Gastric tonometry

We read with interest Drs Trinder and Lavery’s excellent review of the literature related to gastric tonometry (Anuestkcsiu 1996; 51:161-70) but would like to raise a number of points. Firstly the authors do not seem to have answered the question of whether or not the gastric tonometer is a valuable monitor of splanchnic perfusion. Our own conclusion is that based on currently available evidence the answer is no. As the authors quite rightly point out the only large outcome study of therapy guided by gastric intramucosal pH (pHi) was seriously flawed. In any case a survival benefit was only shown for patients with an admission APACHE I1 score of 15-25 and an admission pHi≥7.35 [I]. Drs Trinder and Lavery’s claim that gastric tonometry seems to be useful in highlighting inappropriate times to begin enteral feeding appears to be based on their own work demonstrating a temporal relationship between commencement of feeding and a decrease in pHi in eight of 60 patients [2]. No attempt was made to examine the relationship between this decrease and outcome and thus the clinical significance of the fall is unknown. In order for a monitor to be useful it is necessary to know the normal ranges of values. This has never been clearly established in humans and indeed values associated with poor outcome have varied in different studies from <7.35 to <7.1 [3,4]. Also the blood gas analyser used to measure balloon saline carbon dioxide tension has a marked effect on the values of pHi obtained [5]. The authors of this study concluded that it may be necessary to establish a normal range for each blood gas analyser. Secondly, we presume that the authors would wish to revise the statement in the section on regional supply and demand that ‘low intramucosal pH indicates inadequate mucosal oxygenation’ in the light of recent data demonstrating ileal intramucosal acidosis in the presence of supranormal ileal mucosal tissue oxygen tension in a porcine model of resuscitated septic shock [6]. Finally, the arterial to intramucosal pH gradient and the ‘standard’ intramucosal pH differ only by the constant 7.4 [7]. There are no data supporting the superiority over pHi or either of these measures as an indicator of splanchnic ischemia or as prognostic indicators. Indeed we have shown that the pHa-pHi gradient is of no prognostic significance in a group of emergency admissions to our ICU `[8].