Outcomes of emergency laparotomy in patients on extracorporeal membrane oxygenation for severe respiratory failure: A retrospective, observational cohort study.

PURPOSE: There is a paucity of literature to support undertaking emergency laparotomy when indicated in patients supported on ECMO. Our study aims to identify the prevalence, outcomes and complications of this high risk surgery at a large ECMO centre. MATERIALS AND METHODS: A single centre, retrospective, observational cohort study of 355 patients admitted to a university teaching hospital Severe Respiratory Failure service between December 2011 and January 2017. RESULTS: The prevalence of emergency laparotomy in patients on ECMO was 3.7%. These patients had significantly higher SOFA and APACHE II scores compared to similar patients not requiring laparotomy. There was no difference in the duration of ECMO or intensive care unit (ICU) stay post decannulation between the two groups. 31% of laparotomy patients survived to hospital discharge. Major haemorrhage was uncommon, however emergency change of ECMO oxygenator was commonly required. CONCLUSION: Survival to hospital discharge is possible following emergency laparotomy on ECMO, however the mortality is higher than for those patients not requiring laparotomy, this likely reflects the severity of underlying organ failure rather than the surgery itself. Our service's collocation with a general surgical service has made this development in care possible. ECMO service planning should consider general surgical provision.

Acceptance and transfer to a regional severe respiratory failure and veno-venous extracorporeal membrane oxygenation (ECMO) service: predictors and outcomes.

The use of extracorporeal membrane oxygenation for respiratory failure is high risk and resource intensive. In England, five centres provide this service and patients who are referred have four possible outcomes: declined transfer due to perceived futility; accepted in principle but remain at the referring centre with ongoing surveillance; retrieved using conventional ventilation; or retrieved on extracorporeal support. The decision-making process leading to these outcomes has not previously been examined. We evaluated referrals to one centre and identified factors associated with each decision outcome. Five hundred and sixty-four patients were analysed from January 2012 to October 2015. One hundred and fifty-seven patients were declined; multivariate analysis demonstrated associated factors to be: age (odds ratio (95% confidence interval) 1.05 (1.04-1.07)); immunocompromise (4.95 (2.58-9.67)); lactate (1.11 (1.01-1.22)); duration of ventilation (1.08 (1.04-1.14)); and cardiac failure (3.22 (1.04-10.51)). Factors associated with the decision to retrieve an accepted patient were: plateau pressure (1.05 (1.01-1.10)); ratio of arterial oxygen partial pressure to fractional inspired oxygen (0.89 (0.85-0.93)); partial pressure of carbon dioxide in arterial blood (1.13 (1.03-1.25)); and the absence of non-pulmonary infection (0.31 (0.15-0.61)). Only pH was independently associated with the decision to transfer on extracorporeal support (0.020 (0.002-0.017)). Six-month survival in the declined, non-retrieved, conventionally retrieved and extracorporeal-retrieved groups was 16.6%, 71.1%, 76.7% and 72.1%, respectively, substantially supporting the decision-making model. Survival in the accepted group exceeds that reported previously. However, a proportion of those declined do survive and some remotely managed patients die. This suggests the approach does not account for some important survival-determining factors.

Hypotension and anaemia--a blinding combination.

We present a case of complete blindness following severe dengue haemorrhagic fever complicated by anaemia and a dialysis-related episode of profound hypotension. The clinical and radiological features indicated an optic neuropathy, most likely ischaemic in aetiology. The features of posterior ischaemic optic neuropathy and differential diagnosis are discussed.