New AAN Guideline Evaluates Bell Palsy Treatments

Read the new practice guideline

Oral steroids can improve the likelihood of full facial recovery in people with new-onset Bell palsy, according to “Evidence-based Guideline Update:  Steroids and Antivirals for Bell Palsy,” that was published electronically ahead of print on November 7, 2012, and appears in the November 27, 2012, issue of Neurology®. The efficacy of oral steroids is supported by well-designed, high-quality studies.

Efficacy of Antiviral Therapy Questionable

Antiviral therapy alone has not been shown in well-designed studies to increase the likelihood of full facial recovery. Physicians might offer antiviral drugs as an addition to oral steroid treatment, but they should inform their patients that a benefit from this drug combination has not been strongly demonstrated by well-designed studies. Patients also should be informed that if there is an added benefit of combination therapy even in severe cases, it will be marginal at best. 

Read the guideline and access PDF summaries for clinicians and patients, a slide presentation, and a clinical example. For more information, contact Julie Cox at jcox@aan.com or (612) 928-6069.

Usefulness of Test Depends on Clinician’s Judgment of Probability of sCJD Before Testing

Testing for 14-3-3 protein in spinal fluid may support the clinical diagnosis and other diagnostic tests used to diagnose sporadic Creutzfeldt-Jakob disease (sCJD) in patients who present with rapidly progressive dementia and are suspected of having sCJD. This is the primary finding in “Diagnostic Accuracy of CSF 14-3-3 Protein in Sporadic Creutzfeldt-Jakob Disease,” a new guideline from the AAN that was published electronically ahead of print on September 19, 2012, and appears in the October 2, 2012, print edition of Neurology^®. 

While the test may help when used in cases where doctors suspect sCJD may be present, the test is not accurate enough either to diagnose the disease or to rule out the disease with absolute certainty.

The usefulness of the 14-3-3 test will largely depend on a clinician’s judgment of the pretest probability of sCJD for a given patient. Such judgments will reasonably consider the rarity of sCJD (incidence 1 per million per year), the patient’s clinical presentation, and the results of already obtained ancillary tests such as brain MRI. However, how the test should be used in conjunction with EEG and MRI findings suggestive of sCJD needs further investigation. The authors contend that only physicians experienced in diagnosing dementia should determine whether the 14-3-3 protein test is needed and how results should be understood.

Read the guideline and access PDF summaries for clinicians and patients, a slide presentation, and a clinical example. For more information, contact Julie Cox at jcox@aan.com or (612) 928-6069.

Realizing that there is more than evidence supporting clinical decision making helps explain why randomized controlled trials (RCTs) are not always necessary for making all decisions. Many decisions can be made based solely on the base pillar—principles. These are the axioms of medicine that have been already established. Principles are the facts we know—the anatomy of the medulla, for example. They are often quite useful and sometimes support a clinical decision in and of themselves. Thus, when we see a patient with ipsilateral miosis, ptosis, decreased facial thermal sensation, and contralateral decreased body thermal sensation, we know with a high degree of certainty that the patient has a lesion in the lateral medulla—even if an MRI doesn’t show this. We know things about anatomy, physiology, pathology, and much more—principles.

Can we make a decision regarding the efficacy of parachutes on the sole basis of principles? Of course we can. We know jumping from a great height without a parachute is almost always lethal. It is an established fact that jumping out of an airplane with a parachute increases your chance of survival by a lot. An RCT is unnecessary and foolish in this situation because we already know the answer.

It is, of course, obvious in the parachute and the lateral medullary syndrome scenarios that principles are sufficient to support our decisions. However, it is not always that clear. Would you, for example, think it necessary to perform an RCT in patients with acute appendicitis which compares the efficacy of surgical appendectomy to that of medical therapy (i.e., antibiotics)? Many would say no—that we already know on the basis of principles that surgical appendectomy is the preferred treatment.

When trying to determine whether a decision can be supported completely by established principles, it is helpful to ask two questions. First, is there a plausible alternative inference from first principles? In the appendectomy scenario, the inference that removing the infected appendix will make the patient better makes sense. Does it not also make sense that treating the infected appendix with antibiotics will work? It is difficult to develop a compelling principle-based argument that appendectomy would automatically be better than antibiotics. By contrast, I can come up with no alternative principle-based arguments favoring the choice not to use a parachute. I doubt anyone could.

The second question to ask to determine whether principles are enough is, do reasonable people disagree? When asked, most clinicians I know would not favor an RCT to answer the appendicitis question—but some think it a reasonable idea. There is genuine disagreement. An alternative way of determining whether reasonable people disagree is to ask whether an institutional review board (IRB) would approve an RCT to answer the question. If, in fact, you find that there is a trial under way, you have your answer—an IRB somewhere approved that study. Reasonable people disagree about the best course of action in many situations. In these circumstances, it is a safe bet that the answer cannot be inferred from principles alone as it can with parachutes. I can find no reasonable person who would volunteer for the RCT regarding parachutes.

So what about antibiotics for appendicitis? An RCT comparing appendectomy to antibiotics actually was conducted.¹ Patients randomized to surgery or antibiotics both got better. Patients undergoing surgery had surgical complications that patients treated with antibiotics did not have. Patients receiving medical therapy were more likely to have a recurrent bout of appendicitis than patients undergoing surgical appendectomy. Which is better? It is not at all obvious from the evidence.

What about judgment—the other non-evidence-based pillar of clinical decision making? This is what we rely on when principles and evidence are not enough. Judgment involves making educated guesses. It is a large part of the art of medicine. When we do not know the answer, we have to make our best guess as to what is best for our patients. Guessing is necessary in many situations in medicine—some would say in most situations.

Knowing when we are guessing is a critical skill. It is not always easy to distinguish our seemingly clever, principle-based inferences from our informed opinions. I know some colleagues who genuinely believe that their opinions should be considered established principles of medicine. I suspect many readers of this entry know such colleagues too. The presence of alternative clever, principle-based inferences and genuine disagreements with other colleagues should tell them that they are guessing.

What of evidence? This is the middle, or central, pillar. It separates principles from judgment. It will be the major topic of many future entries in this blog. We will not consider it further for now except to say that, when discussing nuances of evidence such as bias and random error, it is easy to forget that clinical decision making is not just about evidence. We will strive not to become “radical protagonists” of “evidence-only” medicine. We will remember our parachutes.

¹Eriksson S, Granström L. Randomized controlled trial of appendicectomy versus antibiotic therapy for acute appendicitis. British Journal of Surgery 1995;82:166–169.

You may be familiar with an article published in 2003 in the British Medical Journal where the authors conducted a systematic review searching for randomized controlled trials (RCTs) that tested the efficacy of parachute use in preventing death and major trauma related to gravitational challenge.¹ Unsurprisingly, they failed to find a single RCT demonstrating the effectiveness of parachutes. On the basis of this absence of evidence they made the following tongue-in-cheek recommendation:

We think that everyone might benefit if the most radical protagonists of evidence based medicine organised and participated in a double blind, randomised, placebo controlled, crossover trial of the parachute.

Critics of evidence-based medicine (EBM) frequently reference this article. They believe it highlights the absurdity of always requiring an RCT in order to conclude that any intervention works—they are right. Sometimes that requirement is absurd.

When is an RCT unnecessary? The answer to this question will dispel a pervasive myth regarding EBM. A myth espoused by some of its most “radical protagonists.” A myth reinforced by the parachute article. That myth is that “evidence-based medicine” is actually “evidence-only medicine.”

If EBM is not “evidence-only” what else does it include? Two other things: principles and judgment. These two elements, taken together with the evidence, form three pillars that support clinical decision making. Each pillar stacked one atop the other—their relative importance dependent on the clinical circumstances.

The three pillars are more than just a useful metaphor. The concepts are fundamental. They map directly to identical concepts used in all disciplines employing the scientific method. Principles map to the established theoretical framework of what we know. Evidence maps to critical empirical inquiry. Judgment maps to hypothesis generation. Indeed, it is no exaggeration to state that EBM is the explicit application of the scientific method to clinical decision making.

¹Smith GCS, Pell JP. Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials. BMJ 2003;327:1459–1461.