SESSION TITLE: Critical Care Posters
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 29, 2014 at 01:30 PM - 02:30 PM
PURPOSE: The objective of this study is to determine whether transcutaneous CO2 (TCO2) measurements correlate with arterial CO2 (PaCO2) in the setting of critically ill patients in shock.
METHODS: An institutional review board approved, prospective, observational study was performed in adult subjects in two intensive care units at our center. After obtaining informed consent, arterial blood gases were collected at various time points while TCO2 was continuously measured using a SENTEC™ monitor (Therwil, Switzerland) up to 8 hours daily, up to five different days for each subject. Presence of shock was determined at each time point using any of the following criteria: SBP<90 mmHg for >10 minutes, need for vasopressor therapy or lactate level>2 mmol/L. Linear regression was performed to correlate TCO2 to PaCO2 for both non-shock and shock states.
RESULTS: During a 3 month period, 6 subjects were enrolled, providing 15 study days and a total of 38 time point comparisons between TCO2 and PaCO2. Shock was present at 17 of 38 time points. In the overall group, linear regression showed good correlation between TCO2 and PaCO2 (r2=0.793). In both shock and non-shock patients, there was a significant correlation between TCO2 and PaCO2 (r2=0.925 & 0.672) respectively. Mean bias assessment by Bland Altman analysis showed that TCO2 slightly overestimated PaCO2 at values in the high range.
CONCLUSIONS: Our preliminary analysis revealed that TCO2 may reliably correlate with PaCO2 even in states of shock. Continued enrollment to increase power is planned.
CLINICAL IMPLICATIONS: A non-invasive, continuous correlate of arterial CO2 (PaCO2) could improve the management of the critically injured, as in the setting of traumatic brain injury (TBI) where sudden increases in PaCO2 could have a deleterious impact. End tidal CO2 (ETCO2) has been shown to be a poor correlate of PaCO2 in the setting of altered perfusion or shock, limiting its utility in this setting. This study demonstrated good correlation of PaCO2 and TCO2 in both normal and low perfusion states. Transcutaneous CO2 (TCO2) monitoring is a potential alternative to ETCO2, demonstrating persistent correlation with PaCO2 even in shock.
DISCLOSURE: The following authors have nothing to disclose: Leslie Wood, James Aden, Michael Morris, David Bell, John Hunninghake, Victor Convertino, Kevin Chung
No Product/Research Disclosure Information