Cardiopulmonary resuscitation-associated major liver injury

AIM: To evaluate the frequency, presentation, treatment and outcome of cardiopulmonary resuscitation-associated major liver injury in patients after non-traumatic in- or out-of-hospital cardiac arrest. MATERIALS AND METHODS: Retrospective analysis of a cardiac arrest registry in a tertiary care hospital emergency department. We reviewed patients charts, laboratory data, diagnostic imaging studies and autoptic findings. RESULTS: Within 14.5 years, major liver injury (rupture/laceration, haemorrhage/haematoma) was found in 15 of 2558 cardiac arrest victims (0.6%). Eleven were male (73%), median age was 58 (IQR 53-67). In seven, resuscitation was started out-of-hospital. In 9 of the 15 patients, liver injury was correctly diagnosed intra vitam. In 5, haematocrit level was low on admission, in 8 haematocrit dropped significantly during observation; haemostasis was compromised in 13 patients, 4 of them receiving thrombolytic therapy. Bedside abdominal sonography showed free intra-peritoneal fluid in 8 of 9 cases examined. In 11 patients, we found liver rupture/laceration, in 4 liver haemorrhage/haematoma. The site of injury was the left liver lobe in 11, six underwent emergent surgery. Two of 15 patients survived to 6 months in good neurological condition, 1 after emergency surgery. No patient died from bleeding due to liver injury. CONCLUSION: Our single centre observation confirms that resuscitation-associated major liver injury is infrequent and shows that most patients had compromised haemostasis. Low or dropping haematocrit should trigger suspicion. Bedside sonography reveals intra-peritoneal fluid or liver injury. A conservative therapeutic approach or emergency surgery may be warranted. Major liver injury alone scarcely appears to influence overall outcome.