Liver rupture after cardiopulmonary resuscitation during peri‐operative cardiac arrest
2001; Wiley; Volume: 56; Issue: 4 Linguagem: Inglês
10.1046/j.1365-2044.2001.01976-27.x
ISSN1365-2044
Autores Tópico(s)Trauma Management and Diagnosis
ResumoLiver rupture is an uncommon complication of cardiopulmonary resuscitation (CPR). We wish to report the case of a patient who arrested in theatre, prior to induction of anaesthesia and sustained traumatic liver rupture as a result of CPR. A 52-year-old schizophrenic man was admitted to our A & E Department with a history of lower back and abdominal pain. There was no history of trauma. He had undergone cataract surgery some years previously but had no other medical history and he was on no medication. On examination, he was found to have a distended and moderately tender abdomen with absent bowel sounds. Pulse rate was 108 beat.min−1 and blood pressure was 90/50 mmHg. Blood tests were sent to the lab and revealed a haemoglobin of 9.4 g.dl−1 and a white cell count (WCC) of 1.55 × 109.l−1. Urea was 19.1 mmol.l−1 and creatinine 289 µmol.l−1. The patient was catheterised and 10 ml of urine was returned. While awaiting surgical review, his blood pressure dropped to 60/40 mmHg and this was treated vigorously with 3 l crystalloid and 500 ml Gelofusine®. This increased the blood pressure back to 90/60 mmHg. The patient was reviewed by a senior surgical SpR, who made a diagnosis of a leaking abdominal aortic aneurysm and a decision was then made to proceed to theatre with resuscitation in progress. The patient was seen by the anaesthetic team, which consisted of a consultant, an SpR1 and a SHO, and was transferred directly to theatre. In theatre, a further two 14G cannulae were sited and preparations were made for a rapid sequence induction. As the patient was being pre-oxygenated, he became asystolic and, after checking ECG leads and gain, CPR was commenced. The patient was intubated with a 9.0 cuffed oral tracheal tube. Atropine 3 mg and epinephrine 1 mg were given intravenously followed by 1 min of CPR in accordance with Resuscitation Council (UK) guidelines. Re-evaluation revealed that the patient was in pulseless electrical activity (PEA). The surgeons performed a laparotomy immediately and a perforated 2-cm gastric ulcer was found on the lesser curve of the stomach, thought to be at least 24–48 h old. The precipitating cause for the patient's asystolic arrest appeared more likely to be severe acidosis secondary to prolonged hypovolaemia than due to acute blood loss, although no arterial blood gases had been performed. Sepsis secondary to a prolonged visceral perforation was thought to be the cause of his low WCC. The wound was covered and left by the surgeons while resuscitation was in progress. A further 8 mg of epinephrine, 200 ml of 8.4% sodium bicarbonate, Gelofusine® 5 l and fentanyl 250 µg were given. CPR continued for a further 15 min, during which time pulseless electrical activity was present. An anaesthetic SHO and an anaesthetic nurse, both of whom were ALS trained, performed chest compressions. At this point, the abdomen was re-examined by the surgical registrar and a large volume of blood was noted in the peritoneum, which had not previously been present. Further examination of the abdomen revealed a large 17-cm hepatic split under the diaphragm, which was bleeding profusely. A total of 8 units of blood plus 1200 ml of fresh frozen plasma were given and finally an output was restored. The total time taken to restore an output was approximately 35 min. The liver was packed with 10 large abdominal packs and the gastric ulcer was oversewn. Anaesthesia was maintained with 0.5% isoflurane in air/O2 mix and central venous and arterial cannulation were performed. Epinephrine was commenced at 50 µg.min−1 to maintain a blood pressure of 120/80 and heart rate of 109 beat.min−1 despite a CVP of 18 mmHg. The estimated total blood loss was 6 l. Surgery was concluded about 30 min after the restoration of an output and the patient returned to ITU postoperatively, where he was fully sedated and ventilated overnight. Here bloods were rechecked revealing a haemoglobin of 10.1 g.dl−1, WCC 0.97 × 109.l−1 and platelet count of 99 × 109.l−1. His clotting was deranged with an INR of 2.6 and APTT ratio of 2.01. A chest X-ray was performed but no rib fractures were seen then or indeed on subsequent X-rays. Overnight, he was given a further 2 units of blood to transfuse him to a haemoglobin of 13.0 g.dl−1 and 1200 ml of fresh frozen plasma to improve his coagulation. However, he remained in oliguric renal failure. The following day, he was transferred to our regional liver unit where he underwent a pack change 2 days later. A further two laparotomies were carried out over the following 2 weeks but the patient was eventually discharged from intensive care, albeit with some mild cognitive deficit, and eventually returned to Greenwich Hospital about 8 weeks after the original incident. Damage to visceral structures is a rare but documented complication of cardiopulmonary resuscitation, usually with devastating consequences. Damage to spleen, stomach and liver have been reported. We conducted a Medline search from 1966 to 2000 and are aware of two reports of liver rupture following CPR; however, both of these were in association with the administration of thrombolysis [1, 2]. One of these cases had a fatal outcome. We also found a further two reports of liver laceration [3, 4], one of which was in a pregnant woman who also died. We found no instances of this condition occurring during an intra-operative cardiac arrest. In cases where the abdomen is not open, diagnosis is much more difficult and the condition is probably under reported. Two of the above cases were diagnosed on clinical grounds. Of the remaining two, one was diagnosed during thoracotomy for a pulmonary embolectomy and one at autopsy. It is important to use the correct technique while performing CPR to avoid damage to adjacent structures. In this case, it is worth noting that the injury to this patient's liver was not associated with overlying rib fractures. The possibility of visceral damage due to CPR should be considered in cases of prolonged PEA. In this case, the blood loss associated with this patient's liver injury was probably the cause of his persistent PEA and it was only when it was discovered that the patient had profuse bleeding from a ruptured liver that effective therapy was given and an output restored. Finally, in the case of intra-operative cardiac arrest when the abdomen is open, some authorities have recommended the early use of internal cardiac massage and also of thoracotomy and cardiopulmonary bypass (if available) even in non-traumatic intra-operative cardiac arrest [5]. This may have been beneficial in his case and certainly may have prevented the liver rupture.
Referência(s)