Carrier detection in haemophilia A families: comparison of conventional coagulation parameters with DNA polymorphism analysis – first report from India S. SHETTY, K. GHOSH, A. PATHARE* and D. MOHANTY Institute of Immunohaematology (ICMR), 13th floor, KEM Hospital, Parel, Mumbai 400 012, India, *Department of Hematology, KEM Hospital, Parel, Mumbai 400 012, India In a developing country like India, where a routine coagulation profile is not available to all the haemo- philiacs and their family members, the DNA metho- dology of carrier detection investigation is not viable for most people. Thus any simple investigation that would be comparatively inexpensive but fairly accu- rate and available to a large number of patients would be ideal, not only for the patients but also for the community at large. Carrier detection in X-linked recessive disorders by phenotypic assay is often difficult because of a combination of the effects of individual variability and chance differences in the extent of lyonization [1]. However, laboratory data can be used to calculate an odds ratio favouring carriership, this can then be combined with the probability of carriership derived from the pedigree analysis to obtain a final probability of carriership. Currently, however, the prefered method for carrier detection by phenotypic analysis is the bivariate linear dis- criminant analysis based on factor VIII:C and von Willebrand factor antigen (vWF:Ag) measurements [2]. Since DNA technology now allows carriership to be diagnosed definitely, the efficacy of laboratory data in the classification of carriers can now be analysed. We were also prompted to undertake this study by the paucity of reports from India as well as our previous work [3] on these aspects. We sought to determine the relative contribution of coagulation and DNA marker testing in an unselected haemo- philia population who had requested for carrier diagnosis, and a healthy age matched control group considering the effect of age and ABO blood group on factor VIII:C and vWF:Ag level. Materials and methods Study group The carrier group consisted of 69 carriers whose carrier status was confirmed by DNA analysis and Summary. Linear discriminants that include data on factor VIII:C and von Willebrand factor antigen levels are well-established tools in estimating the probabil- ity of carriership in haemophilia A families. A comparison between the conventional coagulation data, i.e. the ratio of factor VIII:C and von Willebrand factor antigen, and the DNA analysis techniques was made in 98 confirmed carriers (39 obligatory, 69 detected by gene tracking analysis) and 71 normal age matched females who did not have any history of bleeding and were not taking any drug. The lowest misclassification rate, i.e. 7% among the carriers, was seen when a cut-off value of 0.7 was chosen. In the case of normals, all were outside this cut-off value. Thus, it was considered as a workable reference value for classifying the carriers in haemophilia A families in our laboratory. We conclude that the optimal service for haemophilia A carrier diagnosis must include above coagulation test probabilities as well as DNA marker studies. However, it is recommended that the smaller laboratories in developing countries can benefit immensely by only establishing factor VIII:C and von Willebrand factor antigen estimation. Keywords: carrier, coagulation, haemophilia A, polymorphism. Correspondence: Dr Dipika Mohanty, MD, PhD, FRCPath, Director, Institute of Immunohematology (ICMR) 13th floor, KEM Hospital Parel, Mumbai 400 012, India. Received 28 October 1998; accepted 15 February 1999 Haemophilia (1999), 5, 243–246 Ó 1999 Blackwell Science Ltd 243