Ten‐year clinical outcomes in patients with intermediate coronary stenosis according to the combined culprit lesion

Abstract Background We assessed the long‐term clinical outcomes of an intermediate lesion (IL) according to the presence of a combined culprit lesion (CCL). Hypothesis Long‐term clinical outcomes of IL may be affected by the presence of a CCL. Methods Angiographic findings (n = 1096) and medical chart were reviewed. Patients with IL were divided into two groups: IL without CCL group (n = 383, 64.5%) and IL with CCL group (n = 211, 35.5%). Results The major adverse cardiovascular events (MACE) in the IL with CCL group were significantly higher than those in the IL without CCL group (death: 12.3% vs. 7.0%, myocardial infarction: 3.3%vs. 0.5%, stroke: 6.6% vs. 2.6%, and revascularization [RVSC]: 25.1% vs. 7.6%) during a mean follow up period of 118.4 ± 5.5 months. IL related RVSC rate in the IL with CCL group was higher than that in the IL without CCL group (5.7% vs. 2.1%, p = 0.020). RVSC rate related to IL in total subjects was lower than that related to stented lesion (3.4% vs. 6.4%). The important predictors of total MACE in total subjects were the presence of CCL, IL percent diameter stenosis, hypertension, history of percutaneous coronary intervention, blood glucose and ejection fraction. The predictors of IL related RVSC were IL percent diameter stenosis and IL located in the right coronary artery. Conclusion 10‐year clinical outcomes of an IL (especially IL without CCL) were better than those of stented lesions. This study suggests that the IL can be safely followed up in sites that do not have ability to assess functional study.


| INTRODUCTION
The prevalence of an intermediate lesion (IL) is relatively high, ranging from 39.4% to 79.3%. [1][2][3][4] Previous studies have evaluated the value of functional status assessment and prognostic factors of the intermediate lesion. [5][6][7] Fractional flow reserve (FFR) index is considered as a gold standard to assess whether the intermediate stenosis is responsible for inducible ischemia and whether the patient would benefit from revascularization (RVSC). Therefore, current guidelines recommend FFR measurements for the assessment of the IL. 8,9 The DEFER (deferral versus performance of balloon angioplasty in patients without documented ischemia) and FAME (fractional flow reserve versus angiography for multi-vessel evaluation) studies showed better 5-year and 2-year clinical outcomes in patients treated by FFR guided therapeutic decision than those treated by angiography guided decision. 1,2,6,10 However, recent analysis of the FAME study showed that the improved clinical outcomes of FFR-guided percutaneous coronary intervention (PCI) group, which were observed in the first 2 years after the index procedure, did not persist until 5 years. 11 And 65% (402/620) of the intermediate lesions showed FFR > 0.80 in other study. 6 Moreover, functional stenosis severity of non-culprit lesions is frequently overestimated. 12 And the utilization of FFR is increasing, however, the many cardiac laboratories still making clinical decision based on coronary angiography alone due to the equipment availability, reimbursement policies, and other financial considerations. 13,14 According to many studies, conservative medical treatment of IL might be safe and justified; therefore, performing PCI may be safely deferred in patients with such lesions. [15][16][17] We previously reported that 10-year clinical outcomes of the IL was favorable compared with those of the significantly stenotic culprit lesion; however long-term outcomes of deferred IL based on angiography alone remain unclear 18 Our previous study was performed in patients with multi-vessel disease including an IL. Therefore long-term outcomes of the IL without combined culprit lesion are unclear. Furthermore, one needs to consider the risk and benefit of PCI in those lesions such as resource use (FFR, PCI), procedure related problems and antiplatelet related problems, et al. [19][20][21] The objectives of this study were to find the real world very longterm (10 years) clinical outcomes of an angiographically intermediate lesion (IL), to find the difference of the clinical outcomes of an IL according to the combined culprit lesion (CCL) and to find predictors of the adverse clinical outcomes in a large number of patients with IL.

| Study population
This study was a non-randomized, retrospective, single center study.
We analyzed the medical records of 1096 patients who underwent coronary angiogram (CAG) in Konyang University Hospital, Daejeon, South Korea between January 2008 and December 2008. We reviewed patient demographics and laboratory and angiographic findings. IL was defined as 30% to 70% angiographic stenosis visually estimated from baseline CAG. 22 Two cardiologists (KYK and JCW) reviewed and determined angiographic stenosis. All disagreements between two cardiologists were resolved by a senior cardiologist (BJH). We excluded 502 patients without IL. Finally, a total of 594 patients were enrolled in this study. Enrolled patients were followed-up over a pre-defined follow-up period to be 10 years. The study subjects were divided into 2 groups according to the angiographically presence or absence of CCL in other major epicardial coronary arteries; IL without CCL group and IL with CCL group. The IL without CCL group included patients who had IL without other significant CCL (angiographic percent stenosis ≥70%), which was mostly treated with PCI.
The study was approved by the Institutional Review Board of Konyang University Hospital and was performed in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.

| Study endpoint
The primary outcome of this study was the occurrence of major adverse cardiovascular event (MACE), which was defined as all-cause death, myocardial infarction (MI), stroke, and revascularization (RVSC; defined as PCI or coronary artery bypass grafting involving the target lesion). Adverse events were reviewed and adjudicated by at least two independent investigators blinded to the study groups. MI was defined as chest pain with or without a ST-segment elevation ≥2 mm in ≥2 contiguous precordial leads, or ≥ 1 mm in ≥2 limb leads, or a new left bundle branch block on electrocardiogram, and elevation of cardiac enzymes at least 3 times the upper limit of the normal range.
Stroke was defined as neurological symptoms associated with radiologic findings based on computed tomography or magnetic resonance imaging. The simultaneous or sequential occurrence of two or more MACE was counted as one incidence of MACE, and time-to-event duration was defined as the duration between enrollment and the first event.

| Statistical analysis
Continuous variables are presented as mean ± standard deviation and categorical variables are presented as numbers and percentages. The Chi-square test or Fisher's exact test was performed to analyze categorical variables, and an independent t test was performed to analyze continuous variables in the univariate analysis. The Kaplan-Meier method was used to assess cumulative event rates. Multivariate Coxproportional hazard analysis was performed to investigate the independent predictors of MACE, RVSC, and IL related RVSC. Clinically relevant coronary artery disease risk factors and angiographic finding variables that showed significant relationships with MACE and RVSC in the univariate analysis (p < 0.10) were included into the multivariate analysis.
All statistical analyses were conducted using the SPSS (version 18.0), and a p-value <0.05 was considered statistically significant.

| Patient demographics
The study population including 330 (55.6%) men had a mean age of 64.4 ± 10.8 years (

| Clinical outcomes
During a mean follow-up period of 118.4 ± 5.5 months, 140 MACE occurred, including 61 events in the IL without CCL group and 79 events in the IL with CCL group ( The most common cause of RVSC was in-stent restenosis (ISR, 6.4% during 10 years of follow up) in both the IL without CCL group and IL with CCL group, but RVSC rate due to ISR was significantly higher in the IL with CCL group than in the IL without CCL group (n = 26, 12.3% vs. n = 12, 3.1%, p < 0.001) ( Table 2).
Both the rates of RVSC related to IL ( Figure S1A) and RVSC related to a new lesion, which initially appeared normal or showed less than 30% stenosis on baseline angiogram, ( Figure S1B) were also higher in the IL with CCL group than those in the IL without CCL In the IL with CCL group, the rate of RVSC due to CCLs, which was initially treated with optimal medication, was 1.9% during 10 years of follow up ( Table 2).
The Kaplan-Meier curves for the primary endpoint during 10 years of follow up are shown in Figure 1. There was significant difference in all event-free survival between the IL without CCL group and IL with CCL group. As presented in Figure 1, the 10-year cumulative total MACE-free survival was significantly higher in the IL without CCL group compared with that in the IL with CCL group (78.4% vs. 54.4%, p < 0.001) (Figure 1.A). The total RVSC-free survival in the IL without CCL group was also significantly higher than that in the IL with CCL group (89.0% vs. 66%, p < 0.001) (Figure 1.B). IL related RVSC-free survival in the IL without CCL group was higher than that in the IL with CCL group (96.9% vs. 91.0%, p = 0.015) ( Figure 1C).  (Table 3).

| Predictors of MACE and RVSC
Receiver operating characteristic (ROC) curve analysis ( Figure S2) was used to determine the performance of IL percent diameter stenosis to predict total MACE and IL related RVSC. The area under the ROC curve was 0.67 and 0.70, respectively. At a cutoff value of IL percent diameter stenosis of 47.5%, the sensitivity and specificity for predicting total MACE were 73% and 57%, respectively ( Figure S2A).
Additionally, at a cutoff value of IL percent diameter stenosis of 47.5%, the sensitivity and specificity for predicting IL related RVSC were 86% and 52%, respectively ( Figure S2B).

| DISCUSSION
The main findings of this study are described below. This study demonstrated that coronary angiographic degree of stenosis could be a predictor of MACE. However, there is some debate as to whether the severity of coronary stenosis is associated with future cardiovascular events. 3,4,23 According to previous observational studies, RCA stenosis is frequently observed in patients with coronary artery disease and RCA disease progresses more rapidly; therefore, patients with RCA stenosis may be more vulnerable than those with LAD stenosis. 24,25 In line with these studies, our study showed that the IL located in the RCA was also an important predictor of RVSC.
In the present study, the rate of new lesion related RVSC was simi- Currently, FFR and instantaneous wave-free ratio (iFR) are used as a gold standard in making therapeutic decisions for IL. 8,9,26 However, these technologies may not be available in all cardiac laboratories. In addition, the clinical application of functional study has been variable and remains underused. 13,14 The FAME and DEFER studies showed better clinical outcomes in patients treated by FFR guided therapeutic decision than those in patients treated by angiography guided decision. The FAME study showed that the benefit of PCI resulted from the routine measurement of FFR, which allowed the judicious use of stents. 2,10,11 The DEFER study showed that PCI of a functionally non-significant stenosis had no advantage and even resulted in more MI in the stented artery. 5 These findings are in accordance with those of previous studies reporting poorer prognosis of multi-vessel coronary artery disease than that of single-vessel disease. 28,29 The presence of CCL may imply a diffuse nature of coronary atherosclerosis or pathophysiological inflammatory process that involves the entire systematic cardiovascular system. Therefore, our study suggests that IL, with CCL, requires more careful surveillance and an optimal treatment strategy to prevent future cardiovascular events than IL without CCL.

| STUDY LIMITATION
This was a retrospective study; therefore, the sources of selection bias in this study could not be excluded. Only 32.8% of patients were lost to follow up but the long-term follow-up partially compensated the limitation to evaluate the natural history of IL. Our study analyzed lesion stenotic severity but not lesion length, which is also an important geometric parameter. In addition, tandem lesions are considered as one lesion which is most stenosis. Although the entire stenotic segment contributes to the resistance, the most stenotic lesion is an important factor in limiting coronary flow.
This study was conducted over 10 years. The patients were followed up and managed by a dedicated doctor, usually in outpatients clinics. The patients were managed with conventional medication, which could be changed during the long follow up period.
Another limitation of our study was that, the impact of medication on clinical outcomes was not analyzed. However, it was practically difficult for us to determine the impact of medications on clinical outcomes, especially in this long-term retrospective clinical study, because of the variations in medications over time. These limitations require confirmation in a prospectively designed study.

| CONCLUSION
Our study showed favorable very long-term (10 years) outcomes of angiographically IL, especially IL without CCL and even better than those of stented lesions. Therefore, this study suggests that optimal medical treatment without any invasive work up or intervention may be a possible therapeutic option for IL in sites that do not have facility to assess FFR.