Through the cubicle’s curtains she looked like a middle-aged Doris Day, pink housedress buttoned to the neck and every blond hair sprayed in place. But her posture spoiled the effect. She sat cross-legged on the emergency-room stretcher, bent forward like an ancient pagan at morning ablutions. As I watched, she placidly, almost dutifully, delivered into the plastic basin cradled in her lap a stream of clear yellow vomit.
I looked around for Terry to get a bit of the story. Terry was one of the best residents--low-key, smart, and very conscientious.
Hey, thanks for letting me grab lunch, I said. Anything up?
Not much. Just this patient of Dr. Morgan’s. I saw her quickly, but he had already called Dr. Summers, the gastroenterology consult, so I sort of backed off.
Terry sounded apologetic, but she shouldn’t have. Morgan was the chief of medicine, and the lines of responsibility blur a bit when private patients show up in the emergency room. This was the boss’s patient. The fewer cooks in the kitchen, the better.
What’s her story? I asked.
Her name is Mrs. Fratelli. She’s 59 years old, with a history of insulin-dependent diabetes. Says she’s been throwing up for two days, since she ate some bad cottage cheese. Hasn’t been able to keep anything down, but swears she hasn’t missed any insulin doses. No fever, no diarrhea, no headache, no chest pain. No nothing, really. Just vomiting. Summers wants to admit her and give her intravenous fluids. He thinks it’s just garden- variety food poisoning. Morgan is on his way down to see her, too.
Get any lab tests back yet?
Just her finger-stick glucose. It’s 415.
The 415 thudded like a speed bump. Though normal glucose levels shouldn’t go much over 100, diabetics can hit levels as high as 200 or even 300, especially when they’re sick. But 415 was too high for garden-variety anything. I decided to take a look.
Mrs. Fratelli still had the plastic basin perched on her knees. Her look of calm concentration hadn’t wavered.
Hello, Mrs. Fratelli, I said. I’m Dr. Dajer. I’m in charge of the emergency room.
She nodded, her eyes fixed on her offering bowl. I crossed my arms on top of her stretcher railing and considered the evidence. So far, my clues were high glucose and a plastic emesis basin. Only one slightly ridiculous question occurred to me.
Mrs. Fratelli, do you feel sick?
Mercifully, she didn’t roll her eyes or guffaw. Instead she thought for a moment, patted her belly gently, and replied, Well, I’ve got the heaves, of course. And I feel so very tired. But nothing else, I think.
Having any trouble urinating?
No, not really. Maybe a little more than usual.
The rest of her answers were equally unrevealing. I glanced at her chart: Temp 98.9, normal. Pulse 88, normal. Respiratory rate 18, a touch high. Blood pressure 140/60, diastolic of 60 a touch low. No help there.
When patients can’t point and say where it hurts, doctors’ brains can skid like ten-wheelers hitting a slick curve. The diagnostic refuge of the perplexed becomes viral syndrome or food poisoning, which doctors are fond of calling gastroenteritis.
I rearranged my clues. Unexplained high glucose in a middle-aged diabetic woman equals--?
Equals urinary infection, dummy, said a little voice in my head.
Doctors pay too much tuition to be taught rules of thumb; instead we are bequeathed clinical pearls. And this pearl’s corollary gave me another clue. Because diabetics can suffer nerve damage from their high glucose levels, they may not experience the discomfort that is the most common sign of bladder infection.
Mrs. Fratelli, I know the nurses have already taken a urine sample, but would you mind giving me another one? There’s something I’d like to check right away. Maybe we can start making you feel a little better, quicker.
She smiled for the first time. Can’t argue with that now, can I?
Five minutes later the little dipstick that detects white cells in the urine was turning a dark purple. No doubt about it, Mrs. Fratelli had a urinary infection. A second pearl slid into place: simple bladder infections don’t make you throw up. I spied Barb, the nurse taking care of Mrs. Fratelli, across the ER. She was rushing somewhere, holding two IV bags.
Barb, hold up, I half-shouted, walking quickly after her. Quick question: Mrs. Fratelli, the lady in 6A--is her temp oral or rectal?
Barb thought for a second. Oral, why?
I think it’s more than food poisoning. She’s got a definite bladder infection. Maybe more. Any chance of a rectal temp?
Barb gave me a wry smile. I was notorious for requesting them. But a third pearl states: no temp is a real temp until it’s a rectal temp. Sick patients breathe quickly, making the mouth too cool and dry to reflect the body’s true temperature.
A bit later Barb was back. She said, Okay, okay, just don’t say I told you so. Her rectal temp is 101.7.
Clinical pearl number four clicked into place: infected urine plus fever plus vomiting equals sepsis.
As feared as it is deceptive, sepsis takes its name from the Greek word meaning to putrefy. Known for generations as blood poisoning, it generally means that bacterial invaders have managed to breach the natural barriers of bladder, lung, and skin and enter the bloodstream. Once there, they begin spreading a witches’ brew of toxins produced by the bacteria or formed from the bacterial cell wall. In either case, they trigger a devastating immune response, the biological equivalent of calling a napalm strike on your own position. The body’s inflammatory proteins, called cytokines, cause blood vessels to collapse and leak, tissues to swell, lungs to swamp, and kidneys to fail. The result can be septic shock so severe that no amount of intravenous fluid or artery- constricting medication can bring the sagging blood pressure back up. Roughly half of all patients with septic shock die.
For such a tidal wave of a disease, it is astonishing how subtle the first ripple of trouble can be. Even worse, those at greatest risk-- older patients--are the toughest to diagnose. They may, like Mrs. Fratelli, come in because they are vomiting. Or they may feel confused, short of breath, or somehow just not right. Their complaints often sound nothing like the classic sepsis doctors are trained to recognize. Because we spend so much of our residencies caring for patients in the intensive care unit, doctors tend to become acquainted with sepsis only in the gravely ill. These patients spend weeks and even months with ventilators and multiple intravenous lines providing direct access for invasive bacteria. It is easy to forget that sepsis can afflict patients who walk in off the street and seem okay until, quite literally, they crash against death’s door.
To make matters worse, doctors have been hamstrung by a clinical imperative: avoid bacterial overkill. Before blasting an infection with antibiotics, we are supposed to pinpoint the source with surgical precision. To do otherwise exposes bacteria to antibiotics needlessly, increasing the chance that they may develop resistance to the drugs we depend on to destroy them. So if we don’t immediately find, say, pneumonia or an infection in the urine or skin, we are tempted to wait until a patient’s illness declares itself through a lab test. But the only test that proves sepsis is a blood culture, which takes two days to turn positive. By then, the only declaration would be on the patient’s death certificate.
Though it may seem appallingly obvious that one should treat blood infections posthaste, some early evidence misleadingly suggested that antibiotics were ineffective. It wasn’t until 1980 that a landmark clinical trial clearly showed that giving antibiotics--and the sooner the better-- cut deaths from sepsis in half. But medical thinking changes slowly; the new hurry-up philosophy has not yet reached the mainstream.
Moreover, with the rise of biotechnology over the past decade, researchers tried to design antibodies that could intercept bacterial toxins or the body’s kamikaze cytokines, stopping the septic response in its tracks. Yet hundreds of millions of dollars in research and dozens of clinical trials have yielded only dismal results. Worse yet, dazzled by the prospect of designer antibodies, doctors have neglected the basics. One large and very expensive study found that it took an average of 19 hours to start treatment once sepsis had been diagnosed. Emergency room doctors still routinely defer antibiotic orders to admitting doctors who, in turn, complacently wait for nurses to get around to it when they can.
The key, as the brilliant Confederate general Nathan Bedford Forrest put it, is to get there first with the most men.
Mrs. Fratelli, I was convinced, had sepsis. I walked over to where Dr. Summers sat writing admission orders. Dr. Morgan stood at his side.
Hello, I said. What do you think about Mrs. Fratelli?
Oh, hi, Tony, Morgan replied. Seems this cottage cheese gave her quite a turn. Probably gastroenteritis.
Gastroenteritis, Summers nodded. We weren’t sure whether to cover her with antibiotics.
I tried to keep my voice casual. Did you know she has a fever? Rectal temp’s 101.7. And her urine’s infected. I dipped it myself.
Summers was unmoved. Well, she did eat that spoiled cottage cheese, you know. Been vomiting ever since. It’s gastroenteritis.
My mind flashed back two months. A 60-year-old woman had come into the ER. Her symptoms were vomiting and low blood pressure. When flecks of blood showed up in her vomit, everyone figured her blood pressure was low because of bleeding from a stomach ulcer. A gastroenterologist examined her with a fiber-optic scope and saw only a few lesions in the stomach wall. Her oral temp read 98. No one took a rectal temp for six hours, but when they did, it was 104. She was vomiting up blood because a bladder infection had spread to her bloodstream, triggering sepsis. Vomiting in that weakened state had produced lesions in the stomach wall. Everyone had been fooled.
I came on duty after she’d already gone into cardiac arrest several times. We resuscitated her again and again, to no avail. Her husband, his face frozen with grief, kept asking me, How could this happen? She was fine yesterday. We went for a long walk. How could this happen?
Two days later her blood cultures grew E. coli, the bacterium that causes most bladder infections. The delay in starting antibiotics had squandered her only chance of survival.
Gastroenteritis, Summers said again, as if repetition would mollify me. What else could it be?
Look, I’ve seen this before, I said, trying to keep my voice even and authoritative. She’s vomiting because she’s septic, and she’s septic from her bladder infection. She needs amp and gent, now. Right away.
When used in combination, the antibiotics ampicillin and gentamicin kill the bacteria that are the likely cause of urinary infections. I was ready to order them myself. Private patient or no private patient, this was still my ER.
We argued back and forth for a while, with both Morgan and Summers gently trying to suggest I was crying wolf. True, the robust woman sitting up in 6A didn’t look ready to succumb to a catastrophic illness. But I remembered that other woman going into repeated cardiac arrest. And I remembered her shattered husband. Finally, more to get me off their backs than because they were convinced, they gave in.
I headed down the hallway to Mrs. Fratelli’s room, where I found Barb already at her bedside.
Amp one gram and gent 120 milligrams stat, I told her.
Boy, one little rectal temp and you start calling in the heavy artillery, she tut-tutted.
I took up my position at Mrs. Fratelli’s railing. We think we know what’s causing the trouble, I began. If we’re right, you might even be able to get rid of that basin soon.
My patient stared into her lap. And just in time, too. You know, I just realized it doesn’t go with my robe.
Two days later I met Morgan outside Mrs. Fratelli’s room.
How’s she doing? I asked.
Fine.
Seen the lab results?
Not yet. Anything growing in the blood cultures?
I’d just been to the lab. E. coli, I answered. It was clear now that Mrs. Fratelli was suffering from more than a case of food poisoning from bad cottage cheese.
Morgan’s face brightened like a naturalist’s before a rare and brilliantly colored butterfly.
No! Good for you! He smiled sheepishly, patted me on the back, and added jokingly, You know, I just knew we should have taken a culture of that cottage cheese.
