Impact of the caFFR-Guided Functional SYNTAX Score on Ventricular Tachycardia/Fibrillation Development in Patients With Acute Myocardial Infarction

Aims To explore the relationship between the severity of coronary artery disease (CAD) and the occurrence of ventricular tachycardia/ventricular fibrillation (VT/VF) in patients with acute myocardial infarction (AMI). Methods We retrospectively enrolled 705 patients with AMI, who were hospitalized and underwent percutaneous coronary intervention (PCI), in Nanfang Hospital from July 2017 to July 2020. Logistic regression analysis and backward stepwise approach were taken to select the correlation factors. The left and the receiver operating characteristic curves (ROC) analysis were plotted to observe the discriminative power of the SYNTAX score (SS)/caFFR-guided functional SS (FSScaFFR) on the incident VT/VF. Results About 58 (8.2%) patients experienced life-threatening VT/VF. The FSScaFFR (OR: 1.155; 95% CI: 1.047 to 1.273; p = 0.004) was an independent predictor of VT/VF after AMI. The ROC analysis showed that the discriminative power of FSScaFFR on the incident VT/VF was significantly better than SS (0.759 vs.0.695, p < 0.0001). Patients with VT/VF were categorized into 2 groups according to the interval between the onset of AMI and the VT/VF. The logistic regression analysis revealed that FSScaFFR was a significant independent correlation of early- and late-VT/VF. Conclusion The incident VT/VF in patients with AMI is closely associated with the severity of CAD evaluated by SS and FSScaFFR. Compared to SS, FSScaFFR has a higher correlation with VT/VF, and FSScaFFR was demonstrated to be an independent correlation factor of incident VT/VF after AMI.


INTRODUCTION
Ventricular tachycardia/ventricular fibrillation (VT/VF) is one of the most severe complications in patients with acute myocardial infarction (AMI). Patients with AMI who develop VT/VF often show long-term hospitalization and a poor prognosis (1). Early identification of individuals at high risk of VT/VF may improve the prognosis of patients with AMI. Although many recent studies have described the risk factors for VT/VF after AMI, the relationship between the severity of coronary artery disease (CAD) and the occurrence of VT/VF in patients with AMI remains unclear.
The SYNTAX score (SS) is an anatomical scoring system that indicates the severity of coronary artery lesions according to the number of lesions, their functional role, location, and complexity (2). However, in many cases, the anatomic lesion severity does not match the coronary physiology using fractional flow reserve (FFR). Therefore, SS, based solely on anatomical information, may lead to an erroneous assessment of lesion complexity, limiting the discriminant ability of the SS. Thus, in 2011, the concept of the functional SS (FSS) was developed (SS is only calculated in vessels with low FFR) by Nam et al., and they demonstrated that the FSS has a better prognostic implication compared to classic SS (3).
Computational pressure-flow dynamics-derived FFR (caFFR) is derived without using a pressure wire or hyperemia induction. A recent study has validated that caFFR has high accuracy, sensitivity, and specificity, with FFR as the reference standard (4). Considering that caFFR is more economical and safer than traditional FFR, FSS guided by caFFR (FSS caFFR ) is a faster and more accurate indicator to assess the severity of the patient's coronary artery. To our best knowledge, no literature report has evaluated the relationship between FSS caFFR and VT/VF. Therefore, this study is aimed to investigate the impact of CAD severity on the incident VT/VF by using SS and FSS caFFR to evaluate the coronary conditions in patients with AMI. It also aimed to explore which index could better guide the clinical practice to reduce the occurrence of VT/VF and improve the prognosis of patients with AMI.

Study Population
We retrospectively enrolled 759 patients with AMI, who were hospitalized and had undergone percutaneous coronary intervention (PCI) at Nanfang Hospital of Southern Medical University from July 2017 to July 2020. Patients who met any of the following conditions were excluded: received thrombolysis or stent implantation or coronary artery bypass grafting (CABG) treatment in the past; developed AMI that was not caused by coronary stenosis or occlusion (such as myocardial infarction with no obstructive coronary atherosclerosis (MINOCA), acute myocarditis, aortic dissection); diagnosed with atrial or ventricular arrhythmia before admission; had severe valvular heart disease or congenital heart disease; had a severe infection or advanced cancer, malignant tumors, and other diseases at the time of admission; or had missed or incomplete clinical data. The study protocol was reviewed and approved by the Local Ethics Committee of the Southern Medical University in accordance with the Declaration of Helsinki.
The acute myocardial infarction was defined as cardiomyocyte necrosis in a clinical setting, consistent with acute myocardial ischemia. A combination of criteria is required to meet the diagnosis of AMI-i.e., the detection of an increase and/or decrease in a cardiac biomarker, preferably high-sensitivity cardiac troponin (hs-cTn) T or I, with at least one value above the 99th percentile of the upper reference limit and at least one of the following: symptoms of myocardial ischemia; new ischemic ECG changes; development of pathological Q waves on ECG; imaging evidence of loss of viable myocardium or new regional wall motion abnormality in a pattern consistent with an ischemic etiology; intracoronary thrombus detected on angiography or autopsy (5). The VT/VF was defined as a fatal ventricular tachycardia or fibrillation that lasted longer than 30 s or required electrical cardioversion and/or antiarrhythmic drugs. The observation time was the period from the onset of AMI to hospitalization. Additionally, to study the related factors affecting the onset time of VT/VF, we divided the patients with VT/VF into two groups according to the interval between the onset of AMI and the VT/VF: within 0-48 h after the onset of AMI (the early-VT/VF group) and 48 h after the onset of AMI (the late-VT/VF group). The clinical features were compared with those of the patients with non-VT/VF for each group to clarify the independent correlation factors of VT/VF.

Data Collection
In this study, the baseline characteristics, medical history data, and treatment strategy of patients were acquired through hospitalized medical records. Venous blood samples were collected from all the patients within 2 h after admission, and complete blood counts and blood biochemical parameters were measured. The peak N-terminal precursor brain natriuretic peptide (NT-proBNP), peak creatine kinase myocardial band (CK-MB), and peak cTnI of the patient were determined according to the blood biochemical results monitored during hospitalization. The estimated glomerular filtration rate (eGFR) was calculated by using the modification of diet in renal disease (MDRD) equation (6).
All the patients had completed echocardiographic testing within 48 h after admission and had undergone coronary angiography using the femoral or radial approaches. The procedure was performed by interventional cardiologists according to published guidelines, institutional policy, and routine practice.
Off-line caFFR assessment was performed by blinded experienced analysts certified in using the software with the FlashAngio system (including the FlashAngio console, FlashAngio software, and Flash Pressure transducer; Rainmed Ltd., Suzhou, China). The details of caFFR measurement were the same as those of the previously reported method (4). Fluoroscopic visualizations were analyzed by 2 independent and experienced interventional cardiologists who were blinded to all the data. In the case of disagreement on visual evaluation, the final decision was made by consensus. Each coronary (including the infarct-related artery) lesion that constituted luminal obstruction ≥ 1.5 mm in diameter and ≥ 50% stenosis was scored to count the SS using the online calculator version of 2.28 (www.syntaxscore.com). The FSS caFFR was defined as modified SS measured only in lesions with a caFFR of ≤0.8.

Statistical Analysis
The normal distribution of the data was tested using the one-sample Kolmogorov-Smirnov test. Continuous variables conforming to a normal distribution were expressed as means ± SD; otherwise, they were expressed as medians with an interquartile range. Categorical variables were summarized using proportions. Characteristics between the patients with and without VT/VF were analyzed using the chi-squared test for categorical variables and the t-test or the Mann-Whitney U test for continuous variables.
Variables that were statistically significant in the univariable logistic analysis were included in the multivariable logistic analysis. Stepwise backward likelihood ratio regression was performed to screen the variables by successively removing non-significant (p > 0.05) covariates until all the remaining variables were statistically significant. Notably, variables such as intra-aortic balloon pump (IABP), mechanical ventilation, temporary pacing, hemodialysis, and continuous renal replacement therapy (HD/CRRT) were not included in this analysis because they were more likely to be caused by VT/VF. Receiver operating characteristic curves (ROC) were plotted to show the power of SS and FSS caFFR to predict incident VT/VF. In all analyses, p < 0.05 was considered statistically significant. All analyses were conducted using SPSS (V. 22.0) and MedCalc (V.15).

Population Characteristics
Among the 705 included patients with AMI, 58 (8.2%) patients developed VT/VF (Figure 1). The details of these patients are listed in Table 1 and Supplementary Table 1.
Compared with the non-VT/VF group, the patients with VT/VF had a higher Killip class and heart rate on admission and lower systolic blood pressure (SBP) and diastolic blood pressure (DBP). Patients with ST-elevation myocardial infarction (STEMI) and patients with diabetes were more likely to develop VT/VF. Regarding laboratory examinations, higher peak CK-MB, peak cTnI, peak NT-proBNP, white blood cell (WBC), C-reactive protein (CRP), uric acid, potassium (K + ), magnesium (Mg 2+ ), and lower eGFR values were found among patients who had developed VT/VF. Additionally, the patients with VT/VF had a lower LVEF and a higher incidence of the left ventricular aneurysm.
Regarding the therapies conducted in these patients, β-blockers and angiotensin-converting enzyme inhibitors/angiotensin receptor blockers (ACEIs/ARBs) were used less frequently among the patients with VT/VF. By contrast, the use of diuretics, IABP, mechanical ventilation, temporary pacing, and HD/CRRTwas higher in the patients with VT/VF.
Concerning coronary angiographic characteristics, the incidence of left main coronary artery (LMCA) disease and post-PCI TIMI Grade < 3 was significantly higher in the VT/VF group. The patients who developed VT/VF had more coronary artery lesions and received more stents during the procedure. Notably, SS and FSS caFFR were significantly higher in the patients with VT/VF than those without VT/VF [32.6 ± 13.5 vs. 24.4 ± 11.7 (p <0.001) and 31.8 ± 9.1 vs. 18. ± 7.6 (p <0.001), respectively].

Correlation Factors of Incident VT/VF
The univariable logistic analysis date is detailed in Table 2

The Discriminative Power of SS/FSS caFFR and Risk Stratification of VT/VF
To further explore the relationship between SS/FSS caFFR and VT/VF, 705 patients were divided into tertiles (intertertile range: 19 to 28.5) of risk based on SS, i.e., low, medium, and high SS  (33.6%, n = 237; 32.8%, n = 231; 33.6%, n = 237, respectively), and were analyzed. After calculating FSS caFFR , 22.4% of the highrisk SS group moved to the medium-risk FSS caFFR group and 13.1% moved to the low-risk FSS caFFR group, whereas 37.6% of the medium-risk SS group moved to the low-risk FSS caFFR group (Supplementary Figure 1). The VT/VF occurred in 3.4, 6.9, and 14.3% of patients with low, medium, and high SS, respectively (p < 0.0001). A similar trend was observed in the FSS caFFR group (2.3, 10.2, and 19.6% in the low-, medium-, and high-FSS caFFR groups, respectively) (p < 0.0001) (Supplementary Figure 2).
The area under the curve (AUCs) of SS and FSS caFFR were 0.695 and 0.759, respectively, and both showed statistically significant differences (p <0.001). Additionally, FSS caFFR demonstrated better discriminative power than SS,with a z-statistic of 6.349 (p < 0.0001) (Figure 2). Furthermore, we divided the participants into two score categories: the low-risk group (scores <25, risk = 4.20%) and the high-risk group (scores ≥ 25, risk = 19.70%) according to the cutoff value of FSS caFFR (Figure 3). The sensitivity and specificity were 67.2% and 74.2%, respectively ( Table 4).

DISCUSSION
The present study is the first to show the correlation between FSS based on caFFR and incident VT/VF in patients with AMI. FSS caFFR was an independent relevant factor of incident VT/VF after AMI. The patients with a higher FSS caFFR of more than 25 showed a higher risk of VT/VF. We also found a significant independent correlation between FSS caFFR and early-and lateonset VT/VF. Overall, our study indicated that patients with AMI with higher FSS caFFR should be alerted to the incident VT/VF, and timely intervention measures should be conducted to prevent VT/VF in clinical practice. In our study, 8.2% of patients with AMI developed lifethreatening VT/VF, a value approximately similar to that reported in the previous studies (1,7). Some clinical indices associated with incident VT/VF have been recognized in patients with AMI. Consistent with the results of previous studies, patients with STEMI (8-10), patients with lower SBP (1,11,12), and patients with reduced LVEF (13) showed an increased risk of incident VT/VF. We found that a higher WBC value showed a higher occurrence rate of VT/VF. Early studies have indicated that, in the acute phase of AMI, neutrophils begin to gather, denature, and infiltrate in the infarcted area under the action of cytokines and chemokines while releasing many inflammatory mediators, reactive oxygen species, and leukotrienes. These events cause the disturbance of local microcirculation and the disorder of cardiac electrophysiological activity, leading to the occurrence of VT/VF (14)(15)(16). The positive correlation between ventricular arrhythmias and diuretic use in patients with AMI is also described in the present study (17).
To date, no studies have elaborated on the relationship between SS and VT/VF in patients with AMI. Although SS in the present study was correlated with VT/VF, no statistical significance was found after adjusting for various confounding factors. The AUC value of SS is lower than 0.7, which might be explained by previous studies reporting that many significantly anatomic lesions had no significant functional stenosis (18,19).
In 2010, the Fractional Flow Reserve vs. Angiography for Multivessel Evaluation (FAME) study demonstrated that FFRguided PCI could improve the long-term prognosis of patients with multivessel CAD (20). Additionally, FSS as a combination concept of FFR and SS was introduced by Nam et al. (3). The FSS is not only concerned with the anatomic severity but also provided information on coronary physiology. According to the literature, FSS is more instructive in assessing coronary artery conditions and has better prognostic prediction power than SS (21,22). The caFFR is a new FFR evaluation method. Its assessment is primarily based on angiography images and does not require a pressure wire or hyperemia induction. A recent study has shown that caFFR with FFR as the reference standard has good diagnostic accuracy and consistency (4). Therefore, caFFR has more comprehensive applicable indications than FFR, and it has good clinical application prospects.
Our study is the first to demonstrate that FSS based on caFFR is an independent predictor for VT/VF after AMI. According to the current theory, the occurrence of VT/VF after AMI is not only related to sympathetic excitatory factors caused by acute myocardial ischemia but also associated with abnormal myocardial metabolism, abnormal automaticity, and reentrant formation caused by reperfusion injury in the ischemic area (13). Early studies have shown that the degree of coronary atherosclerosis is more severe in patients with complex coronary artery disease, leading to a weakening of the coronary response to vasodilator factors such as nitric oxide, prostacyclin, and adenosine (23). This finding might explain why patients with AMI with higher FSS caFFR are more likely to develop VT/VF. We showed that the average of FSS caFFR was similar to the SS among the patients with VT/VF, but FSS caFFR was significantly lower than SS among the patients without VT/VF. Grouping patients according to their SS/FSS caFFR can explain this phenomenon more clearly. After calculating FSS caFFR , patients whose coronary disease severity was overestimated were reassigned to the lower risk group. Although both SS and FSS caFFR had significant differences in the incidence of VT/VF between the lowrisk, medium-risk, and high-risk groups, the difference in the incidence of VT/VF was greater in the FSS caFFR group than in the SS group (Supplementary Figure 2), which indicated that FSS caFFR can better distinguish patients with different levels of severity of CAD. Additionally, ROC analysis showed that FSS caFFR has better discriminative power concerning the incident VT/VF than SS. Thus, SS combining anatomic and functional information is more relevant to incident VT/VF in patients with AMI than anatomic assessments alone.
Early studies have found that the mechanism and clinical characteristics of lethal VT/VF vary at different stages after AMI. The occurrence of VT/VF within 48 h after the onset of AMI may be more related to cardiac electrical and hemodynamic instability (24), while the occurrence of late VT/VF may be more related to overactivated inflammation and cardiac scar formation (25,26). This finding indicates that the long-term prognosis of patients with early VT/VF may be different from that of patients with late VT/VF. To further investigate the relationship between FSS caFFR and the two types of VT/VF, the patients with VT/VF were categorized into two groups. Our results revealed that high FSS caFFR is an independent risk factor for both early and late VT/VF. Thus, the more severe the functional ischemia of the coronary artery, the more likely it is to cause electrical disturbances, thereby inducing early VT/VF. Similarly, the degree of functional ischemia of the coronary artery will also affect inflammation and the formation of cardiac scars, increasing the incidence of late VT/VF. Therefore, high FSS caFFR indicates that the incidence of VT/VF will increase in both early and late

LIMITATIONS
Our study had several limitations. First, this study is a singlecenter retrospective study, and the study population was relatively small. Therefore, some bias may exist in the study population. Second, because of the limitation of retrospective data, our study did not consider the factor of microcirculation dysfunction during the period of AMI that might temporarily increase FFR of the coronary artery slightly (27,28), thereby affecting the calculation of FSS caFFR . However, notably, the current mainstream condition to determine coronary ischemia in patients with AMI is still FFR ≤0.8 (29). Therefore, we believe that the results of this study are still reasonable and reliable. Third, the aortic root pressure was not measured in real-time, and it was obtained from the interventional database. Additionally, the hemodynamics of patients with AMI were very unstable. Therefore, it might slightly influence the results of caFFR because a real-time measured aortic root blood pressure can help obtain a more accurate result of caFFR (30). We believe that more prospective studies are needed to verify these findings in the future.
Frontiers in Cardiovascular Medicine | www.frontiersin.org that FSS caFFR has a higher correlation with VT/VF than SS. Additionally, FSS caFFR was an independent correlation factor of VT/VF, regardless of whether VT/VF occurred in the early or late stage after AMI.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of the Southern Medical University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AUTHOR CONTRIBUTIONS
JP and JX contributed to the conception or design of the work. JP, QZ, LL, YC, GL, HL, JL, XZ, YT, JP, YY, and DM contributed to the acquisition, analysis, or interpretation of data for the work. JP drafted the manuscript. JP, QZ, LL, and JX critically revised the manuscript. All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation, gave final approval and agreed to be accountable for all aspects of work, ensuring integrity, and accuracy.