Identifying the factors that influence energy deficit in the adult intensive care unit: a mixed linear model analysis

Critically ill patients frequently receive inadequate nutrition support as a result of under- or overfeeding. Malnutrition in intensive care unit (ICU) patients is associated with increased morbidity and mortality. The present study aimed to identify the significant factors that influence energy deficit in the ICU.

ICU patients with a length of stay of ≥3 days were studied for 30 days over two consecutive years at a large university teaching hospital. Fifty-six Patients were studied, with a total of 530 records of feeding days. Information was collected for: day when feed initiated, age, gender, length of stay, Acute Physiological and Chronic Health Evaluation score (APACHE II), fed within 24 h, speciality, type of ventilation, feeding route, outcome (survived/died), diarrhoea (yes/no), aspirate volume, dietitian observed nutritional status (malnourished/not), sedation, estimated energy requirements and energy received. Mixed linear models for longitudinal data were used with energy deficit (energy received - energy requirements) as the dependent variable.

Factors that were found to have a significant association with energy deficit were: day feeding was initiated (P<0.001), whether fed within 24 h (P<0.001) and whether sedated (P<0.001). Furthermore, three combined effects were found: ventilation mode and aspirate volume (P<0.007), fed within 24 h and ventilation mode (P<0.001), fed within 24 h and sedation (P<0.017).

The number of days after feeding was initiated, initiation of feeding within 24 h and sedation have been identified as factors that predict energy deficit during ICU stay. Efforts to initiate feeding as soon as possible and minimise interruptions to feeding may reduce energy deficits in these vulnerable patients.