There are many things in the ED which I would rather not do. Packing a bloody nose was highest on my list until the rhinorocket made this easy. Currently topping my list of procedures I do not enjoy is the insertion of the NG tube. It has been rated the most painful procedure we perform on a patient. There are many ways to lesson this pain like local or systemic analgesia but it still stinks for the patient. The gagging and spitting are not great for the provider who is trying to keep the fragile patient doctor bond intact.
Can I simply avoid this procedure?
This review of the question provides some articles to guide your practice and give you a basis or discussions with others who may not share your views.
An ng tube is too small a tube for gastric emptying in overdose and I will let the surgeon who is admitting the patient decide if the SBO needs decompression but the decision we make in the ED is whether a GI bleeding patient needs an ng tube. I am happy that the gastric lavage with water or ice in the setting of GI bleeding went the way of leeching and rotating tourniquets so we are only talking about diagnostic ng tubes.
Clearly there are patients that I do not need to place an NG tube on. If you vomit blood in from of me or EMS or bring in an example of your vomit for me to see (happens more often than you think) then I have my diagnosis and I do not need to place an ng.
I would love to argue that, if there is red blood in the stool (hematochesia) then this is undigested and therefore lower gi bleeding and I do not need to place an ng tube. Unfortunately this is not true. The study- Incidence of occult upper gastrointestinal bleeding in patients presenting to the ED with hematochezia. Am J Emerg Med. 2007 found that 10% of patients with hematochesia in fact had upper Gi bleeding. So much for that argument.
So maybe we are going about this the wrong way. Does ng lavage change outcome? A randomized controlled trial of gastric lavage prior to endoscopy for acute upper gastrointestinal bleeding. J Clin Gastroenterol. 2004 as well as other studies have shown that there is no improved outcome. There may be improved visualization during endoscopy but no patient related outcome benefit. So that is hopeful and a reasonable argument for not performing this painful procedure on patients.
How good a diagnostic test is an ng tube? Usefulness and Validity of Diagnostic Nasogastric Aspiration in Patients Without Hematemesis. Ann Emerg Med 2004 gives us a sensitivity of 42% and a specificity of 91%. A 42% sensitivity stinks. So if you are doing this test to make sure that there is no upper Gi bleeding, a negative test would not rule this out. If you do get blood back then it is probably an upper Gi bleed. So the next obvious question is how does this change management? If I get blood back from my ng tube do I force GI to come in and scope the patient immediately? To answer this question I need to split my Gi bleeders into 2 groups; esophageal variceal bleeders and non-esophogeal varicies bleeders. We know that the vast majority of non-variceal bleeding will stop on its own. The patient needs adequate resuscitation and correction of coagulation factors but the bleeding stops on its own. Studies have not shown that these patients as a group benefit from emergent endoscopy. So getting a positive aspirate and trying to force your Gi service to come in is not warranted unless the patient is crashing and you do not have other options. If the patient is crashing then the lavage is probably not helpful anyway.
This is not the case with variceal bleeding. Early endoscopy has been shown to help and the bleeding will not consistently stop on its own. But that raises other questions. A couple of sources still refer to esophageal varices as a contraindication to ng tube placement. I could find no references to validate this concern and in fact found a few studies like Placement of Nasogastric Tubes and Esophageal Stethoscopes in Patients with Documented Esophageal Varices. Anesth Analg 1988 which demonstrated the safety of this procedure. This is the population where a positive aspirate would warrant a change in management and allow you to push for early endoscopy. In this population I would want an nasogastric tube placed. Note: how do you know if a patient has variceal bleeding versus other bleeding? Assume that all patients with a history of advanced liver disease or physical findings of portal hypertension are variceal bleeders.
So lets review:
Known non-varicieal upper Gi bleeding (vomiting blood)- you have your answer already – no tube.
Not sure if heme + patient is upper Gi bleeding–sensitivity too low to rule out upper gi bleeding- no tube.
Possible variceal bleed – ng tube to determine early vs late endoscope. – yes tube
Hope these articles help.
Dr. Tom Perera