a Department of Internal Medicine and b Clinical Institute of Medical and Chemical Received 23 April 2004 Accepted 2 November 2004

Transcription

1 Original article 371 Evaluation of a new DNA test compared with the lactose hydrogen breath test for the diagnosis of lactase non-persistence Christoph Högenauer a, Heinz F. Hammer a, Karin Mellitzer a, Wilfried Renner b, Günter J. Krejs a and Hermann Toplak a Background and aims Recent publications have found that the CC genotype of the DNA variant T/C upstream of the LCT gene is associated with lactase non-persistence. We therefore compared the value of DNA testing for this variant (DNA test) with the lactose hydrogen breath test (H 2 test), which is the clinical standard for the diagnosis of lactase non-persistence. Patients and methods One hundred and twenty-three consecutive patients with suspected lactose were tested for the presence of the T/C variant by polymerase chain reaction-restriction fragment length polymorphism analysis. These patients also underwent the H 2 test after ingestion of 50 g lactose. Results Thirty-seven subjects had a CC genotype of the T > C polymorphism suggesting lactase non-persistence; 36 (97%) had also a positive H 2 test. Eighty-six subjects had either a TC or a TT genotype suggestive of lactase persistence. Seventy-four (86%) of these tested negative on the H 2 test, while 12 patients had a positive H 2 test. In eight of these 12 patients duodenal biopsies showed no evidence of small bowel disease. One patient carrying a CC genotype had a negative H 2 test. In this patient the rise in serum glucose after oral lactose was normal, furthermore H 2 non-excretion was also excluded. Conclusions An excellent correlation is observed between a CC genotype and a positive H 2 test, whereas the correlation between a TC or TT genotype and a negative H 2 test result is less strong. Analysis of the T/C variant can be considered a good test for predicting the presence of lactase non-persistence in a patient population with suspected lactose. Eur J Gastroenterol Hepatol 17: c 2005 Lippincott Williams & Wilkins. European Journal of Gastroenterology & Hepatology 2005, 17: Keywords: lactase non-persistence, lactose intolerance, lactose, syndromes, genotype, diarrhea a Department of Internal Medicine and b Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Austria. Correspondence and requests for reprints to Christoph Högenauer, M.D., Department of Internal Medicine, Medical University of Graz, Auenbruggerplatz 15, A-8036 Graz, Austria. Tel: ; fax: ; christoph.hoegenauer@meduni-graz.at Received 23 April 2004 Accepted 2 November 2004 Introduction Lactase phlorizin hydrolase [lactase (LPH)] is an intestinal brush border enzyme necessary for the digestion of the disaccharide lactose to glucose and galactose, which can be absorbed by the intestinal mucosa. Deficiency of lactase results in of lactose, which is contained mainly in milk and dairy products. Lactose leads to symptoms of lactose intolerance like diarrhea, bloating, excessive flatus and abdominal pain. Provocation studies found that g lactose, equivalent to one glass of milk, results in these symptoms in 41 75% of subjects with lactose [1]. The most common cause of lactase deficiency is lactase non-persistence (adult hypolactasia). Lactase non-persistence is an autosomal recessive inherited trait, which is present depending on the geographic region in 3 70% of the Caucasian population in Europe. Its incidence in other populations such as Asians is higher, X c 2005 Lippincott Williams & Wilkins reaching up to 100% [1,2]. The onset of lactase nonpersistence is usually during adolescence [3]. Secondary lactase deficiency occurs in diseases that affect large areas of the small intestinal mucosa like celiac disease or giardiasis [1,4]. Currently the clinically most widely used test for lactase deficiency is the lactose hydrogen breath test (lactose H 2 test). The disadvantages of the lactose H 2 test are that it is time consuming, 2 6 h depending on the protocol, and that its sensitivity is influenced by the prevalence of hydrogen non-excretion, which is 18% of the population in this study [5], but has been reported to be as high as 20% [6]. Recently two studies, mainly performed on Finnish subjects, found that two single nucleotide polymorphisms

2 372 European Journal of Gastroenterology and Hepatology 2005, Vol 17 No 3 (SNPs) about 14 and 22 kb upstream of the lactase gene (LCT) are associated with lactase non-persistence [7,8]. One of these polymorphisms ( T > C) was found to be almost perfectly associated with lactase non-persistence: 100% of subjects homozygous for the C allele of T > C (CC genotype) were found to be lactase deficient, as determined by lactase activity of intestinal biopsies. Subjects with a TC or a TT genotype at this polymorphism had normal lactase levels. The second polymorphism ( G > A) was found to be less consistently linked tothepresenceoflactasedeficiency[7]. The purpose of the current study was to compare the DNA analysis for the T > C polymorphism with the lactose H 2 test for the diagnosis of lactase nonpersistence. Patients and methods Patient characteristics One hundred and twenty-three consecutive patients who were referred to our outpatient clinic for the evaluation of suspected lactose were tested. Forty-three patients were male and 80 were female; the mean age was 42 ± 14 years (range, years). All patients were Caucasians and 119 of the patients were of Austrian descent, while four were non-austrians. Symptoms for referral were diarrhea in 44%, bloating and excessive flatus in 72%, abdominal pain in 41% and milk intolerance in 16% of patients. Written informed consent was obtained from the patients, and the study was conducted according to the guidelines of the local ethics committee. Lactose H 2 breath test Patients were instructed to fast and avoid smoking for at least8hbeforethetest.thelactoseh 2 test was performed after oral ingestion of 50 g lactose dissolved in 100 ml water. Breath hydrogen (H 2 ) excretion was recorded every 15 min within the first hour after lactose ingestion, and thereafter every 30 min. An increase in the breath H 2 concentration of at least 20 ppm within 4 h after lactose ingestion was considered as a positive test for lactose. Symptoms after lactose ingestion like diarrhea, bloating and abdominal pain were also recorded. DNA testing For DNA typing, genomic DNA was isolated from venous blood by standard methods and was stored at 41C. A 174 base pair (bp) fragment containing the polymorphic site was amplified by polymerase chain reaction using GTTGTTAGACGGAGACGATCACGT-3 0 as the forward primer and 5 0 -AGGAGAGTTCCTTTGAGGCCAGGT-3 0 as the reverse primer (the mismatching nucleotide of the primer is underlined). The polymerase chain reaction product was digested with restriction endonuclease RsaI (New England Biolab, Frankfurt, Germany), and fragments were analyzed on 3.0% agarose gels stained with ethidium bromide. The T allele was cut into fragments sized 88, 44, 24 and 18 bp, whereas the C allele was cut into fragments of 88, 68 and 18 bp. For each set of reactions, a negative control containing H 2 O instead of DNA and three positive controls (TT, TC and CC genotype) were included. A DNA test suggestive for lactase non-persistence was defined as a CC genotype, while a TC or TT genotype was interpreted as indicative for lactase persistence [7]. Additional testing Subjects with discrepant results between the DNA test and the lactose H 2 test were invited for additional clinical examinations to exclude causes of secondary lactase deficiency or H 2 non-excretion. In subjects with the suspicion of secondary lactase deficiency, small intestinal biopsies were obtained from the third part of the duodenum. In subjects with suspicion of H 2 non-excretion, breath H 2 concentration was measured after 25 g lactulose following the same test protocol as for the lactose H 2 test. H 2 non-excretion was defined as a rise in breath H 2 concentration of less than 20 ppm. In addition, serum glucose was measured 30 min after the ingestion of 50 g lactose [5]. Results Results of lactose H 2 breath test Forty-eight patients (39%) had a positive and 75 (61%) had a negative lactose H 2 test. In patients with a positive lactose H 2 test 34 of 48 (71%) had symptoms after ingestion of 50 g lactose, while in the patient-group with a negative lactose H 2 test only 6/75 (8%) were symptomatic after oral lactose. Results of DNA testing Thirty-seven subjects had a CC genotype of the T > C polymorphism suggesting the genetic disposition for lactase non-persistence. In 86 subjects the DNA analysis suggested lactase persistence: 56 were found to have a TC genotype and 30 had a TT genotype. The results of the DNA testing compared with the results of the lactose H 2 test are presented in Table 1. Ninetyseven percent of patients with a CC allele had a positive lactose H 2 test, while 86% of subjects with either a TC or Table 1 Results of DNA testing for adult-type hypolactasia compared with results of the lactose H 2 test (positive predictive value, 97%; negative predictive value, 86%; sensitivity, 75%; specificity, 99% for lactase non-persistence) Positive lactose H 2 test Negative lactose H 2 test Total CC genotype 36/37 (97%) 1/37 (3%) 37 (lactase non-persistence) TC or TT genotype 12/86 (14%) 74/86 (86%) 86 (lactase persistence) Total

3 DNA testing for lactase non-persistence Högenauer et al. 373 a TT allele were found to have a negative result of the lactose H 2 test. Twenty-seven of 37 patients (73%) with a CC genotype and 14 of 86 patients (16%) with a TT or a TC genotype reported symptoms after lactose ingestions during the H 2 test and were considered to be intolerant to lactose. The peak H 2 increase during the lactose H 2 test compared with the genotype in the subjects is shown in Fig. 1. Only two of 13 subjects with discrepant results between the two tests had a peak hydrogen increase close to the cut off level of the lactose H 2 test of 20 ppm. No difference in the peak hydrogen increase between the TC and the TT genotype was observed. Subjects with discrepant results between the DNA and the H 2 test Twelve subjects had a positive lactose H 2 test while the DNA analysis suggested lactase persistence. Seven out of 12 (58%) of these patients reported symptoms after lactose ingestion. Nine patients had a TC genotype and three had a TT genotype of the T > C polymorphism. Eight patients agreed to an upper gastrointestinal endoscopy with histological examination of small intestinal biopsy; these results and further tests are presented in Table 2. Two additional patients agreed to non-invasive blood tests but refused endoscopy. In none of these patients was a secondary cause of lactase deficiency found. The proportion of subjects with a DNA test suggesting lactase persistence and a positive H 2 test tended to be higher in subjects with the TC genotype than in subjects with the TT genotype. In the group of patients with a TC genotype 9/56 (16%) had a positive lactose H 2 test, whereas in the group with a TT genotype 3/30 (10%) had a positive lactose H 2 test; the difference was, however, not significant as determined by the chi-squared test. One subject, a 39-year-old female, carrying a CC genotype had a negative lactose H 2 test, and reported no symptoms after lactose ingestion. This patient had a normal increase in the H 2 breath excretion after oral 25 g lactulose ruling out H 2 non-excretion. Furthermore, blood glucose levels increased over baseline by 64 mg/dl after 30 min following an oral load of 50 g lactose, which is consistent with normal lactose digestion to glucose and galactose [5]. Discussion The results of this study demonstrate that analysis for the T > C polymorphism is a good indicator for the presence of lactose as defined by a positive lactose H 2 test. In the current study the correlation of a CC genotype suggesting lactase nonpersistence to a positive lactose H 2 test was excellent (36 of 37 subjects, 97%), while the correlation of a TC or a Table 2 Results of further testing and clinical information in subjects with a positive lactose H 2 test and a DNA test suggestive for lactase persistence O&P Symptoms for test Lactose-free diet Clinical diagnosis EMA Ileal biopsy Small bowel enteroclysis Small intestinal biopsy Genotype Symptoms after oral lactose Age (years), gender 52, male TT + Normal n.d. n.d. Rapid transit Negative Bloating No follow-up Lactose intolerance 47, female TC Normal Negative n.d. n.d. Negative Bloating, milk Response Lactose intolerance intolerance 39, female TT + Normal Negative n.d. Rapid transit Negative Diarrhea No response Functional diarrhea, lactose No response Lactose, IBS 50, female TC Normal Negative n.d. n.d. Negative Bloating, abdominal pain 59, male TC + n.d. Negative n.d. Negative Bloating Response Lactose intolerance 45, female TC + Normal Negative Normal Normal n.d. Diarrhea No follow-up Postinfectious IBS, lactose 24, female TC Normal Negative Normal n.d. Negative Diarrhea No response Functional diarrhea, lactose 28, male a TC + n.d. n.d. n.d. n.d. n.d. Abdominal pain Response Lactose intolerance 66, male a TT n.d. n.d. n.d. n.d. n.d. Diarrhea n.d., spontaneous Infectious diarrhea, lactose improvement 61, female TC + Normal Negative Normal Normal n.d. Diarrhea No response NSAID abuse, lactose 34, female TC + Normal Negative n.d. n.d. Negative Diarrhea No follow-up Lactose intolerance 50, female TC n.d. Negative n.d. n.d. n.d. Abdominal pain No response Lactose, functional dyspepsia Biopsies were assessed histologically by standard hematoxylin and eosin stain and light microscopy. n.d., not done; EMA, endomysial antibodies; O&P, stool testing for ova and parasites; IBS, irritable bowel syndrome; NSAID, non-steroidal anti-inflammatory drug. a Patients refused additional examinations.

4 374 European Journal of Gastroenterology and Hepatology 2005, Vol 17 No 3 TT genotype suggesting lactase persistence to a negative lactose H 2 testwaslower(74of86subjects,86%). Thirteen subjects in this study had discrepant results between the DNA and the H 2 test.whereasthednatest gives a clear cut yes/no answer, the interpretation of the H 2 testdependsonacut-offlevel.however,onlytwoofthe13 patients with discrepant results had a peak hydrogen increase close to the cut-off level at 20 ppm (Fig. 1). Therefore, a change in the cut-off level of the lactose H 2 test would not have a major impact on the correlation between the results of the DNA and the H 2 test. Twelve of 48 (25%) patients with a positive lactose H 2 test had a TC or a TT genotype suggesting lactase persistence; one-half of these patients were also symptomatic after lactose ingestion. In none of the patients who returned for further clinical evaluation a secondary cause of lactose could be established, despite use of various tests including histological examination of small bowel biopsy. Small bowel biopsy can pick up a variety of diffuse small intestinal lesions including villous atrophy or giardiasis [3]. Although patchy or distal small bowel disease may be missed, these would most probably not result in secondary lactase deficiency. In addition, causes Fig. 1 H 2 - increase (ppm) Discrepant results Consistent results CC TC TT Comparison of the peak increase of the hydrogen breath test in the different genotypes. Dashed line, cut-off level for a positive H 2 breath test at 20 ppm. Open points, consistent results of the H 2 breath test with the DNA test; filled points, discrepant results. Of the patients with inconsistent results of the tests only two had a peak H 2 increase close to the cut-off level, whereas the rest had either clearly positive or negative results. The bars show the mean peak increase in hydrogen breath concentration with no difference between the TC and the TT genotypes (TC versus TT, P = 0.97; CC versus TT, P < 0.001; CC versus TC, P < P values assessed by one-way analysis of variance, and all pairwise multiple comparisons by the Tukey test). of secondary lactase deficiency are infrequent as compared with lactase non-persistence. There are three probable explanations for these discrepant results. Most of the patients (9/12) were found to be heterozygous for the C allele. In a previous study heterozygotes with a TC genotype were found to have about one-half of the intestinal lactase activity as compared with subjects with a TT genotype [8]. The 50 g dose of lactose used in this study for the lactose H 2 test, which is a commonly recommended dose [4,9], is rather high and is about the amount of lactose contained in 1 l milk. Therefore it is possible that in heterozygous subjects with lower lactase activities high doses of lactose are malabsorbed, while lower doses of lactose might be tolerated. However, no significant differences were detected in the proportion of patients with a positive result on the H 2 test between subjects with a TC or a TT genotype. It remains to be evaluated if lower doses of lactase as 20 g, which are recommended by some authors [10], would influence the correlation between the H 2 test and the DNA test. Second, other factors like accelerated intestinal transit could have contributed to lactose and intolerance in these patients by decreasing contact time despite normal intestinal lactase activity. This is suggested by the results of enteroclysis in two of the patients. However, intestinal transit in enteroclysis has to be interpreted with caution. Finally, it is possible that in these patients lactase nonpersistence was associated with genetic factors other than the T > C polymorphism. One option is the second SNP G > A, in which a GG genotype is associated with lactase non-persistence. However in two previous studies reporting 236 and 53 subjects, respectively, all subjects who had a GG genotype at position carried also the CC genotype at position [7,8]. Two recent studies found that the T allele at the locus has a strong enhancing activity on the LPH promotor [11,12], while the SNP does not influence LPH promotor activity significantly [11]. It is therefore unlikely that mutations at the SNP explain the discrepant results. Another recent study indicates that these two SNPs cannot explain all variation of LPH expression, suggesting genetic heterogeneity in the regulation of LPH expression [13]. Furthermore, both SNPs seem to be located within a megabase stretch of linkage disequilibrium, thus it cannot be excluded that the genuine genetic cause of lactase non-persistence lies even much farther away from the LCT gene [13]. Only one of 71 patients (1%) with a negative lactose H 2 test had a CC genotype suggesting lactase non-persistence. Further tests done in this patient excluded H 2

5 DNA testing for lactase non-persistence Högenauer et al. 375 non-excretion and showed normal intestinal lactose digestion. It is theoretically possible that the adult decline in intestinal lactase levels in this patient was delayed and that this patient would develop hypolactasia later in life. It is interesting that in none of the 37 subjects with a CC genotype was hydrogen non-excretion observed, despite a previously reported frequency of 18% in this population. We were unable to confirm in the Austrian population the perfect correlation of the tested SNP to the presence of lactose demonstrated by lactose H 2 test, as suggested in a previous study in Finish subjects [7]. This difference might be explained by several factors. In the present study the genetic analysis for lactase nonpersistence was compared with the lactose H 2 test, which is an indirect test for the presence of lactase non-persistence. The reason for using the lactose H 2 test in this study is that it is the clinical standard for the diagnosis of lactase non-persistence in most centers, since the measurement of lactase activity of intestinal biopsies requires an invasive procedure to obtain tissue. Furthermore, the measurement of lactase activity in intestinal biopsies cannot assess whether a patient is symptomatic to lactose ingestion (i.e. lactose intolerant). The physiologic background for the lactose H 2 test is bacterial metabolism of malabsorbed lactose to hydrogen, which is absorbed by the colonic mucosa and measured in the exhaled air. Previous studies comparing the lactose H 2 test with lactase activity of intestinal biopsies have found different results regarding sensitivity and specificity. In one study the sensitivity and specificity was 100% [14], while a second study using a different protocol for the lactose H 2 test has reported a sensitivity of only 69% and a specificity of 96% [15]. Another possible reason for the discrepancy of results is the genetic heterogenicity between our population and the Finnish population studied by Enattah et al. All patients with discrepant test results in this study were of Austrian descent. Finally, patient selection could also have influenced the results. In this study consecutive patients referred for the evaluation of lactose intolerance were studied, while in the study of Enattah et al. only patients with established hypolactasia or lactase persistence were tested. We did not test for lactase activity in intestinal biopsies in the patients of this study. Therefore we cannot distinguish whether the discrepant results discussed here were due to hypolactasia or other factors influencing lactose digestion. Further studies comparing measurement of lactase activity in intestinal biopsies, lactose hydrogen breath test and DNA testing for the T > C polymorphism are needed to address this issue. In conclusion, analysis for the T > C polymorphism may be a useful clinical tool in the diagnosis of patients with lactase non-persistence resulting in lactose, with a positive predictive value for lactase non-persistence of 97% as compared with the clinically most widely used lactose H 2 test. The negative predictive value of a DNA test result suggesting lactase persistence as compared with a negative lactose H 2 test is, however, lower with 86%. The DNA test therefore cannot rule out the presence of lactose. If one considers the lactose H 2 test as the clinical gold standard in diagnosing lactase deficiency, the sensitivity and the specificity of the DNA test are 75% and 99%, respectively, compared with the H 2 test. The advantages of the DNA test compared with the lactose H 2 test are that it is less expensive (about 50), less time consuming for the patient and less cumbersome since a vial of venous blood needed for analysis can be easily shipped to the laboratory. The major disadvantages as is also the case for the measurement of lactase activity in intestinal biopsies is that a test result suggestive of lactase non-persistence does not provide information of whether the patient has symptoms after an oral lactose load and that secondary causes of hypolactasia cannot be ruled out. Conflict of interest None declared. Authors contributions C.H., H.F.H. and H.T. contributed to the design of the study and acquisition of the data, and wrote the manuscript. K.M. contributed to acquisition of the data. W.R. performed the DNA analysis. G.J.K. contributed to writing the manuscript. References 1 Gudmand-Hoyer E, Skovbjerg H. Disaccharide digestion and maldigestion. Scand J Gastroenterol 1996; 31 (suppl 216): Sahi T. Genetics and epidemiology of adult-type hypolactasia. Scand J Gastroenterol 1994; 29 (suppl 202): Högenauer C, Hammer HF. Maldigestion and. Sleisenger & Fordtran s Gastrointestinal and Liver Disease. 7th edition. In: Feldman M, Friedman LS, Sleisenger MH (eds). Philadelphia, PA: Saunders; pp Saavedra JM, Perman JA. Current concepts in lactose and intolerance. Annu Rev Nutr 1989; 9: Hammer HF, Petritsch W, Pristautz H, Krejs GJ. Assessment of the influence of hydrogen nonexcretion on the usefulness of the hydrogen breath test and lactose tolerance test. Wien Klin Wochenschr 1996; 108: Gilat T, Ben Hur H, Gelman-Malachi E, Terdiman R, Peled Y. Alterations of the colonic flora and their effect on the hydrogen breath test. Gut 1978; 19: Enattah NS, Sahi T, Savilahti E, Terwilliger JD, Peltonen L, Jarvela I. Identification of a variant associated with adult-type hypolactasia. Nat Genet 2002; 30: Kuokkanen M, Enattah NS, Oksanen A, Savilahti E, Orpana A, Jarvela I. Transcriptional regulation of the lactase-phlorizin hydrolase gene by polymorphisms associated with adult-type hypolactasia. Gut 2003; 52: Buller HA, Rings EH, Montgomery RK, Grand RJ. Clinical aspects of lactose intolerance in children and adults. Scand J Gastroenterol 1991; 26 (suppl 188): Di Stefano M, Veneto G, Malservisi S, Cecchetti L, Minguzzi L, Strocchi A, et al. Lactose and intolerance and peak bone mass. Gastroenterology 2002; 122: Troelsen JT, Olsen J, Moller J, Sjostrom H. An upstream polymorphism associated with lactase persistence has increased enhancer activity. Gastroenterology 2003; 125:

6 376 European Journal of Gastroenterology and Hepatology 2005, Vol 17 No 3 12 Olds LC, Sibley E. Lactase persistence DNA variant enhances lactase promoter activity in vitro: functional role as a cis regulatory element. Hum Mol Genet 2003; 12: Poulter M, Hollox E, Harvey CB, Mulcare C, Peuhkuri K, Kajander K, et al. The causal element for lactase persistence/non-persistence polymorphism is located in a 1 mb region of linkage disequilibrium in Europe. Ann Hum Genet 2003; 67: Newcomer AD, McGill DB, Thomas PJ, Hofman AF. Prospective comparison of indirect methods for detecting lactase deficiency. N Engl J Med 1975; 293: Arola H, Koivula T, Jokela H, Jauhiainen M, Keyrilainen O, Hola T, et al. Comparison of indirect diagnostic methods for hypolactasia. Scand J Gastroenterol 1988; 23: