Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Original article

Validation of point-of-care testing for coeliac disease in children in a tertiary hospital in north India Prashant Singh,1 Nitya Wadhwa,1 Mona K Chaturvedi,1 Vidyut Bhatia,1 Savita Saini,1 Nikhil Tandon,2 Govind K Makharia,3 Markku Maki,4 Tarcisio Not,5 Alan Phillips,6 Shinjini Bhatnagar1 1

Department of Pediatrics, Center for Diarrheal Research, All India Institute of Medical Sciences, New Delhi, India 2 Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, New Delhi, India 3 Department of Gastroenterology and Human Nutrition, All India Institute of Medical Sciences, New Delhi, India 4 Tampere Center for Child Health Research, University of Tampere and Tampere University Hospital, Tampere, Finland 5 Institute for Maternal and Child Health-IRCCS Burlo Garofolo, Trieste, and University of Trieste, Trieste, Italy 6 UCL Institute of Child Health, London, UK Correspondence to Dr Shinjini Bhatnagar, Professor and Head, Pediatric Biology Center, Translational Health Science and Technology Institute, 496, UdyogVihar, Phase III, Gurgaon Haryana 122 016, India; [email protected] PS and NW contributed equally. Received 28 October 2013 Revised 21 May 2014 Accepted 3 June 2014 Published Online First 18 June 2014

ABSTRACT Objective Some of the conventional serological tests for coeliac disease (CD) are expensive, time-consuming and not readily available in developing countries, leading to a delay in diagnosis. Recently, point-of-care tests (POCT) have been manufactured and tested in Europe but have not been validated in our setting. We therefore aimed to study the diagnostic accuracy of the POCT ‘Biocard’ test in diagnosing CD in Indian children. Design Cross-sectional study. Setting Tertiary care centre in north India. Patients Children, aged 2–18 years, with chronic diarrhoea, short stature or refractory anaemia underwent serological testing for CD with antiendomysial antibodies (AEA), antitissue transglutaminase (tTG) antibodies and Biocard test followed by duodenal biopsy irrespective of serological results. CD was diagnosed with positive AEA and duodenal biopsy showing >grade 2 changes using modified Marsh criteria. Those who were both AEA negative and had normal histology were considered CD negative. Results Of 319 children who underwent the serological testing, 170 agreed for biopsy. Of these, 110 were diagnosed with CD and 30 were found to be CD negative. Remaining 30 had discordant AEA and histology results and were not included in analysis. Biocard test agreed with 92/110 positive and 27/30 negative diagnoses based on reference tests (83.6% sensitivity and 90% specificity). tTG was found to be 93.8% sensitive and 96.4% specific. Conclusions We successfully validated the POCT for CD in our setting. It could be used to increase case detection rates in developing countries with a large undiagnosed CD burden.

INTRODUCTION

To cite: Singh P, Wadhwa N, Chaturvedi MK, et al. Arch Dis Child 2014;99:1004–1008. 1004

Coeliac disease (CD) is characterised by small intestinal mucosal damage and circulating autoantibodies against tissue transglutaminase (tTG).1–3 Although the diagnosis of CD requires duodenal biopsy, serological tests such as tTG antibody and antiendomysial antibody (AEA) tests are frequently used as an aid in diagnosis and gluten free diet (GFD) monitoring.2 3 The diagnosis of CD in India and other developing countries poses a challenging task, even though it has come to be recognised as an important disease entity during the past decade.4–8 Villous atrophy in developing countries is frequently attributed to chronic infestation/infection, malnutrition with bacterial overgrowth or tropical sprue, and

What is already known on this topic? ▸ Some of the conventional serological tests for coeliac disease are expensive, time-consuming and not readily available in developing countries, leading to a delay in diagnosis. ▸ Recently, point-of-care tests for diagnosis of coeliac disease have been manufactured and tested in Europe. ▸ There is need to validate these tests in different ethnic populations and our clinical settings to be able to use them in developing countries.

What this study adds? ▸ This is the first study successfully validating the point-of-care ‘Biocard test’ in a developing country setting. ▸ This rapid test would be very useful in developing countries or in remote areas, where there are no centralised laboratories or sample storing facilities. ▸ This would in turn help increase the case detection rates of coeliac disease in developing countries that still have a large undiagnosed burden.

therefore, many cases of CD are missed or misdiagnosed.9 Furthermore, facilities for immunological assays and intestinal biopsies are limited. In addition, it is apparent that many cases of CD can present with atypical symptoms like anaemia, short stature, etc.10–12 Knowledge about atypical symptoms in CD remains poor.10 The initial serological testing in CD diagnosis requires a well-equipped laboratory with suitably experienced staff, giving a relatively high cost of diagnosis and a long turnaround time. Recently, point-of-care tests (POCT) have been produced in Europe that can potentially reduce costs, hasten management and simplify the presumptive diagnosis.13–18 There is a need to develop and validate such POCT in different ethnic populations and clinical settings. Our clinical setting is quite different from developed countries where these tests have been developed and validated till now because a signification proportion of chronic diarrhoea burden in our setting could still be attributed to other causes mentioned above.9 In

Singh P, et al. Arch Dis Child 2014;99:1004–1008. doi:10.1136/archdischild-2013-305567

Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Original article this regard, a commercially available POCT for CD (Biocard test) was considered as the ‘index test’, and the aim of the study was to determine its diagnostic accuracy in identifying CD in Indian children attending a tertiary hospital in north India. It was important to conduct this validation in a setting where the reference tests were feasible and could be done reliably.

METHODS Site and participants The study was conducted in the paediatric outpatient clinic of the All India Institute of Medical Sciences, New Delhi, over a period of 3 years (2009–2012). Children, aged 2–18 years, fulfilling any one of the following inclusion criteria were screened further for CD; (i) chronic or recurrent diarrhoea ( persisting for more than 8 weeks), (ii) isolated short stature (height for age grade 2 changes using modified Marsh criteria and (ii) positive AEA. In children with DM, the results of index test were compared against AEA as a reference test as none of the patients with negative serology gave consent for a biopsy.

Histopathology Biopsy was attempted in all the enrolled cases where consent could be obtained irrespective of serology status. The pathologist independently assessed the biopsies without knowledge of the clinical or serological status of the patient and classified them according to modified Marsh criteria, which had been validated at our centre previously.9 19

antibodies are present in the sera, they form complexes with the self-liberated tTG (obtained by haemolysing an anticoagulated whole blood sample or blood drops from a finger prick) and the colloidal gold labelled mouse IgG antihuman IgA antibodies. The complexes bind to the test line and are made visible due to the localised concentration of colloidal gold. The control reaction occurs between the colloidal gold-labelled mouse IgG antihuman IgA antibodies that pass the test line and react with antimouse IgG antibodies on the control line. This is situated beyond the test line, and thus the appearance of a coloured control line indicates that both the sample and reagents have moved over the test line. Absence or presence of faint control line indicates IgA deficiency. One drop of fresh blood was applied on the test card. The test was performed independently by two research fellows. In case of any discrepancy, the test was repeated by the chief technical assistant within 10–15 min of the initial test and this interpretation was considered final. The readers of the index and AEA tests were blind to clinical, serological or histological data. A positive result was read as soon as the test and the control lines were clearly visible and not beyond 10–15 min. The test was positive if a red control line appeared in the control field and a light to dark line formed in the test field. It was negative if a red control line appeared in the control field and not in the test field. However, the index test did not include positive or negative controls.

Testing of sera for tTG As an additional objective measure, sera collected for the reference test was further tested for anti-tTG antibodies (Euroimmun, Germany) as per the standard methods prescribed by the manufacturers.

Statistical methods We considered the data from a previous study using tTG antibody as the acceptable levels of sensitivity and specificity for calculating the sample size.20 The sensitivity and specificity of tTG assay reported in the study by Poddar et al were 94% and 91.5%, respectively. Using sensitivity of 94% as standard and considering the acceptable differences from this standard to be 5%, the number of CD cases needed in terms of sensitivity was 87. Based on their study, we calculated the number of patients needed to be enrolled to achieve the same was 182. Similarly, the number of cases of true negatives needed (considering the acceptable difference of 5%) in terms of specificity was 126, and the number of patients needed to be enrolled to achieve the same was 242. With this calculation, we needed to enrol at least 242 children for the study to be able to have adequate sample size to comment on diagnostic accuracy of the test. Sensitivity and specificity of index test were measured against the reference test described above.

AEA assay AEA was tested by an indirect immunofluorescence assay using umbilical cord as substrate. The characteristic fluorescent honeycombing around the smooth muscle bundles was taken as a positive test.9 19

Index test Biocard (Ani Biotech Oy, Vantaa, Finland) test uses a lateral flow immunochromatographic strip system. When tTG-specific IgA Singh P, et al. Arch Dis Child 2014;99:1004–1008. doi:10.1136/archdischild-2013-305567

Table 1

Clinical characteristics of patients enrolled in the study

Clinical symptoms

Number of patients

Chronic diarrhoea Abdominal pain Abdominal distention Isolated short stature Failure to thrive Refractory anaemia Constipation

194 91 88 84 76 56 16

(60.8%) (28.5%) (27.6%) (26.3%) (23.8%) (17.6%) (5.01%)

1005

Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Original article RESULTS Participants In total, 319 children fulfilling the study criteria were enrolled in the study. The mean age at presentation was 85.80 months (SD 48.86 months). Of these, 179 children (56.1%) presented with chronic or recurrent diarrhoea, 84 (26.3%) with short stature and 56 (17.6%) with refractory anaemia as the main symptom. The most common symptom was chronic diarrhoea, which was present in 194 (60.8%) children. Table 1 lists the clinical manifestations of these 319 children. Index and AEA were performed in all of the 319 children. Intestinal biopsy was performed in 170 (53.3%) children as consent for biopsy was withdrawn in 145 children and 4 children were placed on a GFD by outside hospital before a biopsy could be performed at our centre. The mean age of these children who withdrew consent was 86.49 months (SD 49.6 months). Of these 145 children, 31 presented with short stature (21.4%), 16 with refractory anaemia (11%) and remaining with chronic diarrhoea. In the primary analysis, the index test was evaluated in only 170 subjects in whom the reference tests (duodenal histology and AEA) were also done. Of these 170 children, 110 were diagnosed to have CD based on the reference test. There were 30 children who had negative AEA and normal histology; they were diagnosed not to have CD (CD negative). Of the remaining 30, duodenal biopsy was inconclusive in 9 children and 21 children had only one positive reference criterion, either AEA (n=6) or histology with marsh grade 2 as gold standard without considering the serology or response to GFD status.14 The histological evidence of villous atrophy is, however, not a finding specific for CD.9 21 Moreover, one of the studies has used anti-tTG serological test as gold standard for calculating diagnostic accuracy of the index POCT.13 The sensitivity and specificity reported for the Biocard test are higher in other studies. The Biocard test has been found to be 96– 96.7% sensitive and 93.5–100% specific in other studies.16 18 This difference might be due to difference in the gold standard used. While we have used a positive serology plus histology suggestive of CD as positive reference test, others have used the histology and response to GFD as gold standard.16 18 Similarly, the difference in the negative controls used could also have led to the

Table 2 Comparison of our index (Biocard) test with the reference test (positive antiendomysial antibody along with histology showing >2 grade changes as per modified Marsh criteria)

Table 3 Comparison of tissue transglutaminase test (tTG) test with the reference test (positive antiendomysial antibody along with histology showing >2 grade changes as per modified Marsh criteria)

Test results The index test agreed with 92/110 positive and 27/30 negative diagnoses based on the reference tests. The sensitivity of the index test was 83.6% (95% CI 76.7% to 90.5%) and the specificity was 90% (95% CI 79.3% to 100%) (table 2). The positive predictive value (PPV) and the negative predictive value (NPV) were 96.8% and 60%, respectively. When compared with AEA results alone, index test results were concordant with AEA results in 274 of 319 cases (85.9%).

tTG results

Index test positive Index test negative

1006

Reference test positive

Reference test negative

92 18

3 27

tTG test positive tTG test negative

Reference test positive

Reference test negative

90 6

1 27

Singh P, et al. Arch Dis Child 2014;99:1004–1008. doi:10.1136/archdischild-2013-305567

Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Original article Table 4 Performance of the Biocard test in our study as compared with other reports Name of test

Principle

Biocard test (Ani Biotech Oy, Vantaa, Finland) Nunc-Immunostick (Nunc A/S, Roskilde, Denmark) Nunc-Immunostick (Nunc A/S, Roskilde, Denmark) Biocard (AniBiotech Oy, Vantaa, Finland) Biocard

Self-tTG-based (whole blood) Self-tTG-based rapid test

Stick CD Biocard Simtomax (Augurix SA, Monthey, Switzerland)

Self-tTG-based point-of-care test Self-tTG-based (whole blood) Self-tTG-based (whole blood) Serum-based Self-tTG-based (whole blood) Self-tTG-based (whole blood)

Sensitivity (%)

Specificity (%)

Turnaround time (min)

83.6

90

10–15

97

96.9

30

82

100

35

96.7

93.5

5





5–10 min

Done as an office procedure

17

100 90.2*

94.9 100

5 5

Study done as comparison between the two tests

18

93.1

95

20

Comments

Reference Our study 14

In adult untreated coeliac patients

15

16

18

13

*96% if IgA-deficient subjects are excluded. CD, coeliac disease; tTG, tissue transglutaminase.

variation in the results. In addition, as described above, both these studies have also not commented on sample size calculation.16 18 Our study showed the Biocard test to be 83.6% sensitive (95% CI 76.7% to 90.5%) and 90% specific (95% CI 79.3% to 100%). In the worst case, the test could give a false negative result in about 23% and false positive in about 21%. A false positive test result should not be considered a problem as the diagnosis of CD could be ruled out later with a normal biopsy. The NPV of index test is low (60%), and thus if the text is negative and index of suspicion is high, additional testing should be carried out. This test was developed to detect IgA-class tTG antibodies, so it is not suitable for the determination of CD autoantibodies in patients with IgA deficiency.22 We found five IgA-deficient subjects, and in such cases, IgG class antibodies should be measured with the conventional serum tTG and AEA assay.23 There are a few limitations of our study. It was not adequately powered to comment on specificity. This was because a large number of patients who were negative for all three serological tests withdrew consent for biopsies. This highlights a very important issue of finding true negatives while validating any such test. Our study included only children with chronic diarrhoea, short stature and refractory anaemia that could have resulted in the high diagnostic yield. In practice, however, this test could even be used for ambiguous indications (eg, unexplained abdominal distention) and high specificity of test would be important to prevent unnecessary biopsies. This blood-based Biocard test kit contains all the equipment needed, and the results can be read visually within a few minutes. For these reasons, and on the basis of our findings, this test should be useful in developing countries or in remote areas, where there are no centralised laboratories or sample storing facilities. Indeed, our results indicate that this rapid test can greatly help the physician to make a preliminary diagnosis of CD among children suffering from chronic diarrhoea in situations where diarrheal disease and malnutrition are common. In such circumstances, making the diagnosis of CD and instigating a GFD would make a significant contribution to reduce morbidity and mortality, particularly when the diagnosis of CD is not yet considered.24 Contributors PS and NW conceptualised the study, collected data and analysed the data, wrote initial manuscript and revised it. MKC and VB collected data, critically reviewed the manuscript and revised it. SS provided technical support for all

the laboratory tests including standardisation and supervising the tests. NT, GKM, MM and TN critically reviewed the manuscript, gave expert inputs and revised it. AP and SB conceptualised the study, supervised the data collection, data analysis and critically reviewed the manuscript. Competing interests AniBiotech Oy, Ltd, Vantaa, Finland, from whom the authors bought the point-of-care test kits used in the present study, did not take part in designing or executing the study, was not involved in reading and interpreting the results or in writing the manuscript. MM is one inventor of the patent ’Methods and Means for Detecting Gluten-Induced Diseases’, US States Patent Number 7,361,480—USA, Patent Granted 22.4.2008; European Patent No. 1390753, European Patent Office 22.10.2008. Finn Medi Oy Ltd, Tampere, Finland, owned by the University of Tampere and Tampere University Hospital, has commercialised the innovation and licensed it to AniBiotech, which used it to develop the Biocard Celiac Test. MM is not an employee of Anibiotech, nor a stock owner. FinnMedi and AniBiotech were not involved in the study. Ethics approval AIIMS Institutional Ethics Committee. Provenance and peer review Not commissioned; externally peer reviewed.

REFERENCES 1 2

3 4 5

6 7 8 9

10 11 12

13

Singh P, et al. Arch Dis Child 2014;99:1004–1008. doi:10.1136/archdischild-2013-305567

Green PHR, Cellier C. Celiac disease. N Engl J Med 2007;357:1731–43. Husby S, Koletzko S, Korponay-Szabó IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr 2012;54:136–60. Rubio-Tapia A, Hill ID, Kelly CP, et al. ACG clinical guidelines: diagnosis and management of celiac disease. Am J Gastroenterol 2013;108:656–76. Sood A, Midha V, Sood N, et al. Prevalence of celiac disease among school children in Punjab, North India. J Gastroenterol Hepatol 2006;21:1622–5. Makharia GK, Verma AK, Amarchand R, et al. Prevalence of celiac disease in the northern part of India: a community based study. J Gastroenterol Hepatol 2011;26:894–900. Kochhar R, Sachdev S, Kochhar R, et al. Prevalence of coeliac disease in healthy blood donors: a study from north India. Dig Liver Dis 2012;44:530–2. Dehghani SM, Haghighat M, Mobayen A, et al. Prevalence of celiac disease in healthy Iranian school children. Ann Saudi Med 2013;33:159–61. Wu J, Xia B, von Blomberg BME, et al. Coeliac disease: emerging in China? Gut 2010;59:418–9. Bhatnagar S, Gupta SD, Mathur M, et al. Celiac disease with mild to moderate histologic changes is a common cause of chronic diarrhea in Indian children. J Pediatr Gastroenterol Nutr 2005;41:204–9. Sharma M, Singh P, Agnihotri A, et al. Celiac disease: a disease with varied manifestations in adults and adolescents. J Dig Dis 2013;14:518–25. Bhattacharya M, Kapoor S, Dubey AP. Celiac disease presentation in a tertiary referral centre in India: current scenario. Indian J Gastroenterol 2013;32:98–102. Bhat AS, Chaturvedi MK, Saini S, et al. Prevalence of celiac disease in Indian children with Down syndrome and its clinical and laboratory predictors. Indian J Pediatr 2013;80:114–7. Bienvenu F, Besson Duvanel C, Seignovert C, et al. Evaluation of a point-of-care test based on deamidated gliadin peptides for celiac disease screening in a large pediatric population. Eur J Gastroenterol Hepatol 2012;24:1418–23.

1007

Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Original article 14 15

16

17

18

1008

Korponay-Szabó IR, Raivio T, Laurila K, et al. Coeliac disease case finding and diet monitoring by point-of-care testing. Aliment Pharmacol Ther 2005;22:729–37. Raivio T, Korponay-Szabó I, Collin P, et al. Performance of a new rapid whole blood coeliac test in adult patients with low prevalence of endomysial antibodies. Dig Liver Dis 2007;39:1057–63. Raivio T, Kaukinen K, Nemes E, et al. Self transglutaminase-based rapid coeliac disease antibody detection by a lateral flow method. Aliment Pharmacol Ther 2006; 24:147–54. Kotze LM, Brambila Rodrigues AP, Kotze LR, et al. A Brazilian experience of the self transglutaminase-based test for celiac disease case finding and diet monitoring. World J Gastroenterol 2009;15:4423–8. Nemec G, Ventura A, Stefano M, et al. Looking for celiac disease: diagnostic accuracy of two rapid commercial assays. Am J Gastroenterol 2006;101: 1597–600.

19

20 21 22 23

24

Singh P, Chaturvedi MK, Rangan P, et al. Patients of celiac disease with mild villous atrophy are clinically similar to those with moderate to severe atrophy. Indian J Gastroenterol 2013;32:413–4. Poddar U, Thapa BR, Nain CK, et al. Celiac disease in India: are they true cases of celiac disease? J Pediatr Gastroenterol Nutr 2002;35:508–12. Yadav P, Das P, Mirdha BR, et al. Current spectrum of malabsorption syndrome in adults in India. Indian J Gastroenterol 2011;30:22–8. Cataldo F, Marino V, Bottaro G, et al. Celiac disease and selective immunoglobulin A deficiency. J Pediatr 1997;131:306–8. Korponay-Szabó IR, Dahlbom I, Laurila K, et al. Elevation of IgG antibodies against tissue transglutaminase as a diagnostic tool for coeliac disease in selective IgA deficiency. Gut 2003;52:1567–71. Byass P, Kahn K, Ivarsson A. The global burden of childhood coeliac disease: a neglected component of diarrhoeal mortality? Plos One 2011;6:e22774.

Singh P, et al. Arch Dis Child 2014;99:1004–1008. doi:10.1136/archdischild-2013-305567

Downloaded from http://adc.bmj.com/ on December 9, 2014 - Published by group.bmj.com

Validation of point-of-care testing for coeliac disease in children in a tertiary hospital in north India Prashant Singh, Nitya Wadhwa, Mona K Chaturvedi, Vidyut Bhatia, Savita Saini, Nikhil Tandon, Govind K Makharia, Markku Maki, Tarcisio Not, Alan Phillips and Shinjini Bhatnagar Arch Dis Child 2014 99: 1004-1008 originally published online June 18, 2014

doi: 10.1136/archdischild-2013-305567 Updated information and services can be found at: http://adc.bmj.com/content/99/11/1004

These include:

References Email alerting service

Topic Collections

This article cites 24 articles, 2 of which you can access for free at: http://adc.bmj.com/content/99/11/1004#BIBL Receive free email alerts when new articles cite this article. Sign up in the box at the top right corner of the online article.

Articles on similar topics can be found in the following collections Clinical diagnostic tests (980) Diarrhoea (157) Pathology (198) Radiology (846) Surgery (246) Surgical diagnostic tests (234) Child health (3334) Metabolic disorders (667) Epidemiologic studies (1545)

Notes

To request permissions go to: http://group.bmj.com/group/rights-licensing/permissions To order reprints go to: http://journals.bmj.com/cgi/reprintform To subscribe to BMJ go to: http://group.bmj.com/subscribe/

Validation of point-of-care testing for coeliac disease in children in a tertiary hospital in north India.

Some of the conventional serological tests for coeliac disease (CD) are expensive, time-consuming and not readily available in developing countries, l...
314KB Sizes 4 Downloads 4 Views