Machine learning prediction model of acute kidney injury after percutaneous coronary intervention

Acute kidney injury (AKI) after percutaneous coronary intervention (PCI) is associated with a significant risk of morbidity and mortality. The traditional risk model provided by the National Cardiovascular Data Registry (NCDR) is useful for predicting the preprocedural risk of AKI, although the scoring system requires a number of clinical contents. We sought to examine whether machine learning (ML) techniques could predict AKI with fewer NCDR-AKI risk model variables within a comparable PCI database in Japan. We evaluated 19,222 consecutive patients undergoing PCI between 2008 and 2019 in a Japanese multicenter registry. AKI was defined as an absolute or a relative increase in serum creatinine of 0.3 mg/dL or 50%. The data were split into training (N = 16,644; 2008–2017) and testing datasets (N = 2578; 2017–2019). The area under the curve (AUC) was calculated using the light gradient boosting model (GBM) with selected variables by Lasso and SHapley Additive exPlanations (SHAP) methods among 12 traditional variables, excluding the use of an intra-aortic balloon pump, since its use was considered operator-dependent. The incidence of AKI was 9.4% in the cohort. Lasso and SHAP methods demonstrated that seven variables (age, eGFR, preprocedural hemoglobin, ST-elevation myocardial infarction, non-ST-elevation myocardial infarction/unstable angina, heart failure symptoms, and cardiogenic shock) were pertinent. AUC calculated by the light GBM with seven variables had a performance similar to that of the conventional logistic regression prediction model that included 12 variables (light GBM, AUC [training/testing datasets]: 0.779/0.772; logistic regression, AUC [training/testing datasets]: 0.797/0.755). The AKI risk model after PCI using ML enabled adequate risk quantification with fewer variables. ML techniques may aid in enhancing the international use of validated risk models.

Trends in Ischemic Mitral Regurgitation Following ST-Elevation Myocardial Infarction Over a 20-Year Period

Background: Ischemic mitral regurgitation (IMR) is a common complication of acute ST-elevation myocardial infarction (STEMI). Little is known regarding the impact of IMR over a long period of follow up.Methods: Of 3,208 consecutive STEMI patients from a prospective registry, full echocardiographic information was available for 2,985 patients between the years 2000 and 2020. We compared the two decades- 2001 to 2010 and 2011 to 2020, and assessed for the presence of IMR at baseline, 3 (range 2–6) months and 12 (range 10–14) months after the index event.Results: One thousand six hundred and sixty six patients were included in the first decade, 1,319 in the second. Mean patient age was 61.3 ± 12.3 years, 21.1% female patients in the first decade vs. 60.9 ± 12.0 years and 22.2% female in the second (p = 0.40 and p = 0.212, respectively). Rates of moderate IMR or above during the index admission were 17.2% in the first period and 9.3% in the second one (p < 0.001). After 3 months, the rate of IMR was 48.5% for those who suffered from IMR at baseline, vs. 9.5% for those without IMR at baseline (HR- 4.2, p < 0.001). Death rates for those with moderate IMR or above were 14.7% and 17.8% after 1 and 2 years, respectively, vs. 7.3 and 9.6% for those without (p < 0.001 for both). IMR was associated with 1 year mortality in multivariate analysis (HR-1.37; 1.09–2.20, p = 0.009), as well as in propensity score matched analysis (HR 1.29; CI: 1.07–1.91; p < 0.001).Conclusions: IMR is a common complication following acute STEMI, impacting prognosis. Rates of IMR have declined significantly over the years.

Pericoronary Fat Attenuation Index Is Associated With Vulnerable Plaque Components and Local Immune‐Inflammatory Activation in Patients With Non‐ST Elevation Acute Coronary Syndrome

Background The pericoronary fat attenuation index (FAI) is assessed using standard coronary computed tomography angiography, and it has emerged as a novel imaging biomarker of coronary inflammation. The present study assessed whether increased pericoronary FAI values on coronary computed tomography angiography were associated with vulnerable plaque components and their intracellular cytokine levels in patients with non‐ST elevation acute coronary syndrome. Methods and Results A total of 195 lesions in 130 patients with non‐ST elevation acute coronary syndrome were prospectively included. Lesion‐specific pericoronary FAI, plaque components and other plaque features were evaluated by coronary computed tomography angiography. Local T cell subsets and their intracellular cytokine levels were detected by flow cytometry. Lesions with pericoronary FAI values >−70.1 Hounsfield units exhibited spotty calcification (43.1% versus 25.0%, P =0.015) and low‐attenuation plaques (17.6% versus 4.2%, P =0.016) more frequently than lesions with lower pericoronary FAI values. Further quantitative plaque compositional analysis showed that increased necrotic core volume (Pearson’s r=0.324, P <0.001) and fibrofatty volume (Pearson’s r=0.270, P <0.001) were positively associated with the pericoronary FAI, and fibrous volume (Pearson’s r=−0.333, P <0.001) showed a negative association. An increasing proinflammatory intracellular cytokine profile was found in lesions with higher pericoronary FAI values. Conclusions The pericoronary FAI may be a reliable indicator of local immune‐inflammatory response activation, which is closely related to plaque vulnerability. Registration URL: https://www.clinicaltrials.gov ; Unique identifier: NCT04792047.

Characteristics and Outcomes of MI with Non-obstructive Coronary Arteries in a South-east Asian Cohort

Background: MI with non-obstructive coronary arteries (MINOCA) is caused by a heterogenous group of conditions with clinically significant sequelae. Aim: This study aimed to compare the clinical characteristics and prognosis of MINOCA with MI with obstructive coronary artery disease (MICAD). Methods: Data on patients with a first presentation of MI between 2011 and 2014 were extracted from the Singapore Cardiac Longitudinal Outcomes Database and patients were classified as having either MINOCA or MICAD. The primary outcomes were all-cause mortality (ACM) and major adverse cardiac events (MACE), defined as a composite of ACM, recurrent MI, heart failure hospitalisation and stroke. Results: Of the 4,124 patients who were included in this study, 159 (3.9%) were diagnosed with MINOCA. They were more likely to be women, present with a non-ST-elevation MI, have a higher left ventricular ejection fraction and less likely to have diabetes, previous stroke or smoking history. Over a mean follow-up duration of 4.5 years, MINOCA patients had a lower incidence of ACM (10.1% versus 16.5%) and MACE (20.8% versus 35.5%) compared with MICAD. On multivariable analysis, patients with MINOCA had a lower risk of ACM (HR 0.42; 95% CI [0.21–0.82]) and MACE (HR 0.42; 95% CI [0.26–0.69]). Within the MINOCA group, older age, higher creatinine, a ST-elevation MI presentation, and the absence of antiplatelet use predicted ACM and MACE. Conclusion: While patients with MINOCA had better clinical outcomes compared with MICAD patients, MINOCA is not a benign entity, with one in five patients experiencing an adverse cardiovascular event in the long term.

Exploiting the Transformation Temperature to Reform an Infolded Nitinol Self-Expanding Peripheral Stent

Purpose: Nickel-titanium (nitinol) alloys possess a special set of properties that allow for a wide range of applications. Specifically, the transformation temperature for self-expanding nitinol peripheral stents allows for easy crimping at or below room temperature and reformation at body temperature becoming superelastic. Case Report: We report the case of an elderly man with iliac stenting 1 month prior, who presented several weeks after recovering from coronavirus disease 2019 with recurrent anterior-wall ST-elevation myocardial infarction. This was complicated by deformation and infolding of the previously implanted nitinol self-expanding stent in his right common iliac artery (CIA). Understanding nitinol’s specific properties, we proceeded with rapid injections of iced saline to cool the nitinol stent to its transformation temperature while nudging the distal end of the stent with a partially inflated balloon. This maneuver softened the nitinol stent, allowing us to “unfold” and reappose it against the wall of the right CIA, resulting in successful restoration of the original shape of the nitinol self-expanding stent. Conclusion: This represents the first reported case describing treatment of an infolded nitinol self-expanding peripheral stent by exploiting the transformation temperature of nitinol using iced cold saline to successfully restore the stent’s original shape and structure.

The utility of SYNTAX score predictability by electrocardiogram parameters in patients with unstable angina

Background SYNTAX score is one of the risk assessment systems to predict cardiac events in acute coronary syndrome patients. Despite the large number of SYNTAX score benefits, invasive methods such as coronary angiography are necessary to perform the scoring. We hypothesized that ECG parameters could predict the SYNTAX score in unstable angina patients. Methods During the retrospective cohort study, a total number of 876 patients were diagnosed with unstable angina. After applying the exclusion criteria, 600 patients were divided into tertiles based on the SYNTAX scores as low (0–22), intermediate (23–32), and high (≥ 33). The association between ECG parameters and SYNTAX score was investigated. Results The study included 65% men and 35% women with a mean age of 62.4 ± 9.97 years. The delayed transition zone of QRS complex, ST-depression in inferior-lateral territories or/and in all three territories, and T-wave inversion in lateral territory were significant (p < 0.05) independent predictors of intermediate SYNTAX score. High SYNTAX score was predicted by the presence of prolonged P wave duration, ST-depression in lateral territory or/and anterior-lateral territories, ST-elevation in aVR–III leads or/and aVR–III–V1 leads. Among those, all three territories ST-depression (AUC: 0.611, sensitivity: 75%, specificity: 51%) and aVR + III ST-elevation (AUC: 0.672, sensitivity: 50.12%, specificity: 80.50%) were the most accurate parameters to predict intermediate and high SYNTAX scores, respectively. Conclusion The present study demonstrates that accompanying the STE in the right side leads (aVR, III, V1) with ST-depression in other leads indicates the patients with high SYNTAX score; meanwhile, diffuse ST-depression without ST-elevation is a marker for intermediate SYNTAX score in unstable angina patients and can be applied for early risk stratification and intervention.