Impact of interlaboratory differences on the measurement of serum creatinine and symmetric dimethylarginine values: does this affect clinical decision-making?

Abstract

ObjectivesThis study aimed to assess the impact of interlaboratory differences (in analysis technique and reference intervals [RIs]) on the measurement of serum creatinine (sCr) and serum symmetric dimethylarginine (sSDMA) concentrations in cats and on classification by stage of chronic kidney disease (CKD).MethodsFor the interlaboratory discrepancy study, samples from 30 client-owned cats (healthy or diagnosed with CKD) with sCr in the range of 130-200&#x2009;&#xb5;mol/l and urine specific gravity <1.035 were retrospectively selected. A single batch of samples was analysed for sCr and sSDMA with one in-house laboratory machine and in three commercial laboratories. In addition, sCr values from 63 healthy cats aged &#x2a7e;10 years were used to calculate an age-specific RI for the in-house analyser.ResultsDespite a relatively good correlation (interclass correlation coefficient [ICC]&#x2009;=&#x2009;0.875), the in-house analyser consistently reported lower sCr values, while for sSDMA, systematically higher values were obtained in one commercial laboratory (ICC&#x2009;=&#x2009;0.707). A total of 22/30 (73%) cases had at least one laboratory reporting inconsistent sSDMA values in relation to the RI, and 23/30 (77%) for sCr. For subsequent staging using sSDMA, at least 1/4 laboratories indicated a different stage in 22/30 (73%) cases. For sCr, all laboratories classified the cats as stage 2; however, the in-house analyser classified 8/30 (27%) cats as stage 1. The RI (71-212&#x2009;&#xb5;mol/l) provided by the company for sCr measurement on the in-house machine was subsequently recalculated, resulting in a considerably lower upper limit (163&#x2009;&#xb5;mol/l) and a significantly (adjustedvalue&#x2009;=&#x2009;0.016) different proportion of cats with measurements outside the RI.Conclusions and relevanceClinicians must be careful when interpreting sCr and sSDMA values, as they may lead to discrepant outcomes due to interlaboratory variability. Moreover, age-specific RIs should be applied when available, facilitating the early detection of CKD in older cats.