S132 Late failure and relapse in patients receiving non-invasive ventilation for exacerbations of COPD: A UK prospective study

Introduction In patients treated with non-invasive ventilation (NIV) for exacerbations of COPD (ECOPD), late failure, defined as recurrent respiratory acidaemia during NIV but after 24 hours following normalisation of pH, or relapse, defined as recurrence of respiratory acidaemia during admission but after 24 hours following cessation of NIV, are associated with poor survival and frailty.1 However, the UK prevalence and outcome of late failure and relapse is unknown. Methods Unique, consecutive patients who had spirometry confirmed ECOPD requiring NIV for respiratory acidaemia were prospectively recruited to the NIV Outcomes study (ISRCTN22921168) in 10 NHS trusts. The prevalence and outcomes of patients with late failure and/or relapse was investigated. Results Of the 733 patients recruited, 41(5.6%) developed late failure alone, 48(6.5%) experienced relapse alone and 4 (0.5%) developed both. The time to late failure was 2.8(1.8–5.8) days from NIV starting and 2.3(1.5–3.6) days from initial pH correction. Population descriptors are shown in the table 1. Following late failure, NIV was continued in 95.6% of patients, with 60% having changes to ventilatory settings and 31.1% having changes to improve patient-ventilator synchrony. Only 2.2% were invasively ventilated following late failure. Overall hospital mortality was 31.7% in late failure patients, 37.5% in patients who relapsed and 75% if both occurred. This compares with 17.7% mortality in patients without late failure or relapse. Discussion Late failure and/or relapse is relatively common, occurring in around 10% of NIV patients and is associated with worse hospital mortality. Both the prevalence of late failure, and inpatient mortality are lower here than described in a previous single centre study.2Patients with late failure/relapse tend towards worse lung function, lower BMI, more LTOT and left ventricular systolic dysfunction. The majority continued on NIV following optimisation of ventilatory settings, and 60–70% survived to discharge. It may be that in patients who have optimised NIV (ventilator settings or synchrony) failure to improve is not due to failure of the ventilatory interface (NIV rather than invasive ventilation), but rather failure of the lungs due to the severity of the underlying disease and acute insult. References Lancet Respir Med 2018;6:935–47 Thorax 2000;55(10):819–25