Discrepancies between histology and serology for the diagnosis of coeliac disease (2)/In response by ProQuest

VIEWS: 6 PAGES: 3

More Info
									LETTERS TO THE EDITOR



termed early clamping. Anecdotally,                   Discrepancies between histology                           In all, we agree that it is important not to
there has been no evidence of significant             and serology for the diagnosis                         rely on serology alone for the diagnosis of
harm to these infants. Scientifically, the            of coeliac disease (1)                                 CD, but to allow serology to increase or
changes that have been noted are of                                                                          decrease your estimation of risk of disease.
arguable long-term medical conse-                     Editor – Sweis and colleagues showed                   However, considering the lifelong implica-
quence. If it were clearly detrimental to             discrepancies between histology and                    tions of a diagnosis of CD, one should still
the health of the child to perform                    serology in the diagnosis of coeliac disease           maintain a degree of suspicion and also take
clamping in one fashion or another,                   (CD) (Clin Med August 2009 pp 346–8),                  great care in interpreting villous atrophy in
surely legislation would have been insti-             and suggest we reduce our reliance on                  the absence of autoantibodies in any patient.
tuted by now. Therefore, we leave the deci-           serology testing in diagnosing and
sion of when to clamp and collect the                 excluding CD. However, we feel there are                                  DEVASENAN DEVENDRA
cord blood up to the preferences of the               major reasons to reconsider this.                          Consultant physician and honorary senior lecturer
mother and her physician or midwife. In                  The numbers reported here must be                                 NHS Brent, Central Middlesex Hospital
our experience, we have not found that                interpreted carefully: 10 out of 26 CD                                        and Imperial College, London
the difference in time involved makes for             patients who received serologic testing were
                                                      seronegative. This 38.5% occurrence of                                        CHUKWUMA UDUKU
a significant difference in collection.
                                                      seronegative CD is misleading. In the spirit                Research assistant and 6th year medical student
                                                      of Bayes theorem, the more common the                                   Department of Investigative Sciences
                      MOHAMED GABALLA
                                                      condition we are testing, the greater the per-                                     Imperial College, London
                      Director, Cardiology Research
                      Sun Health Research Institute
                                                      centage of false negative results.1 In this case,                                            EDWIN LIU
                                 Phoenix, AZ, USA
                                                      all 26 patients were selected due to the diag-                                 Associate professor, pediatrics
                                                      nosis of CD, meaning the prevalence in this                                      Section of Gastroenterology,
                            DAVID T HARRIS            group was already 100%. Therefore, this                                           Hepatology and Nutrition
                        Professor of immunobiology    group is bound to have a high number of                                             The Children’s Hospital,
                           Director, Stem Cell Bank   false negative tests. The authors correctly                                University of Colorado at Denver
                          The University of Arizona   state that a small number of cases of CD will
                                  Tucson, AZ, USA                                                                                   YANNOULLA WILSON
                                                      be missed by relying on serology alone, but
                                                                                                                      Laboratory manager, Autoimmune Serology
                       NATHAN COPELAND                the true prevalence is unknown, and this
                                                                                                                          Northwest London Hospitals NHS Trust
      Medical student, MD-Candidate Class of 2012     number is likely to be much lower than 38.5%.
                          The University of Arizona      In addition, the predictive value of using
                                  Tucson, AZ, USA     an ELISA-based method to detect tissue                 References
                                                      transglutaminase autoantibody (tTG)
                                                                                                             1    Shapiro D. The interpretation of diagnostic
                                                      remains open to discussion. There are                       tests. Stat Methods Med Res 1999;
                                                      currently numerous tTG assays available,                    8(2):113–34.
References                                            all with varying performances. The                     2    Li M, Yu L, Tiberti C et al. A report
1    Meier C, Middelanis J, Wasielewski B et al.      International tTG Workshop for CD                           on the International Transglutaminase
     Spastic paresis after perinatal brain damage                                                                 Autoantibody Workshop for Celiac Disease.
                                                      performed head-to-head comparisons of
     in rats is reduced by human cord blood                                                                       Am J Gastroenterol 2009;104:154–63.
                                                      various commercial and laboratory-based                3    Bonamico M, Mariani P, Thanasi E et al.
     mononuclear cells. Pediatr Res
     2006;59:244–9.
                                                      tTG assays. For this workshop, assays                       Patchy villous atrophy of the duodenum in
2    Ceriani Cernadas JM, Carroli G, Pellegrini L     reported sensitivities ranging from 82% to                  childhood celiac disease. J Pediatr
     et al. The effect of timing of 
								
To top