doi: 10.1161/CIRCIMAGING.108.792572
Original Articles |
Diagnostic Accuracy and Impact of Computed Tomographic Coronary Angiography on Utilization of Invasive Coronary Angiography
From the Department of Medicine (Cardiology), University of Ottawa Heart Institute (B.J.W.C., A.A., G.A.W., L.C., T.D.R., Y.Y., N.G., R.S.B.), Ottawa, Ontario, Canada; Department of Radiology, Ottawa Hospital (B.J.W.C., T.D.R., C.D., R.S.B.), Ottawa, Ontario, Canada; and Department of Cardiac Sciences, University of Calgary (P.D.G.), Calgary, Alberta, Canada.
Correspondence to Benjamin Chow, MD, FRCPC, University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, Ontario, Canada K1Y 4W7. E-mail bchow{at}ottawaheart.ca
Received May 16, 2008; accepted November 19, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Background— Computed tomographic coronary angiography (CTA), given its high negative predictive value, is a potential gatekeeper for invasive coronary angiography (ICA). Before CTA can be further accepted into clinical practice, its impact on healthcare resources needs to be better understood. We sought to determine the clinical impact of CTA on ICA referrals, CTA accuracy, and normalcy rate.
Methods and Results— To determine the impact of CTA, consecutive patients (n=7017) undergoing ICA before and after implementing a dedicated cardiac CT program were reviewed and compared with 3 other centers (n=11 508). To determine CTA accuracy, we evaluated consecutive CTA patients who underwent ICA. For normalcy rate, we identified patients with a low pretest probability for obstructive coronary artery disease. With the implementation of a cardiac CT program, the frequency of normal ICA decreased from 31.5% (1114 of 3538 patients) to 26.8% (932 of 3479 patients) (P<0.001). These findings were significantly different (P=0.003) from the 3 centers, in which normal ICAs were unchanged (30.0% [1870 of 6224 patients] to 31.0% [1642 of 5284 patients]). CTA had excellent per-patient sensitivity (99% [CI, 95% to 100%]), positive predictive value (92% [CI, 86% to 96%]) and negative predictive value (95% [CI, 72% to 100%]). Because of referral bias, specificity (64% [CI, 44% to 81%]) was low; however, the normalcy rate of CTA was 94% (CI, 90% to 97%). After adjusting for referral bias, the adjusted sensitivity was 90% (CI, 89% to 91%), and the adjusted specificity was 95% (CI, 94% to 96%), with positive and negative predictive values of 92% (CI, 91% to 93%) and 93% (CI, 92% to 94%), respectively.
Conclusion— The clinical implementation of CTA appears to positively impact ICA by reducing the frequency of normal ICA. The operating characteristics of CTA support its potential role as a tool useful in ruling out obstructive coronary artery disease.
Key Words: computed tomography • coronary angiography • clinical impact • accuracy • normalcy
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Computed tomographic coronary angiography (CTA) is a rapidly emerging diagnostic tool for the detection of coronary artery disease (CAD).1–9 However, data supporting widespread utilization of this modality are lacking.
Editorial see p 1
Clinical Perspective see p 16
Single center studies have compared the accuracy of CTA to invasive coronary angiography (ICA) and have reported very good sensitivity and negative predictive value.8–10 CTA may be an effective gatekeeper to ICA by identifying patients without significant coronary stenoses thus eliminating their need for ICA and reserving ICA for those patients who may benefit most from coronary revascularization. The clinical impact of CTA on healthcare resource utilization and health economics has not been well studied. Because cardiac CT and cardiac catheterization services are centralized to a single tertiary-care center (University of Ottawa Heart Institute [UOHI]) servicing a population of 1.5 to 1.8 million, our center is in a unique position to accurately assess the direct clinical impact of cardiac CT. We sought to understand the potential impact of CTA on referrals to ICA.
As CTA becomes accepted clinically, it is anticipated that future accuracy studies will demonstrate a decline in specificity and rise in sensitivity, as observed with other imaging modalities.11 Future analyses demonstrating declines in specificity will be attributed to verification and referral bias resulting from ICA performed primarily in patients with abnormal CTA findings.12 To account for such bias, the concept of normalcy rate (NR) has been developed and is defined as the frequency of normal studies in a population with a low pretest probability for CAD.11 Such NRs provide a more accurate estimate of the true specificity of a diagnostic test in clinical practice.11 As CTA is accepted into clinical practice, studies defining the accuracy and NR of CTA are needed to understand their potential impact in real-life clinical practice. We sought to assess the diagnostic characteristics and NR of CTA in real-life clinical practice.
The objectives of this study were to (1) understand the clinical impact of CTA on referrals to ICA and (2) determine the accuracy and NR of CTA.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
We evaluated the following 3 cohorts:
- To understand the impact of CTA on ICA, all left heart catheterizations beginning January 1, 2005, were retrospectively screened. Screening of left heart catheterizations was continued for 13 months after the implementation of the CTA program.
- To ensure that the results at the UOHI were related to establishment of cardiac CT and not because of the changes in ICA referral patterns, ICA patterns (using the same time periods) at 3 academic tertiary-care Canadian institutions (without a dedicated cardiac CT program) were screened using the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) database. To ensure that there was no overlap of patients, the 3 chosen institutions were in different provinces and were more than 1000 km away from the UOHI.
- To assess CTA accuracy, all clinical CTA studies were prospectively screened and NR was calculated using all patients prospectively enrolled in a cardiac CT registry. Indications for CTA included chest pain, dyspnea, equivocal stress test result, or heart failure.
Impact of CTA on Referrals to ICA
Between January 1, 2005 and February 28, 2007, 12 108 consecutive patients underwent left heart catheterization at the UOHI. Excluding patients referred for percutaneous coronary intervention and/or emergent angiography, a total of 7017 patients were referred for diagnostic ICA. During this time a dedicated cardiac CT was installed and operational in February 2006. Angiographic information was collected for 13 months before and 13 months after the establishment of cardiac CT. Because the waiting list for ICA was 1 month, we anticipated a 1 month delay in appreciating the clinical effects of CT. Thus, we compared 3538 patients who had ICA between January 1, 2005, and February 28, 2006 (14 months), and 3479 patients between March 1, 2006, and February 28, 2007 (12 months). During the same time period, 1029 patients underwent clinical cardiac CT at the UOHI (Figure 1).
|
Diagnostic ICA
All ICA reports were reviewed at the UOHI. When reports were inadequate, an observer blinded to the clinical data reviewed the ICA images. Results were categorized as normal if there was no significant obstructive CAD (diameter stenosis <50%) or abnormal if there was
1 coronary segment with diameter stenosis 50%.13,14 The same criteria were used to query the APPROACH database.
Computed Tomographic Coronary Angiography
Patients, without contraindications, received metoprolol or diltiazem targeting a heart rate of 
65 bpm and nitroglycerin 0.8 mg sublingually before image acquisition. A biphasic timing bolus (15 to 25 cc contrast and 40 cc saline) was used to calculate the time interval between intravenous contrast (Visipaque 320 or Omnipaque 350, GE Healthcare, Milwaukee, Wis) infusion and image acquisition. Final images were acquired with a triphasic protocol (100% contrast, 40%/60% contrast/saline, and 40 cc saline). The contrast volume and infusion rate (5 to 8 cc/s) were individualized according to scan time and patient body habitus.
Retrospective ECG-gated data sets were acquired with the GE Volume CT (GE Healthcare) with 64x0.625 mm slice collimation and a gantry rotation of 350 msec (mA=300 to 800, kV=120). Pitch (0.16 to 0.24) was individualized to the patient’s heart rate. The CT data sets were reconstructed with an increment of 0.4 mm using the cardiac phase with the least cardiac motion.
CTA Image Analysis
Images were processed using the GE Advantage Volume Share Workstation (GE Healthcare) and visually interpreted by 1 of 2 expert observers blinded to all clinical data. A 17 segment model of the coronary arteries and 4 point grading score (normal, mild [<50%], moderate [50% to 69%], severe [70%]) was used for the evaluation of coronary stenosis.15 Similar to ICA, obstructive CAD was defined as coronary diameter stenosis 50%.
CTA Accuracy
Of the 1029 consecutive cardiac CT patients, those without a history of revascularization, who underwent ICA were analyzed. The diagnostic characteristics of CTA (sensitivity, specificity, negative predictive value, and positive predictive value were calculated on a per-patient basis.8,15,16
NR Population
Between February 2006 and September 2007, consecutive patients with a low pretest probability (5.5%) for obstructive CAD, imaging heart rate 70 and BMI 40 kg/m2 were identified. Patient pretest probability for obstructive CAD was calculated using age, gender, and symptoms.13,14,17 Two patients were excluded because of repaired anomalous origin of the left coronary artery from the pulmonary artery.
Measurement of NR
Normal was defined as the absence of obstructive CAD,11,13,14 and NRs were calculated using 2 definitions for obstructive CAD (50% and 70% coronary artery diameter stenosis), as defined by Diamond-Forrester and the CASS Registry.13,14
The study was approved by the institutional human research ethics board. The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Statistical Analysis
Statistical analyses were performed (G.W. and L.C.) using SAS (version 9.1.3, SAS Institute Inc, Cary, NC), and statistical significance was defined as P<0.05. Continuous variables are presented as means and standard deviations, and categorical variables are presented as frequencies with percentages. The Wilcoxon rank sum test was used to compare continuous variables, and Fisher exact test was used for categorical variables.
Impact of CTA on ICA Referrals
To assess the impact of CTA on ICA referrals, an autoregressive integrated moving-average model with binary intervention term (0, <March 2006 and 1, March 2006) was used to compare the monthly frequency of normal ICA before and after the establishment of a dedicated cardiac CT in the time series of January 1, 2005, to February 28, 2007. The seasonal pattern was also considered in the autoregressive integrated moving-average model. The change of the frequency of normal ICA was compared between our institution and the controls at 3 APPROACH institutions using the ANOVA.
CTA Accuracy
Diagnostic test characteristics (sensitivity, specificity, negative predictive value, and positive predictive value) are reported with 95% CIs. Positive and negative likelihood ratios were weighted for the prevalence of CAD in the study cohort. Adjusted sensitivity, specificity, negative predictive value, and positive predictive value were generated to correct for referral bias using the Diamond18 method and 95% CI for the adjusted estimates were obtained using bootstrap approach.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Impact of CTA on Referrals to ICA
Over a 26-month period, a total 7017 patients had elective diagnostic ICA. Data were collected for 13 months after the implementation of cardiac CT. At the time of CT implementation, the waiting time for elective coronary angiography was 1 month. As such, it was estimated that any impact of CTA would become evident after 1 month. Thus, the patient population was divided into the 14 months before the effects of CTA (3538 patients) and the 12 months after the effects of CTA (3479 patients) (Figure 1).
The impact assessment model showed that there was a reduction in the frequency of normal angiograms (31.5% versus 26.8%; P<0.001) at the UOHI with the implementation of cardiac CT (Figure 2A). Similarly, there was a reduction in the frequency of normal angiograms performed in elective patients without prior revascularization (39.4% versus 34.7%; P=0.006) (Figure 2B). Comparing the 2 populations (before versus after CT), ICA patients after CTA implementation were older and had a greater frequency of cardiac risk factors (Table 1).
|
|
During the exact same time period, a total of 11 508 patients underwent elective ICA at 3 APPROACH institutions (Table 2). The frequency of normal angiograms was unchanged for all elective patients (30.0% [1870 of 6224] versus 31.0% [1642 of 5284]; P=0.473) and for elective patients without a history of revascularization (38.1% [1690 of 4440] versus 38.9% [1503 of 3861]; P=0.584) (Figure 2).
|
The reduction of normal angiograms at the UOHI was significantly different from the control group (all elective patients [P=0.003] and elective patients without prior revascularization [P=0.01]).
CTA Accuracy
A total of 1029 patients who underwent clinically indicated cardiac CT. Of these, 482 (46.8%) patients had stress tests performed within the preceding 3 months. CTA was requested to rule out CAD in 705 patients and/or to clarify equivocal stress test results in 251 patients.
A total of 543 (52.8%) CTAs were normal (no significant coronary stenoses), 37.7% were abnormal (1 coronary stenoses [50%]), and 9.5% had incomplete CTA evaluation or were nondiagnostic (1 nonevaluable coronary segment in the absence of coronary stenoses in evaluable segments). A total of 148 patients underwent both CTA and clinically indicated ICA. The prevalence of obstructive CAD was 81%. CTA had excellent per-patient sensitivity (99% [CI, 95% to 100%]), positive predictive value (92% [CI, 86% to 96%]) and negative predictive value (95% [CI, 72% to 100%]) for CAD (Tables 3 and 4
). As expected, the specificity was lower (64% [CI, 44% to 81%]). The positive likelihood and negative likelihood ratios (weighted for prevalence) were 11.9 (CI, 6.5 to 21.6) and 0.06 (CI, 0.01 to 0.38), respectively. After adjusting for referral bias, the adjusted sensitivity was 90% (CI, 89% to 91%), the adjusted specificity was 95% (CI, 94% to 96%) with positive and negative predictive values of 92% (CI, 91% to 93%) and 93% (CI, 92% to 94%), respectively.
|
|
NR
Of the 210 patients in the normalcy study, 9 (4.3%) were excluded from analysis for
1 unevaluable coronary segment because of patient motion (1 patient), premature ventricular complexes (2 patients), and cardiac motion (6 patients). Of the remaining 201 patients, 68 (34%) patients had evidence of coronary atherosclerosis with 10 patients having moderate stenoses and 3 with severe stenoses (Table 5). Using thresholds of 50% and 70% stenoses, the NR of CTA was 94% (CI, 90% to 97%) and 99% (CI, 96% to 100%), respectively. Two patients with severe stenoses had their CTA findings confirmed by ICA and were revascularized. The third patient, with a severe stenosis in a diagonal branch, had a corresponding myocardial perfusion defect.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Impact of CTA on Referrals to ICA
Cardiac CTA has a very high negative predictive value and thus may be effective in ruling out obstructive CAD.1–6,9 The high negative predictive value and NR in our study supports this previously reported diagnostic characteristics1–6,9 and may account for the reduction of normal ICA observed following the implementation of cardiac CT. The ability of CTA to rule out disease is extremely desirable, enabling clinicians to restrict the use of ICA to patients requiring revascularization. Because cardiac CT and cardiac catheterization services are centralized to a single tertiary care center servicing a population of 1.5 to 1.8 million, our center is in a unique position to accurately assess the direct clinical impact of cardiac CT.
These observations are further supported by the fact that our cardiac CT program had restricted access and is primarily limited to cardiologists, internists, and cardiac surgeons who refer patients to ICA. To ensure that our observations were unique to our institution, normal angiogram rates were assessed at 3 academic tertiary-care Canadian institutions without dedicated cardiac CT programs. These institutions demonstrated that the frequency of normal angiograms were unchanged during the same time periods.
To further ensure that our control population was not biased by regional practices, 3 other institutions from a different province were queried using the APPROACH database. Because of unavailable patient baseline characteristics, these data were not included in our primary analysis. However, these 3 centers reported an increase in normal angiograms from 27.8% (1290 of 4640) to 30.2% (1234 of 4088) during the same time periods.
We also observed that following the implementation of cardiac CT program, patients who subsequently were referred to ICA had greater risk factors. One may speculate that once CTA has been implemented, lower risk individuals would be excluded from ICA and thus explaining the apparent rise in cardiac risk factors.
Thus CTA is available as an alternative to ICA and its use may have facilitated better patient selection for ICA. Whether or not such patient selection improves overall population outcomes warrants further evaluation.
Accuracy of CTA
The results of this study confirm that CTA, when accepted in to clinical practice has high sensitivity, negative predictive value, and positive predictive value. Additionally, when weighted for the CAD prevalence in our study cohort, the high positive likelihood ratio and low negative likelihood ratio, suggest that CTA may be an effective diagnostic test. As expected due to referral and verification bias, the specificity of CTA fell to 64%; however, the adjusted specificity was 95%. To address this fall in specificity, the NR was also calculated.
NR
Single center studies have demonstrated that CTA has excellent operating characteristics for the identification of patients with significant CAD. However, these accuracy studies are subject to referral bias because the study patients had clinical indications for ICA. CTA has been quickly adopted into clinical practice and it is anticipated that future accuracy studies may demonstrate a decline in specificity.11 Such changes in specificity are attributed to verification and referral bias because ICA would typically be performed in patients with abnormal CTA and uncommonly in patients with nonobstructive CAD or normal CTA.12 To account for such bias, the notion of NR has been developed.
The NR of other noninvasive imaging modalities has been reported. The majority of these studies used single photon emission computed tomography and demonstrated a mean NR of 91%.11,19–23 Bach et al showed that the NR of dobutamine stress echocardiography was 94% in patients with a pretest probability 5% and 92% in patients with a pretest probability 10%.24 Sampson et al determined that the NR of positron-emission tomography/CT in 64 patients was 100%.25 Our study demonstrates that CTA has a very good NR and is similar to other noninvasive imaging modalities.
Though it is desirable to attain high NR, the NR should not be 100%. By screening large populations with a pretest probability >0% for obstructive CAD, one would expect to diagnose patients with CAD. Because our study population had a mean pretest probability of 3.1%, it was not unexpected to have made the diagnosis of obstructive CAD in some patients.
Role of CTA in the Diagnostic Pathway for Patients With Suspected CAD
The role of CTA in the diagnostic algorithm has yet to be determined; however, the diagnostic characteristics and NR are encouraging. Given the high sensitivity, negative predictive value, and NR of CTA, one may speculate that CTA is useful in ruling out CAD. Such benefit would be further supported if there was also a cost-benefit associated with CTA. Though a cost-analysis was beyond the scope of this study and requires further study, preliminary analysis at our institution suggests that 1029 CTAs may have eliminated the need for ICA in 164 patients and may have resulted in a cost-saving.
Limitations
Examining the impact of CTA at a single center could result in an overestimation of its beneficial effects, especially if CTA had a negative impact on other catheterization laboratories which was not accounted for. Though a potential limitation at other centers, coronary catheterization services are centralized to our institution which serves a population of 1.5 to 1.8 million. The negative effects of cardiac CT on distant catheterization laboratories are not an issue. Similarly, as the only cardiac CT program for the same population, the impact of CT is directly attributable to the local cardiac CT program.
We observed some variability in the baseline characteristics between the UOHI and the 3 control centers but attribute them to possible differences in the definitions used to identify risk factors. Because the population across Canada is relatively homogeneous, one would have expected similar changes in baseline characteristics across all sites. The increased prevalence of cardiac risk factors at the UOHI after the implementation of CT, suggests that a lower risk population was appropriately screened and the need for ICA eliminated.
It is well accepted that the NPV of CTA is very high and we would anticipate that a similar effect would be expected at other centers. However, we recognize that interobserver variability and patient referral patterns may limit these effects.
Our study did not evaluate upstream or downstream utilization of other noninvasive modalities after CTA nor did we compare CTA to other noninvasive methods. Other ongoing studies such as study of Perfusion Versus Anatomy’s Role in CAD (SPARC) are considering these questions.26 However, these studies will not compare cohorts before and after the availability of CTA as has been done in the current study.
Our CTA accuracy data are limited by referral and verification bias and therefore we reported the NR of CTA which is believed to be an indirect measure of specificity after a test has been adopted into clinical practice. Though some have advocated the use of normal volunteers to determine NR, the radiation exposure associated with CTA does not permit the screening of normal volunteers. The American College of Cardiology Practice Guidelines for Radionuclide Imaging has indicated that patients with a low pretest probability should be preferentially chosen because they reflect the clinical population.11
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
The implementation of a cardiac CT program reduced the rate of normal angiography compared with controls in this study population. The ability of CTA to rule out disease may have a positive impact on clinical practice, specifically the patient selection for ICA, and may be accompanied by a potential cost savings. Further studies examining the cost-effectiveness of cardiac CT are required. The diagnostic characteristics and NR of CTA suggest that it may be a clinically useful diagnostic tool.
|
Acknowledgments |
|---|
The authors extend their gratitude to Christine Bohan, Joanne Bussell, Kathryn Callidine, Stephanie Colpitts, Micheala Garkisch, Debbie Gauthier, Patricia Grant, Sarah Harkness, Sandina Jamieson, Matt Raegele, Richard Tessier, and the APPROACH investigators for their expertise and dedication to cardiac CT research.
Sources of Funding
Dr Chow is supported by the Canadian Institute of Health Research New Investigator Award No. MSH-83718. Dr Abraham is supported by the Heather and Whit Tucker Research Fellowship in Cardiology.
Disclosures
Dr Chow receives research and fellowship support from GE Healthcare. None of the remaining authors have conflicts of interest to disclose.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
1. Hoffmann MH, Shi H, Schmitz BL, Schmid FT, Lieberknecht M, Schulze R, Ludwig B, Kroschel U, Jahnke N, Haerer W, Brambs HJ, Aschoff AJ. Noninvasive coronary angiography with multislice computed tomography. JAMA. 2005; 293: 2471–2478.
2. Hoffmann U, Nagurney JT, Moselewski F, Pena A, Ferencik M, Chae CU, Cury RC, Butler J, Abbara S, Brown DF, Manini A, Nichols JH, Achenbach S, Brady TJ. Coronary multidetector computed tomography in the assessment of patients with acute chest pain. Circulation. 2006; 114: 2251–2260.
3. Ropers D, Rixe J, Anders K, Kuttner A, Baum U, Bautz W, Daniel WG, Achenbach S. Usefulness of multidetector row spiral computed tomography with 64-x0.6-mm collimation and 330-ms rotation for the noninvasive detection of significant coronary artery stenoses. Am J Cardiol. 2006; 97: 343–348.[CrossRef][Medline]
4. Leber AW, Knez A, von Ziegler F, Becker A, Nikolaou K, Paul S, Wintersperger B, Reiser M, Becker CR, Steinbeck G, Boekstegers P. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol. 2005; 46: 147–154.
5. Raff GL, Gallagher MJ, O'Neill WW, Goldstein JA. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol. 2005; 46: 552–557.
6. Mollet NR, Cademartiri F, van Mieghem CAG, Runza G, McFadden EP, Baks T, Serruys PW, Krestin GP, de Feyter PJ. High-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography. Circulation. 2005; 112: 2318–2323.
7. Chow BJ, Hoffmann U, Nieman K. Computed tomographic coronary angiography: an alternative to invasive coronary angiography. Review article. Can J Cardiol. 2005; 21: 933–940.[Medline]
8. Beanlands RSB, Chow BJW, Dick A, Friedrich MG, Gulenchyn KY, Kiess M, Leong-Poi H, Miller RM, Nichol G. CCS/CAR/CANM/CNCS/CanSCMR joint position statement on advanced noninvasive cardiac imaging using positron emission tomography, magnetic resonance imaging and multidetector computed tomographic angiography in the diagnosis and evaluation of ischemic heart disease — executive summary. Can J Cardiol. 2007; 23: 107–119.[Medline]
9. Schuijf JD, Bax JJ, Shaw LJ, de Roos A, Lamb HJ, van der Wall EE, Wijns W. Meta-analysis of comparative diagnostic performance of magnetic resonance imaging and multislice computed tomography for noninvasive coronary angiography. Am Heart J. 2006; 151: 404–411.[CrossRef][Medline]
10. Di Carli MF, Hachamovitch R. New technology for noninvasive evaluation of coronary artery disease. Circulation. 2007; 115: 1464–1480.
11. Klocke FJ, Baird MG, Lorell BH, Bateman TM, Messer JV, Berman DS, O'Gara PT, Carabello BA. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging-executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines (ACC/AHA/ASNC Committee to revise the 1995 guidelines for the clinical use of cardiac radionuclide imaging). J Am Coll Cardiol. 2003; 42: 1318–1333.
12. Rozanski A, Diamond GA, Berman D, Forrester JS, Morris D, Swan HJ. The declining specificity of exercise radionuclide ventriculography. N Engl J Med. 1983; 309: 518–522.[Abstract]
13. Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis of coronary-artery disease. N Engl J Med. 1979; 300: 1350–1358.[Abstract]
14. Chaitman BR, Bourassa MG, Davis K, Rogers WJ, Tyras DH, Berger R, Kennedy JW, Fisher L, Judkins MP, Mock MB, Killip T. Angiographic prevalence of high-risk coronary artery disease in patient subsets (CASS). Circulation. 1981; 64: 360–367.
15. Hoffmann U, Moselewski F, Cury RC, Ferencik M, Jang IK, Diaz LJ, Abbara S, Brady TJ, Achenbach S. Predictive value of 16-slice multidetector spiral computed tomography to detect significant obstructive coronary artery disease in patients at high risk for coronary artery disease: patient-versus segment-based analysis. Circulation. 2004; 110: 2638–2643.
16. Chow BJ, Dennie C, Hoffmann U, So D, de Kemp RA, Ruddy TD, Beanlands RS. Comparison of computed tomographic angiography versus rubidium-82 positron emission tomography for the detection of patients with anatomical coronary artery disease. Can J Cardiol. 2007; 23: 801–807.[Medline]
17. Gibbons RJ, Chatterjee K, Daley J, Douglas JS, Fihn SD, Gardin JM, Grunwald MA, Levy D, Lytle BW, O'Rourke RA, Schafer WP, Williams SV, Ritchie JL, Gibbons RJ, Cheitlin MD, Eagle KA, Gardner TJ, Garson A Jr, Russell RO, Ryan TJ, Smith SC Jr. ACC/AHA/ACP-ASIM guidelines for the management of patients with chronic stable angina: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With Chronic Stable Angina). J Am Coll Cardiol. 1999; 33: 2092–2197.
18. Diamond GA. Reverend Bayes’ silent majority; an alternative factor affecting sensitivity and specificity of exercise electrocardiography. Am J Cardiol. 1986; 57: 1175–1180.[CrossRef][Medline]
19. Azzarelli S, Galassi AR, Foti R, Mammana C, Musumeci S, Giuffrida G, Tamburino C. Accuracy of 99mTc-tetrofosmin myocardial tomography in the evaluation of coronary artery disease. J Nucl Cardiol. 1999; 6: 183–189.[CrossRef][Medline]
20. Zaret BL, Rigo P, Wackers FJT, Hendel RC, Braat SH, Iskandrian AS, Sridhara BS, Jain D, Itti R, Serafini AN, Goris ML, Lahiri A. Myocardial perfusion imaging with 99mTc tetrofosmin: comparison to 201Tl imaging and coronary angiography in a phase III multicenter trial. Circulation. 1995; 91: 313–319.
21. Iskandrian AS, Heo J, Kong B, Lyons E, Marsch S. Use of technetium-99m isonitrile (RP-30A) in assessing left ventricular perfusion and function at rest and during exercise in coronary artery disease, and comparison with coronary arteriography and exercise thallium-201 SPECT imaging. Am J Cardiol. 1989; 64: 270–275.[CrossRef][Medline]
22. Berman DS, Kiat H, Friedman JD, Wang FP, van Train K, Matzer L, Maddahi J, Germano G. Separate acquisition rest thallium-201/stress technetium-99m sestamibi dual-isotope myocardial perfusion single-photon emission computed tomography: a clinical validation study. J Am Coll Cardiol. 1993; 22: 1455–1464.[Abstract]
23. Amanullah AM, Kiat H, Friedman JD, Berman DS. Adenosine technetium-99m sestamibi myocardial perfusion SPECT in women: diagnostic efficacy in detection of coronary artery disease. J Am Coll Cardiol. 1996; 27: 803–809.[Abstract]
24. Bach DS, Hepner A, Marcovitz PA, Armstrong WF. Dobutamine stress echocardiography: prevalence of a nonischemic response in a low-risk population. Am Heart J. 1993; 125: 1257–1261.[CrossRef][Medline]
25. Sampson UK, Dorbala S, Limaye A, Kwong R, Di Carli MF. Diagnostic accuracy of rubidium-82 myocardial perfusion imaging with hybrid positron emission tomography/computed tomography in the detection of coronary artery disease. J Am Coll Cardiol. 2007; 49: 1052–1058.
26. Di Carli, Marcelo F. CT coronary angiography: where does it fit? J Nucl Med. 2006; 47: 1397–1399.
Related Articles
CLINICAL PERSPECTIVE
Currently, invasive coronary angiography (ICA) is the gold standard for the diagnosis of obstructive coronary artery disease (CAD). However, ICA is best reserved for patients who benefit most from coronary revascularization. Identifying a noninvasive modality that can accurately diagnose obstructive CAD is desirable. Computed tomographic coronary angiography (CTA) is an emerging modality that has a high negative predictive value and, given its ability to rule out CAD, may have a role as a gatekeeper for ICA. To better understand the impact of cardiac CT on clinical practice, a study was undertaken to determine the diagnostic accuracy and clinical impact of CT coronary angiography on the utilization of ICA. CTA had excellent per-patient sensitivity (99%), positive predictive value (92%), and negative predictive value (95%) for obstructive CAD. As expected (due to referral and verification bias), the specificity (64%) was lower than that of previously published studies; however, the normalcy rate of CTA was 94%. After adjusting for referral bias, the adjusted sensitivity was 90%, and the adjusted specificity was 95%. With the implementation of a cardiac CT program, the frequency of normal ICA decreased significantly from 31.5% to 26.8% which was significantly different from centers without dedicated cardiac CT (30.0% to 31.0%) during the same time periods. The results of this study confirm the potential utility of CTA for the diagnosis of obstructive CAD and its potential for improving utilization of ICA by reducing the frequency of patients with nonobstructive CAD requiring ICA.
Circ Cardiovasc Imaging 2009 2: 1-3.
Circ Cardiovasc Imaging 2009 2: 16-23.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||