Between 1998 and 2002, 59 couples at risk for trans­mitting β hemoglobinopathies were counseled about PGD. Forty-one couples initiated 63 PGD cycles (one cycle in 22 couples, two cycles in 16 couples, and three cycles in three couples), of which 20 cycles with <4 IVF embryos were canceled before biopsy and genotype analysis. Amongst 43 completed cycles, 302 cleavage stage em­bryos were biopsied, 236 (78%) gave a genotype result, of which 125 clearly had at least one normal allele with DGGE analysis (unaffected). Transfer of 100 embryos (1-4/cycle) established 16 pregnancies (37% for each com­pleted cycle), including 13 singletons, two sets of twins and one set of triplets (20% implantation per embryo transferred). Six pregnancies were lost within the first trimester but 10 underwent second trimester PND. Nine pregnancies (13 babies) were confirmed unaffected, but one singleton was a PGD misdiagnosis, and was selec­tively terminated. The misdiagnosis was attributed to ex­traneous contamination or a tube switch, and not to inac­curacy of the genotyping method. The triplet pregnancy was selectively reduced to twins, and all pregnancies went to term, resulting in the birth of 12 healthy babies [19].
Although accurate, the DGGE-based protocol is tech­nically demanding, time consuming, and not as sensitive as methods involving fluorescent PCR. To simplify PGD analysis, reduce diagnosis time, improve sensitivity, whilst maintaining accuracy and monitoring of ADO, we recently established a protocol based on real-time PCR, using fluo­rescent hybridization probes for mutation detection. We designed, standardized and validated mutation detection probes for the common β-thal mutations worldwide (and Hb S), through mutation analysis in >200 carriers, and additionally, 25 prenatal diagnoses. We adapted the method to PGD using nested PCR, with the second reac­tion in LightcyclerÔ capillaries (Roche Diagnostics GmbH, Mannheim, Germany), including fluorescent de­tection probes for melting curve analysis and allele assign­ment. So far, we have applied the LightcyclerÔ (Roche Diagnostics) protocol in 10 PGD cycles. Results (avail­able within 5 hours) were obtained in 81/89 blastomeres (91%), of which 69 blastomeres were also analyzed with the DGGE-PGD protocol, and genotypes were completely concordant. Thus, PGD using mutation detection with real-time PCR is accurate, but additionally, it is more sensitive and rapid, when compared to the DGGE-based method [20,21].