Meta-analysis of the effects of supplemental parenteral nutrition on prognosis of critically ill patients

Objective To compare the effects of supplemental parenteral nutrition (SPN) and enteral nutrition (EN) on prognosis of critically ill patients in intensive care unit (ICU) using meta-analysis. Methods Foreign language databases including PubMed, Embase, Cochrane Library, and Cochrane Central Register of Controlled Trials were retrieved with the search terms of " supplemental parenteral nutrition, parenteral nutrition, enteral nutrition, critically ill" , and Chinese database SinoMed database was retrieved with the search terms of "补充性肠外营养,肠外营养,肠内营养,重症" to obtain the publicly published randomized controlled trials about the effects of SPN and EN supportive treatment on prognosis of critically ill patients in ICU from the establishment of each database to December 2018. The Google Scholar was retrieved for supplement. The outcome indexes included the infection rate, anti-infection time, antibiotic-free time, ICU overall mortality, overall mortality during hospitalization, mechanical ventilation time, length of ICU stay, and length of hospital stay. RevMan 5.3 and Stata 14.0 statistical software were used to conduct a meta-analysis of eligible studies. Results A total of 794 patients were included in 8 studies, including 387 patients in SPN group who received SPN and EN and 407 patients in EN group who only received EN. The bias risks of the eight studies included were uncertain. Compared with that of EN group, the infection rate of patients in SPN group was significantly decreased (relative risk=0.79, 95% confidence interval=0.66-0.94, P<0.01). However, there were no statistically significant differences in ICU overall mortality, overall mortality during hospitalization, mechanical ventilation time, length of ICU stay, and length of hospital stay of patients between SPN group and EN group. The subgroup analysis showed that the risks of bias of studies and follow-up time might be sources of the heterogeneity of mechanical ventilation time. There was publication bias in ICU overall mortality (P<0.05), while no publication bias in the other outcome indexes (P>0.05). Conclusions SPN supportive treatment can decrease the infection rate of critically ill patients in ICU, but it has no obvious influences on overall mortality, mechanical ventilation time, and length of hospital stay.