Numerous clinical studies using coronary computed tomography angiography (CTA) and conventional invasive coronary angiography (ICA) confirmed the strong relation between atherosclerotic disease burden and risk of adverse events. Few studies have compared coronary CTA and ICA regarding semiquantitative plaque burden measurements, reproducibility, and cardiovascular risk assessment. We enrolled 71 consecutive patients (mean age 62 ± 9 years, 37% women) from the Genetic Loci and the Burden of Atherosclerotic Lesions study ( NCT01738828 ), who underwent 256-slice multidetector row coronary CTA and ICA at a single site. On average, 42 ± 32 days passed between the 2 examinations. A total of 1,016 coronary segments were imaged by both CTA and ICA according the 18-segment Society of Cardiovascular Computed Tomography classification. We excluded 16 segments treated with coronary stents. Overall, 1,000 segments were evaluated for the presence of stenosis severity (<25%: minimal, 25% to 49%: mild, 50% to 70%: moderate, 70% to 99%: severe, 100%: occlusion). We calculated the segment involvement score (SIS) and segment stenosis score. Patients were classified into 4 groups: extensive obstructive (SIS >4 and ≥50% stenosis), extensive nonobstructive (SIS >4 and <50% stenosis), nonextensive obstructive (SIS ≤4 and ≥50% stenosis), or nonextensive nonobstructive (SIS ≤4 and <50% stenosis). CTA detected coronary artery plaques in 49%, whereas ICA showed coronary plaques in 24% of the analyzed 1,000 segments (p <0.001). CTA detected atherosclerotic plaque in 35% of coronary segments where ICA was negative, whereas ICA detected plaque only in 3% of segments where CTA was negative. CTA-based segment scores were significantly greater, SIS: 6.9 ± 3.0 versus 3.3 ± 2.0, segment stenosis score: 16.4 ± 8.8 versus 9.4 ± 6.8 (p <0.001 for both). In conclusion, coronary CTA detected approximately twice as many coronary segments with plaque compared to ICA, which resulted in 52% of the patients being assigned to a greater risk category.
Numerous anthropometric and biomarker-based multivariate risk models have been established to predict cardiovascular disease (CVD). Multiethnic evaluation of such risk models has shown systematic overestimation of CVD risk by up to 115%, indicating the need for more precise risk estimation. In the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation trial, plaque burden assessment based on invasive coronary angiography (ICA) provided a strong prognostic value, whereas ischemic myocardium burden failed to predict major adverse cardiovascular events. This paradigm shift from the individual lesions to overall atherosclerotic plaque burden necessitates an accurate plaque detection method in large patient populations. Cardiac computed tomography angiography (CTA) has emerged as a useful and highly reliable imaging technique for examination of the coronary arteries. Numerous studies have validated the diagnostic performance of CTA for the detection of obstructive coronary artery disease (CAD) compared to ICA as reference standard, whereas only a few studies have compared these 2 methods regarding semiquantitative plaque burden measurements. Therefore, our objective was to compare 256-slice CTA and ICA regarding semiquantitative plaque burden assessment and to assess the effect of imaging method on cardiovascular risk classification.
Methods
The Genetic Loci and the Burden of Atherosclerotic Lesions (GLOBAL) study enrolled patients who were referred to coronary CTA because of suspected CAD ( NCT01738828 ). Detailed description of the patient population, including the inclusion and exclusion criteria, was reported previously. The study was approved by the institutional ethical review board, and all participants provided written informed consent for the GLOBAL study.
Of the 883 patients enrolled by our institution into the GLOBAL study, all patients who underwent both coronary CTA and ICA within 120 days were included in our study. Only patients with diagnostic image quality for all coronary segments were included in the study. Coronary segments with a diameter of >1.5 mm were analyzed. In total, 71 patients were included in our analysis. In 58 patients, ICA followed CTA based on clinical findings, whereas in 13 cases, ICA was carried out before CTA. These patients were either referred to CTA after revascularization for stent patency assessment (7 patients) or referred to left atrial angiography before radiofrequency ablation (6 patients). Patient characteristics are summarized in Table 1 .
| Variable | n = 71 |
|---|---|
| Age (years) | 61.6 ± 9.0 |
| Female | 26 (37 %) |
| Body Mass Index (kg/m 2 ) | 27.9 ± 4.3 |
| Hypertension | 51 (72 %) |
| Diabetes Mellitus | 13 (18 %) |
| HDL (mmol/l) | 1.34 ± 0.4 |
| LDL (mmol/l) | 3.0± 1.3 |
| Triglyceride (mmol/l) | 1.7 ± 0.8 |
| Cholesterol (mmol/l) | 4.8 ± 1.6 |
| Time between coronary computed tomography angiography and invasive coronary angiography (days) | 40.2 ± 32.1 |
All patients underwent a prospectively electrocardiogram-triggered coronary CTA scan using a 256-slice multidetector row computed tomography (Brilliance iCT; Philips HealthTech, Best, The Netherlands). Coronary CTA images were acquired in axial mode with 270-ms rotation time, 128 × 0.625-mm collimation, tube voltage of 100 to 120 kVp at 78% of the RR interval.
All images were randomly and independently analyzed. Semiquantitative plaque burden quantification of ICA images was performed by an interventional cardiologist (I.F.É. with 10 years of experience). A minimum of 5 projections of the left and right coronary systems were acquired in each patient. All coronary segments were analyzed blinded to CTA results, using a minimum of 2 projections. Coronary CTA images were analyzed by a cardiologist (P.M-.H. with 12 years of experience). Images were analyzed using axial thin-slice and multiplanar reformations ( Figure 1 ). For interreader reproducibility measurements, ICA and CTA images were also analyzed by second readers (S.N. and B.S. with 6 and 4 years of experience, respectively).
A total of 1,016 segments were assessed based on the 18-segment Society of Cardiovascular Computed Tomography classification with both methods. We excluded 16 segments because of the presence of coronary stents leading to overall 1,000 analyzed segments. All segments were scored for the presence or absence of plaque (0: absent; 1: present) and the degree of stenosis (0: none; 1: minimal [<25%]; 2: mild [25% to 49%]; 3: moderate [50% to 69%]; 4: severe [70% to 99%]; or 5: occlusion [100%]). In case multiple lesions were present in a segment, the observers recorded the highest degree of stenosis for that segment. In each patient, segment involvement score (SIS) was used to quantify the number of segments with any plaque, whereas segment stenosis score (SSS) was calculated by summing the stenosis scores of each segment. Indexed values were calculated by dividing the SIS and SSS scores by the number of segments: SIS index (SISi) = SIS/number of segments; SSS index (SSSi) = SSS/number of segments.
Based on the findings of Bittencourt et al., patients were classified into 4 groups: extensive obstructive (SIS >4 and ≥50% stenosis), extensive nonobstructive (SIS >4 and <50% stenosis), nonextensive obstructive (SIS ≤4 and ≥50% stenosis), or nonextensive nonobstructive (SIS ≤4 and <50% stenosis) using both ICA and CTA.
All continuous variables are expressed as mean ± standard deviation, whereas categorical variables are expressed as frequencies and percentages. Presence of plaque was compared using the chi-square test between methods. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated to assess the diagnostic accuracy of CTA compared to ICA as reference standard. SIS, SSS, SISi, and SSSi were compared using the paired t test between the methods. Intrareader and interreader reproducibility was assessed based on 20 randomly selected subjects’ images using Cohen κ for stenosis categories and intraclass correlation (ICC) for segment scores. Cohen κ and ICC values are interpreted as: 0.81 to 1.00: excellent; 0.61 to 0.80: good; 0.41 to 0.60: moderate; 0.21 to 0.40: fair; 0.00 to 0.20: poor. Reclassification rate was calculated by dividing the number of people who shifted groups based on the 2 methods by the total study population. All statistical calculations were performed using SPSS software (SPSS version 23; IBM Corp., Armonk, New York). A p value of ≤0.05 was considered significant.
Results
Coronary CTA detected coronary artery plaque in 49% (487 of 1,000) of the segments, whereas ICA showed coronary plaques in 24% (235 of 1,000) of all segments (p <0.001, Figure 2 ). Of the 235 positive segments with ICA, corresponding segments on CTA was also positive in 94%. CTA detected atherosclerotic plaque in 35% (266 of 765) of coronary segments where ICA was negative; 36% (95 of 266) of these plaques were noncalcified, 38% (102 of 266) were mixed, and 26% (69 of 266) were calcified plaques on CTA. When considering the severity of coronary stenosis only seen by CTA, 79% of plaques caused minimal or mild luminal stenosis (211 of 266). Conversely, ICA detected plaque only in 3% (14 of 513) of segments where CTA was negative. Detailed distribution and quantification of coronary plaques only seen by CTA are summarized in Table 2 . Regarding segment scores, CTA showed more than 2 times as many segments with plaque compared to ICA, and also the overall degree of stenosis caused by the plaques was almost twice. Summary of segment score analysis is given in Table 3 .
| Plaque quantity | Coronary segments | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LM | prox LAD | mid LAD | dist LAD | D1 | D2 | IM | prox LC | md LC | OM1 | OM2 | plb LC | pda LC | prox R | mid R | dist R | pda R | plb R | Total | |
| Minimal | 22 | 9 | 5 | 8 | 3 | 0 | 1 | 11 | 7 | 3 | 3 | 0 | 0 | 12 | 7 | 10 | 3 | 3 | 107 |
| Mild | 15 | 9 | 9 | 8 | 5 | 0 | 1 | 13 | 2 | 5 | 0 | 0 | 1 | 10 | 7 | 12 | 6 | 1 | 104 |
| Moderate | 1 | 4 | 8 | 4 | 5 | 2 | 0 | 4 | 3 | 3 | 0 | 0 | 1 | 7 | 4 | 1 | 0 | 0 | 47 |
| Severe | 0 | 0 | 1 | 0 | 3 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 2 | 0 | 0 | 0 | 8 |
| Occlusion | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Total number of involved segments only on coronary computed tomography | 38 | 22 | 23 | 20 | 16 | 2 | 2 | 28 | 12 | 11 | 4 | 1 | 2 | 29 | 20 | 23 | 9 | 4 | 266 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
