One of the benefits, if you can call it that, of being in the United States is the ability to see previews of the upcoming episodes on American television, something we don't get in Canada. So I wasn't terribly surprised to learn that ER was doing an episode on residents and sleep depravation. What did surprise me, however, was how long it took them to do it -- the lack of sleep residents get and the dangers associated with it are very well known and fairly well documented, and all we've seen on ER so far is passing references to how little sleep someone got. This was obvious in the pilot episode in the conversation Mark and Jenn have in the cafeteria, but it's sorta become something of a "background" issue for the producers and writers.

Sleep is important. It turns out there's a biological need for sleep, and if you don't get enough, you'll die. One study deprived rats of sleep for two weeks, and sure enough, they died. In humans (who don't usually go two weeks without sleep), sleep depravation results in weakened immune responses, head and body aches, muscle stiffness, changes in mood, and, perhaps most obviously, impaired thinking. If you want more information about sleep and what happens to us when we don't get enough, I strongly recommend "Sleep Thieves" by Dr. Stanley Coren, who has done a whole pile of research into sleep depravation and the need for it.

That said, Elizabeth's basic mistake was giving her patient 50% magnesium instead of 5% for asthma. Magnesium is in a lot of the asthma treatment guidelines as an adjunct to prevent bronchospasm, but it's important to note that the therapeutic benefit that was noted in retrospective studies has never been reproduced in prospective ones. What that means is that there is no good evidence that magnesium has any benefit in treating asthma, but a lot of people give it anyway. As a result, the patient arrested, with all the usual complications. Because his rhythm was pulseless electrical activity (PEA), they couldn't shock him, instead having to rely on external pacing and atropine to get the heart started again. Calcium gluconate was also administered to reverse the effects of the magnesium; this is standard treatment for a condition that can occur naturally as well (hypermagnesmia). This happened; it's a mostly successful resuscitation, but I think it's a good example of why physicians on-call should be given more opportunities for sleep and rest.

I am of the opinion that all pediatric patients undergoing painful procedures, whether they be in the emergency department or elsewhere should be sedated or otherwise given some kind of analgesia. My belief in analgesia is very strong and comes from a fundamental belief in the goal of medicine: we have the tools available to relieve the pain, so why not use them? I don't know I'd go as far as Doug's use of general anesthesia in the emergency department (you'll have to forgive me, as I only saw the episode once and didn't pay too much attention; I don't remember exactly what agent he used), but I would certainly consider the administration of painkilling drugs to any patient undergoing a painful procedure to be well-justified. (To quote from an episode of Trauma: Life in the ER, "You give a guy analgesia if you're going to do that. . . I mean, this isn't the Civil War or anything; we don't tell people to gnaw on a piece of wood.")

Kerry can get excited about the successful resuscitation of her first study patient, and good for her (and the patient, lest we forget what the real focus of medicine is). But one successful case does not a standard of care make. Last week, I noted what I thought to be the fundamental problem with her hypothesis (recapped, I don't see how something that has a rotten outcome to begin with is going to have a better outcome when it's done in an even more invasive manner), and this week we're starting to see some of the holes in her methodology. Did you notice how she said the surgeons got the sternal saw every other day? If this statement means what I think it does, patients will either get midline or lateral thoracotomies on alternative days in the emergency department -- it looks like random assignment (remember Doug's PCA study?), which is crucial to determine whether there's an active effect here, but it's really not.

Consider the following real-world example: In 1994, Bickell et al did a study on fluid resuscitation. The standard of care, up to that point, was that patients who had suffered a traumatic injury and were hypovolemic should receive fluids (notably isotonic saline) in the field and be aggressively managed in the emergency department upon arrival. Bickell theorized that by restoring vascular volume before you patch the holes in the body is inviting a whole host of problems, including the loss of red and white blood cells and clotting factors. So he did a study in Houston wherein some patients received fluids in the field prior to their arrival, and others didn't get any fluids until they'd made it to the operating room and they'd had all the bleeding stopped. Patients were enrolled in this study on an alternating day basis. In the end, the study concluded that there was some benefit to withholding fluids, although it cautioned that these results couldn't necessarily be extrapolated to other environments.

It was a very good study -- prospective, randomized, controlled. The problem is that things change on an every-other-day basis. If your patient comes in on a day where they're getting fluids in the field, you're likely to do things a bit differently than if they haven't received anything at all -- they may wait a bit longer in the ED before being shipped off for surgery, interventions may be delayed in favor of diagnostic procedures, that sort of thing. One of the things you're always looking for in research are confounds, and the criteria for randomization is one of them in Bickell's study.

(A note to the more technical reader and to my colleagues: I don't know how much of an effect the confound I just mentioned had on the study. My personal take on it is that the results are probably valid *if* you have an EMS system like the one in Houston, where you are never more than fifteen minutes away from a major trauma center. Delayed resuscitation has its place, but my personal take on it is that the idea will fall apart when faced with my trauma patients, who tend to be elderly and hit by cars, not young and shot on the street.)

The hazards here are obvious: if you know when you're going to have access to the sternal saw, you may be tempted to be a bit more aggressive in your approach to the patient; the reverse is also true if you know you won't have access to it. Eliminating this problem is tricky, but it can be done -- keep the saw in the department full-time, and randomly decide on arrival whether the patient is going into the experimental condition or the control group -- flip a coin or something like that. It's a bit more work, but ensures you're not open to accusations of bias from the alternate day "randomization." All this puts a lot of weight on a single statement by Dr. Weaver, which may or may not be valid. In the event I'm wrong, I like my hats with dark yellow mustard and relish. Felt fedoras, please.

My last word this week is going to be on Mark, who desperately needs to lighten up about the whole spine board thing. I can appreciate why he might not like it very much (it can be uncomfortable and, if you're claustrophobic, an unnerving experience), but to stalk off like that shows a serious need to mellow out. It isn't that bad, at least from my point of view, and he should thank his lucky stars they didn't stick him on the board and then leave him at the bus stop down the street.

You'd probably be better off not asking where that came from. Trust me.