A novel machine learning model to predict high on-treatment platelet reactivity on clopidogrel in Asian patients after percutaneous coronary intervention

BACKGROUND

Various genetic and nongenetic variables influence the high on-treatment platelet reactivity (HTPR) in patients taking clopidogrel.

AIM

This study aimed to develop a novel machine learning (ML) model to predict HTPR in Chinese patients after percutaneous coronary intervention (PCI).

METHOD

This cohort study collected information on 507 patients taking clopidogrel. Data were randomly divided into a training set (90%) and a testing set (10%). Nine candidate Machine learning (ML) models and multiple logistic regression (LR) analysis were developed on the training set. Their performance was assessed according to the area under the receiver operating characteristic curve, precision, recall, F1 score, and accuracy on the test set. Model interpretations were generated using importance scores by transforming model variables into scaled features and representing in radar plots. Finally, we established a prediction platform for the prediction of HTPR.

RESULTS

A total of 461 patients (HTPR rate: 19.52%) were enrolled in building the prediction model for HTPR. The XGBoost model had an optimized performance, with an AUC of 0.82, a precision of 0.80, a recall of 0.44, an F1 score of 0.57, and an accuracy of 0.87, which was superior to those of LR. Furthermore, the XGBoost method identified 7 main predictive variables. To facilitate the application of the model, we established an XGBoost prediction platform consisting of 7 variables and all variables for the HTPR prediction.

CONCLUSION

A ML-based approach, such as XGBoost, showed optimum performance and might help predict HTPR on clopidogrel after PCI and guide clinical decision-making. Further validated studies will strengthen this finding.

Ticagrelor versus Clopidogrel in the Dual Antiplatelet Regimen for Unruptured Intracranial Aneurysm Treated with Stent-Assisted Coil Embolization: A Single-Center Cohort Study

BACKGROUND

Dual antiplatelet therapy (DAPT) of aspirin plus clopidogrel is commonly used in patients with unruptured intracranial aneurysms treated with stent-assisted coil (SAC) embolization. However, the unpredictable clopidogrel efficacy of the 5%-55% nonresponders limits its use. Ticagrelor, as a potential alternative of clopidogrel, is an antiplatelet agent with low resistance rates but uncertain efficacy and safety in these patients.

METHODS

A single-center cohort study was performed to compare the efficacy and safety of ticagrelor with clopidogrel in the DAPT regimen in patients with unruptured intracranial aneurysms and treated with SAC. The patients with clopidogrel resistance identified as inadequate adenosine diphosphate inhibition rate determined by thromboelastography were treated with ticagrelor instead, and both drugs achieved adequate suppression of platelet aggregation when stents were implanted. The occurrence of major adverse cardiovascular and cerebrovascular events (MACCE) and bleeding events was recorded through 6 months follow-up.

RESULTS

Data from 86 patients with 99 unruptured intracranial aneurysms and treated by SAC with clopidogrel were compared with those from 108 patients with 111 aneurysms and treated with ticagrelor. Neither the baseline characteristics nor the incidence of the MACCE or bleeding events differed between the groups. Ticagrelor exerted significantly higher adenosine diphosphate inhibition rate than that of the clopidogrel. Multivariable logistic regression analysis showed that the incidence of MACCE was related to hematocrit and fibrinogen levels.

CONCLUSIONS

Ticagrelor seemed to be as effective and safe as clopidogrel for SAC in unruptured intracranial aneurysms. Hematocrit and fibrinogen levels were independent risk factors for the incidence of MACCE.

Machine Learning to Dynamically Predict In-Hospital Venous Thromboembolism After Inguinal Hernia Surgery: Results From the CHAT-1 Study

BACKGROUND

The accuracy of current prediction tools for venous thromboembolism (VTE) events following hernia surgery remains insufficient for individualized patient management strategies. To address this issue, we have developed a machine learning (ML)-based model to dynamically predict in-hospital VTE in Chinese patients after hernia surgery.

METHODS

ML models for the prediction of postoperative VTE were trained on a cohort of 11 305 adult patients with hernia from the CHAT-1 trial, which included patients across 58 institutions in China. In data processing, data imputation was conducted using random forest (RF) algorithm, and balanced sampling was done by adaptive synthetic sampling algorithm. Data were split into a training cohort (80%) and internal validation cohort (20%) prior to oversampling. Clinical features available pre-operatively and postoperatively were separately selected using the Sequence Forward Selection algorithm. Nine-candidate ML models were applied to the pre-operative and combined datasets, and their performance was evaluated using various metrics, including area under the receiver operating characteristic curve (AUROC). Model interpretations were generated using importance scores, which were calculated by transforming model features into scaled variables and representing them in radar plots.

RESULTS

The modeling cohort included 2856 patients, divided into 2536 cases for derivation and 320 cases for validation. Eleven pre-operative variables and 15 combined variables were explored as predictors related to in-hospital VTE. Acceptable-performing models for pre-operative data had an AUROC ≥ 0.60, including logistic regression, support vector machine with linear kernel (SVM_Linear), attentive interpretable Tabular learning (TabNet), and RF. For combined data, logistic regression, SVM_Linear, and TabNet had better performance, with an AUROC ≥ 0.65 for each model. Based on these models, 7 pre-operative predictors and 10 combined predictors were depicted in radar plots.

CONCLUSIONS

A ML-based approach for the identification of in-hospital VTE events after hernia surgery is feasible. TabNet showed acceptable performance, and might be useful to guide clinical decision making and VTE prevention. Further validated study will strengthen this finding.

Condensation derivatives of 4-isopropylbenzaldehyde with acetophenone from the red alga Laurencia tristicha

Seven undescribed condensation derivatives of 4-isopropylbenzaldehyde with acetophenone, including one 1,3,5-trisubstituted pentane-1,5-dione, two 1,3,4,5,7-pentasubstituted heptane-1,7-diones and four 1,2,3,4,5-pentasubstituted cyclohexanols, together with two known flavonoids, were obtained from the red alga Laurencia tristicha. The relative configurations were elucidated by extensive spectroscopic data analysis of MS, 1D and 2D NMR, while the absolute configurations were determined by comparing the experimental and calculated electronic circular dichroism spectra. All the isolates were proven to be naturally occurring in the red alga by LC-MS analysis, and these 1,3,5-trisubstituted-pentane-1,5-dione, 1,3,4,5,7-pentasubstituted-heptane-1,7-diones and 1,2,3,4,5-pentasubstituted-cyclohexanols were reported from natural sources for the first time. The proposed biogenetic pathway of the isolates was also discussed.

Anticoagulation Resumption in a Patient With Mechanical Heart Valves, Antithrombin Deficiency, and Hemorrhagic Transformation Following Thrombectomy After Ischemic Stroke

Anticoagulation is essential for patients undergoing mechanical heart valve replacement; however, the timing to reinitiate the anticoagulant could be a dilemma that imposes increased risk for bleeding events in patients suffering from the life-threatening hemorrhagic transformation (HT) after ischemic stroke. Such a situation was presented in this case report. A 71-year-old woman was transferred directly to the Neurocritical Care Unit because of a HT that occurred following the mechanical thrombectomy for ischemic stroke. Since she had a history of prosthetic metallic valve replacement, how the anticoagulating therapy could balance the hemorrhagic and thrombotic risks was carefully evaluated. On day 6 after the onset of hemorrhage transformation, the laboratory results of coagulation and fibrinolysis strongly suggested thrombosis as well as antithrombin deficiency. The short-acting and titratable anticoagulant argatroban was immediately initiated at low dose, and thrombosis was temporarily terminated. On day 3 of anticoagulation resumption, argatroban was discontinued for one dose when the prothrombin time and activated partial thromboplastin time significantly prolonged after argatroban infusion. Aortic valve thrombosis was detected the next day. The anticoagulation was then strengthened by dose adjustment to keep mitral valve intact, to stabilize the aortic valve thrombosis, and to decrease the aortic flow rate. The intravenous argatroban was transited to oral warfarin before the patient was discharged. This study is the first report of administering argatroban and titrating to its appropriate dose in the patient with valve thrombosis, antithrombin deficiency, and HT after mechanical thrombectomy for acute ischemic stroke. Notably, the fluctuations argatroban brings to the coagulation test results might not be interpreted as increased bleeding risk. This case also suggested that the reported timing (day 6 to day 14 after hemorrhage) of anticoagulant resumption in primary intracerebral hemorrhage with mechanical valves might be late for some patients with HT.

Antioxidant terpenoids from the red alga Laurencia tristicha

One new laurane-type sesquiterpene, 10-hydroxyepiaplysinol (1), along with two known related sesquiterpenes, epiaplysinol (2) and ar-bisabol-9-en-7,11-diol (3), and one carotenoid metabolite, loliolide (4), were obtained from the red alga Laurencia tristicha. Their structures were elucidated based on comprehensive spectroscopic data analysis. Of the known compounds, 4 was isolated from the genus Laurencia for the first time. Additionally, the antioxidant activities of eleven laurane-type sesquiterpenes from L. tristicha were evaluated.

Diterpenes, sesquiterpenes, and a C15-acetogenin from the marine red alga Laurencia mariannensis

In addition to 10 known compounds (7- 16), one new brominated diterpene, 10-hydroxykahukuene B (1), two new sesquiterpenes, 9-deoxyelatol (2) and isodactyloxene A (3), one new brominated C 15-acetogenin, laurenmariallene (4), and two new naturally occurring halogenated sesquiterpenes (5 and 6) that were previously obtained as intermediates in a biomimetic synthetic study of rhodolaureol and rhodolauradiol have been isolated and identified from the organic extract of the marine red alga Laurencia mariannensis. The structures of these compounds were established by spectroscopic methods. The antibacterial and antifungal activities of new compounds 1-4 were evaluated.

A heterogeneous model for gas transport in carbon molecular sieves

A dual resistance model with distribution of either barrier or pore diffusional activation energy is proposed in this work for gas transport in carbon molecular sieve (CMS) micropores. This is a novel approach in which the equilibrium is homogeneous, but the kinetics is heterogeneous. The model seems to provide a possible explanation for the concentration dependence of the thermodynamically corrected barrier and pore diffusion coefficients observed in previous studies from this laboratory on gas diffusion in CMS. The energy distribution is assumed to follow the gamma distribution function. It is shown that the energy distribution model can fully capture the behavior described by the empirical model established in earlier studies to account for the concentration dependence of thermodynamically corrected barrier and pore diffusion coefficients. A methodology is proposed for extracting energy distribution parameters, and it is further shown that the extracted energy distribution parameters can effectively predict integral uptake and column breakthrough profiles over a wide range of operating pressures.