Article
Long-term stability of C-peptide levels in 24-h urine samples of healthy children participating in the DONALD Study
Search Medline for
Authors
Published: | September 8, 2005 |
---|
Outline
Text
Background
Urinary C-peptide (UCP) is a non-invasive measure of insulin secretion. Information on its long-term stability is required for retrospective analyses, e.g. examining long-term patterns of insulin secretion in healthy children and adolescents.
However, assessment of UCP long-term stability with split samples appears inappropriate since the long-term stability of the immunoassay system itself cannot be guaranteed [Ref. 1]. Alternatively, the concentration of UCP measured in similar groups of subjects according to standardized procedures can be compared between different time points [Ref. 2].
This study therefore compares the UCP excretions in 24-h urine samples stored at – 20° C between 7-8 year old healthy children in 1990, 1996 and 2002.
Participants and methods
Eligible were participants of the DOrtmund Nutrition and Anthropometric Longitudinally Designed Study with 24-h urine collections and a simultaneously recorded dietary diary. Among these, three comparable samples (n=40) aged 7-8 years in 1990, 1996 and 2002, were randomly selected (total n=120). UCP excretions were measured using a enzyme-linked immonosorbent assay (ELISA). Equivalence was tested applying the conservative critical values by Wiens and Iglewicz, which base on the ratio of the maximum pairwise difference in sample means to the standard error [Ref. 3].
Results
Applying ANOVA statistics, urinary data did not differ significantly between the three sampling periods (median (Q1, Q3) UCP 1990, 1996 and 2002: 6.9 (5.0,10.4) vs 7.2 (5.6, 10,0) vs 7.5 (5.5, 10.8) nmol/day, p=0.2 based on log-transformed data). However, according to the conservative equivalence test, log-transformed UCP excretions were not equivalent between the three sampling periods (mean (SD) log-transformed UCP: 1.91 (0.63) vs 1.97 (0.44) vs 2.04 (0.40) nmol/day, p>0.10). Yet, consideration of potential anthropometric and dietary confounders revealed that the slight increase of UCP with time was explained by body weight and dietary glycemic load. Thus, adjustment of UCP excretions for both body weight and glycemic load resulted in bioequivalence (mean (SD) log-transformed UCP: 0.92 (0.68) vs 0.91 (0.45) vs 0.96 (0.43), p<0.05).
Discussion
This study demonstrates that 24-h UCP excretions in healthy children remained stable for 12 years when corrected for the children’s body weight and their dietary glycemic load. These retrospective estimates of UCP can thus be regarded as reasonably precise.
References
- 1.
- Clark PM. Assays for insulin, proinsulin(s) and C-peptide. Ann Clin Biochem 1999; 36: 541-64.
- 2.
- Griefahn B, Remer T, Blaszkewicz M, Brode P. Long-Term stability of 6-hydroxymelatonin sulfate in 24-h urine samples stored at -20 degrees C. Endocrine 2001; 15: 199-202.
- 3.
- Wiens BL, Iglewicz B. On testing equivalence of three populations. J Biopharm Stat 1999; 9: 465-83