September 2012 E-journal Club


We hosted a wonderful group of trainees in September, who hailed from Nashville, TN; Philadelphia, PA and Cedar Rapids, IA.  Our visitors were treated to some glorious end of summer weather – any visitors in the spring and fall can understand my fondness for Thomas Jefferson’s city. Our journal club this month deals with feeding tube position in the adult ICU patient.

September Citation:

Davies AR, Morrison SS, Bailey MJ, et al.  A multicenter, randomized controlled trial comparing early nasojejunal with nasogastric nutrition in critical illness.  Crit Care Med. 2012;40(8):2342-2348.


This was a multicenter, randomized, unblinded study comparing nasogastric (NG) with nasojejunal (NJ) enteral nutrition (EN) in a mixed medical-surgical ICU population.  Patients that were allocated to NJ placement received feeding through a spontaneously migrating frictional tube (Tiger Tube, Cook Critical Care).  The primary outcome was energy delivery from enrollment until the end of the ICU stay, calculated as proportion of estimated energy needs.  Secondary outcomes were ventilator-associated pneumonia, duration of mechanical ventilation, hospital length of stay and mortality.  Tube feeding formula selection was at the discretion of the clinical team and energy needs were estimated by the same method in both groups.

Although the study was not blinded, the patient’s energy requirements were calculated before randomization and the interpretation of chest films were conducted in a blinded fashion with feeding tube position concealed.

The study nutrition protocol was continued until 28 days after enrollment, death, ICU discharge, or until the patient began eating, whichever came first.  Nutritional management after the tube feeding protocol ended was at the discretion of the clinical team. The patients were followed until hospital discharge.

Inclusion and Exclusion Criteria were:

Inclusion criteria:

Initial criteria were: age ≥18 yrs; ICU stay <48 hrs; receiving mechanical ventilation of ≥48 hrs anticipated duration; receiving narcotic infusion above specified amounts, elevated gastric residuals (single measurement >150 mL or 12 hr cumulative volume >500 mL). Due to slow initial recruitment, inclusion criteria were expanded to age ≥16 yrs, ICU stay <72 hrs, and receiving narcotic infusion at any dose.

Exclusion criteria:

Previous anatomy – altered upper gastrointestinal surgery, gastric malignancy, esophageal varices, current peptic ulceration, mechanical bowel obstruction, presence of gastrostomy or jejunostomy, nutrition therapy prior to ICU admission, severe coagulopathy, pregnancy, suspected brain death, death expected within 24 hrs, and suspected hypoxic-ischemic encephalopathy.

Major Results reported by authors:

Of the 3,623 patients screened there were 680 patients that met inclusion, but not exclusion criteria.  Ultimately only 181 patients could be enrolled and 180 patients were analyzed, with 91 NJ and 89 NG.  In the NJ group, 79 patients (87%) achieved small bowel placement after a median of 15 hours, with 21% of tubes in the duodenum and 66% in the jejunum.  NJ tubes were ultimately placed in 9% of the NG group patients due to persistent EN intolerance.

Metoclopramide was administered to a similar number of patients in both groups (82 NJ, 81 NG).  Significantly more patients in the NJ group received erythromycin compared to the NG group (87% NJ, 55% NG, P = .001), in part because the protocol for NJ tube advancement included erythromycin.

There was no significant difference between the two groups in the proportion of estimated energy needs delivered during the entire study period, or during the first 10 days.  There was also no significant difference in VAP, rates of vomiting, “witnessed aspiration”, abdominal distension, diarrhea, accidental tube displacement, ICU stay, hospitalization, mortality or major GI hemorrhage.  The incidence of minor hemorrhage was significantly greater in the NJ group compared to the NG group (13% NJ, 3% NG, P = 0.02).  Subgroup analysis based on APACHE II scores, or those achieving earlier NJ access, revealed no significant outcome differences.

Author’s Conclusions:

Early nasoje­junal nutrition did not increase energy delivery or reduce the occurrence rate of pneumonia in ICU patients with mildly elevated GRVs who were already receiv­ing nasogastric nutrition.

NJ feeding with a self-advancing feeding tube increased the risk of minor gastroin­testinal hemorrhage.  Routine placement of NJ tubes in ICU patient’s with mildly elevated gastric residuals is not recommended.


The strengths of this study include the fact that it was a multicenter study and decisions regarding feeding amount, formula. and diagnosis of pneumonia were done without knowing what group the patient was in.

The authors point out a number of the study limitations in the discussion section, including the fact that bedside caregivers and physicians were not blinded to study group.  One major limitation that the authors noted was the fact that 90% of the patients randomized to receive NJ tube had already received NG feeding before enrollment into the study.  Patients were in the ICU an average of 42 hours at randomization, so it is possible that some patients could already have had pneumonia “brewing” when they entered the study.

The authors also point out that the use of physiologic volumes of gastric residuals (>150 mL) as an indicator for study entry meant that they did not only study those patients with significant impairment to gastric emptying.

Our group also noted that only approx. 27% of the patients that met the inclusion/exclusion criteria were ultimately able to be enrolled/complete the protocol, raising the possibility that this sample may not truly represent the average ICU patient.  It is also worthwhile to consider that these study results may only reflect NJ feeding with the type of tube and NJ advancement protocol used in this study.  Some hospitals are able to rapidly achieve NJ placement, while NJ placement in this study required a median of 15 hours, with only 76% of the patients having achieved jejunal position after 48 hours.

An early version of the self-advancing feeding tube was withdrawn from the market due to problems related to GI bleeding.  It is notable that although the current self-advancing tube did not cause significantly greater major GI bleeding, the rate of minor GI bleeding was significantly increased in a population without ulcer disease, severe coagulopathy or varices.

Our Take Home Message (s)

1.   The use of a self-advancing NJ tube does not appear to increase nutrition delivery or improve outcomes in a population of ICU patients with mild (perceived) feeding intolerance.

2.   Frictional self-advancing feeding tubes appear to increase minor GI bleeding in the ICU population.

Other News on the UVAHS GI Nutrition Website: (

Upcoming Webinars for Spring 2012:

Tuesday, October 23:  Growing Power for Preterm and High Risk Infants–Patti Perks, MS, RD

–Tuesday, November 13:  Nutritional Anabolics–Joe Krenitsky, MS, RD

–Tuesday, December 11:  Information Mastery:  Evidence-Based Medicine in the Real World–David Slawson, MD

Check out What’s New:

–“Nutrition Support Blog”

–“ Resources for the Nutrition Support Clinician

Latest Practical Gastroenterology article:

–Krenitsky J.  Nutrition and Nutraceuticals for Muscle Maintenance and Recovery: Hero or Hokum? Practical Gastroenterology 2012;XXXVI(9):27.


Joe Krenitsky MS, RD

Carol Rees Parrish MS, RD

PS – Please feel free to forward on to friends and colleagues.