Background and Objectives: A machine learning technique for identifying hidden coronary artery disease (CAD) might be useful. We developed and validated machine learning models to predict patients with hidden CAD and assess long-term outcomes in patients with acute ischemic stroke. Methods: Multidetector coronary computed tomography was performed for patients without known history of CAD. Primary outcomes were defined as having any degree of CAD and having obstructive CAD (≥50% stenosis). Demographic variables, risk factors, laboratory results, Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification, NIH Stroke Scale score, blood pressure, and carotid artery stenosis were used to develop and validate machine learning models to predict CAD. Area under the receiver operating characteristic curves (AUC) was calculated for performance analysis, and Kaplan–Meier and Cox survival analyses of long-term outcomes were performed. Major adverse cardiovascular events (MACE) were defined as ischemic stroke, myocardial infarction, unstable angina, urgent coronary revascularization, and cardiovascular mortality. Results: Overall, 1,710 patients were included for the training dataset and 348 patients for the validation dataset. An Extreme Gradient Boosting model was developed to predict any degree of CAD, which showed an AUC of 0.763 (95% CI 0.711–0.814) on validation. A logistic regression model was used to predict obstructive CAD and had an AUC of 0.714 (95% CI 0.692–0.799). During the first 5 years of follow-up, MACE occurred more frequently when predicted of any CAD (P = 0.022) or obstructive CAD (P < 0.001). Cox proportional analysis showed that the hazard ratio of MACE was 1.5 (95% CI 1.1–2.2; P = 0.016) when predicted of any CAD, whereas it was 1.9 (95% CI 1.3–2.6; P < 0.001) for obstructive CAD. Discussion: We demonstrated that machine learning may help identify hidden CAD in patients with acute ischemic stroke. Long-term outcomes were also associated with prediction results. Classification of Evidence: This study provides Class II evidence that in patients with acute ischemic stroke with CAD risk factors but no known history of CAD, a machine learning model predicts CAD on multidetector coronary computed tomography with an AUC of 0.763 (95% CI 0.711-0.814).

Background: Periodontal disease can activate an immune process linked to systemic diseases, including atrial fibrillation. However, the relationship between periodontal disease and atrial fibrillation remains unclear. Aim: This study aimed to investigate whether changes in periodontal disease status are associated with the risk of atrial fibrillation. Methods: Using the National Health Insurance Database Korea, participants who underwent the first oral health examination in 2003 and a second examination in 2005–2006 without a history of atrial fibrillation were included. Participants were grouped according to changes in periodontal disease status during two oral examinations: periodontal disease-free, periodontal disease-recovered, periodontal disease-developed, and periodontal disease-chronic. The outcome was the incidence of atrial fibrillation. Results: The study included 1,254,515 participants, with a median follow-up of 14.3 years and 25,402 (2.02%) cases of atrial fibrillation occurred. During follow-up, the risk of atrial fibrillation was highest in the periodontal disease-chronic group, followed by the periodontal disease-developed, periodontal disease-recovered, and periodontal disease-free groups (p for trend < 0.001). Moreover, recovery from periodontal disease was associated with a reduced risk of atrial fibrillation compared to a chronic periodontal disease status (Hazard ratio: 0.97, 95% Confidence interval: 0.94—0.99, p = 0.045). The development of periodontal disease was associated with an increased risk of atrial fibrillation compared to being periodontal disease-free (Hazard ratio: 1.04, 95% Confidence interval: 1.01—1.08, p = 0.035). Conclusion: Our findings suggest that changes in periodontal disease status impact the risk of atrial fibrillation. Management of periodontal disease may help prevent atrial fibrillation.

Background: A high and low estimated glomerular filtration rate (eGFR) could affect outcomes after reperfusion therapy for ischemic stroke. This study aimed to determine whether renal function based on eGFR affects mortality risk in patients with ischemic stroke within 6 months following reperfusion therapy. Methods: This prospective registry–based cohort study included 2266 patients who received reperfusion therapy between January 2000 and September 2019 and were registered in the SECRET (Selection Criteria in Endovascular Thrombectomy and Thrombolytic Therapy) study or the Yonsei Stroke Cohort. A high and low eGFR were based on the Chronic Kidney Disease Epidemiology Collaboration equation and defined, respectively, as the 5th and 95th percentiles of age- and sex-specific eGFR. Occurrence of death within 6 months was compared among the groups according to their eGFR such as low, normal, or high eGFR. Results: Of the 2266 patients, 2051 (90.5%) had a normal eGFR, 110 (4.9%) a low eGFR, and 105 (4.6%) a high eGFR. Patients with high eGFR were younger or less likely to have hypertension, diabetes, or atrial fibrillation than the other groups. Active cancer was more prevalent in the high-eGFR group. During the 6-month follow-up, there were 24 deaths (22.9%) in the high-eGFR group, 37 (33.6%) in the low-eGFR group, and 237 (11.6%) in the normal-eGFR group. After adjusting for variables with P <0.10 in the univariable analysis, 6-month mortality was independently associated with high eGFR (hazard ratio, 2.22 [95% CI, 1.36–3.62]; P =0.001) and low eGFR (HR, 2.29 [95% CI, 1.41–3.72]; P =0.001). These associations persisted regardless of treatment modality or various baseline characteristics. Conclusions: High eGFR as well as low eGFR were independently associated with 6-month mortality after reperfusion therapy. Kidney function could be considered a prognostic factor in patients with ischemic stroke after reperfusion therapy.

BACKGROUND: We aimed to develop and validate machine learning models to diagnose patients with ischemic stroke with cancer through the analysis of histopathologic images of thrombi obtained during endovascular thrombectomy. METHODS: This was a retrospective study using a prospective multicenter registry which enrolled consecutive patients with acute ischemic stroke from South Korea who underwent endovascular thrombectomy. This study included patients admitted between July 1, 2017 and December 31, 2021 from 6 academic university hospitals. Whole-slide scanning was performed for immunohistochemically stained thrombi. Machine learning models were developed using transfer learning with image slices as input to classify patients into 2 groups: cancer group or other determined cause group. The models were developed and internally validated using thrombi from patients of the primary center, and external validation was conducted in 5 centers. The model was also applied to patients with hidden cancer who were diagnosed with cancer within 1 month of their index stroke. RESULTS: The study included 70 561 images from 182 patients in both internal and external datasets (119 patients in internal and 63 in external). Machine learning models were developed for each immunohistochemical staining using antibodies against platelets, fibrin, and erythrocytes. The platelet model demonstrated consistently high accuracy in classifying patients with cancer, with area under the receiver operating characteristic curve of 0.986 (95% CI, 0.983–0.989) during training, 0.954 (95% CI, 0.937–0.972) during internal validation, and 0.949 (95% CI, 0.891–1.000) during external validation. When applied to patients with occult cancer, the model accurately predicted the presence of cancer with high probabilities ranging from 88.5% to 99.2%. CONCLUSIONS: Machine learning models may be used for prediction of cancer as the underlying cause or detection of occult cancer, using platelet-stained immunohistochemical slide images of thrombi obtained during endovascular thrombectomy.

Proteinuria is associated with cardiovascular disease. However, the relationship between changes in proteinuria status and hypertension remains unclear. This study aimed to explore the association between changes in proteinuria status and the risk of developing hypertension with the data from the Korean National Health Insurance Database. We included participants without prior hypertension history who underwent their first health examination in 2003–2004 and a second examination in 2005–2006. Based on their proteinuria status during these two examinations, participants were classified into four groups: the proteinuria-free, proteinuria-resolved, proteinuria-developed, and chronic proteinuria groups. The study outcome was the incidence of hypertension. The study included 935,723 participants followed for a median of 14.2 years (mean age: 40.96 ± 11.01, 62.5% male participants). During this period, 346,686 (37.1%) cases of hypertension were reported. The chronic proteinuria group had the highest hypertension risk, followed by the proteinuria-developed, proteinuria-resolved, and proteinuria-free groups (p < 0.001). Those who recovered from proteinuria had a lower risk of developing hypertension than those with chronic proteinuria (hazard ratio: 0.58; 95% confidence interval: 0.53–0.63, p < 0.001). In contrast, individuals who developed proteinuria had a higher risk of hypertension than proteinuria-free individuals (hazard ratio: 1.31; 95% confidence interval: 1.26–1.35, p < 0.001). Our findings suggest a significant association between proteinuria status changes and hypertension. Effective management of proteinuria may potentially decrease the risk of developing hypertension.

Cha MJ, et al. Yonsei Med J. 2022 Nov;63(11):1016-1026. https://doi.org/10.3349/ymj.2022.0328