Role of interventional radiology in pregnancies complicated by placenta accreta spectrum disorders: a systematic review and meta-analysis

OBJECTIVE: To determine the potential benefit of the interventional radiology (IR) in improving the outcome of women undergoing surgery for placenta accreta spectrum (PAS) disorders. METHODS: Medline, Embase and CINAHL databases were explored searching a robust cadre of terms relating to PAS. The primary outcome was the intraoperative estimated blood loss (EBL,L); Secondary outcomes were: units of packed red blood cells (PRBC) fresh frozen plasma (FFP), platelets (PLT) cryoprecipitate transfused, operative time (min), length of hospital stay (d), EBL ≥2.5 L, PRBC transfused ≥5 units, surgical complications, bladder ureteral injuries, re-laparotomy, infection, disseminated intravascular coagulation (DIC) and complications related to endovascular catheter placement. Only studies reporting the incidence of or the mean difference in the observed outcomes in women affected by PAS disorders who had compared to those who did not have IR procedures before surgery were considered for the inclusion. All these outcomes were explored in the overall population of women with a prenatal diagnosis of PAS and in those undergoing hysterectomy. Quality assessment of each included study was performed using the Risk of Bias In Non-randomized Studies-of Interventions tool (ROBINS-I). The GRADE methodology was used to assess the quality of the body of retrieved evidence. RESULTS: Fifteen studies (955 women) were included. Mean EBL ((MD -1.01, 95% CI-1.59; -0.43); p<0.001) and the risk of EBL ≥2.5 L (OR 0.18, 95% CI 0.04-0.78, p=0.02) were significantly less in the cases compared to controls. There was no significant difference in the other outcomes explored. In the subgroup analysis of pregnancies complicated by PAS undergoing hysterectomy, the EBL (MD -0.68; 95% CI -1.24, -0.12, p=0.02) and the number of transfused FFP units (MD -1.66; 95% CI -2.71, -0.61, p=0.02) were significantly less in women undergoing endovascular IR procedures compared to controls. Furthermore, women undergoing IR had a significantly lower risk of requiring transfusion of ≥5 PRBC units (OR 0.10, 95% CI 0.02-0.47, p=0.04). Overall, the complications related to the placement of endovascular catheter occurred in 5.3% (95% CI 2.6-8.9; I2 65.3%) of the pregnancies undergoing IR. Overall quality of evidence, as qualified by GRADE, was low. CONCLUSION: The current data available provide encouraging evidence that IR procedures may be associated with lower EBL and rates of transfusion in pregnancies complicated by PAS disorders. However, given the overall very low quality of the evidence, as reflected in the GRADE assessment, further large studies are needed in order to confirm the beneficial role of IR in improving the outcome of women undergoing surgery for PAS disorders.