Abstract
Background
Coronary lipid core plaque may be associated with the incidence of subsequent cardiovascular events.
Methods
We analyzed outcomes of 239 patients who underwent near-infrared spectroscopy (NIRS) coronary imaging between 2009–2011. Multivariable Cox regression was used to identify variables independently associated with the incidence of major adverse cardiovascular events (MACE; cardiac mortality, acute coronary syndromes (ACS), stroke, and unplanned revascularization) during follow-up.
Results
Mean patient age was 64 ± 9 years, 99% were men, and 50% were diabetic, presenting with stable coronary artery disease (61%) or an acute coronary syndrome (ACS, 39%). Target vessel pre-stenting median lipid core burden index (LCBI) was 88 [interquartile range, IQR 50–130]. Median LCBI in non-target vessels was 57 [IQR 26–94]. Median follow-up was 5.3 years. The 5-year MACE rate was 37.5% (cardiac mortality was 15.0%). On multivariable analysis the following variables were associated with MACE: diabetes mellitus, prior percutaneous coronary intervention performed at index angiography, and non-target vessel LCBI. Non-target vessel LCBI of 77 was determined using receiver-operating characteristic curve analysis to be a threshold for prediction of MACE in our cohort. The adjusted hazard ratio (HR) for non-target vessel LCBI ≥77 was 14.05 (95% confidence interval (CI) 2.47–133.51, p = 0.002). The 5-year cumulative incidence of events in the above-threshold group was 58.0% vs. 13.1% in the below-threshold group.
Conclusion
During long-term follow-up of patients who underwent NIRS imaging, high LCBI in a non-PCI target vessel was associated with increased incidence of MACE.
Highlights
- •
We analyzed the long-term outcomes of 239 patients who underwent coronary near-infrared spectroscopy during clinically indicated cardiac catheterization.
- •
Baseline clinical and procedural characteristics as well as non-PCI target vessel lipid core burden index were associated with the incidence of major adverse cardiovascular events.
- •
Using receiver-operating characteristic analysis, a non-PCI target vessel lipid core burden index of 77 was the optimal threshold for prediction of 5-year incidence of events (58% for lipid core burden index ≥77 vs. 13% for lipid core burden index <77).
1
Introduction
Near-infrared spectroscopy (NIRS) is an intravascular imaging modality that was developed to identify vulnerable plaques and vulnerable patients . Observational data suggest that NIRS could help identify plaques associated with subsequent coronary events . Recent reports from NIRS studies in non-culprit vessels and non-target vessel segments indicate that NIRS may be useful for patient-level risk stratification . We sought to identify factors associated with major adverse cardiovascular events (MACE) during long-term follow-up of patients who underwent NIRS imaging.
2
Methods
2.1
Patient population
We examined the baseline clinical and angiographic characteristics of 239 patients who underwent NIRS imaging at our institution between 2009–2011 as part of the Lipid cORe Plaque Association with Clinical Events Near-InfraRed Spectroscopy (ORACLE-NIRS) Registry ( NCT02265146 ). Data were collected and entered into a dedicated database. Patients who underwent clinically indicated cardiac catheterization and NIRS imaging were enrolled in the registry. NIRS was performed in vessels that contained target lesions at index angiography (“target vessels”) as well as vessels that did not contain a target lesion at index angiography (“non-target vessels”).
Femoral access was used in all cases and intravenous heparin was administered for anticoagulation during the catheterization procedure. NIRS imaging was performed after intracoronary or intragraft administration of nitroglycerin (100-200 mcg). Periprocedural myocardial infarction (MI) was defined as creatine-kinase myocardial band (CK-MB) to more than three times the upper limit of normal (ULN; 6.3 ng/dL at our institution). The study was approved by the institutional review board.
2.2
Near-infrared spectroscopy
NIRS images were acquired with a 3.2 French coronary catheter (InfraRedx, Burlington, MA). A built-in automated pullback and rotation device allowed a uniform speed of 0.5 mm per second and 240 rotations per minute. Data were transferred to DVDs for storage and offline analysis using the LipiScan Data Analysis Tool (InfraRedx).
Raw NIRS data were processed by an NIRS algorithm developed using histology as the gold standard for lipid detection . This algorithm has been validated for LCP detection in prospective studies with an associated receiver-operating characteristic area-under-the-curve of 0.80 (95% CI: 0.76–0.85). A virtual color image called a chemogram is constructed for each pullback, which maps the probability of the presence of lipid in the vessel wall, with yellow color representing higher probability of lipid and red representing low probability of lipid ( Fig. 1 ).
The lipid core burden index (LCBI) is a measure of the lipid burden for a given length of vessel, calculated by dividing the number of yellow pixels by the total number of pixels available, multiplied by 1000 (LCBI range: 0–1000).
2.3
Study endpoints
The primary endpoint was the incidence of MACE, defined as the composite of cardiac death, acute coronary syndrome (ACS), unplanned coronary revascularization, and stroke during follow-up after discharge from the index hospitalization. Endpoints were adjudicated on the basis of original source documentation obtained by blinded chart review. Non-fatal ACS consisted of ST-segment myocardial infarction (STEMI), non-STEMI, and unstable angina, defined according to the Third Universal Definition of Myocardial Infarction . Unplanned coronary revascularization included percutaneous coronary intervention (PCI) or coronary artery bypass graft surgery (CABG) that was not planned after the index coronary angiography and NIRS imaging procedure. When possible, events during follow-up were classified as related or unrelated to the index PCI target vessel (if there was an index PCI target vessel), unrelated, or indeterminate.
2.4
Statistical analysis
Normally distributed continuous variables are reported as mean ± standard deviation. Non-normally distributed continuous variables are reported as median [interquartile range, IQR] and categorical variables are reported as percentages. Normally distributed continuous variables were compared using Student’s t-test and non-parametric continuous variables were compared using the Wilcoxon signed-rank test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Cox regression was performed to test for variables associated with the primary outcome. Because there is no validated LCBI threshold for prediction of clinical events, median LCBI was used, as well as LCBI 77, which was determined using receiver-operating characteristic analysis.
Patients were considered at risk until the date of last contact, at which time point they were right-censored. Only the first event was counted for each patient. Cumulative event rates were estimated according to the Kaplan–Meier method or the cumulative incidence function and compared using the log-rank test. Competing risks of the events comprising MACE were accounted for using the cumulative incidence function . Multivariable Cox regression was used to identify independent predictors of MACE at follow-up. All statistical analysis was performed using JMP Software, Version 12.0 (SAS Institute, Cary, NC). A two-sided p-value of <0.05 was considered statistically significant.
2
Methods
2.1
Patient population
We examined the baseline clinical and angiographic characteristics of 239 patients who underwent NIRS imaging at our institution between 2009–2011 as part of the Lipid cORe Plaque Association with Clinical Events Near-InfraRed Spectroscopy (ORACLE-NIRS) Registry ( NCT02265146 ). Data were collected and entered into a dedicated database. Patients who underwent clinically indicated cardiac catheterization and NIRS imaging were enrolled in the registry. NIRS was performed in vessels that contained target lesions at index angiography (“target vessels”) as well as vessels that did not contain a target lesion at index angiography (“non-target vessels”).
Femoral access was used in all cases and intravenous heparin was administered for anticoagulation during the catheterization procedure. NIRS imaging was performed after intracoronary or intragraft administration of nitroglycerin (100-200 mcg). Periprocedural myocardial infarction (MI) was defined as creatine-kinase myocardial band (CK-MB) to more than three times the upper limit of normal (ULN; 6.3 ng/dL at our institution). The study was approved by the institutional review board.
2.2
Near-infrared spectroscopy
NIRS images were acquired with a 3.2 French coronary catheter (InfraRedx, Burlington, MA). A built-in automated pullback and rotation device allowed a uniform speed of 0.5 mm per second and 240 rotations per minute. Data were transferred to DVDs for storage and offline analysis using the LipiScan Data Analysis Tool (InfraRedx).
Raw NIRS data were processed by an NIRS algorithm developed using histology as the gold standard for lipid detection . This algorithm has been validated for LCP detection in prospective studies with an associated receiver-operating characteristic area-under-the-curve of 0.80 (95% CI: 0.76–0.85). A virtual color image called a chemogram is constructed for each pullback, which maps the probability of the presence of lipid in the vessel wall, with yellow color representing higher probability of lipid and red representing low probability of lipid ( Fig. 1 ).
The lipid core burden index (LCBI) is a measure of the lipid burden for a given length of vessel, calculated by dividing the number of yellow pixels by the total number of pixels available, multiplied by 1000 (LCBI range: 0–1000).
2.3
Study endpoints
The primary endpoint was the incidence of MACE, defined as the composite of cardiac death, acute coronary syndrome (ACS), unplanned coronary revascularization, and stroke during follow-up after discharge from the index hospitalization. Endpoints were adjudicated on the basis of original source documentation obtained by blinded chart review. Non-fatal ACS consisted of ST-segment myocardial infarction (STEMI), non-STEMI, and unstable angina, defined according to the Third Universal Definition of Myocardial Infarction . Unplanned coronary revascularization included percutaneous coronary intervention (PCI) or coronary artery bypass graft surgery (CABG) that was not planned after the index coronary angiography and NIRS imaging procedure. When possible, events during follow-up were classified as related or unrelated to the index PCI target vessel (if there was an index PCI target vessel), unrelated, or indeterminate.
2.4
Statistical analysis
Normally distributed continuous variables are reported as mean ± standard deviation. Non-normally distributed continuous variables are reported as median [interquartile range, IQR] and categorical variables are reported as percentages. Normally distributed continuous variables were compared using Student’s t-test and non-parametric continuous variables were compared using the Wilcoxon signed-rank test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Cox regression was performed to test for variables associated with the primary outcome. Because there is no validated LCBI threshold for prediction of clinical events, median LCBI was used, as well as LCBI 77, which was determined using receiver-operating characteristic analysis.
Patients were considered at risk until the date of last contact, at which time point they were right-censored. Only the first event was counted for each patient. Cumulative event rates were estimated according to the Kaplan–Meier method or the cumulative incidence function and compared using the log-rank test. Competing risks of the events comprising MACE were accounted for using the cumulative incidence function . Multivariable Cox regression was used to identify independent predictors of MACE at follow-up. All statistical analysis was performed using JMP Software, Version 12.0 (SAS Institute, Cary, NC). A two-sided p-value of <0.05 was considered statistically significant.
3
Results
3.1
Baseline characteristics
During the study period a total of 239 patients underwent NIRS imaging during angiography. The mean age was 63.5 years, and 99% were male. Indications for coronary angiography were as follows: 145 (61%) had stable coronary artery disease (CAD) and 94 (39%) had ACS (2% STEMI, 22% non-STEMI, 16% UA). Of those with stable CAD, index angiography was performed as part of another research protocol in three patients. PCI was performed in 206 (86%) patients during the index procedure. Baseline characteristics are summarized in Table 1 .
| N = 239 | |
|---|---|
| Baseline clinical characteristics | |
| Age ⁎ | 63.5 ± 8.6 |
| Male (%) | 99 |
| Body mass index ⁎ | 30.7 ± 6.2 |
| Diabetes mellitus (%) | 50 |
| Hypertension (%) | 95 |
| Hyperlipidemia (%) | 93 |
| Current or recent smoker (%) | 32 |
| Family history of CAD (%) | 41 |
| Prior MI (%) | 36 |
| Prior PCI (%) | 42 |
| Prior CABG (%) | 23 |
| Prior stroke (%) | 11 |
| Peripheral arterial disease (%) | 21 |
| History of heart failure (%) | 23 |
| Left ventricular ejection fraction ⁎ | 52 ± 12 |
| Total cholesterol ⁎ | 140 ± 54 |
| Low-density lipoprotein cholesterol ⁎ | 74 ± 40 |
| High-density lipoprotein cholesterol ⁎ | 31 ± 10 |
| Triglycerides ⁎ | 187 ± 160 |
| Statin at presentation (%) | 89 |
| Non-statin anti-lipidemic medication at presentation (%) | 6 |
| Statin at discharge (%) | 97 |
| Non-statin anti-lipidemic medication at discharge (%) | 10 |
| Procedural characteristics | |
| Indication for coronary angiography | |
| STEMI (%) | 2 |
| Non-STEMI (%) | 22 |
| Unstable angina (%) | 15 |
| Stable CAD (%) | 61 |
| Extent of coronary artery disease | |
| 0-vessel disease (%) | 6 |
| 1-vessel disease (%) | 30 |
| 2-vessel disease (%) | 34 |
| 3-vessel disease (%) | 30 |
| PCI performed at index procedure (%) | 86 |
| Percent pre-stenosis at target lesion ⁎ | 83 ± 13 |
| Percent post-stenosis at target lesion ⁎ | 3 ± 10 |
| Number of lesions treated ⁎ | 1.6 ± 0.7 |
| NIRS characteristics of target vessel | |
| Pre-stenting LCBI ⁎ | 88[50–130] |
| Post-stenting LCBI ⁎ | 64[37–94] |
| NIRS characteristics of non-target vessel | |
| LCBI ⁎ | 57[26–94] |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
