OBJECTIVE: To analyze the influencing factors of postoperative thrombocytopenia in critically ill patients with heart disease and construct a nomogram prediction model.
METHODS: From October 2022 to October 2023, 319 critically ill patients with heart disease who visited our hospital were collected and separated into postoperative thrombocytopenia group (n = 142) and no postoperative thrombocytopenia group (n = 177) based on their postoperative thrombocytopenia, Logistic regression analysis was applied to screen risk factors for postoperative thrombocytopenia in critically ill patients with heart disease; R software was applied to construct a nomogram for predicting postoperative thrombocytopenia in critically ill patients with heart disease, and ROC curves, calibration curves, and Hosmer-Lemeshow goodness of fit tests were applied to evaluate nomogram.
RESULTS: A total of 142 out of 319 critically ill patients had postoperative thrombocytopenia, accounting for 44.51%. Logistic regression analysis showed that gender (95% CI 1.607-4.402, P = 0.000), age ≥ 60 years (95% CI 1.380-3.697, P = 0.001), preoperative antiplatelet therapy (95% CI 1.254-3.420, P = 0.004), and extracorporeal circulation time > 120 min (95% CI 1.681-4.652, P = 0.000) were independent risk factors for postoperative thrombocytopenia in critically ill patients with heart disease. The area under the ROC curve was 0.719 (95% CI: 0.663-0.774). The slope of the calibration curve was close to 1, and the Hosmer-Lemeshow goodness of fit test was χ2 = 6.422, P = 0.491.
CONCLUSIONS: Postoperative thrombocytopenia in critically ill patients with heart disease is influenced by gender, age ≥ 60 years, preoperative antiplatelet therapy, and extracorporeal circulation time > 120 min. A nomogram established based on above multiple independent risk factors provides a method for clinical prediction of the risk of postoperative thrombocytopenia in critically ill patients with heart disease.

UNASSIGNED: Patients with heart failure (HF) with preserved ejection fraction (HFpEF) are more prone to atrial fibrillation (AF) compared to those with heart failure with reduced ejection fraction (HFrEF). Nevertheless, a risk prediction model for new-onset atrial fibrillation (NOAF) in HFpEF patients remains a notable gap, especially with respect to imaging indicators.
UNASSIGNED: We retrospectively analyzed 402 HFpEF subjects reviewed at the Affiliated Hospital of Qingdao University from 2017 to 2023. Cox regression analysis was performed to screen predictors of NOAF. A nomogram was constructed based on these factors and internally validated through the bootstrap resampling method. A performance comparison between the nomogram and the mC2HEST score was performed.
UNASSIGNED: Out of the 402 participants, 62 (15%) developed atrial fibrillation. The risk factors for NOAF were finally screened out to include age, chronic obstructive pulmonary disease (COPD), hyperthyroidism, renal dysfunction, left atrial anterior-posterior diameter (LAD), and pulmonary artery systolic pressure (PASP), all of which were identified to create the nomogram. We calculated the bootstrap-corrected C-index (0.819, 95% CI: 0.762-0.870) and drew receiver operator characteristic (ROC) curves [3-year areas under curves (AUC) = 0.827, 5-year AUC = 0.825], calibration curves, and clinical decision curves to evaluate the discrimination, calibration, and clinical adaptability of the six-factor nomogram. Based on two cutoff values calculated by X-tile software, the moderate- and high-risk groups had more NOAF cases than the low-risk group (P < 0.0001). Our nomogram showed better 3- and 5-year NOAF predictive performance than the mC2HEST score estimated by the Integrated Discriminant Improvement Index (IDI) and the Net Reclassification Index (NRI) (P < 0.05).
UNASSIGNED: The nomogram combining clinical features with echocardiographic indices helps predict NOAF among HFpEF patients.

OBJECTIVE: Parathyroidectomy (PTX) is an effective treatment for refractory secondary hyperparathyroidism (SHPT), but it can lead to hungry bone syndrome (HBS), significantly threatening the health of maintenance haemodialysis (MHD) patients. While previous studies have analyzed the risk factors for HBS post-PTX, the predictive performance and clinical applicability of these risk models need further validation. This study aims to construct and validate a risk prediction model for HBS in MHD patients with SHPT post-PTX.
METHODS: A retrospective analysis was conducted on 368 MHD patients with SHPT who underwent PTX at Changsha Jieao Nephrology Hospital from January 2020 to December 2021. Patients were divided into a HBS group and a non-HBS group based on the occurrence of HBS. General data, surgical information, and biochemical indicators were compared between the 2 groups. Multivariate logistic regression was used to identify factors influencing HBS, and a risk prediction model was established. The model\'s performance was evaluated using receiver operator characteristic (ROC) curves, decision curves, and calibration curves. External validation was performed on 170 MHD patients with SHPT who underwent PTX at the Third Xiangya Hospital of Central South University from January to December 2022.
RESULTS: The incidence of HBS post-PTX in MHD patients with SHPT was 60.60%. Logistic regression analysis identified preoperative bone involvement (OR=3.908, 95% CI 2.179 to 7.171), preoperative serum calcium (OR=7.174, 95% CI 2.291 to 24.015), preoperative intact parathyroid hormone (iPTH) (OR=1.001, 95% CI 1.001 to 1.001), preoperative alkaline phosphatase (ALP) (OR=1.001, 95% CI 1.000 to 1.001), and serum calcium on the first postoperative day (OR=0.006, 95% CI 0.001 to 0.038) as independent risk factors for HBS (all P<0.01). The constructed risk prediction model demonstrated good predictive performance in both internal and external validation cohorts. The internal validation cohort showed an accuracy of 0.821, sensitivity of 0.890, specificity of 0.776, Youden index of 0.666, and area under the curve (AUC) of 0.882 (95% CI 0.845 to 0.919). The external validation cohort showed an accuracy of 0.800, sensitivity of 0.806, specificity of 0.799, Youden index of 0.605, and AUC of 0.863 (95% CI 0.795 to 0.932).
CONCLUSIONS: Preoperative bone involvement, serum calcium, iPTH, ALP, and serum calcium on the first postoperative day are influencing factors for HBS in MHD patients with SHPT post-PTX. The constructed risk prediction model based on these factors is reliable.
目的: 甲状旁腺切除术(parathyroidectomy，PTX)是治疗难治性继发性甲状旁腺功能亢进(secondary hyperparathyroidism，SHPT)的有效方法，但PTX后极易出现骨饥饿综合征(hungry bone syndrome，HBS)，严重威胁维持性血液透析(maintenance hemodialysis，MHD)患者的生命健康。目前已有研究分析PTX后并发HBS的风险因素，但风险预测模型的预测性能和临床适用性仍待进一步验证。本研究旨在构建MHD伴SHPT患者PTX后并发HBS的风险预测模型，并验证其预测效果。方法: 回顾性收集2020年1月至2021年12月在长沙捷奥肾病医院行PTX的MHD伴SHPT的368例患者为训练集，按照是否发生HBS分为HBS组和non-HBS组，对2组的一般资料、手术相关信息、生化指标等进行比较，应用多因素logistic回归筛选HBS的影响因素，建立风险预测模型。采用受试者操作特征(receiver operator characteristic，ROC)曲线、决策曲线、校准曲线对模型进行评价。收集2022年1至12月在中南大学湘雅三医院行PTX的MHD伴SHPT的170例患者为验证集进行外部验证。结果: MHD伴SHPT患者PTX后HBS发生率为60.60%，logistic回归分析结果显示：术前骨骼受累(OR=3.908，95% CI 2.179~7.171)、术前血钙(OR=7.174，95% CI 2.291~24.015)、术前全段甲状旁腺激素(intact parathyroid hormone，iPTH)(OR=1.001，95% CI 1.001~1.001)、术前碱性磷酸酶(alkaline phosphatase，ALP)(OR=1.001，95% CI 1.000~1.001)、术后第1天血钙(OR=0.006，95% CI 0.001~0.038)是MHD患者伴SHPT行PTX后并发HBS的独立危险因素(均P<0.01)。构建的风险预测模型在内部训练集和外部验证集中均表现出良好的预测结果，内部验证集的准确度为0.821，灵敏度为0.890，特异度为0.776，约登指数为0.666，曲线下面积(area under curve，AUC)为0.882(95% CI 0.845~0.919)；外部验证集的准确度为0.800，灵敏度为0.806，特异度为0.799，约登指数为0.605，AUC为0.863(95% CI 0.795~0.932)。结论: 术前骨骼受累、术前血钙、术前iPTH、术前ALP、术后第1天血钙水平是MHD伴SHPT患者行PTX后并发HBS的影响因素，基于上述因素构建的风险预测模型可靠。.

UNASSIGNED: Early prediction and intervention are crucial for the prognosis of unexplained recurrent spontaneous abortion (uRSA). The main purpose of this study is to establish a risk prediction model for uRSA based on routine pre-pregnancy tests, in order to provide clinical physicians with indications of whether the patients are at high risk.
UNASSIGNED: This was a retrospective study conducted at the Prenatal Diagnosis Center of Henan Provincial People\'s Hospital between January 2019 and December 2022. Twelve routine pre-pregnancy tests and four basic personal information characteristics were collected. Pre-pregnancy tests include thyroid-stimulating hormone (TSH), free triiodothyronine (FT3), free thyroxine thyroid (FT4), thyroxine (TT4), total triiodothyronine (TT3), peroxidase antibody (TPO-Ab), thyroid globulin antibody (TG-Ab), 25-hydroxyvitamin D [25-(OH) D], ferritin (Ferr), Homocysteine (Hcy), vitamin B12 (VitB12), folic acid (FA). Basic personal information characteristics include age, body mass index (BMI), smoking history and drinking history. Logistic regression analysis was used to establish a risk prediction model, and receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were employed to evaluate the performance of prediction model.
UNASSIGNED: A total of 140 patients in uRSA group and 152 women in the control group were randomly split into a training set (n = 186) and a testing set (n = 106). Chi-square test results for each single characteristic indicated that, FT3 (p = 0.018), FT4 (p = 0.048), 25-(OH) D (p = 0.013) and FA (p = 0.044) were closely related to RSA. TG-Ab and TPO-Ab were also important characteristics according to clinical experience, so we established a risk prediction model for RSA based on the above six characteristics using logistic regression analysis. The prediction accuracy of the model on the testing set was 74.53%, and the area under ROC curve was 0.710. DCA curve indicated that the model had good clinical value.
UNASSIGNED: Pre-pregnancy tests such as FT3, FT4, TG-Ab, 25-(OH)D and FA were closely related to uRSA. This study successfully established a risk prediction model for RSA based on routine pre-pregnancy tests.

UNASSIGNED: This study aimed to develop an integrative dynamic nomogram, including N-terminal pro-B type natural peptide (NT-proBNP) and estimated glomerular filtration rate (eGFR), for predicting the risk of all-cause mortality in HFmrEF patients.
UNASSIGNED: 790 HFmrEF patients were prospectively enrolled in the development cohort for the model. The least absolute shrinkage and selection operator (LASSO) regression and Random Survival Forest (RSF) were employed to select predictors for all-cause mortality. Develop a nomogram based on the Cox proportional hazard model for predicting long-term mortality (1-, 3-, and 5-year) in HFmrEF. Internal validation was conducted using Bootstrap, and the final model was validated in an external cohort of 338 consecutive adult patients. Discrimination and predictive performance were evaluated by calculating the time-dependent concordance index (C-index), area under the ROC curve (AUC), and calibration curve, with clinical value assessed via decision curve analysis (DCA). Integrated discrimination improvement (IDI) and net reclassification improvement (NRI) were used to assess the contributions of NT-proBNP and eGFR to the nomogram. Finally, develop a dynamic nomogram using the \"Dynnom\" package.
UNASSIGNED: The optimal independent predictors for all-cause mortality (APSELNH: A: angiotensin-converting enzyme inhibitors/angiotensin receptor blockers/angiotensin receptor-neprilysin inhibitor (ACEI/ARB/ARNI), P: percutaneous coronary intervention/coronary artery bypass graft (PCI/CABG), S: stroke, E: eGFR, L: lg of NT-proBNP, N: NYHA, H: healthcare) were incorporated into the dynamic nomogram. The C-index in the development cohort and validation cohort were 0.858 and 0.826, respectively, with AUCs exceeding 0.8, indicating good discrimination and predictive ability. DCA curves and calibration curves demonstrated clinical applicability and good consistency of the nomogram. NT-proBNP and eGFR provided significant net benefits to the nomogram.
UNASSIGNED: In this study, the dynamic APSELNH nomogram developed serves as an accessible, functional, and effective clinical decision support calculator, offering accurate prognostic assessment for patients with HFmrEF.

UNASSIGNED: The prognosis of trauma patients is highly dependent on early medical diagnosis. By constructing a nomogram model, the risk of adverse outcomes can be displayed intuitively and individually, which has important clinical implications for medical diagnosis.
UNASSIGNED: To develop and evaluate models for predicting patients with adverse outcomes of trauma that can be used in different data availability settings in China.
UNASSIGNED: This was a retrospective prognostic study using data from 8 public tertiary hospitals in China from 2018. The data were randomly divided into a development set and a validation set. Simple, improved and extended models predicting adverse outcomes were developed, with adverse outcomes defined as in-hospital death or ICU transfer, and patient clinical characteristics, vital signs, diagnoses, and laboratory test values as predictors. The results of the models were presented in the form of nomograms, and performance was evaluated using area under the receiver operating characteristic curve (ROC-AUC), precision-recall (PR) curves (PR-AUC), Hosmer-Lemeshow goodness-of-fit test, calibration curve, and decision curve analysis (DCA).
UNASSIGNED: Our final dataset consisted of 18,629 patients (40.2% female, mean age of 52.3), 1,089 (5.85%) of whom resulted in adverse outcomes. In the external validation set, three models achieved ROC-AUC of 0.872, 0.881, and 0.903, and a PR-AUC of 0.339, 0.337, and 0.403, respectively. In terms of the calibration curves and DCA, the models also performed well.
UNASSIGNED: This prognostic study found that three prediction models and nomograms including the patient clinical characteristics, vital signs, diagnoses, and laboratory test values can support clinicians in more accurately identifying patients who are at risk of adverse outcomes in different settings based on data availability.

UNASSIGNED: Non-suicidal self-injury (NSSI) is a significant social issue, especially among adolescents with major depressive disorder (MDD). This study aimed to construct a risk prediction model using machine learning (ML) algorithms, such as XGBoost and random forest, to identify interventions for healthcare professionals working with adolescents with MDD.
UNASSIGNED: This study investigated 488 adolescents with MDD. Adolescents was randomly divided into 75% training set and 25% test set to testify the predictive value of risk prediction model. The prediction model was constructed using XGBoost and random forest algorithms. We evaluated the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, recall, F Score of the two models for comparing the performance of the two models.
UNASSIGNED: There were 161 (33.00%) participants having NSSI. Compared without NSSI, there were statistically significant differences in gender (P=0.035), age (P=0.036), depressive symptoms (P=0.042), sleep quality (P=0.030), dysfunctional attitudes (P=0.048), childhood trauma (P=0.046), interpersonal problems (P=0.047), psychoticism (P) (P=0.049), neuroticism (N) (P=0.044), punishing and Severe (F2) (P=0.045) and Overly-intervening and Protecting (M2) (P=0.047) with NSSI. The AUC values for random forest and XGBoost were 0.780 and 0.807, respectively. The top five most important risk predictors identified by both machine learning methods were dysfunctional attitude, childhood trauma, depressive symptoms, F2 and M2.
UNASSIGNED: The study demonstrates the suitability of prediction models for predicting NSSI behavior in Chinese adolescents with MDD based on ML. This model improves the assessment of NSSI in adolescents with MDD by health care professionals working. This provides a foundation for focused prevention and interventions by health care professionals working with these adolescents.

UNASSIGNED: There are almost 2 million adult patients with congenital heart disease in China, and the number of moderate and severe patients is increasing. However, few studies have investigated the risk of serious adverse events (SAE) after catheterization among them. The aim of this study was to identify risk factors for SAE related to cardiac catheterization and to provide the risk scoring model for predicting SAE.
UNASSIGNED: A total of 690 patients with moderate and severe adult patients with congenital heart disease (ACHD) who underwent cardiac catheterization in Wuhan Asian Heart Hospital Affiliated to Wuhan University of Science and Technology from January 2018 to January 2022 were retrospectively collected and subsequently divided into a modeling group and a verification group. A univariate analysis was performed on the identified SAE risk factors, and then significant factors were included in the multivariate logistic regression model to screen for independent predictors of SAE. The receiver operating characteristic curve (ROC) and the Hosmer-Lemeshow test were used to evaluate the discrimination and calibration of the model, respectively.
UNASSIGNED: A SAE occurred in 69 (10.0%) of the 690 catheterization procedures meeting inclusion criteria. The established SAE risk calculation formula was logit(p) = -6.134 + 0.992 × pulmonary artery hypertension (yes) + 1.459 × disease severity (severe) + 2.324 × procedure type (diagnostic and interventional) + 1.436 × cTnI ( ≥ 0.028 μ g/L) + 1.537 × NT-proBNP ( ≥ 126.65 pg/mL). The total score of the final risk score model based on the effect size of each predictor was 0 to 7, involving pulmonary artery hypertension (1 point), disease severity (1 point), procedure type (2 points), cTnI (1 point) and NT-proBNP (2 points), and the score greater than 3 means high risk. The C-statistic of the area under the ROC curve was 0.840 and 0.911 for the derivation and validation cohorts, respectively. According to the Hosmer-Lemeshow test, the p values in the modeling group and the verification group were 0.064 and 0.868, respectively.
UNASSIGNED: The risk prediction model developed in this study has high discrimination and calibration, which can provide reference for clinical prediction and evaluation of SAE risk after cardiac catheterization in patients with moderate and severe ACHD.

OBJECTIVE: The main aim of this study is to synthesize the prevalent predictive models for pressure injuries in hospitalized patients, with the goal of identifying common predictive factors linked to pressure injuries in hospitalized patients. This endeavour holds the potential to provide clinical nurses with a valuable reference for providing targeted care to high-risk patients.
BACKGROUND: Pressure injuries (PIs) are a frequently occurring health problem throughout the world. There are mounting studies about risk prediction model of PIs reported and published. However, the prediction performance of the models is still unclear.
METHODS: Systematic review and meta-analysis: The Cochrane Library, PubMed, Embase, CINAHL, Web of Science and Chinese databases including CNKI (China National Knowledge Infrastructure), Wanfang Database, Weipu Database and CBM (China Biology Medicine).
METHODS: This systematic review was conducted following PRISMA recommendations. The databases of Cochrane Library, PubMed, Embase, CINAHL, Web of Science, and CNKI, Weipu Database, Wanfang Database and CBM were searched for all studies published before September 2023. We included studies with cohort, case-control designs, reporting the development of risk model and have been validated externally and internally among the hospitalized patients. Two researchers selected the retrieved studies according to the inclusion and exclusion criteria, and critically evaluated the quality of studies based on the CHARMS checklist. The PRISMA guideline was used to report the systematic review and meta-analysis.
RESULTS: Sixty-two studies were included, which contained 99 pressure injuries risk prediction models. The AUC (area under ROC curve) of modelling in 32 prediction models were reported ranged from .70 to .99, while the AUC of verification in 38 models were reported ranged from .70 to .98. Gender (OR = 1.41, CI: .99 ~ 1.31), age (WMD = 8.81, CI: 8.11 ~ 9.57), diabetes mellitus (OR = 1.64, CI: 1.36 ~ 1.99), mechanical ventilation (OR = 2.71, CI: 2.05 ~ 3.57), length of hospital stay (WMD = 7.65, CI: 7.24 ~ 8.05) were the most common predictors of pressure injuries.
CONCLUSIONS: Studies of PIs risk prediction model in hospitalized patients had high research quality, and the risk prediction models also had good predictive performance. However, some of the included studies lacked of internal or external validation in modelling, which affected the stability and extendibility. The aged, male patient in ICU, albumin, haematocrit, low haemoglobin level, diabetes, mechanical ventilation and length of stay in hospital were high-risk factors for pressure injuries in hospitalized patients. In the future, it is recommended that clinical nurses, in practice, select predictive models with better performance to identify high-risk patients based on the actual situation and provide care targeting the high-risk factors to prevent the occurrence of diseases.
CONCLUSIONS: The risk prediction model is an effective tool for identifying patients at the risk of developing PIs. With the help of risk prediction tool, nurses can identify the high-risk patients and common predictive factors, predict the probability of developing PIs, then provide specific preventive measures to improve the outcomes of these patients.
UNASSIGNED: CRD42023445258.

OBJECTIVE: To conduct a systematic review to assess drug exposure handling in diabetic retinopathy (DR) risk prediction models, a network-meta-analysis to identify drugs associated with DR and a meta-analysis to determine which drugs contributed to enhanced model performance.
METHODS: Systematic review and meta-analysis.
METHODS: We included studies presenting DR models incorporating drug exposure as a predictor. We searched EMBASE, MEDLINE, and SCOPUS from inception to December 2023. We evaluated the quality of studies using the Prediction model Risk of Bias Assessment Tool and certainty using GRADE. We conducted network meta-analysis and meta-analysis to estimate the odds ratio (OR) and pooled C-statistic, respectively, and 95% confidence intervals (CI) (PROSPERO: CRD42022349764).
RESULTS: Of 5,653 records identified, we included 28 studies of 678,837 type 1 or 2 diabetes participants, of which 38,579 (5.7%) had DR. A total of 19, 3, and 7 studies were at high, unclear, and low risk of bias, respectively. Drugs included in models as predictors were: insulin (n = 24), antihypertensives (n = 5), oral antidiabetics (n = 12), lipid-lowering drugs (n = 7), antiplatelets (n = 2). Drug exposure was modelled primarily as a categorical variable (n = 23 studies). Two studies handled drug exposure as time-varying covariates, and one as a time-dependent covariate. Insulin was associated with an increased risk of DR (OR = 2.50; 95% CI: 1.61-3.86). Models that included insulin (n = 9) had a higher pooled C-statistic (C-statistic = 0.84, CI: 0.80-0.88), compared to models (n = 9) that incorporated a combination of drugs alongside insulin (C-statistic = 0.79, CI: 0.74-0.84), as well as models (n = 3) not including insulin (C-statistic = 0.70, CI: 0.64-0.75). Limitations include the high risk of bias and significant heterogeneity in reviewed studies.
CONCLUSIONS: This is the first review assessing drug exposure handling in DR prediction models. Drug exposure was primarily modelled as a categorical variable, with insulin associated with improved model performance. However, due to suboptimal drug handling, associations between other drugs and model performance may have been overlooked. This review proposes the following for future DR prediction models: (1) evaluation of drug exposure as a variable, (2) use of time-varying methodologies, and (3) consideration of drug regimen details. Improving drug exposure handling could potentially unveil novel variables capable of significantly enhancing the predictive capability of prediction models.