Hand-carried ultrasound performed by hospitalists: does it improve the cardiac physical examination?

OBJECTIVE: The traditional physical examination of the heart is relatively inaccurate. There is little information regarding whether cardiac hand-carried ultrasound performed by noncardiologists adds to the accuracy of physical examinations. The purpose of this study was to determine whether hand-carried ultrasound can add to the accuracy of hospitalists' cardiac physical examinations. METHODS: During a focused training program in hand-carried echocardiography, 10 hospitalists performed cardiac examinations of 354 general medical inpatients first by physical examination and then by hand-carried ultrasound. Eligible inpatients included those for whom a conventional hospital echocardiogram was ordered. We measured how frequently the hospitalists' cardiac examination with or without hand-carried ultrasound matched or came within 1 scale level of an expert cardiologist's interpretation of the hospital echocardiogram. RESULTS: Adding hand-carried ultrasound to the physical examination improved hospitalists' assessment of left ventricular function, cardiomegaly, and pericardial effusion. For left ventricular function, using hand-carried ultrasound increased the percentage of exact matches with the expert cardiologist's assessment from 46% to 59% (P=.005) and improved the percentage of within 1-level matches from 67% to 88% (P=.0001). The addition of hand-carried ultrasound failed to improve the assessments of aortic stenosis, aortic regurgitation, and mitral regurgitation. CONCLUSION: Adding hand-carried ultrasound to physical examination increases the accuracy of hospitalists' assessment of left ventricular dysfunction, cardiomegaly, and pericardial effusion, and fails to improve assessment of valvular heart disease. The clinical benefit achieved by improved immediacy of this information has not been determined. An important limitation is that the study assessed only 1 level of training in hand-carried ultrasound.

The rate at which residents learn to use hand-held echocardiography at the bedside.

PURPOSE: Because there is little information about the training that general internists require to perform hand-carried cardiac ultrasonography (HCU), we studied the rate of learning of a group of medical residents performing HCU after minimal formal training. METHODS: Medical residents on the inpatient services at Johns Hopkins Bayview Medical Center received formal training in HCU consisting of 15-30 minutes of didactic instruction about the principles of echocardiography, followed by ongoing one-on-one instruction in performing HCU and subsequent ongoing one-on-one training from a certified echocardiography technician as they were doing scans. The residents were shown how to position the patient to obtain 2-dimensional echo images from the parasternal short and long axes and apical 4-chamber views, and how to obtain color-flow Doppler images across the mitral and aortic valves. Residents were asked to determine whether pericardial effusion was present and to assess left ventricular size, left ventricular function, and the mitral and aortic valves. The residents performed cardiac physical examination and HCU independently on patients who had a conventional transthoracic echocardiogram (CTTE) performed within 24 hours of the HCU. The residents' HCU results were compared with the CTTE results by a cardiologist specializing in echocardiography. The rates at which residents gained technical proficiency and skills in interpreting their studies were measured by linear regression to fit various outcome variables against their experience at scanning as gauged by the number of scans performed. RESULTS: Thirty medical residents performed a total of 231 HCU studies. Linear regression models showed that the residents' overall technical proficiency skills improved at the rate of 0.79 (95% confidence interval [CI] 0.53-1.04) points on an overall assessment index (0-3 scale) per 10 scans completed. Interpretation accuracy improved at a rate of 1.01 (95% CI 0.69-1.39) points per 10 scans as measured by an interpretation accuracy index (0-3 scale). Because scanning efforts and instruction in HCU occurred during residents' usual rotation duties, some residents gathered experience in HCU slowly and sporadically. CONCLUSION: This study, the first prospective, experimental effort of its kind, shows that residents as a group learned important aspects of HCU scanning and interpretation at a reasonably rapid rate.