Complications
There is collateral damage. To the doctor, and to the way people understand risk.
One thing applicants to medical school generally don’t consider is the thought that someday they might hurt somebody.
I think it was about 2000, maybe 2001. This was before we used ultrasound devices to direct the needles we place into a patient’s pleural space. The lung sits inside our chest like an inflated balloon inside a box. The pleural space is the “potential” space that is between the outside of the balloon and the inside of the box. Sometimes fluid accumulates in this space. That is generally not a good thing. So, the first step is to get a sample of the fluid in order to make a diagnosis.
That’s how I happened to meet Mr. G. His chest X ray showed a small amount of fluid around his left lung. It wasn’t a lot but enough that it needed investigating. So I “thumped” around his chest, listening for the hollow sound that indicated air (Lung) and the “thud” that meant fluid. I still remember the sight of pale yellow fluid coming back into the syringe- good news. Then there was a little cloud of red. I moved the needle slightly and the red went away, followed by more yellow fluid. Thanks to the miracle of lidocaine, Mr. G didn’t feel anything. It seemed like that little bit of red was not significant. I sent him home with a promise to call him as soon as the test results on the fluid came back. It was a Friday.
The main memory I have of this event is of the next day, Saturday. I was skiing with my wife and two young children. The sun was out, the snow was good, the kids were smiling; a great day. Then my phone (I’m guessing a Motorola flip) rang. It was one of my surgical colleagues.
“I’ve got Mr G here. And I might have his spleen in a bucket pretty soon,” she said.
“What?”
“Yeah, he showed up in the ER last night in a lot of pain. CT scan shows a big bleed around his spleen. I’m not sure it’s gonna stop. I’m getting ready to take him to the OR.”
The spleen is the most vascular (i.e. likely to bleed) organ in our body. I must have gone just past the pleural fluid (meaning I went 14 millimeters instead of 12), poked the underlying spleen (which sits under the left lung), then pulled back out into the correct space. But that poke, even with a small needle, was enough to cause dangerous bleeding. If it didn’t stop, Mr G would need emergency surgery to save his life.
The surgeon assured me she would keep me posted and hung up. Eventually I heard back- the bleeding had stopped. Mr G ended up doing ok— after a few days in the hospital. But it didn’t matter. Although at that early/middle point of my career I had seen plenty of tragedy (as readers of this stack know), I had never directly hurt anyone. I was distraught. I can remember the run we were on. Copper Mountain. Windsong, off the Timberline Chair. In case you’re wondering if doctors ever get over these things.
But the point of this story is not my struggle. My specialty has it easy— we do relatively minor (unless it’s you!) procedures that are generally low risk. It is to remind us that almost every physician, and certainly every surgeon, has hurt someone. Not from negligence, but because it happens. There is simply no getting around it.
Why share this story? There was no doubt then, and remains no doubt now, that the procedure was called for. I had done hundreds of them without a hitch. If I, or worse all doctors, stopped doing that procedure- lots of people would endure sickness or death unnecessarily.
See where I’m going here?
If someone is harmed by something that the medical community has deemed “safe and effective”— does that mean the medical community is wrong? Incompetent? Lying?
The truth is, a fair number of Americans would say “yes” to that question. That, of course, makes me very sad. A person that views every negative personal event as an indictment of the structure, and even the ideals of the system in which that event occurred is likely to be a very unhappy one, and a very unhealthy one as well.
Unfortunately, Americans don’t like grey zones. Either a thing is perfectly safe, or it is dangerous. But that’s not how risk works. Whether it involves a surgery, a medication, a vaccine— or a car ride. It’s complicated. We weigh the risk of doing something differently than we weigh the risk of not doing something. We also weigh the risk of something we choose to do ( like go skiing, or smoke, or drink) differently from things that we are told to do (wear a helmet, take a pill). I’ll have to come back to that in future posts.
Addendum:
Very recently, a young mom in our neighborhood faced a scary situation. Her 5 year old daughter had a severe infection that required surgery to address. Treatment with antibiotics alone was unlikely to help her. The surgery was considered “low risk” but certainly not zero.
Here’s the thing. That same mom had recently undergone a very low risk orthopedic procedure, done by an extremely experienced surgeon, that had gone wrong- left her with months of pain and disability. Imagine her distress! “Low risk” hits a lot differently when you’ve been that 1 in a thousand (or more).
She agreed to the surgery. It went well. The little one is on the mend. But damn, I admire that mom.


The definitive answer to how a human contracted HIV and it progressed to AIDS was never definitive enough for far too many people, including healthcare professionals when I was a resident. A friend of mine had a niece who was a physician's assistant, and she was very much opposed to her applying for a position with an out-patient clinic that served HIV/AIDS patients not far from the hospital where I worked. She knew well my first year experience was almost exclusively on the AIDS unit and the Emergency Dept., and I told her of the abundance of safety precautions, etc. When she had enough, she just abruptly got up and walked out on me, stating what I imagined summarized the "risk analysis" of many: "She's not your niece."
Bill, this gets at something consent forms almost never solve: “low risk” changes tense after the injury.
Before the procedure, it means the choice is reasonable. After the complication, it can sound to the person harmed like: your pain was inside the acceptable margin. Same statistic, different room.
That is why these stories matter. They keep the complication from becoming either proof that the system is corrupt or proof that the patient should simply accept it. Sometimes the procedure was indicated, the clinician was competent, the consent was real, and someone was still hurt.
The detail of remembering the ski run stayed with me. Patients rarely see that part. The complication enters the chart, but it also enters the doctor's memory, with a place, a weather, a phone call, a sentence.
The cleanest language may be the least defensive: this was reasonable medicine, and it hurt you anyway. Both are true. Neither makes the other disappear.