GOLD 2007 and 2011 fail to accurately predict mortality risk

By Lucy Piper, Senior MedwireNews Reporter

Neither the 2007 nor the 2011 Global Initiative for Chronic Obstructive Lung Disease (GOLD) classification systems have sufficient prognostic accuracy to predict mortality, a pooled analysis of individual patient data shows.

The findings showed that the classification systems were significantly related to total mortality up to 10 years, but neither had a striking discriminatory power, with area under the curves ranging from 0.62 to 0.65.

“Thus, these classification schemes might be useful to guide therapy but are unlikely to be clinically useful to identify patients at high risk of mortality, even in the short term”, observe Joan Soriano (Universidad Autónoma de Madrid, Spain) and colleagues.

The researchers also note in The Lancet Respiratory Medicine that the simpler 2007 GOLD classification, based solely on spirometry, had the same predictive ability as the 2011 GOLD classification scheme, despite it also including patient symptoms and exacerbations.

Data were collected from 22 published COPD cohorts with 15,632 patients, totalling 70,184 person–years of follow-up. The patients were aged an average of 63.9 years and 10,751 were men.

Based on forced expiratory volume in 1 second (FEV1) data, 16% were classified as having mild (stage I) disease, 46% moderate (stage II) disease, 28% severe (stage III) disease and 11% very severe disease (stage IV) according to the 2007 GOLD criteria.

A similar proportion of patients were classified as having mild disease (grade A) with the 2011 GOLD scheme, at 38%. But whereas with the 2007 GOLD scheme most patients had moderate or severe disease, there was a shift to more patients having very severe disease with the 2011 GOLD scheme. Indeed, nearly three times more patients were classified as having very severe grade D COPD, at 31%.

“This change has the potential to increase therapeutic intensity in these patients, which might have important economic and clinical impact”, the team explains.

Indeed, commenting on this finding in the journal Afroditi Boutou (Aristotle University of Thessaloniki, Greece) advises that “before completely replacing the older, simpler, and widely applicable GOLD 2007 classification scheme, prospective cohort studies are needed to establish whether more intense treatment, according to the 2011 GOLD categorisation, could offer a long-term survival benefit among patients with COPD.”

In predicting mortality, the GOLD 2007 showed a severity-dependent relationship, with very severe disease associated with a 5.7 times higher risk of death than mild disease, while moderate and severe disease conferred a risk that was 1.9 times and 3.5 times higher, respectively.

Similarly, patients with very severe disease on GOLD 2011 had the worst prognosis, with a risk of death 3.5 times that of patients with mild disease. But the researchers note that the mortality risk did not differ significantly between grade B and C, with the risk increased by 1.7 to 2.2 times, compared with grade A, suggesting that “class C might be superfluous”.

They were able to confirm, however, that the current % predicted FEV1 thresholds used in the GOLD classification systems of 80%, 50% and 30% are very close to the optimum thresholds for predicting mortality and should be sustained.

medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2015

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.