Emphysema detected on computed tomography and risk of lung cancer: A systematic review and meta-analysis

Abstract

Background

Studies exploring the association between emphysema detected on chest computed tomography (CT) and lung cancer have yielded mixed results. Our objective was to systematically review the evidence for this association.

Methods

We searched MEDLINE, EMBASE and the Cochrane Library for the terms “lung cancer”, “emphysema” and “computed tomography” without language restriction. Bibliographies were also reviewed and authors contacted for additional information. Human studies in which CTs were performed and assessed for emphysema and in which subjects were evaluated systematically for lung cancer were included. Qualitative synthesis of evidence was performed followed by pooling of effect estimates using a random-effects model.

Results

Of 187 citations, 7 were included in the qualitative synthesis and 5 in the meta-analysis. Three studies assessing emphysema visually observed an association with lung cancer, independent of smoking history and airflow obstruction. Three studies using densitometry to detect emphysema found no association with lung cancer. Another study directly comparing automated and visual emphysema detection techniques found only the latter to associate with lung cancer. Among 7368 subjects included in the meta-analysis, 2809 had emphysema on CT and 870 were diagnosed with lung cancer. The pooled adjusted odds ratio for lung cancer in the presence of emphysema on CT was 2.11 (95% CI 1.10–4.04); stratification by detection method yielded OR of 3.50 (95% CI 2.71–4.51) with visually detected emphysema and 1.16 (95% CI 0.48–2.81) with densitometric emphysema.

Conclusion

Systematic literature review shows emphysema detected visually on CT to be independently associated with increased odds of lung cancer. This association did not hold with automated emphysema detection.

Introduction

Lung cancer is the leading cause of cancer death globally [1], [2]. Despite recent advances in targeted therapies over the past decade, lung cancer outcomes remain poor with an overall 5-year survival of 15.7% [3]. Identification of individuals at highest risk for lung cancer will help to focus preventive interventions (e.g. smoking cessation) and lung cancer screening strategies [4], [5].

Pulmonary emphysema is defined morphologically as the enlargement of airspaces distal to the terminal bronchiole from dilatation or destruction of alveolar walls [6], [7]. The advent of chest computed tomography (CT) has greatly advanced our understanding of emphysema by affording accurate pre-mortem detection and prospective study [8], [9], [10]. Several studies have validated qualitative and quantitative methods of detecting presence and severity of emphysema on CT with pathologic specimens (reviewed by Thurlbeck et al. [10]) [8], [9]. More recently, studies have explored the association between emphysema detected on chest CT and lung cancer [11], [12], [13], [14], [15].

The two largest prospective studies were derived from screening trials; after adjustment for smoking history and airflow obstruction, they reported a significant risk of lung cancer when emphysema was detected visually on chest CT [11], [12]. In contrast, two case–control studies found no association with lung cancer when emphysema was assessed quantitatively using a continuous densitometry measure [13], [14]. A subsequent case–control study observed an even stronger effect measure than previous reports when scans were assessed visually by thoracic radiologist [15]. Finally, a study comparing quantitative and qualitative emphysema detection methods by Wilson et al. only found an association with lung cancer when the latter technique was used [16].

A recent systematic review of lung disease and lung cancer risk reported a significant association with emphysema [17]. However, the review included several case–control studies in which the diagnosis of emphysema was based on interview. This may have inflated the observed associations. For example, subjects diagnosed with lung cancer may be more concerned about the health consequences of previous smoking, so they may tend to report emphysema more often than controls. In addition, the accuracy of self-reported diagnosis of emphysema is limited [15]. Furthermore, several of the included studies did not account for airflow obstruction, which is a potential confounder.

Motivated by the mixed conclusions of recent studies with respect to CT-detected emphysema and lung cancer risk, we conducted a systematic review of the literature in an effort to synthesize the available evidence. Particular attention was paid to the method used to detect emphysema on CT and consideration of airflow obstruction. Our objective was to provide an objective synthesis of the evidence, as well as increase precision beyond any single study by pooling data, where appropriate, to provide the most accurate estimate of association between emphysema on CT and lung cancer.