Introduction
This study examined national‐level trends, characteristics, and perioperative outcomes of women who had intra‐arterial balloon occlusion at cesarean hysterectomy for placenta accreta spectrum (PAS).
Material and methods
This was a population‐based retrospective observational study that queried the National Inpatient Sample from October 2015 to December 2018. Study population was women who underwent hysterectomy at cesarean delivery for PAS (n = 6440 in 806 centers). Exposure allocation was the use of intra‐arterial balloon occlusion. Main outcome measures were (a) characteristics associated with intra‐arterial balloon occlusion use, and (b) perioperative outcome including hemorrhage, blood transfusion, coagulopathy, shock, urinary tract injury, intra‐arterial balloon occlusion‐related complication (arterial injury, arterial thrombosis, and lower extremities ischemia), and death, assessed in multivariable analysis.
Results
Intra‐arterial balloon occlusion was used in 420 (6.5%) women in 64 (7.9%) centers. Utilization of intra‐arterial balloon occlusion during cesarean hysterectomy for placenta accreta decreased significantly over time (from 6.3% to 3.1%, p < 0.001), but not in placenta increta (from 12.8% to 9.3%, p = 0.204) or placenta percreta (from 21.3% to 17.5%, p = 0.344). In a multivariable analysis, patient factors (younger age, earlier year, obesity, diabetes mellitus), pregnancy factors (placenta increta/percreta, previous cesarean delivery, placenta previa, and early gestational age), and facility factors (large bed capacity, urban teaching status, and Northeast/West regions) represented the independent characteristics for using the intra‐arterial balloon occlusion (all, p < 0.05). In a classification‐tree model, the absolute difference in intra‐arterial balloon occlusion use among 18 utilization patterns was 48% (range, 0%–48%). In perioperative outcome analysis, women who received intra‐arterial balloon occlusion were more likely to have coagulopathy (adjusted odds ratio [aOR] 3.43) and arterial thrombosis (aOR 9.82) in placenta accreta, but less likely to have hemorrhage (aOR 0.25) in placenta increta, and blood transfusion (aOR 0.60) and urinary tract injury (aOR 0.28) in placenta percreta compared with those who did not (all, p < 0.05).
Conclusions
There is a wide range in the utilization of intra‐arterial balloon occlusion at cesarean hysterectomy for PAS based on patient, pregnancy, and facility factors, which implies that there is a lack of universal practice guidelines in this surgical procedure. Whether the use of intra‐arterial balloon occlusion in the severe forms of PAS improves surgical outcome merits further investigation.