Contrast-enhanced echocardiography (CE) helps to improve image quality in patients with suboptimal acoustic windows. Despite current recommendations, contrast use remains low. The aim of this study was to identify populations that would benefit more from contrast use.
A total of 176 subjects (137 men; mean age, 60.8 ± 13.7 years) with technically difficult transthoracic echocardiographic studies who received clinically indicated intravenous contrast were prospectively studied. The impact on clinical decision making (including alterations in medical therapy, referral, imaging, or clinical procedures) was evaluated.
The use of CE enabled biplane left ventricular (LV) ejection fraction measurement in 97.2% of studies and the interpretation of regional wall motion in 95% of studies. CE allowed definitive assessment of the presence or absence of LV thrombus in 99% of the cases. In the 174 patients whose ordering physicians could be reached at the time of image interpretation, changes in management occurred in 51% of subjects. There was no difference in the proportion of management changes between inpatients and outpatients (60.0% vs 48.1%, P = .225). Subjects with heart failure, cardiomyopathy, and arrhythmia had a higher proportion of changes (61.4% vs 44.2% [ P = .031], 62.5% vs 45.0% [ P = .028], and 72.0% vs 47.7% [ P = .030], respectively). The proportion of management change after CE increased as pre-CE estimated ejection fraction decreased. Logistic regression showed that pre-CE estimated LV ejection fraction < 50% was the only significant predictor of change of management after contrast ( P = .004).
The use of CE has a significant impact on clinical decision making in patients with suboptimal acoustic windows, especially in those with depressed pre-CE LV ejection fractions.
Contrast use in patients with suboptimal images affects patient care, especially in patients who have reduced precontrast estimated LVEF.
The impact of contrast use is seen more significantly in patients with HF, cardiomyopathy, and arrhythmias.
The impact of contrast is not influenced by the results of prior studies, whether outpatient or inpatient, or whether the patient has a defibrillator or a pacemaker.
Contrast-enhanced echocardiography (CE) improves image quality in patients with poor acoustic windows, which occurs in ≥10% to 15% of patients undergoing transthoracic echocardiography (TTE). CE allows more accurate assessment of cardiac structure and function, as well as better identification of left ventricular (LV) mass and thrombus. The use of CE improves the diagnostic value of TTE and decreases the utilization of additional imaging tests, without obvious side effects or prolonged examination time. The value of CE has been validated in intensive care unit and emergency department patients. The impact of CE on clinical decision making has been studied in a primarily inpatient-based population, in which the most significant impact was found in the intensive care unit. A low incidence of clinical management change after CE was found in the smaller outpatient group. The American Society of Echocardiography recommends the use of contrast for examinations in which two segments are not visualized in the apical views. Despite this recommendation, the penetrance of contrast use in TTE remains low compared with the number of examinations with suboptimal images. Compared with inpatients, the reluctance to use contrast in outpatient echocardiographic examinations may be due to a lack of intravenous access, although it has been shown that the use of CE in outpatients improves image quality without prolonging total procedure time. We prospectively conducted a study aiming to understand if there are subgroups of patients who would benefit more in terms of clinical decision making from the improved image quality with CE use.
Three level 3–trained experienced echocardiographers (Y.H., R.O., and M.A.) prospectively interpreted transthoracic echocardiograms from consecutive patients presenting to our institution from January 2016 to January 2017 for clinically indicated TTE who had technically difficult pre-CE transthoracic images and received intravenous echocardiographic contrast (Optison; GE Healthcare, Princeton, NJ). As per American Society of Echocardiography guidelines, a technically difficult study was defined as one in which at least two segments were not visualized despite efforts to optimize image quality. The decision to use contrast was made by the clinical sonographers at the time of image acquisition with input from the echocardiographer on duty if needed. Pregnant patients and those with known or suspected hypersensitivity to blood products or albumin were excluded. Stress echocardiographic studies were also excluded.
A complete two-dimensional and Doppler echocardiographic study was performed using standardized protocols from the parasternal, apical, subcostal, and suprasternal windows using standard equipment (iE33 and EPIQ [Philips Medical Systems, Bothell, WA] and Vivid 7 and 9 [GE Medical Systems, Milwaukee, WI]). Optison was diluted in normal saline and given intravenously in boluses according to the manufacturer’s specifications. Dose and frequency of readministration were adjusted as needed to achieve optimal LV opacification by the sonographer, working with a nurse who administrated the contrast. Contrast-specific imaging modes were used with low mechanical index, which was adjusted as necessary to achieve a balance between ultrasound penetration, intensity, and duration of myocardial opacification with contrast.
Baseline clinical characteristics including age, gender, indication for TTE, and the presence of an implantable cardioverter-defibrillator (ICD) or pacemaker were recorded for each study. If the patient had undergone prior TTE at our institution, the most recent LV ejection fraction (LVEF) was also recorded.
The precontrast images were interpreted by one of the three investigators; LVEF was estimated visually according to the available views obtained without contrast when feasible. The estimated LVEF, the ability to assess regional wall motion abnormality, and the ability to rule out LV thrombus were recorded before visualization of any contrast-enhanced images. An uninterpretable study was defined as a study in which >75% of the endocardium not visualized from any window, and no estimation of LVEF could be reported. After recording these findings, contrast-enhanced images were then interpreted by the same investigator with biplane measurement of LVEF when feasible. The findings on contrast-enhanced echocardiographic images were then recorded before contacting referral physicians.
Assessment of the Impact of Contrast on Clinical Management
After recording the sequential interpretations of the precontrast and contrast-enhanced images, the same investigator called the referring physician. The referring physician was first asked for verbal consent to participate in the study and, when agreeing, was given the LVEF from the most recent study in our system if available, the estimated precontrast LVEF, segmental wall motion abnormality if any, and the assessment of thrombus if possible from precontrast transthoracic echocardiographic images, and was then asked about his or her management plan, on the basis of the reported results. The clinical decision making included (1) any change in medication therapy, including the initiation or discontinuation of hemodynamically active drugs (diuretics, intravenous fluids, inotropic agents, vasodilators, and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers), antiarrhythmic drugs, and anticoagulation therapy; (2) whether any referral to cardiology, heart failure (HF), or electrophysiology would be ordered; (3) the need to order any additional cardiac imaging tests; and (4) clinical decisions involving other invasive procedures (ICD, LV assist device, cardiac resynchronization therapy, endocardial biopsy, catheterization, etc) or rearrangement of follow-up intervals. After recording these management decisions, the results from the contrast-enhanced echocardiographic images were then provided to the clinician, who was asked the same management questions again. The flowchart of the study procedure is detailed in Figure 1 .
The study was approved by the University of Pennsylvania institutional review board. A waiver of Health Insurance Portability and Accountability Act authorization and a waiver of patient consent were approved for clinical procedures and data, and verbal consent was obtained from the referring physicians before they were asked about the clinical managements on the patients.
Statistical analyses were performed using SPSS version 17.0 (SPSS, Chicago, IL). Demographic data and results are provided as mean ± SD for continuous variables and percentage for categorical variables. With regard to clinical decision change, groups with or without a specific baseline characteristic were compared using χ 2 tests. P values < .05 were considered to indicate statistical significance. Univariate logistic regression was then performed using all covariates, and covariates with P values < .05 were entered into multivariate logistic regression to identify independent baseline characteristics associated with the change of clinical management after CE.
A total of 176 patients (137 men; mean age, 60.8 ± 13.7 years) were included in the study, including 131 outpatients (74.4%) and 45 inpatients (25.6%). These represented 6.2% of a total of 2,887 transthoracic echocardiographic studies performed during the same period and interpreted by the three study investigators. Baseline characteristics of patients are listed in Table 1 . The top three indications for TTE were HF ( n = 70 [40.0%]), cardiomyopathy ( n = 64 [36.4%]), and coronary artery disease ( n = 57 [32.4%]). All indications were separately counted when more than one indication was present. Ninety-five patients (54%) had prior transthoracic echocardiographic results in our system.
|Age (y)||60.8 ± 13.7|
|Patients with prior transthoracic echocardiographic results||95 (54.0)|
|Patients with ICDs/PMs implanted||28 (15.9)|
|Medical history/indications for TTE|
|Coronary artery disease||57 (32.4)|
|Valvular diseases||21 (11.9)|
|Dyspnea/chest pain||12 (6.8)|
|Preoperative evaluation||10 (5.7)|
|Heart transplantation||10 (5.7)|
Impact of CE on LV Function and Thrombus Assessment
Estimated LVEFs were obtained in 173 patients (98.3%) before CE. After CE, biplane LVEF measurements were feasible in 171 patients (97.2%). Representative pre- and post-CE images are shown in Figure 2 . The impact of CE on the diagnostic value of echocardiographic studies is shown in Figure 3 . Segmental wall motion could not be assessed in 120 cases (68.2%) before CE but only in five cases (2.8%) after CE. Before CE, the presence of LV thrombus could not be ruled out in 163 (92.6%). CE helped diagnose LV thrombus in five of these patients, while ruling out thrombus in 157 subjects. A representative LV apical thrombus detected by CE is shown in Figure 4 . There was only one case (0.6%) in which LV thrombus could not be excluded after CE. Other new echocardiographic findings detected by CE included regional hypertrophy ( n = 4), LV noncompaction ( n = 4), and right ventricular abnormality ( n = 4).
Impact of CE on LVEF Assessment in Different LVEF Categories
The average pre- and post-CE LVEF values were 43.6 ± 16.9% and 47.0 ± 17.7%, respectively. The number of patients with estimated pre-CE LVEFs <30%, 30% to 50%, and >50% were 35 (20.5%), 66 (38.6%), and 70 (40.9%), respectively. The number of patients in different LVEF categories before and after CE is shown in Table 2 . Pre-CE LVEF > 50% was associated with a lower proportion of LVEF category changes after the administration of contrast compared with pre-CE LVEF 30% to 50% and pre-CE LVEF < 30% (7.2% vs 27.3% and 18.8%, respectively, P = .009).
|Pre-CE LVEF||Post-CE LVEF||Total|
Impact on Clinical Management
We failed to reach the referring providers in two cases, which were excluded from further analysis. A total of 92 physicians were reached by telephone for their management decisions. All physicians gave verbal consent. One hundred twenty of our cases (69.0%) were managed by physicians specializing in cardiology. The number of providers in each category and the total number of their cases are listed in Table 3 .
|Cardiology||39 (42.4%)||120 (69.0%)|
|General||8 (8.7%)||10 (5.7%)|
|Imaging||9 (9.8%)||26 (14.9%)|
|HF||12 (13.0%)||67 (38.5%)|
|Electrophysiology||7 (7.6%)||10 (5.7%)|
|Interventional||3 (3.3%)||7 (4.0%)|
|Noncardiology||53 (57.6%)||54 (31.0%)|
|General medicine||9 (9.8%)||9 (5.2%)|
|Surgical specialties||16 (17.4%)||16 (9.2%)|
|Other specialties ∗||28 (30.4%)||29 (16.7%)|
In the cohort of 174 patients, changes in overall management occurred in 89 subjects (51%). Medication therapy changes occurred in 25 patients, including 15 adjustments for anticoagulation, six for hemodynamic drugs, and one for antiarrhythmic medication.
Forty-three patients had procedural or referral changes after CE. Before CE, additional imaging was noted to be indicated in 42 patients. After CE, the referring physicians stated that 38 of these patients (90%) no longer needed additional imaging tests. Two patients had new indications for cardiac magnetic resonance (CMR) imaging because of the abnormalities revealed by CE (hypertrophic cardiomyopathy in one case and suspected sarcoidosis in the other). Other procedural changes involved ICDs ( n = 6), LV assist devices ( n = 3), life vest ( n = 1), right heart catheterization ( n = 1), endomyocardial biopsy ( n = 1), and rearrangement of follow-up intervals ( n = 3). Referral changes occurred in 27 cases, including 13 new or cancellations of referrals for electrophysiology or HF, eight cases with referral to catheterization for assessing myocardial ischemia or right heart pressures, and six cases with new or cancellation of referral to cardiology.
Impact of CE on Clinical Management in Patients with Different Baseline Characteristics
The impact of CE on overall clinical management in patients with different pre-CE baseline characteristics is summarized in Table 4 . There was no difference in the proportion of clinical management changes in inpatients versus outpatients (60.0% vs 48.1%, P = .225). There was no difference in patients with or without prior echocardiographic examinations (52.6% vs 49.4%, P = .761). There was a trend toward significance in patients with ICD or pacemaker implantation (67.9% vs 47.9%, P = .064). Indications did not significantly influence changes in management, although subjects with HF, cardiomyopathy, and arrhythmia had a statistically significantly higher proportion of changes (61.4% vs 44.2% [ P = .031], 62.5% vs 45.0% [ P = .028], and 72.0% vs 47.7% [ P = .030], respectively). The most significant impact was seen with the lower pre-CE LVEF. Changes in clinical management in different pre-CE LVEF groups are shown in Figure 5 . The incidence of clinical management change increased as pre-CE LVEF decreased. LVEF < 30% and LVEF 30% to 50% were associated with a larger proportion of management changes compared with LVEF > 50% (71.4% and 59.1% vs 31.4%, P < .0001). Logistic regression showed that pre-CE LVEF < 50% was the only significant predictor of change of management after contrast ( P = .004; Table 5 ).
|Baseline characteristic/indication||Number of subjects||Percentage of management change||χ 2 P value|
|Prior transthoracic echocardiographic results|
|History of ICD/PM implantation|
|Coronary artery disease|