How Much Risk Is Too Much Risk
Under The Knife?
I. Why Sane People Don’t Play Russian Roulette
Imagine this little scenario if you will. You’re sitting on a plane which, for the time being, remains parked at the gate as your fellow passengers board. You get yourself situated and start to look around. Let’s say you’re in Los Angeles and need to get to New York, which is maybe four hours by air. As you look about, you notice a small crack in your window. Maybe two inches long. Okay, you think, I can deal with that. The window is at least double-paned, after all. But you look past it and notice some corrosion as well, on the leading edge of the wing. And maybe a few popped rivets besides. More uncomfortable now? Do you bring it to someone’s attention? Do you get off the plane? How much risk are you willing to take for the convenience of getting to New York in a few hours, rather than a few days (by train or car)? What if you are only flying to San Fran, about sixty minutes? Can you abide the risk then? Does the benefit of getting to your destination quickly outweigh the risk of the plane falling out of the sky? What is the risk/benefit ratio?
Another scenario. You’re at home in the rural countryside and realize you forgot to pick up something for dinner. There’s nothing convenient in the kitchen. You know there’s a supermarket ten miles down the road. But it’s snowing. Hard. And you don’t have snow tires. Do you eat the few crumbs left in the kitchen or do you go out in the snow? How bad would the weather have to be for you not to take that risk? Would it matter if you had not eaten anything that day? Or for three days? What if you had only a motorcycle? Or a front wheel drive vehicle? Or a Hummer? Or even just a horse? Would you go if it was sunny? What if it wasn’t a supermarket, but a 5 star restaurant at the end of that snowy road? Would a greater benefit cause you to assume a greater risk?
In other words, how much risk is too much? At what point does the benefit outweigh whatever risk there is?
In each of these scenarios—in all of these scenarios—the risk, the benefit, or both, are subject to change. Most every decision we encounter in our daily lives has both some inherent risk and some tangible (or intangible) benefit. Judging the ratio of these two entities is what allows us to arrive at the decisions we make. We do this a hundred times a day without thinking about it.
Do you walk out to the curbside mailbox to check your mail? Tangible benefit and low risk—unless it’s raining. Or snowing. Or sleeting. If the ground is a sheet of ice, is the risk of falling and breaking your hip worth it? What if you’re waiting on a letter from a long lost friend? Or a job offer?
Do you change the light bulb over the sink? Can you reach it or do you have to stand on a ladder? Is there water in the sink? Is the ladder a sturdy one? Is there someone to hold it? Is the light just burnt out or actually shattered (the glass is gone, you have only the wire remnants in the socket)?
One more example: you want a drink of water. Benefit: your throat will be soothed and, ultimately, so will your kidneys. But what if you have to cross a wet and slippery floor to do so? You could trip and break your leg. Unlikely, of course, but it happens more often than you think. Still, maybe this one heavily favors getting up and going after a drink of water. The risk/benefit ratio if favorable—that is, the benefit far outweighs any risk involved. Maybe by a thousand or hundred thousand times. If we set the risk at 1, we can show the benefit like this: 1/1000 or 1/100,000. The ratio is a very small fraction. So we see that the smaller the ratio, the better or more favorable things are. We lean towards recouping the benefit. We get that drink.
But what if the risk is high and the benefit very low? Ever play russian roulette? I hope not. In that unfortunate parlor game, whether you load the the six-shooter with one or five bullets, the benefit is hard to see, about as low as you can imagine. For reference, let’s set the benefit at 1. The risk, death or something worse (permanent coma, for instance) might be 100,000 or 1,000,000. 1,000,000/1. So the higher the risk/benefit ratio, the worse things are and the less likely we are to pursue the cause. The revolver stays in the drawer or, better yet, the locked case.
II. This Might Be A Wash—Or It Might Not
So, risk/benefit ratios are a part of everyday thinking. If the risk/benefit is low (1/10 or 1/100), the equation favors the benefit and we act accordingly. If the risk/benefit is high (10/1 or 100/1), the opposite is true. Too much risk makes the benefit seem paltry and we don’t care to play. The six-shooter stays in the drawer.
But what if the risk/benefit ratio is not immediately obvious? For example, I like The Walking Dead. I know it’s going to give me nightmares—that’s the risk I take watching it—but it’s pure fun to couch-out and let my brain vege for an hour during each episode. The benefit is intangible in this case. I’m not sure what the risk/benefit is here, but for me that’s ok. I’ll take the chance on a nightmare.
We make these kinds of decisions all the time. Decisions in which the risks and benefits are approximately equal (or at least we think they are). Most of the time these are a wash. Do I drink the milk even though it smells kinda bad? Do I do the laundry now or later? Do I have spaghetti or tacos for dinner? Do we take a shower first and then screw, or vice versa? Do I have that walnut sized brain tumor operated on now, or do I wait and see if it’s going to grow before having it removed?
Did you say brain tumor?
Yes, I did.
I also asked what happens when the risk/benefit ratio is not immediately obvious. I said that most of the time this means the benefits are approximately equal. And that’s true. But it doesn’t mean the risks and benefits are small.
And that brings me to surgery—and to what I really want to talk about: the risks and benefits surrounding a given operation, which are often huge. Both the risks—and the benefits.
Welcome to my world.
So what happens when a patient comes in with a benign brain tumor? (Never mind how we know it’s benign, as in noncancerous and not immediately life-threatening or significantly threatening in any other way; that, perhaps, is a subject for another post). Obviously the risks of brain surgery are high (not, in general, as high as you might think, but high enough). But aren’t the benefits high too? Would not being rid of that tumor be pretty damn beneficial?
So, does one outweigh the other in this situation? What exactly are the risks of brain surgery? And what is the tangible benefit of having the tumor removed? And therein lies the rub, because quantifying these risks and benefits are not always as easy as you would think. In other words, sometimes it’s a wash.
Take the risks, for instance. The risks of brain surgery might include (this is not an exhaustive list) stroke, seizure, infection, hydrocephalus, subarachnoid hemorrhage, deep venous thrombosis, pulmonary embolus, heart attack, pneumonia, aphasia, mutism, incoordination, coma, death, etc., etc., etc. And each of these, take infection for instance, might include others, like superficial wound infection, skin flap sloughing, loss of the bone flap, epidural abscess, subdural empyema, meningitis, ventriculitis, cerebritis, intraparenchymal brain abscess, etc. To be fair, most of these things are rare (some quite rare), but their incidence is not zero so they should at least be acknowledged in the calculations.
Ok, that’s not really true. A digression. It can be shown that if a certain risk (or benefit) is so rare as to be statistically improbable in a given population, one really shouldn’t base their decision on that particular risk or benefit. One important example comes immediately to mind. In elective surgery outside the brain and heart, intra-operative deaths today are exceedingly rare and so the idea one might die during, say, a breast reduction, is not reasonable and should not be considered when deciding for or against surgery. Even with elective heart or brain surgery, this particular risk is often vanishingly small.
But, returning to our example, what about the benefits of having that tumor gone? Well, one benefit is you know it’s gone. There is that. But even under the best of circumstances (a benign tumor that is easily and totally resected by an experienced surgeon), most folks who have a brain tumor removed continue to get surveillance MRIs or CTs for years to come, looking for recurrence. Still, for most folks there is some peace of mind at having the ‘brain tumor’ out. How do you quantify peace of mind? If the tumor was at all painful, hopefully the pain is gone. In some situations, say a tumor of the fifth cranial nerve (never mind what that is), that pain may have been severe and that alone might really sway the risk/benefit ratio in the first place. But we aren’t talking about that here. This is a benign tumor, not immediately threatening to the patient, remember? We aren’t talking about something pushing on the optic nerves and threatening blindness either. That would really push the risk/benefit ratio down towards benefit.
So, the list of risks is long and torturous. The list of tangible benefits not so much (many, if not most, benign brain tumors take years to grow to any significant size before they might cause trouble). The intangible benefit—peace of mind—is something else altogether. Starting to get the picture?
High risk/(low or moderate tangible benefit + unquantifiable intangible benefit) = ?
Now, this might be a wash, or it might not. Very few neurosurgeons would operate an 85 year old man with a walnut-sized tumor they knew or strongly suspected to be benign pressing against the frontal lobe. But the same tumor in a 24 year old woman with her entire life ahead of her—including another 20+ years of fertility and hormonal fluctuations (which can affect even benign tumors): operate.
One more note. You can see from the example in the preceding paragraph how the risk/benefit ratio is often fleeting and always dependent on multiple factors, only some of which it is possible to take into account. Like age (younger often, but not always, favors surgery). Or the presence of co-morbidities, such as diabetes, smoking, end stage heart or liver disease, or morbid obesity, all conditions which always increase the risk, sometimes prohibitively so such that surgery is very unlikely to have a favorable outcome—in such instances the surgeon will decline to operate. Why operate when you know the result will be worse than what you started with?
But given all of that, what if the benefits of surgery are high and the risks are only small or moderate? Should we operate then? Should we always operate then?
III. The Bottom Line
We’ve seen that when the benefits are very high, saving a life for instance, or eliminating intolerable pain, or preventing paralysis or blindness, surgeons are willing to operate even when the risks are very high as well. This is because one can reasonably presume that the higher the benefit of surgery, the greater the downside of no surgery. Operating on a tumor against the brainstem (arguably the highest priced real estate in the body) is dangerous, but the alternative, the downside of not operating—a slow and agonizing decline in which the patient will be severely incapacitated long before they are anywhere near death)—is unthinkable. Or how about a tumor growing inside the heart of a young adult? The risks of surgery in such a condition are extraordinary, but the potential benefit is as well—regaining an entire lifetime. The downside of no surgery in this situation is no future at all, which, for most of us, is unacceptable.
In such circumstances, most surgeons—and indeed most patients—are willing to go towards heroic measures. Fortunately, such cases are the minority, even for neurosurgeons and cardiac surgeons.
But what if the benefits of surgery are high and the risks are only small or moderate? Should we operate then? Should we always operate then?
Human beings are fallible creatures. We make mistakes. Mistakes are intangible and can’t be foreseen. If they could be, we might reasonably be able to quantitate them on an individual level. But we can’t quantitate them in that way. What we can do is quantitate them on a population-based statistical level.
We know, simply based on statistics, that no matter how good a person is, if they do something enough times, they are going to make mistakes. That’s part of the definition of being human. Humans make mistakes. They make more mistakes early on, while learning something, a procedure or an operation, but even later, when they have perfected it, they make mistakes. Otherwise, we’d all be robots.
Ever watch a major league umpire call balls and strikes. They’ve seen thousands of pitches, but still only get it right most of the time.
So back to the question. What if the benefits of surgery are high and the risks are only small or moderate? Should we operate then? Should we always operate then?
Consider the individual with true sciatica, pain down the leg, and the usual cause of true sciatica, a pinched nerve in the lower back. This patient is in agony. There is an operation that can fix that, a lumbar microdiskectomy. It takes less than an hour, is done as an outpatient so our patient won’t even have to spend a night in the hospital, and the pain will be gone immediately (or very close to immediately). Neurosurgeons see these patients everyday. Now, assuming I could operate on everyone of these (I couldn’t, there isn’t enough time in all the world under the sun for that), should I?
The risk is minor. Wound infection, a rare diskitis (more serious infection), recurrent pinching, the rare nerve injury, a transient spinal fluid leak, and the small but not zero risks of being put under and then awakened again.
So, high benefit, low to moderate risk. What’s not to like? Why not operate?
Two reasons. First, neurosurgeons know most of these pinched nerves get better on their own in four to six weeks or so. Operating on them at two weeks means I’d be subjecting a lot of folks to unnecessary risks, no matter how small. One fundamental tenant of surgery is to never operate unless you can improve upon the natural history of the disease or condition. No matter how good a surgeon I or anybody else might be, we will leave a scar in our wake. While not usually a problem, it could be. If the condition resolves on its own, there’s no scar to deal with. This is why surgeons in general and neurosurgeons in particular prefer pain patients to have tried nonoperative (conservative) measures before operating. To give time for mother natural to cure the condition without surgery, which is usually a better deal for the patient—if the patient can wait (which may or may not be the case).
And second? Because humans are fallible creatures. That’s right, that’s what it comes down to in the end. Humans make mistakes. That’s why the pilot of your airplane uses printed checklists for just about every phase of flight, so he or she won’t forget something. Surgeons use checklists too. It’s called a timeout. We use it to make sure we have the correct patient, the correct side and part of the body for surgery (wouldn’t want to take out the wrong kidney or off the wrong leg—don’t laugh, it’s happened), a correct list of allergies, the correct instruments to do the case (don’t want to get halfway into something and realize somebody left the crucial hammer or bolt at home), etc.
But, unfortunately, there is no checklist for the meat of the case. And here’s the real clincher: you don’t even have to make a mistake. Humans are not stamped out as carbons of one another. Even though the anatomy books say it should be one way, it may actually be another way in a given patient. 5% of folks have an extra lumbar vertebrae, six instead of the usual five. Sometimes, two nerve roots exit together at the same level, rather then the rule of one nerve root per exit. Things like this mean a surgeon can do their best on any given day and still have a lousy outcome. A patient can have an unusual susceptibility to infection, or be an unrecognized bleeder, or have an unanticipated allergic reaction to a drug. Anesthesiologists live in fear of something called malignant hyperthermia, which can drive a patient’s temperature to 106 in a matter of minutes and kill him in a matter of hours—even though they did everything right.
I can do this operation 1,000 times, and if I am perfect 99% of the time ten patients will still suffer some type of adverse outcome.
These sorts of things are the stuff of nightmares—and malpractice suits.
Most elective operations, probably better than 95%, go well in this modern age. Probably 90% of the remainder, those in which something out of the ordinary does happen, don’t cause any permanent harm. But sometimes, even under the best of circumstances and with the best of intentions, shit happens.
So the bottom line is this. Don’t live in fear of surgery. If the benefits are high but the risks are just as great, much of the time surgery is reasonable and appropriate. Organ transplants, for instance, often fall into this category. If the benefit is marginal or small and the risks high, surgery is rarely offered.
But if the benefits are high and the risks only small or moderate, take a moment and reassess. Talk to your surgeon. Does the benefit really significantly cover the risk? If it does, and you feel the same, and you can accept the very small chance of some adversity creeping in where it isn’t wanted, then go for it.
Otherwise, understand it’s a wash and wait it out.