Mrs T, a 36-year-old mother of two young children, attended as an inpatient for an elective vaginal hysterectomy and repair of prolapse. She had no relevant past history and her preoperative assessment was unremarkable.
During surgery, blood loss was greater than usual at 800 ml but no other problems were noted. In the recovery room she was well but noted to be pale and agitated, complaining of abdominal pain.
She received patient-controlled opiate analgesia. She was returned to the ward just under an hour after surgery, but nursing staff called the anaesthetics registrar, Dr F, an hour later as she had become unwell, pale and hypotensive with a borderline bradycardia (BP 100/60 mm Hg, pulse 52 bpm). Dr F prescribed 40% oxygen and 500 ml of colloid fluid over an hour.
A failed attempt to take venous blood was abandoned by her gynaecology registrar, Dr K, who called for Dr F to assist him. Dr F noted that the patient’s vital signs were unchanged but her veins were collapsed.
He asked the nurse in charge to give Mrs T one unit of whole blood over the next hour and to transfuse another unit of blood over the following four hours. Dr F reviewed the patient several times over the next few hours.
Two hours after the blood transfusion had been started, Mrs T had a BP of 95/55 mm Hg and a heart rate of 52 bpm. A urinary output of 100 ml since surgery was recorded. Dr F conferred with his consultant, Dr J, and they decided to return Mrs T to the recovery room to put her on a monitor and insert a CVP line. Before they could do this, however, the patient collapsed and stopped breathing.
During resuscitation she was intubated and received adrenaline, atropine and defibrillation. She was transferred to the recovery room, resuscitation continuing during the transfer.
On arrival in recovery Mrs T regained cardiac output and some respiratory effort. She remained hypotensive and had a heart rate of 130 bpm. Dr J was unsuccessful in attempting central venous and peripheral arterial cannulation.
Another anaesthetic consultant, Dr Q, was able to insert a Swan-Ganz cannula and commence rapid transfusion. An enquiry into the possibility of coagulopathy was made at this stage, but laboratory confirmation of this only became available later. The patient did not respond well to the rapid transfusion so an infusion of adrenaline was started and there was a moderate improvement in her vital signs.
The team planned to move Mrs T to the ITU of a nearby hospital and to give fresh frozen plasma once clotting results were known.
Haematological and biochemical results showed a severe coagulopathy with no obvious bleeding. A vaginal examination revealed no haematoma or other abnormality, and a chest x-ray was reported as normal.
The fresh frozen plasma was instituted but during the transfusion Mrs T had a fit, developed bradycardia and cardiac arrest. She did not respond to attempts to resuscitate her.
At autopsy, the cause of death was given as ‘haemorrhagic shock due to an intra-abdominal haemorrhage from pelvic operative site following hysterectomy and vaginal repair for uterine prolapse.’
Mrs T’s family brought a claim for compensation.
We asked an expert in anaesthetics for his opinion. He concluded that the doctors could be criticised for failing to appreciate that hypovolaemia was the cause of Mrs T’s hypotension, and for not taking earlier, more aggressive measures to locate the site of the bleeding and initiate surgical repair. He said, ‘Significant intra-abdominal bleeding must be the number one differential diagnosis of sustained postoperative hypotension, in the absence of other differential diagnoses such as sepsis, anaphylaxis and myocardial depression.’
The claim was settled for a substantial sum.
Sustained clinical indicators of hypovolaemia must not be ignored in a postoperative patient. Following surgery, in a scenario such as this, loss of blood at the operation site must be excluded, as early as possible, as the cause of the patient’s symptoms and signs, in order to avoid tragedy.