Clinical summary: This meta-analysis suggests that the clinical intervention is effective; however, meta-regression indicates that the benefit inversely correlates with the mortality rate in the control group. While benefit also correlates with year of publication, only the the mortality rate in the control group is an independent predictor. The moderate heterogeneity was explained by variations in the mortality rate in the control groups of the trials (see meta-regression below) and the heterogeneity was 'unimportant' after the rate of outcomes among controls was added to the meta-regression (residual I² = 0%). Accordingly, the two most recent trials both had less than 20% mortality in the control group, and the intervention did not reduce mortality. Alternatively, 'Meta-regression dooes not find modulators on the effect of the intervention."

Meta-regression dooes not find modulators (year of publication, study size, event rate in the control groups) on the effect of the intervention.

The quality of evidence supporting benefit in trials with high mortality is pending due to all trials being unblinded and the benefit being in a small number of subjects studied and an unplanned subgroup analysis. We included randomized controlled trials of treating patients with septic shock with early-goal directed therapy. Additional interpretations may be in the references in the 'Cited by' section below.

• Reconciliation of conclusions with prior meta-analyses (source files)
• Reconciliation of studies included with prior meta-analyses (source files)

Acknowledgement: we acknowledge the essential work by the authors of the prior systematic review(s) listed below.

Methods overview: This repository is an openMetaAnalysis that combines methods of scoping, rapid, and living systematic reviews. This analysis updates one or more previously published review(s)(1).

  Newer studies included are listed in the references. Rationale for newer trials excluded may be listed at the end of the references.

A comparison of studies included in this review compared to prior reviews are in the table, reconciliation of studies included with prior meta-analyses. Studies were abstacted into standardized tables of , and results.

• Methods for openMetaAnalysis

Results: Details of the studies included are in the:

• PRISMA documentation (under construction)
• Reconciliation of studies included with prior meta-analyses (source files) (under construction)
• PICO table (study descriptions) (source files) (under construction)
• Risk of bias table (source data) (under construction)
• Forest plot(s) (source data) (under construction)

Below is the forest plot for the primary outcome. Additional Forest plots may be available.

Below is the network meta-analysis for the primary outcome. Additional network plots may be available.

Below is the metaregression for the primary outcome. Additional metaregression plots may be available.

Below is the network meta-analysis for the primary outcome. Additional network plots may be available.

The Summary of Findings Table (SoF) from the GRADE Profile is below (source files).