| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 Department of Radiology, Division of Pediatric Radiology, LKH-University
Hospital, Auenbruggerplatz, Graz A-8036, Austria.
2 Department of Pediatrics, University Hospital Graz, Graz, Austria.
3 Department of Pediatric Urology, Barmherzige Schwestern, Linz, Austria.
Received December 15, 2003; accepted after revision March 1, 2004.
Address correspondence to Michael Riccabona
(michael.riccabona{at}klinikum-graz.at).
OBJECTIVE. The purpose of this study was to evaluate the potential of MR urography in the assessment of children with a suspected "functional single kidney."
SUBJECTS AND METHODS. Sixty patients (age range, 2.7 weeks to 15.7 years) who had been referred for assessment of a suspected functional single kidney underwent MR urography in addition to detailed sonography of the urinary tract and the currently indicated standard imaging. The results of the conventional imaging (99mTc-dimer captosuccinic acid scintigraphy, voiding cystourethrography, and genitography) were compared with the results of sonography and MR urography; surgical findings served as the gold standard if available.
RESULTS. Twenty-six patients had a single kidney. The other diagnoses were six contralateral multicystic dysplastic kidneys, two normal ectopic kidneys, one crossfused double system, and 25 ectopic or dysplastic renal buds. Scintigraphy detected all normal kidneys, two ectopic kidneys, and two dysplastic renal buds. Detailed sonography missed two ectopic kidneys and two orthotopic dysplastic renal buds, but one additional renal bud that could not be confirmed on other imaging techniques (accuracy, 91.7%; sensitivity, 88.2%; specificity, 96.2%) was suspected. MR urography results were correct in all patients and verified in all 13 with surgical correlation.
CONCLUSION. MR urography allows a reliable assessment of renal and ureteral anatomy and of dysplastic or ectopic renal buds, even in non- or poorly functioning systems. MR urography therefore has the potential to replace the currently used excretory urography and scintigraphy. In patients with a suspected functional single kidney a detailed sonographic study and MR urography should be considered the diagnostic algorithm of choice.
The existence of a "functional single kidney" is a rare but important question in pediatric uroradiology. The suspicion is generally based on either prenatal or early postnatal sonography identifying only one kidney or a contralateral multicystic dysplastic kidney. This query also arises from sonographic examinations of older patients who have presented with clinical symptoms such as urinary tract infection or wetting; rarely do typical symptoms that hint at ectopic ureteral insertions, such as dribbling, lead to a focused investigation. This question may arise coincidentally at an investigation performed for other reasons. These suspected functional single kidneys hold the risk of another poorly functioning or ectopic renal unit not yet discovered, potentially with an ectopic ureteral insertion, because single system ectopic ureters draining an often small and dysplastic kidney are rare [1–6]. Early and reliable assessment of a small and dysplastic (contralateral) renal bud enables early and adequate treatment and helps to avoid inappropriate and unsuccessful management due to delayed diagnosis.
The suspicion of a functional single kidney currently leads to a variety of investigations: sonography, voiding cystourethrography and genitography, cystourethroscopy and vaginoscopy, renal scintigraphy, and occasionally excretory urography or CT [7–11]. Until recently, 99mTc-dimer captosuccinic acid (DMSA) static renal scintigraphy was considered the best technique for locating an ectopic kidney [12, 13]. However, some drawbacks exist to all these methods listed: DMSA scanning and excretory urography have limited precision to explore a poorly or nonfunctioning kidney because they depend on functioning renal tissue. DMSA static renal scintigraphy, excretory urography, CT, and voiding cystourethrography share the hazard of ionizing radiation; excretory urography and CT additionally hold the risk of allergic reactions to the contrast medium. Voiding cystourethrography or genitography depends on a refluxing ureter to enable diagnosis. Limited-scope sonography has restricted potential to depict and assess small, dysplastic, and particularly ectopic renal buds; in addition, ureteral anatomy cannot be evaluated and assessment of a potential ectopic ureteral insertion is difficult even when using modern techniques. Furthermore, infants with a suspected multicystic dysplastic kidney need evaluation for potential function and a reliable assessment of the ipsilateral genitalia, and older patients who present with a single kidney may well have had a multicystic dysplastic kidney that has spontaneously regressed. All this adds up to the sometimes limited potential of available imaging in evaluating or assessing a suspected small, dysplastic, even ectopic and usually malfunctioning kidney as well as for preoperative confirmation and anatomic assessment. This establishes an unsatisfactory situation for clinical needs, because missing a residual contralateral kidney or ectopically inserting a ureter from such a unit may cause significant delay in appropriate therapy. Therefore, a reliable, preferably noninvasive, imaging technique with the lowest possible radiation burden is necessary to improve the imaging workup in this setting.
MR urography has successfully been applied to the pediatric urinary tract [14–18]. Though sedation of the patients is necessary for achieving diagnostic-quality images, particularly in young children and infants, it holds vast potential for evaluating the child's urogenital tract by a nonionizing technique that offers both anatomic and functional information [14, 16, 17, 19–23]. To our knowledge, only a few reports focusing on the potential of MR urography for evaluating a suspected single kidney in small patient groups have been published [15, 24–26]. Consequently, we conducted a prospective study at two referral centers to reevaluate the potential of MR urography for this query in a larger patient group.
The aim of our study was to assess the potential of MR urography as a single imaging technique for evaluating infants and children with a sonographically suspected functional single kidney and patients with dysplastic or ectopic kidneys (including multicystic dysplastic kidney) and its potential impact on existing imaging algorithms.
Subjects and Methods
Sixty infants and children (28 boys and 32 girls; mean age, 3.6 years; age range, 2.7 weeks to 15.7 years; median age, 2.5 years) who had been referred for evaluation of a suspected functional single kidney or contralateral dysplastic kidney were included in this study. All patients initially underwent a physical examination with the recording of an extensive medical history. For imaging, they all underwent detailed sonography of the entire urinary tract as well as 99mTc-DMSA scintigraphy, voiding cystourethrography, and MR urography. Excretory urography was available in only a few patients and was not included in the evaluation; CT was never performed because of its high radiation burden. The results of conventional imaging were compared with the MR urography findings and the final diagnosis as well as with intraoperative and vaginoscopy–cystoscopy results, when available.
The detailed sonographic study was performed with either an Elegra or a Sequoia 512 system (Siemens) using a range of linear and curved array multifrequency transducers (14–2-MHz), depending on patient size and age as well as on the anatomic site. Sonography of the entire urinary tract was performed after physiologic oral hydration with a sufficiently distended urinary bladder, including application of color Doppler sonography (e.g., to look for an ostial jet or renal vessels), harmonic imaging, as well as a perineal approach when indicated.
All patients underwent voiding cystourethrography performed according to standard guidelines using either transurethral or suprapubic access.
For static renal scintigraphy, the physiologically hydrated patients received 0.3–0.4 mCi/kg 99mTc-DMSA IV. Static acquisition was started 2–3 hr later after emptying the urinary bladder. All images were acquired using a dual-head large-field-of-view camera with low-energy general purpose collimators, with 150,000 counts/120 sec each in posterior, anterior, left anterior oblique, and right anterior oblique projections and loaded in a 256 x 256 matrix. Reporting was performed by visual image interpretation as well as on the basis of a computerized calculation of relative renal uptake by independent nuclear medicine specialists.
MR urography was performed on a 1.5-T unit (Magnetom Symphony, Siemens) using body coils or, in infants, a head coil. Adequate hydration was granted by the same infusion regime as for scintigraphy, including simultaneous IV application of gadolinium (0.1 mL/kg) and furosemide (0.5 mg/kg; maximum dose, 20 mg) for dynamic evaluation. Imaging sequences consisted of axial, coronal, and sagittal acquisitions using T2-weighted true fast imaging with steady-state free procession (TR/TE, 5.08/2.5; flip angle, 70°), T2-weighted turbo spin-echo (51,790/86; flip angle, 189°), T2-weighted HASTE (1,070/400; flip angle, 150°), transversal T1-weighted fast low-angle shot (126/2.94; flip angle, 70°), T2-weighted turbo spin-echo fat-saturtation PACE (Siemens) (2,060/86; flip angle, 180°), and coronal T1-weighted gradient-recalled echo 3D (4.78/7.46; flip angle, 70°) sequences for dynamic urography after administration of paramagnetic contrast material. No semiquantitative assessment of washout curves was performed. Sedation, particularly in all infants and children younger than 6 years old, was achieved by midazolam hydrochloride (0.1 mg/kg), propofol (3–6 mg/kg), or ketamine hydrochloride (1 mg/kg) administered IV as decided by the anesthesiologist in charge, after informed parental consent had been obtained.
All symptomatic patients with dysplastic renal buds and all patients with ectopic ureteral insertion underwent open nephroureterectomy with pre- or intraoperative vaginocystoscopy. The conclusive results of available imaging interpretations in consensus or operative findings established the final diagnosis. Statistically, accuracy and sensitivity as well as specificity and the predictive value were calculated.
Results
Twenty-six of the 60 patients truly had a unilateral renal agenesis with a contralateral single kidney. Twenty-one of them were healthy except for physiologic contralateral hypertrophy and mild nonobstructive dilatation. Five had additional disease in the single kidney: one had a mild, partial obstructive ureteropelvic junction obstruction, and four had an ipsilateral vesicoureteral reflux. Two girls had a paravaginal cyst; no other associated genital malformation was detected. Scintigraphy, sonography, and MR urography findings were normal in all these patients, with no dysplasia or parenchymal defect in all these kidneys detected on any technique.
Six patients had a contralateral multicystic dysplastic kidney (two small and ectopic, four orthotopic cystic masses, with vesicoureteral reflux in one ureter of a multicystic dysplastic kidney), three with ipsilateral genital malformations (two cystic dysplastic seminal vesicles and one paravaginal cyst); all were recognized by both MR urography and sonography (Figs. 1A and 1B). Two patients had an ectopic normal kidney that had not been found on sonography, but could be seen on scintigraphy and MR urography. A crossfused double system in one child with an inconclusive DMSA scan was recognized on sonography and MR urography.
|
|
Twenty children had a more or less dysplastic renal bud, partially with cystic components (Table 1 and Figs. 2A and 2B). Five of them were positioned ectopically. Five ureters draining a renal bud had vesicoureteral reflux. Additionally, one blind refluxing ureteral stump without a renal bud was detected on sonography, voiding cystourethrography, and MR urography, but no nonrefluxing ureteral stumps were observed in our study population. Six children had an ectopic ureteral insertion (three in the vagina, three in the urethra). These ureters all were suspected from the detailed sonography study and then confirmed on MR urography and at surgery; in the two patients with a refluxing ureter draining into the urethra, this was also depicted by voiding cystourethrography. An ureterocele at the side of the renal bud was present in four patients, all recognized on MR urography and sonography (Figs. 3A, 3B, 3C, 3D, and 3E). Twenty-three of the confirmed renal buds were depicted on sonography; sonography furthermore suggested some residual dysplastic renal tissue in an additional patient, but this could not be confirmed on any other imaging. Scintigraphy depicted three renal buds that still had some residual function. Size measurements of sonography, MRI, and the 13 available surgical specimens correlated well: the mean length of the residual renal bud was 3.0 cm on sonography, 3.2 cm on MRI, and 3.1 cm at surgery.
|
|
|
|
|
|
|
|
Overall accuracy of scintigraphy was 53.3%, with a low sensitivity for a renal bud of 17.6%, a low specificity of 61.5%, a poor negative predictive value of 48.1%, but an excellent positive predictive value of 100%. The detailed sonographic examinations performed at the referral centers had an accuracy of 91.7%, a sensitivity of 88.2%, and a specificity of 96.2%, with a high positive and a slightly lower negative predictive value of 96.8% and 86.2%, respectively. The limited-scope sonography ("sonoscopy") performed at an external site could not be evaluated in detail, but obviously missed many renal buds and performed much worse. MR urography was accurate in all patients, particularly in all 13 who underwent surgery. No other gold standard was available for the evaluation of MR urography results.
Discussion
Sonography has been and probably will remain the primary imaging tool for evaluating urogenital tract disease in neonates, infants, and children. Using modern techniques, sonography has become a highly specialized imaging method that theoretically allows a reliable and comprehensive diagnosis of most conditions [7, 10, 11]. This is supported by our results: most ectopic kidneys, most ectopic ureteral insertions, and all genital malformations were correctly diagnosed or were at least suspected on a detailed and extended sonographic examination of the urogenital system. Sonography even allowed accurate evaluation of vessel anatomy and residual perfusion as well as reliable size measurements. Together with amplitude-coded color Doppler sonography, sonography can be used to evaluate not only the size but also the perfusion of the healthy kidney. However, because of its intrinsic restrictions and a considerable investigator dependency, some doubt may remain. The limitations of sonography are partially a result of potentially interfering overlying structures and partially to problems with standardization and documentation; another important factor is varying investigator education and training. Additionally, sonographic quality is influenced by the varying capabilities of different sonography systems that may lead to missing some diagnoses. Our observations indicate these limitations as inherent to an external orienting sonography examination, sonoscope— that is, from a quick orienting study without any specific patient preparation, with only a glance at the bladder and the kidneys, without evaluation of any other abdominal region, and without Doppler studies; many ectopic renal buds were missed. Even the investigations at one referral center missed two ectopic kidneys (Table 1). Therefore, confirmation of sonographic diagnosis, evaluation of equivocal or questionable negative sonography findings, assessment of renal function and ureteral anatomy, as well as preoperative assessment indicate additional imaging, particularly considering the relatively poor confidence in sonography. This additional imaging at present uses a variety of methods that are partially invasive, partially use ionizing radiation, and still do not always answer all relevant questions; therefore, an improved imaging workup for those children would be beneficial.
MR urography has been successfully used for the evaluation of pediatric urinary tract disease [14–26]. Our study confirms the feasibility of MR urography even in neonates and infants. MR urography not only established the diagnosis in all patients but clearly was superior to DMSA scintigraphy for the assessment of all renal buds or multicystic dysplastic kidney without function, thus increasing conspicuity of the diagnosis also in those patients who in fact had renal agenesis and simultaneously providing all necessary information on the parenchyma and size of the contralateral hypertrophic single kidney. MR urography proved to be an ideal problem-solving tool for evaluation of patients with unclear sonographic results as observed in six of our patients, showed anatomy also in those five patients with false sonography results, and provided all preoperatively necessary information. Thus our results indicate that MR urography may be superior to the conventionally used imaging protocol (consisting of sonography, DMSA scintigraphy, voiding cystourethrography, and excretory urography) for many reasons: MR urography provides all necessary anatomic information about the healthy kidney, a renal bud, or a multicystic dysplastic kidney and, in most conditions (not in completely collapsed ureters), also about the ureters (including the site of insertion), even in poorly or nonfunctioning systems. MR urography additionally provides information on the renal parenchyma of the healthy kidney and on renal perfusion, excretion, and drainage of the healthy as well as dysplastic or ectopic kidney. MR urography allows volume calculation of the hypertrophic functional single kidney and depicts residual function in an ectopic or dysplastic renal bud or residual tissue and perfusion in a multicystic dysplastic kidney. Using a functional approach, MR urography allows the calculation of split renal function and creates drainage curves comparable to scintigraphy [21–24]. However, this approach was not performed in our study. MRI allows assessment of potentially associated genital malformations. Finally, MR angiography can be performed in the same session, allowing a reliable evaluation of vascular anatomy. Therefore, MR urography appears to be the ideal investigation that enables a comprehensive and reliable assessment of most relevant aspects in patients with a suspected functional single kidney or an ectopic–dysplastic renal bud including multicystic dysplastic kidney.
However, some drawbacks to MR urography must be discussed: In the pediatric age group MRI may require sedation, particularly to avoid failure and the consequent waste of precious investigation time for other patients and queries in busy places [20, 27, 28]. Though sedation may also be required in some infants for scintigraphy, sedation for MR urography appears less problematic: invasiveness is only slightly increased, overall radiation burden is reduced, and additional anatomic information as well as visualization of even nonfunctioning systems can be gained. MR urography currently cannot be used to evaluate vesicoureteral reflux; only indirect signs such as ureteral retrograde filling with contrast urine (e.g., draining a nonfunctioning system) or increasing or gross ureteral dilatation with increasing bladder filling may hint at the presence of vesicoureteral reflux. In the future, adapted techniques and protocols may potentially enable vesicoureteral reflux assessment on MRI. Vesicoureteral reflux, however, appears to be less important for the initial diagnosis but may be important particularly in patients with urinary tract infection. In these patients, vesicoureteral reflux in the healthy kidney may reliably be assessed on echo-enhanced cystosonography [29–34]. The role of vesicoureteral reflux in an ectopic ureter or in a ureter that drains a malfunctioning dysplastic renal bud depends on the varying surgical approach. If necessary, voiding cystourethrography can be still performed as part of the preoperative assessment or as indicated by clinical needs. Sonography, excretory urography, and scintigraphy are also readily available, whereas, at present, access to MRI is still limited, particularly for the pediatric age group. MR urography is considered an expensive technique; discussions may arise for economic reasons. One may argue that applying this technique to patients with conditions that can be diagnosed with other conventional methods and less expensive techniques is a waste of resources and only increases health care costs. Our results show that MR urography is superior to conventional imaging for the evaluation of patients with ectopic ureteral insertion, poorly functioning systems, or associated genital malformations. Therefore, MR urography appears to be indicated not only for its superior diagnostic potential but also because early MR urography may shorten the course of the disease in some patients and may replace some conventional imaging (e.g., excretory urography and scintigraphy). Thus, MR urography may even reduce costs (at least in the Austrian reimbursement system) by replacing several techniques with limited diagnostic potential by just one reliable imaging technique. Additionally, by reducing delay in adequate therapy, MR urography may reduce overall health care expenses.
In conclusion, diagnostic-quality MR urography can be performed in infants and neonates with relatively little invasiveness. MR urography properly depicts renal and ureteral anatomy and allows assessment of ectopic renal buds better than sonography and DMSA, even in poorly or nonfunctioning systems or multicystic dysplastic kidney. MR urography provides additional functional information and enables the assessment of potentially associated genital anomalies without using radiation. It therefore has the potential to replace some imaging tools of the currently used imaging algorithm and should be considered the primary technique of choice in spite of the need to sedate patients. If prompt access and availability of MRI are present, one could consider consecutively changing the present imaging algorithm for a suspected functional single kidney: after the initial extended and detailed sonographic study of the genitourinary system (perhaps supplemented by echo-enhanced cystosonography), MR urography could be the single investigation to follow if the detailed sonography study did not answer all clinically relevant questions. Some particularly symptomatic patients may additionally need voiding cystourethrography or genitography individually indicated by the clinical presentation, the potential therapeutic impact, as well as sonography and MR urography findings.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
