?Flu, GI bleed answer

A 30s M with a hx of asthma comes in with cough, fever, myalgias, nausea for three days. His VS are HR 118, RR 16, BP 122 / 82, T 102, RA O2 sat 97. He appears mildly ill, HEENT is normal, and lungs are clear. It is the middle of the flu epidemic and you have many patients with the same symptoms.

Do you get a CXR?
Do you give IV fluids?
Do you send a rapid flu test or flu PCR?
Do you send a lactate, give a bolus of saline at 30 ml/kg, and start empiric abx?
Do you give tamiflu based on the flu test or duration of symptoms?
Do you put the pt in isolation while in the ED?

—

GI bleed answer

For me, the key to managing potential upper GI bleeds (ugib) is whether the patient (pt) is actively bleeding. Like all bleeding, active bleeding requires aggressive treatment to stop the pt from crashing. In general, this means calling GI to come in to scope the pt and supportive measures such as oxygen, blood products, and to a lesser extent – iv fluids and medications.

If blood is pouring out a patient’s mouth, he is actively bleeding. Otherwise, the only way to tell for sure if a patient is still bleeding is an NG tube. NG tubes are generally painful and unnecessary, but it tells you how quickly you need GI. In most cases, you will get the answer when you hook it up to suction. If there’s no blood at all, the patient is not actively bleeding (in all likelihood). If there’s blood and it doesn’t stop when you hook it up to suction, the pt is still bleeding. If there’s blood and it stops when it’s all sucked out, a lavage is generally recommended. If it clears, the bleeding has probably stopped. If the lavage fluid doesn’t clear, the pt is still bleeding. (** most ugib occur in the esophagus and stomach, but this algorithm doesn’t work if the ugib is downstream from the stomach **).

Having said that, you usually get a pretty good sense of whether the pt is actively bleeding based on the h&p. This pt tells you that the bleeding is not acute.

A common decision point is whether to give patients an immediate transfusion with uncrossed RBCs. The downside of an unnecessary transfusion is the loss of a valuable unit of type O blood. In a patient who reports blood loss, hypotension, and paleness, we gave the uncrossed blood. Since we were giving blood, we held off on a saline bolus. Although the pt’s INR is suspected to be high, we did not give kcentra or ffp immediately.  (On a funny note, the medicine rotator gave uncrossed blood to a gi bleed the day before, and the pt’s hct turned out to be 45, so he hesitated this second time around.)

If an NG tube didn’t hurt, I would put it in everyone with a potential ugib. We didn’t put one in because I had a sense that the pt was not actively bleeding. The rectal exam is largely unimportant in this case but necessary because everyone will ask you about it. Regardless whether it shows nothing, guaiac+ stools, melena, or bright red blood, it is an upper GI bleed. Lower GI bleeds do not come in with hematemesis.
GI was called. Because I didn’t think the pt was actively bleeding, I felt he could be admitted and wait for the morning. His Hb and his INR were both 6. GI will rarely scope someone emergently when the INR is 6.

If the patient makes trop, don’t start an aspirin (or heparin). It is a common error. The patient is making trops because of an ugib, not because of acute coronary syndrome. An aspirin does more harm than good.

Don’t forget that occasionally, the definitively treatment for an ugib is an operation. Surgery may need to be involved sometimes.

The patient looked better after a couple of units of blood in the ED. He went to the ICU and received vitamin K, ffp, and iv proton pump inhibitor (helpful though not life-saving). His upper endoscopy showed a small gastric ulcer that wasn’t bleeding along with gastritis. He did well and walked out of the hospital.