Mutations in hepatocyte nuclear factor 1B (HNF1B), which is a transcription factor expressed in tissues including renal epithelia, associate with abnormal renal development. While studying renal phenotypes of children with HNF1B mutations, we identified a teenager who presented with tetany and hypomagnesemia. We retrospectively reviewed radiographic and laboratory data for all patients from a single center who had been screened for an HNF1B mutation. We found heterozygous mutations in 21 (23%) of 91 cases of renal malformation. All mutation carriers had abnormal fetal renal ultrasonography. Plasma magnesium levels were available for 66 patients with chronic kidney disease (stages 1 to 3). Striking, 44% (eight of 18) of mutation carriers had hypomagnesemia (Ͻ1.58 mg/dl) compared with 2% (one of 48) of those without mutations (P Ͻ 0.0001). The median plasma magnesium was significantly lower among mutation carriers than those without mutations (1.68 versus 2.02 mg/dl; P Ͻ 0.0001). Because hypermagnesuria and hypocalciuria accompanied the hypomagnesemia, we analyzed genes associated with hypermagnesuria and detected highly conserved HNF1 recognition sites in FXYD2, a gene that can cause autosomal dominant hypomagnesemia and hypocalciuria when mutated. Using a luciferase reporter assay, we demonstrated HNF1B-mediated transactivation of FXYD2. These results extend the phenotype of HNF1B mutations to include hypomagnesemia. HNF1B regulates transcription of FXYD2, which participates in the tubular handling of Mg 2ϩ , thus describing a role for HNF1B not only in nephrogenesis but also in the maintenance of tubular function.

Nephrotic syndrome at presentation and subnormal eGFR at 1 year were adverse features. The finding that structural disease at onset predicted poor renal outcome at 5 years has implications for the design of therapeutic trials. Treatment of MCGN was variable and not evidence-based.

Growth failure is an important complication for children on dialysis. One possible influence on growth is renal bone disease. We reviewed the case notes of 35 children (23 boys), mean (range) age at inclusion 2.8 (0.25-8.9) years (17 children age <2 years), on dialysis for 2.0 (1-4.8) years, for growth, PTH, calcium and phosphate levels and medications. Data collection ended at age 10 years, commencement of growth hormone (rhGH) or renal transplantation. The mean (range) height standard deviation score (HtSDS) at the start of dialysis was -2.06 (-5.90 to 0.63). No change in HtSDS per year was observed; the median was -0.06 (-1.07 to 2.39). Children aged <2 years showed catch-up growth in the first year on dialysis; median change in HtSDS was 0.31 (-0.78 to 3.13). Mean plasma calcium and ionised calcium were approximately at the mid-point and phosphate just above the mid-point of the respective normal ranges. The median PTH level was 1.52 times the upper limit of normal and levels did not correlate with growth. Our results indicate that intensive nutritional therapy and phosphorus control aiming to keep PTH within the normal range prevents further loss of HtSDS in short children on dialysis. In some children under 2 years of age catch-up growth can be observed in the first dialysis year.