The patient was an elderly Chinese immigrant with a history of gastric cancer. She had come into the emergency room in New York City complaining of abdominal pain.

“Tung, tung?” I asked in my laughable Cantonese.

She pinwheeled a hand over her belly. “Pain everywhere.”

When I pressed on her abdomen, she grimaced. But she had no fever or evidence of peritonitis, an inflammation of the membrane that lines the abdominal cavity.

“How long?” I asked her English-speaking daughter.

“Two days.”

“Getting worse?”

“Yes.”

Her abdominal X-rays showed no obstruction. But the X-rays couldn’t tell the whole story. Older people can harbor abdominal catastrophes—appendicitis, abscesses, gangrenous gallbladders—without mounting a temperature or yielding the usual physical exam clues. 

So, as happens 10 million times a year in this country, I ordered a contrast CT scan—an imaging technique that employs a spinning X-ray beam to take multiple images that create a portrait of internal organs. In a contrast CT scan, additional substances are given to the patient that allow radiologists to better tell one organ from another or detect signs of infection.

First, the patient drinks a quart of what we call an oral contrast—an iodine-based concoction (comes in a tasty lemon-vanilla flavor) whose high molecular density coats the intestines. Next, we inject six ounces of an intravenous material, also iodinated, to enhance the outline of the blood vessels. These materials weaken the passage of X-rays and make the treated regions appear white against the gray of surrounding tissues. The substances are usually excreted in a few hours.

Sometimes the injected material can be dangerous because of the processing burden it puts on the kidneys. And it can trigger a life-threatening allergic reaction. That occurs some 4,000 times a year in the United States. And what of the oral contrast? Usually, the worst that happens is a little diarrhea. It’s about as scary as a milk shake.

The results of the CT scan showed nothing dire. “We will admit you to the hospital,” I told her, “and do more tests.”

She neither changed her expression nor met my eye. 

“The good news is, the tests show nothing seriously wrong,” I said.

No reaction.

“How do you feel?”

A slight nod.

“The admitting doctors will be down in a little while.”

The daughter formed a thin smile. “Thank you.”

An hour later, the patient’s nurse, Nina, grabbed me.

“She’s bad. She can’t breathe. Sounds like airway obstruction.”

“Nina, it’s her belly, not her lungs.”

“You’d better look at her.”

I hurried over.

“Hrahaahr, hrahaahr” came the noise out of her throat. Above the rattle was a high-pitched tone that sounded like stridor, the sound of air being forcibly inhaled through swollen vocal cords or throat membranes. But my patient was making noise as she exhaled. When I listened through the stethoscope, her lungs sounded clear. The trouble was in her throat, but her breathing didn’t make sense. A straw is supposed to collapse when you suck on it, not when you blow out.

Nina and I quickly moved her to a fully monitored bed. The pulse oximeter showed the level of blood oxygen to be 90 percent—terrible, no; marginal, yes. I considered the possibilities. One was an allergic reaction to the scanning material she was given intravenously, which could cause vocal cord and laryngeal tissues to swell. But that would trigger stridor, not this bizarre-sounding exhalation. Another scenario: If the oral scanning material gets into the bronchial tubes that lead into the lungs, it can absorb a large amount of water and clog the airway. But the patient had drunk the concoction hours ago.

The larynx, however, is a bit like a Rube Goldberg contraption. Made up of nine different cartilages, it lies behind the tongue, jutting up like a half-buried, trumpet-shaped pitcher plant. The vocal cords, deep within, stretch from top to bottom. With every swallow, the trapdoor of the epiglottis drops over the opening, shutting off the entrance to the larynx. During the swallow, the whole laryngeal apparatus moves up and forward under the tongue, which with a powerful back kick propels the contents of the mouth into the esophagus. It requires split-second coordination of muscle, nerve, and cartilage to catapult food, drink, and saliva over the voice box and into the gullet.