• A López, B García, A Gómez, L González, N González, L Martín, and G Jaime.
• Grupo de Investigación en Anestesia y Cuidados Críticos, Instituto de Investigación Sanitaria Galicia SUR (ISS Galicia Sur), SERGAS-UVIGO, Vigo, Pontevedra, España; Servicio de Anestesia y Reanimación, Complexo Hospitalario Universitario de Ourense (CHUO), SERGAS, Ourense, España. Electronic address: angel.orense@gmail.com.
• Med Intensiva. 2019 Dec 1; 43 (9): 521-527.

ObjectiveTo evaluate the differences observed in ion and GAP anion determinations obtained by point-of-care (POC) blood gas versus laboratory biochemical testing, and to analyze the possible errors according to the limits of normality.Material And MethodsA descriptive, cross-sectional retrospective study was made to assess concordance between two diagnostic tests in patients admitted to the Critical Care Unit of Ourense University Hospital Complex (Spain), between July and November 2015, involving at least one coinciding biochemical test and POC determination. Patients under 18years of age were excluded.ResultsA total of 1,073 samples were analyzed. Lin's concordance correlation coefficients for sodium, potassium and chlorine were 0.87, 0.84 and 0.72, respectively. Kappa concordance of the normality limits for sodium, potassium and chlorine was 0.63, 0.74 and 0.32. The results indicated poor correlation of the anion GAP and null concordance between POC and biochemical testing, including the value corrected for albumin.ConclusionsPoor concordance was observed between the ion values as determined by biochemistry and blood gases; the two methods are therefore not interchangeable. Kappa agreement with normality limits was good for sodium and potassium, and weak for chlorine. Possible validity was noted in orienting the classification within the ion limits, with the exception of chlorine. No agreement was recorded in relation to the anion GAP, even that corrected for albumin.Copyright © 2018 Elsevier España, S.L.U. y SEMICYUC. All rights reserved.

32 keywords...

hide…