University of Virginia Health System
Nutrition Support E-Journal Club
November and December have been a bit of a blur due to our annual state enteral and parenteral association meeting (VASPEN), Thanksgiving with time off for travel or entertaining guests, and preparations for the next round of holiday cheer.
Despite all of the holiday hubris, we did take time to discuss a new article that is hot off the press - here at UVA we just could not resist the opportunity to dissect a new study about gastric residuals.
Poulard F, Dimet J, Martin-Lefevre L, et al. Impact of Not Measuring Residual Gastric Volume in Mechanically Ventilated Patients Receiving Early Enteral Feeding: A Prospective Before-After Study. JPEN J Parenter Enteral Nutr. 2009 Oct 27.
This was a prospective, before-after study investigating the effect of instituting an enteral feeding protocol that discontinued monitoring gastric residual volume (GRV) in a med-surg ICU. The study compared one year of patient outcomes (enteral feeding delivery, vomiting and ventilator-associated pneumonia [VAP]) in a medical-surgical ICU before and after the change in protocol (July 2004-June 2005 was the control period and July 2005-June 2006 was the study period).
Patients received continuous enteral nutrition (EN) of a polymeric 1.2 calorie/mL formula (Isosource) via a 14-french nasogastric tube. Chest radiographs were checked each day to monitor the position of the feeding tube. Feedings were started at 25mL/hr and increased by 25mL/hr every 6 hours to a goal of 85mL/hr (presumably the goal for all patients). The study period was 7 days after the start of EN. Feeding intolerance was defined as GRV > 250mL or vomiting in the "before" group and by vomiting in the "after" group. Patients that experienced feeding intolerance received IV erythromycin (250mg q 6 hours), which was continued until patients tolerated feedings > 48 hours. Backrest elevation was maintained at > 45 degrees, and episodes of prone position were recorded. In the before group, GRV that were < 250 mL were returned to the patient. A suspected diagnosis of VAP was confirmed by broncoscopy and quantitative culture.
Inclusion and Exclusion Criteria were:
Inclusion criteria were patients that received EN via a nasogastric tube within 48 hours after the start of mechanical ventilation.
Exclusion criteria were those with a history of esophageal or gastric surgery; bleeding from the esophagus, stomach, or bowel; administration of prokinetic agents within 48 hours before starting EN; EN via a jejunostomy or gastrostomy; acute pancreatitis; and pregnancy.
Major Results reported by authors:
There were 205 patients that met inclusion criteria: 102 in the before group that measured GRV (July 2004-June 2005) and 103 in the after group that did not measure GRV (July 2005-June 2006). The groups were similar in terms of patient characteristics, disease states and severity, with no statistically significant differences between the groups.
The median volume of enteral feeding per day was higher in the intervention group (1489 mL; IQR, 1349-1647 mL) than in the control group (1381 mL; IQR, 1151-1591 mL; P = .002). The benefit of increased enteral formula delivery was significantly different only during the first 2 days of feeding. The mean cumulative erythromycin dose over the 7-day study period was higher in the control group (1593 ± 1775 mg) than in the intervention group (888 ± 1515 mg) (P < 0.05).
Feeding intolerance (GRV over 250mL and/or emesis) was significantly greater in the before group than feeding intolerance (just emesis) in the after group 46.1% vs 26.2% respectively (p = 0.004). However, the rate of vomiting was not significantly different between the 2 groups, nor was there a significant difference in the incidence of VAP between the 2 groups.
The primary limitation of this study is the non-randomized, unblinded, before-after design which increases the chance that bias (such as from the passage of time and new treatment methods) can be introduced. Positive aspects of this study include the objective and quantitative diagnosis of VAP, and the relatively large number of subjects that were enrolled in each group, and routine checks of feeding tube position during the study, and monitoring of backrest position during the study. A notable aspect of this study is that it is the first prospective controlled trial that has reported outcomes related to a policy that completely discontinues the use of checking GRV.
At first glance the primary benefit of abolishing the practice of checking GRV appears modest: an increase in formula delivery in the first 2 days of feeding, with an overall median increase of formula delivery of only 108 mL (130 calories) per day and decreased erythromycin administration. Additionally, the difference in feeding tolerance between the groups does not appear to be a fair comparison because intolerance in the before group was defined as a high GRV or emesis, while in the after group intolerance was defined only as emesis. However, our discussion of these points centered on the fact that in the clinical setting feeding intolerance is frequently defined as an increased GRV, which can lead in many settings to the abandonment of EN and the start of parenteral nutrition. We also noted that although the before group had increased prokinetic medication provided, they had no decreased incidence of emesis. This suggests that prokinetics administered due to elevated GRV may only be treating the GRV, and do not necessarily translate into patient benefit. Our consensus was that the lack of any difference in emesis or outcomes between the two groups after stopping the practice of checking GRV was the most "significant" result of this study.
Our group also discussed the fact that although the practice of checking GRV made a certain amount of intuitive sense as a putative safety measure back when there was no data about enteral feeding, it is interesting to note that there are NO prospective data to support the use of GRV. The data that we do have (primarily cohort) all suggest that GRV checks are not a valid marker of enteral feeding tolerance.
Our Take Home message:
This study adds further support to the notion that the practice of checking GRV is neither a sensitive, nor specific indicator for enteral feeding tolerance. It is possible that the use of GRV as an indicator of feeding tolerance may unnecessarily limit enteral nutrition and lead to overuse of prokinetic medications in critically ill populations. Multi-center, randomized trials to test the validity of GRV in different critically ill populations is warranted, although currently available evidence does not support the practice of checking gastric residuals during enteral nutrition.
See the latest Practical Gastroenterology article available at:
Corbello J, Rosner M. Intradialytic Total Parenteral Nutrition (IDPN): Evidence-Based Recommendations. Practical Gastroenterology 2009;XXXIII(12):13.
Information on our 2010 Training Programs is now posted on the website:
- March 13-14, 2010-Weekend Warrior Mini-Traineeship in Charlottesville
- Spring week-long traineeship dates
- 2010 Webinar topics and dates
- September 11-12, 2010-Weekend Warrior Mini-Traineeship in Toronto, Canada!
Joe Krenitsky MS, RD
Carol Rees Parrish MS, RD
PS - Please feel free to forward this on to friends and colleagues.