| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, RWTH Aachen University Hospital,
Pauwelsstr. 30, Aachen D 52057, Germany.
2 Institute for Biometrics, RWTH Aachen University Hospital, Aachen D 52057,
Germany.
Received May 21, 2002;
accepted after revision November 15, 2002.
Address correspondence to T. Schmidt.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. For our prospective study, 114 patients underwent sonography of the kidneys in both modes, fundamental B-mode sonography and phase-inversion tissue harmonic imaging, in a randomly chosen scanning order. Imaging parameters were standardized. Sonographic diagnoses were made under real-time conditions by the examining radiologist. All sonographic diagnoses were compared with a diagnostic reference modality: contrast-enhanced CT, contrast-enhanced MR imaging, or histopathology. Three radiologists different from the examiners evaluated overall image quality, lesion conspicuity, and fluid–solid differentiation for both modalities using hard-copy images.
RESULTS. In 70 patients, fundamental B-mode sonography as the first technique depicted 73 of 111 lesions 10 mm or larger and enabled 71 lesions to be correctly characterized (sensitivity, 65.8%; accuracy, 64.0%). As the first mode, phase-inversion tissue harmonic imaging depicted 57 of 65 focal lesions and enabled 54 lesions to be accurately classified in 44 patients (sensitivity, 87.7%; accuracy, 83.1%). The differences in sensitivity and accuracy were statistically significant (95% confidence interval). For overall image quality, lesion conspicuity, and fluid–solid differentiation phase-inversion harmonic imaging was superior to fundamental B-mode sonography (p < 0.0001).
CONCLUSION. Phase-inversion tissue harmonic imaging is superior to fundamental B-mode sonography in the sonography of focal kidney lesions because phase-inversion tissue harmonic imaging has better overall image quality, lesion conspicuity, and fluid–solid differentiation. In six cases, phase-inversion tissue harmonic imaging added crucial diagnostic information that changed patient management.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Harmonic frequencies are whole-numbered multiples of the fundamental frequencies that are transmitted by the sonography probe. These frequencies originate from the nonlinear propagation of the fundamental frequencies through human tissue. In tissue harmonic imaging, the echoes of the harmonic frequencies are used for imaging, whereas fundamental B-mode sonography uses the echoes of only the fundamental frequencies [5]. Theoretic advantages of tissue harmonic imaging are better lateral and axial resolution, enhanced signal-to-noise ratio, and reduced artifacts, which results in less degradation of sonographic images. In tissue harmonic imaging, the fundamental portion of the sonographic echoes is separated from the harmonic portion by frequency filters to exclusively use the harmonic echoes for image creation [5].
Frequency filters have some limitations, however. In the wide frequency band, the fundamental and harmonic frequencies overlap substantially in the midportion of the frequency spectrum. This overlap causes degradation of low harmonic frequencies by high fundamental frequencies, which results in a lower signal-to-noise ratio. For this frequency overlap to be avoided, the bandwidths of the fundamental and, consequently, the harmonic ultrasound beam are narrowed. However, narrowing the frequency band for image creation results in lowered dynamic and gray-scale range. In tissue harmonic imaging, this narrowed frequency band accounts for deficits in dynamic range, which is the system's capability to differentiate structures with only slightly differing echogenicities. This limitation results in an image impression of a hard, partly blurring contrast between anatomic structures.
Some authors have identified reduced dynamic range as a problem in analyzing the internal details of solid processes [6, 7, 8, 9]. For the drawbacks of narrow bandwidth and low dynamic range to be overcome, phase-inversion tissue harmonic imaging is performed using the phase-inversion technique. This technique adds two incoming pulses that are 180° phase-shifted, the linear fundamental portions of the sonography pulse eliminate each other, and the nonlinear harmonics add to a signal that can be used for image creation. This strategy allows wideband fundamental frequencies to be transmitted and wideband harmonic frequencies to be received [10].
In previous studies, tissue harmonic imaging and phase-inversion tissue harmonic imaging were proven superior for sonography of the heart, vascular system, liver, biliary system, and pancreas and in gynecologic examinations [6, 7, 9, 11, 12, 13, 14, 15]. Harmonic imaging modes provide superior image quality, fewer artifacts, and better delineation of normal and abnormal structures than fundamental B-mode sonography. These improved capabilities add diagnostic confidence and relevant information [6, 11, 12, 13, 14]. Studies have shown that phase-inversion tissue harmonic imaging is superior to tissue harmonic imaging [8, 15]. Jang et al. [8] found better detectability and resolution for internal characteristics of cystic and solid hepatic lesions on phase-inversion tissue harmonic imaging than on tissue harmonic imaging and fundamental B-mode sonography. Hong et al. [15] showed that phase-inversion tissue harmonic imaging offers the best image quality and the fewest artifacts compared with tissue harmonic imaging and fundamental B-mode sonography when imaging the gallbladder.
To our knowledge, no systematic study to date has compared fundamental B-mode sonography with phase-inversion tissue harmonic imaging in the diagnosis of focal renal lesions. In this study, we compared the two modalities in terms of sensitivity, accuracy of lesion classification, and specificity. We also looked for differences between phase-inversion tissue harmonic imaging and fundamental B-mode sonography in patients with clinically relevant processes—malignant or inflammatory—that required further diagnostic or therapeutic measures (i.e., contrast-enhanced CT or contrast-enhanced MR imaging). Furthermore, overall image quality, lesion conspicuity, and fluid–solid differentiation were evaluated to characterize the differences in diagnostic performance of both modalities.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sonographic Techniques, Evaluations, and Statistics
In the prospective part of the study, patients were examined by one of two
radiologists experienced in abdominal sonography. Informed consent was
obtained. The kidneys of all patients were examined with both fundamental
B-mode sonography and phase-inversion tissue harmonic imaging in a randomly
chosen order. Seventy patients were examined with fundamental B-mode
sonography first and phase-inversion tissue harmonic imaging second; in 44
patients, the scanning order was reversed. The examiner had no knowledge of
the diagnoses or any clinical data.
A Sonoline Elegra unit (software version 6.0; Siemens, Erlangen, Germany) with a 3.5-MHz curved array sonography probe was used for all the examinations. A frequency of 3.4 MHz was used for fundamental B-mode sonography, and a frequency combination of 2.0 and 4.0 MHz was used for phase-inversion tissue harmonic imaging. Imaging parameters were standardized. Gain and focal location could be adjusted. The imaging mode was changed by a button on the sonography machine. Focal processes were documented regarding localization, size, and diagnosis based on the sonographic examination. For each modality, we sonographically classified lesions using generally accepted criteria [16, 17, 18]. Especially in the differentiation between complicated and simple cysts, we used characteristics as previously described [17, 19, 20] (Table 1).
|
The sonographic diagnoses were compared with the gold standard in diagnostic imaging for the characterization of focal renal abnormalities—either contrast-enhanced CT or contrast-enhanced MR imaging—or with the histopathologic diagnosis, when available [17, 18].
The sensitivity of fundamental B-mode sonography and the sensitivity of phase-inversion tissue harmonic imaging were determined by counting all detected lesions regardless of the accurate diagnoses. To determine the accuracy of each classification (i.e., simple cyst, complicated cystic lesion, solid tumor [Table 1], inflammatory process, calcification, renal scar, urinary obstruction, or fluid collection), we determined whether the abnormality was categorized correctly in terms of sonographic criteria and whether the classification was consistent with the gold standard. To calculate specificity, we counted the number of kidneys without any focal abnormality accurately shown on fundamental B-mode sonography and phase-inversion tissue harmonic imaging. Thus, a kidney without a focal lesion was included even if the contralateral kidney showed a focal abnormality.
On the basis of each individual sonographic diagnosis of fundamental B-mode sonography and of phase-inversion tissue harmonic imaging, the examiner determined the need for further diagnostic or therapeutic measures for each lesion. This decision was made separately for both sonographic modalities.
Focal processes were regarded as requiring additional diagnostic or therapeutic measures if lesions that had solid tumor characteristics could not safely be classified as benign—that is, a complicated cyst (Table 1)—or if the kidneys showed changes possibly caused by another abdominal disease, such as perirenal fluid in cases of pancreatitis. In turn, unequivocally benign processes such as simple cysts, parenchymal calcifications without signs of obstruction, and renal scars were assigned to the group described as "no necessity for further workup." These decisions were made by the examining radiologist immediately after the examination and evaluated using the results of the gold standard.
In the statistical analysis, we compared the results for sensitivity, accuracy of classification, and specificity of fundamental B-mode sonography and phase-inversion tissue harmonic imaging as a first scanning modality. Subsequently, we compared fundamental B-mode sonography and phase-inversion tissue harmonic imaging when they were used consecutively as the first and second modality and vice versa. We determined whether diagnostic information was gained when fundamental B-mode sonography or phase-inversion tissue harmonic imaging was used as the second modality. For lesions of all sizes and lesions 10 mm or larger in diameter, we used a confidence interval (CI) of 95% for the analysis of sensitivity, accuracy of classification, specificity, and correct decision about whether further workup was needed. For the analysis of subgroups differentiated by size (1–9, 10–19, 20–29, and 30 mm or larger), we used a CI of 98.75% taking into consideration that we analyzed four subgroups of the basic study group.
In the retrospective part of the study, three radiologists different from the sonographic examiners evaluated hard copies of the examinations. The radiologists were unaware of which imaging modality was performed and of patient data; images of the same patient obtained with phase-inversion tissue harmonic imaging and with fundamental B-mode sonography were independently evaluated without direct comparison. Using three-point grading schemes, three radiologists reviewed the images and evaluated the following criteria: overall image quality, characterizing whether a diagnosis could sufficiently be made with the performed examination; lesion conspicuity, measuring contrast between anatomic and pathologic structures; and fluid–solid differentiation within the lesion (Table 2). Each qualitative grading scheme was evaluated using the symmetry test of Bowker, which is a McNemar test not restricted to a fourfold or 2 x 2 table data analysis. Probabilities were calculated with a significance level of 0.05.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
First Scanning Mode
When lesions 10 mm or larger in diameter were considered, phase-inversion
tissue harmonic imaging as the first scanning modality was significantly
better than fundamental B-mode sonography as the first mode regarding
sensitivity (Table 3) and
accuracy (Table 4) of
classification (95% CI). Including lesions 9 mm or smaller, the differences
were not statistically significant. Additionally, Tables
3 and
4 show that the greatest
advantage for phase-inversion tissue harmonic imaging in sensitivity and
accuracy of classification was in lesions measuring 10–19 mm. Both
modalities showed low sensitivities and proportions of accurate
classifications for processes smaller than 10 mm (Tables
3 and
4).
|
|
First modality, fundamental B-mode sonography and second modality, phase-inversion tissue harmonic imaging.—When phase-inversion tissue harmonic imaging was used as the second mode after scanning with fundamental B-mode sonography, phase-inversion tissue harmonic imaging was significantly better in depicting and allowing classification of lesions 10 mm or larger in diameter. The greatest advantage for phase-inversion tissue harmonic imaging was in lesions 10–19 mm in diameter (Tables 3 and 4).
With fundamental B-mode sonography as the first mode, the examiners detected two uncomplicated cysts (5 and 10 mm in diameter) not seen on phase-inversion tissue harmonic imaging. In turn, phase-inversion tissue harmonic imaging as the second mode discovered 28 additional focal processes with a mean diameter of 10.9 mm (SD, 5.1 mm; range, 4–25 mm) that had not been revealed on fundamental B-mode sonography. Among these 28 lesions were 25 uncomplicated cysts and three renal lesions needing further diagnostic workup.
Three lesions with the necessity for further diagnostic workup that were discovered but falsely classified on the basis of fundamental B-mode sonography as the first mode were accurately diagnosed on phase-inversion tissue harmonic imaging. The diagnoses based on both modes—fundamental B-mode sonography and phase-inversion tissue harmonic imaging—resulted in recommendations for contrast-enhanced CT.
First modality, phase-inversion tissue harmonic imaging and second modality, fundamental B-mode sonography.—When lesions 10 mm or larger were considered, phase-inversion tissue harmonic imaging as the first sonographic mode had a higher sensitivity and accuracy of classification than fundamental B-mode sonography as the second mode (Tables 3 and 4), but the difference was not statistically distinct. With phase-inversion tissue harmonic imaging, the radiologist examiners discovered and correctly classified 16 processes (mean diameter, 13.3 mm; SD, 8.8 mm; range, 4–35 mm) including three malignant processes that could not be found afterward on fundamental B-mode sonography. One parapelvic cyst diagnosed on phase-inversion tissue harmonic imaging was misinterpreted as a regional dilatation of the renal pelvis using fundamental B-mode sonography. One uncomplicated cyst was correctly diagnosed on phase-inversion tissue harmonic imaging, whereas internal echoes on fundamental B-mode sonography led the examiner to diagnose this finding as a complicated cyst. Fundamental B-mode sonography as the second scanning mode did not depict any lesion that was not shown on phase-inversion tissue harmonic imaging.
Lesions Requiring Additional Diagnostic or Therapeutic Measures
In 41 patients, 50 lesions with a mean diameter of 38.6 mm (SD, 33.9 mm;
range, 3–140 mm) needed further diagnostic or therapeutic measures. This
group comprises 26 solid tumors, 15 complicated cysts, one renal abscess, and
three collections of perirenal fluid. In five patients, we found unilateral
urinary obstruction with dilatation of the renal pelvis. Five patients had two
lesions and two patients had three lesions that required further diagnostic or
therapeutic measures.
Of the 32 lesions requiring further diagnostic workup that first were examined on fundamental B-mode sonography, the examining radiologists detected 27 lesions and accurately characterized 23 with fundamental B-mode sonography (sensitivity, 84.4%; accuracy of classification, 71.9%). With phase-inversion tissue harmonic imaging as the first mode, 17 of 18 processes were detected, and 16 were classified accurately (sensitivity, 94.4%; accuracy of classification, 88.9%). With fundamental B-mode sonography, the examiners described the need for further workup in 25 lesions (78.1%), whereas further workup was recommended in 17 of 18 with phase-inversion tissue harmonic imaging (94.4%). These differences were not statistically significant (95% CI).
For lesions smaller than 30 mm, the difference between fundamental B-mode sonography and phase-inversion tissue harmonic imaging was greater: a sensitivity of 73.3% for fundamental B-mode sonography versus 88.9% for phase-inversion tissue harmonic imaging. The accuracy of classifications was 53.3% for fundamental B-mode sonography versus 77.8% for phase-inversion tissue harmonic imaging. A correct recommendation for a further workup was given in 66.7% with fundamental B-mode sonography and in 88.9% with phase-inversion tissue harmonic imaging. These results did not show statistical significance (95% CI).
Because seven patients had more than one clinically relevant process, we also analyzed the decision for a diagnostic workup with respect to the number of affected patients. With fundamental B-mode sonography, 22 (84.6%) of 26 patients who needed further workup were identified, whereas 15 (100%) of 15 patients who needed further workup were correctly identified with phase-inversion tissue harmonic imaging. This difference was not statistically distinct (95% CI). However, even when phase-inversion tissue harmonic imaging was used as the second mode, reviewers accurately recommended further diagnostic workup for the four patients who were missed when fundamental B-mode sonography was used as the first mode.
Lesions Not Detected or Falsely Classified by Both Modalities
Altogether 82 lesions with a mean diameter of 8.6 mm (SD, 7.8 mm; range,
3–50 mm) were not revealed by both modalities. In three
lesions—one renal cell carcinoma, one metastasis, and one complicated
cyst—further diagnostic evaluation was necessary. However, with
fundamental B-mode sonography, the examiners found additional suspicious renal
lesions in two of these three patients resulting in a recommendation for
further diagnostic workup. With phase-inversion tissue harmonic imaging, the
examiners referred all three patients for further diagnostic measures because
they found additional suspicious processes of the kidneys.
Five uncomplicated cysts were diagnosed as complicated cysts in both modes because of internal echoes. One complicated cyst that had a calcification without distal shadowing was falsely classified as a hyperechoic tumor. One solid tumor (10 mm in diameter) of unknown origin (breast cancer metastasis or hypernephroma) was falsely described as a complicated cyst using fundamental B-mode sonography and phase-inversion tissue harmonic imaging. However, CT was recommended with both modalities.
Specificity
Fundamental B-mode sonography and phase-inversion tissue harmonic imaging
showed a high specificity. As the first mode, fundamental B-mode sonography
identified 49 of 58 kidneys without focal abnormalities (specificity, 84.5%).
Phase-inversion tissue harmonic imaging as the first mode had a specificity of
97.2% and enabled correct description of 35 of 36 kidneys without any lesions.
The difference was not significant using a 95% CI.
Image Quality
All three radiologists rated phase-inversion tissue harmonic imaging as
better than fundamental B-mode sonography in terms of overall image quality,
lesion conspicuity, and fluid–solid differentiation. For several
examinations, the radiologists' ratings were the same for fundamental B-mode
sonography and phase-inversion tissue harmonic imaging. However, ratings of
overall image quality, lesion conspicuity, and fluid–solid
differentiation were more often excellent or good for phase-inversion tissue
harmonic imaging and unsatisfactory for fundamental B-mode sonography than
vice versa. Results were more frequently rated excellent for phase-inversion
tissue harmonic imaging and good for fundamental B-mode sonography than vice
versa. Overall image quality of phase-inversion tissue harmonic imaging was
rarely rated unsatisfactory, whereas fundamental B-mode sonography was
considered unsatisfactory in a considerable number of examinations. For all
three radiologists, the asymmetry of the distribution with a tendency for
better ratings for phase-inversion tissue harmonic imaging showed statistical
significance in the symmetry test of Bowker (p < 0.0001).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sensitivity and Accuracy of Classification
Previous clinical studies comparing tissue harmonic imaging with
fundamental B-mode sonography concentrated on criteria for image quality.
Hard-copy images were retrospectively reviewed for organ visibility, lesion
conspicuity, and diagnostic confidence for a focal abnormality
[6,
7,
8,
11,
15,
22]. Only a few reports have
directly compared sensitivity and accuracy of classification of tissue
harmonic imaging or phase-inversion tissue harmonic imaging with fundamental
B-mode sonography in a routine clinical setting. For focal abnormalities of
the liver, Tanaka et al. [9]
showed that tissue harmonic imaging is significantly better in depicting focal
masses in the liver. These researchers found the correct characterization of
focal lesions was also more often achieved with tissue harmonic imaging than
with fundamental B-mode sonography, especially for hepatocellular carcinoma in
cirrhotic parenchyma. Besides finding better image quality, Hann et al.
[12] found that tissue
harmonic imaging added relevant diagnostic data in 14 of 48 patients examined
for suspected liver disease.
In our study, we prospectively examined patients under real-time imaging conditions. The scanning order of fundamental B-mode sonography and phase-inversion tissue harmonic imaging was randomly chosen. Thus, we acquired independent figures for sensitivity and accuracy of classification for phase-inversion tissue harmonic imaging and fundamental B-mode sonography as the first scanning modality. We found that phase-inversion tissue harmonic imaging was significantly better than fundamental B-mode sonography for the detection and characterization of focal abnormalities of the kidney. Examining patients with phase-inversion tissue harmonic imaging as the second mode after scanning with fundamental B-mode sonography also provided a statistically significant gain in sensitivity and accuracy of classification. Conversely, fundamental B-mode sonography did not add diagnostic information after an examination with phase-inversion tissue harmonic imaging (Tables 3 and 4).
The lesions that were not detected with fundamental B-mode sonography but that were seen on phase-inversion tissue harmonic imaging as the second mode were relatively small (mean diameter, 10.9 mm; SD, 5.1 mm). The same applied to processes revealed by phase-inversion tissue harmonic imaging that fundamental B-mode sonography as second modality could not show (mean diameter, 13.3 mm; SD, 8.8 mm). Accordingly, the advantages of phase-inversion tissue harmonic imaging in sensitivity and accuracy of classification were greatest for lesions 10–19 mm in diameter. For fundamental B-mode sonography, our results were in accordance with those of Jamis-Dow et al. [23] and Warshauer et al. [24]. These researchers found sensitivities between 28% and 60% for renal parenchymal masses 10–20 mm in diameter. In our study, phase-inversion tissue harmonic imaging showed a higher sensitivity (81.8%) than these previous results using fundamental scanners. Furthermore, using phase-inversion tissue harmonic imaging, we found that the detection of more lesions was not associated with a loss of accuracy. The accuracy of classification in lesions 10–19 mm was 78.8% for phase-inversion tissue harmonic imaging, which means that 96.3% of the detected lesions were correctly characterized.
These results suggest that phase-inversion tissue harmonic imaging is a real advance in sonographic technique for imaging the kidney because of its higher sensitivity without a loss in accuracy. For renal lesions between 20 and 29 mm or larger than 30 mm in diameter, the differences between the two modalities were not significant. This finding might partly be due to the small number of cases. For the size range of from 20 to 29 mm, 21 processes were imaged on fundamental B-mode sonography as the first mode and 12 focal lesions were imaged on phase-inversion tissue harmonic imaging as the first mode.
We included all focal lesions in our study being aware of the problems analyzing patients with different numbers of lesions. Using CIs, we were able to adequately analyze and present the data of this explorative study.
The reason for analyzing a subgroup of lesions 10 mm or larger was the low sensitivity for both modalities in processes smaller than 10 mm (fundamental B-mode sonography, 28.6%; phase-inversion tissue harmonic imaging, 39.6%) without a statistically significant difference between fundamental B-mode sonography and phase-inversion tissue harmonic imaging. These findings are consistent with those of former studies by Jamis-Dow et al. [23] and Warshauer et al. [24]. These researchers found sensitivities between 0% and 26% for focal renal lesions 1–10 mm. We concluded that sonography—even phase-inversion tissue harmonic imaging—is currently not sufficient for assessing very small lesions. Defining a subgroup of lesions 10 mm or larger, however, enabled us to compare fundamental B-mode sonography and phase-inversion tissue harmonic imaging in processes for which sonography is in fact an adequate imaging modality.
Lesions Requiring Additional Diagnostic or Therapeutic Measures
For phase-inversion tissue harmonic imaging and tissue harmonic imaging, an
improvement of diagnostic information and of confidence has been shown for the
evaluation of abdominal organs, such as the liver, gallbladder, and pancreas,
and for intrahepatic vascular structures
[7,
9,
12,
15]. In some patients, this
additional information resulted in an alteration in clinical management.
Lesions visualized more clearly or exclusively on tissue harmonic imaging
included not only cysts but also metastatic lesions to the liver or
intrahepatic vessels, cyst septations, debris in the hepatocholedochal duct,
and solid masses of the pancreas.
The advantages of phase-inversion tissue harmonic imaging that we observed also applied to processes in which further diagnostic or therapeutic action was necessary. Phase-inversion tissue harmonic imaging detected six such lesions that were not discovered using fundamental B-mode sonography. Nine processes could be characterized correctly only with the help of phase-inversion tissue harmonic imaging. For six patients, the examiners were able to recommend further diagnostic or therapeutic steps only by using phase-inversion tissue harmonic imaging. The differences regarding these lesions were not statistically significant. One limiting factor for the results of this subgroup could be that these processes were rather large (mean diameter, 50.3 mm; SD, 38.2 mm) compared with the processes seen in the entire study population.
Specificity
Both fundamental B-mode sonography and phase-inversion tissue harmonic
imaging showed a high specificity without any distinct difference. These
results should be interpreted with prudence because we referred to accurately
characterized kidneys without a focal abnormality even when the contralateral
organ in the same patient showed a focal lesion. Additionally, only 26 of the
patients in our study group lacked a renal lesion. This factor is one
limitation of our study that makes generalizing our results for specificity
difficult.
Image Quality
By evaluating criteria for image quality, we tried to find reasons why
phase-inversion tissue harmonic imaging is superior in terms of detection and
characterization of renal abnormalities. The overall image quality and hence
diagnostic value were better with phase-inversion tissue harmonic imaging than
with fundamental B-mode sonography. One reason for this difference is that
harmonic frequencies originate in the human body and have to pass through the
body wall only once. Moreover, the harmonic beam does not form in the
subcutaneous tissue where a substantial portion of the beam gets distorted.
Most of the harmonic portion of the beam forms close to the focal zone.
Finally, because harmonic frequencies originate in a nonlinear way, the echoes
from side lobes and reverberations in relation to the main lobe are smaller
than in fundamental B-mode sonography
[5,
25,
26]. As a result, the main
ultrasound beam is less degraded before it reaches the region of interest, and
there are fewer echoes from the sides to interfere with the echoes of the main
beam. These factors improve signal-to-noise ratio and reduce scattering
artifacts, thus resulting in better contrast. We noted these effects because
images of the kidney with typical sonomorphologic features were often better
when using phase-inversion tissue harmonic imaging than the fundamental B-mode
sonography.
The mean body mass index of patients did not show significant differences for images with fundamental B-mode sonography and phase-inversion tissue harmonic imaging when subgroups of patients with the same rating for overall image quality, lesion conspicuity, and fluid–solid differentiation were compared with each other. Because only 15 patients had a body mass index of more than 30, the low number of obese patients might be the reason for the lack of statistical proof. However, our subjective impression was that phase-inversion tissue harmonic imaging offers advantages in some heavy patients up to a certain body mass index. As in previous works, obese patients (body mass index, < 35) were not suitable for phase-inversion tissue harmonic imaging because of reduced penetration of the beam. Harmonic ultrasound beams have lower energies than fundamentals and use higher frequencies that are attenuated more rapidly when they propagate through human tissue [5, 12]. These factors seem to be limiting in very obese patients.
One theoretic advantage of phase-inversion tissue harmonic imaging is that fluidfilled structures such as cysts (Figs. 1A, 1B and 2A, 2B), the renal pelvis, and subtle perirenal fluid are more often anechoic because of reduced reverberation artifacts [27]. The superior fluid–solid differentiation of phase-inversion tissue harmonic imaging might have contributed to the high sensitivity for lesions 10–19 mm because many of these smaller lesions were cysts or fluid-containing processes.
|
|
|
|
On phase-inversion tissue harmonic imaging, internal solid components of lesions were diagnosed with confidence because they were not confused so easily with reverberation artifacts in cystic portions (Figs. 3A, 3B). This increased confidence is reflected by better ratings for phase-inversion tissue harmonic imaging concerning fluid–solid differentiation within detected lesions. Our impression that phase-inversion tissue harmonic imaging produces images with great dynamic range that enable examiners to differentiate the internal architecture of even solid processes is in accordance with the findings of Jang et al. [8]. They have shown that phase-inversion tissue harmonic imaging is superior to tissue harmonic imaging and fundamental B-mode sonography regarding detectability of and resolution for internal characteristics of cystic and solid hepatic lesions. Although we did not directly compare dynamic resolution in solid lesions, the examining radiologists had the impression that the drawbacks that have been described for tissue harmonic imaging did not apply for phase-inversion tissue harmonic imaging. The radiologist examiners did not mention a deficit in dynamic range for phase-inversion tissue harmonic imaging compared with fundamental B-mode sonography; instead, they had the impression that the dynamic range for echogenic structures was sometimes higher with phase-inversion tissue harmonic imaging (Figs. 4A, 4B and 5A, 5B).
|
|
|
|
|
|
With reference to the subjective operator-dependent nature of sonographic examinations, our study is limited. For reasons of standardization, we did not account for changes in imaging and postprocessing parameters except for image gain and location of focal zone. We had to compromise in our definitions of fixed parameters, such as using the same gray-scale presetting and dynamic range for both sonographic modalities. Thus, parameters might have been in favor of one or the other imaging modality. Under realtime conditions the examiners sometimes, especially in the beginning of the study, had the subjective impression of a hard, blurring contrast when using phase-inversion tissue harmonic imaging in comparison with fundamental B-mode sonography. However, after obtaining more images with phase-inversion tissue harmonic imaging, the examiners became accustomed to these image characteristics, indicating a positive learning curve. Another solution in a routine clinical setting might be to adjust the postprocessing parameters so that phase-inversion tissue harmonic images look similar to examinations performed using fundamental B-mode sonography. However, we cannot predict whether this potential solution covers the advantages of phase-inversion tissue harmonic imaging in the present study in which no adjustments except image gain and location of focal zone were allowed.
In conclusion, phase-inversion tissue harmonic imaging is an alternative sonographic modality using whole-numbered harmonics of the fundamental sonographic frequencies that are transmitted by the sonography probe. In the present study, phase-inversion tissue harmonic imaging proved superior to fundamental B-mode sonography in the detection and accurate classification of focal renal abnormality, especially in smaller lesions. This better performance of phase-inversion tissue harmonic imaging in comparison with fundamental B-mode sonography was achieved as a result of enhanced overall image quality, better lesion visibility, and the advantages in fluid–solid differentiation. Additionally, previously described drawbacks of harmonic imaging modes with filtering techniques seem to be avoided by the phase-inversion technique, which offers wide dynamic range and good evaluation of the internal morphology of solid lesions. In several lesions requiring further diagnostic or therapeutic measures, phase-inversion tissue harmonic imaging added crucial information that altered the clinical management of patients. For focal abnormalities of the kidney, phase-inversion tissue harmonic imaging is significantly better than fundamental B-mode sonography, with the greatest advantages being seen in the evaluation of lesions 10–19 mm in diameter. However, phase-inversion tissue harmonic imaging seems to be limited in the examination of very obese patients; this limitation has been described for tissue harmonic imaging before.
We recommend phase-inversion tissue harmonic imaging, which is activated by simply pushing a button on the scanner, as the first sonographic modality for imaging the kidneys. In examinations of very obese patients in whom penetration of the ultrasound beam in phase-inversion tissue harmonic imaging is reduced, obtaining additional images using fundamental B-mode sonography can complete an examination. Future studies will clarify whether the advantages of phase-inversion tissue harmonic imaging are independent of manufacturers' implementations. The introduction of contrast agents in abdominal sonography might further increase the diagnostic potential of phase-inversion tissue harmonic imaging and enable contrast-enhanced study of focal renal abnormalities, as has been established for focal hepatic lesions [28, 29].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
3 cm) renal masses: detection with CT versus US and
pathologic correlation. Radiology
1996;198:785
–788
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. I. Jonisch, A. N. Rubinowitz, P. G. Mutalik, and G. M. Israel Can High-Attenuation Renal Cysts Be Differentiated from Renal Cell Carcinoma at Unenhanced CT? Radiology, May 1, 2007; 243(2): 445 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R H Scott, L Walker, O E Olsen, G Levitt, I Kenney, E Maher, C M Owens, K Pritchard-Jones, A Craft, and N Rahman Surveillance for Wilms tumour in at-risk children: pragmatic recommendations for best practice Arch. Dis. Child., December 1, 2006; 91(12): 995 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Hartman, P. L. Choyke, and M. S. Hartman From the RSNA Refresher Courses: A Practical Approach to the Cystic Renal Mass RadioGraphics, October 1, 2004; 24(suppl_1): S101 - S115. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
