a. The blood tests recently developed for the detection and screening of celiacs and their relatives are a useful aid to diagnosis and monitoring. The tests have a clinical utility in:
The actual tests are done using blood from the patient. The blood cells are removed. The gliadin test is usually an automated machine read test. While this means there is little room for interpreter error, there are no standardized tests, normal ranges, or even standardized methods in use in the US.
The endomysial tests are more dependent on the experience and ability of a pathologist in looking at a pattern of staining produced by the patient's serum on a slice of monkey esophagus or human umbillical cord. While this test is done in a similar way in most labs there are differences in how these are interpreted.
c. How good are these tests?
If all of the tests are positive then they are pretty accurate, GT 95% right. However, there are several reasons and circumstances when they are not so accurate. IgA and IgG are two different varieties of antibodies we have in our immune systems. The IgA gliadin and IgA endomysial tests are the most accurate and also become negative relatively quickly after stopping gluten (3-6 months). The IgG is not as specific (it can be positive in non celiacs). However it is important to do both, as about 4% of celiacs have low enough levels of IgA to make the IgA tests inaccurate.
These tests can only be given to a patient that is consuming gluten. The amount and time require for a gluten challenge varies and is a factor for consideration when scheduling the test.
Negative results do not indicate that the patient will never develop Celiac Disease, only that it is not currently a factor. HLA typing can be used to determine if a person has the genetic factors that are associated with celiac disease; without these genetic factors celiac disease is very unlikely to ever develop. For more information, access the file CELIAC CEL-HLA with the GET command from listserv@LISTSERV.ICORS.ORG.
Needless to say the interpretation of mixed results (i.e. some positive and some negative) is complicated. Also the interpretation and use of these tests in infants may be different, due to lower levels of IgA.
d. The blood samples required for the tests can be taken at your physician's office and shipped to a lab for processing.
e. The following instructions are for mail-in, but limited on-site collection is also available. If a third party is collecting the samples, remember to determine the costs associated with collection and handling before the tests are executed.
Two hospitals and two laboratories are noted for their expertise in performing the serological test for patients around the country. Please note that familiarity with the tests and multiple executions of the tests reduces the potential for misdiagnosis.
Sending blood (send the following items): This is a rather sketchy description of the process, exact procedures should be requested from the processing laboratory. For the test you need the following:
The sensitivities and specificities of the IgA and IgG antigliadin antibody and the IgA antireticulin antibody have been compared with the recently described endomysial antibody directed against the basement membrane of smooth muscle in monkey esophagus or human umbillical cord. Endomysial antibodies were found in all patients with untreated CD and subtotal villous atrophy and in 47% of patients on a non-strict gluten- free diet. One patient on a strict gluten-free diet was positive and had partial villous atrophy while all patients in disease control groups were negative. Results were variable with the antireticulin and antigliadin antibodies. Sensitivity and correlation with subtotal villous atrophy in the untreated patient was 100%. It is concluded that the endomysial antibody is superior to other current antibody tests and should be used in preference for the diagnosis of CD. The blood test is only a screening tool. The biopsy remains the "Gold Standard".
Dr. Vijay Kumar presented background information on serology at the Mt. Sinai Medical Centre, on 9 Nov. 1996: "Predictive Value of Serology Testing in Celiac Disease".
2. Saliva Test
The following section represents questions and comments on the saliva test and its possibility as an antibody detection tool.
Back on 7 Feb 1996 Elizabeth Welch asked:
EW) Has anyone heard of IgA antibody testing done using saliva ("spit EW) test")? ..(rest deleted) And (name removed at the request of the poster) replied: > ....I know about that test--I have had the secretory IgA test from > Diagnos-Techs(it is specific for anti-gliadin antibodies). Dr. > Busher, M.D. of Bellevue, WA regards it as an excellent test. (snip) > > My ex-husband is very good friends with Dr. Ilias, > the research biochemist who developed this test and owns > Diagnos-Techs....I was told was that if my diet was scrupulously > gluten free, my IgA levels would drop, but probably not below the > 40's. When the levels get high, they never drop to normal ranges > again....All people have a titer of anti-gliadin anti-bodies on this > test, because gluten causes a small amount of villi damage in > everybody. (snip) > > I asked hard questions about the reliability of this test, and was > assured that it is highly reliable, with essentially no false > positives (well below 1%). I can't remember the rate of false > negatives....but it is also very low....It is a relatively new test. The Listowners forwarded this information to cel-pro and asked for advise. We received several replies which will be summarized below. In addition, Dr. Ilias provided me with enough test kits to try out my whole family. I can report that despite the above statement " When the levels get high, they never drop to normal ranges again"-- My Celiac son, whose endomysial, reticulin and gliadin antibodies were very high prior to his GF diet, had a single digit SIgA level on Dr. Ilias' saliva test. My whole family was well within the normal range, but his was the lowest! For those of you who want more information about the test, the labs' phone number is (206) 251-0596. The test only costs $30. It was more difficult to get my children to properly deliver the saliva than I had expected. Unlike a blood draw, this requires active cooperation. Dr. Ilias told me that in the 12,000 tests he has done to date he consistently gets about 14% to 15% positive results, which is about 35 times greater than the incidence of celiac disease in Europe (1:250). Thus, the test may be a screen for general allergy or intolerance to gluten, not just formal Celiac.
Before reviewing some of the comments about the test itself, I wanted to forward a comment to this statement in the reply post:
LW> ...All people have a titer of anti-gliadin anti-bodies on this test, LW> because gluten causes a small amount of villi damage in everybody. On this point Don Kasarda replied: DK> I don't know anything about the saliva test, but I am not personally DK> aware of any scientific evidence for the above statement that gluten DK> causes a small amount of damage to villi in everybody. As for the test itself, those who commented on it made it clear that the test has not been validated as being either sensitive or specific enough for diagnostic work in Celiac Disease. It appears to be one of a number of a number of unproven, and not-completely-tested alternatives to the current state of the art endoymsial/reticulin/gliadin panel, which approaches 100% accuracy when done by an experienced lab. >From Erkki Savilahti, M.D., who was one of the authors of a study comparing the saliva test to the current serology tests in DH: ES> Our study (Eur J Oral Sci 1995;103:280-4 Patinen et al.; Salivary ES> and serum IgA antigliadin antibodies in dermatitis herpetiformis) ES> did not find salivary antibodies very useful in dermatitis ES> herpetiformis; in fact IgA antigliadin antibodies in saliva were not ES> increased in any of the 10 untreated patients and there was no ES> change after the introduction of gluten free diet. ES> ES> Altogether the measurement of salivary antibodies is plagued with ES> many drawbacks; it is difficult to stimulate and collect resulting ES> in very variable levels. Serum tests for example to study ES> endomysium antibodies may be done on a small sample taken by skin ES> puncture and is acceptable for children. Moreover, the specificity ES> and sensitivity of these antibodies are superior to determination ES> of gliadin antibodies: in a submitted manuscript we found the ES> sensitivity for umbilical cord antibodies to be 0.94 and specificity ES> 1.00 in a large material of pediatric patients. Dr. Karoly Horvath called Dr. Ilyia to get more information about the saliva test, and send the following email: KH> I talked with Dr. Ilyia about the saliva antibody test. Evidently, KH> he states that this is a very helpful test for people looking for KH> any help for the gastrointestinal problems. Several patients who had KH> increased saliva antibodies reacted well for gluten withdrawal. KH> This sounds great. However, from methodological point of view there KH> are several steps to validate the test: KH> KH> - There is no data about the specificity and sensitivity of this KH> test in celiac patients. If somebody introduces a new test it KH> should be compared with the clinical diagnosis based on other KH> test(s) with high specificity and sensitivity or with a KH> pathologic diagnosis (e.g. intestinal histology in celiac KH> patients or skin biopsy in patients with DH). This saliva test KH> was compared only with the serum antigliadin antibody levels of KH> the same patients. The clinical correlation of the results KH> based on the patients subjective response to elimination diets. KH> Dr. Ilyia states that the saliva test is positive when the serum KH> antibody level is not elevated (returned to the normal level). KH> KH> - They have not compared yet their test results with endomysium KH> antibody titers. They have not compared the results with the KH> histology of celiac patients. KH> KH> Dr. Ilyia mentioned to me that he has talked with Joe Murray, KH> and they are planning to perform studies with the participation KH> of 22 patients with celiac disease. ... KH> KH> - They did not evaluate the antibodies in healthy people, and KH> there is no data on the percentage of normal population, who may KH> have increased levels without celiac disease (normal control KH> group)... KH> KH> - There is no data whether people with other documented KH> gastrointestinal diseases (e.g. Crohn's disease) have salivary KH> antibodies more frequently than the healthy controls (GI control KH> group).... KH> KH> I tried to focus on questions which are addressed when somebody KH> comes out with a new test. All these issues have been clarified for KH> the serum serological tests between 1985 and 1991. That is why we KH> have data about the specificity, sensitivity, positive and negative KH> predictive value of the celiac serum serological tests. For a real KH> judgement of the saliva test it should undergo the same steps of KH> evaluation.
3. Screening Tool
A strip-Anti Gliadin Aantibody test was developed in Europe. It allows a rapid detection of serum IgG and IgA AGA from a single drop of whole blood and a dot immunobinding assay to detect anti-alpha-gliadin-specific antibodies. Additional details are available in the published article "A new, rapid, noninvasive screening test for celiac disease" by Not T, M.D., Ventura A, M.D., Peticarari S, Basile S, Tore G, M.D., Dragovic D, M.D. in The Journal of Pediatrics, Vol 123, Num 3, Sept 93, pg. 425.
This test has not been discussed on the Celiac List to date, presumably it suffers from the same incomplete accuracy problems as the regular gliadin serology tests do, i.e. some false negatives and positives that can be avoided by also running the reticulin and endomysial antibody test at the same time. However, this test appears much easier to administer.
It is critical that the physician and pathologists communicate regarding the suspicion of celiac disease. A new variation of the test has been devised, which may prove to be simpler, safer, and more reliable. This test involves challenging the patient with a rectal dose of gluten rather than an oral dose.
2. Rectal biopsy is a concept that has shown some interesting results in British tests, but is not generally accepted at this time.
A rectal biopsy, far easier to obtain than a jejunal biopsy, is then examined for the diagnosis. The test was assessed in 11 patients thought to have CD and 21 patients with other types of bowel disorder. Six hours after rectal challenge with gluten digest, a rectal biopsy was obtained and the lymphocytes within the epithelial tissue were counted. An increase of 10 percent or more was taken as indicative of CD. Among patients tested, the procedure produced one false positive and one false negative. The result demonstrated that the rectal challenge is effective; it is also more rapid than the traditional oral test. In addition, the test is preferred by some patients, who may refuse an oral test that must be followed by a jejunal biopsy.
It is often necessary to prescribe how much gluten to eat. I Ask patients to eat increasing amounts of gluten, starting from a cracker building up to 4 slices of whole bread a day and to keep taking that so long as they do not get too ill, but should get definite symptoms. When I left it up to the patient, it seems they ate less gluten and more sporadically. It is sometimes hard to wait to persuade the patient to wait long enough for damage to occur. While the standard 4 week challenge will pick up most it will not pick up all patients. In those patients who remain asymptomatic I continue to wait and monitor the antibodies as well as the development of symptoms and then biopsy. I also find it useful to review the original biopsy material and other primary information on which the original diagnosis was based. Sampling problems and interpretation are also issues.
There a few people I will not challenge, People who are so nutritionally deplete that might not tolerate the challenge, people who give a history of anaphylactic response that could be life threatening and possibly those with a history of sever neurologic problems.
Not Medical Advise
Q. How long must gluten be taken for the serological tests to be meaningful?
K (Dr. Kumar). There is no simple answer to this question as the susceptibility of the patient to developing CD is dependent upon several factors. One factor is the amount of gluten intake. Another is the genetic makeup of the individual. However, we feel that several weeks of gluten intake, especially in doses of 2 gm gluten/day, should result in positive serology in patients with CD.
H (Dr. Horvath). The result of serological tests depends on the diet. Generally, three to six months of a gluten-free diet may result in normal antibody levels in a new patient. A strict gluten-free diet for more than three months may result in inconclusive serological tests in patients, who have started a diet without any diagnostic test. In this case a gluten challenge should be introduced for a proper diagnosis.
Each patient has different sensitivity to gluten for reasons that are unclear. The period of gluten challenge and the amount of gluten necessary to provoke serological immune response are individually different.
A 0.3 g/kg body weight/day of single gluten challenge causes immunological changes (cellular immunity) in the intestine (J Pediatr Gastroenterol Nutr 1989; 9:176-180) in patients on a gluten-free diet, however, the serological response is much slower.
Our recommendation is to ingest at least 0.3 g/kg/day of gluten for two months prior to the serological tests. However, if somebody experiences symptoms during the gluten challenge we recommend to perform serological tests earlier.
The protein content of wheat flour is between 7-15% and approximately 90% of the protein content is gluten. That means a slice of bread may have 2-3 g of gluten.
Q. What is the probability of false positive and false negative results from the serological tests?
K. The three serological tests that are used for diagnosing CD are:
Each of these three tests provide a certain degree of reliability for diagnosing CD. Of these, endomysial antibody is the most specific test. The following table is taken from our studies (Lerner, Kumar, Iancu, Immunological diagnosis of childhood coeliac disease: comparison between antigliadin, antireticulin and antiendomysial antibodies).
% Sensitivity % Specificity Predictive Value % Pos. % Neg. EMA 97 98 97 98 ARA 65 100 100 72 AGA IgG 88 92 88 92 IgA 52 94 87 74The following definitions related to sensitivity, specificity, positive and negative predictive values may help.
H. The summary below shows the results of the main serological tests based on several publications including 388 patients with CD, and 771 healthy subjects.
SENSITIVITY- the proportion of subjects with the disease who have a positive test. It indicates how good a test is at identifying the diseased.
IgA AGA: average: 78% range: 46-100% IgG AGA: average: 79% range: 57-94% IgA EMA: average: 97% range: 89-100%
SPECIFICITY- the proportion of subjects without the disease who have a negative test. It indicates how good a test is at identifying the nondiseased.
IgA AGA: average: 92% range: 84-100% IgG AGA: average: 84% range: 52-98% IgA EMA: average: 98.5% range: 97-100%
POSITIVE PREDICTIVE VALUE- the probability that a person with positive results actually has the disease.
IgA AGA: average: 72% range: 45-100% IgG AGA: average: 57% range: 42-76% IgA EMA: average: 92% range: 91-94%
NEGATIVE PREDICTIVE VALUE- the probability that a person with negative results does not have the disease.
IgA AGA: average: 94% range: 89-100% IgG AGA: average: 94% range: 83-99% IgA EMA: average: 100% range: 100%
Q. One case I know of had elevated gliadins (both types) but normal EMA and ARA, plus an inconclusive biopsy. Do you see this often?
K. If the tests are performed using well standardized tests with known positive and negative predictive values then you can make the statement that if the serological tests are negative CD can virtually be ruled out. The problem is that some of these assays, especially the gliadin, can give you false positive results. In our laboratory we rarely see positive AGA results in the absence of EMA and ARA antibodies.
Q. Are there any unique factors to be considered for children? I've heard that the serology has a lower predictive value for children under age two, since IgA may be depressed, or with anyone who has a condition which depresses IgA.
K. Not really. It is not true that the serological methods have lower predictive value in children less than two years of age. In all the studies that we did, there was 100% correlation of the EMA to the disease activity irrespective of the age.
H. There are age dependent changes in several blood parameters during childhood. It is well known that immunoglobulin levels depend on the age of children. E.g. the IgA class immunoglobulins reach the adult level only by 16 years of age, and the blood level of IgA immunoglobulins is only 1/5th of adult value below two years of age. A large study from Europe (Brgin-Wollf et al. Arch Dis Child 1991;66:941-947) showed that the endomysium antibody test is less specific and sensitive in children below two years of age. They found that the sensitivity of the EmA test decreased from 98% to 88% in children younger than 2 years of age. It means that 12% of their patients with celiac disease, who were younger than two years of age, did not have an increase in their endomysium antibody levels.
Q. How important is it for a confirmed celiac to have repeat biopsies or serology when on a gluten free diet?
K. It is important for the serum tests to be negative in patients with CD. These tests provide strong indicators that the gluten free diet followed is effective and is free of gluten. Sometimes drugs or other intakes may be contaminated with gluten that may continue sensitization and the disease process which may be subclinically. We and others believe once the diagnosis of CD is confirmed and the patient is on a gluten free diet, repeat tests once in 3-6 months may be sufficient.
H. If a patient has histologically (endoscopy) and serologically (antibody tests) proved celiac disease, and his/her symptoms disappeared on a gluten-free diet, a repeat biopsy is not necessary. The serological tests are useful tools for estimating the effectiveness of the diet after 3-6 months on a gluten-free diet. The disappearance of antibodies from the blood takes months, if there was not any accidental gluten challenge (dietary mistake).
Q. There are different practices amongst g/i's on repeat biopsies vs. serology, and on gluten challenges. My son's g/i, for example, took the position that since my son's symptoms stopped on a GF diet, and his previously sky-high EMA and ARA went back to normal, that it was unnecessary to do either a repeat biopsy or a gluten challenge. From the celiac list correspondence, I now see that my g/i is rather liberal.
K. I think your son's GI is doing the right thing. That is, if the EMA, ARA are normal (LT1:2.5) and he is on a gluten free diet then there is no need to perform biopsy studies. The previous studies relating the EMA to biopsy studies tend to confirm this impression.
Q. Should my child have general anesthesia or conscious sedation prior to the biopsy?
H. The biopsy is a small piece of tissue, such as from the inside lining of the intestine, that has been removed to look for diseases. The biopsy itself is not painful, because there are no pain-sensitive nerves inside the small intestine. An intestinal biopsy can be done in either of two ways depending on the age of the children and the tradition of the institution. Sometimes a blind biopsy procedure is performed by a biopsy capsule. This is thin flexible tube with a capsule at the tip, which has a hole and a tiny knife inside the capsule. This capsule is introduced into the intestine under fluoroscopy (X-ray) control. Alternatively, with an endoscopy the doctor can see inside the digestive tract without using an x-ray to obtain biopsies. The biopsy specimens are processed and viewed under the microscope to identify or exclude celiac disease. An important basic rule is that the biopsy should be performed safely. For a safe procedure children (and adults) should be sedated. There are two methods of sedation: unconscious (general anesthesia) and conscious sedation. During both kinds of sedation the vital parameters (heart rate, blood pressure, oxygen saturation) of patients are continuously monitored. The method of choice depends on the child.
Conscious sedation is performed with two different intravenous medications. One of them is a sedative medication (e.g. Versed), which causes amnesia in 80-90% of children, and even older children do not recall the procedure. The second medication is a pain-killer type medication (e.g. Fentanyl), which further reduces the discomfort associated with the procedure. In addition, the throat is sprayed with a local anesthetic in older children, which makes the throat numb and prevents retching at the introduction of the endoscope.
During general anesthesia the anesthesiologist uses sleep-gases (e.g. halothan) and intravenous medications and then places a tube into the trachea. Children are completely unconscious. This is a safer way to perform endoscopy, because the patients are fully relaxed and their airway is protected. However, the anesthesia itself has certain complications.
Q. Should I just test endomysial antibodies or also do gliadin/reticulin?
H. Serological tests are performed at the time of diagnosis of celiac disease and they are repeated later to estimate the efficacy of the gluten-free diet.
It is recommended to perform a full serological test-panel in patients with suspected celiac disease. These tests measure antibodies belonging to both the IgA and IgG classes of immunoglobulins. The incidence of selective IgA deficiency is much higher in celiac patients than in the general population. In patients with selective IgA deficiency only the IgG antigliadin antibody may be present, however, this antibody is less specific. It means that the IgG-type antigliadin antibody may be present in otherwise normal individuals.
If somebody had a positive endomysial antibody test at the time of diagnosis he/she may choose to use only this antibody test to monitor the effect of the diet. There are individual differences in the disappearance of serum antibodies.
Q. Is it important to use experienced laboratories for reliable test results?
K. Absolutely yes. For the test to provide meaningful results, it must be validated using a large number of clinical documented subjects. In addition, the two tests, endomysial and reticulin are immunofluorescent tests where the readings are subjective. Experienced laboratory personnel are needed to read such tests.
H. There are several advantages to use a laboratory experienced with the celiac serological tests:
Q. How can I convince my doctor to do these tests, and do them at an experienced lab?
K. Convincing the doctor initially depends upon the patient. However, the laboratory to which the test is sent should be available to answer questions the doctor may have. Our laboratory always encourages such questions.
H. Lot of physicians in the USA did not get appropriate training to recognize the protean manifestations of celiac disease. However, if the classical symptoms are present--chronic diarrhea, weight loss, protuberant abdomen, foul-smelling stools, etc.--it is absolutely indicated to test the patients serum for antigliadin and antiendomysium antibodies.
Professionals participating in this discussion group are educating physicians on an almost daily basis. Generally, it is useful to supply the physician with a review article or a textbook chapter describing the values of serological tests and protean manifestations of celiac disease. If that does not help, you can ask the help of professionals participating in the Cel-Pro list. They have helped several patients by calling physicians and convincing them about the necessity of serological testing.
Q. How often must a negative test be repeated in suspect individuals? This question has two aspects: for an individual with existing symptoms, and for a sibling of a known celiac.
K. If the test is negative and there is a strong suspicion of CD, it must be repeated after several weeks (3-4 weeks), especially after a high gluten intake. We did a study of two cases with DH who were serologically negative. However, a gluten challenge 1g/Kg body wt/day resulted in positive serology; the results became normal on a gluten free diet.
If you are a relative of a CD patient and are on a regular diet and the serology performed by an experienced laboratory is negative then there may not be any need for retesting until and unless clinically justified.
H. There is no rule for it. If a family member with previous negative tests experiences any gastrointestinal symptoms associated with CD, he/she should undergo serological testing as soon as possible. It is well known that up to 15% of the family members of a patient with celiac disease may have the asymptomatic (latent or silent) form of celiac disease, although they have positive serological tests and have the pathological changes in the upper part of the small intestine. It is also evident that there are at least three developmental stages of mucosal lesions (Marsh MN. Gastroenterology 1992;102:330-354) and celiac disease may manifest at each period of life. That is why we recommend a repeat test every 2-3 years in first degree relatives of celiac patients.
Q. Suppose the biopsy or serum tests are inconclusive. What do you do?
K. The biopsy may be inconclusive. Serum, if tested for gliadin, endomysial and reticulin antibodies, should provide unequivocal information. Ours and other studies have provided a strong reliability of the serum tests.
H. The biopsy may be inconclusive in a small percentage of patients with so-called patchy lesions in the duodenum. It means that there are histologically normal looking spots with finger like villi and pathologic spots showing flattened mucosa in the upper half of the duodenum. If CD is suspected, the gastroenterologist should obtain several biopsies from different spots of the whole duodenum. Most of the endoscopists routinely examine only the upper half of the duodenum (duodenal bulb and the descending part). The transverse segment of the duodenum is not viewed routinely. Few endoscopic centers have an enteroscope, which is a longer and more flexible endoscope for examining the entire duodenum and jejunum. The enteroscopy allows you to obtain biopsies even from the jejunum. The histological examination of a single biopsy specimen may increases the risk of false negative diagnosis.
The experience of the pathologist in the interpretation of small intestinal histology is important. In centers specializing in celiac disease the gastroenterologist routinely reviews the histologic slides together with the pathologist.
There is still a possibility of inconclusive results if multiple biopsies are obtained and the histological interpretation is appropriate. All disease has a developmental process. It means that it takes time for the pathological changes to be evident. There are cases when the symptoms suggest CD, however, the histology is not conclusive. This problem occurs in only a few cases. A repeated biopsy may be necessary after a period of higher gluten intake. However, if the antiendomysium antibody test is positive and the histology is not conclusive a gluten-free diet is recommended.
The serology test may be inconclusive if:
-patients with celiac disease have a wide variety of manifestations from classical symptoms to the asymptomatic cases.
-the individual sensitivity to gluten is very variable from unclear reasons. The period of gluten challenge resulting in mucosal damage and the amount of gluten necessary to provoke serological immune response/histological changes are individually different.
There is no simple explanation for these observations:
- One possibility that patients have the combination of different "celiac genes". Although we have significant progress in our knowledge in the molecular genetics of CD, we are still far from a full understanding. We know the higher frequency of certain genes but we still do not have a full knowledge.
- The second possible explanation is a difference in the adaptation of the small intestine. The adaptation means that when a smaller or larger part of intestine is removed, chronically damaged or non-functioning the remaining segments are able to compensate the decrease of absorptive surface. This adaptation process is regulated by lot of different factors, which include different trophic factors, hormones and the composition of the diet is also important. To illustrate the complexity of problem there are lot of different local intestinal hormones and other known hormones (insulin, thyroid hormones, steroid hormones) influencing the function and regeneration of intestine. Certain dietary components such as glutamine (an amino acid) or fibers also have beneficial effect on this process.
- There are differences in the local immune response to gluten and other antigens, toxins and pathogens. The primary function of intestinal immune system is to protect us from food and bacterial antigens. If this defense is not functioning well it may result in an abnormal response and further damage. Alternatively, if it functions well it will protect from evident damaging factors.
I only mention three main possible components for the differences we experience in the manifestations of celiac disease. All of them are discussed in thick textbooks and thousands of papers.
THE DIAGNOSIS OF CELIAC DISEASE is not simply determined by a biopsy, but rather based on a combination of:
In my experience, there are different possible combinations of the above components, which I will lay out in a table below, with different cases labeled by the letters A through G.
A B C D E F G Symptoms + + + - - + - Serology/immunology + ? + + + ? - Biopsy + + ? + ? ? + Response to diet + + + + + + +
Although this table can provoke a long scientific discussion, the main message I am trying to get across is that if somebody has symptoms or serological/immunological reactions to gluten or abnormal intestinal mucosa which disappears on a gluten-free diet, this individual should be considered as gluten sensitive.
I think most of the patients belong to the first group [A]. There is no any difficulty in diagnosing these cases. However, we all know that the wide diversity of possible symptoms (from mild anemia to classic celiac appearance) is frequently not appreciated by physicians practicing in this country.
If somebody has symptoms suggestive of celiac disease, but the serology is not conclusive [B] and other diseases resulting in similar symptoms are excluded we should perform intestinal biopsy. If the biopsy is positive, and the patient has favorable response to gluten withdrawal the diagnosis of CD is considered proven.
If both the symptoms and serology are suggestive of celiac disease [C], but the histology is not confirmative a trial with gluten-free diet should be started. Cases belonging to this group are reported recently (Picarelli A et al: Gluten sensitive disease with mild enteropathy. Gastroenterology 1996;111:608-616)
If there are no symptoms but the serology positive (it frequently happens during screening studies) [D] intestinal biopsy should be performed. Typical cases are the asymptomatic family members.
The group [E] is a theoretical possibility. If the serology positive during a screening test but the biopsy is not conclusive there are two options. First, to perform sophisticated tissue culture and immunhistochemical studies on the biopsy specimens to prove the gluten sensitivity. Alternatively, a trial with a gluten-free diet and a following the changes in serology titers can be considered.
We may have patients with symptoms suggestive of celiac disease, who do not have positive serology and the histology is not absolutely confirmative [F], but other possible causes are ruled out . In these cases, I believe a trial with a gluten-free diet can be started. I would like to emphasize here that I do not accept this possibility without a very thorough medical work-up. Self-diagnosed patients may have a high risk for other possible harmful conditions (e.g. intestinal lymphoma).
The group [G] is a theoretical possibility. There are publications when teenagers undergoing upper GI endoscopy had flat intestinal mucosa and they responded well to diet. Serological studies were not reported. They did have recurrent gastric pain, which was the main indication of endoscopy (Group B?). The intestinal atrophy was an "accidental" finding. I created this group for teenagers primarily, because based on my experience this is a "silent period" of celiac disease. I had a body builder, extremely muscular boy with flat mucosa and without abnormal serology. He was on diet temporarily when he was infant and that time responded well to GF diet.
I would like to emphasize that there are no 100% perfect tests in the medicine. The diagnosis of every disease requires several different laboratory and other diagnostic procedures.
THE VALUE OF SEROLOGY TESTS:
In patients without selective IgA deficiency we can accept the following parameters from the serologic tests:
SENSITIVITY- the proportion of subjects with the disease who have positive test. It indicates how good a test is at identifying the disease.
IgA antigliadin antibody test: average: 78% range: 46-100%
IgG antigliadin antibody test: average: 79% range: 57-94%
IgA antiendomysium test: average: 97% range: 89-100%
POSITIVE PREDICTIVE VALUE- is the probability that a person with a positive results actually has the disease.
IgA antigliadin antibody test: average: 72% range: 45-100%
IgG antigliadin antibody test: average: 57% range: 42-76%
IgA antiendomysium test: average: 92% range: 91-94%
I would like to make two comments:
-patients with selective IgA deficiency are not able to produce IgA antigliadin and IgA antiendomysium antibodies. Among my pediatric patients 1 out of 13 patients with CD had selective IgA deficiency (7.69%). False negatives test results may occur in these patients.
-It was clearly shown in large study from Europe (B|rgin-Wollf et al. Arch Dis Child 1991;66:941-947) that the endomysium antibody (EmA) test is less specific and sensitive in children below two years of age. They found that the sensitivity of the EmA test decreased from 98% to 88 % in children younger than 2 years of age. It means that 12% of their patients with celiac disease, who were younger than two years of age, but were not selective IgA deficient, did not have increase in their endomysium antibody levels. Thus, this group is also more at risk for false negative test results.
VALUE OF INTESTINAL BIOPSY
The intestinal biopsy is an important part of the diagnosis. However, the routine histological examination has never been considered as specific itself for celiac disease. There are sophisticated, but not routinely used tests on biopsy tissues which can be considered more specific than the simple histological test. These are computerized morphometric analyzes and specialized immunological tests on biopsy specimens in the presence of gliadin. They can be done in cases when there is any doubt regarding the diagnosis.
The routine histological examination of intestinal biopsy specimen may be inconclusive in small percentage of patients with so-called "patchy lesions" in the duodenum. It means that there are histologically normal looking spots with finger like villi and pathologic spots showing flattened mucosa in the upper half of the duodenum. If CD is suspected, the gastroenterologist should obtain several biopsies from different spots of the whole duodenum. Most of the endoscopist routinely examine only the upper half of the duodenum (duodenal bulb and the descending part). The transverse segment of duodenum is not viewed routinely. Few endoscopic centers have enteroscope, which is a longer and more flexible endoscope for the examination of entire duodenum and jejunum. The enteroscopy allows to obtain biopsies even from the jejunum. The histological examination of a single biopsy specimen may increases the risk of false negative diagnosis.
The experience of pathologist in the interpretation of small intestinal histology is important. In Centers with expertise in celiac disease, the gastroenterologist routinely reviews the histologic slides together with the pathologist.
There is still a possibility of inconclusive results if multiple biopsies are obtained and the histological interpretation is appropriate (see again: Picarelli A et al: Gluten sensitive disease with mild enteropathy. Gastroenterology 1996;111:608-616). All disease has a developmental process. It means that it takes time for the pathological changes to be evident. There are cases when the symptoms suggest CD, however, the histology is not conclusive. This problem occurs in only few of the cases. A repeated biopsy may be necessary after a period of higher gluten challenge. However, if the antiendomysium antibody test is positive and the histology is not conclusive a gluten-free diet is recommended.
Karoly Horvath, M.D., Ph.D.
Associate Professor of Pediatrics
Pediatric Gastroenterology and Nutrition
University of Maryland at Baltimore
The test for anti-endomysial antibodies is more subjective and more complicated for the lab to perform than the anti-gliadin assays. It involves serially diluting some of the patients serum, that is, diluting it by 1/2 then 1/4, 1/8, 1/16, etc. and putting these dilutions on a glass slide that has some sort of tissue affixed to it. The slide is then processed with various solutions and examined under a fluorescent microscope to determine if any of that serum binds to any of the proteins in the tissue. If so, then that patient is confirmed as having antibodies to that particular protein. This method of testing is called an IFA or sometimes IIFA. It stands for Indirect Immuno-Fluorescent Assay.
The selection of which tissue slide to use is determined by what specific protein, hence which antibody, you are specifically looking for. Endomysial antibodies react with the endomysium, which is a sheath of reticular fibrils that surround each muscle fiber. Therefore, to detect endomysial antibodies, you would want to use a tissue substrate that contains a lot of muscle tissue. The substrate used most often for this assay is distal sections of the esophagus. These are very thinly sliced and fixed to the slide. They contain muscle fibers and not much else so there is a lot of endomysium available to react with the anti-endomysial antibodies.
Reading this test involves viewing the reacted slides with a fluorescent microscope to make the determination. This requires a highly skilled and trained eye and, of necessity, is somewhat subjective. You are looking for a green fluorescence in the endomysium covering the muscle fibers. The test is reported as the "titer" or final dilution in which the fluorescence can still clearly be seen. As you can imagine, this is very subjective. There are no standardized values and it is up to the judgement of the particular technician what the endpoint titer is. Recently, (1998) the endomysial antigen targeted by the anti-endomysial antibodies was identified as the protein cross-linking enzyme known as tissue transglutaminase (tTG). This has enabled the production of an antigen specific ELISA assay incorporating tTG as a reliable and objective alternative to the traditional and subjective.
Immunofluorescence based assays. In clinical trials, the correlation with the endomysial IFA assay has been shown to be close to 100%. This is a test that has been very well received in the professional community. It is an ELISA, like the anti-gliadin antibody test and, as such, is not subject to interpretation like the IFA. That is the greatest advantage to this new test! With this or any ELISA, the response is measured on an instrument that calculates the amount of light of a particular wavelength that is absorbed by the solution and prints out a numerical result. There is no chance of human error skewing the results because there is no judgement call involved. The ELISA plate, regardless of what you are testing for, is processed with at least three control sera (sometimes as many as eight) in addition to the unknown sample being tested. There is a negative serum and at least two positive sera containing different levels of the antibody being tested. There are specific requirements for the absorption levels of these three controls. That is, each of them has a minimum or maximum (or both) number that must be seen by the instrument in order for it to be a valid test. If there is any variance from these expected numbers, it is an indication that something went wrong and the test results are discarded and the test repeated. There is therefore no way the20 technician could report inaccurate results, (assuming they diluted the sample correctly). Either the test was valid, and you can rely upon the accuracy of the result, or the test is invalid, and the entire result discarded. If any error was made during the processing of the ELISA plate, it would result in the control sera numbers being out of range and the entire test result would be thrown out.
In summary, the tTG ELISA is measuring the same thing that the endomysial IFA is measuring but with a method that is more sensitive and specific and not subject to interpretation.
IgA class Reticulin antibodies are found only in Celiac disease and dermatitis herpetiformis. These antibodies are found in approximately 60% of CD patients and 25% of DH patients. This test is falling into disuse because of the limited utility and the availability of better tests. It is an IFA performed on a tissue substrate with all the attendant problems that go along with it.
The development of all of these serum assays has tremendously simplified the diagnosis of CD and improved the accuracy as well. The original criteria for diagnosis according to the European Society for Pediatric Gastroenterology and Nutrition, (ESPGAN), involved a year of arduous studies with: a) an initial positive gut biopsy, b) 6 months on a gluten free diet, c) a second, negative gut biopsy, d) a gluten challenge for 6 months and e) a third, positive gut biopsy. The revised ESPGAN criteria call for positive results in two of the serological tests confirmed by a single positive biopsy. In practice, many gastroenterologists are utilizing the serologies in conjunction with a controlled diet and the clinical presentation to form a basis for diagnosis without the need for the invasive procedure.
Through the auspices of the Celiac Disease Foundation and others, a professional symposium and workshop was organized earlier this year in Marina Del Rey, California with participants from Europe as well as the U.S. to establish standards for reporting test results. This should improve testing and diagnosis even more. At the conclusion of this conference a Celiac Disease Standardization Committee was formed to investigate and make recommendations on a standardized method of reporting results.
The following detailed explanation of celiac tests is from Tom Ryan, Technical Service Specialist, INOVA Diagnostics, Inc.
There has been a lot of discussion about serological testing for celiac disease recently, specifically regarding tTG (tissue Transglutaminase) testing. I will try to answer some of the many questions that have appeared on this list about all of the tests. First, and this applies to any of the blood tests, you must currently be on a gluten containing diet for the tests to be accurate. Antibodies are produced by the immune system in response to substances that the body perceives as threatening. The immune response that your body produces is it's response to being exposed to gluten in the diet and it's subsequent effect on the intestinal mucosa. If there is no gluten in the diet, then there is no response that we can measure. A brief change in diet will not have a noticeable effect. If you have been gluten free for a week or so, it won't make any great difference. The response might be marginally less but the difference is insignificant because the body has not had time to respond to the change. Conversely, if you have been gluten free for a protracted period of time and decide to be tested, a brief challenge of a couple of weeks is not enough to elicit a response and get an accurate test. There are several steps that take place to generate an immune response and it takes time both for the positive reaction when gluten is present and to clear the antibodies when gluten is eliminated. There has been a great deal of discussion about how much and how long a challenge should be and there is no consensus. Talk with your Doctor. My personal feeling is that the minimum is 2 slices of bread per day for 6 weeks to get an accurate test but I would not try to second guess the Doctor. There are basically four tests that can be performed to aid in diagnosing celiac disease. Notice that I say they will "aid" in diagnosing celiac disease. Immunology is fairly accurate but it is far from being an exact science. All of the lab tests, regardless of the type or source, are presented as aids to diagnosis. They should not be used alone as a basis for diagnosis but rather are intended to be considered in conjunction with the physical examination of the patient as well as the reported symptoms, etc. by a trained physician. There has been a great deal of confusion about what the tests are and I hope to alleviate some of the misunderstandings. There are many terms that we hear. tTG, IgA, IgG, ELISA, etc. What are all of these? Some contributors to the list make reference to the "IgA" or "IgG" test or to the "ELISA" test. These labels are incomplete for our purposes and could be referring to any number of different tests.
We all have, within our bodies, a family of closely related although not identical proteins that are capable of acting as antibodies. These are collectively referred to as "immunoglobulins". Five major types of immunoglobulins are normally present in the human adult. They are IgG, IgA, IgM, IgE and IgD. Each of these is a shorthand way of writing "immunoglobulin gamma G" (or A or M, etc.) and they each perform a different function in our systems. IgG is the principal immunoglobulin in human serum. It is important in providing immunity in a developing fetus because it will pass across the placental barrier. IgA is the principal immunoglobulin in secretions from respiratory and intestinal mucosa. IgE is a gamma globulin produced by cells lining the intestinal and respiratory tracts. It produces the antibodies associated with most hypersensitivity (allergic) responses. It is associated with asthma, hay fever, etc. IgM is a globulin formed in almost every immune response in the early part of the reaction. IgD is a rare protein present in normal sera in a tiny amount. These designations refer to the type of protein that is carrying the antibody in question. Both IgG and IgA subtypes of anti-gliadin antibody are produced, hence we refer to them as "IgG gliadin" or "IgA gliadin". Collectively they are anti-gliadin antibodies.
Both IgA and IgG anti-gliadin antibodies (AGA) are detected in sera of patients with gluten sensitive enteropathy (celiac disease). IgG anti-gliadin antibodies are more sensitive but are less specific markers for disease compared with IgA class antibodies. IgA anti-gliadin antibodies are less sensitive but are more specific. In clinical trials, the IgA antibodies have a specificity of 97% but the sensitivity is only 71%. That means that, if a patient is IgA positive, there is a 97% probability that they have CD. Conversely, if the patient is IgA negative, there is only a 71% probability that the patient is truly negative for CD. Therefore, a positive result is a strong indication that the patient has the disease but a negative result doesn't necessarily mean that they don't have it. False positive results are rather uncommon but false negative results can occur.
On the other hand, the IgG anti-gliadin antibodies are 91% specific and have an 87% sensitivity. This means that they will show positive results more readily but there isn't as strong a correlation with CD. It is less specific. Patients with other conditions but not afflicted with CD will occasionally show positive results. IgG anti-gliadin antibodies are detectable in approximately 21% of patients with other gastrointestinal disorders. This test might yield false positive results but is less likely to yield false negative results.
A sensitive testing protocol includes testing for both IgA and IgG anti-gliadin antibodies since a significant portion of celiac patients (approx. 2-5%) are IgA deficient. This combined IgA and IgG anti-gliadin antibody assay has an overall sensitivity of 95% with a specificity of 90%.
The type of test used to detect the anti-gliadin antibodies is called an ELISA. This is an acronym and it stands for Enzyme Linked Immuno- Sorbent Assay. "ELISA" is not a test in itself. It is a method of testing and it is a relatively simple test to perform. It involves putting a measured amount of diluted patient serum into the wells of a specially constructed and prepared plate and incubating it for a period of time with various chemicals. The end result is a color change, the intensity of which is dependent upon the concentration of anti-gliadin antibody (or other protein being measured) in the patient serum. The ability of this colored solution to absorb light at a particular wavelength can be measured on a laboratory instrument and mathematically compared with solutions that contain a known amount of anti-gliadin antibody to arrive at a number for the amount of antibody present. The sample can then be classified as negative, (0-20 units); weak positive, (21-30 units); or moderate to strong positive if greater than 30 units.
The purpose of testing for anti-gliadin antibodies includes, in addition to diagnosis of gluten sensitive enteropathy, monitoring for compliance to a gluten free diet. IgA gliadin antibodies increase rapidly in response to gluten in the diet and decrease rapidly when gluten is absent from the diet. The IgA anti-gliadin antibodies can totally disappear in 2-6 months on a gluten free diet, so they are useful as a diet control. By contrast, IgG anti-gliadin antibodies need a long time, sometimes more than a year, to become negative. The reverse is also true. That is, a patient with CD who has been on a gluten free diet and tests negative for IgA anti-gliadin antibodies, will show a rapid increase in antibody production when challenged by gluten in the diet. Approximately 90% of challenged patients will yield a positive IgA anti-gliadin result within 14-35 days after being challenged. The IgG antibodies are somewhat slower.
I got a very concise reply from Tom Ryan, Technical Service Specialist at INOVA Diagnostics, Inc. who make the test. He answered with just about everything you could want to know about the test, so I will quote his message here:
"Celiac disease and dermatitis herpetiformis, more specifically termed gluten sensitive enteropathy are essentially the same disease but with different clinical presentations. It is an inability of the body to process the protein gliadin properly. Gliadin is a fraction of the gluten protein found in wheat and some other cereal grains. The body produces IgA and IgG antibodies to this protein.
"This results in an inability to digest and absorb it properly and gives rise to the production of antibodies to the sheaf of fibers that surround the smooth muscle layer of the small intestine. This is called the endomysium and it is the presence of these IgA type antibodies that is being tested for with the endomysial antibody test. The IgA antibodies are very specific for celiac disease. IgA type endomysial antibodies are not found in any other condition. This test is very subjective, however. It involves applying some of the patient's serum to a section of tissue fixed onto a glass slide, treating it with various solutions, and viewing it under a fluorescent micrscope. If the endomysium surrounding the muscle fibers shows a green, glowing fluorescence, the test is positive. This involves a great deal of expertise and judgement viewing the slide and is subject to interpretation. Tissue transglutaminase has been identified as the protein that is actually being targeted in the endomysium and the tTG test has been developed to measure antibodies to this protein directly rather than looking for its visual effect on the endomysium so it's easier, faster and less subjective than the endomysial test.
"A very small percentage of celiac patients are IgA deficient, not all of them. Only about 2-4% of them show this deficiency. Still, - - this is significantly higher than the general population where only about one half of a percent of the people are IgA deficient. It is significant enough to look for an additional tool to diagnose this disease. It has been found that those who are IgA deficient are usually IgG positive and that is the reason for testing for IgG. To catch the rare patient who is IgA deficient.
"None of these tests, or any others, can report accurate results if the patient has been following a GF diet because the challenge by gluten in the diet is necessary to trigger the body to produce these antibodies or to cause a change in the jejunum that would be observable by endoscopy or detectable by biopsy."
My thanks to Tom Ryan for the detailed answer, and to the many of you who shared what they knew about the test, too.
Ferreira M, Lloyd Davies S, Butler M, Scott D, Clark M, Kumar P. Endomysial antibody: is it the best screening test for coeliac disease? Gut 1992;33:1633-1637.
Khoshoo V, Bhan MK, Puri S, Jain R, Jayashree S, Bhatnagar S, Kumar R, Stintzing G. Serum antigliadin antibody profile in childhood protracted diarrhea due to coeliac disease and other causes in a developing country. Scand J Gastroenterol 1989;24:1212-1216.
Chan KN, Phillips AD, Mirakian R, Walker-Smith JA. Endomysial antibody screening in children. J Pediatr Gastroenterol Nutr 1994;18:316-320.
Bode S, Weile B, Krasilnikoff PA, Gdmand-Hyer E. The diagnostic value of the gliadin antibody testing celiac disease in children: a prospective study. J Pediatr Gastroenterol Nutr 1993;17:260-264.
Calabuig M, Torregosa R, Polo P, Tom s C, Alvarez V, Garcia-Vila A, Brines J, Vilar P, Farr C, Varea V. Serological markers and celiac disease: a new diagnostic approach ? J Pediatr Gastroenterol Nutr 1990;10:435-442.
The purpose of this copyright is to protect your right to make free copies of this paper for your friends and colleagues, to prevent publishers from using it for commercial advantage, and to prevent ill-meaning people from altering the meaning of the document by changing or removing a few paragraphs.