Kinda critically ill patients seemed much less common in medical school and to a lesser extent in residency. Not that they didn't exist, of course; they did, but it was easier — and the standard of care — to admit almost anyone who was borderline. (Sorry, internal medicine residents!) This was probably because of a few factors: training in New York with a hugely uninsured population with low health literacy; the acuity of the patients overall; the ease of admitting a patient; finding a diagnosis and knowing, “Oh, yeah, that gets admitted.”
It's tougher in some ways now as an attending working in a health system where almost everyone has a primary care doctor. Do I admit the young “probably fine” patient? Or do I discharge them with close follow-up? I swear, I'm seeing much more pathology that isn't so cut-and-dried now that I'm the one calling the shots.
Let's take cholangitis. I was taught Charcot's triad and Reynold's pentad, and that these patients are Sick. They go to the unit. They do not do well. Even Rosen's mentions that “sepsis is a common complication and may be heralded by tachycardia, tachypnea, and frank hypotension…. Treatment of cholangitis includes hemodynamic stabilization with crystalloid fluid and, if necessary, vasopressors.”
But I can't be the only one who has patients with clear-cut cholangitis — fever, a robust white count, imaging and labs suggesting common bile duct obstruction — who look great. They get their antibiotics, some fluids, and the next morning, their ERCP snatches the stone, and voilà, home they go. What about the huge mortality? The sepsis? Where's the altered mental status? Where's the hypotension?
I'm not saying that individual patients can't be sick as hell; of course, they can, but overall, it seems like critical illness is becoming less and less … prevalent.
Our patients are changing on us because we're changing on them.
Better Techniques: We've come up with — and continue to come up with — better techniques for managing diseases. ERCP has made it possible to access the common bile duct less invasively. That and improved diagnostics have dropped its mortality from about 50 percent to less than 10 percent, depending on the study you read. The Seldinger technique improved central venous access from cutdowns; laparoscopes allowed much smaller incisions to be made to perform common, open surgical procedures less invasively. You can now get a portable pulse oximetry device for less than $100 or even just by using the camera on your smartphone.
Better Diagnostics: We've also added better, faster diagnostic studies to our workflow that help in at least two ways. They allow us to find potential disease earlier, and perhaps we're catching it before it becomes more severe. I can't think of better examples than the urine pregnancy test or the bedside ultrasound. Within a few minutes, I can rule out ectopic pregnancy, and I can differentiate the kidney stone from the ruptured or leaking AAA. Last week, a colleague found a cardiac tamponade; last year, a colleague found an atraumatic positive FAST: spontaneous renal artery rupture. Neither were “ultrasound gurus.”
These improved diagnostics also allow us to confirm the diagnosis we suspect. We're all taught to fear the big, bad pulmonary embolism, but how much of this was based on confirmed PEs? “Pleuritic chest pain” and “right heart strain on EKG” can be lots of things. Newer, better diagnostics can make sure that we're dealing with a PE instead of a lobar collapse, atypical pneumonia, or pleural effusion.
Better Data: Because PE is one of everyone's favorite boogeymen, let me posit this: we will have more reliable data in a few years on which patients with pulmonary emboli can be safely discharged. And we will be discharging them. (Right now there are some data, but I'm not sold just yet.) By studying patients, we can learn more about who does well and who doesn't, and who needs a workup and who needs reassurance.
One can similarly look at pediatric head injuries and now make a much more confident assessment of which patients to image — and which not to. Thanks to Nate Kuppermann's monumental PECARN study and a number of studies of radiation research, we can much more easily identify the rare child who needs a head CT, and educate parents about the vast majority who don't.
Better Screening: Better data beget better screening. It's obviously not cost-effective to screen all older men for AAA, but with knowledge of risk factors for certain diseases, we can target populations to screen for disease. When we find it, we can then do risk modification, stratification, and prognosis, and manage it faster. Imagine your workup of the nonspecific belly pain with normal vitals with and without “6 cm abdominal aortic aneurysm” in the patient's problem list or past medical history.
Better Understanding: Our treatments improve as our medical knowledge and understanding of disease processes grow. The classic case is Barry Marshall, drinking his own cultured concoction of H. pylori, giving himself severe gastritis to help prove its role in peptic ulcer disease. What was once treated with invasive vagotomies and valium for stress is now treated with antibiotics and acid blockade.
This is still happening. Thrombotic thrombocytopenic purpura was a rare and mysterious disease. We now know the role of ADAMTS13 in its pathophysiology (and more importantly for us in emergency medicine, better knowledge of its treatment and management).
Better, better, better. Next month I'll talk about a few other areas of betterment, and quantify our improvement since we started measuring.
Click and Connect!Access the links in EMN by reading this issue on our website or in our iPad app, both available onwww.EM-News.com.
Copyright © 2013 Wolters Kluwer Health, Inc. All rights reserved.
- Use Dr. Walker's medical calculator at www.mdcalc.com and his number-needed-to-treat tool at www.thennt.com. Follow him on Twitter.com/grahamwalker for daily thoughts on emergency medicine.
- Read all of Dr. Walker's past columns at http://bit.ly/WalkerEmergentology.
- Comments about this article? Write to EMN at email@example.com.