ABSTRACT
Background
Percutaneous coronary intervention (PCI) outcomes can vary due to various factors, including patient clinical condition, complexity of coronary lesions, expertise of operators, and quality of the PCI center.
Aims
This study evaluated the influence of PCI center volume and operator experience on patient outcomes after the procedure.
Methods
Retrospective data on demographic, clinical details, and outcomes for all patients undergoing PCI across 39 hospitals in Thailand from 2018 to 2019 were retrieved. PCI center volume was categorized based on annual number of interventions: low (< 200), intermediate (200–499), and high (≥ 500). Operator experience was assessed by years of practice (low [< 5] and high [≥ 5]) and the number of PCI cases performed annually (low [< 75] and high [≥ 75]). The evaluated PCI outcomes were: PCI failure; procedural complications; PCI‐related in‐hospital mortality; 1 year post‐intervention all‐cause mortality.
Results
A total of 19,701 patients who underwent PCI were included in the analysis, of whom 17,432 had follow‐up data available after 1 year. Of these, 58.1% presented with either ST‐elevation or non‐ST elevation myocardial infarction/unstable angina, while 41.9% had stable CAD. Nearly half of the patients had triple‐vessel or left‐main disease, and 8.7% presented with cardiogenic shock. The percent with PCI failure, procedural complications, PCI‐related in‐hospital death, and 1‐year all‐cause mortality were 4.9%, 5.1%, 2.7%, and 11.8%, respectively. Despite patients in higher‐risk profiles being treated at high‐volume PCI centers and by experienced operators, there were no significant differences in PCI failure, PCI‐related in‐hospital mortality nor 1‐year all‐cause mortality compared to those treated at low or intermediate volume PCI centers. However, high‐volume PCI centers had procedural complications more frequently (4.7%) than did intermediate (3.9%) and low‐volume (2.5%) centers (p < 0.001). After adjusting for confounding factors, no significant associations were found between PCI center volume and PCI outcome. Similarly, no significant relationship was found between operator experience and procedural complications, nor 1‐year all‐cause mortality. Nevertheless, operators with more years of practice were associated with lower PCI‐related in‐hospital mortality (odds ratio [95% CI] of 0.75 (0.57, 0.98); p < 0.038). Additionally, operators conducting a higher number of PCIs annually tended to have less PCI failures (odds ratio [95% CI] of 0.76 (0.57, 1.01); p = 0.062).
Conclusion
A center's PCI volume did not significantly impact PCI outcome. In contrast, operator experience did impact outcomes. This result highlights areas for improvement and can help reform strategies for national PCI systems at both center and operator levels.
Keywords: hospital volume, operator experience, operator volume, percutaneous coronary intervention
1. Introduction
Outcomes of percutaneous coronary intervention (PCI) procedures can vary due to multiple factors, including patient clinical condition, complexity of coronary lesions, expertise of operators, and quality of the PCI center. The significance of operator skill and of a center's PCI volume becomes more important when managing complex anatomical conditions (e.g., left main disease or chronic total occlusion), critical clinical scenarios (like primary PCI), or procedural complications [1, 2, 3].
Previous studies have suggested that there is a direct relationship between favorable PCI outcomes and high‐volume operators and centers, often referred to as the “volume‐outcome” relationship [4, 5, 6]. In, the American College of Cardiology Foundation (ACCF), American Heart Association (AHA), Society for Cardiovascular Angiography and Interventions (SCAI) [7, 8] and the European Society of Cardiology (ESC)/European Association for Cardio‐Thoracic Surgery (EACTS) [9] have all endorsed recommendations for minimum institutional and operator annual volumes to maintain satisfactory outcomes, particularly in patients with ST elevation myocardial infarction (STEMI). These volume‐outcome guidelines have been globally adopted to maximize quality and safety of the PCI procedure.
While past studies indicated that there are positive correlations between both operator experience and PCI center volume, and PCI outcome, more recent studies find this association to be weaker than in those earlier studies [10, 11, 12]. Such a trend could be attributed to the increased standardization of PCI techniques, advancements in stent technology enhancing deliverability and performance and/or the development of specialized equipment enabling easier completion of complex PCI procedures. Thus, the procedure may have become less reliant on individual operator skills and more influenced by technical advancements. Notably, a study from the US Veterans Affairs’ healthcare system finds no association between operator nor facility volumes and 30‐day mortality [13].
Similarly, prior publications involving Asian populations presented inconclusive findings regarding an association between operator volume and in‐hospital outcome following PCI [14, 15]. The direct mechanisms by which operator experience and hospital volume may impact outcomes remains an area of research and debate within the interventional community, both in Asia and the West. Additionally, short‐term PCI outcomes, including in‐hospital mortality, might not reflect an operator's skill level, particularly when assessed in isolation [16]. Thus, there is a need to identify or refine variables which better predict patient outcome. Extending the period of outcome observation may better reflect an operator's skill and the influence of a PCI center's volume.
Before 2010, qualified interventional cardiologists in Thailand typically gained their skills through overseas training or self‐directed learning, as formal training programs were not established until later. Since then, numerous cardiac catheterization laboratories have been operating across the country over the past decade. The government's policy also enhances accessibility to PCI treatment by leveraging various health insurance schemes. These efforts aimed to enhance patient accessibility to PCI treatment. To further improve PCI outcomes and patient safety, ensuring operator proficiency and PCI center volume remain important. Therefore, this study utilized data from the Thai PCI Registry to assess the relationships of PCI center volume and operator experience to PCI outcomes, specifically PCI failure, procedural complications, and both PCI‐related in‐hospital and 1‐year all‐cause mortalities.
2. Methods
This retrospective cohort study utilized a nationwide, prospective multicenter PCI Registry initiated in 2018 by the Cardiac Intervention Association of Thailand. The study protocol was detailed in a previous publication [17]. Briefly, all Thai PCI facilities were invited to participate in the Registry and 39 out of 72 facilities voluntarily joined in this registry initiative. These included university, government, and private PCI centers which were situated across all five regions of the country. Patients included in this study were aged 18 years or older and underwent PCI during the periods May 1, 2018–April 2, 2019, and June 21–August 1, 2019. Patients involved in other clinical trials, unable to complete follow‐up, or who declined participation were deemed ineligible. The study received approval from the Central Research Ethics Committee of Thailand (COA‐CREC # 006/2018) and the Ethics Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University (COA‐MURA 2024/129). Written informed consent was given by each patient (or his/her legal representative) before the PCI intervention. Note that patients who underwent procedures and subsequently died were maintained in this analysis.
2.1. Data Collection
All electronic databases were stored at a central data management unit (DMU), in the Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital. Data were initially recorded on case record forms (CRFs), and then entered into the electronic databases by trained staff from the catheterization labs. The authors audited data from all study sites to validate accuracy and completeness, randomly selecting 10% of the total number of PCI patients from each site. Medical records and hard copies of CRFs were meticulously reviewed during these audits. Follow‐up audits were conducted in cases where the central DMU had questions regarding the correctness of the data.
Following demographic, clinical, and angiographic characteristics, along with procedural data were extracted from the DMU. The patient data analyzed encompassed various factors: age, gender, health insurance (universal coverage, government service/state enterprise, social security service, uninsured or self‐pay), body‐mass index (BMI), presence of cardiovascular risk factors (diabetes mellitus [DM], hypertension, dyslipidemia, smoking, and chronic kidney disease [CKD, defined as eGFR < 60 ml/min/1.73m2]), and history of relevant underlying cardiovascular diseases (cerebrovascular disease [CVD], peripheral arterial disease [PAD], myocardial infarction [MI], previous PCI/coronary artery bypass graft [CABG]). The clinical and angiographic data compiled included: clinical presentation (ST‐elevation myocardial infarction [STEMI], non‐ST‐elevation myocardial infarction [NSTEMI]/unstable angina [UA] and stable CAD), left ventricular ejection fraction (LVEF), presence of cardiogenic shock before PCI, prior thrombolytic treatment, type of PCI (urgent/emergent, primary PCI, rescue PCI), number of diseased vessels [single‐vessel disease (SVD), double vessel disease (DVD), triple vessel disease (TVD), left main (LM)], SYNTAX score (< 22, 23–32, > 33%), lesion complexity types (A, B1, B2, C), number of lesions treated, and stent deployed, mechanical support devices (intra‐aortic balloon pump [IABP] insertion, extracorporeal membrane oxygenation [ECMO], ventricular assist device [VAD]), radial access, stent type (drug eluting stent [DES], bare metal stent, bioabsorbable stent), mode of lesion severity assessment (intravascular ultrasound study [IVUS], optical coherence tomography [OCT] or fractional flow reserve [FFR] wire), and use (or not) of plaque modification devices (rotational atherectomy, cutting/scoring balloon or laser atherectomy), glycoprotein IIb/IIIa inhibitor, and post‐PCI medication treatments such as aspirin, statin, clopidogrel, prasugrel, and ticagrelor.
2.2. Definitions of PCI Center and Operator Volume
The determinations of PCI center and operator volumes (number of PCIs performed annually) were extracted from the Thai PCI Registry databases. The duration of operator experience was calculated by subtracting the year of current practice from the year of completion of interventional cardiology training.
We established the stratification for PCI center and operator volumes in accordance with the ACCF/AHA/SCAI 2011 (and 2013 Update) Clinical Competence Statement on Coronary Artery Interventional Procedures [7, 8]. For PCI center volume, groupings of annual PCI volumes were defined as follows: low (< 200), intermediate (200–499), and high (≥ 500). Operator experience was evaluated based on years of practice (low [< 5] and high [≥ 5]) and number of PCI cases performed per year (low [< 75] and high [≥ 75]).
2.3. PCI Outcomes of Interest
The PCI outcomes of primary interest were PCI‐related in‐hospital mortality and 1‐year all‐cause mortality. Additionally, instances of PCI failure (defined as the inability to cross lesions with a wire, or to advance a balloon or stent, resulting in > 50% residual stenosis) and procedural complications were carefully documented. Procedural complications included nonfatal MI, nonfatal stroke, cardiogenic shock, heart failure, new requirements for dialysis, major bleeding requiring blood transfusion, endotracheal intubation, cardioversion/defibrillation, and in‐hospital CABG. Over the 1‐year follow‐up period, instances of nonfatal MI, nonfatal stroke, and unplanned repeat PCI were also recorded. In cases where a patient experienced multiple cardiovascular events, only the first event was considered. The definition of MI was an increase in cardiac troponin (cTn) coupled with one of the following: (1) evidence of prolonged ischemia characterized by chest pain lasting over 20 min; (2) ischemic ST‐segment changes or the emergence of new pathological Q waves; (3) angiographic evidence of coronary occlusion or no‐reflow/slow flow; (4) imaging demonstrating new loss of viable myocardium or new regional wall motion abnormality. Stroke was defined as a history of ischemic or hemorrhagic stroke, or a transient ischemic attack, validated through CT or MRI scans.
2.4. Statistical Analysis
Data are presented as mean and standard deviation (SD), or median and interquartile range (IQR) where appropriate for continuous data; frequency and percentage for categorical data. Characteristics of patients, PCI centers, and operators were then compared between groups using t‐test or the Mann‐Whitney U test where appropriate for continuous data; using Chi‐square test for categorical data. Multivariate logistic regression was applied to assess associations between PCI center and operator factors, and PCI outcomes after adjustment for co‐variables. All analysis were performed using STATA version 17.0. P‐values of less than 0.05 were considered statistically significant.
3. Results
A total of 19,701 patients who underwent PCI were included in the analysis; 1‐year follow‐up data were available for 17,432. Their average age was 64.1 ± 11.7 years; 69.1% were male and 60% classified as overweight (Table S1). Atherosclerotic risk factors were common, with 67.4% having hypertension, 65.3% dyslipidemia, 44.2% DM, 28.9% CKD without dialysis, and 3.5% CKD requiring dialysis; 55.4% were current or ex‐smokers. Notably, 23.3% had a previous history of MI, and 20.7% had undergone a prior PCI. The majority (63.6%) of patients were covered by universal health coverage, followed by those covered under government service/state enterprise (26.6%) or social security service (6.7%).
Regarding clinical presentation, about 58.1% of patients showed acute coronary syndrome, comprising either STEMI or NSTEMI/UA, while the remainder exhibited stable CAD. Nearly half of the patients had triple‐vessel or left main disease, and 22% were found to have impaired left ventricular systolic function (LVEF < 40%), with 8.7% presenting with cardiogenic shock.
Of the 39 participating hospitals, the distribution of PCI volumes was as follows: 17 hospitals (43.6%) were categorized as high‐volume PCI centers, nine hospitals (23.1%) as intermediate‐volume centers, and 13 hospitals (33.3%) as low‐volume centers. There were 135 operators involved; of them, 115 (85.2%) had over 5 years of interventional practice. When considering the number of PCIs done annually, 79 operators (58.5%) performed a high number (> 75) of cases.
3.1. Comparison of Baseline Characteristics Among Groups
The majority of patients underwent PCI procedures at high‐volume PCI centers (77.2%). Likewise, most patients received their PCI from experienced operators with over 5 years of interventional practice (80.3%) and a high number of PCIs annually (93.6%). A comparative analysis of baseline demographic, clinical, and angiographic characteristics, and procedural data stratified by both PCI center and operator volume, is presented in Table 1. In summary, no specific patterns were observed among groups in regard to age, gender and comorbidities. However, distinct angiographic and procedural features were found in each group. For instance, patients with high‐risk profiles and referred cases (e.g., STEMI, cardiogenic shock, rescue PCI, emergency/urgent PCI, post‐thrombolytic therapy, high SYNTAX score, and complex lesions [i.e., type B2 and C, including chronic total occlusion (CTO)]) tended to be treated at high‐volume centers by operators with more years of experience and a higher annual PCI volume. Conversely, primary PCI and emergent/urgent PCI procedures tended to be conducted at low‐volume PCI centers, by interventionists with fewer years of practice and lower annual PCI volumes.
TABLE 1.
Baseline characteristics stratified by PCI center volume and operator experience.
| PCI center volume (PCIs/yr) | P‐value | Experience of operator (yrs) | P‐value | Number of PCIs/operator/yr | P‐value | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| < 200 | 200‐499 | > 500 | < 5 | > 5 | < 75 | > 75 | ||||
| n = 908 | n = 3574 | n = 15,219 | n = 3881 | n = 15,820 | n = 1267 | n = 18,434 | ||||
| Male, % | 64.5 | 68.5 | 69.6 | 0.004 | 68.5 | 69.3 | 0.319 | 69.3 | 69.1 | 0.890 |
| Age, mean ± SD | 64.6 ± 12.6 | 63.8 ± 11.9 | 64.2 ± 11.6 | 0.052 | 63.6 ± 11.8 | 64.3 ± 11.7 | < 0.001 | 64.9 ± 12.3 | 64.1 ± 11.7 | 0.017 |
| BMI, mean ± SD | 25.0 ± 4.5 | 24.4 ± 4.2 | 24.2 ± 4.2 | < 0.001 | 24.2 ± 4.2 | 24.3 ± 4.2 | 0.344 | 24.5 ± 4.0 | 24.3 ± 4.2 | 0.084 |
| Admission SBP, mmHg, mean ± SD | 136.6 ± 25.5 | 141.0 ± 27.6 | 136.5 ± 26.5 | < 0.001 | 136.4 ± 27.3 | 137.5 ± 26.6 | 0.016 | 136.8 ± 25.6 | 137.3 ± 26.8 | 0.498 |
| Admission HR, bpm, mean ± SD | 76.9 ± 17.2 | 76.3 ± 16.8 | 75.9 ± 16.6 | 0.129 | 74.8 ± 16.5 | 76.4 ± 16.7 | < 0.001 | 76.7 ± 17.1 | 76.0 ± 16.6 | 0.159 |
| Refer case, % | 20.5 | 55.0 | 56.5 | < 0.001 | 58.5 | 53.6 | < 0.001 | 35.3 | 55.9 | < 0.001 |
| Health insurance, % | ||||||||||
| UC | 34.0 | 59.7 | 66.3 | < 0.001 | 68.4 | 62.5 | < 0.001 | 41.3 | 65.2 | < 0.001 |
| Government service/state enterprise | 28.1 | 31.2 | 25.4 | 21.8 | 27.8 | 38.5 | 25.8 | |||
| Social security service | 7.3 | 8.4 | 6.3 | 9.0 | 6.2 | 6.6 | 6.8 | |||
| Uninsured or self‐pay | 30.6 | 0.8 | 1.9 | 0.8 | 3.6 | 13.6 | 2.3 | |||
| DM, % | 46.6 | 44.5 | 44.0 | 0.282 | 45.1 | 43.9 | 0.176 | 43.8 | 44.2 | 0.783 |
| Hypertension, % | 72.0 | 70.8 | 66.4 | < 0.001 | 72.0 | 66.3 | < 0.001 | 70.0 | 67.3 | 0.044 |
| DLP, % | 75.0 | 63.9 | 65.0 | < 0.001 | 68.9 | 64.4 | < 0.001 | 69.2 | 65.0 | 0.002 |
| Current/Ex smoking, % | 37.8 | 53.3 | 56.9 | < 0.001 | 57.8 | 54.8 | 0.001 | 45.9 | 56.1 | < 0.001 |
| CKD, % | 31.5 | 33.9 | 30.7 | < 0.001 | 30.2 | 31.5 | 0.111 | 31.9 | 31.2 | 0.635 |
| Previous CVD, % | 6.1 | 5.2 | 5.8 | 0.345 | 4.7 | 5.9 | 0.003 | 5.8 | 5.7 | 0.884 |
| Previous MI, % | 19.2 | 22.5 | 23.6 | 0.005 | 23.2 | 23.2 | 0.924 | 24.7 | 23.1 | 0.197 |
| PAD, % | 2.8 | 1.2 | 1.8 | 0.002 | 1.7 | 1.7 | 0.765 | 2.1 | 1.7 | 0.365 |
| Prior CABG, % | 1.8 | 1.5 | 1.6 | 0.763 | 1.3 | 1.6 | 0.201 | 2.0 | 1.5 | 0.231 |
| Known CAD, % | 34.4 | 34.5 | 33.7 | 0.623 | 34.3 | 33.8 | 0.579 | 36.2 | 33.7 | 0.079 |
| Prior heart failure, % | 11.8 | 12.1 | 14.1 | 0.001 | 10.4 | 14.3 | < 0.001 | 13.6 | 13.6 | 0.950 |
| Previous PCI, % | 29.1 | 29.2 | 29.9 | 0.658 | 29.5 | 29.8 | 0.724 | 30.1 | 29.7 | 0.771 |
| CAD presentation, % | < 0.001 | 0.103 | 0.010 | |||||||
| STEMI | 18.1 | 22.5 | 29.6 | 26.5 | 28.1 | 24.2 | 28.1 | |||
| NSTEMI/UA | 36.6 | 31.0 | 29.8 | 31.3 | 30.1 | 32.3 | 30.2 | |||
| Stable CAD | 45.4 | 46.5 | 40.5 | 42.2 | 41.8 | 43.6 | 41.7 | |||
| PCI status, % | ||||||||||
| Elective | 64.7 | 65.8 | 60.4 | < 0.001 | 62.4 | 61.4 | 0.253 | 64.2 | 61.4 | 0.101 |
| Urgent | 18.3 | 14.9 | 15.7 | 15.9 | 15.6 | 13.9 | 15.8 | |||
| Emergency | 17.1 | 19.4 | 23.8 | 21.7 | 23.0 | 21.9 | 22.8 | |||
| LVEF, mean ± SD | 52.1 ± 15.8 | 50.4 ± 15.6 | 51.6 ± 15.6 | 0.003 | 52.0 ± 15.2 | 51.3 ± 15.7 | 0.021 | 51.0 ± 15.3 | 51.5 ± 15.6 | 0.390 |
| Cardiogenic shock before PCI,% | 4.5 | 4.1 | 8.6 | < 0.001 | 6.6 | 7.9 | 0.010 | 5.9 | 7.8 | 0.017 |
| Thrombolytic, % | 20.4 | 32.0 | 36.5 | < 0.001 | 30.6 | 36.5 | < 0.001 | 25.4 | 35.9 | < 0.001 |
| Primary PCI, % | 75.9 | 64.3 | 59.4 | < 0.001 | 65.8 | 59.4 | < 0.001 | 69.9 | 60.1 | 0.004 |
| Rescue PCI, % | 7.4 | 7.4 | 12.8 | < 0.001 | 10.7 | 12.1 | < 0.001 | 9.6 | 11.9 | 0.004 |
| Emergency/urgent PCI, % | 35.4 | 34.3 | 39.6 | < 0.001 | 37.6 | 38.6 | 0.278 | 35.8 | 38.6 | 0.052 |
| No. of diseased vessels, % | ||||||||||
| SVD | 25.7 | 29.2 | 25.8 | < 0.001 | 27.2 | 26.2 | 0.420 | 28.2 | 26.3 | 0.045 |
| DVD | 29.6 | 29.3 | 28.6 | 28.6 | 18.9 | 30.7 | 28.7 | |||
| TVD | 32.2 | 31.2 | 33.1 | 32.8 | 32.7 | 30.7 | 32.9 | |||
| LM | 12.4 | 10.3 | 12.5 | 11.4 | 12.2 | 10.4 | 12.2 | |||
| SYNTAX score, % | ||||||||||
| < 22 | 80.0 | 76.4 | 71.1 | 0.002 | 77.2 | 70.9 | < 0.001 | 82.4 | 72.3 | 0.036 |
| 23‐32 | 11.4 | 16.4 | 18.0 | 16.7 | 17.8 | 9.2 | 17.8 | |||
| > 33 | 8.6 | 7.2 | 10.9 | 6.1 | 11.3 | 8.4 | 9.9 | |||
| No. of lesion treated, median (range) |
1.0 (1.0, 4.0) |
1.0 (1.0, 5.0) |
1.0 (1.0, 5.0) |
< 0.001 |
1.0 (1.0, 5.0) |
1.0 (1.0, 5.0) |
0.956 |
1.0 (1.0, 4.0) |
1.0 (1.0, 5.0) |
0.047 |
| No. of stent deployed, median (range) |
1.0 (0.0, 6.0) |
1.0 (0.0, 9.0) |
1.0 (1.0, 5.0) |
0.004 |
1.0 (0.0, 9.0) |
1.0 (0.0, 9.0) |
0.598 |
1.0 (0.0, 7.0) |
1.0 (0.0, 9.0) |
0.010 |
| IABP, % | 3.9 | 2.3 | 3.6 | < 0.001 | 3.3 | 3.4 | 0.601 | 3.5 | 3.4 | 0.849 |
| Radial access, % | 51.1 | 33.9 | 46.0 | < 0.001 | 34.5 | 46.4 | < 0.001 | 42.6 | 44.2 | 0.166 |
| DES, % | 89.8 | 90.6 | 89.0 | 0.013 | 90.6 | 89.1 | 0.004 | 89.0 | 89.4 | 0.612 |
| IVUS assessment, % | 10.8 | 10.7 | 14.4 | < 0.001 | 10.1 | 14.4 | < 0.001 | 14.1 | 13.5 | 0.575 |
| Plaque modification devices, % | 4.5 | 3.7 | 5.6 | < 0.001 | 4.4 | 5.4 | 0.015 | 2.6 | 5.4 | < 0.001 |
| Medications used, % | ||||||||||
| GP IIb/IIIa inhibitor | 2.9 | 4.4 | 6.6 | < 0.001 | 4.3 | 6.4 | < 0.001 | 6.2 | 6.0 | 0.754 |
| Aspirin | 96.6 | 99,4 | 99.3 | < 0.001 | 99.1 | 99.2 | 0.679 | 97.9 | 99.3 | < 0.001 |
| Statin | 92.2 | 92.9 | 93.0 | 0.062 | 93.5 | 92.8 | 0.151 | 92.7 | 93.0 | 0.678 |
| Clopidogrel | 76.0 | 92.7 | 93.8 | < 0.001 | 93.5 | 92.6 | 0.058 | 84.9 | 93.3 | < 0.001 |
| Ticagrelor/Prasugrel | 24.0 | 7.3 | 6.2 | < 0.001 | 6.5 | 7.4 | 0.160 | 15.1 | 6.7 | < 0.001 |
| Arrhythmia, % | ||||||||||
| No | 96.6 | 96.3 | 95.0 | 0.001 | 96.1 | 95.1 | 0.034 | 96.1 | 95.3 | 0.003 |
| Arrhythmia without treatment | 1.3 | 1.2 | 1.3 | 1.2 | 1.3 | 1.9 | 1.2 | |||
| with treatment | 2.1 | 2.5 | 3.7 | 2.8 | 3.6 | 2.1 | 3.5 | |||
| ET intubation, % | 4.0 | 2.4 | 4.6 | < 0.001 | 3.3 | 4.4 | 0.002 | 3.3 | 4.2 | 0.123 |
| IABP, % | 3.9 | 2.3 | 3.6 | < 0.001 | 3.3 | 3.4 | 0.601 | 3.5 | 3.4 | 0.849 |
| FFR‐guided, % | 2.7 | 1.8 | 1.9 | 0.260 | 1.4 | 2.1 | 0.003 | 2.5 | 1.9 | 0.171 |
| Vascular complication required treatment, % | 0.1 | 0.1 | 0.4 | 0.043 | 0.4 | 0.3 | 0.614 | 0.3 | 0.3 | 0.979 |
| Lesion complexity types, % | ||||||||||
| A | 10.5 | 4.6 | 3.8 | < 0.001 | 7.6 | 3.4 | < 0.001 | 5.1 | 4.2 | < 0.001 |
| B1 | 29.2 | 17.4 | 13.5 | 16.9 | 14.4 | 22.5 | 14.4 | |||
| B2 | 27.3 | 18.0 | 18.2 | 18.1 | 18.7 | 24.4 | 18.2 | |||
| C | 33.0 | 60.1 | 64.5 | 57.4 | 63.5 | 48.0 | 63.2 | |||
Abbreviations: BMI, body mass index (kg/m2); CAD, coronary artery disease; CKD, chronic kidney disease (eGFR < 60 mL/min/1.73m2); CVD, cerebrovascular disease; DES, drug‐eluting stent; DM, diabetes mellitus; DLP, dyslipidemia; DVD, double vessel disease; ET, Endotracheal tube; IABP, intra‐aortic balloon pump; IVUS, intravascular ultrasound; LM, left main; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NSTEMI, non‐ST elevation MI; PAD, peripheral arterial disease; PCI, percutaneous coronary intervention; SD, standard deviation; STEMI, ST‐elevation myocardial infarction; SVD, single vessel disease; TVD, triple vessel disease; UA, unstable angina; UC, universal coverage.
Regarding procedural features, approaches seemed inconsistent, potentially based on operator discretion. However, radial access was commonly used across all groups of operators and PCI centers, accounting for 30%–40% of cases. Lesion severity assessment primarily utilized IVUS (14.4%); the use of plaque modification devices (5.6%) was common in high‐volume PCI centers and among operators with greater years of practice. Drug‐eluting stents were used in nearly 90% of procedures, while the remaining 10% deployed either bare‐metal or bioabsorbable stents.
Most patients were administered aspirin and statins (99%, and 93.0%, respectively) after PCI. Interestingly, clopidogrel was more frequently prescribed in high‐volume PCI centers (93.8%), whereas ticagrelor and prasugrel were commonly prescribed in low‐volume centers and by operators conducting fewer annual PCIs. Glycoprotein IIb/IIIa inhibitor was more frequently prescribed in high‐volume PCI centers (6.6%) followed by intermediate and low‐volume centers (4.4% and 2.9%, respectively).
3.2. In‐Hospital and 1 Year PCI Outcomes
Overall, the percentage of PCI failures was 4.9%; with 5.1% experiencing procedural complications and 1.0% requiring blood transfusion due to bleeding. The in‐hospital all‐cause mortality rate was 2.7%. Despite patients with higher‐risk profiles being treated at high‐volume PCI centers and by experienced operators, there were no significant differences in PCI failure and in‐hospital mortality based on center volumes (Table 2). However, high‐volume PCI centers had a higher rate of procedural complications (4.7%) compared to intermediate‐ (3.9%) and low‐volume centers (2.5%) (p < 0.001). Similarly, high‐volume PCI centers experienced more major bleeds (1.1% compared to 0.4% and 0.7%, respectively; p < 0.001), especially among operators who conducted a higher number of PCIs annually. Additionally, there was a trend toward increased procedural complications among operators with more years of practice (4.6% vs. 3.9%, p = 0.055).
TABLE 2.
In‐hospital and 1‐year clinical outcomes stratified by PCI center volume and operator experience.
| PCI center volume (PCIs/yr) | P‐value | Experience of operator (yrs) | P‐value | Number of PCIs/operator/yr | P‐value | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| < 200 | 200‐499 | > 500 | < 5 | > 5 | < 75 | > 75 | ||||
| n = 908 | n = 3574 | n = 15,219 | n = 3881 | n = 15,820 | n = 1267 | n = 18,434 | ||||
| In‐hospital outcomes | ||||||||||
| Death, % | 3.4 | 2.3 | 2.8 | 0.106 | 2.8 | 2.7 | 0.551 | 3.0 | 2.7 | 0.490 |
| PCI failure, % | 3.0 | 3.5 | 4.0 | 0.110 | 3.6 | 4.0 | 0.215 | 4.5 | 3.9 | 0.226 |
| Procedural complications, % | 2.5 | 3.9 | 4.7 | < 0.001 | 3.9 | 4.6 | 0.055 | 3.9 | 4.5 | 0.343 |
| Major bleeding, % | 0.7 | 0.4 | 1.1 | < 0.001 | 0.8 | 1.0 | 0.128 | 3.4 | 5.0 | 0.012 |
| ET intubation, % | 4.0 | 2.4 | 4.6 | < 0.001 | 3.3 | 4.4 | 0.002 | 3.3 | 4.2 | 0.443 |
| Vascular complication required treatment, % | 0.1 | 0.1 | 0.4 | 0.043 | 0.4 | 0.3 | 0.614 | 0.3 | 0.3 | 0.979 |
| New requirement for Dialysis, % | 0.2 | 0.2 | 0.6 | 0.001 | 0.6 | 0.5 | 0.357 | 0.7 | 0.5 | 0.339 |
| 1‐year outcomes | ||||||||||
| Death, % | 9.7 | 10.0 | 10.6 | 0.374 | 10.4 | 10.5 | 0.815 | 9.1 | 10.6 | 0.096 |
| Nonfatal MI, % | 11.8 | 4.8 | 7.6 | < 0.001 | 6.5 | 7.5 | 0.036 | 7.6 | 7.3 | 0.673 |
| Nonfatal stroke, % | 1.1 | 1.1 | 1.1 | 0.968 | 0.8 | 1.2 | 0.027 | 1.3 | 1.1 | 0.408 |
| Unplanned‐repeat PCI, % | 2.5 | 1.1 | 1.9 | 0.002 | 1.6 | 1.9 | 0.260 | 2.8 | 1.7 | 0.009 |
Note: Abbreviations as in Table 1.
At the 1‐year follow‐up, all‐cause mortality was increased (from 2.7% to 11.8%). Again, there were no significant differences in all‐cause of mortality among the groups based on PCI center volume nor operator experience. Patients treated at low‐volume PCI centers were more likely to develop nonfatal MI (11.8%) and experience unplanned repeat revascularization (2.5%) compared to those treated at intermediate and high‐volume centers (4.8%, 7.6%, p < 0.001, and 1.1%, 1.9%, respectively; p = 0.002). Interestingly, patients undergoing PCI by operators with more years of practice showed a higher incidence of nonfatal MI (7.5 vs. 6.5%, p = 0.036) and nonfatal stroke (1.2 vs. 0.8%, p = 0.027) compared to those with fewer years. Unplanned repeat revascularizations were also more common among operators conducting fewer PCIs annually (2.8 vs. 1.7%, p = 0.009).
3.3. Impact of PCI Center Volume and Operator Experience on PCI Outcomes
Univariate analyses were conducted for all variables associated with procedural failure, procedural complications, PCI‐related in‐hospital mortality, and 1‐year all‐cause mortality (Table S2–S5). No significant associations were found between PCI center volume or operator experience and PCI failure or 1‐year all‐cause mortality. However, high‐ and intermediate‐ volume PCI centers were associated with increased procedural complications. But notably, intermediate PCI volume centers were associated with decreased in‐hospital mortality.
After adjusting for confounding factors, no significant association was found between PCI center volume and any aspect of PCI outcome. Similarly, regardless of the number of years of practice and PCIs performed annually, operator experience was not significantly related to procedural complications nor 1‐year all‐cause mortality. However, operators performing a higher number of PCIs annually did tend to have fewer PCI failures (odds ratio [95% CIs] of 0.77 (0.57, 1.02), p = 0.066) (Table 3). And those operators with more years of practice had lower PCI‐related in‐hospital mortality rates (odds ratio [95% CIs] of 0.76 (0.58, 1.00), p = 0.045) (Table 4).
TABLE 3.
Comparison of factors associated with PCI failure and procedural complications: multivariate analysis.
| Factors | PCI failure | Procedural complications | ||
|---|---|---|---|---|
| OR (95% CI) | P‐value | OR (95% CI) | P‐value | |
| PCI center volume, PCIs/yr | ||||
| ≥ 500 versus < 200 | 1.12 (0.76, 1.65) | 0.576 | 1.41 (0.93, 2.12) | 0.109 |
| 200 − 499 versus < 200 | 0.99 (0.66, 1.49) | 0.939 | 1.39 (0.90, 2.14) | 0.141 |
| Experience of operator, yrs | ||||
| ≥ 5 versus < 5 | 1.00 (0.84, 1.20) | 0.961 | 1.06 (0.88, 1.26) | 0.589 |
| Number of PCIs/operator/yr | ||||
| ≥ 75 versus < 75 | 0.77 (0.57, 1.02) | 0.066 | 1.08 (0.80, 1.47) | 0.629 |
| Age, per 5 years | 1.010 (1.004, 1.016) | 0.001 | ||
| Male | 0.81 (0.70, 0.93) | 0.003 | ||
| BMI, kg/m2 | 1.02 (1.01, 1.04) | 0.031 | ||
| Hypertension | 0.77 (0.67, 0.89) | < 0.001 | ||
| Previous MI | 1.25 (1.08, 1.46) | 0.005 | 1.41 (1.20, 1.65) | < 0.001 |
| Prior CABG | 2.03 (1.40, 2.94) | < 0.001 | 1.68 (1.08, 2.63) | 0.024 |
| CAD presentation | ||||
| STEMI versus NSTEMI/UA | 0.55 (0.42, 0.73) | < 0.001 | 1.27 (0.98, 1.65) | 0.078 |
| NSTEMI/UA versus stable CAD | 1.09 (0.92, 1.30) | 0.350 | 0.77 (0.63, 0.94) | 0.009 |
| Emergency/urgent versus elective | 1.18 (0.95, 1.45) | 0.144 | 1.28 (1.02, 1.60) | 0.036 |
| Radial versus femoral access | 0.76 (0.66, 0.88) | < 0.001 | 1.03 (0.89, 1.19) | 0.732 |
| Cardiogenic shock before PCI | 1.39 (1.11, 1.73) | 0.005 | ||
| IABP | 1.64 (1.25, 2.14) | < 0.001 | ||
| GP IIb/IIIa inhibitor | 1.69 (1.36, 2.10) | < 0.001 | ||
| Vascular complication required treatment | 2.72 (1.28, 5.76) | 0.009 | 15.82 (9.23, 27.13) | < 0.001 |
| Arrhythmia | ||||
| with treatment versus no arrhythmia | 2.98 (2.36, 3.75) | < 0.001 | ||
| without treatment versus no arrhythmia | 3.38 (2.39, 4.78) | < 0.001 | ||
| Disease vessel | ||||
| LM versus SVD | 1.34 (1.04, 1.72) | 0.027 | ||
| TVD versus SVD | 1.48 (1.20, 1.82) | < 0.001 | ||
| DVD versus SVD | 1.31 (1.06, 1.64) | 0.017 | ||
| ET intubation | 2.68 (2.05, 3.49) | < 0.001 | ||
| Lesion complexity types | ||||
| C versus A | 7.74 (3.45, 17.4) | < 0.001 | 2.21 (1.39, 3.51) | 0.001 |
| B2 versus A | 2.88 (1.25, 6.62) | 0.013 | 1.59 (0.98, 2.59) | 0.062 |
| B1 versus A | 1.05 (0.43, 2.59) | 0.919 | 1.07 (0.64, 1.79) | 0.804 |
Note: Abbreviations as in Table 1.
TABLE 4.
Comparison of factors associated with in‐hospital mortality and 1‐year all‐cause mortality: multivariate analysis.
| Factors | In‐hospital mortality | One‐year all‐cause mortality | ||
|---|---|---|---|---|
| OR (95% CI) | P‐value | HR (95% CI) | P‐value | |
| PCI center volume, PCIs/yr | ||||
| ≥ 500 versus < 200 | 0.71 (0.42, 1.19) | 0.189 | 0.87 (0.69, 1.1) | 0.228 |
| 200 − 499 versus < 200 | 0.89 (0.50, 1.57) | 0.669 | 1.00 (0.77, 1.29) | 0.959 |
| Experience of operator, yrs | ||||
| ≥ 5 versus < 5 | 0.76 (0.58, 1.00) | 0.045 | 0.94 (0.83, 1.05) | 0.234 |
| Number of PCIs/operator/yr | ||||
| ≥ 75 versus < 75 | 0.85 (0.54, 1.32) | 0.448 | 1.11 (0.91, 1.35) | 0.342 |
| Age, per 5 years | 1.21 (1.15, 1.27) | < 0.001 | 1.16 (1.13, 1.19) | < 0.001 |
| Male | 0.77 (0.61, 0.97) | 0.024 | 0.91 (0.83, 1.00) | 0.048 |
| BMI, kg/m2 | 0.96 (0.95, 0.97) | < 0.001 | ||
| Admission SBP, mmHg | 0.95 (0.93, 0.97) | < 0.001 | 0.98 (0.98, 0.99) | < 0.001 |
| Admission HR, bpm | 1.16 (1.10, 1.21) | < 0.001 | 1.14 (1.12, 1.17) | < 0.001 |
| Refer case | 0.50 (0.39, 0.64) | < 0.001 | ||
| Health coverage scheme | ||||
| Government service/state enterprise versus UC | 1.05 (0.80, 1.38) | 0.752 | 0.87 (0.78, 0.96) | 0.007 |
| Social security service versus UC | 1.41 (0.83, 2.40) | 0.210 | 1.15 (0.92, 1.45) | 0.237 |
| Uninsured or self‐pay versus UC | 1.31 (0.58, 2.94) | 0.522 | 0.80 (0.56, 1.13) | 0.191 |
| Hypertension | 1.20 (1.07, 1.34) | 0.002 | ||
| CKD | 1.82 (1.44, 2.32) | < 0.001 | 2.00 (1.80, 2.21) | < 0.001 |
| DLP | 0.72 (0.57, 0.90) | 0.004 | 0.76 (0.69, 0.84) | < 0.001 |
| PAD | 2.67 (1.53, 4.67) | 0.001 | 2.06 (1.66, 2.57) | < 0.001 |
| DM | 1.12 (0.89, 1.41) | 0.346 | 1.21 (1.10, 1.33) | < 0.001 |
| Previous MI | 0.82 (0.73, 0.93) | 0.002 | ||
| Previous CVD | 1.26 (1.08, 1.47) | 0.004 | ||
| Known CAD | 1.26 (1.10, 1.43) | 0.001 | ||
| Prior heart failure | 1.43 (1.29, 1.60) | < 0.001 | ||
| CAD presentation | ||||
| STEMI versus NSTEMI/UA | 3.35 (1.99, 5.65) | < 0.001 | 1.51 (1.26, 1.81) | < 0.001 |
| NSTEMI/UA versus stable CAD | 2.24 (1.38, 3.63) | 0.001 | 1.14 (0.99, 1.32) | 0.072 |
| Emergency/urgent versus elective | 2.68 (1.79, 4.01) | < 0.001 | 1.30 (1.13, 1.50) | < 0.001 |
| Radial versus femoral access | 0.75 (0.58, 0.98) | 0.035 | 0.80 (0.72, 0.88) | < 0.001 |
| Cardiogenic shock before PCI | 2.56 (1.96, 3.34) | < 0.001 | 1.58 (1.38, 1.81) | < 0.001 |
| IABP | 2.73 (2.06, 3.62) | < 0.001 | 1.49 (1.28, 1.74) | < 0.001 |
| Clopidogrel | 0.70 (0.54, 0.90) | 0.005 | ||
| Prasugrel | 0.28 (0.14, 0.56) | < 0.001 | ||
| Ticagrelor | 0.60 (0.48, 0.76) | < 0.001 | ||
| Arrhythmia | ||||
| with treatment versus no arrhythmia | 2.00 (1.47, 2.73) | < 0.001 | 1.64 (1.40, 1.92) | < 0.001 |
| without treatment versus no arrhythmia | 1.72 (1.10, 2.70) | 0.019 | 1.52 (1.18, 1.95) | 0.001 |
| Disease vessel | ||||
| LM versus SVD | 1.44 (1.03, 2.01) | 0.037 | 1.29 (1.11, 1.49) | 0.001 |
| TVD versus SVD | 0.85 (0.62, 1.15) | 0.277 | 1.02 (0.90, 1.16) | 0.785 |
| DVD versus SVD | 1.02 (0.76, 1.37) | 0.925 | 1.05 (0.92, 1.19) | 0.505 |
| FFR‐guided | 0.52 (0.31, 0.87) | 0.013 | ||
| ET intubation | 5.42 (4.17, 7.04) | < 0.001 | 2.55 (2.21, 2.93) | < 0.001 |
| PCI failure | 2.17 (1.49, 3.17) | < 0.001 | 1.57 (1.35, 1.84) | < 0.001 |
| Procedural complications | 2.02 (1.46, 2.78) | < 0.001 | 1.21 (1.04, 1.40) | 0.017 |
| New requirement for dialysis | 3.07 (1.78, 5.29) | < 0.001 | 1.92 (1.50, 2.45) | < 0.001 |
| Vascular complication required treatment | 1.88 (1.13, 3.15) | 0.016 | ||
| Temporary pacemaker | 0.75 (0.58, 0.98) | 0.030 | ||
Note: Abbreviations as in Table 1.
4. Discussion
This study investigated the correlations among PCI center volumes, operator experience, and short‐ and long‐term clinical outcomes by utilizing a nationwide PCI registry. Our findings revealed no significant associations between PCI center volume and PCI outcome such as PCI failure, procedural complications, PCI‐related in‐hospital mortality, and 1‐year all‐cause mortality. However, the study highlighted the significance of operator experience as a pivotal factor. We observed that more years of practice was associated with lower in‐hospital mortality. Additionally, operators performing a higher number of PCIs annually tended to have lower PCI failure rates. Despite minor absolute differences in risk among operators, which might partly be explained by unmeasured variations in case complexity, an inverse relationship persisted even in risk‐adjusted analyses.
Several other countries have established registries to enable the analysis of clinical outcomes, which have subsequently guided their local recommendations and contributed to improvements in their national programs [18, 19, 20, 21]. Recent analyses have shown that operators with greater experience tend to have fewer complications, improved procedural success rates, and reduced mortality [13, 22]. Additionally, experienced operators often exhibit better proficiency in handling complex lesions (e.g., left main or CTO) resulting in better outcomes for challenging cases [1, 23]. PCI volume has frequently served as a surrogate for measuring quality due to its ease of measurement, with previous studies indicating correlations with outcomes at both operator and center levels [24, 25, 26, 27]. Analyzing volume outcomes also offers a means of benchmarking performance at both the center and operator levels. Nevertheless, it is crucial to acknowledge that discrepancies may occur in center‐ or operator‐volume outcomes at regional levels. Disparities in PCI volumes across different geographical regions might explain variation in the volume‐outcome relationships in existing literature [4, 6, 10, 14, 15], especially since what defines a high‐volume center in one country may not represent the same volume status in another country.
Many PCI operators in the United States have been performing fewer PCI procedures annually than originally recommended, which led to revised ACCF/AHA/SCAI guidelines in 2013, replacing the 2011 recommendations [8], reducing the threshold for PCI center volume from > 400/year to > 200/year, and the annual number of PCIs per operator from > 75/year to > 50/year. In contrast, the ESC/EACTS guidelines in 2018 for myocardial revascularization [28] state that a minimum of 75 PCIs per year is necessary to maintain proficiency. Meanwhile, the number of PCI procedures done annually in Thailand rises steadily. To uniformly classify PCI centers, we have adjusted the criteria for PCI center volume, setting it higher (> 500 PCIs/year) for high‐volume centers while maintaining < 200 PCIs/year for low‐volume PCI centers. Furthermore, regarding operator case volume, we have retained the criterion of > 75 PCIs/year for high volume instead of > 50/year as outlined in the 2013 guidelines. This adjustment aligns with a previous report that showed an association between center volumes of < 400 PCIs/year and operator volumes of < 75 PCIs/year to be consistently linked to higher rates of inpatient mortality and adverse events [24]. In our study, skilled operators performing more than 75 PCIs/year tended to have lower PCI failure rates. However, there were no differences in procedural complications, PCI‐related in‐hospital mortality, nor 1‐year all‐cause mortality.
In addition to considering operator case volume, our analysis addresses a gap in current data by examining the interplay between volume‐outcome relationships and the lifetime experience of operators. The existing evidence lacks insight into how an operator's lifetime experience impacts PCI outcomes [13]. Therefore, we included years of experience as an interventionist as one of the factors to assess PCI outcomes. It seems logical to assume that lifetime learning may play an important role in reducing adverse events and complications, potentially beyond the experience gained through procedural volume. Our findings indicated a direct correlation between operators with more years of experience and reduced PCI‐related in‐hospital mortality. However, these experienced operators, more than those with less experience, may have encountered more challenging cases involving complex lesions or patients with unstable hemodynamics which necessitated hemodynamic support devices or plaque modification devices combined with lesion severity assessment [29, 30]. Consequently, patients treated by these experienced operators may have included more with procedural complications and major bleeding, and potentially more negative long‐term outcomes such as nonfatal MI and stroke. Nevertheless, after adjusting for all confounding risk factors, the PCI‐related in‐hospital mortality of patients treated by these experienced operators remained less, as compared to those treated by other operators.
In terms of PCI center volume, evidence supports a relationship between volume and PCI outcome [31, 32, 33]. Hospitals handling higher volumes of PCI cases generally demonstrate lower mortality rates, fewer complications, and improved procedural success. Furthermore, high‐volume PCI centers often possess experienced operators, specialized resources, experienced teams, established protocols, and streamlined processes. Conversely, low‐volume PCI centers typically had less experienced operators, often younger interventionists who had just finished their training programs. However, in our study, we observed no significant impact of either the main effect of PCI center volume or the interaction between PCI volume and operator experience on any aspect of PCI outcomes after adjusting for confounding factors.
Several explanations may account for positive PCI outcomes in our low‐volume PCI centers. First, case selection played a pivotal role. Young operators often initiated their practice with simpler cases and referred more complex cases to high‐volume PCI centers with more experienced operators. Second, the presence of on‐site proctorship was noteworthy. Recently graduated interventionists frequently maintained relationships with senior mentors and invited them to assist with challenging cases, providing invaluable guidance. Our results support a recent article highlighting the value of operator experience and the added safety benefits for less experienced operators due to considerable support from more senior operators [34]. Third, an efficient training program contributes significantly. The integration of advanced stent technology, specialized equipment, and increased utilization of radial access can streamline procedures [35, 36], even within low‐volume centers. Fourth, swift access to treatment is facilitated by a rapid referral system and universal health insurance coverage, which is particularly important for initiating treatment of prevalent STEMI cases in low‐volume centers. Despite the fact that a higher annual hospital volume of primary PCI (> 36/year) is usually associated with lower mortality compared to those with lower volumes [37], our findings revealed no differences of the in‐hospital and 1‐year mortality rates based on either PCI volume or operator experience. Lastly, the frequent use of potent P2Y12 inhibitors [38, 39], more common in low‐volume PCI centers, is associated with the younger interventionists who conduct fewer PCIs. The better outcome after primary PCIs is likely due to a combination of these explanations.
Our study identified room for improvement for operators in low‐volume PCI centers. Over the 1‐year follow‐up period, the incidence of recurrent nonfatal MI and unplanned repeat PCIs (excluding staged PCI) was higher in low‐volume centers compared to intermediate‐ and high‐volume centers. Patients in these centers may have experienced in‐stent restenosis or stent thrombosis, necessitating repeat revascularization. Unfortunately, due to the need for further adjudication, comprehensive information regarding subsequent events could not be definitely established. Furthermore, not all centers had onsite cardiac surgery. Although there are data suggesting this may not influence outcome [40, 41], Thailand's national policy encourages all hospitals to have onsite surgery, which can handle the more complex cases. Moreover, the primary mechanical support device used is the IABP, while less than 0.1% utilized ECMO or VAD; the Impella® heart pump was not used, as it was not available in the country. Such devices should be accessible to provide backup support when performing PCIs on high‐risk patients and those who develop unstable hemodynamics.
4.1. Study Limitations
While studies utilizing PCI registries can provide valuable insights into the impact of operator experience and hospital volume on PCI outcome, there are some limitations that need to be considered when interpreting the results. First, the study focused on a Thai population and hospital system; generalizability of findings may be limited to where health and hospital system contexts are similar to those in Thailand. Second, not all PCI facilities are required to participate, and among those that do, not all consecutive procedures were registered. This lack of comprehensive registration could introduce selection bias. Third, the study's criteria regarding PCI center volume and the number of PCIs/operator limit comparability with other studies. Additionally, being a retrospective cohort study, there was some incomplete data and the analysis could not consider variables such as socioeconomic status, psychosocial factors, and concurrent treatments which may have confounded PCI outcomes. Fourth, the Registry primarily gathered quantitative procedural and outcome data. It tended to lack qualitative insights into procedural intricacies, operator proficiency, hospital resources, and team dynamics that may have impacted outcomes. Lastly, 1.2% of subjects did not complete the 12‐month assessment despite the team's effort and this incomplete follow‐up, though a small percentage, may have introduced bias into the findings.
5. Conclusions
In this national cohort of patients who underwent PCIs, center volumes did not significantly influence PCI outcomes after risk adjustment. However, it was evident that operator experience played a pivotal role. Operators with more years of practice were associated with decreased in‐hospital mortality rates, while those performing a greater number of PCIs annually tended to have fewer PCI failures. These findings offer valuable insights that can highlight areas for improvement and aid in the refinement of strategies within the Thai national PCI system, at both the center and operator level.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting information.
Acknowledgments
The study team thanks Dr. Arthur Brown for reviewing the manuscript and providing valuable comments. This project received a research grant from the Health System Research Institute, Ministry of Public Health, Thailand.
[Correction added on 21 January 2025, after first online publication: The name of the 6th author has been changed from “Songsak Kiatchoosak” to “Songsak Kiatchoosakun” in this version.]
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Xu B., Redfors B., Yang Y., et al., “Impact of Operator Experience and Volume on Outcomes After Left Main Coronary Artery Percutaneous Coronary Intervention,” JACC: Cardiovascular Interventions 9 (2016): 2086–2093. [DOI] [PubMed] [Google Scholar]
- 2. Karacsonyi J., Tsiafoutis I., Alaswad K., et al., “Association of Annual Operator Volume With the Outcomes of Chronic Total Occlusion Percutaneous Coronary Intervention,” Journal of Invasive Cardiology 34 (2022): E645–E652. [DOI] [PubMed] [Google Scholar]
- 3. Krishnamurthy A., Keeble C. M., Anderson M., et al., “Association Between Operator Volume and Mortality in Primary Percutaneous Coronary Intervention,” Open Heart 9 (2022): e002072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hannan E. L., “Coronary Angioplasty Volume‐Outcome Relationships for Hospitals and Cardiologists,” JAMA: The Journal of the American Medical Association 277 (1997): 892–898. [PubMed] [Google Scholar]
- 5. Jollis J. G., Peterson E. D., DeLong E. R., et al., “The Relation Between the Volume of Coronary Angioplasty Procedures at Hospitals Treating Medicare Beneficiaries and Short‐Term Mortality,” New England Journal of Medicine 331 (1994): 1625–1629. [DOI] [PubMed] [Google Scholar]
- 6. Badheka A. O., Patel N. J., Grover P., et al., “Impact of Annual Operator and Institutional Volume on Percutaneous Coronary Intervention Outcomes,” Circulation 130 (2014): 1392–1406. [DOI] [PubMed] [Google Scholar]
- 7. Levine G. N., Bates E. R., Blankenship J. C., et al., “2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention,” Circulation 124 (2011): 574–651. [Google Scholar]
- 8. Harold J. G., Bass T. A., Bashore T. M., et al., “ACCF/AHA/SCAI 2013 Update of the Clinical Competence Statement on Coronary Artery Interventional Procedures,” Circulation 128 (2013): 436–472. [DOI] [PubMed] [Google Scholar]
- 9. Windecker S., Kolh P., Alfonso F., et al., “2014 ESC/EACTS Guidelines on Myocardial Revascularization: The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio‐Thoracic Surgery (EACTS) Developed With the Special Contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI),” European Heart Journal 35 (2014): 2541–2619. [DOI] [PubMed] [Google Scholar]
- 10. Hulme W., Sperrin M., Curzen N., et al., “Operator Volume is Not Associated With Mortality Following Percutaneous Coronary Intervention: Insights From the British Cardiovascular Intervention Society Registry,” European Heart Journal 39 (2018): 1623–1634. [DOI] [PubMed] [Google Scholar]
- 11. Fanaroff A. C., Zakroysky P., Wojdyla D., et al., “Relationship Between Operator Volume and Long‐Term Outcomes After Percutaneous Coronary Intervention,” Circulation 139 (2019): 458–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Fanaroff A. C., Zakroysky P., Dai D., et al., “Outcomes of PCI in Relation to Procedural Characteristics and Operator Volumes in the United States,” Journal of the American College of Cardiology 69 (2017): 2913–2924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kovach C. P., O'Donnell C. I., Swat S., et al., “Impact of Operator Volumes and Experience on Outcomes After Percutaneous Coronary Intervention: Insights From the Veterans Affairs Clinical Assessment, Reporting and Tracking (CART) Program,” Cardiovascular Revascularization Medicine 40 (2022): 64–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Inohara T., Kohsaka S., Yamaji K., et al., “Impact of Institutional and Operator Volume on Short‐Term Outcomes of Percutaneous Coronary Intervention,” JACC: Cardiovascular Interventions 10 (2017): 918–927. [DOI] [PubMed] [Google Scholar]
- 15. Lee J.‐H., Eom S.‐Y., Kim U., et al., “Effect of Operator Volume on In‐Hospital Outcomes Following Primary Percutaneous Coronary Intervention for ST‐Elevation Myocardial Infarction: Based on the 2014 Cohort of Korean Percutaneous Coronary Intervention (K‐PCI) Registry,” Korean Circulation Journal 50 (2020): 133–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Hannan E. L. and Wu C., “Assessing Quality and Outcomes for Percutaneous Coronary Intervention: Choosing Statistical Models, Outcomes, Time Periods, and Patient Populations,” American Heart Journal 145 (2003): 571–574. [DOI] [PubMed] [Google Scholar]
- 17. Sansanayudh N., Srimahachota S., Chandavimol M., et al., “Multi‐Center, Prospective, Nationwide Coronary Angioplasty Registry in Thailand (Thai PCI Registry): Registry Design and Rationale,” Journal of the Medical Association of Thailand 104 (2021): 1678–1685. [Google Scholar]
- 18. Aikawa T., Yamaji K., Nagai T., et al., “Procedural Volume and Outcomes After Percutaneous Coronary Intervention for Unprotected Left Main Coronary Artery Disease — Report From the National Clinical Data (J‐PCI Registry),” Journal of the American Heart Association 9 (2020): e015404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Inohara T., Kohsaka S., Spertus J. A., et al., “Comparative Trends in Percutaneous Coronary Intervention in Japan and the United States, 2013 to 2017,” Journal of the American College of Cardiology 76 (2020): 1328–1340. [DOI] [PubMed] [Google Scholar]
- 20. Rashid M., Ludman P. F., and Mamas M. A., “British Cardiovascular Intervention Society Registry Framework: A Quality Improvement Initiative on Behalf of the National Institute of Cardiovascular Outcomes Research (NICOR),” European Heart Journal ‐ Quality of Care and Clinical Outcomes 5 (2019): 292–297. [DOI] [PubMed] [Google Scholar]
- 21. Shin D.‐H., Kang H.‐J., Jang J.‐S., et al., “The Current Status of Percutaneous Coronary Intervention in Korea: Based on Year 2014 & 2016 Cohort of Korean Percutaneous Coronary Intervention (K‐PCI) Registry,” Korean Circulation Journal 49 (2019): 1136–1151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Fanaroff A. C., Zakroysky P., Wojdyla D., et al., “Relationship Between Operator Volume and Long‐Term Outcomes After Percutaneous Coronary Intervention,” Circulation 139 (2019): 458–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Zein R., Seth M., Othman H., et al., “Association of Operator and Hospital Experience With Procedural Success Rates and Outcomes in Patients Undergoing Percutaneous Coronary Interventions for Chronic Total Occlusions,” Circulation: Cardiovascular Interventions 13 (2020): e008863. [DOI] [PubMed] [Google Scholar]
- 24. Hannan E. L., Wu C., Walford G., et al., “Volume‐Outcome Relationships for Percutaneous Coronary Interventions in the Stent Era,” Circulation 112 (2005): 1171–1179. [DOI] [PubMed] [Google Scholar]
- 25. McGrath P. D., “Relation Between Operator and Hospital Volume and Outcomes Following Percutaneous Coronary Interventions in the Era of the Coronary Stent,” Journal of the American Medical Association (Chicago, IL) 284 (2000): 3139–3144. [DOI] [PubMed] [Google Scholar]
- 26. Spaulding C., Morice M.‐C., Lancelin B., et al., “Is the Volume‐Outcome Relation Still an Issue in the Era of PCI With Systematic Stenting? Results of the Greater Paris Area PCI Registry,” European Heart Journal 27 (2006): 1054–1060. [DOI] [PubMed] [Google Scholar]
- 27. Vakili B. A., Kaplan R., and Brown D. L., “Volume‐Outcome Relation for Physicians and Hospitals Performing Angioplasty for Acute Myocardial Infarction in New York State,” Circulation 104 (2001): 2171–2176. [DOI] [PubMed] [Google Scholar]
- 28. Neumann F.‐J., Sousa‐Uva M., Ahlsson A., et al, “2018 ESC/EACTS Guidelines on Myocardial Revascularization,” European Heart Journal 40 (2019): 87–165. [DOI] [PubMed] [Google Scholar]
- 29. Kinnaird T., Gallagher S., Sharp A., et al., “Operator Volumes and in‐Hospital Outcomes,” JACC: Cardiovascular Interventions 14 (2021): 1423–1430. [DOI] [PubMed] [Google Scholar]
- 30. Choi K. H., Lee S. Y., Song Y. B., et al., “Prognostic Impact of Operator Experience and IVUS Guidance on Long‐Term Clinical Outcomes After Complex PCI,” JACC: Cardiovascular Interventions 16 (2023): 1746–1758. [DOI] [PubMed] [Google Scholar]
- 31. Badheka A. O., Patel N. J., Grover P., et al., “Impact of Annual Operator and Institutional Volume on Percutaneous Coronary Intervention Outcomes,” Circulation 130 (2014): 1392–1406. [DOI] [PubMed] [Google Scholar]
- 32. Srinivas V. S., Hailpern S. M., Koss E., Monrad E. S., and Alderman M. H., “Effect of Physician Volume on the Relationship Between Hospital Volume and Mortality During Primary Angioplasty,” Journal of the American College of Cardiology 53 (2009): 574–579. [DOI] [PubMed] [Google Scholar]
- 33. O'Neill W. W., “A Case Against Low‐Volume Percutaneous Coronary Intervention Centers,” Circulation 120 (2009): 546–548. [DOI] [PubMed] [Google Scholar]
- 34. Rymer J. A., Narcisse D. I., Chen A., et al., “Case Volumes and Outcomes Among Early‐Career Interventional Cardiologists in the United States,” Journal of the American College of Cardiology 83 (2024): 1990–1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Bajraktari G., Rexhaj Z., Elezi S., et al., “Radial Access for Coronary Angiography Carries Fewer Complications Compared With Femoral Access: A Meta‐Analysis of Randomized Controlled Trials,” Journal of Clinical Medicine 10 (2021): 2163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Doll J. A., Beaver K., Naranjo D., et al., “Trends in Arterial Access Site Selection and Bleeding Outcomes Following Coronary Procedures, 2011–2018,” Circulation: Cardiovascular Quality and Outcomes 15 (2022): e008359. [DOI] [PubMed] [Google Scholar]
- 37. Kontos M. C., Wang Y., Chaudhry S. I., Vetrovec G. W., Curtis J., and Messenger J., “Lower Hospital Volume Is Associated With Higher in‐Hospital Mortality in Patients Undergoing Primary Percutaneous Coronary Intervention for ST‐Segment–Elevation Myocardial Infarction,” Circulation: Cardiovascular Quality and Outcomes 6 (2013): 659–667. [DOI] [PubMed] [Google Scholar]
- 38. Wiviott S. D., Antman E. M., Gibson C. M., et al., “Evaluation of Prasugrel Compared With Clopidogrel in Patients With Acute Coronary Syndromes: Design and Rationale for the TRial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet InhibitioN With Prasugrel Thrombolysis In Myocardial Infarction 38 (TRITON‐TIMI 38),” American Heart Journal 152 (2006): 627–635. [DOI] [PubMed] [Google Scholar]
- 39. Wallentin L., Becker R. C., Budaj A., et al., “Ticagrelor Versus Clopidogrel in Patients With Acute Coronary Syndromes,” New England Journal of Medicine 361 (2009): 1045–1057. [DOI] [PubMed] [Google Scholar]
- 40. Aversano T., Lemmon C. C., and Liu L., “Outcomes of PCI at Hospitals With or Without On‐Site Cardiac Surgery,” New England Journal of Medicine 366 (2012): 1792–1802. [DOI] [PubMed] [Google Scholar]
- 41. Moady G., Ovdat T., Rubinshtein R., et al., “The Impact of On‐Site Cardiac Surgical Backup on Clinical Outcomes of Acute Coronary Syndrome‐Analysis of the ACSIS National Registry,” Frontiers in Cardiovascular Medicine 10 (2023): 1207473. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.