DISCUSSION
The present study showed that the most common indication for PPH was placental invasion anomalies. The planned peripartum hysterectomies provided lower morbidities and better neonatal outcomes compared to emergent cases whereas the emergent peripartum hysterectomies needed higher blood product requirements.
The incidence of PPH varies in a wide range. In a large-scaled meta-analysis, the incidence was found as 0.9 per 1000 deliveries6. A retrospective cohort study from Pakistan showed higher incidence as 4.01 per 1000 deliveries12. We have also observed PPH yielding an incidence of 1.7 per 1000 deliveries in a previous study from our tertiary center13. There are several studies that assessed the peripartum hysterectomies and the incidence may change among countries and centers whether they have sufficient antenatal care for pregnancies. In some studies from Turkey, the incidence of PPH was established between 0.3 and 5.38 per 1000 deliveries7,14-17. Sharma et al. found much higher incidence as 6.9 per 1000 deliveries3. Our incidence was similar to the previous literature and accordance with another Turkish studies. The most common indication was uterine atony in the past7,8. However, the main indication has been evolved over recent years from uterine atony to abnormal placentation18. The rising cesarean delivery rates may result to the placental pathologies so it increases the rates of peripartum hysterectomy6,19-21. Van den Akker et al. evaluated approximately 8 million deliveries and they reported that placental abnormalities were the most common indication for PPH and followed by uterine atony6. In a recent study, Kazi found that emergent peripartum hysterectomy was performed for hemorrhage primarily due to uterine atony12. Senturk et al. put forward that the incidence of PPH was higher in the east of Turkey and the main indication was uterine atony and rupture17. The increasing use of uterotonics and cesarean rate may explain the conversion of main indication from uterine atony to abnormal invasive placentation for PPH3,6,19,22-24. The morbidly adherence of placenta become prominent as an indication especially in planned cesarean hysterectomy3. Briery et al. reported that uterine atony was responsible in over half of the patients for emergent peripartum hysterectomy and placenta accreata was the second frequent indication in this group11. In a retrospective study, Sharma et al. showed that placenta accreata was found in all of elective peripartum hysterectomy patients3. We found similar results in accordance to the recent literature. The cesarean rate has incrased over the years and this rate was observed as 63.6% in our study population. Therefore, the placental abnormalities consisted of 56% of patients. We have also determined that the reason of peripartum hysterectomy was only abnormal placental pathologies in planned hysterectomy group. We have performed total abdominal hysterectomy(63.6%) for patients and there was no significant differences between emergent and planned hysterectomy group in terms of surgical type. Subtotal hysterectomy is more desirable for surgeons because removal of the cervix may be difficult due to possible dilated cervix in the peripartum hysterectomy cases. Total abdominal hysterectomy was performed more frequently in our study. Some studies demonstrated that subtotal abdominal hysterectomy is more suitable especially in placental invasion abnormalities, and the morbidities was lower than total abdominal hysterectomy cases7,20. However, some researchers proposed to make total abdominal hysterectomy if the patient status is in good condition and they indicated that total abdominal hysterectomy should be considered to prevent hemorrhage from the cervix5,8.
It was established that intra-operative bleeding was higher in emergent peripartum hysterectomy group compared to scheduled cases8,11. In a recent prospective-cohort study, Seoud et al. have observed lower intra-operative bleeding in elective surgery group and they also found that lower blood prodcuts were transfused in the elective cases25. In parallel with the higher blood loss, the transfused blood products increase in peripartum hysterectomies. In our study, we observed that red blood cell transfusion was given to all of the patients in the emergent group, although 83.9% of the patients needed to get transfusion in the planned surgery group(p=0.014). We have also determined that lower fresh frozen plasma transfusion was required in the planned hysterectomy cases. Wei et al. showed the red blood cell transfusion with a rate of 95%26. Sak et al. found that the red blood cell was transfused to 62.2% of the placenta accreata patients27. Briery et al. compared the transfusion of red blood cell between emergent and planned cesarean hysterectomies and they observed 66% vs. 33% transfusion rate with a mean transfused units of 4.5 vs 1.6, respectively(p<0.05)11. A prospective-cohort study has also found that elective surgery was associated with lower blood transfusion rate compared to emergent cases25. In another retrospective study, authors have demonstrated lower post-operative hemoglobin values in the emergent surgery but it did not reach any significance(7.8±1.6 vs. 8.9±2.2, p=0.08)3. Our study also showed significanly lower hemoglobin levels in the emergent perpiartum hysterectomy group. The transfused units of red blood cell and fresh frozen plasma were higher in the emergent group. Similar to our study, Seoud et al. established that the transfusion rate and mean transfused units were higher in the emergent casses25. We have also analyzed the difference between pre-operative and post-operative hemoglobin levels and it revealed lower differences in the planned surgery compared to emergent cases. The higher complication rate is expected in the emergent cases than scheduled surgery. The bladder injury that is the most common complication was observed as 3-20% in several studies3,7,12,17,23. A higher incidence of bladder injury(27.2%) was stated in our study in reference to the literature. The planned hysterectomy groups had higher bladder injury rate than emergent group but not significantly. We thought that the higher rate may be related to the higher incidence of abnormal placental invasion in planned surgery group. Briery et al. showed higher incidence of post-operative complications in the emergent cesarean hysterectomy11. Two studies 11,25have established that the number of hospital stays were similar between groups but Pettit et al. have found fewer hospital duration in the planned surgery28. We have also reported slightly higher hospital duration in emergent hysterectomy group, but it did not differ significantly similar to the literature findings.
The neonatal outcomes are important in the peripartum hysterectomies. In emergent situations, these outcomes may be affected negatively, so we can improve the neonatal outcomes by performing planned peripartum hysterectomies in selected patients with proper timing. Seoud et al. demonstrated similar birth weight and apgar scores among elective and emergent cases25. Pettit et al. also compared the neonatal outcomes and showed similar apgar scores between emergent and planned surgeries28. Otherwise, they found higher gestational weeks and birth weight in the planned hysterectomy. Briery et al. obtained that the patients underwent planned cesarean hysterectomy had higher gestational weeks, higher fetal birth weight and apgar scores compared to emergent group, but not significantly11. On the contrary, we have observed that the gestational weeks, and birth weight were higher in the emergent peripartum hysterectomy cases significantly. We also reported that the apgar scores were significantly higher in the planned surgery although the gestational week and birth weight were lower than the emergent surgery group. It may be associated that we administered antenatal corticosteroids to all of the planned surgery prior to the delivery.
The emergent peripartum hysterectomy is a life-saving procedure but it has some post-operative problems. Thus, the planned peripartum hysterectomy may improve the maternal and neonatal outcomes and decreases the complication rates. The prenatal diagnosis of the abnormal placental invasion become significant for performing scheduled surgery in these cases. One third of the placental accreata cases that antenatally diagnosed still delivered in an unplanned manner28. We think that it is not possible to avoid the emergent cases completely. The ideal delivery time for suspected abnormal placentation cases is still controversial. There was more optimal outcomes in the placenta accreata cases that delivered at the 34 gestational weeks29. American College of Obstetrics and Gynecology(ACOG) recently recommended delivery time at 34 weeks to 35 weeks 6 days especially in suspected case for placenta accreata30. ACOG also suggests to make the deliveries of placenta accreata cases with expert team in a tertiary center10.
The main limitation of our study was the retrospective nature. The datas of the study population covers a very wide time interval, so the uterine atony was more in the old datas and placental pathologies were more prominent in the recent datas. It might make a selection bias for our study. However, the present study comprehensively compared the emergent and planned peripartum hysterectomies and the sample size of the study was enough according to the previous studies. One of the limitations is the lack of information about the expertise level of surgeons in both study groups. Another important limitation is that we compared the different indications between emergent and planned surgeries such as uterine atony and placental invasion anomalies; since in uterine atony, anatomy is not distorted than placental invasion anomalies.
In conclusion; we showed that antenatally planning of the peripartum hysterectomy improved the maternal and neonatal outcomes. Prenatal diagnosis of suspected cases provide some modifications to the surgeons during the surgery. According to the above ACOG recommendations, we make an effort to diagnose the suspected cases antenatally and we also currently perform the planned peripartum hysterectomies at 34 weeks to 35 weeks 6 days with an expert multi-disciplinary team. Further prospective studies are needed to investigate the correlation of planned peripartum hysterectomy with perinatal outcomes.