435

Introduction: Physiologic dead space (comprised of alveolar and airway dead space) is associated with mortality and injurious mechanical ventilation practices in patients with respiratory failure. Time based capnography (time versus PCO2) is routinely used to monitor mechanically ventilated patients. The end tidal alveolar dead space fraction (AVDSF), often used at the bedside to estimate alveolar dead space, is calculated using the end tidal carbon dioxide level from time based capnography (PaCO2-PETCO2)/PaCO2. However volumetric capnography (volume versus PCO2) measurements of physiologic dead space ((VD/VT) = (PaCO2 – PeCO2 (mean expired CO2))/PaCO2) and alveolar dead space (Vdalv/Vtalv measured through analysis of the volumetric capnography curve) are considered more accurate. The correlation of AVDSF with volumetric capnography VD/VT and Vdalv/Vtalv has not been previously described. Hypothesis: AVDSF will correlate with both physiologic dead space and alveolar dead space measured by volumetric capnography. Methods: Prospective study of mechanically ventilated children < 21 y/o admitted to a single PICU. Volumetric capnography dead space measurements and the AVDSF were calculated with each arterial blood gas (ABG). Pearson within subjects correlation coefficients were analyzed, accounting for repeated measurements. Results: 149 ABGs were obtained from 14 children (median age 18 mths, range 11- 62 mths) with a mean PaO2/FiO2 ratio of 244 +/- 21.3. Mean VD/VT was 0.54 +/- 0.11, mean AVDSF was 0.17 +/- 0.14, and mean Vdalv/Vtalv was 0.25 +/- 0.14. The correlation between AVDSF and Vdalv/Vtalv was strong with r=0.79 (p<0.001). The correlation between VD/VT and AVDSF was less r=0.55 (p<0.001) however similar to the correlation between VD/VT and Vdalv/Vtalv (r=0.54, p<0.001). Conclusions: The AVDSF correlated well with Vdalv/Vtalv. The correlation between VD/VT and either AVDSF or Vdalv/Vtalv is weaker possibly related to the influence of patient specific effects on airway dead space.