Gastric tonometry has proved to be a sensitive but not specific predictor of outcome in the critically ill. The data accumulated to date indicate that those patients able to achieve or maintain a normal gastric mucosal pH do better than those who do not. In addition, therapy aimed at improving an abnormal gastric mucosal pH has proved to be less successful. These findings may simply indicate that tonometry identifies those "responders" and "nonresponders," as becomes increasingly apparent in populations of critical care patients receiving interventional therapy. Gastric tonometry has undergone a number of methodologic changes over the last decade, seeing a switch from saline to automated gas tonometry. Along with this switch of methodology has come a deeper scrutiny of the indices used to assess gut perfusion. Most studies (including all the interventional ones) have used gastric mucosal pH. The newer indices of gut luminal PCO2 (PgCO2) referenced to arterial CO2 (PgCO2-PaCO2) or end tidal CO2 (PgCO2-PeCO2), although relatively well validated, remain to be proven as predictors of outcome or guides to interventional therapy. If we take a fresh look at the interventional trials in intensive care patients, there is a very definite trend toward benefit in the protocol groups, although they are generally reported as negative studies. There is much to be accomplished, however, before we accept the gastric tonometer as a routine tool with which to guide therapy based on gastrointestinal perfusion, including a greater understanding of gastrointestinal physiology and, as ever, the call for an adequately powered prospective randomized controlled trial to evaluate the clinical utility of gas tonometry.