Machine learning performance for a small dataset: random oversampling improves data imbalances and fairness

Background Patients with prior coronary artery bypass graft (CABG) surgery often require percutaneous coronary intervention (PCI). Data are still limited in regards to the outcomes of native versus graft PCI after CABG. Methods We performed a retrospective study in a tertiary reference cardiac centre of consecutive patients who underwent PCI after CABG. The data were collected for patients who underwent either native or graft PCI from January 2008 to December 2018. Major adverse cardiac events (MACE) included death or myocardial infarction (MI) or revascularization. All outcomes were assessed at 1-year after each index procedure. Results A total of 445 PCI were performed in 410 patients (209 had native PCI and 201 had graft PCI). The groups of patients with native vessel PCI and graft PCI were statistically comparable regarding their baseline characteristics. In multivariable Cox regression graft PCI relative to native PCI was an independent risk factor for MACE (hazard ratio [HR] 1.818, 95% confidence interval [CI] 1.148–2.878). Conclusion Compared with native coronary PCI, bypass graft PCI was significantly associated with higher incidence of MACE at 1-year and this was mainly driven by MI and revascularization. MACE outcomes Funding Acknowledgement Type of funding source: None

There exist limited outcomes data for women and minorities after contemporary percutaneous coronary intervention (PCI). To examine 1-year outcomes in women and minorities vs white men after PCI with everolimus-eluting stents. The PLATINUM Diversity study was a single-arm study enrolling women and minorities. Patient-level pooling with the PROMUS Element Plus Post-Approval Study was prespecified. Data on social determinants of health and language were collected in the PLATINUM Diversity cohort, which included 1501 patients at 52 US sites. The PROMUS Element Plus Post-Approval study enrolled 2681 patients at 52 US sites with some site overlap and included an "all-comers" population. All patients were enrolled beginning in October 2014 and were followed for 12 months. Analyses began in August 2016. Patients received 1 or more everolimus-eluting stent implantation. The primary end point was 1-year major adverse cardiac events (MACE), which included death/myocardial infarction (MI)/target vessel revascularization. Secondary ischemic end points were also evaluated. The pooled study consisted of 4182 patients: 1635 white men (39.1%), 1863 women (white and minority) (44.5%), and 1059 minority patients (women and men) (25.3%). Women and minorities had a higher prevalence of diabetes, prior stroke, hypertension, renal disease, and congestive heart failure than white men but lower rates of multivessel disease, prior coronary artery bypass graft surgery, prior MI, and smoking. Unadjusted 1-year MACE rates (white men, 7.6%; women, 8.6%; minorities, 9.6%) were similar between groups with no significant differences after risk adjustment. The adjusted risk of death/MI was higher among women (odds ratio, 1.6; 95% CI, 1.1-2.4) and minorities (odds ratio, 1.9; 95% CI, 1.2-2.8) compared with white men and the adjusted risk of MI was higher in minorities (odds ratio, 2.6; 95% CI, 1.4-4.8). These differences were driven primarily by nonstent-related MIs. Within the PLATINUM Diversity cohort, the independent predictors of MACE were cardiogenic shock, renal disease, history of peripheral vascular disease, multivessel disease, widowhood, and lack of private insurance. After contemporary everolimus-eluting stent implantation, women and minorities experience a similar risk of 1-year MACE but a higher adjusted risk of recurrent ischemic events primarily because of nonstent-related MIs. Both clinical and angiographic factors and social determinants of health, including widowhood and insurance status, contribute to 1-year MACE among women and minorities.

Incidence, Range, and Clinical Effect of Hemoglobin Changes Within 24 Hours After Transradial Coronary Stenting

Objectives. This study evaluated angiographic success and in-hospital outcomes of percutaneous coronary intervention (PCI) with rotational atherectomy (RA) in patients with low left ventricular ejection fraction (LVEF). Design. Between January 2010 and March 2014, 272 consecutive patients with heavily calcified lesions underwent elective PCI with RA. Of these, 33 patients had LVEF ≤35% (low LVEF group), whereas 237 patients had LVEF >35% (preserved LVEF group). The primary endpoint was angiographic success and in-hospital major adverse cardiac events (MACE). MACE included death from any cause, postprocedure onset MI, emergency coronary artery bypass grafting, and target vessel revascularization. The secondary endpoints were MACE and the components within 30days after PCI. The components of MACE were evaluated. Results. Angiographic success, defined as <30% residual stenosis with thrombolysis in myocardial infarction flow 3 at final angiography, was achieved in all patients without fatal complications. Intra-aortic ballon pumping (IABP) was used significantly more frequently in the low LVEF group compared with the preserved LVEF group (15.2% vs. 2.1%, p = .003). There were no significant differences between groups regarding in-hospital and clinical outcomes within 30 days following PCI. Conclusion. If medications and mechanical support were appropriately performed, the angiographic success rate and in-hospital MACE rate of PCI with RA in patients with low LVEF could be expected to have good outcomes similar to those for patients with preserved LVEF.

To evaluate percutaneous coronary intervention (PCI) practice trends and 12-month outcomes in Australia in the era of drug-eluting stents (DES). Prospective study of consecutive patients undergoing 9204PCIs between 1April 2004and 31March 2008at seven Victorian public hospitals. Temporal trends in baseline characteristics and in-hospital and 12-month clinical outcomes including death, myocardial infarction (MI), target vessel revascularisation (TVR) and composite major adverse cardiac events (MACE), from year to year. Between 2004-2005and 2007-2008, the mean age of patients undergoing PCI was stable (65± 12years), and comorbidities such as hypertension, hyperlipidaemia, peripheral arterial disease and stroke increased (P<0.05). There were fewer elective and more urgent PCIs, especially for MI <24hours (17.6% in 2004-2005to 27.2% in 2007-2008, P<0.01). Overall stent use remained high (mean, 94.6%), but use of DES declined steadily (53.9% in 2004-2005to 32.0% in 2007-2008, P<0.01), despite increases in complex lesions. Planned clopidogrel therapy of ≥12months after insertion of DES increased from 54.7% in 2004-2005to 98.0% in 2007-2008(P<0.01). The overall procedural success rate was high (mean, 95.9%), and 12-month rates of mortality (3.8%), MI (4.8%), TVR (6.8%) and stent thrombosis (1.8%) remained low. Selective use of DES was an independent predictor of freedom from MACE at 12months (odds ratio, 0.68; 95% CI, 0.56-0.81). Use of DES declined steadily from 2004-2005to 2007-2008, despite increasing patient risk profile and lesion complexity. Procedural success remained high and 12-month adverse outcomes remained low, with increasing use of prolonged dual antiplatelet therapy.

ObjectivesLiving in a military environment, as a unique job and lifestyle, may affect the physical and mental status of military personnel. Coronary artery disease (CAD) status and outcomes of percutaneous...

BackgroundPerformance of percutaneous coronary intervention (PCI) is usually restricted to hospitals with cardiac surgery on site. We conducted a noninferiority trial to compare the outcomes of PCI performed at hospitals without and those with on-site cardiac surgery.MethodsWe randomly assigned participants to undergo PCI at a hospital with or without on-site cardiac surgery. Patients requiring primary PCI were excluded. The trial had two primary end points: 6-week mortality and 9-month incidence of major adverse cardiac events (the composite of death, Q-wave myocardial infarction, or target-vessel revascularization). Noninferiority margins for the risk difference were 0.4 percentage points for mortality at 6 weeks and 1.8 percentage points for major adverse cardiac events at 9 months.ResultsA total of 18,867 patients were randomly assigned in a 3:1 ratio to undergo PCI at a hospital without on-site cardiac surgery (14,149 patients) or with on-site cardiac surgery (4718 patients). The 6-week mortality rate was 0.9% at hospitals without on-site surgery versus 1.0% at those with on-site surgery (difference, −0.04 percentage points; 95% confidence interval [CI], −0.31 to 0.23; P=0.004 for noninferiority). The 9-month rates of major adverse cardiac events were 12.1% and 11.2% at hospitals without and those with on-site surgery, respectively (difference, 0.92 percentage points; 95% CI, 0.04 to 1.80; P=0.05 for noninferiority). The rate of target-vessel revascularization was higher in hospitals without on-site surgery (6.5% vs. 5.4%, P=0.01).ConclusionsWe found that PCI performed at hospitals without on-site cardiac surgery was noninferior to PCI performed at hospitals with on-site cardiac surgery with respect to mortality at 6 weeks and major adverse cardiac events at 9 months. (Funded by the Cardiovascular Patient Outcomes Research Team [C-PORT] participating sites; ClinicalTrials.gov number, NCT00549796.)

Percutaneous coronary intervention (PCI) is a widely used treatment for coronary artery disease. While direct stenting and predilation are commonly practiced techniques in PCI, their comparative efficacy and outcomes need further investigation to guide optimal treatment strategies. Objective: To compare the outcomes of direct stenting versus predilation in patients undergoing percutaneous coronary interventions (PCI). Methods: This randomized controlled trial (RCT) was conducted at the National Institute of Cardiovascular Diseases (NICVD) in Karachi, Pakistan, from August 2023 to January 2024, following ethical approval. One hundred fifty patients undergoing PCI were enrolled and randomly assigned into Group A (Direct Stenting) and Group B (Predilation). Group A received PCI with direct stent placement, while Group B underwent PCI with balloon predilation before stent deployment. Patients were followed up at one month and six months post-PCI for major adverse cardiovascular events (MACE), including death, myocardial infarction, target vessel revascularization (TVR), and stent thrombosis. Angiographic follow-up was performed as clinically indicated at six months. Data were analyzed using SPSS Version 25. Results: 150 patients participated, with a mean age of 50.25 ± 10.14 years. The Direct Stenting group had a shorter mean hospital stay (2.04 ± 0.44 days) than the Predilation group (3.08 ± 0.56 days). Gender distribution showed a higher percentage of males in the Direct Stenting group (54.7%) and more females in the Predilation group (60.0%). The Direct Stenting group demonstrated a lower incidence of MACE (5.3% vs. 12.0%) than the Predilation group. The rates of myocardial infarction, TVR, stent thrombosis, and complications such as coronary dissection and distal embolization were comparable between both groups. Mortality rates were identical at 1.3% in both groups. Conclusion: Both direct stenting and predilation are effective strategies in PCI, with direct stenting showing a slight advantage in reducing MACE and hospital stay. However, further large-scale randomized trials are required to define better the optimal use of these techniques across diverse patient populations and lesion types, ultimately improving patient outcomes.

In-stent restenosis (ISR) is highly prevalent and leads to repeat revascularisation. Long-term implications of ISR are poorly understood. This study aimed to evaluate the long-term outcomes of patients undergoing percutaneous coronary intervention (PCI) for ISR. National Cardiovascular Data Registry CathPCI records for individuals aged ≥65 years undergoing PCI from July 2009 to December 2014 were linked to Medicare claims. Baseline characteristics and long-term rates of death, myocardial infarction (MI), repeat revascularisation including target vessel revascularisation (TVR), and major adverse cardiovascular and cerebrovascular events (MACCE) were compared between ISR PCI versus de novo lesion PCI. Of 653,304 individuals, 10.2% underwent ISR PCI and 89.8% underwent de novo lesion PCI. The median duration of follow-up was 825 days (quartile 1: 352 days-quartile 3: 1,379 days). The frequency of MACCE (55.6% vs 45.0%; p<0.001), all-cause mortality (27.8% vs 25.5%; p<0.001), MI (19.0% vs 12.3%; p<0.001), repeat revascularisation (31.9% vs 18.6%; p<0.001), TVR (22.4% vs 8.0%; p<0.001), and stroke (8.8% vs 8.3%; p=0.005) was higher after ISR PCI. After multivariable adjustment, ISR PCI remained associated with worse long-term outcomes than after de novo lesion PCI (hazard ratio [HR] for MACCE 1.24 [95% CI: 1.22, 1.26], mortality 1.07 [95% CI: 1.05, 1.09], MI 1.44 [95% CI: 1.40, 1.48], repeat revascularisation 1.55 [95% CI: 1.51, 1.59], and TVR 2.50 [95% CI: 2.42, 2.58]). ISR PCI was common and was associated with a significantly higher risk of recurrent long-term major ischaemic events compared to patients undergoing de novo lesion PCI. There remains a need for new strategies to minimise ISR.

We present real world experience from a single center registry comparing the 6-month outcome of percutaneous coronary intervention (PCI) in unselected high-risk individuals using either sirolimus-eluting (SES) or paclitaxel-eluting stents (PES). We compared clinical outcome at 6 months follow-up in two cohorts of 156 consecutive patients (total n = 312) who underwent SES (June 2002-February 2003) and PES (march 2003-July 2003) implantation. The primary endpoint was a composite of major adverse cardiac events (MACE). Baseline clinical characteristics were well matched. The 6-month target vessel revascularization (TVR) rates were 1.9% (SES) and 2.6% (PES) and MACE rates were similar in the two groups (SES 4.5% vs. PES 3.2%, P = NS). In the PES group, intervention for multivessel disease, bifurcation lesions and in small vessels was more common, and for in-stent restenosis less common, reflecting the impact of drug eluting stents on indications for PCI. The incidence of sub-acute stent thrombosis, related to inadequate antiplatelet therapy in 3 of the 6 cases, was 0.95% with no difference between the two groups. This study confirms the safety and efficacy of SES and PES in unselected high risk patients undergoing PCI. Clinical outcomes of both stents are equivalent at 6 months with low rates of MACE and TVR. These data provide important complementary information to forthcoming randomized studies.

Outcomes of complex percutaneous coronary interventions (PCIs) in older patients are still debated. The aim of the study was to evaluate clinical outcomes of Octogenarian patients treated with ultrathinstents on left main or on coronary bifurcations, compared with younger patients. All consecutive patients presenting a critical lesion of an unprotected left main (ULM) or a bifurcation and treated with very thin stents were included in the RAIN (veRy thin stents for patients with left mAIn or bifurcatioN in real life) registry and divided into octogenarians group (OG, 551 patients) and nonoctogenarians (NOGs, 2,453 patients). Major adverse cardiovascular event (MACE), a composite end point of all-cause death, nonfatal myocardial infarction (MI), target lesion revascularization (TLR), and stent thrombosis (ST), was the primary endpoint, while MACE components, cardiovascular (CV) death, and target vessel revascularization (TVR) were the secondary ones. Indication for PCI was acute coronary syndrome in 64.7% of the OG versus 53.1% of the NOG. Severe calcifications and a diffuse disease were significantly more in OG. After a follow-up of 15.2 ± 10.3 months, MACEs were higher in the OG than in the NOG patients (OG 19.1% vs. NOG 11.2%, p < .001), along with MI (OG 6% vs. NOG 3.4%, p = .002) and all-cause death (OG 14% vs. NOG 4.3%, p < .001). In contrast, no significant difference was detected in CV-death (OG 5.1% vs. NOG 4%, p = .871), TVR/TLR, or ST. At multivariate analysis, age was not an independent predictor of MACE (OR 1.02 CI 95% 0.76-1.38), while it was for all-cause death, along with diabetes, GFR < 60 ml/min, and ULM disease. Midterm outcomes of complex PCI in OG are similar to those of younger patients. However, due to the higher non-CV death rate, accurate patient selection is mandatory.

In-Stent CTO Percutaneous CoronaryIntervention: Individual PatientData Pooled Analysis of 4MulticenterRegistries.

High-sensitivity C-reactive protein (hs-CRP) has been used to predict the risk of adverse cardiac events in patients with coronary artery disease (CAD) after percutaneous coronary intervention (PCI). Less is known, however, about the association between hs-CRP and long-term outcome after PCI in the Japanese population. We studied 3,039 all-comer patients with CAD who underwent their first PCI and had data available for preprocedural hs-CRP at Juntendo University between 2000 and 2011. Patients were assigned to tertiles based on preprocedural hs-CRP concentration. We evaluated the incidence of major adverse cardiac events (MACE) including all-cause death, acute coronary syndrome (ACS), and target vessel revascularization (TVR). Patients with higher hs-CRP had a higher prevalence of current smoking, chronic kidney disease and ACS, and a lower prevalence of statin use. During a median follow-up period of 6.5 years, ongoing divergence in MACE with hs-CRP tertile was noted on Kaplan-Meier curves (hs-CRP <0.08 mg/L, 26.4%; 0.08-0.25 mg/L, 38.2%; >0.25 mg/L, 45.6%; log-rank P<0.001). After adjustment for established cardiovascular risk factors, hs-CRP was associated with higher incidence of MACE (hazard ratio [HR], 1.10; 95% CI: 1.04-1.16, P<0.001) and higher all-cause mortality (HR, 1.14; 95% CI: 1.06-1.22, P<0.001). Preprocedural hs-CRP measurement is clinically useful for long-term risk assessment in Japanese patients with established CAD and undergoing PCI.

Long-Term Outcomes With Drug-Eluting Stents Versus Bare Metal Stents in the Treatment of Saphenous Vein Graft Disease (Results from the REgistro Regionale AngiopLastiche Emilia-Romagna Registry)

Background and ObjectivesHyperinsulinemia plays a key role in the development of cardiovascular impairment in patients with metabolic syndrome. The aim of this study was to evaluate the influence of hyperinsulinemia on long-term clinical outcomes of percutaneous coronary intervention (PCI) in patients without diabetes mellitus who have acute myocardial syndrome (ACS).MethodsBetween March 2016 and January 2019, we enrolled 468 patients with ACS and without diabetes mellitus who received primary PCI. Patients were divided into a low-insulin group (n = 157), medium-insulin group (n = 154), and high-insulin group (n = 157) according to tertiles of fasting insulin level. The primary endpoint was major adverse cardiac events (MACE; all-cause death, non-fatal myocardial infarction, target vessel revascularization [TVR]) at 24 months. The secondary endpoint was angina hospitalization.ResultsPatients in the high-insulin group had an unfavorable prognosis, with a higher rate of MACE (34.39%) than the low-insulin group (22.29%) and medium-insulin group (23.37%) at 24 months (P < 0.05). This difference was mainly driven by the increase in TVR. The high-insulin group also had a higher rate of angina hospitalization than the low-insulin group. Cox proportional hazards regression showed that high-insulin level (hazard ratio [HR] 1.870, 95% confidence interval [CI] 1.202–2.909), small-vessel lesion (HR 1.713, 95% CI 1.111–2.642), bifurcation lesion (HR 3.394, 95% CI 2.033–5.067), SYNTAX score (HR 1.084, 95% CI 1.039–1.130), and stent length (HR 1.017, 95% CI 1.002–1.032) increased the incidence of MACE in patients with ACS and without diabetes who underwent PCI.ConclusionHyperinsulinemia might be a valid predictor of clinical outcomes in patients with ACS undergoing PCI.