Comparative efficacy of antioxidant therapies for sepsis and septic shock in the intensive care unit: A frequentist network meta-analysis

Abstract

Background: Antioxidant therapy is gaining traction in managing sepsis and septic shock, owing to its perceived positive impact on patient outcomes. This study sought to compare the efficacy of five antioxidant therapies (melatonin, vitamin C, vitamin E, selenium, and N-acetylcysteine, both individually and in combination with other compounds such as vitamin B1, hydrocortisone, propolis, and glutamine) in treating sepsis or septic shock in the intensive care unit (ICU).

Methods: The study involved randomized and multi-arm trials with sepsis or septic shock patients using melatonin, vitamin C, vitamin E, selenium, or N-acetylcysteine. Studies were sourced from PubMed, Embase, Cochrane Library, ClinicalTrials.gov, and WHO - Clinical Trials Registry Platform for the frequentist network meta-analysis on 28-day mortality and Sequential Organ Failure Assessment (SOFA) scores. The risk of bias was assessed using the Physiotherapy Evidence Database scale. Therapies were compared directly and indirectly using R software.

Results: The study of 56 trials involving 9,366 patients was included. Bias assessment revealed that 89.3 % of trials achieved excellent or good quality. Based on treatment ranking and pairwise comparisons, melatonin with propolis (SUCRA = 93.29 %) is effective in improving SOFA scores, statistically significant, with no publication bias (p= 0.73). High-dose vitamin C (SUCRA = 83.97 %), vitamin C with vitamin B1 (SUCRA = 78.72 %), and melatonin (SUCRA = 67.03 %) are potential therapies for organ dysfunction. Melatonin (SUCRA = 88.22 %) and high-dose vitamin C (SUCRA = 80.75 %) were the most effective in reducing 28-day mortality rates. However, analysis indicated that the results for 28-day mortality rates were not statistically significant. Also, these results contained publication bias (p= 0.02).

Conclusion: The study offers fresh perspectives on antioxidant therapy treatments for sepsis or septic shock in ICU, emphasizing the combination of melatonin and propolis notably reduces SOFA scores for those patients.

Keywords: Antioxidant; High-dose vitamin C; Intensive care unit; Melatonin; Sepsis; Septic shock.

Fig. 1

Schematic representation of the network meta-analysis workflow using R software, detailing the steps from data extraction to analysis with ‘netmeta’ and ‘netmetabin’ packages.

Fig. 2

PRISMA flow diagram illustrating the study selection process, from identification to inclusion in the meta-analysis, detailing numbers at each stage.

Fig. 3

Descriptive statistics of included studies categorized by (A) geographic location, (B) control group types (C) intervention types, and (D) Summarizing the risk of bias within included studies according to the PEDro scale, including individual criteria assessments. PEDro: Physiotherapy Evidence Database.

Fig. 4

Network graphs of pairwise comparisons for (A) 28-day mortality and (B) SOFA scores among antioxidant therapies, annotated with the number of contributing trials. SOFA: Sequential Organ Failure Assessment; Se ST: selenium standard therapy; Se: selenium high-dose therapy; VTM C: vitamin C; VTM B1: vitamin B1; VTM E: vitamin E; MLT: melatonin; NAC: N-acetylcysteine; HC: hydrocortisone; GLN: glutamine.

Fig. 5

Forest plots illustrating the comparative effects of antioxidant therapies on (A) 28-day mortality and (B) SOFA scores, with corresponding odds ratios or mean differences and confidence intervals. SOFA: Sequential Organ Failure Assessment; Se ST: selenium standard therapy; Se: selenium high-dose therapy; VTM C: vitamin C; VTM B1: vitamin B1; VTM E: vitamin E; MLT: melatonin; NAC: N-acetylcysteine; HC: hydrocortisone; GLN: glutamine.

Fig. 6

Rankograms visualizing the probability rankings of antioxidant therapies based on their effectiveness for (A) 28-day mortality and (B) SOFA scores from the network meta-analysis. SOFA: Sequential Organ Failure Assessment. Se ST: selenium standard therapy; Se: selenium high-dose therapy; VTM C: vitamin C; VTM B1: vitamin B1; VTM E: vitamin E; MLT: melatonin; NAC: N-acetylcysteine; HC: hydrocortisone; GLN: glutamine.

Fig. 7

Heatmap of net heat plots assessing the consistency across direct and indirect comparisons for treatments influencing (A) 28-day mortality and (B) SOFA scores. Plcb: placebo; Se: selenium high-dose therapy; MLT: melatonin; NAC: N-acetylcysteine; VTMC: vitamin C; VChd: high-dose vitamin C; VCld: low-dose vitamin C.

Fig. 8

Funnel plot for the assessment of publication bias in studies reporting 28-day mortality, with asymmetry indicative of potential bias. Se ST: selenium standard therapy; Se: selenium high-dose therapy; VTM C: vitamin C; VTM B1: vitamin B1; VTM E: vitamin E; MLT: melatonin; NAC: N-acetylcysteine; HC: hydrocortisone; GLN: glutamine.

Fig. 9

Funnel plot for the assessment of publication bias in studies reporting SOFA scores, demonstrating the distribution of studies around the mean effect size. SOFA: Sequential Organ Failure Assessment. Se ST: selenium standard therapy; Se: selenium high-dose therapy; VTM C: vitamin C; VTM B1: vitamin B1; VTM E: vitamin E; MLT: melatonin; NAC: N-acetylcysteine; HC: hydrocortisone; GLN: glutamine.