I. INTRODUCTION

Cardiovascular disease remains the leading cause of death worldwide, with at least 19 million deaths in 2021, and ischemic heart disease continues to be a major cause of premature mortality and disability [1]. Over the past four decades, percutaneous coronary intervention (PCI) has become a routine revascularization strategy, supported by major advances in devices, techniques, pharmacotherapy, and procedural technology, allowing treatment of increasingly complex lesions and higher-risk patients [2].

In Vietnam, interventional cardiology has expanded rapidly after 25 years of development, with more than 115 centers, nearly 400 principal operators, and nearly 30,000 PCI procedures annually [3]. However, PCI-related complications still occur, including stent loss, coronary perforation, rupture, dissection, and aortic dissection, as reported in previous Vietnamese data [4].

Because PCI outcomes depend on baseline risk and procedural complexity, simple risk scores are useful for predicting complications and mortality. Therefore, this study aimed to describe clinical/paraclinical characteristics, procedural features, complications, mortality, and the distribution of MCRS and NYRS, and to evaluate their associations and predictive value for PCI-related complications and in-hospital mortality.

II. MATERIALS AND METHODS

2.1. Study population, study design and sampling

All inpatients admitted via the Emergency Department at Vietnam-Cuba Friendship Hospital, Dong Hoi, from June 2024 to January 2025, diagnosed with ACS (STEMI, NSTEMI, unstable angina) or CCS (stable angina/chronic ischemic heart disease). This was a cross-sectional descriptive study with follow-up from PCI initiation to discharge, using convenience sampling of all eligible patients.

2.2. Inclusion and exculsion criteria

Inclusion criteria: Eligible patients underwent PCI and were classified as Emergency PCI (urgent/emergent PCI for ACS during index admission) or Elective PCI (planned PCI without emergency indication). Adverse events were categorized as intra-procedural or post-procedural in-hospital complications. Exclusion criteria: Patients were excluded if they had hematologic/coagulation disorders, acute liver failure with encephalopathy, end-stage renal disease on maintenance hemodialysis, diagnostic coronary angiography without PCI, or refusal to participate.

2.3. Data collection and follow-up

Data were collected using standardized case report forms, including demographics, symptom onset for Emergency PCI, hemodynamic status, heart failure severity, comorbidities, and baseline investigations. Patients were monitored during PCI and followed until discharge for clinical status, hemodynamics, ECG, access-site complications, urine output, and clinically indicated tests.

2.4. Risk factors, assessments, and outcomes

Analyzed risk factors included cardiogenic shock, Killip/NYHA class, hypertension, diabetes, anemia, and renal dysfunction. Complication risk was assessed using MCRS and mortality risk using NYRS. Baseline and follow-up assessments, as clinically indicated, included 12-lead ECG, transthoracic echocardiography, hs-troponin T, serum creatinine, blood glucose, and complete blood count. Outcomes included in-hospital complications, mortality, and the predictive value of MCRS/NYRS. Coronary perforation, dissection, and no-reflow were classified using the Ellis system, NHLBI A-F grading, and TIMI flow grade, respectively. Periprocedural myocardial infarction, reperfusion syndrome, and contrast-induced nephropathy were also assessed, with CIN risk evaluated using the Mehran score.

2.5. Statistical analysis

Data were analyzed using SPSS 26.0. Continuous and categorical variables were compared using t-test and χ² test, respectively. ORs with CIs were calculated, with p < 0.05 considered significant. ROC curves and AUC were used to assess MCRS and NYRS performance, interpreted as excellent (0.90-1.00), good (0.80-0.89), fair (0.70-0.79), poor (0.60-0.69), or no value (0.50-0.59).

2.6. Ethics

The study was approved by the Biomedical Research Ethics Committee of Hue University of Medicine and Pharmacy, conducted to improve care without added risk. PCI indications were clinically consulted, approved by hospital leadership, and performed after informed agreement from patients’ family members.

III. RESULTS

The study included 102 PCI patients, with a mean age of 70.1 ± 11.8 years; most were ≥60 years old (82.4%) and male (72.5%). Hypertension was the most common comorbidity (55.9%), followed by diabetes (24.5%) and chronic coronary artery disease (23.5%). Before PCI, 31.4% had NYHA III-IV and 8.8% had Killip III-IV. Among myocardial infarction patients, the most common symptom-to-admission time was 6-24 hours (41.1%). Most patients had preserved or moderately reduced EF, while baseline tests showed mild renal impairment, hyperglycemia, and elevated hs-troponin T (Table 1).

Table 1: Baseline characteristics & pre-PCI assessments

Radial access was used in almost all procedures (97.1%). LAD was the most common stenotic vessel, while total LAD/RCA occlusion was significantly more frequent in emergency PCI. Emergency PCI also had a longer procedure time. Stenting after balloon predilation was the main technique, and overall PCI success was high (98.0%) without significant group difference (Table 2).

Table 2: Angiographic findings and procedural characteristics

Overall, 18.6% of patients had in-hospital complications, with no significant difference between emergency and elective PCI. The most common events were reperfusion syndrome, periprocedural myocardial infarction, side-branch occlusion, and access-site bleeding. Reperfusion syndrome occurred only after emergency PCI, while periprocedural myocardial infarction occurred only after elective PCI. In-hospital mortality was low (2.9%) and occurred only in emergency PCI patients (Table 3).

Table 3: In-hospital outcomes, complications, and mortality

The overall mean scores were 7.8 ± 3.6 for MCRS and 9.2 ± 5.3 for NYRS. Median MCRS and NYRS were both significantly higher in the emergency PCI group than in the elective PCI group (11 vs. 5 and 10 vs. 5, respectively; both p < 0.001), indicating a substantially higher risk profile in emergency PCI patients (Table 4).

Table 4: Distribution of risk scores (MCRS and NYRS) and comparison by PCI type

Both MCRS and NYRS were associated with older age, worse heart failure status, lower eGFR, higher glucose and hs-troponin T, emergency PCI, and longer procedure time. Both scores also correlated with mortality, reperfusion syndrome, and complication burden. MCRS and NYRS were strongly correlated with each other (r = 0.819, p < 0.001) (Table 5).

Table 5: Associations of MCRS and NYRS with baseline, procedural, and outcome variables

MCRS modestly predicted in-hospital complications (AUC 0.640; cut-off >9), while NYRS excellently predicted mortality (AUC 0.907; cut-off >11). However, direct ROC comparison showed no significant difference between the two scores for predicting complications or mortality. In multivariable analysis, NYHA III-IV independently predicted complications, whereas cardiogenic shock independently predicted in-hospital mortality (Tables 6-8; Figures 1).

Table 6: ROC performance of MCRS and NYRS

Table 7: Logistic regression for in-hospital complications

Table 8: Logistic regression for in-hospital mortality

Figure 1: Comparison of ROC curves of the MCRS and NYRS for predicting in-hospital complications (A) and in-hospital mortality (B) in PCI patients

IV. DISCUSSION

The main findings were that MCRS and NYRS were higher in emergency than elective PCI, correlated with greater clinical severity, and had different predictive strengths: MCRS modestly predicted in-hospital complications, whereas NYRS excellently predicted in-hospital mortality. These findings are clinically plausible, as both scores use simple pre-procedural variables and have been validated for short-term PCI risk stratification [5-9]>.

Higher scores were associated with older age, NYHA III-IV, renal dysfunction, hyperglycemia, elevated hs-troponin T, and more complex emergency PCI features such as total LAD/RCA occlusion and longer procedures. This is consistent with ACS guidelines, which identify hemodynamic instability, heart failure, renal impairment, and infarct severity as key determinants of early PCI risk [10]. In multivariable analysis, NYHA III-IV independently predicted complications, while cardiogenic shock predicted mortality, highlighting the central role of bedside instability. The strong correlation between MCRS and NYRS suggests overlapping assessment of patient frailty and procedural vulnerability, although their performance differs by outcome [5-10]>.

Practically, these scores may support PCI triage, counseling, and monitoring. NYRS >11 may help identify patients needing intensive hemodynamic monitoring, critical care support, and early risk communication, while MCRS >9 may flag patients requiring closer surveillance for complications, renal protection, and bleeding prevention. However, CIN still requires dedicated assessment, such as the Mehran score [11]. Limitations include the single-center design, modest sample size, low event counts, especially deaths, and lack of external recalibration, which may limit model stability, cut-off accuracy, and generalizability. Nevertheless, the findings support previous evidence that MCRS and NYRS remain clinically useful, although performance may vary across populations and settings [5,6,8,9,11].

V. CONCLUSION

In this single-center PCI cohort, emergency PCI patients had higher risk and worse in-hospital outcomes than elective PCI patients. MCRS and NYRS were associated with clinical severity and provided useful prognostic information: MCRS modestly predicted complications, while NYRS excellently predicted mortality. These findings support using simple pre-procedural risk scores, combined with bedside assessment of heart failure and shock, for early PCI risk stratification.

Declaration of conflicting interests: The authors declare no conflicts of interest with respect to the research, authorship, and/or publication of this article.