The association of plasma homocysteine levels with short-term mortality in sepsis patients: A meta-analysis

Abstract

The association between plasma homocysteine (Hcy) levels and short-term mortality in sepsis patients remains unclear. This meta-analysis aimed to clarify this potential relationship. Following PRISMA 2020 and Cochrane Handbook guidelines, we conducted a comprehensive literature search in the PubMed, Embase, and Web of Science databases up to June 24, 2024. We included cohort studies that assessed the association between plasma Hcy levels and all-cause mortality in adult sepsis patients. Standardized mean differences (SMDs) and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using a random-effects model to account for potential heterogeneity. Nine cohort studies involving 771 sepsis patients were included. Overall, no significant difference in plasma Hcy levels was observed between survivors and non-survivors (SMD: -0.23, 95% CI: -0.84 to 0.37, P = 0.45), with substantial heterogeneity (I² = 86%). Subgroup analysis revealed lower plasma Hcy levels among survivors in Chinese patients (SMD: -1.56, 95% CI: -1.98 to -1.13, P < 0.001) but not in non-Asian patients. Plasma Hcy levels were not significantly associated with all-cause mortality (OR per 1-unit increment: 1.03, 95% CI: 0.95 to 1.11, P = 0.51), with notable heterogeneity (I² = 72%). However, a significant association was found in Chinese patients (OR: 1.09, 95% CI: 1.03 to 1.15, P = 0.003), but not in non-Asian patients. In conclusion, plasma Hcy levels were not generally associated with short-term mortality in sepsis patients. However, significant associations were observed in Chinese patients, suggesting potential ethnic differences that warrant further investigation.

Figure 1.

PRISMA flowchart of database search and study identification.

Figure 2.

The forest plots for this meta-analysis illustrate the differences in plasma homocysteine levels at enrollment between survivors and non-survivors of sepsis, divided into several analyses. (A) Overall meta-analysis; (B) Subgroup analysis according to the study countries; (C) Subgroup analysis according to the diagnostic criteria for sepsis.

Figure 3.

The forest plots for this meta-analysis examine the difference in plasma homocysteine levels at enrollment between survivors and non-survivors of sepsis, broken down by specific subgroup analyses. (A) Subgroup analysis according to the mean age of the patients; (B) Subgroup analysis according to the proportion of men.

Figure 4.

The forest plots for this meta-analysis illustrate the differences in plasma homocysteine levels at enrollment between survivors and non-survivors of sepsis, divided into several analyses. (A) Subgroup analysis according to the follow-up durations; (B) Subgroup analysis according to the analytic models.

Figure 5.

The forest plots for this meta-analysis display the odds ratios for the association between plasma homocysteine levels (per 1-unit increment) at enrollment and all-cause mortality in patients with sepsis. (A) Overall meta-analysis; (B) Subgroup analysis according to the study countries.

Figure 6.

Funnel plots for meta-analyses. (A) Funnel plots for the meta-analysis of the difference of plasma Hcy between survivors and non-survivors of sepsis; (B) Funnel plots for the meta-analysis of the OR for the association between plasma Hcy (per 1-unit increment) and mortality of patients with sepsis. Hcy: Homocysteine; OR: Odds ratio.