Main findings and interpretation
The first abdominal pregnancy was reported in 1708 as an autopsy finding and many cases have been reported worldwide ever since (3). While in a primary abdominal pregnancy the fertilized ovum implants directly in the abdominal cavity (10), in a secondary abdominal pregnancy the embryo or fetus continues to grow in the abdominal cavity after explusion from its primary site of development (11). There are several theories about the pathophysiology of an abdominal pregnancy. Berghella V et al hypothesised that an omental pregnancy is secondary to a primary tubal or ovarian pregnancy that subsequently implants on the omentum (12), while Cavanagh stated that fertilization may occur in the posterior cul-de-sac where sperms accumulate and the ovum is transported via dependent flow of peritoneal fluid (13). Paternoster DM et al suggested delayed ovulation associated with a reversed tubal course by retrograde menstrual flow as a possible mechanism and further emphasised the role of intraperitoneal fluid flow in carrying the zygote from the cul-de-sac to different intraperitoneal sites (14). Dmowski et al and Hall et al postulated a spontaneous migration of the embryo from the uterus to the retroperitoneal space along lymphatic channels (15,16). Abdominal pregnancies post hysterectomy have been explained by migration of the fertilised ovum via fistulous tracts (17).
Though the ectopic pregnancy rates are on an incline, the risk of abdominal pregnancies, with ruptured ectopic pregnancy among one of its many causes, is on a declining trend due to improved antenatal care and diagnostic methods (11). Despite the availability of defined criteria by Studdiford, its diagnosis proves challenging and requires a high degree of suspicion. In addition to its vague non-specific clinical presentation, initial misdiagnosed ultrasound reports may interfere with timely diagnosis, like in our case. The most common physical examination findings include abdominal tenderness (100%), abnormal fetal lie (70%), easily palpable fetal parts and a displaced cervix (9). Our patient presented with symptoms of generalised pricking abdominal pain, mild chest pain and spotting per vagina at 13 weeks of gestation. This vague presentation was misleading and resulted in a delay in diagnosis till evaluation at our centre. More often than not a transvaginal ultrasound confirms the diagnosis but in case of a dilemma, MRI can provide further clarity regarding the diagnosis.
Gestational age at diagnosis and maternal hemodynamic stability are the main factors dictating management. The two main modalities of treatment are medical and surgical, surgical being more common as patients usually present as an acute emergency with intra-abdominal bleeding (4). The choice between laparoscopy and laparotomy depends on multiple factors including period of gestation at diagnosis, maternal hemodynamic stability and availability of resources and skill (18). With increasing early ultrasonography, ectopic pregnancies are being identified prior to complications like rupture, providing an opportunity to attempt medical management with KCl or methotrexate individualised for each case (4). Less commonly used agents include hyperosmolar glucose, prostaglandins, danazol, etoposide and mifepristone (18).
The management of the placenta ranges from complete or partial removal of the placenta to leaving the complete placenta in situ (18). The use of adjuvant methotrexate for the placental remnants is controversial. While some believe that it helps accelerate the involution of the placenta others say that rapid degeneration results in higher amounts of necrotic tissue accumulation forming a nidus for infection thereby increasing sepsis rates (18). The use of methotrexate at advanced gestation is debatable, as it works best on rapidly dividing cells and is less likely to help in case of a mature placenta (19). We did not administer methotrexate to our patient as very minimal placental tissue <1 cm was left behind.
Abdominal pregnancy, especially when advanced, is known to be associated with high maternal and neonatal morbidity and mortality (18). The maternal mortality rate ranges from 0.5-20% (20). Severe hemorrhage, perforation, bowel obstruction, fistula, disseminated intra-vascular coagulation and sepsis contribute to the increased morbidity and mortality. Our patient was transfused with 3 units of packed red blood cells and 2 units of fresh frozen plasma and did not have any other post operative morbidity. Perinatal mortality rates of 40%-95% have been reported and 21-90% of surviving fetuses have serious birth defects due to anhydramnios and subsequent compression and also vascular disruption (20). Surviving neonates frequently develop torticollis, flattening of the head, facial or cranial asymmetry, thoracic malformations, limb defects, joint abnormalities and CNS malformations (9,20).
In a systematic review done by Poole A et al on 225 abdominal pregnancies diagnosed/treated at less than 20 weeks of gestation, from 1965 to 2009, 8% of the early abdominal ectopics occurred with an intrauterine device in situ. Mean gestational age at treatment was 10 weeks. The commonest sites of early abdominal ectopics were pouches around the uterus (24.3%), serosal surface of uterus and tubes (23.9%) and other sites (12.8%). Primary surgical management was done in 208 cases (87.8%). Non-surgical adjuvant or primary therapy included intramuscular methotrexate, intralesional methotrexate, intracardiac KCl and uterine artery embolization. The overall maternal mortality was 3% (4).
Nkusu ND et al reviewed 163 cases of advanced abdominal pregnancy after 20 weeks of gestation from 1946 to 2008. Forty five percent cases were diagnosed preoperatively. Fetal/ perinatal mortality was 72% and pressure deformities were common among survivors. Maternal postoperative complications like hemorrhage and infection were noted in 55%, transfusion of blood products in 80%. Maternal mortality was 12% (20).
In the current review of 314 cases of early and advanced abdominal pregnancy by various authors from 1930-2021, we observed a maternal morbidity in terms of transfusion of blood and blood products in 74 cases (23.6%) and maternal deaths in 15 cases (4.8%). The maternal mortality rate we noted in our review is marginally higher than the systematic review done by Poole A et al(4) but is significantly lower compared to the rate quoted by Nkusu ND (12%)(20). This difference could be attributed to the period of gestation used as inclusion criteria. Early versus advanced abdominal pregnancy is an important factor influencing the varied maternal mortality and morbidity rates reported by different authors and higher rates have been reported at advanced gestations. We also noted that 93.9% of the cases were surgically managed and 6% required hysterectomy. These rates are comparable to that noted by Poole A et al. In addition, multiple other factors may have contributed to the reduction in maternal mortality rates over the years like early diagnosis, advanced imaging modalities, increased patient and physician awareness, improved surgical and non-surgical techniques, blood product availability and better access to tertiary care centres over the years. Regarding the overall neonatal outcome, 8.6% were live born, 76.8% were fetal loses <20 weeks of gestation, 14.6% were IUFD/ stillborn and 1.9% were neonatal deaths.