No abstract

Objective To determine the incidence, risk factors, presentation, treatment and morbidity associated with secondary postpartum haemorrhage. Design Analysis of 132 consecutive women presenting with secondary postpartum haemorrhage occurring over a three-year period. Setting The maternity unit in a district general teaching hospital serving an annual delivery rate of around 6500 women. Main outcome measures Factors associated with the cause of the haemorrhage and the resulting morbidity.Results Most women presented during the second week after delivery. A history of primary postpartum haemorrhage (OR 9.3; 95% CI 6.2-14.0) and manual removal of placenta (OR 3.5; 95% CI 1.6-7.5) were the only signi®cant risk factors identi®ed. There was a high associated morbidity, with 84% requiring hospital admission, 63% surgical evacuation, 17% blood transfusion, with three women suffering a uterine perforation, one managed by hysterectomy. In women undergoing evacuation only, 37% had retained placental tissue con®rmed after surgery; pre-operative ultrasound examination did not provide a better discrimination over clinical assessment for this ®nding. Conclusions Secondary postpartum haemorrhage occurs in just under 1% of women, is associated with primary postpartum haemorrhage and retained placenta, and may result in signi®cant maternal morbidity. This problem deserves more attention than it has received in recent years.

No abstract

Case report A 32 year old healthy nulliparous woman had a normal delivery at term. Primary postpartum haemorrhage of 500 mL occurred within minutes of delivery and settled with an infusion of syntocinon and suturing of a vaginal tear. Bleeding recurred within hours and she underwent examination in theatre with evacuation of clots and placental tissue and resuturing. The estimated blood loss was 2600 mL and she had a three-unit blood transfusion. On the following two days, she was transfused a further six units. She went home on day three. She had further bleeding and was commenced on antibiotics and readmitted on day six when she underwent general anaesthesia and resuturing. By day seven, her clotting results showed an activated partial thromboplastin time (APPT) of 78.2, ratio 2.8 and a prothrombin time (PT) of 10.9, ratio 0.9. Tests revealed her factor VIII level at <1% of normal. A diagnosis of acquired factor VIII inhibitor was made at a level of 15 Bethesda units. She was commenced on prednisolone 1 mg/kg/day and had two doses of factor VIIa 360 IU. Subsequently, her loss fluctuated and she had factor VIIa and blood transfused. By day 16, loss increased and she was given desmopressin (DDAVP) 0.3 Ag/kg, giving initial improvement. Continued bleeding, despite further DDAVP and factor VIIa, prompted transfer to the regional unit. At the time of transfer, 19 days since delivery, she had received more than 20 units of packed cells, five doses of factor VIIa and two doses of DDAVP. She was on tranexa-mic acid 1 g qds, prednisolone 60 mg od and ferrous sul-phate. Over the subsequent three days, repeated attempts to control the bleeding with vaginal packing and resuturing had failed. She was transferred to the angiography department and selective bilateral internal pudendal artery embolisation was performed via a right common femoral artery catheter under local anaesthetic. Arterial blushing was seen in the vaginal wall and both internal pudendal arteries were embolised with gelatine sponge particles. Local pressure was applied to the puncture site in the groin and factor VII infusions were needed both at the end of the embo-lisation procedure and on day three to control local oozing. The following day, she started mobilising and the aprotinin was discontinued. Inspection of the perineal area showed no evidence of ischaemic damage. She was discharged home, 31 days after she had given birth, on prednisolone 60 mg od and tranexamic acid 500 mg tds. Discussion Acquired factor VIII deficiency has an incidence of approximately one per million. 1 It is due to development of antibodies against factor VIII. Most cases occur in healthy individuals without discernible risk factors, but it is also associated with autoimmune conditions such as rheumatoid arthritis and systemic lupus erythematosus, inflammatory bowel disease, multiple sclerosis and malignancies. 2 In up to 11% of cases, the associated factor is a recent or ongoing pregnancy. 2,3 Acquired haemophilia may occur in relation to any pregnancy but the risk appears to...

A 32-year-old para 4, prese nted to the A&E Department 48 hours aft er a successful home delivery. Her antenatal history was unremarkable and had four spontaneous vaginal deliveries in the past. On arrival, she complained of worsening abdominal pain over the past 48 h which was now unbearable. Her labour was precipitous with fi rst stage lasting 2 h and assisted third stage 30 min. Blood loss was 150 ml.On admission she was apyrexial, with blood pressure 89/42 mmHg and pulse 82 b.p.m. On examination, there was generalised abdominal tenderness with guarding and rebound. Bimanual examination revealed cervical excitation tenderness and mild bilateral adnexal tenderness. A diagnosis of endometritis was made and intravenous antibiotics and fl uids commenced.An ultrasound revealed a left adnexal mass. Haemoglobin on admission was 10.1 g/dl and this declined to 5 g/dl, prompting an emergency laparoscopy, which revealed a haemoperitoneum of 700 ml and a large left retroperitoneal haematoma extending to the splenic fl exure.An explorative laparotomy was performed in the presence of the vascular surgeons. Th ere was a high index of suspicion that the haematoma was possibly due to a ruptured uterine artery and a decision was made to ligate the left uterine artery using haemostatic sutures. Postoperatively, despite 7 units of blood transfusion, her haemoglobin remained at 6 g/dl. An urgent CT with contrast suggested a left ovarian artery abnormality. She remained tachycardic (pulse 106 b.p.m.) and the haemoglobin declined to 4.3 g/dl. She was transferred to the tertiary centre for angiography. Aortogram and selective catheterisation of the left ovarian artery confi rmed active bleeding into the pelvis. Th ere was contrast extravasating from the pelvic segment of the left ovarian artery, which was dilated and tortuous ( Figure 1). Aft er prolonged catheter manipulation due to arterial tortuosity and recurrent spasm, successful embolisation was performed using a total of fi ve embolisation coils of 5 mm diameter. Figure 2 shows the check angiogram.She was further transfused 6 units of blood and her haemoglobin stabilised at 9.9 g/dl. She had an unremarkable recovery and was discharged 3 days later. DiscussionTh e actual incidence of ovarian artery rupture is unknown. In 1904, Williams was the fi rst to report a spontaneous haemoperitoneum resulting from rupture of utero-ovarian vessels. A recent literature review in 2009 found 12 reported cases of pregnancy-related ruptured ovarian artery aneurysm. Some 10 of these cases occurred at the time of delivery or up to 4 days postpartum and two cases were in the 3rd trimester (Chao and Chen 2009). However, haemorrhage from the utero-ovarian vessels can occur from as early as the 10th week of pregnancy and as late as 3 weeks ' postpartum (Tsoutsoplides 1967).Maternal mortality is approximately 49% but with improvement in blood banking and surgical techniques, this may be 4% (Hodgkinson and Christenson 1950). Th e perinatal mortality is approximately 31%.