 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Carol Lynn Berseth, MD
Mead-Johnson Nutritionals
Evansville, IN 47721-0001
Despite necrotizing enterocolitis (NEC) being a significant contributor to the mortality and morbidity associated with premature birth, no clinical variables have been found to definitively predict the onset of NEC.1,2 Hence, we were very interested to read the article by Cobb et al3 that attempts to correlate gastric residuals and feeding intolerance with a higher risk of NEC. Neonatologists in different regions of practice arbitrarily attribute different gastric aspirate volumes and colors as indicative of feeding intolerance and possible development of NEC. Mihatsch et al,4 for instance, investigated whether 2 mL of gastric residual volume (or 
20% of a 100 mL/kg feeding total in <750-g infants) would have a significant effect in delaying the achievement of full feeds; they concluded that 2 mL failed to identify infants who were subsequently delayed in reaching full feedings. Others, including our own group,4–8 have used a volume of 50% of a total 3-hour bolus feed as a marker indicative of bowel pathology, but no studies have clearly substantiated this practice. In their retrospective case-control study, Cobb et al found that those infants who went on the develop NEC had maximal gastric residual volumes that constituted 24% to 61% of bolus feeding volumes over the 6 days prior to the diagnosis of NEC versus 14% in control infants over a similar time period.3
Although their findings provide support for many of the published empiric feeding practices, we advise caution in interpreting these results. We speculate that the overall design of the study may have biased the outcome reported. Their review of control medical records was limited to those infants who had not had feedings withheld for 
1 day (in an attempt to exclude those infants who had NEC stage 1). We believe a better control group would have included all infants who did not have a confirmed diagnosis of NEC (Bell's stage II and III). Because this was a retrospective study, the authors could have confirmed which patients went on to have NEC (defined as those with pneumatosis, portal venous air, or bowel pathology, indicating disease) and thus avoided unknowingly excluding infants who had feeding intolerance for reasons unrelated to NEC, such as ileus due to sepsis, meconium plug syndrome, gastroesophageal reflux disease, Hirschsprung's disease, cow-milk sensitivity, and poor gastrointestinal motility. It is evident from the studies of Berseth et al9–12 that many premature infants have initial gastrointestinal dysmotility, which can result in feeding intolerance, but they do not all necessarily develop NEC. By ensuring that controls had no "feeding intolerance," the authors eliminated many patients who could have had their feeds held for reasons other than NEC. The authors' selection of control infants could have increased the reported differences in gastric residual volumes artificially and is likely the reason for the higher percentage of infants in the control group in achieving full feeds when compared with those infants in the NEC group (91% vs 69%; P = .01). We speculate that by using our definition of control subjects, the overlap of residual volumes among control infants would have been wider and decreased the differences compared with the infants with NEC, as reported in the current study.
In addition to the major concern noted above, we are also curious as to why only 6 days before the onset of NEC were studied. It is possible that these infants had feeding intolerance throughout their early hospital course right up to the time of diagnosis; however, the authors do not clarify or justify their selection of the time frame before the onset of NEC. Also, it is unclear if any of the infants were begun on hypocaloric feeds such as half-strength formula. Studies by Baker and Berseth12 have shown that feeding very low birth weight infants a hypocaloric formula versus full-strength formula will result in delaying gastric emptying.
Although we applaud the authors' efforts attempting to make sense out of the nonsensical nature by which neonatologists define feeding intolerance, we do not believe that this report supports current arbitrary practices. We, like the authors, look forward to prospective studies that will help define the significance, if any, of gastric residuals obtained while feeding very low birth weight infants.
REFERENCES
In Reply.—
We thank Kenton at al1 for their thoughtful comments on our study. They have correctly indicated that a control group consisting of infants who did not have a confirmed diagnosis of necrotizing enterocolitis (NEC) (stages II or III), rather than a control group without any possibility of NEC (stages I, II, or III), would have decreased the overlap of residual volumes between the cases and controls. It was precisely to maximize the differences and minimize the overlap between the groups that we chose a priori to exclude infants with stage I NEC. The proportion of infants with preclinical NEC who would progress to stages II or III when feeds are stopped and antimicrobial agents are administered is not known, and inclusion of these infants in the control group may contaminate the results. Despite our efforts to maximize the differences, substantial overlap continued to be evident as we mentioned in the discussion of our article2: "Although these differences were statistically significant, there was much overlap of these variables with those of the control infants, limiting the clinical utility of these observations." The marked degree of overlap is also apparent from Fig 1 and Table 3 of our article.2
It is important also to note that infants with NEC show an increase in maximum residuals only on the day before the diagnosis of NEC (Fig 2 of ref. 2), indicating that the residuals are probably an early sign of NEC rather than a predisposing or risk factor. Although we only analyzed data in the week before the onset of NEC, it is possible but less likely that earlier feeding intolerance was present in these infants.
Very low birth weight infants in our institution are invariably started on full-strength formula feeds, and hence hypocaloric formula is unlikely as a cause of delayed gastric emptying. We agree with Kenton et al1 that additional prospective studies are required to develop rational clinical practices in neonatal feeding and nutrition.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  A. B. Kenton, C. J. Fernandes, C. L. Berseth, B. A. Cobb, W. A. Carlo, and N. Ambalavanan Gastric Residuals in Prediction of Necrotizing Enterocolitis in Very Low Birth Weight Infants Pediatrics, June 1, 2004; 113(6): 1848 - 1849. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
