Repetitive transcranial magnetic stimulation in drug‐resistant idiopathic epilepsy of dogs: A noninvasive neurostimulation technique

Abstract Background Although repetitive transcranial magnetic stimulation (rTMS) has been assessed in epileptic humans, clinical trials in epileptic dogs can provide additional insight. Objectives Evaluate the potential antiepileptic effect of rTMS in dogs. Animals Twelve client‐owned dogs with drug‐resistant idiopathic epilepsy (IE). Methods Single‐blinded randomized sham‐controlled clinical trial (dogs allocated to active or sham rTMS) (I) and open‐labeled uncontrolled clinical trial (dogs received active rTMS after sham rTMS) (II). Monthly seizure frequency (MSF), monthly seizure day frequency (MSDF), and number of cluster seizures (CS) were evaluated for a 3‐month pre‐TMS and post‐rTMS period and safety was assessed. The lasting effect period of rTMS was assessed in each dog treated by active stimulation using the MSF ratio (proportion of post‐TMS to pre‐rTMS MSF) and treatment was considered effective if the ratio was <1. Results No adverse effects were reported. In trial I, MSF and MSDF decreased significantly (P = .04) in the active group (n = 7). In the sham group (n = 5), no significant changes were found (P = .84 and .29, respectively). Cluster seizures did not change significantly in either group. No significant differences were detected between the groups. In trial II, previously sham‐treated dogs (n = 5) received active rTMS and significant decreases in MSF and MSDF were noted (P = .03 and .008, respectively). The overall effect of rTMS lasted for 4 months; thereafter, the MSF ratio was >1. Conclusions and Clinical Importance Repetitive transcranial magnetic stimulation may be a safe adjunctive treatment option for dogs with drug‐resistant IE, but large‐scale studies are needed to establish firm conclusions.

K E Y W O R D S dog, epileptic seizures, management, nonpharmacological, refractory

| INTRODUCTION
Idiopathic epilepsy (IE) is a common neurological disorder, with an estimated prevalence of 0.5% to 0.82% in the general canine population, and up to 33% in certain families of genetically predisposed breeds. [1][2][3][4][5][6][7] Drug resistance occurs in up to 30% of the dogs with IE leading to a grave prognosis and eventually euthanasia because of limited nonpharmacological treatment options. 8 Repetitive transcranial magnetic stimulation (rTMS) has received attention the recent years as a treatment method that can have neuromodulatory effects on the brain that last longer than the duration of the neurostimulation. 9 Although the specific antiepileptic mechanism of action still remains unclear, 10 it might be related to the disruption of networks related to cortical hyperexcitability. 11,12 Clinical trials of low-frequency rTMS in drug-resistant epilepsy in humans however had conflicting outcomes with regard to the decrease in seizure frequency. [13][14][15] Because dogs with spontaneous epilepsy are similar in etiology, clinical manifestation, treatment response, and drug resistance to epileptic humans, [16][17][18][19] a preliminary veterinary study was conducted to (a) investigate this new, noninvasive, and nonpharmacological treatment option for dogs with drug-resistant IE and (b) provide preliminary information for future large-scale clinical trials in dogs that could establish firm conclusions regarding its effect in IE of dogs and a potential use as a translational model for epileptic humans.

| MATERIAL AND METHODS
Dogs with drug-resistant IE without age, breed, or sex limitations were considered for enrollment in the study. The classification, definition, and diagnosis of IE were based on the recommendations of the International Veterinary Epilepsy Task Force (IVETF) consensus reports. 20,21 Drug-resistant IE, in particular, was defined as epilepsy with <50% decrease in monthly seizure frequency (MSF) compared to baseline after treatment with at least 2 antiepileptic drugs (AEDs) despite optimal dose, serum drug concentrations, or both. 20,21 The study consisted of 2 trials: (I) a single-blinded randomized period of epileptic seizure frequency and adverse events related to treatment). After the initial evaluation period of 3 months, dogs treated using active rTMS were followed as long as possible until study termination. During phases (a) and (c), owners recorded epileptic seizure events in a diary.
The study was approved by the university's ethical committee (EC 2016/30). Owner consent forms were provided and signed by the owners. The overall timespan of recruitment was 12 months.

| Trial I
At the end of phase (a), dogs were randomly assigned to the active or sham rTMS group by using sealed envelopes. Equal numbers of entries indicating either active or sham rTMS were created and placed in envelopes. The envelopes were sealed, mixed, and randomly numbered. They were opened for each included dog following a numerical sequence starting from envelope number 1. The investigators did not know the randomization order. Owners were blinded to the chosen treatment (ie, owners were not informed about which treatment their dog would receive). The dogs were hospitalized for 5 days or on consecutive afternoons and all received sedation and IV catheters while the procedure was initiated and after the owners had left the hospital.
Blood samples for CBC and serum biochemistry as well as AED serum concentration assessment were collected from all dogs at that time.
During phase (b), the dogs received rTMS (active or sham) for 1 hour daily for 5 consecutive days. Overall, stimulation parameters and study environment were exactly the same in both groups, the only difference being that the sham group received inactive stimulation by placing the operating (round) coil perpendicular to the skull and with a distance of 20 cm above the head in order to circumvent brain stimulation. Dogs in both groups were sedated using dexmedetomidine 1 μg/kg (Dexdomitor; Orion pharma, Finland) and

| Trial II
After the 3-month evaluation period of trial I, dogs from the trial I sham group were included in trial II to receive active rTMS. The effect of active rTMS on these dogs was assessed for an additional 3-month evaluation period. Owners were not blinded to the treatment in trial II. The procedure followed in each phase was the same as described earlier for the active rTMS group in trial I.   Table 2 and Figures 2 and 3. When comparing the 2 groups, no significant differences were found in MSF (P = .14), MSDF (P = .25), and number of CS (P = .61) post-rTMS. Based on the MSF ratio, the median rTMS effect lasted for 4 months (range, 2-10).

| Statistical analysis
Median follow-up of dogs treated with active rTMS from inclusion to termination of the study was 4 months (range, 2-12). In the active group, no adverse effects were related to the 5-day treatment of lowfrequency rTMS.    29 However, no consensus currently exists on the optimal stimulation parameters for specific diseases, such as epilepsy. 10 Cortical excitability can be increased or decreased using highfrequency (>1 Hz) and low-frequency (≤1 Hz) neurostimulation, respectively, which likely is caused by the potential induction of long-term potentiation and depressive mechanisms, respectively. 30 Although there is an overall agreement that low-frequency rTMS supresses epileptic discharges and leads to a decrease in seizure frequency, 31-33 the remaining stimulation parameters are quite variable in trials of human patients. 10 With regard to the pulses, more pulses per rTMS session were associated with higher efficacy. 34 As far as the duration of treatment is concerned, rTMS has an effect that outlasts the duration of the treatment, which can be attributed to consecutive sessions. 14,35,36 In 1 study, neuroplastic long-lasting changes were observed when at least 2 rTMS sessions were administered within 24 hours, but not when administered 1 week apart. 35 In addition, the effects of low-frequency rTMS on cortical excitability were dependent on the intensity used. 37 In 1 study, high rTMS intensity (90%) was found to be superior and significantly decreased seizure frequency, compared to low-intensity rTMS (20%), 38 whereas another study found good antiepileptic rTMS effect with relatively high intensity (70%). 14

| CONCLUSION
We presented preliminary evidence on the potential antiepileptic effect of rTMS in epileptic dogs that received active stimulation because such an effect was not shown in dogs that received sham rTMS, although the small study population did not allow sufficiently powered results to detect a statistical difference between the groups.
Evidence also was provided to support the safety profile of this tech-

ACKNOWLEDGMENTS
No funding was received for this study. The authors thank all the participating owners, veterinarians, nurses, technicians and veterinary students who helped with the study.