Safety and efficacy of nasogastric intubation for gastrointestinal bleeding after myocardial infarction: an analysis of 125 patients at two tertiary cardiac referral hospitals.

Our purpose was to analyze risks versus benefits of nasogastric (NG) intubation for gastrointestinal (GI) bleeding performed soon after myocardial infarction (MI). While NG intubation and aspiration is relatively safe, clinically beneficial, and routinely performed in the general population for recent GI bleeding, its safety after MI is unstudied and unknown. In addition to the usual complications of NG tubes, patients status post-MI may be particularly susceptible to myocardial ischemia or cardiac arrhythmias from anxiety or discomfort during intubation. We studied NG intubation within 30 days of MI in 125 patients at two hospitals from 1986 through 2001. Indications for NG intubation included melena in 55 patients; fecal occult blood with an acute hematocrit decline, severe anemia, or sudden hypotension in 37; hematemesis in 18; bright red blood per rectum in 8; and dark red blood per rectum in 7. The intubation was performed on average 5.3 +/- 7.2 (SD) days after MI. NG aspiration revealed bright red blood in 38 patients, "coffee grounds"-appearing blood in 45, and clear (or bilious) fluid in 42. Among 114 of the patients undergoing esophagogastroduodenoscopy (EGD), EGD revealed the cause of bleeding in 79 (95%) of 83 patients with a grossly bloody NG aspirate versus 12 (39%) of 31 patients with a clear aspirate (P < 0.0001, OR = 31.3, OR CI = 9.4-103.1). Among 85 patients undergoing EGD within 16 hr of NG intubation, stigmata of recent hemorrhage were present in 28 (42%) of 66 with a bloody NG aspirate versus 3 (16%) of 19 with a clear aspirate (P = 0.06, OR = 3.93). Among 35 patients undergoing lower GI endoscopy, lower endoscopy revealed the cause of bleeding in 14 (56%) of 25 patients with a clear NG aspirate versus 1 (10%) of 10 patients with a grossly bloody aspirate (P < 0.04, OR = 11.46, OR CI = 1.55-78.3). The two NG tube complications (epistaxis during intubation and gastric erosions from NG suctioning) were neither cardiac nor major (requiring blood transfusions). This study suggests that short-term NG intubation is relatively safe and may be beneficial and indicated for acute GI bleeding after recent MI. Aside from improving visualization at EGD, the potential benefits include providing a rational basis for the timing of endoscopy (urgent versus semielective), for prioritizing the order of endoscopy (EGD versus colonoscopy), and for avoiding or deferring endoscopy in low-yield situations (e.g., colonoscopy when the NG aspirate is bloody). These benefits may be particularly relevant in patients after recent MI due to their increased endoscopic risks.