|
|
||||||||
1 First and
2 Second Divisions of Internal Medicine and
3 Surgery Department, University Hospital of Palermo, 90127 Palermo, Italy.
aAddress correspondence to this author at: via Coffaro 25, 90124 Palermo, Italy. Fax 39-091-6552936; e-mail liwcar{at}tin.it.
| Abstract |
|---|
|
|
|---|
Methods: We studied 207 consecutive patients (99 men, 108 women; age range, 1784 years) who underwent intestinal biopsy for suspected CD. Patients presented with one or more of the following: weight loss, anemia, chronic diarrhea, abdominal pain, dyspepsia, alternating bowel habits, constipation, pain in the joints, and dermatitis. At entry to the study, an intestinal biopsy was performed and a serum sample was taken for IgA EmAs, anti-gp-tTG, and anti-h-tTG.
Results: Intestinal histology showed that 24 patients had partial or total villous atrophy; in these patients the diagnosis of CD was confirmed by follow-up. The remaining 183 patients had villous/crypt ratios that were within our laboratorys reference values and were considered controls. Serum EmAs, anti-gp-tTG, and anti-h-tTG were positive in all 24 CD patients; in the control group, none were positive for serum EmAs, but 15 of 183 (8.2%) were positive for anti-gp-tTG, and 6 of 183 (3.3%) were positive for anti-h-tTG. Sensitivity was 100% for all assays, whereas specificity was 100% for the EmA, 92% for the anti-gp-tTG, and 97% for the anti-h-tTG assay. The negative predictive value was 100% for all assays; the positive predictive value was 100% for the EmA, 80% [95% confidence interval (CI), 6595%] for the anti-h-tTG (P = 0.03 vs EmA) and 60% (95% CI, 4476%) for the anti-gp-tTG assay (P = 0.0002 vs EmA). Areas (95% CIs) under the ROC curves were 0.987 (0.971.0) for anti-h-tTG and 0.965 (0.940.99) for anti-gp-tTG. Most of the patients testing false positive for anti-tTG had Crohn disease or chronic liver disease.
Conclusions: Although both anti-tTG ELISAs showed optimum sensitivity, their lack of specificity yielded positive predictive values significantly lower than those for the EmA assay.
| Introduction |
|---|
|
|
|---|
The aim of the present prospective study was to compare the diagnostic accuracy of the EmA assay and two commercially available anti-tTG ELISAs, one based on gp-tTG and the other on h-tTG as antigen, in consecutive patients prospectively investigated for suspected CD, all of whom underwent intestinal biopsies.
| Patients and Methods |
|---|
|
|
|---|
At entry in the study, on the same day as the intestinal biopsy sampling, a serum sample was taken from each patient for IgA anti-tTG and anti-EmA assays. Between January and December 1999, only the gp-tTG ELISA was performed immediately, and all the sera were stored at -80 °C. After December 1999, the new anti-tTG ELISA, based on h-tTG as the antigen, was commercially available, and this assay was performed on the stored sera of the patients recruited from January to December 1999. From January to April 2000, both the gp-tTG and the h-tTG IgA ELISA were performed immediately on entry to the study. Total immunoglobulin values (IgA, IgM, and IgG) were also measured by ELISA.
The diagnosis of CD was based on evidence of clinical symptoms and partial or total intestinal villous atrophy on a gluten-containing diet, disappearance of the symptoms and normalization of the intestinal histology on a gluten-free diet, and reappearance of the symptoms on gluten challenge.
According to the individual clinical presentation and laboratory data, the diagnostic work-up of the patients may also have included abdominal ultrasonography and/or computed tomography, colonoscopy, small intestine barium examination, liver histology, a H2 breath test, and duodenal fluid microbiological evaluation.
The protocol was approved by the Ethics Committee of the University Hospital of Palermo, and informed consent was obtained from the patients involved in the study.
SERUM ANTI-tTG IgA ELISA USING gp-tTG AS ANTIGEN
The anti-gp-tTG assay was performed with a commercial ELISA (Eu.tTG IgA; Eurospital) in accordance with the method described by Troncone et al. (5), modified according to Sulkanen et al. (3). The antigen used in the ELISA was gp-tTG (cat. no. T5398; Sigma). Values were expressed as a percentage of the values obtained for positive reference sera [obtained from untreated celiac patients diagnosed according to the criteria of the European Society of Paediatric Gastroenterology and Nutrition (11) and which tested positive for EmAs in all cases]. Anti-tTG values greater than the 95th percentile of a control group including >100 healthy controls and negative for serum EmAs were considered positive (7% of the reference serum). The intraassay CV was 8.7% (n = 15), and the interassay CV was 10% (n = 15).
SERUM ANTI-tTG IgA ELISA USING RECOMBINANT h-tTG AS ANTIGEN
The anti-h-tTG ELISA was performed in our laboratory on serum samples from 60 patients immediately after blood collection and on serum samples, from 147 patients, that had been kept frozen at -80 °C. A control test performed in our laboratory on 30 serum samples in which anti-tTG antibodies were first assayed on fresh serum and then 8 months later after storage at -80 °C showed that storage at -80 °C did not significantly alter the results obtained: the mean value (± SD) was 5.23% ± 0.71% for fresh samples and 4.92% ± 0.75% for frozen samples. The interassay CV was 7.9%.
Serum IgA recombinant anti-h-tTG antibody concentrations were determined using a commercially available assay (Eu.tTG IgA umana; Eurospital). The purity of the recombinant protein was assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Recombinant h-tTG antigen diluted in phosphate-buffered saline was used to coat the wells; serum samples diluted 1:26 (20 µL of serum plus 500 mL of phosphate-buffered saline) in phosphate-buffered saline containing 1 mL/L Tween 20 were incubated for 1 h at room temperature. The plates were washed three times and subsequently incubated for 1 h at room temperature with horseradish peroxidase-labeled sheep anti-human IgA. The excess conjugate was removed by washing, and a chromogenic substrate was added. The absorbance was read in a microplate reader at 450 nm. Results were expressed as a percentage of the positive control serum. Normal values were taken as <7%, which represented a value > 2 SD above the mean of 850 healthy individuals. The intraassay CV for the IgA h-tTG autoantibody ELISA was 2.2% (n = 30), and the interassay CV was 5.8% (n = 30).
SERUM IgA EmA ASSAY
IgA-class EmA values were determined with a commercially available indirect immunofluorescence method on monkey esophagus (Anti-endomisio; Eurospital Pharma), as described previously (12)(13).
intestinal biopsy and histology
Biopsy specimens were taken during gastroduodenoscopy at the second duodenal portion. In all cases, at least two biopsy specimens were obtained. In accordance with our previous studies (12)(13)(14), specimens adequate in size were immediately oriented with the aid of a stereomicroscope and subsequently embedded in paraffin. The slides were stained with hematoxylin and eosin and graded by conventional histology as (a) normal, (b) partial villous atrophy, or (c) subtotal/total villous atrophy. In addition, the villous/crypt ratio was evaluated (reference values in our laboratory
2.5 for children,
3 for adults), and the number of intraepithelial lymphocytes (IELs) per 100 villous epithelial cells was assessed as described by Ferguson and Murray (15); the upper limit of the reference values in our laboratory is 30 IELs/100 epithelial cells.
In all cases, histologic analysis was performed by an examiner unaware of the clinical condition of the patients and of the laboratory test results.
statistical analysis
We followed standard criteria for methodologic studies on the diagnostic accuracy of tests (16). The sensitivities and specificities of the methods examined and their positive and negative predictive values were calculated by standard statistical methods (17). The Fisher exact test was used to compare the different positive and negative predictive values of the three assays. In addition, ROC curves were plotted to show the discriminative ability of the tests. The model plots sensitivity (the proportion of CD patients attributed by the test to the CD group) vs 1 - specificity (the proportion of control patients attributed by the tests to the CD group). The areas under the ROC curves and their 95% confidence intervals were also calculated using the nonparametric method described by Hanley and McNeil (18), developed as a statistical program.
| Results |
|---|
|
|
|---|
3 and, consequently, were considered as controls to evaluate the specificity of the diagnostic tests for CD. Table 1
|
None of the patients included in the study showed serum IgA deficiency. Serum EmAs and anti-tTG antibodies, assayed with both the gp-tTG and the h-tTG ELISA, were positive in all 24 CD patients (Table 2
). In the control group (non-CD patients), none were positive for serum EmAs, but 15 of 183 (8.2%) were positive for anti-gp-tTG and 6 of 183 (3.3%) were positive for anti-h-tTG (Table 2
). In the 15 controls positive for anti-gp-tTG, values ranged between 9% and 24%; in the controls positive by the anti-h-tTG ELISA, values ranged between 8.5% and 28%. These values were in the same range as the values observed in the 24 CD patients with partial or total mucosa atrophy (range of values, 831% for gp-tTG and 934% for h-tTG). After performing a cumulative analysis of the data from patients and controls, we plotted the ROC curves for anti-h-tTG and anti-gp-tTG; the areas under curves were 0.987 [95% confidence interval (CI), 0.971.0] for anti-h-tTG and 0.965 (95% CI, 0.940.99) for anti-gp-tTG.
|
The patients with false-positive anti-tTG ELISA results based on guinea pig antigen included seven with Crohn disease (diagnosis was confirmed by histologic examination of ileal/colon biopsies and positive barium x-ray of ileum examination and/or colonoscopy), three with multiple food intolerance (intolerance to cows milk and its derivatives, fish, tomato, chocolate, and/or orange, diagnosed by means of an elimination diet that documented the disappearance of the symptoms and a subsequent blind challenge that demonstrated the reappearance of the symptoms), three with chronic hepatitis attributable to a hepatitis virus C infection (with moderate/severe liver histology damage), one patient with biopsy-confirmed collagenous colitis, and one with psoriasis. False-positive results with human tTG antigen all were in the set of those positive in the anti-gp-tTG ELISA: two patients with ileal Crohn disease, two with chronic hepatitis, one with psoriasis, and one with collagenous colitis. Three false-positive results were obtained on fresh sera and three on samples stored at -80 °C. Further examinations performed on these six patients showed that they did not carry the DQ2 or DQ8 HLA haplotypes, which characterize CD patients, and were negative for serum anti-gliadin antibodies.
Because all three assays had a sensitivity of 100%, without any false-negative results, they all had a negative predictive value of 100% (Table 3
). In our series (CD prevalence, 11.6%), the positive predictive value was 100% only for the EmA assay. The two anti-tTG assays that we evaluated had lower positive predictive values. The anti-tTG assay based on guinea pig antigen had a positive predictive value of 60%, significantly lower than that of the EmA assay (P = 0.0002, Fisher test), whereas the anti-tTG assay based on human recombinant antigen had a positive predictive value of 80%, also significantly lower than that of EmA assay (P = 0.03).
|
| Discussion |
|---|
|
|
|---|
In the present study, we considered a wide group of individuals referred to three different centers and followed as outpatients for clinical symptoms and/or signs compatible with a diagnosis of CD. We compared the anti-tTG ELISAs with the anti-EmA assay, which can be considered the "gold standard" in the serology of CD (25). Although the three assays identified all the CD patients, this cannot be considered a generally reproducible result because we had no patients with a serum IgA deficiency, a condition more common in CD than in the general population, which limits the performance of the serologic tests for CD based on IgA antibody determinations.
The positive predictive value of the anti-gp-tTG assay (60%) is too low to warrant subjecting a patient to intestinal biopsy for suspected CD. This relatively low positive predictive value of anti-gp-tTG, however, cannot be considered completely unexpected. In fact, we obtained false-positive results for seven patients with Crohn disease, a condition in which a high prevalence of anti-tTG antibodies has been reported previously (26), and in four patients with intestinal disease characterized by moderate/severe mucosal inflammation (three patients with multiple food intolerance and one patient with biopsy-confirmed collagenous colitis). Three other patients who were false positive for anti-gp-tTG antibodies had chronic hepatitis attributable to a hepatitis virus C infection, a liver disease in which we have previously demonstrated a 14% frequency of false-positive results (27).
All six false positives in the anti-h-tTG assay were also positive in the anti-gp-tTG assay. Although we had no doubts about excluding a CD diagnosis in the two patients with Crohn disease and in the two with chronic liver disease, both collagenous colitis (28) and psoriasis (29) have been reported to be associated with CD. However, the intestinal biopsies of these two patients, performed at the second portion of the duodenum, did not show alterations of the villi or crypts (ratio of villi to crypts >3), and serum IgG and IgA anti-gliadin antibodies were negative. Moreover, they did not have the classical HLA DQ2 or DQ8 alleles, which are typical of CD patients. The low positive predictive values of the anti-tTG assays will be even lower at lower prevalences of CD, such as those commonly reported in studies of consecutive patients with dyspepsia (30)(31).
The false-positive results observed in our study could arise from impurities in the gp-tTG for the ELISA system based on this antigen, as we have previously suggested in other studies (9)(27). For the false-positive results obtained with the specific recombinant h-tTG in the ELISA system, the hypothesis of protein impurities in the tTG preparation can also be considered, but we do not exclude the possibility that high serum concentrations of anti-tTG antibodies were really present in some "controls". Immunoblot or immunoprecipitation studies with false-positive sera are needed to determine whether the reactive band is really tTG.
In conclusion, the EmA assay has a higher positive predictive value than the anti-tTG assays in CD diagnosis. On this basis, the simpler anti-h-tTG ELISA could be suggested as a first-step examination, with positive results to be confirmed by the EmA assay. We suggest that discordant anti-tTG/EmA results merit further evaluation, e.g., by means of HLA DQ determination.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
D. Basso, G. Guariso, P. Fogar, A. Meneghel, C.-F. Zambon, F. Navaglia, E. Greco, S. Schiavon, M. Rugge, and M. Plebani Antibodies against Synthetic Deamidated Gliadin Peptides for Celiac Disease Diagnosis and Follow-Up in Children Clin. Chem., January 1, 2009; 55(1): 150 - 157. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Rashtak and J. A. Murray Tailored Testing for Celiac Disease Ann Intern Med, September 4, 2007; 147(5): 339 - 341. [Full Text] [PDF] |
||||
![]() |
A. Carroccio, L. Di Prima, G. Pirrone, C. Scalici, A. M. Florena, M. Gasparin, G. Tolazzi, A. Gucciardi, C. Sciume, and G. Iacono Anti-Transglutaminase Antibody Assay of the Culture Medium of Intestinal Biopsy Specimens Can Improve the Accuracy of Celiac Disease Diagnosis Clin. Chem., June 1, 2006; 52(6): 1175 - 1180. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Zintzaras and A. E. Germenis Performance of Antibodies against Tissue Transglutaminase for the Diagnosis of Celiac Disease: Meta-Analysis Clin. Vaccine Immunol., February 1, 2006; 13(2): 187 - 192. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. E. Prince Evaluation of the INOVA Diagnostics Enzyme-Linked Immunosorbent Assay Kits for Measuring Serum Immunoglobulin G (IgG) and IgA to Deamidated Gliadin Peptides Clin. Vaccine Immunol., January 1, 2006; 13(1): 150 - 151. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Carroccio, I. Brusca, G. Iacono, L. Di Prima, S. Teresi, G. Pirrone, A. M. Florena, S. M. La Chiusa, and M. R. Averna Anti-Actin Antibodies in Celiac Disease: Correlation with Intestinal Mucosa Damage and Comparison of ELISA with the Immunofluorescence Assay Clin. Chem., May 1, 2005; 51(5): 917 - 920. [Full Text] [PDF] |
||||
![]() |
I. Dahlbom, M. Olsson, N. K. Forooz, A. G. Sjoholm, L. Truedsson, and T. Hansson Immunoglobulin G (IgG) Anti-Tissue Transglutaminase Antibodies Used as Markers for IgA-Deficient Celiac Disease Patients Clin. Vaccine Immunol., February 1, 2005; 12(2): 254 - 258. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Van Meensel, M. Hiele, I. Hoffman, S. Vermeire, P. Rutgeerts, K. Geboes, and X. Bossuyt Diagnostic Accuracy of Ten Second-Generation (Human) Tissue Transglutaminase Antibody Assays in Celiac Disease Clin. Chem., November 1, 2004; 50(11): 2125 - 2135. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Baldas, T. Not, A. Tommasini, F. Ansaldi, S. Demarini, D. Sblattero, R. Marzari, L. Torelli, A. Burlina, C. Tiribelli, et al. Anti-Transglutaminase Antibodies and Age Clin. Chem., October 1, 2004; 50(10): 1856 - 1860. [Full Text] [PDF] |
||||
![]() |
J.-J. Baudon, C. Johanet, Y. B. Absalon, G. Morgant, S. Cabrol, and J.-F. Mougenot Diagnosing Celiac Disease: A Comparison of Human Tissue Transglutaminase Antibodies With Antigliadin and Antiendomysium Antibodies Arch Pediatr Adolesc Med, June 1, 2004; 158(6): 584 - 588. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Lumbreras-Lacarra, J. M. Ramos-Rincon, and I. Hernandez-Aguado Methodology in Diagnostic Laboratory Test Research in Clinical Chemistry and Clinical Chemistry and Laboratory Medicine Clin. Chem., March 1, 2004; 50(3): 530 - 536. [Abstract] [Full Text] [PDF] |
||||
![]() |
A A Fisher, M W Davis, and M M Budge Should we screen adults with osteoporotic fractures for coeliac disease? Gut, January 1, 2004; 53(1): 154 - 155. [Full Text] |
||||
![]() |
A. Picarelli, M. Di Tola, L. Sabbatella, S. Vetrano, M. C. Anania, A. Spadaro, M. L. Sorgi, and E. Taccari Anti-Tissue Transglutaminase Antibodies in Arthritic Patients: A Disease-specific Finding? Clin. Chem., December 1, 2003; 49(12): 2091 - 2094. [Full Text] [PDF] |
||||
![]() |
A. Carroccio, G. Iacono, M. Cottone, L. Di Prima, F. Cartabellotta, F. Cavataio, C. Scalici, G. Montalto, G. Di Fede, G. Rini, et al. Diagnostic Accuracy of Fecal Calprotectin Assay in Distinguishing Organic Causes of Chronic Diarrhea from Irritable Bowel Syndrome: A Prospective Study in Adults and Children Clin. Chem., June 1, 2003; 49(6): 861 - 867. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |