Risk of Placenta Accreta Spectrum Following Myomectomy: A Nationwide Cohort Study

ABSTRACT Placenta accreta spectrum (PAS) is described as a partial or total lack of decidua with adherence or invasion of the placenta to the myometrium and is a major cause of postpartum hemorrhage and peripartum hysterectomy. The risk of PAS is approximately 1 in 272 in the United States. Prior uterine surgeries including myomectomy disrupt the integrity of the endometrium and myometrium and may increase the risk of PAS. However, the association between myomectomy and PAS remains controversial. This retrospective nationwide cohort study aimed to investigate the effect of myomectomy, stratified by method of myomectomy, on the risk of PAS. Data were obtained from the Taiwan National Health Insurance Research Database, which includes prenatal, perinatal, demographic, and treatment data on nearly all pregnant patients in Taiwan. All pregnant people who gave birth after 20 weeks of gestation between January 2008 and December 2017 were included. Patients with incomplete data, previous history of PAS before myomectomy, robotic-assisted myomectomy, or a diagnosis of adenomyosis were excluded. Patients were classified into 3 groups according to method of myomectomy: laparotomic, laparoscopic, and hysteroscopic. A 1:1 propensity-score estimation matching with logistic regression was used for patients with and without history of myomectomy to minimize selection bias. The primary outcome was placenta accreta, whereas secondary outcomes included placenta previa, postpartum hemorrhage, placenta abruption, uterine rupture, and preterm delivery. Logistic regression models were used to analyze the association between myomectomy and adverse pregnancy outcomes and the risk of placenta accreta for the entire population. Analysis of variance and post hoc tests were used to analyze differences in clinical characteristics and adverse pregnancy outcomes according to different methods of myomectomy. A total of 1,393,628 patients were included in this study. Among them, 11,255 patients had a history of myomectomy. Placenta accreta spectrum occurred in 0.96% of those with a history of myomectomy and 0.20% of those without. The risk of PAS was significantly higher in patients with a history of myomectomy, compared with those without (adjusted odds ratio, 2.28; confidence interval [CI], 1.85–2.81; P < 0.01). There was an elevated risk of other adverse pregnancy outcomes including placenta previa, postpartum hemorrhage, cesarean hysterectomy, uterine rupture, preterm delivery, and placenta abruption among those with a history of myomectomy compared with those without. The presence of placenta previa significantly increased risk of PAS, as did the number of prior myomectomies. The propensity score–matched cohort analysis found a similar increase in PAS risk associated with myomectomy. Compared with subjects without a history of myomectomy, the risk of PAS was greater after hysteroscopic (adjusted odds ratio, 3.88; 95% CI, 2.68–5.63), laparoscopic (2.02; 95% CI, 2.79–5.62), and laparotomic myomectomy (1.86; 95% CI, 1.33–2.60). The results of this study find that a history of myomectomy is significantly associated with increased risk of PAS in subsequent pregnancies. Hysteroscopic myomectomy was associated with the highest risk of PAS, and the presence of placenta previa or history of multiple myomectomies further increased the risk.