I Knew. I Hesitated. She Almost Died. 

What a near-fatal near-miss taught me about trusting myself as a physician

What’s worse than making a mistake?

Knowing that you’re probably making one, and doing it anyway.

I’m sure you’ve done this at some point in your life: Overrode your own instincts and made a bad decision.

Maybe you didn’t buy the trip insurance, and then had to eat the cost when a hurricane hit. Or maybe you put off investigating a strange sound coming from the car engine, and ended up on the side of the road waiting for a tow truck.

We’ve all had those moments where we ignored our gut instinct, and then had to deal with the consequences.

For a physician, the stakes of that moment can be someone’s life.

After twenty years of practicing medicine, I learned to listen to my gut.

I didn’t always.

This is the story of the day I almost let my patient talk me out of saving her life.

It was Fall 2011, and I was seeing patients in my internal medicine clinic. Kelly had made an urgent early morning visit for shortness of breath.

She was an athletic woman with a history of asthma, and also, she was a nurse. Her seasonal allergies had been acting up, and she had already called asking for a refill of her Flonase and her Albuterol inhaler. “But it’s not working, I’m just so short of breath!” she told me. As we chatted, I could see that she couldn’t finish a full sentence without needing to stop and take in another breath.

I’d treated her for asthma exacerbations before, but this time, when I listened to her lungs, she only had a few scattered wheezes.

“I guess the allergies are really bad this year, huh? My lungs are doing new and strange things!” She was convinced that she was having a bizarre asthma attack.

Then a thought occurred to me: Could this be a PE?

PE, or Pulmonary Embolism, is when a blood clot forms in the deep veins of the leg, gets loose, travels up the large veins, and then blocks blood flow through the lungs. If the deoxygenated blood from the veins can’t be pumped through the lungs and get oxygenated, the body becomes oxygen-starved. The bigger the blood clot, the worse the blockage, and the more serious the oxygen starvation.

A PE can be deadly serious: it’s consistently one of the top 3 causes of cardiovascular death in the U.S. (the other 2 being heart attacks and strokes). There are almost a million cases of PE in the U.S every year. So, it’s not rare, and, it’s very dangerous.

I asked her about common risk factors and symptoms: “Kelly, have you had any recent plane trips? Any leg pains or cramping?” No, she hadn’t. She didn’t smoke, she hadn’t had surgery. No one in her family had had clots. But she was on the Pill for birth control, which can raise the risk of clots.

I told her what I was worried about. “You seem too short of breath for the degree of wheezing you have,” I said. “This could be something more serious, like a PE.”

The wise words of one of my medical school teachers echoed in my mind: If PE is on the list, rule it out quick.

Hence, my next sentence was going to be: Kelly, I’d like to send you to the emergency room for a workup.

But Kelly laughed. 

“Oh, I don’t think this is a PE at all. I’ve seen patients with PEs, and I know I’d be feeling alot worse than this.”

So I walked her up and down the hall on the oxygen monitor, and her oxygen saturation was 97%. We did a 12-lead EKG, and it was normal.

I knew that sometimes, patients with genuine asthmatic pathophysiology do not have marked wheezing. Given this, her normal oxygen saturation, and normal EKG, I was somewhat reassured.

But, PE wasn’t ruled out. It was still on my list of possible diagnoses.

PE was most definitely not on Kelly’s list. “I really don’t want to go to the ER, I need to get back to work. Can’t I just have a nebulizer treatment and some steroids?” She was polite, but firm.

Kelly was an experienced and knowledgable nurse who knew her body, and I wanted to respect her opinion. We negotiated a bit and came up with a compromise:

We would escalate treatment for asthma. But, she would have a set of labs before leaving the office, including a D-dimer blood test, which is sometimes a good test for a clot. It’s good in that if it’s negative, there is no clot. If it’s positive, it can mean a million things, one of which is a clot.

She agreed and off she went. 

I had a nagging sense that I was screwing up.

I continued seeing patients in my busy morning clinic, but I was distracted. As soon as I had a break for lunch, I looked up Kelly’s results.

The D-Dimer was 2500! That was really, really positive. I dropped my almond-butter-and-jelly sandwich and called Kelly’s cell phone.

“Kelly, the D-dimer is positive, you’ve got to go to the ER for a CT scan,” I told her. I was trying to sound calm, thinking CRAP CRAP CRAP I should have sent her to the ER in the first place.

Kelly was unperturbed. She was obviously also eating lunch, and replied between munches: “But I think the nebs helped. If it was a PE, the nebs wouldn’t have done anything. I really don’t want to have to spend hours in the ER when I don’t have to.”

I had to admit, the fact that she had had improvement after the nebulizer treatment fit more with her asthma than a PE. And she sounded good, speaking full sentences while eating. This was a real difference from the morning visit to my office.

But that little voice from the past kept whispering in my ear: 

If PE is on the list, rule it out quick.

So again we negotiated and again we compromised. I ordered the CT scan of the chest myself, but I insisted that she have it done right away, and I spent a lot of time on the phone coordinating that. I never finished my lunch. Charting and calls to other patients went out the window.

I also had to forego my lunchtime phone call to home, to check in with my mom, who was watching my toddler son. (That was painful. Even though the conversation usually consisted of chatting about what he ate, if he napped, and if he pooped, it was my chance to hear his little squeals and burbles, to get grounded.)

One-o’clock came and I had patients waiting for me. I tried to keep my mind on my afternoon clinic. It was late when I called down to outpatient radiology to see if Kelly had come in for her scan yet– and she had not!

I called her cell. I was furious– with myself, not the patient. CRAP CRAP CRAP I should have sent her to the ER in the first place, I thought, again.

Kelly didn’t sound calm anymore. 

She answered on the first ring. She was scared, breathless: “I’m here, I’m in the ER, I got really short of breath all of a sudden… I thought, Oh my God, this IS a PE, and I called 911 on myself.” 

I could hear the blipping and beeping of the monitors in the background as she relayed what happened in the ER: 

Her oxygen levels were much lower, and she had been put on oxygen. She explained to the staff that her primary care doc had been worried about a PE and her D Dimer had been elevated. So they did a CT scan right away, and it showed not one, but several blood clots to the lungs (in other words, several PEs). 

She was put on intravenous blood thinners to help melt away those clots and prevent new ones. She would be admitted at least overnight, until she no longer required oxygen. 

I apologized. 

I felt badly that I hadn’t insisted that she go to the emergency room right away, and I told her that.

“Don’t feel bad, I probably wouldn’t have gone anyways,” she laughed. 

I think I could have tried harder to convince her. We were lucky that those clots hadn’t been large enough to completely cut off her circulation. Very, very lucky. She really could have died. Thankfully, Kelly was my patient for many years after that incident, and I think we both learned alot from the whole thing. 

I always aim to learn from my mistakes. 

The first lesson here is: If you are at all considering a life-threatening diagnosis for a patient, you need to rule that out FIRST before considering more benign diagnoses.

And, of course, it also underscored the old classic: If PE is on the list, rule it out quick!

Of course, on the flip side, this approach could also be considered “trigger-happy”. Some people would say that cautious approach only contributes to overflowing ERs, as well as increased healthcare costs.

I know for a fact that I have sent people to the ER out of legitimate concern for a serious diagnosis, but it ended up being something benign. (I have written about a case of chest pain that I was worried was a heart attack, and ended up being a pulled muscle.)

I don’t feel bad about that, however. 

The ultimate lesson here:

Better to be wrong and the patient is alive after an unnecessary emergency room evaluation, than to be wrong and the patient is dead after a missed diagnosis.

---

An earlier version of this post was first published on my blog in 2011.