[Extravasal position of central venous catheters despite unsuspicious ECG-guidance]

Does the electrocardiographic method for central venous catheter positioning distinguish between a correct intravasal and a malpositioned extravasal position?

24 cardiac surgical patients were enrolled in this prospective observational study. In 18 patients the left, in another 6 patients the right internal jugular vein was cannulated. Using a J-wire within a triple-lumen catheter the amplitude of the P-wave was measured at 3 different intravasal sites: Intra-1: (intravasal baseline electrocardiogram), i. e. 10 cm marking of the catheter on skin level; Intra-2: clear rise of the P-wave amplitude upon further insertion of the catheter; Intra-3: maximum P-wave amplitude. At this position the control of the catheter tip was achieved by means of transoesophageal echocardiography (TOE). Intraoperatively, another J-wire within a triple-lumen catheter was placed by the heart surgeon on 3 extravasal sites and the ECG was recorded: Extra-1: extravasal at the left innominate vein above the pericardial reflection; Extra-2: extravasal on the superior vena cava below the pericardial reflection; Extra-A: extravasal on ascending aorta below the pericardial reflection. The catheter was suture fixed with its tip in position Intra-3. Post surgery a chest radiograph was taken.

All catheter tips were visualised at the basis of the Crista terminals (border between right atrium and superior vena cava) by TOE control. The rise of the P wave amplitude at Intra-2, Extra-2 and Extra-A was highly significant compared to the base line at Intra-1 (Intra-1/Intra-2, Intra-1/Extra-2, Intra-1/Extra-A: p in each case < 0.001). The P wave amplitudes of the corresponding intra- and extravasal positions of the left innominate vein (Intra-1/Extra-1, n = 18, p = 0.096)) as well as those of the superior vena cava (Intra-2/Extra-2, n = 24, p = 0.859) did not differ.

The electrocardiographic method can not differentiate between intra- and extravasal position of a central venous catheter, and thus, presumably fails to identify malpositioning as a result of vascular perforation.