Skip to main content

01.12.2018 | Research article | Ausgabe 1/2018 Open Access

Arthritis Research & Therapy 1/2018

Accuracy of the HumaSensplus point-of-care uric acid meter using capillary blood obtained by fingertip puncture

Arthritis Research & Therapy > Ausgabe 1/2018
Stéphanie Fabre, Pierre Clerson, Jean-Marie Launay, Jean-François Gautier, Tiphaine Vidal-Trecan, Jean-Pierre Riveline, Adam Platt, Anna Abrahamsson, Jeffrey N. Miner, Glen Hughes, Pascal Richette, Thomas Bardin
Wichtige Hinweise
A comment to this article is available online at https://​doi.​org/​10.​1186/​s13075-019-1891-1.



The uric acid (UA) level in patients with gout is a key factor in disease management and is typically measured in the laboratory using plasma samples obtained after venous puncture. This study aimed to assess the reliability of immediate UA measurement with capillary blood samples obtained by fingertip puncture with the HumaSensplus point-of-care meter.


UA levels were measured using both the HumaSensplus meter in the clinic and the routine plasma UA method in the biochemistry laboratory of 238 consenting diabetic patients. HumaSensplus capillary and routine plasma UA measurements were compared by linear regression, Bland-Altman plots, intraclass correlation coefficient (ICC), and Lin’s concordance coefficient. Values outside the dynamic range of the meter, low (LO) or high (HI), were analyzed separately. The best capillary UA thresholds for detecting hyperuricemia were determined by receiver operating characteristic (ROC) curves. The impact of potential confounding factors (demographic and biological parameters/treatments) was assessed. Capillary and routine plasma UA levels were compared to reference plasma UA measurements by liquid chromatography-mass spectrometry (LC-MS) for a subgroup of 67 patients.


In total, 205 patients had capillary and routine plasma UA measurements available. ICC was 0.90 (95% confidence interval (CI) 0.87–0.92), Lin’s coefficient was 0.91 (0.88–0.93), and the Bland-Altman plot showed good agreement over all tested values. Overall, 17 patients showed values outside the dynamic range. LO values were concordant with plasma values, but HI values were considered uninterpretable. Capillary UA thresholds of 299 and 340 μmol/l gave the best results for detecting hyperuricemia (corresponding to routine plasma UA thresholds of 300 and 360 μmol/l, respectively). No significant confounding factor was found among those tested, except for hematocrit; however, this had a negligible influence on the assay reliability. When capillary and routine plasma results were discordant, comparison with LC-MS measurements showed that plasma measurements had better concordance: capillary UA, ICC 0.84 (95% CI 0.75–0.90), Lin’s coefficient 0.84 (0.77–0.91); plasma UA, ICC 0.96 (0.94–0.98), Lin’s coefficient 0.96 (0.94–0.98).


UA measurements with the HumaSensplus meter were reasonably comparable with those of the laboratory assay. The meter is easy to use and may be useful in the clinic and in epidemiologic studies.
Über diesen Artikel

Weitere Artikel der Ausgabe 1/2018

Arthritis Research & Therapy 1/2018 Zur Ausgabe