Risk factors for recurrent urolithiasis in children

To identify risk factors associated with recurrent kidney stones in a paediatric cohort in a Belgian tertiary centre.

Medical records of children with the first episode of urolithiasis between 1998 and 2016, followed at Ghent University Hospital initially and at least one-year follow-up were retrospectively reviewed. Patient characteristics, past medical history, presenting symptoms, the results of laboratory investigations and the applied management strategy were analysed. The significant variables from the univariate analysis were integrated into a backward conditional multivariate model.

Ninety-seven children were included in the analysis, of which 33 (34%) presented with at least one episode of stone recurrence. In the univariate analysis, body mass index (BMI) > 85th percentile and asymptomatic stones at initial presentation were associated with 1.8 and 0.1 times lower risk of recurrent stones, respectively (p = 0.020, 95% confidence interval (CI):0.368-8.749 and p = 0.017, 95% CI:0.014-0.921). In contrast, immobilization resulted in a 10-times higher risk (p = 0.002, 95% CI:1.968-50.005) and the need for technical intervention was associated with a 3.2- times higher risk (p = 0.017, 95% CI:1.297-8.084) of developing recurrent stones. On multivariate analysis only BMI >85th percentile was associated with a 15 times lower risk of stone recurrence (p = 0.030, 95% CI:0.006-0.739).

A possible explanation of reduced risk in patients with a BMI > 85th percentile may lie in a different metabolic profile. Immobilization as a risk factor can be explained by calcium metabolism, which is influenced by immobilization due to fractures, paralysis or motor disability because it causes resorption of the skeleton resulting in elevated blood calcium levels. This study showed that patients who presented without symptoms when the stones first occurred were less likely to have recurring kidney stones compared with patients with symptoms at initial presentation. When technical intervention was needed, we believe this is partly due to a larger stone burden, however we could not find an evidence-based explanation. The institutional protocol, which allowed to create a database with a limited number of patients, was lost to follow-up. Despite the retrospective setting some data were missing. There might also be a bias because the patients were followed-up at a tertiary centre. Possibly, our conclusions cannot be generalized toward the entire paediatric population.

Of all the factors investigated in our cohort, BMI >85 th percentile and asymptomatic stones are associated with a lower risk of stone recurrence. Conversely, immobilized patients and those who require technical intervention at initial presentation may benefit from an intense follow-up after the first episode of urolithiasis.