Spontaneous rupture of uterine artery in a non-pregnant woman with adenomyosis: a case report and review of current literature

A 40-year-old woman, para 1, presented overnight to the emergency department of her local hospital with vaginal bleeding. It was her second attendance within 24 hours. She reported onset of menstruation six days previously, and the bleeding was increasingly heavy with clots, associated with intermittent lower abdominal cramps and fainting episodes. Her last pregnancy was more than 15 years ago. She had stopped a combined oral contraceptive pill two months before for a planned discectomy. On examination, she was pale with mildly tender suprapubic region, but her vital signs were stable. Initial blood results were as follows: haemoglobin 110 g/L, white cell count 11.5x10 9 /L, platelet count 261x10 9 /L. She was initially managed conservatively with tranexamic acid and medroxyprogesterone. In the next few hours, she had multiple episodes of hypotensive shock and continued to bleed despite medical management. On vaginal examination, cervix felt normal and a significant amount of blood clot was evacuated with active bright red bleeding. Bedside transvaginal ultrasound was performed, and it showed an endometrial thickness of 2 mm, clots in uterus and vagina, and no fibroids or free fluid in the pelvis. Her haemoglobin dropped to 62 g/L. The massive haemorrhage protocol was activated, and she received 8 units of packed red cells, 2 units of fresh frozen plasma and 2 units of cryoprecipitate in total. She therefore underwent a computed tomography (CT) of abdomen and pelvis with abdominal angiography. CT showed active extravasation of contrast from a branch of the right uterine artery into the endometrial cavity, tracking through the cervix into the vagina vault, but no evidence of a discrete vascular abnormality ( Figure 1 ). Trans-arterial embolisation was considered, but due to the patient’s ongoing bleeding and worsening pain, compounded by the delay to transfer to a tertiary centre with interventional radiology, a decision was made to undergo transabdominal hysterectomy. Intra-operative findings included haemoperitoneum, mild features of endometriosis on the uterus and normal adnexae. The procedure was uncomplicated, and her post-operative recovery was largely unremarkable, except for ileus and atelectasis. Histopathological examination of the specimen confirmed adenomyosis and no evidence of malignancy.

Spontaneous rupture of uterine artery is a rare occurrence. As such, to date, no systemic review or large studies have been conducted on this. The electronic database PubMed was searched looking for the following terms ‘spontaneous uterine artery rupture’, ‘spontaneous uterine vessel rupture’ and ‘spontaneous uterine artery erosion’. A number of cases have been reported and patients with this condition presented during pregnancy, labour, puerperium 1 - 8 , or up to two years following a gynaecological intervention 9 . In relation to pregnancy, the exact aetiology is not well understood 3 , but factors including pressure dynamics, anatomic disruptions and hormonal changes were proposed 10 . Other causes that have been identified 3 include aneurysm and pseudoaneurysm, congenital abnormality, vascular defect caused by inflammation or oestrogen 7 , erosion from endometriosis 11 - 14 , iatrogenic damage 15 - 18 , and termination of pregnancy 9 , 19 . As described in Williams Obstetrics, uterine artery pseudoaneurysm is a known rare cause of postpartum haemorrhage 20 . A case series in China concluded that pseudoaneurysm can develop after traumatic pelvic operations and non-traumatic delivery/abortion 21 . Another observational study in Japan estimated the incidence of uterine artery pseudoaneurysm to be 0.3-0.6% of deliveries and suggested that some may be undetected due to the absence of massive bleeding and/or spontaneous resolution 22 . A case series in Pakistan 23 suggested that bleeding as a result of acquired and iatrogenic uterine vascular abnormalities can be attributed to previous caesarean section, gestational trophoblastic disease, pelvic tumours 24 - 28 and malignancies 29 , 30 , pelvic infection, and dilatation and curettage. Of all reported cases, uterine artery rupture that are truly spontaneous (i.e. unprovoked by pregnancy or trauma) are exceptionally rare. There was one case where no cause was identified 31 , and few cases where the haemoperitoneum occurred as a complication of rupture of a massive fibroid 24 , 32 , 33 . In some case reports, local erosion of the uterine artery by an endometriotic lesion was described in non-gravid women 11 , 12 , 34 . Given the similar histopathology, we speculate that the adenomyosis in our case behaves similarly to this endometriosis case, causing local ulceration to the adjacent uterine artery, leading to eventual rupture and subsequent haemorrhagic event. In one case in Japan with similar presentation, a patient had a cardiac arrest secondary to severe haemorrhage from severe adenomyosis with fibroid 33 . Based on the literature and cases discussed, we have summarised the possible aetiology of spontaneous uterine artery rupture in Figure 2 . In patients with severe haemorrhage such as ours, resuscitation and maintaining haemodynamic stability is the initial management. In terms of definitive treatment, trans-arterial embolisation is the treatment of choice for uterine artery rupture secondary to pseudoaneurysm 35 - 38 , and this would also have been an appropriate treatment for the patient described in this case report. However, due to the lack of interventional radiology facilities in our local hospital, deterioration of the patient and completion of child-bearing, the more invasive management of hysterectomy was performed.

Spontaneous rupture of uterine artery is a rare event but an emergency with high mortality rate. Clinical suspicion in patients with unexplained ongoing severe vaginal bleeding should prompt further investigations and urgent input from obstetrics and gynaecology and interventional radiology. We suggest that, in patients with history of adenomyosis and endometriosis, the threshold for further investigations should be lower to allow rapid recognition and treatment.