Date of Award

January 2018

Document Type

Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Chris L. Moore

Abstract

Cardiopulmonary complaints are common in the emergency department (ED) with causes ranging from benign to life-threatening. Point-of-care ultrasound (POCUS) is a non-invasive test that can be quickly performed at the bedside to identify conditions requiring intervention. We sought to determine the impact of early POCUS on time to first appropriate treatment. We conducted a prospective randomized trial of a convenience sample of adult ED patients presenting with cardiopulmonary chief complaints during pre-defined enrollment periods. Eligible patients were randomized to early POCUS vs usual care. Early POCUS included focused cardiac, thoracic, and abdominal views with preliminary results presented to the treating team (findings ultimately confirmed by the clinical attending). Subjects randomized to usual care could have an ultrasound performed by the clinical team at their discretion. Medical records were reviewed by two separate physicians blinded to the randomization group to determine the final diagnosis and timing of a condition-specific treatment (if administered). The primary outcome was time to first appropriate treatment defined a priori. Secondary outcomes were time to final disposition and diagnostic accuracy and certainty. A total of 200 patients completed the study (101 in early POCUS and 99 in usual care) with the most common chief complaints of chest pain in 122 and dyspnea in 67. Mean age was 55 +/- 18 years and 39% were male. Appropriate condition-specific treatment was administered to 63 patients, with an average time of 2.7 hours (IQR 21.8) in the POCUS group vs 3.4 (IQR 12.4) in the usual care group, a difference of 0.7 hours (95% CI -1.7 to 2.7). Time to disposition in the POCUS was 2.5 hours (IQR 2.0) vs 2.2 (IQR 2.0) in the usual care group with a difference of 0.2 hours (95% CI -0.07 to 0.7), p=0.1. There was no significant difference in length of stay (p=0.07) or 30-day mortality (p=0.2). Primary providers had similar levels of diagnostic certainty on a Likert scale of 1-5 (4.1 ± 1.0 vs 3.9 ± 1.1) at disposition but had a higher rate of accurate final diagnosis in 79% vs 63% (difference 17%, 95% CI 4-29%) of patients for the POCUS and usual care groups, respectively. Of note, 1 pulmonary embolism was missed in the ED from the usual care group (with evidence of right heart strain on inpatient echocardiogram), and 1 pneumothorax would have been missed if the patient had not been randomized to the POCUS group. In conclusion, early POCUS shows promise in improving time to appropriate treatment although this has not reached statistical significance. It also significantly improves diagnostic accuracy. Larger numbers will need to be enrolled to determine if there is a significant impact on intervention, disposition, and outcome.

Comments

This thesis is restricted to Yale network users only. It will be made publicly available on 06/25/2100

Share

COinS