Background
The basic reservation concerning the use of hand-carried echocardiographic devices by general practitioners has been a lack of data demonstrating that they would provide clinically important information on patients’ outcomes.
Methods
One hundred seventy-five patients with heart failure and/or heart failure risk factors were selected from a general community outpatient clinic. They underwent simplified hand-carried echocardiography performed by an internist with basic echocardiographic experience and N-terminal pro–B-type natriuretic peptide measurements. Patients were followed for the occurrence of the combined end point, consisting of death or hospitalization for cardiovascular causes.
Results
Hand-carried echocardiographic results were abnormal in 90 patients (55%). During 48 ± 7 months of follow-up, the combined end point occurred in 41 patients (25%). On multivariate analysis, only abnormal echocardiography (hazard ratio, 5.55; 95% confidence interval, 2.04–14.28; P = .0004) was an independent predictor of outcomes.
Conclusions
Hand-carried echocardiographic examinations performed by an internist with basic echocardiographic training can provide important prognostic information, independent of N-terminal pro–B-type natriuretic peptide levels.
In recent years, progress in ultrasound technology has led to the development and dissemination of fully portable and truly hand-carried devices, or “echocardiographic stethoscopes.” The accuracy and reproducibility of hand-carried echocardiography for the detection of left ventricular (LV) systolic dysfunction, LV hypertrophy, and valvular regurgitation by experienced sonographers or cardiologists have been demonstrated both in the hospital setting and in the community. However, their use by noncardiologists has been debated. One of the basic reservations concerning the use of hand-carried devices by general practitioners and internists—so-called anyone, anytime, anywhere echocardiography—has been a lack of data demonstrating that examinations performed by inexperienced physicians would provide reliable and clinically important information on patients’ outcomes. Therefore, we hypothesized that use of an echocardiographic stethoscope by a primary care provider would provide useful prognostic data on clinical outcome in addition to standard clinical measures and serum B-type natriuretic peptide levels in patients with heart failure and heart failure risk factors. We undertook a community-based, prospective study to evaluate the clinical utility and prognostic value of hand-carried echocardiography, with a simplified imaging protocol, performed by an internist with basic training in echocardiography.
Methods
Study Population
The medical records of 5,521 patients, all Caucasians, from a community outpatient clinic, which provides health care services for the population of inhabitants of a small town (Iłża, Poland) were screened. The study selection criteria were age ≥ 55 years and either a diagnosis of heart failure ( n = 56 [32%]) or at least one of the following heart failure risk factors: ischemic heart disease ( n = 32 [34%]), hypertension with complications ( n = 73 [78%]), or diabetes mellitus ( n = 32 [34%]). Two hundred eighteen patients who met the selection criteria were invited to participate in the study. One hundred seventy-five patients (80.3%) responded to a written invitation. During an initial visit, all patients underwent comprehensive physical and simplified echocardiographic examinations with the hand-carried device (OptiGo; Philips Medical Systems, Andover, MA), performed by an internist with basic training in echocardiography. In addition N-terminal pro–B-type natriuretic peptide (NT-pro-BNP) levels were measured. A study flowchart is presented in Figure 1 .
The study complied with the Declaration of Helsinki, and the research protocol was approved by the ethics committee at our institution. Each patient gave written informed consent to participate in the study.
Hand-Carried Echocardiography
The hand-carried machine was equipped with a 2.5-MHz phased-array broadband transducer and weighed 3 kg. The device provided basic two-dimensional imaging, color Doppler, freeze frame, and linear distance measurement. The machine was equipped neither with spectral Doppler nor image-storing capability.
The physician (M.L.), without previous experience in echocardiography, underwent a 4-week training in a busy referral echocardiographic laboratory that performs approximately 1,500 examinations/month. The training covered theory, operating a hand-carried device, and performing 75 complete echocardiographic examinations under the supervision of an experienced cardiologist, which corresponded to basic-level training recommended by the American Society of Echocardiography.
The trainee focused in particular on a simplified imaging protocol that included (1) three standard measurements in the long-axis parasternal view (LV diastolic diameter, left atrial diameter, and LV posterior wall thickness, according to the recommendations of the American Society of Echocardiography ); (2) assessment of LV function in a binary way, as preserved or impaired (on the basis of a visually estimated LV ejection fraction); and (3) assessment of normal versus abnormal valvular function. Valvular function was defined as abnormal when color Doppler showed at least moderate regurgitation or two-dimensional imaging demonstrated visibly abnormal valve morphology and movement or when prosthetic valves were present. A ratio of jet area to left atrial area < 20% and no or minimal flow convergence defined nonsignificant mitral regurgitation. Larger jets were considered significant.
Results were defined as abnormal if at least one of the following was present: a visually estimated ejection fraction < 50%, LV enlargement (end-diastolic dimension ≥ 56 mm), LV hypertrophy (posterior wall thickness ≥ 12 mm), left atrial enlargement > 40 mm, or any valve pathology.
In 11 of 175 patients (6%), the quality of the hand-carried echocardiographic images was considered inadequate for the analysis, so the presented data are from the remaining 164 patients. The average duration of the hand-carried echocardiographic studies was 10 ± 3 minutes.
Serum NT-Pro-BNP Measurement
Blood samples for assessment of serum NT-pro-BNP levels were collected during the initial visit. The specimen collection and preparation were performed with kits available commercially, following the manufacturers’ instructions. Plasma was separated and stored at −70°C until analysis. Serum NT-pro-BNP levels were analyzed by an independent laboratory, blinded to the study results, using an Elecsys NT-pro-BNP sandwich-type immunoassay (Roche Diagnostics GmbH, Mannheim, Germany). The measuring range was 5 to 35,000 pg/mL, with an analytical sensitivity of 5 pg/mL and a functional sensitivity of 50 pg/mL. Receiver operating characteristic analyses were used to define the best cutoff value of NT-pro-BNP level to predict the occurrence of the combined end point.
Clinical Follow-Up
Patients were followed up prospectively for 48 ± 7 months (range, 5–50 months) by their primary care physicians. Follow-up was complete in all patients. The combined end point of the study was defined as death of any cause, hospital admission for heart failure, acute coronary syndromes, or any cardiac surgery (not planned at the time of baseline evaluation). The events were verified and adjudicated from chart review by an independent researcher, blinded to the results of echocardiographic assessment and NT-pro-BNP levels.
Statistical Analysis
Results are presented as mean ± SD for continuous variables and as percentages for categorical data. Categorical data were analyzed using χ 2 tests and normally distributed continuous data using Student’s t tests (or Mann-Whitney U tests when data were not normally distributed). Cox regression analysis using a multivariate logistic regression model was used to examine the independent prognostic value of abnormal hand-carried echocardiographic examination and NT-pro-BNP levels. Variables statistically significant ( P < .05) on univariate analyses and additionally sex and age (as important demographic variables) were included in multivariate analysis. Event-free (combined end point) survival curves for patients with normal versus abnormal echocardiographic results and normal versus abnormal NT-pro-BNP levels were constructed using the Kaplan-Meier method. The comparisons were made using the log-rank test. P values < .05 were considered significant.
Results
Hand-Carried Echocardiography
Hand-carried echocardiographic results were abnormal in 90 of 164 patients (54%; Table 1 ). The most frequent abnormalities were LV hypertrophy ( n = 47 [29%]) and LV systolic dysfunction ( n = 45 [27%]). Of the 45 patients with impaired LV systolic function, 11 patients did not have clinically overt heart failure. Of the 14 patients with valvular pathology, seven had significant mitral regurgitation, two had abnormal mitral valve leaflet mobility (corresponding to mitral stenosis), and two had abnormal aortic valve morphology and movement (significant aortic valve calcification). In one case, significant aortic regurgitation was accompanied by an ascending aortic aneurysm (6 cm in diameter). Two patients had prosthetic valves (one mitral prosthesis and one both mitral and aortic valve prostheses). In all patients with suspected valve problems ( n = 11), the diagnoses were confirmed by standard echocardiographic examinations, after which three patients underwent cardiac surgery. Baseline characteristics of patients with normal and abnormal echocardiographic results are presented in Table 2 . Standard echocardiographic examinations had been performed previously in 41 patients (25%), including 25 of 56 patients (45%) with heart failure and 16 of 108 patients (15%) without overt heart failure but with heart failure risk factors.
| Echocardiographic abnormality | n (%) |
|---|---|
| Impaired LV function | 45 (27) |
| Enlarged left ventricle | 21 (13) |
| Enlarged left atrium | 31 (19) |
| LV hypertrophy | 47 (29) |
| Valvular pathology | 14 (8) |
| Additional findings ∗ | 1 (0.6) |
∗ A 6-cm ascending aortic aneurysm in a patient with severe aortic regurgitation.
| Clinical characteristic | Normal echocardiographic results ( n = 74) | Abnormal echocardiographic results ( n = 90) | P |
|---|---|---|---|
| Age (y) | 64.8 ± 6.6 | 67.2 ± 8.1 | .04 |
| Women | 43 (58%) | 52 (58%) | .96 |
| Smokers | 26 (35%) | 30 (33%) | .80 |
| Heart failure | 0 (0%) | 56 (62%) | <.0001 |
| Hypertension | 52 (70%) | 81 (90%) | .001 |
| Ischemic heart disease | 24 (32%) | 43 (48%) | .047 |
| Diabetes | 26 (35%) | 29 (32%) | .69 |
| Peripheral vascular disease | 5 (7%) | 11 (12%) | .24 |
| Cerebrovascular disease | 3 (4%) | 11 (12%) | .05 |
| Atrial fibrillation/flutter | 3 (4%) | 19 (21%) | .001 |
| Abnormal NT-pro-BNP | 26 (36%) | 49 (57%) | .009 |
Follow-Up
Follow-up data were available for all 164 participants with diagnostic echocardiogram. During 48 ± 7 months of follow-up, the combined end point occurred in 41 patients (25%). Twenty-nine (70%) were cardiovascular hospitalizations: 18 for heart failure, five for acute coronary syndromes, and six for cardiac surgery. Twelve patients (7%) died during the follow-up. Clinical outcomes in those with normal versus abnormal echocardiographic results are presented in Table 3 .
| Outcome | Echocardiographic results ( n = 164) | |
|---|---|---|
| Normal ( n = 74) | Abnormal ( n = 90) | |
| Death | 1 | 11 |
| Hospitalization for | ||
| Acute coronary syndromes | 1 | 4 |
| Cardiac surgery | 1 | 5 |
| Heart failure | 3 | 15 |
| Combined end point | 6 | 35 |
According to receiver operating characteristic analysis, the best cutoff for NT-pro-BNP to predict the combined end point was found to be 206 pg/mL (area under the curve, 0.62; P = .019). Abnormal NT-pro-BNP levels and abnormal hand-carried echocardiographic results were significantly more common among patients with the combined end point in comparison with patients without events (61% vs 39%, P = .006, and 79% vs 41%, P < .0001, respectively). Among 41 patients with the combined end point, two (5%) had abnormal NT-pro-BNP levels and normal hand-carried echocardiographic results, 10 (24%) had abnormal echocardiographic results and normal NT-pro-BNP levels, 24 (58%) had abnormal results on both tests, and five (12%) had normal results on both tests. The sensitivity, specificity, and positive and negative predictive values of abnormal hand-carried echocardiographic results for the prediction of the combined end point were 85%, 52%, 37%, and 91%, respectively.
On univariate analysis, heart failure, cerebrovascular disease, atrial fibrillation or flutter, abnormal hand-carried echocardiographic results, and abnormal NT-pro-BNP levels were significant predictors of the occurrence of the combined end point. On multivariate analysis only abnormal hand-carried echocardiographic results remained a significant predictor of outcomes. The clinical characteristics of the patients with versus without events and the univariate and multivariate predictors of clinical outcome are shown in Table 4 .
| Clinical characteristic | Events ( n = 41) | No events ( n = 123) | Univariate P | Multivariate | |
|---|---|---|---|---|---|
| P | HR (95% CI) | ||||
| Age (y) | 66.9 ± 6.8 | 65.6 ± 7.7 | .46 | .72 | 1.01 (0.97–1.05) |
| Women | 25 (61%) | 70 (57%) | .60 | .26 | 1.49 (0.73–3.12) |
| Smokers | 13 (32%) | 43 (35%) | .64 | — | — |
| Heart failure | 22 (54%) | 34 (28%) | .003 | .51 | 1.30 (0.35–1.69) |
| Hypertension | 37 (90%) | 96 (78%) | .06 | — | — |
| Ischemic heart disease | 19 (46%) | 48 (39%) | .35 | — | — |
| Prior myocardial infarction | 12 (29%) | 35 (28%) | .87 | — | — |
| Diabetes | 10 (24%) | 45 (37%) | .18 | — | — |
| Peripheral vascular disease | 5 (12%) | 11 (9%) | .55 | — | — |
| Cerebrovascular disease | 8 (19%) | 6 (5%) | .01 | .20 | 1.88 (0.74–5.00) |
| Atrial fibrillation/flutter | 11 (27%) | 11 (9%) | .008 | .38 | 1.42 (2.04–3.12) |
| Abnormal echocardiographic results | 35 (85%) | 55 (45%) | <.0001 | .0004 | 5.55 (2.04–14.28) |
| Abnormal NT-pro-BNP | 26 (67%) | 51 (41%) | .006 | .07 | 1.86 (0.93–4.16) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
