Pre-hospital thrombolysis in ST-segment elevation myocardial infarction: a regional Australian experience

The known Pre-hospital thrombolysis in patients with ST-segment elevation myocardial infarction (STEMI) has been found to be an effective strategy in randomised controlled trials.

The new Pre-hospital thrombolysis is safe and effective in the real world setting, especially in a region where transport distances to the cardiac catheterisation laboratory are great.

The implications Pre-hospital thrombolysis can be implemented in regions where timely primary percutaneous coronary intervention for STEMI patients is not available.

Primary percutaneous coronary intervention (PCI), if performed in a timely manner, is the preferred reperfusion strategy for patients with an ST-segment elevation myocardial infarction (STEMI).1 However, physicians in regions geographically isolated from a primary PCI centre are unable to perform PCI rapidly, and the optimal reperfusion strategy is therefore undefined. Recent guidelines have highlighted the importance of systems-based approaches to STEMI management, taking into account regional characteristics.2 In Australia, many patients are a long way from a primary PCI centre, and these patients usually receive thrombolysis in hospitals that are not PCI-capable. However, if reperfusion with thrombolysis fails or a contraindication prevents its application, there can be long delays in reaching a cardiac catheterisation laboratory (CCL) for PCI.3 Further, a pharmaco-invasive (PI) strategy, including routine cardiac catheterisation within 24 hours of thrombolysis, was found to achieve better outcomes than ischemia-driven angiography.4 This strategy is especially relevant in areas where geographic access and logistical delays that increase the time to treatment (eg, rural areas, large urban areas with traffic congestion) may reduce some of the benefits of primary PCI in clinical practice.5

The international Strategic Reperfusion Early After Myocardial Infarction (STREAM) trial, in which STEMI patients who were unable to undergo primary PCI within one hour of presentation were randomised to either primary PCI or a PI strategy, confirmed that outcomes of the PI strategy at one year were not inferior to those of primary PCI.6,7 However, the STREAM trial did not include all patients who presented with STEMI; although most of those randomised underwent randomisation in the ambulance setting (81%), the patients included in the trial had presented within 3 hours of symptom onset (compared with 6–12 hours for the STEMI patients who were not included), and the median time from presentation to PCI in the PI arm was 10 hours, which may not be achievable in a real world setting. Similarly, geographical and logistical constraints that are important factors in different parts of the world, such as distance and regional resources, were not considered in the STREAM trial.

The case series described in this article was from the Hunter New England Local Health District (HNELHD), which has a complex geography and limited resources. The area it covers (130 000 km²) is comparable in size with England, but it has only one continuously operating CCL; it is located in one corner of the health district (in Newcastle), and patients with a STEMI may be as far as 600 km away (Box 1). In 2008, we implemented a system of care in HNELHD for providing rapid reperfusion to STEMI patients throughout this large district. Modelled on the Emergency Triage of Acute Myocardial Infarction study in the Northern Sydney Area Health Service, which achieved reduced delays and improved outcomes for patients with primary angioplasty,9 our system involved pre-hospital diagnosis and triage by paramedic staff, followed by their allocating the patients to either PHT or primary PCI according to their travel time to the CCL in our large regional health district. The analysis reported in this article compares the safety and effectiveness of pre-hospital assessment and allocation to pre-hospital thrombolysis (PHT) or primary angioplasty by paramedics.10

Methods

Study design

Hunter AMI was a prospective, non-randomised, consecutive, single-centre case series of STEMI patients who had been diagnosed on the basis of a pre-hospital ECG performed by paramedics. Data were collected electronically through the collaboration of the New South Wales Ambulance and the HNELHD, and data were independently analysed by the Clinical Research Design, IT and Statistical Support (CReDITSS) unit at the Hunter Medical Research Institute.

Study patients

For all patients with symptoms or signs suggesting myocardial infarction, paramedics recorded a 12-lead electrocardiogram (ECG; LIFEPAK 15 monitor system, Physio-Control Australia). Posterior and right-sided leads were not recorded by the paramedics for infero-posterior or right ventricular myocardial infarction. If the ECG indicated a possible STEMI according to the Glasgow algorithm,11,12 it was transmitted to the on-call cardiologist or cardiology advanced trainee at the John Hunter Hospital in Newcastle; the doctor and the paramedic then discussed the case, and reperfusion of the patient started. For patients who could reach a CCL within 60 minutes of first medical contact (FMC) or for whom there was any contraindication to thrombolysis (irrespective of FMC-to-CCL door time), reperfusion therapy consisted of primary PCI at the John Hunter Hospital. PHT (fibrinolysis with tenecteplase administered by paramedics) was administered by paramedics to patients for whom the FMC-to-CCL door time exceeded 60 minutes, and was followed by transfer to the PCI-capable centre (Appendix 1). All consecutive STEMI patients who had been diagnosed by pre-hospital ECG were included in our analysis.

FMC was defined as the first face-to-face contact made by paramedics with the patient. Total ischaemic time was defined as the time from symptom onset to balloon inflation for the primary PCI group, and to injection of tenecteplase for the PHT group. Hypertension was diagnosed if blood pressure greater than 140/90 mmHg was measured on two or more occasions in the hospital, or if the patient was already being treated with antihypertensive medication.13,14 Diabetes mellitus was diagnosed if two of the following applied: symptoms of hyperglycaemia; fasting sugar level > 7 mmol/L; 2-hour post-prandial sugar level > 11.1 mmol/L; HbA_1c level > 6.5 mmol/mol.15 Cardiogenic shock was defined as persistent hypotension (systolic blood pressure < 90 mmHg or mean arterial pressure 30 mmHg lower than baseline).16 Left ventricular ejection fraction was recorded during the index hospitalisation by transthoracic echocardiography, using Simpson’s biplane method.

Study therapies

We present the outcomes of PHT and primary PCI performed according to our local guideline-based protocol, which includes concomitant dual antiplatelet and anticoagulant therapy. Adjunct administration of a glycoprotein IIb/IIIa antagonist was at the discretion of the intervening cardiologist. In the PHT arm, tenecteplase was administered at a weight-based dose (55 to < 60 kg, 30 mg; 60 to < 70 kg, 35 mg; 70 to < 80 kg, 40 mg; 80 to < 90 kg, 45 mg; ≥ 90 kg, 50 mg). The tenecteplase dose administered to patients aged 75 years or more has since been reduced by half. Anticoagulant therapy consisted of a 30 mg intravenous bolus of enoxaparin (omitted for patients aged 75 years or more) followed by subcutaneous injection of 1 mg/kg bodyweight (0.75 mg/kg for patients aged 75 years or more) twice each day. Antiplatelet therapy consisted of a 300 mg loading dose of clopidogrel (omitted for patients aged 75 years or more) followed by 75 mg daily, together with an initial 300 mg aspirin dose, immediately followed by 100 mg daily. Patients in the PHT arm with electrical or haemodynamic instability, ongoing ischaemic symptoms, or less than 50% ST-segment resolution within 90 minutes of thrombolysis underwent rescue PCI.

Primary endpoints

The primary efficacy endpoint was 12-month all-cause mortality. The primary safety endpoint was bleeding, categorised according to Thrombolysis in Myocardial Infarction (TIMI) study group criteria.17

Statistical analysis

Summary statistics for patient characteristics in each of the two groups are presented. Normality of distribution was assessed with the Shapiro–Wilk test. Between-group comparisons of continuous variables used independent sample t tests for parametric data and pairwise comparison median tests for non-parametric data. The Pearson χ² test assessed differences between categorical variables. Nelson–Aalen survival distributions were estimated, and adjusted using Cox proportional hazards regression models. Predictors of 12-month mortality were assessed using logistic regression. All statistical analyses were conducted in Stata 14 (StataCorp) and SAS 9.4 (SAS Institute).

Ethics approval

The study protocol was approved by the Hunter New England Human Research Ethics Committee in August 2014 (reference, 14/06/18/5.09).

Results

From 1 August 2008 to 31 August 2013, 484 patients with STEMI diagnosed by pre-hospital ECG were allocated to PHT (150 patients) or primary PCI (334 patients) on the basis of estimated FMC-to-CCL door times or any contraindication to thrombolysis (Appendix 2). There were more patients in the primary PCI group with diabetes mellitus or hypertension, and fewer with hypercholesterolaemia (Box 2).

The median time from FMC to the start of reperfusion was 35 minutes (IQR, 28–43 min) for bolus tenecteplase and 130 minutes (IQR, 100–150 min) for balloon inflation (P = 0.001). The median time between symptom onset and treatment was correspondingly shorter in the PHT group (94 [IQR, 65–127] v 180 [IQR, 140–265] minutes) (Box 2). The median distance of patients in the PHT group from the CCL was 119 km (range, 8–483 km). The median time from symptom onset to angiography was thus longer in the PHT group than in the primary PCI group, with a delay of 4 hours for the 27% of patients who required rescue or urgent intervention, and of 49 hours for the other patients in this group. Significantly, more open vessels (TIMI flow grade of 2 or more) were found on first angiography in the PHT than in the primary PCI group (45% v 8%; P < 0.001) (Box 3).

There were three ischaemic strokes and two transient ischaemic attacks in the primary PCI group and none in the PHT group. Four of the five patients who suffered a transient ischaemic attack or stroke had atrial fibrillation, either before or detected during the admission. Bleeding was significantly more frequent in the PHT group than in the primary PCI group (9.3% v 5.1%; P = 0.001) with two (1.3%) TIMI major bleeds and one (0.7%) intracranial haemorrhage in the PHT group (Box 4).

Of the total cohort of 484 patients, 34 (7.0%) died within 12 months (the primary efficacy endpoint): 10 of 150 patients (6.7%) in the PHT group and 24 of 334 (7.2%) in the primary PCI group (relative risk in the PHT group, 0.93; 95% CI, 0.45–1.9; P = 0.84) (Box 5). Anterior STEMI, cardiogenic shock, and hypertension were independent predictors of mortality for the total cohort of 484 patients in the multivariate analysis (Box 6). Age, bleeding, ejection fraction and femoral access were significant only in the univariate analysis (data not shown).

Discussion

Our study produced several important findings. First, delivery of PHT by paramedics, based on algorithm assessment of 12-lead ECGs, is feasible and safe in regional Australia. Second, 12-month mortality for patients remote from a CCL who receive PHT and are then transferred to a PCI-capable centre is similar to that for patients near the CCL who receive primary PCI.

Our practice is to administer fibrinolysis to STEMI patients who are remote from the CCL if there is no contraindication, and to then assess clinical and ECG signs of reperfusion. If they were successfully reperfused, we did not mandate immediate cardiac catheterisation but instead brought them to the CCL as soon as convenient, provided they remained clinically stable in the meantime. Those for whom reperfusion was not successful or chest pain recurred were urgently transferred to the CCL for rescue PCI.

To place our findings in context, we compared them with the outcomes of the STREAM trial.7 Our PHT group had some notable differences from the STREAM PI group: our patients were older, if not statistically significantly (62 years [SD, 13] v 60 years [SD, 12]), and a greater proportion of our patients were aged 75 years or more (18% v 14%); we also had a higher proportion of patients with cardiogenic shock (5.3% v 0.1%). Sex balance, rates of diabetes, hypertension, and prior coronary artery bypass graft surgery, and symptom onset-to-treatment times were similar for our PHT patients and the STREAM fibrinolysis group. The median time to angiography for our PHT group was 28 hours, compared with 10 hours in the STREAM fibrinolysis arm. Despite these differences, which indicate a higher risk level for our patients, survival at 12 months in the two fibrinolysis groups was identical (93.3%).

All-cause mortality at 12 months was similar for both treatment groups in our study, but this was not a randomised controlled trial, so that conclusions about treatment efficacy should not be drawn. The safety of the pre-hospital assessment and treatment, on the other hand, is reassuring. As expected, there was a significantly higher bleeding complication rate in the PHT group, but only one intracranial haemorrhage. There was a higher stroke rate in the primary PCI group; most events occurred in association with atrial fibrillation, but this finding requires further follow-up assessment of potential contributors.

The optimal ECG algorithm for the detection of STEMI is unknown. For ease of use in the field, we chose to perform standard 12-lead ECGs without right ventricular or posterior leads. For their interpretation, we use the Glasgow algorithm because of its specificity.10 It remains unknown whether the addition of right ventricular or posterior leads, or the use of another algorithm, would improve the sensitivity or specificity of detection in this setting.

The transport times for the primary PCI group in our study were considerably longer than the 60-minute FMC-to-CCL goal. The decision to administer or withhold PHT was driven mainly by the estimated FMC-to-CCL time. In our protocol, the approximate time to the CCL is estimated by the paramedics, as their knowledge of prevailing weather and traffic conditions means they are in the best position to do so. Their decision is then discussed with the cardiologist or advanced trainee. There may be some bias towards underestimating transport times, but, more importantly, there are inherent difficulties in adhering to these guidelines. First, patients who had any contraindication to fibrinolysis underwent primary PCI, regardless of the estimated FMC-to-CCL time, and this would increase overall transport times for the primary PCI group. Second, clinical factors (eg, treatment of arrhythmias, resuscitation from cardiac arrest) sometimes interfere with transport, also prolonging transfer times. It is therefore important that actual transport times are reported, not just the goal transfer times. It is hoped that knowledge of both will help when devising systems of care for other Australian regions.

The long distances to the CCL for the patients in our PHT group underscore the importance of implementing systems of care in areas with complex geography that achieve optimal outcomes for STEMI patients. The median times from symptom onset to treatment in both our treatment groups were comparable with those of the STREAM groups. This finding suggests that total ischaemic time is the interval with the greatest prognostic impact, and that we should focus on reducing this period, not than just door-to-balloon times. While medical practitioners can influence treatment intervals after FMC, community education campaigns that encourage patients to call an ambulance as soon as they have chest pain are needed for reducing the time from symptom onset to FMC.

Our study had some important limitations. First, it was a non-randomised, consecutive case series study. We present the mortality associated with two therapies, but baseline and other differences between the two groups may have contributed to the different outcomes. Second, our study was underpowered for detecting small differences in mortality, so that our results should be interpreted with caution. It is nevertheless hoped that our data can inform systems of care for patients with STEMI throughout Australia.

In conclusion, our real world experience showed that PHT delivered by paramedics followed by early transfer to a PCI-capable centre is a safe and effective reperfusion strategy for patients remote from primary PCI centres.

Box 1 –
The Hunter New England Local Health District, about 600 km from north to south. The John Hunter Hospital is located in Newcastle, in the bottom right hand corner of the map*

* Adapted from: Eastwood et al (2010).8

Box 2 –
Baseline characteristics of the pre-hospital thrombolysis (PHT) and primary percutaneous coronary intervention (PCI) groups

	All patients	PHT group	Primary PCI group	P

Total number	484	150	334
Age (years), mean (SD)	64 (13)	62 (13)	65 (13)	0.3
≥ 75 years	115 (24%)	27 (18%)	88 (26%)	0.05
Sex (males)	365 (75%)	114 (7%)	251 (75%)	0.8
Systolic blood pressure (mmHg), mean (SD)	130 (24)	125 (25)	131 (24)	0.1
Cardiogenic shock	30 (6.2%)	8 (5.3%)	22 (6.5%)	0.8
Anterior STEMI	188 (39%)	55 (37%)	133 (40%)	0.5
Cardiovascular history
Coronary artery disease	96 (20%)	28 (19%)	68 (20%)	0.9
Prior coronary artery bypass graft surgery	13 (2.7%)	2 (1.3%)	11 (3.3%)	0.2
Hypertension	267 (55%)	64 (43%)	203 (61%)	< 0.001
Diabetes mellitus	97 (20%)	20 (13%)	77 (23%)	0.01
Smoking	213 (44%)	67 (45%)	146 (44%)	0.8
Hypercholesterolaemia	198 (41%)	76 (51%)	122 (36%)	0.002
Body mass index (kg/m²), mean (SD)	29 (5)	28 (6)	29 (5)	0.08
First medical contact to treatment (min), median (IQR)	105 (49–140)	35 (28–43)	130 (100–150)	0.001
Symptom to treatment (min), median (IQR)	155 (107–235)	94 (65–127)	180 (140–265)	< 0.001
Symptom to first medical contact > 3 h	83 (17%)	14 (9.3%)	69 (21%)	0.002
Ejection fraction, mean (SD)	48% (8)(n = 332)	49% (10)(n = 236)	47% (7)(n = 96)	0.01
Peak troponin (ng/mL), mean (SD)	48 (32)	44 (28)	50 (35)	0.04
Haemoglobin (g/L), mean (SD)	144 (48)	145 (43)	143 (56)	0.2
Creatinine (μmol/L), mean (SD)	95 (31)	89 (20)	98 (34)	0.006
Length of stay (days), mean (SD)	4 (3)	4 (3)	4 (3)	1.0

STEMI = ST-segment elevation myocardial infarction.

Box 3 –
Angiographic characteristics of the pre-hospital thrombolysis (PHT) and primary percutaneous coronary intervention (PCI) groups

	PHT group	Primary PCI group	P

Total number	150	334
Coronary angiography undertaken	138 (92%)	334 (100%)	< 0.001
Rescue PCI undertaken	37 (27%)	NA	—
Symptom onset to angiography (h), median (IQR)	28 (6–70)	3.5 (2.2–4.2)	< 0.001
Time to rescue PCI (h), median (IQR)	4 (3–5)	NA	—
Initial TIMI flow score ≥ 2	67 (45%)	27 (8.1%)	< 0.001
PCI/CABG undertaken	97 (65%)	307 (92%)	< 0.001
Femoral access	52 (38%)	153 (46%)	0.08
IIb/IIIa antagonist administered	7 (4.7%)	151 (45.2%)	0.004

CABG = coronary artery bypass graft surgery; NA = not applicable; TIMI = Thrombolysis in Myocardial Infarction study group.

Box 4 –
Primary efficacy and safety outcomes for the pre-hospital thrombolysis (PHT) and primary percutaneous coronary intervention (PCI) groups

PHT group

Primary PCI group

Primary efficacy outcome

12-month all-cause mortality

10 (6.7%)

24 (7.2%)

0.84

Primary safety outcomes

Intracranial haemorrhage

1 (0.7%)

—

Total bleeding (TIMI bleeding criteria)

14 (9.3%)

17 (5.1%)

0.001

TIMI major bleeding

2 (14%*)

—

TIMI minor bleeding

5 (36%*)

9 (53%*)

0.005

TIMI minimal bleeding

7 (50%*)

8 (47%*)

0.5

TIMI = Thrombolysis in Myocardial Infarction study group. * Percentage of all bleeding.

Box 5 –
Nelson–Aalen cumulative hazard estimates for patients receiving pre-hospital thrombolysis or primary percutaneous coronary intervention

Box 6 –
Independent predictors of mortality in 484 patients with ST-segment elevation myocardial infarction (multivariate analysis)

Odds ratio (95% CI)

Anterior STEMI

2.4 (1.2–4.8)

0.01

Cardiogenic shock

8.5 (3.4–21.3)

< 0.001

Hypertension

4.2 (1.8–10)

0.001

STEMI = ST-segment elevation myocardial infarction.

Variation in coronary angiography rates in Australia: correlations with socio-demographic, health service and disease burden indices

The known Angiography rates vary across Australia. Whether this variation is correlated with indices of socio-economic deprivation, chronic disease, acute coronary syndrome (ACS) incidence, or health service characteristics is uncertain.

The new Social disadvantage and remoteness were correlated with ACS incidence and mortality, but not with angiography rates. Private hospital cardiac admissions were strongly correlated with angiography rates; the relationship with public hospital cardiac admissions was less marked. Socio-economic indicators, regional location, and ACS and chronic disease burden were not significantly associated with angiography rates.

The implications A focus on clinical care standards and better health service distribution is needed to reduce the variation.

Managing coronary artery disease (CAD) with coronary angiography is informed by an extensive evidence base.1–5 Variation in the rates of coronary angiography has been described in Australia.6,7 This variation may be explained by heterogeneity in clinical need (ie, variations in disease incidence or prevalence), but differences unexplained by disease burden highlight inequities in access to health services, and differential over- and underuse of health care resources. These disparities are potentially targets for policy interventions that aim to improve population health and achieve a high quality health system.

Health care in Australia faces a combination of challenges. The geographic distribution of the population and substantial cultural diversity give rise to complexity in providing access to clinical expertise and procedures such as angiography. Clinical audits of acute coronary syndrome (ACS) practice have identified variations in applying components of care advocated in guidelines, particularly invasive management.8–12 Australia’s demography may contribute to heterogeneity in access to health services, differential clinical needs and variation in care. The relative contributions of demographic factors have implications both for national policy and for local efforts to re-design health services.

In our study we explored the associations of selected socio-economic, geographic, and chronic disease factors with ACS incidence and mortality rates, and examined whether rates of coronary angiography across the Australian population are correlated with indicators of disease burden, health access, and clinical activity.

Methods

Data sources

Socio-economic and health workforce data

Our analysis included data for the entire Australian population of about 23.5 million people. Between 2010 and 2015, federal government support for primary health care services was organised into 61 geographic divisions (Medicare Locals). Social, economic and health service characteristics of the Medicare Locals were obtained from a publicly accessible website that publishes age- and sex-standardised rates for these features, including the estimated proportions of privately insured residents and of Indigenous residents, based on data from the 2011 Australian census (conducted by the Australian Bureau of Statistics [ABS]).13 Data from the Socio-Economic Indexes for Areas (SEIFA),14 a relative measure of social and economic disadvantage (normalised to 1000; higher numbers denote more advantaged areas), were also used as a socio-economic indicator. Medical workforce data were drawn from the annual survey of the Australian Health Practitioner Registration Authority, which records the primary locations of practice and specialisations of all registered medical practitioners. Further, modelled rates of chronic cardiovascular conditions (prior myocardial infarction and angina, heart failure, stroke and rheumatic heart disease), estimates derived from the Australian Health Survey 2011–2013 (conducted by the ABS), were available for all but three Medicare Locals (Tasmania, Northern Territory, Australian Capital Territory).

As an indicator of local clinical practice in each region, the likelihood that a suspected ACS patient received coronary angiography, compared with the national average, was estimated from the SNAPSHOT ACS clinical audit.9,10

Rates of ACS, angiography, revascularisation and mortality

The National Hospital Morbidity Database (NHMD) was used to identify ACS separations (International Classification of Diseases, revision 10, Australian modification [ICD-10-AM] principal diagnosis codes I20 and I21) and catheter procedures during the calendar year 2011. Data on procedures undertaken as ambulatory care (eg, outpatient angiography) were obtained by the Australian Institute of Health and Welfare (AIHW) from the Medicare Benefits Schedule. The combined total rate of angiography is presented in this article. All cardiac-specific admissions were included in the analysis. Three of the eight state or territory jurisdictions did not report data for private hospital admissions, so that private hospital and total admissions were not available for three Medicare Locals (Tasmania, NT, ACT). Data on deaths attributed to CAD were drawn from the National Mortality Database (NMD).

Statistical analysis

Standardised age- and sex-specific rates — of ACS separations, inpatient and outpatient angiograms, percutaneous coronary interventions (PCI), coronary artery bypass graft (CABG) procedures, and CAD-related deaths of people aged 35 years or more grouped in 5-year age intervals (to 85 years or more) and by sex — were calculated by dividing by the corresponding Australian population figure (at 30 June 2001) for the relevant age and sex group, and expressed as numbers per 100 000 population.

Socio-economic data, health service information, procedure and ACS incidence, and mortality rates were stratified by geographic location of the Medicare Local as defined by the AIHW, and compared using Kruskal–Wallis tests. Univariate correlations between potential explanatory factors and rates of angiography, ACS and mortality were assessed using partial correlations; the strength of the adjusted correlation coefficient (CC) was defined as strong (> 0.70), moderate (0.50–0.69), or weak (0.30–0.49). Separate correlation plots of ACS v CAD mortality rates and of angiography v ACS rates were generated, and fitted linear predictions superimposed. To assess the potential overuse of angiography, the ratio of the numbers of coronary revascularisations to those of coronary angiograms was calculated for each Medicare Local, and plotted as a function of the rate of coronary angiography, using the fractional polynomial estimate.

Given the small numbers of Medicare Locals and the availability of earlier data to evaluate these relationships, a Bayesian linear regression approach was used (Appendix). All analyses were undertaken in Stata 14.0 (StataCorp); P < 0.05 was deemed statistically significant.

Ethics approval

This SNAPSHOT ACS study was conducted with local ethics approval for opt-out consent, except for two hospitals where opt-in consent was applied (lead ethics committee: Cancer Council NSW Human Research Ethics Committee; reference, 2011/06/334). The AIHW work program and any release of data from AIHW datasets are subject to the oversight of the AIHW Ethics Committee and handled in accordance with the AIHW Act 1987 (as amended), the Privacy Act 1988, and any terms and conditions set by data providers. Separate ethics committee approval is not required for the analysis of AIHW datasets that does not involve data linkage, and was thus not needed for our study.

Results

Characteristics of the Medicare Locals

Of the 61 Medicare Locals, 27 were categorised by AIHW criteria as metropolitan, 24 as regional and ten as rural. Regional and rural locations were significantly more disadvantaged than metropolitan areas, with higher proportions of the population on long term unemployment support, greater reported delays in seeking medical consultation because of the associated costs, and lower proportions of residents with private health insurance. Similarly, the prevalence of smoking, obesity and chronic cardiovascular disease was higher in regional and rural areas than in metropolitan areas. There was no difference in the rates of total hospital admissions by geographic location, but there was a significantly lower rate of private hospital cardiac admissions in non-metropolitan locations, where fewer private hospitals exist (Box 1).

ACS rates were higher in non-metropolitan areas, but this was not reflected in a significantly higher rate of coronary angiography. PCI rates were significantly lower in rural areas, while rates of CABG were significantly higher in non-metropolitan areas. Overall, there was no significant difference in the combined coronary revascularisation rates according to geographic location. Premature death from CAD and total mortality were higher in regional and rural areas. There were 3.7-fold (ACS), 5.3-fold (angiography), 2.2-fold (revascularisation) and 2.3-fold (CAD mortality) differences between the lowest and highest rates for individual Medicare Locals. Box 2 shows the variation in ACS, angiography, revascularisation, and mortality rates for each Medicare Local by SEIFA index score.

Correlation of socio-economic, chronic health and health service data with ACS admission mortality rates

The ACS admission rates for individual Medicare Locals were significantly correlated with all-cause mortality (CC, 0.52; P < 0.001). There were strong correlations between socio-economic measures and ACS admission rates and mortality (Box 3). Similarly, rates of smoking, obesity and chronic cardiovascular conditions were all correlated with mortality; smoking and obesity rates were also correlated with ACS admission rates. Strong negative correlations between the proportion of insured people in the Medicare Local and the ACS admission and all-cause mortality rates were also observed. Conversely, there were negative correlations between local availability of specialist physicians and ACS admissions and mortality, as well as positive correlations between delays in consultations and these outcomes. From a health service use perspective, emergency department and public hospital admission rates were correlated with ACS admission and mortality rates, while private hospital cardiac admission rates were negatively associated with total mortality. An increased likelihood of patients admitted with suspected ACS undergoing angiography was associated with lower mortality (Box 3).

An adjusted analysis of mortality was performed to explore relationships with the indicators SEIFA score, regional location, local cardiovascular health status (chronic cardiovascular conditions), ACS rates, workforce capacity (access to specialist physician care), and health service provision (cardiac admission rates to public and private hospitals). Rural location was associated with increased mortality (19 additional deaths [95% CI, 10–27] per 100 000 population). Interestingly, the likelihood of coronary angiography in the context of ACS appeared to be associated with a modest reduction in mortality rates (three fewer deaths [95% CI, 1–5] per 100 000 population for each 10 percentage point increase in the likelihood of angiography for a suspected ACS admission). After considering these two factors, socio-economic index, disease burden, health service indicators, and angiography rates were not significantly correlated with the Medicare Local CAD mortality rate.

Correlation of socio-economic status, chronic health status and health service with angiography rates

There was no correlation between measures of social disadvantage or health service availability and coronary angiography rates (Box 3). A positive correlation between all cardiac admissions and angiography rates was evident, particularly private hospital cardiac admissions. This correlation was weaker when analysis was restricted to public hospital admissions. The likelihood of angiography for acute patients was not correlated with the overall rate of angiography in Medicare Locals. There was a weak correlation between the rates of angiography and of ACS (CC, 0.31; P = 0.018), but no correlation with premature ischaemic heart disease deaths (CC, 0.13; P = 0.315) or total CAD mortality (CC, 0.06; P = 0.671) (Box 4).

In the adjusted model, private hospital cardiac admissions had a large influence on the angiography rate (71 additional angiograms [95% CI, 47–93] per 1000 admissions). The relationship between public hospital admission rates and the angiography rate was more modest (44 angiograms [95% CI, 25–63] per 1000 admissions). Socio-economic indicators, regional location, and background ACS or chronic disease rate burden were not significantly associated with angiography rates.

Progression from angiography to revascularisation

The angiography rate was correlated with those of PCI (CC, 0.54; P < 0.001), CABG surgery (CC, 0.44; P < 0.001) and any revascularisation (CC, 0.65; P < 0.001). Revascularisation rates as a proportion of angiography rates varied greatly (17–61%). There was a striking negative correlation between the angiography rate and the proportion of patients undergoing angiography who proceeded to any form of coronary revascularisation (CC, −0.71; P < 0.001) (Box 5). This was also seen with the individual modes of revascularisation (PCI: CC, −0.62; P < 0.001; CABG: CC, –0.62; P < 0.001). There was no correlation between PCI rates and the incidence of myocardial infarction (CC, −0.11; P = 0.402) or ACS (CC, −0.12; P = 0.345). However, rates of CABG were correlated with those of myocardial infarction (CC, 0.53; P < 0.001) and of ACS (CC, 0.45; P < 0.001).

Discussion

We observed that:

increasing socio-economic disadvantage, rural location, and chronic disease burden were each correlated with rates of ACS and of total mortality;
local availability of specialist physicians and admissions to private hospitals were negatively correlated with ACS admissions and mortality rates;
there was no association between coronary angiography rates and the burden of chronic cardiovascular disease, and a modest positive association with ACS rates, but there was a positive association between angiography rates and those of cardiac admissions to private hospitals; and
the rates of angiography and coronary revascularisation were correlated, but there was a negative correlation between the local angiography rates and the proportion of these procedures proceeding to revascularisation.

These findings suggest that health reforms aimed at the appropriate use of diagnostic coronary angiography may be required to improve consistency and equity of access, and consequently to deliver positive outcomes for the Australian community more efficiently.

As expected, a correlation between the incidence of ACS and CAD mortality was evident. Similarly, higher ACS rates in regional and rural locations were confirmed, as was the relationship between indicators of socio-economic disadvantage and chronic cardiac disease burden, and between disease incidence and outcomes. However, the distribution of acute care services was negatively correlated with the incidence of disease in the population. This geographic mismatch has also been described elsewhere, including the United States, and highlights the influence of factors such as funding and workforce proficiency in delivering services to non-metropolitan communities.15–17

A solid evidence base supports using coronary angiography, with subsequent revascularisation where deemed appropriate, in ACS patients with elevated troponin levels.18 However, the superiority of angiography and revascularisation to medical management of patients with stable CAD is less robust.5,19 This clinical evidence base contrasts strongly with several of our findings. Firstly, the indicator most associated with variation in angiography appeared to be cardiac admission rates to private hospitals, but an inverse correlation with ACS rates implies that these procedures are mainly being undertaken for indications other than ACS. Secondly, while angiography rates were correlated with revascularisation rates, neither angiography nor PCI rates (unlike the CABG rate) were correlated with that of ACS. Thirdly, angiography rates were negatively correlated with progression to revascularisation. These observations are inconsistent with the evidence base for invasive management for non-ACS indications, with private institutions accounting for a higher proportion of the variation. Computed tomography coronary angiography for investigating non-acute CAD may reduce the use of invasive angiography for this indication, but caution should be exercised, as there is ample evidence that all forms of cardiac testing tend to motivate further investigations.19,20

These comparisons suggest certain policy targets for improving the clinically appropriate application of coronary angiography. At the higher end of patient risk, the Australian Commission on Safety and Quality in Health Care has developed clinical care standards for the management of ACS.21 These standards may appropriately increase the use of angiography for ACS patients, for whom the benefits of invasive management have been established. Linking the funding of hospitals to ACS performance measures may be an approach for achieving changes in practice and outcomes. Further, resourcing and implementing clinical support networks that serve regional and remote areas (eg, telemedicine) would enable the appropriate selection of patients who would benefit most from being transferred for angiography and revascularisation, and this would also be an opportunity for improving access to angiography. While there is no current guidance on stable CAD in Australian, criteria for the appropriateness of angiography and revascularisation have been developed in the US.22 It is notable that re-imbursement by Medicare and Medicaid in the US for the costs of invasive procedures is now linked to these appropriateness criteria. It is suggested that funding linked to the appropriateness of care or the achievement of clinical care standards, and limiting re-imbursement for angiography in low value clinical situations, should be focuses of debate in any health care reform discussion in Australia.

Limitations

Given the ecological study design, pockets of excellence in care and outcomes probably exist but are obscured by data aggregation. Further, in view of the small number of Medicare Locals, a linear relationship between variables was assumed, although curvilinear relationships are possible. The small sample of the SNAPSHOT ACS study is acknowledged, with caution accordingly exercised when interpreting the association with mortality rates. While this relationship should be examined in larger studies, our finding is consistent with international large scale data.23 Similarly, we cannot fully exclude the possibility of under- (misclassification) or over-reporting (double counting secondary to inter-hospital transfers) of ACS admissions or procedures; systematic under-reporting is, however, unlikely, given the funding implications of these coding practices. Detailed interrogation of the system would require documentation of patient-level characteristics and care, combined with an evaluation of the health service infrastructure beyond the availability of catheter laboratories, extending to other modalities of cardiac investigation, such as computed tomography coronary angiography and functional imaging. Such information is not currently available in Australia.

Conclusion

Significant variation in providing coronary angiography, not related to clinical need, is evident across Australia. A greater focus on clinical care standards and better distribution of health services will be required if these disparities are to be reduced.

Box 1 –
Socio-economic and health service characteristics, and age- and sex-standardised rates of death, diagnosis, and coronary procedures (per 100 000 population), by Medicare Locals stratified according to metropolitan, regional and rural locations, for the calendar year 2011

	Total	Metropolitan	Regional	Rural	P

Number	61	27	24	10
Socio-economic indicators
SEIFA score, mean (SD)	992 (42)	1022 (39)	976 (21)	955 (33)	0.001
Indigenous population, mean (SD)	3.9% (5.5)	1.1% (0.7)	3.1% (2.1)	13.2% (8.1)	0.001
Long term unemployed, mean (SD)	3.5% (1.4)	2.6% (1.1)	4.1% (1.0)	4.6% (1.5)	0.001
Private insurance, mean (SD)	44.3% (9.9)	51.7% (9.3)	40.8% (4.3)	33.0% (5.2)	0.001
Chronic health status indicators
Diabetes, mean (SD)	5.3% (1.0)	5.7% (1.1)	4.8% (0.7)	5.5% (0.9)	0.004
Hypertension, mean (SD)	10.2% (0.6)	10.2% (0.6)	10.3% (0.6)	10.1% (0.5)	0.994
Smokers, mean (SD)*	19.1% (3.6)	16.2% (2.9)	21.4% (1.8)	22.8% (1.5)	0.001
Obesity, mean (SD)*	28.3% (4.2)	25.5% (4.4)	30.6% (1.9)	31.4% (2.1)	0.001
Hypercholesterolaemia, mean (SD)	33.1% (1.7)	32.9% (1.4)	33.6% (1.8)	32.1% (2.2)	0.186
Chronic cardiovascular condition, mean rate (SD)*	88 (14)	81 (11)	91 (9)	102 (21)	< 0.001
Premature ischaemic heart disease, mean rate (SD)	28.4 (10.2)	22.9 (5.1)	27.6 (3.4)	45.3 (13.3)	< 0.001
Access and health workforce indicators
Delay in medical consultation because of cost, mean (SD)	14.6% (3.6)	13.2% (3.8)	15.3% (2.9)	16.8% (3.1)	0.006
Primary care physicians, mean rate (SD)	110.8 (17.4)	113.9 (22.1)	109.1 (10.8)	106.3 (15.8)	0.776
Specialist physicians, mean rate (SD)*	22.9 (21.6)	34.2 (26.7)	13.4 (6.5)	10.8 (6.4)	0.002
Health service provision indicators
Primary care health check, mean rate (SD)	4266 (1181)	4336 (1034)	4548 (1239)	3401 (1111)	0.032
Public cardiac admissions, mean rate (SD)	1684 (451)	1366 (274)	1826 (316)	2201 (479)	< 0.001
Private cardiac admissions, mean rate (SD)^†	752 (264)	861 (202)	717 (283)	527 (227)	0.021
All emergency department presentations, mean rate (SD)	30 881 (11 358)	24 939 (11 358)	32 400 (9837)	43 277 (17 082)	< 0.001
Likelihood of angiogram in suspected acute coronary syndrome, mean (SD)	40.6% (16.7)	49.2% (17.2)	30.9% (8.9)	41.4% (18.2)	< 0.001
Coronary events and procedures
Myocardial infarction, mean rate (SD)	250 (63)	225 (48)	250 (46)	316 (90)	0.003
Acute coronary syndrome, mean rate (SD)	419 (108)	375 (93)	423 (84)	532 (120)	< 0.001
Coronary angiography, mean rate (SD)	849 (236)	803 (167)	895 (300)	863 (218)	0.742
Percutaneous coronary intervention, mean rate (SD)	212 (47)	222 (38)	215 (58)	178 (23)	0.009
Coronary artery bypass surgery, mean rate (SD)	70 (15)	64 (13)	74 (15)	75 (18)	0.040
Revascularisation, mean rate (SD)	278 (49)	284 (41)	283 (61)	250 (25)	0.089
Premature ischaemic heart disease deaths, mean rate (SD)^‡	28.4 (10.2)	22.9 (5.1)	27.6 (3.4)	45.3 (13.3)	< 0.001
Premature cerebrovascular accident deaths, mean rate (SD)^‡	9.2 (2.4)	8.2 (1.6)	9.4 (1.7)	11.5 (3.7)	< 0.001
Total mortality, mean rate (SD)	88 (14)	81 (11)	91 (9)	102 (21)	< 0.001
Population, mean (SD)	296 666 (165 595)	414 767 (144 115)	232 023 (110 353)	132 939 (94 442)	< 0.001

SEIFA = Socio-Economic Indexes for Areas. * Estimates from modelled data: not available for three Medicare Locals (all rural). † Data not released for three Locals (one each for metropolitan, regional and rural). ‡ Annualised rates per 100 000 individuals for 2008–2011.

Box 2 –
Variation in angiography, revascularisation, acute coronary syndrome and mortality rates according to Socio-Economic Index for Australia score for the location of the Medicare Local*

SEIFA = Socio-Economic Indexes for Areas score; a higher SEIFA value indicates lesser disadvantage. * The size of the symbol is proportional to the size of the population served by the Medicare Local.

Box 3 –
Correlations between indicators of socio-economic status, health status, health workforce, health care access and clinical practice, and coronary angiography, acute coronary syndrome and mortality rates in Medicare Locals*

Coronary angiography rate

Acute coronary syndrome admission rate

Total mortality rate

Socio-economic indicators

SEIFA score

–0.11 (0.42)

–0.62 (< 0.001)

–0.54 (< 0.001)

Indigenous population^†

–0.08 (0.53)

0.53 (0.002)

0.30 (0.019)

Long term unemployed^†

–0.07 (0.62)

0.60 (< 0.001)

0.46 (0.002)

Private insurance^†

–0.15 (0.24)

–0.65 (< 0.001)

–0.62 (< 0.001)

Chronic health status indicators

Diabetes^†

–0.22 (0.10)

–0.05 (0.72)

0.001 (0.94)

Hypertension^†

–0.27 (0.03)

–0.17 (0.20)

0.16 (0.22)

Smokers^†

0.25 (0.05)

0.61 (< 0.001)

0.62 (< 0.001)

Obesity^†

0.15 (0.26)

0.51 < 0.001)

0.65 (< 0.001)

Hypercholesterolaemia^†

0.16 (0.23)

–0.39 (0.002)

–0.08 (0.54)

Chronic cardiovascular condition^†

–0.21 (0.12)

0.05 (0.70)

0.38 (0.003)

Premature ischaemic heart disease

0.13 (0.32)

0.59 (< 0.001)

0.58 (< 0.001)

Access and health workforce indicators

Delay in medical consultation because of cost^†

0.05 (0.69)

0.61 (< 0.001)

0.45 (< 0.001)

Primary care physicians

–0.07 (0.59)

–0.26 (0.04)

–0.39 (0.002)

Specialist physicians

0.12 (0.37)

–0.41 (0.002)

–0.47 (< 0.001)

Health service provision indicators

Primary care health check (45 years)

0.28 (0.03)

0.02 (0.88)

–0.12 (0.35)

Public cardiac admissions

0.30 (0.02)

0.65 (< 0.001)

0.49 (< 0.001)

Private cardiac admissions

0.44 (0.006)

–0.13 (0.32)

–0.42 (< 0.001)

Emergency presentations

0.14 (0.28)

0.47 (0.001)

0.35 (0.005)

Likelihood of angiogram in suspected acute coronary syndrome^†

0.06 (0.69)

–0.01 (0.96)

–0.40 (0.002)

SEIFA = Socio-Economic Indexes for Areas. * Expressed as correlation coefficient, with significance (P) in parentheses. † Correlation between percentage of population positive for indicator and angiography, acute coronary syndrome admissions and mortality rates.

Box 4 –
Correlation between acute coronary syndrome and mortality rates, angiography and acute coronary syndrome admissions rates, and angiography and mortality rates*

ACS = acute coronary syndrome. * The size of the symbol is proportional to the size of the population served by the Medicare Local.

Box 5 –
Revascularisation rates as proportions of angiography rates in each Medicare Local,* with fractional polynomial estimate and 95% confidence band

* The size of the symbol is proportional to the size of the population served by the Medicare Local.

Ensuring access to invasive care for all patients with acute coronary syndromes: beyond our reach?

We need to ensure that those who need care most receive it

Coronary artery disease (CAD) remains the leading cause of death and disability in Australia, with suspected acute coronary syndromes (ACS) being the most common reason for acute presentation to hospital.1 A substantial body of evidence supports the early use of invasive care — coronary angiography and, if appropriate, revascularisation (either by percutaneous coronary intervention [PCI] or coronary artery bypass grafting [CABG]) — in patients presenting with ST-elevation myocardial infarction to reduce mortality and re-infarction rates.2 Evidence and expert opinion also favour invasive management of patients with high risk, troponin-positive non-ST-elevation ACS (NSTEACS).3,4 In patients with stable CAD, there is no evidence for any benefit from invasive care if optimal medical therapy is administered.5 Access to invasive care should be in accordance with clinical need, and this is likely to be greater in populations with a higher prevalence of CAD, CAD-related deaths, and coronary risk factors.

In this issue of the MJA, a large ecological study encompassing the entire population of 61 (former) Medicare Locals and using information from several databases identified associations between socio-economic, geographic and chronic cardiovascular disease factors and ACS incidence and mortality rates.6 Chew and colleagues also examined whether rates of invasive care (coronary angiography being the key measure) were correlated with indicators of disease burden, access to care, and clinician practice.

The study had some limitations. Data were analysed and associations between different variables (presumed to be linear) defined at the population rather than at the individual level; data on private hospital admissions were missing for three of eight state or territory jurisdictions; rates of ACS, invasive care and CAD-related deaths were only adjusted for age and sex, not for other risk factors or measures of disease severity; the likelihood of a patient with suspected ACS receiving coronary angiography was based on a relatively small national snapshot audit (n = 4398) from 8 years ago;7 the timing of invasive care after the ACS event was not provided; and rates of non-invasive co-interventions for ACS were not included in the analysis.

Despite these limitations, some of the key findings are interesting. Rates of angiography and of ACS were only weakly correlated, with no correlation between ACS and PCI rates; these trends were most evident in non-metropolitan areas where CAD mortality was highest. While ACS rates varied 3.7-fold between Medicare Locals, angiography rates varied 5.3-fold. The strongest predictor of angiography being undertaken was admission as a cardiac patient to a private hospital (71 additional angiograms per 1000 admissions), despite lower rates of ACS among private patients, suggesting that many procedures were for non-ACS indications. Coronary revascularisation rates as a proportion of angiography rates varied between 17% and 61%; higher angiograms rates were associated with reduced likelihood that revascularisation followed. There was a reasonably strong positive association between rates of ACS and CABG, suggesting that less invasive PCI is more vulnerable to unwarranted use. Depressingly, the study also reconfirmed the higher ACS rates in non-metropolitan locations where the prevalence of smoking, obesity and chronic cardiovascular disease is higher.

The disparity between rates of invasive care and those of ACS and overall CAD burden probably means that some patients are receiving interventions they do not need, while, more worryingly, patients who have real need for them are missing out. Without data on clinical indications and criteria of appropriateness for individual patients, overuse cannot be distinguished from underuse. Nevertheless, such variations in invasive care, seemingly unexplained by variations in clinical indications, are of concern, especially as they have been documented since 2005.7–11

So why is universal access to invasive care according to need seemingly beyond our reach? It is not for want of trying on the part of national professional bodies that develop, disseminate and promote evidence-based recommendations and implementation frameworks.12 The answer lies with front line health care delivery systems. Cardiology service networks at the state level should collect data on ACS incidence and rates of invasive care in public and private facilities, identify locations where the mismatch is greatest, and seek to understand and mitigate the relevant factors. Networked, hub-and-spoke support systems of rapid diagnostic, referral and transfer procedures are needed, whereby patients with ACS presenting to any emergency centre have rapid access to invasive care in angiography-capable facilities, in accordance with clinical indications and socio-cultural context, and without logistical barriers.13,14 Hospitals should report on their provision of appropriate ACS care according to agreed care standards and data collection methods, for benchmarking against peers and sharing improvement strategies.15

Invasive care prevents about 10% of all CAD-related deaths, whereas medical treatments and reducing risk factors account for at least 80% of saved lives.16 As Chew and colleagues report, a higher probability of undergoing coronary angiography was associated with only a modest reduction in ACS mortality rates (three fewer deaths per 100 000 population for each 10 percentage point increase in likelihood of angiography). The maximal gain in lives saved after ACS onset will require optimisation of all care modalities along the entire patient trajectory. However, while non-invasive care is not consistently employed across Australia,7–10 differences in the use of invasive care are more marked and cannot be allowed to persist into the next decade.

Improving outcomes in coronary artery disease

Systems, procedures and policies are needed to further reduce the toll of cardiovascular disease

Although major advances have been made in many aspects of the treatment of heart disease, rates of mortality and morbidity from acute coronary syndromes (ACS) in Australia and New Zealand remain significant. The SNAPSHOT ACS study reported outcomes of patients presenting with ACS.1 The 18-month mortality for patients presenting with an ST-elevation myocardial infarction was 16.2%, 16.3% for those presenting with a non-ST-elevation myocardial infarction, and 6.8% for those with unstable angina. Although these outcomes are a substantial advance on historical data, there is still much room for improvement.2 It is therefore timely that the National Heart Foundation (NHF) and the Cardiac Society of Australia and New Zealand (CSANZ) have revised the Australian guidelines for the management of ACS (summarised in this issue of the MJA).3 The NHF, CSANZ and all involved should be congratulated on the development of these guidelines.

The revised ACS guidelines simplify the target time for myocardial reperfusion in patients presenting with an ST-elevation myocardial infarction. Primary percutaneous coronary intervention (PCI) should be performed within 120 minutes of first medical contact or within 90 minutes of presentation to a PCI-enabled hospital. Otherwise thrombolysis should be administered unless there are contraindications. Patients’ outcomes are maximised when the duration of myocardial ischaemia (ie, time from symptom onset to reperfusion) is minimised, with optimal outcomes occurring when this time is less than 120 minutes.4 Protocols to expedite thrombolysis or PCI are required to minimise treatment delays.5 Continuing public education programs are needed to raise awareness of the symptoms of a heart attack. Such campaigns have been shown to reduce Australian patients’ time to presentation.6 Pre-hospital electrocardiograms recorded by the ambulance service further reduce delays, and pre-hospital thrombolysis programs, especially for patients in remote locations, can significantly reduce ischaemia times, as reported by Khan and colleagues in this issue of the Journal.7 Additionally, the revised guidelines provide clinical assessment protocols to expedite troponin testing using highly sensitive assays to rule out myocardial infarction within 2 hours of presentation. This will particularly benefit compliance with the 4-hour National Emergency Access Target for presentations to emergency departments.8

In this issue of the Journal, May and colleagues indirectly address the appropriate use of coronary stenting (PCI) in patients with stable coronary artery disease.9 The DEFER study confirmed that subjective operator assessment of the severity of intermediate coronary artery lesions is a poor predictor of myocardial ischaemia as assessed by fractional flow reserve (FFR), an index of reduced myocardial perfusion.10 The FAME 2 trial demonstrated that the benefit of PCI in stable patients is confined to stenting lesions with an FFR < 0.80.11 FFR was used in 4.8% of coronary angiograms funded by Medicare in 2015. It is not known how many of these patients were investigated for stable coronary disease or whether these patients underwent pre-procedural stress testing as an alternative to FFR. PCI should be carefully targeted in stable coronary artery disease; it is of benefit when there is limiting angina due to a proximal coronary artery stenosis > 50%, a positive stress test demonstrating ischaemia in at least 10% of the myocardium, or an FFR < 0.80.12

Following a cardiac event, secondary prevention strategies improve outcomes.13 Secondary prevention is traditionally overseen in a cardiac rehabilitation program. Only a minority of patients attend such programs, and a significant number do not continue with guideline-recommended therapy and lifestyle modifications.14 New technologies including video conferencing and various apps may provide a mechanism to extend the reach of cardiac rehabilitation, as Gallagher and Neubeck note in this issue of the Journal.15 These technologies may enhance or replace traditional services. It is yet to be demonstrated that they will deliver on their promise. Strategies that enhance compliance with the Australian Commission on Safety and Quality in Health Care ACS standard are required.16

There is a need to enhance quality assurance by establishing registries to assess patient outcomes. The SNAPSHOT ACS study provided valuable insights into the nationwide treatment of ACS patients, and allowed the identification of treatment gaps and compliance with guidelines and standards. Such information is required to assess the impact of various treatment strategies and health care policies. The CSANZ has established the Australasian Cardiac Outcomes Registry (http://www.acor.net.au). Sustainable funding models need to be established to support the collection of risk-adjusted patient, procedural and device outcomes. It is incumbent on the cardiac community to ensure that systems, procedures and policies are developed to further reduce the toll of cardiovascular disease in an environment where outcomes are monitored and continuously assessed.

Cardiac tamponade in undiagnosed systemic lupus erythematosus

A 22-year-old woman presented with a 3-day history of fever, retrosternal chest pain and exertional dyspnoea. Her heart rate was 130 bpm with a blood pressure level of 109/68 mmHg. Physical examination suggested tamponade: distended jugular veins, pulsus paradoxus and muffled heart tones. The chest radiography was notable for the characteristic water-bottle sign (Figure, A).1 Contrast-enhanced chest computed tomography demonstrated a massive pericardial effusion (Figure, B) associated with venous engorgement of the superior and inferior vena cava (SVC, IVC), prevascular space (arrows), and bilateral axillary veins (arrowheads). An emergency thoracoscopic pericardial window was performed and 620 mL of bloody fluid was drained.

The presence of anti-nuclear, anti-double-stranded DNA, anti-Smith antibodies and hypocomplementaemia supported the diagnosis of systemic lupus erythematosus.2 The patient recovered after 1 week of intravenous methylprednisolone pulse therapy. At an 8-month follow-up, there have been no recurrences.

Figure

How health technology helps promote cardiovascular health outcomes

Health technology in the hands of cardiac patients — helpful, hindrance, or hesitate to say?

Cardiovascular disease (CVD) is a leading cause of death and hospital admissions in Australia.1 Almost a third of patients will have a recurrence such as myocardial infarction (MI), stroke, heart failure and death within 5 years.2 Reductions of at least 80% in these events can be achieved through secondary prevention behaviours, including taking cardioprotective medications as prescribed, ceasing smoking, increasing physical activity and consuming a healthy diet.3 The most comprehensive and proven of strategies to support secondary prevention is cardiac rehabilitation. This is an evidence-based, cost-effective method to reduce CVD deaths, assist recovery and promote secondary prevention through reduction of cardiovascular risk factors.

The strength of secondary prevention success in cardiac rehabilitation ultimately depends on patients’ awareness, willingness and capacity to make lifestyle changes and to engage in the required behaviours. Patients must have a strong commitment to their health to sustain secondary prevention behaviours for the rest of their lives, often without direct evidence of benefit, as CVD is asymptomatic in the majority of cases.3 However, cardiac rehabilitation is widely underutilised, with less than a third of eligible patients attending the sessions and dropout rates estimated at 25%. In the absence of support, many patients struggle with sustaining the requisite behaviours.

Medications are a prime example of this struggle. Less than two-thirds of patients are reported to persist with all prescribed medications by 1 year following MI.4 The reasons given for discontinuation are largely self-determined, emphasising the importance of the patients’ understanding and engagement.5 Similar issues are present for most secondary prevention behaviours. At 6 months after MI, 27% of patients smoked, 26% consumed an unhealthy diet and 59% did not exercise enough.3 The key reasons identified for patients’ struggles are a lack of awareness that treatments must be long term to achieve effective prevention, and forgetting to follow recommendations.5 Patient education is necessary, but specialised support is often required for sustained behaviour change. Technology, particularly mobile technology, offers a solution for support for long term behaviour change and may also augment existing secondary prevention programs, such as cardiac rehabilitation.

Adoption of mobile technologies, such as mobile phones, smart phones and tablets that provide internet access, has been widespread in Australia.6 Rapid growth in popularity and technological advances have also occurred in wearable devices for tracking behaviours, such as fitness activities, that connect with mobile phones. These technologies and related applications can efficiently enable long term support for patients and provide memory prompts for behaviours. However, there has been such rapid evolution of technologies that we ask: are health technologies helpful, a hindrance, or should we hesitate to say?

Are health technologies helpful?

The evidence indicates that well designed and often simple technologies can improve patient outcomes in multiple cardiovascular risk factors.7 The most consistent evidence of benefit from health technologies is for coronary heart disease compared with other cardiovascular diseases.8 Simple short-message service (SMS) interventions delivered regularly can improve awareness and prompt actions, which may otherwise be forgotten. Text messages double the odds of adhering to medications and the Australian TEXT ME program has improved cardiovascular disease risk factor profiles.9 Emerging evidence indicates that many more risk factors may also be addressed through text messages and mobile phone apps, and potentially decrease dropout from cardiac rehabilitation.8,10 In addition, wearable activity trackers provide reliable information on steps and time spent in moderate to vigorous physical activity, which can be monitored, used for goal-setting and shared with treating doctors.11

There may be a temptation to assume that health technology is not applicable to cardiac patients because of their older age and lack of experience; however, such patients may be more ready and willing to use health technology than previously suspected. At least two-thirds of cardiac patients (67%) use the internet and at least half (50%) use mobile technologies for health, with higher rates of acceptance of health technology in cardiac rehabilitation participants (74%).12

Are health technologies a hindrance?

Several key issues hinder uptake and complicate recommendations for patient-facing technology in regular clinical practice. There is a proliferation of publicly available health technologies suitable for cardiac patients, particularly mobile phone apps, alongside a paucity of research-based evidence to support selection.7 Quality too is generally low, and popularity is a poor indicator. For instance, an evaluation of patient engagement, quality and safety of 1046 health care-related, patient-facing apps for chronic disease found that only 43% (iOS, Apple) and 27% (Android, Google) were actually likely to be useful.13 A recent review of mobile phone interventions for secondary prevention of cardiovascular disease did not identify any intervention that resulted in a negative impact, but six of the 28 interventions reviewed had no benefit.8 Owing to the poor quality of the studies included, it is difficult to distinguish the features that independently predict success; however, the use of text messaging, telemonitoring and interaction may have been important. In a similar way, popular weight loss apps do not incorporate successful behaviour change techniques and many have inaccurate content. The motivation of app developers must be carefully considered given the rise of pro-smoking apps disguised as educational games. At best, patients may experience no benefits from using health technology, and at worst, patients may question and disregard credible advice given by health professionals. Patients have few sources of advice for selection and use of health technologies, as physicians, like patients, have varying capacity and interest in using them. Previous experience, education, confidence and interaction with early adopters to understand how to interface with the health app and with the patient using the app are all required for success.14 Patients who are older, have little or no prior experience with technology and who have lower levels of education are less likely to be interested and have lower expectations of success or benefit from using technology for their health.12 The same could be argued for health professionals. This is important because, for health technologies that do prove useful, positive staff attitudes may offset the lower levels of expertise or interest present in older people and in those with lower education.

Should we hesitate to say?

The rapid advances in health technology, the mass of publicly available apps and programs, and the inability of research to keep pace are major factors affecting our ability to definitively determine whether health technologies are a help or a hindrance. These factors make it difficult to identify the technologies that are evidence-based and free from unintended negative consequences.15 However, there are a few shortcuts. For instance, there is no single repository for patient-based health technologies for cardiovascular secondary prevention, nor is there an accrediting or regulatory system for quality or effectiveness. There has been some attempt to catalogue apps, but the process is slow and apps are often modified by the time of publication. Regulation of health technologies, such as mobile phone apps, is poorly developed. A balance is needed between two polar possibilities: government regulation, as used for implanted medical devices, which is rigorous but slow and expensive; and self-regulation by technology companies through checklist certification, which is dependent on the company ethics.16 App stores such as those of Apple and Google currently vet apps sold in their stores, but the emphasis is on security, not on the quality or credibility of the app.17 Ultimately consumers must decide. One shortcut to ensuring credible content is to only use or recommend health technologies and health apps developed by reliable institutions, such as the Heart Foundations of Australia, Britain and Canada. However, the gains in credibility and time-saving from using this shortcut occur at a trade-off in diversity, innovation and personalisation for patients.

A principles-based approach may be more appropriate and feasible to discriminate the features of technology that encourage consumer uptake and promote appropriate behaviour change effectively. These principles have been highlighted for apps for the secondary prevention of CVD in a recent review12 and are relevant to most health technologies. Key elements to consider include simplicity, credibility of content, behaviour change components, real-time tracking, reward system, personalisation, social features and attention to privacy of data collected. A positive, but discerning attitude to patient-facing health technology is required, much the same as with any new health treatments, to maximise patient outcomes and ensure equitable access. Regular screening to identify effective health technology has become a necessary component of health professional life, equivalent to keeping up to date with effective medications and techniques. Evidence-based practice was introduced to medical curricula to help ensure the most effective treatment for an individual patient. Likewise, it is now time to introduce screening and selection systems for health technologies into health professional education.

In summary, the evidence suggests that mobile technologies will improve access to and participation in secondary prevention activities, but careful consideration is needed to ensure that the most effective and acceptable technologies are incorporated into patient care.

“Congenital heart health”: how psychological care can make a difference

An integrated approach incorporating both physical and mental health is critical to “congenital heart health”

Congenital heart disease (CHD) affects more than 2400 Australian babies each year. It is the most common cause of admission to paediatric intensive care in the neonatal period,1 a leading cause of infant death2 and one of the leading causes of disease-related disability in children under 5 years of age.3 In Australia and around the world, the landscape of CHD care is rapidly evolving. With advances in medicine, survival has markedly improved over the past two decades4 and the best estimates suggest that the total population, from newborns through to adults living with CHD, now represents well over 65 000 Australians.5 These gains in survival are a triumph, but paradoxically, they bring new challenges. Earlier diagnosis, more complex treatment choices, longer survivorship and a need for transition from paediatric to adult cardiac services lead us into new territory. Embedded in each of these challenges are a range of psychological complexities, foremost of which is how to best support the wellbeing of people with CHD across a lifetime. “Congenital heart health” requires an integrated, life course approach — with equal emphasis on the physical and the psychological — beginning before birth, through infancy to adulthood. This article presents the evidence for such an approach, highlighting the areas where evidence is lacking and calling for national standards of mental health care in CHD.

Beginning before birth: the mental health of expectant parents

Fetal diagnosis of CHD has changed clinical care. Half of all babies who need heart surgery in the first year of life are now diagnosed while in the womb.6 A major advantage of fetal diagnosis is that potentially unstable newborns can be delivered close to specialised paediatric cardiac services, reducing morbidity and improving survival.7,8 Fetal diagnosis also facilitates early family counselling regarding the nature of the baby’s heart abnormality and expected prognosis. But when a parent hears words such as “your baby has tetralogy of Fallot” their world stops. Fetal diagnosis of any abnormality, major or minor, precipitates an emotional crisis for expectant parents. Threats to the health of the fetus have long been recognised as an important risk factor for psychological disturbance during pregnancy, which in turn indicates a high risk of ongoing psychiatric disorders postpartum. Research has shown that one in three mothers and fathers of infants with complex CHD report symptoms that meet clinical criteria for depression, and about 50% of parents report severe stress reactions consistent with a need for clinical care up to 1 year after their baby’s diagnosis.9 These rates far exceed documented rates of perinatal depression and anxiety in the general community10 and as health care providers, we often markedly underestimate the severity and potential consequences of these symptoms.

There is now overwhelming evidence that depression and anxiety in the perinatal period (from conception to the end of the first postnatal year) can have devastating consequences, not only for the person experiencing it, but also for his or her children and family. Parents with high distress report poorer physical health,11 greater parenting burden,12 higher health service use13 and more suicidal ideation14 compared with parents of sick children with lower distress. At least 14 independent prospective studies have demonstrated a link between maternal antenatal stress and child neurocognitive, behavioural and emotional outcomes.15 Acknowledging these findings is not to blame parents — not in any way, not at all. But to not talk about this, to not address the significant emotional toll of CHD, is to not provide optimal care.

Early life experiences can have profound consequences

Diagnosis of a life-threatening illness during a child’s formative years can have far-reaching effects that ripple through the family and across a lifetime. Infants with complex CHD experience a range of uncommon and painful events, such as separation from their mother at birth, urgent transfer to specialised intensive care, frequent invasive medical procedures, feeding difficulties and withdrawal from narcotic pain relief. These early life experiences can have profound consequences for the developing child, shaping brain development, the body’s immune system and responses to stress. Studies of individuals exposed to high levels of stress early in life consistently show that the experience of early adversity is associated with disrupted child–parent attachment and with alterations in the developmental trajectories of networks in the brain associated with emotion and cognition.16

From a neurodevelopmental perspective, children with CHD experience greater difficulties compared with their healthy peers. The risk and severity of neurodevelopmental impairment increases with greater CHD complexity, the presence of a genetic disorder or syndrome, and greater psychological stress. During infancy, the most pronounced difficulties occur in motor functioning. By early childhood, studies show that children with complex CHD have an increased risk of neurodevelopmental impairment, characterised by difficulties in fine and gross motor skills, speech and language, attention, executive functioning, emotion regulation and behaviour.17 Emotionally, 15–25% of parents report significant internalising (eg, anxiety, depression, somatisation) or externalising (eg, aggression) difficulties in their child.18 Few studies have used validated clinical assessments, but those studies that have included a clinical assessment report psychiatric illness in more than 20% of adolescents and young adults with CHD. There is also a diagnostic rate of attention deficit and hyperactivity disorder up to four times higher than in the general population.18

Navigating the transitions of adolescence

Young people must not only navigate the typical transitions of adolescence and early adulthood, they also face the challenges that come with developing a sense of identity and autonomy in the context of CHD while, at the same time, breaking the bonds formed with their paediatric team and transitioning to adult cardiovascular care. Clinicians lose their “clinical hold” on so many patients during this process, with an alarmingly high proportion of adults with CHD (24–61%) not receiving the recommended cardiac care.5,19,20 This can have catastrophic health consequences,5 yet our understanding of the factors that influence transition is wanting. In 2013, the Cardiac Society of Australia and New Zealand published a statement outlining best practices in managing the transition from paediatric to adult CHD care.21 This statement recognises the role that psychological factors play in shaping young people’s capacity to manage their heart health care, in addition to clinical, familial and practical factors.

Living with CHD into adulthood

Our understanding of the psychology of adult CHD lags decades behind our knowledge of children’s experiences. While neurodevelopmental clinics have been established at several paediatric cardiac centres across Australia, little attention is paid to neurocognitive health in adult CHD care. Without effective intervention, hardships encountered during infancy and childhood can persist for years after diagnosis and treatment. It is also possible for difficulties to emerge for the first time in adulthood, with heart failure, atrial fibrillation, cardiac surgery and recurrent strokes increasing vulnerability to neurocognitive impairment in adults with CHD. A meta-analysis of 22 survey-based studies of emotional functioning in adolescents and adults with CHD found no differences from the norm;22 however, studies using clinical interviews have found that one in three adults with CHD report symptoms of anxiety or depression warranting intervention.23 The vast majority of these adults go untreated.

An integrated psychology service dedicated to CHD

In 2010, we established an integrated clinical psychology service dedicated to CHD. Located across the Sydney Children’s Hospitals Network, we provide a statewide mental health service for children and young people with all forms of heart disease and their families (Box). Evidence-based, psychologically informed care and a variety of interventions can prevent or relieve psychological morbidity.24 National (beyondblue, 2011) and international (International Marcé Society, 2013) authorities now recognise and endorse psychosocial assessment for depression and anxiety in the perinatal period as part of routine clinical care. Integrating psychosocial assessment within existing health services has been shown to improve treatment uptake, as well as mental health and parenting outcomes, up to at least 18 months postpartum.24 Indeed, integrating psychosocial assessment within a clinical setting with which families are already engaged is a key factor distinguishing successful and unsuccessful early mental health interventions. Providing tailored skills-based training and support to staff also has the potential to overcome a range of patient (eg, stigma, cost), provider (eg, limited awareness, low confidence), and health system (eg, cost, under-prioritisation) factors that can prevent access to care. Moreover, there are costs to not offering mental health care. In neonatal intensive care, for example, targeted interventions significantly reduce maternal depression and anxiety, increase positive parent–infant interactions, and reduce health service costs by reducing the length of stay at the neonatal intensive care unit and total hospital admission.25

A call for national standards of mental health care in CHD

While there may be deep sadness associated with many aspects of CHD, there may also be deep hope and, for many, good health. Early intervention can make a profound difference for families and influence a lifetime of outcomes for a child. To transition CHD to “congenital heart health”, equal emphasis must be placed on both physical and mental health so that the successes in physical care are not undermined by the absence of adequate emotional care. Across the CHD continuum, there are still huge gaps in access to mental health services. To start to address these gaps, we call for the formation of a multidisciplinary working group to develop national standards of mental health care in CHD. We also advocate strongly for more Australian research across the discovery-to-translation pipeline such that, from early life in the womb through to adulthood, we can better understand — and prevent or assuage — the difficulties that our patients and their families experience.

Box –
Integrated model of psychological care dedicated to congenital heart disease and based at the Heart Centre for Children

National Institute for Health and Care Excellence. Antenatal and postnatal mental health: clinical management and service guidance. NICE clinical guidelines, no. 192 (updated edition). London: NICE, 2014.

Time to bury “hypertension”

An absolute cardiovascular risk approach will better target patients who need pharmacotherapy

The publication of the Systolic Blood Pressure Intervention Trial (SPRINT), sponsored by the United States National Institutes of Health, has left physicians claiming that systolic blood pressure targets of < 120 mmHg are too low and unobtainable, and that the results are not generalisable to their “real” patients.1 Although it was ostensibly a “hypertension optimal treatment” trial, it was also, in effect, a quasi-trial of the treatment for elevated blood pressure in high risk individuals who would otherwise remain untreated. This can be argued because the study population were all high risk patients determined by age, clinical conditions or Framingham risk score, and the entry-level systolic blood pressure was 130 mmHg rather than the normal treatment threshold of 140 mmHg. The low entry level, with a mean systolic blood pressure of 139.7 mmHg, probably explains the low targets achieved in the intensive treatment group. The study demonstrated not only that the reduction of systolic blood pressure leads to benefits in decreasing the rates of all-cause mortality and cardiovascular morbidity and mortality, but also that this reduction could be achieved with relative safety, even for older patients, as there was no overall difference in serious adverse event rates between the intensive treatment group and the standard treatment group. This conclusion had also been arrived at in the Hypertension in the Very Elderly Trial, a randomised controlled study of blood pressure lowering in very old patients, where serious adverse events were actually lower in the treatment group compared with the placebo group.2 The accepted wisdom of “it only takes one broken hip to wipe out all that gain in cardiovascular risk” does not seem to hold true.3

The predictable criticism of the findings reflects the entrenched clinical concept of hypertension — that there is a magic figure above which you have the condition and below which you do not. SPRINT reinforces that lowering blood pressure to at least 120 mmHg may be beneficial for a high risk individual as no J-curve nadir was demonstrated. It is opportune to return elevated blood pressure to its continuous variable risk factor status rather than treat it as a dichotomous disease. After all, the very term “hypertension” is confusing to patients.4

Who should we treat with blood pressure-lowering drugs?

Initiation of pharmacotherapy should be reserved for those who will probably benefit in terms of preventing a major adverse cardiovascular event, and when this benefit clearly outweighs the potential harms of side effects and costs of treatment. Candidates are therefore those at moderate to high risk of such events in what is an asymptomatic condition. A simple algorithm populated by the most important determinants of cardiovascular disease risk is sufficient to identify the individuals who do not have a manifest disease. This is readily accessible with the Australian absolute cardiovascular risk calculator.5 Such an approach recognises that drug therapy should be considered in the context of the whole person, while acknowledging that action on risk stratification can be challenging and complex for many.

Implementing the absolute cardiovascular risk factor approach in Australian health care

Australian clinical practice has not yet widely adopted the absolute cardiovascular risk factor approach,6 despite the development of evidence-based guidelines by the National Vascular Diseases Prevention Alliance (NVDPA). These guidelines are the result of a collaboration of four peak bodies — Kidney Health Australia, the National Heart Foundation, the National Stroke Foundation and Diabetes Australia — and have the endorsement of the National Health and Medical Research Council (NHMRC).7,8 The NHMRC has also recently adopted the absolute cardiovascular risk approach as one of its priority cases for research translation.9

With such an approach, the recommended pharmacotherapy regimen will need to be changed for high risk “normotensive” patients and for low risk “hypertensive” patients. The first group comprises individuals who are at a high risk due to a clinical manifestation of cardiovascular disease or clustering of risk factors, but whose blood pressure has not crossed the 140/90 mmHg threshold. The National Prescribing Service, as part of its MedicineWise program, and the NVDPA have addressed this group through educational programs which use case vignettes of the unexpected fatal myocardial infarction of a late middle-aged male smoker who did not have hypertension or hypercholesterolaemia (http://www.nps.org.au/media-centre/media-releases/repository/latest-program-from-nps-medicinewise-targets-blood-pressure). SPRINT reinforces the benefits of treatment in this group.

The second group comprises those (often younger) patients with mildly elevated blood pressure, who are low risk but hypertensive and for whom drug treatment is not recommended. For this group, there is a concern that such individuals may be harmed due to a delay or absence of treatment, allowing irreversible pathological damage to occur, that is, to accrue adverse legacy effects. While research in this area is ongoing and has yet to demonstrate such effects, it does contrast with another clinical concern, that of overdiagnosis.10,11 Diagnosing an individual with a medical condition has adverse effects for those who have an asymptomatic condition where intermediate benefit is very unlikely; this also has opportunity costs to society because of the misdirection of limited resources. Such individuals do not remain untreated, just unmedicated, as attention is paid to adverse health behaviours, which, when addressed, have benefits beyond the cardiovascular system. In practice, such individuals are likely to delay rather than avoid drug therapy, as age is the most important determinant of risk; however, their years on therapy will be truncated without affecting their lifespan and quality of life. Evidence suggests that reassessment of risk is not required for most low to moderate risk individuals within 8–10 years of the diagnosis, with the exception of those close to treatment thresholds, for whom annual review is recommended.12

Given the limited uptake of the absolute risk approach to date, how can we encourage its increased use? One possible way is through the Pharmaceutical Benefits Scheme (PBS). The current PBS indication for blood pressure-lowering agents is hypertension, rather than a specified blood pressure threshold. Hence, prescribing would seem to be unimpeded by the absolute risk approach, given that such a definition is likely to be deferred to expert guidelines. Cholesterol-lowering agents, on the other hand, have a complex set of criteria for eligible prescribing on the scheme that are not solely based on a single serum cholesterol threshold (http://www.pbs.gov.au/info/healthpro/explanatory-notes/gs-lipid-lowering-drugs). To advance cardiovascular health care in Australia, the NHMRC Primary Health Care Steering Group recognised that uptake of the absolute risk approach could be enhanced by changing PBS criteria for statins from these criteria to one based more simply on an absolute risk threshold.9 To this end, the NHMRC is asking the Pharmaceutical Benefits Advisory Committee to consider aligning their prescribing conditions to the NHMRC-approved absolute cardiovascular disease risk guidelines.8 This would mean that all physicians would need to become familiar with the Australian cardiovascular risk calculator in order to access statins for their primary prevention patients. Once habituated, they may be more willing and able to apply it in the setting of treating elevated blood pressure.

With benefit demonstrated at lower thresholds and to lower targets, there is a greater imperative to move away from the hypertensive model of care as these thresholds and targets approach the ideal blood pressure of 115 mmHg,13 which would capture most of the population. Taking the absolute risk route will, on the other hand, target those who have a covert cardiovascular disease most likely to manifest clinically in the foreseeable future and, therefore, benefit from pharmacotherapy.

Relieving the pressure: new Australian hypertension guideline

The National Heart Foundation guideline has been updated to reflect recent evidence and Australian conditions

Some might think there are too many guidelines on hypertension.1–4 However, given that hypertension is a major risk factor for premature death and disability from cardiovascular disease in Australia and globally,5,6 practitioners need a practical, contemporary and localised guide to best practice. Most countries and their hypertension societies publish their own guidelines. However, these differ for a number of reasons. The data available at the particular time of publication vary. Experts interpret these data differently. To varying extents, guideline development is subject to vested interests, such as governments and other funders wishing to keep down costs, or industry wishing to make treatments available to the widest groups possible.7 The scope of the review can vary from people with uncomplicated hypertension to those with a broad range of complications and comorbidities.

In Australia, we are fortunate that the National Heart Foundation has produced an excellent new guideline (http://heartfoundation.org.au/for-professionals/clinical-information/hypertension) adapted for Australian conditions at a time when knowledge of the field has been moving rapidly.8

One of the first questions concerns who is at risk from hypertension and who warrants active treatment. For some time now, Australian recommendations have followed an absolute risk approach, rather than determining risk on blood pressure alone. The recommended way of assessing this is according to the national guideline on absolute risk developed by the National Vascular Disease Prevention Alliance.9 This predicts risk of a major cardiovascular event or death in the next 5 years based on a modified Framingham equation. Unless blood pressure is very high, it is a better way of identifying someone who will benefit from treatment than blood pressure alone as a single risk factor. It also helps deal with the well known epidemiological paradox that most people who have heart attacks or strokes caused by elevated blood pressure do not meet the conventional definition of hypertension. Systolic blood pressure in the high normal range (130/85–139/89 mmHg) carries risk and this includes a substantial proportion of the population. This applies equally to other risk factors such as blood glucose or lipids. Levels below arbitrary cut-points capture only some of the relevant risk.

However, risk assessed this way is very much influenced by age and is not applicable in young adults (aged under 45 years, or under 35 years in Indigenous Australians) or older adults (aged over 74 years).9

Recent hypertension trials and other guidelines have used several methods of assessing risk, including the presence or absence of target organ damage, clinical indicators such as abdominal girth, tobacco use, low levels of high density lipoprotein.10 What can be concluded is that there are many ways of assessing risk and, if risk is high, people benefit from modification of standard risk factors including blood pressure.

The 2016 guideline offers advice on new areas including out-of-clinic blood pressure measurement using ambulatory or home procedures, white coat hypertension and blood pressure variability. It includes updated evidence on the management of hypertension with comorbidities including chronic kidney disease, diabetes and peripheral arterial disease. There are minor updates to recommendations on first-line and combination pharmacotherapies.

A key area of debate surrounds the revision of treatment targets based on new evidence for a target blood pressure of < 120 mmHg in patients at high risk for cardiovascular events but without diabetes.11 Where supported by evidence, a target of < 140 mmHg is recommended. In other groups where there is supporting evidence, the recommendation is to aim for < 120 mmHg but be cautious about adverse events, especially in older, frail people. The debate is fuelled by the lack of data in particular patient groups but increasingly the clinical trial data are aligning with the epidemiology, with lower blood pressure being associated with better outcomes across the spectrum.12

The holes in the evidence base and shifting ground as new evidence comes forward are not helpful to the individual clinician who wants advice on what to do for a particular patient on a particular day. Further, we know that despite the proliferation of guidelines, most people with hypertension do not achieve the goals of their therapy, irrespective of what country they live in and what guideline is being followed. In future, we need to move the emphasis from large tomes written by expert groups to providing decision support individualised to the patient.

The new guideline also covers emerging evidence on diagnostic and therapeutic aspects of hypertension, including the uncertain role of renal denervation. It will guide management of hypertension in Australia for the immediate future. I commend it to readers of the MJA and thank all those who contributed to its preparation.

Detecting ascites

Most cases can be diagnosed by good clinical assessment at the bedside

The presence of ascites is a common physical finding and the detection of ascites is important for both diagnostic and prognostic reasons. Ascites is defined as the pathological accumulation of fluid in the peritoneal cavity.1 It may be due to a number of causes (Box 1). The most common is portal hypertension as a result of cirrhosis (> 75%) but malignancy (10%), heart failure (3%) and infection (2%) are other possibilities.1

Patients with ascites usually present to clinicians with increasing abdominal distension, weight gain and discomfort. However, ascites may be detected incidentally in patients developing other complications of their cirrhosis, such as variceal haemorrhage and encephalopathy. Moreover, a patient may present when their underlying heart failure or malignancy progresses.

Initial assessment of the patient will involve taking a history to determine the risk of liver disease. This will include questions about alcohol consumption and risk factors for chronic hepatitis, especially hepatitis C. Cardiac symptoms (shortness of breath and orthopnoea) can be assessed and patients should also be asked about symptoms that might indicate an underlying malignancy, such as weight loss and decreased appetite. Patients who do not complain of ankle swelling or abdominal distension are unlikely to have significant ascites.2

The examination of a patient with ascites includes two main components: inspection and palpation.

General inspection of the patient should include looking for signs of chronic liver disease, such as jaundice, spider naevi, palmar erythema, gynaecomastia and loss of body hair. Prominent collateral veins in the abdominal wall may also be present. Bulging flanks (Box 2) may be due to ascites or obesity, but the absence of this sign makes the presence of ascites unlikely.2

Palpation of significant hepatomegaly and splenomegaly may help the diagnostic process, but can be difficult to perform in a patient with a large volume of ascites or tenderness. With the patient lying supine with their head on a single pillow, there will be a tympanic area to percussion in the midline of the abdomen, which will normally be bordered by an area of dullness in either flank (Box 3). To assess this, the clinician should place their hand in the midline, parallel to the direction of the expected change in resonance, and percuss away from themselves until the percussion note changes from tympanic to dull. It is possible to repeat this manoeuvre in each direction. However, to demonstrate shifting dullness, the clinician will need to ask the patient to roll towards them, and on to the patient’s right side, while keeping their hand on the patient at the location of the dull percussion note. After waiting up to a minute for fluid to shift, the clinician can percuss once again from the tympanic area, which will now be in the flank, to the dull area, which will be in the midline. The sensitivity and specificity of this test for ascites is more than 70%.3

With a significant volume of ascites, it may be possible to elicit a fluid thrill or wave. This may require two people. The second person places the ulnar border of their forearm, or both hands, on the midline of the patient’s abdomen to prevent transmission of the thrill through the subcutaneous fat. The clinician then taps the flank firmly and feels for an impulse on the opposite side. This sign lacks sensitivity but is highly specific.3

Box 4, adapted from a review by Williams and Simel,3 summarises the significant symptoms and signs to consider when evaluating a patient with ascites, and their accuracy and precision. The absence of reported ankle swelling and abdominal distension, combined with no bulging flanks, flank or shifting dullness, are most helpful in excluding ascites. A diagnosis can be positively made when a patient has a fluid thrill together with shifting dullness and ankle swelling. Studies suggest high levels of agreement between clinicians, especially those with experience, on the detection of ascites using these techniques.2,4,5 Using ultrasound or computed tomography, the volume of ascites that can be detected is probably as small as 100 mL. However, it is unlikely that volumes of this magnitude will be detectable on clinical assessment. For flank dullness, more than 1 L of ascitic fluid needs to be present.

The gold standard for detecting ascites is aspiration of fluid after visualisation with imaging.3 However, with good clinical assessment, using the presence of a fluid thrill, shifting dullness and peripheral oedema as the positive indicators, the majority of cases can be diagnosed by a clinician at the bedside.

Box 1 –
Frequency of causes of ascites

Frequency

Cause

Very common

Cirrhosis

Common

Right-sided heart failure

Malignancy

Rare

Tuberculosis

Pancreatitis

Nephrotic syndrome

Very rare

Constrictive pericarditis

Budd–Chiari syndrome

Protein-losing enteropathy

Chylous ascites

Serositis (lupus, familial Mediterranean fever)

Box 2 –
Bulging flanks and abdominal distension secondary to ascites

Box 3 –
Tympanic area of the abdomen in the presence of ascites

Diagram courtesy of Rebecca Veysey.

Box 4 –
Significant pooled results for the accuracy of clinical history and physical examination in the detection of ascites*

Symptom or sign	Sensitivity	Specificity	Likelihood ratio^†
Symptom or sign	Sensitivity	Specificity	Positive	Negative

Abdominal distension	87%	77%	4.2	0.2
Ankle swelling	93%	66%	2.8	0.1
Bulging flanks	81%	59%	2.0	0.3
Flank dullness	84%	59%	2.0	0.3
Shifting dullness	77%	72%	2.7	0.3
Fluid thrill	62%	90%	6.0	0.4

* Adapted from Williams and Simel.3 † The likelihood that a symptom or sign would be expected if a patient had ascites: positive likelihood ratio indicates how much more likely the presence of ascites will be if the sign or symptom is present; negative likelihood ratio indicates how unlikely ascites would be if the symptom or sign were absent.