Purpose: We investigated the natural history of nephrocalcinosis in premature infants treated with furosemide and attempted to identify factors to predict infants most at risk.
Materials and methods: We evaluated 13 preterm infants in this longitudinal pilot study. During hospitalization and while receiving a loop diuretic nephrocalcinosis developed in each patient. Patients were divided into groups based on resolution (6) and nonresolution (7) according to spontaneous resolution of nephrocalcinosis at any point during followup. The 2 groups were compared to each other and to a control group.
Results: Mean followup after discontinuation of furosemide in the resolution versus nonresolution groups was 10.3 and 7.7 months, respectively. Between the 2 groups there was no significant difference in average gestational age, birth weight, number of days hospitalized or on furosemide, or total furosemide dose. Mean calcium-to-creatinine ratio while receiving furosemide at the time nephrocalcinosis developed was 0.38 in the resolution group but 2.23 in the nonresolution group (p < 0.005). Initial calcium-to-creatinine ratio in age matched infants who did not have nephrocalcinosis was 0.4. Frank renal stones developed in 2 of the 7 patients without resolution and 0 of the 6 with resolution. When nephrocalcinosis resolved, it was at a mean of 5.2 months following discontinuation of the diuretic.
Conclusions: Early data indicate that nephrocalcinosis resolves in approximately 50% of premature infants 5 to 6 months after discontinuation of furosemide. The only factor that appears to be predictive of the infants who will have resolution is the calcium-to-creatinine ratio when nephrocalcinosis is diagnosed. In patients without resolution this ratio is much higher than in age adjusted normal controls, while in those with resolution it appears normal for age.