Background/Purpose: Circulating urate is an important biomarker, not only in the detection and management of gout, but also in assessing the risk of related comorbidity. The impact of biobanking processes and assay type on uric acid (UA) measurement has been the subject of limited study but is important given the role of biorepositories in biomarker assessments. Thus, we analyzed banked samples of citrated pUA (CpUA) and serum UA (sUA) using two separate assays and compared results to heparinized plasma UA (HpUA) from electronic health data. 

Methods: Two clinical laboratories, one based at an academic medical center (Lab A) and the other at a VA affiliate (Lab B), measured CpUA and sUA using banked samples from 30 rheumatic disease patients. Samples were collected using acid citrate dextrose ACD (plasma) and SST tubes (serum) between 2003 and 2017 and were stored at -70 °C until analysis. Lab A used the Beckman Coulter AU 5800 analyzer while Lab B used the VITROS URIC Slide method for UA measurement. UA measures were compared to lifetime average HpUA as part of routine care and abstracted from electronic health data from 15 of the 30 patients.

Results: There was no association of CpUA or sUA with year of sample collection. In contrast, mean (±SD) sUA was higher than CpUA at Lab A (5.9 ± 1.7 mg/dl vs. 3.5 ± 1.5, p≤0.01) and Lab B (6.0 ± 1.7 vs. 3.7 ± 1.9, p≤0.01). Serum UA values from Lab A and Lab B were highly correlated (r=0.98, p<0.01) with less than 5% variability overall. In contrast, CpUA values from the same samples differed on average by nearly 39% between Lab A and Lab B. Percent differences and correlations for the 4 measures are shown by lab and sample type (Table). Lifetime average HpUA concentrations measured as part of routine care were similar to sUA values obtained from both labs using banked samples, but were on average 50 to 80% higher than CpUA levels measured on banked samples (Table).

Conclusion: We observed a nearly 2-fold difference in UA values measured from banked serum compared to banked citrated plasma. Moreover, CpUA values showed greater variability when measured in different laboratories with different assays, variability that did not appear to be related to the time elapsed from sample procurement. Compared to CpUA values, sUA measures were much more strongly associated with HpUA values measured in ‘real-time’ as part of routine care. Although additional study will be needed to understand the mechanisms driving these observations, these results suggest that UA measurement of banked citrated plasma leads to marked assay variability in addition to falsely low concentrations.