The “CVP Sono” Can Safely Eliminate the Need for Post-Insertion Radiographs for Central Venous and Peripherally Inserted Catheters (CVC/PICC)

Abstract

Introduction: Real-time ultrasound guidance of CVC/PICC insertion improves safety and efficacy. We hypothesized that a more robust ultrasound surveillance technique incorporating echocardiographic and vascular views– the “CVP sono”– would avoid the need for chest radiography to realize cost and efficiency gains. Methods: We conducted a 5-month prospective data collection in a high-volume, urban, academic SICU of consecutive RN-placed PICCs and resident-placed CVCs. A single surgical intensivist, blinded to the results of chest radiography, performed all CVP sono's post-insertion. Catheter malposition was defined as location extrinsic to the superior vena cava and determined by a board-certified radiologist on portable chest radiography. “CVP sono” consisted of: 1) Mechanical Complications Screen (hemo-, pneumothorax): low frequency transducer view of applicable anterior and lateral hemithorax in B(rightness) and M(otion) modes 2) Superior Vena cava tip position screen: low frequency echo of the and heart/cava in parasternal short axis, four chamber and subcostal views 3) Position anomaly screen: high frequency view of bilateral internal jugular/innominate veins Statistical analysis was performed using Chi-square or Fisher's exact tests for categorical variables and Student's t-test for continuous variables. Results: “CVP sono” evaluated 83 catheters (42CVC:13 internal jugular, 29 subclavian and 41PICC) and considered technically adequate in 59 (71%). Incomplete studies were significantly more common in those with chest tubes (p=0.02), but not in those with cervical collars (p=0.07), an open abdomen (p=0.28) and BMI>40 (p=0.33). Mean “CVP sono” time was 10.8 min, compared to CXR (75.3 min; p <0.001). No hemo- pneumothoraces developed. Presence of multiple indwelling central catheters (>1CVCs) trended for inaccurate “CVP sono” for catheter malposition (accuracy: 79% vs 93%, p=0.11).