Gastric residual volume: end of an era.

GASTRIC DYSMOTILITY IS COMMON IN CRITICALLY ILL patients. The pathophysiology is multifactorial including the severity and etiology of the underlying critical illness, use of narcotic analgesia and other sedatives, decreased blood flow from shock, and use of vasopressors. Gastric dysmotility results in delayed gastric emptying that may place patients at risk of developing complications such as vomiting, aspiration, and ventilatorassociated pneumonia (VAP). To manage this risk, guidelines recommend monitoring gastric residual volumes (GRVs) on an intermittent schedule and holding enteral feedings when residual volumes exceed certain limits. The practice of holding or interrupting enteral feedings for elevated GRV developed from a desire to detect intolerance to enteral feeding early and potentially prevent complications from vomiting or aspiration. Two decades ago, McClave et al reported that 30% of critically ill patients experienced GRV greater than 200 mL compared with none of 20 normal control patients. These data convinced many that 200 mL was a reasonable GRV threshold for interrupting enteral feedings. A decade later, Pinilla et al found similar rates of vomiting with a 250-mL compared with a 150-mL GRV threshold although the 150-mL threshold resulted in more than twice as many enteral feeding interruptions (53% vs 23% of patients). Numerous studies have demonstrated that elevated GRV represents the most common reason for interrupting enteral nutrition and not reaching goal enteral feeding rates. Consequently, McClave et al used calorimetric spheres and food coloring to demonstrate that rates of aspiration and regurgitation did not differ between patients randomized to 200 mL and 400 mL of GRV thresholds. Again, enteral feedings were interrupted significantly more with lower thresholds. Mentejo et al took the concept of higher GRV thresholds further by comparing clinical outcomes of patients randomized to 200vs 500-mL thresholds. Patients managed with higher thresholds received a higher percentage of prescribed enteral nutrition over the first week and reached goal enteral feeding rates faster without experiencing increased rates of VAP. Other clinical outcomes, including duration of mechanical ventilation and ventilator-free days, intensive care unit (ICU) lengths of stay, and ICU and hospital mortality, were also similar. These data prompted many to increase their GRV threshold to between 300 mL and 500 mL or to require additional signs of gastrointestinal intolerance before interrupting enteral feedings. However, it still was not clear that GRVs alone were clinically important, that they were correlated with gastrointestinal intolerances, or that holding enteral feedings for some arbitrary volume provided any protection from feeding complications. Mentec et al found that more than half of critically ill patients who vomited never had a GRV higher than 150 mL, whereas the patients who vomited did so before their GRV had increased to 150 mL (ie, elevated GRVs occurred after vomiting and could not be used to predict vomiting). However, GRVs higher than 500 mL correlated with vomiting but not with increased VAP rates. In addition, GRVs are dependent on caliber, position, and number of openings of the gastric tube and on patient positioning and, as such, lack reliable reproducibility and do not correlate with either abdominal x-ray or with examination findings. Physiologically, the stomach does not empty continuously. A certain volume of gastric content is necessary to stimulate contractions to facilitate emptying, and that volume varies from person to person. As such, an elevated GRV may simply be physiologic, as suggested by a study demonstrating that 80% of critically ill patients who experienced a GRV greater than 200 mL never had a second episode, despite continuing enteral feeding after the first episode. Given the data demonstrating safety of higher GRV thresholds and the uncertainty of their clinical utility, the next logical question was whether monitoring GRVs conferred any clinical benefit. In this issue of JAMA, the clinical trial by Reignier and colleagues provides an answer to this question. The investigators randomized 449 adults receiving enteral nutrition via gastric tubes within 36 hours of initiation of mechanical ventilation, 222 of whom were randomized to a protocol in which GRV was checked every 6 hours, with adjustment of enteral feeding rates if the