S57 Predicting outcome from exacerbations of copd requiring assisted ventilation: results from the NIV Outcome (NIVO) study (original) (raw)

The failing lung in COPD, 2019

Abstract

Introduction Exacerbations of COPD account for approximately 12% of UK hospital admissions. Over 20% will be complicated by respiratory acidaemia, which has high mortality. Non-Invasive ventilation (NIV) confers a 2–3 fold mortality reduction, but practice is sub-optimal; the intervention is underused, infrastructure is lacking, and complex decisions are made by a wide range of clinicians.1 It is likely that prognostic pessimism contributes to underuse. We aimed to derive and separately validate a simple, bedside, clinical tool to predict in-hospital mortality in exacerbations of COPD complicated by respiratory acidaemia requiring assisted ventilation. Methods Derivation was single-centre and retrospective. Consecutive patients meeting selection criteria were identified and clinical data collected. Multivariable regression identified independent predictors of in-hospital death and a simple model created. For validation, consecutive patients were prospectively recruited from 10 sites and model performance assessed. Results 489 patients were identified in the derivation study and 733 in the validation (in-hospital mortality 25.4 and 20.1% respectively). Key validation descriptors: 70% hospitalised during previous year, Mean (SD) age 70.5 (9.3) years and FEV1 % predicted 37.2 (15.4). 56% were unable to leave the house unassisted (eMRCD 5a or 5b) and 29% prescribed LTOT. 36% had previously required NIV and 9% were receiving home ventilation. Median (IQR) pH at onset of ventilation 7.27 (7.22–7.30), with CO210.2 (2.7) kPa. The final prognostic (NIVO) score comprised: Atrial fibrillation, chest X-ray consolidation, pH <7.25, Glasgow coma scale ≤14 (all 1 point), timing of acidaemia >12 hours from admission time (2 points) and eMRCD (1–4=0, 5a=2, 5b=3) yielding a maximum score of 9 using 6 indices. Stepwise increase in mortality was observed with an area under the receiver operated curve of 0.79 in the validation cohort (0.83 derivation). The NIVO score outperformed pre-identified comparator scores (APACHE II, CAPS, Confalonieri risk chart) in both its derivation and validation studies.Abstract S57 Table 1 NIVO Score Survived Died Total Mortality 0 87 3 90 3.3% 1 70 6 76 7.9% 2 134 7 141 5.0% 3 121 25 146 17.2% 4 95 23 118 19.5% 5 48 39 87 44.8% 6 23 26 49 53.1% 7 7 10 17 58.8% 8 1 7 8 87.5% 9 0 1 1 100% Total 586 147 733 20.1% Risk category (score) Survived Died Total Mortality Low (0 – 2) 291 16 307 5.2% Medium (3–4 ) 216 48 264 18.2% High (5–7 ) 78 75 153 49.0% Very High (8–9 ) 1 8 9 88.9% Discussion Using only simple, readily available indices good prediction of in-hospital mortality is feasible. Potential practical applications include but are not limited to guiding level of care, setting treatment limitations and objectifying both clinician decision making and discussion with patients/family members. Reference Inspiring change 2017

Keir Lewis hasn't uploaded this paper.

Let Keir know you want this paper to be uploaded.

Ask for this paper to be uploaded.