Biomarkers enhance the long-term predictive ability of the KAMIR risk score in Chinese patients with ST-elevation myocardial infarction

Interpretable machine learning for in-hospital mortality risk prediction in patients with ST-elevation myocardial infarction after percutaneous coronary interventions

BACKGROUND AND OBJECTIVE

Despite the constant improvement of coronary heart disease (CHD) diagnostics and treatment methods it remains one of the main causes of death in most countries around the world. And myocardial infarction with ST segment elevation on the electrocardiogram (STEMI) still is one of the most dangerous clinical variants of CHD. This study aims to develop an explainable machine learning model for in-hospital mortality (IHM) risk prediction in STEMI patients after myocardial revascularization by percutaneous coronary intervention (PCI).

METHODS

A single-center observational retrospective study was conducted, enrolling 4677 electronic medical records of patients with STEMI after PCI, which were analyzed using statistical analysis and machine learning methods. A pool of potential IHM predictors was identified, and prognostic models were developed and validated based on multivariate logistic regression, random forest, and stochastic gradient boosting methods at two stages of hospital treatment: during the initial physicians examination in the emergency department and immediately after PCI surgery. To explain the IHM prognosis, threshold values of IHM risk factors were determined using 3 grid search methods for optimal cut-off points, calculating centroids and SHapley Additive exPlanations (SHAP).

RESULTS

IHM prognostic models were developed using clinical and functional status data of STEMI patients during two stages of hospital treatment. The IHM prediction accuracy according to the first scenario was AUC = 0.85, and according to the second - AUC = 0.9. Predictors identified and validated in the models were converted into risk factors. Models whose parameters were risk factors demonstrated high forecast accuracy (AUC = 0.87), with the best model formed using the SHAP method.

CONCLUSIONS

For the forecast result interpretation risk factors obtained by categorizing continuous variables can be used by assessing the impact of the latter on the end point using the SHAP method.

Performance of the Models Predicting In-Hospital Mortality in Patients with ST-Segment Elevation Myocardial Infarction with Predictors in Categorical and Continuous Forms

The aim of the study is to assess the performance of predictive models developed on the basis of predictors in the continuous and categorical forms to predict the probability of in-hospital mortality (IHM) in patients with ST-segment elevation myocardial infarction (STEMI) after percutaneous coronary intervention (PCI).

Materials and Methods

A single-center retrospective study has been conducted, within the framework of which data from 4674 medical records of patients with STEMI after PCI, treated at the Regional Vascular Center of Vladivostok (Russia), have been analyzed. Two groups of patients were identified: group 1 consisted of 318 (6.8%) individuals who died in the hospital, group 2 included 4356 (93.2%) patients with a favorable outcome of treatment. IHM prognostic models were developed using multivariate logistic regression (MLR), random forest (RF), and stochastic gradient boosting (SGB). 6-metric qualities were used to evaluate the accuracy of the models. Threshold values of IHM predictors were determined using a grid search to find the optimal cut-off points, calculating centroids, and Shapley additive explanations. The latter helped evaluate the degree to which the model predictors influence the endpoint.

Results

Based on the results of the multi-stage analysis of indicators of clinical and functional status of the STEMI patients, new predictors of IHM have been identified and validated, complementing the factors of the GRACE scoring system, their categorization has been carried out and prognostic models with continuous and categorical variables based on MLR, RF, and SGB have been developed. These models had a high (AUC - 0.88 to 0.90) and comparable predictive accuracy, but their predictors differed in various degrees of influence on the endpoint. The comparative analysis has shown that the Shapley additive explanation method has advantages in categorizing predictors compared to other methods and allows for detailing the structure of their relationships with IHM.

Conclusion

The use of modern data mining methods, including machine learning algorithms, categorization of predictors, and assessment of the degree of their effect on the endpoint, makes it possible to develop predictive models possessing high accuracy and the properties of explanation of the generated conclusions.

Comparison of Mortality Prediction by the GRACE Score, Multiple Biomarkers, and Their Combination in All-comer Patients with Acute Myocardial Infarction Undergoing Primary Percutaneous Coronary Intervention

Objective This study examined the ability of a combination of biomarkers, including N-terminal pro-B-type natriuretic peptide (N-BNP) and high-sensitivity C-reactive protein (hs-CRP), to better predict mortality than the Global Registry of Acute Coronary Events (GRACE) score in acute myocardial infarction (AMI) patients who received primary percutaneous coronary intervention (PPCI). Methods The in-hospital mortality in 754 all-comer patients with AMI who underwent successful PPCI over 8 years was examined. A receiver operating characteristic (ROC) analysis was performed to determine the in-hospital mortality in a single center. A logistic regression analysis was used to compare the predictive accuracy of the GRACE score and biomarkers. The incremental predictive value of those biomarkers beyond the GRACE score was also examined. Results The mean age was 66±13 years old, and 609 patients with ST-elevated AMI (80.8%) were included. The in-hospital mortality was 6.8%. The GRACE score (in-hospital survivor/non-survivor: 106±33/161±32; p<0.05,) and N-BNP (in-hospital survivor/non-survivor: 2,458±7,058/8,880±1,1331 pg/mL; p<0.05) were significantly lower in survivors than in non-survivors. The area under the ROC curve (AUC) of in-hospital mortality of the GRACE score was significantly higher than that of the dual-biomarker combination (0.868/0.720; p<0.05). The AUC of the combination of the GRACE score and dual-biomarkers was not significantly higher than that of the GRACE score alone (0.870/0.868; p=0.747). Conclusion The measurement of representative cardiovascular biomarkers did not provide any additional benefit for mortality prediction beyond the GRACE score in AMI patients who received PPCI.

Systematic review and meta-analysis of prognostic models in Southeast Asian populations with acute myocardial infarction

Background

The cultural and genetic diversity of the Southeast Asian population has contributed to distinct cardiovascular disease risks, incidence, and prognosis compared to the Western population, thereby raising concerns about the accuracy of predicted risks of existing prognostic models.

Objectives

We aimed to evaluate the predictive performances of validated, recalibrated, and developed prognostic risk prediction tools used in the Southeast Asian population with acute myocardial infarction (AMI) events for secondary events

Methods

We searched MEDLINE and Cochrane Central databases until March 2022. We included prospective and retrospective cohort studies that exclusively evaluated populations in the Southeast Asian region with a confirmed diagnosis of an AMI event and evaluated for risk of secondary events such as mortality, recurrent AMI, and heart failure admission. The CHARMS and PRISMA checklists and PROBAST for risk of bias assessment were used in this review.

Results

We included 7 studies with 11 external validations, 3 recalibrations, and 3 new models from 4 countries. Both short- and long-term outcomes were assessed. Overall, we observed that the external validation studies provided a good predictive accuracy of the models in the respective populations. The pooled estimate of the C-statistic in the Southeast Asian population for GRACE risk score is 0.83 (95%CI 0.72–0.90, n = 6 validations) and for the TIMI risk score is 0.80 (95%CI: 0.772–0.83, n = 5 validations). Recalibrated and new models demonstrated marginal improvements in discriminative values. However, the method of predictive accuracy measurement in most studies was insufficient thereby contributing to the mixed accuracy effect. The evidence synthesis was limited due to the relatively low quality and heterogeneity of the available studies.

Conclusion

Both TIMI and GRACE risk scores demonstrated good predictive accuracies in the population. However, with the limited strength of evidence, these results should be interpreted with caution. Future higher-quality studies spanning various parts of the Asian region will help to understand the prognostic utility of these models better.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?%20RecordID=228486.

NT-proBNP and Major Adverse Cardiovascular Events in Patients with ST-Segment Elevation Myocardial Infarction Who Received Primary Percutaneous Coronary Intervention: A Prospective Cohort Study

Background

The prognostic significance of the amino-terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP) in patients with ST-segment elevation myocardial infarction (STEMI) after percutaneous coronary intervention (PCI) has not been fully elucidated. Major adverse cardiovascular events (MACEs) are clinically viable indicators for the accurate, rapid, and safe evaluation of patients with STEMI. This study was designed to investigate the relationship between NT-proBNP levels and the occurrence of short-term MACEs in patients with STEMI who underwent emergency PCI.

Methods

This prospective cohort study included 405 patients with STEMI aged 20–90 years who underwent emergency PCI at the First People's Hospital of Changde City from April 6, 2017, to May 31, 2019. Stent thrombosis, reinfarction, congestive heart failure, unstable angina, and cardiac death were considered as MACEs in this study. The target-independent and -dependent variables were NT-proBNP at baseline and MACE, respectively.

Results

There were 28.25% of MACEs. Age, number of implanted stents, Killip class, infarction-related artery, applied intra-aortic balloon pump (IABP), creatine kinase (CK) peak value, CK-MB peak value, TnI peak value, and ST-segment resolution were independently associated with MACE (P < 0.05). In a multivariate model, after adjusting all potential covariates, Log2 NT-proBNP levels remained significantly associated with MACE, with an inflection point of 11.66. The effect sizes and confidence intervals of the left and right sides of the inflection point were 1.07 and 0.84–1.36 (P=0.5730) and 3.47 and 2.06–5.85 (P < 0.0001), respectively.

Conclusions

In patients with STEMI who underwent PCI, Log2 NT-proBNP was positively correlated with MACE within 1 month when the Log2 NT-proBNP was >11.66 (NT-proBNP >3.236 pg/mL).

Changes of Ocular Surface Before and After Treatment of Blepharoptosis With Combined Fascial Sheath Suspension and Frontal Muscle Flap Suspension

OBJECTIVE

To explore the ocular surface changes after ptosis surgery in patients with severe congenital blepharoptosis.

METHODS

The patients were divided into group A and group B, Group A received conjoint fascial sheath suspension, and group B received frontal muscle flap suspension. The ocular surface changes were followed upon the 7th day and in the 1st and 3rd month after operation, which included Schirmer test (ST), break-up time (BUT), tear meniscus height (TMH). For normal distribution variables, t test was used before and after operation, and the Wilcoxon test was used for variables with abnormal distribution.

RESULTS

Compared to preoperative status, the ST and TMH were not significantly changed after surgery (P > 0.05), but BUT higher on the 7th day and in the 1st and 3rd month after operation (P < 0.05). The fluorescence staining (FL) score was higher in the 3rd month than that in the 1st month (P < 0.05), but was not significantly different between the 7th and in the 1st month after operation (P > 0.05). The ST, TMH, and BUT after surgery were not significantly different between groups of A and B (P > 0.05), but the FL score lower in the 3rd month than on the 7th day and in the 1st month after operation (P < 0.05).

CONCLUSIONS

The ST, BUT, and TMH were not significantly changed after surgery between groups of A and B, but the FL score was lower in group A than that in group B in early postoperative time.

Admission Hyperglycemia is Associated with Global Registry of Acute Coronary Events Score and Complications Following Acute Myocardial Infarction During 1-Year Follow-Up

We explored the effect of admission hyperglycemia (AH) on the Global Registry of Acute Coronary Events (GRACE) risk score and major adverse cardiovascular and cerebrovascular event (MACCE) incidence during 1-year follow-up in acute myocardial infarction (AMI) patients. In this retrospective observational study enrolling 1098 AMI patients, hyperglycemia was defined as blood glucose level ≥180 mg/dl at admission. Overall, 158 and 84 patients of 272 diabetic and 826 non-diabetic patients were diagnosed with AH, respectively. Glucose levels at admission were closely associated with the GRACE score in patients with/without diabetes. MACCEs occurred in 222 patients; patients with AH showed significantly higher MACCE incidence (28.1%). Multivariate Cox logistic regression analysis indicated that AH was an independent risk factor for 1-year MACCEs. Subgroup analysis demonstrated that hyperglycemia increases MACCE risk in non-diabetic patients but not in diabetic patients. The admission glucose level combined with GRACE risk score showed a certain predictive value for MACCE incidence according to ROC analysis (OR = 0.798, p < .001). AH was strongly associated with a higher GRACE risk score in ST-segment elevation myocardial infarction patients. Thus, AH was an independent risk factor and had a high predictive value for MACCE during 1-year follow-up after AMI.

Impacts of smoking status on the clinical outcomes of coronary non-target lesions in patients with coronary heart disease: a single-center angiographic study

Background

Coronary atherosclerotic plaque could go through rapid progression and induce adverse cardiac events. This study aimed to evaluate the impacts of smoking status on clinical outcomes of coronary non-target lesions.

Methods

Consecutive patients with coronary heart disease who underwent two serial coronary angiographies were included. All coronary non-target lesions were recorded at first coronary angiography and analyzed using quantitative coronary angiography at both procedures. Patients were grouped into non-smokers, quitters, and smokers according to their smoking status. Clinical outcomes including rapid lesion progression, lesion re-vascularization, and myocardial infarction were recorded at second coronary angiography. Multivariable Cox regression analysis was used to investigate the association between smoking status and clinical outcomes.

Results

A total of 1255 patients and 1670 lesions were included. Smokers were younger and more likely to be male compared with non-smokers. Increase in percent diameter stenosis was significantly lower (2.7 [0.6, 7.1] % vs. 3.5 [0.9, 8.9]%) and 3.4 [1.1, 7.7]%, P = 0.020) in quitters than those in smokers and non-smokers. Quitters tended to have a decreased incidence of rapid lesions progression (15.8% [76/482] vs. 21.6% [74/342] and 20.6% [89/431], P = 0.062), lesion re-vascularization (13.1% [63/482] vs. 15.5% [53/432] and 15.5% [67/431], P = 0.448), lesion-related myocardial infarction (0.8% [4/482] vs. 2.6% [9/342] and 1.4% [6/431], P = 0.110) and all-cause myocardial infarction (1.9% [9/482] vs. 4.1% [14/342] and 2.3% [10/431], P = 0.128) compared with smokers and non-smokers. In multivariable analysis, smoking status was not an independent predictor for rapid lesion progression, lesion re-vascularization, and lesion-related myocardial infarction except that a higher risk of all-cause myocardial infarction was observed in smokers than non-smokers (hazards ratio: 3.00, 95% confidence interval: 1.04–8.62, P = 0.042).

Conclusion

Smoking cessation mitigates the increase in percent diameter stenosis of coronary non-target lesions, meanwhile, smokers are associated with increased risk for all-cause myocardial infarction compared with non-smokers.

Prognostic value of the combination of GRACE risk score and mean platelet volume to lymphocyte count ratio in patients with ST-segment elevation myocardial infarction after percutaneous coronary intervention

The Global Registry of Acute Coronary Events (GRACE) risk score and the mean platelet volume to lymphocyte count ratio (MPVLR) can be used independently to predict adverse outcomes in patients with acute coronary syndromes. However, the level of MPVLR in relation to the GRACE score, and whether a combination of these methods can better predict the clinical adverse outcome of patients with ST-segment elevation myocardial infarction (STEMI), have not been previously examined. Therefore, the aim of the present study was to investigate whether the combination of GRACE risk score and MPVLR is a good predictor of a 30-day major adverse cardiovascular events (MACE) in patients with STEMI. A total of 464 patients with STEMI undergoing percutaneous coronary intervention (PCI) were enrolled, and divided into four groups based on the optimal cut-off values for GRACE score and MPVLR. GRACE score and MPVLR levels were separately recorded during admission. Spearman's rank correlation analysis showed a positive correlation between GRACE score and MPVLR (ρ=0.304; P<0.001). Both GRACE score [hazard ratio (HR), 1.706; 95% CI, 1.435-3.058; P<0.001] and MPVLR level (HR, 1.668; 95% CI, 1.202-2.170; P<0.001) were found to be independent predictors of a 30-day MACE. Additionally, the high MPVLR + high GRACE score group of patients had an HR of 2.455 (95% CI, 1.736-3.188) for a 30-day MACE, when using the low MPVLR + low GRACE score group as a reference. Based on the area under the curve, MPVLR combined with GRACE scores achieved an improved performance in differentiating angiographic no-reflow during a 30-day MACE, compared with individual MPVLR and GRACE scores. Therefore, the present results suggested that the GRACE score may be positively correlated with MPVLR and that their combination accurately predicted the occurrence of short-term MACE in patients with STEMI after PCI.