Cordocentesis‐associated fetal loss and risk factors: experience of 6650 cases from a single center

OBJECTIVE: To identify the risk factors of fetal loss associated with cordocentesis METHODS: Based on our prospective database, a retrospective cohort study was conducted on pregnancies undergoing cordocentesis at mid-pregnancy from a single center (a tertiary hospital, teaching school). All consecutive cases were validated to retrieve those meeting the inclusion criteria, including 1) low risk pregnancies; 2) singleton pregnancies; 3) availability of the data of pregnancy outcomes; and 4) normal fetuses (no structural / chromosomal abnormalities and severe disorders). The cases with therapeutic termination of pregnancy were excluded. The various potential risk factors of fetal loss secondary to cordocentesis were tested, including effect of cordocentesis model training, procedure difficulty, placenta penetration, prolonged bleeding, fetal bradycardia, puncture site, and early gestation. RESULTS: A total of 10,343 procedures were performed during the study period. Of them, 6,650 met the inclusion criteria and were available for analysis. In evaluation of influence of model training, the fetal loss rate in the first 60 cases of practice (early learning curve) of 6 performers who had no prior model training (n=360 procedures) was significantly higher than that of 18 performers with prior model training (n=1,080 procedures) (6.9% vs 1.6%, p<0.001), whereas the loss rate during the next 60 cases of practice was comparable between both groups. After excluding the first 60 procedures in the groups of no prior model training, the overall fetal loss rate was still significantly higher than that of the control (1.6% vs 1.0%; p < 0.001). Penetration of the placenta (OR=2.65;95%CI:1.71-4.10), prolonged bleeding (OR=10.85;95%CI:5.27-22.36) and fetal bradycardia (OR=3.32;95%CI:1.83-6.04) were an independent risk factor of fetal loss. CONCLUSIONS: Cordocentesis model training, the only one modifiable factor, markedly reduces fetal loss in early learning curve of practice. Thus, practice without prior model training is no longer acceptable. Other significant risk factors are placental penetration, prolonged bleeding and bradycardia following the procedures. Cordocentesis-related fetal loss may be only 0.6%, much lower than that previously reported. This article is protected by copyright. All rights reserved.