What is the Present State of PA in PWE?

The data on PA and sports in PWE are limited but tend to indicate lower levels of engagement compared with the general population. A survey conducted in Ohio, United States (US) revealed that while 47% of PWE were instructed to be more physically active versus 35% of controls, only 58% of PWE performed PA compared with 76% of controls.Reference Elliott, Moore and Lu ⁹

Similarly, a Norwegian study reported that sedentary lifestyle was more prevalent in PWE compared with controls (25% vs. 13%, p<0.05).Reference Nakken ¹⁰ Another study from the US Midwest documented that PWE performed low-intensity PA ≤3 times per week.Reference Ablah, Haug and Konda ¹¹ A German study demonstrated that controls were more likely to engage in sports on a regular basis (42% of controls vs. 25% of PWE) and less likely to state that they never played sports (15% of controls vs. 31% of PWE). The authors indicated that PWE were instructed to refrain from sports by teachers, instructors and even doctors.Reference Steinhoff, Neusüss, Thegeder and Reimers ¹²

In their Brazilian study, Arida et al reported that 49% of the PWE did not perform PA on a regular basis. The reasons for this phenomenon included recommendations by relatives, friends and physicians, fear and embarrassment of having a seizure in public, lack of time or motivation, fatigue and absence of company, among others.Reference Arida, Scorza, de Albuquerque, Cysneiros, de Oliveira and Cavalheiro ¹³ Identical results were seen in a Korean investigation which documented low PA participation owing to factors such as anxiety, polypharmacy and having experienced a seizure during exercise.Reference Han, Choi-Kwon and Lee ¹⁴

A Canadian population-based study revealed that 60% of PWE reported being sedentary. Individuals with epilepsy had a higher likelihood of being physically inactive compared with the general population (odds ratio [OR]: 1.4, confidence interval [CI] 95% 1.1-1.7).Reference Hinnell, Williams and Metcalfe ¹⁵ The 2010 US National Health Survey revealed that PWE were less likely to follow health guidelines concerning PA. For instance, when asked whether participants had walked at least 10 minutes in the past week, only 39% of PWE indicated they had versus 50% in the general population.Reference Cui, Zack, Kobau and Helmers ¹⁶

Wong and Wirrell determined that adolescents with epilepsy were less likely to be involved in sports groups or engaged in PA and were more likely to be overweight and obese than siblings without the disease.Reference Wong and Wirrell ¹⁷ In Thailand, 38% of PWE surveyed did not perform PA on a regular basis.Reference Saengsuwan, Boonyaleepan and Tiamkao ¹⁸

This trend may be reversible as evidenced by an epilepsy program which enrolled PWE in physical activities in Arizona and was able to achieve a reduction in activity-limited days.Reference Chong, Kudrimoti, Lopez and Labiner ¹⁹

Some studies have failed to document differences in activity levels. In Finland, Jalava et al did not detect any significant difference between PWE and controls in terms of PA frequency. Nonetheless, 9% of PWE reported being physically inactive compared with 2% of controls. The same study found performance below expected levels in tests of muscle strength in people with a history of seizures.Reference Jalava and Sillanpää ²⁰ Similar results were reported in another California-based study. However, in this study, PA was not addressed during routine medical consultation in the preceding year in 56% of appointments.Reference Elliott, Lu, Moore, McAuley and Long ²¹ A Nova Scotia-based Canadian investigation did not relate any difference in PA levels but again revealed that PWE were less likely to participate in sports such as hockey or weight lifting and less likely to engage in PA at home.Reference Gordon, Dooley and Brna ²²

The preponderance of data would suggest that PWE are less active than their peers due to a variety of reasons such as prejudice, stigmatization, fear, shame, lack of knowledge or medical advice.

What are the Clinical Effects of PA in This Population?

Animal models confirm a positive effect of PA on mood and seizure control. In a murine model of depression, PA was correlated with a decrease in negative mood symptoms as well as a delay in the development of epilepsy.Reference Epps, Kahn, Holmes, Boss-Williams, Weiss and Weinshenker ²³ This finding was associated with the production of galanin, a neuropeptide with antidepressant and anticonvulsive effects.Reference Lerner, Sankar and Mazarati ²⁴

PA and sports participation can have a positive impact on overall health status and quality of life of people living with chronic diseases such as depression,Reference Babyak, Blumenthal and Herman ²⁵ arthritis,Reference Gupta and Aggarwal ²⁶ asthma,Reference Westergren, Fegran, Nilsen, Haraldstad, Kittang and Berntsen ²⁷ hypertension and diabetes.Reference Haxhi, Leto and di Palumbo ²⁸ People with epilepsy tend to be more sedentary than the general population, a finding associated with higher body mass index, minor physical resistance, low self-esteem, and increased likelihood of anxiety and depression. A Brazilian study documented that physical inactivity constitutes a risk factor for the development of depression and anxiety in PWE.Reference de Lima, de Lira and Arida ²⁹

In an Ohio-based study, a 12-week sports intervention program specifically designed for PWE generated enhanced quality of life scores in the PA group (p<0.031) compared with controls (p=0.943). Improvement was seen in variables such as self-image, vitality and emotional state.Reference McAuley, Long and Heise ³⁰ Similarly, a study from the University of Alabama confirmed that adult PWE who practiced regular PA had lower levels of depression.Reference Roth, Goode, Williams and Faught ³¹ In a Korean study, it was also observed that a program of regular PA in children with benign epilepsy with centrotemporal spikes (BECTS) generated improvements in attention, psychomotor speed, impulse control, inhibition/disinhibition and problem-solving skills.Reference Eom, Lee and Park ⁵ Overall, exercise seems to exert favorable effects on quality of life, neurocognitive domains and psychosocial functionReference Götze, Kubicki, Munter and Teichmann ^{32 ,} Reference Horyd, Gryziak, Niedzielska and Zieliński ³³ (Table 1).

Table 1

Studies on physical exercise and epilepsy

AED=antiepileptic drug; BECTS=benign epilepsy with centro-temporal spikes; BMI=body mass index; HARCES=the Hague Restrictions in Childhood Epilepsy Scale; JME=juvenile myoclonic epilepsy; PA=physical activity; PWE=people with epilepsy.

The pleiotropic effects of antiepileptic drugs (AEDs) must also be considered. Some AEDs such as carbamazepine, lamotrigine, oxcarbazepine, valproate and clobazam can positively impact mood, while others such as levetiracetam, zonisamide, perampanel and phenobarbital can do the opposite. Moreover, some AEDs are known to promote weight gain (i.e., valproic acid, carbamazepine, vigabatrin, gabapentin and phenobarbital), whereas others may be associated with weight loss (i.e., topiramate, zonisamide and felbamate).Reference Ben-Menachem ⁴⁶ Appropriate AED selection requires joint decision-making between patient and prescriber and a drug’s side-effect profile should be discussed and fully considered before initiation.

Does PA Reduce the Number of IED Seen on EEG and Impact Seizure Frequency?

Electroencephalogram recordings in rats show IEDs decrease or disappearance during PA with returning of the interictal discharges at rest. One hypothesis for this finding is that increased vigilance and attention required in PA may reduce seizure frequency.Reference Hellier and Dudek ⁴⁷ Basic research has shown that short-duration swimming exerciseReference Tutkun, Ayyildiz and Agar ⁴⁸ and short-, moderate- and long-duration treadmill exercisesReference Kayacan, Tutkun, Arslan, Ayyildiz and Agar ⁴⁹ consistently decreased the frequency of penicillin-induced epileptiform activity in male Wistar rats. Likewise, a reduction in the number of clinical and electrographic seizures has been observed with both strengthening (i.e., weightlifting) and aerobic exercises.Reference Radak, Chung and Goto ^{50 ,} Reference Peixinho-Pena, Fernandes and de Almeida ⁵¹

Findings in animal models may have direct observable clinical implications in humans. For instance, an analysis of video EEG recordings from 26 Norwegian children with epilepsy who attended an exercise program demonstrated a 25% decrease in IEDs in approximately 77% (20/26) of patients during exercise. IEDs increased in relation to baseline following cessation of exercise.Reference Nakken, Løyning, Løyning, Gløersen and Larsson ⁴ High-intensity exercise has been shown to reduce seizure occurrence and paroxysmal EEG activity in people with temporal lobe and juvenile myoclonic epilepsy.Reference Vancini, de Lira and Scorza ^{34 ,} Reference De Lima, Vancini and Arida ³⁵ Moreover, a population-based Swedish study demonstrated that low cardiovascular fitness status at 18 years of age correlated with an increased likelihood of developing epilepsy. The association remained even after controlling for several confounders such as family history, personal history of diabetes, stroke and traumatic head injury. The authors concluded that behaviors that increase cardiovascular fitness may act as positive disease-modifiers against the future development of epilepsy.Reference Nyberg, Aberg, Toren, Nilsson, Ben-Menachem and Kuhn ⁵²

Which Mechanisms Related to PA Could Explain Improved Seizure Control?

Data from animal studies would suggest that the putative neuroprotective effects of PA might be ascribed to various genetic, molecular, biochemical and structural changes.Reference Epps, Kahn, Holmes, Boss-Williams, Weiss and Weinshenker ^{23 ,} Reference Hellier and Dudek ^{47 ,} Reference Peixinho-Pena, Fernandes and de Almeida ^{51 ,} Reference Setkowicz and Mazur ^{53 ,} Reference Arida, Scorza, dos Santos, Peres and Cavalheiro ⁵⁴ The proposed mechanisms include:

1. 1. Release of β-endorphins from the opioid systemReference Contet, Gavériaux-Ruff, Matifas, Caradec, Champy and Kieffer ⁵⁵

2. 2. Release of steroids secondary to stressReference Arida, Scorza, Toscano-Silva and Cavalheiro ⁵⁶

3. 3. Increase in melatonin concentrationsReference Mevissen and Ebert ⁵⁷

4. 4. Increase of parvo albumin in affected cells after seizures. This molecule has been linked to antiepileptogenic effects, cytoprotection and prevention of neuronal death in the affected cells.Reference Arida, Scorza, Scorza, Gomes da Silva, da Graça Naffah-Mazzacoratti and Cavalheiro ⁵⁸

5. 5. CA1 cells hyperreactivity reduction and generation of structural changes within the hippocampus, which may have an inhibitory effect on the occurrence of abnormal electrical discharges.Reference Arida, Scorza, Scorza, Gomes da Silva, da Graça Naffah-Mazzacoratti and Cavalheiro ^{58 ,} Reference Arida, Sanabria, da Silva, Faria, Scorza and Cavalheiro ⁵⁹

A study using a murine epileptogenic model by Arida et alReference Arida, de Jesus Vieira and Cavalheiro ⁶⁰ suggested that greater effort was needed to induce epileptogenesis in physical trained animals. The delay in seizure occurrence was attributed to the inhibitory effect of noradrenaline and GABA released during exercise.Reference Westergren, Fegran, Nilsen, Haraldstad, Kittang and Berntsen ^{27 ,} Reference Yoneda, Kanmori, Ida and Kuriyama ⁶¹ Additionally, exercise resulted in decreased production of oxidants and free radicals.Reference Radak, Chung and Goto ⁵⁰

In a pentylenetetrazol murine model used to assess the effect of swimming on epileptogenesis after 6 weeks of practice, the exercise group had a greater latency to first seizure, shorter seizure duration, lower amplitude and frequency of IEDs, increased superoxide dismutase activity and non-protein sulfhydryl levels, and greater attenuation of oxidant production.Reference Souza, Oliveira and Furian ⁶²

Similarly, a Polish study that used a pilocarpine murine model of focal epilepsy determined that animals undergoing a PA program had a longer latency in the appearance of status epilepticus and demonstrated lower intensity and shorter seizure duration.Reference Setkowicz and Mazur ⁵³ Possible hypotheses for these observations include increased angiogenesis resulting in decreased excitotoxicity,Reference Black, Isaacs, Anderson, Alcantara and Greenough ^{63 ,} Reference Kleim, Cooper and VandenBerg ⁶⁴ release of neuroprotective trophic factors and expression of neuronal growth factors.Reference Nisticò, Ciriolo, Fiskin, Iannone, de Martino and Rotilio ^{65 ,} Reference Carro, Trejo, Busiguina and Torres-Aleman ⁶⁶

Can PA Adversely Affect Seizure Control?

Sporadic case reports of PA-induced seizures should be weighed against the relatively numerous and well-established benefits of PA (Table 2). Based on scant data, up to 2-10% of PWE may have exercise-induced seizures (defined as seizures occurring in >50% of training sessions).Reference Nakken ^{10 ,} Reference Bjørholt, Nakken, Røhme and Hansen ⁶⁹ These case reports describe patients with both genetic generalized epilepsiesReference Ogunyemi, Gomez and Klass ^{67 ,} Reference Werz ⁷³ as well as symptomatic focal epilepsiesReference Eriksen, Ellertsen, Grønningsaeter, Nakken, Løyning and Ursin ⁷¹ of frontalReference Simpson and Grossman ⁶⁸ and temporal lobe origin.Reference Sturm, Fedi, Berkovic and Reutens ^{72 ,} Reference Kamel, Badawy and Cook ⁷⁴

Table 2

Seizure induction by exercise

PA=physical activity; PWE=people with epilepsy, TLE=temporal lobe epilepsy.

A review of the limited literature on this subject would indicate that the occurrence of exercise-induced seizures is overall quite rare.Reference Nakken, Løyning, Løyning, Gløersen and Larsson ^{4 ,} Reference Nakken ¹⁰ Of interest, PA-induced seizures have been mainly attributed to high-intensity exercise such as ball games, jogging, running and hiking. In addition, high altitude may also provoke seizures, due in part to hypoxia and hypocapnic hyperventilation.

Some patients are reported to present with a high frequency of seizures happening either often (>80% of the time) or exclusively occurring during PA. The authors surmise that these patients may have a form of temporal lobe reflex epilepsyReference Schmitt, Thun-Hohenstein, Vontobel and Boltshauser ^{70 ,} Reference Sturm, Fedi, Berkovic and Reutens ^{72 ,} Reference Kamel, Badawy and Cook ⁷⁴ and that the temporal lobe might be uniquely more sensitive to the generation of exercise-induced IEDs compared with other cortical areas. In these cases, complex partial seizures might be most susceptible to activation during exercise.Reference Bennett ⁷⁵

Arida et alReference Arida, Cavalheiro, da Silva and Scorza ⁷⁶ correlated the occurrence of PA-induced seizures with homeostatic alterations linked to general fatigue, psychic stress of competition, hypoxia, hypoglycemia, hyperhydration, hyponatremia and hyperthermia. It is known, for instance, that hyperventilation may frequently induce absence seizures and some focal seizures in patients at rest. It has been theorized that resting hyperventilation might trigger seizures due to respiratory alkalosis. However, this explanation is unlikely to be compelling in the case of exercise, which tends to result in metabolic acidosis.