Critical Gastrointestinal Bleeding at an Inpatient Rehabilitation Center: Incidence, Risk Factors, and the Role of Gastrointestinal Prophylaxis

Abstract

To determine the incidence of critical gastrointestinal bleeding at an inpatient rehabilitation center, the risk factors associated with said bleeding, and the role of gastrointestinal prophylaxis. This was a retrospective case-control study. Medical records and databases from November 7, 1997, to July 1, 2008, were reviewed. Inpatient rehabilitation center. All adult patients admitted from November 7, 1997, to July 1, 2008, who developed critical gastrointestinal bleeding were considered case patients. Gastrointestinal bleeding was defined as "critical" if it directly led to hemodynamic instability requiring the patient's transfer from inpatient rehabilitation to a higher level of care for close monitoring and/or emergent management. The control for each case patient was a randomly chosen adult patient who was admitted on the same day or the day before but who did not develop critical gastrointestinal bleeding. Clinical information was extracted by chart review. A total of 70 patients (35 case patients and 35 control patients) were included in this study. Patient demographics, comorbidities, and medication regimens were compared between case patients and control patients. Of 11,645 adult patients admitted to an inpatient rehabilitation center during a 10-year and 8-month period, 35 developed critical gastrointestinal bleeding, which translates to an incidence of 0.3%. Significant risk factors for critical gastrointestinal bleeding included diabetes, anticoagulant treatment, treatment with glucocorticoids, renal insufficiency, and colonic disease. Gastrointestinal prophylaxis was not found to be protective. The overall incidence of critical gastrointestinal bleeding was low in the inpatient rehabilitation setting. This finding is similar to a study in which the authors reported a small incidence of gastrointestinal bleeding, 0.4%, within the acute care setting. Significant risk factors included diabetes, anticoagulant treatment, treatment with glucocorticoids, renal insufficiency, and colonic disease. The mechanism of gastrointestinal bleeding from this inpatient setting may be different from what has been proposed among critically ill patients in the intensive care unit setting. Some of the emerging data now indicate the potential risks associated with long-term proton pump inhibitors and histamine-2 receptor antagonist therapy, including variations in the bioavailability of common medications, thrombocytopenia, interstitial nephritis, Clostridium difficile colitis, and nosocomial pneumonia, not to mention the expense of such medications. Gastrointestinal prophylaxis was not found to provide significant protection in the inpatient rehabilitation setting, and its routine use among most rehabilitation inpatients appears potentially to be unwarranted. Although our study does not support the use of pharmacologic treatment for gastrointestinal prophylaxis, further studies would be warranted before making a final conclusion regarding gastrointestinal prophylaxis within the inpatient rehabilitation setting.