120 Cardiopulmonary Ultrasound in Sepsis: A Pilot Study

Abstract

Sepsis-induced cardiomyopathy (SICM) is a recognized and often reversible form of cardiac dysfunction in patients with sepsis. Few studies have assessed the utility of point-of-care ultrasound in the management of patients with sepsis, especially early in the disease course. The purpose of this study is to evaluate echocardiogram and lung ultrasound findings and correlation with clinical outcomes. This was a prospective, pilot study enrolling adult patients (>18 years) with sepsis at a single academic emergency department (ED). Patients were enrolled within one hour of sepsis onset, defined as the time at which our sepsis screening tool in the electronic medical record alerted the provider of potential and/or severe infection. Echocardiogram and lung ultrasounds were performed and reviewed by an ultrasound-fellowship trained member of the research team. Ultrasound findings included left ventricular (LV) function, right ventricular (RV) function, RV size, and presence of anterior thoracic B lines. Demographics, clinical outcomes, sepsis measures, and ultrasound findings were analyzed using descriptive statistics. Multivariate logistic regression was performed with variables chosen a priori and included: hyperdynamic LV function, reduced right ventricular (RV) function or abnormal RV size, inferior vena cava (IVC) collapsibility with respiration, > 5 B-lines on lung ultrasound (LUS), >1 pleural effusion on LUS, and Mortality in ED sepsis score. Primary outcomes were need for positive pressure ventilation (PPV) or vasopressors at 12 hours. In total, 92 patients were enrolled, including 43 patients with hyperdynamic LV function and 49 with normal/reduced LV function. 10 patients required vasopressor and 11 required PPV at 12 hours from the time of ED arrival. Demographics and sepsis management variables were well matched between groups, with the exception of supplemental oxygen use being more prevalent in the normal/reduced group (28 v. 11, p = 0.002). Hyperdynamic LV function did not correlate with vasopressor or PPV use at 12 hours (OR 2.51—95% CI 0.32-19.88 and OR 4.63—95% CI 0.37-57.24, respectively). Abnormal RV function or size, IVC collapsibility, and lung US findings did not demonstrate statistical significance (Table 1). In this small cohort of patients, hyperdynamic LV function did not correlate with need for vasopressors or PPV at 12 hours. However, abnormal RV size and >5 B-lines did trend towards significance for both primary outcomes. Future studies with a larger sample size are needed to better elucidate these findings.Table 1Multivariate Regression Analysis for primary outcomes.VariablePPV Use at 12HRVasopressor Use at 12HRORp95% CIORp95% CIHyperdynamic LV4.630.2330.37-57.242.510.3820.32-19.88RV Function- abnormalCannot Calculate0.0880.3430.00-13.38RV Size- abnormal1.380.8580.04-47.067.770.0810.78-77.70IVC Collapsible0.040.0600.00-1.150.140.1390.01-1.90>5 B-lines on LUS11.570.0530.96-138.720.200.2450.01-3.10>1 Pleural effusion on LUS0.060.1320.00-2.340.940.9570.11-8.41MEDS Score1.480.0051.13-1.951.400.0031.12-1.74 Open table in a new tab