• D McGillivray, F M Ducharme, Y Charron, C Mattimoe, and S Treherne.
• Departments of Pediatrics, Epidemiology, Biostatistics, and Biochemistry, Montreal Children's Hospital, McGill University, Montreal, Quebec, Canada. mdmg@musica.mcgill.ca
• Ann Emerg Med. 1999 Jul 1;34(1):58-63.

Study ObjectiveIn children aged 1 month to 18 years, we sought to examine the correlation between venous and arterialized capillary blood gas values, and to determine whether the source of blood sample influenced the interpretation of the acid-base status and clinical management.MethodsIn a cross-sectional study, venous and capillary blood gas values were simultaneously obtained in acutely ill well-perfused patients treated in a pediatric emergency department. Intraclass correlation coefficients for capillary and venous measured gas values were calculated. Crude agreement and intraobserver concordance were calculated for responses of 2 intensivists to the interpretation and clinical management questions, based on capillary and venous gas results.ResultsIntraclass correlation coefficients for 78 capillary and venous paired measured gas values were.92 (pH), .80 (PCO 2 ), and .67 (PO 2 ). The alpha of concordance values between capillary and venous blood gas values, with 95% confidence intervals (CIs) were as follows, respectively, for physician A and B: interpretation, .61 (.47 to .73) and .48 (.41 to .55); need for bicarbonate,.85 (.73 to.97) and.80 (.72 to.88); and need for intubation .73 (.64 to .82), and .83 (.75 to .91).ConclusionIn the well-perfused patient, we believe that venous samples are an acceptable alternative to capillary blood samples for determination of blood gas values and for making clinical management decisions.

21 keywords...

hide…