University of Virginia Nutrition Support
E-Journal Club
May 2006


We had another great traineeship week, with trainees hailing from Greenville NC, Norfolk VA, and Green Bay WI.  The trainees were treated to Blue Ridge weather at its mornings and warm afternoons. We were able to enjoy a wonderful dinner outside on our night out.  

May Citation: 

Metheny NA, et al.  Tracheobronchial aspiration of gastric contents in critically ill tube-fed patients: frequency, outcomes, and risk factors.  Crit Care Med 2006;34:1007-1015.


This was a prospective descriptive study of 360 patients, conducted in five ICUs in a Midwestern university-affiliated medical center with level I trauma status.  Inclusion criteria were: admission to one of the five ICUs, mechanical ventilation, feeding by a nasal or oral placed feeding tube or a gastrostomy tube, and completed data for all the variables of interest.  Exclusion criteria were pneumonia at the time of enrollment and surgical jejunostomy.

The study objectives were:

  • To describe the frequency of aspiration of gastric contents (defined as pepsin-positive tracheal secretions)
  • The outcomes associated with aspiration
  • The risk factors associated with aspiration and pneumonia

Tracheal secretions obtained during suctioning were collected from 0800 to 2400 hrs (no collections from 2400-0800) on days 1, 2 and 3 of the study, and analyzed for the presence of pepsin.  Presence of pneumonia was determined from a simplified clinical pulmonary infection score (CPIS) on days 1, 2, 3 and 4 of the study.

Major Results reported by authors:

The researchers analyzed 5,857 tracheal secretions for the presence of pepsin; 31.3% of all measurements were positive for pepsin.  Most patients (88.9%) had at least one tracheal secretion positive for pepsin.  Twenty-four percent of the patients were diagnosed with pneumonia (via CPIS) on the first day of the study, and 48% of the patients had pneumonia on day 4 of the study.  The patients who were diagnosed with pneumonia on day 4 had a significantly greater percentage of their tracheal secretions test positive for pepsin (42.2% vs 21.1%, p < 0.001).  A number of factors were significantly associated with more frequent detection of pepsin in secretions including a low backrest elevation (p = < 0.024), emesis (p = 0.007), gastric feedings (p = < 0.009), G-E reflux (p = <0.033), and a Glascow Coma Score < 9 (p = < 0.021).

The most significant independent risk factors for pneumonia were identified as:

1) Frequency of pepsin-positive secretions (p < 0.001)

2) Use of paralytic agents (p = < 0.002) and,

3) A high sedation level (p = < 0.039).

Authors Conclusions:

Aspiration of gastric contents (defined as presence of pepsin in tracheal secretions) is common in critically ill patients, and frequent aspiration is significantly associated with increased incidence of pneumonia.


This is an extensive study that looks at a number of variables.  It will not be possible to do it full justice in a concise manner, so we encourage everyone to read the full text.  The group agreed that the methodology of this study is inspired. In most studies it is not possible to differentiate aspiration of gastric contents from aspiration of oral and pharyngeal secretions.  Previously the only methods available were to label feedings with a tracer such as technetium99m, or add colored microscopic beads into the feeding.  The measurement of pepsin in tracheal secretions appears to be a sensitive measure to detect gastric contents that migrate into the pulmonary tract.  However, there were several limitations to this study that are worth considering:

The first point is that this is not a randomized study, it is observational in nature, and therefore, any associations are just that, they can only imply cause-and-effect.  The second point is that the study did not appear to consider (or control for) the nasal versus oral route for endotracheal intubation, or type of feeding tubes used in the analysis.  Although nasal versus oral route may not affect detection of pepsin, it may affect the incidence of ventilator-associated pneumonia by influencing the development of sinusitis (1). 

This study does highlight both the importance of keeping the head of the bed elevated, AND the practical concern that even in the best of circumstances (study nurse checking daily) that most facilities do not do an impeccable job of keeping the head of the bed elevated (2).

Pepsin was detected in tracheal secretions more often in patients fed with gastric-placed tubes compared to small bowel tubes; increased frequency of pepsin-detection was strongly associated with pneumonia incidence.  However, we did note that the position of feeding tubes was not independently associated with pneumonia occurrence.  Furthermore, it would have been very helpful to include a third group of patients who were not fed to determine how they differ in  "baseline" in frequency of aspiration based on pepsin positive aspirates.

Clinical diagnosis of pneumonia with CPIS has been criticized as being inadequate to differentiate ventilator-associated pneumonia (VAP) from systemic inflammatory response syndrome (SIRS) in trauma patients (3), and the authors themselves admit that this is a potential limitation of the study.  Considering that some aspects of the CPIS score are affected by subjective interpretation, it would be useful to know if the investigator that determined the CPIS score was blinded to the position of the feeding tube and pepsin results.

The author's conclusions include recommendations that consider strategies to reduce aspiration risk when residuals are greater than 200mL, and discuss a small bowel placed feeding tube as a possible option.  However, increased residuals (mean = 41 mL vs 31 mL !) did not significantly correlate with aspiration or pneumonia in this study, and there is no evidence that gastric residuals are a sensitive or specific indicator of aspiration risk.  Although the route of feeding was associated with the detection of pepsin in secretions, it did not correlate with incidence of pneumonia in this study.  There are a number of randomized studies that demonstrate no significant difference between gastric and jejunal feeding in terms of pneumonia incidence, and there is inadequate evidence generated from an observational study that would override evidence from randomized trials.  Finally, it is important to note that 24% of patients had pneumonia on day 1 which might very well have been present prior to initiation of TF. 

Some other points raised during the journal club were that the paper does not provide details regarding how the patients were fed (bolus vs continuous) and that 72 % of patients entered the study on day 1 of enteral feeds, while 28% entered the study after TF had begun.

Observational studies such as this do not provide adequate evidence to make recommendations on clinical policies and practices for improving patient outcome.  The data generated are useful to develop theories for strategies, which then must be tested in a randomized fashion to determine if they actually result in an improvement in patient outcome.

Take home message: 

This study provides useful information on factors associated with aspiration and pneumonia.  The results provide encouragement for the use of pepsin detection to help differentiate between aspiration of gastric contents versus other contributors to pneumonia in future studies.  Observational studies should be used as the basis for hypothesis generation for prospective studies.  This study does not provide adequate evidence to make recommendations on strategies to improve patient outcome during nutrition support due to the multiple sources of bias found in any observational study.


  1. Holzapfel L, Chastang C, Demingeon G, et al.  A randomized study assessing the systematic search for maxillary sinusitis in nasotracheally mechanically ventilated patients. Influence of nosocomial maxillary sinusitis on the occurrence of ventilator-associated pneumonia.  Am J Respir Crit Care Med. 1999;159(3):695-701.
  2. van Nieuwenhoven CA, Vandenbroucke-Grauls C, Van Tiel FH, et al.  Feasibility and effects of the semirecumbent position to prevent ventilator-associated pneumonia: a randomized study.  Crit Care Med 2006;34:396-402.
  3. Croce MA, Swanson JM, Magnotti LJ, et al. The futility of the clinical pulmonary infection score (CPIS) in trauma patients.  J Trauma. 2006;60(3):523-527.

Other News:

1)Check out the latest Practical Gastroenterology article at:  

Scroll down to GI Nutrition on the far left column and click on link
Then scroll down to box with links within the nutrition site
Nutrition Articles in Practical Gastroenterology is in the right column.

Krenitsky J.  Immunonutrition-Fact, Fancy or Folly?  Practical Gastroenterology 2006;XXX(4):47.

Joe Krenitsky, MS, RD                                                                                      

Carol Parrish, RD, MS

PS - Please feel free to forward this on to friends and colleagues.