NE-AUA 2006 Annual Meeting, September 28 - 30, 2006, The Westin Hotel & Rhode Island Convention Center Providence, Rhode Island
Back to Scientific Program
Back to Annual Meeting
Establishment of New Nomograms for Predicting Resolution of Vesicoureteral Reflux
Carlos R. Estrada, M.D.1, Carlo C. Passerotti, M.D.1, Dionne Graham, Ph.D.1, Craig A. Peters, M.D.2, Ilina Rosoklija, M.S.1, Aimee Baysinger, B.A.1, Samuel H. Eaton, M.D.1, Alan B. Retik, M.D.1, Hiep T. Nguyen, M.D.1.
1Children's Hospital Boston, Boston, MA, USA, 2University of Virginia, Charlottsville, VA,

Background: The AUA guidelines on vesicoureteral reflux (VUR) are widely utilized for counseling and clinical decision making, but their data are incomplete and not validated. Using a longitudinal VUR database, we aimed to formulate nomograms based on age, sex, grade, and laterality (uni- v. bilateral) to predict the probability of resolution by a given time point (cumulative probability).
Methods: 1876 patients from the database with follow-up periods of up to 10 years were included in the analysis. Demographic data and results of yearly radiological studies were recorded until VUR resolution, surgery, or loss to follow-up occurred. In bilateral VUR, the higher grade was used. Resolution was defined as no reflux on radiological imaging. Cox proportional hazards regression was used to model time to resolution as a function of the demographic and clinical variables. Subjects who underwent surgery were treated as censored at the time of surgery. The resulting model was used to construct nomograms predicting the cumulative probability of VUR resolution over time.
Results: Multivariate analyses demonstrated that age at diagnosis (<2y, 2-5y, >5y), sex, grade ( 2, 3, 4), and laterality were significantly associated with time to VUR resolution. In general, younger ages (hazard ratio (HR): age<2y 2.1; age 2-5y 1.5), male sex (HR: 1.2), lower grades (HR: grade 2 2.3; grade=3 1.5), and unilateral VUR (HR: 1.5) were significantly associated with earlier times to resolution. Specific predicted cumulative probabilities of VUR resolution at yearly intervals from diagnosis (1-6 years) were calculated for every combination of grade, sex, laterality and age.
Conclusions:
Our analysis demonstrates that time to VUR resolution is dependent on age, sex, grade, and laterality. Using a Cox proportional hazard model we provide nomogram tables which contain estimates of the cumulative probability of yearly VUR resolution as a function of these variables over time. These data are useful for clinical counseling as well as clinical management decisions.


Back to Scientific Program
Back to Annual Meeting