Abstract
Transthoracic echocardiography is recommended in all patients with acute coronary syndrome but is time-consuming and lacks an evidence base. We aimed to assess the feasibility, diagnostic accuracy, and time efficiency of hand-held echocardiography in patients with acute coronary syndrome and describe the impact of echocardiography on clinical management in this setting.
Patients with acute coronary syndrome underwent both hand-held and transthoracic echocardiographies with agreement between key imaging parameters assessed using kappa statistics. The immediate clinical impact of hand-held echocardiography in this population was systematically evaluated. Overall, 262 patients (65 ± 12 years, 71% male) participated. Agreement between hand-held and transthoracic echocardiographies was good-to-excellent (kappa 0.60–1.00) with hand-held echocardiography having an overall negative predictive value of 95%. Hand-held echocardiography was performed rapidly (7.7 ± 1.6 min) and completed a median of 5 (interquartile range 3–20) h earlier than transthoracic echocardiography. Systematic hand-held echocardiography in all patients with acute coronary syndrome identified an important cardiac abnormality in 50%, and the clinical management plan was changed by echocardiography in 42%. In 85% of cases, hand-held echocardiography was sufficient for patient decision-making, and transthoracic echocardiography was no longer deemed necessary.
In patients with acute coronary syndrome, hand-held echocardiography provides comparable results to transthoracic echocardiography, can be more rapidly applied, and gives sufficient imaging information for decision-making in the vast majority of patients. Systematic echocardiography has clinical impact in half of patients, supporting the clinical utility of echocardiography in this population and providing an evidence base for current guidelines.
Introduction
Acute coronary syndrome is a challenging clinical and public health problem worldwide. Despite major therapeutic advances, acute coronary syndrome remains the foremost cause of death in developed countries and one of the most important risk factors for the development of heart failure.1–3 Transthoracic echocardiography is important in the diagnosis, prognostication, and management of patients with suspected or diagnosed acute coronary syndrome. Although a comprehensive transthoracic echocardiography examination is recommended in every patient presenting with acute coronary syndrome in both the European and American guidelines,4–8 there is a lack of evidence to demonstrate the efficacy of this approach. Indeed, these recommendations are given a Class 1 indication but with a level of evidence C, indicating this is based upon expert opinion in the absence of scientific evidence. Furthermore, performing a dedicated comprehensive transthoracic echocardiography in all patients with acute coronary syndrome is a labour-intense endeavour, especially in resource-constrained settings.
In this context, the development of portable hand-held ultrasound devices, also known as hand-held echocardiography, provides an attractive, easy-to-use, and low-cost method that could facilitate rapid bedside clinical assessment of patients presenting with acute coronary syndrome. This approach might be more readily deliverable across diverse healthcare systems and might avoid the need for a comprehensive transthoracic echocardiogram in all patients. Previous studies have shown the ability of hand-held echocardiography devices to perform a focused echocardiographic evaluation and to assess specific imaging parameters including ventricular enlargement, systolic dysfunction, hypertrophy, and pericardial effusions, outperforming physical examinations and reducing the use of other imaging examinations.9–12 However, studies demonstrating the utility of this approach in specific clinical scenarios where multiple different pathologies and diagnoses might be identified are limited, and in particular, we lack studies investigating the role of hand-held echocardiography in patients with acute coronary syndrome.
The aim of the present study was two-fold. First, we sought to investigate the feasibility of routine hand-held echocardiography in patients with acute coronary syndrome, its time efficiency, and its ability to identify important pathology compared to more comprehensive transthoracic echocardiography. Second, we wanted to evaluate the clinical impact of systematic echocardiography in sequential patients with acute coronary syndrome to support the central role of this imaging technique in this condition. Our hypothesis was that hand-held echocardiography would provide similar diagnostic information to comprehensive transthoracic echocardiography, but in a more time-efficient manner, and that routine echocardiography in patients who present with acute coronary syndrome would have an important impact on diagnosis and management.
Methods
Study population
Patients admitted with suspected acute coronary syndrome to the coronary care unit or cardiology wards of the Royal Infirmary of Edinburgh, UK, were prospectively enrolled in the study. Patients with ST-segment elevation or non–ST-segment elevation acute coronary syndrome were eligible for inclusion. Patients presenting with acute chest pain with an alternative diagnosis that was clearly apparent before the need for echocardiography were excluded from the study. The present study was conducted as a quality improvement (QI) project, after discussion with and approved by the QI team at the National Health Service Lothian Health Board.
Study design
The study consisted of two study phases. During Phase 1, the feasibility of hand-held echocardiography to perform a diagnostic imaging evaluation in patients hospitalized with suspected acute coronary syndrome was tested by four study investigators (J.G., P.M., A.B., and S.B.). The imaging protocol was optimized so that reliable measurements of key imaging assessments driving decision-making in patients with acute coronary syndrome could be delivered at the bedside in <10 min (see Supplementary data online, Figure S1). Key imaging assessments were chosen based upon clinical priority and recommendations by an expert group of imaging and interventional cardiologists (R.E., H.S.A., B.C., D.E.N., and M.R.D.)13–15 and comprised left ventricular, right ventricular, and left atrial size, left ventricular wall thickness, visually estimated left ventricular ejection fraction (≥55%, 54–40%, <40%), left ventricular regional wall motion abnormality (present vs. absent; if present, anterior, inferior, or lateral), right ventricular function (normal vs. impaired), significant valve disease (at least moderate regurgitant valve disease or stenotic aortic valve disease; present vs. absent), pericardial effusion (absent, <2 cm, ≥2 cm), and dilatation of the ascending aorta (normal vs. dilated) according to the European Association of Cardiovascular Imaging (EACVI) recommendations.16 Furthermore, the scan screened for the presence of acute coronary syndrome–related complications (left ventricular thrombus, left ventricular aneurysm or pseudoaneurysm, ventricular septal defect, and ruptured mitral valve chordae) or evidence of an alternative diagnosis to explain the patient’s clinical presentation (e.g. myocarditis, pericarditis, takotsubo cardiomyopathy, and pulmonary embolism).
Patients underwent both bedside hand-held echocardiography and comprehensive transthoracic echocardiography, and the scans were assessed to compare the ability of both techniques to identify the key predetermined imaging findings as listed above. Both echocardiographic examinations were performed by different experienced sonographers blinded to the results of the other examination, with time interval between admission and performance of the two scans recorded. A maximum time delay between the hand-held echocardiography and transthoracic echocardiography examinations of 24 h for patients with ST-segment elevation acute coronary syndrome and 4 days for those with non–ST-segment elevation acute coronary syndrome was considered acceptable.
In Phase 2, bedside hand-held echocardiography was routinely performed in all patients hospitalized with acute coronary syndrome, and the clinical impact of hand-held echocardiography was evaluated prospectively.
Imaging
Hand-held echocardiography
The hand-held echocardiography examination was performed by cardiologists (J.G., P.M., and S.B.) or a sonographer (A.B.) with prior training and experience in echocardiography (3–5 years) with a Lumify S4-1 broadband sector array transducer with 4–1 MHz extended frequency range (Philips Healthcare) and which obtains 2D greyscale, colour Doppler, M-mode, and pulsed wave Doppler imaging. No electrocardiogram (ECG) interface was present, 3-s clips were automatically stored. Patients were imaged at the bedside in left lateral decubitus position. The hand-held echocardiography examination was performed in a standardized manner including 2D and colour Doppler images from parasternal and apical windows, using an optimized imaging protocol (see Supplementary data online, Figure S1).
Transthoracic echocardiography
A comprehensive transthoracic echocardiography examination, using a Vivid E9 (GE Healthcare) or Affiniti 70 (Philips Healthcare) scanner, was performed by experienced and EACVI- or British Society of Echocardiography (BSE)–accredited sonographers. Patients were imaged in the echocardiography department or at the bedside depending on their clinical stability and at the discretion of the attending cardiologist. The transthoracic echocardiography examination included 2D, M-mode, spectral, tissue, and colour Doppler imaging with ECG-gated acquisition executed in different standard acoustic windows according to international guidelines17 in a BSE-accredited echocardiography laboratory. Additional modalities, such as three-dimensional, contrast-enhanced, and strain imaging, were used at the discretion of the sonographer and clinical team according to clinical need.
Scan duration and time to diagnosis
Scan duration, measured from time the ultrasound probe was placed on the patient’s chest until the last image was saved, was recorded for both hand-held echocardiography and transthoracic echocardiography examinations. The exact date and time of the patient’s admission and when their hand-held echocardiography and transthoracic echocardiography examinations were performed were registered, allowing assessments of whether hand-held echocardiography or transthoracic echocardiography was performed more promptly after the patient’s admission.
Image analysis
Independent analysis of anonymized hand-held echocardiography and transthoracic echocardiography images presented in random order was performed by a blinded accredited expert team (J.G., A.B., J.M., and M.D.) to determine the key predetermined imaging findings in patients presenting with acute coronary syndrome, as outlined above. The review also included a global assessment of image quality on a four-level scale (good, diagnostic, poor, and non-diagnostic). The transthoracic echocardiography images were reviewed for a second time for the same findings by the same team after an interval of 4 weeks to reduce recall bias and to allow intra-observer assessment of transthoracic echocardiography agreement as our reference standard.
Clinical findings and impact
Across the total study population, we collected information regarding the frequency of detecting key imaging findings (as specified above) with hand-held echocardiography. The clinical impact of systematic echocardiography in this population was investigated and documented using a standardized proforma (see Supplementary data online, Table S1). This was completed by the attending physician, directly after reviewing the hand-held echocardiography examination. Assessment of clinical impact was based on the hierarchical model proposed by Fryback and Thornbury.18 This six-level model represents a framework by which the diagnostic performance of any imaging modality can be evaluated (see Supplementary data online, Table S2). The overall clinical impact was defined as new diagnosis (Level 3: diagnostic thinking efficacy) or change in management (Level 4: therapeutic efficacy) (Figure 1). A new diagnosis was defined as occurring only if had not already been demonstrated by other investigations or clinical assessments and was previously unknown to the physician. An impact on management was defined as change in strategy prompted by the results of the echocardiography, reflected by one of the following criteria: change in medical therapy, change in investigations performed, or change in hospital duration. The impact on vehicular driving advice and the need to perform a comprehensive transthoracic echocardiography was also assessed, independent of the above criteria.
Flowchart representing the definition of significant clinical impact of hand-held echocardiography in patients hospitalized with acute coronary syndrome, including new diagnosis or change in management. The impact on vehicular driving advice and the need to perform a comprehensive transthoracic echocardiography after hand-held echocardiography was also assessed, independent from the criteria of overall clinical impact. Numbers illustrate the percentage of patients in whom these effects were found.
Statistical analysis
Categorical variables are presented as number (%), continuous variables as median (interquartile range), or mean ± standard deviation. Distributions of data were tested for normality with Shapiro–Wilk test and quantile–quantile plots. Comparisons between parameters and techniques were performed using independent samples Student’s t-test or Mann–Whitney U-test as appropriate for continuous variables or χ2 test or Fisher exact test for categorical variables. Agreement of hand-held echocardiography vs. transthoracic echocardiography was assessed with Cohen’s kappa statistics and compared to agreement of reviewing the comprehensive transthoracic echocardiography images twice (transthoracic echocardiography vs. transthoracic echocardiography) as reference standard. Kappa values of 0.41–0.60 were considered moderate agreement, 0.61–0.80 good agreement, and ≥ 0.81 excellent agreement as described previously.19 The diagnostic sensitivity, specificity, negative predictive value, and positive predictive value for hand-held echocardiography vs. transthoracic echocardiography and for transthoracic echocardiography vs. transthoracic echocardiography were calculated. Univariable and multivariable logistic regression models were used to identify potential clinical characteristics associated with a clinical impact and the need to perform transthoracic echocardiography. Analysis was performed using IBM SPSS Statistics (v28.0) and Python (v3.12). A two-sided P < 0.05 was considered statistically significant.
Results
Baseline characteristics
Overall, 295 patients admitted with acute coronary syndrome were scanned by hand-held echocardiography, of whom 33 were excluded (incomplete hand-held echocardiography scan in 1 patient, maximum predetermined duration between hand-held echocardiography and transthoracic echocardiography exceeded in 1 patient, and no transthoracic echocardiography performed during hospitalization in 31 patients), leaving 262 patients eligible for inclusion. The study cohort consisted of predominantly male patients with mean age of 65 ± 12 years, with 63% diagnosed with ST-segment elevation acute coronary syndrome (Table 1). An abnormal ECG was observed in all patients with ST-segment elevation acute coronary syndrome and in half of those with non–ST-segment elevation acute coronary syndrome. Invasive coronary angiography was performed in >90% of the overall study population.
Baseline clinical characteristics of the study population
| Characteristic | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Age, years | 65 ± 12 | 64 ± 12 | 67 ± 11 | 0.028 |
| Women | 75 (29%) | 46 (28%) | 29 (30%) | 0.727 |
| BMI, kg/m2 | 29 ± 6 | 29 ± 6 | 28 ± 5 | 0.674 |
| Hypertension | 102 (39%) | 102 (39%) | 42 (43%) | 0.283 |
| Type 2 diabetes | 50 (19%) | 27 (16%) | 23 (24%) | 0.144 |
| Current or ex-smoker | 144 (55%) | 94 (57%) | 50 (52%) | 0.394 |
| Known IHD | 68 (26%) | 33 (20%) | 35 (36%) | 0.004 |
| Baseline ECG changes | 213 (81%) | 165 (100%) | 50 (52%) | <0.001 |
| Peak troponin I (ng/L) | 153 (50 to 671) | 260 (67 to 988) | 92 (34 to 266) | <0.001 |
| Cardiac murmur | 13 (5%) | 3 (2%) | 10 (10%) | 0.002 |
| Angiography performed | 253 (97%) | 162 (98%) | 91 (94%) | 0.061 |
| MI territory | ||||
| Anterior | 86 (33%) | 66 (40%) | 20 (21%) | 0.001 |
| Inferior | 89 (34%) | 83 (50%) | 6 (6%) | <0.001 |
| Lateral | 14 (5%) | 10 (6%) | 4 (4%) | 0.501 |
| Multivessel/undetermined | 73 (28%) | 6 (4%) | 67 (69%) | <0.001 |
| Treatment | ||||
| PCI | 198 (76%) | 151 (92%) | 47 (48%) | <0.001 |
| CABG | 13 (5%) | 1 (1%) | 12 (12%) | <0.001 |
| Medical therapy | 45 (17%) | 13 (8%) | 32 (33%) | <0.001 |
| Other | 4 (2%) | 0 | 4 (4%) | 0.009 |
| Characteristic | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Age, years | 65 ± 12 | 64 ± 12 | 67 ± 11 | 0.028 |
| Women | 75 (29%) | 46 (28%) | 29 (30%) | 0.727 |
| BMI, kg/m2 | 29 ± 6 | 29 ± 6 | 28 ± 5 | 0.674 |
| Hypertension | 102 (39%) | 102 (39%) | 42 (43%) | 0.283 |
| Type 2 diabetes | 50 (19%) | 27 (16%) | 23 (24%) | 0.144 |
| Current or ex-smoker | 144 (55%) | 94 (57%) | 50 (52%) | 0.394 |
| Known IHD | 68 (26%) | 33 (20%) | 35 (36%) | 0.004 |
| Baseline ECG changes | 213 (81%) | 165 (100%) | 50 (52%) | <0.001 |
| Peak troponin I (ng/L) | 153 (50 to 671) | 260 (67 to 988) | 92 (34 to 266) | <0.001 |
| Cardiac murmur | 13 (5%) | 3 (2%) | 10 (10%) | 0.002 |
| Angiography performed | 253 (97%) | 162 (98%) | 91 (94%) | 0.061 |
| MI territory | ||||
| Anterior | 86 (33%) | 66 (40%) | 20 (21%) | 0.001 |
| Inferior | 89 (34%) | 83 (50%) | 6 (6%) | <0.001 |
| Lateral | 14 (5%) | 10 (6%) | 4 (4%) | 0.501 |
| Multivessel/undetermined | 73 (28%) | 6 (4%) | 67 (69%) | <0.001 |
| Treatment | ||||
| PCI | 198 (76%) | 151 (92%) | 47 (48%) | <0.001 |
| CABG | 13 (5%) | 1 (1%) | 12 (12%) | <0.001 |
| Medical therapy | 45 (17%) | 13 (8%) | 32 (33%) | <0.001 |
| Other | 4 (2%) | 0 | 4 (4%) | 0.009 |
Significant p-values are highlighted in bold. BMI, body mass index; IHD, ischaemic heart disease; CCU, coronary care unit; ECG, electrocardiogram; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft.
Baseline clinical characteristics of the study population
| Characteristic | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Age, years | 65 ± 12 | 64 ± 12 | 67 ± 11 | 0.028 |
| Women | 75 (29%) | 46 (28%) | 29 (30%) | 0.727 |
| BMI, kg/m2 | 29 ± 6 | 29 ± 6 | 28 ± 5 | 0.674 |
| Hypertension | 102 (39%) | 102 (39%) | 42 (43%) | 0.283 |
| Type 2 diabetes | 50 (19%) | 27 (16%) | 23 (24%) | 0.144 |
| Current or ex-smoker | 144 (55%) | 94 (57%) | 50 (52%) | 0.394 |
| Known IHD | 68 (26%) | 33 (20%) | 35 (36%) | 0.004 |
| Baseline ECG changes | 213 (81%) | 165 (100%) | 50 (52%) | <0.001 |
| Peak troponin I (ng/L) | 153 (50 to 671) | 260 (67 to 988) | 92 (34 to 266) | <0.001 |
| Cardiac murmur | 13 (5%) | 3 (2%) | 10 (10%) | 0.002 |
| Angiography performed | 253 (97%) | 162 (98%) | 91 (94%) | 0.061 |
| MI territory | ||||
| Anterior | 86 (33%) | 66 (40%) | 20 (21%) | 0.001 |
| Inferior | 89 (34%) | 83 (50%) | 6 (6%) | <0.001 |
| Lateral | 14 (5%) | 10 (6%) | 4 (4%) | 0.501 |
| Multivessel/undetermined | 73 (28%) | 6 (4%) | 67 (69%) | <0.001 |
| Treatment | ||||
| PCI | 198 (76%) | 151 (92%) | 47 (48%) | <0.001 |
| CABG | 13 (5%) | 1 (1%) | 12 (12%) | <0.001 |
| Medical therapy | 45 (17%) | 13 (8%) | 32 (33%) | <0.001 |
| Other | 4 (2%) | 0 | 4 (4%) | 0.009 |
| Characteristic | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Age, years | 65 ± 12 | 64 ± 12 | 67 ± 11 | 0.028 |
| Women | 75 (29%) | 46 (28%) | 29 (30%) | 0.727 |
| BMI, kg/m2 | 29 ± 6 | 29 ± 6 | 28 ± 5 | 0.674 |
| Hypertension | 102 (39%) | 102 (39%) | 42 (43%) | 0.283 |
| Type 2 diabetes | 50 (19%) | 27 (16%) | 23 (24%) | 0.144 |
| Current or ex-smoker | 144 (55%) | 94 (57%) | 50 (52%) | 0.394 |
| Known IHD | 68 (26%) | 33 (20%) | 35 (36%) | 0.004 |
| Baseline ECG changes | 213 (81%) | 165 (100%) | 50 (52%) | <0.001 |
| Peak troponin I (ng/L) | 153 (50 to 671) | 260 (67 to 988) | 92 (34 to 266) | <0.001 |
| Cardiac murmur | 13 (5%) | 3 (2%) | 10 (10%) | 0.002 |
| Angiography performed | 253 (97%) | 162 (98%) | 91 (94%) | 0.061 |
| MI territory | ||||
| Anterior | 86 (33%) | 66 (40%) | 20 (21%) | 0.001 |
| Inferior | 89 (34%) | 83 (50%) | 6 (6%) | <0.001 |
| Lateral | 14 (5%) | 10 (6%) | 4 (4%) | 0.501 |
| Multivessel/undetermined | 73 (28%) | 6 (4%) | 67 (69%) | <0.001 |
| Treatment | ||||
| PCI | 198 (76%) | 151 (92%) | 47 (48%) | <0.001 |
| CABG | 13 (5%) | 1 (1%) | 12 (12%) | <0.001 |
| Medical therapy | 45 (17%) | 13 (8%) | 32 (33%) | <0.001 |
| Other | 4 (2%) | 0 | 4 (4%) | 0.009 |
Significant p-values are highlighted in bold. BMI, body mass index; IHD, ischaemic heart disease; CCU, coronary care unit; ECG, electrocardiogram; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft.
Level of agreement and time efficiency
During Phase 1, both transthoracic echocardiography and hand-held echocardiography were performed in 108 patients admitted with acute coronary syndrome. The hand-held echocardiography scan was graded as poor image quality in 20% compared to 9% for transthoracic echocardiography (P = 0.020) and graded as non-diagnostic in 2% compared to 1% for transthoracic echocardiography (P = 0.890).
The negative predictive value of hand-held echocardiography vs. transthoracic echocardiography for detecting the key predetermined imaging findings was 95%; the positive predictive value was 78% (Table 2). This was similar to repeated review of transthoracic echocardiography images (negative predictive value 95%, positive predictive value 86%). Agreement in detecting the majority of pre-specified clinical features was good to excellent (kappa > 0.60–1.00) and similar for hand-held echocardiography vs. transthoracic echocardiography compared to repeated transthoracic echocardiography assessments (Table 2 and Figure 2). Agreement between hand-held echocardiography and transthoracic echocardiography was less good than repeated transthoracic echocardiography analysis for the detection of valvular heart disease (at least moderate mitral or aortic regurgitation, or aortic stenosis), lateral regional wall motion abnormalities, right ventricular dysfunction, and left ventricular thrombus (Table 2).
Hand-held echocardiography (left) and transthoracic echocardiography images (right) for some of the key echocardiographic findings. (A) Apical four-chamber view of patient with ischaemic mitral regurgitation. (B) Apical four-chamber view of patient with apical akinesis and evidence of apical thrombus (arrow). (C) Parasternal long-axis image of patient with takotsubo cardiomyopathy with apical ballooning. (D) Short-axis view at the level of the mitral valve in patient with hypertrophic cardiomyopathy.
Accuracy of hand-held echocardiography vs. transthoracic echocardiography compared to reviewing the transthoracic echocardiography images twice (transthoracic echocardiography vs. transthoracic echocardiography) as reference
| Variables | Hand-held echocardiography vs. transthoracic echocardiography | Transthoracic echocardiography vs. transthoracic echocardiography | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | |
| Left ventricular dilatation | 0.85 (0.64–1.00) | 100 | 98 | 100 | 75 | 0.90 (0.72–1.00) | 83 | 100 | 99 | 100 |
| Left ventricular hypertrophy | 0.79 (0.68–0.91) | 84 | 95 | 87 | 93 | 0.83 (0.72–0.94) | 86 | 96 | 89 | 96 |
| Sign valve disease | 0.74 (0.56–0.92) | 67 | 99 | 94 | 92 | 0.90 (0.78–1.00) | 89 | 99 | 98 | 94 |
| Right ventricular dysfunction | 0.33 (0.01–0.64) | 33 | 96 | 94 | 43 | 0.60 (0.32–0.87) | 67 | 96 | 97 | 60 |
| Right ventricular dilatation | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 |
| Left ventricular ejection fraction < 40% | 0.66 (0.51–0.80) | 83 | 85 | 91 | 73 | 0.62 (0.46–0.78) | 64 | 94 | 84 | 85 |
| Presence of regional wall motion abnormality | 0.69 (0.52–0.87) | 96 | 68 | 83 | 92 | 0.73 (0.58–0.89) | 93 | 83 | 76 | 95 |
| Anterior | 0.89 (0.80–0.97) | 96 | 93 | 96 | 92 | 0.89 (0.80–0.97) | 90 | 98 | 92 | 98 |
| Inferior | 0.74 (0.61–0.87) | 92 | 86 | 95 | 77 | 0.65 (0.50–0.79) | 89 | 80 | 94 | 70 |
| Lateral | 0.37 (0.13–0.62) | 33 | 97 | 88 | 67 | 0.79 (0.63–0.95) | 78 | 98 | 96 | 88 |
| Left ventricular thrombus | 0.31 (−0.18–0.81) | 33 | 98 | 98 | 33 | 0.85 (0.57–1.00) | 100 | 99 | 100 | 75 |
| Pericardial effusion | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Ventricular septal defect | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Takotsubo cardiomyopathy | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Average test performance | 0.74 (0.59–0.92) | 80 | 94 | 95 | 78 | 0.84 (0.66–0.93) | 88 | 96 | 95 | 86 |
| Variables | Hand-held echocardiography vs. transthoracic echocardiography | Transthoracic echocardiography vs. transthoracic echocardiography | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | |
| Left ventricular dilatation | 0.85 (0.64–1.00) | 100 | 98 | 100 | 75 | 0.90 (0.72–1.00) | 83 | 100 | 99 | 100 |
| Left ventricular hypertrophy | 0.79 (0.68–0.91) | 84 | 95 | 87 | 93 | 0.83 (0.72–0.94) | 86 | 96 | 89 | 96 |
| Sign valve disease | 0.74 (0.56–0.92) | 67 | 99 | 94 | 92 | 0.90 (0.78–1.00) | 89 | 99 | 98 | 94 |
| Right ventricular dysfunction | 0.33 (0.01–0.64) | 33 | 96 | 94 | 43 | 0.60 (0.32–0.87) | 67 | 96 | 97 | 60 |
| Right ventricular dilatation | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 |
| Left ventricular ejection fraction < 40% | 0.66 (0.51–0.80) | 83 | 85 | 91 | 73 | 0.62 (0.46–0.78) | 64 | 94 | 84 | 85 |
| Presence of regional wall motion abnormality | 0.69 (0.52–0.87) | 96 | 68 | 83 | 92 | 0.73 (0.58–0.89) | 93 | 83 | 76 | 95 |
| Anterior | 0.89 (0.80–0.97) | 96 | 93 | 96 | 92 | 0.89 (0.80–0.97) | 90 | 98 | 92 | 98 |
| Inferior | 0.74 (0.61–0.87) | 92 | 86 | 95 | 77 | 0.65 (0.50–0.79) | 89 | 80 | 94 | 70 |
| Lateral | 0.37 (0.13–0.62) | 33 | 97 | 88 | 67 | 0.79 (0.63–0.95) | 78 | 98 | 96 | 88 |
| Left ventricular thrombus | 0.31 (−0.18–0.81) | 33 | 98 | 98 | 33 | 0.85 (0.57–1.00) | 100 | 99 | 100 | 75 |
| Pericardial effusion | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Ventricular septal defect | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Takotsubo cardiomyopathy | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Average test performance | 0.74 (0.59–0.92) | 80 | 94 | 95 | 78 | 0.84 (0.66–0.93) | 88 | 96 | 95 | 86 |
Κ, Cohen kappa; CI, confidence interval; Sen, sensitivity; Spe, specificity; NPV, negative predictive value; PPV, positive predictive value.
Accuracy of hand-held echocardiography vs. transthoracic echocardiography compared to reviewing the transthoracic echocardiography images twice (transthoracic echocardiography vs. transthoracic echocardiography) as reference
| Variables | Hand-held echocardiography vs. transthoracic echocardiography | Transthoracic echocardiography vs. transthoracic echocardiography | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | |
| Left ventricular dilatation | 0.85 (0.64–1.00) | 100 | 98 | 100 | 75 | 0.90 (0.72–1.00) | 83 | 100 | 99 | 100 |
| Left ventricular hypertrophy | 0.79 (0.68–0.91) | 84 | 95 | 87 | 93 | 0.83 (0.72–0.94) | 86 | 96 | 89 | 96 |
| Sign valve disease | 0.74 (0.56–0.92) | 67 | 99 | 94 | 92 | 0.90 (0.78–1.00) | 89 | 99 | 98 | 94 |
| Right ventricular dysfunction | 0.33 (0.01–0.64) | 33 | 96 | 94 | 43 | 0.60 (0.32–0.87) | 67 | 96 | 97 | 60 |
| Right ventricular dilatation | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 |
| Left ventricular ejection fraction < 40% | 0.66 (0.51–0.80) | 83 | 85 | 91 | 73 | 0.62 (0.46–0.78) | 64 | 94 | 84 | 85 |
| Presence of regional wall motion abnormality | 0.69 (0.52–0.87) | 96 | 68 | 83 | 92 | 0.73 (0.58–0.89) | 93 | 83 | 76 | 95 |
| Anterior | 0.89 (0.80–0.97) | 96 | 93 | 96 | 92 | 0.89 (0.80–0.97) | 90 | 98 | 92 | 98 |
| Inferior | 0.74 (0.61–0.87) | 92 | 86 | 95 | 77 | 0.65 (0.50–0.79) | 89 | 80 | 94 | 70 |
| Lateral | 0.37 (0.13–0.62) | 33 | 97 | 88 | 67 | 0.79 (0.63–0.95) | 78 | 98 | 96 | 88 |
| Left ventricular thrombus | 0.31 (−0.18–0.81) | 33 | 98 | 98 | 33 | 0.85 (0.57–1.00) | 100 | 99 | 100 | 75 |
| Pericardial effusion | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Ventricular septal defect | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Takotsubo cardiomyopathy | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Average test performance | 0.74 (0.59–0.92) | 80 | 94 | 95 | 78 | 0.84 (0.66–0.93) | 88 | 96 | 95 | 86 |
| Variables | Hand-held echocardiography vs. transthoracic echocardiography | Transthoracic echocardiography vs. transthoracic echocardiography | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | κ (95% CI) | Sen (%) | Spe (%) | NPV (%) | PPV (%) | |
| Left ventricular dilatation | 0.85 (0.64–1.00) | 100 | 98 | 100 | 75 | 0.90 (0.72–1.00) | 83 | 100 | 99 | 100 |
| Left ventricular hypertrophy | 0.79 (0.68–0.91) | 84 | 95 | 87 | 93 | 0.83 (0.72–0.94) | 86 | 96 | 89 | 96 |
| Sign valve disease | 0.74 (0.56–0.92) | 67 | 99 | 94 | 92 | 0.90 (0.78–1.00) | 89 | 99 | 98 | 94 |
| Right ventricular dysfunction | 0.33 (0.01–0.64) | 33 | 96 | 94 | 43 | 0.60 (0.32–0.87) | 67 | 96 | 97 | 60 |
| Right ventricular dilatation | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 | 0.66 (0.22–1.00) | 100 | 98 | 100 | 50 |
| Left ventricular ejection fraction < 40% | 0.66 (0.51–0.80) | 83 | 85 | 91 | 73 | 0.62 (0.46–0.78) | 64 | 94 | 84 | 85 |
| Presence of regional wall motion abnormality | 0.69 (0.52–0.87) | 96 | 68 | 83 | 92 | 0.73 (0.58–0.89) | 93 | 83 | 76 | 95 |
| Anterior | 0.89 (0.80–0.97) | 96 | 93 | 96 | 92 | 0.89 (0.80–0.97) | 90 | 98 | 92 | 98 |
| Inferior | 0.74 (0.61–0.87) | 92 | 86 | 95 | 77 | 0.65 (0.50–0.79) | 89 | 80 | 94 | 70 |
| Lateral | 0.37 (0.13–0.62) | 33 | 97 | 88 | 67 | 0.79 (0.63–0.95) | 78 | 98 | 96 | 88 |
| Left ventricular thrombus | 0.31 (−0.18–0.81) | 33 | 98 | 98 | 33 | 0.85 (0.57–1.00) | 100 | 99 | 100 | 75 |
| Pericardial effusion | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Ventricular septal defect | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Takotsubo cardiomyopathy | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 | 1.00 (1.00–1.00) | 100 | 100 | 100 | 100 |
| Average test performance | 0.74 (0.59–0.92) | 80 | 94 | 95 | 78 | 0.84 (0.66–0.93) | 88 | 96 | 95 | 86 |
Κ, Cohen kappa; CI, confidence interval; Sen, sensitivity; Spe, specificity; NPV, negative predictive value; PPV, positive predictive value.
The mean hand-held echocardiography scanning time was 7.7 ± 1.6 min, and the median time from first presentation with chest pain to echocardiography was 5-h shorter for hand-held echocardiography compared to transthoracic echocardiography [15 (7–32) vs. 21 (15–46) h, P = 0.040].
Imaging abnormalities
Across the total study population of 262 patients hospitalized with acute coronary syndrome, echocardiographic evaluation identified an important cardiac abnormality in 50% (Table 3). An important abnormality was observed in 53% of patients with ST-segment elevation acute coronary syndrome and 45% of those with non–ST-segment elevation acute coronary syndrome. Across all patients, left ventricular systolic dysfunction with left ventricular ejection fraction < 40% was observed in 28%, left ventricular hypertrophy in 36%, left ventricular dilatation in 7%, and right ventricular dysfunction in 8%, whilst valve disease was observed in 10%, and left ventricular thrombus was suspected in 3%.
Diagnostic findings of systematic echocardiography in a contemporary population of patients presenting with an acute coronary syndrome
| Imaging findings | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Important cardiac abnormalitya | 131 (50%) | 87 (53%) | 44 (45%) | 0.222 |
| Left ventricular dilatation | 18 (7%) | 12 (7%) | 6 (6%) | 0.744 |
| Left atrial dilatation | 51 (20%) | 27 (17%) | 24 (25%) | 0.104 |
| Aortopathy | 5 (1.9%) | 5 (3%) | 0 | 0.085 |
| Significant valve diseaseb | 25 (10%) | 12 (7%) | 13 (13%) | 0.106 |
| Right ventricular dysfunction | 20 (8%) | 15 (9%) | 5 (5%) | 0.241 |
| Right ventricular dilatation | 10 (3.8%) | 9 (5.5%) | 1 (1.0%) | 0.070 |
| Left ventricular ejection fraction < 40% | 73 (28%) | 53 (32%) | 20 (21%) | 0.042 |
| Suspicion of left ventricular thrombus | 9 (3.4%) | 6 (3.7%) | 3 (3.1%) | 0.813 |
| Left ventricular aneurysm | 3 (1.1%) | 2 (1.2%) | 1 (1.0%) | 0.893 |
| Pericardial effusion | 4 (1.5%) | 2 (1.2%) | 2 (2.1%) | 0.597 |
| Ventricular septal defect | 1 (0.4%) | 1 (0.6%) | 0 | 0.442 |
| Takotsubo cardiomyopathy | 1 (0.4%) | 0 | 1 (1.0%) | 0.192 |
| Left ventricular hypertrophy | 93 (36%) | 58 (36%) | 35 (36%) | 0.916 |
| Presence of regional wall motion abnormality | 202 (77%) | 150 (92%) | 52 (54%) | <0.001 |
| Imaging findings | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Important cardiac abnormalitya | 131 (50%) | 87 (53%) | 44 (45%) | 0.222 |
| Left ventricular dilatation | 18 (7%) | 12 (7%) | 6 (6%) | 0.744 |
| Left atrial dilatation | 51 (20%) | 27 (17%) | 24 (25%) | 0.104 |
| Aortopathy | 5 (1.9%) | 5 (3%) | 0 | 0.085 |
| Significant valve diseaseb | 25 (10%) | 12 (7%) | 13 (13%) | 0.106 |
| Right ventricular dysfunction | 20 (8%) | 15 (9%) | 5 (5%) | 0.241 |
| Right ventricular dilatation | 10 (3.8%) | 9 (5.5%) | 1 (1.0%) | 0.070 |
| Left ventricular ejection fraction < 40% | 73 (28%) | 53 (32%) | 20 (21%) | 0.042 |
| Suspicion of left ventricular thrombus | 9 (3.4%) | 6 (3.7%) | 3 (3.1%) | 0.813 |
| Left ventricular aneurysm | 3 (1.1%) | 2 (1.2%) | 1 (1.0%) | 0.893 |
| Pericardial effusion | 4 (1.5%) | 2 (1.2%) | 2 (2.1%) | 0.597 |
| Ventricular septal defect | 1 (0.4%) | 1 (0.6%) | 0 | 0.442 |
| Takotsubo cardiomyopathy | 1 (0.4%) | 0 | 1 (1.0%) | 0.192 |
| Left ventricular hypertrophy | 93 (36%) | 58 (36%) | 35 (36%) | 0.916 |
| Presence of regional wall motion abnormality | 202 (77%) | 150 (92%) | 52 (54%) | <0.001 |
aDefined as presence of at least one of the imaging parameters summed below with exception of left ventricular hypertrophy and presence of regional wall motion abnormality.
bSignificant valve disease is defined as at least moderate mitral or aortic regurgitation or suspicion of aortic stenosis.
n (%). Significant p-values are highlighted in bold.
Diagnostic findings of systematic echocardiography in a contemporary population of patients presenting with an acute coronary syndrome
| Imaging findings | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Important cardiac abnormalitya | 131 (50%) | 87 (53%) | 44 (45%) | 0.222 |
| Left ventricular dilatation | 18 (7%) | 12 (7%) | 6 (6%) | 0.744 |
| Left atrial dilatation | 51 (20%) | 27 (17%) | 24 (25%) | 0.104 |
| Aortopathy | 5 (1.9%) | 5 (3%) | 0 | 0.085 |
| Significant valve diseaseb | 25 (10%) | 12 (7%) | 13 (13%) | 0.106 |
| Right ventricular dysfunction | 20 (8%) | 15 (9%) | 5 (5%) | 0.241 |
| Right ventricular dilatation | 10 (3.8%) | 9 (5.5%) | 1 (1.0%) | 0.070 |
| Left ventricular ejection fraction < 40% | 73 (28%) | 53 (32%) | 20 (21%) | 0.042 |
| Suspicion of left ventricular thrombus | 9 (3.4%) | 6 (3.7%) | 3 (3.1%) | 0.813 |
| Left ventricular aneurysm | 3 (1.1%) | 2 (1.2%) | 1 (1.0%) | 0.893 |
| Pericardial effusion | 4 (1.5%) | 2 (1.2%) | 2 (2.1%) | 0.597 |
| Ventricular septal defect | 1 (0.4%) | 1 (0.6%) | 0 | 0.442 |
| Takotsubo cardiomyopathy | 1 (0.4%) | 0 | 1 (1.0%) | 0.192 |
| Left ventricular hypertrophy | 93 (36%) | 58 (36%) | 35 (36%) | 0.916 |
| Presence of regional wall motion abnormality | 202 (77%) | 150 (92%) | 52 (54%) | <0.001 |
| Imaging findings | Overall (n = 262) | ST-segment elevation acute coronary syndrome (n = 165) | Non–ST-segment elevation acute coronary syndrome (n = 97) | P-value |
|---|---|---|---|---|
| Important cardiac abnormalitya | 131 (50%) | 87 (53%) | 44 (45%) | 0.222 |
| Left ventricular dilatation | 18 (7%) | 12 (7%) | 6 (6%) | 0.744 |
| Left atrial dilatation | 51 (20%) | 27 (17%) | 24 (25%) | 0.104 |
| Aortopathy | 5 (1.9%) | 5 (3%) | 0 | 0.085 |
| Significant valve diseaseb | 25 (10%) | 12 (7%) | 13 (13%) | 0.106 |
| Right ventricular dysfunction | 20 (8%) | 15 (9%) | 5 (5%) | 0.241 |
| Right ventricular dilatation | 10 (3.8%) | 9 (5.5%) | 1 (1.0%) | 0.070 |
| Left ventricular ejection fraction < 40% | 73 (28%) | 53 (32%) | 20 (21%) | 0.042 |
| Suspicion of left ventricular thrombus | 9 (3.4%) | 6 (3.7%) | 3 (3.1%) | 0.813 |
| Left ventricular aneurysm | 3 (1.1%) | 2 (1.2%) | 1 (1.0%) | 0.893 |
| Pericardial effusion | 4 (1.5%) | 2 (1.2%) | 2 (2.1%) | 0.597 |
| Ventricular septal defect | 1 (0.4%) | 1 (0.6%) | 0 | 0.442 |
| Takotsubo cardiomyopathy | 1 (0.4%) | 0 | 1 (1.0%) | 0.192 |
| Left ventricular hypertrophy | 93 (36%) | 58 (36%) | 35 (36%) | 0.916 |
| Presence of regional wall motion abnormality | 202 (77%) | 150 (92%) | 52 (54%) | <0.001 |
aDefined as presence of at least one of the imaging parameters summed below with exception of left ventricular hypertrophy and presence of regional wall motion abnormality.
bSignificant valve disease is defined as at least moderate mitral or aortic regurgitation or suspicion of aortic stenosis.
n (%). Significant p-values are highlighted in bold.
Clinical impact of routine echocardiography
During Phase 2, systematic echocardiographic evaluation using hand-held echocardiography was performed in 154 patients with acute coronary syndrome. The patient’s attending cardiologist found that routine echocardiography had an immediate clinical impact in 48% of cases (new clinical diagnosis or change in management; Table 4), which was more frequent in patients with ST-segment elevation acute coronary syndrome compared to those with non–ST-segment elevation (55 vs. 37%, P = 0.025). Vehicular driving advice was also altered by the results of the echocardiography in 66% of patients.
Clinical impact of systematic echocardiography with hand-held echocardiography device in patients admitted with acute coronary syndrome
| Impact of echocardiography | Overall (n = 154) | ST-segment elevation acute coronary syndrome (n = 92) | Non–ST-segment elevation acute coronary syndrome (n = 62) | P-value |
|---|---|---|---|---|
| Overall clinical impact | 74 (48%) | 51 (55%) | 23 (37%) | 0.025 |
| New diagnosis | 43 (28%) | 32 (35%) | 11 (18%) | 0.021 |
| Left ventricular ejection fraction < 40% | 30 (20%) | 23 (25%) | 7 (11%) | 0.035 |
| Congestive heart failure | 7 (4%) | 4 (4%) | 3 (5%) | 0.886 |
| Aortopathy | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Valvular heart disease | 9 (6%) | 3 (3%) | 6 (10%) | 0.662 |
| Hypertrophic cardiomyopathy | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Right ventricular infarction | 2 (1%) | 2 (2%) | 0 | 0.243 |
| Left ventricular aneurysm | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Suspicion of left ventricular thrombus | 6 (4%) | 3 (3%) | 3 (5%) | 0.620 |
| Impact on management | 65 (42%) | 43 (47%) | 21 (34%) | 0.112 |
| Cancellation of transthoracic echocardiogram | 131 (85%) | 79 (86%) | 52 (84%) | 0.733 |
| Cancellation of other investigations | 2 (1%) | 1 (1%) | 1 (2%) | 0.777 |
| Extra investigations requested | 25 (16%) | 14 (15%) | 11 (18%) | 0.677 |
| Departmental transthoracic echocardiography | 19 (12%) | 11 (12%) | 8 (13%) | 0.861 |
| Contrast transthoracic echocardiography | 3 (2%) | 1 (1%) | 2 (3%) | 0.346 |
| Chest X-ray | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Cardiac magnetic resonance | 6 (4%) | 4 (4%) | 2 (3%) | 0.724 |
| Initiation/uptitration of heart failure drugs | 37 (24%) | 25 (27%) | 12 (19%) | 0.265 |
| Initiation of other drugs | 5 (3%) | 3 (3%) | 2 (3%) | 0.990 |
| Discontinuation of drugs | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Defibrillator might be considered | 12 (8%) | 10 (11%) | 2 (3%) | 0.083 |
| Change in length of hospital stay | 26 (17%) | 16 (17%) | 10 (16%) | 0.838 |
| Shorter | 17 (11%) | 11 (12%) | 6 (10%) | 0.658 |
| Longer | 9 (6%) | 5 (5%) | 4 (6%) | 0.792 |
| Impact on vehicular driving advice | 101 (66%) | 69 (75%) | 32 (52%) | 0.003 |
| Impact of echocardiography | Overall (n = 154) | ST-segment elevation acute coronary syndrome (n = 92) | Non–ST-segment elevation acute coronary syndrome (n = 62) | P-value |
|---|---|---|---|---|
| Overall clinical impact | 74 (48%) | 51 (55%) | 23 (37%) | 0.025 |
| New diagnosis | 43 (28%) | 32 (35%) | 11 (18%) | 0.021 |
| Left ventricular ejection fraction < 40% | 30 (20%) | 23 (25%) | 7 (11%) | 0.035 |
| Congestive heart failure | 7 (4%) | 4 (4%) | 3 (5%) | 0.886 |
| Aortopathy | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Valvular heart disease | 9 (6%) | 3 (3%) | 6 (10%) | 0.662 |
| Hypertrophic cardiomyopathy | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Right ventricular infarction | 2 (1%) | 2 (2%) | 0 | 0.243 |
| Left ventricular aneurysm | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Suspicion of left ventricular thrombus | 6 (4%) | 3 (3%) | 3 (5%) | 0.620 |
| Impact on management | 65 (42%) | 43 (47%) | 21 (34%) | 0.112 |
| Cancellation of transthoracic echocardiogram | 131 (85%) | 79 (86%) | 52 (84%) | 0.733 |
| Cancellation of other investigations | 2 (1%) | 1 (1%) | 1 (2%) | 0.777 |
| Extra investigations requested | 25 (16%) | 14 (15%) | 11 (18%) | 0.677 |
| Departmental transthoracic echocardiography | 19 (12%) | 11 (12%) | 8 (13%) | 0.861 |
| Contrast transthoracic echocardiography | 3 (2%) | 1 (1%) | 2 (3%) | 0.346 |
| Chest X-ray | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Cardiac magnetic resonance | 6 (4%) | 4 (4%) | 2 (3%) | 0.724 |
| Initiation/uptitration of heart failure drugs | 37 (24%) | 25 (27%) | 12 (19%) | 0.265 |
| Initiation of other drugs | 5 (3%) | 3 (3%) | 2 (3%) | 0.990 |
| Discontinuation of drugs | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Defibrillator might be considered | 12 (8%) | 10 (11%) | 2 (3%) | 0.083 |
| Change in length of hospital stay | 26 (17%) | 16 (17%) | 10 (16%) | 0.838 |
| Shorter | 17 (11%) | 11 (12%) | 6 (10%) | 0.658 |
| Longer | 9 (6%) | 5 (5%) | 4 (6%) | 0.792 |
| Impact on vehicular driving advice | 101 (66%) | 69 (75%) | 32 (52%) | 0.003 |
n (%). Significant p-values are highlighted in bold.
Clinical impact of systematic echocardiography with hand-held echocardiography device in patients admitted with acute coronary syndrome
| Impact of echocardiography | Overall (n = 154) | ST-segment elevation acute coronary syndrome (n = 92) | Non–ST-segment elevation acute coronary syndrome (n = 62) | P-value |
|---|---|---|---|---|
| Overall clinical impact | 74 (48%) | 51 (55%) | 23 (37%) | 0.025 |
| New diagnosis | 43 (28%) | 32 (35%) | 11 (18%) | 0.021 |
| Left ventricular ejection fraction < 40% | 30 (20%) | 23 (25%) | 7 (11%) | 0.035 |
| Congestive heart failure | 7 (4%) | 4 (4%) | 3 (5%) | 0.886 |
| Aortopathy | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Valvular heart disease | 9 (6%) | 3 (3%) | 6 (10%) | 0.662 |
| Hypertrophic cardiomyopathy | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Right ventricular infarction | 2 (1%) | 2 (2%) | 0 | 0.243 |
| Left ventricular aneurysm | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Suspicion of left ventricular thrombus | 6 (4%) | 3 (3%) | 3 (5%) | 0.620 |
| Impact on management | 65 (42%) | 43 (47%) | 21 (34%) | 0.112 |
| Cancellation of transthoracic echocardiogram | 131 (85%) | 79 (86%) | 52 (84%) | 0.733 |
| Cancellation of other investigations | 2 (1%) | 1 (1%) | 1 (2%) | 0.777 |
| Extra investigations requested | 25 (16%) | 14 (15%) | 11 (18%) | 0.677 |
| Departmental transthoracic echocardiography | 19 (12%) | 11 (12%) | 8 (13%) | 0.861 |
| Contrast transthoracic echocardiography | 3 (2%) | 1 (1%) | 2 (3%) | 0.346 |
| Chest X-ray | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Cardiac magnetic resonance | 6 (4%) | 4 (4%) | 2 (3%) | 0.724 |
| Initiation/uptitration of heart failure drugs | 37 (24%) | 25 (27%) | 12 (19%) | 0.265 |
| Initiation of other drugs | 5 (3%) | 3 (3%) | 2 (3%) | 0.990 |
| Discontinuation of drugs | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Defibrillator might be considered | 12 (8%) | 10 (11%) | 2 (3%) | 0.083 |
| Change in length of hospital stay | 26 (17%) | 16 (17%) | 10 (16%) | 0.838 |
| Shorter | 17 (11%) | 11 (12%) | 6 (10%) | 0.658 |
| Longer | 9 (6%) | 5 (5%) | 4 (6%) | 0.792 |
| Impact on vehicular driving advice | 101 (66%) | 69 (75%) | 32 (52%) | 0.003 |
| Impact of echocardiography | Overall (n = 154) | ST-segment elevation acute coronary syndrome (n = 92) | Non–ST-segment elevation acute coronary syndrome (n = 62) | P-value |
|---|---|---|---|---|
| Overall clinical impact | 74 (48%) | 51 (55%) | 23 (37%) | 0.025 |
| New diagnosis | 43 (28%) | 32 (35%) | 11 (18%) | 0.021 |
| Left ventricular ejection fraction < 40% | 30 (20%) | 23 (25%) | 7 (11%) | 0.035 |
| Congestive heart failure | 7 (4%) | 4 (4%) | 3 (5%) | 0.886 |
| Aortopathy | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Valvular heart disease | 9 (6%) | 3 (3%) | 6 (10%) | 0.662 |
| Hypertrophic cardiomyopathy | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Right ventricular infarction | 2 (1%) | 2 (2%) | 0 | 0.243 |
| Left ventricular aneurysm | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Suspicion of left ventricular thrombus | 6 (4%) | 3 (3%) | 3 (5%) | 0.620 |
| Impact on management | 65 (42%) | 43 (47%) | 21 (34%) | 0.112 |
| Cancellation of transthoracic echocardiogram | 131 (85%) | 79 (86%) | 52 (84%) | 0.733 |
| Cancellation of other investigations | 2 (1%) | 1 (1%) | 1 (2%) | 0.777 |
| Extra investigations requested | 25 (16%) | 14 (15%) | 11 (18%) | 0.677 |
| Departmental transthoracic echocardiography | 19 (12%) | 11 (12%) | 8 (13%) | 0.861 |
| Contrast transthoracic echocardiography | 3 (2%) | 1 (1%) | 2 (3%) | 0.346 |
| Chest X-ray | 1 (1%) | 0 | 1 (2%) | 0.222 |
| Cardiac magnetic resonance | 6 (4%) | 4 (4%) | 2 (3%) | 0.724 |
| Initiation/uptitration of heart failure drugs | 37 (24%) | 25 (27%) | 12 (19%) | 0.265 |
| Initiation of other drugs | 5 (3%) | 3 (3%) | 2 (3%) | 0.990 |
| Discontinuation of drugs | 4 (3%) | 4 (4%) | 0 | 0.096 |
| Defibrillator might be considered | 12 (8%) | 10 (11%) | 2 (3%) | 0.083 |
| Change in length of hospital stay | 26 (17%) | 16 (17%) | 10 (16%) | 0.838 |
| Shorter | 17 (11%) | 11 (12%) | 6 (10%) | 0.658 |
| Longer | 9 (6%) | 5 (5%) | 4 (6%) | 0.792 |
| Impact on vehicular driving advice | 101 (66%) | 69 (75%) | 32 (52%) | 0.003 |
n (%). Significant p-values are highlighted in bold.
We explored if, based on clinical characteristics, we could predict which patients with acute coronary syndrome would not have received a clinical impact from routine echocardiography. All baseline clinical variables were tested, but none were independent predictors in multivariable analysis (see Supplementary data online, Table S3).
Clinical decision-making with hand-held echocardiography
In 85% of patients, the attending cardiologist felt that transthoracic echocardiography was no longer required after reviewing the hand-held echocardiography images, and the transthoracic echocardiography was subsequently cancelled (Table 4). When follow-up transthoracic echocardiography was required, the most important reasons were to further evaluate valve disease (n = 9), because of suspicion of left ventricular thrombus (n = 5) and due to insufficient hand-held echocardiography image quality (n = 5). Performing a comprehensive transthoracic echocardiography after hand-held echocardiography changed diagnosis in 76% of these cases: estimation of left ventricular ejection fraction was clarified in 33%, valve disease was excluded in 33%, and left ventricular thrombus was excluded in 19%. One patient with left ventricular thrombus, one patient with right ventricular dilatation, and one patient with left ventricular hypertrophy diagnosed on transthoracic echocardiography were missed by hand-held echocardiography, but in all these cases, the hand-held echocardiography image quality was graded as suboptimal, and the clinical team decided that transthoracic echocardiography was required for diagnostic imaging.
We aimed to assess whether certain clinical characteristics could predict patients in whom hand-held echocardiography would not be considered sufficient and who should therefore go straight to transthoracic echocardiography. On univariable logistic regression, presence of a cardiac murmur was the only clinical characteristic at presentation associated with the need to perform a transthoracic echocardiography (odds ratio, 10.83; 95% confidence interval, 1.69–69.30; P = 0.012; see Supplementary data online, Table S4). This association remained after adjustment for age (P = 0.023), body mass index (P = 0.007), presence of ECG changes (P = 0.007), and troponin concentration (P = 0.009).
Discussion
Our data demonstrate the feasibility of hand-held echocardiography imaging in patients hospitalized with acute coronary syndrome and that this technique is accurate in providing similar diagnostic information to transthoracic echocardiography for the majority of key echocardiographic findings. However, hand-held echocardiography can deliver this information in a more time-efficient manner than transthoracic echocardiography, providing rapid diagnostic images at the patient bedside. Importantly, attending cardiologists were happy to base their decision-making upon the information provided by hand-held echocardiography without the need for further imaging in 85% of acute coronary syndrome patients. The use of hand-held echocardiography to assess patients presenting with acute coronary syndrome therefore has major potential to improve the efficient delivery of current acute coronary syndrome guidelines across cardiology departments worldwide.
Modern hand-held echocardiography devices provide comparable image quality to cart-based transthoracic echocardiography machines, but are four- to five-fold less expensive, require less energy, and are more portable, allowing for faster bedside imaging. Whilst hand-held echocardiography devices have reduced functionality compared to cart-based transthoracic echocardiography, for example, the lack of strain imaging and Doppler quantification in most hand-held echocardiography machines, these additional assessments are not always required for clinical decision-making, particularly in patients with acute coronary syndrome. Hand-held echocardiography is therefore attractive as a cost-effective, sustainable imaging technology that might streamline bedside patient assessment and reduce demand on busy echocardiography departments. This approach is of particular interest in the large population of patients with acute coronary syndrome in whom echocardiography during hospitalization is recommended by current international guidelines, despite a previous lack of underlying scientific evidence. Our study has confirmed the feasibility of performing hand-held echocardiography in these patients, demonstrating that this approach allows more rapid evaluation with similar diagnostic accuracy and a high overall negative predictive value of 95% compared to transthoracic echocardiography. This indicates that hand-held echocardiography could be implemented as a reliable screening test that can rule out important imaging abnormalities in patients presenting with acute coronary syndrome and distinguish patients in whom transthoracic echocardiography can be avoided. Beyond acute coronary syndrome management, hand-held echocardiography has several other valuable indications. Some examples are evaluation of volume status and guiding fluid management in heart failure patients, real-time imaging guidance during bedside procedures like pericardiocentesis, and rapid exclusion of pericardial effusion after cardiac interventions.
Our hand-held echocardiography evaluation took less than 8 min and accurately identified significant left ventricular impairment, most regional wall motion abnormalities, and left ventricular and right ventricular dilatation. Moreover, hand-held echocardiography was able to pick up most important complications following myocardial infarction including ventricular septal defects and pericardial effusions. It was also able to diagnose pathologies other than acute coronary syndrome that might underlie a presentation with acute chest pain, such as takotsubo cardiomyopathy. Importantly, when shown the hand-held echocardiography scans, attending cardiologists felt confident in the information these provided, using them to make clinical decisions without the need for transthoracic echocardiography in more than four-fifths of patients.
The two key advantages of hand-held echocardiography are its low cost and time efficiency. The focused hand-held echocardiography took a median of 7.7 min, compared to transthoracic echocardiography which takes a minimum of 20 min to perform. In this study, after taking into account the number of patients who required both hand-held echocardiography and a transthoracic echocardiography evaluation, the hand-held echocardiography first approach saved more than 16 h of scanning time per 100 patients with acute coronary syndrome. This is of clinical importance as many echocardiography departments face increasing workloads and staffing pressure. This time difference is based purely on the difference in scanning time. Hand-held echocardiography has other time advantages related to its portability and the fact that it renders bedside imaging more readily feasible. This avoids additional delays and the labour required to transport patients to and from echocardiography departments and might explain why in our study bedside hand-held echocardiography could be performed 5 h more quickly after admission than comprehensive transthoracic echocardiography. Such time gains might have important implications, for example, allowing for earlier initiation of therapy and more rapid discharge from hospital. Rapid bedside imaging is also of particular potential values in patients with ST-segment elevation acute coronary syndrome where focused imaging immediately before or after percutaneous coronary intervention is frequently useful in decision-making and establishing an accurate diagnosis.
For these reasons, the use of hand-held echocardiography provides an attractive, easy-to-use, and low-cost method that could facilitate rapid bedside clinical assessment of acute coronary syndrome patients. This approach might be more readily deliverable across stretched healthcare systems than current strategies, avoiding the need for comprehensive transthoracic echocardiography and patient transfer to the echocardiography department in the large majority of patients.
Despite these encouraging findings, hand-held echocardiography has its limitations that need to be considered if this approach is rolled out more broadly in clinical practice. Our data showed lower agreement between hand-held echocardiography and transthoracic echocardiography for lateral regional wall motion abnormalities and right ventricular dysfunction. The left ventricular lateral wall and right ventricular are frequently hard to image with echocardiography in general. Indeed, this study demonstrated relatively low agreement for these features when the same transthoracic echocardiography images were assessed twice. The further reduction in agreement with hand-held echocardiography may be related to decreased image quality and limited ability to optimize images. There was also a low sensitivity of 33% for hand-held echocardiography to detect left ventricular thrombus compared to transthoracic echocardiography. The difficulty of assessing left ventricular thrombus more generally with non-contrast transthoracic echocardiography is already well described in the literature with low sensitivity ranging from 21 to 35%.20,21 Keeping this in mind, our threshold for requesting an additional transthoracic echocardiography in Phase 2 was low, and a transthoracic echocardiography was scheduled whenever there was any suspicion of left ventricular thrombus on hand-held echocardiography or when the apex was not fully visualized. Consequently, possible left ventricular thrombus was overestimated with hand-held echocardiography in this second phase of our study. Indeed, transthoracic echocardiography only confirmed a diagnosis of left ventricular thrombus in one of the six patients where it was suspected by hand-held echocardiography. Another limitation is that our hand-held echocardiography device lacked continuous wave Doppler, meaning the operator had to rely on visual estimation of colour Doppler, resulting in limited ability to accurately assess valve disease severity, as previously reported.22–24 Again, with this limitation in mind, a transthoracic echocardiography was requested whenever hand-held echocardiography suggested any valvular lesion, again resulting in overestimation of the presence of significant valve disease on hand-held echocardiography vs. transthoracic echocardiography in Phase 2. Our data indicate that in patients who present with heart murmurs, hand-held echocardiography may not be a good initial option. These patients will ultimately require comprehensive transthoracic echocardiography assessment for heart valve disease and should therefore move straight to this test.
Having established the feasibility and accuracy of hand-held echocardiography, we were then able to investigate more generally whether routine echocardiography is of clinical utility in all patients with acute coronary syndrome, what abnormalities it detects, and whether it changes management. Do all acute coronary syndrome patients really need an echocardiogram or is it only of value in particular subgroups, for example, those with ST-segment elevation acute coronary syndrome or an abnormal baseline ECG?
European and American guidelines strongly recommend performing a complete echocardiogram in every hospitalized patient with acute coronary syndrome, but evidence supporting this guideline has been lacking.4–8 Indeed, current recommendations are based on expert opinion and clinical experience and are not supported by any clinical studies (Class 1, Level of Evidence C). We did not consider it ethical to perform a randomized controlled trial investigating the clinical utility of echocardiography in patients with acute coronary syndrome. Instead, we have therefore performed this prospective observational study where we sought to investigate systematically and prospectively the abnormalities identified by echocardiography in patients with acute coronary syndrome and to evaluate their clinical impact. To our knowledge, this is the first observational study of this kind in an acute coronary syndrome population receiving contemporary interventional and pharmaceutical treatments. In 267 patients with a broad range of clinical presentations ranging from large ST-segment elevation acute coronary syndrome to small non–ST-segment elevation acute coronary syndrome, we have demonstrated that systematic echocardiography reveals a relevant underlying imaging abnormality in half of the patients. These findings influenced diagnostic thinking or clinical management in a similar proportion as determined by the opinion of their attending cardiologist. None of the baseline clinical characteristics were able to predict the likelihood of echocardiography to have clinical impact. It thus remains difficult to select patients or subgroups who are more likely to benefit from echocardiography or indeed patients in whom echocardiography is not necessary. Of importance, echocardiography also had an important impact on advice for driving in more than two-thirds of patients, further underlining its general utility across this patient population. In totality, we believe our data provide strong contemporary support to the current guidance of performing echocardiographic evaluation in all patients with acute coronary syndrome.
Limitations
This is a single-centre study with relatively small sample size. For the assessment of agreement between modalities, time between hand-held echocardiography and transthoracic echocardiography was kept to a minimum but could be as long as 24 h for patients with ST-segment elevation acute coronary syndrome and 96 h for those with non–ST-segment elevation acute coronary syndrome. Therefore, it is possible that evolution in haemodynamic and cardiac function could account for some of the observed variation between hand-held echocardiography and transthoracic echocardiography findings. The hand-held examination was performed by examiners with prior training and experience in echocardiography. Its performance in the hands of less-experienced operators is uncertain, but echocardiography should always be performed by appropriately trained individuals. The hand-held echocardiography protocol was more targeted than the transthoracic echocardiography protocol. Streamlining the transthoracic echocardiography examination could therefore also improve its time efficiency although contemporary guidelines do not recommend this approach.4–7 Most patients had an ST-segment elevation acute coronary syndrome that could potentially bias the clinical impact of echocardiography, as one might assume these patients would benefit more from echocardiography. However, in multivariable analysis, an ST-segment elevation acute coronary syndrome diagnosis was not able to distinguish between patients who would or would not benefit from echocardiography. Although we demonstrated clinical impact, subsequent influence on patient outcomes was not evaluated—this would require a randomized controlled trial that as stated we did not consider ethical to perform. Finally, the small numbers of infrequent complications, such as pericardial effusions or ventricular septal defects, limits the power and interpretability of our findings.
Conclusion
Echocardiographic evaluation of patients hospitalized with acute coronary syndrome detects an important imaging finding and consequently influences diagnostic thinking and clinical management in half of the cases. A bedside hand-held echocardiography in patients hospitalized with acute coronary syndrome is time-efficient and reliable, providing similar diagnostic information to transthoracic echocardiography for most key echocardiographic findings in a shorter timeframe. Systematic echocardiographic evaluation of patients with acute coronary syndrome is recommended, and hand-held echocardiography could be an effective gatekeeper to transthoracic echocardiography, reducing both time-to-scan and scan duration as well as decreasing the need for transthoracic echocardiography.
Supplementary data
Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.
Acknowledgements
The authors would like to acknowledge the medical and paramedical teams of the cardiology and cardiac imaging departments of the Royal Infirmary of Edinburgh for their contributions to the conduct of the study.
Funding
J.G. is supported by the European Association of Cardiovascular Imaging (research grant 2022) and by the Research Foundation Flanders (FWO Travel Grant 2023). M.R.D. is supported by the British Heart Foundation (FS/SCRF/21/32010) and is the recipient of the Sir Jules Thorn Award for Biomedical research 2015 (15/JTA). D.E.N. is supported by the British Heart Foundation (CH/09/002, RG/F/22/110093, RE/18/5/34216) and was the recipient of a Wellcome Trust Senior Investigator Award (WT103782AIA).
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
Author notes
Conflict of interest: None declared.
