Influence of short-term venous stasis on clinical chemistry testing

Abstract

Control and standardization of preanalytical variability is a critical factor for achieving accuracy and precision in laboratory testing. Although venous stasis from tourniquet placement during venepuncture should be minimized, as it has been claimed to account for spurious and significant variations for several analytes in plasma, there is controversy surrounding its real impact on laboratory testing. The aim of the present study was the investigation of the influence of short-term venous stasis on routine biochemical testing, by measuring the plasma concentration of 12 common analytes, including proteins, protein-bound substances, enzymes and electrolytes, in plasma specimens collected either without venous stasis or following the application of standardized external pressure of 60mm Hg using a sphygmomanometer for 1 and 3min. Although the overall correlation between measures was acceptable, the pattern of change was mostly dependent on the length of stasis, size and protein-binding characteristics of the analytes, achieving clinical significance for albumin, calcium and potassium after 1-min stasis, and alanine aminotransferase, albumin, calcium, chloride, total cholesterol, glucose and potassium after 3-min stasis. Statistically significant differences could be observed in seven (alanine aminotransferase, albumin, calcium, total cholesterol, creatine kinase, iron and potassium) and ten (alanine aminotransferase, albumin, calcium, chloride, total cholesterol, creatine kinase, creatinine, glucose, iron and potassium) out of the 12 analytes tested, after 1- and 3-min venous stasis, respectively. The most clinically significant changes from standard venepuncture, when compared to the current analytical quality specifications for desirable bias, occurred for potassium (1-min stasis, −2.8%; 3-min stasis, −4.8%, both p<0.001), calcium (1-min stasis, +1.6%, p<0.05; 3-min stasis, +3.6%, p<0.001) and albumin (1-min stasis, +3.5%; 3-min stasis, +8.6%, both p<0.001). As most of these effects are dependent on the stasis time during venepuncture and biochemical or physiological characteristics of the analyte, these variations could likely be anticipated, allowing the most appropriate preventive measures to be adopted.