Objective: To evaluate the relationship between PaCO2 and end-tidal CO2 tension (PetCO2) before weaning and during a weaning trial and to determine the ability of PetCO2 to identify clinically relevant episodes of hypercapnia.
Design: Open, prospective study.
Setting: General intensive care unit.
Patients: 30 critically ill patients (mean age 63 +/- 2 years; Acute Physiology And Chronic Health Evaluation (APACHE) II of 18.4 +/- 3) who underwent a weaning trial during the recovery phase of acute respiratory failure requiring mechanical ventilation (MV) (8.9 +/- 1 days on MV).
Interventions: Weaning trial consisted of 2 h breathing on 5 cmH2O of Continuous Positive Airway Pressure (CPAP).
Measurements and results: Arterial blood gas values, PetCO2 register and pulse oximetry determinations were recorded in assist/control ventilation before CPAP, after 1 h on CPAP and after 2 h on CPAP (immediately before extubation) or immediately before returning to assist/control mode in patients who failed the weaning trial. Clinically relevant hypercapnic episodes were described as: (1) an increment in PaCO2 > 42 mm Hg in previously normocapnic patients and (2) an increment of > 8 mm Hg from previous PaCO2 in previously hypercapnic patients. Changes in PaCO2 and changes in PetCO2 between MV and the first and second hour of CPAP showed a significant correlation (r = 0.74; p < 0.01). Clinically relevant hypercapnic episodes were detected by increments of > 3 mm Hg in PetCO2 with a sensitivity of 82%, a specificity of 76% and a negative predictive value of 94%. The area under the receiver operating characteristic curve for increments in PetCO2 was 0.90.
Conclusions: Capnography provided good assessment of hypercapnic episodes during weaning, although the high number of false positives may result in arterial blood sampling in patients who do not present with ventilation failure.