Is omega-3 fatty acids enriched nutrition support safe for critical ill patients? A systematic review and meta-analysis

Abstract

Objective: To systematically review the effects of omega-3 poly unsaturated fatty acids (FA) enriched nutrition support on the mortality of critically illness patients.

Methods: Databases of Medline, ISI, Cochrane Library, and Chinese Biomedicine Database were searched and randomized controlled trials (RCTs) were identified. We enrolled RCTs that compared fish oil enriched nutrition support and standard nutrition support. Major outcome is mortality. Methodological quality assessment was conducted based on Modified Jadad's score scale. For control heterogeneity, we developed a method that integrated I2 test, nutritional support route subgroup analysis and clinical condition of severity. RevMan 5.0 software (The Nordic Cochrane Centre, Copenhagen, Denmark) was used for meta-analysis.

Results: Twelve trials involving 1208 patients that met all the inclusion criteria. Heterogeneity existed between the trials. A random model was used, there was no significant effect on mortality RR, 0.82, 95% confidence interval (CI) (0.62, 1.09), p = 0.18. Knowing that the route of fish oil administration may affect heterogeneity, we categorized the trials into two sub-groups: parenteral administration (PN) of omega-3 and enteral administration (EN) of omega-3. Six trials administered omega-3 FA through PN. Pooled results indicated that omega-3 FA had no significant effect on mortality, RR 0.76, 95% CI (0.52, 1.10), p = 0.15. Six trials used omega-3 fatty acids enriched EN. After excluded one trial that was identified as source of heterogeneity, pooled data indicated omega-3 FA enriched EN significant reduce mortality, RR=0.69, 95% CI [0.53, 0.91] (p = 0.007).

Conclusion: Omega-3 FA enriched nutrition support is safe. Due to the limited sample size of the included trials, further large-scale RCTs are needed.

Figure 1

Literature searching and selection.

Figure 2

Effect of Omega-3 FA administration on mortality of severe patients.

Figure 3

Sensitivity analysis: high quality trials.

Figure 4

Heterogeneity Analysis: Linear Regression Curve for Impact on Mortality and APACHE III Scores: The horizontal axis shows the APACHE III scores, and the vertical axis shows the mortality (%). According to the data of ALI studies, a linear regression model is set up based on the APACHE III scores and the mortality. The regression equation is expressed by the straight line in the figure.

Figure 5

Heterogeneity Analysis: Observed/Expected Ratio (O/E Ratio) of Acute Lung Injury Studies: The vertical axis shows the O/E Ratio. The O/E Ratio of both Omega-3 group (0.48, 95% CI: 0.34–0.67) and control group (0.30, 95% CI: 0.19–0.46) in Rice T. et al. [13] study are far below 1, which indicate that the patients in the Rice T. et al. study are in significantly more severe condition, but have significantly lower mortality.