T Lindberg

Epidemic of coeliac disease in Swedish children.

Coeliac disease has emerged as a public health problem. The aim of the present study was to analyse trends in the occurrence of symptomatic coeliac disease in Swedish children from 1973 to 1997, and to explore any temporal relationship to changes in infant dietary patterns. We established a population‐based prospective incidence register of coeliac disease in 1991, and, in addition, retrospective data from 1973 were collected. A total of 2151 cases fulfilled the diagnostic criteria. Furthermore, we collected national data on a yearly basis on duration of breastfeeding, intake of gluten‐containing cereals and recommendations on when and how to introduce gluten into the diet of infants. From 1985 to 1987 the annual incidence rate in children below 2 y of age increased fourfold to 200‐240 cases per 100 000 person years, followed from 1995 by a sharp decline to the previous level of 50‐60 cases per 100 000 person years. This epidemic pattern is quite unique for a chronic disease of immunological pathogenesis, suggesting that prevention could be possible. The ecological observations made in this study are compatible with the epidemic being the result, at least in part, of a change in and an interplay among three factors within the area of infant feeding, i.e. amount of gluten given, age at introduction of gluten, and whether breastfeeding was ongoing or not when gluten was introduced. Other factor(s) may also have contributed, and the search for these should be intensified.

Increasing incidence of childhood coeliac disease in Sweden. Results of a national study

A survey of the incidence of coeliac disease was carried out by asking all 43 paediatric departments in Sweden to report the number of children born between 1978 and 1987 in whom coeliac disease had been diagnosed. Thirty‐four departments representing a population of 7.18 million reported 1944 cases of coeliac disease among 804935 children born between 1978 and 1987. The cumulative incidence of coeliac disease was 1.7 per 1000 live births in children born between 1978 and 1982 and doubled to 3.5 per 1000 live births in children born after 1982. The highest incidence was found in the southern and south‐eastern regions of the country. The observed increase may have been influenced by changes in infant feeding practices such as the postponed age of introduction of gluten from four to six months of age and an increase in gluten content of proprietary baby foods.

Macromolecular Absorption in Preterm and Term Infants

ABSTRACT. Human o‐lactalbumin (α‐LA) has been used as a marker for measuring macromolecular absorption. The serum concentration of human α‐LA after a human milk feed has been studied in 32 healthy very low birthweight infants (VLBW), fed human milk (gestational age 26–32 weeks) and in 56 term, breast‐fed infants, age 3–140 days. At 31 weeks of gestation the serum concentration of human α‐LA was more than 10 times higher (mean value 3000 and median value 2101 μg/1 serum/1 human milk/kg body weight, n= 11) than in the term infants aged 3–30 days (mean value 257 and median value 152, n= 29). The serum concentration of o‐LA decreased with increasing maturity in the VLBW‐infants. At a postconceptional age of 37 weeks the values were similar (mean value 200 and median value 99, n= 8) to those found for term infants during the first month. In the term infants a decreasing absorption of α‐LA was found with increasing postnatal age.

Motilin and infantile colic. A prospective study.

ABSTRACT. Two hundred and nineteen infants, consecutively born, took part in a prospective study of infantile colic from birth to 12 weeks of age. The prevalence of infantile colic in this group was 17.4%. S‐motilin was studied in 78 term infants (19 with and 59 without infantile colic) at birth (cord blood), at one day, and at 6 and 12 weeks of age, respectively. Basal motilin levels were raised both in cord blood (p<0.01) and in blood from neonates (p<0.001) who developed colic as well as in 6 week (p<0.05) and in 12‐week‐old infants with colic (p<0.01). Formula‐fed infants with colic had higher basal motilin levels than formula‐fed controls at 6 and 12 weeks of age (p<0.05). Breast‐fed infants with colic had higher basal motilin levels than breast‐fed controls at 12 weeks of age (p<0.05). The raised levels of motilin from the first day of life in infants who develop infantile colic might indicate that the gastrointestinal tract is affected in infants with infantile colic, before any symptoms of colic appear.

Macromolecular Absorption in Infants with Infantile Colic

ABSTRACT. Intestinal absorption of macromolecules, using human α‐lactalbumin (α‐LA) as a marker, was studied in breast‐fed and formula‐fed infants with infantile colic. Serum samples taken at 30 and 60 min after an intake of human milk were analyzed for α‐LA by a competitive radioimmunoassay technique. Breast‐fed infants with infantile colic had significantly higher s‐α‐LA levels compared with age‐matched breast‐fed control infants 0‐1 month of age: median value 926 μg α‐LA/I serum/I human milk/kg bodyweight (n= 11) versus 150 (n= 34); 1–2 months of age: 173 (n= 22) versus 31 (n= 16); 2–3 months of age: 132 (n= 8) versus 11 (n= 16). Similarly, formula‐fed colicky infants had significantly higher s‐α‐LA levels than agematched formula‐fed control infants 1‐2 months of age: median value 126 (n= 12) versus < 10 (n= 14); 2–3 months of age: 156 (n= 11) versus < 10 (n= 10). The increased absorption of the macromolecule human α‐lactalbumin in infantile colic suggests that the gut mucosa is affected in infants with infantile colic.

Infantile colic: aetiology and prognosis

Infantile colic is one of the most commonly encountered problems within the first 3 mo of life in the Western world. The condition has been known for hundreds of years and was, for instance, described in detail by Niels Rosén von Rosenstein in the first Swedish paediatric textbook in 1764 (1). The prevalence, using Wessel’s criteria (2) (> 3 h dÿ1 of crying and fussing, occurring on more than 3 d wk ÿ1 and continuing for more than 3 wks in an otherwise healthy infant), varies between 9 and 20% (3). The condition is profoundly disturbing both to the infant and to the family. In spite of these facts the aetiology of the disorder is still an enigma. Moreover, our knowledge of the prognosis is incomplete and earlier studies have given controversial results (for references see Ref. 4). The natural course of infantile colic, with a spontaneous disappearance of symptoms at 3–4 mo of age, makes research on this disorder more difficult. In this issue ofActa Paediatricatwo papers deal with aetiology (pp. 18 and 22) and one paper is a follow-up study at 4 y of age (p. 13). Wessel’s criteria for the disorder were used in the studies. Three main types of theory cover the various explanations for the origin of infantile colic presented over the years: (i) psychological theories, with inadequate interaction between mother and infant; (ii) gastrointestinal theories, such as hypertonicity and immaturity of the gastrointestinal tract; and (iii) allergic theories, with reaction to cow’s milk or other food constituents. The papers by Jakobsson et al. (p. 18) and Estep and Kulczycki (p. 22) both present results supporting the third theory. Jakobsson et al. studied the effectiveness of 2 formulae with extensively hydrolysed casein in 22 infants with severe colic. As many as 7 infants were considered as failures (1 treatment and 6 protocol failures), while the remaining 15 infants showed a significant decrease in crying time and crying intensity on both formulae. After commencing the 2 wks of open trial the infants were challenged in a double-blind design. Eleven infants reacted with an increase in crying time to cow’s milk protein or bovine whey protein. Thus, at the end of the study period, the colicky symptoms could be elicited by cow’s milk protein(s) in 50% of the infants. The preliminary study by Estep and Kulczycki (p. 22) included six formula-fed colicky infants chosen consecutively. In all infants the crying and fussing time improved on a diet with an amino acid-based formula. At the end of the study period (after about 2 wk) an unblinded challenge with bovine immunoglobulin G (IgG) increased the crying and fussing time in the six infants. The results from this preliminary study are promising but no conclusions can be drawn because of the small number of infants studied. We are looking forward to a study including a larger number of infants and preferably performed in a double-blind design. It is still not clear which of the various bovine protein(s) elicit the colicky symptoms in some colicky infants. The IgG preparation used was not pure but contained at least 10% of other whey proteins which can play a role in this context. Again, the challenges need to be performed in a double-blind design. Finally, it must be emphasized that the infants included in these two studies were selected and were classified as severe colics. The third paper in this issue, by Canivet et al. (p. 13), is a follow-up study of colicky infants at 4 y of age. The study, using questionnaires, included 50 colicky infants and 102 controls. The colicky infants more often refused certain foods and they did not enjoy the meals as much as the controls did. They also complained more often of stomach ache. Unfortunately, the authors did not ask about the presence of any allergic disease, in spite of the knowledge that food allergy, for example, is more common in children who have had infantile colic (4, 5). On the behaviour and temperament scales the only difference was that the colicky infants were reported to be more emotional than the controls. Recently, an Italian follow-up study (6) showed that in childhood abdominal pain and allergic disorders were significantly associated with infantile colic. There was also an ssociation with sleep disorders, fussiness and aggressiveness. One could discuss endlessly whether the behaviour and temperament problems reported in colicky infants are primary or secondary phenomena in the disorder. The study presented in this issue did not help to answer this question. Several studies have shown that intestinal function, determined by various methods, is abnormal in infants with infantile colic (for references see Ref. 7). The nature of this abnormality is still not known. It is a great pity that intestinal function has not been monitored over

In vitro digestion of proteins in human milk fortifiers and in preterm formula.

BACKGROUND Knowledge about the digestibility of the proteins in new products designed for feeding preterm infants is limited. The purpose of this study was to observe in vitro the hydrolysis of the bovine and human whey proteins in such products. METHODS Proteins in human milk, in human milk fortifiers (Presemp [Semper AB, Stockholm, Sweden] and Enfamil [Mead Johnson, Evansville, IN, U.S.A.] human milk fortifiers), in preterm formulas (Similac Special Care [Ross, Columbus, OH, U.S.A.] and Enfalac [Mead Johnson]), and whey protein concentrates with varying degrees of denaturation were digested by duodenal juice from healthy preterm infants, from a 3-year-old child, and from adults. Digestion was studied in vitro using polyacrylamide gradient gel electrophoresis, electroimmunoassay, and nonprotein nitrogen analysis. RESULTS Casein was the protein most rapidly degraded in all products. Human and bovine whey proteins were more slowly digested; as much as 68% of human lactoferrin was still immunoreactive after 40 minutes of digestion. The corresponding figure for bovine serum albumin was 24-69%; for B-lactoglobulin, 20-40%; for bovine alpha-lactalbumin, 20-51%; and for human alpha-lactalbumin, 41%. Contrary to common belief, digestibility of bovine whey proteins decreased with a high degree of denaturation of the proteins. CONCLUSIONS Bovine whey proteins in human milk fortifiers and in preterm formulas are relatively slowly digested in vitro by normal duodenal juice. The results may have implications for the design of products for feeding preterm infants.

Human α-Lactalbumin as a Marker of Macromolecular Absorption in Early Infancy

It is now generally accepted that antigenic macromolecules can penetrate the small intestinal mucosal membranes in quantities that may be of immunologic importance ( I ) . We do know, that food antigens like bovine proteins are penetrating the intestinal barrier of the mother and are secreted into the human milk in radioimmunologically measurable amounts ( 2 ) . We d o also know from animal experiments that food antigens are passing through the placenta to the foetus ( 3 ) . In experimental animals a period with a high macromolecular permeability is seen during the neonatal period. This is followed by a considerably reduced transmission (4) called “gut closure”, considered to represent an intestinal maturation, affecting the mucosal barrier (4). In humans an increased absorption of macromolecules in infants below three months of age has been shown indirectly by higher concentrations of antibodies to food antigens. Recent studies (5, 6) have also shown higher concentrations of food antigens in the serum of preterm and term infants. Several methods have been used to study gut permeability in man. Heterologous food proteins such as bovine serum albumin, ovalbumin, cow’s milk proteins have been used as markers. When serum concentrations of heterologous proteins are measured local intestinal and systemic immune responses must be considered. The molecules used in most permeability studies are not proteins but-for example, lactulose, rhamnose or polyethylene glycols with relatively low molecular weights. The transfer of these substances through the gut membranes does not reflect the sitution for transfer of food proteins (7). By using a human protein as a marker of macromolecular absorption, we avoid the drawbacks mentioned above. We have chosen human a-lactalbumin (MW=14000), the dominant whey protein of human milk. This protein is not present in the blood except in pregnant and lactating women and in infants after a human milk feeding. This report presents the method and the results from infants with varying maturity and age (8, 9).

Plasma Somatostatin and Cholecystokinin Levels in Preterm Infants during the First Day of Life

Our knowledge about regulatory gut peptides in preterm infants is scanty. We therefore began a study of plasma somatostatin (SS) and cholecystokinin (CCK) in preterm infants at birth and during the neonatal period. Plasma SS and CCK levels were assessed in 77 mothers and in 91 preterm infants immediately after birth (umbilical cord) and during the first day of life (1F) (n = 69, median age 5 h). The gestational age ranged from 23 to 36 weeks and the birth weight from 460 to 3,350 g. After Sep-Pak C18 semichromatography of plasma, SS and CCK were analyzed by RIA. Both plasma SS and CCK levels increased significantly during the first hours of life. Plasma SS levels were negatively correlated to gestational age, birth weight and birth length. When the SS-1F levels were adjusted for gestational age in a multivariate analysis there was no independent association with birthweight but a weak association with birth length. Plasma CCK-1F levels were not correlated with any of these variables. Plasma SS-1F levels were lower after cesarean section. Plasma SS and CCK levels during the first day were not correlated to multiple birth, mode of anesthesia, umbilical pH, Apgar score and blood glucose level before first meal.

Digestion of Milk Proteins in Infancy

Digestion and absorption of protein is a complex matter. It comprises the combined action of several proteolytic enzymes and different mechanisms for absorption. When discussing the quantity and quality of proteins to be given to infants, especially low birth weight infants, it is important to consider the infants’ capacity to digest and absorb the proteins. There are remarkably few reports about this in literature (1, 2 , 3 , 4) and it is rarely a subject of consideration when early infant nutrition is discussed. However, in recent years some studies have been performed to elucidate this question of digestion and absorption of proteins ( 5 , 6).