Development and validation of a clinical prediction model for prognostic factors in patients with primary pontine hemorrhage

Abstract We aimed to develop a prognostic model for primary pontine hemorrhage (PPH) patients and validate the predictive value of the model for a good prognosis at 90 days. A total of 254 PPH patients were included for screening of the independent predictors of prognosis, and data were analyzed by univariate and multivariable logistic regression tests. The cases were then divided into training cohort (n=219) and validation cohort (n=35) based on the two centers. A nomogram was developed using independent predictors from the training cohort to predict the 90-day good outcome and was validated from the validation cohort. Glasgow Coma Scale score, normalized pixels (used to describe bleeding volume), and mechanical ventilation were significant predictors of a good outcome of PPH at 90 days in the training cohort (all P<0.05). The U test showed no statistical difference (P=0.892) between the training cohort and the validation cohort, suggesting the model fitted well. The new model showed good discrimination (area under the curve=0.833). The decision curve analysis of the nomogram of the training cohort indicated a great net benefit. The PPH nomogram comprising the Glasgow Coma Scale score, normalized pixels, and mechanical ventilation may facilitate predicting a 90-day good outcome.

Application progress of predictive models in the prevention and control of myopia in children and adolescents / 国际眼科杂志(Guoji Yanke Zazhi)

In medical research,predictive models have been widely used to predict disease progression and identify high-risk populations in advance, especially in the prevention and diagnosis of chronic diseases. In ophthalmology, the predictive and diagnostic models for fundus diseases such as age-related macular degeneration and diabetic retinopathy have demonstrated expert-level accuracy. However, the application of predictive models is still in the exploratory stage as for myopia prevention and control. The establishment of a predictive model is helpful to identify the high-risk myopic children in advance, so that preventive measures such as adequate outdoor activities and reducing near work can be taken in time, which is of great significance to prevent or slow down the occurrence and development of myopia. Because the mechanism of myopia has not been fully elucidated, there are still challenges and limitations in the selection of application objects, predictors and predictive outcomes. This paper reviews the research progress of different types of myopia predictive models in order to provide reference for further development and improvement.

Risk prediction models of dangerous behaviors among patients with severe mental disorder in community / 四川精神卫生

BackgroundThe occurrence rate of dangerous behaviors in patients with severe mental disorders is higher than that of the general population. In China， there is limited research on the prediction of dangerous behaviors in community-dwelling patients with severe mental disorders， particularly in terms of predicting models using data mining techniques other than traditional methods. ObjectiveTo explore the influencing factors of dangerous behaviors in community-dwelling patients with severe mental disorders and testing whether the classification decision tree model is superior to the Logistic regression model. MethodsA total of 11 484 community-dwelling patients with severe mental disorders who had complete follow-up records from 2013 to 2022 were selected on December 2023. The data were divided into a training set （n=9 186） and a testing set （n=2 298） in an 8∶2 ratio. Logistic regression and classification decision trees were separately used to establish predictive models in the training set. Model discrimination and calibration were evaluated in the testing set. ResultsDuring the follow-up period， 1 115 cases （9.71%） exhibited dangerous behaviors. Logistic regression results showed that urban residence， poverty， guardianship， intellectual disability， history of dangerous behaviors， impaired insight and positive symptoms were risk factors for dangerous behaviors （OR=1.778， 1.459， 2.719， 1.483， 3.890， 1.423， 2.528， 2.124， P<0.01）. Being aged ≥60 years， educated， not requiring prescribed medication and having normal social functioning were protective factors for dangerous behaviors （OR=0.594， 0.824， 0.422， 0.719， P<0.05 or 0.01）. The predictive effect in the testing set showed an area under curve （AUC） of 0.729 （95% CI： 0.692~0.766）， accuracy of 70.97%， sensitivity of 59.71%， and specificity of 72.05%. The classification decision tree results showed that past dangerous situations， positive symptoms， overall social functioning score， economic status， insight， household registration， disability status and age were the influencing factors for dangerous behaviors. The predictive effect in the testing set showed an AUC of 0.721 （95% CI： 0.705~0.737）， accuracy of 68.28%， sensitivity of 64.46%， and specificity of 68.60%. ConclusionThe classification decision tree does not have a greater advantage over the logistic regression model in predicting the risk of dangerous behaviors in patients with severe mental disorders in the community. ［Funded by Chengdu Medical Research Project （number， 2020052）］

Modelo predictivo del fracaso de la terapia eléctrica en la fibrilación auricular paroxística

Introducción: La fibrilación auricular es la arritmia recurrente más habitual en la práctica clínica. Su prevalencia se multiplica en la población actual y tiene diferentes causas fisiopatológicas que la convierten en una pandemia mundial. Objetivos: Diseñar un modelo predictivo de fracaso de la terapia eléctrica en pacientes con fibrilación auricular paroxística. Métodos: Se realizó un estudio de casos y controles, con 33 casos y 66 controles. Variables predictoras: edad, fracción de eyección ≤ 40 %, volumen de aurícula izquierda ≥ 34 mL/m2. A partir de la regresión logística se obtuvo un modelo en el que fueron incluidos el valor predictivo positivo, valor predictivo negativo, la sensibilidad y especificidad. Resultados: Los factores de riesgo predictores fueron: edad ≥ 55 años (p= 0,013; odds ratio (OR)= 3,58; intervalo de confianza -IC- 95 %: 1,33-9,67); la fracción de eyección del ventrículo izquierdo (FEVI) ≤ 40 % se observó en 20 pacientes (22,7 %) (p= 0,004; OR= 4,45; IC95 %: 1,54-12,8); presión de aurícula izquierda elevada, volumen de aurícula izquierda elevado (p= 0,004; OR= 3,11; IC95 %: 1,24-8,77), según el modelo de regresión logística. Se realizó la validación interna por división de datos; se confirmó que el modelo pronostica bien los que van a tener éxito en el resultado terapéutico. Conclusiones: El modelo predictivo elaborado está compuesto por los predictores edad > 55 años, FEVI; volumen de aurícula izquierda; presenta un buen ajuste y poder discriminante, sobre todo valor predictivo positivo.

Introduction: Atrial fibrillation is the most common recurrent arrhythmia in clinical practice. Its prevalence is multiplying in the current population and has different pathophysiological causes that make it a global pandemic. Objectives: To design a predictive model for failure of electrical therapy in patients with paroxysmal atrial fibrillation. Methods: A case-control study was carried out with 33 cases, and 66 controls. Predictor variables: age, ejection fraction ≤ 40%, left atrial volume ≥ 34 mL/m2. From logistic regression, a model was obtained in which the positive predictive value, negative predictive value, sensitivity and specificity were included. Results: The predictive risk factors were: age ≥ 55 years (p= 0.013; odds ratio (OR)= 3.58; 95% confidence interval -CI-: 1.33-9.67); left ventricular ejection fraction (LVEF) ≤ 40% was observed in 20 patients (22.7%) (p= 0.004; OR= 4.45; 95% CI: 1.54-12.8); elevated left atrial pressure, elevated left atrial volume (p= 0.004; OR= 3.11; 95% CI: 1.24-8.77), according to the logistic regression model. Internal validation was carried out by data division; It was confirmed that the model predicts very well those who will be successful in the therapeutic result. Conclusions: The predictive model developed is composed of the predictors age > 55 years, LVEF; left atrial volume; It presents a good fit and discriminating power, especially positive predictive value.

Prognostic predictive model for PONV in cesarean delivery

Introduction: Apfel simplified risk score for postoperative nausea and vomiting (PONV) has shown to be useful in anesthesia; however, since it has not been calibrated in regional anesthesia or in pregnant patients, its use in cesarean section is limited. Objective: To develop a prognostic predictive model for postoperative nausea and vomiting in pregnant patients undergoing cesarean section under spinal anesthesia. Methods: In a cohort of 703 term pregnant patients scheduled of cesarean section, 15 variables were prospectively assessed, to design a prognostic predictive model for the development of postoperative nausea and vomiting. A logistic regression analysis was used to construct the model and its calibration and discrimination were based on the Hosmer-Lemeshow test, the calibration curves, and C statistic. Additionally, the internal calibration was performed with the Bootstrap resampling method. Results: Postoperative nausea and vomiting were experienced by 27% of the patients during the first six hours after surgery. The model included as prognostic variables the development of intraoperative nausea and vomiting, age under 28 years, a history of PONV, the mother's BMI and the weight of the newborn baby. The model showed an adequate calibration (x2: 4.65 p: 0.5888), though a low discrimination (Statistic C = 0.68). Conclusions: A prognostic predictive model was created for the development of PONV in cesarean section. This model was used to build a prognostic scale for the classification of patients into risk groups.

Introducción: La escala de riesgo simplificada de Apfel para náuseas y vómito posoperatorio (NVPO) ha mostrado utilidad en anestesia; sin embargo, al no haber sido calibrada en anestesia regional o en pacientes embarazadas, su utilidad en cesárea es limitado. Objetivo: Desarrollar un modelo de predicción pronóstica para náuseas y vómito posoperatorios en pacientes embarazadas, llevadas a cesárea bajo anestesia espinal. Métodos: En una cohorte de 703 pacientes con embarazo a término programadas para cesárea, se evaluaron 15 variables de forma prospectiva para construir un modelo de predicción pronóstica para el desarrollo de náuseas y vómito posoperatorio. Se utilizó el análisis de regresión logística para la construcción del modelo y se calculó su calibración y discriminación con la prueba de Hosmer-Lemeshow, las curvas de calibración y el estadístico C. Además, se realizó la calibración interna con el método de remuestreo Bootstrap. Resultados: Las náuseas y vómito posoperatorio se presentaron en el 27% de las pacientes durante las primeras seis horas después de la cirugía. El modelo incluyó como variables pro-nósticas el desarrollo de náuseas y vómito en el intraoperatorio, edad menor de 28 años, antecedentes de NVPO, índice de masa corporal (IMC) de la madre y el peso del recién nacido. El modelo mostró una adecuada calibración (x2: 4,65 p: 0,5888), aunque una baja discriminación (Estadístico C = 0,68). Conclusiones: Se construyó un modelo de predicción pronóstica para el desarrollo de NVPO en cirugía cesárea, y con este se construyó una escala pronóstica que permite clasificar a las pacientes por grupos de riesgo.

Modelo predictivo de enfermedad cardiovascular basado en inteligencia artificial en la atención primaria de salud / A Predictive Model for Cardiovascular Disease based on Artificial Intelligence in Primary Health Care

Introducción: En Cuba y en el resto del mundo, las enfermedades cardiovasculares son reconocidas como un problema de salud pública mayúsculo y creciente, que provoca una alta mortalidad. Objetivo: Diseñar un modelo predictivo para estimar el riesgo de enfermedad cardiovascular basado en técnicas de inteligencia artificial. Métodos: La fuente de datos fue una cohorte prospectiva que incluyó 1633 pacientes, seguidos durante 10 años, fue utilizada la herramienta de minería de datos Weka, se emplearon técnicas de selección de atributos para obtener un subconjunto más reducido de variables significativas, para generar los modelos fueron aplicados: el algoritmo de reglas JRip y el meta algoritmo Attribute Selected Classifier, usando como clasificadores el J48 y el Multilayer Perceptron. Se compararon los modelos obtenidos y se aplicaron las métricas más usadas para clases desbalanceadas. Resultados: El atributo más significativo fue el antecedente de hipertensión arterial, seguido por el colesterol de lipoproteínas de alta densidad y de baja densidad, la proteína c reactiva de alta sensibilidad y la tensión arterial sistólica, de estos atributos se derivaron todas las reglas de predicción, los algoritmos fueron efectivos para generar el modelo, el mejor desempeño fue con el Multilayer Perceptron, con una tasa de verdaderos positivos del 95,2 por ciento un área bajo la curva ROC de 0,987 en la validación cruzada. Conclusiones: Fue diseñado un modelo predictivo mediante técnicas de inteligencia artificial, lo que constituye un valioso recurso orientado a la prevención de las enfermedades cardiovasculares en la atención primaria de salud(AU)

Introduction: In Cuba and in the rest of the world, cardiovascular diseases are recognized as a major and growing public health problem, which causes high mortality. Objective: To design a predictive model to estimate the risk of cardiovascular disease based on artificial intelligence techniques. Methods: The data source was a prospective cohort including 1633 patients, followed for 10 years. The data mining tool Weka was used and attribute selection techniques were employed to obtain a smaller subset of significant variables. To generate the models, the rule algorithm JRip and the meta-algorithm Attribute Selected Classifier were applied, using J48 and Multilayer Perceptron as classifiers. The obtained models were compared and the most used metrics for unbalanced classes were applied. Results: The most significant attribute was history of arterial hypertension, followed by high and low density lipoprotein cholesterol, high sensitivity c-reactive protein and systolic blood pressure; all the prediction rules were derived from these attributes. The algorithms were effective to generate the model. The best performance was obtained using the Multilayer Perceptron, with a true positive rate of 95.2percent and an area under the ROC curve of 0.987 in the cross validation. Conclusions: A predictive model was designed using artificial intelligence techniques; it is a valuable resource oriented to the prevention of cardiovascular diseases in primary health care(AU)

Aprendizaje automático aplicado a la predicción de diabetes mellitus, utilizando información socioeconómica y ambiental de usuarios del sistema de salud

Objetivo: Se propuso aplicar modelos basados en técnicas de aprendizaje automático como apoyo para el diagnóstico temprano de la diabetes mellitus, utilizando variables de datos ambientales, sociales, económicos y sanitarios, sin la dependencia de la toma de muestras clínicas. Metodología: Se utilizaron datos de 10 889 usuarios afiliados al régimen subsidiado de salud de la zona suroccidental en Colombia, diagnosticados con hipertensión y agrupados en usuarios sin (74,3 %) y con (25,7 %) diabetes mellitus. Se entrenaron modelos supervisados utilizando k vecinos más cercanos, árboles de decisión y bosques aleatorios, así como modelos basados en ensambles, aplicados a la base de datos antes y después de balancear el número de casos en cada grupo de diagnóstico. Se evalúo el rendimiento de los algoritmos mediante la división de la base de datos en datos de entreno y de prueba (70/30, respectivamente), y se utilizaron métricas de exactitud, sensibilidad, especificidad y área bajo la curva. Resultados: Los valores de sensibilidad aumentaron considerablemente al utilizar datos balanceados, pasando de valores máximos del 17,1 % (datos sin balancear) a valores de hasta 57,4 % (datos balanceados). El valor más alto de área bajo la curva (0,61) fue obtenido con los modelos de ensambles, al aplicar un balance en el número de datos por cada grupo y al codificar las variables categóricas. Las variables de mayor peso estuvieron asociadas con aspectos hereditarios (24,65 %) y con el grupo étnico (5.59 %), además de la dificultad visual, el bajo consumo de agua, una dieta baja en frutas y verduras, y el consumo de sal y azúcar. Conclusiones: Aunque los modelos predictivos, utilizando información socioeconómica y ambiental de las personas, surgen como una herramienta para el diagnóstico temprano de la diabetes mellitus, estos aún deben ser mejorados en su capacidad predictiva.

Objective: The objective was to apply models based on machine learning techniques to support the early diagnosis of diabetes mellitus, using environmental, social, economic and health data variables, without dependence on clinical sample collection. Methodology: Data from 10,889 users affiliated with the subsidized health system in the southwestern area of Colombia, diagnosed with hypertension and grouped into users without (74.3%) and with (25.7%) diabetes mellitus, were used. Supervised models were trained using k-nearest neighbors, decision trees, and random forests, as well as ensemble-based models, applied to the database before and after balancing the number of cases in each diagnostic group. The performance of the algorithms was evaluated by dividing the database into training and test data (70/30, respectively), and metrics of accuracy, sensitivity, specificity, and area under the curve were used. Results: Sensitivity values increased significantly when using balanced data, going from maximum values of 17.1% (unbalanced data) to values as high as 57.4% (balanced data). The highest value of area under the curve (0.61) was obtained with the ensemble models, by applying a balance in the amount of data for each group and by coding the categorical variables. The variables with the greatest weight were associated with hereditary aspects (24.65%) and with the ethnic group (5.59%), in addition to visual difficulty, low water consumption, a diet low in fruits and vegetables, and the consumption of salt and sugar. Conclusions: Although predictive models, using people's socioeconomic and environmental information, emerge as a tool for the early diagnosis of diabetes mellitus, their predictive capacity still needs to be improved.

Objetivo: Propôs-se aplicar modelos baseados em técnicas de aprendizagem automática como apoio para o diagnóstico precoce da diabetes mellitus, utilizando variáveis de dados ambientais, sociais, econômicos e sanitários, sem a dependência da coleta de amostras clínicas. Metodologia: Usaram-se dados de 10.889 usuários filiados ao regime subsidiado de saúde da zona sudoeste da Colômbia, diagnosticados com hipertensão e agrupados em usuários sem (74,3%) e com (25,7%) diabetes mellitus. Foram treinados modelos supervisionados utilizando k vizinhos mais próximos, árvores de decisão e florestas aleatórias, assim como modelos baseados em montagens, aplicados à base de dados antes de depois de equilibrar o número de casos em cada grupo de diagnóstico. Avaliou-se o desempenho dos algoritmos por meio da divisão da base de dados de treino e teste (70/30, respectivamente), e utilizaram-se métricas de exatidão, sensibilidade, especificidade e área sob a curva. Resultados: Os valores de sensibilidade aumentaram de maneira significativa ao utilizar dados equilibrados, passando de valores máximos de 17,1% (dados sem equilibrar) a valores de até 57,4% (dados equilibrados). O valor mais elevado de área sob a curva (0,61) foi obtido com os modelos de montagens, ao aplicar um balanço no número de dados por cada grupo e codificar as variáveis categóricas. As variáveis de maior peso estiveram associadas com fatores hereditários (24,65%) e com o grupo étnico (5,59%), além da dificuldade visual, o baixo consumo de água, um regime baixo em frutas e vegetais e o consumo de sal e açúcar. Conclusões: Embora os modelos preditivos, utilizando informação socioeconômica e ambiental das pessoas, surgem como uma ferramenta para o diagnóstico precoce da diabetes mellitus, ainda devem ser melhorados em sua capacidade preditiva.

Applicability of the acute leukemia (AL) - EBMT score as a prognostic model for allogeneic hematopoietic stem cell transplantation: a single-center, prospective, cohort study at a reference transplant center in Brazil

Abstract Introduction The Acute Leukemia-European Society for Blood and Marrow Transplantation (AL-EBMT) risk score was recently developed and validated by Shouval et al. Objective To assess the ability of this score in predicting the 2-year overall survival (OS-2), leukemia-free survival (LFS-2) and transplant-related mortality (TRM) in acute leukemia (AL) adult patients undergoing a first allogeneic hematopoietic stem cell transplant (HSCT) at a transplant center in Brazil. Methods In this prospective, cohort study, we used the formula published by Shouval et al. to calculate the AL-EBMT score and stratify patients into three risk categories. Results A total of 79 patients transplanted between 2008 and 2018 were analyzed. The median age was 38 years. Acute myeloid leukemia was the most common diagnosis (68%). Almost a quarter of the cases were at an advanced stage. All hematopoietic stem cell transplantations (HSCTs) were human leukocyte antigen-matched (HLA-matched) and the majority used familial donors (77%). Myeloablative conditioning was used in 92% of the cases. Stratification according to the AL-EBMT score into low-, intermediate- and high-risk groups yielded the following results: 40%, 12% and 47% of the cases, respectively. The high scoring group was associated with a hazard ratio of 2.1 (p= 0.007), 2.1 (p= 0.009) and 2.47 (p= 0.01) for the 2-year OS, LFS and TRM, respectively. Conclusion This study supports the ability of the AL-EBMT score to reasonably predict the 2-year post-transplant OS, LFS and TRM and to discriminate between risk categories in adult patients with AL, thus confirming its usefulness in clinical decision-making in this setting. Larger, multicenter studies may further help confirm these findings.

Factors influencing pulmonary infection in elderly neurocritical patients and their predictive values: a data analysis for consecutive four-year / 中华危重病急救医学

Objective:To analyze the factors influencing pulmonary infections in elderly neurocritical patients in the intensive care unit (ICU) and to explore the predictive value of risk factors for pulmonary infections.Methods:The clinical data of 713 elderly neurocritical patients [age ≥ 65 years, Glasgow coma score (GCS) ≤ 12 points] admitted to the department of critical care medicine of the Affiliated Hospital of Guizhou Medical University from 1 January 2016 to 31 December 2019 were retrospectively analyzed. According to whether or not they had HAP, the elderly neurocritical patients were divided into hospital-acquired pneumonia (HAP) group and non-HAP group. The differences in baseline data, medication and treatment, and outcome indicators between the two groups were compared. Logistic regression analysis was used to analyze the factors influencing the occurrence of pulmonary infection.The receiver operator characteristic curve (ROC curve) was plotted for risk factors and a predictive model was constructed to evaluate the predictive value for pulmonary infection.Results:A total of 341 patients were enrolled in the analysis, including 164 non-HAP patients and 177 HAP patients. The incidence of HAP was 51.91%. According to univariate analysis, compared with the non-HAP group, mechanical ventilation time, the length of ICU stay and total hospitalization in the HAP group were significantly longer [mechanical ventilation time (hours): 171.00 (95.00, 273.00) vs. 60.17 (24.50, 120.75), the length of ICU stay (hours): 263.50 (160.00, 409.00) vs. 114.00 (77.05, 187.50), total hospitalization (days): 29.00 (13.50, 39.50) vs. 27.00 (11.00, 29.50), all P < 0.01], the proportion of open airway, diabetes, proton pump inhibitor (PPI), sedative, blood transfusion, glucocorticoids, and GCS ≤ 8 points were significantly increased than those in HAP group [open airway: 95.5% vs. 71.3%, diabetes: 42.9% vs. 21.3%, PPI: 76.3% vs. 63.4%, sedative: 93.8% vs. 78.7%, blood transfusion: 57.1% vs. 29.9%, glucocorticoids: 19.2% vs. 4.3%, GCS ≤ 8 points: 83.6% vs. 57.9%, all P < 0.05], prealbumin (PA) and lymphocyte count (LYM) decreased significantly [PA (g/L): 125.28±47.46 vs. 158.57±54.12, LYM (×10 9/L): 0.79 (0.52, 1.23) vs. 1.05 (0.66, 1.57), both P < 0.01]. Logistic regression analysis showed that open airway, diabetes, blood transfusion, glucocorticoids and GCS ≤ 8 points were independent risk factors for pulmonary infection in elderly neurocritical patients [open airway: odds ratio ( OR) = 6.522, 95% confidence interval (95% CI) was 2.369-17.961; diabetes: OR = 3.917, 95% CI was 2.099-7.309; blood transfusion: OR = 2.730, 95% CI was 1.526-4.883; glucocorticoids: OR = 6.609, 95% CI was 2.273-19.215; GCS ≤ 8 points: OR = 4.191, 95% CI was 2.198-7.991, all P < 0.01], and LYM, PA were the protective factors for pulmonary infection in elderly neurocritical patients (LYM: OR = 0.508, 95% CI was 0.345-0.748; PA: OR = 0.988, 95% CI was 0.982-0.994, both P < 0.01). ROC curve analysis showed that the area under the ROC curve (AUC) for predicting HAP using the above risk factors was 0.812 (95% CI was 0.767-0.857, P < 0.001), with a sensitivity of 72.3% and a specificity of 78.7%. Conclusions:Open airway, diabetes, glucocorticoids, blood transfusion, GCS ≤ 8 points are independent risk factors for pulmonary infection in elderly neurocritical patients. The prediction model constructed by the above mentioned risk factors has certain predictive value for the occurrence of pulmonary infection in elderly neurocritical patients.

Establishment and validation of a model for predicting recurrent wheezing in children with bronchiolitis / 中国医师进修杂志

Objective:To analyze the risk factors of recurrent wheezing in children with bronchiolitis and to construct a predictive model.Methods:Prospective research methods was used. One hundred and eighty children with bronchiolitis who were treated in Hefei Eighth People's Hospital from February 2017 to February 2019 were selected as the study subjects, and the included children were separated into a modeling group (126 cases) and a validation group (54 cases) according to 7∶3. The children were followed up for 3 years, and then the modeling group was divided into wheezing group (48 cases) and no wheezing group (78 cases) according to whether the children had recurrent wheezing. The Hosmer-Lemeshow fitting curve and receiver operating characteristic (ROC) curve were drawn to evaluate the validity and accuracy of the constructed prediction model.Results:Multivariate Logistic regression analysis showed that artificial feeding ( OR = 8.838, 95% CI 2.601 to 30.027), family history of allergies ( OR = 6.709, 95% CI 1.825 to 24.665), underlying diseases ( OR = 8.114, 95% CI 1.638 to 40.184), and higher IgE level ( OR = 1.020, 95% CI 1.012 to 1.029) were the independent risk factors for recurrent wheezing in children with bronchiolitis ( P<0.05). The area under the curve of the modeling group was 0.917 (95% CI 0.855 to 0.959), and the sensitivity and specificity were 83.33% and 85.90%, respectively; the area under the curve of the validation group was 0.911 (95% CI 0.847 to 0.954), and the sensitivity and specificity were 89.58% and 79.49%, respectively. Conclusions:Artificial feeding, family history of allergies, underlying diseases, and higher IgE level are the independent risk factors for recurrent wheezing in children with bronchiolitis. The constructed prediction model has good accuracy and validity, and can be used as an effective tool for clinical prediction of recurrent wheezing in children with bronchiolitis.

The value of preoperative inflammatory markers in predicting recurrence of hepatocellular carcinoma after liver transplantation / 中华消化外科杂志

Objective:To investigate the value of aspartate aminotransferase/lymphocyte ratio (ALR), γ-glutamyltranspeptidase/lymphocyte ratio (GLR) and aspartate aminotransferase/alanine aminotransferase ratio (AAR) in predicting the recurrence of hepatocellular carcinoma after liver transplantation.Methods:The retrospective cohort study was conducted. The clinicopathological data of 178 patients with hepatocellular carcinoma who underwent liver transplantation in Tianjin First Central Hospital from July 2014 to June 2018 were collected. There were 156 males and 22 females, aged (54±9)years. All patients received the first time of orthotopic liver transplantation. Observation indicators: (1) follow-up; (2) the predictive value and cutoff value of each index for tumor recur-rence of patients with hepatocellular carcinoma after liver transplantation; (3) analysis of risk factors for tumor recurrence of patients with hepatocellular carcinoma after liver transplantation; (4) cons-truction and evaluation of the predictive model for tumor recurrence of patients with hepatocellular carcinoma after liver transplantation. Measurement data with normal distribution were represented as Mean± SD, and comparison between groups was conducted using the t test. Measurement data with skewed distribution were represented as M(range), and comparison between groups was conducted using the Mann-Whitney U test. Count data were expressed as absolute numbers, and comparison between groups was conducted using the chi-square test or Fisher exact probability. The Kaplan-Meier method was used to draw survival curve and the Log-rank test was used for survival analysis. Factors with P<0.05 in univariate analysis were included in multivariate analysis. Univariate analysis and multivariate analysis were performed by COX proportional risk regression model with forward method. The regression coefficient was used to build the prediction model. The receiver operating characteristic curve was drawn, and the area under curve (AUC) was used to evaluate the predictive ability of prediction model. Results:(1) Follow-up. All 178 patients with hepatocellular carcinoma were followed up for 36(range, 1?74)months after liver transplantation. During the follow-up, there were 41 patients died, 61 patients with tumor recurrence and 117 cases without tumor recurrence. The 3-, 5-year overall survival rates and 3-, 5-year tumor recurrence free survival rates of patients after liver transplantation were 72.8%, 69.9% and 57.3%, 52.8%, respectively. (2) The predictive value and cutoff value of each index for tumor recurrence of patients with hepatocellular carcinoma after liver transplantation. The AUC of preoperative serum alpha fetoprotein (AFP), tumor diameter, ALR, GLR, neutrophil to lymphocyte ratio, AAR in recipients were 0.76, 0.70, 0.69, 0.65, 0.64, 0.65 (95% confidence interval as 0.68?0.83, 0.61?0.79, 0.61?0.77, 0.57?0.74, 0.56?0.73, 0.56?0.74, P<0.05), and the corresponding best cutoff value of each index were 228.00 μg/L, 5.25 cm, 92.90, 122.40, 3.00, 2.42. (3) Analysis of risk factors for tumor recurrence of patients with hepato-cellular carcinoma after liver transplantation. Results of multivariate analysis showed the preoperative serum AFP >228.88 μg/L, number of tumor as multiple, tumor diameter >5.25 cm, ALR >92.90, AAR >2.42 were indepen-dent risk factors for tumor recurrence of hepatocellular carcinoma after liver transplantation ( hazard ratio=3.13, 1.90, 2.66, 2.40, 2.75, 95% confidence interval as 1.81?5.41, 1.08?3.35, 1.49?4.74, 1.40?4.11, 1.54?4.91, P<0.05). (4) Construction and evaluation of the predictive model for tumor recurrence of patients with hepatocellular carcinoma after liver transplantation. According to the results of multivariate analysis, the preoperative serum AFP, number of tumor, tumor diameter, ALR, AAR were used to construct the predictive model for tumor recurrence of hepatocellular carcinoma after liver transplantation. The AUC, best cutoff value, specificity and sensitivity of the predictive model were 0.83 (95% confidence interval as 0.76?0.89, P<0.05), 5.5, 80.3% and 73.8%. Of the 178 patients, there were 110 patients with low risk of tumor recurrence (scoring as 0?5) and 68 patients with high risk of tumor recurrence (scoring as 6?16) after liver transplantation. The 1-, 3-, 5-year tumor recurrence free survival rates and 1-, 3-, 5-year overall survival rates of patients with high risk of tumor recurrence were 27.7%, 18.2%, 18.2% and 63.7%, 48.9%, 48.9%, respectively. The above indicators of patients with low risk of tumor recurrence were 92.3%, 82.4%, 74.6% and 90.4%, 87.7%, 83.6%, respectively. There were significant differences of the above indicators between patients with high risk of tumor recurrence and low risk of tumor recurrence ( χ2=67.83, 21.95, P<0.05). Conclusions:The preoperative serum AFP, number of tumor, tumor diameter, ALR, AAR are independent influencing factors for tumor recurrence of hepato-cellular carcinoma after liver transplantation. The predictive model constructed based on the above indexes has a good prediction efficiency.

Construction of a hypoglycemia prediction model for older adults with type 2 diabetes based on random forest algorithm / 中国实用护理杂志

Objective:To construct a hypoglycemia random forest prediction model for older adults with type 2 diabetes, and assess the model′s prognostication performance through internal and external verification.Methods:From August 2022 to January 2023, 300 older adults with type 2 diabetes in Beijing Hospital were selected. The demographic characteristics, medical history, laboratory tests, and other data of the patients were collected, and the data set was randomly divided into the training set and verification set in a ratio of 7∶3. The hypoglycemia prediction model for older adults with type 2 diabetes was constructed and optimized based on the random forest algorithm. The calibration curve was used to evaluate the model′s calibration, and the ROC was used to evaluate the model′s discrimination. The clinical applicability of the model was assessed by the decision curve analysis. The risk factors for hypoglycemia in the older adults were explored by prioritizing the contributions of variables in prediction. The Bootstrap method was used for internal validation, and the validation set was used for external validation.Results:Among the 300 older adults with type 2 diabetes, 128 cases (42.67%) experienced hypoglycemia within one week. The predictive contributions of risk factors in the model were ranked as follows: the number of episodes of hypoglycemia in one month, HDL-C, heart disease, diabetes knowledge and education, combination therapy, age, duration of diabetes, staple food restriction, glycosylated hemoglobin, and gender. The internal and external calibration curves of the hypoglycemia random forest model for the older adults with type 2 diabetes fluctuated around the diagonal, indicating that the calibration degree of the predictive model is good. The AUROC of internal verification was 0.823 (95% CI 0.752-0.894), the sensitivity and specificity were 0.867 and 0.698, respectively. The external verification was 0.859 (95% CI 0.817 - 0.902), and sensitivity and specificity were 0.789 and 0.804, respectively, showing that the overall discrimination of the prediction model was good. The DCA curves were far from the all-positive line and all-negative line, which indicated that the prediction model had good clinical applicability. Conclusions:The predictive effect of this model is good, and it is suitable for predicting the risk of hypoglycemia in older adults with type 2 diabetes, and it provides a reference for early hypoglycemia screening and predictive intervention for this kind of patients.

Value of CT-based model in differential diagnosis of papillary thyroid carcinoma / 中华内分泌外科杂志

Objective:To evaluate the differential diagnosis of papillary thyroid carcinoma (PTC) based on CT signs.Methods:Retrospective analysis was performed on 156 patients with PTC confirmed by surgery and pathology in the Imaging Department of Jinhua Hospital Affiliated to Zhejiang University College of Medicine from Jan. 2017 to Jan. 2022 as PTC group, and 132 patients with nodular goiter (NG) as NG group. There were 112 females and 45 males in the PTC group. The age was (49.32±3.25) years. There were 104 females and 52 males in NG group. The age was (50.12±3.27) years. Preoperative plain and contrast-enhanced CT scans were performed to analyze the features of the images, and univariate analysis was performed on the morphologic features, high tension, plain "bite cake sign" , enhanced "bite cake sign" , microcalcification, blurred/reduced scope after enhancement, nodule density, and asymmetric diffuse enlargement of thyroid gland. Statistically significant factors were included in the multivariate Logistic regression analysis, and the differential model of PTC was established according to the selected risk factors. The value of the model in the differential diagnosis of PTC was evaluated by ROC curve.Results:The percentages of irregular shape, no high tension, plain scan "bite cake sign" , enhanced "bite cake sign" , microcalcification, enhanced blur/reduced scope, uniform nodule density, completely slightly low tissue density, no cystic degeneration, and asymmetric thyroid diffuse enlargement in PTC group were higher than those in NG group ( χ2=161.014, 3.387, 95.885, 151.331, 60) . 200, 18.104, 105.260, 16.855, 89.064, 16.913, P<0.05) , suggesting that the above CT signs had important diagnostic value in differentiating PTC and NG. Among the single CT signs, plain scan "bite cake sign" had the highest sensitivity, specificity and accuracy. The sensitivity of PTC diagnosis combined with other signs gradually decreased, while the specificity gradually increased. At the same time, plain scan "bite cake sign" and microcalcification signs had high specificity in PTC identification, and the specificity of PTC identification reached 100.00% when any 4 or more signs were present. Multivariate Logistic regression analysis. The results showed that irregular morphology ( OR=15.831, 95% CI: 7.444-33.670) , high tension ( OR=0.162, 95% CI: 0.108-0.242) , plain scan "bite cake sign" ( OR=5.601, 95% CI: 2.691-11.659) , microcalcification ( OR=4.031, 95% CI: 2.062-7.880) , edge blur/range reduction after enhancement ( OR=4.761, 95% CI: 3.126-7.260) , uniform density of nodules ( OR=4.778, 95% CI: 3.299-6.290) and increased asymmetric diffusion ( OR=3.758, 95% CI: 1.911-7.391) were important signs for distinguishing NG from PTC ( P<0.05) . The above factors were incorporated into the Logistic regression equation to construct the model, and then the ROC curve was drawn. The results showed that the area under the curve of the model established based on CT signs was 0.94 (0.925-0.983) , and the sensitivity and specificity were 90.37% and 91.45%, respectively. Conclusions:In CT signs, irregular shape, high tension, "biting cake sign" on plain scan, microcalcification, blurred edge/scope reduction signs after enhancement, and uniform nodule density are important signs for differentiating papillary thyroid carcinoma from nodular goiter. The constructed model has good predictive value for identifying papillary thyroid carcinoma.

Development and validation of a nomogram to predict in-hospital cardiac arrest in critically ill patients / 中华急诊医学杂志

Objective:To explore the independent risk factors of in-hospital cardiac arrest (IHCA) in critically ill patients and construct a nomogram model to predict the risk of IHCA based on the identified risk factors.Methods:Patients who were admitted to the intensive care units (ICUs) from 2008 to 2019 were retrospectively enrolled from the Medical Information Mart for Intensive Care -Ⅳ database. The patients were excluded if they (1) were younger than 18 years old, (2) had repeated ICU admission records, or (3) had an ICU stay shorter than 24 h. The patients were randomly divided into the training and internal validation cohorts (7 : 3). Univariate and multivariate logistic regression models were used to identify independent risk factors of IHCA, and a nomogram was constructed based on these independent risk factors. Calibration curve, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA) were used to evaluate the nomogram model. Finally, the nomogram was externally validated using the emergency ICU collaborative research database.Results:A total of 41,951 critically ill patients were enrolled (training cohort, n=29 366; internal validation cohort, n=12 585). Multivariate analysis showed that myocardial infarction, pulmonary heart disease, cardiogenic shock, respiratory failure, acute kidney injury, respiratory rate, glucose, hematocrit, sodium, anion gap, vasoactive drug use, and invasive mechanical ventilation were independent risk factors of IHCA. Based on the above risk factors, a nomogram for predicting IHCA was constructed. The area under the ROC curve (AUC) of the nomogram was 0.817 (95% CI: 0.785–0.847). The calibration curve showed that the predicted and actual probabilities of the nomogram were consistent. Moreover, DCA showed that the nomogram had clinical benefits for predicting IHCA. In the internal validation cohort, the nomogram had a similar predictive value of IHCA (AUC=0.807, 95% CI: 0.760–0.862). In an external validation cohort of 87,626 critically ill patients, the nomogram had stable ability for predicting IHCA (AUC=0.804, 95% CI: 0.786–0.822). In addition, the nomogram also had predictive value for in-hospital mortality (AUC=0.818, 95% CI: 0.802-0.834). Conclusions:The nomogram is constructed based on identified independent risk factors, which has good predictive value for IHCA. Moreover, the performance of the nomogram in the external validation cohort is robust. The study findings may help clinicians to assess the risk of IHCA in critically ill patients.

Risk Factors and Predictive Model for Severe Myelosuppression due to Chemotherapy in Triple-negative Breast Cancer / 中山大学学报(医学科学版)

ObjectiveTo investigate the risk factors and construct a predictive model for severe myelosuppression due to chemotherapy in triple negative breast cancer （TNBC）. MethodsPatients with TNBC who received anthracycline combined with cyclophosphamide sequential paclitaxel chemotherapy regimen at the Second Affiliated Hospital of Nanchang University from September 2， 2016 to September 2， 2021 were selected and assigned to severe myelosuppression group and no/mild myelosuppression group. The χ2 test and binary logistic regression were used to analyze the risk factors for severe myelosuppression due to chemotherapy and to develop a prediction model. Hosmer-Lemeshow test and receiver operating characteristic （ROC） curve were used to evaluate the predictive efficiency of the regression model. Kappa consistency test was used to verify the regression model externally. ResultsA total of 207 patients who met the inclusion were enrolled and 106 patients （51%） had severe myelosuppression. Binary logistic regression multivariate analysis showed that age 40 to 60 years （OR = 3.463， 95% CI： 1.144 to 10.486， P = 0.028）， age >60 years （OR = 3.474， 95% CI： 1.004 to 12.020， P = 0.049）， body mass index （BMI） 18.5 to 24.0 （OR = 1.445， 95% CI： 0.686 to 3.087， P = 0.328）， BMI <18.5 （OR = 3.582， 95% CI： 1.260 to 10.182， P = 0.017）， tumor TNM stage Ⅱ （OR = 1.698， 95% CI： 0.831 to 3.468， P = 0.146）， tumor TNM stage Ⅲ （OR = 2.943， 95% CI： 1.199 to 7.227， P = 0.019）， previous diabetes （OR = 2.441， 95% CI： 1.076 to 5.539， P = 0.033）， low pre-treatment albumin level （OR = 2.759， 95% CI： 1.141 to 6.669， P = 0.024） and low pre-treatment lymphocytes （OR = 3.428， 95% CI： 1.689 to 6.958， P = 0.001） were independent risk factors for severe myelosuppression due to chemotherapy. The χ2 value for the logistic regression model Hosmer-Lemeshow test was 11.507， P= 0.175， the area under the ROC curve was 0.763， standard error 0.033， 95% CI： 0.698-0.828， P=0.000. External validation showed that the prediction model had a specificity of 88% and a sensitivity of 80%； the kappa value was 0.679， standard error 0.081， P=0.000. conclusionThis logistic regression model had high predictive efficacy and is useful for clinicians to predict whether patients with TNBC develop severe myelosuppression.

Construction of a new model for evaluating insulin resistance in newly diagnosed type 2 diabetic patients using anthropometry parameters / 中华内分泌代谢杂志

Objective:To construct a new model for assessing insulin resistance(IR) in newly diagnosed type 2 diabetic patients by combining anthropometry parameters and biochemical parameters.Methods:A total of 677 newly diagnosed type 2 diabetic patients were included in this study. Clinical data, biochemical indicators, and body composition measurements were collected, and a predictive model was constructed using logistic regression analysis.Results:The IR prediction model was constructed based on five indicators: triglycerides(TG), fasting plasma glucose(FPG), visceral fat area(VFA), alanine aminotransferase(ALT), and uric acid(UA). The formula for the new predictive model was as follows: y=-17.765+ 1.389×ln VFA+ 1.045×ln UA+ 0.91×ln ALT+ 2.167×ln FPG+ 0.805×ln TG. The receiver operating characteristic curve(ROC) area under the curve(AUC) for the model was 0.82, with an optimal cutoff value of 1.67, sensitivity of 0.80, and specificity of 0.71. The AUC values for the triglyceride glucose(TyG) index, lipid accumulation product(LAP), and triglyceride/high-density lipoprotein cholesterol ratio(THR) were 0.75, 0.75, and 0.70, respectively. The corresponding sensitivities were 0.66, 0.84, and 0.71, and the specificities were 0.71, 0.59, and 0.60. The optimal cutoff values were 1.81, 30.31, and 1.14, respectively. Conclusion:The new model constructed using TG, FPG, VFA, ALT, and UA as indicators showed high predictive value and can serve as a new model for assessing IR in newly diagnosed type 2 diabetic patients.

Risk factors for postoperative cognitive dysfunction and the predictive model in elderly patients undergoing lumbar surgery under general anesthesia / 中华麻醉学杂志

Objective:To identify the risk factors for postoperative cognitive dysfunction (POCD) and develop the prediction model in elderly patients undergoing lumbar surgery under general anesthesia.Methods:The elderly patients undergoing elective lumbar surgery under general anesthesia in our hospital from July 2021 to July 2022 were enrolled. Cognitive function was assessed at 7 days after surgery using Mini-Mental State Examination and Montreal Cognitive Assessment. When the decrease in both scales≥ 1 standard deviation, the patients were considered as having POCD. The patients were divided into POCD group and non-POCD group according to whether POCD developed. The propensity score matching was used to balance the confounding bias between two groups. The multivariate logistic regression analysis was used to identify the risk factors for POCD. The prediction model was constructed, and a nomogram was drawn for visualization of the model. The receiver operating characteristic curve, calibration plot and decision curve analysis (DCA) were drawn to evaluate the differentiation, consistency and clinical validity of the model, respectively.Results:A total of 159 patients were enrolled in this study, and the incidence of POCD was 31.4%. There were statistically significant differences in the ratio of intraoperative blood transfusion, cumulative time of hypotension, total infusion volume and operation time between two groups ( n=32 each) after propensity score matching ( P<0.05). The results of multivariate logistic regression showed that age, educational levels, diabetes mellitus, previous two or more operations under general anesthesia, APTT and cumulative time of hypotension were independent risk factors for POCD in elderly patients undergoing lumbar surgery under general anesthesia ( P<0.05). A model was developed based on the risk factors mentioned above: LogitP=-15.878+ 0.263 × Age (years) - 0.122 × Educational Level (years)+ 1.601 × Diabetes Mellitus+ 1.468 × History of General Anesthesia for 2 or more times+ 0.608 × Cumulative Time of Hypotension(min) - 0.140 × APTT (s). The area under the receiver operating characteristic curve was 0.930 (95% CI 0.887-0.973), the sensitivity was 0.920, specificity was 0.798 and Youden index was 0.718. After visualizing the model via nomogram, the model was verified by Hosmer-Lemeshow test, P=0.403, C index was 0.930, and corrected C index was 0.914. Conclusions:Age, educational levels, diabetes mellitus, previous multiple operations under general anesthesia, APTT and cumulative time of hypotension are independent risk factors for POCD in elderly patients undergoing lumbar surgery under general anesthesia, and the established risk prediction model can effectively predict the occurrence of POCD in elderly patients undergoing lumbar surgery under general anesthesia.

An acute kidney injury risk prediction model for elderly patients with sepsis based on the intensive care medicine information database-Ⅳ / 中华老年医学杂志

Objective:To explore the risk factors of acute kidney injury(stage 3)developed within 48 hours in elderly patients with sepsis, and to use them to develop a risk prediction model and then evaluate and externally validate the model.Methods:Clinical data of all elderly patients(age≥ 60 years)with sepsis in the intensive care medicine information database(MIMIC-Ⅳ v1.0)were extracted.Independent risk factors were determined by multivariate logistic regression analysis.A risk prediction model was constructed, a nomogram was drawn, and the receiver operating characteristic curve(ROC)and the Hosmer-Lemeshow(H-L)test were used to evaluate the model's prediction accuracy and R-squared.Clinical data of elderly patients(age≥ 60 years)with sepsis admitted to the Department of Critical Care Medicine of the Second People's Hospital of Hefei from May 2019 to October 2021 were retrospectively collected and fed into the prediction model to conduct external validation.Results:A total of 1 977 elderly patients with sepsis were screened out from the MIMIC-IV database and included in the training set, of whom, 544 developed AKI-stage 3 within 48 hours.Univariate analysis was performed for factors that might be associated with acute kidney injury in elderly patients with sepsis.Compared with the normal group that did not progress to AKI stage 3, there were statistically significant differences in 28 indicators, such as the duration of ICU stay, intravenous fluid intake in 24 hours, and use of vasoactive drugs[5(3, 9)d vs.7(4, 12)d; 2.05(1.17, 3.27)ml·kg -1·h -1vs.2.37(1.47, 4.10)ml·kg -1·h -1; 761(53.11%) vs.375(68.93%), P<0.001]. Based on the results of multivariate logistic regression analysis, a prediction model was finally constructed with 9 variables: albumin( OR=0.983, 95% CI: 0.966-0.999, P=0.040), aspartate transaminase( OR=1.000, 95% CI: 1.000-1.000, P<0.001), APTT( OR=1.005, 95% CI: 1.001-1.009, P=0.028), total bilirubin( OR=1.003, 95% CI: 1.001-1.004, P=0.001), serum creatinine( OR=1.005, 95% CI: 1.004~1.007, P<0.001), Charlson score( OR=1.117, 95% CI: 1.061-1.177, P<0.001), intravenous fluid intake in 24 hours( OR=1.101, 95% CI: 1.034-1.173, P=0.003), weight( OR=1.023, 95% CI: 1.018-1.029, P<0.001), and mechanical ventilation( OR=2.412, 95% CI: 1.843-3.157, P<0.001). Then a nomogram was generated.The area under the ROC curve(AUC)of the prediction model was 0.755(95% CI: 0.731-0.780), and the H-L test was conducted( χ2=10.89, P=0.208>0.05), indicating a good fit.Data from 102 elderly patients were included in the validation set, with 27 cases that had developed AKI-stage3 within 48 hours, and were fed into the prediction model, with an AUC of 0.778(95% CI: 0.676-0.880)and χ2=3.72 and P=0.882>0.05 from the H-L test, consistent with the results of the training set. Conclusions:The model has some predictive value for acute kidney injury in elderly patients with sepsis.

Risk factors of carbapenem-resistant Klebsiella pneumoniae infection in ICU patients and construction of a prediction model for infection / 中华临床感染病杂志

Objective:To analyze the risk factors of carbapenem-resistant Klebsiella pneumoniae (CRKP) infection in intensive care unit (ICU) patients and to construct a prediction model for infection. Methods:The clinical data of 204 patients with Klebsiella pneumoniae infection admitted in ICU of Jining First Hospital during January 2020 to December 2022 were retrospectively analyzed. Patients admitted during January 2020 to December 2021 were selected as model set ( n=150), and patients admitted during January to December 2022 were selected as validation set ( n=54). In model set, there were 59 cases infected with CRKP (CRKP group) and 91 cases infected with carbapenem-sensitive Klebsiella pneumonia (CSKP group). The risk factor of CRKP infection in ICU patients were analyzed with multivariate Logistic regression, based on which an infection prediction model was constructed. The predictive value of the model was evaluated by ROC, and verified in the validation group. Results:Multivariate Logistic regression analysis showed that empirical use of beta-lactam antibiotics( OR=6.985, 95 % CI 1.658-29.423, P=0.008), central vein catheterization( OR=7.486, 95 % CI 2.776-20.186, P<0.001)and tracheal intubation/incision( OR=10.695, 95 % CI 2.701-42.351, P=0.001)were risk factors for CRKP infection in ICU patients. The regression equation for predicting the risk of infection was -4.851+ empirical use of beta-lactam antibiotics×1.944+ central vein catheterization×2.013+ tracheal intubation/incision×2.370. The area under the ROC curve (AUC) of the model for predicting infection in the model group was 0.905, with sensitivity and specificity of 79.7% and 90.1%, respectively. The AUC of the model for predicting infection in validation group was 0.881, with sensitivity and specificity of 84.2% and 85.7%, respectively. Conclusion:The constructed infection prediction model in the study can effectively predict CRKP infection in ICU patients.

Development and validation of a predictive model for the assessment of potassium-lowering treatment among hyperkalemia patients / 世界急诊医学杂志（英文）

@#BACKGROUND: Hyperkalemia is common among patients in emergency department and is associated with mortality. While, there is a lack of good evaluation and prediction methods for the efficacy of potassium-lowering treatment, making the drug dosage adjustment quite difficult. We aimed to develop a predictive model to provide early forecasting of treating effects for hyperkalemia patients. METHODS: Around 80% of hyperkalemia patients (n=818) were randomly selected as the training dataset and the remaining 20% (n=196) as the validating dataset. According to the serum potassium (K+) levels after the first round of potassium-lowering treatment, patients were classified into the effective and ineffective groups. Multivariate logistic regression analyses were performed to develop a prediction model. The receiver operating characteristic (ROC) curve and calibration curve analysis were used for model validation. RESULTS: In the training dataset, 429 patients had favorable effects after treatment (effective group), and 389 had poor therapeutic outcomes (ineffective group). Patients in the ineffective group had a higher percentage of renal disease (P=0.007), peripheral edema (P<0.001), oliguria (P=0.001), or higher initial serum K+ level (P<0.001). The percentage of insulin usage was higher in the effective group than in the ineffective group (P=0.005). After multivariate logistic regression analysis, we found age, peripheral edema, oliguria, history of kidney transplantation, end-stage renal disease, insulin, and initial serum K+ were all independently associated with favorable treatment effects. CONCLUSION: The predictive model could provide early forecasting of therapeutic outcomes for hyperkalemia patients after drug treatment, which could help clinicians to identify hyperkalemia patients with high risk and adjust the dosage of medication for potassium-lowering.