Micronutrient supplementation for critically ill adults : a systematic review of the evidence

[摘要] BackgroundCritical illness is associated with increased production of reactive oxygen speciesand oxidative stress, and low levels of most micronutrients with resultant diminishedendogenous antioxidant defences. Micronutrient supplementation is thought to bebeneficial to the critically ill patient by ameliorating oxidative stress and by improvingclinical outcome.ObjectivesThis systematic review assessed the effects of micronutrient supplementation onadults recovering from critical illness. Primary outcomes included clinical endpoints[mortality, infectious complications, length of intensive care unit and hospital stay(LICU and LOS)]. Secondary outcomes included descriptions of practice issues,micronutrient status, morbidity, course of the acute phase response and oxidativestress.Search strategyAn electronic bibliographic database search was carried out, bibliographies ofretrieved articles were reviewed and personal files searched to obtain additionalcitations. Databases were searched from inception until 29 February 2008.Selection criteriaRandomized controlled trials (RCTs) of micronutrient supplementation (by any route)in adult critically ill patients, given in addition to their routine care, were included.Data collection and analysisTwo authors independently extracted data and assessed trial quality. For the primaryoutcomes the random-effects model was used to estimate overall relative risk /mean difference and effect size due to the presence of study heterogeneity.Selected exploratory analyses were undertaken. Differences at the level of p<0.05was considered to be statistically significant. The secondary outcomes were sparseand variably recorded such that this data was not formally aggregated.Main resultsFifteen RCTs involving 1714 participants and 18 RCTs involving 1849 participants wereincluded for the primary and secondary objectives respectively. The quality of theRCTs, as reported, was disappointing, particularly for allocation concealment.Fourteen trials (n=1468) of micronutrient supplementation showed a statistically significant reduction in overall mortality [relative risk (RR) 0.78, 95% confidenceinterval (CI) 0.67-0.90, I2=0%, p=0.0009]. An asymmetrical funnel plot necessitatescaution when directly interpreting these results. Six RCTs (n=1194) indicated astatistically significant reduction in 28 day mortality (RR 0.75, 95% CI 0.63-0.88, I2=0%,p=0.0006) (symmetrical funnel plot). Micronutrient supplementation in this systematicreview was not associated with a reduction in infectious complications, LICU or LOS.In sub-group analyses, single nutrients were associated with borderline statisticalsignificance (RR 0.82, 95% CI 0.66-1.01, I2=0%, p=0.06) in terms of mortality, whist asensitivity analysis of combined micronutrients indicated a significant reduction inmortality (RR 0.69, 95% CI 0.54-0.90, I2=2%, p= 0.006). This review did not find clearevidence that parenteral is superior to enteral administration in terms of clinicaloutcomes. The secondary outcomes confirmed that timing, duration and dosing arekey factors to ensure optimal clinical benefit.ConclusionThis review does suggest potential benefit of micronutrient supplementation incritically ill adults for some clinical outcomes (especially mortality), but also highlightsthat caution is warranted as nutrient interactions and risk of toxicity are not clearlydefined in critical illness. More large multi-centre randomized trials are necessary toassess the effects of different types and doses of micronutrient supplementation inselected groups of patients with different types of critical illness.

[发布机构] Stellenbosch University