Article Text

Download PDFPDF

Abdominal pain and dysuria in pregnancy: urinary tract infection or life threatening haemorrhage?
  1. M J Lamyman,
  2. H Connor,
  3. R Brown
  1. Accident and Emergency Department, St. Mary’s Hospital NHS Trust, London, UK
  1. Correspondence to:
 Mr M J Lamyman
 8B Merridian Way, Holtye Road, East Grinstead, West Sussex, RH19 3GB, UK;


This report describes the case of a 27 year old woman presenting at 19 weeks’ gestation with epigastric pain and dysuria. Initially diagnosed with a urinary tract infection, she re-presented 10 days later with acute abdominal pain and haemoperitoneum. The diagnosis of placenta percreta was not made until laparotomy. This case highlights placenta percreta as a rare but serious complication of pregnancy that may become increasingly frequent as the rates of caesarean delivery rise. Early diagnosis, close monitoring, and prompt surgical management are essential as massive blood loss can occur. This can be challenging, as clinical presentation can be unusual.

Statistics from

A 27 year old woman presented to the emergency department 19 weeks into her pregnancy with epigastric pain and dysuria. There was no history of vaginal bleeding. Her obstetric history included one termination and six first trimester miscarriages. She had two healthy children, born by caesarean section. Her observations were normal and abdominal examination revealed mild suprapubic tenderness. Urine dipstick was positive for leucocytes and she was diagnosed with a urinary tract infection. Although subsequent microscopy and culture were negative, her symptoms improved following a course of amoxycillin.

She re-presented 10 days later with acute onset, severe epigastric pain radiating to both clavicles. The pain was worse on inspiration and lying flat. On examination, she was pale and distressed, with a pulse of 110/min and blood pressure 115/50 mmHg. Her abdomen was generally tender with guarding in the epigastric region. Vaginal and speculum examination were unremarkable. Haemoglobin was 93 g/l, with normal white cell count, urea and electrolytes, liver function and amylase. Arterial blood gas analysis was normal and erect chest revealed no air under the diaphragm.

A transabdominal ultrasound demonstrated a large quantity of fluid in the abdomen and pelvis. Abdominal organs appeared normal, and the fetus was active in adequate liquor. No retroplacental collection was seen, and the ovaries could not be visualised.

She was presumed to have an intra-abdominal bleed, although the source at this point was uncertain. She was admitted under joint surgical and obstetric care. The initial management strategy was conservative. She was closely monitored overnight and remained haemodynamically stable following initial fluid resuscitation. Over the next 6 hours, her symptoms did not settle, and as her haemoglobin dropped to 61 g/l, she required transfusion. The decision was made to proceed to laparotomy, which revealed 2.5 litres of blood in the abdomen. There was extensive bleeding from the lower segment of the uterus, which had ruptured at the site of a previous caesarean section scar. There was placental penetration through the defect with adherence to the bladder. To control the bleeding, it was necessary to proceed to termination of pregnancy and hysterectomy. The bladder was dissected free and cystoscopy revealed no invasion of the mucosal layer. She made a good postoperative recovery. Counselling was arranged to help the patient with her loss and she was discharged from the ward after 7 days.


Placenta percreta is the most extreme form of placenta accreta and is a rare but serious complication of pregnancy. Placenta accreta is abnormal placental attachment or invasion into the myometrium.1 Three degrees have been described: true placenta accreta, where villi are adherent to myometrial bed without invasion; placenta increta, where the villi invade the muscle; and placenta percreta, where there is full thickness penetration of the myometrium.2

The incidence of placenta accreta increased from 1 per 30739 between 1930 and 1950 to 1 per 7270 in the 1970s.3 The incidence of placenta percreta has been estimated to be around 1 per 90 000 deliveries,4 which would equate to around six cases per year in the UK.

Predisposing factors to the development of placenta accreta include: previous caesarean section, uterine curettage, manual removal of the placenta, and placenta praevia.5 The increasing incidence of placenta accreta may be due to the increased frequency of caesarean delivery.6,7

The clinical presentation of placenta accreta depends on the degree of placental penetration into the uterus and includes vaginal bleeding, uterine rupture, intra-abdominal haemorrhage, and bladder invasion.7 Ultrasound imaging has been used to assess the uterus in pregnancies at risk for placenta accreta. However, the extent of extrauterine involvement could not be consistently demonstrated.1,8 Thus, ultrasound is useful in the acute setting for detection of intraperitoneal fluid, but cannot reliably diagnose placenta percreta.

In retrospect, the urinary tract symptoms experienced by our patient could be considered a “warning” and may have been due to bladder irritation from the invading placenta. However, if no bleeding had occurred at this stage, it is unlikely that ultrasound imaging would have diagnosed the problem.

This case also demonstrates how well a young patient can compensate for substantial blood loss. Rapid decompensation can occur and close monitoring is essential if ongoing haemorrhage is suspected. The presence of placenta accreta increases maternal mortality by 10%. In the report on maternal deaths published in 2001,9 two deaths were attributable to placenta accreta. In one of these cases, despite anticipation of haemorrhage and exemplary care, the patient still died from massive blood loss.

Haemoperitoneum and uterine rupture due to placenta percreta usually requires hysterectomy. Conservative surgery has been described and includes localised resection of the placental implantation site, oversewing, blunt dissection, and packing techniques.1,7 However, such techniques have been associated with higher mortality, and their use is probably limited to less extensive accreta, where bleeding is minimal.7

Placenta percreta is rare, but as the number of caesarean deliveries increases, so will the number of women at risk. Placenta percreta should be considered in the gravid patient with risk factors for abnormal placental attachment presenting with abdominal pain and evidence of haemorrhage. Early diagnosis and prompt surgical management may reduce maternal mortality.



  • Competing interests: none declared

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.