Heart International 2013; volume 8:e1 Potential role of coronary computed tomography-angiography for guiding perioperative cardiac management for non-cardiac surgery Amit K. Pahwa,1 Armin Arbab-Zadeh,2 Daniel J. Brotman,1 Leonard S. Feldman1 1Hospitalist Program, Department of Medicine; 2Division of Cardiology, Department of Medicine, The Johns Hopkins University, Baltimore, USA increasing patient morbidity, mortality, and studies with 8119 patients to determine the Abstract length of stay.1 According to the American operating characteristics (sensitivity and College of Cardiology Foundation/American specificity) of a variety of non-invasive stress Perioperative cardiac events can be a major Heart Association (ACCF/AHA) 2007 tests performed before surgery. When evalu- consequence of surgery. The American College Guidelines on Perioperative Cardiovascular ating for the likelihood of perioperative car- of Cardiology Foundation/American Heart Evaluation, non-invasive cardiac stress test- diac death and non-fatal myocardial infarc- Association has set out guidelines to aid ing should be considered for patients under- tion in vascular surgery patients, dobutamine physicians in identifying patients at the high- going moderate or high-risk surgery with stress echocardiography had the lowest est risk for these events. The guidelines do poor functional capacity and at least one clin- (most favorable) negative likelihood ratio recommend for some patients to undergo non- ical risk factor. The intent of the testing is to (LR), 0.21, while radionuclide ventriculogra- invasive cardiac stress testing for further risk further refine risk assessment for periopera- phy had the highest positive LR, 5.56.5 stratification, but their sensitivity and speci- tive cardiac events and to determine the need Beattie et al. in their meta-analysis of 68 ficity for predicting cardiac events is not opti- for interventions prior to surgery aimed at studies of 10,049 patients compared thallium mal. With more data emerging of the superior reducing such a risk. The presence of imaging to stress echocardiography and performance of computed coronary tomogra- obstructive coronary artery disease increases determined that stress echocardiography had phy angiography (CCTA) compared to non- the risk of perioperative cardiac events. a better positive LR (4.09 vs.1.83) and nega- invasive stress testing, CCTA could be more Although the obstructive lesion itself is often tive LR (0.23 vs.0.44) for predicting a postop- useful in risk stratification for these patients. not the cause of the cardiac event, the lesion erative cardiac event.6Based on these num- is a marker for advanced coronary atheroscle- bers, a patient with a moderate perioperative rotic plaque burden and, among them, unsta- cardiac event risk of 6% would have a post- ble plaques that are thought to lead to many test probability of 21% with a positive preop- Introduction of these events.2 In addition to plaque type, erative stress echocardiography study and a other factors including alterations in coro- risk of 1% with a negative study. Major perioperative cardiac events occur in nary blood flow, changes in hemostasis, neu- More recently, computed coronary tomogra- approximately 1% of non-cardiac surgeries, rohormonal dysregulation, or other environ- phy angiography (CCTA) has emerged as a mental stressors are typically required to more sensitive test to non-invasively evalu- incite an acute coronary event.3Surgery pro- ate cardiac risk through imaging of the coro- duces inflammation, catecholamine release, nary arteries. Although there has only been and a hypercoagulable state which can pre- limited experience with this modality in the Correspondence: Amit K. Pahwa, Nelson 219, 600 cipitate plaque rupture, thrombosis, and perioperative setting, CCTA is able to accu- North Wolfe Street, Baltimore, MD 21287, USA. myocardial infarction (MI).1 Dawood et al. rately detect and exclude significant coronary E-mail: [email protected] showed, through autopsy review, that the artery disease (CAD) in many other clinical Key words: computed coronary tomography pathophysiology of a fatal perioperative settings. A systematic review of 41 studies angiography, myocardial perfusion imaging, car- myocardial infarction is very similar to non- and 2500 patients with stable chest pain com- diac event. operative events. They compared the paring CCTA to the gold standard diagnostic histopathology of myocardial infarction in test, coronary angiography, found CCTA to Conflict of interests: the authors declare no patients who underwent surgery as well as have a sensitivity of 99% and specificity of potential conflicts of interests. those who did not. Plaque hemorrhage/rup- 89% for detecting flow-limiting CADdefined ture occurred in 55% of the perioperative as greater than 50% stenosis by quantitative Received for publication: 1 September 2012. group versus 40% of the non-operative analysis. This translates to a positive likeli- Accepted for publication: 28 November 2012. patients (P=0.32), while thrombus without hood ratio of 9 and negative likelihood ratio This work is licensed under a Creative Commons plaque disruption was noted in 29% versus of 0.01.7 Attribution NonCommercial 3.0 License (CC BY- 36% (P=0.59) of the perioperative versus NC 3.0). non-operative patients, respectively.4 A review of the literature reveals only fair ©Copyright A.K. Pahwa et al., 2013 Outcome: obstructive disease sensitivity and specificity of current non- Licensee PAGEPress, Italy invasive cardiac stress modalities to predict Heart International 2013; 8:e1 doi:10.4081/hi.2013.e1 cardiac events in perioperative patients. In a recent review, pooled results from 7 stud- Kertai et al.performed a meta-analysis of 58 ies and 483 patients showed a discrepancy [Heart International 2013; 8:e1] [page 1] Review between SPECT and CCTA for detecting more to five years after testing.16Such negative pre- (while acknowledging uncertainty given the than 50% coronary artery stenosis. Using inva- dictive value is unmatched by other non-inva- current body of evidence) using CCTA for pre- sive coronary angiography as the reference sive tests. On the other hand, the specificity of operative evaluation on patients with less standard, SPECT was 66% sensitive and 69% only 41% produces a low positive LR (1.70) than 4 METS of functional capacity and at specific compared to the 96% sensitivity and which means that a positive CCTA, using simply least one clinical risk factor undergoing inter- 88% specificity of CCTA.8 the presence of obstructive coronary artery dis- mediate risk or vascular surgery. Given the A few studies have explored the use of CCTA in ease, is a poor predictor of an impending major CCTA data discussed above, it may be reason- patients undergoing non-coronary cardiac sur- cardiac event.15 To further investigate the nega- able to substitute CCTA for the other stress gery.9-13 In the largest cohort of 70 patients tive predictive value of CCTA in low- to interme- testing modalities, such as MPI. A recent preparing for valvular surgery, CCTA was 100% diate-risk patients (thrombolysis in myocardial study by Goldstein et al.looked at efficiency, sensitive and 92% specific for detecting more infarction, TIMI 0-2) presenting to the ED with cost, and safety of CCTA versus MPI in than 50% stenosis in the coronary arteries when chest pain, Litt et al.randomized patients to a patients with no coronary artery disease pre- compared to invasive coronary angiography.9 CCTA or usual carepathway. Of those studied in senting to the emergency department with the CCTA pathway, 640 had a negative CCTA acute chest pain and TIMI risk score below 4. (less than 50% stenosis), and none of them died There was no significant difference in num- or had an MI within 30 days after presentation.17 ber of major adverse cardiac events between Outcome: cardiac events In addition, Hoffman et al. performed a multi- the two groups. The costs of both tests are center study randomizing 1000 patients present- very similar, while the exposure to radiation ing to the ED with chest pain with any TIMI is significantly lower with CCTA.20 The prognostic value of CCTA has been well score to either standard ED evaluation or CCTA studied in emergency department (ED) litera- as first evaluation. Patients were followed for 28 ture. Gallagher et al. enrolled 85 low-risk days after discharge from the hospital. In the patients who presented to the ED with chest Conclusions usual carepathway, there were 6 cases of acute pain. All had negative electrocardiogram coronary syndrome and only 2 in the CCTA (EKG) and serum markers for myocardial group. Of note, the CCTA was positive for clini- Based on available data, CCTA is effective at ischemia. All patients underwent myocardial cally significant disease in both of these excluding the presence of flow-limiting CAD perfusion imaging (MPI) and CCTA. MPI was patients, but they both had negative stress tests and in identifying those patients with chest considered abnormal if it showed reversible and were managed medically. In addition, there pain who are unlikely to have near-term coro- perfusion defects, and CCTA was deemed was no significant difference in number of nary events. Absence of flow-limiting lesions abnormal if there was greater than 50% steno- catherizations, percutaneous intervention (PCI) appears to be a surrogate marker for low coro- sis. Patients were followed for 30 days to deter- or coronary artery bypass surgery (CABG) nary atherosclerotic plaque burden and mine whether the patient had an acute coro- between each group.18 unstable coronary plaques. Given that stress nary event such as unstable angina, non-ST echocardiography and MPI performance is, at elevation myocardial infarction (NSTEMI) or best, only fair in risk-stratifying preoperative ST elevation myocardial infarction STEMI (by patients, and in other settings is inferior to record review/questionnaire or invasive Can we use computed coronary CCTA, we suggest considering CCTA as an angiography). No events occurred in 78 tomography angiography in the appropriate, and perhaps better way to strati- patients while a major cardiac event occurred fy perioperative patients at risk for CAD if in 7 patients. In those who did not have an perioperative setting? testing is indeed indicated (Table 1). CCTA is event, 70 of 78 had a negative MPI and 72 of 78 both sensitive and specific for detecting had a negative CCTA. Of the 7 who had an The ACCF released appropriate usecriteria for angiographically apparent obstructive CAD. event, MPI and CCTA predicted 5 and 6 of CCTA most recently in 2010. The foundation Although there are no studies on the periop- them, respectively. These data suggest that rates CCTA as appropriate to detect CAD in erative patient, the data on CCTA in other set- CCTA is just as good, if not better than MPI in intermediate risk (10-90% CAD risk) tings suggest it should be considered as a rea- identifying those who will not have a major patients. This refers to patients presenting sonable and possibly superior substitute for cardiac event.14 with non-acute ischemic symptoms despite other non-invasive modalities. A randomized A 2011 systematic review and meta-analysis having an interpretable EKG or ability to exer- control trial comparing CCTA to other modali- included 18 studies evaluating major cardiac cise.19 In addition, the ACCF recommends ties in the preoperative setting is warranted. events in 9592 symptomatic patients after evalu- ation with CCTA. Each patient had known or sus- pected coronary artery disease. Median follow- up time was 20 months. Major cardiac events were defined as death, myocardial infarction, or need for coronary revascularization. A CCTA was considered positive if a greater than 50% steno- sis was identified. CCTA demonstrated high sen- sitivity (99%) and low negative LR (0.08) for a Table 1. Likelihood ratios for various risk-stratification modalities. major cardiac event. Patients with a negative Preoperative cardiac event +LR -LR test had a much lower post-test probability of Dobutamine stress echocardiogram 2.83 0.21 having an event.15 Importantly, most of these Radionuclide ventriculography 5.56 0.55 studies included total mortality, but not cardiac mortality, as end points. If only cardiac mortality Total mortality +LR -LR or myocardial infarction are considered, the CCTA 1.7 0.08 event rate after a normal CCTA was zero even up LR, likelihood ratio; CCTA, computed coronary tomography angiography. [page 2] [Heart International 2013; 8:e1] Review as an alternative to invasive coronary patients. Ann Emerg Med 2007;49:125-36. References angiography in the investigation of coro- 15. Hulten EA, Carbonaro S, Petrillo SP, et al. nary artery disease. Health Technol Assess Prognostic value of cardiac computed 1. Devereaux PJ, Goldman L, Cook DJ, et al. 2008;12:iii-iv,ix-143. tomography angiography a systematic Perioperative cardiac events in patients 8. Arbab-Zadeh A. Stress testing and non- review and meta-analysis. J Am Coll undergoing noncardiac surgery: A review invasive coronary angiography in patients Cardiol 2011;57:1237-47. of the magnitude of the problem, the with suspected coronary artery disease: 16. Andreini D, Pontone G, Mushtaq S, et al. A pathophysiology of the events and meth- time for a new paradigm. Heart Long-Term Prognostic Value of Coronary ods to estimate and communicate risk. International 2012;7:e2. CT Angiography in Suspected Coronary CMAJ 2005;173:627-34. 9. Meijboom WB, Mollet NR, Van Mieghem Artery Disease. JACC Cardiovascular 2. Fleisher LA, Beckman JA, Brown KA, et al. CA, et al. Pre-operative computed tomogra- Imaging 2012;5:690-701. ACC/AHA 2007 guidelines on perioperative phy coronary angiography to detect signif- 17. Litt HI, Gatsonis C, Snyder B, et al. CT cardiovascular evaluation and care for icant coronary artery disease in patients angiography for safe discharge of patients noncardiac surgery: A report of the referred for cardiac valve surgery. J Am with possible acute coronary syndromes. N American College of Cardiology/American Coll Cardiol 2006;48:1658. Engl J Med 2012;366:1393-403. Heart Association Task Force on Practice 10. Gilard M, Cornily JC, Pennec PY, et al. Guidelines (Writing Committee to Revise Accuracy of multislice computed tomogra- 18. Hoffmann U, Truong QA, Schoenfeld DA, et the 2002 Guidelines on Perioperative phy in the preoperative assessment of al; ROMICAT-II Investigators. Coronary CT Cardiovascular Evaluation for Noncardiac coronary disease in patients with aortic angiography versus standard evaluation in Surgery). J Am Coll Cardiol 1007;50:e159- valve stenosis J Am Coll Cardiol 2006;47: acute chest pain. N Engl J Med 2012;367: 241. 2020-4. 299-308. 3. Arbab-Zadeh A, Nakano M, Virmani R, 11. Manghat NE, Morgan-Hughes GJ, Broadley 19. Taylor AJ, Cequeira M, Hodgson JM, et al. Fuster V. Acute coronary events. AJ, et al. 16-detector row computed tomo- ACCF/SCCT/ACR/AHA/ASE/ASNC/SCAI/SC Circulation 2012;125:1147-56. graphic coronary angiography in patients MR 2010 appropriate use criteria for car- 4. Dawood MM, Gutpa DK, Southern J, et al. undergoing evaluation for aortic valve diac computed tomography: a report of the Pathology of fatal perioperative myocardial replacement: comparison with catheter American College of Cardiology infarction: implications regarding patho- angiography Clin Radiol 2006;61:749-57. Foundation Appropriate Use Criteria Task physiology and prevention. Int J Cardiol 12. Reant P, Brunot S, Lafitte S, et al. Force, the Society of Cardiovascular 1996;57:37-44. Predictive value of noninvasive coronary Computed Tomography, the American 5. Kertai MD, Boersma E, Bax JJ, et al. A angiography with multidetector computed College of Radiology, the American Heart meta-analysis comparing the prognostic tomography to detect significant coronary Association, the American Society of accuracy of six diagnostic tests for predict- stenosis before valve surgery. Am J Cardiol Echocardiography, the American Society of ing perioperative cardiac risk in patients 2006;97:1506-10. Nuclear Cardiology, the Society for undergoing major vascular surgery. Heart 13. Scheffel H, Leschka S, Plass A, et al. Cardiovascular Angiography and 2003;89:1327-34. Accuracy of 64-slice computed tomography Interventions, and the Society for 6. Beattie WS, Abdelnaem E, Wijeysundera for the preoperative detection of coronary Cardiovascular Magnetic Resonance. J Am DN, Buckley DN. A meta-analytic compari- artery disease in patients with chronic son of preoperative stress echocardiogra- aortic regurgitation Am J Cardiol 2007; Coll Cardiol 2010;56:1864-94. phy and nuclear scintigraphy imaging. 100:701-6. 20. Goldstein JA, Chinnaiyan KM, Abidov A, et Anesth Analg 2006;102:8-16. 14. Gallagher MJ, Ross MA, Raff GL, et al. The al. The CT-STAT (Coronary Computed 7. Mowatt G, Cummins E, Waugh N, et al. diagnostic accuracy of 64-slice computed Tomographic Angiography for Systematic Systematic review of the clinical effective- tomography coronary angiography com- Triage of Acute Chest Pain Patients to ness and cost-effectiveness of 64-slice or pared with stress nuclear imaging in Treatment) Trial. J Am Coll Cardiol 2011; higher computed tomography angiography emergency department low-risk chest pain 58:1414-22. [Heart International 2013; 8:e1] [page 3]