Cognition, emotional state, and quality of life of survivors after cardiac arrest with rhythmic and periodic EEG patterns

.


Introduction
After cardiac arrest followed by successful resuscitation, 64-80% of patients arrive at the hospital in a comatose state because of diffuse postanoxic encephalopathy. 1,2 During this comatose state, brain activity can be measured with an electroencephalogram (EEG). Different EEG patterns reflect divergent extents of ischemic brain injury, and some are reliable predictors of functional outcome. [3][4][5] Rhythmic and periodic patterns (RPPs), often referred to as electrographic seizures or status epilepticus, are reported in 10-33% of patients, with case fatality rates of 80-100%, despite treatment with anti-epileptic drugs. [6][7][8] Based on the few existing reports on longterm outcomes of survivors, 6-10% may have a good outcome according to the Cerebral Performance Categories (CPC 1 or 2), [8][9][10] but data on cognitive, emotional, and quality of life outcomes are lacking. Studies in cardiac arrest survivors using more sensitive instruments than the CPC found disturbances in the domains of cognition, emotion, and fatigue in approximately half of unselected survivors. 11,12 The incidence and severity of these problems may be higher in the subgroup of survivors with RPPs, since these reflect a more severe postanoxic encephalopathy, but data are so far lacking. 10 Here, we aim to provide insight into the neurological, cognitive, emotional, and quality of life outcomes at one year after cardiac arrest in patients with RPPs in the comatose stage. Knowledge about these patients' long-term outcomes could guide treatment and appropriate care decisions.

Study design
A predefined analysis of prospectively collected one-year outcomes of patients included in the 'treatment of electroencephalographic status epilepticus after cardiopulmonary resuscitation' (TELSTAR) trial was performed. 8 TELSTAR was a multicenter randomized trial in comatose cardiac arrest patients with RPPs on continuous EEG lasting > 30 minutes. The intervention contrast was a step-wise strategy suppressing RPPs with anti-seizure medication for 48 h in addition to standard care versus standard care alone. The primary outcome was neurological recovery according to the CPC at 3 months. Secondary outcomes collected in 5/11 participating centers included cognitive outcome assessed with a cognitive assessment, emotional outcomes with the HADS, and quality of life with the SF-36 at twelve months. The TELSTAR trial was approved by the Medical Research Ethics Committee Twente in the Netherlands (NL46296.044.13). Methods and primary outcomes have been published previously. 13

Participants
The trial population consisted of comatose adult patients after cardiac arrest and successful cardiopulmonary resuscitation, with RPPs on continuous EEG. RPPs comprised periodic discharges, rhythmic delta activity, and spike-and-wave or sharp-and-wave, at a rate of 0.5 Hz, during thirty minutes. One-year survivors that received a one-year follow-up cognitive assessment were included in the current analyses.

Procedure
Written informed consent was obtained from legal representatives. The patients or legal representatives were asked for separate informed consent for the one-year follow-up. At the one-year follow-up, information on cognitive functioning, depression and anxiety, and quality of life were obtained at the local hospital or at the patients' residence.

Cognitive assessment
Depending on the mental capacity of the patient, one of three predefined cognitive test batteries was administered (Supplementary materials, Table 1). The full test battery took 2.5 hours. We analyzed (sub)tests for three cognitive domains: 14

Anxiety and depression
The Hospital Anxiety and Depression Scale (HADS) was used to assess feelings of depression and anxiety. 15 A higher score represents more complaints: 0-7 indicates no anxiety or depression, 8-10 indicates a possible anxiety disorder or depression, and a score of 11-21 indicates a probable anxiety disorder or depression. 16 .

Quality of life
Quality of life was assessed by the 36-item short-form health survey (SF-36), containing 36 questions assessing eight subdomains of quality of life. 17 Items can be scored from 0-100 with higher scores indicating a better health state.

Statistical analysis
Data analyses were performed using SPSS Statistics 25.0. 19 Descriptive statistics were used to describe patient, cardiac arrest, and RPP characteristics. A patient was considered impaired in a cognitive domain if he/she had a score more than 1.5 SD below the mean of the norm group (general population, controlled for sex, age, and education) on 2 (sub)tests within the domain of memory, executive functioning, or attention. 20,21,22 The scores on the SF-36 and the HADS were compared to norm scores of the general population 23,24 , scores of unselected cardiac arrest survivors one-year post-arrest from the Activity and Life After Survival of a Cardiac Arrest trial (ALASCA) 25 , and other patient groups. 26,27 Medians and interquartile ranges (IQRs) were used because of the small sample size and non-normally distributed data. To test for potential selection bias, baseline characteristics of survivors included in this analysis were compared with those not included, using Mann-Whitney and Chi-squared tests, where appropriate. Pvalues < 0.05 were considered statistically significant.

Patient characteristics
Fourteen of the 31 patients who survived to one-year from the total 172 patients in the TELSTAR trial had a cognitive assessment and were included in this analysis. Seventeen of the survivors were not included in this analysis, because they were included in the participating centers that did not collect data on cognitive or emotional outcome at one year (n = 5) or refused to take part in the follow-up (n = 12). Baseline characteristics of the sample (n = 14) compared to the other survivors (n = 17) did not differ in terms of age (median = 58 vs 59 years) or sex (21% vs 32% female). The proportion of patients with a favorable outcome (CPC 1 or 2) was slightly (but not significantly) higher in the current sample (57% (8/14) vs. 41% (7/17)). All but one patient had continuous EEG background activity during RPPs (13/14). The most common RPP type was periodic discharges (10/14). More information about the sample and EEG characteristics can be found in Table 1.

Anxiety and depression (HADS) and quality of life (SF-36)
According to the HADS, that was completed by eight participants, two had a possible depression, one a possible anxiety disorder (8 score 10), one a probable depression, and one a probable anxiety disorder (score 11). The median score of the five patients who completed the SF-36 was 53 on a scale ranging from 0 to 100 (scores = 38, 43, 53, 57, 84). In Table 2, the medians and IQRs of scores on the SF-36 subscales and the HADS are compared with those of the general population (age 55-64 years), an unselected cardiac arrest population at one-year follow-up, ICU survivors after SARS-CoV-2 pneumonia at one-year follow-up, and ST-elevation myocardial infarction (STEMI) patients at 18-month follow-up. [23][24][25][26][27] The patients in the current sample scored lower on most of the SF-36 subscales.

Discussion
This is the first study to describe the cognitive, emotional, and quality of life outcomes of one-year post cardiac arrest patients with RPPs during the comatose state. We found that 13/14 patients (93%) had a cognitive impairment. 20,21 This subgroup with RPPs during coma seems more impaired than unselected one-year survivors of cardiac arrest. In the ALASCA trial, 13-43% of 141 patients remained cognitively impaired one-year post-arrest, in the current sample this is 93%. 21 This may reflect relatively severe postanoxic encephalopathy and a higher risk of poorer cognitive outcome in patients with RPPs during coma, although the sample was too small to draw strong conclusions.
Half of the assessed patients reported a concerning level of symptoms of depression or anxiety. This seems to be a higher proportion than in the general population, unselected one-year cardiac arrest survivors, and STEMI patients. 24,25 The quality of life also seems lower compared to the general population and other patient groups. 23,25   The poor cognitive outcomes are somewhat surprising, since 8/14 patients had a "good outcome" according to the CPC. This confirms previous findings that using more sensitive instruments to test cognition and wellbeing is warranted. 11,12 Strengths of this study include the prospective design. The most important limitation is the small sample size. There was a high case fatality rate of comatose patients with RPPs in TELSTAR. Due to the small sample size, it was not possible to perform statistical analyses, nor to draw strong conclusions from the data. Another limitation is the use of three different cognitive test batteries with different difficulties and workload, necessary for the wide range of mental capacity levels within the study population. The consequent reduction in standardization complicated our analysis. Three out of 14 patients were on anti-seizure medication at one year and had clinically manifest seizures. Although we believe that cognitive dysfunction was primarily caused by postanoxic encephalopathy in our cohort, we cannot exclude an association between anti-seizure medication and our cognitive outcomes.

Conclusions
To conclude, this small-scale analysis provides insights into one-year recovery after cardiac arrest of comatose patients with RPPs. All but one had cognitive impairment at one-year follow-up, often despite a 'good' outcome according to the CPC. The quality-of-life and depression and anxiety scores were worse than those of unselected cardiac arrest survivors, other patient populations, and the general population.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.