Case
The patient was a 32-year-old woman (gravida 3, para 1) with a spontaneously conceived pregnancy. Her history was significant for stage 4 endometriosis, for which she had had 6 previous laparoscopies for resection of bilateral endometriomas and extensive disease. Her most recent surgery was 5 years prior. She underwent induction of labour at 39 + 1 weeks of gestation for polyhydramnios and a baby large for gestational age. She proceeded to have an uncomplicated ventouse delivery due to a delay in the second stage. Her estimated blood loss at delivery was 400 mL, following active management of the third stage of labour with syntometrine. Following the birth she was discharged to a primary birthing unit.
She was transferred back to the hospital unit six hours later with severe bilateral pain in the shoulder tips and generalised abdominal pain. She was tachycardic, with a heart rate of 116 bpm, and her blood pressure was 138/96 mmHg. Her abdomen was distended and generally tender with rebound tenderness worse in the right iliac fossa, and had normal bowel sounds. A bedside ultrasound scan revealed free fluid in both adnexa and no signs of retained products of conception.
A full blood count demonstrated a decrease in haemoglobin count to 98 g/L from 116 g/L pre-delivery. Her white cell and neutrophil counts were also raised at 17.7x10 9 g/L and 15.6x10 9 g/L, respectively. Coagulation studies and renal and liver function tests were normal. Empiric antibiotics and intravenous fluid resuscitation were administered with an improvement in her heart rate to 88 bpm but persisting shoulder tip and abdominal discomfort. An abdominal x-ray was unremarkable. Her haemoglobin continued to decrease to 90 g/L from 98 g/L over four hours prompting a computerised tomography (CT) scan of the abdomen and pelvis due to concerns of an intra-abdominal bleed.
The CT scan identified a large volume of free fluid in keeping with haemoperitoneum, but with no active bleeding vessel or obvious source ( Fig. 1 , Fig. 2 ). The working diagnosis was spontaneous haemoperitoneum in pregnancy. She was managed conservatively with tranexamic acid and a morphine patient-controlled analgesia (PCA) pump. Her haemoglobin reached a nadir of 77 g/L with secondary symptoms of anaemia, and she received two units of red blood cells and an iron infusion. Fig. 1 CT scan of the abdomen and pelvis demonstrating haemoperitoneum and post-partum uterus. Fig. 1 Fig. 2 CT scan of the abdomen and pelvis demonstrating haemoperitoneum. Fig. 2
CT scan of the abdomen and pelvis demonstrating haemoperitoneum and post-partum uterus.
CT scan of the abdomen and pelvis demonstrating haemoperitoneum.
Tranexamic acid and the PCA were stopped after three days due to improvement in her symptoms. A decision was made to withhold prophylactic enoxaparin during this time due to concerns over her bleeding risk. While haemorrhage requiring blood transfusion and being postpartum were risk factors for venous thromboembolism (VTE), she was a non-smoker with a normal body mass index (BMI) and had no family or personal history of VTE, and it was considered that the risk of bleeding was greater than the risk of VTE.
On day 4 postnatally she developed pleuritic chest tightness, shortness of breath and right calf tenderness. A CT pulmonary angiogram demonstrated left pulmonary emboli extending from the bifurcation of lower lobe segmental bronchi into multiple subsegmental branches, with no evidence of right heart strain ( Fig. 3 ). Lower limb ultrasound showed no evidence of deep-vein thrombosis. She was started on subcuticular enoxaparin (80 mg BD). Six months of anticoagulation was recommended by the haematology team as this was classified as a provoked pulmonary embolism. As the patient was breastfeeding, direct oral anticoagulants were contraindicated, and she continued enoxaparin as she preferred this over the monitoring requirements of warfarin. The enoxaparin dose was 80 mg (1 mg/kg) BD for 4 weeks and then 120 mg (1.5 mg /kg) daily for five months. She was monitored closely as an inpatient for two further days to ensure there were no signs of further intra-abdominal bleeding and discharged home on day 7 postpartum. When reviewed at 6 weeks postpartum, she had made a good recovery and was continuing her enoxaparin treatment. Her inpatient clinical path is summarised in Table 1 . Fig. 3 CT pulmonary angiogram showing left-sided pulmonary emboli. Fig. 3 Table 1 Summary of clinical events. Table 1 Clinical events Haemoglobin g/L Management Day 0 Vacuum birth 4180 g baby boy Apgar scores of 7,9, 10. EBL 400mls 116 5 h postnatal: abdominal & shoulder tip pain 98 IV fluid resuscitation 12 h postnatal: CT abdomen & pelvis: haemoperitoneum with no active bleeding source. 90 IV tranexamic acid Morphine PCA Day 1 Symptomatic of anaemia. Pain controlled with analgesia 77 2 units red blood cells. Iron infusion Day 2 90 Day 3 91 Tranexamic acid stopped Day 4 Chest tightness and shortness of breath. CTPA: pulmonary embolus confirmed Therapeutic enoxaparin commenced Day 5 112 Day 7 Discharge from hospital
CT pulmonary angiogram showing left-sided pulmonary emboli.
Summary of clinical events.
Conclusion
SHiP is a rare condition that has the potential for serious maternal and fetal complications. It should be considered in any pregnant patient presenting with acute abdominal pain, shoulder-tip pain or signs of hypovolaemia, especially if there is a history of endometriosis. This case describes successful conservative management of SHiP in the postpartum period. Decisions regarding anticoagulation use in patients with SHiP are difficult, necessitating the balancing of the risks of both anticoagulation and VTE.
Discussion
SHiP is a rare condition that can be associated with maternal and perinatal morbidity and mortality [ 11 , 12 ]. Most cases of SHiP occur in the third trimester of pregnancy, with 61 % of SHiP occurring before labour, 18 % during and 21 % after labour [ 7 ]. A Japanese case series found that 26 % of cases occurred postpartum [ 13 ]. Antenatal cases pose significant fetal risk, with a systematic review of 59 cases of SHiP finding signs of fetal distress in 40.7 % of cases [ 4 ]. Surgical management with a laparotomy is often employed in these situations. A literature review of 54 cases of SHiP found that 85 % required a laparotomy and 6 % an emergency hysterectomy [ 5 ]. Kato et al. present an antenatal case of SHiP where emergency laparotomy was required for delivery for fetal distress [ 6 ]. The case presented here occurred postpartum, which meant that fetal wellbeing did not need to be considered and supported conservative management with careful monitoring, as employed.
SHiP may present with a variety of non-specific symptoms such as acute abdominal pain, shoulder-tip pain or signs of hypovolaemia or anaemia. This can present a diagnostic challenge, especially if presenting antenatally [ 6 , 14 ]. Due to its rarity, SHiP is often misdiagnosed as a placental abruption or uterine rupture and correct diagnoses are rarely made before emergency laparotomy [ 3 , 15 ]. Vuong et al. report 2 cases of SHiP where the diagnosis was unclear pre-operatively and able to be confirmed only at laparotomy [ 3 ]. As this case occurred postpartum, CT imaging was utilised promptly and safely to assist in the diagnosis.
The patient in this report had had six previous laparoscopies for stage 4 endometriosis, which is consistent with endometriosis being a reported risk factor for SHiP [ 3 , 7 , 11 , 16 ]. The pathophysiology of this is thought to be due to chronic inflammation, vascular proliferation and decidualisation of ectopic endometriotic foci being placed under tension from the growing uterus [ 7 ]. A systematic review by Lier et al. identified that 55.9 % of 59 cases of SHiP were associated with endometriosis [ 4 ]. No correlation has been found between the stage of endometriosis and the likelihood of developing SHiP.
Pregnancies conceived with assisted reproductive technology (ART) also have a higher incidence of SHiP; it is hypothesized that higher hormonal levels associated with ART pregnancies potentially exacerbate the ectopic decidualisation process [ 2 ].
This case report illustrates the challenges in the decisions of prophylactic anticoagulation in patients at high risk of bleeding. The patient's risk of VTE was increased by recent pregnancy, haemorrhage and immobility and it is possible that tranexamic acid also contributed to this risk [ 17 , 18 ]. Despite this, the patient was also at high risk of life-threatening intra-abdominal haemorrhage that could have been exacerbated by anticoagulant use. Conversely, her vaginal birth, normal BMI and absence of a family history of VTE were reassuring as to her VTE risk [ 19 ]. Ultimately there is a paucity of data on SHiP complicated by VTE to adequately guide decision-making in this situation (no other reported cases of concurrent PE and SHiP could be found in the literature).
Optimal management of anticoagulation use for patients at high risk of bleeding is very challenging and careful consideration of the risks should be employed on a case-by-case basis [ 20 ]. An American study evaluating complications of postpartum women receiving therapeutic anticoagulation found a higher incidence of wound and haemorrhagic complications with earlier resumption of anticoagulation [ 21 ]. Usually, when deciding the timing of commencing post-natal VTE prophylaxis, anticoagulation would be commenced once there was certainty of haemostasis [ 22 ]. A disadvantage of the non-surgical management employed in this case was that there was no surgical reassurance of confirmed haemostasis.
Introduction
Spontaneous haemoperitoneum in pregnancy (SHiP) is defined as an intraperitoneal haemorrhage that manifests during pregnancy or postpartum in the absence of trauma, excluding bleeding associated with ectopic pregnancy, uterine rupture or caesarean section [ 1 ]. SHiP is uncommon, with an incidence of 1/25000 [ 2 ]. It has high rates of perinatal mortality [ 1 , 3 , 4 ]. Reported risk factors include pregnancies conceived with assisted reproductive technology and a history of endometriosis [ [4] , [5] , [6] , [7] , [8] ]. A diagnosis of SHiP should be suspected in any pregnant woman presenting with acute abdominal pain, signs of hypovolaemia or decreasing haemoglobin levels.
Pulmonary embolism (PE) complicates 1 in 1–3000 pregnancies and is a leading direct cause of maternal death in developed countries [ 9 ]. Most cases of pregnancy-associated PE occur antenatally, but the daily risk is highest in the first 6 weeks postpartum [ 9 ]. Risk factors for PE include prior thromboembolism, caesarean delivery, thrombophilia, postpartum haemorrhage requiring transfusion, obesity, preeclampsia and prolonged hospitalisation [ 10 ]. This report discusses a postpartum case of SHiP complicated by PE.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.