An observational study of feeding practice in ventilated patients with spinal cord injury

Cervical spinal cord injuries (CSCI) are frequently complicated by respiratory failure requiring prolonged invasive mechanical ventilation in the intensive care unit (ICU). Providing adequate nutrition may be an important factor in the recovery of respiratory muscle function for ventilator weaning. The aim of this study was to examine the practical implementation of a multi-disciplinary approach to nutrition combining indirect calorimetry and the modified Harris Benedict equation to achieve target rates of nutrition for patients with CSCI.

A retrospective observational study was performed by collecting data of normal nutrition practice in a series of adult ventilated patients with CSCI admitted between 2014 and 2017 to the ICU. Administered calories by enteral nutrition and measured total energy expenditure (TEE) by indirect calorimetry were recorded for the first 3-weeks. Calculations were performed to measure the adequacy of calorie administration relative to TEE. Average daily temperatures and minute ventilation were recorded to support evidence for hypermetabolism. TEE estimates using the modified Harris Benedict equation were compared to indirect calorimetry measures for each patient.

Sixteen patients were included in the study. Calorie administration rapidly increased to a plateau of 2300-2400 kcal per day over the first four days. The median measured TEE by indirect calorimetry was initially high, starting at 3472 kcal/day and decreasing to 2784 kcal/day at three weeks. However, there was wide variation in energy expenditure amongst patients. Average daily temperatures and minute ventilation were high supporting hypermetabolism. Adequacy of calorie administration to TEE ranged between 76 and 86 percent through the study period. There was relatively wide variation when comparing estimates of TEE using the modified Harris Benedict equation versus indirect calorimetry.

A multi-disciplinary approach to nutrition in ventilated patients with CSCI, incorporating indirect calorimetry, achieves target rates of nutrition. Our findings suggest that a hypermetabolic response may be common but variable in this population and subsequently both under and over feeding may be a risk if nutrition targets are not guided by indirect calorimetry. Further prospective research using protocoled indirect calorimetry measures would be beneficial to ascertain accurate energy requirements for this group of patients and also determine whether feeding to target influences weaning off the ventilator.