Experimental Studies on Diagnostic and Therapeutic Aspects of Intraosseous Access

Reliable access to the circulation is paramount in most medical and surgical emergencies. When venous access cannot be expediently established, intraosseous (IO) access is indicated. This method has a high success rate even in relatively inexperienced hands and there is considerable clinical experience of IO administration of drugs and fluids. There is however limited evidence on the use of IO samples for laboratory analysis. Also, uptake of drugs during shock has not been extensively studied. Further, there have been concerns that analysis of IO samples may damage laboratory equipment. We have studied, in a porcine model, the use of IO samples for point of care analysis of blood gases, acid base parameters and blood chemistries in stable circulation, in experimental septic shock, and in hypovolemia after major hemorrhage, comparing IO samples with arterial and venous samples, and comparing IO samples from different sites. We have also studied coagulation assays on IO samples in stable circulation and after major hemorrhage. Furthermore, we have compared IO and intravenous administration of antibiotics in experimental sepsis. Average differences between IO and arterial/venous samples varied between the studied analytes. During stable circulation, average IO levels of blood gases, acid-base parameters, hemoglobin/hematocrit and several blood chemistries approximated venous levels relatively well. Differences in acid-base and blood gas parameters, and lactate, were more pronounced in hypovolemia, as well as in sepsis. The dispersion of the differences was often relatively large, indicating limited precision. Average differences between two intraosseous sites were small. Intraosseous samples were clinically hypercoagulable with a strong tendency to clot in vitro, and thromboelastography demonstrated shortened reaction times compared with venous samples. Major bleeding and hemodilution moderately affected the studied coagulation parameters. In endotoxemic animals with circulatory instability, concentrations of cefotaxime and gentamicin in samples from the pulmonary artery were comparable at 5 minutes after intraosseous and intravenous administration, and during a 3 hour observation period. In summary, agreement between analytes in intraosseous and conventional blood samples was variable and often unpredictable, especially during circulatory compromise. Intraosseous samples clinically appeared hypercoagulable, and thromboelastography confirmed this. High and comparable concentrations of cefotaxime and gentamicin were found after intraosseous and intravenous administration of equivalent doses, suggesting that uptake is acceptable during septic instability.