Spontaneous haemoperitoneum in the second trimester of pregnancy

This is a case of a spontaneous haemoperitoneum occurring in the second trimester of pregnancy which was managed with interventional radiology to avoid laparotomy and its potential consequences. We aim to raise awareness of this condition in pregnancy because the perinatal mortality rate is as high as 36%. Spontaneous haemoperitoneum in pregnancy (SHiP) has frequently been associated with vascular rupture from pre-existing endometriosis. Most cases of SHiP have been managed with laparotomy. However, transcatheter embolisation can impart lifesaving alternatives to more invasive interventions when caring for pregnant patients. More judicious use of imaging procedures may also help improve diagnostic and therapeutic pathways with SHiP. We recommend that high-risk pregnancies are managed in level IV regional perinatal healthcare centres, when possible, where subspecialists and alternative measures of management exist.

Pregnancy, Materno-fetal medicine, Radiology, Interventional radiology, Urinary and genital tract disorders

Spontaneous haemoperitoneum in pregnancy (SHiP) is a rare and life-threatening complication, characterised by non-traumatic intraperitoneal bleeding occurring from 6 weeks of gestation to 30–42 days post partum. While establishing a national incidence is challenging, a study across 9 regions in Italy reported 0.04 cases per 1000 births. The majority of cases (68.9%) occur in the third trimester, with identified risk factors including maternal age over 34, pregnancies involving assisted reproductive technology, multiple pregnancies, endometriosis and a history of previous abdominal surgeries. Symptoms typically present as abdominal or flank pain in approximately 70% of cases, often leading to sudden deterioration and hypovolaemic shock in 18% of instances, potentially causing fetal distress. Perinatal mortality rates range from 26.9% to 36%. Diagnosing SHiP is complicated by physiological changes in pregnancy, and concerns about radiation exposure often lead to hesitation in performing imaging studies.1–6 Despite improvements in maternal mortality, perinatal morbidity and mortality associated with SHiP persist. Early diagnosis and treatment have historically involved vertical midline laparotomy in 85% of cases. Suspected causes of spontaneous intraperitoneal haemorrhage include the rupture of abnormal vasculature, with endometriosis playing a significant role (27%–57% of cases), often undiagnosed prior to presentation. Chronic inflammation, adhesions and decidualisation associated with endometriosis contribute to blood vessel formation, potentially leading to intraperitoneal bleeding during pregnancy. Adhesive disease resulting from endometriosis, combined with substantial uterine growth in the second trimester, increases tension on adhesions, possibly causing the rupture of aberrant vasculature. This tension is thought to be a consequence of the rapid expansion of the uterus, frequently affecting the posterior uterine surfaces and parametrium, with bleeding sites predominantly in the left pelvis (75%).1–6 Case presentation The patient was a multiparous patient in her 30s, with a history of two previous pregnancies resulting in two live births and no miscarriages. She was pregnant at 17 weeks and 2 days. She sought medical attention in the emergency department due to a sudden onset of chest and left shoulder pain that radiated to her back. She had abdominal pain that radiated to her shoulder. Pain was associated with non-bloody, non-bilious emesis. Pain was intermittent but 10/10 in severity when episodes occurred. She denied fever, chills and shortness of breath. She reported no recent heavy lifting. She denied pelvic pain, vaginal bleeding or leakage of amniotic fluid. She had no recent trauma to the abdomen. Her pregnancy had otherwise been uncomplicated. Her medical, surgical and social history were unremarkable. She had no medical history of thrombosis or endometriosis. The family history was unremarkable. Her obstetrical history was unremarkable with both prior pregnancies ending in vaginal deliveries at term. Her vital signs were pertinent only for tachycardia to 110 beats per minute. The patient’s body mass index was 28. On physical examination, the patient was in acute distress and appeared uncomfortable. She had dry mucous membranes, mild tenderness to palpation at the anterior chest wall, epigastric pain on palpation and a positive Kehr sign (pain referred to the left shoulder with palpation of the abdomen). Investigations In the emergency department, investigations revealed a white cell count of 15.7×103/µL, haemoglobin 12.6 g/L, haematocrit 36.8% and platelet count 328×103/μL. A metabolic panel showed no electrolyte abnormalities, normal kidney function, bilirubin and liver transaminases. Other normal findings included lipase (10 U/L), troponin I (10 pg/mL) and a negative urinalysis. An ECG showed sinus rhythm at a rate of 110 beats per minute. A chest X-ray revealed no cardiopulmonary issues, and a limited cardiac ultrasound showed no structural anomalies. Chest CT angiography was initially recommended by the emergency room physicians due to the patient’s complaint of chest pain. This was done to assess for venous thromboembolism, which is a leading cause of maternal mortality in the USA. Although no evidence of pulmonary embolism was found, abdominal CT revealed moderate haemoperitoneum. There was a small volume of perihepatic and perisplenic fluid noted. The perihepatic fluid demonstrated low to intermediate density. There was a slightly lobulated collection of intermediate-attenuation material in the upper abdomen, adjacent to the stomach and spleen, extending into the left abdominal mesentery. This material measured up to 11.2×6.8 cm (see figures 1–5). Ultrasound confirmed a viable intrauterine pregnancy. Emergency consultations with general surgery, interventional radiology and obstetrics were initiated for management. Differential diagnosis In pregnant women experiencing epigastric and shoulder pain, differential diagnoses include complications such as placental abruption, ruptured haemorrhagic ovarian cyst, acute fatty liver of pregnancy and rupture of a rudimentary uterine horn. Later in pregnancy, preeclampsia, particularly with hemolysis, elevated liver enzymes, and low platelets (HELLP syndrome) or liver rupture, is also a consideration.7 Complications outside of pregnancy, such as hepatic conditions (adenomas, haemangiomas, etc) and splenic conditions (ruptured spleen from infectious aetiologies), may contribute to haemoperitoneum in pregnancy. Vascular pathologies like arterial aneurysms, mycotic aneurysms, and varices are also potential causes.8 The differential diagnosis for epigastric pain encompasses various conditions such as costochondritis, Mallory-Weiss tears, myocardial infarction and pulmonary embolism. Aortic dissection, although rare in pregnancy, should be considered in patients with underlying aortopathic conditions. Concomitant nausea and vomiting may lead to contemplation of cholecystitis, appendicitis, pancreatitis and perforated viscus.7 Imaging in patients with intraperitoneal blood may reveal spontaneous rupture of structures like uterine veins, liver, external iliac vessels, renal hamartoma, uterus, ovarian artery, hepatic or splenic blood vessels or a pedunculated uterine leiomyoma.1 5 Subacute abdominal or flank pain, free peritoneal fluid, Kehr’s sign and decreasing haemoglobin levels are common features in SHiP.1–3 When a pregnant patient presents with acute abdominal pain, especially in the late second or third trimester, consideration should be given to SHiP, particularly if free fluid is noted on imaging along with decreasing haemoglobin.2 9 Pulmonary embolism was a primary consideration for our patient. A chest CT angiogram revealed no pulmonary embolism but showed a haemoperitoneum, prompting subsequent abdominal and pelvic CT scans. Interventional radiology conducted an arterial angiogram, assessing the coeliac, superior mesenteric artery and inferior mesenteric arteries (refer to figures 6–8). No active bleeding, visceral aneurysm or pseudoaneurysm was detected. Despite planned vascular embolisation, no intervention was required. The patient was hospitalised for serial abdominal exams and haemoglobin monitoring, remaining haemodynamically stable. Haemoglobin levels stabilised after reaching a nadir of 8.8 g/L on the morning of hospital day 2. Symptomatic relief was achieved with antiemetics and narcotics. By hospital day 2, the patient was clinically stable for discharge with scheduled outpatient follow-up. Outcome and follow-up At the patient’s follow-up clinic appointment, she reported she had been doing well since discharge and was without recurrence of concerning symptoms or other complaints. The remainder of her pregnancy course was uncomplicated. Routine anatomy ultrasound was done at about 20 weeks of gestation and showed a fetus with normal anatomy. She presented in spontaneous labour at 39 weeks and 3 days of gestation. Her labour course was complicated by a precipitous delivery of a 4310 g live male infant with Apgar scores of 8 and 9. The placenta was abnormally adherent to the lower uterine segment and required manual extraction and uterine curettage. Both the patient and the infant were discharged home without complications. She and the baby did well post partum. She decided on a progesterone intrauterine system for contraception which was placed at her 6-week postpartum visit.

A review of 59 cases published in 2017 revealed the seriousness and adverse pregnancy complications associated with a spontaneous intraperitoneal haemorrhage in pregnancy. Only two cases continued until term, and 93% of the reported cases were treated with laparotomy.5 The maternal death rate was 1.7%, and the risk of fetal or neonatal death approached 27%. Over half of these cases occurred in the third trimester of pregnancy. Almost 76% required blood transfusions. Two women died, and one required a hysterectomy.5 Free intra-abdominal fluid was noted by imaging in almost 63% of cases, but due to the concern about potential adverse fetal effects of ionising radiation, imaging was not always immediately completed.2 3 It has been proposed that more liberal use of imaging procedures could decrease the incidence of unnecessary laparotomies and adverse outcomes with SHiP.5 Ultrasound is the first line of detection of SHiP, but CT scan and MRI may also be used if time permits, to visualise any vascular lesions that may be the cause of bleeding.1 3 Concerns arise regarding fetal exposure to ionising radiation. When selecting diagnostic imaging for pregnant patients, clinicians must thoroughly evaluate the associated risks. This decision should involve collaboration with the patient, considering the clinical scenario. Balancing the potential adverse effects of radiation against the risks of not conducting the necessary procedure and its potential benefits is essential. Considerations when performing interventional radiologic procedures on pregnant women include the difficulty with the interpretation of imaging due to physiological changes in tissue from pregnancy as well as the enlarging uterus. The proliferating glands in the maternal pregnant breasts are more sensitive to radiation; thus, increased exposure may increase the risk of breast cancer. When the fetus is in the direct beam, tailoring the exam to reduce the dose of radiation exposure by limiting fluoroscopic time, collimating the beam and reducing the field of view to only the very area of interest may be considered. Using wide pitch and narrow collimation can reduce the radiation dose. Many institutions have specific protocols that are followed when imaging pregnant patients to reduce radiation exposure to both the mother and the fetus.10 According to the American College of Radiology and Society for Pediatric Radiology Guidelines, radiation exposure below 50 milligray (mGy) from the 11th to the 17th week of gestation is not associated with significant detrimental effects. Exposure between 50 mGy and 100 mGy is uncertain but likely remains insignificant. Levels exceeding 100 mGy present a negligible risk of diminished IQ or mental retardation, with risks increasing with increasing doses. After 15 weeks of gestation, the primary fetal risk of radiation exposure is potential malignancy, with a 0.45% increase in leukaemia risk at 50 mGy exposure. A single-phase abdominal and pelvic CT scan delivers up to 20 mGy, unlikely to cause fetal tissue effects. A chest CT angiogram delivers 5–10 mGy, and an abdominal and pelvic angiogram delivers 1–5 mGy. It is important to remember that radiation exposure can vary significantly depending on several factors, including the specific protocols used, the equipment and the individual patient characteristics. In our patient, she likely received 16–35 mGy of total radiation with her imaging. In the process of conducting radiologic studies, minimising radiation requires either reducing the number of images or limiting the phases involved. While this approach may have an impact on image quality, it is done to uphold diagnostic capability, ensuring that despite the adjustments made to reduce radiation, the medical images remain sufficiently clear and informative for accurate diagnosis.11 12 While our patient did not exhibit any prior symptoms of endometriosis, we suggest that her undiagnosed endometriosis may have contributed to the haemoperitoneum. This insight could influence future discussions about potential bleeding risks in subsequent pregnancies. The mechanism behind the spontaneous stabilisation and resolution of bleeding in our case remains unclear, but the body’s natural processes, such as clot formation and vasoconstriction, may have contributed. Further research is needed to investigate the correlation between the endometriosis stage and the severity of SHiP, aiding in identifying high-risk patients and understanding potential complications. The management of SHiP depends on the patient’s clinical presentation, haemodynamic status, gestational age, the extent of intraperitoneal haemorrhage and the underlying cause of the bleeding. In the third trimester, caesarean delivery has been deemed indicated, and maternal resuscitation with blood products is necessary in many cases. Haemostatic sutures, electrocoagulation, haemostatic agents and compression of the area of bleeding can help intraoperative management of bleeding.1 13 In up to 9% of reported cases, a hysterectomy has been required.3 4 In an article by Vuong et al, emergent laparotomy was done in two cases of spontaneous intraperitoneal haemorrhage in pregnancy. One patient had endometriosis, and one had a previous ovarian surgery. A literature review found the mortality rate to be approaching 27%. Only 15.6% of cases successfully continued the pregnancy when haemorrhage occurred in the second trimester. Endometriosis was present in 55.9% of patients. Many patients went on to develop preterm labour and delivery, with some getting infectious peritonitis.1 In a 2017 systematic review by Lier et al, surgical intervention was deemed necessary in over 76% of cases of SHiP. Maternal reasons accounted for 69.6%, fetal distress for 3.6% and a combination of both for 26.8% of interventions.3 A 2022 article by Mazzocco MI et al detailed 29 cases of spontaneous haemoperitoneum in France, occurring at a mean gestational age of 24 weeks. Treatment modalities included laparotomy (64%), laparoscopy (three cases) and embolisation (one case). The study reported a 36% fetal non-survival rate, with one term case requiring emergent hysterectomy. Typical blood loss ranged from 400 mL to 4000 mL, and some instances, such as twins at 32 weeks, reported 5 L of haemoperitoneum.5 14 Recurrent intraperitoneal bleeding has occurred in some cases, with one patient experiencing three episodes in the same pregnancy, ending with a caesarean section at 25+ weeks due to intrauterine infection after multiple bleeds of 950 cc, 1500 cc, 500 cc and 800 cc.4 Managing SHiP involves various approaches and subspecialists. The primary interventions include laparotomy, undertaken in 64%–93% of cases, and caesarean delivery, performed in 92.6% of cases. However, it is important to note that both procedures carry the potential risk of contributing to additional maternal and neonatal morbidity.4 Laparoscopy is attempted in approximately 10% of cases, but half of these instances require conversion to laparotomy.3 Despite these challenges, alternative treatments have been explored. In the care of pregnant patients, interventional radiology has recently assumed a unique and invaluable role. Successful cases of treating uterine artery aneurysms and embolisation of splenic and renal artery aneurysms during pregnancy have been reported.15 One specific case exemplifies this approach’s success. In a scenario where a ruptured right uterine artery aneurysm and severe anaemia presented in a 22-week pregnancy, the patient underwent embolisation of the right uterine artery. A healthy baby was delivered by caesarean section 12 weeks later. This highlights the potential effectiveness of uterine artery embolisation in specific cases of SHiP.3 The decision regarding the management of spontaneous intraperitoneal haemorrhage in pregnancy involves a comprehensive evaluation of various factors. Angiography, serving both diagnostic and therapeutic purposes, should be considered when clinically feasible for suspected intraperitoneal bleeding.8 Haemodynamic stability plays a pivotal role, with conservative management being considered when the pregnant patient is stable, without ongoing active bleeding, while intervention is warranted in cases of haemodynamic instability marked by symptoms like tachycardia, hypotension or significant bleeding. The source and severity of bleeding also influence the choice, favouring conservative measures for minor or self-limiting bleeding and prompting intervention for active and significant bleeding amenable to procedures like embolisation. Gestational age considerations may lead to a preference for conservative management in early gestation, while intervention may be considered if gestational age allows for preserving the pregnancy. Maternal and fetal status, response to initial measures, diagnostic imaging findings, patient preferences and the availability of resources further guide decision-making. The collaboration of obstetricians, interventional radiologists and surgeons in a multidisciplinary approach is crucial, ensuring individualised care tailored to the specific clinical scenario for optimal outcomes for both the mother and the fetus. Additionally, the limited availability of interventional radiology resources in certain locations may impact the feasibility of embolisation as a treatment option. The therapeutic effectiveness of interventional radiology (IR) is constrained when addressing SHiP originating from specific sites like the uterine wall, broad ligament and adnexa. The intricacies of the vascular anatomy in these areas pose challenges for selective embolisation, especially in cases of diffuse or multiple small-vessel bleeding. Limited accessibility, proximity to critical structures and the potential involvement of complex vascular networks further contribute to the limitations of IR. Additionally, interventions near the uterine arteries or adnexa may jeopardise fetal blood supply, requiring a delicate balance between haemostasis and preserving fetal well-being. Operator experience is crucial in navigating the complex vascular terrain, and despite successful embolisation, the risk of recurrent bleeding remains, particularly if underlying gynaecological pathology persists. While interventional radiology is valuable in managing spontaneous haemoperitoneum, its limitations underscore the necessity for a multidisciplinary approach. Collaboration between surgeons and interventional radiologists becomes paramount to determine the most suitable treatment strategy based on the specific clinical scenario, emphasising the need for haemostasis while preserving maternal and fetal well-being.1 In challenging anatomical sites, surgical intervention may be required for definitive bleeding control. Interventional radiology’s role in diagnosing and treating SHiP hinges on specific indications. Haemodynamic instability, when conventional methods fail, and the need for localisation of the bleeding source through angiography are primary indications. However, the success of interventional radiology is operator dependent, emphasising the importance of experience and skill. Furthermore, its efficacy may be limited when the cause of haemoperitoneum is unclear or multifactorial, necessitating exploratory laparotomy in such instances. The evidence supporting interventional radiology for SHiP is largely anecdotal, relying on case reports and small case series. Although angiographic embolisation appears effective in controlling bleeding and preserving pregnancies, there are currently no large studies or randomised controlled trials due to the condition’s rarity. Success rates vary based on bleeding aetiology, gestational age and the interventional radiology team’s expertise. In conclusion, while interventional radiology is becoming integral in managing spontaneous haemoperitoneum, the need for further evidence highlights the importance of individualised decision-making, carefully weighing risks and benefits for both mother and fetus. Learning points. Spontaneous haemoperitoneum in pregnancy (SHiP) is often not initially suspected but should be considered, especially in the third trimester, when a pregnant patient presents with sudden and severe abdominal and shoulder pain. This condition is more prevalent in patients with endometriosis and is associated with high perinatal morbidity and mortality. Opting for vascular embolisation procedures to address SHiP may present a safer and less invasive alternative for some patients, in contrast to exploratory laparotomy. Whenever feasible, high-risk pregnancies should be overseen in level IV regional perinatal healthcare centres, ensuring the timely availability of a diverse range of subspecialists. In cases of patients with SHiP, interventional radiology embolisation procedures when available should be prioritised as the primary alternative to exploratory laparotomy. This approach aims to minimise maternal and perinatal morbidity and mortality. Footnotes Contributors: The following authors were responsible for drafting of the text, sourcing and editing of clinical images, investigation results, drawing original diagrams and algorithms, and critical revision for important intellectual content: KC and SM. The following authors gave final approval of the manuscript: KC and SM. Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy. Competing interests: None declared. Provenance and peer review: Not commissioned; externally peer reviewed. Ethics statements Patient consent for publication Consent obtained directly from patient(s).

- 1.Vuong ADB, Pham TH, Nguyen XT, et al. Spontaneous hemoperitoneum in the second and third trimester of pregnancy: two uncommon case reports at Tu Du Hospital, in Vietnam and a literature review. Int J Emerg Med 2023;16:26. 10.1186/s12245-023-00498-w [DOI] [PMC free article] [PubMed] [Google Scholar] - 2.Huang LY, Hsu PY, Chiang CT, et al. Endometriosis-related spontaneous hemoperitoneum in the early second trimester: a case report. Taiwan J Obstet Gynecol 2021;60:328–30. 10.1016/j.tjog.2021.01.032 [DOI] [PubMed] [Google Scholar] - 3.Lier MCI, Malik RF, Ket JCF, et al. Spontaneous hemoperitoneum in pregnancy (SHiP) and endometriosis - a systematic review of the recent literature. Eur J Obstet Gynecol Reprod Biol 2017;219:57–65. 10.1016/j.ejogrb.2017.10.012 [DOI] [PubMed] [Google Scholar] - 4.Golfier F, Pleynet L, Bolze PA. Spontaneous hemoperitoneum in pregnancy: a life-threatening maternal and fetal complication of endometriosis. J Gynecol Obstet Hum Reprod 2022;51:102415. 10.1016/j.jogoh.2022.102415 [DOI] [PubMed] [Google Scholar] - 5.Mazzocco MI, Donati S, Maraschini A, et al. Spontaneous hemoperitoneum in pregnancy: Italian prospective population-based cohort study. Acta Obstet Gynecol Scand 2022;101:1220–6. 10.1111/aogs.14431 [DOI] [PMC free article] [PubMed] [Google Scholar] - 6.Brosens IA, Fusi L, Brosens JJ. Endometriosis is a risk factor for spontaneous hemoperitoneum during pregnancy. Fertil Steril 2009;92:1243–5. 10.1016/j.fertnstert.2009.03.091 [DOI] [PubMed] [Google Scholar] - 7.Braverman AC, Mittauer E, Harris KM, et al. Clinical features and outcomes of pregnancy-related acute aortic dissection. JAMA Cardiol 2021;6:58–66. 10.1001/jamacardio.2020.4876 [DOI] [PMC free article] [PubMed] [Google Scholar] - 8.Escobar Vidarte MF, Montes D, Pérez A, et al. Hepatic rupture associated with preeclampsia, report of three cases and literature review. J Matern Fetal Neonatal Med 2019;32:2767–73. 10.1080/14767058.2018.1446209 [DOI] [PubMed] [Google Scholar] - 9.Inoue T, Moriwaki T, Niki I. Endometriosis and spontaneous rupture of utero-ovarian vessels during pregnancy. Lancet 1992;340:240–1. 10.1016/0140-6736(92)90506-x [DOI] [PubMed] [Google Scholar] - 10.Kasotakis G. Spontaneous hemoperitoneum. Surg Clin North Am 2014;94:65–9. 10.1016/j.suc.2013.10.006 [DOI] [PubMed] [Google Scholar] - 11.Sensakovic WF, Royall I, Hough M, et al. Fetal Dosimetry at CT: A Primer. Radiographics 2020;40:1061–70. 10.1148/rg.2020190166 [DOI] [PubMed] [Google Scholar] - 12.American college of Radiology and society of pediatric Radiology; 2023. Available: https://www.acr.org/-/media/ACR/Files/Practice-Parameters/Pregnant-Pts.pdf - 13.Shalaby MA, Maged AM, Al-Asmar A, et al. Safety and efficacy of preoperative tranexamic acid in reducing intraoperative and postoperative blood loss in high-risk women undergoing cesarean delivery: a randomized controlled trial[201]. BMC Pregnancy Childbirth 2022;22:201. 10.1186/s12884-022-04530-4 [DOI] [PMC free article] [PubMed] [Google Scholar] - 14.Rafi J, Mahindrakar G, Mukhopadhyay D. Endometriosis nodule causing spontaneous haemoperitoneum in pregnancy: a case report and literature review. Case Rep Obstet Gynecol 2017;2017:3480287. 10.1155/2017/3480287 [DOI] [PMC free article] [PubMed] [Google Scholar] - 15.Moirano J, Khoury J, Yeisley C, et al. Interventional radiology and pregnancy: from conception through delivery and beyond. Radiographics 2023;43:e230029. 10.1148/rg.230029 [DOI] [PubMed] [Google Scholar]