Impact of refeeding syndrome on short- and medium-term all-cause mortality: a systematic review and meta-analysis.

Abstract

BACKGROUND\nThe refeeding syndrome has been described as a potentially life-threatening complication of re-nutrition. However, moving from single reports to larger population studies, the real impact of refeeding syndrome on all-cause mortality is still unknown.\n\n\nMETHODS\nPubMed/Medline, EMBASE, Cochrane library and CINAHL databases were systematically searched until September 2020 for studies reporting mortality rates in patients who developed the syndrome at re-nutrition, compared to those who did not develop it. Effect sizes were pooled through a random-effect model.\n\n\nRESULTS\nThirteen studies were finally considered in the meta-analysis, for a total of 3846 patients (mean age 64.5 years; 58% males). Pooled data showed a non-significant trend toward an increased short-term (≤ 1 month) mortality in patients developing the refeeding syndrome (OR=1.27, 95% CI 0.93-1.72), mostly driven by studies in which re-nutrition was not prescribed and supervised by a nutritional support team (p=0.01 at subgroup analysis) and by studies published in earlier years (p=0.04 at meta-regression). When examining medium-term (≤ 6 month) mortality, an overall statistical significance towards higher risk was observed (OR=1.54, 95% CI 1.04-2.28).\n\n\nCONCLUSION\nThis was the first meta-analysis that specifically assessed the impact of refeeding syndrome on mortality. Our results suggested a non-significant trend towards increased mortality in the short-term, but a significantly increased mortality in the medium-term. The supervision/management of the refeeding process by a nutrition specialist might be a key factor for the limitation of this mortality excess.