Mortality, overall and specific infectious complications in parenteral nutrition: systematic review with meta-analyses and trial sequential analysis

Objectives: To examine whether parenteral nutrition is associated with increased risk of mortality or infectious complications in hospitalized patients compared with other nutritional support strategies. Design: Systematic review and meta-analysis. Data Sources: MEDLINE, EMBASE, Cochrane Central, SCOPUS, Clinical Trials and Web of Science were searched for randomized controlled trials and observational studies published until July 2020 exploring the effect of parenteral nutrition on mortality and infectious complications. Eligibility Criteria for Selecting Studies: Randomised controlled trials or observational studies with parallel groups comparing parenteral nutrition with any comparator that included data regarding mortality or infections rate. Two independent reviewers extracted the data and assessed the risk of bias of included studies. Fixed effects meta-analysis were performed to compare the parenteral nutrition group with the control group. Trial sequential analysis was used to identify if current results are enough for definitive conclusions. Results: Of 79 included studies from patients receiving parenteral nutrition versus other strategies, we were able to perform meta-analysis from 64 randomised controlled trials. Parenteral nutrition was not associated with higher risk of mortality (risk relative 1.01, 95% confidence interval 0.94 to 1.07), with low inconsistency across trial results (I² = 26%). On the other hand, parenteral nutrition was associated with higher risk of overall infection (risk relative 1.23, 95% confidence interval 1.11 to 1.36), with low inconsistency across trial results (I² = 26%). Regarding specific infectious complications, parenteral nutrition was specifically associated with abdominal infection, catheter infection and fungaemia, but not pneumonia. The trial sequential analysis shows that there is sufficient data to perform numerical conclusions about mortality, overall and abdominal infectious complications, but not for other specific infectious complications. Most of the included trials were ate high or unclear risk of bias in at least one domain. Conclusions: This study suggests that although parenteral nutrition is not associated with greater mortality in hospitalized patients, it is associated with infectious complications. Through trial sequential analysis could reach definite conclusions about survival and overall infection. Study Registration: PROSPERO registration number CRD42018075599.