William Herrington , * , † , Nicola Illingworth * , ‡ , Natalie Staplin * , Aishwarya Kumar * , ‡ , Ben Storey * , † , ‡ , Renata Hrusecka * , ‡ , Parminder Judge * , † , ‡ , Maria Mahmood † , Sarah Parish * , ‡ , Martin Landray * , Richard Haynes * , † , ‡ , Colin Baigent * , ‡ , Michael Hill * , ‡ , Sarah Clark * , ‡
21 September 2016
albuminuria, randomized controlled trials, microalbuminuria, proteinuria, Albumins, Boric Acids, Confidence Intervals, creatinine, Freezing, Humans, Kidney Diseases, Randomized Controlled Trials as Topic, Specimen Handling, Temperature, Urinalysis, boric acid
Because there is substantial biologic intraindividual variation in albumin excretion, randomized trials of albuminuria-reducing therapies may need multiple urine samples to estimate daily urinary albumin excretion. Mailing spot urine samples could offer a convenient and cost-effective method to collect multiple samples, but urine albumin-to-creatinine ratio stability in samples stored at ambient temperatures for several days is unknown.
Patients with kidney disease provided fresh urine samples in two tubes (with and without boric acid preservative). Reference aliquots from each participant were analyzed immediately, whereas remaining aliquots were subject to different handling/storage conditions before analysis, including delayed processing for up to 7 days at three different storage temperatures (4°C, 18°C, and 30°C), multiple freeze-thaw cycles, and long–term frozen storage at −80°C, −40°C, and −20°C. We calculated the mean percentage change in urine albumin-to-creatinine ratio for each condition, and we considered samples stable if the 95% confidence interval was within a ±5% threshold.
Ninety-three patients provided samples with detectable albuminuria in the reference aliquot. Median (interquartile range) urine albumin-to-creatinine ratio was 87 (20–499) mg/g. The inclusion of preservative had minimal effect on fresh urine albumin-to-creatinine ratio measurements but reduced the changes in albumin and creatinine in samples subject to processing delay and storage conditions. The urine albumin-to-creatinine ratio was stable for 7 days in samples containing preservative at 4°C and 18°C and 2 days when stored at 30°C. It was also stable in samples with preservative after three freeze-thaw cycles and in frozen storage for 6 months at −80°C or −40°C but not at −20°C.
Mailed urine samples collected with preservative and received within 7 days if ambient temperature is ≤18°C, or within 2 days if the temperature is higher but does not exceed 30°C, are suitable for the measurement of urine albumin-to-creatinine ratio in randomized trials. Preserved samples frozen to −40°C or −80°C for 6 months before analysis also seem suitable.