One Poke or Two: Can Intravenous Catheters Provide an Acceptable Blood Sample? A Data Set Presentation, Review of Previous Data Sets, and Discussion

IntroductionThe purpose of this study was to show that acceptable blood samples can be collected through intravenous (IV) catheters. Hemolysis of blood samples and appropriate methods of blood sample collection can be a topic of controversy in an emergency department and throughout a hospital. This investigation was conducted by laboratory personnel and ED nursing staff at a moderately sized hospital in the northwestern United States. Nearly 9,000 blood draws were collected and categorized between May and August 2011 to determine hemolysis rates. At the start of this project, there was some question about whether blood collected from IV catheter starts in the emergency department provided an acceptable laboratory sample. By the end of the project, it was clear that low rates of hemolysis were consistently achievable by each of the 3 methods investigated.MethodsInformation was collected electronically as part of the standard laboratory intake and processing of samples. The level of hemolysis in a blood sample was measured spectrophotometrically by automated laboratory equipment as part of the sample intake process. The data were then cataloged and presented monthly. Because of the simplicity of this project and the clarity of the results, only simple summary statistics were performed to assist in interpretation of the data.ResultsThe rates of blood samples rejected because of hemolysis were as follows: 1.1% when collected from an IV catheter start, 0.8% when collected from an existing vascular access, and 0.1% when collected by venipuncture with a steel needle.DiscussionThe data show that low rates of samples rejected because of hemolysis are achievable by the collection of blood from an IV catheter start. With all 3 collection methods studied, the sample rejection rates because of hemolysis were well below the 2% level that has been cited as a benchmark best practice of the American Society of Clinical Pathology. The results of this investigation clearly support the use of blood samples collected from IV starts.