2005 OPEN FORUM Abstracts
MICROSTREAM® END-TIDAL CARBON DIOXIDE (ETCO2) MONITORING AS AN EARLY INDICATOR OF RESPIRATORY COMPROMISE IN PATIENTS RECEIVING PROCEDURAL SEDATION.
Sally Whitten, RRT, Roy Cobean, MD, Roger Collard, RN, John Dziodzio, B.A., Christopher Hirsch, MPH, RRT Maine Medical Center, Portland, Maine
Introduction: At our institution pulse oximetry is recognized as a standard of care for monitoring respiratory status during procedures requiring sedation and analgesia. Concerns over the limitation of pulse oximetry in early identification of respiratory compromise have been raised. There is increasing evidence that a more reliable method for identification of respiratory compromise during these procedures may be capnography, using Microstream® technology. This study was part of a patient safety initiative.
Methods: End-Tidal Carbon Dioxide (ETCO2) measurements were obtained from 17 patients requiring procedural sedation while undergoing elective upper gastrointestinal endoscopy. Carbon dioxide (CO2) measurements were obtained through the use of the Smart BiteBlocTM O2 system (Oridion Capnography, Needham, MA) which allows for delivery of oxygen while simultaneously measuring CO2 levels. Capnography measurements were interfaced with monitoring systems (Philips Medical, Andover, MA.) which allowed continuous CO2 waveforms to be recorded. Patient vital signs were recorded prior to the administration of sedatives and analgesia, and at a minimum of 30 second intervals during and after the procedure by the Respiratory Care Practitioner (RCP). The RCP also independently viewed a capnograph monitor and recorded periods of apnea, respiratory pattern depression, and pulse oximetry values. All other caregivers were blinded from the capnograph monitor. An RN provided routine observation utilizing our standards of pulse oximetry, automated blood pressure cuff, heart rate, and clinical assessment.
Results: Strip-chart waveforms and the RCP's handwritten procedure logs were analyzed to identify events of respiratory compromise, as indicated by low oxygen saturation (SPO2 < 90%), or apnea (lack of ETCO2 waveform for 15 seconds or longer). Thirty-three periods of apnea, ranging from 15 seconds to 70 seconds in duration, were observed by the RCP. Pulse oximetry demonstrated 13 episodes of desaturation, 5 of which were correlated with apneic periods measured by capnography. Eight desaturation events were either uncorrelated, or correlated with hypoventilatory episodes under the 15 second criteria we imposed. Of the 13 desaturation events, only three resulted in intervention by the RN. An additional two events were flagged by the RCP in attendance as clinically significant, who provided out-of-protocol but clinically appropriate intervention. Of the total of 41 events of respiratory compromise, 33 (80.5%) were identified by capnography, while 13 (31.7%) were identified by oximetry. By difference of proportions, the difference in sensitivity between techniques was calculated at a statistically significant level (P< 0.0001).
Conclusions: ETCO2 was a more sensitive indicator of respiratory compromise than pulse oximetry. Our findings were based on observing both numeric value and waveforms. Further analysis of alarm settings without the benefit of waveform observation is needed to determine clinical applicability.