点击显示 收起
1 From the Department of Pediatrics (AvdB, RMvE, EAMW, WPFF) and the Institute of Research in Extramural Medicine (JWRT), VU University Medical Center, Amsterdam, Netherlands
2 Nutricia Nederland BV (Zoetermeer, Netherlands) provided Nenatal preterm formula, glutamine, and placebo supplementation. 3 Reprints not available. Address correspondence to: A van den Berg, Department of Pediatrics, VU University Medical Center, De Boelelaan 1117, 1081 HV, Amsterdam, Netherlands. E-mail: a.vandenberg{at}vumc.nl.
ABSTRACT
Background: Glutamine depletion has negative effects on the functional integrity of the gut and leads to immunosuppression. Very-low-birth-weight (VLBW) infants are susceptible to glutamine depletion because nutrition is limited in the first weeks of life.
Objective: The objective was to determine the effect of glutamine-enriched enteral nutrition on feeding tolerance, infectious morbidity, and short-term outcome in VLBW infants.
Design: In a double-blind randomized controlled trial, VLBW infants (gestational age <32 wk or birth weight <1500 g) were allocated to receive enteral glutamine supplementation (0.3 g · kg–1 · d–1) or isonitrogenous control supplementation (alanine) between days 3 and 30 of life. The supplementations were added to breast milk or to preterm formula. The primary endpoint for the study was time to full enteral feeding. Secondary endpoints were other variables of feeding tolerance, infectious morbidity, and short-term outcome.
Results: Baseline patient and nutritional characteristics were not significantly different in the glutamine-supplemented (n = 52) and the control (n = 50) groups. The median time to full enteral feeding was 13 d (range: 7–31 d) in the glutamine-supplemented group and 13 d (range: 6–35 d) in the control group (hazard ratio: 1.19; 95% CI: 0.79, 1.79; P = 0.40). In the glutamine-supplemented group, 26 of 52 infants (50%) had 1 serious infection compared with 38 of 50 (76%) in the control group (odds ratio: 0.32; 95% CI: 0.14, 0.74; P = 0.008). Other variables of feeding tolerance and short-term outcome were not significantly different between groups.
Conclusions: Glutamine-enriched enteral nutrition did not improve feeding tolerance or short-term outcome in VLBW infants. However, infectious morbidity was significantly lowered in infants who received glutamine-enriched enteral nutrition.
Key Words: Infant • very-low-birth-weight • glutamine • enteral nutrition • randomized controlled trial
INTRODUCTION
Enteral feeding of very-low-birth-weight (VLBW) infants is a challenge because metabolic demands are high and administration of enteral nutrition is limited by immaturity of the gastrointestinal tract. Experimental studies have shown that the amino acid glutamine plays an important role in maintaining the functional integrity of the gut (1-4). Glutamine serves as fuel for enterocytes (1) and provides nitrogen for the synthesis of amino sugars that are involved in the maintenance of tight junctions (2) and mucin synthesis (3). Moreover, glutamine has stimulatory and regulatory effects on mucosal cell proliferation and differentiation (4). Glutamine is used at a high rate not only by the intestinal epithelium but also by cells of the immune system. In vitro studies have shown that increasing the availability of glutamine stimulates proliferation of T lymphocytes in response to T cell mitogens (5), phagocytosis and antigen presentation by monocytes (6), and the Th1 cytokine response (7).
In critically ill patients, glutamine is considered a conditionally essential amino acid because endogenous glutamine synthesis cannot meet the increased demand (8). VLBW infants may be especially susceptible to glutamine depletion because placental supply suddenly ceases at birth, tolerance of enteral nutrition is limited, and parenteral nutrition, for solubility and stability reasons, does not contain glutamine. Glutamine depletion has negative effects on the functional integrity of the gut (9) and leads to immunosuppression (10). Previous studies in VLBW infants have shown a positive effect of glutamine supplementation on feeding tolerance and the incidence of infections (11-13), although 2 recent multicenter trials in VLBW infants did not confirm these findings (14, 15).
The primary aim of this randomized controlled trial was to investigate the effect of glutamine-enriched enteral nutrition on feeding tolerance in VLBW infants. We also evaluated the effects of glutamine-enriched enteral nutrition on infectious morbidity and short-term outcome in these infants.
SUBJECTS AND METHODS
Subjects
Infants with a gestational age (GA) <32 wk or a birth weight (BW) <1500 g who were admitted to the level III neonatal intensive care unit (NICU) of the VU University Medical Center, Amsterdam, were eligible for participation in the study. Exclusion criteria were major congenital or chromosomal anomalies, death within 48 h after birth, transfer to another hospital within 48 h after birth, and admission from an extraregional hospital. The national central committee on research involving human subjects and the medical-ethical review board of our hospital approved the study protocol. Written informed consent was obtained from all parents.
Randomization, blinding, and treatment
After assignment to 1 of 3 birth weight strata (<799 g, 800–1199 g, and >1200 g), the infants were randomly allocated within 48 h of birth to receive either enteral glutamine supplementation or isonitrogenous control supplementation. An independent researcher used a computer-generated randomization table based on blocks of 4 (Nutricia Nederland BV, Zoetermeer, Netherlands) to assign infants to treatment A or B, which corresponded to batch numbers on the nutrition products. Investigators, parents, and medical and nursing staff were unaware of treatment allocation. The code for the batch numbers was broken after data analysis was performed.
The glutamine powder contained 82% L-glutamine and 18% glucose (15.5% N, by wt; 371 kcal/100 g), whereas the isonitrogenous control powder contained 100% L-alanine (15.7% N, by wt; 435 kcal/100 g) (Nutricia Nederland BV). The 2 powders were indistinguishable by appearance, color, and smell. During the study period, both powders were monitored for stability and microbiological contamination.
The supplementation was administered in increasing doses between days 3 and 30 of life to a maximum of 0.3 g · kg glutamine–1 · d–1 in the glutamine-supplemented group. Initially, the supplementation dose was based on birth weight. After 2 wk, the dose was adjusted to the actual weight of the infants. Two members of the nursing staff added the supplementation daily to breast milk or to preterm formula (Nenatal; Nutricia Nederland BV). Each 100 mL of preterm formula provided 78 kcal, 2.1 g protein (casein-whey protein ratio 40:60), 4.4 g fat, and 7.5 g carbohydrate. The preterm formula did not contain free L-glutamine or free L-alanine. When infants were transferred to other hospitals before the end of the study, the protocol was continued under supervision of the principal investigator (AvdB).
Study endpoints
The primary endpoint for the study was time to full enteral feeding, which was defined as a feeding volume 120 mL · kg–1 · d–1. Secondary endpoints included other variables of feeding tolerance, such as age when parenteral nutrition was discontinued, days of no enteral feeding, and the occurrence of Bell stage II or III necrotizing enterocolitis (16).
Furthermore, we assessed the incidence of serious infections, including sepsis, meningitis, pyelonephritis, pneumonia, and arthritis. Sepsis work-up consisted of blood, cerebrospinal fluid, and urine (suprapubic bladder tap) cultures. Sepsis was defined as the combination of a blood culture positive for microorganisms and the presence of 2 clinical signs of infection (body temperature <36.5 °C or >37.5 °C, hypotension, tachycardia, apnoeic attacks, feeding problems, irritability, or apathy). Meningitis was diagnosed when microorganisms were identified in cerebrospinal fluid cultures. Pyelonephritis was diagnosed when a urine culture was positive for microorganisms and a dimercaptosuccinic acid renal scan showed segmental renal damage. Pneumonia was defined as the combination of a positive culture of tracheal aspirate, bronchial secretion, or sputum positive for microorganisms and the presence of either 1 clinical sign in ventilated infants (purulent sputum, changed sputum characteristics, or deterioration of ventilation settings) or 2 clinical signs in nonventilated infants (tachypnea, cyanosis, wheezing, rales, crepitation, purulent sputum, or changed sputum characteristics). Arthritis was defined as the combination of an intraarticular fluid culture positive for microorganisms and the presence of signs of articular inflammation. Medical records were evaluated for the presence of serious infections by one investigator (a neonatologist), who was unaware of treatment allocation (RMvE).
Other secondary endpoints were growth [weight z scores according to Usher et al (17)], need for mechanical ventilation, presence of periventricular-intraventricular hemorrhage (PIVH) (18), patent ductus arteriosus treated with indomethacin or surgical ligation, bronchopulmonary dysplasia (19) or retinopathy of prematurity (20), age at discharge from NICU and at discharge from the hospital, and death.
Nutritional support
Protocol guidelines for the introduction of parenteral and enteral nutrition followed current practice at our NICU. Parenteral nutrition was started at day 2 and was advanced gradually until amino acid intake reached 2.5 g · kg–1 · d–1 at day 6. Parenteral nutrition was discontinued when enteral feeding reached a volume of 150 mL · kg–1 · d–1. Each 100 mL of parenteral nutrition, an all-in-one mixture provided by the hospital pharmacy, contained 54 kcal, 8.5 g glucose, 1.7 g amino acids, and 1.7 g lipids. If necessary, glucose, amino acids, and lipids were given in separate solutions.
Guidelines for the introduction of enteral nutrition were as follows: 1) minimal enteral feeding was started at day 1 (6–12 mL/d); 2) enteral nutrition was advanced either from day 3 or from day 5 in case of a BW <10th percentile (17), GA <26 wk, Apgar score <6 at 5 min, umbilical artery pH <7.10, or base deficit >10 mmol/L; and 3) feeding was advanced at a dose of 20 mL · kg–1 · d–1 to a maximum of 150 mL · kg–1 · d–1 (based on actual weight). Enteral feeding was reduced or withheld in case of gastric residuals (more than the total volume of past 2 feedings), bilious residuals, emesis, ileus, or necrotizing enterocolitis (Bell stage II) (16). When the signs of feeding intolerance resolved, feeding was advanced within 2 d to the volume given before reduction or withholding. A feeding schedule was proposed on the basis of each infant's BW and the feeding guidelines. However, the staff of our NICU had the final responsibility for the administration of parenteral nutrition and the advancement of enteral feeding.
Statistical analysis
We calculated that a sample size of 102 infants was necessary to detect a difference of 2.5 d to full enteral feeding, assuming an SD of 4.5 d (two-tailed t test, = 0.05, ß = 0.20). The SD value was based on a retrospective analysis of time to full enteral feeding in infants with a GA <32 wk or a BW <1500 g admitted to our NICU in 1998.
Normally distributed and nonparametric data are presented as means (± SDs) and medians (range), respectively. Perinatal and nutritional characteristics were analyzed with a Student's t test, a Mann-Whitney U test, or a chi-square test or Fisher's exact test for continuous normally distributed, nonparametric continuous, and dichotomous data, respectively.
A Cox regression analysis was performed to examine the effect of glutamine-enriched enteral nutrition on time to full enteral feeding. A logistic regression analysis was performed to examine whether glutamine-enriched enteral nutrition influenced the incidence of serious infections. In an additional analysis, adjustments were made for maternal age, administration of antenatal corticosteroids, GA, BW, BW <10th percentile, and administration of breast milk.
Crude analyses of secondary outcomes were performed with a Mann-Whitney U test, a chi-square test or Fisher's exact test, or a log rank test for nonparametric continuous, dichotomous, and time-dependent data, respectively.
All statistical analyses were performed on an intention-to-treat basis. In addition, a per-protocol analysis was performed in which we excluded all infants who were not treated according to protocol, which was defined as >3 consecutive days or a total of 5 d on minimal enteral feeding or >3 consecutive days or a total of 5 d without supplementation after the start of enteral feeding.
A two-tailed P value < 0.05 was considered significant. We used SPSS 9.0 (SPSS Inc, Chicago, IL) for data analysis.
RESULTS
Between September 2001 and July 2003, 107 VLBW infants entered the study. The trial profile is shown in Figure 1. Baseline patient and nutritional characteristics were not significantly different between the 2 groups, except for the occurrence of severe PIVH in the glutamine-supplemented group (P = 0.03; Table 1 and Table 2). No adverse effects of glutamine supplementation were observed in any of the infants.
FIGURE 1.. Trial profile. GA, gestational age; BW, birth weight.
View this table:
TABLE 1. Baseline characteristics of the very-low-birth-weight infants1
View this table:
TABLE 2. Nutritional characteristics of the very-low-birth-weight infants
The median time to full enteral feeding was 13 d (range: 7–31 d) in the glutamine-supplemented group and 13 d (range: 6–35 d) in the control group (hazard ratio: 1.19; 95% CI: 0.79, 1.79; P = 0.40; Table 3). The effect of glutamine-enriched enteral nutrition on time to full enteral feeding was not significantly influenced by maternal age, administration of antenatal corticosteroids, GA, BW, BW <10th percentile, or by administration of breast milk, as shown in the adjusted analysis (Table 3).
View this table:
TABLE 3. Time to full enteral feeding for very-low-birth-weight infants1
In the glutamine-supplemented group, 26 of 52 infants (50%) had 1 serious infection compared with 38 of 50 (76%) in the control group (odds ratio: 0.32; 95% CI: 0.14, 0.74; P = 0.008; Table 4). The effect of glutamine-enriched enteral nutrition on the incidence of serious infections remained after adjustments were made for maternal age, administration of antenatal corticosteroids, GA, BW, BW <10th percentile, and administration of breast milk (Table 4). In both groups, the most frequent cultured microorganisms were coagulase-negative Staphylococcus, Staphylococcus aureus, and Klebsiella pneumoniae.
View this table:
TABLE 4. Serious infections in the very-low-birth-weight infants1
Other variables of feeding tolerance and growth were not significantly different between the treatment groups (Table 5). The need for fewer days on ventilator support in the glutamine-supplemented group compared with the control group approached statistical significance (P = 0.06 and P = 0.09 in infants who survived and infants who died, respectively; Table 5). Although 8 of 52 infants (15%) died in the glutamine-supplemented group compared with 4 of 50 infants (8%) in the control group, the difference was not statistically significant (P = 0.25; Table 5). Other major outcomes were not significantly different between the treatment groups (Table 5).
View this table:
TABLE 5. Secondary outcomes in very-low-birth-weight infants1
The results of the per-protocol analysis (n = 74, Figure 1) were comparable with the results of the intention-to-treat analysis. The median time to full enteral feeding was 13 d (range: 7–29 d) in the glutamine-supplemented group and 12 d (range: 7–22 d) in the control group (hazard ratio: 1.03; 95% CI: 0.65, 1.63; P = 0.91). In the glutamine-supplemented group, 17 of 35 infants (49%) had 1 serious infection compared with 29 of 39 (74%) in the control group (odds ratio: 0.33; 95% CI: 0.12, 0.85; P = 0.023).
DISCUSSION
We found that glutamine-enriched enteral nutrition in VLBW infants did not decrease time to full enteral feeding or improve other variables of feeding tolerance. These findings are in contrast with the results of previous studies on both enteral (12, 14) and parenteral (11, 13) glutamine supplementation in VLBW infants. Vaughn et al (14) and Neu et al (12) found that enteral glutamine supplementation decreased gastrointestinal dysfunction (Bell stage I) and decreased the number of days when feeding was withheld, respectively. In our study, we used feeding guidelines to prevent large differences in nutritional support between the infants, which may have contributed to the lack of difference between the treatment groups in time to reach full enteral feeding. Furthermore, a comparison between our study and the studies by Vaughn et al (14) and Neu et al (12) is difficult because those studies did not report data on either time to full enteral feeding or the age when parenteral nutrition was discontinued. Trials of parenteral glutamine supplementation have shown an increase in time to full enteral feeding (15), a decrease in time to full enteral feeding (11, 13), and fewer days on total parenteral nutrition (11). However, some methodologic concerns can be raised: in 2 of these studies (11, 13), an intention-to-treat analysis was not performed, and in the study by Thompson et al (13), the desired sample size was not achieved. In general, comparisons of all studies may be hampered by the use of different feeding guidelines for both the introduction of and the reduction or withholding of enteral nutrition.
We also found that glutamine-enriched enteral nutrition in VLBW infants reduced the risk of serious infections. This was true both in the intention-to-treat analysis, even after adjustment for possible confounding factors, and in the per-protocol analysis. Our finding agrees with the results of studies in VLBW infants (12), adult recipients of bone marrow transplants (22), and patients with multiple traumas (23). However, Vaughn et al (14) and Poindexter et al (15) did not find a decrease in nosocomial sepsis in recent multicenter trials of enteral and parenteral glutamine supplementation in VLBW infants. In the study by Vaughn et al (14), glutamine was administered in water separately from the formula, whereas in both the study of Neu et al (12) and our study, glutamine was added to the formula. The results of our study on the incidence of serious infections are more comparable with the results of the study by Neu et al (12) than with the results of the study by Vaughn et al (14). Thus, the way the supplementation was administered may have played a role in the results of the studies. In addition, in the study by Vaughn et al (14), the supplementation dose was based on birth weight and was not adjusted for actual weight after 2–3 wk. Consequently, the supplementation dose may have been insufficient, given rapid postnatal weight gain. In our study, the rate of infection was relatively high in both treatment groups. This agrees with the results of a recent surveillance study of nosocomial infections in our NICU (24). In both studies we used predefined definitions for nosocomial infections in neonates that were adapted from the current definitions from the Centers for Disease Control and Prevention for nosocomial infections in children <1 y. The relatively high infection rate can be partially explained by our definitions for nosocomial infections and the relatively high device-associated utilization ratios (24). The relatively high infection rate may have influenced the finding that glutamine-enriched enteral nutrition decreased the incidence of serious infections.
Although several studies have shown that infectious morbidity is decreased in patients receiving glutamine-enriched enteral nutrition, the mechanism of the beneficial effect is not fully understood. A recent study in VLBW infants showed that enterally administered glutamine is entirely metabolized by the gut (25), which indicates that the gut is probably the primary site of action. Experimental studies have shown that glutamine plays an important role in maintaining the functional integrity of the gut by serving as fuel for enterocytes (1), stimulating mucosal cell proliferation and differentiation (4), improving mucus quality (3), and maintaining the integrity of tight junctions (26). Improved intestinal integrity in turn leads to decreased bacterial translocation (27, 28), decreased systemic spread of bacteria (29), and consequently may lead to decreased infectious morbidity. However, until now only animal studies have provided direct evidence for this hypothesis. In human studies, the use of glutamine-enriched enteral nutrition has resulted in a lower systemic inflammatory response in patients with multiple traumas (23) and changes in immune-cell subtype distribution in VLBW infants (12), which reflects decreased bacterial challenge in these patients.
In our study, all the infants with severe PIVH (stage III) were randomly assigned to the glutamine-supplemented group. In all of these infants, PIVH occurred before any supplementation was administered. The higher mortality in the glutamine-supplemented group than in the control group can be explained because 3 of the 6 infants with severe PIVH died. Short-term outcomes, including growth, age at discharge from the NICU, and age at discharge from the hospital, were not significantly different between the glutamine-supplemented and the control groups. These findings agree with the results of Neu et al (12) and Vaughn et al (14). The need for fewer days on ventilator support in the glutamine-supplemented group than in the control group approached statistical significance. This agrees with the finding of Lacey et al (11) that parenteral glutamine supplementation results in fewer days on ventilator support in infants with a BW <800 g. An explanation for our finding may be that enteral glutamine supplementation decreases the intestinal inflammatory response and thereby attenuates distal organ injury, as shown in experimental studies (30, 31). In addition, the lower incidence of serious infections in the glutamine-supplemented group may also have contributed to the need for fewer days on ventilator support.
In the present study, we compared the effects of supplementing normal nutrition with either glutamine or isonitrogenous alanine. We recognize that a control group without glutamine or alanine supplementation would have been more representative of daily practice. However, a comparison of glutamine supplementation with no amino acid supplementation limits the possibility of drawing conclusions about the effect of glutamine, because the results may reflect the effect of amino acid supplementation per se. The choice of control supplementation remains controversial, as recently discussed by Neu (32). Although the use of an isonitrogenous control supplementation is a limitation for the external validity of our study, other aspects of the study design support a general application to daily practice. First, the study population was composed of a randomly selected group of VLBW infants that had no severe congenital malformations. Infants with both a high probability of rapid transfer to another hospital and a high probability of short-term death were included. Second, both breastfed and formula-fed infants were included. Finally, the analysis on an intention-to-treat basis supports a general application to daily practice.
In conclusion, we showed that glutamine-enriched enteral nutrition did not improve feeding tolerance in VLBW infants. However, because we found that glutamine-enriched enteral nutrition decreased the incidence of serious infections, the use of glutamine-enriched enteral nutrition in VLBW infants should be seriously considered.
ACKNOWLEDGMENTS
We are indebted to the parents for allowing their infants to participate in the study. Furthermore, we thank the medical and nursing staff of the neonatal intensive care unit of the VU University Medical Center and all participating hospitals in Noord-Holland for their cooperation, J Feenstra and S Garconius for preparing study feedings and their commitment to the study, and medical students J Migchels, L Vermeij, A van Zwol for their indispensable assistance.
RMvE and WPFF formulated the research questions and wrote the study protocol. AvdB, EAMW, and RMvE coordinated the study. AvdB and JWRT analyzed the data. AvdB wrote the draft for this manuscript, and the other authors reviewed the manuscript. All authors approved the final version of the manuscript. None of the authors had financial or personal conflicts of interest.
REFERENCES