- Email: [email protected]
Automated Computer-Assisted Diagnosis of Obstructive Coronary Artery Disease in Emergency Department Patients Undergoing 256-Slice Coronary Computed Tomography Angiography for Acute Chest Pain
Automated Computer-Assisted Diagnosis of Obstructive Coronary Artery Disease in Emergency Department Patients Undergoing 256-Slice Coronary Computed Tomography Angiography for Acute Chest Pain Sharbell Hashoul, MDa,b, Tamar Gaspar, MDa,b, David A. Halon, MBChBa,b, Basil S. Lewis, MDa,b, Yuval Shenkar, MDa,b, Ronen Jaffe, MDa,b, Nathan Peled, MDa,b, and Ronen Rubinshtein, MDa,b,* A 256-slice coronary computed tomography angiography (CCTA) is an accurate method for detection and exclusion of obstructive coronary artery disease (OBS-CAD). However, accurate image interpretation requires expertise and may not be available at all hours. The purpose of this study was to evaluate the usefulness of a fully automated computer-assisted diagnosis (COMP-DIAG) tool for exclusion of OBS-CAD in patients in the emergency department (ED) presenting with chest pain. Three hundred sixty-nine patients in ED without known coronary disease underwent 256-slice CCTA as part of the assessment of chest pain of uncertain origin. COMP-DIAG (CorAnalyzer II) automatically reported presence or exclusion of OBS-CAD (>50% stenosis, ‡1 vessel). Performance characteristics of COMP-DIAG for exclusion and detection of OBS-CAD were determined using expert reading as the reference standard. Seventeen (5%) studies were unassessable by COMP-DIAG software, and 352 patients (1,056 vessels) were therefore available for analysis. COMP-DIAG identified 33% of assessable studies as having OBS-CAD, but the prevalence of OBS-CAD on CCTA was only 18% (66 of 352 patients) by standard expert reading. However, COMP-DIAG correctly identified 61 of the 66 patients (93%) with OBS-CAD with 21 vessels (2%) with OBS-CAD misclassified as negative. In conclusion, compared to expert reading, automated computer-assisted diagnosis using the CorAnalyzer showed high sensitivity but only moderate specificity for detection of obstructive coronary disease in patients in ED who underwent 256-slice CCTA. The high negative predictive value of this computer-assisted algorithm may be useful in the ED setting. Ó 2015 Elsevier Inc. All rights reserved. (Am J Cardiol 2015;116:1017e1021) New-generation computed tomography (CT) scanners now available have high accuracy for detection of obstructive coronary artery disease (OBS-CAD) and a low rate of unassessable segments allowing accurate diagnosis in most of the scanned patients and safe discharge when appropriate.1e10 However, even with new-generation CT scanners accurate image interpretation requires expertise and may not be available at all hours. To promote rapid and accurate diagnosis, automated computer-assisted diagnosis (COMP-DIAG) tools have become prevalent in general medicine and particularly in diagnostic radiology, for example, in diagnosis of breast imaging and lung nodules.11,12 Because accurate assessment of coronary computed tomography angiography (CCTA) is time consuming, requires expertise, and may not be available at all times, an automated tool has been developed and tested to identify potentially OBS-CAD from CCTA. This may be
a
Lady Davis Carmel Medical Center and bRuth and Bruce Rappaport Faculty of Medicine, TechnioneIsrael Institute of Technology, Haifa, Israel. Manuscript received May 12, 2015; revised manuscript received and accepted July 3, 2015. See page 1020 for disclosure information. *Corresponding author: Tel/fax: (þ972)-4-8250153. E-mail address: [email protected] (R. Rubinshtein). 0002-9149/15/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2015.07.014
particularly useful in the ED setting. Hence, the purpose of our study was to evaluate the performance characteristics of a COMP-DIAG tool in patients in ED with acute chest pain. Methods This single-center, retrospective, observational study was approved by the institutional review board with waiver of informed consent. The cohort included consecutive adult patients, without previously known CAD, who presented to the ED with chest pain of uncertain origin and underwent a 256-slice CCTA during ED evaluation. Study exclusion criteria were known CAD and/or previous revascularization or CCTA exclusion criteria (pregnancy, contrast allergy, renal function impairment [estimated glomerular filtration rate <60 ml/min], and unable to cooperate). CCTA was performed using a 256-row scanner (Brilliance iCT; Philips Healthcare, Cleveland, Ohio), which has a longitudinal coverage of 8 cm, rotation time of 0.27 seconds, and a 120-kW generator. CCTA was performed either as prospectively triggered “step-and-shoot” scans or with helical retrospective electrocardiographic (ECG) gating. Oral and/or intravenous b blockers were used to lower heart rate when >70 beats/min. Sublingual nitroglycerine (0.4 mg) was given before CCTA to all patients with systolic blood pressure of 110 mm Hg and www.ajconline.org
1018
The American Journal of Cardiology (www.ajconline.org)
Table 1 Patient characteristics (N¼352) Age (years)
Table 2 256-slice coronary CT angiography scan data SD Range
Men Women Smoker Diabetes mellitus Hyperlipidemia* Hypertension† Family history of coronary artery disease
55.5 12.5 17-89 208 (59%) 41% 33% 21% 50% 54% 19%
* Total serum cholesterol level>5 mmol/l or treatment with lipid lowering drugs. † Systemic blood pressure>140/90 or treatment with antihypertensive medication.
no clinical contraindications (such as aortic stenosis or suspected pulmonary embolism). CCTA was performed using a bolus of iohexol (Omnipaque 350 mg I/ml, GE Healthcare, Princeton, New Jersey) injected into an antecubital vein at a flow rate of 5 to 6 ml/s, followed by a mixed 50% contrast or saline injection and then a 20- to 30-ml saline chaser bolus. Iohexol dose was calculated according to the following formula: (predicted scan time in seconds þ 5) intravenous contrast flow rate. Both modes of scans were performed at 120 kV with a slice collimation of 128 0.625 mm with a dynamic dual focal spot (therefore 256-slice acquisition) and a rotation time of 0.27 or 0.33 seconds. The helical scans (retrospective ECG gating) were performed with an effective tube current (rotation time product normalized by the pitch) in the range of 900 to 1500 mAs (eff) depending on body mass index and body habitus, and a pitch of 0.14 to 0.2. ECG-based tube-current modulation was used when appropriate. The prospectively triggered scans were performed in patients with stable heart rhythm and heart rate <65 beats/min with a tube currentex-ray ON time product of 160 to 300 mAs. Radiation exposure was assessed as doseelength product (product of scan length and CT dose index). Reconstruction was performed using a window centered at 75% of the R-R interval as default. For heart rates >70 beats/min an earlier reconstruction phase (usually 45%) was frequently used with retrospective gating. Two expert readers (>10 years of experience each in CCTA; standard reading) blinded to COMP-DIAG results diagnosed and reported significant coronary stenosis (OBSCAD) (>50% diameter stenosis, 1 vessel) by consensus. The CorAnalyzer II tool (Rcadia Medical Imaging, Haifa, Israel) was used for COMP-DIAG. This system installed in a standard workstation classifies the coronary artery system into 3 main arteries and 10 coronary segments: the left main coronary artery is presented together with the left anterior descending artery (the left anterior descending artery itself is divided to proximal, middle, and distal segments), the left circumflex, and the right coronary arteries are divided into 3 segments (proximal, middle and distal). Each artery is classified by the software to 1 of 3 categories: potentially stenotic (50% diameter stenosis), without significant stenosis (<50%), or indecisive (expert reading required). The latter would usually be reported by the system in the case of an unassessable segment.
Heart rate (BPM) (mean SD) Volume of IV contrast (ml) (mean SD) Dose length product (total) (mGy x cm) (mean SD) Dose length product of CTA (mGy x cm) (mean SD) Step & Shoot mode (percent)
63 10 73.5 10 955 695 713 629 97/333 (29%)
Baseline characteristics and imaging and scanning parameters were recorded using descriptive statistics. Performance characteristics (patient-based and vessel-based) were calculated using CCTA expert reading results as the reference standard. A p value of <0.05 was considered significant. Statistical analysis was performed using Statistix 8 software package (Analytical Software, Tallahassee, Florida). Results Three hundred sixty-nine patients in ED with chest pain and without previously known CAD were originally included. However, 17 of 369 studies (5%) were unassessable by the automated COMP-DIAG and were excluded from analysis. Three hundred fifty-two patients (352 3 ¼ 1,056 vessels) were therefore available for comparative analysis. Demographic data of the 352 patients are presented in Table 1. CCTA imaging parameters are presented in Table 2. Expert reading of CCTA, which served as the reference standard diagnosed OBS-CAD in 107 of 1,056 vessels (10%) among 66 of 352 (18%) of the patients. By COMP-DIAG, OBS-CAD was diagnosed in 301 of 1,056 vessels (28%) in 116 of 352 patients (33%). Using expert reading as the reference standard, the sensitivity of COMP-DIAG to diagnose OBS-CAD was 93% (95% confidence interval [CI] 87 to 99) as 5 patients were incorrectly classified by the software as not having obstructive lesions. One hundred twenty-three patients were misclassified as having OBS-CAD by COMP-DIAG (false positive). Therefore, specificity was 56% (95% CI 50 to 62; Table 3). As reported previously, expert reading diagnosed OBSCAD in 107 of 1,056 (10%) vessels, but COMP-DIAG diagnosed 301 of 1,056 (28%) vessels as having OBSCAD. Thus, 194 vessels were misclassified as having OBS-CAD (false positive), in addition to 21 vessels with OBS-CAD misclassified as nonobstructive (false negative). Therefore, on a vessel-based analysis, the sensitivity of COMP-DIAG to diagnose OBS-CAD was 80% (73% to 88%) and specificity was 77% (75% to 80%; Table 3). An example of a true-positive case is presented in Figure 1. Of note, no significant differences in terms of diagnostic accuracy were apparent among the 3 coronary vessels reported by the system (LM-LAD/LCX/RCA). Discussion Our study showed that in a group of patients in ED presenting with acute chest pain and who underwent 256-slice CCTA, a rapid, fully automatic, computer-based
Coronary Artery Disease/Computer-Assisted Diagnosis of Coronary CTA
1019
Table 3 Performance of the CorAnalyzer (COMP-DIAG) for detection of obstructive coronary artery disease in emergency department patients with acute chest pain Analysis
N
Obstructive coronary disease by COMP-DIAG
Obstructive coronary disease by expert reading
Sensitivity (%, 95% CI)
Specificity (%, 95% CI)
PPV (%, 95% CI)
NPV (%, 95% CI)
Patient based
352
116 (33%)
66 (18%)
Vessel based
1056
301 (28%)
107 (10%)
93% (87-99) 66/71 80% (73-88) 86/107
56% (50-62) 158/281 77% (75-80) 734/949
35% (28-42) 66/189 29% (24-34) 86/301
97% (94-100) 158/163 97% (96-98) 734/755
COMP-DIAG ¼ computer assisted diagnosis using the CorAnalyzer; NPV ¼ negative predictive value; PPV ¼ positive predictive value.
Figure 1. Significant stenosis in the left anterior descending coronary artery by both the COR Analyzer [automated screenshot; (A), “Significant” result marked and (B), segmentation process screen] and expert reading [multiplanar reformation, (C)].
algorithm was able to diagnose most patients with normal coronary arteries or nonobstructive atheroma with a relatively high negative predictive value of 97% on both a patient-based and vessel-based analysis. On a
patient-based analysis, the automated system showed high sensitivity but limited specificity. Our findings and especially the high negative predictive value imply possible usefulness of COMP-DIAG in the ED setting where rapid
1020
The American Journal of Cardiology (www.ajconline.org)
exclusion of OBS-CAD is important and may lead to rapid discharge. Recent trends in ED practice emphasize the usefulness and efficacy of a problem-oriented and sophisticated workup in the ED setting, including cardiac imaging, to prevent unnecessary hospitalizations in patients presenting with chest pain syndromes and other common complaints (such as syncope).13 Several studies in recent years1,2,5,14 including from our institution3 demonstrated the usefulness of CCTA in the ED setting. Of these studies, 3 major multicenter prospective trials established the high negative predictive value of CCTA compared to standard protocol.1,5,14 CCTA-based ED chest pain protocol allows safe discharge of low-to-intermediate risk patients from the ED, and a CCTA-based chest pain protocol is also advantageous in terms of lowering cost and achieving a shorter ED stay compared to standard ED triage.1,2,5,14 Although CCTA has become an established tool for chest pain assessment in the ED,15,16 accurate image interpretation requires expert readers which are not always available, therefore limiting the usefulness and applicability of a CCTA-based acute chest pain protocol in the ED. This is why automated computer-assisted diagnostic tools may be of assistance. We have recently shown that COMP-DIAG using the CorAnalyzer was associated with a low rate of false-negative results even in a group of patients with high prevalence of disease who underwent both CCTA and invasive coronary angiography.17 However, the main disadvantage of the CorAnalyzer system in the present study and in previous publication relates to its “extra safety” features exacerbating the tendency of CCTA for overdiagnosis (high false-positive rate). This high false-positive rate limits the number of patients who will be diagnosed as normal by the system (in whom, analysis by an experienced observer might be postponed to a later date). A further limitation of the present system lies in its method of assessment of the main coronary arteries: left main, left anterior descending, circumflex, and the right coronary artery which it divides into 10 segments. Stenosis in a smaller size coronary branch such as the diagonal, marginal, intermediate, or possibly important right coronary artery posterolateral and posterior descending branches may warrant a comment from the system that further analysis is required. However, for small branches that elicit no comment, the user does not really know whether they were assessed and found to be free from significant stenosis or were not assessed at all because of their small diameter. This is undoubtedly of less importance in very small and distal coronary segments, and in the present study, we observed that in most first-order coronary branches, the system did provide a diagnosis. Moreover, false-positive results in our study were often caused by either unusual coronary anatomy or by reduced image quality (as previously reported).18 One other possible use of the CorAnalyzer in the ED setting could be an automatic assessment of the Agatston calcium score from the contrast enhanced scan (CCTA) which was developed for this tool and was recently shown to be feasible and accurate compared to standard calcium scoring calculated from nonenhanced scans.19 A zero calcium score in the ED has a high negative predictive value for adverse cardiac events although it may not totally exclude the presence of a noncalcified obstructive lesion.20e22
Our study was a single-center retrospective study, and a selection bias may have been introduced; however, the high negative predictive value of the automated system and the (increased) tendency for false-positive results have also been consistent with previous publications.23,24 Of note, the reference standard for analysis was expert reading rather than invasive coronary angiography, but the results are not much different than what we had previously reported for this system in patients with high prevalence of OBS-CAD (unlike patients in ED) undergoing invasive angiography.17 In fact, a comparison of the system performance characteristics with expert reading (as the reference standard) is probably more realistic for assessment of an ED-based technology and estimating how useful it is in allowing rapid discharge rather than waiting for an expert reader report. An additional limitation of our results relate to the 50% diameter stenosis cutoff used by the automated system for making a diagnosis of OBS-CAD rather than a 70% cut-off value suggested by some as the recommended cut-off value for OBS-CAD by CCTA, which if used may have decreased the false-positive rate. Our method therefore includes a similar 50% cut-off value for expert reading (reference standard) to match the automatic system programming and for identifying potentially obstructive CAD. Disclosures Dr. Peled serves on the advisory board of Rcadia Medical Imaging. The other authors have no conflicts of interest to disclose. 1. Litt HI, Gatsonis C, Snyder B, Singh H, Miller CD, Entrikin DW, Leaming JM, Gavin LJ, Pacella CB, Hollander JE. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med 2012;366:1393e1403. 2. Hoffmann U, Bamberg F, Chae CU, Nichols JH, Rogers IS, Seneviratne SK, Truong QA, Cury RC, Abbara S, Shapiro MD, Moloo J, Butler J, Ferencik M, Lee H, Jang IK, Parry BA, Brown DF, Udelson JE, Achenbach S, Brady TJ, Nagurney JT. Coronary computed tomography angiography for early triage of patients with acute chest pain: the ROMICAT (Rule Out Myocardial Infarction using Computer Assisted Tomography) trial. J Am Coll Cardiol 2009;53:1642e1650. 3. Rubinshtein R, Halon DA, Gaspar T, Jaffe R, Karkabi B, Flugelman M, Kogan A, Shapira R, Peled N, Lewis BS. Usefulness of 64-Slice cardiac computed tomographic angiography for diagnosing acute coronary syndromes and predicting clinical outcome in emergency department patients with chest pain of uncertain origin. Circulation 2007;115: 1762e1768. 4. Goldstein JA, Gallagher MJ, O’Neill WW, Ross MA, O’Neil BJ, Raff GL. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49:863e871. 5. Goldstein JA, Chinnaiyan KM, Abidov A, Achenbach S, Berman DS, Hayes SW, Hoffmann U, Lesser JR, Mikati IA, O’Neil BJ, Shaw LJ, Shen MY, Valeti US, Raff GL. The CT-STAT (coronary computed tomographic angiography for systematic triage of acute chest pain patients to treatment) trial. J Am Coll Cardiol 2011;58:1414e1422. 6. LaBounty TM, Earls JP, Leipsic J, Heilbron B, Mancini GB, Lin FY, Dunning AM, Min JK. Effect of a standardized quality-improvement protocol on radiation dose in coronary computed tomographic angiography. Am J Cardiol 2010;106:1663e1667. 7. Salavati A, Radmanesh F, Heidari K, Dwamena BA, Kelly AM, Cronin P. Dual-source computed tomography angiography for diagnosis and assessment of coronary artery disease: systematic review and metaanalysis. J Cardiovasc Comput Tomogr 2012;6:78e90. 8. de Graaf FR, Schuijf JD, van Velzen JE, Kroft LJ, de Roos A, Reiber JH, Boersma E, Schalij MJ, Spanó F, Jukema JW, van der Wall EE, Bax JJ. Diagnostic accuracy of 320-row multidetector computed
Coronary Artery Disease/Computer-Assisted Diagnosis of Coronary CTA
9.
10. 11. 12. 13.
14.
15.
tomography coronary angiography in the non-invasive evaluation of significant coronary artery disease. Eur Heart J 2010;31: 1908e1915. Chao SP, Law WY, Kuo CJ, Hung HF, Cheng JJ, Lo HM, Shyu KG. 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: 1916e1923. Petcherski O, Gaspar T, Halon DA, Peled N, Lewis BS, Rubinshtein R. High diagnostic accuracy of 256-row CT angiography for detection of obstructive coronary artery disease. Am J Cardiol 2013;111:510e515. Doi K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Comput Med Imaging Graph 2007;31:198e211. Giger ML, Chan HP, Boone J. Anniversary paper: history and status of CAD and quantitative image analysis: the role of medical physics and AAPM. Med Phys 2008;35:5799e5820. Shen WK, Decker WW, Smars PA, Goyal DG, Walker AE, Hodge DO, Trusty JM, Brekke KM, Jahangir A, Brady PA, Munger TM, Gersh BJ, Hammill SC, Frye RL. Syncope evaluation in the emergency department study (SEEDS). A multidisciplinary approach to syncope management. Circulation 2004;110:3636e3645. Hoffmann U, Truong QA, Schoenfeld DA, Chou ET, Woodard PK, Nagurney JT, Pope JH, Hauser TH, White CS, Weiner SG, Kalanjian S, Mullins ME, Mikati I, Peacock WF, Zakroysky P, Hayden D, Goehler A, Lee H, Gazelle GS, Wiviott SD, Fleg JL, Udelson JE; ROMICAT-II Investigators. Coronary CT angiography versus standard evaluation in acute chest pain. N Engl J Med 2012;367:299e308. Taylor AJ, Cerqueira M, Hodgson JM, Mark D, Min J, O’Gara P, Rubin GD. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/ SCMR 2010 appropriate use criteria for cardiac computed tomography. A report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2010;56:1864e1894.
1021
16. Rubinshtein R, Wolak A, Goitein O, Kornowski R, Gaspar T, Aviram G, Roguin A, Blinder G, Shemesh J, Konen E. Appropriateness criteria for the use of cardiac computed tomography: position paper of the Israeli Heart Society and the Israeli Society of Radiology. Harefuah 2011;150:801e805. 17. Blank N, Gaspar T, Halon DA, Peled N, Petcherski O, Lewis BS, Rubinshtein R. Computer-assisted diagnosis of obstructive coronary atherosclerosis in patients undergoing 256-slice coronary computed tomography angiography: a comparison with invasive coronary angiography. Int J Cardiol 2014;172:130e131. 18. Halpern EJ, Halpern DJ. Diagnosis of coronary stenosis with CT angiography comparison of automated computer diagnosis with expert readings. Acad Radiol 2011;18:324e333. 19. Rubinshtein R, Halon DA, Gaspar T, Lewis BS, Peled N. Automatic assessment of coronary artery calcium score from contrast-enhanced 256-row coronary computed tomography angiography. Am J Cardiol 2013;113:7e11. 20. Rumberger J, Sheedy P, Breen JF, Schwartz R. Electron beam computed tomographic coronary calcium score cutpoints and severity of associated angiographic lumen stenosis. J Am Coll Cardiol 1997;29: 1542e1548. 21. Rubinshtein R, Gaspar T, Halon DA, Goldstein J, Peled N, Lewis BS. Prevalence and extent of obstructive coronary artery disease in patients with zero or low calcium score undergoing 64-slice cardiac multidetector computed tomography for evaluation of a chest pain syndrome. Am J Cardiol 2007;99:472e475. 22. Gottlieb I, Miller JM, Arbab-Zadeh A, Dewey M, Clouse ME, Sara L, Niinuma H, Bush DE, Paul N, Vavere AL, Texter J, Brinker J, Lima JA, Rochitte CE. The absence of coronary calcification does not exclude obstructive coronary artery disease or the need for revascularization in patients referred for conventional coronary angiography. J Am Coll Cardiol 2010;55:627e634. 23. Kang KW, Chang HJ, Shim H, Kim YJ, Choi BW, Yang WI, Shim JY, Ha J, Chung N. Feasibility of an automatic computer-assisted algorithm for the detection of significant coronary artery disease in patients presenting with acute chest pain. Eur J Radiol 2012;81:640e646. 24. Goldenberg R, Eilot D, Begelman G, Walach E, Ben-Ishai E, Peled N. Computer-aided simple triage (CAST) for coronary CT angiography (CCTA). Int J Comput Assist Radiol Surg 2012;7:819e827.
a
Lady Davis Carmel Medical Center and bRuth and Bruce Rappaport Faculty of Medicine, TechnioneIsrael Institute of Technology, Haifa, Israel. Manuscript received May 12, 2015; revised manuscript received and accepted July 3, 2015. See page 1020 for disclosure information. *Corresponding author: Tel/fax: (þ972)-4-8250153. E-mail address: [email protected] (R. Rubinshtein). 0002-9149/15/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2015.07.014
particularly useful in the ED setting. Hence, the purpose of our study was to evaluate the performance characteristics of a COMP-DIAG tool in patients in ED with acute chest pain. Methods This single-center, retrospective, observational study was approved by the institutional review board with waiver of informed consent. The cohort included consecutive adult patients, without previously known CAD, who presented to the ED with chest pain of uncertain origin and underwent a 256-slice CCTA during ED evaluation. Study exclusion criteria were known CAD and/or previous revascularization or CCTA exclusion criteria (pregnancy, contrast allergy, renal function impairment [estimated glomerular filtration rate <60 ml/min], and unable to cooperate). CCTA was performed using a 256-row scanner (Brilliance iCT; Philips Healthcare, Cleveland, Ohio), which has a longitudinal coverage of 8 cm, rotation time of 0.27 seconds, and a 120-kW generator. CCTA was performed either as prospectively triggered “step-and-shoot” scans or with helical retrospective electrocardiographic (ECG) gating. Oral and/or intravenous b blockers were used to lower heart rate when >70 beats/min. Sublingual nitroglycerine (0.4 mg) was given before CCTA to all patients with systolic blood pressure of 110 mm Hg and www.ajconline.org
1018
The American Journal of Cardiology (www.ajconline.org)
Table 1 Patient characteristics (N¼352) Age (years)
Table 2 256-slice coronary CT angiography scan data SD Range
Men Women Smoker Diabetes mellitus Hyperlipidemia* Hypertension† Family history of coronary artery disease
55.5 12.5 17-89 208 (59%) 41% 33% 21% 50% 54% 19%
* Total serum cholesterol level>5 mmol/l or treatment with lipid lowering drugs. † Systemic blood pressure>140/90 or treatment with antihypertensive medication.
no clinical contraindications (such as aortic stenosis or suspected pulmonary embolism). CCTA was performed using a bolus of iohexol (Omnipaque 350 mg I/ml, GE Healthcare, Princeton, New Jersey) injected into an antecubital vein at a flow rate of 5 to 6 ml/s, followed by a mixed 50% contrast or saline injection and then a 20- to 30-ml saline chaser bolus. Iohexol dose was calculated according to the following formula: (predicted scan time in seconds þ 5) intravenous contrast flow rate. Both modes of scans were performed at 120 kV with a slice collimation of 128 0.625 mm with a dynamic dual focal spot (therefore 256-slice acquisition) and a rotation time of 0.27 or 0.33 seconds. The helical scans (retrospective ECG gating) were performed with an effective tube current (rotation time product normalized by the pitch) in the range of 900 to 1500 mAs (eff) depending on body mass index and body habitus, and a pitch of 0.14 to 0.2. ECG-based tube-current modulation was used when appropriate. The prospectively triggered scans were performed in patients with stable heart rhythm and heart rate <65 beats/min with a tube currentex-ray ON time product of 160 to 300 mAs. Radiation exposure was assessed as doseelength product (product of scan length and CT dose index). Reconstruction was performed using a window centered at 75% of the R-R interval as default. For heart rates >70 beats/min an earlier reconstruction phase (usually 45%) was frequently used with retrospective gating. Two expert readers (>10 years of experience each in CCTA; standard reading) blinded to COMP-DIAG results diagnosed and reported significant coronary stenosis (OBSCAD) (>50% diameter stenosis, 1 vessel) by consensus. The CorAnalyzer II tool (Rcadia Medical Imaging, Haifa, Israel) was used for COMP-DIAG. This system installed in a standard workstation classifies the coronary artery system into 3 main arteries and 10 coronary segments: the left main coronary artery is presented together with the left anterior descending artery (the left anterior descending artery itself is divided to proximal, middle, and distal segments), the left circumflex, and the right coronary arteries are divided into 3 segments (proximal, middle and distal). Each artery is classified by the software to 1 of 3 categories: potentially stenotic (50% diameter stenosis), without significant stenosis (<50%), or indecisive (expert reading required). The latter would usually be reported by the system in the case of an unassessable segment.
Heart rate (BPM) (mean SD) Volume of IV contrast (ml) (mean SD) Dose length product (total) (mGy x cm) (mean SD) Dose length product of CTA (mGy x cm) (mean SD) Step & Shoot mode (percent)
63 10 73.5 10 955 695 713 629 97/333 (29%)
Baseline characteristics and imaging and scanning parameters were recorded using descriptive statistics. Performance characteristics (patient-based and vessel-based) were calculated using CCTA expert reading results as the reference standard. A p value of <0.05 was considered significant. Statistical analysis was performed using Statistix 8 software package (Analytical Software, Tallahassee, Florida). Results Three hundred sixty-nine patients in ED with chest pain and without previously known CAD were originally included. However, 17 of 369 studies (5%) were unassessable by the automated COMP-DIAG and were excluded from analysis. Three hundred fifty-two patients (352 3 ¼ 1,056 vessels) were therefore available for comparative analysis. Demographic data of the 352 patients are presented in Table 1. CCTA imaging parameters are presented in Table 2. Expert reading of CCTA, which served as the reference standard diagnosed OBS-CAD in 107 of 1,056 vessels (10%) among 66 of 352 (18%) of the patients. By COMP-DIAG, OBS-CAD was diagnosed in 301 of 1,056 vessels (28%) in 116 of 352 patients (33%). Using expert reading as the reference standard, the sensitivity of COMP-DIAG to diagnose OBS-CAD was 93% (95% confidence interval [CI] 87 to 99) as 5 patients were incorrectly classified by the software as not having obstructive lesions. One hundred twenty-three patients were misclassified as having OBS-CAD by COMP-DIAG (false positive). Therefore, specificity was 56% (95% CI 50 to 62; Table 3). As reported previously, expert reading diagnosed OBSCAD in 107 of 1,056 (10%) vessels, but COMP-DIAG diagnosed 301 of 1,056 (28%) vessels as having OBSCAD. Thus, 194 vessels were misclassified as having OBS-CAD (false positive), in addition to 21 vessels with OBS-CAD misclassified as nonobstructive (false negative). Therefore, on a vessel-based analysis, the sensitivity of COMP-DIAG to diagnose OBS-CAD was 80% (73% to 88%) and specificity was 77% (75% to 80%; Table 3). An example of a true-positive case is presented in Figure 1. Of note, no significant differences in terms of diagnostic accuracy were apparent among the 3 coronary vessels reported by the system (LM-LAD/LCX/RCA). Discussion Our study showed that in a group of patients in ED presenting with acute chest pain and who underwent 256-slice CCTA, a rapid, fully automatic, computer-based
Coronary Artery Disease/Computer-Assisted Diagnosis of Coronary CTA
1019
Table 3 Performance of the CorAnalyzer (COMP-DIAG) for detection of obstructive coronary artery disease in emergency department patients with acute chest pain Analysis
N
Obstructive coronary disease by COMP-DIAG
Obstructive coronary disease by expert reading
Sensitivity (%, 95% CI)
Specificity (%, 95% CI)
PPV (%, 95% CI)
NPV (%, 95% CI)
Patient based
352
116 (33%)
66 (18%)
Vessel based
1056
301 (28%)
107 (10%)
93% (87-99) 66/71 80% (73-88) 86/107
56% (50-62) 158/281 77% (75-80) 734/949
35% (28-42) 66/189 29% (24-34) 86/301
97% (94-100) 158/163 97% (96-98) 734/755
COMP-DIAG ¼ computer assisted diagnosis using the CorAnalyzer; NPV ¼ negative predictive value; PPV ¼ positive predictive value.
Figure 1. Significant stenosis in the left anterior descending coronary artery by both the COR Analyzer [automated screenshot; (A), “Significant” result marked and (B), segmentation process screen] and expert reading [multiplanar reformation, (C)].
algorithm was able to diagnose most patients with normal coronary arteries or nonobstructive atheroma with a relatively high negative predictive value of 97% on both a patient-based and vessel-based analysis. On a
patient-based analysis, the automated system showed high sensitivity but limited specificity. Our findings and especially the high negative predictive value imply possible usefulness of COMP-DIAG in the ED setting where rapid
1020
The American Journal of Cardiology (www.ajconline.org)
exclusion of OBS-CAD is important and may lead to rapid discharge. Recent trends in ED practice emphasize the usefulness and efficacy of a problem-oriented and sophisticated workup in the ED setting, including cardiac imaging, to prevent unnecessary hospitalizations in patients presenting with chest pain syndromes and other common complaints (such as syncope).13 Several studies in recent years1,2,5,14 including from our institution3 demonstrated the usefulness of CCTA in the ED setting. Of these studies, 3 major multicenter prospective trials established the high negative predictive value of CCTA compared to standard protocol.1,5,14 CCTA-based ED chest pain protocol allows safe discharge of low-to-intermediate risk patients from the ED, and a CCTA-based chest pain protocol is also advantageous in terms of lowering cost and achieving a shorter ED stay compared to standard ED triage.1,2,5,14 Although CCTA has become an established tool for chest pain assessment in the ED,15,16 accurate image interpretation requires expert readers which are not always available, therefore limiting the usefulness and applicability of a CCTA-based acute chest pain protocol in the ED. This is why automated computer-assisted diagnostic tools may be of assistance. We have recently shown that COMP-DIAG using the CorAnalyzer was associated with a low rate of false-negative results even in a group of patients with high prevalence of disease who underwent both CCTA and invasive coronary angiography.17 However, the main disadvantage of the CorAnalyzer system in the present study and in previous publication relates to its “extra safety” features exacerbating the tendency of CCTA for overdiagnosis (high false-positive rate). This high false-positive rate limits the number of patients who will be diagnosed as normal by the system (in whom, analysis by an experienced observer might be postponed to a later date). A further limitation of the present system lies in its method of assessment of the main coronary arteries: left main, left anterior descending, circumflex, and the right coronary artery which it divides into 10 segments. Stenosis in a smaller size coronary branch such as the diagonal, marginal, intermediate, or possibly important right coronary artery posterolateral and posterior descending branches may warrant a comment from the system that further analysis is required. However, for small branches that elicit no comment, the user does not really know whether they were assessed and found to be free from significant stenosis or were not assessed at all because of their small diameter. This is undoubtedly of less importance in very small and distal coronary segments, and in the present study, we observed that in most first-order coronary branches, the system did provide a diagnosis. Moreover, false-positive results in our study were often caused by either unusual coronary anatomy or by reduced image quality (as previously reported).18 One other possible use of the CorAnalyzer in the ED setting could be an automatic assessment of the Agatston calcium score from the contrast enhanced scan (CCTA) which was developed for this tool and was recently shown to be feasible and accurate compared to standard calcium scoring calculated from nonenhanced scans.19 A zero calcium score in the ED has a high negative predictive value for adverse cardiac events although it may not totally exclude the presence of a noncalcified obstructive lesion.20e22
Our study was a single-center retrospective study, and a selection bias may have been introduced; however, the high negative predictive value of the automated system and the (increased) tendency for false-positive results have also been consistent with previous publications.23,24 Of note, the reference standard for analysis was expert reading rather than invasive coronary angiography, but the results are not much different than what we had previously reported for this system in patients with high prevalence of OBS-CAD (unlike patients in ED) undergoing invasive angiography.17 In fact, a comparison of the system performance characteristics with expert reading (as the reference standard) is probably more realistic for assessment of an ED-based technology and estimating how useful it is in allowing rapid discharge rather than waiting for an expert reader report. An additional limitation of our results relate to the 50% diameter stenosis cutoff used by the automated system for making a diagnosis of OBS-CAD rather than a 70% cut-off value suggested by some as the recommended cut-off value for OBS-CAD by CCTA, which if used may have decreased the false-positive rate. Our method therefore includes a similar 50% cut-off value for expert reading (reference standard) to match the automatic system programming and for identifying potentially obstructive CAD. Disclosures Dr. Peled serves on the advisory board of Rcadia Medical Imaging. The other authors have no conflicts of interest to disclose. 1. Litt HI, Gatsonis C, Snyder B, Singh H, Miller CD, Entrikin DW, Leaming JM, Gavin LJ, Pacella CB, Hollander JE. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med 2012;366:1393e1403. 2. Hoffmann U, Bamberg F, Chae CU, Nichols JH, Rogers IS, Seneviratne SK, Truong QA, Cury RC, Abbara S, Shapiro MD, Moloo J, Butler J, Ferencik M, Lee H, Jang IK, Parry BA, Brown DF, Udelson JE, Achenbach S, Brady TJ, Nagurney JT. Coronary computed tomography angiography for early triage of patients with acute chest pain: the ROMICAT (Rule Out Myocardial Infarction using Computer Assisted Tomography) trial. J Am Coll Cardiol 2009;53:1642e1650. 3. Rubinshtein R, Halon DA, Gaspar T, Jaffe R, Karkabi B, Flugelman M, Kogan A, Shapira R, Peled N, Lewis BS. Usefulness of 64-Slice cardiac computed tomographic angiography for diagnosing acute coronary syndromes and predicting clinical outcome in emergency department patients with chest pain of uncertain origin. Circulation 2007;115: 1762e1768. 4. Goldstein JA, Gallagher MJ, O’Neill WW, Ross MA, O’Neil BJ, Raff GL. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol 2007;49:863e871. 5. Goldstein JA, Chinnaiyan KM, Abidov A, Achenbach S, Berman DS, Hayes SW, Hoffmann U, Lesser JR, Mikati IA, O’Neil BJ, Shaw LJ, Shen MY, Valeti US, Raff GL. The CT-STAT (coronary computed tomographic angiography for systematic triage of acute chest pain patients to treatment) trial. J Am Coll Cardiol 2011;58:1414e1422. 6. LaBounty TM, Earls JP, Leipsic J, Heilbron B, Mancini GB, Lin FY, Dunning AM, Min JK. Effect of a standardized quality-improvement protocol on radiation dose in coronary computed tomographic angiography. Am J Cardiol 2010;106:1663e1667. 7. Salavati A, Radmanesh F, Heidari K, Dwamena BA, Kelly AM, Cronin P. Dual-source computed tomography angiography for diagnosis and assessment of coronary artery disease: systematic review and metaanalysis. J Cardiovasc Comput Tomogr 2012;6:78e90. 8. de Graaf FR, Schuijf JD, van Velzen JE, Kroft LJ, de Roos A, Reiber JH, Boersma E, Schalij MJ, Spanó F, Jukema JW, van der Wall EE, Bax JJ. Diagnostic accuracy of 320-row multidetector computed
Coronary Artery Disease/Computer-Assisted Diagnosis of Coronary CTA
9.
10. 11. 12. 13.
14.
15.
tomography coronary angiography in the non-invasive evaluation of significant coronary artery disease. Eur Heart J 2010;31: 1908e1915. Chao SP, Law WY, Kuo CJ, Hung HF, Cheng JJ, Lo HM, Shyu KG. 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: 1916e1923. Petcherski O, Gaspar T, Halon DA, Peled N, Lewis BS, Rubinshtein R. High diagnostic accuracy of 256-row CT angiography for detection of obstructive coronary artery disease. Am J Cardiol 2013;111:510e515. Doi K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Comput Med Imaging Graph 2007;31:198e211. Giger ML, Chan HP, Boone J. Anniversary paper: history and status of CAD and quantitative image analysis: the role of medical physics and AAPM. Med Phys 2008;35:5799e5820. Shen WK, Decker WW, Smars PA, Goyal DG, Walker AE, Hodge DO, Trusty JM, Brekke KM, Jahangir A, Brady PA, Munger TM, Gersh BJ, Hammill SC, Frye RL. Syncope evaluation in the emergency department study (SEEDS). A multidisciplinary approach to syncope management. Circulation 2004;110:3636e3645. Hoffmann U, Truong QA, Schoenfeld DA, Chou ET, Woodard PK, Nagurney JT, Pope JH, Hauser TH, White CS, Weiner SG, Kalanjian S, Mullins ME, Mikati I, Peacock WF, Zakroysky P, Hayden D, Goehler A, Lee H, Gazelle GS, Wiviott SD, Fleg JL, Udelson JE; ROMICAT-II Investigators. Coronary CT angiography versus standard evaluation in acute chest pain. N Engl J Med 2012;367:299e308. Taylor AJ, Cerqueira M, Hodgson JM, Mark D, Min J, O’Gara P, Rubin GD. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/ SCMR 2010 appropriate use criteria for cardiac computed tomography. A report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2010;56:1864e1894.
1021
16. Rubinshtein R, Wolak A, Goitein O, Kornowski R, Gaspar T, Aviram G, Roguin A, Blinder G, Shemesh J, Konen E. Appropriateness criteria for the use of cardiac computed tomography: position paper of the Israeli Heart Society and the Israeli Society of Radiology. Harefuah 2011;150:801e805. 17. Blank N, Gaspar T, Halon DA, Peled N, Petcherski O, Lewis BS, Rubinshtein R. Computer-assisted diagnosis of obstructive coronary atherosclerosis in patients undergoing 256-slice coronary computed tomography angiography: a comparison with invasive coronary angiography. Int J Cardiol 2014;172:130e131. 18. Halpern EJ, Halpern DJ. Diagnosis of coronary stenosis with CT angiography comparison of automated computer diagnosis with expert readings. Acad Radiol 2011;18:324e333. 19. Rubinshtein R, Halon DA, Gaspar T, Lewis BS, Peled N. Automatic assessment of coronary artery calcium score from contrast-enhanced 256-row coronary computed tomography angiography. Am J Cardiol 2013;113:7e11. 20. Rumberger J, Sheedy P, Breen JF, Schwartz R. Electron beam computed tomographic coronary calcium score cutpoints and severity of associated angiographic lumen stenosis. J Am Coll Cardiol 1997;29: 1542e1548. 21. Rubinshtein R, Gaspar T, Halon DA, Goldstein J, Peled N, Lewis BS. Prevalence and extent of obstructive coronary artery disease in patients with zero or low calcium score undergoing 64-slice cardiac multidetector computed tomography for evaluation of a chest pain syndrome. Am J Cardiol 2007;99:472e475. 22. Gottlieb I, Miller JM, Arbab-Zadeh A, Dewey M, Clouse ME, Sara L, Niinuma H, Bush DE, Paul N, Vavere AL, Texter J, Brinker J, Lima JA, Rochitte CE. The absence of coronary calcification does not exclude obstructive coronary artery disease or the need for revascularization in patients referred for conventional coronary angiography. J Am Coll Cardiol 2010;55:627e634. 23. Kang KW, Chang HJ, Shim H, Kim YJ, Choi BW, Yang WI, Shim JY, Ha J, Chung N. Feasibility of an automatic computer-assisted algorithm for the detection of significant coronary artery disease in patients presenting with acute chest pain. Eur J Radiol 2012;81:640e646. 24. Goldenberg R, Eilot D, Begelman G, Walach E, Ben-Ishai E, Peled N. Computer-aided simple triage (CAST) for coronary CT angiography (CCTA). Int J Comput Assist Radiol Surg 2012;7:819e827.