In-Hospital Neurologic Complications, Neuromonitoring, and Long-Term Neurologic Outcomes in Patients With Sepsis: A Systematic Review and Meta-Analysis

Abstract

Objectives: Although delirium is well described in patients with sepsis, there are limited data on other neurologic complications. We aimed to systematically review the prevalence, neuromonitoring tools, and neurocognitive outcomes in sepsis patients with neurologic complications.

Data sources: MEDLINE and six other databases (Embase, Web of Science, Cochrane CENTRAL, and ClinicalTrials.gov ) were searched through January 2023.

Study selection: Studies of adult patients with sepsis reported neurologic complications, use of neuromonitoring tools, neuropathology, and cognitive outcomes.

Data extraction: Two independent reviewers extracted the data. Random-effect meta-analyses were used to pool data.

Data synthesis: Seventy-four studies ( n = 146,855) were included. Neurologic complications were reported in 38 studies ( n = 142,193) including septic encephalopathy (36%, 95% CI, 27-46%; I 2 = 99%), ischemic stroke (5%, 95% CI, 2.1-11.5; I 2 = 99%), intracranial hemorrhage (2%, 95% CI, 1.0-4.4%; I 2 = 96%), seizures (1%, 95% CI, 0.2-7%; I 2 = 96%), posterior reversible encephalopathy syndrome (9%), and hypoxic-ischemic brain injury (7%). In the meta-regression analysis, pulmonary infection, sepsis induced by a gram-positive organism, higher sequential organ failure assessment score, acute physiology and chronic health evaluation II score at admission, and longer ICU length of stay were associated with higher risk of developing septic encephalopathy. Three studies ( n = 159) reported postmortem neuropathological findings, acute brain injury was noted in 47% of patients. Twenty-six studies ( n = 1,358) reported the use of neuromonitoring tools, electroencephalogram was the most used tool for seizure detection. Transcranial Doppler and near infrared spectroscopy were used for monitoring cerebral hemodynamic changes to detect early ischemia. Six studies reported cognitive outcomes ( n = 415) up to 12 months postdischarge and cognitive impairment (≥ one domain) was reported in 30%.

Conclusions: In-hospital neurologic complications are common in patients with sepsis. However, the mechanism and timing of those sepsis-associated complications are poorly understood and there are limited data on standardized neuromonitoring in this population.