We performed this prospective study over a 12-month period in the cardiothoracic section of the University Hospital of Jena following the approval of our institutional ethics committee. Each patient gave informed written consent. Two hundred thirteen patients undergoing elective cardiac surgery were eligible for the study. Patient characteristics recorded were age, weight, height, and type of surgery (Table 2). Exclusion criteria were ECG rhythm other than sinus rhythm after induction of anesthesia, or a contraindication to TEE (eg, gastric or esophageal pathology). Poor quality of CXRs was not considered an exclusion criterion. At our institution, CVC placement and TEE monitoring are routine for these surgeries. The patients were randomly assigned using computer-generated numbers to the right (group R) or left (group L) internal jugular vein approach. After induction of general anesthesia, the internal jugular vein was cannulated with a triple-lumen CVC (Certofix Trio SB 730, length 30 cm, 7F; B. Braun Melsungen AG; Melsungen, Germany) using sterile Seldinger technique. Under ECG guidance (Einthoven, lead II), the CVC was advanced until the P-wave increased noticeably from baseline. At that point, all three lumina of the catheter were tested for free back-flow. If there was any difficulty in aspirating venous blood, the catheter was rotated up to 360°. If problems persisted, the catheter was advanced in steps of 1 cm. Arrhythmias occurring during this procedure were recorded. After suture fixing the CVC, a TEE probe (multiplane probe, 6.2 MHz/HP, Sono 5500; Philips; Andover, MA) was inserted to allow a bicaval view with the patient in the supine position. The RA as well as 3 to 4 cm of the lower SVC are visualized by this view. The echocardiographic correlate of the SVC-right atrial junction is defined as the base of the superior edge of the crista terminalis.12 A CVC tip in the upper RA was accepted, and its position was not corrected. Two anesthetists, well experienced in perioperative TEE, independently analyzed the films offline. The relationship of the CVC tip to the crista terminalis was recorded (Fig 1). Postoperatively, all patients were admitted to the ICU, and a bedside anteroposterior CXR with the patient lying supine was obtained within 3 h of admission. All radiographs were digitally saved via a patient archiving and communication system (Image Devices GmbH; Idstein, Germany). Finally, one senior radiologist in charge of our 52-bed ICU for > 10 years and two radiologists in training (fifth year) independently reviewed the radiographs. The radiologists were unaware that the CVC placement had been ECG guided, and they were blinded to the results of the TEE assessment. First, they assessed the quality of the radiographs (rotation, contrast); second, they had to decide whether the CVC tip was situated intraatrial or extraatrial. For this decision, our study design did not limit the radiologists to certain radiographic landmarks. Our radiologists were not aware of the relevance of TC-distance as a marker for intraatrial placement. First, they had to decide on the position, and then they documented the radiologic markers used for their decision. At last, they had to identify the carina and the right tracheobronchial angle, respectively. In a final step, they drew horizontal lines through the carina as well as the tip of the CVC and measured the TC-distance between these parallels (Fig 2–4).