Article Text

Download PDFPDF

Coronary CTA works for preoperative risk stratification, but do we know when and how to use it?
  1. Márcio Sommer Bittencourt1,2,3,
  2. Danielle Menosi Gualandro4,5
  1. 1 Division of Internal Medicine, University Hospital, University of Sao Paulo, Sao Paulo, Brazil
  2. 2 Hospital Israelita Albert Einstein, Sao Paulo, Brazil
  3. 3 Diagnósticos da América (DASA), Sao Paulo, Brazil
  4. 4 Cardiology, Heart Institute (InCor), University of Sao Paulo Medical School, Sao Paulo, Brazil
  5. 5 Cardiovascular Research Institute Basel (CRIB)Department of Cardiology, University Hospital, Basel, Switzerland
  1. Correspondence to Dr Márcio Sommer Bittencourt, Division of Internal Medicine, University Hospital, University of Sao Paulo, Sao Paulo 05412003, Brazil; msbittencourt{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

More than 300 million surgeries are performed annually worldwide. Despite improvements in surgical techniques and anaesthesia, mortality related to these procedures is still higher than expected. Since cardiovascular complications are important contributors to postoperative morbidity and mortality, it is fundamental to identify patients at increased cardiovascular risk, not only to take measures to decrease risk but also to access the risk–benefit ratio of the surgery. Although the core of preoperative risk stratification are clinical risk scores, such as the Revised Cardiac Risk Index, they are known to have limited accuracy and additional tools to improve risk assessment may be needed for part of the population. Among the various cardiovascular complications commonly seen in the perioperative period, myocardial infarction/injury (PMI) are important contributors to mortality. Although in other clinical scenarios most of those episodes are mostly related to coronary plaque rupture or other acute atherosclerotic plaque instability defining a type 1 myocardial infarction (MI), in the perioperative period a significant proportion of events are also caused by oxygen supply demand mismatch, known as type 2 MI, though this is also most common in individuals with prior coronary artery disease (CAD).1 2

Despite the need to identify higher risk individuals, recommendations for routine preoperative testing have substantially decreased in the last two decades.3 This was motivated by the negative results of several randomised trials evaluating preoperative ‘prophylactic’ myocardial revascularisation, that failed to show benefit,4 5 as well as by the lack of evidence that performing preoperative testing reduces postoperative cardiac complications. In fact, unnecessary preoperative testing could result in harm by delaying surgery and compromising the prognosis of the disease leading to the indication for surgery. Yet, despite those restrictions, higher risk individuals, such as those with severe CAD and ischaemia, may still benefit from additional risk stratification and selective use of preventative interventions prior to surgery to reduce the risk of adverse outcomes, though in other cases identification of excessive risk may also be useful to discuss the expected risk benefit from the surgical procedure with the patient.

According to current guidelines, most patients can have an adequate risk assessment solely based on clinical presentation and risk factors. However, for patients with more than two clinical risk factors (ischaemic heart disease, stroke or transient ischaemic attack, creatinine >2 mg/dL, diabetes mellitus requiring insulin) and poor functional capacity (<4 metabolic equivalent of task (MET)) who have a high-risk surgery planned, current guidelines recommend functional stress testing associated with echocardiography or myocardial perfusion scintigraphy (MPS) imaging. Those recommendations are based on two findings derived from observational studies. First, the rate of mortality and PMI are directly associated with the presence and extent of the perfusion defects in MPS.6 7 Second, a negative stress echocardiography has a high negative predictive value (NPV) for postoperative cardiac events. Due to the limited prior evidence, guidelines do not include coronary CT angiography (CTA) as an alternative test to for the preoperative cardiac risk stratification due to the limited evidence to support its use at the time of publication of those documents.3 8

In their Heart paper, Koshy et al 9 report an interesting systematic review and meta-analysis evaluating the role of CT coronary angiography (CTA) and coronary calcium scoring (CAC) in risk for prediction of major adverse cardiovascular events (MACE). Although neither is currently recommended by guidelines, the authors identified 11 studies, with >3000 patients who underwent either one or both CT-derived imaging modalities in the preoperative risk stratification of individuals undergoing intermediate or high-risk non-cardiac surgery. Unfortunately, due to limitation on data reporting, only four studies reporting coronary CTA and five studies reporting CAC results could be incorporated in the quantitative meta-analysis. Additionally, the included studies were not homogeneous, particularly due to differences in the type of surgery included. Despite those limitations, the study was able to report interesting findings likely to impact the clinical use of those methods in the preoperative setting.

First, the presence, extent and severity of coronary atherosclerosis was directly associated with the risk of cardiovascular events in the perioperative period. While individuals with no coronary atherosclerosis had a 2% rate of MACE, this increased to 4% in those with non-obstructive disease. In individuals with obstructive disease, the risk of events was 7% for those with single vessel disease, and as high as 23% for those with multivessel disease. Although no point estimates are described for CAC, a similar trend is also reported. This incremental risk associated with the CAD atherosclerotic burden is analogous to the reported risk stratification information provided by the previously described stress imaging modalities.

Second, the authors focused on the NPV, or the probability of no MACE given a negative test result. In this analysis, the authors focused only on high clinical risk individuals, and a ‘negative’ result was considered absence of multivessel disease. Even with this rather soft definition of negative result in a higher risk cohort, the NPV of the coronary CTA was impressively high, at 96%. Although no data were provided for CAC, for individuals with low or moderate CAC (<100 or even 300 AU) would likely result in comparable NPV, though the prevalence of ‘abnormal’ results would likely be higher.

Collectively, those two key findings suggest that coronary CTA performance for risk stratification and identification of low-risk individuals is comparable to the findings reported form ischaemia driven risk stratification using stress echocardiography or MPS. Thus, coronary CTA can be considered a safe and effective alternative for those methods in individuals deemed to be candidates for additional risk stratification prior to higher risk surgeries.

Despite those encouraging findings supporting the use of coronary CTA in the preoperative setting, some questions still remain unanswered. First, the vast majority of patients in whom additional risk stratification is recommended includes older individuals with several risk factors. In those individuals a non-normal, non-high-risk anatomy coronary CTA results is expected. In those individuals with non-obstructive CAD, moderate obstructive lesions (50%–70% stenosis) or with single vessel obstructive disease the risk of MACE is clearly increased, but the benefit of any medical or interventional treatment is undefined. Additionally, such results may lead to downstream testing with may negatively impact cost and prognosis, not only due to the inherent risk of additional testing but also due to the potential delay in a needed surgical intervention.

Second, even the benefit of additional interventions required for individuals with multivessel obstructive disease remains to be tested. Should those individuals receive medical therapy such as aspirin or statin prior to surgery? Statins seem reasonable in all patients with documented CAD which may include asymptomatic individuals with more extensive or obstructive atherosclerotic plaques on coronary CTA as we have enough evidence that it decreases the risk of perioperative complications, without serious side effects. Aspirin may pose a more complex issue, not only due to recent evidence questioning its benefit in the primary prevention setting, but particularly in the perioperative period as the risk associated with bleeding needs to be carefully accessed. Although no clear definition of coronary atherosclerosis extent or severity may lead to aspirin use, it is clearly beneficial to withhold its use when no or only minimal atherosclerosis is detected in the preoperative period. Finally, it remains unknown if any subgroup of individuals in whom anatomy poses such a high risk that a percutaneous coronary intervention or a coronary artery bypass surgery can be recommended prior to the scheduled surgical procedure.

Priori evidence already suggests that coronary CTA might be the most effective initial test for the investigation of stable chest pain,10 particularly because it leads to a higher increase in the use of preventative therapies, such as aspirin and statins, than the use of stress testing.11 With the current results one can securely say coronary CTA can also be a safe and effective initial test for the risk stratification in the preoperative setting, with expected similar benefits in the longer term. However, how its findings can be incorporated in the routing preoperative risk stratification in order to improve outcomes at a reasonable cost and limited risk of adverse consequences, including delay in surgical procedures, need to be evaluated in future studies.



  • Contributors Both authors worked on the conceptualisation, writing, drafting, editing and reviewed the final version of the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

  • Patient consent for publication Not required.

Linked Articles