|| Checking for direct PDF access through Ovid
Gastrointestinal promotility agents may improve tolerance to enteral nutrition, reduce gastroesophageal reflux and pulmonary aspiration, and therefore have the potential to improve outcomes of critically ill patients.To systematically review and critically appraise studies of promotility agents in the critical care setting.Computerized bibliographic search of published research (1980–2001), citation review of relevant articles, and contact with primary investigators.Randomized trials of critically ill adult patients that evaluated the effect of promotility agents on measures of gastrointestinal motility were included.Relevant methods and outcome data were abstracted in duplicate by independent investigators.We reviewed 60 citations; 18 articles met the inclusion criteria (six studies of feeding tube placement, 11 studies evaluating gastrointestinal function, and one study of clinical outcomes). The heterogeneity of study methods and outcomes measured precluded a quantitative synthesis of the data. Although there are conflicting studies, the larger and more methodologically robust studies suggest that metoclopramide has no effect on feeding tube placement. Erythromycin has been shown to increase success rates with small-bowel tube placement in two studies. Eight of ten studies evaluating the effect of cisapride, metoclopramide, or erythromycin on measures of gastrointestinal transit demonstrated positive effects; the two studies that did not were relatively small (n = 27 and 10) and likely had inadequate power to detect a difference in treatment effect. No study demonstrated a positive effect on clinical outcomes.As a class of drugs, promotility agents appear to have a beneficial effect on gastrointestinal motility in critically ill patients. A one-time dose of erythromycin may facilitate small-bowel feeding tube insertion. Administration of metoclopramide appears to increase physiologic indexes of gastrointestinal transit and feeding tolerance. Concerns about safety and lack of effect on clinically important outcomes preclude strong treatment recommendations.