Sudden unexpected death in epilepsy patients (SUDEP) is devastating outcome for some patients with epilepsy. It ranks second only to stroke among neurological diseases in years of potential life lost. Patho-mechanisms of SUDEP remain not well understood, however peri-ictal respiratory dysfunction likely plays an important role in many cases.

Literature supports a critical role for the serotonergic system in central control of ventilation. Serotonin neurons in the raphe nuclei of the brainstem sense rising carbon dioxide and low pH, thereby stimulating breathing and arousal. These responses may serve as mechanisms that protect against asphyxia, particularly during sleep or the post-ictal state. In mouse models of seizure-induced sudden death, pre-treatment with selective serotonin reuptake inhibitor (SSRI) agents prevents death following seizures. Hence, the investigators hypothesize that a subset of drug resistant epilepsy patients who have impaired central chemo-responsiveness have a greater degree of peri-ictal respiratory depression, therefore a higher risk of SUDEP. The investigators further hypothesize that fluoxetine will improve central chemo-responsiveness and therefore will reduce peri-ictal respiratory depression.

Sudden unexpected death in epilepsy (SUDEP) refers to the sudden, unexpected, nontraumatic, non-drowning, witnessed or unwitnessed death of an individual with epilepsy. Postmortem examination in such cases fails to reveal an obvious medical or toxicologic cause for the death, and patients who die from SUDEP are typically healthy apart from their epilepsy. The incidence of SUDEP in epilepsy patients is estimated to be 0.1 in 1000 patient years, and this rate increases to >9.3 per 1000 for those with durg resistant epilepsy (DRE) who are candidates for epilepsy surgery. Although thought to be rare, SUDEP is estimated to be responsible for 17% of all deaths in patients with epilepsy, and approximately 50% of all deaths in patients with DRE. It is second only to stroke among neurological diseases in YPLL because many who die of SUDEP are relatively young and therefore it is a major public health concern. While there are some acknowledged risk factors for SUDEP, the actual cause, or causes, of SUDEP is not known. Seizure induced respiratory depression is likely to be a major contributor in SUDEP in many cases.

Preliminary results from the ventilatory response to CO2 or hypercapnic ventilatory response (HCVR) study of patient with epilepsy in epilepsy monitoring unit (EMU) suggests prolonged period of CO2 elevation after seizures correlating with low HCVR. These findings suggest a defect in CO2 responsiveness in this high-risk population that may predispose to SUDEP. Serotonin nerve cells in the brain stem are responsible for detecting increases in CO2, and in response stimulating breathing and arousal from sleep. Fluoxetine is a selective serotonin reuptake inhibitor (SSRI) drug that increases availability of serotonin in the brain. As such, it may serve to stimulate breathing after seizures in patients with epilepsy who exhibit low CO2 sensitivity at baseline and this may alter SUDEP risk.

This study consists of a double blind randomized controlled clinical trial with a 6-week titration of an intervention. It is designed to evaluate primarily feasibility of a larger clinical trial testing efficacy of fluoxetine in modifying HCVR in patients with epilepsy while also collecting important secondary and exploratory outcomes that would be valuable for designing future larger studies.

We will evaluate challenges in screening, enrollment, randomization, and completion of study-related procedures by quantifying the numbers of subjects eligible for screening, the number of subjects enrolled in the study per month, the proportion of patients successfully completing the study, and the specific challenges at each step. We will also assess challenges in setting up and performing outpatient HCVR testing.

Inclusion Criteria:

1. Adult patients aged 18 or older
2. Patients with epilepsy
3. Native English speaker or adequate fluency in English to provide informed consent.
4. Female patients of child-bearing potential must be using an acceptable method of contraception, and willing to refrain from sexual intercourse during the study.

Exclusion Criteria:

1. Progressive neurological disease.
2. Clinical diagnosis of bipolar disease, panic disorder, psychosis or severe depression, or PHQ-9 score > 20
3. Patients with prior hospitalization related to depression or Electroconvulsive therapy.
4. History of suicidal ideation or intent in past or present
5. Clinical history or laboratory evidence of hepatic or renal insufficiency.
6. Pregnant or lactating women.
7. Current heavy alcohol use (>14 drinks per week for men or >7 drinks per week for women) or) known medical disorder related to alcohol use or current illicit drug use, other than marijuana and its derivatives.
8. Patients with recent use (<1 month) or already taking fluoxetine or other selective serotonin reuptake inhibitors (SSRIs).
9. Concurrent use of monoamine oxidase inhibitors, antipsychotic agents, antidepressant agents other than SSRIs or frequent use of triptan agents (>2/week).
10. History of a previous allergic reaction or adverse effects with fluoxetine, hypersensitive reaction-anaphylaxis; laryngeal edema; hives
11. History of serotonin syndrome.
12. History of uncontrolled pulmonary or cardiac illness.
13. Patients with hypercapnic ventilatory response (HCVR) slope of > 2.0
14. Patients with known prolong QT interval
15. Patients with family history of prolong QT interval
16. Patients with family history of sudden cardiac death under the age of 40 in a first degree relative.

Adverse events were monitored and collected after randomization up to 7 weeks from study drug initiation or study dropout which ever was closer.

We used definition of adverse events similar to clinicaltrials.gov.