| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 From the Department of Pediatric Radiology, University Hospital La Paz, Paseo de la Castellana 261, 28046 Madrid, Spain (T.B., F.G., A.A.); and Department of Radiologic Pathology, Armed Forces Institute of Pathology, Washington, DC (G.J.L.). From the 2000 RSNA scientific assembly. Received November 15, 2000; revision requested December 27; revision received March 26, 2001; accepted May 2. Address correspondence to T.B. (e-mail: cprieto@hulp.insalud.es).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Two hundred sixteen pediatric patients underwent cystosonography enhanced with SH U 508A, a galactose-based echo-enhancing agent. Sonograms of the kidneys and bladder were obtained before filling, during bladder filling, and during voiding. This examination was followed immediately with VCUG. Each kidney or portion of a kidney with its own complete collecting system was considered separately, for a total of 440 kidney units.
RESULTS: VUR was detected in 123 kidney units at cystosonography and in 104 at VCUG. In 401 kidney units, there was concordance between results at cystosonography and at VCUG regarding the presence or absence of VUR. Ninety-four kidney units showed VUR with both methods. Seventy-seven of the 94 refluxing units were depicted with the same grade of VUR with both modalities, and in 17 the VUR grade was greater at cystosonography than at VCUG. Twenty-nine of the 94 units showed VUR at only cystosonography, and 10 units at only VCUG. The McNemar test showed that cystosonography depicted a significantly (P = .003) higher number of units with VUR. By patient, VUR was depicted with both studies in 67 and with only one study in 25. VUR was seen at only cystosonography in 16 patients and at only VCUG in nine. The McNemar test for patients showed no significant difference between the two tests in detection of VUR.
CONCLUSION: Cystosonography with SH U 508A appears comparable to VCUG in the depiction of VUR.
Index terms: Bladder, US, 83.12988, 83.12989 • Ultrasound (US), contrast media, 82.12988, 83.12988 • Ureter, reflux, 82.85, 83.85 • Ureter, US, 82.12988, 82.12989 • Voiding cystourethrography, 82.123, 83.123
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The usual diagnostic procedure for the detection of VUR is voiding cystourethrography (VCUG); its reliability depends on the severity of the VUR. Several studies have demonstrated that only grades 4 and 5 VUR are shown at VCUG with 100% reliability (12–15), because VUR may be intermittent and its severity may vary on sequential cystograms (16–18). Another important limitation of VCUG is radiation exposure. It is estimated that as much as 25% of the radiation with potential to produce genetic alterations received by the pediatric population is related to imaging of the urinary system, especially with VCUG (19,20). Cleveland et al (21) studied children aged 1–5 years by using a digital fluoroscope adapted for pediatric use with optimized parameters for dose reduction; the ovarian dose could not be reduced to less than 200–300 mrad (0.002–0.003 Gy) (and the testicular dose to less than 100–250 mrad (0.001–0.003 Gy). Radionuclide cystography imparts a gonadal radiation dose less than that with conventional VCUG (22,23), but it does not provide the same anatomic detail (23).
The first use of ultrasonography (US) in the diagnosis of VUR was published in 1976, to our knowledge (24). Earlier studies showed US was not a sensitive method for the diagnosis of VUR (25–30). The low sensitivity and specificity for both B-mode and Doppler US for the detection of VUR (31–34) turned attention to US echo-enhancement agents which, instilled into the bladder as at VCUG, would improve detection of VUR. After a few unsuccessful attempts with several substances (35–40), in 1992 two sonographic echo enhancers made of galactose suspension (gadoxetate disodium and SH U 508A, Echovist and Levovist, respectively; Schering, Berlin, Germany), proved to be useful in animal models (41,42). In 1997, Darge et al (43) successfully used SH U 508A in pediatric patients. The same substances have been used by Bosio (44) with excellent results, particularly with SH U 508A. The purpose of our study was to evaluate the usefulness of echo–enhanced cystosonography of the urinary tract compared with the accepted standard of VCUG for both detecting and grading VUR.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Patients who were unable to void spontaneously (eg, because of spinal cord or central nervous system lesions) or who had a neurogenic bladder were excluded from the study. The first 20 patients were excluded because they were considered part of the initial experience of the study. Patients in whom one of the two procedures could not be completed for any reason were also excluded. Two patients in whom sonograms could not be obtained during voiding and four in whom voiding images could not be obtained during VCUG were excluded. There were technical problems in videotaping at cystosonography in one patient, and the kidneys could not be properly evaluated due to severe scoliosis in another patient; these patients were also excluded. Three patients were excluded because they had associated abnormalities that made the grading of VUR impossible. Neither obesity nor large size of the patient were an exclusion criterion for cystosonography.
The final number of patients included in the study was 216 (100 [46.3%] male patients and 116 [53.7%] female patients; mean age, 3 years 9 months; age range, 3 days to 18 years [SD, 3.4 years]). For data analysis, results were assessed in terms of patients and in terms of renal units, with both kidneys considered separately. A renal unit was defined as a renal pelvis with its own ureter inserting into the bladder. A normal kidney was considered as one kidney-ureter unit, and a complete duplication was considered as two kidney-ureter units. Incomplete duplications were considered one renal unit. Therefore, 440 renal units of 216 patients were included in the study. Twelve of the 216 patients had a complete unilateral renal duplication, and one patient had a complete bilateral renal duplication (all proved at prior excretory urography). Two patients had single native kidneys, and four patients had single transplanted kidneys. Table 1 lists the indications for the performance of VCUG in the study population.
|
The patients were catheterized transurethrally, and the catheter (5- or 8-F infant feeding tube) was connected to a three-way stopcock to which was connected an intravenous tubing set attached to a saline bottle placed 70 cm above the patient’s head. A 10-mL syringe of SH U 508A (concentration, 300 mg/mL) was also attached to the three-way stopcock. The catheter was secured to the perineum. The bladder capacity was estimated from the patient’s age as follows: volume (in milliliters) = age (in years) + 2 x 30 mL (45,46).
The bladder was filled with saline to half the estimated volume. The saline was infused by means of gravity. The echo enhancer was then instilled very slowly with continuous US guidance. SH U 508A was injected into a half-filled bladder to allow the diagnosis of low-volume VUR and a more complete correlation with VCUG, in which the first images are acquired when the bladder is half filled. After SH U 508A was injected, saline was instilled until the bladder was full. US was performed intermittently to depict the bladder and kidneys during instillation of the saline and SH U 508A. Each kidney and the bladder were imaged at US approximately every 15 seconds. Initially, while microbubbles were still attached to galactose particles, the suspension had a higher specific gravity than that of saline and filled in the bladder from the dorsal region to the ventral region. Gradually, as the bladder filled with saline, SH U 508A achieved a homogeneous distribution.
The US scans of the kidneys and bladder were obtained before filling, during bladder filling, and during voiding. The transurethral catheter was not removed from the bladder during voiding because VCUG was performed afterwards, and a second catheterization was avoided. The examination was finished when US depicted an empty bladder or when the older patients completed voiding. VCUG was then performed, by following the guidelines suggested by the International Reflux Study in Children (IRSC) (47). The bottle with the contrast medium (Plenigraf [iodine concentration, 16.5%]; Juste, Madrid, Spain) was placed at the same height as the saline used at cystosonography. The same volume of liquid used for the previous study was administered by means of gravity, at the same temperature (room temperature) and infusion rate, to reproduce as much as possible the same conditions used at cystosonography. VCUG was performed with a digital fluoroscopic unit (Fluorospot Compact; Siemens Medical Systems, Berlin, Germany). Intermittent fluoroscopy (300 mrad/min [0.003 Gy/min]) was performed, and several abdominal radiographs were obtained as follows: (a) half-filled and (b) completely filled bladder radiographs, (c) anteroposterior and oblique voiding radiographs, and (d) postvoiding and (e) voiding lateral radiographs in male patients.
Vital signs, including arterial blood pressure and electrocardiographic monitoring, were documented during the examination. Special attention was paid to any sign of potential adverse events that could be related to administration of SH U 508A, such as skin rash or abnormal respiratory sounds. Parents were asked to contact us if any adverse events occurred in the 24 hours after the procedures. All patients received prophylactic antibiotic therapy (trimethoprim sulfate, 5 mg per kilogram of body weight, in two doses, at 2 hours before the first procedure and at 12 hours after the second procedure). Each patient was included in the study once, and sedation was not used in any case. During catheterization, baseline US, and cystosonography, either the parents or the nurse assistant immobilized the patients since ionizing radiation was not involved. VCUG was performed with the immobilization devices routinely used in our department.
Cystosonographic studies were performed by one radiologist (T.B.), and VCUG studies were performed by a second radiologist (A.A.). Cystosonograms were interpreted for the presence of VUR and its grade before VCUG was performed. Neither radiologist knew the results with the other technique or at any prior VCUG examination. All the studies were later reinterpreted, by consensus of both radiologists (T.B., A.A.), without knowledge of the original interpretation. First, all cystosonograms were interpreted by both radiologists, and a final consensus diagnosis was obtained for each. Later, VCUG images were interpreted by consensus by the same two radiologists, without knowledge of the cystosonographic interpretation.
VUR was diagnosed at cystosonography when microbubbles appeared in a ureter or renal pelvis. VUR was graded at VCUG by using the international system (47). VUR was graded at cystosonography by using the classification of Atala et al (38): grade 1, echo contrast in only the ureter; grade 2, echo contrast in the pelvicaliceal system with no dilatation; grade 3, mild to moderate dilatation of the pelvicaliceal system; grade 4, moderate to severe dilatation of the pelvicaliceal system; and grade 5, gross dilatation of the pelvicaliceal system with total or partial atrophy.
The time for each step of the procedure varied among the patients. The approximate mean time for cystosonography was 26 minutes (including baseline US, catheterization time, and completion of cystosonography). The approximate mean time for VCUG was 8 minutes (time for catheterization was not included). Table 2 shows the mean duration of each examination.
|
statistic and its standard error. Fleiss (48) suggests that values less than 0.40 indicate poor agreement, values between 0.40 and 0.75 indicate fair to good agreement, and values greater than 0.75 indicate strong agreement. Discordance between the two methods was calculated by means of the McNemar test of symmetry, with statistical software (SPSS, version 9.0; SPSS, Chicago, Ill). |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Results by Renal Units
Cystosonography showed VUR in 123 (27.9%) of the 440 units (Figs 1, 2): grade 1 VUR in two units, grade 2 in 58, grade 3 in 49, grade 4 in seven, and grade 5 in seven. VUR occurred at low volume (before the bladder was filled) in 25 units. VCUG images depicted VUR in 104 (23.6%) renal units: grade 1 VUR in 19, grade 2 in 40, grade 3 in 32, grade 4 in six, and grade 5 in seven. Low-volume VUR was seen in 17 of these units.
|
|
|
|
|
|
|
|
|
|
In 16 (7.4%) patients, VUR was seen at only cystosonography (grade 3 in seven [43.8%] and grade 2 in eight [50.0%]); results were negative at VCUG. In nine (4.2%) patients, VUR was seen at only VCUG (grade 1 in seven). Table 4 shows the distribution of VUR by grade obtained with both techniques.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The reports of Darge et al (43) and Bosio (44) were used as the starting point for our study, although we made several modifications. First, every case was examined with both techniques, with cystosonography performed first and with the patients in identical positions. Both studies were performed with the same liquid volume, rate of infusion, and temperature, as these factors may influence results (13). Jequier and Jequier (12) showed that cyclic cystography enhances the detection and grading of VUR. In our study, we performed only one complete bladder filling (one half first and then the second half) for cystosonography and VCUG, to avoid modifying the results with sequential complete fillings. The transurethral catheter was not removed from the bladder during voiding cystosonography to avoid a second catheterization. It has been demonstrated that this procedure does not modify the presence and the grade of VUR (13,54).
The injection of SH U 508A in a half-filled bladder at cystosonography enables improved correlation with VCUG because it allows the diagnosis of early (low-volume) VUR, which cannot be detected if the enhancer is instilled into a full bladder. VUR was diagnosed whenever the microbubbles were observed in the ureter or pelvicaliceal system. These particles were more echogenic than the renal sinus and were easy to recognize (Figs 1–4). VUR could also be seen as a strong hyperechogenic material occupying the entire renal pelvis, with strong acoustic shadowing similar to that of a staghorn calculus (Figs 1, 4). This appearance might be due to a high concentration of SH U 508A being poorly diluted with little saline. We noticed that the pattern of enhancement was very similar for both kidneys in patients with bilateral VUR. In patients in whom a dilated distal ureter was identified at baseline US, the anechoic ureteral contents became hyperechoic after administration of SH U 508A, and the diagnosis of VUR was easily made. If the distal ureter was not dilated, then the diagnosis of grade 1 VUR was more difficult, owing to the hyperechogenicity of the retrovesical space. US monitoring of the bladder during injection of SH U 508A allowed us to observe its behavior. Initially, while microbubbles were still attached to galactose particles, the echogenicity of the posterior wall was increased, creating a strong posterior acoustic shadowing and obscuring visualization of the retrovesical space, as noted by Darge et al (43). SH U 508A gradually mixed with saline, which allowed better assessment of the distal ureters and retrovesical space.
The comparison of cystosonography with VCUG showed that both techniques were concordant for the presence or absence of VUR and for the grade of VUR in 83.7% of renal units. There was concordance in the detection of VUR but not in the grade of VUR in 3.8% of the units. Cystosonography tended to depict a higher grade of VUR than did VCUG when both tests demonstrated VUR. In fact, in all of the units with different grades of VUR detected with each technique (when both tests showed VUR), a higher grade was depicted at cystosonography than at VCUG. Grades 2 and 3 VUR tended to be underestimated at VCUG: 28 of the 29 units with a positive cystosonogram and a negative VCUG image were depicted with grades 2 and 3 VUR at cystosonography. Data show that cystosonography depicted not only more refluxing units but also VUR with higher grades than did VCUG. Differences in detection rates between the two methods favored cystosonography. These results correlate not only with those in the study of Mentzel et al (55), with SH U 508A, but also with those in studies with different echo enhancers (35–40).
The improved detection of VUR may be due to the greater sensitivity or ease with which US depicts a very small amount of SH U 508A in comparison with fluoroscopy, in which very faint contrast may be quite difficult to identify. Some authors have described this result (positive cystosonogram with negative VCUR image) as a false-positive US result. Recently, Mentzel et al (55), in a series of 46 patients, found five renal units with VUR not depicted at VCUG. The authors suggest that false-positive findings may actually be true-positive findings, and they observed a tendency to detect higher grades of VUR with cystosonography than with VCUG. The tendency for the grade of VUR to be underestimated at VCUG has been suggested by other authors (15,16), in comparative studies with scintigraphic cystography. We believe that the main reason that VUR is diagnosed more frequently at cystosonography than at VCUG is the more frequent monitoring during both filling and voiding of the bladder with the former; in addition, visualization of SH U 508A may be easier than visualization of the faint fluoroscopic contrast material. The intermittent nature of VUR is well known; therefore, the possibility of detecting it increases with the time of visualization (12,16,18). Differences in the grade of VUR detected with both methods are extremely important, since the treatment recommendations of the American Society of Urologists depend primarily on the grade of VUR. Therefore, the tendency of cystosonography to depict a higher grade of disease, when compared with VCUG, may have implications that depend on the practice of the treating physician.
In 16 (7.4%) patients in our series, VUR was detected at only cystosonography. This result implies that these patients would not have been treated if cystosonography had not been performed, even though seven of them had grade 3 VUR and eight had grade 2 VUR. Additionally, 11 of the 16 patients were younger than 5 years, which increases their likelihood of developing renal scars, as the frequency of renal damage is higher in younger patients (8,18). Therefore, we believe the increased ability of cystosonography to depict VUR makes it a valuable tool in the evaluation of possible VUR.
While cystosonography in our study depicted higher grades of VUR than did VCUG in the same patient, this was true for only grades 2–5 VUR. There is a tendency for cystosonography to depict underestimated grade 1 VUR (or miss it entirely); 70% (seven of 10) of the units with a positive VCUG image and a negative cystosonogram had grade 1 VUR. Grade 1 VUR that is missed at cystosonography may have consequences for the patient, although treatment of grade 1 VUR is controversial. In our series, all of the units with grade 1 VUR seen at only VCUG were in patients with unilateral VUR, so these seven patients would not likely have received treatment if only cystosonography had been performed. The poor visualization of grade 1 VUR at cystosonography was likely due to the US appearance of an echogenic retrovesical space that may also contain gas-filled intestine, which makes detection of SH U 508A in nondilated distal ureters difficult. Interestingly, more than half of the VUR classified as grade 1 at VCUG was graded as grades 2 or 3 at cystosonography.
In patients with VUR, not only is the detection and grading of VUR important but also the identification of anomalies that could result in or prevent depiction of VUR. In our series, although not originally a part of our study design, it was our impression that the urethra was considerably more difficult to evaluate anatomically with cystosonography than with VCUG. In our opinion, this constitutes an important limitation of cystosonography when compared with VCUG. This may be less of a consideration in female patients, in whom urethral anomalies occur considerably less frequently than they do in male patients. Therefore, we do not believe that cystosonography should replace VCUG for the initial evaluation of boys suspected of having VUR.
Cystosonography may also help evaluate disease in patients in whom there is a high suspicion for VUR but a negative VCUG image, because it can be repeated without additional radiation exposure and because, in our study, it depicted higher rates of VUR than did VCUG. We also believe that cystosonography is acceptable for the screening of VUR in asymptomatic siblings and the offspring of patients with known VUR. The absence of ionizing radiation may also allow the patient to be accompanied during the examination by parents or a caregiver, even a pregnant caregiver, which is very important in the pediatric population. Both studies (US and cystosonography) can be performed in the same diagnostic session, which compensates somewhat for the longer duration of cystosonography.
The main advantage of cystosonography is the accurate and reliable depiction of VUR without the use of ionizing radiation. This is important in the pediatric population, considering the number of imaging evaluations for VUR that each child may undergo.
The ideal method for diagnosis of VUR should be accurate, safe, and easy to perform; involve no ionizing radiation; and be noninvasive and inexpensive. Cystosonography fulfills all but the last two criteria. It requires bladder catheterization and therefore is invasive. Additionally, it may prove more expensive than fluoroscopic VCUG, depending on the cost of SH U 508A and additional US machine use. In their last report (4), the panel of experts in VUR of the American Society of Urology stressed the importance of developing voiding bladder and upper tract evaluations for VUR that do not use ionizing radiation. Cystosonography is a promising method with which to achieve this objective, and it should prove valuable in the evaluation and monitoring of disease in children with suspected or known VUR.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, T.B.; study concepts and design, T.B.; literature research, T.B., A.A.; clinical studies, T.B., A.A.; data acquisition, T.B., A.A.; data analysis, T.B., G.J.L., F.G.; statistical analysis, F.G.; manuscript preparation, T.B., A.A., G.J.L.; definition of intellectual content, T.B.; manuscript editing, F.G., T.B., G.J.L.; manuscript review, G.J.L.; manuscript final version approval, all authors.
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official nor as reflecting the views of the Departments of the U.S. Air Force or Defense.
See also the editorial by O’Hara (pp 283–284 ) in this issue.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Thompson, S. D. Simon, V. Sharma, and U. S. Alon Timing of Follow-up Voiding Cystourethrogram in Children With Primary Vesicoureteral Reflux: Development and Application of a Clinical Algorithm Pediatrics, February 1, 2005; 115(2): 426 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Berrocal, F. Gaya, and A. Arjonilla Vesicoureteral Reflux: Can the Urethra Be Adequately Assessed by Using Contrast-enhanced Voiding US of the Bladder? Radiology, January 1, 2005; 234(1): 235 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Schoellnast, F. Lindbichler, and M. Riccabona Sonographic Diagnosis of Urethral Anomalies in Infants: Value of Perineal Sonography J. Ultrasound Med., June 1, 2004; 23(6): 769 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Berrocal, P. Lopez-Pereira, A. Arjonilla, and J. Gutierrez Anomalies of the Distal Ureter, Bladder, and Urethra in Children: Embryologic, Radiologic, and Pathologic Features RadioGraphics, September 1, 2002; 22(5): 1139 - 1164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Halpern, F. Frauscher, and G. J. Lonergan Diagnosis and Grading of Vesicoureteral Reflux * Radiology statistical consultant responds: * Dr Lonergan responds: Radiology, July 1, 2002; 224(1): 293 - 294. [Full Text] [PDF]
|
|
|
|
 |
|
 B. P. Wood Using Ultrasound to Evaluate Possible Vesicoureteral Reflux? AAP Grand Rounds, April 1, 2002; 7(4): 42 - 42. [Full Text] [PDF]
|
|
|
|
|
|
S. M. O'Hara Vesicoureteral Reflux: Latest Option for Evaluation in Children Radiology, November 1, 2001; 221(2): 283 - 284. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
