Prediction Models for Exacerbations of COPD: A Qualitative Systematic Review and Critical Appraisal

Abstract

Introduction: Prediction of future exacerbations of chronic obstructive pulmonary disease (COPD) is a major concern for long-term management of this disease. However, it is not clear which models are best to predict exacerbations in patients with COPD. Therefore, our aim was to investigate known prognostic variables and multidimensional scores that predict prognosis in COPD in a systematic review, specifically including variables that contribute to risk assessment of patients for exacerbation.Methods: We conducted a systematic review on prognostic variables, multivariate score or models for COPD. PUBMED, EMBASE, the Cochrane database, CINAHL PLUS, Scopus and the Web of Science were searched up to Sep 2, 2019. Results: A total of 12,014 abstracts were screened, leading to 355 full-text reviews, and , 22 studies with 46 prediction models met the inclusion criteria. Sample sizes ranged from 121 to 14600. The definition and measurement of exacerbations was symptom-based in nine(out of 22 studies), event-based in 11 and unclear definition in one case. Exacerbations were not adjudicated by a committee in any study. The prediction models were mainly based on prospective cohort studies (a retrospective cohort studies for seven models), while prediction models were based on a randomised controlled trial for two models). Follow-up periods ranged from 90 days to up to 8 years (the most common follow-up was up to 1 year). All included studies aim to find a combination of predictors or multidimensional scores strongly associated with exacerbations, while few studies focused on particular predictors but adjusted for age, gender, body mass index (BMI)FEV1 % predicted, smoking and previous exacerbation, age,number of comorbidities. More than 40 different predictors were used across the included prediction models. Airways obstruction (FEV1 % predicted or FEV1 or GOLD stage) was the most common predictor). Conclusion: A number of variables contributing to the prediction of exacerbation in COPD were identified. However, the quality of evidence remains low. Although the prognostic performance of some of the indices has been validated, they all lack sufficient evidence for implementation.