Reliability of Urinary Albumin, Total Protein, and Creatinine Assays after Prolonged Storage: The Family Investigation of Nephropathy and Diabetes Rulan S. Parekh,* † W.H. Linda Kao, ‡ Lucy A. Meoni, †§ Eli Ipp,  Paul L. Kimmel, ¶ Janine La Page,  Carol Fondran,** William C. Knowler, †† Michael J. Klag, †‡ ‡‡ and the Family Investigation of Nephropathy and Diabetes Research Group Departments of *Pediatrics, † Medicine, Johns Hopkins University School of Medicine ‡ Epidemiology, § Biostatistics, and ‡‡ Health Policy and Management, and Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland;  Los Angeles Biomedical Research Institute, Harbor-UCLA Medical Center, Los Angeles, California; ¶ Kidney Urologic and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland, and Department of Medicine, George Washington University Medical Center, Washington, DC; **Department of Epidemiology and Biostatistics, School of Medicine, Case Western Reserve University, Cleveland, Ohio; and †† Diabetes Epidemiology and Clinical Research Section, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Phoenix, Arizona Background and objectives: This study investigated the effect of long-term storage at 70°C on urinary albumin, protein, and creatinine measurements in the Family Investigation of Nephropathy and Diabetes, a multicenter study designed to identify genes for diabetic nephropathy. Design, setting, participants, & measurements: Spot urine samples were collected at eight centers and shipped overnight on ice packs to a central laboratory. Samples were aliquotted and frozen at 70°C for a median of 8 d before initial assay. As part of quality control procedures to determine interassay variability, 351 replicate samples were retrieved from storage at 70°C after a median storage time between original and quality control analyses of 126 d (range 28 to 869 d). Freezer time was characterized as the difference in days between the initial assay and quality control assay. Percentage difference [(quality control  original/original)  100%] between samples was regressed on storage time and adjusted for original value, age, race, gender, hypertension, and diabetes. Results: After adjustment, freezer time per 30 d was associated with small decreases in percentage difference of urinary albumin (0.25%, P  0.02), total protein (0.23%, P  0.02), and albumin-to-creatinine ratio (0.34%, P  0.001). Urinary creatinine levels were not affected by freezer time (P  0.25). Conclusions: Measurements of urinary albumin, total protein, and albumin-to-creatinine ratio are minimally affected by storage at 70°C for approximately 2.5 yr. Prolonged storage results in small decreases of urinary albumin and protein that do not substantially affect phenotype classification of overt renal disease. Clin J Am Soc Nephrol 2: 1156 –1162, 2007. doi: 10.2215/CJN.01030207 U rinary albumin excretion is a sensitive and early marker of diabetic nephropathy, other forms of renal dysfunction, and endothelial permeability throughout the vascular tree. Minimally elevated levels of urinary albumin excretion, commonly termed “microalbu- minuria,” predict progression of kidney disease, develop- ment of cardiovascular disease, and total mortality (1,2). For these reasons, urinary albumin excretion is being estimated increasingly frequently in observational studies of renal and cardiovascular disease, often on frozen urine specimens that have been stored for prolonged periods. Storage of urine at -70°C is recommended when assay of freshly voided spec- imens is not feasible (3). Stability of quantitative measures of albumin in urine stored at -70°C for 180 d, however, is unknown. In addition, it is not known whether urinary proteins are stable across a wide range of chronic kidney disease conditions resulting in microalbuminuria to nephrot- ic-range proteinuria. We addressed the reproducibility of determinations of urinary albumin excretion using data from a study that was designed to identify genes that are associ- ated with diabetic nephropathy and other forms of renal disease: The Family Investigation of Nephropathy and Dia- betes (FIND) (4,5). As part of the quality control (QC) pro- cedures for this study, urine specimens were reassayed for albumin level after being stored for up to 900 d. Received February 27, 2007. Accepted July 6, 2007. Published online ahead of print. Publication date available at www.cjasn.org. Correspondence: Dr. Rulan S. Parekh, Welch Center for Prevention, Epidemiol- ogy & Clinical Research, 2024 Monument Street, Suite 2-511, Baltimore, MD 21205. Phone: 410-502-5745; Fax: 410-502-5746; E-mail: rsparekh@jhmi.edu Copyright © 2007 by the American Society of Nephrology ISSN: 1555-9041/206 –1156