Image enhancement by noncontrast harmonic echocardiography. Part II. Quantitative assessment with use of contrast-to-speckle ratio

Postprocessing Approaches for the Improvement of Cardiac Ultrasound B-Mode Images: A Review

The improvement in the quality and diagnostic value of ultrasound images has been an ongoing research theme for the last three decades. Cardiac ultrasound suffers from a wide range of artifacts such as acoustic noise, shadowing, and enhancement. Most artifacts are a consequence of the interaction of the transmitted ultrasound signals with anatomic structures of the examined body. Structures such as bone, lungs (air), and fat have a direct limiting effect on the quality of the acquired images. Furthermore, physical phenomena such as speckle introduce a granular pattern on the imaged tissue structures that can sometimes obscure fine anatomic detail. Over the years, numerous studies have attempted to address a range of artifacts in medical ultrasound, including cardiac ultrasound B-mode images. This review provides extensive coverage of such attempts identifying their limitations as well as future research opportunities.

Effect of tissue harmonic imaging and contrast upon between observer and test-retest reproducibility of left ventricular ejection fraction measurement in patients with heart failure

AIMS

To investigate the effects of tissue harmonic imaging (THI) and contrast chamber opacification (LVO) upon measurement variability and reproducibility of echocardiographic left ventricular (LV) volume and ejection fraction (EF) measurements in patients with heart failure (HF).

BACKGROUND

Echocardiography is often used in HF patients to determine LV volumes and EF. However, current echo methods are variable and may not be applicable for repeat testing in individual patients. THI and LVO have both been shown to improve endocardial visualisation, but it remains to be determined whether this results in better measurement reproducibility.

METHODS

Thirty-one HF patients and 30 control subjects underwent echocardiography on two separate days. LV volumes were measured under four different imaging conditions: fundamental, THI, LVO and LVO with ECG-triggered Power Doppler. Chamber opacification, pulmonary transit time (PTT), endocardial enhancement, reproducibility and bias were assessed.

RESULTS

Chamber opacification was inferior and the PTT longer in the HF patients. PTT was related to LV volumes, EF, jugular venous pressure and mitral filling pattern. THI improved endocardial visualisation, and although LVO improved endocardial visualisation in the controls, it offered no benefit over THI in the HF patients. LV volumes and EF were different for each method and THI was the least variable method for repeat measurements.

CONCLUSIONS

THI improved endocardial visualisation and was the least variable of the techniques. LVO offered no further advantage in patients with HF and thus cannot be routinely advocated and since LV volumes and EF were different for each, these methods are neither comparable nor interchangeable for follow-up assessments.

Comparison of Echocardiography Using Tissue Harmonics and Contrast Harmonics with Radionuclide Angiography for the Assessment of Left Ventricular Function

Background

Left ventricular ejection fraction (LVEF) is a significant predictor of morbidity and mortality; however, the optimal noninvasive modality for the quantitative determination of LVEF is not apparent.

Hypothesis

We verified the hypothesis that the various echocardiographic methods of assessing LVEF using the Method of Discs with contrast (Optison human albumin microspheres; Amersham Health, Princeton, NJ) and visual assessment of LVEF using tissue harmonics and contrast harmonics compare favorably with radionuclide angiography (RNA).

Methods

In a prospective analysis, 24 consecutive patients scheduled to undergo RNA had an echocardiogram using tissue harmonics and contrast harmonics on the same day. LVEF was assessed by RNA by an experienced, blinded reader using manual determination of the region of interest. LVEF was calculated using the Method of Discs (Simpson's Rule) by a blinded sonographer. LVEF was visually estimated by two blinded readers using echocardiography with tissue harmonics and contrast harmonics on separate occasions.

Results

By linear regression analysis, LVEF determination by echocardiography with contrast using the Method of Discs correlated well with RNA (r = .835, p < .0005). Using Bland-Altman analysis, the second echocardiogram reader had excellent agreement with RNA, whereas the first reader had a mean difference of 5.25% (CI 1.3–9.2; p = .012) with visual assessment using tissue harmonics and a mean difference of 4.67% (CI 0.4–8.8; p = .031) with visual assessment using contrast harmonics compared with RNA. Thus, a small difference in agreement between RNA and echocardiographic visual estimation was noted that appeared to be primarily reader dependent.

Conclusions

LVEF determination with echocardiography with contrast using the Method of Discs correlated well with RNA and provided agreement across a range of cardiac functions. Visual echocardiographic assessment of LVEF with both tissue harmonics and contrast harmonics correlated well with RNA, but contrast harmonics did not appear to offer an advantage over tissue harmonics alone.

Interpretation of Left Ventricular Wall Motion During Stress Testing

This article describes the obstacles to stress echocardiographic interpretation, and reviews the techniques currently available that offer a more objective approach to stress wall motion analysis than the conventional visual methodology. These techniques include Doppler-based methods, such as myocardial Doppler velocity and strain rate imaging, as well as automated border detection techniques, such as acoustic quantification and color kinesis.

Integrated backscatter during harmonic and fundamental frequency imaging--effect of depth, mechanical index, and tissue anisotropy: implications for myocardial tissue characterization

OBJECTIVE

To explore the potential advantages of tissue harmonic imaging (THI) versus fundamental frequency imaging (FFI) when applied to tissue characterization.

METHODS

A Philips Medical Systems Sonos 5500 echocardiograph equipped with a broadband transducer (S4) and an on-line quantitative analysis software package (Acoustic Densitometry) was used for imaging. The effect of mechanical index (MI), imaging depth, and anisotropy on relative backscatter amplitude was evaluated.

RESULTS

This study demonstrated that imaging with tissue harmonics generated relatively greater backscatter values at clinically relevant imaging depths and instrument settings referenced to FFI. This effect was dependent on MI setting. A direct relationship between backscatter amplitude and MI was demonstrated. Additionally, tissue anisotropy had similar effects on integrated backscatter amplitude during both THI and FFI. However, relative backscatter values at each fiber orientation are greater during THI at similar instrument settings when referenced to FFI.

CONCLUSION

Tissue harmonic imaging may offer advantages over FFI for myocardial tissue characterization.

Rapid freehand scanning three-dimensional echocardiography: accurate measurement of left ventricular volumes and ejection fraction compared with quantitative gated scintigraphy

This study was performed to assess clinical feasibility of rapid freehand scanning 3-dimensional echocardiography (3DE) for measuring left ventricular (LV) end-diastolic and -systolic volumes and ejection fraction using quantitative gated myocardial perfusion single photon emission computed tomography as the reference standard. We performed transthoracic 2-dimensional echocardiography and magnetic freehand 3DE using a harmonic imaging system in 15 patients. Data sets (3DE) were collected by slowly tilting the probe (fan-like scanning) in the apical position. The 3DE data were recorded in 10 to 20 seconds, and the analysis was performed within 2 minutes after transferring the raw digital ultrasound data from the scanner. For LV end-diastolic and -systolic volume measurements, there was a high correlation and good agreement (LV end-diastolic volume, r = 0.94, P <.0001, standard error of the estimates = 21.6 mL, bias = 6.7 mL; LV end-systolic volume, r = 0.96, P <.0001, standard error of the estimates = 14.8 mL, bias = 3.9 mL) between gated single photon emission computed tomography and 3DE. There was an overall underestimation of volumes with greater limits of agreement by 2-dimensional echocardiography. For LV ejection fraction, regression and agreement analysis also demonstrated high precision and accuracy (y = 0.82x + 5.1, r = 0.93, P <.001, standard error of the estimates = 7.6%, bias = 4.0%) by 3DE compared with 2-dimensional echocardiography. Rapid 3DE using a magnetic-field system provides precise and accurate measurements of LV volumes and ejection fraction in human beings

Harmonic ultrasound: a review

Harmonic ultrasound is a technique based on the principle of transmitting at frequency f and receiving at frequency 2f (or 1/2f). This technology has become available through the development of wide-bandwidth transducers. Microbubble contrast media produce a large amount of harmonic signal. Contrast harmonic ultrasound provides the opportunity to image patterns of high flow vasculature and overall perfusion. Regions of poor perfusion, including necrosis or infarction, can be identified with contrast harmonic ultrasound. While proportionately lower, tissues also produce harmonic signals. Tissue harmonic ultrasound sequences often improve subjective image quality compared to fundamental ultrasound in echocardiographic and abdominal examinations. This review will discuss the physical principles of harmonic ultrasound signal generation, medical and animal research applications, and an overview of current veterinary experiences.

Radio frequency dual-spectra analysis of regional myocardial perfusion: Comparison with harmonic densitometric method

Our objective was to compare harmonic-to-fundamental frequency ratio peak (HFRp) analysis with conventional harmonic gray-scale densitometric analysis on the basis of ability to eliminate heterogeneity in ultrasound signals. Broadband radio frequency and harmonic data were obtained by using intermittent short-axis scans in 10 open-chest pigs before and during infusion of contrast microbubbles. HFRp and gray-scale intensity values were measured in 6 segments of left ventricular myocardium. In baseline images, the influence of anisotropy on HFRp values was significantly less than that in gray-scale intensities. In perfusion assessment with subtraction, contrast heterogeneity in HFRp values was significantly smaller than that in gray-scale intensities. The increase in HFRp values after subtraction was significantly greater than that in gray-scale intensities in lateral (P <.001), posterior (P <.0001), and inferior (P <.01) myocardium. HFRp analysis can compensate for baseline myocardial anisotropy and regional contrast heterogeneity. With background subtraction, HFRp analysis allows better quantification of myocardial perfusion.

Contrast echocardiography improves the accuracy and reproducibility of left ventricular remodeling measurements: a prospective, randomly assigned, blinded study

OBJECTIVES

We sought to assess the impact of contrast injection and harmonic imaging, on the measure by echocardiography of left ventricular (LV) remodeling.

BACKGROUND

Left ventricular remodeling is a precursor of LV dysfunction, but the impact of contrast injection and harmonic imaging on the accuracy or reproducibility of echocardiography is unclear.

METHODS

We prospectively collected LV images by using simultaneous methods. Then, LV volumes were measured off-line, in blinded manner and in random order. The accuracy of echocardiography was determined in comparison to electron beam computed tomography (EBCT) in 26 patients. The reproducibility of echocardiography was assessed by three blinded observers with different training levels in 32 patients.

RESULTS

End-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV) and ejection fraction (EF), as measured by EBCT (195 +/- 55, 58 +/- 24 and 137 +/- 35 ml and 71 +/- 5%, respectively) and echocardiography with harmonic imaging and contrast injection (194 +/- 51, 55 +/- 20 and 140 +/- 35 ml and 72 +/- 4%, respectively), showed no differences (all p > 0.15) and excellent correlations (all r > 0.87). In contrast, echocardiography using harmonic imaging without contrast injection underestimated the EBCT results (all p < 0.01). Reproducibility was superior with rather than without contrast injection for intraobserver and interobserver variabilities (all p < 0.001). Values measured by different observers were different without contrast injection, but were similar with contrast injection (all p > 0.18). Consequently, intrinsic patient differences represented a larger and almost exclusive proportion of global variability with contrast injection for EDV (94 vs. 79%), ESV (93 vs. 82%), SV (87 vs. 53%) and EF (84 vs. 41%), as compared with harmonic imaging without contrast injection (all p < 0.005).

CONCLUSIONS

For assessment of LV remodeling, echocardiography with harmonic imaging and contrast injection improved the accuracy and reproducibility, as compared with imaging without contrast injection. With contrast injection, variability was almost exclusively due to intrinsic patient differences. Therefore, when evaluation of LV remodeling is deemed important, assessment after contrast injection should be the preferred echocardiographic approach.

Three-dimensional echocardiography with tissue harmonic imaging shows excellent reproducibility in assessment of left ventricular volumes

We studied the reproducibility of repeated measurements of left ventricular (LV) volumes by 2-dimensional (biplane method of disks) and 3-dimensional echocardiography (coaxial scanning) with tissue harmonic imaging. Ten healthy subjects underwent estimation of LV volumes by transthoracic echocardiography twice within 1 week by 2 different operators to investigate interexamination and operator variance. In addition, the analysis of LV volume was done manually by 2 observers to assess both interobserver and intraobserver variances. With 3D echocardiography, observer variation had the greatest impact on variance. Operator variability showed important contributions to total variance with the use of 2D echocardiography. The reproducibility of 3D echocardiography and tissue harmonic imaging is excellent and comparable to magnetic resonance imaging techniques; 3D echocardiography therefore should provide a powerful tool for noninvasive LV volume estimation.

Contrast dobutamine stress echocardiography: clinical practice assessment in 300 consecutive patients

In this study we compared non-contrast imaging with contrast imaging of the left ventricle during dobutamine stress echocardiography (DSE). Wall segment visualization, image quality, and confidence of interpretation were determined with and without the use of intravenous Optison, a second-generation echocardiographic contrast agent, in 300 consecutive patients undergoing rest and peak DSE. At rest and at peak stress, the percentage of wall segments visualized, image quality, and confidence of interpretation were better with contrast compared with non-contrast imaging. No significant decrease was seen in wall segment visualization, image quality, or confidence of interpretation from rest to peak stress in images obtained with contrast, unlike the images obtained without contrast from rest to peak stress. The use of the intravenous echocardiographic contrast agent Optison during DSE significantly improved wall segment visualization and image quality at rest and at peak stress, resulting in improved confidence of interpretation.

Measurement of left ventricular volumes by 3-dimensional echocardiography with tissue harmonic imaging: a comparison with magnetic resonance imaging

We hypothesized that tissue harmonic imaging (THI) in comparison with fundamental imaging (FI) would improve endocardial border detection, and therefore in combination with 3-dimensional echocardiography (3D echo), it would be a precise method for left ventricular (LV) volume measurement. Ten healthy subjects and 18 consecutive patients with dilated hearts underwent estimation of LV volumes by magnetic resonance imaging (MRI) and transthoracic 3D echo with THI and FI. In patients, the agreement between MRI and 3D echo was closer with THI in comparison with FI for assessment of LV volumes. Thus the mean +/- 2 SD of differences between MRI and 3D echo with THI versus FI, respectively, was -6.4 +/- 40.0 mL versus -17.4 +/- 57.6 mL (P <.01) for the end-diastolic volume (EDV), and 0.0 +/- 26.6 mL versus -8.1 +/- 35.6 mL (P <.01) for the end-systolic volume (ESV). In patients, THI in comparison with FI approximately halved observer variation on EDV and ESV. In healthy subjects, only ESV showed significantly reduced observer variation by THI. In conclusion, because THI demonstrated a clinically relevant reduction in observer variation and a closer agreement to the MRI technique in patients with dilated hearts, it should replace FI in LV volume measurements.

Qualitative and quantitative effects of harmonic echocardiographic imaging on endocardial edge definition and side-lobe artifacts

Harmonic imaging is a new ultrasonographic technique that is designed to improve image quality by exploiting the spontaneous generation of higher frequencies as ultrasound propagates through tissue. We studied 51 difficult-to-image patients with blinded side-by-side cineloop evaluation of endocardial border definition by harmonic versus fundamental imaging. In addition, quantitative intensities from cavity versus wall were compared for harmonic versus fundamental imaging. Harmonic imaging improved left ventricular endocardial border delineation over fundamental imaging (superior: harmonic = 71.1%, fundamental = 18.7%; similar: 10.2%; P <.001). Quantitative analysis of 100 wall/cavity combinations demonstrated brighter wall segments and more strikingly darker cavities during harmonic imaging (cavity intensity on a 0 to 255 scale: fundamental = 15.6 +/- 8.6; harmonic = 6.0 +/- 5.3; P <.0001), which led to enhanced contrast between the wall and cavity (1.89 versus 1.19, P <.0001). Harmonic imaging reduces side-lobe artifacts, resulting in a darker cavity and brighter walls, thereby improving image contrast and endocardial delineation.

Does second harmonic imaging improve left ventricular endocardial border identification at higher heart rates during dobutamine stress echocardiography?

BACKGROUND

The increased heart rate during dobutamine stress echocardiography (DSE) may impair endocardial border visualization. Second harmonic imaging (SHI) enhances left ventricular (LV) border visualization compared with conventional fundamental imaging (FI) at rest. However, its role during DSE is not well established yet.

OBJECTIVE

Our objective was to compare the additional value of SHI to FI for the LV endocardial border visualization during various stages of DSE.

METHODS

Eighty patients underwent DSE. Imaging was performed with both FI and SHI at rest and at low-and peak-dose dobutamine infusion. Endocardial border visualization was assessed by using a 16-segment/3-point score (0 = well visualized; 1 = poorly visualized; 2 = not visualized).

RESULTS

Heart rate increased from rest (70 +/- 13 bpm) to low-dose dobutamine (77 +/- 17, P <.01) and showed further increase at peak dose (129 +/- 16, P <.001 versus low dose). There was a higher prevalence of segments with an invisible LV endocardial border with FI compared with SHI at rest (9.4% versus 6.2%, P <.0001), at low dose (10.8% versus 6.3%, P <.0001), and at peak dose (15.0% versus 8.2%, P <.0001). There was an increase in the number of segments with an invisible border from rest to peak stress by FI (P =.0001), whereas the difference was less significant for SHI (P =.07).

CONCLUSION

Second harmonic imaging improves visualization of the LV endocardial border compared with FI during DSE. The advantage of SHI over FI is more marked at higher heart rates than at rest.

Enhanced detection of left atrial spontaneous echo contrast by transthoracic harmonic imaging in mitral stenosis

BACKGROUND

Spontaneous echo contrast (SEC) of the left atrium is associated with increased risk of thromboembolism in patients with mitral stenosis (MS). The determination of the presence and severity of left atrial (LA) SEC is of prognostic importance in these patients. Harmonic imaging (HI), a novel echocardiographic technique that differs from conventional fundamental imaging (FI) in that it involves transmitting ultrasound at one frequency and receiving at twice the transmitted frequency, produces better endocardial border definition and myocardial opacification. However, there are no data about its value for the detection of LA SEC. The purpose of this study was to investigate the utility of transthoracic noncontrast tissue HI in the detection of LA SEC in patients with MS.

METHODS

Seventy-four consecutive patients with MS (49 women, mean age 51 years) underwent standard transthoracic echocardiography (TTE) in both HI and FI modes and transesophageal echocardiography (TEE) to determine the presence and severity of LA SEC. Left atrial SEC was graded by TEE as either mild (only seen at high gain) or severe (visible in the entire left atrium at normal gain control of the equipment). The control group comprised 30 patients randomly selected from patients who did not have LA SEC at the TEE examination.

RESULTS

Atrial fibrillation was found in 46 patients (62.2%). The mean mitral valve area and mean mitral gradient were 1.0+/-0.3 cm(2) and 8.2+/-4.1 mm Hg, respectively. Nine patients (12.2%) had episodes of systemic embolism; 8 had stroke, and 1 had peripheral embolism. Left atrial thrombus was found in 11 patients (14.9%) by TEE. Left atrial SEC was present in all but one patient by TEE (mild in 35 patients, severe in 38). Fundamental imaging with TTE, however, revealed LA SEC in only 5 (6.8%) of the 73 patients. In contrast, with HI, LA SEC could be detected in 63 (86.3%) patients. In the detection of severe LA SEC, the sensitivities of FI and HI were 13.2% (5/38) and 100% (38/38), respectively. Left atrial SEC was not observed in control subjects by either FI or HI.

CONCLUSIONS

Transthoracic HI significantly enhances the detection of LA SEC in patients with MS.

Clinical impact of second harmonic imaging and left heart contrast in echocardiographic stress testing

Second harmonic imaging and left heart contrast agents are recent echocardiographic advancements that enhance the assessment of wall motion. Because little information exists concerning their clinical impact on echocardiographic stress testing in daily practice, this was determined for 9-month periods before (1997) and after (1998) their introduction. Harmonic imaging was used in all patients after its introduction. A second generation intravenous left heart contrast agent (Optison) was used at the discretion of the sonographer and physician team. Both exercise and dobutamine stress tests were included. At the time of study interpretation, diagnostic confidence was assigned as high, medium, or low. For all patients who underwent coronary angiography < or = 6 months after stress testing, the diagnostic accuracy was determined (true positive plus true negative/total studies). There were 574 studies before and 746 studies after implementation. Optison was used in 28% of the harmonic imaging studies. Study cancellations due to uninterpretable images fell from 6.4% to 1.2% (p <0.001) despite a more obese population completing testing (body mass index: 29 +/- 7 to 31 +/- 8 kg/m2, p = 0.02), whereas high diagnostic confidence increased from 55% to 64% (p <0.001). For the 7% of patients who underwent cardiac catheterization, the diagnostic accuracy remained unchanged (74 vs 73%) although a prior negative stress test was less common (40% to 20% p = 0.04). Thus, these new technologies had a favorable clinical impact.

Tissue harmonic imaging: experimental analysis of the mechanism of image improvement

Tissue harmonic scanning visually improves echocardiographic image quality. The aim of the present study was to objectively assess the improvement in harmonic image quality under controlled laboratory conditions. A tissue-mimicking phantom that contained 8-mm-diameter cystic lesions at depths ranging from 2 to 12 cm was used. Harmonic scans (1.7 MHz transmit, 3.4 MHz receive) of the phantom were obtained and lesion detectability was compared to that in scans acquired with 2 fundamental frequencies (2.0 and 3.3 MHz). A 2 cm-thick ethanol layer was also used to simulate the nonlinear effect of human fat. Cyst detectability was quantified by measurement of the contrast-to-speckle ratio (CSR). The results indicated no significant difference in the CSR between harmonic and fundamental images obtained without the ethanol layer. With images obtained with the ethanol layer, a relative increase of the CSR during harmonic imaging was observed with respect to fundamental imaging (p<0.05). In conclusion, a fat layer, here simulated by ethanol, plays a significant role in determining the resulting image quality. Without this layer, the contribution of the second harmonic mode was not significant. Thus, in a slim patient, the harmonic mode may not be as beneficial to image improvement as in an obese patient.

Advances in Exercise Echocardiography Can This Technique Still Thrive in the Era of Pharmacologic Stress Testing?

In the current literature, pharmacologic stress techniques are the focus of interest and excitement regarding new technologies and new indications such as the diagnosis of viable myocardium. In contrast, exercise echocardiography has evolved less and is less amenable to the introduction of new technologies. This article reviews the indications for exercise echocardiography (especially in contrast to pharmacologic stress), its accuracy relative to other testing, and application to clinical decision making. Exercise echocardiography remains to be well accepted as a diagnostic and risk-assessment technique, and in some clinical situations it provides valuable data that are not available during pharmacologic stress testing.