Abstract
Pupose
The newer 256-slice computed tomography coronary angiography (CTCA) has the capability of improving diagnostic performance in the detection of obstructive coronary artery disease (CAD) compared to 64-slice CTCA. The aim of this study was to compare the diagnostic performance of 64- versus 256-slice CTCA in two similar populations.
Methods
Our study included 120 consecutive patients who were referred for CTCA and subsequently underwent conventional coronary angiography (CCA). Sixty patients were studied by 64-slice CTCA, with the other 60 by 256-slice CTCA. We compared the technical characteristics and diagnostic performance of 64- and 256-slice CTCA for the detection of ≥ 50% stenosis of the coronary arteries on CCA.
Results
The 256-slice CTCA had a shorter scanning time (4.4 ± 0.6 sec vs. 5.0 ± 0.7 sec, p < 0.001) compared to 64-slice CTCA. The diagnostic accuracy rates of 256-slice CTCA based on patient analysis (97% vs. 83%, p < 0.05), vessel analysis (95% vs. 85%, p < 0.05), and segment analysis (94% vs. 88%, p < 0.05) were significantly superior to those of 64-slice CTCA. The diagnostic accuracy rates of 64- and 256-slice CTCA were affected by the presence of stent (65% vs. 75%, respectively, p > 0.05) and severe calcifications (75% vs. 82%, respectively, p > 0.05).
Conclusions
In two similar populations, 256-slice CTCA displayed superior diagnostic performance than 64-slice CTCA. However, the performance of 256-slide CTCA is affected in those segments that are severely calcified and/or stented.
Keywords: Computed tomography coronary angiography (CTCA), Conventional coronary angiography, Diagnostic performance, 64-slice helical CTCA, 256-slice helical CTCA
INTRODUCTION
Selective conventional coronary angiography (CCA) remains the “gold standard” for the assessment of coronary artery anatomy, but it is an invasive procedure and carries the risk of major complications.1 Hence, a non-invasive test would be advantageous for patients and clinicians alike. In the past few years, 64-slice multi-detector computed tomography (MDCT) has become a well-documented non-invasive diagnostic tool for the quick exclusion of coronary artery disease (CAD).2-9 However, 64-slice MDCT has limitations, such as the need for a stable and lower heart rate, achieved with β-adrenergic blocking agents to control the heart rate to less than 80 beats per minute when scanning.2,3,10-12 Also, extensive coronary calcification and stent structure affect the evaluation of coronary artery lumen.
Recently, there have been many technological improvements in MDCT. The newer, wide-area 256-slice MDCT has: 1) twice the longitudinal (z) detector co-verage (8 cm) as compared to 64-slice MDCT (4 cm); 2) a flying focal spot technology that doubles the z-sampling providing a simultaneous acquisition of 256 slices per rotation; 3) a fast gantry rotation time of 0.27 sec/rotation; and dose reduction technologies such as dynamic helical collimation which reduces excess radiation dose resulting from overscanning at either end of a helical CT.13-15 This scanner has the capability of improving diagnostic performance for detection of obstructive CAD with fewer of the limitations commonly associated with 64-slice MDCT. We have reported the diagnostic accuracy of 256-slice MDCT compared with CCA in patients with suspected CAD.16 However, there has been no publication to date comparing the diagnostic performance of 64-slice CTCA versus 256-slice CTCA for the detection of significant CAD (> 50%) in two similar populations using CCA as the gold standard. Hence, the purpose of the present study is to compare the technical characteristics and diagnostic performance of 64- versus 256-slice CTCA for the diagnosis of CAD.
METHODS
Patient population
The local hospital ethics committee approved the protocol as a review of an existing database. All patient identifiers were removed prior to analysis. The study population of this retrospective study was composed of 120 patients with any symptoms and intermediate pretest likelihood of CAD (96 men and 24 women, aged 63 ± 11 years), who underwent CTCA and subsequently underwent CCA at our institute. Among the 120 study patients, 60 of them underwent 64-slice CTCA between March 2008 and November 2008, and the remaining 60 underwent 256-slice CTCA between December 2008 and April 2009. All the CTCA were carried out before CCA.
CTCA protocol and image acquisition
In the first cohort, retrospective ECG-gated CTCA image acquisition was performed using a 64-slice MDCT scanner (Brilliance 64, Philips Healthcare, Cleveland, OH, USA), which contains single source, 64 detector rows. All patients were in normal sinus rhythm, and oral β-blocker therapy (20 mg of propranolol HCl; Astra Zeneca, Cheshire, United Kingdom) was administered if baseline heart rate was greater than 70 beats per minute at the time of the 64-slice CTCA. A nonenhanced scan to calculate the total calcium score was performed before the CTCA. This was followed by a retrospective electrocardiogram (ECG)-gated 64 slice CCTA with the following scan parameters: 64 × 0.625 mm collimation, 400 ms rotation time, pitch 0.2, tube voltage 120 kVp, and effective tube current-time product (normalized to pitch) 800-1000 mAs.
In the second cohort, retrospective ECG-gated CTCA data acquisition was performed using a 256-slice MDCT scanner (Brilliance iCT, Philips Healthcare, Cleveland, OH, USA), which contains single source, 128 detector rows with focal spot-shift technology. Similarly, calculation of total calcium scan was performed by a non-enhanced scan before the CTCA. The 256-slice CTCA was conducted according to the following parameters: 128 × 2 × 0.625 mm collimation, 270-ms rotation time, pitch of 0.16, tube voltage 120 kVp, and effective tube current-time product (normalized to pitch) 800-1000 mAs. According to our preliminary experience, in contrast to the recommended use of 64-slice CTCA at heart rate ≤ 70 beats per minute during scanning, 256-slice CTCA can tolerate higher heart rate while maintaining diagnostic image quality.16 Therefore, oral β-blocker therapy (20 mg of propranolol HCl; Astra Zeneca, Cheshire, United Kingdom) was only administered if baseline heart rate was greater than 90 beats per minute at the time of the 256-slice CTCA.
The standard temporal resolution (210 ms for the 64-slice MDCT and 135 ms for the 256-slice MDCT) was improved by employing advanced cardiac multi-cycle reconstruction algorithms that combined data from consecutive cardiac cycles.17 Additionally, the overlapped pitch, along with the use of cardiac gating algorithms (Beat-to-Beat Variable Delay Algorithm, Philips Healthcare, Cleveland, OH, USA) enabled the detection and reconstruction of the same physiological cardiac phase of interest (for example, quiescent phase corresponding to ventricular diastasis).18
A bolus of 65 to 110 ml nonionic contrast (Optiray 350, Tyco Healthcare, Montreal, Quebec, Canada) was injected, depending on the total scan duration, at a rate of 4 ml/s when body weight < 60 kg, or 5 ml/s when body weight > 60 kg, followed by a 20 ml saline flush.
The effective radiation dose of the non-enhanced scan and the CTCA was estimated from the product of the dose-length product and a conversion coefficient [k = 0.017 mSv/(mGy X cm)] for the chest as the investigated anatomical region.
CTCA image evaluation
The total calcium scores of study patients were analyzed using dedicated software [Calcium Scoring, Extended Brilliance Workspace (EBW), Philips Healthcare, Cleveland, OH, USA]. Interpretation of CTCA images was performed by two experienced radiologists (Law WY, Kuo CJ), who were unaware of the results of CCA. Classification of coronary vessels’ segments was according to the 15-segment modified American Heart Association classification and was condensed into 14 segments, including (1) left main coronary artery (LM), (2) proximal left anterior descending artery (LAD), (3) middle LAD, (4) distal LAD, (5) all diagonal branches of LAD, (6) proximal left circumflex artery (LCX), (7) middle LCX, (8) distal LCX, (9) all obtuse marginal branches, (10) proximal right coronary artery (RCA), (11) middle RCA, (12) distal RCA, (13) posterior descending branch, and (14) posterolateral branch.19 Intermediate arteries were classified as additional vessels. All segments, regardless of size, were included for comparison between the 64- and 256-slice CTCA. Segments were classified as normal, and defined as smooth parallel or tapering borders; non-significantly obstructed coronary vessels were defined as wall irregularities with < 50% stenosis, and significantly obstructed coronary vessels were defined as ≥ 50% reduction in the diameter of the lumen. Also, segments with significant obstruction were scored positive for significant CAD. The presence of calcium on a persegment basis was assessed and graded as none (not calcified), moderate (small eccentric calcification in the vessel wall), and high (calcification extending longitudinally along the vessel wall). Disagreements over imaging interpretation were resolved by consensus. The segments that could not be evaluated by either radiologist were rated as positive and were included in the final data sets.
Quantitative coronary angiography
All study patients received CCA with standard pro-tocol within 1 month after CTCA. Interpretation of coronary vessels was made by two experienced car-diologists (Chua SK, Shyu KG), who were blinded to the clinical information and results from CTCA. Coronary artery segments were identified using the same segment classification as used for the CTCA. The stenoses in segments were quantified by a validated quantitative coronary angiography (QCA). Significant angiographic stenoses, evaluated in the worst angiographic view, were defined as a diameter reduction of ≥ 50%.
Statistical analysis
In previous studies, the accuracy rate of each subject group receiving CTCA was normally distributed with standard deviation of around 10%.3-5 If the true difference in the accuracy rate between the 64- and 256-slice CTCA is 5%, it will be necessary to study 52 subjects with 64-slice CTCA and 52 subjects with 256-slice CTCA. Then, we can be in the position to reject the null hypothesis that the accuracy rate of the 64- and 256-slice CTCA groups are equal with probability (power) 0.8. The Type I error probability associated with this null hypothesis test is 0.05.
Quantitative data are expressed as mean ± standard deviation. The chi-square test with Yates’ correction or Fisher’s exact test was used to analyze the nonpara-metric data. If the frequency of any cell was < 5, then a Fisher’s exact test was used. When p was < 0.05, this was considered statistically significant. Diagnostic sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and accuracy rate of CTCA for the detection of significant coronary stenoses on CCA were calculated three ways: 1) on a per-patient analysis (defined as presence of at least one or absence of any significant stenosis per patient); 2) on a per-vessel basis (presence of at least one or absence of any significant stenosis in one coronary artery); and 3) on a per-segment basis (presence of at least one or absence of any significant stenosis in one coronary artery segment) with the corresponding 95% confidence intervals (CI), which is calculated according to the percentile method. Statistical analyses were performed with SPSS (SPSS 16.0 Version for Windows, SPSS. Inc., Chicago, IL, USA).
RESULTS
Patient characteristics
A total of 120 consecutive patients were enrolled in the study. The initial 60 patients (48 men and 12 women, mean age 63.2 ± 11.7 years) underwent 64-slice CTCA, whereas the remaining 60 patients (48 men and 12 women, mean age 63.4 ± 10.7 years) underwent 256-slice CTCA. Patient characteristics are shown in Table 1. Both groups were predominantly male, without sig-nificant differences between them. There were no significant differences between the two groups in body mass index, smoking status, and underlying disease.
Table 1. Baseline characteristics of the patients .
| 64-slice CTCA (n = 60) | 256-slice CTCA (n = 60) | p value | |
| Age, years | 63.2 ± 11.7 | 63.4 ± 10.7 | 0.94 |
| Male | 48 (80) | 48 (80) | 0.50 |
| Body mass index, kg/m2 | 26.6 ± 3.8 | 27.6 ± 2.9 | 0.12 |
| Smoking | 27 (45.0) | 25 (41.7) | 0.43 |
| Hypertension | 42 (70.0) | 43 (71.7) | 0.50 |
| Diabetes, Type II | 20 (33.3) | 19 (31.7) | 0.42 |
| Dyslipidemia | 19 (31.7) | 22 (36) | 0.29 |
| Family history of CAD | 2 (3.3) | 5 (8.3) | 0.22 |
| Prior percutaneous coronary intervention | 19 (31.7) | 13 (21.7) | 0.15 |
| Creatinine, mg/dl | 1.05 ± 0.25 | 1.00 ± 0.23 | 0.22 |
| Characteristics of MDCT | |||
| Previous β-blocker therapy | 37 (61.7) | 36 (60.0) | 0.43 |
| β-blocker requested before scan | 21 (35.0) | 2 (3.3) | < 0.001 |
| Heart rate on CTCA, beats/min | 60.1 ± 2.9 | 67.9 ± 11.3 | < 0.001 |
| Agatston calcium score | |||
| Mean | 348.3 ± 529.7 | 445.80 ± 625.76 | 0.36 |
| < 400 | 43 (71.7) | 38 (63.3) | 0.17 |
| ≥ 400 | 17 (28.3) | 22 (36.7) | |
| Scan time, sec | 5.0 ± 0.7 | 4.4 ± 0.6 | < 0.001 |
| Scan length, cm | 13.5 ± 2.3 | 13.1 ± 2.5 | 0.08 |
| Radiation dose on CTCA, mSv | 12.9 ± 3.0 | 12.7 ± 3.7 | 0.78 |
| Time from CTCA to CCA, days | 16.2 ± 8.8 | 15.5 ± 9.9 | 0.70 |
Values presented as number (%) and mean ± SD. CAD, coronary artery disease; CCA, conventional coronary angiography; CTCA, computed tomography coronary angiography.
Characteristics of MDCT
Additional β-blockers before 64- and 256-slice CTCA scanning were administered in 21 (35%) and 2 (3%) of patients (p < 0.001), respectively, decreasing the mean heart rate to 60.1 ± 2.9 beats per minute and 67.9 ± 11.3 beats per minute (p < 0.001), respectively (Table 1). Both groups had similar Agatston calcium scores (348.2 ± 529.7 vs. 445.8 ± 625.8, p = 0.36). The 256-slice CTCA had a shorter scan duration (4.4 ± 0.6 sec vs. 5.0 ± 0.7 sec, p < 0.001) than the 64-slice CTCA, without difference in scan length (13.5 ± 2.3 cm vs. 13.1 ± 2.5 cm, p = 0.08). Our study showed no difference in mean radiation dose between retrospective ECG-gated 64- versus 256-slice CTCA (12.9 ± 3.0 mSv vs. 12.7 ± 3.7 mSv, p = 0.78). There was no significant difference in the delay interval between CTCA and CCA for either the 64-slice or 256-slice CTCA population (16.2 ± 8.8 days vs.15.5 ± 9.9 days, p = 0.70).
Diagnostic performance of 256-slice CTCA: patient-based analysis
The 256-slice CTCA had significantly better diag-nostic accuracy rate [97% (95% CI: 87% to 99%)] compared to 64-slice CTCA [83% (95% CI: 71% to 91%)] (p < 0.05) for the diagnosis of a patient with at least one coronary artery stenosis of 50% or more as assessed by CCA (Table 2). In all patients with CAD, 256-slice CTCA detected at least 1 significant coronary artery stenosis, which means that in a patient-based analysis, all of these patients were correctly identified using 256-slice CTCA. In addition, using 256-slice CTCA, only 2 patients with non-significant CAD were incorrectly classified as significant CAD.
Table 2. Diagnostic performance of 64- versus 256-slice CTCA for detecting stenosis ≥ 50% on CCA in the per-patient analysis (95% CI) .
| MDCT | Prevalence % | N | TP | TN | FP | FN | SEN, % | SPE, % | PPV, % | NPV, % | Accuracy rate, % | |
| Patient-based analysis | 64 | 81.67 | 60 | 44 | 6 | 5 | 5 | 90 (77-96) | 55 (29-84) | 90 (90-96) | 55 (35-82) | 83 (71-91)* |
| 256 | 90.00 | 60 | 54 | 4 | 2 | 0 | 100 (92-100) | 67 (24-94) | 96 (87-99) | 100 (50-98) | 97 (87-99)* | |
| HR ≤ 70 bpm | 64 | 81.67 | 60 | 44 | 6 | 5 | 5 | 90 (77-96) | 55 (29-84) | 90 (90-96) | 55 (35-82) | 83 (71-91) |
| 256 | 91.89 | 37 | 34 | 2 | 1 | 0 | 100 (87-100) | 67 (13-98) | 97 (83-100) | 100 (20-95) | 97 (84-100) | |
| HR > 70 bpm | 64 | - | - | - | - | - | - | - | - | - | - | - |
| 256 | 86.96 | 23 | 20 | 2 | 1 | 0 | 100 (80-100) | 67 (13-98) | 95 (74-100) | 100 (20-95) | 96 (76-100) | |
| ACS < 400 | 64 | 87.50 | 40 | 32 | 4 | 1 | 3 | 91 (76-98) | 80 (30-99) | 97 (82-100) | 57 (20-88) | 90 (75-97) |
| 256 | 92.11 | 38 | 35 | 3 | 0 | 0 | 100 (88-100) | 100 (31-97) | 100 (88-100) | 100 (31-97) | 100 (89-100) | |
| ACS ≥ 400 | 64 | 70.00 | 20 | 12 | 2 | 4 | 2 | 86 (56-97) | 33 (6-76) | 75 (47-92) | 50 (9-91) | 70 (46-87) |
| 256 | 86.36 | 22 | 19 | 1 | 2 | 0 | 100 (79-100) | 33 (2-87) | 90 (68-98) | 100 (5-89) | 91 (69-98) |
* p < 0.05 for comparison of 64- vs. 256-slice CTCA. ACS, Agatston calcium score; bpm, beats per-minute; CCA, conventional coronary angiography; CI, confidence interval; CTCA, computed tomography coronary angiography; FN, false negative; FP, false positive; HR, heart rate; MDCT, multidetector computed tomography; NPV, negative predictive value; PPV, positive predictive value; SEN, sensitivity; SPE, specificity; TN, true negative; TP, true positive.
Using 256-slice CTCA to detect ≥ 50% coronary artery stenosis at the patient level with heart rates ≤ 70 beats per minute versus > 70 beats per minute, the diagnostic accuracy rate of 256-slice CTCA declined when the heart rate > 70 beats per minute versus ≤ 70 beats per minute, but not with statistical significance.
We examined the diagnostic performance of 64- and 256-slice CTCA based on baseline Agatston calcium score, stratified by < 400 versus ≥ 400 Agatston units. The diagnostic accuracy rates of 64- and 256-slice CTCA were both lower in patients with Agatston calcium score ≥ 400 without difference between them being significant.
Diagnostic performance of 256-slice CTCA: vessel-based analysis
Diagnostic accuracy rate for ≥ 50% coronary artery stenosis in the LM and LAD did not differ between 64- and 256-slice CTCA (Table 3). The diagnostic accuracy rate of 256-slice CTCA based on vessel analysis was better than that of 64-slice CTCA [95% (95% CI: 92% to 98%) vs. 85% (95% CI: 80% to 90%), p < 0.05]. Significant vessel stenoses ≥ 50% in the RCA and LCX were more commonly underestimated than were those in the LM and LAD by 64-slice CTCA. In our study, the 256-slice CTCA had a better diagnostic accuracy rate for detecting significant stenosis in the RCA [95% (95% CI: 85% to 99%) vs. 82% (95% CI: 69% to 90%), p < 0.05] and LCX [92% (95% CI: 81% to 87%) vs. 75% (95% CI: 62% to 85%), p < 0.05] than 64-slice CTCA.
Table 3. Diagnostic performance of 64- versus 256-slice CTCA for detecting stenosis ≥ 50% on CCA in the per-vessel analysis (95% CI) .
| MDCT | Prevalence % | N | TP | TN | FP | FN | SEN, % | SPE, % | PPV, % | NPV, % | Accuracy rate, % | |
| Vessel-based analysis | 64 | 41.25 | 240 | 81 | 124 | 17 | 18 | 82 (73-89)* | 88 (81-93)* | 83 (73-89)* | 87 (90-92)* | 85 (80-90)* |
| 256 | 45 | 240 | 103 | 126 | 6 | 5 | 95 (89-98)* | 95 (90-98)* | 95 (88-98)* | 96 (91-99)* | 95 (92-98)* | |
| RCA | 64 | 40.00 | 60 | 19 | 30 | 6 | 5 | 79 (57-92) | 83 (67-93) | 76 (54-90) | 86 (69-95) | 82 (69-90)* |
| 256 | 51.67 | 60 | 30 | 27 | 2 | 1 | 97 (81-100) | 93 (76-99) | 94 (78-99) | 96 (80-100) | 95 (85-99)* | |
| LM | 64 | 8.33 | 60 | 4 | 54 | 1 | 1 | 80 (30-99) | 98 (89-100) | 80 (30-99) | 98 (89-100) | 97 (87-99) |
| 256 | 10 | 60 | 6 | 53 | 1 | 0 | 100 (52-98) | 98 (89-100) | 86 (42-99) | 100 (92-100) | 98 (90-100) | |
| LAD | 64 | 75.00 | 60 | 42 | 11 | 4 | 3 | 93 (81-98) | 73 (45-91) | 91 (78-97) | 79 (49-94) | 88 (77-95) |
| 256 | 81.67 | 60 | 48 | 10 | 1 | 1 | 98 (88-100) | 91 (57-100) | 98 (88-100) | 91 (57-100) | 97 (87-99) | |
| LC | 64 | 41.67 | 60 | 16 | 29 | 6 | 9 | 64 (43-81) | 83 (66-93) | 73 (50-88) | 76 (59-88) | 75 (62-85)* |
| 256 | 36.67 | 60 | 19 | 36 | 2 | 3 | 86 (64-96) | 95 (81-99) | 90 (68-98) | 92 (78-98) | 92 (81-87)* |
* p < 0.05 for comparison of 64- vs. 256-slice CTCA. LAD, left anterior descending artery; LC, left circumflex artery; LM, left main artery; RCA, right coronary artery; Abbreviations as in Table 2.
Diagnostic performance of 256-slice CTCA: segment-based analysis
Overall, 1560 segments were included for com-parison of diagnostic performance of 64- and 256-slice CTCA with CCA (Table 4). All coronary segments were evaluated, including segments with or without stent. The diagnostic accuracy rates of 256-slice CTCA based on segment analysis [94% (95% CI: 92% to 95%) vs. 88% (95% CI: 86% to 90%), p < 0.05], including non-stented segments [95% (95% CI: 93% to 96%) vs. 90% (95% CI: 88% to 92%), p < 0.05] and segments without calcification [98% (95% CI: 96% to 99%) vs. 93% (95% CI: 91% to 95%), p < 0.05], were significantly better than those of 64-slice CTCA.
Table 4. Analysis of the influence of coronary calcium and stent on the diagnostic performance of 64- versus 256-slice CTCA for detecting stenosis ≥ 50% on CCA in the per-segment analysis (95% CI) .
| MDCT | Prevalence % | N | TP | TN | FP | FN | SEN, % | SPE, % | PPV, % | NPV, % | Accuracy rate, % | |
| Segment-based analysis | 64 | 19.05 | 840 | 112 | 627 | 53 | 48 | 70 (62-77)* | 92 (90-94)* | 68 (60-75)* | 93 (91-95)* | 88 (86-90)* |
| 256 | 22.98 | 840 | 172 | 615 | 32 | 21 | 89 (84-93)* | 95 (93-97)* | 84 (78-89)* | 97 (95-98)* | 94 (92-95)* | |
| Nonstented segments | 64 | 16.51 | 769 | 91 | 602 | 40 | 36 | 72 (63-79)* | 94 (92-95)† | 69 (61-77)* | 94 (92-96)*† | 90 (88-92)*† |
| 256 | 21.18 | 779 | 152 | 589 | 25 | 13 | 92 (87-96)*‡ | 96 (94-97)‡ | 86 (80-90)* | 98 (96-99)*‡ | 95 (93-96)*‡ | |
| Calcium | ||||||||||||
| None | 64 | 11.02 | 599 | 40 | 518 | 15 | 26 | 61 (48-72)*† | 97 (95-98)† | 73 (59-83)* | 95 (93-97)*† | 93 (91-95)*† |
| 256 | 11.62 | 594 | 62 | 518 | 7 | 7 | 90 (80-95)* | 99 (97-99)‡ | 90 (80-95)* | 99 (97-99)*‡ | 98 (96-99)*‡ | |
| Moderate | 64 | 24.75 | 101 | 19 | 64 | 12 | 6 | 76 (54-90)* | 84 (73-91) | 61 (42-77)* | 91 (81-96) | 82 (73-89) |
| 256 | 44.07 | 118 | 49 | 57 | 9 | 3 | 94 (83-99)* | 86 (75-93) | 84 (72-92)* | 95 (86-99) | 90 (83-95) | |
| High | 64 | 52.17 | 69 | 32 | 20 | 13 | 4 | 89 (73-96)† | 61 (42-77)† | 71 (55-83) | 83 (62-95)† | 75 (63-85)† |
| 256 | 65.67 | 67 | 41 | 14 | 9 | 3 | 93 (80-98) | 61 (39-80)‡ | 82 (68-91) | 82 (55-95)‡ | 82 (70-90)‡ | |
| Stented segments | 64 | 46.48 | 71 | 21 | 25 | 13 | 12 | 64 (45-79) | 66 (49-80)† | 62 (44-77) | 68 (50-81)† | 65 (52-76)† |
| 256 | 45.90 | 61 | 20 | 26 | 7 | 8 | 71 (51-86)‡ | 79 (61-90)‡ | 74 (53-88) | 76 (58-89)‡ | 75 (62-85)‡ |
* p < 0.05 for comparison of 64- vs. 256-slice CTCA. † p < 0.05 for comparisons of stented vs. nonstented, and none vs. high calcified segments with 64-slice CTCA. ‡ p < 0.05 for comparisons of stented vs. nonstented, and none vs. high calcified segments with 256-slice CTCA. Abbreviations as in Table 2.
The diagnostic accuracy rates in stented segments were both significantly lower with 64- [65% (95% CI: 52% to 76%) vs. 90% (88% to 92%), p < 0.05] and 256-slice CTCA [75% (95% CI: 62% to 85%) vs. 95% (95% CI: 93% to 96%), p < 0.05] as compared with those in segments without stent. The sensitivities of the 64- and 256-slice CTCA were both increased, but specificities decreased with the presence of severe calcifications. The diagnostic accuracy rates of 64- and 256-slice CTCA were both lower in segments with severe calcification, without any significant difference between them [75% (95% CI: 63% to 85%) vs. 82% (95% CI: 70% to 90%), p > 0.05].
DISCUSSION
Major findings
The 256-slice CTCA tolerated higher heart rates during CTCA without any negative impact on diagnostic performance. Overall, the diagnostic accuracy rates of 256-slice CTCA in per-patient, per-vessel and per-segment analysis were significantly better as compared to those of 64-slice CTCA. The 256-slice CTCA had better diagnostic accuracy rate for significant lesions in the RCA and LCX, which are often underestimated by 64-slice CTCA. Stented segments and the presence of severe calcifications continued to be limitations when using the 256-slice CTCA.
Comparison of technical characteristics between 64- and 256-slice MDCT
Technical improvements with the 256-slice MDCT provide multiple advantages compared to the 64-slice MDCT. The wider coverage of 256-slice MDCT, along with a fast gantry rotation (0.27 sec/rotation) results in shorter acquisition times, and therefore relaxes the breath-hold requirements, reduces the occurrence of respiratory arrhythmias, and leads to less susceptibility to sinus arrhythmia and ectopic beats during the acquisitions. In addition, dynamic helical collimation helps reduce excess radiation dose caused by helical over-scanning, a common occurrence in wide-area MDCT scanners.
256-slice CTCA overcomes several important limitations that are inherent in the 64-slice CTCA, such as the requirement of a regular and low heart rate. In the present study, 256-slice CTCA had similar diagnostic accuracy rate in patients with a heart rate of more than 70 beats per minute compared with a heart rate of less than 70 beats per minute. Also, previous studies demonstrated that 256-slice CTCA could be used over a wider range of heart rates and can tolerate sinus arrhythmia or ventricular premature beats without negatively impacting the overall diagnostic performance.13,14,16,20
Diagnostic performance of 256-slice CTCA compared with 64-slice CTCA
Previous studies demonstrated that sensitivity, specificity, PPV and NPV of 64-slice CTCA for detecting significant CAD were 85-99%, 64-90%, 64-91%, and 83-99% by per-patient analysis, respectively; 75-96%, 77-93%, 51-82%, and 89-99% by per-vessel analysis, respectively; and 88-90%, 90-94%, 47-54%, and 98-99% by per-segment analysis, respectively.2-6,10-12 The diagnostic performance of 64-slice CTCA in the present study was similar to those in the previous studies, except that 5 of 11 patients with insignificant CAD in the present study were incorrectly classified as having significant CAD, resulting in a relatively lower specificity and NPV of patient-based analysis compared with the previous studies. However, the diagnostic performances of vessel- and segment-based analysis in 64-slice CTCA in the present study were similar with those in the previous studies.
In our study, diagnostic accuracy measures improved with 256-slice CTCA compared with 64-slice CTCA in patient-, vessel-, and segment-based analysis. In the more clinically relevant patient-based analysis, none of the patients with significant CAD were incorrectly classified as negative by 256-slice CTCA, thus making 256-slice CTCA more reliable for ruling out significant coronary stenosis. Therefore, 256-slice CTCA is more accurate than 64-slice CTCA in determining obstructive CAD, and consequently can aid in preventing additional and possibly unnecessary invasive coronary angiography.
Previous studies showed that RCA and LCX were more prone to have motion artifacts compared to other coronary arteries.21,22 The lesions in the RCA and LCX were more often undetected by 64-slice CTCA than lesions in the LAD and LM.3 However, results from the current 256-slice CTCA showed a significant improvement of diagnostic performance for coronary artery stenosis in the RCA and LCX. The improvement in stenosis detection for the 256-slice CCTA in this study may be linked to better image quality of those vessels running in the atrioventricular groove, such as RCA (Figure 1) and LCX. The maximal motion of the heart is obtained from base to apex. Since the acquisition of the 256-slice helical CTCA data set is during the same 1-2 heart beats, there is no chance for inter-beat variability to hinder coronary arterial segment reconstruction. The RCA and LCX are quite prone to in-plane motion artifacts as well on the 64-slice CTCA. In addition, most of the patients had right dominant coronary artery, which means the LCX was relatively smaller than other coronary arteries.23,24 The increase in temporal resolution in 256-slice MDCT reduced the occurrence of motion artifacts, and provided a better analysis of more distal vessels.
Figure 1.
256-slice computed tomography coronary angiography (CTCA) of the right coronary artery (RCA). A volume-rendered three-dimensional CTCA image (A) reveals the anatomy of the RCA. A curved multiplanar reconstructed image (B) and a thick maximum-intensity projected image (C) disclose a significant coronary stenosis (arrows) in the proximal RCA, which was corroborated by conventional coronary angiogram (D). Significant calcified plaques can be seen proximally and distally to the significantly obstructive lesion, (C).
Our study showed that the diagnostic accuracy rate was significantly better for 256-slice CTCA in the detection of coronary artery stenosis in segments without stent or calcification compared with 64-slice CTCA. However, 256-slice CTCA has limitations similar to those of 64-slice CTCA, such as reduced efficacy when eva-luating coronary artery stenosis in vessels with severe calcification. By using 64- and 256-slice CTCA, the sensitivity increased and specificity decreased as the degree of calcification grew. Accurately evaluating the severity of a lesion is limited in vessels with calcification because of the blooming effect caused by high-density calcified lesions, routinely causing the severity of stenosis to be overestimated. Also, our study found that the diagnostic performance in detection of stent restenosis with 256-slice CTCA declined compared to segments without stent. Non-invasive assessment of coronary stent restenosis, while possible in CTCA, is still challenged by beam hardening caused by the stent structures.
Study limitations
The initial limitation was that our study was performed on a middle-aged, predominantly male population, which is not representative of patients with a low to intermediate probability for which CTCA is currently recommended.24 However, a study comparing CTCA with CCA in patients at low to intermediate risk would be difficult to perform because CCA is not always indicated. Second, although this scanner enables the use of axial prospective gating over a wider range of heart rates compared to 64-slice MDCT,20 we employed conventional (helical) acquisition techniques to compare the diagnostic performance with 64-slice MDCT. Third, our study population has a higher prevalence of disease than customarily seen in the general outpatient popu-lation, and few patients in our study had insignificant CAD, limiting the analysis of specificity and NPV. Fourth, this study focused on the diagnostic performance of determining obstructive disease, not on establishing the value in defining atherosclerotic plaque burden. Finally, though there was a trend of improved diagnostic performance in 256-slice CTCA for calcified and stented segments as compared to 64-slice CTCA, its statistical insignificance probably resulted from small sample numbers.
CONCLUSION
In two populations with similar characteristics, 256-slice CTCA had superior diagnostic accuracy rates for patient-, vessel- and segment-based analysis than 64-slice CTCA. The 256-slice CTCA had better diagnostic accuracies for detecting significant stenosis in the RCA and LCX, which were more commonly underestimated by 64-slice CTCA. However, severe calcification or stents pose challenges and thus limit the effectiveness of 256-slice CTCA for detection of > 50% coronary artery stenosis in CCA.
REFERENCES
- 1.Laskey W, Boyle J, Johnson LW. Multivariable model for prediction of risk of significant complication during diagnostic cardiac catheterization. The Registry Committee of the Society for Cardiac Angiography & Interventions. Cathet Cardiovasc Diagn. 1993;30:185–190. doi: 10.1002/ccd.1810300302. [DOI] [PubMed] [Google Scholar]
- 2.Miller JM, Rochitte CE, Dewey M, et al. Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med. 2008;359:2324–2336. doi: 10.1056/NEJMoa0806576. [DOI] [PubMed] [Google Scholar]
- 3.Meijboom WB, Meijs MF, Schuijf JD, et al. Diagnostic accuracy of 64-slice computed tomography coronary angiography: a prospective, multicenter, multivendor study. J Am Coll Cardiol. 2008;52:2135–2144. doi: 10.1016/j.jacc.2008.08.058. [DOI] [PubMed] [Google Scholar]
- 4.Abdulla J, Abildstrom SZ, Gotzsche O, et al. 64-multislice de-tector computed tomography coronary angiography as potential alternative to conventional coronary angiography: a systematic review and meta-analysis. Eur Heart J. 2007;28:3042–3050. doi: 10.1093/eurheartj/ehm466. [DOI] [PubMed] [Google Scholar]
- 5.Mowatt G, Cook JA, Hillis GS, et al. 64-Slice computed tomography angiography in the diagnosis and assessment of coronary artery disease: systematic review and meta-analysis. Heart. 2008;94:1386–1393. doi: 10.1136/hrt.2008.145292. [DOI] [PubMed] [Google Scholar]
- 6.Hamon M, Morello R, Riddell JW. Coronary arteries: diagnostic performance of 16- versus 64-section spiral CT compared with invasive coronary angiography -- meta-analysis. Radiology. 2007;245:720–731. doi: 10.1148/radiol.2453061899. [DOI] [PubMed] [Google Scholar]
- 7.Matsumoto N, Nagao K, Hirayama A, Sato Y. Non-invasive assessment and clinical strategy of stable coronary artery disease by magnetic resonance imaging, multislice computed tomography and myocardial perfusion SPECT. Circ J. 2010;74:34–40. doi: 10.1253/circj.cj-09-0791. [DOI] [PubMed] [Google Scholar]
- 8.Jinzaki M, Sato K, Tanami Y, et al. Diagnostic accuracy of angiographic view image for the detection of coronary artery stenoses by 64-detector row CT: a pilot study comparison with conventional post-processing methods and axial images alone. Circ J. 2009;73:691–698. doi: 10.1253/circj.cj-08-0798. [DOI] [PubMed] [Google Scholar]
- 9.Chang HU, Yin WS. Anomalous right coronary artery arising from left coronary sinus in two brothers. Acta Cardiol Sin. 2011;27:124–127. [Google Scholar]
- 10.Budoff MJ, Dowe D, Jollis JG, et al. Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing In-vasive Coronary Angiography) trial. J Am Coll Cardiol. 2008;52:1724–1732. doi: 10.1016/j.jacc.2008.07.031. [DOI] [PubMed] [Google Scholar]
- 11.Meijboom WB, van Mieghem CA, Mollet NR, et al. 64-slice computed tomography coronary angiography in patients with high, intermediate, or low pretest probability of significant coronary artery disease. J Am Coll Cardiol. 2007;50:1469–1475. doi: 10.1016/j.jacc.2007.07.007. [DOI] [PubMed] [Google Scholar]
- 12.Van Mieghem CA, Thury A, Meijboom WB, et al. Detection and characterization of coronary bifurcation lesions with 64-slice computed tomography coronary angiography. Eur Heart J. 2007;28:1968–1976. doi: 10.1093/eurheartj/ehm195. [DOI] [PubMed] [Google Scholar]
- 13.Wm , Guy Weigold MEO, Walker MJ. Low-dose prospectively gated 256-slice coronary computed tomographic angiography. Int J Cardiovasc Imaging. 2009;25:24. [Google Scholar]
- 14.Matthew J, Walker MEO, Desai MY, Halliburton SS, Flamm SD. New radiation dose saving technologies for 256-slice cardiac computed tomography angiography. Int J Cardiovasc Imaging. 2009;25:11. [Google Scholar]
- 15.Tariq A, Hameed SDT, Vembar M, Dharaiya E, Rydberg J. Low radiation dose ECG-gated chest CT angiography on a 256-slice multidetector CT scanner. Int J Cardiovasc Imaging. 2009;25:267–278. [Google Scholar]
- 16.Chao SP, Law WY, Kuo CJ, et al. The diagnostic accuracy of 256-row computed tomographic angiography compared with invasive coronary angiography in patients with suspected coronary artery disease. Eur Heart J. 2010;31:1916–1923. doi: 10.1093/eurheartj/ehq072. [DOI] [PubMed] [Google Scholar]
- 17.Manzke R, Grass M, Nielsen T, et al. Adaptive temporal resolution optimization in helical cardiac cone beam CT reconstruction. Med Phys. 2003;30:3072–3080. doi: 10.1118/1.1624756. [DOI] [PubMed] [Google Scholar]
- 18. Vembar M, Garcia MJ, Heuscher DJ, et al. A dynamic approach to identifying desired physiological phases for cardiac imaging using multislice spiral CT. Med Phys. 2003;30:1683–1693. doi: 10.1118/1.1582812. [DOI] [PubMed] [Google Scholar]
- 19.Austen WG, Edwards JE, Frye RL, et al. A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation. 1975;51:15–40. doi: 10.1161/01.cir.51.4.5. [DOI] [PubMed] [Google Scholar]
- 20.Hou Y, Yue Y, Guo W, et al. Prospectively versus retrospectively ECG-gated 256-slice coronary CT angiography: image quality and radiation dose over expanded heart rates. Int J Cardiovasc Imaging. 2012;28:153–162. doi: 10.1007/s10554-010-9760-7. [DOI] [PubMed] [Google Scholar]
- 21.Johnson KR, Patel SJ, Whigham A, et al. Three-dimensional, time-resolved motion of the coronary arteries. J Cardiovasc Magn Reson. 2004;6:663–673. doi: 10.1081/jcmr-120038086. [DOI] [PubMed] [Google Scholar]
- 22.Hofman MB, Wickline SA, Lorenz CH. Quantification of in-plane motion of the coronary arteries during the cardiac cycle: implications for acquisition window duration for MR flow quantification. J Magn Reson Imaging. 1998;8:568–576. doi: 10.1002/jmri.1880080309. [DOI] [PubMed] [Google Scholar]
- 23.DS B, et al. Grossman’s cardiac catheterization, angiography, and intervention. Philadelphia : Lippincott, Williams & Wilkins ; 2006. pp. 202– 203. [Google Scholar]
- 24.Andreini D, Pontone G, Bartorelli AL, et al. Comparison of feasibility and diagnostic accuracy of 64-slice multidetector computed tomographic coronary angiography versus invasive coronary angiography versus intravascular ultrasound for evaluation of in-stent restenosis. Am J Cardiol. 2009;103:349–358. doi: 10.1016/j.amjcard.2009.01.343. [DOI] [PubMed] [Google Scholar]