The Science Journal of the American Association for Respiratory Care

1995 OPEN FORUM Abstracts

Virtual Abg's Derived From Noninvasive Etco_2 And Spo_2 Data

Whitney L. Schwartz, BA, RRT, Amy Orons, BA, the MRICU Respiratory Care Staff, Herbert Patrick, MD. Department of Respiratory Care, Thomas Jefferson University Hospital, Philadelphia, PA.

Introduction: Having previously reported a method for predicting PaCO_2 from ETCO_2 and Vd/Vt (Resp Care, 1994), we sought to both enhance the accuracy of the predicted PaCO_2 and include predicted SaO_2 using SpO_2, thereby establishing "virtual" ABG's.

Methods: We used data from 27 patients in our MRICU on A-C only in order to derive the line relating PaCO_2 - ETCO_2 vs. Vd/Vt. Although PECO_2 for Vd/Vt was previously determined using the SensorMedics DeltaTrac, we now use a simple eight-liter exhaled gas collection box which was validated against the DeltaTrac (n=6; difference, mean ± SD; r: 0.0007 ± 0.0465; 0.93).


SpO_2 measurements were made using the bedside Hewlett Packard (HP) SaO_2/Pleth M1020A Model 66 with either a HP nondisposable or Nellcor disposable finger probe. SpO_2 was validated against SaO_2 calculated from ABG's (n=12; 0.0866 ± 3.17; 0.67). Using this system, virtual ABG's for patients on A-C were available throughout the day following a single conventional ABG and Vd/Vt measurement by box in the morning.

Results: Data from 13 MRICU patients on A-C were used for virtual PaCO_2 and checked against actual PaCO_2 (n=13; 0.731 ± 8.47; 0.60). Three patients had virtual - actual PaCO_2 exceeding ± 5mmHg. Virtual SaO_2's were checked against actual SaO_2 (n=15; -1.117 ± 2.88; 0.67). All patients had virtual - actual SaO_2 within ± 3%.

Conclusions: 1) Our modified method for virtual PaCO_2 has improved accuracy and precision by using a derived (PaCO_2 - ETCO_2) vs. Vd/Vt line specific for a patient population, i.e., A-C mode. Nevertheless, the inaccurate virtual PaCO_2 in 3 of 13 patients may represent cardiopulmonary diseases altering PETCO_2 and PECO_2, identifying patients who may not be eligible for virtual ABG's. 2) The favorable accuracy and precision of SpO_2 as virtual SaO_2 is not surprising in our patient population. Accurate virtual ABG's should prove to be cost effective and greatly decrease utilization of resources in the ICU.


You are here: » Past OPEN FORUM Abstracts » 1995 Abstracts » Virtual Abg's Derived From Noninvasive Etco_2 And Spo_2 Data