Partial pressure of carbon dioxide/pH interaction and its association with mortality among patients mechanically ventilated in the emergency department.

ABSTRACT

OBJECTIVES: There is currently conflicting data as to the effects of hypercapnia on clinical outcomes among mechanically ventilated patients in the emergency department (ED). These conflicting results may be explained by the degree of acidosis. We sought to test the hypothesis that hypercapnia is associated with increased in-hospital mortality and decreased ventilator-free days at lower pH, but associated with decreased in-hospital mortality and increased ventilator-free days at higher pH, among patients requiring mechanical ventilation in the emergency department (ED). METHODS: Secondary analysis of patient level data from prior clinical trials and cohort studies that enrolled adult patients who required mechanical ventilation in the ED. Patients who had a documented blood gas while on mechanical ventilation in the ED were included in these analyses. The primary outcome was in-hospital mortality, and secondary outcome was ventilator-free days. Mixed-effects logistic, linear, and survival-time regression models were used to test if pH modified the association between partial pressure of carbon dioxide (pCO2) and outcome measures. RESULTS: Of the 2348 subjects included, the median [interquartile range (IQR)] pCO2 was 43 (35-54) and pH was 7.31 (7.22-7.39). Overall, in-hospital mortality was 27%. We found pH modified the association between pCO2 and outcomes, with higher pCO2 associated with increased probability of in-hospital mortality when pH is below 7.00, and decreased probability of in-hospital mortality when pH is above 7.10. These results remained consistent across multiple sensitivity and subgroup analyses. A similar relationship was found with ventilator-free days. CONCLUSIONS: Higher pCO2 is associated with decreased mortality and greater ventilator-free days when pH is >7.10; however, it is associated with increased mortality and fewer ventilator-free days when the pH is below 7.00. Targeting pCO2 based on pH in the ED may be a potential intervention target for future clinical trials to improve clinical outcomes.