Discussion

The meta-analyses from ten RCTs revealed that an enteral immunomodulatory diet (omega-3 fatty acid, γ-linolenic acid and antioxidant supplementation) significantly shortened duration of mechanical ventilation and ICU stays, extended ICU-free days, reduced SOFA and MOD score and overall 28 days mortality in critically ill patients.
Our finding showed that enteral immunomodulatory containing omega-3 fatty acid, γ-linolenic acid and antioxidant improved the respiratory functions in ICU patients. Our results are similar to a review of 6 clinical trials (Lev and Singer, 2012) in ALI/ARDS patients concluding that the use of formula enriched with omega-3 and GLA could improve oxygenation and clinical outcomes. The n-3/n-6 ratio of fatty acids could have important role in the alveolar cytokines release (Lev and Singer, 2012). A mechanisms for a protective effect of omega-3 fatty acids in the lungs might be resulting from the products of EPA and DHA which resolves mediators, including resolvins, protectins and maresins found within the lung and circulation (6). Resolving these mediators regulates neutrophil infiltration, cytokine production, clearance of inflammatory leucocytes and inflammatory response, which ultimately helps to maintain the integrity of the lung membrane (Lemoine et al., 2019).
Evidence support the relationship between imbalanced oxidant homeostasis and lung disease showing a trend for potential benefit of antioxidant supplementation in respiration function (Tashakkor et al., 2011). Actually, antioxidant supplementation increases antioxidant capacity like reduced glutathione levels, which subsequently decreases oxidative stress (Mudway et al., 2004).
The SOAF and MOD score evaluates organ failure in critically ill patients (Zygun et al., 2006). This study showed that immunomodulatory diet reduced organ failure. This finding is consistent with the results of Shirai et al (Shirai et al., 2015) and Elamin et al (Elamin et al., 2012) where these improvements in patient status lead to decreased duration of ICU stays and increased ICU-free days. On the other hand, this study showed significant reduction in duration of mechanical ventilation and increased PaO2/FiO2 and ventilator free days in ARDS patients. These results have been well illustrated in Shirai (Shirai et al., 2015) and Gadek (Gadek et al., 1999) studies.
Another review claimed that the beneficial effect of this immunomodulatory formula related to its omega-3 fatty acids composition and adverse outcome in critically ill patients is due to excess production of proinflammatory cytokines and eicosanoids from other polyunsaturated fatty acids (Das, 2013). When the balance between proinflammatory and anti-inflammatory molecules is upset as in critically ill patients, it would lead to persistence of inflammation and progressive cell/tissue and organ damage (Das, 2013). In sepsis, ARDS, ALI and other systemic inflammatory conditions, not only an extend in proinflammatory molecules such as IL-6, TNFα, but also a decrease in the production and action of anti-inflammatory molecules such as IL-4, IL-10 occurrs (Das, 2013). Therefore, supplementation with formula containing omega-3 fatty acid, γ-linolenic acid and antioxidant may help to maintain the balance between proinflammatory and anti-inflammatory conditions.
According to previous meta-analysis (Li et al., 2015), the enteral immunomodulatory diet did not extend ICU- free days, ventilator- free days and any significant reduction in the risk of all-cause mortality. The beneficial effect of this formula on reducing risk of mortality indicated only in patients with high mortality. Our result is inconsistent with these reports because of addition of 3 more studies (Pontes-Arruda et al., 2011, Shirai et al., 2015, Kagan et al., 2015) compared to the previous meta-analysis. Our study considered all the studies conducted in critically ill patients, while previous meta-analysis was restricted to only ARDS and ALI patients. However, the findings from our subgroup analysis contrary to previous meta-analysis showing a decrease in all cause 28 days mortality in ARDS and ALI patients. The significant impact of this immunomodulatory diet on mortality should be interpreted with caution, because not only a small number of studies have examined the effect of this immunomodulatory formula on mortality but also this reduction is based on estimated raw data and we could not adjust the impact of other important confounders such as age, severity of injury and body weight.
Gastrointestinal dysfunction such as diarrhea, dyspepsia and nausea were reported as the main adverse effects of this formula (Li et al., 2015). However, it may result from intolerance response to the rate of continuous enteral infusions in patients (Li et al., 2015). Rice et al (Rice et al., 2011) solved this problem by using a bolus delivery, namely small-volume approach. Also, the components of this supplement are within the tolerance range and are safe.
To test the robustness of the results, we conducted sensitivity analyses. We excluded each individual study, re-analyzing and comparing with the original results. When excluding the trial conducted by Rice et al (Rice et al., 2011), the overall effect for ICU-free days and ventilator-free days became significant suggesting that these outcomes increase in non- ALI patients. When excluding other trials, the results were consistent with a previous report (Li et al., 2015).
Some limitations in this report should be mentioned. First, the sample sizes of the included trials were small. Second, there was one study for some critical diseases like multiple trauma and head and neck cancer. Therefore, we cannot conclude with certainty about supplementation in these patients. Third, for mortality there were few studies (n=9 and sample size=1104) and there was not any clinical trial for long- term mortality (more than 3 months). Also, we reported the risk ratio of all cause 28 days mortality and we could not adjust the impact of other important confounders such as age, severity of injury and body weight. However, due to the randomness of all the included studies, many confounding variables were similar between the intervention and control groups. The effect of this formula requires a longer-term follow up to show its effect on mortality. Forth, we did not have enough data for biochemical and hematological markers and these outcomes could be important to judge about the overall effect of this formula. Finally, according to NutriGrade the meta- evidence for all outcomes were moderate and for level of oxygenation (PaO2/FiO2) was low. Therefore more well-designed studies with larger sample size and long term follow-up of mortality are recommended.
The strength of this study was the adoption of comprehensive search strategy without language and time restrictions. We used Nutrigrade tool to evaluate the quality of each outcome. Compared to previous meta-analysis (Li et al., 2015), our study included more studies and variables including length of hospital stays, duration of ICU stays, duration of mechanical ventilation, level of oxygenation, SOFA and MOD score. In addition, we performed several subgroup analyses to identify the source of heterogeneity