First seizure management: I can see clearly now?

I n the era of evidence-based medicine the need for guidelines should be diminishing. After all, if the evidence is so clear, why would one expect clinicians to have trouble interpreting and applying it? In reality, the evidence is not so clear. Most evidence is acquired using imperfect tools from populations who have varying degrees of similarity to the specific patient sitting in front of each of us every day. Even the best clinical trials yield subtly flawed evidence, as each requires interpretation of inclusion and exclusion criteria at the time of enrollment and ascertainment of endpoints that are imperfectly reported (e.g., seizure number, adverse events, compliance). Trials are also subject to important population stratification driven by inclusion and exclusion criteria, proximity to centers conducting trials, persuasiveness of investigators in recruiting participants, and receptivity of specific patient subgroups to enrolling and completing trials. Important subgroups of patients are frequently excluded from trials, e.g., women not committed to avoiding pregnancy; patients with a history of suicidal ideation, cardiac disease, hepatic disease, renal disease, or progressive conditions that might occlude trial endpoints; and, perhaps most importantly, children and the elderly. Finally, clinical trialists build academic careers and may earn significant fractions of their income based on their success in meeting recruitment goals and completing trials, likely leading to subtle biases in the day-today decisions that ultimately produce evidence. While blinding and randomization reduce the effect of these issues, they may not eliminate them completely. Furthermore, most epidemiologic and natural history studies are neither blinded nor randomized, and many are not prospective. Not only is evidence flawed, it is often complicated and sometimes contradictory. The neurology community has proactively designed thoughtful processes and standards for the development of guidelines by committees of vetted experts screened for major conflicts of interest and nominated (or self-appointed) on the basis of interest and availability. These panels typically spend tremendous effort sifting through literature and applying their own experience to develop consensus statements describing what they judge to be best practice, usually presented with a statement of strength of the evidence. Guidelines are promulgated with the assumption that they represent a rational reduction of the evidence into digestible bits that can be easily accessed by practicing physicians, and they are typically wrapped in disclaimers denying their generalizability, authority, timeliness, and inclusivity. In a recently published article in Neurology, Krumholz et al. present a guideline on management of an unprovoked first seizure in adults. This guideline specifically addresses questions regarding the risk of recurrence in patients who are treated or untreated after a first event; the likelihood of remission, defined as a prolonged period of seizure freedom regardless of treatment, in patients who are treated early vs late; and the occurrence of side effects of