Screening of young children between 2-24 months of Age with a first febrile UTI for VUR with imaging modalities

Background: Urinary tract infection (UTI) is one of the most common serious bacterial illnesses affecting children. The presence of moderate to severe vesicouretheral reflux (VUR), particularly if bilateral, is an important risk factor for pyelonephritis and renal scarring. Imaging studies, including renal ultrasonography (USG), technetium-99m– labelled dimercaptosuccinic acid (DMSA) scanning, and voiding cystourethrography (VCUG), are part of standard care after diagnosis of a first UTI in young children. This study was done to determine the sensitivity, specificity and predictive values of renal USG and DMSA scan findings for VUR. Methods: All children between 2-24 months of age presenting with first febrile UTI were prospectively studied. All infants were hospitalized and investigated using renal USG, DMSA scan, and VCUG after diagnosis. The association among findings of USG, DMSA scan, and VCUG were evaluated. Results: A total of 48 children were included in the study and screened, out of which there were abnormal results in 22 (45.8%) USG and 29(60.4%) DMSA scan. All grade VUR (I-V) was present in 8(16.7%) patients and high grade VUR (III-V) was present in 7(14.6%) patients. The sensitivities for detecting VUR of renal USG alone (75%) or DMSA scan alone (87.5%) were not as good as that of the “OR rule” strategy, which had 100% sensitivity and 100% negative predictive value. Conclusions: Renal USG and DMSA scanning both should be routinely performed in young children with a first febrile UTI. VCUG is only indicated when abnormalities are apparent on either USG scanning or DMSA scanning or both.


Introduction
UTI is one of the most common serious bacterial illnesses affecting children, particularly infants and young children [1]. It is the third most common infection in paediatric age group, rank next to respiratory and gastrointestinal infections and account for 4 to 10% of febrile children admitted to hospital [2]. UTI is defined by combination of positive urine culture (any growth of bacteria in urine from a suprapubic bladder aspiration or >5×10 4 colony-forming units per millilitre in a catheterized specimen or >10 5 colonyforming units per millilitre from midstream clean catch) and fever ≥38.0°C (100.4°F) [3,4]. UTI should be suspected in an infant or child presenting with unexplained fever beyond 3 days [5]. All febrile UTI are considered to involve the upper urinary tract with potential for renal scarring [6]. Rapid evaluation and treatment of UTI is important to prevent renal parenchymal damage (renal scarring) that can cause hypertension and chronic renal failure later.
UTI is often associated with VUR or urinary tract obstruction, associated with frequent recurrence of UTI and permanent renal damage [7]. VUR is seen in 40-50% infants and 30-50% children with UTI, and resolves with age. Its severity is graded using the International Study Classification from grade I to V, based on the appearance of the urinary tract on VCUG [3].
The presence of moderate to severe VUR, particularly if bilateral, is an important risk factor for pyelonephritis and renal scarring [8].

Material and Methods
This prospective observational study was conducted during Jan 2013 to Dec 2013 at Indira Gandhi Institute of Child Health, Bangalore. All children between 2 months to 24 months of age who present with first febrile UTI confirmed with a positive result on a urine culture were included in the study. Children with proven congenital malformations of the genito-urinary system or neurologic disease, with previous history of UTI, and history of prior antibiotic administration for the present complaints were excluded for the study. The eligible children were enrolled into the study after obtaining informed consent from the parents. A detailed history of the present complaints was taken in a prestructured proforma. A detailed clinical examination was done to rule out any congenital malformations of genito-urinary system or neurologic disease. Blood sample was drawn from all children for complete hemogram and blood CRP (quantitative) estimation and urine sample was collected by suprapubic aspiration for urine microscopy and culture before administration of parental antibiotics. Children with positive urine culture were enrolled in the study and antibiotics changed according to sensitivity pattern and continued for 10-14 days.
Bladder and renal USG was performed within 3 days of admission by an experienced pediatric radiologist. All abnormal USG findings were recorded, including dilatation of the pelvis, calyx, ureter or posterior urethra, thickened pelvic wall (circumferential hypoechoic rim delineated on each side by thin hyperechoic lines, the rim being thicker than 0.8 mm) [9] , or bladder (>3 mm in a full or 5 mm in an empty bladder) [10] , swelling or thinning of renal parenchyma, diffuse or focal increased renal echogenicity with absence of corticomedullary differentiation, renal abscess, and signs of renal hypodysplasia (i.e., small kidney, thinned or hyperechoic cortex, and cortical cysts) , or a duplicated collecting system. Renal length was regularly measured from the back, and the reference value was based on normograms described by Miranda et al. [11]. Renal length >2 standard deviation scores (SDS) or <−2 SDS was considered abnormal.
Technetium-99-m-DMSA scan was performed within 5 days after admission and was considered normal when homogenous uptake of the isotope was evident throughout the kidneys and the renal contour was preserved. APN was diagnosed if there was a focal or multiple areas of diminished isotope uptake with preserved renal contour or when the kidney was enlarged with diffuse decrease in uptake. Renal cortical scars were reported when there were defects in uptake associated with cortical loss and decreased renal volume. A small kidney with diffusely decreased isotope uptake and decreased differential renal function was reported as renal hypodysplasia.
A VCUG was performed 10-14 days after appropriate antibiotic treatment. Series of images was obtained to determine whether any liquid travelled backwards into 1 or both ureters when the patient emptied his or her bladder.
VUR was graded according to recommendations of the International Reflux Study in Children [12]. Children with grades I-II reflux are defined as low grade VUR and grades III-V reflux as high-grade VUR. In Children with bilateral VUR, the higher grade of the two ureters was recorded.
• VUR is the retrograde flow of urine from the bladder into the ureter and toward the kidney secondary to a dysfunctional vesicoureteric junction [13].
To assess the usefulness of imaging studies in screening for VUR, which indicated the need for VCUG, four approaches were compared: • USG alone.
• DMSA scan alone, • "OR rule" -defined as the routine performance of both USG and DMSA scan in all children with first febrile UTI. An abnormality on either study or on both constituted an indication for VCUG.
• "AND rule" -defined as performance of VCUG only if both USG and DMSA scan were abnormal. [14].
The diagnostic significance of renal US, DMSA scan, OR rule, and AND rule strategies in relation to the presence of all (I-V) or high-grade VUR was assessed by odds ratios (ORs) using logistic regression analysis. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV), positive and negative likelihood ratios (LR) for all and high-grade VUR of these methods were calculated. (Fig 1)

Results
A total of 48 eligible patients were admitted and investigated for the study purpose in Indira Gandhi Institute of Child Health, Bangalore. Approval from the institutional ethical committee of IGICH was taken. Baseline parameters of the study subjects were depicted in table 1. The most common causative agent was Escherichia coli, affecting 36 (75%) patients (table1).    (Table 3) The calculated sensitivity, specificity, PPV, NPV, and positive or negative LRs for all and high-grade VUR of the four methods is depicted in Table 4.

Discussion
Several reports suggest that acute DMSA scan with or without renal US may be used as screening test to identify high-grade VUR. The risk of APN and subsequent scarring, is substantially higher in patients with high-grade VUR [15,16]. Our approach aims at predicting children at high risk for VUR and evaluating them more thoroughly, while avoiding VCUGs in children who do not need them.
Ultrasound scanning is a good tool to assess obstructive uropathies, urolithiasis, ectopic kidney or ureter, and duplication of the collecting system with or without ureterocele. It is less reliable, however, for demonstrating APN, renal scarring and detecting VUR, perhaps because reflux is an intermittent, dynamic event [4,17,18].
The sensitivity of ultrasound scanning in our study was 75%, which is comparable with that of other studies. Three signs, including renal hypodysplasia, hydronephrosis, and hydroureter, were the most reliable clues for predicting high grade VUR.
In our study the sensitivity, specificity, positive predictive value and negative predictive value of ultrasound scanning for high grade VUR was 71.4%, 58.5%, 22.7% and 92.3% respectively. DMSA renal scanning is the most sensitive tool for detecting APN and renal scarring, with sensitivity for APN of approximately 90% [19]. DMSA scanning now increasingly plays a prominent role in evaluating the condition of children with a febrile UTI [1,15]. As with ultrasound scanning, however, normal scan results cannot absolutely exclude high-grade VUR.
In our study the sensitivity, specificity, positive predictive value and negative predictive value of DMSA scanning for high grade VUR was 100%, 46.3%, 24.1% and 100% respectively. DMSA scanning results in our study in detecting high grade VUR is comparable with other studies.
In Fouzas et al [20], performance of DMSA scanning was 7-14 days after diagnosis of UTI compared to other studies were its performed earlier (<5 days of admission in Tsai et al [21], within 2 days after diagnosis in Tseng et al [1] and median of 5 days after admission in Preda et al [22]). This could explain our higher sensitivity of DMSA scan for detecting high grade VUR than Fouzas et al [20].
In our study, on the basis of the OR rule strategy, 35 children would have undergone VCUG, of whom 7 had high-grade VUR (grade III, 2; grade IV, 1; grade V, 4) and 1 had low grade VUR(grade II). None of the VUR cases were missed if this strategy was taken as a screening test. The OR rule strategy thus had a highest sensitivity of 100% and a negative predictive value of 100% in screening high grade VUR. This is comparable with other similar studies, Lee et al [23] and Tsai et al [21].
In the present study, on the basis of the AND rule strategy, 17 children would have undergone VCUG, of whom 5 had high-grade VUR (grade IV, 1; grade V, 4). Of the 31 patients with no abnormalities on either test, 2 had grade III VUR and 1 had grade II VUR. The AND rule strategy has a better specificity (73.2%) compared to OR rule strategy, renal USG alone or DMSA scan alone, with the cost of decreased sensitivity (62.5%). If this strategy is taken as screening test then would have missed 3 patients with VUR.
With a high sensitivity and NPV, the OR rule strategy may be used as a good screening tool for detecting high-grade VUR. If VCUG is performed only in children with abnormal results on renal USG, DMSA, or both, 13 (27.1%) fewer VCUGs will be performed. In this study, we only assess the value of acute DMSA and renal USG in revealing high-grade VUR but do not provide information about long-term renal consequences. Because the benefit of identifying highgrade VUR is still in doubt, this screening strategy is useful only when intervention of high-grade VUR can be proved to change a patient's future risk or long-term outcome.

Conclusion
In conclusion, renal USG and DMSA scan have different but complementary roles in evaluating kidneys and urinary tract. The OR rule strategy has the highest sensitivity and NPV for screening high-grade VUR. Hence would recommend to use OR rule strategy as a screening tool for detecting high grade VUR and to perform VCUG in patients with abnormal result in OR rule strategy.