Over the past decade, expenditure on cardiovascular imaging in general, and echocardiography in particular, have increased significantly,1 which has raised concerns about the sustainability of this growth and potential overuse or misuse of imaging tests. In order to improve clinical practice, reduce unnecessary tests and enhance overall cost-effectiveness, the American College of Cardiology in partnership with the American Society of Echocardiography and other subspecialty societies developed appropriate use criteria (AUC) for transthoracic echocardiography (TTE). This document, first published in 20072 and updated in 2011,3 contains recommendations for the rational use of TTE, rating the grade of appropriateness of various clinical indications. Since then, there have been studies of the appropriateness of clinical requests for TTE in different settings in the USA4–8 and Europe.9,10

The aim of this study was to assess whether TTE requests comply with the 2011 AUC at a Portuguese tertiary care center. In addition, we aimed to identify the factors associated with lower adherence to the AUC.

We found that, according to the 2011 AUC document, 78.7% of TTE requests were appropriate, 15.3% were inappropriate and 6.0% were of uncertain appropriateness. These results are similar to those reported by other studies, in which appropriateness rates range from 71.0% to 96.5%.5,6,9,10 Several factors may influence these results, such as the setting in which the study is carried out (university vs. non-university center, tertiary vs. non-tertiary center, in- vs. outpatients), the characteristics of the study population and the specialty of the requesting physician.

Given that in 2013 about 10 000 TTE exams were performed at our echocardiography laboratory (echo lab), the sample included in this study probably reflects the usual pattern of TTE requests at our center. A significant proportion of appropriate exams were performed to evaluate symptoms or conditions potentially related to cardiac etiology, in particular stroke or TIA. Other common appropriate indications were related to initial evaluation of patients with known or suspected heart failure, hypertensive heart disease or valvular or structural heart disease. A few scenarios account for the majority of inappropriate studies. The most significant was routine surveillance of ventricular function in patients with known CAD and no change in clinical status or cardiac exam, which is also reported in other studies as a frequent inappropriate indication.9,10

The study population was mainly composed of outpatients. We found that inappropriate requests were more frequent in outpatients than in inpatients. This is not unexpected, since inpatients commonly present new symptoms or signs suggesting cardiac disease or worsening of known cardiovascular disease, and both scenarios are rated as appropriate. On the other hand, outpatient requests usually refer to routine TTE in patients with no change in clinical status, which is normally rated as inappropriate. Previous studies have also reported a higher proportion of inappropriate exams among outpatients.4,5

Surprisingly, cardiologists presented a higher rate of inappropriateness (19.1%) than most other specialties. This was mainly related to routine evaluation of outpatients with known CAD and no change in clinical status. These findings conflict with other studies, in which cardiologists’ requests were more often appropriate than other specialties.9,10 At our center, cardiologists have close contact with the echo lab and easier access to the scheduling system than other specialties, which could explain some of the overuse of the technique. Interestingly, we found that cardiologists who read TTE had similar rates of inappropriate requests to other cardiologists.

Cardiology residents presented a higher rate of inappropriate requests than cardiology specialists, although the difference was not statistically significant. Recently, Bhatia et al. reported similar rates of inappropriate TTE among cardiology fellows.12

The Portuguese National Health Service provides health coverage for all the population and is mainly funded through general taxation. Physicians are paid a fixed monthly wage, which is not related to services rendered. Despite constant monitoring of effectiveness and quality, the national public health surveillance system does not perform any systematic monitoring of appropriateness of requested tests. The strict application of AUC would have reduced the number of exams performed annually at our echo lab by around 1500.

There are few data on the best strategies to improve AUC compliance. Bhatia et al. reported a significant reduction in the proportion of inappropriate TTE exams in an inpatient academic medical service after implementation of a simple educational program consisting of lectures, pocket cards and feedback on ordering behavior via e-mail.7 However, some methodological caveats have been pointed out regarding this study.11 Recently, the same group of investigators reported the results of the first randomized control trial of an AUC-based educational and feedback intervention designed to reduce inappropriate outpatient TTE ordered by physicians-in-training, mainly cardiology fellows. After implementing an educational intervention (generally similar to that described above), they reported a significantly lower rate of inappropriate TTE in the intervention compared to the control group (13% vs. 34%, p<0.001).12 Interestingly, in both in- and outpatient settings, there was an increase in inappropriate TTE during follow-up in the post-intervention period, suggesting the need for a continued program to achieve sustained improvement in physicians’ ordering behavior.12,13 Given the heterogeneity of most common inappropriate indications between centers, a strategy tailored to the specificities of each practice environment would be desirable.7

Information on the clinical impact of AUC compliance is scarce and inconsistent. In a multicenter community study, Ballo et al. showed that TTE exams with appropriate or uncertain indications were often clinically more useful than those with inappropriate indications (87% vs. 14%, p<0.001).9 Clinical impact was defined as any change in diagnostic workup, therapeutic decisions or follow-up planning induced by TTE results. However, more recently, in a single academic center study, Matulevicius et al., using the same definition, found a markedly lower overall rate of clinical impact (32%).8 In addition, the proportion of exams resulting in an active change in clinical care did not correlate with AUC classification. However, the use of retrospective review of electronic medical records to determine clinical significance has been identified as a major drawback of this study.14 One of the main challenges in future studies will be to define more precise measures of the utility of appropriate vs. inappropriate TTE in clinical practice.

The 2011 update of the 2007 AUC increased the clinical indications for TTE from 59 to 98 and led to a significant reduction in unclassified studies.5,15 Indeed, in this study the AUC indications covered most of the clinical requests, with only 2.5% being unclassifiable. Control TTE after cardiovascular interventions and TAVI follow-up accounted for the majority of unclassifiable tests. Recently, the European Society of Cardiology and the European Association of Cardiovascular Imaging announced the formation of a taskforce to define appropriateness criteria for cardiovascular imaging use in clinical practice in Europe.16 Some of these missing clinical scenarios should be considered in future recommendations.

This is a single tertiary center study, and therefore the results cannot be extrapolated to other settings. Despite extensive review of all patient information, there was no direct contact with the ordering physician, and so the process of determining the AUC indication is not free of bias related to clinical factors not fully considered by the reviewers. We did not collect data on TTE abnormalities or assess its impact on decision-making due to the lack of detailed electronically stored information. Finally, we had a low number of TTE exams in an acute setting, since in this context cardiologists usually perform focused bedside echocardiography rather than complete TTE at the echo lab. This may result in some underestimation of the overall appropriateness of TTE, as in this group the rate of appropriateness is higher than in outpatients.

The majority of TTE requests at a Portuguese tertiary care center were appropriate. Cardiologists’ requests and outpatient referrals presented the highest rates of inappropriateness. Strategies to improve AUC compliance and evaluation of its impact on clinical outcomes should be explored in the near future.

The authors have no conflicts of interest to declare.