death


I admitted a patient from the ER one night over the holidays. The ER called with a CT scan showing diffuse pneumatosis, and the most obvious portal venous air I’ve seen so far. The patient himself looked far better than the scan, and was amazingly comfortable, considering that he had a heart rate of 140 and was already in acute renal failure. He was so comfortable that it was very difficult to persuade either my attending, or the patient, that he needed emergency surgery. (“Pain out of proportion to exam, Alice. You can’t tell me he has ischemic bowel and no pain.” I insisted, so we didn’t really lose any time, but it was a little disconcerting.)

As for the patient, that was the worst conversation I have ever had to have. Telling a family that someone died is easier. Telling a man who’s chatting happily that he’s almost certainly going to be dead within 24 hours is nearly impossible, either to find the words, or to convince the patient. I had to not only convince him that matters were this serious, but also discuss the option of surgery – his only chance of survival, but a very slim one, with a significant chance of a long ICU stay and major morbidities, if he did survive. (Some might say that with that CT scan, we shouldn’t operate. 1) You can see pneumatosis and portal venous air from a bad bowel obstruction, which can be salvageable. 2) He was relatively young, and with few comorbidities. We never did figure out what caused his ischemia.)

In between talking to him, I was calling the chief and the attending and the OR and the ICU, getting iv fluids and antibiotics running, and moving him to preop holding. Not much time. No sooner had I settled him in preop, with a nurse to watch, and the attending about to walk in, than the trauma pager started going off with multiple gunshot wounds, so I had to leave him. Three hours and several traumas later, I found him and the chief resident in the ICU. The operation had been completely unsuccessful; there was absolutely nothing to be done. His body was shutting down, and there was barely time to have the family at the bedside before he died.

I felt awful afterwards. Not just because it was the holidays, and we had lost a previously healthy man suddenly, but because I had spent half an hour talking to him about his death, and had never talked about what would happen to him after death. I had watched somebody dying, and had never even mentioned God or heaven or hell. Which meant that I did him exactly no good at all. He died, as I knew he would, and had to face eternity, and I hadn’t even mentioned it.

Yesterday one of the PACU nurses came up to me. (At night PACU and preop are staffed by the same nurses.) “Remember that man with the ischemic bowel who died? I went to the funeral home. I had to tell his family something he told me that night. He said, ‘I’m not worried about this, because I’m putting it in God’s hands. He took care of me when I had surgery 30 years ago, and he’s taking care of me now. If he wants me to live, I will; and if not, it’s all right. If I don’t make it through surgery, tell my family I’ll see them in heaven.’ ” I started crying in the middle of PACU. He’s safe, after all. I didn’t do anything I should have, but he knew better than me. Next time, I won’t make the same mistake.

(As for the nurse, I have a whole new respect for her, going out of her way to comfort not only the family, but also the other caregivers.)

In one sense it was a bad night, because I didn’t get much sleep, spent a lot of time in the ER, and had my patient die anyway. On the other hand, I surprised myself by handling the problems well. I probably shouldn’t talk about it too much, being just starting into a month of night float, with so many upcoming opportunities to make mistakes and act idiotically. But . . . I talked on here a lot, especially a year or two ago, about how much I admired the senior surgery residents. . . how they took control of bad situations, and knew what to do, and stayed calm. I thought I would never be like them. Much to my astonishment, and I have no idea when or how it happened, I found myself acting like them last night.
 
It was one of those messy situations, where the ER knows the patient being flown in needs surgery of some kind, but the diagnosis is unclear till the patient can be seen and have a CT scan. At least this time they called the surgical services ahead of time; perhaps after the series of fiascos last week our attendings yelled at them enough to impress the importance of calling ahead for ruptured AAAs and such like. (Not that the ER got around to informing the surgery residents, but my radar is getting pretty good.) Which meant that I, also bearing past miscommunications in mind, called the OR ahead and had them getting ready. So everyone was in the right place when the patient arrived.
 
He looked deceptively good for about 30 seconds, and then fell apart. Maybe it was a little longer, because I had time to at least make sure to my own satisfaction that he belonged to my service and no one else’s, so he was mine. Then it was the usual chaos of trying to intubate and do CPR all at once, get iv access, get monitors on, get blood and fluids lined up. . . The ER attending was technically responsible, because we were in the ER, and the patient hadn’t been officially diagnosed yet (I was just extrapolating freehand, taking the most pessimistic interpretation of the available data); but I was responsible too, because I knew if we could stabilize him, my attending needed to operate on him; and if you’re a surgeon and you’re present, you can never blame anyone else for anything that happens. The ER nurses and attendings knew he was ours, so they kept looking to me for directions. Gratifying, but scary.
 
Also I got a central line so fast, despite my hands shaking, that I was almost too surprised to finish threading the catheter. Now the ER folks think I’m magic, which is fine, I guess.
 
I’m not saying I have the calm part completely down; I was certainly pacing back and forth, and – not quite wringing my hands, but touching everything, the ivs, the bags of blood, checking for pulses repetitively. But I didn’t change orders and contradict myself, or give orders too often to be meaningful, and my voice wasn’t squeaking.
 
In the end we stopped. Despite occasional fleeting spontaneous pulses, we weren’t getting anywhere very encouraging with CPR, and it was clear that the patient was never going to be stable enough to move to the OR, let alone even start an operation on, which made further efforts futile.
 
The only thing I really still want to fix is, my voice keeps breaking when I talk to family members. It’s bad enough getting a midnight phone call to say your loved one is dead or dying, you would think the doctor should at least be able to speak coherently and audibly. It doesn’t really do much good for me to call people to give them bad news, and then have my voice be too shaky to communicate anything except that something bad is going on, leaving them to imagine for themselves what that must be. . . I know, sympathy and emotion from the doctor are good. . . but when you don’t know each other from Adam, probably simple transfer of information would be more valuable.
 
I especially hate calling 80-something wives – widows – who you know are home alone in the middle of the night. They’re half deaf, and sleepy, and don’t want to hear that their husband is dead. . . and when they do hear you, you can hear them just about collapsing. . . but so often the wife is the only phone number listed, and if you want to reach the adult children you have to go through her. . . . and you can’t just not tell people, and hope the morning will make it better. . . I have no idea how, but the ER social worker does miracles. She discovered the pastor and sent him over to keep her company. Now that was probably the most useful action of the night (and another profession to add to my list of people besides doctors who don’t get to sleep at night).

One of my friends, an intern, is struggling with the belief that they killed their patient.

I’ve thought that more than once, and in cold reflection I believe it to be true in at least one and two halves. That is, one I’m personally responsible for, and about two others I’m definitely responsible for significant failings. There were several other times that I felt very guilty about for a week, but as time passes I think my responsibility is less weighty in those. I haven’t written about them before because, in close temporal proximity, I was too upset to write, and I didn’t want any time correlation for the lawyers to find.

The one patient that I think of particularly, I personally failed to notice something, and that thing being overlooked led to another thing, and the complications of that other thing led to the patient dying. So it’s not like I directly administered an overdose. But it seems reasonably certain that if I hadn’t overlooked that particular thing, the patient would have been much more likely to survive. Also there were several other doctors, both residents and attendings, from my own and other services, who also had cause to notice that particular thing and act on it, and none of them did, either. But it was my patient, on my service; so I can’t decrease my own fault by saying that others, who were not as directly responsible, although more senior, made the same mistake. It comes down to, my lack of attention led to the patient’s death.

The other times are similar: I didn’t do anything – I didn’t cut a major artery, or cause a laparoscopic injury – I’m sure those are down the road – but I failed to pay close enough attention, or to pay attention soon enough, and then the patient died. If I had done a better job – if I had done well the job that I was supposed to be doing – it probably wouldn’t have ended the way it did.

I don’t know what to tell my friend, though. There’s no way around it. Sometimes I’ve tried to reason with interns (because they’re the ones to whom it happens for the first time; for the rest of us, the feeling of guilt is familiar and feared) and tell them, in this case it wasn’t their fault. But inevitably there comes a time, probably before six months are out, when there is no honest way to reason out of it: it really is my fault – your fault – our fault.

I can’t remember now how I dealt with those times. By not thinking about it, I suppose. I considered the facts enough to realize what I had done, maybe asked a senior resident what they thought about it, and then I closed a door in my mind. I think the phrase is from King Lear: “That way madness lies.” Now, one part of me knows I’ve killed people, and the rest of me is for all practical purposes unaware of that fact. It takes time, though, to get that door closed, and to keep it closed. And so for a week or two, it’s quite miserable. M&M helps a little, to have it out in the open. The attendings’ conclusion, surprisingly enough, has rarely been as harsh as my own. After all, only I know exactly when I knew certain things, and exactly what conclusions I drew from them, and whether I could have taken certain actions sooner than I did. The final picture, in public, is always a bit blurry; the blame never settles very definitely. Inside my own mind, though, I know that I failed – and it will happen again, no matter how careful I am; it will happen again. . .

It’s hard to watch interns learning that.

 Night shift is like a nonstop final exam. Remember how waiting for the test score was sometimes harder than studying for the test? Nights is a series of problem-solving exercises, where you have to come up with your best explanation and plan, then leave the building. You come back twelve hours later, and like it or not, the answer is up in public view. The rest of the residents and attendings on that service have had all day to think about it, and the official position is out: you got it right, or you missed this or that diagnosis or test or medication, and everyone knows.
 
I need a handbook, something like “Medical Spanish for Dummies,” maybe “How to Break Bad News in Three Easy Steps.” Last night was the worst test ever: a CT scan so bad I had to look at it three times before I completely realized how bad it was (and then radiology was overwhelmed, and perhaps felt I’d used up my quota of over-the-phone consults, and couldn’t read it for me till two hours after I needed it). After a few bad experiences early in the year, there’s a couple of conversations I try to avoid having with patients: being the first one to tell them they have cancer, especially as a consultant; giving bad news in the middle of the night; giving bad news without a family member available for support. So I looked at the CT another three times, to see if I could get out of it, and I couldn’t. How do you tell someone, You’re going to die within the next few days; I could try to stop it, but you really don’t want me to. And then, in the textbook scenarios, the patient is supposed to have something to say to that: questions, denial, grief – something. When they don’t say anything except, OK – you can’t even really try to comfort them, because there’s nothing left to say.
 
As if that wasn’t bad enough, then I felt obliged to call their family and explain the momentous decision we’d made. No one answered the phone, so I thought I had escaped at least that difficult conversation. Then, ten minutes before the end of my shift, the family got my message and called back; so I did have to tell them. I could have deferred it to the primary service (we were just consultants), or to the daylight team that I had already signed out to, but although I try not to be the one giving bad news (I think I’m still too junior to be the one making life and death pronouncements), I despise doctors who dodge their responsibility, and let days go by without telling patients and families the bad news that the medical team already knows. I was the one who’d read the scan, talked to the primary service and my attending, and had the discussion with the patient. So I talked to the family, on the phone (even worse than in person; another rule from medical school – don’t give bad news over the phone), stammering and repeating myself and hiding in a forest of medical details. They understood me, though; the only question was, how long do we have?
 
(And how do you answer that question, anyway? I’m in the business of trying to keep people alive. I’m not really familiar with how things go when we decide to give up. All I could do was make a guess, and warn them that I could be off by several days in either direction.)
 
So then I had to go home, and try to sleep, and wait to come back in the evening and find out –if the radiology attending agreed with our preliminary reading of the scan (what if I had made all these dramatic statements, and been wrong on the diagnosis?); if the surgery attending agreed with my assessment of how bad the prognosis was; if discussion with the family in the light of day changed the decision about whether to intervene or not. I couldn’t decide whether to wish that I had been flamingly, humiliatingly incorrect on all points, and the patient would do better than I thought, or that I was correct, with all that implied for patient.
 
I was right.
 
I don’t feel any better.

Medicine is full of superstitions. Anyone reading medical blogs has heard about not saying “quiet night” or “not busy,” for fear that the opposite will immediately happen; or that appys and AAAs and other things come in threes; or that patients who say they’re about to die probably are.

One of the less well-known has to do with specific hospital rooms: When you’ve seen something bad happen in one room, there’s a visceral reluctance to have another of your patients stay in the same room, especially soon afterwards. This is of course irrational, being that we’re in a hospital, and something bad has happened in every single room, more than once.

The floors aren’t too bad. The patient turnover is high, only a couple of days usually per patient per room, so most of the rooms don’t have strong associations with any one person; or if they do, there are several other patients also associated with that room.

But the ICU is different. We had a patient die last week, and a day later another patient, of the same attending, from the same demographic group, was put in the room. The attending frankly begged the charge nurse to change the room, and the chief and I were murmuring agreement. We compromised on transferring the patient out of the ICU within twelve hours.

Our surgical ICU has a little more than two dozen rooms, and as I walk around it in my mind’s eye, I can picture a tragic death in one room after another: the trauma patient from June in that room; the one who died within 24 hours of surgery in that room; another trauma patient here, this one from August; and this last patient in another room. It’s not so much deaths in general that bother me, but the ones that were preventable, the people who weren’t expected or “supposed” to die when they came in; the ones about whom I think that if only I had done something differently, they would have survived. Probably it’s not true, but what wouldn’t I give to be able to rewind the clock and find out.

There’s one room in particular that I don’t really look at at all. Six months later, I still try to ignore its existence; fortunately, I haven’t had a patient in there since it happened. This one wasn’t so much that I thought I could do something differently; but I spent a month fighting for that guy. Literally a month; he’d already been there a while when I came on service, and he died a day or two before the end of the rotation. I spent most of every day in or near his room; I did so many procedures, lines and chest tubes, that I could probably fill out all my procedure requirements just on him, if I’d thought to keep notes. He died anyway; it was pretty hopeless from the beginning, but we kept trying, watching every 5mmHg improvement or decline in his pO2 intensely.

By the end of residency, every room in the unit will have its own ghost. I’ll have to get over it.

Whatever white cloud I once had, it is definitely gone now. My chief is beginning to be over-satisfied with the number of consults I pick up when we’re on call.

Yesterday, the chief called me: “I’m about to scrub in for the next case, but I was just looking at a patient in the ER. He looks pretty sick; his white count is sky high, and his heart rate was at least 140. The ER says he has cholecystitis, but I don’t quite believe them. [our ER's ultrasounds are notoriously unreliable]  If he does, he should come to the OR next. Go figure out what exactly is going on, sort it out, and let me know.”

I was on the other end of the hospital from the ER, so I started walking. By the time I got to the ER, there was a certain electricity in the air – literally. The charge nurse looked up from the front desk. “Whoever you’re here for, they need some help in 5. They’re having a code.” 5 was where I had been heading.

The code went on for an hour, increasingly pessimistically, but it was impossible to stop till we were certain. Usually, in a code, the patient has no cardiac rhythm whatsoever – asystole; and in that case the protocol calls for compressions and drugs, but no defibrillation. This patient remained in vfib forever; and we defibrillated over and over – enough to satisfy any ER junky – to no effect.

Finally, after we called the code and the ER doctors had started trying to figure out contact information, I went up to the OR. “I’m afraid that patient coded. We did CPR for an hour, but we couldn’t get him back.” The chief cocked an eyebrow at me. “I told you to take care of things; I didn’t mean you had to do it that drastically.”

That was one code I found it impossible to feel too badly about. If the patient was sick enough to code in the ER, he would not have survived surgery. Moreover, you don’t develop cardiac arrest from acute cholecystitis, even a severe case. There must have been something else going on.

On the other hand, that makes three patients I’ve lost in the last week – more than in the previous six months together.

This is a problem. I’m studying for the ABSITE, trying to make up for time lost in December, which means reading through the review book every couple of days.

And every single chapter brings up vivid flashbacks of all my worst patients. I’ve only been doing this for a year and a half, and I’ve got quite a collection: patients whose death I feel (irrationally) responsible for; patients whose death I blame myself for even though I couldn’t have done anything different; disfigured faces and limbs; lives ruined by traumatic injuries; abused children; families crying in the ICU. Patients dying of a PE in front of my eyes; burn victims; pancreatic transections; vasculopaths; spinal cord injuries.

I don’t know how Bongi lives with his collection. For mine, for now, stop studying for the ABSITE (one week to go); but when it gets bigger, that won’t be enough.

All quiet on the transplant front again. Seems like as soon as I come near the service the operations disappear.

Which leaves time to study for the ABSITE, a good thing since the test is coming up in. . . 8 days. I got to the neurosurgery part of the review book, and had a flashback to my neurosurgery rotation.

It was far enough in to the month that I was holding the call pager by myself most days. I got called to see a lady in the ER. She’d had a headache for a few days, but that day it was much worse, and her son had finally forced her to come to the ER. Her history of severe, poorly controlled hypertension was a red flag, and the ER doctors got a CT scan. By the time it was done, and they had recognized the subarachnoid hemorrhage, her mental status was deteriorating to the point that, while still fairly alert, she could no longer answer questions coherently. I didn’t waste much time on exam, just verified that her pupils were still equal and reactive, and there were no other gross neurological deficits yet (neurosurgery physical exam is the most abrupt and pointed of any specialty), then called the resident. He concluded that her worsening symptoms were due to a still-active aneurysm, and arranged for her to be taken to radiology for an emergent cerebral angiogram and coiling of the aneurysm. I saw her off, then tried to tackle some of the other pages that had been accumulating (hypotensive post-op patient; tachypneic patient failing a vent wean; rising ICP; hyponatremia in a trauma patient).

About half an hour later I got a stat page from radiology: “Your patient just had a bradycardic arrest on the table, maybe you should come down here.” I asked the nurse to please page my resident as he was the only really useful person, and then ran down the stairs.

I arrived (with the resident soon on my heels) to find that the report was very slightly exaggerated. She hadn’t completely arrested, just become so bradycardic that there had been several 20-second pauses between heartbeats. That had improved with atropine, and she was now awake. So awake, in fact, that she was insisting on leaving AMA.

Which posed a problem, since her vital signs were a classic case of Cushing’s triad, found in impending brain stem herniation: bradycardia, hypertension, and slow respirations. Well, actually, she was breathing just fine, since she was loudly insisting that we let go of her, give her clothes back, and let her leave.

The resident announced that he needed to put in a ventriculostomy drain now – right there, in the middle of the angio suite. He started finding the supplies – some of which had to be brought down from the neuro ICU. I was left to deal with the matter of consent. The patient herself was very dramatically not consenting. By now, it was taking the efforts of two nurses to keep her lying down (which of course wasn’t doing any good for her blood pressure or her intracranial pressure, which was what we intended to relieve by placing a drain). Her son, whom we knew to be somewhere in the hospital, had disappeared: either he was trying to get a bite of lunch, or the move to the maze in the depths of the radiology department had lost him. So when the supplies were assembled, we decided to proceed with the drain as an emergent procedure – no consent required.

Despite all of us knowing quite well that the patient’s protests were further evidence of altered mental status and injury from the blood now surrounding her brain, it was no fun to perform an invasive procedure on patient who spent the entire time protesting that we were kidnapping and abusing her, and who had to be held down by several staff members. Once the drain was placed, we ended up intubating her right there as well, as her level of consciousness continued to decline.

So by the time her son caught up with us, in the neuro ICU, we had the job of explaining that his mother had gotten significantly worse, and was now on a ventilator.

It was all downhill from there. Everything that can go wrong with subarachnoids went wrong with her: her ICP stayed up despite all measures to lower it; she had surgery to remove the aneurysm, but with no improvement; she remained in persistent vasospasm, despite every single treatment in the book being tried; she developed diabetes insipidus (seen in brain injured-patients when the hypothalamus stops producing anti-diuretic hormone, needed for the kidney to concentrate urine). After two weeks in the ICU, she died. So the last her son saw of her, conscious, was in the ER; and the last time any of us had talked to her had been while we were wrestling with her in radiology.

That’s why I hate dealing with subarachnoid hemorrhages, and I could never imagine being a neurosurgeon. Within twenty minutes she went from a pleasant lady with a headache to being delirious, then intubated and critically ill; and nothing we did could help at all.

A few hours into a busy morning – the kind that always develops when, building on a string of slow days, I have a stack of journal articles to read and paperwork to do – I got a nonsense consult. Nonsense as in, all the surgery attendings in the hospital already knew about the patient, and had discussed her condition at length and leisurely among themselves. As a result of this consultation, spread over three days, they had decided that the one attending should officially consult the other attending. Which means his resident, that is, me, needed to go put an official note in the chart to let the poor medicine team which was babysitting this patient know that the surgical attendings have changed.

So my seeing the patient and writing a formal consult was going to contribute absolutely nothing to the patient’s care or to my team’s knowledge of her; but it had to be written.

It didn’t make me any more enthusiastic that the picture I got from the chart before going into the room was of a patient seeking pain medication. Sure, she had a couple genuine chronic conditions with biopsy documentation of their existence; but she was on a lot of narcotics, plus some valium thrown in. She had been on disability for years, even before this most recent, serious problem cropped up.

I was in for a surprise.

She was polite, pleasant, and a very coherent historian (first clue; real seekers try to muddy the waters). She was able to tell me all the studies that had been done, and gave me a timeline of her symptoms and the path to the final diagnosis.

I asked how long she’d been on disability, and all of a sudden she started talking. She’d been injured a few years before, but had kept busy up till last month taking care of her father, whose health had declined precipitously. Last month he died at home.

I didn’t have to say anything at all; she just wanted someone to listen. She told me about her mother’s poor health and inability to care for her husband, about how painful it was to watch her father getting continually worse. She told me about how he joined the army right after Pearl Harbor, flew several bombing missions, and was eventually interned in Switzerland, then came back to get married and start a family.

There was a lot more – his death had hit the family hard, and it sounded like the siblings weren’t relating to each other well now – but I wasn’t looking for holes in the story any more. No slacker takes disability, then works 24/7 caring for a dying parent. Most healthy people don’t do that much.

At the end of that talk, I understand why psychiatrists don’t believe in physical exam. After that much sharing, it’s rather anticlimactic to ask if you can listen to the patient’s lungs.

My patient is dying (again), and I can’t do anything to stop him.

That’s such a horrible feeling. I can’t help him. I can’t stop the disease, I can’t change anything anymore. It’s too late.

At that point, the thought arises, if I can’t cure him, at least maybe I could make this quicker, easier for him and his family.

I never thought I’d understand (dare I say sympathize with) that idea.

I understood today, finally, how doctors, whose purpose is to heal, can end up wanting to kill (because that’s what euthanasia is, in the final analysis). I wanted to do something, anything, for this man; and if I couldn’t fix him, that left only one thing.

The problem is that I’m not God. There’s a very old joke about the difference between God and the surgeons; and I think death is his way of reminding us humans of our place in the world. Death is not under our control. It’s not a thing that we can order around, or organize, or turn off and on at our whim.

Life and death belong to God. He gives life, and he controls its end. The time of death does not belong to us. That’s our human arrogance talking, to think we can control every aspect of our lives, right down to death itself.

So I had to let go. That man was God’s creation. God let me care for him for a while; but ultimately I and my colleagues were never the ones in control. As the psalmist says, “Man returns to dust, and his spirit returns to his Maker.”

Next Page »

Follow

Get every new post delivered to your Inbox.