP危险因素,结合OP主要中医证候——肾虚证典型症状腰膝酸软、下肢抽筋、夜尿频多和耳鸣构建COPT模型。其ROC曲线下面积AUC=0.765,高于OSTA指数AUC;最优判断界值为>-0.79,灵敏度为69.43,特异性为72.60。外部数据验证灵敏度为68.85,特异性为75.00。结论 中医证候表现在OP风险预测中具有一定作用,在本研究中,COPT模型预测效能优于OSTA指数,外部数据验证结果进一步支持COPT具有良好预测效能。     

基于LASSO-Cox回归分析的非轻症急性胰腺炎死亡风险列线图预测模型的建立和临床应用效果分析

目的 构建非轻症急性胰腺炎（NMAP）患者死亡风险的列线图预测模型,并验证其预测效能和临床应用价值,同时分析其对于其他评分系统的优势。方法 纳入大型重症监护数据库MIMIC-Ⅲ中的606例NMAP患者临床资料,按7 3比例随机分为训练集和验证集。采用LASSO-Cox回归分析构建NMAP患者死亡风险列线图预测模型,并通过受试者工作特征（ROC）曲线、校准曲线以及决策曲线分析（DCA）对列线图模型进行评估。然后比较列线图模型与急性胰腺炎严重程度床边指数（BISAP）、序贯器官衰竭评分（SOFA）、快速序贯器官功能衰竭评分（q SOFA）、急性生理评分Ⅲ（APSⅢ）及牛津急性疾病严重程度评分（OASIS）对NMAP患者死亡风险的预测效能。结果 LASSO-Cox回归分析结果表明,年龄以及入院24 h内的收缩压、红细胞分布宽度（RDW）、血清白蛋白、尿素氮（BUN）、总胆红素和国际标准化比值（INR）是与NMAP患者死亡风险相关的独立危险因素（P<0.05）,以此建立的列线图预后模型预测训练集NMAP患者14、30、60、90 d内死亡风险的ROC曲线下面积（AUC）分别为0.76(9...     

FRS与a-FRS对不同胰腺手术术后胰瘘的预测价值分析

背景与目的:胰十二指肠切除术(PD)和胰体尾切除术(DP)是治疗胰腺肿瘤常见的手术方式。术后胰瘘(POPF)是胰腺术后最常见、最严重的并发症之一,若能准确预测POPF的发生将有重要的临床意义。胰瘘危险评分(FRS)和新的胰瘘危险评分(a-FRS)是运用最广的两个POPF预测模型,但这两个预测模型能否有效预测POPF仍需进一步验证。本研究比较FRS和a-FRS对PD与DP的POPF预测价值,以期为临床医师选择合适的预测模型提供理论依据和参考。方法:回顾性收集单中心2018—2019年间行胰腺手术的所有患者的临床资料,经排除标准筛选的入选病例作为研究对象。统计整体与不同手术类型(PD、DP)的POPF发生情况,用受试者工作特征曲线下面积(AUC)分析两种评分模型对整体及不同手术类型的POPF的预测效能。结果:排除不符合的病例后共纳入339例患者,其中193例行PD,146例行DP。全组POPF发生率为17.4%,PD组为18.1%,DP组为16.4%。FRS和a-FRS在全组中预测POPF的能力相似(AUC:0.67 vs.0.65,P=0.412),FRS在PD组中POPF预测价值优于a-FRS(AUC:0.74 vs.0.67,P=0.006),但对DP组的POPF无预测价值(AUC=0.57,95% CI=0.44～0.70,P=0.285),而a-FRS预测DP组POPF的能力好于FRS(AUC:0.66 vs.0.57,P=0.048)。此外,按FRS与a-FRS任何一个模型,POPF的发生率在全组、PD组、DP组的发生率均随着风险等级的上升而增加。FRS的预测因子中,术中失血量和主胰管直径在DP组和PD组间存在明显差异(均P0.05)。结论:FRS和a-FRS均可用于POPF的预测。FRS对PD的POPF预测价值大于a-FRS,但不适用于DP的POPF的预测,而a-FRS对DP的POPF的预测有帮助。术中失血量和主胰管直径是导致FRS对DP的POPF预测效能较低的原因。但由于研究的局限性,结论仍需进一步验证。     

Predictive model for early functional outcomes following acute care after traumatic brain injuries: A machine learning-based development and validation study

IntroductionFew studies on early functional outcomes following acute care after traumatic brain injury (TBI) are available. The aim of this study was to develop and validate a predictive model for functional outcomes at discharge for TBI patients using machine learning methods.Patients and methodsIn this retrospective study, data from 5281 TBI patients admitted for acute care who were identified in the Beijing hospital discharge abstract database were analysed. Data from 4181 patients in 52 tertiary hospitals were used for model derivation and internal validation. Data from 1100 patients in 21 secondary hospitals were used for external validation. A poor outcome was defined as a Barthel Index (BI) score ≤ 60 at discharge. Logistic regression, XGBoost, random forest, decision tree, and back propagation neural network models were used to fit classification models. Performance was evaluated by the area under the receiver operating characteristic curve (AUC), the area under the precision-recall curve (AP), calibration plots, sensitivity/recall, specificity, positive predictive value (PPV)/precision, negative predictive value (NPV) and F1-score.ResultsCompared to the other models, the random forest model demonstrated superior performance in internal validation (AUC of 0.856, AP of 0.786, and F1-score of 0.724) and external validation (AUC of 0.779, AP of 0.630, and F1-score of 0.604). The sensitivity/recall, specificity, PPV/precision, and NPV of the model were 71.8%, 69.2%, 52.2%, and 84.0%, respectively, in external validation. The BI score at admission, age, use of nonsurgical treatment, neurosurgery status, and modified Charlson Comorbidity Index were identified as the top 5 predictors for functional outcome at discharge.ConclusionsWe established a random forest model that performed well in predicting early functional outcomes following acute care after TBI. The model has utility for informing decision-making regarding patient management and discharge planning and for facilitating health care quality assessment and resource allocation for TBI treatment.     

The comparability of models for predicting the risk of a positive prostate biopsy with prostate-specific antigen alone: a systematic review

CONTEXT: The sensitivity and specificity profile of measuring levels of prostate-specific antigen (PSA) to select men for prostate biopsy is not optimal. This has prompted the construction of nomograms and artificial neural networks (ANNs) to increase the performance of PSA measurements. OBJECTIVE: A systematic review of nomograms and ANNs designed to predict the risk of a positive prostate biopsy for cancer was conducted in order to determine their value versus measuring PSA levels alone. EVIDENCE ACQUISITION: Medical Literature Analysis and Retrieval System Online (U.S. National Library of Medicine's life science database; MEDLINE) was searched using the terms "nomogram" "artificial neural network" and "prostate cancer" for dates up to and including July 2007 and was supplemented by manual searches of reference lists. Included studies used an assessment tool to examine the risk of a positive prostate biopsy in a man without a known cancer diagnosis. Intramodel comparisons with evaluation of PSA levels alone, and intermodel comparisons of area under the curve (AUC) from receiver operating characteristic (ROC) curves were conducted. Individual case examples were also used for comparisons. EVIDENCE SYNTHESIS: Twenty-three studies examining 36 models were included. With the exception of two studies, all the models had AUC values of 0.70 or greater, with eight reporting an AUC of >/=0.80 and four (all ANNs) reporting an AUC >/=0.85, with variable validation status. Fourteen studies compared the AUC with PSA levels alone: all showed a benefit from using AUCs which varied from 0.02 to 0.26. Of the 16 external validation comparisons, in 13 the AUC was lower in the external population than in the model population. CONCLUSIONS: Nomograms and ANNs produce improvements in AUC over measurement of PSA levels alone, but many lack external validation. Where this is available, the benefits are often diminished, although most remain significantly better than with evaluation of PSA levels alone. In men without additional risk factors, PSA cutoff values alone provide a relatively precise risk estimate, but if additional risk factors are known, PSA values alone are less accurate.     

Comparison of accuracy between the Partin tables of 1997 and 2001 to predict final pathological stage in clinically localized prostate cancer

PURPOSE: We validated externally the predictive accuracy of the 2001 Partin tables and compared the 1997 and 2001 versions. MATERIALS AND METHODS: We used ROC derived AUC to test the predictive accuracy of organ confinement (OC), extraprostatic extension (ECE), seminal vesicle invasion (SVI) and lymph node involvement (LNI) of 1997 and 2001 Partin tables derived probabilities. These probabilities were defined by the pretreatment clinical stage, serum prostate specific antigen and biopsy Gleason grade of 2,139 patients treated with radical prostatectomy for clinically localized prostate cancer. RESULTS: OC, ECE, SVI and LNI were noted in 63.5%, 23.1%, 10.5% and 2.9% of cases, respectively. AUC of the 2001 tables was 0.787, 0.766, 0.775 and 0.790, for OC, ECE, SVI and LNI, respectively. These values were virtually the same as the respective 1997 Partin table AUC values, namely 0.784, 0.728, 0.791 and 0.799. CONCLUSIONS: This external validation of the 2001 Partin tables confirms good predictive accuracy of the updated tables. However, predictive accuracy in this external validation data set of 2,139 European men is virtually the same as that of the original 1997 tables. Therefore, a transition from the 1997 tables to the updated 2001 version does not appear warranted unless superior accuracy is demonstrated in other external cohorts.     

A Predictive Tool to Estimate the Risk of Axillary Metastases in Breast Cancer Patients with Negative Axillary Ultrasound

Background

Sentinel node biopsy (SNB) is the “gold standard” in axillary staging in clinically node-negative breast cancer patients. However, axillary treatment is undergoing a paradigm shift and studies are being conducted on whether SNB may be omitted in low-risk patients. The purpose of this study was to evaluate the risk factors for axillary metastases in breast cancer patients with negative preoperative axillary ultrasound.

Methods

A total of 1,395 consecutive patients with invasive breast cancer and SNB formed the original patient series. A univariate analysis was conducted to assess risk factors for axillary metastases. Binary logistic regression analysis was conducted to form a predictive model based on the risk factors. The predictive model was first validated internally in a patient series of 566 further patients and then externally in a patient series of 2,463 patients from four other centers. All statistical tests were two-sided.

Results

A total of 426 of the 1,395 (30.5 %) patients in the original patient series had axillary lymph node metastases. Histological size (P < 0.001), multifocality (P < 0.001), lymphovascular invasion (P < 0.001), and palpability of the primary tumor (P < 0.001) were included in the predictive model. Internal validation of the model produced an area under the receiver operating characteristics curve (AUC) of 0.731 and external validation an AUC of 0.79.

Conclusions

We present a predictive model to assess the patient-specific probability of axillary lymph node metastases in patients with clinically node-negative breast cancer. The model performs well in internal and external validation. The model needs to be validated in each center before application to clinical use.     

基于临床病理大数据的早发乳腺癌腋窝淋巴结转移风险预测模型构建

背景与目的：近年来,乳腺癌的发病人群趋于年轻化,并且更容易发生腋窝淋巴结（ALN）转移。本研究通过临床病理大数据平台分析年轻乳腺癌患者ALN转移的影响因素,并建立风险预测模型,为年轻乳腺癌患者的诊断和治疗提供参考依据。 方法：收集SEER数据库中2010—2015年间被诊断为乳腺癌并且接受了ALN手术的年轻患者的临床病理资料,采用单因素和多因素回归分析筛选ALN转移的影响因素,并以列线图的方式可视化。通过AUC/C指数量化列线图区分不同ALN状态患者的能力,采用bootstrap方法（1 000次重复,随机数种子设置为12）进行列线图预测性能内部验证。另外,收集2015—2017年在武汉大学中南医院初诊为乳腺癌的年轻患者资料,对模型行外部验证。 结果：共纳入SEER数据库中23 778例年轻乳腺癌患者,其中39.6%患者存在ALN转移。单因素Logistic回归分析显示,年龄、种族、肿瘤部位、病理学分级、肿瘤大小、胸壁或皮肤是否受侵以及ER、PR、HER-2状态与ALN转移有关（均P<0.001）;多因素Logistic回归分析显示：年龄、种族、婚姻状态、边侧、肿瘤部位、分级、肿瘤大小、胸壁或皮肤是否受侵以及ER与PR状态是ALN转移独立影响因素（均P<0.05）,据此建立风险预测模型。内部验证的校准曲线显示,利用该模型计算的预测值与真实值之间存在良好的一致性（AUC/C指数=0.716）。共纳入391例年轻乳腺癌患者作为外部验证数据集,其中49.9%患者初次手术发现有ALN转移。外部验证提示模型预测能力较好（AUC/C指数=0.798）。 结论：基于SEER数据库建立的年轻乳腺癌患者ALN转移的风险预测模型具有较好的预测能力,可为临床预测患者ALN转移风险提供参考。     

External validation of the RENAL nephrometry score nomogram for predicting high-grade renal cell carcinoma in solid, enhancing, and small renal masses

Purpose

To confirm predictive accuracies of the RENAL nephrometry score (RNS) nomogram for identifying malignancy and high-grade renal cell carcinoma (RCC) in an external cohort of small renal masses (SRMs).

Methods

A total of 1,129 patients who underwent extirpative renal surgery for solid and enhancing cT1 renal tumors between 2005 and 2012 at a single institution were included in the validation cohort. A single uro-radiologist utilized computed tomography image reconstruction to classify tumors according to the RNS. The area under the curve (AUC) and calibration plots were used to determine predictive accuracies of malignancy and high-grade models of the RNS nomogram.

Results

Malignant and high-grade tumors were identified in 1,012 (89.6 %) and 389 (38.4 %) patients with cT1 tumors, and in 658 (87.3 %) and 215 (32.6 %) patients with cT1a tumors, respectively. Predictive performances of the nomogram for malignancy and high-grade models revealed AUCs of 0.722 and 0.574 for cT1 tumors, and 0.727 and 0.495 for cT1a tumors, respectively. The predictive value of the malignancy model was comparable to that of the model-development cohort (AUC = 0.76); however, the predictive value of the high-grade model was inferior to that of the model-development cohort (AUC = 0.73).

Conclusions

Unlike previous validation studies, we report inferior predictive performance of the RNS nomogram for discriminating high-grade RCC in solid and enhancing SRMs. This suggests that the RNS nomogram may be unreliable for preoperatively predicting high-grade RCC in SRMs, in which tumor size, the key determinant of high-grade RCC, is a limiting factor.     

Prediction of non-sentinel lymph node involvement in breast cancer patients with a positive sentinel lymph node

Completion axillary lymph node dissection (cALND) is the golden standard if breast cancer involves the sentinel lymph node (SLN). However, most non-sentinel lymph nodes (NSLN) are not involved, cALND has a considerable complication rate and does not improve outcome. We here present and validate our predictive model for positive NSLNs in the cALND if the SLN is positive.Consecutive early breast cancer patients from one center undergoing cALND for a positive SLN were included. We assessed demographic and clinicopathological variables for NSLN involvement. Uni- and multivariate analysis was performed. A predictive model was built and validated in two external centers.21.9% of 470 patients had at least one involved NSLN. In univariate analysis, seven variables were significantly correlated with NSLN involvement: tumor size, grade, lymphovascular invasion (LVI), number of positive and negative SLNs, size of SLN metastasis and intraoperative positive SLN. In multivariate analysis, LVI, number of negative SLNs, size of SLN metastasis and intraoperative positive pathological evaluation were independent predictors for NSLN involvement. The calculated risk resulted in an AUC of 0.76. Applied to the external data, the model was accurate and discriminating for one (AUC = 0.75) and less for the other center (AUC = 0.58).A discriminative predictive model was constructed to calculate the risk of NSLN involvement in case of a positive SLN. External validation of our model reveals differences in performance when applied to data from other institutions concluding that such a predictive model requires validation prior to use.     

Outcome prediction in coronary artery bypass grafting and valve surgery in the Netherlands: development of the Amphiascore and its comparison with the Euroscore.

OBJECTIVES: (1) To define models that predict in-hospital death, major adverse cardiac events and extended intensive care unit duration for patients who underwent coronary artery bypass grafting (CABG), a heart valve operation or combined; and (2) to validate the Euroscore model in our population. METHODS: Data of all 7282 patient who underwent a CABG and/or heart valve operation in 1997-2001 were prospectively collected. Three outcomes were examined: in-hospital death, major adverse cardiac events (MACE) and extended length of stay on intensive care (ELOS). Predicting models were made by multivariate logistic regression. The patient population was randomly divided in a derivation (two thirds) and a validation (one third) set. Area under the receiver operating characteristics curve (AUC) was used to study the discriminatory abilities of these models and the Euroscore. Hosmer-Lemeshow goodness-of-fit was used to study calibration of the predictive models. RESULTS: 2.4% of the patients died in-hospital, 17% of the patients had a MACE and 14% had ELOS. The models for in-hospital mortality and ELOS had a good validation (AUC 0.84 and 0.79, respectively). The validation for MACE was moderate (receiver-operating characteristic, ROC 0.67). All models were well calibrated. The validation of the Euroscore was as good as our model for in-hospital mortality (ROC 0.84). CONCLUSIONS: The Amphia score performs as well as the Euroscore in discriminating patients with respect to in-hospital death. Our models for predicting major adverse cardiac events and extended length of stay on intensive care may be useful tools in categorising patients in various subgroups of risk for postoperative morbidity.     

Current status of simulation-based training in pediatric surgery: A systematic review

BackgroundSimulation based training enables pediatric surgical trainees to attain proficiency in surgical skills. This study aims to identify the currently available simulators for pediatric surgery, assess their validation and strength of evidence supporting each model.MethodsBoth Medline and EMBASE were searched for English language articles either describing or validating simulation models for pediatric surgery. A level of evidence (LoE) followed by a level of recommendation (LoR) was assigned to each validation study and simulator, based on a modified Oxford Centre for Evidence-Based Medicine classification for educational studies.ResultsForty-nine articles were identified describing 44 training models and courses. Of these articles, 44 were validation studies. Face validity was evaluated by 20 studies, 28 for content, 24 demonstrated construct validity and 1 showed predictive validity. Of the validated models, 3 were given an LoR of 2, 21 an LoR of 3 and 12 an LoR of 4. None reached the highest LoR.ConclusionsThere are a growing number of simulators specific to pediatric surgery. However, these simulators have limited LoE and LoR in current studies. The lack of NoTSS training is also apparent. We advocate more randomized trials to validate these models, and attempts to determine predictive validity.Type of studyOriginal / systematic review.Level of evidence1.     

Current status of simulation and training models in microsurgery: A systematic review

With the prolific uptake of simulation‐based training courses, this systematic review aims to identify the available microsurgical simulation and training models, their status of validation, associated studies, and levels of evidence (LoE) for each training model, thereby establishing a level of recommendation (LoR). MEDLINE, Embase, and the Cochrane Library databases were searched for English language articles, describing microsurgery simulators and/or validation studies. All studies were assessed for LoE, and each model was subsequently awarded a LoR using a modified Oxford Centre for Evidence‐Based Medicine classification, adapted for education, with 1 being the highest and 4 the lowest score. A total of 86 studies were identified describing 64 models and simulators ranging from bench models, cadaveric animal tissue, cadaveric human tissue, live animal models, virtual reality simulators, and training curricula. Of these, 49 simulators had at least one validation study. Models were assessed for face (n = 42), content (n = 31), construct (n = 25), transfer (n = 10), and concurrent validity (n = 1) by these studies. The most commonly identified modality was bench models (n = 28) followed by cadaveric animal tissue (n = 24). The cryopreserved rat aorta model received the highest LoR followed by chicken wing, chicken thigh and practice cardboard models. Microsurgery simulation is a growing field and increasing numbers of models are being produced. However, there are still only a few validation studies with a high LoE. It is therefore imperative that training models and/or programs are evaluated for validity and efficacy in order to allow utilization in microsurgical skills acquisition.