Utilisation of echocardiography in Australia

Appropriate Use Criteria (AUC) for echocardiography were first published in the USA in 2007. AUC for echocardiography have impacted on clinical practice in the USA including provision of services and reimbursement policy. The American College of Cardiology and American Society of Echocardiography (ACC/ASE) recently updated the AUC for echocardiography.1 The 202 indications, derived from both anticipated use and clinical practice guidelines, were scored by a technical panel of 15, including a minority of specialists in echocardiography, to determine whether these indications were appropriate, uncertain or inappropriate. Cost was also considered as a factor. An appropriate imaging study was expected to provide incremental information that together with clinical judgement would exceed the expected negative consequences by a sufficient margin to be considered acceptable care. The negative consequences included risk of the test and the consequences of false-positive or false-negative results (all low for transthoracic echocardiography) and presumably also cost, although there was no defined weighting for this. It was assumed that a complete clinical history and physical examination had been completed by a qualified clinician and that the information in the indication was correct (for example, an asymptomatic patient was truly asymptomatic). This of course may not always be the case in Australian practice, where echocardiography is sometimes incorrectly used as a screening test replacing adequate clinical assessment. The technical panel rated 97 of 202 indications appropriate, 34 uncertain and 71 inappropriate. Predictably, indications for initial diagnosis, change in clinical status or guiding therapy were more likely to be considered appropriate indications for echocardiography. In contrast, indications for routine testing without a change in clinical status or where results were unlikely to change management were more likely to be considered uncertain or inappropriate. The ACC/ASE writing group stated that the AUC are likely to be useful to clinicians, health care facilities, third party payers and informing policy decisions for cardiovascular imaging. This statement will be considered with interest by Australian stakeholders, including review groups involved in a planned government review of the Medicare Benefits Schedule.2 There is interest in adapting the American AUC for transthoracic echocardiography to Australian practice. An Australian group (Fonseca et al.) has evaluated the indications considered appropriate in the 2011 AUC by matching these to class of recommendations in major published clinical guidelines.3 Of the 202 AUC, 90 could be matched to guidelines, and of these, 53 were usually appropriate and 76 were concordant with guidelines (Class I or IIa indication). Fonseca's group thus demonstrated inconsistencies in 15% of the 98 AUC for transthoracic echocardiography and expressed concern that rating of appropriateness in the ACC/ASE document depended on expert opinion rather than levels of evidence. Prior to AUC or a modification being adopted in Australia, improved internal consistency would be required between AUC and guidelines. Al-Kaisey extended the study of AUC applicability to regional Australia by considering the 2011 AUC in Bendigo Hospital, Victoria, in relation to 1000 selected transthoracic echocardiography requests from January to June 2014.4 Of the 976 classifiable requests, 752 (77%) were considered appropriate and 198 (20.3%) inappropriate according to the 2011 AUC. There were 16 unclassifiable requests, which included indications specific to Australia such as commercial licence or defence force assessment and monitoring while on Pharmaceutical Benefits Scheme-subsidised treatment for pulmonary arterial hypertension (seven studies). As previously described in the 2011 AUC, a significant proportion of inappropriate indications was for routine surveillance without a change in clinical status. This group comprised more than 30% of requests, and 55% of these patients had undergone echocardiography within the preceding 12 months. Other common inappropriate indications were routine perioperative evaluation of ventricular function (24 patients) and transient fever without bacteraemia, a new murmur or transient bacteraemia with an atypical organism or endovascular source of infection (11 patients). It was estimated that in Bendigo Hospital alone, adherence to AUC and performing only studies with appropriate indications could reduce transthoracic echocardiography studies by 700 per year and produce significant time and cost savings. Before the AUC could be utilised in Australia, revisions would be required to ensure Australian-specific clinical indications are included as well as improving consistency of AUC and clinical guidelines. In a second manuscript published in this Journal, Fonseca et al. examined growth rates and regional differences in outpatient cardiac imaging in Australia from 2002 to 2013.5 Transthoracic echocardiography had an annualised growth rate of 8.8%, increasing from 1780 to 3497 per 100 000 people. Fonseca showed that the main correlation with this increase was the number of medical practitioners per 100 000 people. (Use of stress echocardiography increased markedly over this time, with a relatively small increase in SPECT, which may reflect increasing use of stress echocardiography and regional variations in access to this test.) Fonseca et al. comment that substantial regional variation in use of imaging modalities (15-fold to 47-fold variation) may be a marker of poor practice. Increases in health expenditure have resulted in attempts to contain this growth. In the USA, development of AUC and changes in reimbursement appear to have slowed growth in spending. A similar exercise in Australia may become inevitable. It may also be possible to constrain use of echocardiography by modifying existing practice protocols and diagnostic pathways. Quinlivan et al. describe a strategy to reduce utilisation of echocardiography in patients with scleroderma, after re-examining diagnostic tests in a group of patients with scleroderma and pulmonary hypertension.6 Early diagnosis of pulmonary arterial hypertension (PAH) in patients with scleroderma is important as this allows earlier initiation of treatment with documented improved functional capacity and survival. The American College of Cardiology Foundation and American Heart Association recommend transthoracic echocardiography if pulmonary hypertension is suspected on history, risk factors or examination.7 Mukerjee et al. note that scleroderma patients develop symptoms at relatively low pulmonary pressures.8 Resting echocardiography is less useful when estimated tricuspid gradient is <40 mmHg, as false-positive results increase. At present, screening for pulmonary arterial hypertension in scleroderma in Australia is based on transthoracic echocardiography and pulmonary function tests. Quinlivan et al. demonstrate that yearly screening with the biomarker N-terminal pro-B-type natriuretic peptide and pulmonary function tests in combination instead of echocardiography could result in cost savings of $1936 (15%) per case of PAH diagnosed. This does not remove the role for echocardiography as a baseline assessment to provide information in addition to right ventricular : right atrial (RV : RA) pressure gradient. Baseline assessment could include right ventricular size and function according to ASE criteria, other measures of pulmonary hypertension such as pulmonary acceleration time and exercise assessment of RV : RA gradient and right ventricular function. However, annual echocardiography could be avoided, with associated cost savings. Such a change in practice should be incorporated into the AUC. There will be increasing pressure to limit costs for diagnostic imaging in Australia, as has been the case in the USA. A modification of the AUC, revised for Australian practice and inpatient as well as outpatient care, is likely to form part of this. Sensible review of existing screening practices, such as in those patients with chronic diseases like scleroderma, or in assessment prior to cardiac and non-cardiac surgery, may also have a place. It remains important to ensure that standards of practice of echocardiography are maintained and improved, particularly with increasing uptake of more specialised tests such as transoesophageal echocardiography and stress echocardiography, even as spending on health care is constrained.