[Prognostic value of difference between peripheral venous and arterial partial pressure of carbon dioxide in patients with septic shock: a pilot study].

OBJECTIVE: To evaluate the prognostic value of the difference between peripheral venous and arterial partial pressure of carbon dioxide in patients with septic shock following early resuscitation.
METHODS: This prospective study was conducted among the patients with septic shock treated in our department during the period from May, 2017 to May, 2018. Peripheral venous, peripheral arterial and central venous blood samples were collected simultaneously and analyzed immediately at bedside after 6-h bundle treatment. Arterial blood lactate concentration (Lac) and the arterial (PaCO2), peripheral venous (PpvCO2) and central venous partial pressure of carbon dioxide (PcvCO2) were recorded. The differences between PpvCO2 and PaCO2 (Ppv-aCO2) and between PcvCO2 and PaCO2 (Pcv-aCO2) were calculated. Pearson correlation analysis was used to test the agreement between Pcv-aCO2 and Ppv-aCO2. Multivariable logistic regression analysis was performed to analyze the possible risk factors for 28-day mortality, and the receiver-operating characteristic curve (ROC) was plotted to assess the prognostic values of these factors for 28-day mortality.
RESULTS: A total of 62 patients were enrolled in this study, among who 35 survived and 27 died during the 28-day period. Compared with the survivor group, the patients died within 28 days showed significantly higher Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score (24.2±6.0 vs 20.5±4.9, P=0.011), sequential organ failure assessment (SOFA) score (14.9±4.7 vs 12.2±4.5, P=0.027), PcvaCO2 (5.5±1.6 vs 7.1±1.7, P < 0.001), PpvaCO2 (7.1±1.8 vs 10.0±2.7, P < 0.001), and arterial lactate level (3.3±1.2 vs 4.2±1.3, P=0.003) after 6-h bundle treatment. Pearson correlation analysis showed that Ppv-aCO2 was significantly correlated with Pcv-aCO2 (r=0.897, R2= 0.805, P < 0.001). Multiple logistic regression analysis identified Ppv-aCO2 (β=0.625, P=0.001, OR=1.869, 95% CI: 1.311-2.664) and lactate level (β=0.584, P=0.041, OR=1.794, 95%CI: 1.024-3.415) as the independent risk factors for 28-day mortality. The maximum area under the ROC (AUC) of Ppv-aCO2 was 0.814 (95%CI: 0.696- 0.931, P < 0.001), and at the best cut- off value of 9.05 mmHg, Ppv-aCO2 had a sensitivity of 70.4% and a specificity of 88.6% for predicting 28-day mortality. The AUC of lactate level was 0.732 (95%CI: 0.607-0.858, P=0.002), and its sensitivity for predicting 28-day mortality was 70.4% and the specificity was 74.3% at the best cut-off value of 3.45 mmol/L; The AUC of Pcv-aCO2 was 0.766 (95%CI: 0.642-0.891, P < 0.001), and its sensitivity was 66.7% and the specificity was 80.0% at the best cut-off value of 7.05 mmHg.
CONCLUSIONS: A high Ppv-aCO2 after early resuscitation of septic shock is associated with poor outcomes. Ppv-aCO2 is well correlated with Pcv-aCO2 and can be used as an independent indicator for predicting 28-day mortality in patients with septic shock.