Blog | Thursday, July 19, 2012

Asking questions to better understand

Recently we made a diagnosis that really mattered for a patient. He came to us for an entirely different problem. Here is the rough sequence of events:

Man in his 50s, no recent hospitalizations, admitted for new atrial fibrillation with rapid ventricular response. His symptoms followed several days for nausea, vomiting and decreased oral intake. His abdomen was minimally tender. All routine labs were interpreted as unremarkable including liver tests and lipase. A CT suggested pancreatitis, but he had BS present and no other symptoms.

His atrial fibrillation responded to IV diltiazem drip and converted to sinus rhythm. He had two more episodes of atrial fibrillation, but eventually maintained sinus rhythm.

I first saw him on day 3 of his hospitalization. On day 4 his Hgb dropped from 13 to 10.5 and he reported a large melanic stool.

Review of his original labs provided a clue that we had missed. His initial BUN was 42 with a creatinine of 1.4. But in the housestaff's defense, they had checked his stool for blood at that time and it was negative.

I have left out things that I did not know until I got to the bedside on day 4. Some questioning led us to a totally different diagnosis.

I asked for questions. As a sign of the problem, 2 commenters suggested tests. One commenter was spot on.

The melanic stool required explanation. Therefore asking questions that might give information towards a diagnosis is the proper course. I picked the right first question: What medications were you taking? He was taking 3 medications including Voltaren (diclofenac), a strong non-steroidal.

"How long have you been taking Voltaren?" I asked.

"10 years."

"Why were you taking it?"

"For my arthritis."

"What kind of arthritis do you have?"

"I don't know, just arthritis."

A careful examination revealed PIP and MCP joint abnormalities, spongy, squishy and tender, while the DIP joints had some mild deformities but no squishiness or sponginess.

Now I knew the sequence of events and the underlying problem. Do you?

The tender spongy PIP and MCP joints raised the likelihood of rheumatoid arthritis. We considered hemochromatosis, but the patient did not have diabetes or cirrhosis and he had a low ferritin.

The rheumatoid factor was positive and both the CRP and ESR were significantly elevated. The hand X-rays confirmed symmetrical inflammatory arthritis.

He responded dramatically in less than 24 hours to low dose steroids.

We postulated that he had had undiagnosed rheumatoid arthritis for 10 years, treated with NSAIDs. The NSAIDs cause GI bleeding and dyspepsia. He became volume contracted, which stimulated his atrial fibrillation.

By discharge he remained in sinus rhythm, and was discharged on low dose steroids for his rheumatoid arthritis.

To me the point of this story was that we ignored or failed to appreciate his arthritis. Evaluating his GI bleed made us focus on the arthritis because we focused on why he was taking diclofenac.

db is the nickname for Robert M. Centor, MD, FACP. db stands both for Dr. Bob and da boss. He is an academic general internist at the University of Alabama School of Medicine, and is the Associate Dean for the Huntsville Regional Medical Campus of UASOM. He also serves as a frequent ward attending at the Birmingham VA Hospital. This post originally appeared as parts one, two and three at his blog, db's Medical Rants.