Original Contributions zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Differences in Drug Delivery with Peripheral and Central Venous Injections: Normal Perfusion WILLIAM G. BARSAN, MD,* JERRIS R. HEDGES, MS, MD,* HIROSHI NISHIYAMA, MD,t STEVEN T. LUKES, BSt The differences between central venous (CV) and peripheral ve- nous (PV) injections with respect to drug delivery times and peak concentrations in the left ventricle are examined. In this drug- delivery model, anesthetized dogs (n = 7) received both PV and CV injections of a radionuclide tracer. Measurements of radio- activity levels were made over each ventricle during normal per- fusion. There was no statistically significant difference in left ventricular delivery times or peak radioactivity levels between the two different routes. When delivery time was corrected for variations in flow rate, a small but statistically significant dlf- ference was seen, the CV route being faster than the PV route (P < 0.05). The magnitude of the observed difference for normal perfusion (2 seconds) is unlikely to be of clinical significance. (Am J Emerg Med 1986;4:1-3) Central venous catheters are frequently used for drug administration in critically ill patients. Intuition suggests that higher peak drug levels can be achieved when drugs are given via the central venous (CV) route rather than the peripheral venous (PV) route. Moreover, the peak concentration of a centrally ad- ministered drug may be reached significantly sooner. Such an effect could contribute to drug toxicity from transient elevations in drug levels when drugs such as lidocaine, aminophylline, or phenytoin are given via the CV route. There are few data from human or an- imal studies on differences in peak injection concen- trations or delivery times with the CV versus the PV route. During closed-chest cardiopulmonary resuscitation (CPR) in dogs, Hedges et crl’ found that the CV route From the *Department of Emergency Medicine and the tE. L. Saenger Radioisotope Laboratory, Department of Radiology, University of Cincinnati Hospital, Cincinnati, Ohio. Manuscript received January 17, 1965; revision received April 4, 1985; revision accepted May 9, 1985. Address reprint requests to Dr. Hedges: Department of Emer- gency Medicine, Pavilion C,, Room 409, 234 Goodman St., Cin- cinnati, OH 45267-0769. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Key Words: Central venous, drug delivery, peripheral venous. of administration provides approximately a 70-second advantage over the PV route for delivery of a radio- active marker to the left ventricle. They also found that the peak left ventricular radioactivity level found during CPR was similar for the two routes. In a study of closed-chest CPR in patients who were not resus- citated, Kuhn et ul’ found that peak marker concen- tration was present within 30 seconds with the CV route, but marker did not become detectable until at least 90 seconds after PV injection. Furthermore, the peak marker concentration was considerably lower after PV injection and had not peaked by 5 minutes. In an animal model of open-chest CPR with cardiac output maintained at 30% of control, Barsan et al3 found that peak levels of lidocaine were approximately 33% greater when administered by the CV route, al- though the time to peak levels was not significantly different. We examined the effect of different routes of radio- nuclide administration on left ventricular delivery time and peak levels of radioactivity during normal perfu- sion in an anesthetized canine model. MATERIALS AND METHODS Seven mongrel dogs weighing 18-22 kg were studied using a methodology reported by Hedges et al.’ All dogs were anesthetized with 30 mg/kg pento- barbital, intubated with a cuffed endotracheal tube, and allowed to breathe spontaneously. For the PV in- jections, a distal forelimb vein was catheterized with an l&gauge catheter 1.5 inches in length. For the CV injections, an external jugular vein was cannulated in the mid-cervical area with a 22-gauge catheter 8 inches in length. The catheter was inserted 4 inches into the external jugular vein. Intrathoracic placement of the CV line in the superior vena cava was later verified visually on the scintillation monitor at the time of ra- dionuclide injection in all animals. Line placement was further verified in several animals at necropsy. The animals were then placed in a supine position