20 Jan 10
Predicting dead space ventilation in critically ill patients using clinically available data
By D Frankenfield, S Alam, E Bekteshi, R Vender
Crit Care Med 2010;38:288-291
To develop and validate an equation to predict dead space to tidal volume ratio (Vd/Vt) from clinically available data in critically ill mechanically ventilated patients.
Design
Prospective, observational study using a convenience sample of patients whose arterial blood gas and respiratory gas exchange had been measured with indirect calorimetry.
Setting
Medical and surgical critical care units of a university medical center.
Patients
Adult, mechanically ventilated patients at rest with Fio2 ≤0.60 and no air leaks who had recent arterial blood gas recordings and end-tidal carbon dioxide concentration monitoring.
Intervention
Observational only.
Measurements and main results
Indirect calorimetry was used to determine carbon dioxide production and expired minute ventilation in 135 patients. Tidal volume and respiratory rate were recorded from the ventilator. End tidal carbon dioxide concentration, body temperature, arterial carbon dioxide partial pressure (Paco2), and other clinical data were recorded. Vd/Vt was calculated using the Enghoff modification of the Bohr equation (Paco2 − PECO2/Paco2). Regression analysis was then used to construct a predictive equation for Vd/Vt using the clinical data: Vd/Vt = 0.32 + 0.0106 (Paco2 − ETCO2) + 0.003 (RR) + 0.0015 (age) (R2 = 0.67). A second group of 50 patients was measured using the same protocol and their data were used to validate the equations developed from the original 135 patients. The equation was found to be unbiased and precise.
Conclusions
Vd/Vt is predictable from clinically available data. Whether this predicted quantity is valuable clinically must still be determined.