Introduction

Significant iatrogenic coronary dissection and perforation are important technical complications that can occur during percutaneous coronary intervention (PCI). These complications are often associated with wiring techniques, increased balloon pressurization, use of inappropriately large balloons or stents, especially in patients with complex anatomical features, such as chronic total occlusion or bifurcation, or use of rotational atherectomy during the procedure1,2. Although rare, these can have serious consequences, including fatalities3,4. Studies have reported poor in-hospital outcomes, including death, owing to these complications3,4,5,6,7. However, limited information is available regarding the long-term consequences for patients who survive these complications during hospitalization.

Investigating the outcomes of patients who have experienced procedural complications is crucial because these patients may be particularly concerned about their prognosis. Accurate long-term information is necessary for clinicians so that they can deliver precise and informed guidance to patients in the event of such complications, plan follow-up visits, and implement preventive measures against complications such as heart failure. Although the benefit of studies on long-term effects of coronary dissection and perforation is expected, limited research is available, especially large-scale observational studies that include consecutive all-comer patients. Therefore, this study aimed to assess the relationship of coronary dissection and coronary perforation with long-term outcomes. The study was conducted in accordance with the principles of the 1964 Declaration of Helsinki.

Methods

Database

This study was conducted as part of the Japan Cardiovascular Database-Keio Interhospital Cardiovascular Studies (JCD-KiCS) PCI registry, which is a multicenter prospective registry that includes data of consecutive patients who have undergone PCI since 2008 at 15 institutions within the Tokyo metropolitan area. The details of this registry have been published previously8,9,10,11,12,13. The participating hospitals were instructed to document and register patient data from consecutive hospital visits for PCI, using an Internet-based data collection system. The registered data were reviewed for completeness and internal consistency. Therefore, the study was prospectively designed and retrospectively collected: we predetermined the variables to be collected beforehand, and the recording of patient data was conducted on average 2–3 months after the index procedure. Long-term outcomes, including subsequent revascularization, were reviewed at the 2-year mark after PCI.

Quality assurance of the data was achieved through automatic system validation, reporting of data completeness, and education and training of clinical research coordinators who were specifically trained to use the present PCI registry. The senior study coordinator (I.U.) and exclusive onsite auditing by the investigator (S.K.) ensured appropriate registration of each patient. All the participants provided written informed consent. Before the launch of the JCD-KiCS registry, the University Hospital Medical Information Network of Japan (UMIN000004736) provided information regarding its objective for clinical trial registration. The present study was approved by the IRB committee of Keio University (Reference number: 20080073).

Studied patients

Of the 8,792 consecutive patients registered between September 2008 and December 2017 with 2-year outcomes, we excluded 58 and 13 patients with missing information on sex and long-term outcomes, respectively, resulting in a final cohort of 8,721 patients.

Definition of outcomes and variables

The clinical variables and outcomes of the JCD-KiCS were aligned using data from the National Cardiovascular Data Registry (CathPCI Registry version 4.1). Significant coronary dissection was defined as persistent contrast medium extravasation or spiral or persistent filling defects with complete distal flow, impaired flow, or total occlusion. Coronary perforation was defined as leakage of contrast medium from the coronary artery into the surrounding tissue or body cavity. In the JCD-KiCS, all major procedural complications (e.g., death, bleeding complications, and cardiac and cerebrovascular events) were recorded by project coordinators, and the details of the procedural complications were adjudicated by an adjudicator according to a pre-defined data dictionary13. A second or third adjudicator was consulted in the event of a disagreement between the opinions of a project coordinator and the first adjudicator.

Acute coronary syndrome (ACS) was defined as ST-segment elevation myocardial infarction, non-ST-elevation myocardial infarction, or unstable angina. Stable coronary artery disease was defined as stable angina, previous myocardial infarction, or silent ischemia. Heart failure was defined as a left ventricular ejection fraction ≤ 35% or documentation of heart failure by the attending physician, regardless of the left ventricular ejection fraction14. Multivessel disease was defined as two or more major coronary arteries with ≥ 75% stenosis. The estimated glomerular filtration rate was calculated using the Modification of Diet in Renal Disease Equation for Japanese Patients proposed by the Japanese Society of Nephrology15,16,17.

After hospital discharge, we followed the participants to identify hospitalizations for cardiovascular or bleeding events and all-cause deaths via medical records, phone calls, or mail. All follow-up data at 730 days after PCI were collected and recorded using a secure Internet-based electronic data capture system by dedicated clinical research coordinators trained by the primary investigator and project coordinators18. The primary outcome was a composite of ACS, heart failure, coronary artery bypass grafting, stroke and bleeding requiring readmission, and all-cause death. The secondary outcome was a component of the primary outcome. Our captured endpoints were aligned with the NCDR CathPCI registry17. Regarding the capture of the revascularization procedure, the intensive utilization of PCI in Japan, involving routine follow-up angiograms until around 2015, led us to incorporate CABG alone as a revascularization endpoint. Our previous collaboration with NCDR CathPCI has indicated that the link between unplanned coronary revascularization with PCI and higher mortality is relatively less significant compared to the association with ACS19. Thus, our incorporation of CABG alone as a revascularization procedure can be considered an effort to address and alleviate these potential biases.

Statistical analyses

Continuous variables are presented as mean ± standard deviation or median (interquartile range), as appropriate, for data distribution. Categorical variables are expressed as percentages. Changes in continuous variables from the baseline were evaluated using Student’s t-test or the Mann–Whitney U test. Χ2 or Fisher’s exact test was used to analyze categorical variables13.

To analyze the long-term outcomes, we utilized Kaplan–Meier estimates and constructed a multivariable Cox proportional hazard model for the primary endpoint, considering the impact of coronary dissection and perforation. The following variables were included in the analysis: age, sex, body mass index, diabetes, hypertension, dialysis, prior myocardial infarction, prior heart failure, indications for PCI, PCI lesions, estimated glomerular filtration rate, hemoglobin level, and puncture site.

All statistical calculations and analyses were performed using R software v. 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria), and p < 0.05 was considered statistically significant.

Results

In our cohort of 8,721 patients, the mean age of the patients was 68.3 ± 11.3 years. Among the studied patients, 68 (0.78%) had significant coronary dissections, and 61 (0.70%) had coronary perforations. Baseline characteristics and in-hospital and long-term outcomes were compared between the patients with and without coronary dissection and between those with and without coronary perforations (Tables 1, 2, 3, 4).

Table 1 Baseline characteristics of patients with dissection and those without.
Full size table
Table 2 In-hospital and long-term outcomes of patients with dissection and those without.
Full size table
Table 3 Baseline characteristics of patients with perforation and those without.
Full size table
Table 4 In-hospital and long-term outcomes of patients with perforation and those without.
Full size table

A significantly lower proportion of males had coronary dissections, and the proportions of chronic lung disease, heart failure at admission, femoral artery approach, right coronary artery culprit vessel, use of an intra-aortic balloon pump, and chronic total occlusion, and bifurcation were higher among those with coronary dissections than among those without (Table 1). There were no significant differences in left ventricular ejection fraction between patients with coronary dissection and those without. Additionally, patients with coronary dissection had higher rates of post-PCI myocardial infarction, cardiogenic shock, heart failure, and transfusion than those without (Table 2).

The proportions of prior myocardial infarction, coronary bypass, heart failure on admission, femoral artery approach, multivessel disease, chronic total occlusion, type C lesions, and use of rotational atherectomy were higher among the patients with coronary perforations than among those without (Table 3). Moreover, the patients with coronary perforations had higher rates of post-PCI cardiogenic shock, cardiac tamponade, and transfusion, however, only 6.6% of patients with coronary perforation suffered from cardiac tamponade (Table 4).

Importantly, the patients with significant coronary dissections had higher rates of the primary endpoint (25.0% versus 14.3%, P = 0.02) and heart failure requiring readmission (10.3% versus 4.2%, P = 0.03) than those without, and there was a marginal difference in the rate of coronary bypass (4.4% vs. 1.2%, P = 0.068) between the former and latter. Interestingly, there were no significant differences in the primary and secondary outcomes between the patients with and without coronary perforations (primary outcome: 8.2% vs. 14.5%, P = 0.23).

The Kaplan–Meier curves of the primary endpoint for patients with and without coronary dissections and coronary perforations are shown in Figs. 1 and 2. The patients with coronary dissection had significantly higher rates of the primary endpoint than those without (unadjusted hazard ratio [HR]: 1.98, 95% confidence interval [CI]: 1.22–3.19; P = 0.005) at the 2-year follow-up, but there was no significant difference in the primary endpoint between those with and without coronary perforations (unadjusted HR 0.56, 95% CI 0.23–1.34; P = 0.19).

Figure 1
figure 1

Kaplan–Meier curve of long-term outcomes (primary endpoint) for patients with and without coronary dissection.

Full size image
Figure 2
figure 2

Kaplan–Meier curve of long-term outcomes (primary endpoint) for patients with and without coronary perforation.

Full size image

After adjusting the multivariable Cox proportional hazard risk model for other variables, patients with coronary dissections showed a significantly higher rate of the primary endpoint than those without (adjusted HR 1.70, 95% CI 1.02–2.84; P = 0.04) at the 2-year follow-up, but there was no significant difference in the primary endpoint rate between those with and without coronary perforation (adjusted HR 0.51, 95% CI 0.21–1.23; P = 0.13).

Discussion

The salient findings of our study are as follows: In a cohort of 8,721 PCI patients with a 2-year follow-up, significant coronary dissection (0.78% of patients) was associated with notable differences in the baseline characteristics, procedural details, and outcomes, including increased rates of post-PCI myocardial infarction, cardiogenic shock, and transfusion. Patients with significant coronary dissection also had higher rates of primary endpoints and heart failure requiring readmission than those without. However, there were no significant differences in the outcomes between the patients with and without coronary perforation. The Kaplan–Meier curves indicated a higher rate of the primary endpoint for patients with coronary dissection than for those without, confirmed by the multivariable Cox proportional hazard risk model. No difference was observed between the patients with and without coronary perforation.

Very limited long-term data are available on the impact of significant coronary dissection in all-comer PCI registries, with only one report exploring iatrogenic catheter-induced coronary dissection. However, this study focused only on catheter-induced coronary dissection of the ostium of the coronary artery. It is crucial to avoid significant coronary dissection during PCI and recognize known risk factors such as female sex, complex PCI such as chronic total occlusion, long lesions and calcification, and the use of a relatively large balloon size4,20. Female patients tend to have smaller coronary arteries, and this may cause procedure-related coronary dissections frequently.

Our findings revealed that patients with coronary dissection experienced worse outcomes, including high rates of primary endpoint and heart failure readmissions and a tendency to require coronary bypass and medication adjustment with close monitoring post-discharge. Interestingly, patients with coronary dissection did not have lower left ventricular ejection fraction compared to those without. We hypothesized that residual coronary dissection may cause myocardium ischemia, which results in diastolic dysfunction21. Since we demonstrated that patients with significant coronary dissection had a higher risk of heart failure readmission, they may need to have an appointment within 1 week after discharge or diuresis adjustment with a heart failure specialist to decrease the risk of readmission for heart failure22.

Coronary perforation is a serious complication that can lead to pericardial effusion, causing cardiac tamponade, which can be fatal if not treated promptly23,24,25. The long-term consequences of coronary perforation are not well understood, but studies on patients with chronic total occlusions have shown that perforation has a legacy effect on mortality, with an odds ratio for 12-month mortality of 1.60 for perforation survivors compared with that for those without perforation24. However, these data are limited to patients with chronic total occlusions, who could have higher risk profiles than other PCI patients because coronary perforation, especially collateral circulation, may worsen the myocardium supply (i.e., due to coil embolization). We found an overall decline in mortality related to coronary artery perforation over time26 and the risk of additional adverse events in the long term may be low, as suggested in our study. Moreover, our all-comer PCI registry showed that only a quarter of patients with coronary perforation had chronic total occlusion, and only 6.6% of the patients had developed cardiac tamponade related to coronary perforation, and this could explain why they did not experience poor long-term outcomes.

Compared to coronary dissection and perforation, which are less frequently encountered procedural complications, other complications, such as bleeding or acute kidney injury, have been extensively studied and shown to impact both in-hospital and long-term mortality27,28. For example, PCI-related bleeding complications are relatively common and well studied29,30,31. These complications can affect not only in-hospital mortality but also long-term mortality27,32. Given these poor outcomes, we implemented bleeding avoidance strategies, such as the transradial approach, to decrease the risk of bleeding complications.

Our study had several limitations. First, this was an observational study, and our analysis could not be adjusted for unmeasured confounders. Second, in our registry, the follow-up survey focused only on clinically driven events. Therefore, subsequent revascularization was retrospectively reviewed, and some revascularization events may not have been captured, especially, in cases of transfer to institutions outside of the JCD‐KiCS network. Third, we did not adjust for left ventricular ejection fraction because almost half of the whole patients and a third of patients with coronary dissection or perforation did not have information on left ventricular ejection fraction; however, left ventricular ejection fractions were similar between patients with and without coronary perforation or dissection. Fourth, we did not collect data on the treatment for coronary dissection or perforation (i.e., covered stent, coil embolization, or cutting balloon) that may have affected future events, such as revascularization6,33. In addition, we did not collect data on the detailed significance of coronary perforation (i.e., Ellis classification)34. Previous studies have suggested a correlation between the Ellis classification and long-term outcomes or the requirement for covered stents33. While our definition of coronary perforation aligns with other internationally recognized registries, the evaluation of the degree of perforation may be necessary in future studies to elucidate its precise impact on long-term implications. However, only a few patients with coronary perforation suffered from cardiac tamponade (6.6%), which may reflect a better prognosis in patients with coronary perforation than expected.

In conclusion, in patients undergoing PCI, significant coronary dissection was associated with poor long-term outcomes, but not in those with coronary perforation. Our long-term data on coronary dissection and perforation can provide valuable insights for physicians who follow patients in the post-discharge setting and promote the use of appropriate procedures (such as the use of intravascular imaging) to eliminate the risk of fatal complications.