Original Article
The application of combined central venous pressure and oxygen metabolism parameters monitoring in diagnosing septic shock–induced left ventricular dysfunction
Keliang Cui, Xiaoting Wang, Hongmin Zhang, Wenzhao Chai, Dawei Liu
Published 2015-10-01
Cite as Chin J Intern Med, 2015, 54(10): 855-859. DOI: 10.3760/cma.j.issn.0578-1426.2015.10.008
Abstract
ObjectiveTo evaluate the value of central venous pressure (CVP), central venous oxygen saturation (ScvO2) and venous–arterial carbon dioxide partial pressure gradient (Pv–aCO2) in the diagnosis of septic shock–induced left ventricular dysfunction.
MethodsConsecutive patients with septic shock were enrolled from September 2013 to September 2014 in ICU at Peking Union Medical College Hospital. The data of CVP, Pv–aCO2 and ScvO2 were recorded and analyzed. According to the left ventricular ejection fraction (LVEF) tested by bedside echocardiography, the patients were divided into two groups: new onset of left ventricular dysfunction (LVEF<50%) group and non–left ventricular dysfunction (LVEF ≥ 50%) group. A diagnostic model was created by logistic regression. The diagnostic performance and cut–off values of CVP, Pv–aCO2, ScvO2 were determined using receiver operating characteristic (ROC) curve analysis.
ResultsAmong 93 patients enrolled, 39 were diagnosed with left ventricular dysfunction. In the new onset group, CVP [(12.5±3.9) mmHg(1 mmHg=0.133 kPa) vs (10.4±2.5)mmHg; P=0.005] and Pv–aCO2 [(7.5±3.9) mmHg vs (4.5±2.6)mmHg; P<0.001] were significantly higher than those in the non–left ventricular dysfunction group, while ScvO2 [(62.4±10.5)% vs (72.6±9.0)%; P<0.001] was significantly lower. As far as the diagnostic value of these three parameters were concerned for left ventricular dysfunction, the sensitivity of CVP≥12.5 mmHg was 46.2%, specificity 81.5% with an area under ROC curve(AUCROC) 0.674; the sensitivity of Pv–aCO2≥5.0 mmHg 76.9%, specificity 37.0%, AUCROC 0.738; the sensitivity of ScvO2≤65.8% 64.1%, specificity 78.6%, AUCROC 0.775. When the cut–off values were determined by ROC, the diagnostic performance of the model was ≥0.377 with the sensitivity, specificity and AUCROC 82.1%, 79.6% and 0.835, respectively.
ConclusionIn patients with septic shock, the logistic regression model established by CVP, Pv–aCO2 and ScvO2 contributes to the diagnosis of septic shock–induced left ventricular dysfunction.
Key words:
Shock, septic; Left ventricular dysfunction; Central venous pressure; Logistic models
Contributor Information
Keliang Cui
Department of Critical Care Medicine, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100730, China
Xiaoting Wang
Hongmin Zhang
Wenzhao Chai
Dawei Liu
