Comparison of oesophageal and gastric air tonometry in patients with circulatory failure

Abstract

Background: Gastric PCO2 measured by balloon tonometry can estimate the adequacy of splanchnic perfusion. However, enteral feeding and gastric content can interfere with gastric PCO2 assessment. Tonometry in other sites of the body could avoid these problems. We therefore tested the hypothesis that oesophageal air tonometry would give results similar to gastric tonometry.

Methods: We studied 20 consecutive patients (mean age 68 (SD 9) [range 49-81] yr, 18 males, SAPS II score 55 (SD 18), ICU mortality 55%) with circulatory disorders during mechanical ventilation in the intensive care unit. Tonometer probes were placed via the nose, one into the stomach and the other in the oesophagus. PCO2 was measured with two automated gas analysers, at admission and 30 min, 1, 2, 3, 32, 40, and 48 h thereafter.

Results: One hundred and forty-eight paired measurements were obtained. Gastric PCO2 was greater than oesophageal PCO2 on admission (7.19 (1.43) vs 5.89 (0.73) kPa, P < 0.01) and subsequently. Differences between the measures correlated (r = 0.67) with the mean absolute value, indicating that overestimation increased as gastric PCO2 increased.

Conclusions: Oesophageal PCO2 is less than gastric PCO2, and the difference is greater when gastric PCO2 levels are greater. Air tonometry may not measure regional PCO2 levels in the oesophagus satisfactorily. Other methods and sites for carbon dioxide tonometry should be examined.