Results
Out of 797 women who were reviewed during data collection, 727 women were eligible for the study. Five hundred ninety-two (81.43%) women were managed by uterus preserving procedures, while 135 (18.56%) underwent caesarean hysterectomy. Their mean ages were 32.99 ± 4.94 and 33.83 ± 4.73 years, respectively. Of women manged by uterus preserving procedures, PAS type was accreta in 293 (49.49%) and 10 (7.41%) in women who had caesarean hysterectomy. Placenta percreta was present 86 (63.70%) of women who had caesarean hysterectomy. In women managed by caesarean hysterectomy, placental site was most commonly central, covering the internal os in 78 (57.78%) women, while it was equally localized anterior low, posterior low, or central in women who had uterus preservation. Among women who were treated with uterus preserving procedures, compression sutures were applied to 195 (32.94%) patients, placental bed swing was done in 115 (19.43%) patients, vascular ligation was performed in 89 (15.03%), and local uterine wall resection was achieved in 33 (5.57%) patients. Uterus preservation was unsuccessful in 20.1% of cases. Massive blood loss was reported in 129 (17.74%) of women. Patient characteristics, PAS features and management approaches are summarized in Table 1.
Incidence of intraoperative complications were significantly higher among women manged by caesarean hysterectomy. Specifically, unintentional cystotomy occurred in 33 (24.44%) women in the “caesarean hysterectomy” group compared to 41 (6.93%) in the “uterus preservation” group (p < 0.0001). Otherwise, bowel and ureteric injuries were comparable in both groups. PAS-associated massive blood loss was more common in “caesarean hysterectomy” group compared to “uterus preservation” group (43 [31.9%] vs. 86 [14.5%], respectively; p < 0.0001). There was no significant difference in the incidence of maternal admission to ICU between the 2 groups. Table 2 shows a summary of clinical outcomes of treatment groups.
On univariate analysis, uterus preservation was associated with lower risk of PAS-associated massive blood loss (unadjusted OR 0.36, 95% CI 0.24 - 0.56). After adjustment for significant or close-to-significance variables, management approach was not associated with higher risk of massive blood loss (aOR 1.71, 95% CI 0.78 - 3.81). Massive blood loss was significantly associated with BMI (aOR 1.09, 95% CI 1.02 - 1.15), gestational age at diagnosis (aOR 1.08, 95% CI 1.02 - 1.14), preoperative haemoglobin (aOR 0.77, 95% CI 0.65 - 0.92), interventional radiology (aOR 5.62, 95% CI 2.38 - 13.29), centrally located placenta (aOR 2.28, 95% CI 1.25 - 4.18), diffuse versus localized invasion (aOR 3.15, 95% 1.61 - 6.16), incision away from the placenta (aOR 0.26, 95% CI 0.13 - 0.51), bladder invasion (aOR 3.08, 95% CI 1.95 – 8.61), parametrial invasion (aOR 5.37, 95% CI 1.21 – 23.79), and intrauterine foetal death (aOR 10.25, 95% CI 1.51 – 69.34) (Table 3).