Association of Vitamin C, Thiamine, and Hydrocortisone Infusion With Long-term Cognitive, Psychological, and Functional Outcomes in Sepsis Survivors

This secondary analysis of a randomized clinical trial reports on the cognitive performance, risk of posttraumatic stress disorder and depression, and functional status 6 months after the administration of a treatment regimen consisting of antioxidants and hydrocortisone infusion in adults with sepsis.

The Weschler Memory Scale IV Logical Memory I & II Subtests 4 examine immediate and delayed declarative memory, respectively. The examiner reads a brief story and instructs the participant to repeat the story immediately and again later during the assessment interview. We used two stories, tailored based on the participants' ages. Scores are based on number of salient items included within each story and range from 0 to 19 after combining stories and scaling for age. Higher scores reflect better performance.
The WAIS-IV Similarities Subtest 3 is an 18-question survey used to assess language conceptualization and verbal abstraction. Participants are given pairs of words and asked to explain how they are similar (e.g. a HORSE and a TIGER are both animals). Up to two points can be given for each response, for a total score ranging from 0 to 36. Higher scores reflect better performance.
The Controlled Oral Word Association (COWA) Test 5 consists of three trials wherein the participant is given a letter prompt (F, A, and S) and requested to list aloud as many words as possible beginning with the specified letter within a 60 second period. This test assesses verbal fluency. One point is given for each unique word (excluding proper nouns), and the scores for the three trials are summed. This total sum is used to find a scaled score and a t-score. Higher scores reflect better performance.
The Hayling Sentence Completion Test 6 assesses response inhibition as a form of executive function. Participants listen to a series of sentences with the last word missing and are instructed to complete the sentence with a word that is congruent with the meaning of the sentence. After 15 trials, the participant is then instructed to complete the next 15 sentences with words that are incongruent with the sentence prompts. For example, in response to the prompt "London is a very busy…" participants might respond with "banana" instead of "city" or "town." Scores are based on response time, number and types of errors, and are scaled on a range from 0 to 10 with higher scores reflecting better performance.

Psychological Assessment Tools
The PTSD-8 7 was derived from the first half of the Harvard Trauma Questionnaire Part IV, 8 which corresponds to the DSM-IV criteria for post-traumatic stress disorder. There are four items evaluating intrusion symptoms, two items evaluating avoidance symptoms and two items evaluating hypervigilance symptoms. Item responses are rated on a four-point Likert scale ('not at all' (1), 'a little' (2), 'quite a bit' (3), and 'all the time' (4)). A score of 3 or greater on any item within each of the 3 categories, in relation to a serious or life-threatening event causing intense fear, helplessness or horror that cause clinically significant distress or impairment is consistent with a diagnosis of PTSD. The summed score across all 8 items provides an indication of symptom severity, with higher numbers indicating worse outcome. In the VICTAS study, participants were asked to evaluate the frequency of symptoms over the preceding 6 months, specifically in relationship to their ICU experience.
The Patient-Reported Outcome Measurement Information System (PROMIS) Depression item banks were developed as part of the NIH Roadmap for Medical Research Initiative. 9 The PROMIS Depression item banks assess self-reported negative mood, views of self and social cognition, as well as decreased positive affect and engagement. Somatic symptoms are not included, which eliminates consideration of these items' confounding effects when assessing patients with comorbid physical conditions. The depression short forms are universal rather than disease-specific. All assess depression over the past seven days.
The six items presented on the PROMIS Depression 6a Short Form are rated on a scale of 1 (Never) to 5 (Always) and the total sum of scores is converted to a T-score for each participant. The T-score rescales the raw total into a standardized score with a mean of 50 and a standard deviation of 10. Higher scores reflect worse symptoms. A T-score of 60 corresponds to 1 standard deviation (SD) about the population mean and, based on crosswalk with analogous depression screening instruments, 10 is consistent with the presence of moderate depression.

Functional Assessment Tools
The Katz Activities of Daily Living Score 11 (ADLS) is a brief inventory designed to assess an individual's current ability to perform routine activities without assistance. Activities addressed include eating, bathing, dressing, toileting, transferring, walking and continence. Activities are rated on a scale of 0 (no dependence) to 2 (total dependence). Higher scores reflect greater level of impairment.
The Functional Activities Questionnaire (FAQ) 12 is a 10-question survey designed to assess level of impairment in instrumental activities of daily living such as buying groceries and managing finances. Items are scored on a scale of 0 (Normal function) to 3 (Completely Dependent). and a total score is derived from the sum of all individual item scores. Higher total scores reflect worse impairment.
The EuroQol-5D 13 is designed to measure health-related quality of life. It can be used in a wide variety of health conditions. It is divided into two parts: 1) descriptive ratings of level of difficulty in 5 domains (mobility, self-care, usual activities, pain/discomfort and anxiety/depression) and 2) a visual analog scale (VAS) of overall self-perception of health. Items in part 1 are rated on a scale of 1 (no problems) to 3 (very severe problems), and the VAS provides a single overall rating from 0 (worst imaginable health) to 100 (best imaginable health). For the purposes of the current analysis, we report the responses on part 2. Higher scores represent better perception of an individual's own health.
In our Healthcare Utilization Survey we asked additional questions about rehospitalization, repeat ICU stays, receipt of mental health care since discharge from the sepsis admission. We also collected information about current use of medications for depression or anxiety.

Ineligible or Lost to Follow-Up
No Premorbid education level was assessed at enrollment and was not known by surrogates in 53% of cases. This information was requested directly from survivors at 6-month follow-up. Participants who died, withdrew, declined or were lost to follow-up did not have the second education assessment.

No. (%)
Overall a Partial follow-up information (ADLs, FAQ and hospital utilization) were gathered from the surrogate for one participant who was hospitalized at the time of follow-up. This participant subsequently died prior to completing cognitive and psychological assessments, which were imputed based on previously gathered data. b Participants designated as Unable to Reach at the individual assessment level were those for whom follow-up assessments had been initiated but not completed in one session, and a second session could not be arranged within the follow-up window. a N = Number of participants completing assessment. This may differ from number in primary analysis due to exclusion of participants who were cognitively unable to complete the assessment (in primary analysis these participants were assigned the lowest score). For dichotomous outcomes (marked with **), N indicates number meeting criteria for the outcome of interest (e.g. number of participants screening positive for PTSD). Positive PTSD screen was defined as a score of 3 or higher on 3 out of 4 symptom categories represented within the PTSD-8 questionnaire (see Hayley et al, . Depression was defined as a T-score of 60 or greater on the PROMIS-6 questionnaire.