Massive pulmonary embolism (PE) can present with extreme physiological dysfunction, characterised by acute right ventricular failure, hypoxaemia unresponsive to conventional therapy and cardiac arrest. Consensus regarding the management of patients with persistent shock following thrombolysis is lacking. Our primary objective was to describe the application of extracorporeal membrane oxygenation (ECMO) in the treatment of acute massive PE. We were unable to identify any randomised controlled trials (RCTs) comparing ECMO with other support systems in the setting of massive PE. We reviewed case reports and case series published in the past 20 years to evaluate the mortality rate and any poor prognostic factors. Overall survival was 70.1% and none of the definitive treatment modalities was associated with a higher mortality (thrombolysis - OR - 0.99, P - 0.9, catheter embolectomy - OR - 1.01, P - 0.99, surgical embolectomy - OR - 0.44, P - 0.20). Patients who had ECMO instituted whilst in cardiorespiratory arrest had a higher risk of death. (OR - 16.71, P - 0.0004). When compared with other causes of cardiac arrest, patients who survived a massive PE presented a good neurological outcome (cerebral performance category 1 or 2).

SummaryAccurate assessment of intravascular fluid status and measurement of fluid responsiveness have become increasingly important in peri-operative medicine and critical care. The objectives of this systematic review and narrative synthesis were to discuss current controversies surrounding fluid responsiveness and describe the merits and limitations of the major cardiac output monitors in clinical use today in terms of usefulness in measuring fluid responsiveness. We searched the MEDLINE and EMBASE databases (2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015); inclusion criteria included comparison with an established reference standard such as pulmonary artery catheter, transthoracic echocardiography and transoesophageal echocardiography. Examples of clinical measures include static (such as central venous pressure) and dynamic (such as stroke volume variation and pulse pressure variation) parameters. The static parameters measured were described as having little value; however, the dynamic parameters were shown to be good physiological determinants of fluid responsiveness. Due to heterogeneity of the methods and patient characteristics, we did not perform a meta-analysis. In most studies, precision and limits of agreement (bias AE1.96SD) between determinants of fluid responsiveness measured by different devices were not evaluated, and the definition of fluid responsiveness varied across studies. Future research should focus on the physiological principles that underlie the measurement of fluid responsiveness and the effect of different volume expansion strategies on outcomes.

No abstract