Background:Primary vesicoureteral reflux (VUR) is commonly managed with continuous antibiotic prophylaxis (CAP) to prevent urinary tract infections until the VUR resolves or if surgical repair is completed. In some cases, CAP may be ceased even in the face of persistent VUR if the risk of recurrent UTI is considered low. We hypothesized that CAP cessation is possible in most children even with persistent VUR, and that a nomogram could reflect the probability of UTI recurrence in these children.Methods: We identified all children diagnosed with primary vesicoureteral reflux (VUR) between January 2012 and December 2018. Our analysis focused on a subset of these patients for whom continuous antibiotic prophylaxis (CAP) was discontinued despite the lack of a VCUG demonstrating VUR resolution. Children were considered for CAP cessation when they remained free of breakthrough UTI’s after toilet training. The absence of febrile urinary tract infection (fUTI) or presence of fUTI once CAP was stopped were defined as success or failure, respectively. Exclusion criteria included secondary VUR, children undergoing anti-reflux surgery or those lost to follow-up after CAP cessation. Univariate followed by multivariate logistic regression model were used to investigate factors associated with success of VUR management post-CAP cessation. Performance of the model was assessed using AUC-ROC curve, and to enhance robustness, we applied bootstrapping with 1000 replicates. Additionally, a nomogram was developed to visually represent the predictive model. All analyses were conducted using R software.Results:Of 637 children with primary VUR, 463 (73%) were selected for CAP discontinuation at a median age of 37 months. A total of 411 (90.8%) were considered successful and 52 (9.2%) failed this management. Median follow-up after CAP cessation was 43 months. The logistic regression analysis identified significant predictors for success after CAP cessation (Table 1). On multivariate regression, presence of BBD was associated with a lower odd of success (OR=0.17, p<0.001), while lower reflux grades (I-III) were linked to higher odds of success (OR=2.33, p=0.016). The ROC curve yielded an AUC of 0.74. Bootstrapping with 1,000 replicates confirmed the stability of the model, with a mean AUC of 0.74. Furthermore, a nomogram was created based on the regression model, providing a visual guide to the predicted probabilities of successful VUR management (Figure 1).Conclusions: We demonstrate that a majority of children with VUR may be candidates for CAP cessation even when lacking a VCUG confirming VUR resolution. Bowel and bladder dysfunction and high VUR grade are predictors of failed management. While the small number of failures limited the power of our prediction model, we offer a nomogram that may assist in identifying children less likely to remain fUTI free after CAP cessation.
| Table 1. Univariate analysis | ||||
| Predictors | p-value | Success Management Odds Ratio | CI Lower | CI Upper |
| Female Gender | 0.037 | 0.4969697 | 0.257 | 0.959 |
| Bowel Bladder Dysfunction | <0.001 | 0.15427 | 0.082 | 0.287 |
| History of Breakthrough UTI | 0.035 | 0.4403226 | 0.205 | 0.945 |
| Low-grade VUR | 0.05 | 1.821429 | 0.998 | 3.322 |
