Morbidity associated with nonemergent hysterectomy for placenta accreta.

OBJECTIVE: The purpose of this study was to report the morbidity of nonemergent hysterectomy for suspected placenta accreta. STUDY DESIGN: This was a retrospective study of all patients who underwent nonemergent hysterectomy for placenta accreta at Tampa General Hospital from June 1, 2003 to May 31, 2009. RESULTS: Twenty-nine patients were identified. Diagnosis was suspected on ultrasound scanning in 26 women (6 women also underwent magnetic resonance imaging) and on direct vision at repeat cesarean section delivery in 3 women. All of the women were multiparous, and 18 women had undergone > or =2 cesarean section deliveries. Twenty-one women had a placenta previa, and 8 women had a low anterior placenta. Final pathologic findings revealed accreta (20 specimens), increta (6 women), and percreta (3 women). Mean total operative time was 216 minutes; blood loss was 4061 mL. Two women had ureteral transection (1 was bilateral); 3 women had cystotomy, and 3 women had partial cystectomy. Postoperative hemorrhage occurred in 5 women; 1 hemorrhage resolved after catheter embolization, and the other 4 hemorrhage required reoperation. CONCLUSION: Nonemergent hysterectomy for placenta accreta is associated with significant morbidity in the forms of hemorrhage and urinary tract insult.

Risk factors for recurrent hemorrhage after successful mesenteric arterial embolization.

The objective of this study was to describe the risk factors and to determine the outcomes after recurrent gastrointestinal hemorrhage after successful mesenteric arterial embolization A retrospective analysis was undertaken of a single-center experience with mesenteric arterial embolization performed for gastrointestinal hemorrhage over a 5-year period. Statistical analyses including Student's t test and Fisher's exact test were used to compare results. For the years 2001 through 2005, 36 patients (10 women; average age, 60.8 years) underwent 37 technically successful mesenteric embolizations for acute gastrointestinal hemorrhage. Two (5.4%) cases required surgical intervention for cessation of hemorrhage, and six (16.2%) patients died during their hospitalization after technically successful embolization. Nine (24.3%) patients experienced in-hospital rehemorrhage, and of these, five (55.6%) died. Risk factors for rehemorrhage included intra-abdominal malignancy (P < 0.05), transfusion requirement greater than 10 units before angiography (P < 0.05), and the source of hemorrhage other than solitary gastroduodenal artery hemorrhage (P < 0.05). The failure of initial embolization was associated with an increased incidence of death (55.6% vs 5.0%; P < 0.05) and operative intervention to cease hemorrhage (P < 0.05). The failure of technically successful mesenteric embolization is not uncommon and is associated with identifiable risk factors. Risk factor awareness should assist in patient selection for and timing of mesenteric embolization.