Tissue perfusion may be assessed at the bedside with ETCO₂-derived dead-space estimation

Introduction

Alveolar dead space is determined by the degree of ventilation/perfusion mismatch in the lung. Since the entire cardiac output must pass through the lungs, we hypothesized that global changes in perfusion would correlate with changes in alveolar dead space. Dead space (VD/VT) can be estimated from the modified Bohr equation, which is further modified by substituting end-tidal CO₂ for PECO₂ – hence the VDVTest (estimated dead space). We used blood lactate as the endpoint of resuscitation.

Methods

Five mechanically ventilated patients with lactic acidosis (lactate >30 mg/dl) were included. Four patients were postlaparotomy, one had multiple trauma. Ages were 36–87 years, APACHE II scores 18–37, and P:F ratio 203–470. We recorded the HR, MAP, CVP and ETCO₂ from the ICU monitor. We took simultaneous arterial and central venous blood samples at 30-min intervals for the first 4 hours of the study, then every 1 hour for a total of 12 hours.

Results

There was a strongly positive linear trend correlating the VDVTest with lactate levels [r(76) = 0.800; P < 0.001] (Fig. 1). Furthermore, when plotted against time, the trends in VDVTest and lactate are parallel (example data from study patient 1 are shown; Fig. 2). There was also a significant correlation between the arterial to end-tidal CO₂ gradient (PaCO₂–PETCO₂) and lactate levels [r(76) = 0.730, P < 0.001]. There was no meaningful correlation between blood lactate and MAP, CVP, ScvO₂ and O₂ER.

Figure 1

Figure 2

Conclusion

In this pilot study we demonstrated linear correlation between estimated dead-space and lactic acid levels. We propose that this technique may be a useful end-point of resuscitation during shock. It is convenient to measure, requires no special equipment and is non-invasive. Further study is required to assess the technique's reliability across different patient populations.