Milk hypersensitivity in young adults

European Journal of Clinical Nutrition (1999) 53, 620±624 ß 1999 Stockton Press. All rights reserved 0954±3007/99 $12.00 http://www.stockton-press.co....
Author: John McDaniel
1 downloads 0 Views 105KB Size
European Journal of Clinical Nutrition (1999) 53, 620±624 ß 1999 Stockton Press. All rights reserved 0954±3007/99 $12.00 http://www.stockton-press.co.uk/ejcn

Milk hypersensitivity in young adults L Pelto1*, O Impivaara2, S Salminen1, T Poussa3, R SeppaÈnen2 and E-M Lilius1 1

Department of Biochemistry and Food Chemistry, University of Turku, Finland; 2Research and Development Centre, Social Insurance Institution, Turku, Finland; and 3Stat-Consulting, Tampere, Finland

Objective: To estimate the prevalence of milk hypersensitivity in Finnish adults. Design: Cross-sectional study. Subjects: Two hundred men and 206 women aged 27 y randomly recruited from the population register in southwestern Finland. Interventions: The subjects were interviewed about their dairy product consumption, abdominal discomfort after dairy product intake and lactose intolerance. From serum samples, serum reactivity to milk protein and milk-speci®c IgG1, IgG2, IgG3 and IgA were measured. Results: About 20% of the subjects reported abdominal discomfort after dairy product intake, whereas only 6.4% had been diagnosed to have lactose intolerance. The amount of milk consumed correlated well with the serum assay results in subjects reporting abdominal discomfort but not in subjects who were free from these symptoms. Among subjects with no record of dairy product restriction or lactose intolerance, those experiencing abdominal discomfort after dairy product intake had signi®cantly higher serum reactivity to milk protein than those without such discomfort. The concentrations of serum milk-speci®c antibodies did not differ between these two groups. The prevalence of milk hypersensitivity in this population was estimated to be 3 ± 6%. Conclusions: Milk hypersensitivity may be as common in adults as in infants. The measurement of serum reactivity to milk protein may prove useful in screening milk hypersensitivity in subjects who have not restricted their dairy product consumption. Sponsorship: The study was supported by the Finnish Cultural Foundation, the Turku University Foundation, and Social Insurance Institution, Finland. Descriptors: antibodies; gastrointestinal symptoms; lactose intolerance; milk hypersensitivity; prevalence; sampling studies

Introduction Adverse reactions to food may be classi®ed as non-immune mediated food intolerance or immune mediated food hypersensitivity (Bruijnzeel-Koomen et al, 1995). Adverse reactions to milk include lactose intolerance caused by lactase (b-galactosidase) de®ciency and milk hypersensitivity. About 17% of Finns have lactase de®ciency (Sahi, 1974). About 70% of these respond to two glasses of milk (Sahi et al, 1983; Savaiano & Levitt, 1987; Vesa et al, 1997) corresponding roughly to the average daily milk consumption in Finland (Anonymous, 1998). Therefore, about 12% of Finns can be considered lactose-intolerant. Milk hypersensitivity is mediated by immunological mechanisms. About 3% of infants suffer from cow milk allergy (Hùst & Halken, 1990; Savilahti et al, 1991; Schrander et al, 1993). Most of them have been reported to become *Correspondence: Dr L Pelto, Department of Biochemistry and Food Chemistry, University of Turku, FIN-20014 Turku, Finland. Guarantor: L Pelto. Contributions: All authors contributed to the design, execution, analysis of data and writing up of this study: Speci®cally, O Impivaara: clinical assessment; S Salminen: overall conduct of the study; T Poussa: statistical data; R SeppaÈnen: nutritional data; and E-M Lilius: immunological methods. Received 13 January 1999; revised 8 February 1999; accepted 25 February 1999

symptom-free by the age of three (Hùst & Halken, 1990), and milk hypersensitivity is therefore considered rare among adults. Nevertheless, our recent study suggests that this may not be the case (Pelto et al, 1998). Undegraded dietary antigens can penetrate the intestinal epithelial surface in immunologically signi®cant quantities in healthy and hypersensitive subjects (Lovegrove et al, 1993). Transferred antigens stimulate the production of speci®c antibodies which is, however, strictly controlled in healthy subjects (Challacombe & Tomasi, 1980). Appearance of IgE is known to correlate with hypersensitivity reactions but the clinical relevance of IgG, IgA and IgM isotypes produced in response to dietary antigens is obscure (Bruijnzeel-Koomen et al, 1995). Most recently published papers suggest that IgG and IgA may have a role in hypersensitivity reactions (Keller et al, 1996; Marinkovich, 1996; Wutrich, 1996; Awazuhara et al, 1997; Duchen et al, 1997; Kolopp-Sarda et al, 1997; Shimojo et al, 1997; Spuergin et al, 1997). In our previous study, IgE was not a major antibody in milk-hypersensitive adults (Pelto et al, 1998). Similar results have been obtained by other groups (Bengtsson et al, 1996; Werfel et al, 1997; Bischoff et al, 1997). We measured serum reactivity to milk protein and milk-speci®c IgG1, IgG2, IgG3 and IgA antibodies in serum and estimated the prevalence of milk hypersensitivity in young adults.

Milk hypersenstivity in young adults L Pelto et al

Methods Study population The subjects were randomly drawn from the population registers of the city of Turku and adjacent rural and urban communities in southwestern Finland. The study population was made up of 200 men and 206 women aged 27 y. Before implementation, the study was approved by the Ethical Committee of the Research Centre of the Social Insurance Institution. Interview The participants were asked whether they had ever been diagnosed by a physician to have lactose intolerance, whether they had experienced abdominal discomfort after the intake of dairy products, and whether they had restricted their consumption of such products (milk, sour milk, yoghurt or ice cream) because of the symptoms. The participants were also asked about their daily consumption of dairy products. The amount of milk, sour milk and yoghurt was recorded as glasses (equalling to 2 dl). Hard and soft cheeses were recorded as slices (equalling to 15 g) and tablespoons (equalling to 20 g), respectively. Consumption of ice cream was recorded as portions (equalling to 1 dl). The records were checked by a dietician. For the results, the amounts were calculated in dl or g. Serum assays Samples. Blood samples were collected in the morning, after an overnight fast. After separation, serum available from 402 study subjects was stored at ÿ70 C. Serum of a healthy (no lactose intolerance, no milk hypersensitivity) 26 y old woman not belonging to the study group served as a negative control. Milk-speci®c antibodies were measured from randomly selected 71 subjects with no abdominal discomfort after dairy product intake and from all reporting abdominal discomfort (n ˆ 82). Reagents. Hank's balanced salt solution (HBSS) was made without phenol red, pH 7.4 and supplemented with 0.1% gelatin (gHBSS). Monoclonal Mouse Anti-Human IgG1, IgG2, IgG3 and IgA and the labelled Fab (AP-F(ab0 )2 Rabbit Anti-Mouse IgG (H‡L)) were purchased from Zymed (San Francisco, CA, USA). The substrate (AMPPD) were from Tropix (Bedford, MA, USA). Dilutions were made in gHBSS unless otherwise stated. Measurement of serum reactivity to milk protein. The capacity of antigen-antibody complexes to activate phagocytes were measured as previously described (Pelto et al, 1998). Measurement of milk-specific antibodies. Milk-specific IgG1, IgG2, IgG3 and IgA were measured with ELISA modified for luminometry. 100 ml of 1 : 10 diluted serum was added to milk protein coated tubes and incubated for 30 min at ‡37 C. Serum was removed by a vacuum and the tube was washed with 500 ml of gHBSS. 100 ml 1 : 100 diluted anti-human immunoglobulin isotype antibody (IgA, IgG1, IgG2 or IgG3) was added and incubated for 30 min at ‡4 C. After vacuum clearance and a wash 100 ml 1 : 200 diluted anti-mouse immunoglobulin Fab-fragment labelled with alkaline phophatase was added and

incubated for 30 min at ‡4 C. After vacuum clearance, the tube was washed three times so that the unspecific label would not interfere. AMPPD was diluted 1 : 250 in 0.1 M carbonate buffer and 100 ml of substrate was added just before the measurement. The reaction was measured with a luminometer for 30 min at ‡37 C and the peak chemiluminescence was recorded. Statistics The levels of serum reactivity to milk protein and the concentrations of milk-speci®c antibodies in serum were adjusted to the negative control, and were expressed as percentage of the negative control. The study population was divided into six groups according to reported abdominal symptoms, restriction of dairy product consumption and lactose intolerance. The analysis of variance (ANOVA) was used to compare the differences between groups, and Fisher's LSD test was used in pairwise comparisons. Due to skewed distributions, the variables were logarithmically transformed before analysis. The associations between serum reactivity to milk protein and milk-speci®c antibodies were analysed by Pearson's correlation analysis, and stepwise regression analysis was performed to ®nd out which of the milkspeci®c antibodies best indicated the level of serum reactivity to milk protein. Differences in dairy product consumption between the de®ned study groups were analysed using Mann ± Whitney U-test. Results Interview The results of the interview (Table 1) showed no major differences between men and women and their results were therefore combined for further analysis. The study population was divided into six groups according to the symptoms, dairy product consumption and reported lactose intolerance (Figure 1). The groups with one and two subjects were not analysed statistically. The remaining four groups were de®ned as group A (no symptoms, no restriction of dairy product consumption, no lactose intolerance), group B (symptoms, no dairy product restriction, no lactose intolerance), group C (symptoms, dairy product restriction, no lactose intolerance) and group D (symptoms, dairy product restriction and diagnosed lactose intolerance). Dairy product consumptions in groups A, B, C and D are presented in Table 2. As expected, consumption of liquid milk products (milk, sour milk and yoghurt) was Table 1 Prevalence (95% CI) of abdominal discomfort, restriction of dairy product consumption and diagnosed lactose intolerance as reported in the interview Men (n ˆ 200)

Women (n ˆ 206)

Total (n ˆ 406)

Abdominal discomfort 19.0 21.4 20.2 after dairy product intake (%) (13.6 ± 24.4) (15.8 ± 27.0) (16.3 ± 24.1) Restriction of dairy product consumption (%) Lactose intolerance diagnosed by a physician (%)

16.0 17.0 16.5 (10.9 ± 21.2) (11.9 ± 22.1) (12.9 ± 20.1) 5.5 (2.3 ± 8.7)

7.3 (3.7 ± 10.8)

6.4 (4.0 ± 8.8)

621

Milk hypersenstivity in young adults L Pelto et al

622

Figure 1 Study population divided into groups on the basis of (1) abdominal discomfort after dairy product consumption; (2) restricted dairy product consumption; and (3) diagnosed lactose intolerance. Study groups are A, B, C and D.

Table 2 Dairy product consumption (median, inter quartile range) in groups A, B, C and D*

Liquid milk products{ (dl=d) Hard cheese (g=d) Soft cheese (g=d) Ice cream (dl=week)

A (n ˆ 323)

B (n ˆ 13)

C (n ˆ 44)

D (n ˆ 23)

4 (2 ± 8)

6 (2 ± 8)

3 (2 ± 6)a

2 (0 ± 2)a,b,c

45 (15 ± 75) 60 (30 ± 90) 60 (15 ± 64) 30 (15 ± 68) 0 (0 ± 20) 0 (0 ± 20) 0 (0 ± 25) 0 (0 ± 10) 1 (0 ± 2) 1 (0 ± 2) 1 (0 ± 2) 0 (0 ± 1)

*A ˆ no abdominal symptoms, no restriction of dairy products, no lactose intolerance. B ˆ abdominal symptoms, no restriction of dairy products, no lactose intolerance. C ˆ abdominal symptoms, restriction of dairy products, no lactose intolerance. D ˆ abdominal symptoms, restriction of dairy products, lactose intolerance. {Milk, sour milk and yoghurt. Statistically signi®cant (P < 0.05) difference: acompared to A; bcompared to B; ccompared to C.

signi®cantly lower in subjects with restricted dairy product consumption due to discomfort (groups C and D) than in those without any abdominal discomfort and with unrestricted dairy product consumption (group A). Subjects with lactose intolerance restricting dairy product consumption (group D) had signi®cantly lower dairy product consumption than subjects with abdominal discomfort after dairy product consumption but without diagnosed lactose intolerance (groups B and C). The consumption of cheeses and ice cream did not differ between the groups. Serum assays Distributions of serum reactivity to milk protein and milkspeci®c antibodies are presented in Figure 2. Serum assays vs interview results The results of serum reactivity to milk proteins and milkspeci®c antibodies in the four groups (A, B, C, D) are presented in Table 3. In groups with abdominal discomfort after dairy product consumption the subjects with restricted dairy product consumption (groups C and D) had signi®cantly lower serum reactivity to milk protein and lower concentrations

Figure 2 Distribution of IgA, IgG1, IgG2, IgG3 and serum reactivity to milk protein (percentage of a negative control) in the whole study population of subjects aged 27 y. Table 3 Serum reactivity to milk protein and milk-speci®c antibodies as percentage of a negative control (geometric means) in the four study groups*

Serum reactivity IgA IgG1 IgG2 IgG3 n

A

B

C

D

193 127 201 122 101 71 ± 319

422a 124 185 124 112 13

149b 106a 116a,b 74a,b 52a,b 44

130b 97a,b 125a,b 64a,b 56a,b 23

*A ˆ no abdominal symptoms, no restriction of dairy products, no lactose intolerance. B ˆ abdominal symptoms, no restriction of dairy products, no lactose intolerance. C ˆ abdominal symptoms, restriction of dairy products, no lactose intolerance. D ˆ abdominal symptoms, restriction of dairy products, lactose intolerance. Statistically signi®cant (P < 0.05) difference: acompared to A; bcompared to B.

of milk-speci®c antibodies than those with unrestricted dairy product consumption (group B). In subjects with abdominal discomfort after dairy product consumption (groups B, C and D), serum reactivity to milk protein and concentrations of IgG2 and IgG3 were signi®cantly lower in subjects consuming less than 2 dl milk per day than in those consuming 2 dl or more milk per day. Geometric mean percentage of serum reactivity to milk protein was 109 in subjects consuming less than 2 dl milk per day and 283 in subjects consuming 2 dl or more milk per day, P < 0.0001. For milk-speci®c IgG2 and IgG3 geometric means were 60 and 103, P < 0.0001; and 51 and 75, P < 0.007, respectively. There was no such correlation in subjects without abdominal discomfort (group A).

Milk hypersenstivity in young adults L Pelto et al

Table 4 Correlation of serum reactivity to milk protein and milkspeci®c antibodies

Serum reactivity vs IgA IgG1 IgG2 IgG3 n

Subjects without abdominal discomfort after milk intake (A)

Subjects with abdominal discomfort after milk intake (BCD)

All subjects (ABCD)

r 0.49 0.36 ÿ0.08 0.32 71

r 0.65 0.73 0.72 0.69 83

r 0.54 0.71 0.67 0.66 154

Of the subjects without milk restriction and lactose intolerance, those with abdominal discomfort after dairy product consumption (group B) had signi®cantly higher serum reactivity to milk protein than those without abdominal discomfort (group A) (p < 0.01). Concentrations of milk-speci®c antibodies did not differ between these groups. Correlation of serum reactivity to milk protein and milkspeci®c antibodies All milk-speci®c antibodies correlated positively with serum reactivity to milk protein in subjects with abdominal discomfort after milk intake (groups B, C and D) (Table 4). Similar correlations excepting milk-speci®c IgG2 were seen in subjects without abdominal discomfort (Group A). In a stepwise regression analysis covering all study groups, all antibodies proved signi®cant predictors of serum reactivity (R2 ˆ 0.74). Discussion The aim of this study was to estimate the prevalence of milk hypersensitivity in a group of young adults. According to the interview, 6.4% of the subjects had had lactose intolerance diagnosed by a physician. This is less than expected since about 12% of Finns are reported to have lactose intolerance with symptoms (Sahi, 1974; Sahi et al, 1983; Savaiano & Levitt, 1987; Vesa et al, 1997). Lactose intolerance is well known and often self-diagnosed and therefore subjects do not necessarily consult a physician for it. On the other hand, the proportions of subjects reporting abdominal discomfort and restricting dairy product consumption were higher than expected, namely 20.2% and 16.5%, respectively. Lactose intolerance alone can not explain this. It seems obvious that in addition to lactose intolerance, milk hypersensitivity can produce symptoms to some adults (Pelto et al, 1998) whereas others probably misconnect their symptoms to milk intake. Serum assay results correlated well with milk product consumption in subjects with abdominal discomfort after milk product intake but not in those free from such symptoms. In children with coeliac disease, gluten challenge is known to signi®cantly increase IgG and IgA levels to gluten, whereas gluten elimination decreases these (Scott et al, 1985; Lindberg et al, 1985; Wauters et al, 1991). In healthy subjects a chronic increase of dietary antigens does not affect food antibody concentrations even if single large loads of dietary antigens may lead to formation of food antibodies as a physiological response (Gallagher et al, 1983). Therefore, in long term consumption, the amount of

dietary antigens may not in¯uence the concentration of antibodies in healthy subjects, whereas this may happen in subjects with gastrointestinal problems. This may result from increased penetration of antigens through the pathologically altered intestinal mucosa. Subjects with lactose intolerance diagnosed by a physician (group D) consumed signi®cantly less milk than those in groups B and C who also had reported discomfort after milk consumption but in whom this diagnosis had not been veri®ed. Therefore, a proper diagnosis by a physician seems to in¯uence the subject's dietary habits. The differences in serum assay results between groups A and D were probably unrelated to lactose intolerance. Immunological serum assays are unlikely to be affected by non-immune mediated intolerance but rather by differences in dairy product consumption. Serum reactivity to milk protein was signi®cantly higher in subjects with abdominal discomfort after dairy product intake but with no milk restriction and no lactose intolerance (group B) than in those without any symptoms during unrestricted dairy product consumption (group A). A similar difference was seen in our previous study between milkhypersensitive and control subjects (Pelto et al, 1998). Therefore, it seems that group B included signi®cantly more milk-hypersensitive subjects than group A. On the other hand, milk restriction reduces serum reactivity to milk protein. Therefore, group C which was comparable to group B, except for restricted dairy product consumption, showed low reactivity similar to that in group A even if there may have been equal proportions of milk-hypersensitive subjects in groups B and C. Milk-speci®c antibodies correlated well with serum reactivity in subjects with abdominal discomfort after milk intake (groups B, C and D) and in healthy subjects (group A) except for milk-speci®c IgG2. The best correlation was obtained when all measured antibodies were taken into account. In addition to the antibodies, also complement compounds participate to the reaction of serum reactivity to milk protein. The concentrations of milk-speci®c antibodies did not differ between groups A and B. Nevertheless, serum reactivity to milk protein was signi®cantly higher in group B than in group A. This difference in serum reactivity to milk protein between groups A and B is probably related to different correlations of IgG2 with serum reactivity, as seen in Table 4. In groups C and D serum reactivity and the concentrations of milk-speci®c antibodies were lower than in groups A and B because of smaller consumption of milk products. We did not measure serum IgG4 and IgM in this study because our pilot study showed limited range of the values and little variations between the study subjects in these antibodies. Conclusions The subjects participating in this study were not originally recruited for screening milk hypersensitivity but for general health monitoring purposes. Lactose tolerance tests or milk challenges were therefore not performed. Groups B and C presumably included both lactose-intolerant and milkhypersensitive subjects as well as subjects misconnecting their symptoms to milk intake. Taking these aspects into account, we can estimate the prevalence of milk hypersensitivity in young Finnish adults. According to epidemiological studies, about 12% of Finns have lactase de®ciency with symptoms (Sahi, 1974; Sahi et al, 1983; Savaiano &

623

Milk hypersenstivity in young adults L Pelto et al

624

Levitt, 1987; Vesa et al, 1997). Out of our 406 study subjects this would be 49 subjects (vs the actual number of 25 subjects who had been diagnosed by a physician, Figure 1). Subtracting 49 subjects from the total of 82 subjects with abdominal discomfort after dairy product intake, we have 33 subjects who can be considered milkhypersensitive subjects. Assuming that 60% of them can be expected to misrelate their symptoms to milk intake (Bengtsson et al, 1996; Bock, 1987; Nùrgaard & Bindslev-Jensen, 1992), there may be 13 subjects with more probable milk hypersensitivity. Using the actual ®gures (Figure 1), we had 57 potential milk-hypersensitive subjects (groups B and C) of whom 23 (40%) may have milk hypersensitivity. The prevalence of milk hypersensitivity can then be estimated to be between 13 ± 23 out of 406, or 3 ± 6%. This compares well with the prevalence of cow milk allergy in children under the age of three which is reported to vary between 2.2 ± 3.5% (Hùst & Halken, 1990; Savilahti et al, 1991;Schrander et al, 1993). On the other hand, also other groups than B and C may include milkhypersensitive subjects who have not connected their symptoms to milk or are both lactose-intolerant and milkhypersensitive. Our estimate is based on a sample of 27 y old adults. It remains to be studied what the prevalences of milk hypersensitivity are in age groups younger and older than 27 y. Serum reactivity to milk protein may be a good method for screening milk hypersensitivity in subjects who have not previously connected their symptoms to milk, and have not restricted their dairy product consumption. This kind of test is at present available only for screening IgE-mediated milk hypersensitivity. The same principle may also be applied to screen for hypersensitivity to other food stuffs. Acknowledgements ÐWe thank Katariina RasinmaÈki, Hanna-Kaisa ErkkilaÈ and Perttu PoÈllaÈnen for laboratory assistance.

References Anonymous (1998): Balanced sheet for food commodities 1996 and 1997 (preliminary). Information Centre of the Ministry of Agriculture and Forestry, Helsinki. Awazuhara H, Kawai H & Maruchi N (1997): Major allergens in soybean and clinical signi®cance of IgG4 antibodies investigated by IgE- and IgG4-immunoblotting with sera from soybean-sensitive patients. Clin. Exp. Allergy 27, 325 ± 332. È & Ahlstedt S (1996): Double Bengtsson U, Nilsson-BalknaÈs U, Hanson LA blind, placebo controlled food reactions do not correlate to IgE allergy in the diagnosis of staple food related gastrointestinal symptoms. Gut 39, 130 ± 135. Bischoff SC, Mayer J, Wedemeyer J, Meier PN, Zeck-Kapp G, Wedi B, Kapp A, Cetin Y, Gebel M & Manns MP (1997): Colonoscopic allergen provocation (COLAP): a new diagnostic approach for gastrointestinal food allergy. Gut 40, 745 ± 753. Bock SA (1987): Prospective appraisal of complaints of adverse reactions to foods in children during the ®rst 3 years of life. Pediatrics 79, 683 ± 688. Bruijnzeel-Koomen C, Ortolani C, Aas K, Bindslev-Jensen C, BjoÈrksten B, Moneret-Vautrin D & WuÈthrich B (1995): Adverse reactions to food. Position paper. Allergy 50, 623 ± 635.

Challacombe SJ & Tomasi TB Jr (1980): Systemic tolerance and secretory immunity after oral immunization. J. Exp. Med. 152, 1459 ± 1472. Duchen K, Einarsson R, Grodzinsky E, Hattevig G & Bjorksten B (1997): Development of IgG1 and IgG4 antibodies against beta-laktoglobulin and ovalbumin in healthy and atopic children. Ann. Allergy Asthma Immunol. 78, 363 ± 368. Gallagher PJ, Goulding NJ, Gibney MJ, Jones DB & Morgan J (1983): Acute and chronic immunological response to dietary antigen. Gut 24, 831 ± 835. Hùst A & Halken S (1990): A prospective study of cow milk allergy in Danish infants during the ®rst 3 years of life. Allergy 45, 587 ± 596. Keller KM, Burgin-Wolff A, Lippold R, Wirth S & Lentze MJ (1996): The diagnostic signi®cance of IgG cow's milk protein antibodies re-evaluated. Eur. J. Pediatr. 155, 331 ± 337. Kolopp-Sarda MN, Moneret-Vautrin DA, Gobert B, Kanny G, Brodschii M, Bene MC & Faure GC (1997): Speci®c humoral immune responses in 12 cases of food sensitization to sesame seed. Clin. Exp. Allergy 27, 1285 ± 1291. Lindberg T, Nilsson LA, Borulf S, Cavell B, Fallstrom SP, Jansson U, Stenhammar L & Stintzing G (1985): Serum IgG and IgG gliadin antibodies and small intestinal mucosal damage in children. J. Pediatr. Gastroenterol. Nutr. 4, 917 ± 922. Lovegrove JA, Osman DL, Morgan JB & Hampton SM (1993): Transfer of cow's milk b-lactoglobulin to human serum after a milk load: a pilot study. Gut 34, 203 ± 207. Marinkovich V (1996): Speci®c IgG antibodies as markers of adverse reactions to foods. Monogr. Allergy 32, 221 ± 225. Nùrgaard A & Bindslev-Jensen C (1992): Egg and milk allergy in adults. Allergy 47, 503 ± 509. Pelto L, Salminen S, Lilius E-M, Nuutila J & Isolauri E (1998): Milk hypersensitivityÐkey to poorly de®ned gastrointestinal symptoms in adults. Allergy 53, 307 ± 310. Sahi T, Launiala K & Laitinen H (1983): Hypolactasia in a ®xed cohort of young Finnish adults. A follow-up study. Scand. J. Gastroenterol. 18, 865 ± 870. Sahi T (1974): Lactose malabsorption in Finnish-speaking and Swedishspeaking populations in Finland. Scand. J. Gastroenterol. 9, 303 ± 308. Savaiano DA & Levitt MD (1987): Milk intolerance and microbe-containing dairy foods. J. Dairy Sci. 70, 397 ± 406. Savilahti E, Tainio VM, SalmenperaÈ L, Arjomaa P, Perheentupa J & Siimes MA (1991): Low colostral IgA associated with cow's milk allergy. Acta Pediatr. Scand. 80, 1207 ± 1213. Schrander JJ, van der Bogart JP, Forget PP, Schrander-Stumpel CT, Kuijten RH & Kester AD (1993): Cow's milk protein intolerance in infants under 1 year of age: a prospective epidemiological study. Eur. J. Pediatr. 152, 640 ± 644. Scott H, Ek J & Brandtzaeg P (1985): Changes of serum antibody activities to various dietary antigens related to gluten withdrawal or challenge in children with coeliac disease. Int. Arch. Allergy Appl. Immunol. 76, 138 ± 144. Shimojo N, Kohno Y, Katsuki T, Honma K, Yamaguchi K, Tsunoo H, Takahashi T, Kaminogawa S & Niimi H (1997): Comparison of speci®city between IgG, IgE and T cells to three casein components: implication for the role of circulating allergen-speci®c T cells in food allergy. Acta Pediatr. Jpn. 39, 416 ± 421. Spuergin P, Walter M, Schiltz E, Deichmann K, Forster J & Mueller H (1997): Allergenicity of alpha-caseins from cows, sheep, and goat. Allergy 52, 293 ± 298. Vesa TH, Lember M & Korpela R (1997): Milk fat does not affect the symptoms of lactose intolerance. Eur. J. Clin. Nutr. 51, 633 ± 636. Wauters EA, Jansen J, Houwen RH, Veenstra J & Ockhuizen T (1991): Serum IgG and IgA anti-gliadin antibodies as markers of mucosal damage in children with suspected celiac disease upon gluten challenge. J. Pediatr. Gastroenterol. Nutr. 13, 192 ± 196. Werfel T, Ahlers G, Schmidt P, Boeker M, Kapp A & Neumann C (1997): Milk-responsive atopic dermatitis is associated with a casein-speci®c lymphocyte response in adolescent and adult patients. J. Allergy Clin. Immunol. 99, 124 ± 133. Wuthrich B (1996): Speci®c IgG antibodies as markers of adverse reactions to food. Contra! Monogr. Allergy 32, 226 ± 227.

Suggest Documents