Background: Prevention and management of gastrointestinal complications in critically ill patients receiving enteral nutrition represent important nursing consideration. These require utilizing techniques and methods for increasing the rate of gastric emptying, and so enhancing tolerance to this nutritional support modality. Among these methods is abdominal massage, which has been proved to improve various digestive functions. Aim of the study: This study aimed to evaluate the effect of abdominal massage on gastrointestinal outcomes among intermittent enterally fed critically ill patients. Design: a quasiexperimental design was utilized. Setting: The study was carried out in General Intensive Care Unit at El Fayoum University Hospital. Subjects: a purposive sample of 110 critically ill patients who met the inclusion criteria and they were divided to equal study and control group (55 patients in each group). Data collection tools: Data were obtained through two main tools; patients’ assessment tool and gastrointestinal outcomes assessment tool. Results: There were highly statistically significant differences between the study and control group subjects regarding mean scores of gastric residual volume and intra-abdominal pressure, abdominal distension and constipation occurrence at the fifth day post implementation of abdominal massage, while no statistically significant difference between the study and control group subjects regarding frequency of vomiting episodes at the fifth day post implementation of abdominal massage. Conclusion: Abdominal massage was an effective intervention on the gastrointestinal outcomes among enterally fed critically ill patients as it leads to reduction in gastric residual volume, abdominal distension, intra-abdominal pressure and constipation. Recommendations: On-going and regular in-service educational programs about assessment of gastrointestinal outcomes among enterally fed critically ill patients and how to measure gastric residual volume and intra-abdominal pressure. The study should be replicated on large sample and in different geographical settings in order to generalize the results.