Diagnosis of Abnormal Diaphragm Motion after Cardiothoracic Surgery: Ultrasound Performed by a Cardiac Intensivist vs. Fluoroscopy


Joan Sanchez de Toledo, MD, Pediatric Critical Care Medicine, Children's Hospital of Pittsburgh, 4401 Penn Avenue, Pittsburgh, PA 15224, USA. Tel: (+1) 412-692-7366; Fax: (+1) 412-692-5169; E-mail: sanchezdetoledoj@upmc.edu; Division of Pediatric Cardiac Intensive Care, Hospital Universitari Vall d'Hebron, Passeig Vall d'Hebron 119-129, 08035 Barcelona, Spain, joansdts@yahoo.es


Objectives.  Abnormal diaphragmatic motion secondary to phrenic nerve injury is not uncommon after pediatric cardiothoracic surgery. Fluoroscopy is the most frequent method of diagnosis but it carries risks associated with transportation of critically ill children and exposure to ionizing radiation. Ultrasonography, a reliable diagnostic method in adults, eliminates both concerns. Since most cardiac intensivists are trained in echocardiography, we tested the hypothesis that chest ultrasound performed by a cardiac intensivist is faster than fluoroscopy, and is highly accurate in predicting fluoroscopy results, therefore serving as an equally useful diagnostic test.

Design.  Prospective study in consecutive pediatric patients with suspected abnormal diaphragmatic motion after cardiothoracic surgery. All patients underwent fluoroscopy and ultrasound study of the diaphragm. Ultrasound was performed by a pediatric cardiac intensivist and a trainee. Kappa statistic was calculated to assess concordance between both ultrasound readings. Sensitivity, specificity, and positive and negative predictive values (PPV and NPV) were calculated to assess accuracy of each ultrasound test in predicting fluoroscopy results.

Results.  Twenty-five patients with median age 3 months (12 days–11 years) and median weight of 3.8 kg (2.5–29 kg) were included. The ultrasound diagnosis of the cardiac intensivist was perfectly accurate (100% sensitivity, specificity, and PPV and NPV) in predicting fluoroscopy results. The ultrasound performed by the trainee achieved 85.7% sensitivity, 94.4% NPV, and 100% specificity relative to fluoroscopy. The interoperator reliability of chest ultrasound was 0.89 (95% confidence interval 0.69–1). Delay between clinical suspicion and the diagnostic tests was 15 minutes (5 minutes–2.5 hours) for ultrasound and 17 hours (60 minutes–82 hours) for fluoroscopy (P < 0.001).

Conclusions.  Chest ultrasound performed by cardiac intensivists allows for an early and accurate diagnosis of abnormal diaphragmatic motion, as evidenced by their ability to predict fluoroscopy findings in pediatric cardiothoracic patients. Training in ultrasound-guided assessment of diaphragmatic motion should be reinforced during pediatric cardiac intensive care fellowship.