Preimplantation genetic testing for aneuploidy: a comparison of live birth rates in patients with recurrent pregnancy loss due to embryonic aneuploidy or recurrent implantation failure.

STUDY QUESTION: Can preimplantation genetic testing for aneuploidy (PGT-A) improve the live birth rate and reduce the miscarriage rate in patients with recurrent pregnancy loss (RPL) caused by an abnormal embryonic karyotype and recurrent implantation failure (RIF)? SUMMARY ANSWER: PGT-A could not improve the live births per patient nor reduce the rate of miscarriage, in both groups. WHAT IS KNOWN ALREADY: PGT-A use has steadily increased worldwide. However, only a few limited studies have shown that it improves the live birth rate in selected populations in that the prognosis has been good. Such studies have excluded patients with RPL and RIF. In addition, several studies have failed to demonstrate any benefit at all. PGT-A was reported to be without advantage in patients with unexplained RPL whose embryonic karyotype had not been analysed. The efficacy of PGT-A should be examined by focusing on patients whose previous products of conception (POC) have been aneuploid, because the frequencies of abnormal and normal embryonic karyotypes have been reported as 40-50% and 5-25% in patients with RPL, respectively. STUDY DESIGN, SIZE, DURATION: A multi-centre, prospective pilot study was conducted from January 2017 to June 2018. A total of 171 patients were recruited for the study: an RPL group, including 41 and 38 patients treated respectively with and without PGT-A, and an RIF group, including 42 and 50 patients treated respectively with and without PGT-A. At least 10 women in each age group (35-36, 37-38, 39-40 or 41-42 years) were selected for PGT-A groups. PARTICIPANTS/MATERIALS, SETTING, METHODS: All patients and controls had received IVF-ET for infertility. Patients in the RPL group had had two or more miscarriages, and at least one case of aneuploidy had been ascertained through prior POC testing. No pregnancies had occurred in the RIF group, even after at least three embryo transfers. Trophectoderm biopsy and array comparative genomic hybridisation (aCGH) were used for PGT-A. The live birth rate of PGT-A and non-PGT-A patients was compared after the development of blastocysts from up to two oocyte retrievals and a single blastocyst transfer. The miscarriage rate and the frequency of euploidy, trisomy and monosomy in the blastocysts were noted. MAIN RESULT AND THE ROLE OF CHANCE: There were no significant differences in the live birth rates per patient given or not given PGT-A: 26.8 versus 21.1% in the RPL group and 35.7 versus 26.0% in the RIF group, respectively. There were also no differences in the miscarriage rates per clinical pregnancies given or not given PGT-A: 14.3 versus 20.0% in the RPL group and 11.8 versus 0% in the RIF group, respectively. However, PGT-A improved the live birth rate per embryo transfer procedure in both the RPL (52.4 vs 21.6%, adjusted OR 3.89; 95% CI 1.16-13.1) and RIF groups (62.5 vs 31.7%, adjusted OR 3.75; 95% CI 1.28-10.95). Additionally, PGT-A was shown to reduce biochemical pregnancy loss per biochemical pregnancy: 12.5 and 45.0%, adjusted OR 0.14; 95% CI 0.02-0.85 in the RPL group and 10.5 and 40.9%, adjusted OR 0.17; 95% CI 0.03-0.92 in the RIF group. There was no difference in the distribution of genetic abnormalities between RPL and RIF patients, although double trisomy tended to be more frequent in RPL patients. LIMITATIONS, REASONS FOR CAUTION: The sample size was too small to find any significant advantage for improving the live birth rate and reducing the clinical miscarriage rate per patient. Further study is necessary. WIDER IMPLICATION OF THE FINDINGS: A large portion of pregnancy losses in the RPL group might be due to aneuploidy, since PGT-A reduced the overall incidence of pregnancy loss in these patients. Although PGT-A did not improve the live birth rate per patient, it did have the advantage of reducing the number of embryo transfers required to achieve a similar number live births compared with those not undergoing PGT-A. STUDY FUNDING/COMPETING INTEREST(S): This study was supported by the Japan Society of Obstetrics and Gynecology and grants from the Japanese Ministry of Education, Science, and Technology. There are no conflicts of interest to declare. TRIAL REGISTRATION NUMBER: N/A.

Anticardiolipin antibodies and recurrent spontaneous abortion in Brazilianwomen / Anticardiolipin antibodies and recurrent spontaneous abortion in Brazilianwomen

Aims: To determine the frequency of anti-cardiolipin antibodies (aCL) in women with previous history of recurrent spontaneous abortion (RSA).Methods: Medical records from pregnant women seen from April 2005 to December 2008 at the High-Risk Pregnancy Unit at the Hospital de Base from FUNFARME (Fundação Faculdade Regional de Medicina), in São José do Rio Preto, São Paulo, Brazil, were revised. Patients olderthan 18 years who had at least two spontaneous abortions and who were tested for aCL wereincluded in the study. Data on maternal age, number of miscarriages and the results of serological tests for aCL were recorded. The exact Fisher's test was used to compare the results. A p value less than 0.05 was considered significant.Results: During the study period a total of 294 pregnant women were seen, from whom 44 consecutive women fulfilled the inclusion criteria. The overall mean age was 33.8±5.4 years (range: 22 to 44; median: 34). Eighteen (40.9%) patients were reagent for aCL and 26 (59.1%) non-reagent,and the difference between their mean age was not statistically significant (reagent: 34.6±5.8 years; non reagent: 33.2±5.2 years, p=0.4001). Fourteen (77.8%) patients presented IgM aCL and six (33.2%), IgG aCL. Two patients (11%) were reagent for both IgM and IgG aCL. In the most of cases the aCL antibody titers were compatible with low risk for pregnancy morbidity. The number of abortions ranged from two to six. The average number of abortions among those reagent for aCL was 3.5±1.1 and in those non-reagent was 2.9±1.1 (p=0.0813). Conclusions: The frequency of aCL was elevated among patients with a history of RSA,especially those having higher number of fetal losses. Among women with at least two spontaneousabortions, the mean number of abortions was not significantly different between those reagent for aCL and those non reagent.

Objetivos: Avaliar a frequência de anticorpos anticardiolipina (aCL) em mulheres comhistória prévia de aborto espontâneo recorrente (AER).Métodos: No período de abril de 2005 a dezembro de 2008 foram avaliados os dados deprontuários de gestantes atendidas no Ambulatório de Gestação de Alto Risco do Hospital de Baseda Fundação Faculdade Regional de Medicina (FUNFARME), em São José do Rio Preto, SãoPaulo, Brasil. Foram incluídas neste estudo pacientes com idade acima de 18 anos que tiveram pelomenos dois abortos espontâneos e foram avaliadas para aCL. O teste exato de Fisher foi usado paracomparar os resultados. Valor de p menor que 0,05 foi considerado significante.Resultados: Um total de 294 mulheres gestantes foram avaliadas durante o período doestudo, das quais 44 atendiam aos critérios de inclusão. A média de idade foi 33,8±5,4 anos(variação: 22 a 44; mediana: 34). Dezoito pacientes (40,9%) foram reagentes para aCL e 26 (59,1%)não reagentes, e a diferença entre a média de idade não foi estatisticamente significante (reagentes:34,6±5,8 anos; não reagentes: 33,2±5,2 anos, p=0,4001). Quatorze (77,8%) pacientes apresentaramaCL IgM e seis (33,2%) aCL IgG. Duas pacientes (11%) foram reagentes para ambas as classes deaCL, IgM e IgG. Na maioria dos casos os títulos de anticorpos aCL foram compatíveis com baixorisco para morbidade na gravidez. O número de abortos variou de dois a seis. O número médio deabortos entre as reagentes para aCL foi de 3,5±1,1 e entre as não reagentes foi de 2,9±1,1(p=0,0813).Conclusões: A frequência de aCL foi alta entre as pacientes com história de AER,especialmente entre aquelas que tiveram um maior número de perdas fetais. Entre as mulheres compelo menos dois abortos espontâneos, a média do número de abortos não foi significativamentediferente entre aquelas reagentes e não reagentes para aCL.