Current evidence for a role of serotonin in SUDEP
[Part 2: Questions; Buchanan GF & Richerson GB]
There is a large amount of research currently being directed at understanding why sudden unexpected death in epilepsy (SUDEP) occurs, because defining the mechanisms of SUDEP may lead to ways to identify patients at highest risk, discover specific treatments to prevent death, or validate biomarkers to allow definitive diagnosis when SUDEP does occur. A leading theory proposes that a subset of SUDEP results when a seizure causes respiratory dysfunction, such as apnea and/or hypoventilation (Massey CA et al., 2014). The profound unresponsiveness that typically follows generalized seizures may increase the risk of SUDEP due to the inability to wake up and relieve any airway obstruction such as lying face down in a pillow (Richerson GB & Buchanan GF, 2011). The recent MORTEMUS series supports this theory because terminal respiratory arrest was found to be an early event more often than previously realized where 81% of patients died in the prone position, and all fatal events occurred after generalized seizures (Ryvlin P et al., 2013).
In recent years there has been increasing interest in a possible role of serotonin in SUDEP. Serotonin is a neurotransmitter found in only a small number of neurons in the brainstem, but with effects on the rest of the brain. Serotonin is commonly associated with mood, but also stimulates breathing, causes wakefulness and raises seizure threshold (Bagdy G et al., 2007; Richerson GB, 2004). Therefore, a defect in serotonin could contribute to SUDEP by increasing seizure frequency, and causing impaired breathing and arousal. It might also explain why many patients with seizures have co-morbid depression (Richerson GB & Buchanan GF, 2011).
Animal models of SUDEP have been very important in understanding the possible role of serotonin in SUDEP. The first evidence that serotonin might be linked to SUDEP came from experiments in laboratory mice. When one of the receptors for serotonin (2c) was deleted in mice using genetic engineering, spontaneous seizures occurred followed by death due to respiratory arrest (Brennan TJ et al., 1997). More recently, a series of studies were performed in DBA/2 mice, in which loud sounds cause seizures followed by respiratory arrest (Venit EL et al., 2004). Respiratory arrest could be prevented if mice were treated with the anti-depressant fluoxetine, a serotonin selective reuptake inhibitor (SSRI) (Tupal S & Faingold CL, 2006), and was made more likely if mice were treated with cyproheptidine, a drug that blocks serotonin receptors (Tupal S & Faingold CL, 2006). In a third mouse model, the DBA/1 mouse, which also exhibits seizures with respiratory arrest when exposed to loud sounds (Faingold CL, Randall M & Tupal S, 2010), the SSRIs fluoxetine (Faingold CL, Tupal S & Randall M, 2011; Zeng C et al., 2015), sertraline (Faingold CL & Randall M, 2013), paroxetine, and fluvoxamine (Faingold CL et al., 2014) can prevent the respiratory arrest. In these mice, activating the serotonin 3 receptor (Faingold CL et al., 2016) or giving the animal the chemical precursor for serotonin, 5-hydroxytryptophan (Zhang H et al., 2016), can also prevent the respiratory arrest. Both DBA/1 and DBA/2 mouse were found to have abnormalities in serotonin receptors in the brain (Uteshev VV, Tupal S, Mhaskar Y & Faingold CL, 2010; Faingold CL, Randall M, Mhaskar Y & Uteshev VV, 2011). In a fourth mouse model, Lmx1b conditional knockout mice, serotonin neurons never develop (Zhao ZQ et al., 2006). These mice are more likely to have seizures, and when they do they are more likely to die from respiratory arrest. In this model, death can be prevented by mechanical ventilation during a seizure, or by pre-treating with the drugs TCB-2 or DOI, which activate serotonin 2A receptors (Buchanan GF et al., 2014). Interestingly, many patients with epilepsy have hypoxia after seizures (Bateman LM et al., 2008), which could possibly increase the risk of SUDEP. Taking SSRIs can reduce this hypoxia (Bateman LM et al., 2010).
Interest in the relationship between SUDEP and serotonin has been increased by work on sudden infant death syndrome (SIDS). Theories of SIDS have also centered on defects in breathing and arousal, and SIDS has also been linked to an abnormal serotonin system (Kinney HC et al., 2009). A majority of SIDS cases are found in the prone position. Thus, there may be shared mechanisms between SIDS and SUDEP. In fact, it has even been suggested that some cases of SIDS might be due to seizures that have gone undiagnosed (Brownstein CA et al., 2018; Kinney HC et al., 2013; Rodriguez ML et al., 2012).
The possibility that some cases of SUDEP are due to serotonin-dependent respiratory arrest after a seizure is intriguing, though not surprising (Massey CA, Sowers LP, Dlouhy BJ & Richerson GB, 2014; Richerson GB, 2013). Serotonin neurons are known to regulate breathing and cause arousal from sleep in order to maintain normal CO2 and pH in the blood (Richerson, 2004). In animal models, treatments that increase serotonin decrease seizure susceptibility. Conversely, decreasing serotonin function increases seizure susceptibility (Badgy et al., 2007). Interestingly, patients with epilepsy who take SSRIs, such as the aforementioned fluoxetine, tend to have improved seizure control (Kanner AM, 2009). A recent study demonstrated that serum serotonin levels rise following generalized seizures and the serum serotonin level is inversely proportional to the duration of suppression of brain activity following a seizure (Murugesan A et al., 2018). The duration of this suppression may correlate with SUDEP risk (Lhatoo SD et al., 2010).
It is likely that further studies in animal models, and validation in human patients, will lead to better definition of the mechanisms of SUDEP and the potential link to serotonin. This may then allow identification of those individuals at the highest risk of SUDEP in whom preventive measures can be implemented. One of these preventive measures may one day include drugs that, at least in part, increase brain serotonin activity.
Gordon F Buchanan, Associate Professor, Department of Neurology, University of Iowa, USA
George B Richerson, Professor & Head, Department of Neurology, University of Iowa, USA
Updated December 2018; Updated June 2016; Original Dec 2014
How to cite:
Buchanan GF & Richerson GB. Current evidence for a role of serotonin in SUDEP. In: Hanna J, Panelli R, Jeffs T, editors. Continuing the global conversation [online]. SUDEP Action & SUDEP Aware; 2018 [retrieved day/month/year]. Available from: www.sudepglobalconversation.com.