Life after intensive care-post : intensive care syndrome, inflammation, and HMGB1

Survival rates of intensive care unit (ICU) patients are consistently increasing. This new patient population has recently gained increased recognition as they suffer from psychological, physical, and cognitive impairments in the years to come after ICU discharge. These symptoms are collectively referred to as the post-intensive care syndrome (PICS). Importantly, the underlying pathophysiology of PICS is unclear, but in many ways consistent with a persistent, non-resolving inflammation. In this context, high mobility group box 1 (HMGB1), a prototypical alarmin involved in both sterile and infectious inflammation, has gained interest, particularly since animal studies indicate that HMGB1 promotes neuroinflammation and cognitive impairment. This thesis investigates aspects of PICS in ICU survivors, including subjective and objective cognitive function, physical performance, and markers of inflammation. Patients were included in two cohorts designed for ICU follow-up studies.

In study I, the association between sepsis and delirium during ICU stay and symptoms of psychological distress (Hospital Anxiety and Depression Scale (HADS) and Post-traumatic Stress Symptoms Scale-10 (PTSS-10)) and self-rated cognitive function (Cognitive Failures Questionnaire (CFQ)) after three months were investigated. There was no significant association between sepsis or delirium at ICU stay and self-rated cognitive function at the three months follow-up. In contrast, there was a strong significant correlation between patients’ self-rated cognitive function and symptoms of depression, anxiety, and PTSD.

Studies II-IV investigate a 12 months follow-up cohort of ICU survivors. Patients were examined at the ICU follow-up clinic at three, six, and twelve months after ICU discharge. They underwent formal neuropsychological testing, performed physical tests, and responded to three questionnaires on psychological distress and subjective cognitive function (HADS, PTSS-10, CFQ). Blood samples were collected at the three and six-month follow-up visits. Study II investigates whether patients ́ subjective cognitive function correlates to objectively measured cognitive function. Answers on the CFQ were analyzed together with outcomes on the Cambridge Neuropsychological Test Automated Battery (CANTAB). There was no clinically relevant correlation between subjective and objective cognitive function as measured here. We conclude that subjective cognitive function tests in ICU survivors must be interpreted with caution and that both subjective and objective testing may be necessary to adequately ascertain cognitive function in ICU survivors. In study III, the association between plasma HMGB1 and objective cognitive function measured in four different cognitive domains (executive function, visual memory, sustained attention, working memory) was investigated. Interestingly, plasma levels of HMGB1 were significantly elevated in the ICU, at discharge, and at the three- and six-months follow-up visits as compared with reference populations. Elevated plasma levels of HMGB1 were associated with reduced sustained attention at the three- and six-month follow-up visits. Based on these findings, further follow-up studies on HMGB1 biology in ICU survivors are warranted to investigate the potential for therapeutic targeting of HMGB1 function in prevention of cognitive impairment in ICU survivorsIn study IV we explore the association between plasma HMGB1 and physical performance at ICU follow-up. We observed no significant association between levels of plasma HMGB1 and the three physical tests performed (i.e., 6-min walk test, timed stands test, handgrip strength test).

In conclusion, this thesis provides new insights on objective and subjective cognitive function, psychological distress, and markers of inflammation in ICU survivors. Future work may build on this knowledge to improve the identification and treatment of at-risk subjects in this vast and growing patient population.

List of scientific papers

I. The impact of sepsis, delirium, and psychological distress on self-rated cognitive function in ICU survivors-a prospective cohort study. Brück E, Schandl A, Bottai M, Sackey P. J Intensive Care. 2018 Jan 8;6:2.
https://doi.org/10.1186/s40560-017-0272-6

II. Lack of clinically relevant correlation between subjective and objective cognitive function in ICU survivors: a prospective 12-months follow-up study. Brück E, Larsson JW, Lasselin J, Bottai M, Hirvikoski T, Sundman E, Eberhardson M, Sackey P, Olofsson PS. Crit Care. 2019 Jul 12;23(1):253.
https://doi.org/10.1186/s13054-019-2527-1

III. Prolonged elevation of plasma HMGB1 is associated with cognitive impairment in intensive care unit survivors. Brück E, Lasselin J, Caravaca AS, Gallina AL, Bottai M, Andersson U, Sackey P, Olofsson PS. [Submitted]

IV. Plasma HMGB1 levels and physical performance in ICU survivors. Brück E, Svensson-Raskh A, Larsson JW, Caravaca AS, Gallina AL, Eberhardson M, Sackey P, Olofsson PS. [Manuscript]