Breastfeeding, Diet, and Neonatal Hyperbilirubinemia
2000; American Academy of Pediatrics; Volume: 1; Issue: 2 Linguagem: Inglês
10.1542/neo.1-2-e25
ISSN1526-9906
Autores Tópico(s)Hemoglobinopathies and Related Disorders
ResumoAfter completing this article, readers should be able to:Among the many factors related to neonatal hyperbilirubinemia is the composition of an affected infant’s diet. In 1879, Frerichs suggested that “bad nursing” could “exercise a powerful influence” on neonatal hyperbilirubinemia. Much has been learned since this suggestion was made.Many investigations have documented that the consumption of human milk is related to neonatal hyperbilirubinemia, including one review of 12 studies involving more than 8,000 infants in the first week of life and controlled for factors such as hemolysis to enable comparison of dietary effects alone. Moderate hyperbilirubinemia (total serum bilirubin [TSB], 205 mcmol/L [12 mg/dL]) was present in 12.9% of the breastfed infants and 4% of the formula-fed infants (P<0.00001). Severe hyperbilirubinemia (TSB, 256 mcmol/L [15 mg/dL]) was present in 2% of the breastfed infants and 0.3% of the formula-fed infants (P<0.00001). Breastfed infants have higher serum bilirubin levels on each of the first 5 days of life, and this hyperbilirubinemia can persist for weeks to months. The association between feedings of human milk and neonatal hyperbilirubinemia has been reported in preterm infants fed banked human milk or mixtures of human milk and formula and among various races.More recent studies of otherwise healthy newborns have used noninvasive transcutaneous devices to assess hyperbilirubinemia daily (Figs. 1 and 2 ). These studies agree with earlier conclusions that otherwise healthy infants exclusively fed human milk will have higher levels of hyperbilirubinemia than infants who consume formula. The difference begins to become significant at approximately 2 days of age, depending on the formula.The neonatal hyperbilirubinemia associated with breastfeeding is characterized by an elevation in the unconjugated bilirubin level. Data vary, but as many as 6.8% of breastfed infants develop unexplained serum bilirubin levels greater than 256 mcmol/L (15 mg/dL). This form of hyperbilirubinemia has been divided into early or late types, based on the age of onset. The early-onset variety has been termed “breastfeeding jaundice,” and the later-onset condition is called “breast-milk jaundice” or the “breast milk jaundice syndrome” (Table 1 ). Breastfeeding hyperbilirubinemia often is ascribed to inadequate intake related to early problems in initiating feeding and has been reported to result in TSB elevations greater than 204 mcmol/L (12 mg/dL) in 12.9% of breastfed infants. Breast milk jaundice syndrome generally becomes obvious after the first week of life in otherwise healthy and thriving infants, 2% to 4% of whom have TSB levels greater than 170 mcmol/L (10 mg/dL) at 3 weeks of age.It is unclear how these two types of hyperbilirubinemia are related. Kivlahan and James interpret data from their study to suggest that the separation into early- and late-onset may be artificial. On the other hand, Grunebaum et al studied the natural history of 60 infants who had breast milk jaundice and found two serum bilirubin peaks occurring on the fourth to fifth and fourteenth to fifteenth days of life. Hyperbilirubinemia could be detected in some of these infants 12 weeks after birth. Six of 43 otherwise normal siblings (13.9%) had prolonged indirect hyperbilirubinemia of unknown etiology. This suggests that other genetic or environmental factors may be involved in the pathophysiology of this hyperbilirubinemia.The fact that hyperbilirubinemia is associated so commonly with breastfeeding raises the question of whether bilirubin is good for humans. In fact, it is a significant antioxidant believed to be of physiologic importance. However, it is clear that severe neonatal hyperbilirubinemia has the serious and life-threatening risk of kernicterus, and no detrimental effects have been reported for neonatal hypobilirubinemia.Why do breastfed infants experience more hyperbilirubinemia than formula-fed infants? Theoretically, hyperbilirubinemia could result from effects on any of the steps involved in the production and clearance of bilirubin. Bilirubin is produced from the catabolism of heme-containing proteins, of which hemoglobin is the largest component. Following production, bilirubin is transported to the liver bound to serum albumin. After hepatic uptake, bilirubin is conjugated by the microsomal enzyme bilirubin glucuronosyltransferase, and bilirubin mono- and diglucuronides are excreted into the bile. These bilirubin conjugates can be converted back to bilirubin and absorbed from the intestine, undergoing enterohepatic circulation. Alternatively, they can be metabolized to urobilinoids or excreted in the feces. Urobilinoid production requires intestinal bacteria.Hyperbilirubinemia results from either an increase in bilirubin production, a decrease in bilirubin excretion, or a combination of these two mechanisms. There are no data to suggest that bilirubin production increases in neonates fed human milk rather than formula. Carbon monoxide excretion, a measure of bilirubin production, has been shown to be similar between breast- and formula-fed infants. Hence, theories to explain hyperbilirubinemia related to breastfeeding have focused on differences in bilirubin excretion between breast- and formula-fed infants.There are many possible etiologic factors in the development of hyperbilirubinemia in otherwise healthy breastfed infants (Table 2 ). Starvation or fasting is associated with increased serum bilirubin levels in adults and experimental animals, although the mechanism is still debated. Neonates ingesting human milk receive fewer calories in the first days of life than do formula-fed infants, averaging 12 kcal/kg during the first day and increasing to 66 kcal/kg by 72 hours of age. Studies differ on the impact of the number of feedings per day and the relationship to serum bilirubin levels. One study showed that increasing the frequency of feedings to more than eight times per day was associated with significantly lower serum bilirubin levels, but a subsequent larger study showed no relationship between frequency of breastfeeding and serum bilirubin levels. Breastfed babies appear to lose more weight in the first several days of life than do formula-fed babies, although studies differ on whether there is a relationship between serum bilirubin levels or hyperbilirubinemia and neonatal weight loss. Some studies show no relationship; others show that breastfeeding and poor weight recovery or weight loss are significantly associated with hyperbilirubinemia. In summary, breastfed infants generally lose more weight in the first days of life, which often is associated with hyperbilirubinemia. However, the many studies showing no relationship between weight loss and neonatal hyperbilirubinemia suggest that other factors frequently must be more important. There is no difference in hyperbilirubinemia levels between exclusively breastfed infants and those who receive supplements with water or dextrose.Another possible etiology of hyperbilirubinemia associated with breastfeeding is a substance in the human milk that is absorbed by the infant and subsequently inhibits the hepatic excretion of bilirubin. Proposed substances have included pregnanediol, fatty acids, and lipase. Results of studies conflict, and no single substance has been able to stand up to the scrutiny of scientific investigation. Hence, at present there is no accepted agent in human milk that adversely affects hepatic bilirubin glucuronidation or other hepatic steps in bilirubin metabolism. However, there are many factors related to the control of glucuronosyltransferase activity and compounds potentially passed in human milk (eg, metal ions, steroids, or nucleotides) that possibly could inhibit bilirubin excretion and result in hyperbilirubinemia.Another etiologic group of hypotheses involves bilirubin absorption from the intestine, the enterohepatic circulation of bilirubin. Meconium, the blackish-greenish contents of the intestinal tract that have accumulated throughout gestation, has been reported to contain significant concentrations of bilirubin. Accordingly, it is not surprising that delayed meconium passage is associated with hyperbilirubinemia, and early passage is associated with low bilirubin levels. There is a significant relationship between the amount of stool passed and the total serum bilirubin level in breastfed infants. Serum bilirubin levels are higher when fecal weights are lower, and the decrement in serum bilirubin during the first 3 weeks of life is greater if more stool is excreted.Another factor related to the enterohepatic circulation of bilirubin is urobilinoid formation. In adults, intestinal bacteria convert bilirubin glucuronides to urobilinoids, thus protecting against bilirubin absorption. Urobilinoids (eg, urobilinogens, urobilins, violins) comprise a large family of related nontoxic compounds that result from hydrogenation and dehydrogenation of the various unsaturated bonds in bilirubin and its conjugates. Conversion of bilirubin conjugates to urobilinoids blocks the enterohepatic circulation in adults. At birth, the intestine of the neonate is sterile; intestinal flora develop over the first days to weeks of life and are related to the type of feeding. Therefore, urobilinogen production in the neonate is limited, and fecal bile pigment profiles demonstrate the same bilirubin conjugates that are seen in duodenal bile. These bilirubin conjugates present the potential to be absorbed by the intestine until the intestinal bacteria develop enough to produce urobilinoids. Formula-fed infants excrete urobilin in their stools earlier than breastfed infants. The composition of the diet obviously influences the development of intestinal bacteria.Another important factor related to the enterohepatic circulation of bilirubin is beta-glucuronidase, an enzyme that cleaves the ester linkage of bilirubin glucuronide, producing unconjugated bilirubin. Unconjugated bilirubin can be absorbed more easily from the intestinal lumen than can bilirubin glucuronides. In utero, beta-glucuronidase is believed to be important for the absorption of bilirubin into the fetal circulation, which allows clearance via the placenta. Following birth, however, persistent beta-glucuronidase in the intestinal lumen will continue to potentiate the enterohepatic circulation, thereby delaying bilirubin clearance from the body. Beta-glucuronidase is present in intestinal tissue, bacteria, and human milk, but it is not present in infant formula. Some studies have shown that human milk beta-glucuronidase is related to neonatal hyperbilirubinemia, although other studies have not confirmed this finding. Fecal beta-glucuronidase also has been shown to be related to serum bilirubin levels.Alkaline pH can hydrolyze bilirubin glucuronides to unconjugated bilirubin. However, stools from breastfed infants are more acid than stools from formula-fed infants. Hence, alkaline hydrolysis is not responsible for enhanced enterohepatic circulation in breastfed infants. Although breast milk jaundice generally is not a cholestatic condition, occasional studies have reported serum bile acids to be elevated or the ratio of glycine- to taurine-conjugated bile acids to be varied. The significance of these findings remains unclear.Because hyperbilirubinemia is such a common problem among breastfed infants, the American Academy of Pediatrics (AAP) has developed some recommendations. When jaundice begins developing in a breastfed infant, breastfeeding should not be interrupted initially. Rather, feeding should be increased in frequency (8 to 10 times per day). Although excess fluid such as water does not lower total serum bilirubin levels, some infants are dehydrated and require supplemental fluid. Because bilirubin photoproducts are excreted in urine and bile, adequate hydration is important. If jaundice continues to increase, treatment options include continued observation, initiation of phototherapy, or use of formula (supplementation or exclusive feeds) with the acknowledgment that formula use can inhibit the enterohepatic circulation of bilirubin. If breastfeeding will be stopped temporarily, it is very important to educate and encourage mothers to pump their breasts so that the milk supply will be maintained in anticipation of resumption of breastfeeding. Tan recently showed that the addition of supplementary formula to the feedings of breastfed infants enhances the efficacy of phototherapy and reduces exposure time.Temporary interruption of breastfeeding and substitution with formula is advocated by the AAP as one therapy for hyperbilirubinemia associated with breastfeeding. Hence, it would be useful to know if any particular formula is associated with lower levels of hyperbilirubinemia. A few studies have compared infant formulas and their affect on neonatal hyperbilirubinemia. In 1992, Gourley et al examined the relationship between fecal output and neonatal hyperbilirubinemia in infants exclusively fed either human milk or one of three infant formulas (whey-predominant [Enfamil®], casein-predominant 3305H [the pre-1982 version of Enfamil®], and casein hydrolysate [Nutramigen®]), with approximately 15 infants in each group. They meticulously measured all stool output during the first 3 weeks of life and assessed hyperbilirubinemia by taking transcutaneous jaundice index readings (Fig. 1 ). The jaundice index data showed differences between the groups at all times after day 3 (P=0.04 [day 4] to 0.0001 [days 10 to 21]). As expected, the breastfed group had the highest levels of hyperbilirubinemia. The two routine formulas (pre- and post-1982 Enfamil®) were next in levels of hyperbilirubinemia and did not differ significantly from each other. The lowest levels of hyperbilirubinemia were seen in the Nutramigen®-fed group. This group had significantly lower bilirubin levels than the whey-predominant or casein-predominant routine formulas on days 10 to 18 and 10 to 21, respectively.Fecal output was significantly related to the decrease (from day 3 to 21) in both serum bilirubin level and jaundice index in only the breastfed group. Infants in the breast- and Nutramigen®-fed groups had fecal output characteristics that were most similar, including mean cumulative wet and dry stool weight, total number of stools per week, and percent water. The routine formula groups produced a larger volume of stool that had a higher percentage of water. The fact that the breastfed and Nutramigen®-fed groups were the most disparate in hyperbilirubinemia levels suggested strongly that the difference in hyperbilirubinemia was not related to stool output or water content. The authors speculated that the greater stool weights and lower serum bilirubin levels of infants fed routine formula might be related to differences in fat absorption. The coefficient of fat absorption in breastfed infants is higher than in formula-fed infants and because of the lipophilic nature of bilirubin, fat malabsorption would be expected to enhance bilirubin clearance.A similar but independent and more rigorous study of a different group of infants was published by Gourley et al in 1999. In this study, infants were fed exclusively human milk, Nutramigen®, or Enfamil® (20 infants per group). The infants were followed carefully with daily transcutaneous measurements of jaundice during the first week of life and every 2 to 3 days thereafter. Transcutaneous data were collected, with careful notation of time to the minute, which enabled computer modeling of the data (Fig. 2 ). The results were very similar to those of the 1992 study, with the group of Nutramigen®-fed infants demonstrating significantly lower hyperbilirubinemia levels than the Enfamil®-fed infants on days 6 to 16. Similarly, the highest hyperbilirubinemia levels were recorded in the breastfed infants.One reason that the Nutramigen®-fed infants might have lower hyperbilirubinemia levels than those fed routine formulas is the presence of a recently identified beta-glucuronidase inhibitor in Nutramigen®. This inhibitor, found in the casein hydrolysate fraction of Nutramigen®, remains inhibitory to beta-glucuronidase after passage through the infant’s gastrointestinal tract. The observation that both intake and output of Nutramigen®-fed infants exhibit inhibitory activity increases the likelihood that this beta-glucuronidase inhibition also is present in the gastrointestinal tracts of these infants. The inhibition occurs over a pH range of 4 to 7, similar to that found in the neonatal intestine. If active, this inhibitor would be predicted to result in lower beta-glucuronidase activity in the gastrointestinal tract, which would result in less hydrolysis of bilirubin glucuronides and less production of unconjugated bilirubin. Because unconjugated bilirubin is absorbed more easily from the intestine than are bilirubin conjugates, the expected net result would be a decrease in bilirubin absorption and an increase in bilirubin excretion. Although it remains unproven that this mechanism is operative, it is intriguing to note that human milk is rich in beta-glucuronidase activity, routine formula (Enfamil®) has no beta-glucuronidase activity, and Nutramigen® contains a beta-glucuronidase inhibitor. This coincides exactly with the observation that serum bilirubin levels are highest in breastfed infants, intermediate in Enfamil®-fed infants, and lowest in Nutramigen®-fed infants.These observations raise the question about the most appropriate formula to use if following AAP guidelines. In some infants, a simple change to routine formula for a short time may be all that is necessary to treat hyperbilirubinemia. In these infants, the more expensive casein hydrolysate formula may not be cost-effective. However, in other infants, a change to formula is the step that precedes other treatment, such as phototherapy or exchange transfusion. In these infants, it may be possible that change to the casein hydrolysate might prevent or lessen the additional therapy. The cost-effectiveness of such maneuvers can be determined only through controlled clinical trials. Hence, the use of Nutramigen® for this purpose is not recommended at this time. Alternatively, if this effect on hyperbilirubinemia is due to a specific component of Nutramigen®, it might be more effective to identify and isolate this component and administer it separately while breastfeeding without interruption.The current trend is increasingly earlier hospital discharge of newborns as third-party payers attempt to reduce health-care costs. Several articles have been written regarding the effects on the infant of short postnatal hospital stays. Although many of these studies have methodologic flaws, a recent investigation of more than 310,000 newborns in the state of Washington has shown that those discharged early (<30 h of age) were more likely to be rehospitalized for hyperbilirubinemia within 7 days of discharge (adjusted odds ratio, 1.31; 95% confidence interval, 1.16 to 1.54) compared with infants discharged at 30 to 78 hours after birth. There is no doubt that early hospital discharge is a potential risk factor for the development of severe hyperbilirubinemia and kernicterus. It has been recognized that hyperbilirubinemia requiring phototherapy is the most commonly reported readmission diagnosis and can account for approximately 75% of all causes. As many as 94% of infants readmitted for hyperbilirubinemia or dehydration have been breastfed. Recognition and follow-up of hyperbilirubinemia has become more difficult because of earlier discharge.It has been said that the hyperbilirubinemia associated with breastfeeding never results in brain damage, but it is clear that rare breastfed infants who have no other cause for their hyperbilirubinemia can develop severe hyperbilirubinemia and kernicterus. Such infants frequently are discharged early from the hospital and do not return for evaluation until severe hyperbilirubinemia and kernicterus already have developed. Most commonly these infants are breastfed.The AAP and the American College of Obstetricians and Gynecologists have published guidelines for infants being considered for early discharge. They suggest the following are necessary for early discharge: attendance at prenatal classes, instruction regarding common neonatal problems, uncomplicated term delivery, normal physical examination, thermal homeostasis, a negative Coombs test, and demonstration of basic maternal skills, including successful infant feeding within 48 to 72 hours after discharge. It is unlikely that the minimum requirements can be accomplished in fewer than 48 hours. These criteria include: “There is no evidence of significant jaundice in the first 24 hours of life,” and “Laboratory data are available and reviewed, including: … cord or infant blood type and direct Coombs’ test result as clinically indicated.” The AAP recommends that “follow-up should be provided to all neonates discharged less than 48 hours after birth by a health care professional in an office, clinic, or at home within 2 to 3 days of discharge.”If there were some way to identify infants at risk for developing severe hyperbilirubinemia, high-risk infants could be followed more closely, but at present there are no widely accepted screening methods to identify high-risk infants. One method recently described by Bhutani et al involves obtaining a serum bilirubin level on every infant prior to hospital discharge and employing an algorithm by which infants can be categorized for risk of subsequent hyperbilirubinemia. The effectiveness of such a screening program awaits validation.The author gratefully acknowledges the technical assistance of Bill Kreamer and Brenda Egan.
Referência(s)