Structured triglycerides versus physical mixtures of medium- and long-chain triglycerides for parenteral nutrition in surgical or critically ill adult patients: Systematic review and meta-analysis

Abstract

Introduction: New generations of parenteral lipid emulsions combine Long Chain Triglycerides (LCTs) with Medium Chain Triglycerides (MCTs) either by physically mixing MCT- and LCT-containing oils or by using synthetically structured triglycerides (STGs). In order to clarify some open issues relating to their comparative effect, in particular in terms of clinical outcomes, pertinent evidence was systematically identified, reviewed and meta-analyzed.

Methods: PubMed, Scopus, Wanfang Data, China Hospital Knowledge Database and Google Scholar were searched for published clinical trials comparing STGs vs. MCTs/LCTs PN regimens administered over 5-7 days in surgical and/or critically ill patients. Two independent investigators performed screening and data extraction using a predefined list of parameters. Data were pooled using RevMan^® 5.2. Quality of evidence was assessed according to Cochrane's risk of bias tool. Pre-specified high quality (HQ), incremental analyses and a post hoc subgroup analysis were performed.

Results: 21 studies were included. The meta-analysis revealed a significantly better cumulative nitrogen balance (Std. mean difference [95% CI]) (1.34 [0.98-1.7], p < 0.00001), as well as higher values for pre-albumin (24.99 mg/L [6.71-43.27], p < 0.000001), and albumin (1.22 g/L [0.66-1.77] p < 0.0001), while plasma triglycerides were significantly lower (-0.28 mmol/L [-0.41 to -0.15], p < 0.0001) in the STG vs. MCT/LCT group. ALT, AST, and GGT were significantly lower with STGs than with MCTs/LCTs, while for total bilirubin and ALP only a trend was observed. STGs were also associated with a trend to a shorter hospital length of stay (LOS) (-1.74 days [-3.49 to 0.01] p = 0.05). Quality of evidence was affected by an unclear risk of selection bias, mostly due to the lack of detailed reporting (random sequence generation, allocation concealment). For the other domains, most of the weighted information was judged at low risk of bias. HQ estimated effects, incremental and subgroup analyses were consistent with the main analysis.

Conclusions: In postsurgical and/or critically ill patients, the administration of STGs vs. MCT/LCTs was significantly associated with improved protein economy, better liver tolerance and a more efficient triglyceride elimination. With regard to clinical outcomes a strong trend towards reduced LOS was observed for STG patients.

Keywords: Adverse events; Meta-analysis; Nitrogen balance; Parenteral nutrition; Pre-albumin; Structured triglycerides.