Introduction

      Epilepsy is the most common chronic neurological condition in people with an estimated incidence of 0.05–0.1% and prevalence of 0.4–1% (Sander and Shorvon, 1996; Cowan, 2002). Despite treatment with two adequate anti-epileptic drugs (AEDs), 23% of human patients continue to have seizures (Picot et al., 2008). Epilepsy has also been suggested in dogs to be the most common chronic neurological disorder (Chandler, 2006; Fluehmann et al., 2006) with an estimated prevalence of 1–2% in a referral hospital population (Schwartz-Porsche, 1986) and 0.6% in first opinion practice (Kearsley-Fleet et al., 2013). Around 75–85% of dogs with idiopathic epilepsy will continue to have seizures (Heynold et al., 1997; Berendt et al., 2002, 2007; Arrol et al., 2012) and around 20–30% will remain poorly controlled (<50% reduction of seizure frequency) despite adequate treatment with phenobarbitone (PB) and/or potassium bromide (KBr) (Schwartz-Porsche et al., 1985; Podell and Fenner, 1993; Trepanier et al., 1998). Thus, with an estimated current UK companion dog population of 9.4 million, around 55,000 dogs have idiopathic epilepsy of which 12500 will be classified as poorly controlled. Dogs with epilepsy, especially poorly controlled epilepsy with a high seizure frequency, have an increased risk of premature death, behaviour changes and a reduced quality of life (Chang et al., 2006; Berendt et al., 2007; Shihab et al., 2011; Wessmann et al., 2012). Seizures do not only affect the quality of life for the affected dogs, but also for the pet owner (Chang et al., 2006; Wessmann et al., 2012). Improving our understanding of why some patients respond to treatment while others do not is therefore of key importance.

      Epilepsy is caused by a heterogeneous group of chronic conditions with seizures being its clinical manifestation. Despite the brain’s complex structure, there are only limited ways in which it can demonstrate its function and dysfunction. A wide variety of disturbances of brain structure and function can result in seizures and epilepsy. However, similar pathophysiological pathways can lead to other neurodevelopmental and neuro-behavioural disorders (Johnson and Shorvon, 2011; Shihab et al., 2011). Many of these disturbances share commonalities at various levels, which can help to understand, diagnose, monitor and treat these disorders.

      Investigations into why treatment could fail can therefore:

      • Shed light on the mechanisms causing a refractory state to drug therapy and how to overcome them; and

      • Deepen our understanding of the patho-physiology of epilepsy and its natural clinical course.

       Definition

      The definition of pharmacoresistant epilepsy has been a matter of debate in human medicine. Formerly, epilepsy was considered responsive to medical treatment if seizures were reduced by ≥50% to one or two adequate AED which achieved therapeutic serum concentration and steady state (Regesta and Tanganelli, 1999). Most current epilepsy drug trials in veterinary medicine have used this definition to determine AED efficacy (Dewey et al., 2004, 2009; Platt et al., 2006; von Klopmann et al., 2007; Volk et al., 2008; Muñana et al., 2010, 2012b). However, this does not take into account that even a reduction of seizure frequency by more than 50% may still entail a high seizure frequency, which negatively impacts on the animal’s and the owner’s quality of life (Chang et al., 2006; Wessmann et al., 2012). A good quality of life is however best improved by seizure freedom, one of the main outcome measures nowadays in epilepsy trials in human medicine (Elsharkawy et al., 2009).

      A task force of the International League against Epilepsy (ILAE) has recently agreed on a global consensus on granted outcome measures for therapeutic interventions (Level 1) and definition of pharmacoresistant epilepsy (Level 2) taking the occurrence of drug-induced side effects and seizure control into account (Kwan et al., 2010):