• Energy Research

Interruption of enteral nutrition (EN) in the intensive care unit (ICU) occurs frequently for various reasons including feeding intolerance and the conduct of diagnostic and therapeutic procedures. However, few studies have investigated the details of EN interruption practices including reasons for and duration of interruptions. There is no standard protocol to minimize EN interruptions.This is a retrospective review of 100 patients in the ICU staying more than 72 h and receiving EN in a 12-bed, medical/surgical ICU in a tertiary care center in 2013. Data collected include total time designated for EN; the number of EN interruption episodes; reason for each interruption categorized as diagnostic study, therapeutic intervention, or gastrointestinal (GI) event, and their individual subcategories; duration of each interruption; and the presence of written orders for interruptions.One hundred patients staying in the ICU for at least 72 h and receiving EN were included. There were 567 episodes of EN interruption over a median ICU length of stay of 17.1 (interquartile range 8.0-22.0) days. There were a median of three EN interruption episodes per patient. EN interruption was performed for undetermined reasons (166 episodes, 29%), airway manipulation (103 episodes, 18%), GI events (78 episodes, 14%), and intermittent dialysis (71 episodes, 13%). Median duration of EN interruption in all patients was 5.5 (3.0-10.0) h. The cumulative interruption time corresponds to 19% of the total time designated for EN. Duration of EN interruption varied according to reason, including airway manipulation (9.0 [5.0-21.0] h), tracheostomy (9.5 [7.5-14.0] h), and GI events (6.5 [3.0-14.0] h). The average calorie deficits due to interruptions were 11.5% of daily target calories. Only 60 episodes (12%) had clear written orders for interruption.Based on this single-center retrospective chart review, interruption of EN in the ICU is frequent, reasons for and duration of interruption varied, and airway procedures are associated with a relatively longer duration of interruption. Documentation and orders were frequently missing. These results warrant development of a protocol for EN interruption.