Low Risk for Urinary Tract Infection After Voiding Cystography in Children Without Dilated Vesikoureteral Reflux

Authors: Emelie Widhe, Mette Hambraeus, Erik Hedström, Anna Börjesson, Martin Salö

Abstract

Aim

Urinary tract infection (UTI) is the most common and significant complication after voiding cystourethrography (VCUG) but the reported incidence in children varies highly. Therefore, the aim of this study was to evaluate the rate and possible risk factors for UTIs after VCUG.

Method

A multicentre retrospective cohort study of children < 15 years of age undergoing VCUG between 2017 and 2022 was conducted. Lower UTI (cystitis) and febrile UTI (pyelonephritis) within 14 days of VCUG were registered, defined by clinical assessment and a positive urinary culture.

Results

One thousand and one examinations were included (median age 1 year, 52% boys). Previous febrile UTI was present in 58%. Pathology was found in 47% of children, of which 83% had vesicoureteral reflux (VUR). Post VCUG UTI occurred in 34 (3.4%) children within 14 days, of which 60% were within 7 days and 88% were febrile. Only dilating VUR increased the risk for post VCUG UTI (adjusted odds ratio 5.4 [2.6–8.7] p < 0.001).

Conclusion

There was a low rate of post VCUG UTIs. Since children with dilated VUR are at risk for chronic kidney injury, targeted interventions after VCUG could possibly lower additional UTIs in this group.

Introduction

A voiding cystourethrogram (VCUG) is used to evaluate the anatomy and function of the urinary tract, including visualisation of the urethra, bladder, ureters and renal pelvis [1]. The most common indication is evaluating the presence of vesicoureteral reflux (VUR) in children with previous febrile urinary tract infections (UTI). However, VCUG is invasive and comes with risks such as development of a post‐procedural UTI.

Febrile UTIs post VUCG may result in urosepsis and increases the long‐term risk of kidney damage [2, 3]. Urinary tract infections occur after 0%–30% of VCUG examinations [4, 5, 6, 7, 8, 9, 10, 11] with more recent studies reporting a lower rate of 1.0%–4.3% [9, 10, 11]. Reported UTI rates and risk factors for post‐VCUG UTI may not be directly comparable across studies given variations in bacterial spectra, resistance patterns, and VCUG protocols. Larger studies from different continents are therefore needed and may serve as a benchmark for clinicians and help improve pre‐procedural information to children and parents.

This study therefore aimed to investigate the rate and risk factors for UTI after VCUG in children under 15 years old.

Materials and Methods

The study was approved by the regional ethical board before the start of the study (Ref. no 2010/49). Informed consent was waived by the ethical board.

Study Design

This is a retrospective multicentre study of all children in Skåne county, Sweden, who underwent VCUG between the 1st of March 2017 and 31st of December 2022. Exclusion criteria were age over 15 years and incomplete examinations. Repeated VCUG examinations of the same children were included. All VCUGs were performed by radiology nurses and evaluated by paediatric radiologists. Most pathological VCUGs are also re‐evaluated and demonstrated on paediatric radiology rounds, often by another paediatric radiologist.

The following variables were examining hospital, age, sex, underlying known urogenital malformation or other condition, presence of any type of bladder drainage before the examination such as an urinary or suprapubic catheter, constipation, previous febrile UTI, continuous antibiotic prophylaxis (CAP) and type, pre‐VUCG antibiotic prophylaxis and type, VCUG findings, post‐VCUG infection and type (lower UTI or febrile UTI) and how many days after the procedure the infection occurred.

Settings and Population

The VCUGs were performed in five different hospitals in the county of Skåne, Sweden: Malmö, Lund, Helsingborg, Ystad, and Kristianstad; serving a total of around 1.4 million inhabitants, of which 20% are < 15 years. Lund and Malmö are university hospitals and the rest are county hospitals. Highly specialised paediatric surgical and urologic care are centralised to Lund.

The VCUG was performed according to local protocol similar to international guidelines [12] during the included years and was the same for all included centres. No cyclic studies were performed. Per protocol, all children should receive oral antibiotic prophylaxis (Trimetroprim or Furadantin) at the time of placement of the urinary catheter. A child already on antibiotic prophylaxis, received the other type of antibiotics (Trimetroprim or Furadantin 2 mg/kg).

Primary Outcome

The primary outcome was rate of post‐VUCG UTI which was defined as an UTI within 14 days after the VCUG. Lower UTI and febrile UTI were defined based on clinical assessment and a positive urine culture. Lower UTI was included to capture the full spectrum of morbidity following VCUG in children. The period of 14 days was chosen since symptoms of UTI can be more subtle in children leading to parent and/or doctor delay, but also to be as transparent as possible when using the results of this study for informing children and families.

Definitions

Underlying urogenital malformations/conditions including findings at the VCUG and their definitions hydronephrosis (anterior–posterior measurement > 10 mm), hydroureteronephrosis (distal ureter > 6 mm), VUR (graded accordingly to the international classification system, grade I‐V; the highest grade of VUR was counted when bilateral), complete duplex anomalies (seen on any radiological examination), other kidney malformations (horseshoe kidney, renal dysplasia, renal aplasia, polycystic kidney disease, multicystic dysplastic kidney and ectopic pelvic kidney), anorectal malformations, Hirschsprung disease, bladder trabeculation, spina bifida, kidney stone disease, any malignant urogenital tumour, any known genetic syndrome and urethral and genital malformations (posterior urethral valves, urethral stricture, epispadias and hypospadias), bladder diverticula and vaginal reflux. Presence of a urinary or suprapubic catheter was defined as a catheter in the bladder for > 4 weeks before the VCUG. Constipation was defined as having ongoing treatment with laxatives. A febrile UTI was defined as clinical symptoms and blood tests consistent with febrile UTI, together with a positive urinary culture. A urinary culture with growth > 100.000 CFU/mL was considered significant.

Statistics

The continuous variable (age) had a clear non‐normal distribution and is therefore presented as median (range). Categorical variables are presented as the absolute number and percentage, n (%). For comparison between the two groups, i.e., with and without urinary tract infection after VCUG, the Mann–Whitney U test was used for continuous data and the chi‐square test for the categorical variables. Variables with p < 0.05 were further analysed with univariate and multivariate logistic regression analysis and presented with unadjusted and adjusted odds ratios (OR) and 95% confidence intervals (95% CI). Independent variables in the regression analysis were age, sex, underlying urogenital condition, presence of urinary or suprapubic catheter, previous febrile UTI, CAP, pre‐VCUG antibiotics, any pathological VCUG finding, and the presence of VUR on VCUG. A sensitivity analysis was performed to assess risk for post‐VCUG UTI as related to non‐dilating (grade I–II) versus dilating (grade III–V) VUR. All analyses were performed in the software SPSS version 29 (IBM Corp). p values < 0.05 were considered to show significant differences.

Results

A total of 1018 VCUG examinations were performed during the study period, of which 17 (1.7%) were excluded due to incomplete examinations related to issues with the catheter or technical failures, leaving 1001 examinations on 912 patients for analysis.

Cohort Characteristics and VCUG Findings

The median age at the time of VCUG was 1 year (1 day–14.7 years) and approximately half (51.5%, n = 516) of the examinations were performed in boys. Previously known urogenital malformations or conditions were present in 64% (n = 639) of the examinations. Hydronephrosis was most common (28%, n = 276) followed by hydroureteronephrosis (19%, n = 194), VUR (17%, n = 165) and duplication anomalies (12%, n = 113). A long‐term urinary or suprapubic catheter drainage was present in 8% (n = 75). The majority of the VCUGs were performed in children who had a previously diagnosed febrile UTI (Table 1).

The majority of examinations were performed in children on CAP (58%, n = 575) and administration of pre‐VCUG prophylaxis was recorded in 82% (n = 824) of cases. Both CAP and pre‐VUCG prophylaxis were given in 46% (n = 459), and no prophylaxis at all was recorded for 3% (n = 28) of examinations. Trimethoprim and Nitrofurantoin were the most common antibiotics as both CAP and prophylaxis (Table 2).

Pathological VCUG findings were evident in 47% (n = 470) of examinations, of which VUR made up 84% (n = 395). The most common grade of VUR was IV (33%, n = 131) and bilateral VUR was found in 45% (n = 177). Other common pathological findings were posterior urethral valve/urethral stricture and bladder trabeculation/megacystis (each 3%) (Table 3).

Post‐VCUG Urinary Tract Infection

Post‐VCUG UTI was seen in 3.4% (n = 34) cases, of which 88% were febrile (n = 29). In almost 60% of cases (n = 20) the post‐VCUG UTI occurred within 7 days (Figure 1). Post VCUG UTI was associated with younger age (0.53 vs. 1.01 years, p = 0.014), higher rate of urinary/suprapubic catheter (18% vs. 7%, p = 0.019), higher rate of CAP (82% vs. 57%, p = 0.001), higher rate of any pathological VCUG finding (88% vs. 46%, p < 0.001), and higher rate of VUR (74% vs. 38%, p < 0.001). No differences were found between the two groups regarding sex, frequency of urogenital malformations, frequency of previous febrile UTI, or pre‐VCUG prophylaxis (Table 4). A post‐VCUG UTI was only seen in 5 (0.01%) of examinations in children without pre‐existing conditions or absence of VUR. No child over ≥ 1 year of age without a known pre‐existing urogenital malformation had an infection.

When significant variables from the univariate analysis were included in a logistic regression, only VUR remained significant; aOR 4.56 (95% CI, 2.0–10.4), p < 0.001 (Table 5). In the sensitivity analysis, only dilated VUR (aOR 5.4 [2.6–8.7], p < 0.001) remained significant.

Discussion

This is the largest European study evaluating the rate of post‐VCUG UTI in children. It shows an overall UTI rate of 3.4%, a febrile UTI rate of 2.9% and almost 60% of the former occurred within 7 days. Dilated VUR was the only risk factor associated with UTI after VCUG in adjusted analyses.

The post‐VCUG UTI rate in the current study was 3.4%, which is in accordance with the most recent studies showing a rate between 1.0% and 4.3% [9, 10, 11]. The current study differs from previous studies as it included 14 days follow‐up instead of the commonly used 7 days and also included both lower UTI and febrile UTI as separate endpoints. Febrile UTI within 7 days post VCUG was 2% in the present study. We consider it a strength to include infections up to 14 days after VCUG since symptoms of UTI can be more subtle in children, which may lead to parent and/or doctor delay and may at least in part explain the lower incidence found in some studies. With inclusion of lower UTI in the UTI rate, a more complete picture of the post‐VCUG morbidity is provided. A 4% rate of post‐VCUG febrile UTI was found in a study of 300 children < 3 years of age [9]. They found no associated risk factors or decreased risk with antibiotic prophylaxis [9]. Another study that also included older children also reported a similar UTI rate and found that VUR grade IV–V was associated with an increased risk [10]. The lowest post‐VCUG febrile UTI rate (1%) was reported from a large study of 1200 American children [11]. Nearly all cases of febrile UTI occurred in children with pre‐existing urologic diagnoses [11].

Older studies report UTI rates between 0% and 30% [4, 5, 6, 7, 8] but with more heterogeneous cohorts than more recent papers [9, 10, 11]. To summarise, the differences between findings in the current study and previous studies likely relate to patient age, number of included children, definition of UTI, post‐VCUG time frame, rate of underlying conditions and probably differences in causative bacteria and bacterial resistance patterns between countries. The present study includes a large and diverse multicentre cohort, with a high rate of underlying conditions and pathological VCUG findings. This leads us to conclude that the febrile UTI rate of almost 3% < 14 days found in the current study is acceptable.

Univariate analyses showed that children who developed a post‐VCUG UTI were younger and more likely to have a urinary/suprapubic catheter, use CAP and have VUR. However, in the adjusted regression analysis, only dilating VUR was associated with increased risk of post‐VCUG UTI. This probably reflects the higher prevalence of VUR in lower ages, that many patients with VUR are on CAP and that some children with severe (often bilateral) VUR are drained consistently with a catheter during a few months. An association between VUR and post‐VCUG UTI was found in two of three of the most recent previous studies [10, 11]. In the study by Johnson et al. [11], all children with UTI had VUR on VCUG, while Martins et al. [10] illustrated an association between UTI and VUR grade IV and V. The association between VUR and post‐VCUG UTI is most likely explained by these children generally having a higher overall infection rate and that a lower UTI more easily develops into a febrile UTI with refluxing urine. Another likely reason is that these children lack one of the most important protective mechanisms against infection—complete bladder emptying. While it may be argued that a 3% UTI rate is low and that interventions trying to reduce this rate may be unnecessary, these children often experience several febrile UTIs and have a long‐term risk of chronic kidney disease. Therefore, we believe that the outcome could be improved further by specific interventions such as better parental counselling, early clinical follow‐up with laboratory testing, selective post‐procedural antibiotic treatment and possibly in select cases bladder drainage a few days following VCUG.

The univariate analysis showed a higher risk for post‐VUCG UTI in younger children but is as mentioned probably a proxy for a higher rate of VUR in this age group. However, there were no infections in children ≥ 1 year of age and without any known urogenital malformation. This age limit could therefore probably serve as an age cut‐off in otherwise healthy children when considering not giving prophylactic antibiotics.

The retrospective design makes data regarding lower UTI cases less reliable, but we believe this limitation does not apply to cases of febrile UTI. voiding cystourethrography images were not re‐evaluated for the current study specifically, but most children with pathological findings had their images demonstrated at paediatric radiology rounds as part of the clinical routine. Most of the other independent variables are less sensitive to the retrospective design, except for the administration of prophylactic antibiotics, which sometimes was difficult to determine with certainty.

Conclusion

The overall rate of post‐VCUG UTI was 3.4% and the rate of post‐VCUG febrile UTIs was 2.9%. This is in line with previous studies. Since dilated VUR is a known risk factor for the development of chronic kidney disease and this was the only statistically significant risk factor of post‐VCUG UTI in the current study, targeted interventions could potentially reduce additional UTIs in this patient group.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.