The ICU can be a terrifying place for an intern. Of course, the patients are probably a bit more frightened, but then, many of them are unconscious. Rounds are endless in the ICU; new admissions, work rounds, teaching rounds, evaluations for transfer, afternoon rounds, lasix rounds (don't ask), all punctuated by CPR codes on the floor, and your own patients trying their hardest to die.
One of the more sobering ICU experiences is the gastrointestinal (GI) bleed. One night while making some sort of middle-of-the-night rounds, I saw a patient sitting up in bed watching TV. She was middle aged, hair done but mussed from plasticized hospital pillows, and quite awake. I grabbed her chart and stopped in, happy to have someone conscious to talk to. She came to the hospital when she noticed dark black stools, and in the ER she was found to have a low blood count. But she looked fine, and given how busy our ICU was, I wondered how the hell she warranted a bed.
We chatted casually for a bit about nothing in particular, she as happy as I to have someone to talk to, and in mid-sentence her eyes rolled back and her telemetry alarms went off. Her blood pressure tanked, her heart rate soared, and she was clearly in shock. Just like that: fine one moment, dying the next. Thankfully, we saved her, but that taught me a valuable lesson: every GI bleed is potentially life-threatening, at least until you have a good idea about what's going on inside the patient.
The gut, from mouth to anus, is a continuous tube, and a long one. Blood can come from any part of it, but some problems are more common and some more dangerous than others. In general, we tend to divide GI bleeds into "upper" (from the mouth to upper part of the small intestine) and "lower" (all the rest).
This division helps us develop a differential diagnosis, a list of possible causes of the problem. In general, upper GI bleeding shows up in vomitus either as red blood or as vomit mixed with "coffee-grounds". If the blood makes it all the way through, it generally turns black and comes out as black, tarry stool (melena).
Lower GI bleeding generally shows up as bright red blood from the rectum. The list of potential diagnoses is huge, but a few of them are most common. Upper GI bleeds are often the result of ulcers, and lower GI bleeds are often caused by diverticulosis, arteriovenous malformations, or growths, either benign or malignant.
Once we manage to stabilize a patient, which often involves transfusion of saline and/or blood, we can take a look into the stomach and the colon via endoscopy. Usually we can find a cause, but in somewhere between 5-25% of cases, endoscopy doesn't give us an answer.
EGD, or esopaphagogastroduodenoscopy, allows direct visualization of the upper GI tract. Not only can an expert take a good look around, but she has many tools for stopping ongoing bleeding, including cauterization and injection of drugs. All of this can be done without opening up a patient.
The same specialist can do a colonoscopy, where a scope is passed from the rectum all the way to the cecum, where the small intestine meets the large. But these two procedures leave a lot of gut unexplored. Sometimes an specialized EGD can be used to look a little bit further into the small bowel, but for the most part, the small bowel is a dark zone when it comes to direct visualization. If the bleeding is bad enough, and a source isn't found by scope, the surgeons can dig in and try to locate and stop the bleeding, but people with hemorrhagic shock are not the easiest to operate on. They can develop blood clotting disorders, and start bleeding from literally everywhere.
The gold standard for identifying severe, obscure GI bleeding is angiography. A catheter can be threaded into the arterial system, dye can help localize bleeding, and then the bleeding can be stopped using various angiography-related techniques. This is very invasive, and only useful during severe bleeding. We have two other high-tech, relatively non-invasive tools to look for obscure GI bleeding: tagged red blood cell scan, and capsule endoscopy. In tagged RBC scans, RBCs are literally tagged with a radioactive isotope and if there is enough bleeding, the radioactive RBCs can be detected by the appropriate equipment. This technique has several limitations. Another technique is capsule endoscopy. This one is pretty cool.
In capsule endoscopy, the patient swallows a small capsule with a camera inside. This is then, um, recovered, and the films viewed. It can be quite successful. Another new technique, called "push enteroscopy", can also be quite effective, but requires very specialized training and a lot of time.
As I proofread this piece (not all that carefully...why change now?) I'm wondering what the point here is, and I suppose it's this:
The human body is full of surprises, and so is human invention. In my career I've watched several people bleed to death, and I'll never be able to erase those memories. Properly-developed medical technology saves lives, and techniques available today would have saved some of the people I've seen die, people whose blood drawn into a syringe looked like dilute cranberry juice before their hearts stopped, people who complained about how cold they felt as they died of shock.
CAVE, D. (2005). Obscure Gastrointestinal Bleeding: The Role of the Tagged Red Blood Cell Scan, Enteroscopy, and Capsule Endoscopy Clinical Gastroenterology and Hepatology, 3 (10), 959-963 DOI: 10.1016/S1542-3565(05)00716-0
PENNAZIO, M. (2004). Outcome of patients with obscure gastrointestinal bleeding after capsule endoscopy: Report of 100 consecutive cases Gastroenterology, 126 (3), 643-653 DOI: 10.1053/j.gastro.2003.11.057