Long-term risk and predictors of cerebrovascular events following sepsis hospitalization: A systematic review and meta-analysis

Abstract

Background: Long-term risk and predictors of cerebrovascular events following sepsis hospitalization have not been clearly elucidated. We aim to determine the association between surviving sepsis hospitalization and cerebrovascular complications in adult sepsis survivors.

Method: We searched MEDLINE, Embase, Scopus, Web of Sciences, Cochrane library, and Google scholar for studies published from the inception of each database until 31 August 2022.

Results: Of 8,601 screened citations, 12 observational studies involving 829,506 participants were analyzed. Surviving sepsis hospitalization was associated with a significantly higher ischemic stroke [adjusted hazard ratio (aHR) 1.45 (95% CI, 1.23-1.71), I ² = 96], and hemorrhagic stroke [aHR 2.22 (95% CI, 1.11-4.42), I ² = 96] at maximum follow-up compared to non-sepsis hospital or population control. The increased risk was robust to several sensitivity analyses. Factors that were significantly associated with increased hazards of stroke were: advanced age, male gender, diabetes mellitus, hypertension, coronary artery disease, chronic heart failure, chronic kidney disease, chronic obstruction pulmonary disease, and new-onset atrial fibrillation. Only diabetes mellites [aHR 1.80 (95% CI, 1.12-2.91)], hypertension [aHR 2.2 (95% CI, 2.03-2.52)], coronary artery disease [HR 1.64 (95% CI, 1.49-1.80)], and new-onset atrial fibrillation [aHR 1.80 (95% CI, 1.42-2.28)], were associated with > 50% increase in hazards.

Conclusion: Our findings showed a significant association between sepsis and a subsequent risk of cerebrovascular events. The risk of cerebrovascular events can be predicated by patient and sepsis-related baseline variables. New therapeutic strategies are needed for the high-risk patients.

Keywords: cardiovascular events; cerebrovascular events; meta-analysis; sepsis; stroke.

FIGURE 1

Prisma flowchart showing studies selection process.

FIGURE 2

Forest plot for meta-analysis of cohort studies comparing effect of sepsis with no sepsis on stroke. The association between sepsis and each stroke subtype at maximum reported follow-up is pooled and displayed. Weight are from random effect analysis. CI, confidence interval; HR; hazard ratio.

FIGURE 3

Forest plot for cumulative random effects meta-analysis of cohort studies comparing effect of sepsis with no sepsis on stroke. The association between sepsis and each stroke subtype at maximum reported follow-up is pooled and displayed. The studies are shorted by study size, starting with the largest sized. CI, confidence interval.

FIGURE 4

Risk factors for stroke following sepsis hospitalization. FM, fixed effect model; RM, random effect model; CI, confidence interval; HR; hazard ratio.