Prevalence and associated factors of abnormal liver values in children with celiac disease

G Model YDLD-3162; No. of Pages 7

ARTICLE IN PRESS Digestive and Liver Disease xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld

Alimentary Tract

Prevalence and associated factors of abnormal liver values in children with celiac disease Linnea Äärelä a,b , Samuli Nurminen a,b , Laura Kivelä a,b , Heini Huhtala c , Markku Mäki a , Anna Viitasalo e , Katri Kaukinen b,d , Timo Lakka e,f,g , Kalle Kurppa a,∗ a

Tampere Centre for Child Health Research, University of Tampere and Tampere University Hospital, Tampere, Finland School of Medicine, University of Tampere, Finland c Tampere School of Health Sciences, University of Tampere, Tampere, Finland d Department of Internal Medicine, Tampere University Hospital, Tampere, Finland e Department of Physiology, Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland f Kuopio Research Institute of Exercise Medicine, Kuopio, Finland g Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital, Kuopio, Finland b

a r t i c l e

i n f o

Article history: Received 9 April 2016 Accepted 24 May 2016 Available online xxx Keywords: Alanine aminotransferase Gluten-free diet Paediatrics Transaminases

a b s t r a c t Background: The prevalence and factors associated with transaminasemia in celiac disease are poorly known. Aims: To investigate these issues in paediatric celiac patients and controls. Methods: Alanine aminotransferase (ALT) was studied in 150 children with untreated celiac disease, 161 disease controls and 500 population-based controls. The association between ALT and clinical and histological variables and the effect of a gluten-free diet were investigated in celiac patients. Results: ALT was >30 U/l: celiac disease 14.7%, ulcerative colitis 37.2%, Crohn’s disease 16.7%, reflux disease 16.2%, functional gastrointestinal symptoms 8.9%, and controls 3.6%. Factors associated with increased ALT were poor growth (45.5% vs 24.2%, P = 0.039) and severe villous atrophy (median 23.0 U/l vs partial atrophy 19.0 U/l, P = 0.008), but not age, sex, body-mass index, type or severity of symptoms and comorbidities. ALT had a moderate correlation with endomysial (r = 0.334, P < 0.001) and transglutaminase antibodies (r = 0.264, P = 0.002) and ferritin (r = −0.225, P = 0.03), but not with other laboratory values. On gluten-free diet median ALT decreased from 22.0 U/l to 18.0 U/l (P = 0.002) and 80% of the high values normalized. Conclusion: Increased ALT is associated with more advanced serological and histological celiac disease. Adherence to a gluten-free diet appears to result in normalization or reduction of ALT levels. © 2016 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

1. Introduction The estimated prevalence of celiac disease is as high as 1–2% in many Western countries and even rising [1,2]. Currently most patients remain undiagnosed, but the clinical prevalence has also increased rapidly, being now for example in the Scandinavian countries more than 0.5% [3–5]. The major reason for the improved diagnostic efficacy is recognition of the diversity of the clinical picture, including a variety of extra-intestinal symptoms [6,7]. One of the best-characterized of these is hypertransaminasemia [8–11],

∗ Corresponding author at: Tampere Center for Child Health Research, University of Tampere and Tampere University Hospital, Finn Medi 3, 33520 Tampere, Finland. Tel.: +358 3 3551 8403; fax: +358 3 3551 8402. E-mail address: kalle.kurppa@uta.fi (K. Kurppa).

the prevalence of which in untreated adults has been up to 42% [12–14]. Only a limited number of studies have been conducted in children, but the percentages have been generally comparable to those reported in adults [9,15,16]. Interestingly, we recently observed only 11% of Finnish adults with celiac disease to have elevated transaminases at diagnosis [17]. A plausible explanation for such a low prevalence is early diagnosis of celiac disease due to active case-finding and increased at-risk group screenings. Supporting such a conception, even normal transaminase values improved on a gluten-free diet, indicating the presence of early gluten-dependent liver damage [17]. It is unclear whether a similar decrease in transaminasemia is under way in children, where the diagnostic approach and clinical picture have also changed markedly during the 2000s [3]. Moreover, it remains obscure what patient-related factors are associated with increased liver values in celiac disease.

http://dx.doi.org/10.1016/j.dld.2016.05.022 1590-8658/© 2016 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model YDLD-3162; No. of Pages 7 2

ARTICLE IN PRESS L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

Here we further elucidated these issues by comparing the median levels and prevalence of increased alanine aminotransferase (ALT) values between children with untreated celiac disease, children with other chronic gastrointestinal diseases, and population-based controls. Further, we investigated the association between ALT levels and different clinical, laboratory and histological parameters and the effect of a gluten-free diet in celiac disease patients. We hypothesized that the prevalence of increased ALT has decreased also in children with celiac disease, and that increased values are associated with more severe disease.

2. Materials and methods 2.1. Study design and participants The study was conducted at the Tampere Centre for Child Health Research, Tampere University Hospital and the University of Tampere, and at the Institute of Biomedicine, University of Eastern Finland. It involved well-defined cohorts of children with celiac disease and those with other gastrointestinal diseases, and healthy children from a general population. Inclusion criteria were age below 18 years and available ALT value either at diagnosis (disease groups) or at study enrolment (population controls). Exclusion criteria were unclear diagnosis of celiac disease and other gastrointestinal diseases and presence of medication (such as liver-toxic drugs, supplements or herbal products) known to affect the liver metabolism. Children with known co-existing condition (such as congenital liver disease, malignancy, viral or autoimmune hepatitis, primary sclerosing cholangitis, alpha-1-antitrypsin deficiency, Wilson’s disease, other metabolic liver disease) possibly affecting liver values, were excluded from all study groups. In addition, celiac disease patients were further investigated for possible underlying liver diseases in case of persistently high liver values despite a gluten-free diet. Children with biopsy-proven (Marsh IIIa-c) celiac disease were collected from our regularly updated database, comprising a large cohort of children diagnosed at the Department of Paediatric Gastroenterology [3]. Medical information on the patients has been collected systemically from the medical records and subsequently complemented with personal/parent interviews by an experienced study nurse or a paediatrician. The hospital is tertiary centre with catchment area of around one million inhabitants. In our settings primary screening of celiac disease is conducted mainly in primary care, but regardless of the disease severity children with a celiac disease suspicion are referred to Department of Paediatric Gastroenterology for further investigations. All disease patients for the present study were diagnosed at the same site between the years 2002–2014. From the database we retrospectively selected all consecutive celiac disease patients who had ALT value recorded at diagnosis. However, from the year 2012 onwards more than 90% of the children have been enrolled prospectively, and at the same time ALT has been included in the routine diagnostic evaluation, while before 2012 it was measured irregularly without systematic recommendations. The disease control groups comprised 79 untreated children with inflammatory bowel disease (IBD) (36 with Crohn’s disease and 43 with ulcerative colitis), 37 untreated children with gastrointestinal reflux disease (GERD) and 45 children with functional abdominal complaints. The patients were diagnosed in 2007–2014 at the same site as the celiac disease patients. The diagnoses of IBD and GERD were based on the criteria set by the European Society for Pediatric Gastroenterology Hepatology and Nutrition [18–20]. The diagnosis of functional gastrointestinal disorder was set for children who had undergone thorough clinical, laboratory

and endoscopic investigations due to unspecific abdominal symptoms without established organic cause [21]. The population-based control group was obtained from the Physical Activity and Nutrition in Children (PANIC) study [22]. Altogether 736 children aged 6–8 years who started primary school in the Kuopio region in 2007–2009 were invited to participate in a physical activity and dietary intervention study. Of these 512 children participated and underwent analyses of general health, growth and metabolic laboratory values as described in detail elsewhere [22]. After exclusion criteria the final group comprised 500 children. After collection of the study data (see below), all groups underwent comparisons of ALT levels and percentages of elevated values. Further, the association between baseline ALT values and different clinical, serological, laboratory and histological parameters was investigated in celiac patients, as well as the effect of the gluten-free diet. The study was approved by the Ethics Committees of Pirkanmaa Hospital District and the Research Ethics Committee of the Hospital District of Northern Savo, and by the Department of Pediatrics, Tampere University Hospital. In addition, all children and/or their parents participating in prospective study enrolment or personal interviews gave written informed consent. 2.2. General assessments The following clinical information was recorded on children in the disease groups at the time of diagnosis and on the population controls during the study visit: age, gender, presence of other diseases and medications, body mass index (BMI, kg/m2 ) and BMI standard deviation score and ALT value (U/l). In addition, weight loss, baseline haemoglobin value and possible presence of anaemia were recorded in the disease groups, as well as gastrointestinal symptoms such as diarrhoea, melena/hematochezia, abdominal pain, constipation, gastroesophageal reflux, vomiting and dysphagia. Of the different liver biochemical tests only ALT was used, since it is more specific for hepatocyte injury than aspartate aminotransferase and almost exclusively used in the first-line screening test for liver damage in children [23]. In the present study ALT values higher than >30 U/l were considered increased in both genders [24–26]. 2.3. Celiac disease-specific data 2.3.1. Clinical parameters In addition to the aforesaid variables, the following clinical data were recorded on the children with celiac disease at diagnosis: type (gastrointestinal, extra-intestinal and screendetected) and severity (asymptomatic, mild, moderate and severe) of clinical symptoms, possible presence of concomitant celiac disease-associated conditions (type 1 diabetes, autoimmune thyroidal disease, Down’s syndrome), and growth parameters. Poor growth was defined as a significant deviation from the expected height or abnormal growth velocity [27]. 2.3.2. Laboratory values, celiac disease serology and histology The following blood parameters in addition to ALT and haemoglobin were collected on each celiac disease patient when available at the time of diagnosis: alkaline phosphatase (U/l), mean corpuscular volume (fl), total iron (␮mol/l), transferrin receptor 1 (mg/l), ferritin (␮g/l), albumin (g/l), thyroid-stimulating hormone (TSH) (mU/l) and thyroxin (pmol/l). These laboratory values were recorded in order to further estimate the overall severity of the disease. Serum endomysial antibodies (EmA) were measured by an indirect immunofluorescence method using human umbilical cord as substrate [28]. Titres 1:5 or lower were considered positive and

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model YDLD-3162; No. of Pages 7

ARTICLE IN PRESS L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

100

A

90

ALTabove reference (>30 U/l), %

further diluted up to 1:4000 until negative. Either conventional ELISA (Phadia, Uppsala, Sweden) or an automatized EliA assay (Phadia) was used to determine serum transglutaminase 2 (TG2ab) antibodies. In our hospital TG2ab value 7.0 U/l or higher is considered positive and the maximum reported value is 120.0 U/l. At least four forceps biopsies were taken from the duodenum upon gastroscopy in all children with celiac disease suspicion [29]. From 2012 onwards biopsies have also been systemically taken from the duodenal bulb [30]. The specimens were carefully orientated and processed in the pathology department according to our standard operating procedures [31] and the severity of histological damage further categorized into partial (PVA), subtotal (SVA) and total (TVA) villous atrophy. These correspond approximately to Marsh-Oberhuber scores IIIa, IIIb, and IIIc, respectively.

3

2.3.3. Gluten-free diet and follow-up ALT All children with proven celiac disease started a gluten-free diet shortly after the diagnosis. Compliance was routinely evaluated at each follow-up visit by combining self- and/or parent-reported dietary adherence and the results of celiac serology [32]. The effect of the dietary treatment in ALT was assessed by recording the follow-up values after a median of six (range 4–30) months after the diagnosis.

80 70 60 50 40 30 20 10 0 Celiac disease 500

Ulcerative colitis

Crohn’s disease

Reflux disease

Functional Population symptoms controls

B

400 300

2.4. Statistical analysis

200 100

ALT, U/l

Independent sample Kruskal–Wallis test, Chi-square test and Fisher’s exact test were used to compare baseline differences between the study groups in ALT values and proportion of increased values. The association between ALT and other disease variables was tested either. The correlation between ALT values and continuous parameters was analyzed with Spearman correlation test. Wilcoxon signed-rank test was used to assess significance in the change in ALT values on a gluten-free diet. Statistical significance was defined as P value <0.05. All analyses were performed using SPSS version 22.0 (SPSS Inc., Chicago, IL, USA).

80

60

40

20

3. Results Altogether 156 children with celiac disease had ALT value recorded at diagnosis and, after applying the exclusion criteria, the final cohort consisted 150 children. Of these, 92 (61%) had been diagnosed after the year 2011 and were thus enrolled prospectively. Children with celiac disease were somewhat younger than those in the other study groups except the population controls, and also more often girls (Table 1). Furthermore, celiac disease patients as well as those with IBD had lower median haemoglobin values and presented more often with anaemia at diagnosis than those in the other disease groups (Table 1). Elevated ALT was less common among the population controls than in the disease groups; in children with a specific diagnosis those with ulcerative colitis had ALT above reference range more often than seen in association with other gastrointestinal conditions (Fig. 1A). There was also a trend towards a lower prevalence of increased ALT in those with functional symptoms compared with the other disease groups, but this was not significant (Fig. 1A). The differences between the groups in median ALT levels were fairly small, but patients with ulcerative colitis had the widest dispersion of the values, this explaining their higher percentage of increased values (Fig. 1B). Characteristics of the celiac children divided into two subgroups based on the presence of normal or elevated ALT are presented in Table 2. Poor growth was significantly more common and EmA, TG2ab and TSH values higher in patients with increased ALT, whereas there was no difference in age, gender, presence of

Celiac Ulcerative Crohn’s disease, disease, colitis, n=150 n=43 n=36

Reflux Functional Population disease, symptoms, controls, n=500 n=37 n=45

Fig. 1. (A) Percentage of increased alanine aminotransferase (ALT) values (>30 U/l) in different gastrointestinal diseases at the time of diagnosis and in populationbased control children. Patients with ulcerative colitis and population controls were significantly different from the other groups (P < 0.001). (B) Median (quartiles, range) values of alanine aminotransferase (ALT) in different gastrointestinal diseases and in population-based control children. When adjusted for multiple comparisons there was a significant difference between celiac disease vs controls (P = 0.001), functional symptoms (P = 0.002) and Crohn’s disease (P = 0.029) and functional symptoms vs ulcerative colitis (P = 0.036).

anaemia and other than the above-mentioned laboratory parameters (Table 2). There were also trends for associated conditions and weight loss to be more common in those with increased ALT, but these were not significant (Table 2). The corresponding median ALT values for binary variables were: boys 21.0 U/l vs girls 20.0 U/l, P = 0.388; poor growth 23.0 U/l vs normal growth 19.5 U/l, P = 0.048; weight loss 29.0 U/l vs no loss 17.5 U/l, P = 0.032; anaemia 22.0 U/l vs no anaemia 20.0 U/l, P = 0.237; and associated disease 24.0 U/l vs no disease 20.0 U/l, P = 0.207. Altogether 53% of the celiac disease patients had been diagnosed due to gastro-intestinal symptoms, 30% because of extra-intestinal presentation and 17% had been screen-detected. Six percent were asymptomatic, 48% had mild, 38% moderate and 8% severe

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model

ARTICLE IN PRESS

YDLD-3162; No. of Pages 7

L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

4

Table 1 Clinical characteristics and haemoglobin at diagnosis in children with different gastrointestinal diseases and in population-based controls.

Age, median (quartiles), years BMI, mean (95% CI), kg/m2 BMI SDS, mean (95% CI), Haemoglobin, mean (95% CI), g/l Anaemia, N (%) Girls, N (%) Symptoms, N (%) Diarrhoea Melena/hematochezia Abdominal pain Weight loss Constipation Reflux Vomiting Dysphagia Othera

Celiac disease n = 150

Ulcerative colitis n = 43

Crohn’s disease n = 36

Gastrointestinal reflux disease n = 37

Functional symptoms n = 45

Population controls n = 500

7.3 (4.3,11.8) 17.2 (16.4–17.9) −0.3 (−0.6 to 0.0) 119 (117–121) 37 (25) 103 (69)

12.7 (9.7,14.4) 18.1 (16.7–19.6) −0.3 (−0.8 to 0.3) 120 (114–126) 17 (46) 18 (42)

12.4 (10.3,14.1) 16.4 (14.9–17.9) −1.1 (−1.8 to −0.4) 117 (109–125) 13 (43) 17 (47)

10.3 (5.0,13.0) 18.8 (16.4–21.2) 0.4 (−0.3 to 1.0) 129 (122–135) 4 (14) 14 (38)

11.3 (8.2,14.0) 18.5 (16.9–20.1) −0.1 (−0.6 to 0.4) 134 (131–137) 0 (0) 23 (51)

7.6 (7.4,7.9) 16.1 (15.9–16.3) −0.2 (−0.3 to −0.1) ND ND 240 (48)

54 (40) 6 (4) 76 (56) 32 (53) 37 (27) 5 (4) 7 (5) 0 (0) 62 (41)

20 (47) 36 (84) 26 (61) 12 (28) 2 (5) 2 (5) 2 (5) 0 (0) 10 (23)

17 (47) 15 (42) 18 (50) 11 (31) 4 (11) 0 (0) 3 (8) 2 (6) 12 (33)

13 (35) 4 (11) 23 (62) 3 (8) 5 (14) 21 (57) 14 (38) 2 (5) 9 (24)

21 (47) 3 (7) 40 (89) 6 (13) 9 (20) 4 (9) 5 (11) 0 (0) 13 (24)

ND ND ND ND ND ND ND ND ND

BMI, body mass index; CI, confidence interval; ND, no data; SDS, standard deviation score. a Fever, recurrent aphtous ulcers, poor appetite, short stature, hair loss, tiredness, proctalgia, mouth pain/swelling, cough, dizziness, headache, arthralgia, excessive belching or flatulence.

symptoms at diagnosis. There were no significant associations between ALT values and either the type (Fig. 2A) or severity (Fig. 2B) of clinical presentation. Also no association was seen between ALT levels and gastrointestinal symptoms divided into different subcategories. Thirty-eight percent of the celiac patients had PVA, 40% SVA and 21% TVA at the diagnostic biopsy, and children with TVA had significantly higher median ALT than those with PVA (Fig. 2C). Overall, increased ALT was present in 7% of the PVA, 17% of the SVA and 26% of the TVA patients (P = 0.058). Taking ALT as a continuous variable there was a moderate negative correlation between ALT and age in Crohn’s patients with borderline significance (r = −0.309, P = 0.067), whereas in other groups this was non-significant (r = −0.15–0.1, P > 0.05). There was also a significant correlation between ALT and BMI in the population group, but with a low correlation coefficient (r = 0.231, P < 0.001), in the Crohn’s disease group r was higher (0.353) but

not significant, and in other groups r was between −0.05 and 0.23 and P > 0.05. In celiac patients ALT showed a moderate correlation with EmA (r = 0.396, P < 0.001) and a negative correlation with total iron (r = −0.307, P = 0.112); there were also statistical significances but lower correlation coefficients with TG2ab and ferritin (r = 0.264, P = 0.002 and r = −0.225, P = 0.033, respectively), but not with other laboratory values (r = −0.14 to 0.18, P > 0.05). Altogether 72% of the celiac patients reported being on a strict gluten-free diet, while 28% had minor/occasional lapses. Celiac autoantibody values decreased or normalized in all children after 12 (3–25) months on diet (change in median: EmA from 1:500 to negative; TG2ab from 120.0 U/l to 4.5 U/l). After six months on the diet ALT values decreased in 73% (80% of those with abnormal value at baseline) and the median value for the whole group was significantly lower compared with baseline (Fig. 3). The median also decreased in subjects with normal value at diagnosis

Table 2 Different clinical, histological and laboratory parameters at the time of diagnosis in 150 celiac disease children with normal or increased (>30 U/l) alanine aminotransferase (ALT) value. Normal ALT

Girls Associated diseasesb Anaemia Poor growth Weight loss

n = 128

Increased ALT

P value

n

%

n

%

89 4 30 30 26

69.5 3.2 23.4 24.2 49.1

14 3 7 10 6

63.6 14.3 33.3 45.5 85.7

Increased ALT n = 22

Normal ALT n = 128

Age, years Height, SD Body mass index, kg/m2 EmA, titre TG2ab, U/l Alkaline phosphatase, U/l Haemoglobin, g/l MCV, fl Transferrin receptor, mg/l Albumin, g/l TSH, mU/l

n = 22

0.582 0.062 0.331 0.039 0.109 P value

na

Median (Q1 , Q3 )

na

Median (Q1 , Q3 )

128 76 72 100 117 80 128 123 63 76 95

7.4 (4.5, 11.6) 0.1 (−0.7, 1.0) 16.2 (14.9, 17.7) 1:500 (1:100, 1:2000) 120.0 (32.0, 120.0) 198 (163, 234) 122 (111, 129) 80.0 (76.0, 83.0) 4.3 (3.6, 5.9) 38.0 (37.0, 40.8) 2.5 (1.7, 3.2)

22 14 14 14 18 11 21 20 9 10 15

5.4 (3.4, 13.0) −0.3 (−1.6, 0.7) 17.0 (15.5, 19.2) 1:2000 (1:1000, 1:4000) 120.0 (120.0, 120.0) 163 (97, 314) 123 (101, 134) 77.5 (72.5, 82.0) 5.3 (3.6, 9.8) 38.0 (34.8, 43.0) 3.1 (2.4, 3.7)

0.601 0.187 0.374 0.006 0.003 0.503 0.836 0.287 0.261 0.989 0.038

EmA, endomysial antibodies; MCV, mean corpuscular volume; SD, standard deviation; TG2ab, transglutaminase 2 antibodies; TSH, thyroid stimulating hormone. a Data available. b Type 1 diabetes, autoimmune thyroidal disease, Down’s syndrome.

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model

ARTICLE IN PRESS

YDLD-3162; No. of Pages 7

L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

P = 0.556 P =0.201

100

100

90

90

80

80

70

70

60 50

30

30

20

20

10

10

Extraintestinal n=45

P =0.481

50 40

Gastrointestinal n=78

P =0.238

60

40

0

P = 0.181

B

P =0.672

ALT, U/l

ALT, U/l

A

5

0

Screen detected n=25

Mild n=69

Moderate n=55

Severe n=12

P = 0.008

C

P =0.320

P =0.279

100 90 80

ALT, U/l

70 60 50 40 30 20 10

0

PVA n=56

SVA n=59

TVA n=31

Fig. 2. Median (quartiles, range) values of alanine aminotransferase (ALT) at celiac disease diagnosis in 150 children divided into three subgroups based on (A) baseline clinical presentation, (B) severity of the symptoms before diagnosis and (C) degree of small-bowel mucosal villous atrophy.

(at diagnosis 19.0 U/l vs on diet 16.0 U/l, P = 0.012). ALT remained elevated on longer follow-up in two children, of whom one was persistently seropositive and reported dietary lapses while the other the value was decreasing but still above reference after two years. 4. Discussion There were several important findings in the present study. First, the prevalence of increased ALT in children with celiac disease appears to be becoming rarer and is nowadays comparable with that in other common gastrointestinal diseases. Further, elevated ALT was associated with more advanced histological and serological disease, and even initially normal values may improve on a gluten-free diet. Together these results indicate that an early diagnosis of celiac disease may prevent future liver problems in children. Here only 14.7% of the celiac patients evinced increased ALT at diagnosis, and the general level of values was also low. Although

still higher than in the population, the figure is much lower than in most previous studies. For example, Farre and colleagues [33] reported increased ALT in 30% of children, Biase et al. [34] in 38% and Bonamico et al. in as many as 57% [16]. Similarly, in celiac adults Bardella et al. found 42% [35] to have increased liver values, and in a recent study by Castelli et al. this was seen in 41% [14]. Making the difference even more conspicuous, we used a rather low ALT value (30 U/l) for the upper limit of normal (ULN) instead of 40 U/l [34]. In fact, using a ULN of 40 U/l the prevalence of elevated ALT here would have been only 7%. The rationale for a lower cut-off was recent evidence indicating that previous values have been set too high to reveal early developing liver damage [24,25]. The fact that here only 3.6% of the unselected population were found to have elevated ALT further substantiates the cut-off used. Of the abovementioned studies Farre et al. set the same cut-off as we [33]. It has been advocated that the reference ranges should be defined even more precisely depending on age and gender [36], but for simplicity we decided to use the same limit for both sexes; this decision would

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model

ARTICLE IN PRESS

YDLD-3162; No. of Pages 7 6

L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

P =0.002 100 90 80

ALT, U/l

70 60 50 40 30 20 10 0 Untreated

GFD

Fig. 3. Alanine aminotransferase (ALT) values in individual celiac disease children (N = 41) at the time of the diagnosis and after a median of six (range 4–30) months on a gluten-free diet (GFD). Solid lines denote decreased values and dotted lines increased values, and thick solid lines denote median values for all cases.

appear justified as there was no association between ALT and sex or age in celiac patients. A likely explanation for the falling prevalence of transaminasemia is the changing clinical picture and earlier diagnosis of celiac disease [3,37]. Physicians have become increasingly alert to the heterogeneous presentation, and serological tests enable noninvasive screening of at-risk children. As a result, figures for celiac disease are increasing rapidly in developed countries, and up to 40% of patients might have been detected by screening [3,37–39]. Even if we found no significant difference in ALT between clinicallyand screen-detected children, none of the latter had a value above 50 U/l. Another factor promotive of reduced diagnostic delay is decentralization of the first-line case-finding from tertiary to primary care [40]. Most previous studies [33,35,41–43] have been conducted in special centres and severe presentation has probably been overrepresented. Accordingly, in our earlier adult study, which also showed a low prevalence of increased liver values, most patients had mild presentation [17]. The few other studies reporting a low prevalence for transaminasemia have also been conducted mostly in countries in which celiac disease is familiar among physicians [12]. Interestingly, the presentation might be ameliorating not solely because of improved diagnostics but also due to changing environmental factors [3]. Obviously the decreasing prevalence of transaminasemia might also be explained for example by population-level changes in general health and nutrition, but this hardly explains the differences between developed countries. Besides milder clinical presentation, the earlier detection of celiac disease has resulted in less severe histological damage [3]. In fact, nowadays forthcoming celiac patients are often found even before the development of villous atrophy [41,44]. This is relevant in the context of liver function, as we now discovered increased ALT to be more common in children with advanced enteropathy. In line, these children presented more often with poor growth and there was also a trend towards a higher frequency of anaemia and weight loss. Accordingly, Zanini and colleagues also reported increased ALT to be associated with more severe villous atrophy in adults [13]. In contrast, an older paediatric study found no such association, but it involved only 27 children [45]. We also detected an association between high autoantibody values and increased ALT, which is in line with the known correlation between serology and histology in celiac disease [46]. However, there is also evidence to indicate that the antibodies might in fact participate in

the pathogenesis [10,11,47]. If proven, this would further emphasize the importance of early diagnosis. Of other associations, the fact that a link between ALT and BMI was seen only in the population was unexpected, but this is probably explained by the low prevalence of obesity in the disease groups. A similar absence of association has been reported by Zanini et al. in adults [13]. Interestingly, TSH was also higher in celiac patients with elevated ALT, and there is a known link between thyroid and liver diseases [48]. Practically all celiac children here evinced decrease in ALT on a gluten-free diet. This beneficial response is in accord with previous studies in children and adults [8,43,45,49]. In fact, the reason for persistently abnormal values has usually been either a concomitant liver-affecting disease or dietary lapses [17,34,35,50], showing that transaminasemia is reversible on a strict diet. Interestingly, ALT also decreased in many patients with a normal value at diagnosis. Although this remains speculative, such a finding could indicate the presence of incipient liver damage which might have progressed without the diet. This constitutes further evidence that the ULNs currently used might be set too high; in fact they should perhaps be even lower than 30 U/l [24]. In any case, in view of the rapid response of ALT to the diet one might ask whether early diagnosis is so important. However, progression of hepatic damage in individual patients is difficult to predict, and untreated celiac disease may even predispose to fulminant liver failure [51]. Further, chronic liver injury may lead to fibrosis or cirrhosis possibly irreversible even on a gluten-free diet [52]. Conversely, it could be argued that it is unnecessary to control only mildly elevated values in children with proven adherence and no signs of other hepatic disease, but we cannot define the precise cutoff and this must be decided independently in each case. Apparently children with known dietary lapses or imperfect response to the diet should be subjected to careful follow-up. Our main strengths were the well-defined and large cohorts of celiac patients and controls. Celiac patients were also diagnosed according to harmonized nationwide guidelines and we had a wide selection of their medical data available. The main limitations were the partly retrospective design and the lack of systematic collection of follow-up ALT values. Moreover, even if re-measured, the timing of this varied substantially and in some cases it was probably too soon for ALT to normalize. Also, possible concomitant liver diseases were not meticulously excluded in all children, but we considered this unnecessary in cases of mildly elevated ALT and dietary response. Although laboratory values other than serology and ALT were collected mainly to estimate disease severity, the fact that in many cases they were lacking may have caused a selection bias. There can also be age-related differences in paediatric ALT values which might hamper comparisons between the study groups, but these are mostly seen in very young children not included in the present study. Finally, although we consider the use of the modern low cut-off value for increased ALT necessary, this inevitably limits the comparability of our results with previous studies. In conclusion, the present study indicates that early diagnosis and treatment of celiac disease reduces the risk of liver damage. Nevertheless, physicians should remember celiac disease as a potential cause of unexplained hypertransaminasemia and screen such cases with a low threshold. Possible liver manifestations should be excluded especially in patients with severe serological and histological presentation. An excellent dietary response may obviate the need for routine control of mild transaminasemia in children with a strict diet and no signs of other disease.

Conflict of interest None declared.

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022

G Model YDLD-3162; No. of Pages 7

ARTICLE IN PRESS L. Äärelä et al. / Digestive and Liver Disease xxx (2016) xxx–xxx

Acknowledgements This study was supported by the Academy of Finland Research Council for Health, the Competitive State Research Financing of the Expert Responsibility Areas of Tampere University Hospital, the Foundation for Pediatric Research, the Finnish Medical Foundation, the Sigrid Juselius Foundation and the Päivikki and Sakari Sohlberg Foundation. References [1] Lohi S, Mustalahti K, Kaukinen K, et al. Increasing prevalence of coeliac disease over time. Alimentary Pharmacology and Therapeutics 2007;26:1217–25. [2] Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology 2009;137:88–93. [3] Kivelä L, Kaukinen K, Lähdeaho ML, et al. Presentation of celiac disease in Finnish children is no longer changing: a 50-year perspective. Journal of Pediatrics 2015;167:1109–15. [4] Ivarsson A, Myléus A, Norström F, et al. Prevalence of childhood celiac disease and changes in infant feeding. Pediatrics 2013;131:e687–94. [5] Mårild K, Kahrs CR, Tapia G, et al. Infections and risk of celiac disease in childhood: a prospective nationwide cohort study. American Journal of Gastroenterology 2015;110:1475–84. [6] Green PH, Cellier C. Celiac disease. New England Journal of Medicine 2007;357:1731–43. [7] Leffler DA, Green PH, Fasano A. Extraintestinal manifestations of coeliac disease. Nature Reviews Gastroenterology & Hepatology 2015;12:561–71. [8] Hagander B, Berg NO, Brandt L, et al. Hepatic injury in adult coeliac disease. Lancet 1977;2:270–2. [9] Anania C, De Luca E, De Castro G, et al. Liver involvement in pediatric celiac disease. World Journal of Gastroenterology 2015;21:5813–22. [10] Rubio-Tapia A, Murray JA. Liver involvement in celiac disease. Minerva Medica 2008;99:595–604. [11] Marciano F, Savoia M, Vajro P. Celiac disease-related hepatic injury: insights into associated conditions and underlying pathomechanisms. Digestive and Liver Disease 2016;48:112–9. [12] Sainsbury A, Sanders DS, Ford AC. Meta-analysis: coeliac disease and hypertransaminasaemia. Alimentary Pharmacology and Therapeutics 2011;34:33–40. [13] Zanini B, Baschè R, Ferraresi A, et al. Factors that contribute to hypertransaminasemia in patients with celiac disease or functional gastrointestinal syndromes. Clinical Gastroenterology and Hepatology 2014;12:804–10. [14] Castillo NE, Vanga RR, Theethira TG, et al. Prevalence of abnormal liver function tests in celiac disease and the effect of a gluten-free diet in the US population. American Journal of Gastroenterology 2015;110:1216–22. [15] Vajro P, Paolella G, Maggiore G, et al. Pediatric celiac disease, cryptogenic hypertransaminasemia, and autoimmune hepatitis. Journal of Pediatric Gastroenterology and Nutrition 2013;56:663–70. [16] Bonamico M, Pitzalis G, Culasso F, et al. Hepatic damage in celiac disease in children. Minerva Pediatrica 1986;38:959–62. [17] Korpimäki S, Kaukinen K, Collin P, et al. Gluten-sensitive hypertransaminasemia in celiac disease: an infrequent and often subclinical finding. American Journal of Gastroenterology 2011;106:1689–96. [18] IBD Working Group of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition. Inflammatory bowel disease in children and adolescents: recommendations for diagnosis – the Porto criteria. Journal of Pediatric Gastroenterology and Nutrition 2005;41:1–7. [19] Vandenplas Y, Rudolph CD, Di Lorenzo C, et al. Pediatric gastroesophageal reflux clinical practice guidelines: joint recommendations of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN). Journal of Pediatric Gastroenterology and Nutrition 2009;49:498–547. [20] Levine A, Koletzko S, Turner D, et al. ESPGHAN revised Porto criteria for the diagnosis of inflammatory bowel disease in children and adolescents. Journal of Pediatric Gastroenterology and Nutrition 2014;58:795–806. [21] Chiou E, Nurko S. Functional abdominal pain and irritable bowel syndrome in children and adolescents. Therapy 2011;8:315–31. [22] Viitasalo A, Pihlajamäki J, Lindi V, et al. Associations of I148M variant in PNPLA3 gene with plasma ALT levels during 2-year follow-up in normal weight and overweight children: the PANIC study. Pediatric Obesity 2015;10:84–90. [23] Green RM, Flamm S. AGA technical review on the evaluation of liver chemistry tests. Gastroenterology 2002;123:1367–84. [24] Schwimmer JB, Dunn W, Norman GJ, et al. SAFETY study: alanine aminotransferase cutoff values are set too high for reliable detection of pediatric chronic liver disease. Gastroenterology 2010;138:1357–64.

7

[25] Stirnadel-Farrant HA, Galwey N, Bains C, et al. Children’s liver chemistries vary with age and gender and require customized pediatric reference ranges. Regulatory Toxicology and Pharmacology 2015;73:349–55. [26] Fraser A, Longnecker MP, Lawlor DA. Prevalence of elevated alanine aminotransferase among US adolescents and associated factors: NHANES 1999–2004. Gastroenterology 2007;133:1814–20. [27] Nurminen S, Kivelä L, Taavela J, et al. Factors associated with growth disturbance at celiac disease diagnosis in children: a retrospective cohort study. BMC Gastroenterology 2015;15:125. [28] Kurppa K, Paavola A, Collin P, et al. Benefits of a gluten-free diet for asymptomatic patients with serologic markers of celiac disease. Gastroenterology 2014;147:610–7. [29] Pais WP, Duerksen DR, Pettigrew NM, et al. How many duodenal biopsy specimens are required to make a diagnosis of celiac disease. Gastrointestinal Endoscopy 2008;67:1082–7. [30] Husby S, Koletzko S, Korponay-Szabó IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. Journal of Pediatric Gastroenterology and Nutrition 2012;54:136–60. [31] Taavela J, Koskinen O, Huhtala H, et al. Validation of morphometric analyses of small-intestinal biopsy readouts in celiac disease. PLOS ONE 2013;8: e76163. [32] Kurppa K, Lauronen O, Collin P, et al. Factors associated with dietary adherence in celiac disease: a nationwide study. Digestion 2012;86:309–14. [33] Farre C, Esteve M, Curcoy A, et al. Hypertransaminasemia in pediatric celiac disease patients and its prevalence as a diagnostic clue. American Journal of Gastroenterology 2002;97:3176–81. [34] Di Biase AR, Colecchia A, Scaioli E, et al. Autoimmune liver diseases in a paediatric population with coeliac disease – a 10-year single-centre experience. Alimentary Pharmacology and Therapeutics 2010;31:253–60. [35] Bardella MT, Fraquelli M, Quatrini M, et al. Prevalence of hypertransaminasemia in adult celiac patients and effect of gluten-free diet. Hepatology 1995;22:833–6. [36] England K, Thorne C, Pembrey L, et al. Age- and sex-related reference ranges of alanine aminotransferase levels in children: European paediatric HCV network. Journal of Pediatric Gastroenterology and Nutrition 2009;49:71–7. [37] Tapsas D, Hollén E, Stenhammar L, et al. The clinical presentation of coeliac disease in 1030 Swedish children: changing features over the past four decades. Digestive and Liver Disease 2016;48:16–22. [38] Kinos S, Kurppa K, Ukkola A, et al. Burden of illness in screen-detected children with celiac disease and their families. Journal of Pediatric Gastroenterology and Nutrition 2012;55:412–6. [39] Virta LJ, Kaukinen K, Collin P. Incidence and prevalence of diagnosed coeliac disease in Finland: results of effective case finding in adults. Scandinavian Journal of Gastroenterology 2009;44:933–8. [40] Fuchs V, Kurppa K, Huhtala H, et al. Factors associated with long diagnostic delay in celiac disease. Scandinavian Journal of Gastroenterology 2014;49:1304–10. [41] Volta U, Caio G, Giancola F, et al. Features and progression of potential celiac disease in adults. Clinical Gastroenterology and Hepatology 2015. [42] Dickey W, McMillan SA, Collins JS, et al. Liver abnormalities associated with celiac sprue. How common are they, what is their significance, and what do we do about them? Journal of Clinical Gastroenterology 1995;20:290–2. [43] Novacek G, Miehsler W, Wrba F, et al. Prevalence and clinical importance of hypertransaminasaemia in coeliac disease. European Journal of Gastroenterology and Hepatology 1999;11:283–8. [44] Kurppa K, Ashorn M, Iltanen S, et al. Celiac disease without villous atrophy in children: a prospective study. Journal of Pediatrics 2010;157:373–80. [45] Arslan N, Büyükgebiz B, Oztürk Y, et al. The prevalence of liver function abnormalities in pediatric celiac disease patients and its relation with intestinal biopsy findings. Acta Gastroenterologica Belgica 2005;68:424–7. [46] Agardh D, Lee HS, Kurppa K, et al. Clinical features of celiac disease: a prospective birth cohort. Pediatrics 2015;135:627–34. [47] Korponay-Szabó IR, Halttunen T, Szalai Z, et al. In vivo targeting of intestinal and extraintestinal transglutaminase 2 by coeliac autoantibodies. Gut 2004;53:641–8. [48] Malik R, Hodgson H. The relationship between the thyroid gland and the liver. QJM 2002;95:559–69. [49] Alavi Moghaddam M, Rostami Nejad M, Shalmani HM, et al. The effects of gluten-free diet on hypertransaminasemia in patients with celiac disease. International Journal of Preventive Medicine 2013;4:700–4. [50] Casella G, Antonelli E, Di Bella C, et al. Prevalence and causes of abnormal liver function in patients with coeliac disease. Liver International 2013;33: 1128–31. [51] Kaukinen K, Halme L, Collin P, et al. Celiac disease in patients with severe liver disease: gluten-free diet may reverse hepatic failure. Gastroenterology 2002;122:881–8. [52] Volta U. Pathogenesis and clinical significance of liver injury in celiac disease. Clinical Reviews in Allergy & Immunology 2009;36:62–70.

Please cite this article in press as: Äärelä L, et al. Prevalence and associated factors of abnormal liver values in children with celiac disease. Dig Liver Dis (2016), http://dx.doi.org/10.1016/j.dld.2016.05.022