Abstract
INTRODUCTION: Exacerbation of COPD complicated by respiratory acidaemia is the commonest indication for noninvasive ventilation (NIV). The NIV outcomes (NIVO) score offers the best estimate of survival for those ventilated. Unfortunately, two-thirds of cases of COPD are unrecognised, and patients may present without COPD having been confirmed by spirometry. METHODS: In the 10-centre NIVO validation study there was no pre-admission spirometry in 111 of 844 consecutive patients (termed "clinical diagnosis" patients). We compared the performance of the NIVO, DECAF and CURB-65 scores for in-hospital mortality in the clinical diagnosis cohort. Usual clinical practice was not influenced, but confirmation of COPD in the year following discharge was captured. RESULTS: In the clinical diagnosis cohort, in-hospital mortality was 19.8% and rose incrementally across the NIVO risk categories, consistent with the NIVO validation cohort. NIVO showed good discrimination in the clinical diagnosis cohort: area under the receiver operating curve 0.724, versus 0.79 in the NIVO validation cohort. At 1 year after discharge, 41 of 89 clinical diagnosis patients had undertaken diagnostic spirometry; 33 of 41 had confirmation of airflow obstruction (forced expiratory volume in 1 s/(forced) vital capacity <0.7), meaning the diagnosis of COPD was incorrect in 19.5% of cases. DISCUSSION: These data support the use of the NIVO score in patients with a "clinical diagnosis" of COPD. NIVO can help guide shared decision-making, assess risk-adjusted outcomes by centre and challenge prognostic pessimism. Accurate diagnosis is critical to ensure that acute and long-term treatment is optimised; this study highlights failings in the follow-up of such patients.