The Wall Street Journal recently contacted me regarding an upcoming article on Sedasys, the new gadget that is supposed to be able to infuse propofol by computer while monitoring vital signs.
If you’ve read anything I’ve written previously, you’ll know that I am NOT a big proponent of technology as a means of “improving” patient care. To me, the more technology you put between the patient and the caregiver, the less medicine you’re practicing, and the more data-entry and computer programming you’re doing.
Sedasys is designed specifically to administer propofol. Propofol is a milk-like substance that produces a range of effects from sedation to general anesthesia. For sedation you just use less, for general anesthesia you use more. Its very quick onset and very quick recovery make it great for outpatient sedation. It has to be given in a continuous drip because its effect goes away so fast. GI docs love it because its so effective. I suspect they also love it because propofol comes with an anesthesiologist to give it.
The only problem is the one Michael Jackson encountered: it has this pesky side effect of causing you to stop breathing. And you can’t tell by looking at a person how much will sedate them and how much will make them stop breathing.
A little old lady with a million health problems might sedate at, say, 40 mg and stop breathing at 60 mg, while an 19-year-old could probably take 150 mg and still be fighting you. It’s not necessarily weight-based.
What nurses usually use is a combination of midazolam, which is like an IV version of valium, and fentanyl, a short-acting versionof morphine. Either one of those can cause you to stop breathing too, but especially in combination. Necessary doses of these are extremely variable as well. People who drink or who are super anxious might still be talking to you at 8 mg while the little old lady snores with 0.5 mg. A lot of the cases anesthesia gets are those in which this sort of sedation has been tried and either the patient was too sleepy or, more likely, not sleepy enough. Nurses use these in a weight-based algorithm with upper limits for dosage. People getting this type of sedation generally take longer to wake up because the drugs are metabolized more slowly.
In both sedation methods, a real human is generally required to titrate dosages based on each individual patient. There’s a bit of an art to it. I don’t think you need four years of med school and four years of residency to give either type. You kinda need to be human though.
The tricky part about the specific procedures Sedasys is being proposed for are that the level of discomfort or what we call “stimulation” is not static over the length of the case. In a colonoscopy, the initial introduction of the scope is stimulating, and sometimes getting around the bend toward the cecum is as well.
You often need a little extra sedation for those parts, but can lighten the sedation for the rest of it, which doesn’t hurt at all. An esophogastroduodenoscopy (EGD) is even trickier, because you’re “sharing the airway” with a big tube, and most people, even asleep, will object to having the big tube shoved down their throats. You need a lot of sedation for that part, but once the tube is in it doesn’t bother people at all. Plus, every doctor is different in their speed and technique. It is hard to see how a computer can monitor all those factors. Again, you kinda need to be warm-blooded.
As to the study that was cited by the makers of Sedasys as “pivotal” proof of the machines effectiveness, I have two main problems.
First, they compared Sedasys giving propofol with nurses giving midazolam and fentanyl. Of COURSE the doctors like the Sedasys better – they were getting propofol, which GI doctors like because it works so well. Of COURSE the recovery time is quicker – that’s the nature of propofol. You can’t compare the two sedation methods this way. Now if you had a Sedasys propofol vs anesthesiologist propofol study, then we could talk.
Secondly, the study’s primary end-point was oxygen desaturation. Now, oxygen saturation is measured by a little clip that goes on the patients finger. It measures the percent of hemoglobin bound with oxygen.
Most people walking around have oxygen saturations of 97-100%. When you stop breathing or obstruct your airway in some way, your oxygen saturation obviously goes down. A healthy person can maintain a saturation of 100% even when not breathing for 2-3 minutes.
But here’s the tricky part: the oxygen saturation machine reading lags behind the actual oxygen saturation by 10-20 seconds. So, for a healthy patient, oxygen desaturation is a VERY LATE sign that something is not right. 2 to 3 1/2 minutes late. Not the most awesome endpoint to be using.
Let’s just say I don’t think anesthesiologists need to hone their late-night comedy skills any time soon.
Shirie Leng, MD is an anesthesiologist in the Boston area. She blogs regularly at medicine for real.