Objective: To update the 2004 American Academy of Neurology (AAN) guideline for treating new-onset focal or generalized epilepsy (GE) with second- and third-generation antiepileptic drugs (AEDs). Methods: The 2004 AAN criteria was used to systematically review literature (January 2003 to November 2015), classify pertinent studies according to the therapeutic rating scheme, and link recommendations to evidence strength. Results: Several second-generation AEDs are effective for new-onset focal epilepsy. Data are lacking on efficacy in new-onset generalized tonic–clonic seizures, juvenile myoclonic epilepsy, or juvenile absence epilepsy, and on efficacy of third-generation AEDs in new-onset epilepsy. Recommendations: Lamotrigine (LTG) should (Level B) and levetiracetam (LEV) and zonisamide (ZNS) may (Level C) be considered in decreasing seizure frequency in adults with new-onset focal epilepsy. LTG should (Level B) and gabapentin (GBP) may (Level C) be considered in decreasing seizure frequency in patients ≥60 years with new-onset focal epilepsy. Unless there are compelling adverse-effect–related concerns, ethosuximide (ETS) or valproic acid (VPA) should be considered before LTG to decrease seizure frequency in treating absence seizures in childhood absence epilepsy (Level B). No high-quality studies suggest clobazam, eslicarbazepine, ezogabine, felbamate, GBP, lacosamide, LEV, LTG, oxcarbazepine, perampanel, pregabalin, rufinamide, tiagabine, topiramate, vigabatrin, or ZNS is effective in treating new-onset epilepsy because no high-quality studies exist in adults of various ages. A recent FDA strategy allows extrapolation of efficacy across populations; therefore, for focal epilepsy, eslicarbazepine and lacosamide (oral only for pediatric use) as add-on or monotherapy in persons ≥4 years old and perampanel as monotherapy received FDA approval.

Objective: To systematically review the evidence and make recommendations with regard to diagnostic utility of cervical and ocular vestibular evoked myogenic potentials (cVEMP and oVEMP, respectively). Four questions were asked: Does cVEMP accurately identify superior canal dehiscence syndrome (SCDS)? Does oVEMP accurately identify SCDS? For suspected vestibular symptoms, does cVEMP/oVEMP accurately identify vestibular dysfunction related to the saccule/utricle? For vestibular symptoms, does cVEMP/oVEMP accurately and substantively aid diagnosis of any specific vestibular disorder besides SCDS? Methods: The guideline panel identified and classified relevant published studies (January 1980-December 2016) according to the 2004 American Academy of Neurology process. Results and Recommendations: Level C positive: Clinicians may use cVEMP stimulus threshold values to distinguish SCDS from controls (2 Class III studies) (sensitivity 86%-91%, specificity 90%-96%). Corrected cVEMP amplitude may be used to distinguish SCDS from controls (2 Class III studies) (sensitivity 100%, specificity 93%). Clinicians may use oVEMP amplitude to distinguish SCDS from normal controls (3 Class III studies) (sensitivity 77%-100%, specificity 98%-100%). oVEMP threshold may be used to aid in distinguishing SCDS from controls (3 Class III studies) (sensitivity 70%-100%, specificity 77%-100%). Level U: Evidence is insufficient to determine whether cVEMP and oVEMP can accurately identify vestibular function specifically related to the saccule/utricle, or whether cVEMP or oVEMP is useful in diagnosing vestibular neuritis or Meniere disease. Level C negative: It has not been demonstrated that cVEMP substantively aids in diagnosing benign paroxysmal positional vertigo, or that cVEMP or oVEMP aids in diagnosing/managing vestibular migraine.

To systematically review the evidence and make recommendations with regard to diagnostic utility of cervical and ocular vestibular evoked myogenic potentials (cVEMP and oVEMP, respectively). Four questions were asked: Does cVEMP accurately identify superior canal dehiscence syndrome (SCDS)? Does oVEMP accurately identify SCDS? For suspected vestibular symptoms, does cVEMP/oVEMP accurately identify vestibular dysfunction related to the saccule/utricle? For vestibular symptoms, does cVEMP/oVEMP accurately and substantively aid diagnosis of any specific vestibular disorder besides SCDS?

Goodin and Reder misstate the AAN’s rules for the classification of therapeutic evidence.1 A randomized controlled trial (RCT) that documents substantial balance between treatment groups in known baseline confounding characteristics is eligible for a Class I rating. A non-randomized matched cohort study that documents substantial balance between treatment groups in known baseline confounders between treatment groups is at best eligible for a Class II rating. In this way the AAN classification system incorporates the methodological superiority of randomization over matching.2 However, as Goodin and Reder correctly point out, randomization can fail. To determine whether randomization (or matching) succeeded in balancing treatment groups on baseline confounding characteristics, well-reported studies report the distribution of baseline confounders between treatment groups.3 Matched cohort studies or RCTs failing to document the distribution of baseline confounders are at best rated Class III by the AAN system. The authors of the long-term follow-up (LTF) study4 failed to report baseline confounding characteristics related to mortality. Thus, the LTF study was rated Class III. Responding to Goodin and Reder’s other points:

Factors influencing medical decision-making