Introduction: Hypoxemia is difficult to describe quantitatively over âme and across modes of ventilation due to varying fraction of inspired oxygen (FiO2) and interventions causing change in mean airway pressure (Paw). Our objective is to introduce and preliminarily validate an objective, intuitive measure of hypoxemia in adult mechanically ventilated patients. Methods: This is a concurrent observational study of all patients admitted to the SICU and treated with mechanical ventilation between 6/1 97 and 8.1 97. Daily record was made of all measured PaO2, FiO2, and Paw. The PaO2/FiO2 ratio was divided by the Paw to yield the Oxygénation Dysfunction Index (ODI). Endpoints included extubation, death, and change of mode of ventilation. Ventilatory modes used in our SICU include CPAP, PSV, SIMV, CMV, PCV, PCIRV, and UHFJV (listed here in ascending order of our perceived degree of "invasivencss"). Change of the ODI over time was quantified as the slope of the linear regression of wpKrmal ODIs prior to reaching one of the above defined endpomts. Slopes were compared (t-test, significance set at .05) for sequential ODIs approaching improvement (extubation or less invasive mode) vs. worsening (death or more invasive mode). Frequency of positive and negative slopes was analyzed with Chi-squared (significance at .OS). Results: 23 patients (mean APACHE n 13.3±6.6) reached 82 endpomts. Slopes of sequential ODIs prior to improved endpoints were significantly greater than slopes prior to worsened endpoints for 1 and 2 days prior to the endpoints(Tible 1), and were more likely to positive than negative (1 day p"l.6x10-5,2 day p 005). Iabk_L 1 day Tdays Mean slope prior to "improved" endpoints 5.3 n=34 4.9 n=24 Mean slope prior to "worsened" endpoints -17.3 n=15 1.0 v?Q P value .03 .04 Conclusions: The ODI provides a quantifiable, intuitive measure of hypoxemia indexed to FiO2 and Paw. "Improvement" is characterized by an increase in ODI, and "worsening" is characterized by a decrease in ODI over 24 and (less markedly) 48 hours prior to an endpoint event This may be a useful tool to quantify "improvement" or "worsening" in research protocols where patients are converted from one mode of ventilation to another.
|Original language||English (US)|
|Journal||Critical care medicine|
|Issue number||1 SUPPL.|
|State||Published - Dec 1 1998|
ASJC Scopus subject areas
- Critical Care and Intensive Care Medicine