Dyslipidaemia in children on renal replacement therapy

37. Ueda H, Miyazaki Y, Matsusaka T et al. Bmp in podocytes is essential for normal glomerular capillary formation. J Am Soc Nephrol 2008; 19: 685–694 38. Madhavan SM, O’Toole JF, Konieczkowski M et al. APOL1 Localization in normal kidney and nondiabetic kidney disease. J Am Soc Nephrol 2011; 22: 2119–2128 39. Divers J, Vaughan LK, Padilla MA et al. Correcting for measurement error in individual ancestry estimates in structured association tests. Genetics 2007; 176: 1823–1833 Received for publication: 27.7.2013; Accepted in revised form: 31.8.2013 Nephrol Dial Transplant (2014) 29: 594–603 doi: 10.1093/ndt/gft429 Advance Access publication 29 October 2013 Dyslipidaemia in children on renal replacement therapy Marjolein Bonthuis1, Karlijn J. van Stralen1, Kitty J. Jager1, Sergey Baiko2, Timo Jahnukainen3, Guido Franz Schaefer10 and Enrico Verrina11 1 ESPN/ERA-EDTA Registry, Department of Medical Informatics, Academic Medical Center, University of Amsterdam, Amsterdam, The ORIGINAL ARTICLE Netherlands, 22nd Children’s Hospital, Minsk, Belarus, 3Children’s Hospital, University of Helsinki, Helsinki, Finland, 4Department of Nephrology, University Children’s Hospital, Zurich, Switzerland, 5Faculty of Medicine, PJ Safarik University, Kosice, Slovak Republic, 62nd School of Medicine, University Hospital Motol, Charles University Prague, Prague, Czech Republic, 7Leeds General Infirmary, Leeds, UK, 8 Armand Trousseau Hospital, Assistance Publique–Hôpitaux de Paris (APHP), and University Pierre and Marie Curie, Paris, France, 9 Department of Pediatric Nephrology, Emma Children’s Hospital, Academic Medical Center, Amsterdam, The Netherlands, 10University of Heidelberg, Heidelberg, Germany and 11Nephrology, Dialysis, and Transplantation Unit, Gaslini Children’s Hospital, Genoa, Italy Correspondence and offprint requests to: Karlijn J. van Stralen; E-mail: recipients, use of cyclosporin was associated with significantly higher non-HDL and HDL levels than tacrolimus usage (P < 0.01). In transplant patients with eGFR < 29 mL/min/1.73 m2, the mean triglyceride level was 137 mg/dL (99% confidence interval (CI): 119–159) compared with 102 mg/dL among those with eGFR > 90 mL/min/1.73 m2 (P < 0.0001). Conclusions. Dyslipidaemia is common among paediatric ESRD patients in Europe. Young age and PD treatment are associated with worse lipid profiles. Although lipid levels generally improve after transplantation, dyslipidaemia may persist due to decreased graft function, high BMI or to the use of certain immunosuppressants. A B S T R AC T Background. Information on lipid abnormalities in end-stage renal disease (ESRD) mainly originates from adult patients and small paediatric studies. We describe the prevalence of dyslipidaemia, and potential determinants associated with lipid measures in a large cohort of paediatric ESRD patients. Methods. In the ESPN/ERA-EDTA registry, lipid measurements were available for 976 patients aged 2–17 years from 19 different countries from the year 2000 onwards. Dyslipidaemia was defined as triglycerides >100 mg/dL (2–9 years) or >130 mg/dL (9–17 years), high-density lipoprotein (HDL) cholesterol <40 mg/dL or non-HDL cholesterol >145 mg/dL. Missing data were supplemented using multiple imputation. Results. The prevalence of dyslipidaemia was 85.1% in peritoneal dialysis (PD) patients, 76.1% in haemodialysis (HD) patients and 55.5% among renal allograft recipients. Both low and high body mass index (BMI) were associated with a less favourable lipid profile. Younger age was associated with a worse lipid profile among PD patients. HDL levels significantly improved after transplantation, whereas no significant improvements were found for triglyceride and non-HDL levels. In transplant © The Author 2013. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. Keywords: children, dialysis, dyslipidaemia, renal replacement therapy, transplantation INTRODUCTION Cardiovascular disease is a major cause of morbidity and mortality in children with end-stage renal disease (ESRD) [1]. Paediatric dialysis patients are estimated to have an up to 1000fold increased cardiovascular mortality risk compared with age594 F. Laube4, Ludmila Podracka5, Tomás Seeman6, Kay Tyerman7, Tim Ulinski8, Jaap W. Groothoff9, Subjects The ESPN/ERA-EDTA registry collects annual data of paediatric patients undergoing renal replacement therapy (RRT) in Europe. Within the registry, individual patient data are collected regarding date of birth, gender, primary renal diagnosis, the initial and any subsequent RRT treatment modalities, as well as a variable set of anthropometric, biochemical and medicationrelated data. For the present study, only those countries providing data on total cholesterol, high-density lipoprotein (HDL) cholesterol and triglyceride levels from the year 2000 onwards were included. This included information for the following countries and periods: Belarus (2008–10), Czech Republic (2007–11), Denmark (2008), Estonia (2008–10), Finland (2000–10), Greece (2009–12), Croatia (2009), Hungary (2005–11), Iceland (2005– 11), Lithuania (2008–12), FYR Macedonia (2008–12), Norway (2008–10), Poland (2007–12), Portugal (2007–11), Serbia (2007– 12), Slovakia (2005–10), Slovenia (2007–10), Switzerland (2009) and the Netherlands (2007–12). Missing data for total cholesterol (3.4%), HDL (31.6%), triglycerides (14.3%), serum creatinine (16.7%), height (8.3%) and weight (7.9%) were imputed using a multiple imputation method as recommended by the STROBE guidelines [13, 14]. To test whether associations were similar in patients with complete information we performed sensitivity analyses among complete cases only, these analyses did not reveal different associations compared with the associations found in patients in whom missing data were imputed. Definition of variables Non-HDL cholesterol was calculated as total cholesterolHDL cholesterol. We defined dyslipidaemia by the presence of at least one of the following: hypertriglyceridaemia (triglycerides Dyslipidaemia in children on RRT Statistical analyses The number of lipid measurements recorded in the registry differed largely per patient. To correct for the correlations of measurements within the same patient, we used multinomial generalized estimating equations models [22] to estimate prevalence estimates of dyslipidaemia. In this way, a patient who had an elevated triglyceride level at one measurement and a normal triglyceride level at the second measurement, this patient contributed as ½ of a patient to the group of patients with elevated triglyceride levels and ½ to the group with normal triglyceride levels. To study factors associated with different lipid levels, lipid concentrations were log transformed and analysed with linear mixed models with both a random intercept and a random slope to account for the time between successive measurements within a patient, and adjustments were made for possible confounders. (...truncated)