Randomized, controlled clinical trial of point-of-care limited ultrasonography assistance of central venous cannulation: The Third Sonography Outcomes Assessment Program (SOAP-3) Trial*

Abstract

CONTEXT: A 2001 Agency for Healthcare Research and Quality Evidence Report on patient safety addressed point-of-care limited ultrasonography guidance for central venous cannulation and strongly recommended real-time, dynamic guidance for all central cannulas. However, on the basis of one limited study, the report dismissed static assistance, a "quick look" with ultrasound to confirm vein location before preparing the sterile field, as unhelpful. OBJECTIVE: The objective of this trial was to compare the overall success rate of central cannula placement with use of dynamic ultrasound (D), static ultrasound (S), and anatomical landmarks (LM). DESIGN AND SETTING: A concealed, randomized, controlled, clinical trial conducted from September 2003 to February 2004 in a U.S. urban teaching hospital. PATIENTS: Two-hundred one patients undergoing internal jugular vein central venous cannulation. INTERVENTIONS: Patients were randomly assigned to three groups: 60 to D, 72 to S, and 69 to LM. An iLook25 SonoSite was used for all imaging. MEASUREMENTS AND MAIN RESULTS: Cannulation success, first-attempt success, and number of attempts were noted. Other measures were vein size and clarity of LM. Results, controlled for pretest difficulty assessment, are stated as odds improvement (95% confidence interval) over LM for D and S. D had an odds 53.5 (6.6-440) times higher for success than LM. S had an odds 3 (1.3-7) times higher for success than LM. The unadjusted success rates were 98%, 82%, and 64% for D, S, and LM. For first-attempt success, D had an odds 5.8 (2.7-13) times higher for first success than LM, and S had an odds 3.4 (1.6-7.2) times higher for first success than LM. The unadjusted first-attempt success rates were 62%, 50%, and 23% for D, S, and LM. CONCLUSIONS: Ultrasound assistance was superior to LM techniques. D outperformed S but may require more training and personnel. All central cannula placement should be conducted with ultrasound assistance. The 2001 Agency for Healthcare Research and Quality Evidence Report dismissing static assistance was incorrect.