Facebook Twitter YouTube

18561

Recurrence Free Survival Following the First Episode of Upper-Tract Calculi In a Pediatric Cohort

Saturday, October 20, 2012
Grand Ballroom A/B (Hilton Riverside)
Matthew Christman, Gregory E. Tasian, Angela Kalmus and Pasquale Casale, The Children's Hospital of Philadelphia, Philadelphia, PA

Purpose:   Few data are available regarding stone recurrence free survival in a pediatric cohort.  We aimed to determine the recurrence free survival following the first episode of upper-tract calculi.  We hypothesized that a model of demographic and stone factors could be developed to predict recurrence in subjects treated at a children's hospital.

Methods:   A database of 902 patients with ICD-9 codes consistent with urolithiasis from 2004–2012 was reviewed.  Subjects with a confirmed diagnosis of upper-tract urolithiasis and whose entire stone history had been abstracted were eligible.  Patients were excluded if clearance of their first stone episode was not confirmed by CT scan or renal/bladder ultrasound.  Information pertaining to demographics and individual stone episodes was analyzed.  Patients were censored at their last follow-up visit to our institution.  Kaplan-Meier survival estimates were generated.  Potential predictors of recurrence analyzed via a Cox proportional hazards model included:  age at initial presentation, gender, race, BMI, family history of stones, history of neurogenic bladder and/or bladder augmentation, and stone composition.  Factors that trended towards significance (α<0.10) on univariate analysis were used to construct a multivariate model of recurrence.

Results:   Radiologic clearance of the first episode was verified in 218 subjects, of whom 22% (48/218) presented with a recurrent stone.  Median [IQR] age was 11.9 years [6.9, 15.5].  The cohort underwent 288.4 person-years of observation.  Median upper-tract stone recurrence free survival was 4.4 (3.0, 5.7) years (Fig1).  14.9% recurred by the end of the first year.  Age at the initial presentation, gender, history of bladder augmentation, and a history of calcium oxalate stone/s predicted recurrence on univariate analysis; the remaining variables were not statistically significant predictors.  On multivariate analysis, no factor was a significant predictor of recurrence (Table1).

Fig1

Stone RFS_13Apr12..tif

Table1

Univariate Analysis

Multivariate Analysis

Hazard Ratio (HR)

Hazard Ratio (HR)

Variable

Estimate (95% CI)

p-value

Estimate (95% CI)

p-value

Age at Initial Episode

1.04 (1.01, 1.08)

0.020

1.06 (0.99, 1.13)

0.084

Gender (Female:Male)

2.12 (1.14, 3.97)

0.018

1.66 (0.72, 3.84)

0.238

Race (Relative to Caucasians)

     African-American

0.67 (0.21, 2.17)

0.500

     Asian

0 (0, …)

1.000

Family History of Stones

0.89 (0.45, 1.74)

0.727

Neurogenic Bladder

1.47 (0.71, 3.05)

0.299

Augmented Bladder

3.06 (1.08, 8.69)

0.036

1.22 (0.29, 5.11)

0.789

BMI

1.02 (0.97, 1.07)

0.486

Calcium oxalate

0.44 (0.21, 0.93)

0.033

0.51 (0.22, 1.19)

0.119

Calcium phosphate

1.09 (0.48, 2.45)

0.840

Struvite

1.70 (0.64, 4.49)

0.286

Uric Acid

0.79 (0.17, 3.67)

0.763

Conclusion:   Over half of pediatric patients with upper-tract urolithiasis may recur within 5 years.  Patients with a neurogenic bladder do not appear to be at increased risk of upper-tract stone recurrences.  Furthermore, stone composition cannot be used to predict recurrence.  We suspect that metabolic factors play the larger role in determining recurrence free survival.