Application of net reclassification index to non-nested and point-based risk prediction models: a review

Machine Learning in Hypertrophic Cardiomyopathy: Nonlinear Model From Clinical and CMR Features Predicting Cardiovascular Events

BACKGROUND

The cumulative burden of hypertrophic cardiomyopathy (HCM) is significant, with a noteworthy percentage (10%-15%) of patients with HCM per year experiencing major adverse cardiovascular events (MACEs). A current risk stratification scheme for HCM had only limited accuracy in predicting sudden cardiac death (SCD) and failed to account for a broader spectrum of adverse cardiovascular events and cardiac magnetic resonance (CMR) parameters.

OBJECTIVES

This study sought to develop and evaluate a machine learning (ML) framework that integrates CMR imaging and clinical characteristics to predict MACEs in patients with HCM.

METHODS

A total of 758 patients with HCM (67% male; aged 49 ± 14 years) who were admitted between 2010 and 2017 from 4 medical centers were included. The ML model was built on the internal discovery cohort (533 patients with HCM, admitted to Fuwai Hospital, Beijing, China) by using the light gradient-boosting machine and internally evaluated using cross-validation. The external test cohort consisted of 225 patients with HCM from 3 medical centers. A total of 14 CMR imaging features (strain and late gadolinium enhancement [LGE]) and 23 clinical variables were evaluated and used to inform the ML model. MACEs included a composite of arrhythmic events, SCD, heart failure, and atrial fibrillation-related stroke.

RESULTS

MACEs occurred in 191 (25%) patients over a median follow-up period of 109.0 months (Q1-Q3: 73.0-118.8 months). Our ML model achieved areas under the curve (AUCs) of 0.830 and 0.812 (internally and externally, respectively). The model outperformed the classic HCM Risk-SCD model, with significant improvement (P < 0.001) of 22.7% in the AUC. Using the cubic spline analysis, the study showed that the extent of LGE and the impairment of global radial strain (GRS) and global circumferential strain (GCS) were nonlinearly correlated with MACEs: an elevated risk of adverse cardiovascular events was observed when these parameters reached the high enough second tertiles (11.6% for LGE, 25.8% for GRS, -17.3% for GCS).

CONCLUSIONS

ML-empowered risk stratification using CMR and clinical features enabled accurate MACE prediction beyond the classic HCM Risk-SCD model. In addition, the nonlinear correlation between CMR features (LGE and left ventricular pressure gradient) and MACEs uncovered in this study provides valuable insights for the clinical assessment and management of HCM.

Nomogram for predicting pathologic complete response to neoadjuvant chemoradiotherapy in patients with esophageal squamous cell carcinoma

PURPOSE

A pathologic complete response (pCR) to neoadjuvant chemoradiotherapy (nCRT) is seen in up to 40% of the patients with esophageal squamous cell carcinoma (ESCC). No nomogram has been constructed for the prediction of pCR for patients whose primary chemotherapy was a taxane-based regimen. The aim is to identify characteristics associated with a pCR through analyzing multiple pre- and post-nCRT variables and to develop a nomogram for the prediction of pCR for these patients by integrating clinicopathological characteristics and hematological biomarkers.

MATERIALS AND METHODS

We analyzed 293 patients with ESCC who underwent nCRT followed by esophagectomy. Clinicopathological factors, hematological parameters before nCRT, and hematotoxicity during nCRT were collected. Univariate and multivariate logistic regression analyses were performed to identify predictive factors for pCR. A nomogram model was built and evaluated for both discrimination and calibration.

RESULTS

After surgery, 37.88% of the study patients achieved pCR. Six variables were included in the nomogram: sex, cN stage, chemotherapy regimen, duration of nCRT, pre-nCRT neutrophil-to-lymphocyte ratio (NLR), and pre-nCRT platelet-to-lymphocyte ratio (PLR). The nomogram indicated good accuracy and consistency in predicting pCR, with a C-index of 0.743 (95% confidence interval: 0.686, 0.800) and a p value of 0.600 (>0.05) in the Hosmer-Lemeshow goodness-of-fit test.

CONCLUSIONS

Female, earlier cN stage, duration of nCRT (< 62 days), chemotherapy regimen of taxane plus platinum, pre-nCRT NLR (≥2.199), and pre-nCRT PLR (≥99.302) were significantly associated with a higher pCR in ESCC patients whose primary chemotherapy was a taxane-based regimen for nCRT. A nomogram was developed and internally validated, showing good accuracy and consistency.

Development and validation of survival nomograms in elder triple-negative invasive ductal breast carcinoma patients

BACKGROUND

We aimed to develop a nomogram to predict the overall survival of elderly patients with Triple-negative invasive ductal breast carcinoma (TNIDC).

RESEARCH DESIGN AND METHODS

12165 elderly patients with nonmetastatic TNIDC were retrieved from the SEER database from 2010 to 2019 and were randomly assigned to training and validation cohorts. Stepwise Cox regression analysis was used to select variables for the nomogram based on the training cohort. Univariate and multivariate Cox analyses were used to calculate the correlation between variables and prognosis of the patients. Survival analysis was performed for high- and low-risk subgroups based on risk score.

RESULTS

Eleven predictive factors were identified to construct our nomograms. Compared with the TNM stage, the discrimination of the nomogram revealed good prognostic accuracy and clinical applicability as indicated by C-index values of 0.741 (95% CI 0.728-0.754) against 0.708 (95% CI 0.694-0.721) and 0.765 (95% CI 0.747-0.783) against 0.725 (95% CI 0.705-0.744) for the training and validation cohorts, respectively. Differences in OS were also observed between the high- and low-risk groups (p < 0.001).

CONCLUSION

The proposed nomogram provides a convenient and reliable tool for individual evaluations for elderly patients with M0_stage TNIDC. However, the model may only for Americans.

Development and evaluation of nomograms and risk stratification systems to predict the overall survival and cancer-specific survival of patients with hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer, and patients with HCC have a poor prognosis and low survival rates. Establishing a prognostic nomogram is important for predicting the survival of patients with HCC, as it helps to improve the patient's prognosis. This study aimed to develop and evaluate nomograms and risk stratification to predict overall survival (OS) and cancer-specific survival (CSS) in HCC patients. Data from 10,302 patients with initially diagnosed HCC were extracted from the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2017. Patients were randomly divided into the training and validation set. Kaplan-Meier survival, LASSO regression, and Cox regression analysis were conducted to select the predictors of OS. Competing risk analysis, LASSO regression, and Cox regression analysis were conducted to select the predictors of CSS. The validation of the nomograms was performed using the concordance index (C-index), the Akaike information criterion (AIC), the Bayesian information criterion (BIC), Net Reclassification Index (NRI), Discrimination Improvement (IDI), the receiver operating characteristic (ROC) curve, calibration curves, and decision curve analyses (DCAs). The results indicated that factors including age, grade, T stage, N stage, M stage, surgery, surgery to lymph node (LN), Alpha-Fetal Protein (AFP), and tumor size were independent predictors of OS, whereas grade, T stage, surgery, AFP, tumor size, and distant lymph node metastasis were independent predictors of CSS. Based on these factors, predictive models were built and virtualized by nomograms. The C-index for predicting 1-, 3-, and 5-year OS were 0.788, 0.792, and 0.790. The C-index for predicting 1-, 3-, and 5-year CSS were 0.803, 0.808, and 0.806. AIC, BIC, NRI, and IDI suggested that nomograms had an excellent predictive performance with no significant overfitting. The calibration curves showed good consistency of OS and CSS between the actual observation and nomograms prediction, and the DCA showed great clinical usefulness of the nomograms. The risk stratification of OS and CSS was built that could perfectly classify HCC patients into three risk groups. Our study developed nomograms and a corresponding risk stratification system predicting the OS and CSS of HCC patients. These tools can assist in patient counseling and guiding treatment decision making.

Prognostic nomograms for breast cancer with lung metastasis: a SEER-based population study

BACKGROUND

Lung metastasis is a significant adverse predictor of prognosis in patients with breast cancer. Accurate estimation for the prognosis of patients with lung metastasis and population-based validation for the models are lacking. In the present study, we aimed to establish the nomogram to identify prognostic factors correlated with lung metastases and evaluate individualized survival in patients with lung metastasis based on SEER (Surveillance, Epidemiology, and End Results) database.

METHODS

We selected 1197 patients diagnosed with breast cancer with lung metastasis (BCLM) from the SEER database and randomly assigned them to the training group (n = 837) and the testing group (n = 360). Based on univariate and multivariate Cox regression analysis, we evaluated the effects of multiple variables on survival in the training group and constructed a nomogram to predict the 1-, 2-, and 3-year survival probability of patients. The nomogram were verified internally and externally by Concordance index (C-index), Net Reclassification (NRI), Integrated Discrimination Improvement (IDI), Decision Curve Analysis (DCA), and calibration plots.

RESULTS

According to the results of multi-factor Cox regression analysis, age, histopathology, grade, marital status, bone metastasis, brain metastasis, liver metastasis, human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), progesterone receptor (PR), surgery, neoadjuvant therapy and chemotherapy were considered as independent prognostic factors for patients with BCLM. The C-index in the training group was 0.719 and the testing group was 0.695, respectively. The AUC values of the 1-, 2-, and 3-year prognostic nomogram in the training group were 0.798, 0.790 and 0.793, and the corresponding AUC values in the testing group were 0.765, 0.761 and 0.722. The calculation results of IDI and NRI were shown. The nomograms significantly improved the risk reclassification for 1-, 2-, and 3-year overall mortality prediction compared with the AJCC 7th staging system. According to the calibration plot, nomograms showed good consistency between predicted and actual overall survival (OS) values for the patients with BCLM. DCA showed that nomograms had better net benefits at different threshold probabilities at different time points compared with the AJCC 7th staging system.

CONCLUSIONS

Nomograms that predicted 1-, 2-, and 3-year OS for patients with BCLM were successfully constructed and validated to help physicians in evaluating the high risk of mortality in breast cancer patients.

Development and validation of a clinical prediction model for glioma grade using machine learning

BACKGROUND

Histopathological evaluation is currently the gold standard for grading gliomas; however, this technique is invasive.

OBJECTIVE

This study aimed to develop and validate a diagnostic prediction model for glioma by employing multiple machine learning algorithms to identify risk factors associated with high-grade glioma, facilitating the prediction of glioma grading.

METHODS

Data from 1114 eligible glioma patients were obtained from The Cancer Genome Atlas (TCGA) database, which was divided into a training set (n= 781) and a test set (n= 333). Fifty machine learning algorithms were employed, and the optimal algorithm was selected to construct a prediction model. The performance of the machine learning prediction model was compared to the clinical prediction model in terms of discrimination, calibration, and clinical validity to assess the performance of the prediction model.

RESULTS

The area under the curve (AUC) values of the machine learning prediction models (training set: 0.870 vs. 0.740, test set: 0.863 vs. 0.718) were significantly improved from the clinical prediction models. Furthermore, significant improvement in discrimination was observed for the Integrated Discrimination Improvement (IDI) (training set: 0.230, test set: 0.270) and Net Reclassification Index (NRI) (training set: 0.170, test set: 0.170) from the clinical prognostic model. Both models showed a high goodness of fit and an increased net benefit.

CONCLUSION

A strong prediction accuracy model can be developed using machine learning algorithms to screen for high-grade glioma risk predictors, which can serve as a non-invasive prediction tool for preoperative diagnostic grading of glioma.

Padua prediction score may be inappropriate for VTE risk assessment in hospitalized patients with acute respiratory conditions: A Chinese single-center cohort study

Background

The Padua Prediction Score (PPS) recommended by the guidelines lacks effective external validation in a Chinese cohort. This study sought to assess the accuracy of the PPS to predict venous thromboembolism (VTE) risk in medical inpatients with acute respiratory conditions.

Methods

This consecutive cohort study included 1,574 inpatients from January to August 2019. The occurrence rate of VTE in patients classified at high-risk and low-risk groups according to PPS and Caprini risk assessment model (RAM) was compared. The discriminatory capability of the RAMs was evaluated in all the patients and the subgroup without pharmacological prophylaxis. Reclassification parameters were also used to assess the clinical utility.

Results

170 (10.8%) patients were objectively confirmed as having VTE during hospitalization. The incidence rate of VTE in low-risk patients was 6.3% by PPS, which was significantly higher than that by Caprini RAM (2.6%, p < 0.001). The area under the curve (AUC) for PPS and Caprini RAM was 0.714 (95%CI, 0.672-0.756) and 0.760 (95%CI, 0.724-0.797), respectively (p = 0.003). The AUC of Caprini RAM was larger than PPS even in subgroups without pharmacological prophylaxis (0.774 vs 0.709, p = 0.002). Compared with Caprini RAM, the net reclassification index was estimated at 0.037 (p = 0.436), and integrated discrimination improvement was 0.015 (p = 0.495) by PPS.

Conclusions

According to our cohort study, PPS may not be appropriate to predict VTE risk in hospitalized patients with acute respiratory conditions. An accurate, widely applicable, validated RAM needs to be further constructed in Chinese medical inpatients.

Association of remnant cholesterol with intracranial atherosclerosis in community-based population: The ARIC study

OBJECTIVE

To evaluate the association between remnant cholesterol (remnant-C) and intracranial atherosclerotic disease (ICAD) in the Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS).

METHODS

We studied 1,564 participants with data on lipid profiles and high-resolution vessel wall MRI (VWMRI) from the ARIC-NCS. Remnant-C was computed as total cholesterol minus high-density lipoprotein cholesterol minus low-density lipoprotein cholesterol (LDL-C). The primary outcomes were the presence of intracranial plaques and luminal stenosis. Contributors were separated into four different groups based on remnant-C (22 mg/dL) and LDL-C (100 mg/dL) levels to investigate the function of remnant-C vs. LDL-C on ICAD. Multivariable logistic regression models were utilized to estimate the correlation among the discordant/concordant remnant-C and LDL-C, and ICAD.

RESULTS

A total of 1,564 participants were included (age 76.2 ± 5.3). After multivariable adjustment, log remnant-C was correlated with greater ICAD risk [odds ratio (OR) 1.36, 95% confidence interval (CI) 1.01 to 1.83]. The lower remnant-C/higher LDL-C group and the higher remnant-C/lower LDL-C group manifested a 1.53-fold (95% CI 1.06 to 2.20) and 1.52-fold (95% CI 1.08 to 2.14) greater risk of ICAD, relative to those having lower remnant-C/low LDL-C. Additionally, remnant-C ≥ 22 mg/dL distinguished participants at a greater risk of the presence of any stenosis compared to those at lower levels, even in participants with optimal levels of LDL-C.

CONCLUSIONS

Elevated levels of remnant-C were connected to ICAD independent of LDL-C and traditional risk factors. The mechanisms of remnant-C association with ICAD probably offer insight into preventive risk-factor of ischemic stroke.

A Nomogram for Predicting Survival in Advanced Non-Small-Cell Lung Carcinoma Patients: A Population-Based Study

Non-small-cell lung cancer (NSCLC) remains the most common malignant cancer. We identified 43140 advanced NSCLC patients from the SEER database to develop and validate a new prognostic model. The prognostic performance was evaluated by P value, concordance index, net reclassification index, integrated discrimination improvement, and decision curve analysis. The following variables were contained in the final prognostic model: age, sex, race, TNM stage, and grade and treatment options. Compared to the AJCC staging system, this prognostic model is conducive to the implementation of individualized clinical treatment schemes and can be an important part of the precise medical care of NSCLC tumors.

Association of preoperative glucose concentration with mortality in patients undergoing craniotomy for brain tumor

OBJECTIVE

Hyperglycemia is associated with worse outcomes in ambulatory settings and specialized hospital settings, but there are sparse data on the importance of preoperative blood glucose measurement before brain tumor craniotomy. The authors sought to investigate the association between preoperative glucose level and 30-day mortality rate in patients undergoing brain tumor resection.

METHODS

This retrospective cohort study included patients undergoing craniotomy for brain tumors at West China Hospital, Sichuan University, from January 2011 to March 2021. Surgical mortality rates were evaluated in patients who had normal glycemia (< 5.6 mmol/L) as well as mild (5.6-6.9 mmol/L), moderate (7.0-11.0 mmol/L), and severe hyperglycemia (> 11.0 mmol/L).

RESULTS

The study included 12,281 patients who underwent tumor resection via craniotomy. The overall 30-day mortality rate was 2.0% (242/12,281), whereas the rates for normal glycemia and mild, moderate, and severe hyperglycemia were 1.5%, 2.5%, 3.8%, and 6.5%, respectively. Compared with normal glycemia, the odds of mortality at 30 days were higher in patients with mild hyperglycemia (adjusted odds ratio [OR] 1.44, 95% confidence interval [CI] 1.05-2.00), moderate hyperglycemia (OR 2.04, 95% CI 1.41-2.96), and severe hyperglycemia (OR 3.76, 95% CI 1.96-7.20; p < 0.001 for trend). When blood glucose was analyzed as a continuous variable, for each 1 mmol/L increase in blood glucose, the adjusted OR of 30-day mortality was 1.13 (95% CI 1.08-1.19). The addition of a preoperative glucose level significantly improved the area under the curve and categorical net reclassification index for prediction of mortality.

CONCLUSIONS

In patients undergoing craniotomy for brain tumors, even mild hyperglycemia was associated with an increased mortality rate, at a glucose level that was much lower than the commonly applied level.

Risk assessment for major adverse cardiovascular events after noncardiac surgery using self-reported functional capacity: international prospective cohort study

BACKGROUND

Guidelines endorse self-reported functional capacity for preoperative cardiovascular assessment, although evidence for its predictive value is inconsistent. We hypothesised that self-reported effort tolerance improves prediction of major adverse cardiovascular events (MACEs) after noncardiac surgery.

METHODS

This is an international prospective cohort study (June 2017 to April 2020) in patients undergoing elective noncardiac surgery at elevated cardiovascular risk. Exposures were (i) questionnaire-estimated effort tolerance in metabolic equivalents (METs), (ii) number of floors climbed without resting, (iii) self-perceived cardiopulmonary fitness compared with peers, and (iv) level of regularly performed physical activity. The primary endpoint was in-hospital MACE consisting of cardiovascular mortality, non-fatal cardiac arrest, acute myocardial infarction, stroke, and congestive heart failure requiring transfer to a higher unit of care or resulting in a prolongation of stay on ICU/intermediate care (≥24 h). Mixed-effects logistic regression models were calculated.

RESULTS

In this study, 274 (1.8%) of 15 406 patients experienced MACE. Loss of follow-up was 2%. All self-reported functional capacity measures were independently associated with MACE but did not improve discrimination (area under the curve of receiver operating characteristic [ROC AUC]) over an internal clinical risk model (ROC AUC_baseline 0.74 [0.71-0.77], ROC AUC_{baseline+4METs} 0.74 [0.71-0.77], ROC AUC_{baseline+floors climbed} 0.75 [0.71-0.78], AUC_{baseline+fitnessvspeers} 0.74 [0.71-0.77], and AUC_{baseline+physical activity} 0.75 [0.72-0.78]).

CONCLUSIONS

Assessment of self-reported functional capacity expressed in METs or using the other measures assessed here did not improve prognostic accuracy compared with clinical risk factors. Caution is needed in the use of self-reported functional capacity to guide clinical decisions resulting from risk assessment in patients undergoing noncardiac surgery.

CLINICAL TRIAL REGISTRATION

NCT03016936.

Systemic Immune-Inflammation Index and Long-Term Mortality in Patients with Stroke-Associated Pneumonia

Background

Systemic immune inflammation has been investigated as a prognostic marker of different diseases. This study is designed to assess the association of systemic immune-inflammation index (SII) with long-term mortality of stroke-associated pneumonia (SAP) patients.

Methods

Patients aged ≥18 years with SAP were selected from the Nanjing Stroke Registry Program in China. We retrospectively evaluated systemic immune-inflammation response with SII and pneumonia severity with the pneumonia severity index and the confusion, uremia, elevated respiratory rate, hypotension, and aged 65 years or older score. To explore the correlation between SII and mortality in SAP patients, multivariable Cox regressions and competing risk regressions were conducted. Mediation analysis was also performed to assess the role of pneumonia severity.

Results

Among 611 patients in the SAP population, death occurred in 164 patients (26.8%) during the median follow-up of 3.0 (1.2-4.6) years. In multivariate analysis, higher SII scores could predict increased mortality in patients with SAP (adjusted hazard ratio 2.061; 95% confidence interval, 1.256-3.383; P = 0.004), and the association was mediated by pneumonia severity. Moreover, adding SII to traditional models improved their predictive ability for mortality.

Conclusion

Our study displayed that SII was characterized in SAP patients with different prognoses. Elevated SII scores increased the risk of mortality. Further research is required for the clinical practice of the index among SAP patients.

Leveraging Machine Learning Techniques to Forecast Chronic Total Occlusion before Coronary Angiography

BACKGROUND

Chronic total occlusion (CTO) remains the most challenging procedure in coronary artery disease (CAD) for interventional cardiology. Although some clinical risk factors for CAD have been identified, there is no personalized prognosis test available to confidently identify patients at high or low risk for CTO CAD. This investigation aimed to use a machine learning algorithm for clinical features from clinical routine to develop a precision medicine tool to predict CTO before CAG.

METHODS

Data from 1473 CAD patients were obtained, including 1105 in the training cohort and 368 in the testing cohort. The baseline clinical characteristics were collected. Univariate and multivariate logistic regression analyses were conducted to identify independent risk factors that impact the diagnosis of CTO. A CTO predicting model was established and validated based on the independent predictors using a machine learning algorithm. The area under the curve (AUC) was used to evaluate the model.

RESULTS

The CTO prediction model was developed with the training cohort using the machine learning algorithm. Eight variables were confirmed as 'important': gender (male), neutrophil percentage (NE%), hematocrit (HCT), total cholesterol (TC), high-density lipoprotein cholesterol (HDL), ejection fraction (EF), troponin I (TnI), and N-terminal pro-B-type natriuretic peptide (NT-proBNP). The model achieved good concordance indices of 0.724 and 0.719 in the training and testing cohorts, respectively.

CONCLUSIONS

An easy-to-use tool to predict CTO in patients with CAD was developed and validated. More research with larger cohorts are warranted to improve the prediction model, which can support clinician decisions on the early discerning CTO in CAD patients.

Development and validation of novel nomograms to predict survival of patients with tongue squamous cell carcinoma

BACKGROUND

There is no unified standard to predict postoperative survival in patients with tongue squamous cell carcinoma (TSCC), hence the urgency to develop a model to accurately predict the prognosis of these patients.

AIM

To develop and validate nomograms for predicting overall survival (OS) and cancer-specific survival (CSS) of patients with TSCC.

METHODS

A cohort of 3454 patients with TSCC from the Surveillance, Epidemiology, and End Results (SEER) database was used to develop nomograms; another independent cohort of 203 patients with TSCC from the Department of Oral and Maxillofacial Surgery, First Affiliated Hospital of Zhejiang University School of Medicine, was used for external validation. Univariate and multivariate analyses were performed to identify useful variables for the development of nomograms. The calibration curve, area under the receiver operating characteristic curve (AUC) analysis, concordance index (C-index), net reclassification index (NRI), and decision curve analysis (DCA) were used to assess the calibration, discrimination ability, and clinical utility of the nomograms.

RESULTS

Eight variables were selected and used to develop nomograms for patients with TSCC. The C-index (0.741 and 0.757 for OS and CSS in the training cohort and 0.800 and 0.830 in the validation cohort, respectively) and AUC indicated that the discrimination abilities of these nomograms were acceptable. The calibration curves of OS and CSS indicated that the predicted and actual values were consistent in both the training and validation cohorts. The NRI values (training cohort: 0.493 and 0.482 for 3- and 5-year OS and 0.424 and 0.402 for 3- and 5-year CSS; validation cohort: 0.635 and 0.750 for 3- and 5-year OS and 0.354 and 0.608 for 3- and 5-year CSS, respectively) and DCA results indicated that the nomograms were significantly better than the tumor-node-metastasis staging system in predicting the prognosis of patients with TSCC.

CONCLUSION

Our nomograms can accurately predict patient prognoses and assist clinicians in improving decision-making concerning patients with TSCC in clinical practice.

Radiomics for prediction of intracerebral hemorrhage outcomes: A retrospective multicenter study

BACKGROUND

Accurate risk stratification of patients with intracerebral hemorrhage (ICH) could help refine adjuvant therapy selection and better understand the clinical course. We aimed to evaluate the value of radiomics features from hematomal and perihematomal edema areas for prognosis prediction and to develop a model combining clinical and radiomic features for accurate outcome prediction of patients with ICH.

METHODS

This multicenter study enrolled patients with ICH from January 2016 to November 2021. Their outcomes at 3 months were recorded based on the modified Rankin Scale (good, 0-3; poor, 4-6). Independent clinical and radiomic risk factors for poor outcome were identified through multivariate logistic regression analysis, and predictive models were developed. Model performance and clinical utility were evaluated in both internal and external cohorts.

RESULTS

Among the 1098 ICH patients evaluated (mean age, 60 ± 13 years), 703 (64 %) had poor outcomes. Age, hemorrhage volume and location, and Glasgow Coma Scale (GCS) were independently associated with outcomes. The area under the receiver operating characteristic curve (AUC) of the clinical model was 0.881 in the external validation cohort. Addition of the Rad-score (combined hematoma and perihematomal edema area) improved predictive accuracy and model performance (AUC, 0.893), net reclassification improvement, 0.140 (P < 0.001), and integrated discrimination improvement, 0.050 (P < 0.001).

CONCLUSIONS

The radiomics features of hematomal and perihematomal edema area have additional value in prognostic prediction; moreover, addition of radiomic features significantly improves model accuracy.

Nomogram Including Serum Ion Concentrations to Screen for New-Onset Hypertension in Rural Chinese Populations Over a Short-Term Follow-up

BACKGROUND

This study aimed to establish a clinically useful nomogram to evaluate the probability of hypertension onset in the Chinese population.

METHODS AND RESULTS

A prospective cohort study was conducted in 2012-2013 and followed up in 2015 to identify new-onset hypertension in 4,123 participants. The dataset was divided into development (n=2,748) and verification (n=1,375) cohorts. After screening risk factors by lasso regression, a multivariate Cox regression risk model and nomogram were established. Among the 4,123 participants, 818 (19.8%) developed hypertension. The model identified 10 risk factors: age, waist-to-hip ratio, systolic blood pressure, diastolic blood pressure, high pulse rate, history of diabetes, family history of hypertension and stroke, intake frequency of bean products, and intensity of physical labor. The C-indices of the model in the development and validation cohorts were 0.744 and 0.768, respectively. After the inclusion of serum calcium and magnesium concentrations, the C-indices in the development and validation cohorts were 0.764 and 0.791, respectively, with areas under the curve for the updated model of 0.907 and 0.917, respectively. The calibration curve showed that the nomogram accurately predicted the probability of hypertension. The updated nomogram was clinically beneficial across thresholds of 10-60%.

CONCLUSIONS

The newly developed nomogram has good predictive ability and may effectively assess hypertension risk in high-risk rural areas in China.

Cohen's Kappa Coefficient as a Measure to Assess Classification Improvement following the Addition of a New Marker to a Regression Model

The need to search for new measures describing the classification of a logistic regression model stems from the difficulty in searching for previously unknown factors that predict the occurrence of a disease. A classification quality assessment can be performed by testing the change in the area under the receiver operating characteristic curve (AUC). Another approach is to use the Net Reclassification Improvement (NRI), which is based on a comparison between the predicted risk, determined on the basis of the basic model, and the predicted risk that comes from the model enriched with an additional factor. In this paper, we draw attention to Cohen's Kappa coefficient, which examines the actual agreement in the correction of a random agreement. We proposed to extend this coefficient so that it may be used to detect the quality of a logistic regression model reclassification. The results provided by Kappa's reclassification were compared with the results obtained using NRI. The random variables' distribution attached to the model on the classification change, measured by NRI, Kappa, and AUC, was presented. A simulation study was conducted on the basis of a cohort containing 3971 Poles obtained during the implementation of a lower limb atherosclerosis prevention program.

Association of inflammatory markers with cerebral small vessel disease in community-based population

Background

This study investigated the relationships of neutrophil count (NC), neutrophil-to-lymphocyte ratio (NLR) and systemic immune-inflammation index (SII) with cerebral small vessel disease (CSVD).

Methods

A total of 3052 community-dwelling residents from the Poly-vasculaR Evaluation for Cognitive Impairment and vaScular Events (PRECISE) study were involved in this cross-sectional study. CSVD burden and imaging markers, including white matter hyperintensity (WMH), lacunes, cerebral microbleeds (CMBs) and enlarged perivascular spaces in basal ganglia (BG-EPVS), were assessed according to total CSVD burden score. The associations of NC, NLR and SII with CSVD and imaging markers were evaluated using logistic regression models. Furthermore, two-sample Mendelian randomization (MR) analysis was performed to investigate the genetically predicted effect of NC on CSVD. The prognostic performances of NC, NLR and SII for the presence of CSVD were assessed.

Results

At baseline, the mean age was 61.2 ± 6.7 years, and 53.5% of the participants were female. Higher NC was suggestively associated with increased total CSVD burden and modified total CSVD burden (Q4 vs. Q1: common odds ratio (cOR) 1.33, 95% CI 1.05–1.70; cOR 1.28, 95% CI 1.02–1.60) and marginally correlated with the presence of CSVD (OR 1.29, 95% CI 1.00–1.66). Furthermore, elevated NC was linked to a higher risk of lacune (OR 2.13, 95% CI 1.25–3.62) and moderate-to-severe BG-EPVS (OR 1.67, 95% CI 1.14–2.44). A greater NLR was related to moderate-to-severe BG-EPVS (OR 1.68, 95% CI 1.16–2.45). Individuals with a higher SII had an increased risk of modified WMH burden (OR 1.35, 95% CI 1.08–1.69) and moderate-to-severe BG-EPVS (OR 1.70, 95% CI 1.20–2.41). MR analysis showed that genetically predicted higher NC was associated with an increased risk of lacunar stroke (OR 1.20, 95% CI 1.04–1.39) and small vessel stroke (OR 1.21, 95% CI 1.06–1.38). The addition of NC to the basic model with traditional risk factors improved the predictive ability for the presence of CSVD, as validated by the net reclassification index and integrated discrimination index (all p < 0.05).

Conclusions

This community-based population study found a suggestive association between NC and CSVD, especially for BG-EPVS and lacune, and provided evidence supporting the prognostic significance of NC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12974-022-02468-0.

Incremental Prognostic Value of Pericoronary Adipose Tissue Thickness Measured Using Cardiac Magnetic Resonance Imaging After Revascularization in Patients With ST-Elevation Myocardial Infarction

Background and Aim

Pericoronary adipose tissue (PCAT) reflects pericoronary inflammation and is associated with coronary artery disease. We aimed to identify the association between local PCTA thickness using cardiac magnetic resonance (CMR) and prognosis of patients with ST-elevation myocardial infarction (STEMI), and to investigate the incremental prognostic value of PCAT thickness in STEMI after reperfusion.

Methods

A total of 245 patients with STEMI (mean age, 55.61 ± 10.52 years) who underwent CMR imaging within 1 week of percutaneous coronary intervention therapy and 35 matched controls (mean age, 53.89 ± 9.45 years) were enrolled. PCAT thickness indexed to body surface area at five locations, ventricular volume and function, infarct-related parameters, and global strain indices were evaluated using CMR. Associations between PCAT thickness index and 1-year major adverse cardiovascular events (MACE) after STEMI were calculated. The prognostic value of the standard model based on features of clinical and CMR and updated model including PACT thickness index were further assessed.

Results

Patients with MACE had a more significant increase in PCAT thickness index at superior interventricular groove (SIVGi) than patients without MACE. The SIVGi was significantly associated with left ventricular ejection fraction (LVEF), infarct size, and global deformation. SIVGi &gt; 4.98 mm/m2 was an independent predictor of MACE (hazard ratio, 3.2; 95% CI: 1.6–6.38; p &lt; 0.001). The updated model significantly improved the power of prediction and had better discrimination ability than that of the standard model for predicting 1-year MACE (areas under the ROC curve [AUC] = 0.8 [95% CI: 0.74–0.87] vs. AUC = 0.76 [95% CI: 0.68–0.83], p &lt; 0.05; category-free net reclassification index [cfNRI] = 0.38 [95% CI: 0.1–0.53, p = 0.01]; integrated discrimination improvement [IDI] = 0.09 [95% CI: 0.01–0.18, p = 0.02]).

Conclusions

This study demonstrated SIVGi as an independent predictor conferred incremental value over standard model based on clinical and CMR factors in 1-year MACE predictions for STEMI.

A Survival Prediction for Acute Heart Failure Patients via Web-Based Dynamic Nomogram with Internal Validation: A Prospective Cohort Study

Purpose

The current study aimed to develop a convenient and accurate prognostic dynamic nomogram model for the risk of all-cause death in acute heart failure (AHF) patients that incorporates clinical characteristics including N-terminal pro-brain natriuretic peptide (NT-pro BNP) and growth stimulation expresses gene 2 protein (ST2).

Patients and Methods

We prospectively studied 537 consecutive AHF patients and derived a clinical prediction model. The least absolute shrinkage and selection operator regression model combined with clinical characteristics were used for dimensional reduction and feature selection. Multivariate Cox proportional hazard analysis and “Dynnom” package were used to build the dynamic nomogram for prediction of 1-,2-,and 5-year overall survival for AHF. With bootstrap validation, the time-dependent concordance index (C-index) and calibration curves were used to assess predictive discrimination and accuracy. The contributions of NT-pro BNP and ST2 to the nomogram were evaluated using integrated discrimination improvement (IDI) and net reclassification improvement (NRI), while decision curve analysis (DCA) was used to assess clinical value.

Results

Patients were randomly divided into derivation (74.9%, n=402) and validation (25.1%, n=135) cohorts. Optimal independent prognostic factors for 1-,2-, and 5-year all-cause mortality were BS-ACMR (B: NT-pro BNP; S: ST2; A: age; C: complete right bundle branch block; M: mean arterial pressure; and R: red cell distribution width >14.5%); these were incorporated into the dynamic nomogram (https://bs-acmr-nom.shinyapps.io/dynnomapp/) with bootstrap validation. The C-indexes in the derivation (0.793) and validation (0.782) cohorts were consistent with comparable performance parameters. The calibration curve showed good agreement between the nomogram-predicted and actual survival. Adding NT-pro BNP and ST2 provided a significant net benefit and improved performance over other less adequate schemes in terms of DCA of survival probability compared to those neglecting either of these two factors.

Conclusion

The study constructed a dynamic BS-ACMR nomogram, which is a convenient, practical and effective clinical decision-making tool for providing accurate prognosis in AHF patients.

A nomogram integrating non-ECG factors with ECG to screen left ventricular hypertrophy among hypertensive patients from northern China

OBJECTIVE

We aimed to establish and validate a user-friendly and clinically practical nomogram for estimating the probability of echocardiographic left ventricular hypertrophy (echo-LVH) indexed to BSA among hypertensive patients from northern China.

METHODS

A total of 4954 hypertensive patients were recruited from a population-based cohort study from January 2012 to August 2013. The dataset was randomly split into two sets: training (n = 3303) and validation (n = 1651). Three nomograms were initially constructed. That is the Cornell product nomogram, the non-ECG nomogram, and the integrated nomogram which integrated non-ECG risk factors and Cornell-voltage duration product. The least absolute shrinkage and selection operator strategies were employed to screen for non-ECG features. The performance of the nomograms was evaluated using discrimination, calibration, and decision curve analysis (DCA). The net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were also calculated.

RESULTS

The AUCs, NRIs, IDIs, and DCA curves of the nomograms demonstrated that the integrated nomogram performed best among all three nomograms. The integrated nomogram incorporated age, sex, educational level, hypertension duration, SBP, DBP, eGFR, sleep duration, tea consumption, and the Cornell-voltage duration product. The AUC was 0.758 and had a good calibration (Hosmer-Lemeshow test, P = 0.73). Internal validation showed an acceptable AUC of 0.735 and good calibration was preserved (Hosmer-Lemeshow test, P = 0.19). The integrated nomogram was clinically beneficial across a range of thresholds of 10-50%.

CONCLUSION

The integrated nomogram is a convenient and reliable tool that enables early identification of hypertensive patients at high odds of LVH and can assist clinicians in their decision-making.

HFA-PEFF scores: prognostic value in heart failure with preserved left ventricular ejection fraction

BACKGROUND/AIMS

The Heart Failure Association (HFA)-PEFF score is recognized as a simple method to diagnose heart failure (HF) with preserved ejection fraction (HFpEF). This study aimed to evaluate the relationship between HFA-PEFF scores and cardiovascular outcomes in HFpEF patients.

METHODS

A total of 502 consecutive HFpEF patients were prospectively observed for up to 1,500 days. Cardiovascular outcomes were compared between two groups of patients, defined by their HFA-PEFF scores: those who scored 2-4 (the intermediate-score group) and those who scored 5-6 group (the high-score group). Overall, 236 cardiovascular events were observed during the follow-up period (median, 1,159 days).

RESULTS

Kaplan-Meier analysis showed that there were significant differences in composite cardiovascular events and HF-related events between the intermediate-score group and the high-score group (p = 0.003 and p < 0.001, respectively). Multivariate Cox proportional hazards analysis showed that the HFA-PEFF scores significantly predicted future HF-related events (hazard ratio, 1.66; 95% confidence interval [CI], 1.11 to 2.50; p = 0.014); receiver operating characteristic analysis confirmed this relationship (area under the curve, 0.633; 95% CI, 0.574 to 0.692; p < 0.001). The cutoff HFA-PEFF score for the identification of HF-related events was 4.5. Decision curve analysis revealed that combining the HFA-PEFF score with conventional prognostic factors improved the prediction of HF-related events.

CONCLUSION

HFA-PEFF scores may be useful for predicting HF-related events in HFpEF patients.

Construction of a risk assessment model of cardiovascular disease in a rural Chinese hypertensive population based on lasso-Cox analysis

Many assessments have been used to predict cardiovascular risks in the general population, but their applicability in patients with hypertension needs to be further evaluated. In the current study, a cardiovascular risk assessment model was constructed in a hypertensive population. This prospective cohort study was conducted with cardiovascular examinations in rural northeast China in 2012 and 2013, and followed up to collect cardiovascular events in 2015 and 2018. Data were derived from 4763 hypertensive patients who were free of cardiovascular disease (CVD) at baseline and completed follow‐up. After lasso regression was used to screen for risk factors of CVD at baseline, a multivariate Cox regression risk model was established and a nomogram was developed. The model was validated using an independent test set (one third of data not used for model building). Among 4763 patients, 354 (7.43%) had a cardiovascular event during a median follow‐up of 4.66 years. Nine risk factors were screened by lasso regression, including sex, age, current smoking, body mass index (BMI), history of transient ischemic attack (TIA), family history of hypertension, family history of stroke, physical labor intensity, and high low‐density lipoprotein cholesterol (LDL‐C). The c‐index of the CVD model was 0.707, and that of an updated model with baseline blood pressure was 0.732. In the validated cohort the respective c‐indexes were 0.665 and 0.714. An assessment model of CVD risk was established in a hypertensive population which may provide an original prevention strategy for hypertensive populations in rural China, and further reduce the CVD burden.

A machine learning–based clinical decision support algorithm for reducing unnecessary coronary angiograms

Background

Conventional clinical risk scores and diagnostic algorithms are proving to be suboptimal in the prediction of obstructive coronary artery disease, contributing to the low diagnostic yield of invasive angiography. Machine learning could help better predict which patients would benefit from invasive angiography vs other noninvasive diagnostic modalities.

Objective

To reduce patient risk and cost to the healthcare system by improving the diagnostic yield of invasive coronary angiography through optimized outpatient selection.

Methods

Retrospective analysis of 12 years of referral data from a provincial cardiac registry, including all patients referred for invasive angiography of more than 1.4 million individuals in Ontario, Canada. Stable outpatients undergoing coronary angiography during the study period were included in the analysis. The training set (80% random sample, n = 23,750) was used to develop 8 prediction models in Python using grid-search cross-validation. The test set (20% random sample, n = 5938), evaluated the discrimination performance of each model.

Results

The machine-learning model achieved a substantially better performance (area under the receiver operating characteristics curve: 0.81) than existing models for predicting obstructive coronary artery disease in patients referred for invasive angiography. It significantly outperformed both the reference model and current clinical practice with a net reclassification index of 27.8% (95% confidence interval [CI]: [24.9%–30.8%], P value <.01) and 44.7% (95% CI: [42.4%–47.0%], P value <.01), respectively.

Conclusion

This prediction model, when coupled with a point-of-care, online decision support tool to be used by referring physicians, could improve the diagnostic yield of invasive coronary angiography in stable, elective outpatients, thus improving patient safety and reducing healthcare costs.

Sex-Specific Metabolic Pathways Were Associated with Alzheimer's Disease (AD) Endophenotypes in the European Medical Information Framework for AD Multimodal Biomarker Discovery Cohort

BACKGROUND

physiological differences between males and females could contribute to the development of Alzheimer's Disease (AD). Here, we examined metabolic pathways that may lead to precision medicine initiatives.

METHODS

We explored whether sex modifies the association of 540 plasma metabolites with AD endophenotypes including diagnosis, cerebrospinal fluid (CSF) biomarkers, brain imaging, and cognition using regression analyses for 695 participants (377 females), followed by sex-specific pathway overrepresentation analyses, APOE ε4 stratification and assessment of metabolites' discriminatory performance in AD.

RESULTS

In females with AD, vanillylmandelate (tyrosine pathway) was increased and tryptophan betaine (tryptophan pathway) was decreased. The inclusion of these two metabolites (area under curve (AUC) = 0.83, standard error (SE) = 0.029) to a baseline model (covariates + CSF biomarkers, AUC = 0.92, SE = 0.019) resulted in a significantly higher AUC of 0.96 (SE = 0.012). Kynurenate was decreased in males with AD (AUC = 0.679, SE = 0.046).

CONCLUSIONS

metabolic sex-specific differences were reported, covering neurotransmission and inflammation pathways with AD endophenotypes. Two metabolites, in pathways related to dopamine and serotonin, were associated to females, paving the way to personalised treatment.

Multiple Blood Biomarkers and Stroke Risk in Atrial Fibrillation: The REGARDS Study

Background

Atrial fibrillation is associated with increased stroke risk; available risk prediction tools have modest accuracy. We hypothesized that circulating stroke risk biomarkers may improve stroke risk prediction in atrial fibrillation.

Methods and Results

The REGARDS (Reasons for Geographic and Racial Differences in Stroke) study is a prospective cohort study of 30 239 Black and White adults age ≥45 years. A nested study of stroke cases and a random sample of the cohort included 175 participants (63% women, 37% Black adults) with baseline atrial fibrillation and available blood biomarker data. There were 81 ischemic strokes over 5.2 years in these participants. Adjusted for demographics, stroke risk factors, and warfarin use, the following biomarkers were associated with stroke risk (hazard ratio [HR]; 95% CI for upper versus lower tertile): cystatin C (3.16; 1.04–9.58), factor VIII antigen (2.77; 1.03–7.48), interleukin‐6 (9.35; 1.95–44.78), and NT‐proBNP (N‐terminal B‐type natriuretic peptide) (4.21; 1.24–14.29). A multimarker risk score based on the number of blood biomarkers in the highest tertile was developed; adjusted HRs of stroke for 1, 2, and 3+ elevated blood biomarkers, compared with none, were 1.75 (0.57–5.40), 4.97 (1.20–20.5), and 9.51 (2.22–40.8), respectively. Incorporating the multimarker risk score to the CHA₂DS₂VASc score resulted in a net reclassification improvement of 0.34 (95% CI, 0.04–0.65).

Conclusions

Findings in this biracial cohort suggested the possibility of substantial improvement in stroke risk prediction in atrial fibrillation using blood biomarkers or a multimarker risk score.

Free Hemoglobin Ratio as a Novel Biomarker of Acute Kidney Injury After On-Pump Cardiac Surgery: Secondary Analysis of a Randomized Controlled Trial

BACKGROUND

Cardiac surgery with cardiopulmonary bypass (CPB) is associated with a high risk of postoperative acute kidney injury (AKI). Due to limitations of current diagnostic strategies, we sought to determine whether free hemoglobin (fHb) ratio (ie, levels of fHb at the end of CPB divided by baseline fHb) could predict AKI after on-pump cardiac surgery.

METHODS

This is a secondary analysis of a randomized controlled trial comparing the effect of nitric oxide (intervention) versus nitrogen (control) on AKI after cardiac surgery (NCT01802619). A total of 110 adult patients in the control arm were included. First, we determined whether fHb ratio was associated with AKI via multivariable analysis. Second, we verified whether fHb ratio could predict AKI and incorporation of fHb ratio could improve predictive performance at an early stage, compared with prediction using urinary biomarkers alone. We conducted restricted cubic spline in logistic regression for model development. We determined the predictive performance, including area under the receiver-operating-characteristics curve (AUC) and calibration (calibration plot and accuracy, ie, number of correct predictions divided by total number of predictions). We also used AUC test, likelihood ratio test, and net reclassification index (NRI) to compare the predictive performance between competing models (ie, fHb ratio versus neutrophil gelatinase-associated lipocalin [NGAL], N-acetyl-β-d-glucosaminidase [NAG], and kidney injury molecule-1 [KIM-1], respectively, and incorporation of fHb ratio with NGAL, NAG, and KIM-1 versus urinary biomarkers alone), if applicable.

RESULTS

Data stratified by median fHb ratio showed that subjects with an fHb ratio >2.23 presented higher incidence of AKI (80.0% vs 49.1%; P = .001), more need of renal replacement therapy (10.9% vs 0%; P = .036), and higher in-hospital mortality (10.9% vs 0%; P = .036) than subjects with an fHb ratio ≤2.23. fHb ratio was associated with AKI after adjustment for preestablished factors. fHb ratio outperformed urinary biomarkers with the highest AUC of 0.704 (95% confidence interval [CI], 0.592-0.804) and accuracy of 0.714 (95% CI, 0.579-0.804). Incorporation of fHb ratio achieved better discrimination (AUC test, P = .012), calibration (likelihood ratio test, P < .001; accuracy, 0.740 [95% CI, 0.617-0.832] vs 0.632 [95% CI, 0.477-0.748]), and significant prediction increment (NRI, 0.638; 95% CI, 0.269-1.008; P < .001) at an early stage, compared with prediction using urinary biomarkers alone.

CONCLUSIONS

Results from this exploratory, hypothesis-generating retrospective, observational study shows that fHb ratio at the end of CPB might be used as a novel, widely applicable biomarker for AKI. The use of fHb ratio might help for an early detection of AKI, compared with prediction based only on urinary biomarkers.

Key Experimental Factors of Machine Learning-Based Identification of Surgery Cancellations

This study aimed to provide effective methods for the identification of surgeries with high cancellation risk based on machine learning models and analyze the key factors that affect the identification performance. The data covered the period from January 1, 2013, to December 31, 2014, at West China Hospital in China, which focus on elective urologic surgeries. All surgeries were scheduled one day in advance, and all cancellations were of institutional resource- and capacity-related types. Feature selection strategies, machine learning models, and sampling methods are the most discussed topic in general machine learning researches and have a direct impact on the performance of machine learning models. Hence, they were considered to systematically generate complete schemes in machine learning-based identification of surgery cancellations. The results proved the feasibility and robustness of identifying surgeries with high cancellation risk, with the considerable maximum of area under the curve (AUC) (0.7199) for random forest model with original sampling using backward selection strategy. In addition, one-side Delong test and sum of square error analysis were conducted to measure the effects of feature selection strategy, machine learning model, and sampling method on the identification of surgeries with high cancellation risk, and the selection of machine learning model was identified as the key factors that affect the identification of surgeries with high cancellation risk. This study offers methodology and insights for identifying the key experimental factors for identifying surgery cancellations, and it is helpful to further research on machine learning-based identification of surgeries with high cancellation risk.

Prediction of Cardiovascular Events by Type I Central Systolic Blood Pressure: A Prospective Study

Supplemental Digital Content is available in the text.

Compared with brachial blood pressure (BP), central systolic BP (SBP) can provide a better indication of the hemodynamic strain inflicted on target organs, but it is unclear whether this translates into improved cardiovascular risk stratification. We aimed to assess which of central or brachial BP best predicts cardiovascular risk and to identify the central SBP threshold associated with increased risk of future cardiovascular events. This study included 13 461 participants of CARTaGENE with available central BP and follow-up data from administrative databases but without cardiovascular disease or antihypertensive medication. Central BP was estimated by radial artery tonometry, calibrated for brachial SBP and diastolic BP (type I), and a generalized transfer function (SphygmoCor). The outcome was major adverse cardiovascular events. Cox proportional-hazards models, differences in areas under the curves, net reclassification indices, and integrated discrimination indices were calculated. Youden index was used to identify SBP thresholds. Over a median follow-up of 8.75 years, 1327 major adverse cardiovascular events occurred. The differences in areas under the curves, net reclassification indices, and integrated discrimination indices were of 0.2% ([95% CI, 0.1–0.3] P<0.01), 0.11 ([95% CI, 0.03–0.20] P=0.01), and 0.0004 ([95% CI, −0.0001 to 0.0014] P=0.3), all likely not clinically significant. Central and brachial SBPs of 112 mm Hg (95% CI, 111.2–114.1) and 121 mm Hg (95% CI, 120.2–121.9) were identified as optimal BP thresholds. In conclusion, central BP measured with a type I device is statistically but likely not clinically superior to brachial BP in a general population without prior cardiovascular disease. Based on the risk of major adverse cardiovascular events, the optimal type I central SBP appears to be 112 mm Hg.

Predicting the risk for major bleeding in elderly patients with venous thromboembolism using the Charlson index. Findings from the RIETE

Old patients receiving anticoagulant therapy for venous thromboembolism (VTE) are at an increased risk for bleeding. We used data from the RIETE registry to assess the prognostic ability of the Comorbidity Charlson Index (CCI) to predict the risk for major bleeding in patients aged > 75 years receiving anticoagulation for VTE beyond the third month. We calculated the area under the receiver-operating characteristic curve (AUC), the category-based net reclassification index (NRI) and the net benefit (NB). We included 4303 patients with a median follow-up of 706 days (interquartile range [IQR] 462-1101). Of these, 147 (3%) developed major bleeding (27 died of bleeding). The AUC was 0.569 (95% CI 0.524-0.614). Patients with CCI ≤ 4 points were at a lower risk for adverse outcomes than those with CCI > 10 (major bleeding 0.81 (95% CI 0.53-1.19) vs. 2.21 (95% CI 1.18-3.79) per 100 patient-years; p < 0.05; all-cause death 1.9 (95% CI 1.45-2.44) vs. 15.67 (95% CI 12.63-19.22) per 100 patient-years; p < 0.05). A cut-off point of 4 points (CCI4) had a sensitivity of 82% (95% CI 75-89) and a specificity of 30% (95% CI 29-31) to predict major bleeding beyond the third month. CCI4 reclassification improved the NB of the RIETE bleeding score to predict bleeding beyond the third month (CCI4 NB 1.78% vs. RIETE NB 0.44%). Although the AUC of the CCI to predict major bleeding was modest, it could become an additional help to select patients aged > 75 years that obtain more benefit of extended anticoagulation, due to a lower risk for bleeding and better survival.

Nomograms predicting long-term survival in patients with invasive intraductal papillary mucinous neoplasms of the pancreas: A population-based study

BACKGROUND

There are few effective tools to predict survival in patients with invasive intraductal papillary mucinous neoplasms of the pancreas.

AIM

To develop comprehensive nomograms to individually estimate the survival outcome of patients with invasive intraductal papillary mucinous neoplasms of the pancreas.

METHODS

Data of 1219 patients with invasive intraductal papillary mucinous neoplasms after resection were extracted from the Surveillance, Epidemiology, and End Results database, and randomly divided into the training (n = 853) and the validation (n = 366) cohorts. Based on the Cox regression model, nomograms were constructed to predict overall survival and cancer-specific survival for an individual patient. The performance of the nomograms was measured according to discrimination, calibration, and clinical utility. Moreover, we compared the predictive accuracy of the nomograms with that of the traditional staging system.

RESULTS

In the training cohort, age, marital status, histological type, T stage, N stage, M stage, and chemotherapy were selected to construct nomograms. Compared with the American Joint Committee on Cancer 7^th staging system, the nomograms were generally more discriminative. The nomograms passed the calibration steps by showing high consistency between actual probability and nomogram prediction. Categorial net classification improvements and integrated discrimination improvements suggested that the predictive accuracy of the nomograms exceeded that of the American Joint Committee on Cancer staging system. With respect to decision curve analyses, the nomograms exhibited more preferable net benefit gains than the staging system across a wide range of threshold probabilities.

CONCLUSION

The nomograms show improved predictive accuracy, discrimination capability, and clinical utility, which can be used as reliable tools for risk classification and treatment recommendations.

GARFIELD-AF model for prediction of stroke and major bleeding in atrial fibrillation: a Danish nationwide validation study

OBJECTIVES

To externally validate the accuracy of the Global Anticoagulant Registry in the FIELD-Atrial Fibrillation (GARFIELD-AF) model against existing risk scores for stroke and major bleeding risk in patients with non-valvular AF in a population-based cohort.

DESIGN

Retrospective cohort study.

SETTING

Danish nationwide registries.

PARTICIPANTS

90 693 patients with newly diagnosed non-valvular AF were included between 2010 and 2016, with follow-up censored at 1 year.

PRIMARY AND SECONDARY OUTCOME MEASURES

External validation was performed using discrimination and calibration plots. C-statistics were compared with CHA₂DS₂VASc score for ischaemic stroke/systemic embolism (SE) and HAS-BLED score for major bleeding/haemorrhagic stroke outcomes.

RESULTS

Of the 90 693 included, 51 180 patients received oral anticoagulants (OAC). Overall median age (Q1, Q3) were 75 (66-83) years and 48 486 (53.5%) were male. At 1-year follow-up, a total of 2094 (2.3%) strokes/SE, 2642 (2.9%) major bleedings and 10 915 (12.0%) deaths occurred. The GARFIELD-AF model was well calibrated with the predicted risk for stroke/SE and major bleeding. The discriminatory value of GARFIELD-AF risk model was superior to CHA₂DS₂VASc for predicting stroke in the overall cohort (C-index: 0.71, 95% CI: 0.70 to 0.72 vs C-index: 0.67, 95% CI: 0.66 to 0.68, p<0.001) as well as in low-risk patients (C-index: 0.64, 95% CI: 0.59 to 0.69 vs C-index: 0.57, 95% CI: 0.53 to 0.61, p=0.007). The GARFIELD-AF model was comparable to HAS-BLED in predicting the risk of major bleeding in patients on OAC therapy (C-index: 0.64, 95% CI: 0.63 to 0.66 vs C-index: 0.64, 95% CI: 0.63 to 0.65, p=0.60).

CONCLUSION

In a nationwide Danish cohort with non-valvular AF, the GARFIELD-AF model adequately predicted the risk of ischaemic stroke/SE and major bleeding. Our external validation confirms that the GARFIELD-AF model was superior to CHA₂DS₂VASc in predicting stroke/SE and comparable with HAS-BLED for predicting major bleeding.