Comparison of results between karyotyping and microarray testing in postnatal and prenatal specimens: karyotyping is not dead yet. D. Warburton1, V. Jobanputra2, V. Aggarwal2, A. Sobrino3, M. Macera3, O. Nahum2, B. Levy2 1) Gen & Development, and Pediatrics, Columbia University, New York, NY; 2) Pathology, Columbia University, New York, NY; 3) N.Y. Presbyterian Hospital, Columbia University Medical Center, New York, NY.

   The advent of microarray analysis (CMA) to determine dosage changes in chromosomal complements has led to differences of opinion about whether standard cytogenetic analysis by G-banding should remain a routine part of clinical testing. We have compared results from approximately one year of laboratory testing when both procedures were performed on the same specimen. Aneuploidy is underrepresented among these cases, since no microarray was done if initial FISH detected aneuploidy or trisomy was indicated clinically. We divided results into those where (1) both tests gave normal results (2) an abnormality had the same diagnosis by either technique (3) a clinically significant abnormality was found only by CMA (4) both techniques were required to interpret the results and (6) an abnormality was found only by karyotype. CMA findings of uncertain significance have been scored as normal. These occurred in 9.8% of blood specimens and 4.6% of prenatal specimens. Of 480 blood specimens, studied because of clinical abnormalities in the patients,71 (14.8%)were abnormal. Of these 36 (7.5%) had dosage changes detectable only by CMA, demonstrating the power of this technique. In 20 specimens (4.2%) there was an abnormality diagnosable by both methods (mostly aneuploidy). In 9 cases both karyotype and CMA were necessary for a complete interpretation, and in 6 cases only the karyotype revealed the abnormality. Thus karyotyping was important for diagnosis in 15/71 (21.1%) of abnormal specimens. In prenatal specimens the abnormality rate, of course, was much lower. In 284 specimens there were 19 (6.7%) abnormalities, 4 diagnosable by either method, 3 diagnosable by CMA only, 8 where both methods were required and 4 where karyotype alone revealed the abnormality. Thus karyotyping was important for diagnosis in 12/19 = 63.2% of abnormal specimens. The abnormalities detected only by karyotyping consisted of a reciprocal translocation, a Robertsonian translocation, four complex rearrangements, and 4 mosaics, including an XY line in a girl with Turner syndrome. All are abnormalities expected to be missed by CMA but which had major consequences for the patient. Cases where both karyotype and CMA were necessary for interpretation were chiefly those where an abnormal karyotype was better delineated by CMA results (e.g. markers, visible deletions and derived chromosomes.) Our analysis clearly demonstrates that karyotyping is still a valuable part of cytogenetic analysis.

You may contact the first author (during and after the meeting) at