Year in review 2013: Lung cancer, respiratory infections, tuberculosis, cystic fibrosis, pleural diseases, bronchoscopic intervention and imaging


  • Chi Chiu Leung,

    Corresponding author
    1. Tuberculosis and Chest Service, Department of Health, Hong Kong, China
    • Correspondence: Chi Chiu Leung, Wanchai Chest Clinic, 99 Kennedy Road, Wanchai, Hong Kong, China. Email:

    Search for more papers by this author
  • José M. Porcel,

    1. Pleural Diseases Unit, Department of Internal Medicine, Arnau de Vilanova University Hospital, Biomedical Research Institute of Lleida, Lleida, Spain
    Search for more papers by this author
  • Kazuhisa Takahashi,

    1. Department of Respiratory Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan
    Search for more papers by this author
  • Marcos I. Restrepo,

    1. South Texas Veterans Health Care System Audie L. Murphy Division, San Antonio, Texas, USA
    2. University of Texas Health Science Center at San Antonio, San Antonio, Texas, USA
    Search for more papers by this author
  • Pyng Lee,

    1. Department of Medicine, National University Hospital, Singapore
    Search for more papers by this author
  • Claire Wainwright

    1. Queensland Children's Medical Respiratory Centre, Royal Children's Hospital, Brisbane, Queensland, Australia
    2. Queensland Children's Medical Research Institute, University of Queensland, Brisbane, Queensland, Australia
    Search for more papers by this author


community acquired pneumonia


cystic fibrosis


chronic obstructive pulmonary disease


cancer stem cells


computed tomography


radial endobronchial ultrasound


pEBUS-guide sheath


epidermal growth factor receptor


electromagnetic navigation bronchoscopy


health care-associated pneumonia


human immunodeficiency virus


inhaled corticosteroids


intensive care unit




lung clearance index




non-small cell lung cancer


non-tuberculous mycobacteria


N-terminal pro brain natriuretic peptide


parapneumonic effusion


maximum standardized uptake value




This Year in Review examines the most relevant articles published on lung cancer, respiratory infections, tuberculosis (TB), cystic fibrosis (CF), pleural diseases, and interventional pulmonology and imaging in Respirology and other respiratory medicine journals during 2013.

Important findings are highlighted with comments in the following sections, each of which was contributed by an author as named in the beginning of each section.

Lung Cancer

Kazuhisa Takahashi

Relationship between lung cancer and chronic obstructive pulmonary disease

Lung cancer is the leading cause of cancer-related death worldwide.[1] The economic and social burden of chronic obstructive pulmonary disease (COPD) and lung cancer is dramatically increasing worldwide.[2, 3] The World Health Organization estimated that COPD and lung cancer will become the third and fifth causes of death in the year 2020, respectively.[4] COPD is well known to be a risk factor and important coexisting disease for lung cancer. However, the current status of the epidemiological relationship between COPD and lung cancer and the management of COPD in lung cancer patients are not fully understood. Therefore, Zhang et al. investigated several clinical questions described as follows: (i) the rate of diagnosis of COPD in patients with lung cancer; (ii) conformity to the Global Initiative for Chronic Obstructive Lung Disease treatment guidelines for COPD; and (iii) factors that may promote the timely diagnosis of COPD in the lung cancer population.[5] The authors retrospectively reviewed the medical records of hospitalized lung cancer patients, calculated the rate of diagnosis of COPD based on Global Initiative for Chronic Obstructive Lung Disease's spirometric criteria and investigated conformity to the Global Initiative for Chronic Obstructive Lung Disease treatment guidelines. The prevalence of COPD in hospitalized lung cancer patients was 21.6%. The overall rate for the diagnosis of COPD was 7.1%. Very interestingly, the treatment conformity for stable and acute exacerbation of COPD was 27.1% and 46.8%, respectively. Moreover, the rate of diagnosis of COPD was higher among patients with a history of smoking, respiratory diseases and symptoms. These results suggest that COPD is substantially underdiagnosed and undertreated in hospitalized patients with lung cancer. Documenting tobacco exposure and respiratory symptoms would help to improve COPD management.

Cancer stem cells in lung cancer

An excellent review with regard to lung cancer stem cells (CSC) was provided by Alamgeer et al.[6] CSC are believed to be a phenotypically distinct population that possesses a tumourigenic potential. CSC are potential therapeutic targets, as they are considered to be involved in relapse after intensive treatment and innate resistance to chemotherapy and molecular target agents. Alamgeer et al. reviewed the evidence with regard to the existence and function of lung CSC in the context of experimental and clinical studies, and discussed the controversies regarding these models.[6] The remarkable ability of lung cancer to recur despite definitive local and/or systemic therapy suggests that minimal residual disease contains a population of cells with an enormous capacity for self-renewal and regeneration, a biological function generally limited to somatic CSC. In order to detect CSC, identifying specific cell surface markers is essential. However, no universal lung CSC markers have been discovered to date. Although several CSC markers, such as the CD133, CD44 and aldehyde dehydrogenase activity, have been reported to be good candidates for CSC markers,[7-9] none appears to be satisfactory with regard to specificity and sensitivity.

Molecular diagnosis and personalized treatment based on the molecular profile

Over the past few decades, the management of advanced-stage lung cancer has been revolutionized from systemic chemotherapy to the present stage of testing for molecular biomarkers (epidermal growth factor receptor (EGFR) gene mutations, the echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase fusion gene and so on) to select appropriate therapies and improve the efficacy of treatment. Molecular biomarkers for lung cancer may be used in the following situations, in addition to guiding treatment decisions: (i) risk identification using risk-associated markers; (ii) early detection or diagnosis; and (iii) prediction of the treatment response and prognosis. Lam published an excellent review article with regard to the current understanding of molecular targets that are important for personalized therapy for lung cancer, specifically how testing for these molecular targets, namely EGFR, Kirsten rat sarcoma viral oncogene homolog and anaplastic lymphoma kinase, guides therapeutic decision making in the setting of advanced stage lung cancer.[10] He emphasized that multigene analyses could be of use in future individualized therapy for non-small cell lung cancer (NSCLC) and that several mass mutation assays, such as Raindance (RainDance Technologies, Lexington, KY, USA), SNaOshot (Applied Biosytems, Foster City, CA, USA), are already in place. Large prospective studies are ongoing among multiple institutions, including the Lung Cancer Mutation Consortium, performing mass mutation analyses in patients with advanced NSCLC, with specific clinical trials linked to different mutations. Whole genome deep sequencing is coming into application and will provide a more thorough and comprehensive view of lung cancer.[11] The molecular profile of lung cancer, particularly adenocarcinoma, however, differs between patients in Asian and Western countries. For example, the rate of positive EGFR activation mutations is approximately 50% in Japanese patients and less than 30% in patients in the United States.[12] Therefore, it is quite important to establish a comprehensive screening strategy to detect EGFR mutations in Asian patients with lung adenocarcinoma. Ho et al. reported a molecular diagnostic algorithm for EGFR gene mutation detection in Taiwanese patients with adenocarcinoma.[13] A total of 445 adenocarcinoma patients were recruited for an analysis of tumour EGFR mutations using immunohistochemistry with mutation-specific antibodies, polymerase chain reaction-direct sequencing and Scorpion/amplified refractory mutation system methods. In this algorithm, immunohistochemistry with mutation-specific antibodies is used to screen EGFR mutations. Immunohistochemistry-negative samples are subjected to polymerase chain reaction-direct sequencing methods to identify EGFR mutations. polymerase chain reaction-wild samples are further analysed using Scorpion/amplified refractory mutation system technology. Samples negative for EGFR mutations using all methods are further examined for the status of other driver mutations, such as human epidermal growth factor receptor 2 and echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase translocation.

Diagnostic imaging using positron emission tomography scans

A recent publication indicated a relationship between the preoperative maximum standardized uptake value (SUVmax) in positron emission tomography and the EGFR mutation status in lung cancer.[14] In this previous study, EGFR-mutated adenocarcinomas of the lungs presented with low SUVmax values, and the authors suggested that EGFR mutation testing may be omitted in patients with a high SUVmax. However, the relationship between the SUVmax on positron emission tomography and the EGFR mutation status remains controversial. Putora PM et al. investigated the SUVmax values in 14 patients with EGFR mutations and 14 patients with wild-type EGFR adenocarcinomas.[15] The mean SUVmax value was 10.7 in the EGFR-mutated tumours and 9.9 in the wild-type adenocarcinomas. There were no correlations between the SUV values and EGFR mutation status. Therefore, excluding EGFR testing for lung cancer in patients with a low SUVmax is not feasible.


Pemetrexed plus cisplatin has been shown to be efficacious as a first-line treatment for advanced NSCLC. In a recent randomized, non-inferiority, phase III trial, pemetrexed plus cisplatin and gemcitabine plus cisplatin were compared in patients with advanced NSCLC.[16] Based on this study, the pemetrexed plus cisplatin regimen is effective and well tolerated and has therefore been approved by the US Federal Drug Administration for use in patients with non-squamous NSCLC. However, little is known about its efficacy and safety in the East Asian population. Zhang X et al. reported the final analysis of overall survival from a multicenter, randomized phase II study in chemotherapy-naïve Chinese patients with advanced NSCLC.[17] In that study, a total of 254 patients with NSCLC were randomized to receive either pemetrexed (500 mg/m2) plus cisplatin (75 mg/m2) (day 1) or gemcitabine (1000 mg/m2) (day 1 and 8) plus cisplatin (75 mg/m2) (day 1). The median overall survival in the pemetrexed/cisplatin arm was 15.3 months, while that in the gemcitabine/cisplatin arm was 16.9 months (hazard ratio: 1.09; 95% confidence interval: 0.80–1.48, P = 0.488). The patients in the pemetrexed/cisplatin arm exhibited better tolerance than those in the gemcitabine/cisplatin arm. Secondary analysis suggested that the pemetrexed/cisplatin treatment was associated with longer survival among females and patients with non-squamous cell carcinoma.

Respiratory Infections

Marcos I. Restrepo


Much interesting data have been reported during the past year in Respirology focusing on different pneumonia related issues, such as diagnosis, pneumonia characterization and treatment studies.

Pneumonia diagnosis

Two interesting studies evaluated the use of a biomarker and Streptococcus pneumoniae urinary antigen test to improve the diagnosis and assess the prognosis of patients with pneumonia. Horita N et al.[18] used a meta-analysis to assess the sensitivity and specificity of the Binax Now S. pneumoniae urinary antigen test (Binax, Portland, ME, USA) for unconcentrated urine from adult patients with pneumonia. The reference standard was the presence of at least one positive culture and/or smear for S. pneumoniae. The authors suggest that the S. pneumoniae urinary antigen is an important test to rule in (95% specificity), but not perfect to rule out (75% sensitivity) pneumococcal infection, respectively. Lin SC et al.[19] evaluated whether N-terminal pro-brain natriuretic peptide (NT-proBNP) could be used to assess prognosis among pneumonia patients who required intensive care unit (ICU) admission. NT-proBNP levels were statistically significantly higher in patients who died compared with survivors. Further studies are needed to evaluate the clinical application of this novel biomarker in patients with severe community-acquired pneumonia (CAP).

Pneumonia phenotypes

Several manuscripts published during last year focused on health care-associated pneumonia (HCAP), aspiration pneumonia and necrotizing pneumonia. An elegant and thorough review performed by Dobler and Waterer[20] exposed the similarities and differences of the term ‘HCAP’ and their clinical relevance between the United States and the Asia Pacific region. The authors explored the controversies that emerged over the past few years, questioning the risks related to overcall this condition and the need to cover empirically for multidrug-resistant (MDR) pathogens, which may potentially induce overtreatment. The authors suggested that for the Asia Pacific region, an individual MDR risk-factor assessment should be performed rather than applying the broad HCAP classification in order to inform the appropriate selection of empirical antimicrobial therapy. In the same area of research, Taylor and Taylor[21] performed an important evaluation of HCAP patients who received haemodialysis according to the treatment with narrow versus broad-spectrum antibiotic therapy. HCAP patients were more likely to have longer time (by 6 days) to switch to oral therapy and longer length (by 6 days) of hospital stay. In addition, there was a trend towards longer time to achieve clinical stability and longer continuation of intravenous antibiotics after reaching clinical stability in HCAP patients compared with patients treated with CAP-concordant antibiotic regimens. The authors suggested that narrow-spectrum antibiotics might be safe in haemodialysis patients with no other HCAP risk factors. Komiya K et al.[22] evaluated the impact of aspiration pneumonia in patients with CAP and HCAP. Aspiration pneumonia was defined as having both risk factors for aspiration (dysphagia due to a neurological disorder, or disturbance of consciousness) and evidence of gravity-dependent opacity on chest computed tomography (CT). Aspiration pneumonia occurred in 18% of the patients and it was independently associated with increased 30-day mortality. Prompt identification of aspiration risk among patients with pneumonia may predict a poorer outcome. Finally, Seo H et al.[23] described the clinical characteristics and clinical course of patients with focal necrotizing pneumonia as a distinct entity from lung abscess. ‘Focal necrotizing pneumonia’ was defined as a localized type of necrotizing pneumonia characterized by a single or few cavities of low density without rim enhancement on CT scan. The authors concluded that compared with lung abscess, focal necrotizing pneumonia occurred more frequently in non-gravity-dependent segments, had higher rates of aerobic bacteria and had lower incidence of risk factors for aspiration. However, similar to lung abscess, patients with focal necrotizing pneumonia had a high rate of clinical success with appropriate treatment of 97%.

Corticosteroids treatment

Recent studies have raised concerns about the association between the use of inhaled corticosteroids (ICS) and the risk of developing pneumonia in patients with COPD. Zhang L et al.[24] evaluated in a cross-sectional study the association between ICS and S. pneumoniae oropharyngeal colonization among asthmatic children. The authors identified that after adjusting for potential confounders, asthmatic children had a fourfold higher risk of oropharyngeal S. pneumoniae carriage when they received ICS therapy as compared with non-ICS exposed asthmatic children. Lee CH et al.[25] evaluated the risk of hospital admission or emergency room visit for pneumonia in patients using respiratory inhalers such as long-acting β2 agonist and ICS. A large case-crossover Korean national claims database study was used to assess 186 018 inhaler users who visited the emergency department or were admitted to the hospital during a 3-year period. The results showed that the use of an ICS without a long-acting β2 agonist was associated with a 73% increased risk of hospital admission or emergency room visit for pneumonia. In contrast, the use of an ICS and a long-acting β2 agonist was associated with a 37% decreased risk for pneumonia-related hospital admission or emergency room visit. Polverino E et al.[26] described the frequency of the use of systemic corticosteroids in patients with CAP, the reasons for use and the clinical impact on outcomes. The main reasons for administering systemic steroids were the presence of chronic respiratory comorbidity and severe clinical presentation, but therapy did not influence mortality or clinical stability. In contrast, systemic corticosteroid administration was associated with prolonged hospital stay. These results suggest that well-designed randomized clinical trials are needed to evaluate the influence of systemic corticosteroids on clinical outcomes of patients with CAP.

Non-tuberculous mycobacteria (NTM)

Marras TK et al.[27] assessed the associations among CT features, symptoms and microbiological disease criteria in patients with pulmonary Mycobacterium xenopi infection. Seventy consecutive immunocompetent patients with pulmonary M. xenopi isolation and classified according to the American Thoracic Society diagnostic criteria for disease were included in the study. Among them, cavitation and large pulmonary nodules predicted fulfilling microbiological disease criteria, while bronchiectasis and small pulmonary nodules predicted active symptoms. Jeon K et al.[28] evaluated in an age- and gender-matched case–control study the prevalence and degree of severity of vitamin D deficiency in patients with NTM lung disease. After adjusting for potential confounding factors, the authors identified that severe vitamin D deficiency with levels below 10 ng/mL was independently associated with NTM lung disease. However, they suggest that further studies are needed to assess for causality.

Lung transplant infection

Burguete SR et al.[29] performed an elegant review on lung transplant-related infections. The authors emphasized that despite improved survival rates over the decades, lung transplant recipients have lower survival rates compared with other solid organ transplant recipients. The morbidity and mortality following lung transplantation is largely due to infection and rejection-related complications. This comprehensive review focused on the most common infections that develop in lung transplant recipients, including the epidemiology and risk factors, microbiology, diagnosis, prophylaxis, treatment and outcomes.


Chi Chiu Leung

Pathogen and vaccines

A recent analysis of the genomic sequences of Mycobacterium tuberculosis complex isolates from different parts of world suggested that the tubercle bacillus emerged about 70 000 years ago, accompanying migrations of anatomically modern humans out of Africa.[30] The pathogen has developed various mechanisms to allow its successful co-existence with the human host during the long period of co-evolution, and the gap in our current knowledge on the complex microbe–host interactions has been hindering research and development in various aspects of TB control.[31]

Bacillus Calmette–Guérin vaccine offers only partial and unreliable protection against pulmonary TB in adults, the crucial transmission link for this airborne infection.[31] Wang CC et al. reviewed new discoveries in latency antigens and new next-generation candidate vaccines that could promise the possibility of sterile eradication.[32] In the development of TB vaccine, the conventional ‘Identify, Inactivate/Attenuate, Administer’ approach is limited the lack of long-lasting immunity following recovery from natural infection/disease.[31, 32] Systems biology and vaccinomics may help to shorten development of an efficacious TB vaccine targeting both growing bacteria and non-growing persisters through utilization of high-throughput technology, computer modelling and integrative approaches.[32]

Latent TB infection

Targeted screening and treatment of latent TB infection with the currently available immunodiagnostic tools and treatment regimens aims more for personal protection than major epidemiological impact or elimination.[30, 33] Positive predictive values of the immunodiagnostic tests for the development of TB are substantially influenced by the type of test, the age of the person being tested, the prevalence of TB in the society and the risk group to which the person belongs.[33] Although interferon-γ release assay, not being affected by cross-reaction to the bacillus Calmette–Guérin vaccine, has improved the diagnosis of latent TB infection in immunocompetent individuals, its respective sensitivity/specificity values in patients with immune-mediated inflammatory diseases to be started on biologics are unknown owing to variable pretest probability and lack of confirmatory test.[34] The use of interferon-γ release assay for such screening is therefore variable among different countries. Serial interferon-γ release assay, being done in vitro, has a theoretical advantage of not affecting subsequent test response in situations with ongoing TB exposure. However, there is question on its intrinsic reproducibility as well as the actual criteria and implication for test conversion.[31, 35]

Randomized controlled trials have established the protective efficacy of isoniazid therapy for 6–12 months among human immunodeficiency virus (HIV)-infected and non-HIV-infected subjects with latent TB infection, even though controversies remain on the optimal treatment duration.[31, 34] Directly observed weekly therapy with rifapentine plus isoniazid for 12 weeks offers an attractive alternative in both HIV-infected and non-HIV-infected subjects.[34, 36] Uncertainty remains on the role of prolonged isoniazid therapy among HIV-infected individuals.[36] Controversies also persist over the use and the choice of preventive treatment for MDR-TB contacts, largely as a result of the variable risks of developing either drug-susceptible TB or MDR-TB and the currently limited treatment options.[31, 34]

Diagnosis of active disease

An accurate and affordable point-of-care test is needed to address the currently unmet needs in the diagnosis of TB.[37] Controversies remain over TB screening among immigrants from high-prevalence countries, regarding when, where and in whom radiographs are best done for transborder TB control.[38] Enhanced symptom screening and wider application of affordable molecular diagnostics are urgently needed to facilitate early diagnosis of TB in an HIV-infected person.[36]

While some genotypic diagnostic tools, such as the Xpert MTB/RIF assay (Cepheid, Sunnyvale, CA, USA), may show promise as a potential point-of-care test for diagnosis of both TB and drug resistance,[37] genomic sequencing of M. tuberculosis isolates discovered a large number of new genes, intergenic regions and non-synonymous single nucleotide polymorphisms, showing consistent associations with drug resistance, which indicates that the genetic basis of drug resistance is more complex than previously anticipated.[39] In contrast with conventional phenotypic drug susceptibility testing using automated liquid culture system (Bactec MGIT960, Becton, Dickinson and Company, Franklin Lakes, NJ, USA), neither line probe assays nor sequencing were able to detect the clinically important 1% resistant bacteria in simulated patient samplings containing mixtures of rifampicin-resistant and susceptible M. tuberculosis strains.[40] However, highly discordant drug susceptibility test results were observed between Löwenstein–Jensen medium and MGIT960 for M. tuberculosis isolates with certain rpoB mutations (511Pro, 516Tyr, 533Pro, 572Phe and several 526 mutations), with MGIT-drug susceptibility test failing to give a result or declaring the strains susceptible.[41] In another study comparing the MGIT 960 System against Löwenstein–Jensen proportion method, the critical concentrations of moxifloxacin (0.5 μg/mL), levofloxacin (1.0 μg/mL), kanamycin (2.5 μg/mL) and capreomycin (2.5 μg/mL) were concordant and reliable for testing second line drug resistance, while further studies are required for ethionamide and ρ-aminosalicylic acid.[42]

Treatment of active disease

In an observational study by Horita N et al. in Japan, decreased activity of daily living was found to be a strong risk factor for liver injury among adult inpatients treated with a standard regimen for newly diagnosed smear-positive pulmonary TB.[43] To balance the competing risks related to delayed antiretroviral therapy and antiretroviral therapy-related immune reconstitution inflammatory syndrome, early initiation of antiretroviral therapy is indicated, preferably within 2 weeks after starting TB treatment, for patients with a cluster of differentiation 4 cell count of <50 cells/μL, except in case of central nervous system involvement.[36]

Globally, 3.7% of new cases and 20% of previously treated cases have MDR-TB, but the scale-up of programmatic management of drug-resistant TB remains slow.[42] The overall treatment success rate of MDR-TB cases notified to World Health Organization in 2009 was only 48%, far below the global target of 75%.[44] Management strategies for established cases of difficult MDR-TB/extensively drug-resistant TB rely on harnessing existing drugs (notably newer generation fluoroquinolones, high-dose isoniazid, linezolid and pyrazinamide with in vitro activity) in the best combinations and dosing schedules, together with adjunctive surgery in carefully selected cases.[45] Linezolid 300 mg daily[46] or intermittent administration of higher doses[47] may help to reduce its often treatment-limiting toxicity. Bedaquiline has recently been approved by the US Federal Drug Administration to treat adults with MDR-TB when other alternatives are not available, but concerns remain over possible fatal arrhythmia risk associated with prolonged Q-T interval.[48] The use of delamanid for 6 months or more significantly improved outcomes and reduced mortality in MDR-TB as compared with its use for 2 months or less in an observational study, but the European Medicines Agency considers the currently available evidence inadequate for substantiating use of delamanid for treating MDR-TB.[49]

Cystic Fibrosis

Claire E. Wainwright

Genes and gene modifier therapies

In January 2012, the US Federal Drug Administration approved the registration of ivacaftor for patients with CF aged ≥6 years carrying at least one copy of the G551D mutation. Ivacaftor is a potentiator and increases the open probability of the G551D protein. The G551D mutation is a gating mutation that results in a normal quantity of protein available in the plasma membrane but with reduced open probability of the channel. The impressive and sustained improvements in lung function and weight and reduction in sweat chloride levels reported in the first phase 3 trial were mirrored in the second of the phase 3 trials which recently report the paediatric results that supported the Federal Drug Administration registration of ivacaftor.[50] While the prevalence of patients carrying a G551D mutation varies, on average, approximately 4% of patients carry the mutation. This therapy has heralded a new dawn of personalized medicine specific for genetic mutations for patients with CF. The most common, the CF transmembrane regulator mutation, a deletion of phenylalanine at position 508 in the nucleotide binding domain 1 of CF transmembrane regulator (ΔF508) is carried on at least one allele by around 66% of patients with CF worldwide. This mutation leads to impairment and disruption of protein folding leading to degradation in the ubiquitin proteasome system, and as a result, there is almost no functional protein expression at the plasma membrane. Complex approaches appear to be required to chaperone and rescue the ΔF508 protein. Okiyoneda T et al. describe the molecular targets of available correctors and how a combination of mechanism-based approaches will be needed to more fully correct the ΔF508 protein.[51]

While the ΔF508 is the most common mutation, there are now almost 2000 variants of CF transmembrane regulator reported and understanding which of these variants are actually disease causing and the functional consequences of the mutations will be key in both establishing appropriate diagnoses through newborn screening and enabling mutation specific treatment approaches in the future. Sosnay PR et al. used a combination of genotype and phenotype data from 39 696 individuals with CF in registries from Europe and North America.[52] They evaluated 159 variants with allele frequency ≥ 0.01% and assessed for clinical and functional criteria to identify 127 disease-causing mutations. For 32 mutations that were unable to be characterized, they performed a novel disease penetrance study in 2188 fathers of patients with CF and were able to assign 12 as non-disease variants.

Development, measurement and prevention of lung disease

End-stage lung disease in CF is characterized by both bronchiectasis which increases in prevalence and severity with age, and air trapping which occurs frequently in early infancy.[53, 54] 2013 has seen two landmark studies that shed new understanding of early lung disease in CF. Sly PD et al. reported that free neutrophil elastase activity in bronchoalveolar lavage at 3 months of age was associated with persistent bronchiectasis at 12 and 36 months of follow up.[55] These important results provide new opportunities for the development of interventions aimed at preventing early lung disease. In a paradigm shifting paper, Adam RJ et al. have demonstrated that air trapping and airflow obstruction occur in newborn CF piglets prior to any airway infection and inflammation suggesting that CF is associated with congenital airway abnormalities which are likely to impact on the development of lung disease.[56]

Clinical trials in young children are complicated by the difficulty in finding objective, non-invasive outcome measures. Lung clearance index (LCI) measured using a multiple breath washout detects ventilation inhomogeneity which is a recognized consequence of CF lung disease. LCI has been shown to predict subsequent pulmonary exacerbations in children aged 5–19 years with CF.[57] A statistically significant treatment effect in favour of inhaled hypertonic saline on LCI performed at a single centre as a substudy of the randomized controlled Infant Study of Inhaled Saline trial suggested that LCI may be a feasible outcome measure in young children.[58] The United Kingdom CF Gene Therapy Consortium conducted a multicentre observational study of patients aged ≥10 years at the start and after completion of treatment of a physician-defined pulmonary exacerbation using intravenous antibiotic therapy. The study was designed to identify and investigate changes in 46 different assays spread across domains of lung structure, lung function, clinical symptoms, and pulmonary and systemic inflammatory markers.[59] Improvements were detected predominantly in large airway parameters and systemic inflammation in response to the acute intervention, and the use of LCI was extended to a multicentre setting with suggestion that it may be better suited as an outcome measure in milder lung disease.

A recent randomized controlled trial of inhaled glutathione given twice daily over 6 months in patients ≥8 years aimed at reducing oxidant stress and inflammation in the airways showed no clinically relevant benefits in lung function or markers of oxidative stress or inflammation.[60]


Survival has improved dramatically over the last 30 years in patients with CF, and in many countries, there are now more adults than children with CF. As a result, there are increasing numbers of adult patients with severe lung disease and limited respiratory reserve who are living much longer making the timing of lung transplantation more difficult and acute respiratory deterioration in such patients present significant challenges for health-care teams. A retrospective study from the Royal Brompton Hospital in London that looks after 750 adult patients with CF reported their experience, in 30 patients requiring endotracheal intubation and ventilation over 34 occasions in the ICU predominantly for pneumothorax, massive haemoptysis or an acute infective exacerbation over a 6-year period.[61] There was an overall mortality of 53% either in ICU or before hospital discharge. Survival was better for those who were intubated for pneumothorax or massive haemoptysis (13/22) compared with those admitted for an acute infective exacerbation (4/12). Interestingly, acute disease markers did not predict survival and only osteoporosis and the fall in body mass index over the preceding 2 years was associated with reduced survival highlighting the importance of nutritional outcomes in the long-term management of patients with CF. An editorial by Reid and Bell in the same edition of Respirology[62] discussed the similar experience of ICU care in a large Australian CF centre noting that ICU care is more commonly used for post transplant management and in some centres extracorporal membrane oxygenation is being used as a bridge to lung transplant.

Chest physiotherapy is a mainstay of treatment in CF; however, there has been little evidence in the literature to date to assist in choosing the optimal airway clearance method. McIlwaine MP et al. have greatly contributed to our understanding in their study which reported that the use of positive expiratory pressure mask was associated with reduced numbers of pulmonary exacerbations along with an increased time to first exacerbation and a shorter treatment time compared with high-frequency chest wall oscillation in a randomized multicentre trial.[63] Variation in health outcomes across different centres and institutions is well recognized in CF and for many chronic diseases. In 2008, the Physiotherapy for Cystic Fibrosis in Australia Consensus Statement was developed to assist in providing an evidence-based framework with the goal of improving health outcomes. Holland and Button undertook a study to assess both knowledge and uptake of the recommendations. Physiotherapists who were providing care to CF patients in Australia were invited to complete an online survey and 68 (49%) took part. There was excellent knowledge and consensus around common areas of treatment including infant airway clearance techniques; however, there was a wide range in knowledge (16–94%) across different questionnaire items. Both specialized knowledge such as the need to measure blood glucose measurements with exercise in patients with CF-related diabetes using insulin and knowledge outside CF Specialist centres was also poorer.[64] This study will greatly assist in the development of improved dissemination strategies for clinical guidelines and consensus statements in the future.

Airway infection

Flight WG et al. highlighted the importance of respiratory viral infections in CF.[65] They followed 100 adults with CF over 12 months and demonstrated that respiratory viral infections were associated with increased morbidity and an increased risk of exacerbation.[65] The emerging problem of infection with NTM and in particular with Mycobacterium abscessus complex does not appear to be related to either chronic use of macrolides[66] or the use of inhaled therapies including inhaled steroids, dornase alpha and antibiotics.[67] On the other hand, shared strains of bacteria despite the careful adoption of conventional cross-infection measures continue to be a concern and suggest between-patient transmission. Bryant JM et al. reported likely between-patient transmission of M. abscessus complex based on whole genome sequencing of 168 consecutive isolates from 31 patients at the Papworth CF Centre in Cambridge.[68] Kidd TJ et al. reported the widespread detection of shared strains of Pseudomonas aeruginosa across 18 Australian CF Centres with 62% of the 983 patients who provided 2887 sputum isolates having one of the 38 shared genotypes. There were two predominant genotypes detected one of which was associated with increased health care utilization.[69] Understanding transmission of these infections will be key in developing more effective cross-infection policies in the future.

The future

Moodley et al. recently reviewed the progress made and ongoing challenges in understanding and developing the use of stem cells for lung repair. While therapeutic use is still some way into the future, enormous progress has been made in differentiating pluripotent cells into lung cells and provides hope for the future.[70]

Pleural Diseases

José M. Porcel

Pleural effusions in the ICU

An excellent review described the limited evidence that exists on the aetiology and physiological effects of pleural effusions in patients admitted to ICU.[71] For those who are mechanically ventilated, especially, many factors may alter normal pleural fluid turnover and lead to pleural fluid accumulation. About half of the critically ill patients show evidence of pleural effusions by ultrasonography.[71] The range of potential causes is wide, among which hypoalbuminaemia, fluid overload, heart failure, atelectasis, postoperative effusions and pleural infections are thought to account for many cases. Pleural effusions in this group of patients can have significant deleterious effects on respiratory mechanics and gas exchange and, when large in size, may also affect hemodynamics. Therapeutic thoracenteses under ultrasound guidance improve symptoms, lung function and oxygenation in the majority of cases.[71] However, the effects of systematic pleural drainage on the duration of mechanical ventilation await clinical trials.

Pleural procedures

Pleuroscopy, which is performed under conscious sedation and local anaesthesia, is usually the last step in establishing the diagnosis of unexplained exudative effusions. It is traditionally performed with rigid instruments, but in the last decade semi-rigid thoracoscopes, which are handled similarly to a bronchoscope, have attracted the interest of pulmonologists. A prospective randomized pilot study assigned patients with pleural effusions of undetermined aetiology and/or suspicious of having malignancy to a rigid (38 patients) or semi-rigid (41 patients) diagnostic thoracoscopy.[72] An average of 10 biopsies were taken from the parietal pleural surfaces in each procedure. Attempts to obtain sufficient tissue slightly prolonged the examination when using the semi-rigid as compared with the rigid instrument (36 vs 27 min). Forty-seven patients were ultimately demonstrated to have malignant disease, of whom 32 (68%) had mesothelioma. The diagnostic accuracy of both techniques was similar (100% for the rigid and 97.6% for the semi-rigid thoracoscope).[72] Biopsy specimens obtained by the semi-rigid technique were smaller (11.7 vs 24.7 mm2), but of comparable quality and interpretability by the pathologists. Larger comparative studies are needed before firm recommendations can be made on the procedure of choice for the workup of undiagnosed pleural effusions. Even so, rigid pleuroscopes are still considered the gold standard.

The use of transthoracic ultrasonography to guide pleural procedures (e.g. thoracentesis, chest tube thoracostomy, closed pleural biopsy and pleuroscopy) has lessened the risk of iatrogenic complications. A recent prospective study suggests that the completeness of a 14-point checklist before transthoracic ultrasonography-guided procedures may further increase patient safety.[73] The checklist encouraged a systematic review of the procedural indication, chest imaging, coagulation studies, patient monitoring and complications. The implementation of this pleural checklist decreased the overall procedural complication rate from 8.3% to 1.5%.[73]

Parapneumonic effusions

Pleural bacterial infection is characterized by neutrophil migration and activation of pro-inflammatory cytokines within the pleural space. One study examined the role of interleukin (IL)-17A in the pathophysiology of pleural inflammation caused by bacterial products.[74] It was found that pleural fluid expression of IL-17A was uncommon in patients with effusions other than parapneumonic effusions (PPE). Specifically, IL-17A was detected by using a commercially available ELISA kit in 62% of 29 and 46% of 52 patients with PPE from two different cohorts.[74] Notably, IL-17A expression could not be demonstrated in any patient with neutrophil-predominant effusions unrelated to pneumonia, which raises the possibility of using IL-17A as a specific biomarker of PPE in this setting. In a complementary experiment, intrapleural injections of IL-17A in mice elicited a neutrophil-predominant inflammatory response, while IL-17A neutralization partially abrogated pleural neutrophilia induced by bacterial products.[74]

A prospective study investigated long pentraxin-3, which is produced locally in response to pro-inflammatory stimuli, in patients with pleural effusions of different aetiologies (24 PPE, 40 malignant and 20 miscellaneous exudates).[75] Pleural fluid pentraxin-3 levels ≥12 ng/mL yielded 88% sensitivity, 73% specificity, likelihood ratio positive of 3.3 and negative of 0.17 for diagnosing PPE, with an area under curve of 0.855.[75] In a logistic regression model, pentraxin-3 remained as the only significant predictor of PPE in comparison with other pleural fluid analytes (including neutrophil predominance) and serum C-reactive protein. Unfortunately, neither the IL-17A[74] nor the pentraxin-3[75] studies were powerful enough to evaluate the role of these molecules in the most challenging clinical setting, that is, the identification of those non-purulent PPE needing chest tube drainage.

Heart failure-related effusions

Natriuretic peptides are biomarkers of heart failure. Previous studies had shown that pleural fluid levels of the NT-proBNP were useful in establishing the aetiology of a pleural effusion as being heart failure.[76] In addition, the diagnostic accuracy of NT-proBNP for this purpose was shown to be superior to that of the BNP.[76] Recently, the levels of another natriuretic peptide, the mid-region pro-atrial natriuretic peptide, were measured in the pleural fluid of 95 patients with heart failure and 90 with non-cardiac effusions, and compared with those of NT-proBNP.[77] Pleural NT-proBNP and mid-region pro-atrial natriuretic peptide had similar operating characteristics for the diagnosis of heart failure-related effusions, with respective area under curve of 0.935 and 0.918.[77] The optimal cut-off values for NT-proBNP (1700 pg/mL) and mid-region pro-atrial natriuretic peptide (260 pmol/L) were influenced by age, renal function and body mass index. Pleural fluid levels above these cut-offs argued for heart failure (likelihood ratio positive >5), while levels below significantly decreased the probability of having cardiac effusions (respective likelihood ratio negative of 0.10 and 0.19). Finally, NT-proBNP and mid-region pro-atrial natriuretic peptide correctly identified 83% and 69% of 29 cardiac effusions misclassified as exudates by Light's criteria, respectively.[77] In summary, if clinicians choose pleural fluid specimens for natriuretic peptide testing, it is recommended that NT-proBNP be selected based on supporting evidence and technical reasons.

Bronchoscopic Intervention and Imaging

Pyng Lee


Czarnecka and Yasufuku published a comprehensive review on diagnostic bronchoscopic techniques for targeting peripheral pulmonary nodules or masses, mediastinal staging of thoracic malignancies and detection of early central airway cancers as well as pleuroscopy for the evaluation of pleural effusions.[78] In the largest postal survey about bronchoscopic practice in Japan,[79] questionnaires were mailed to 538 facilities accredited by Japan Society for Respiratory Endoscopy, and 95% responded. Bronchoscopy was performed by pulmonologists in 64%; surgeons in 5% and both pulmonologist and surgeons in 31% of the facilities surveyed. Majority of diagnostic bronchoscopy was conducted in an outpatient setting. Written informed consent was obtained in 96.8%; verbal consent in 2.8% and not obtained in 0.4%. Prebronchoscopy tests were checked in 80%, fluoroscopy utilized in 99.8% and post-transbronchial lung biopsy chest X-ray in 70%. While intravenous sedatives were used in 36%, atropine was administered in 67.5% of facilities. Most bronchoscopies were performed via the oral route and biopsy performed after antiplatelet drugs, and anticoagulants have been discontinued in 97%. Semiflexible pleuroscopy and convex endobronchial ultrasound-guided transbronchial needle aspiration under local anaesthesia were commonly performed except rigid bronchoscopy in 18.5% of facilities. The authors noted that there were improvements from 2006 survey[80] with use of atropine decreased from 92% to 67.5%, cessation of antiplatelet drugs and anticoagulants before biopsy, and intravenous sedatives in 36%. Higher frequency of antibiotic administration, prebronchoscopy testing and via oral route signalled practice differences between Japan and other countries.

Stather DR et al. conducted a retrospective review of consecutive pulmonary procedures performed by an interventional pulmonologist from July 2007 to April 2011.[81] More than 50% of 1100 procedures were advanced diagnostic procedures such as endobronchial ultrasound-guided transbronchial needle aspiration, electromagnetic navigation bronchoscopy (ENB) and/or peripheral endobronchial ultrasound. A trainee participated in 84% of these procedures with a trend towards increased complication rates in the trainee versus no trainee groups (4.7% vs 1.1%, P 0.076). Significant differences in procedural length in favour of the no-trainee group (58.32 min vs 37.69 min, P 0.001) as well as dose of propofol (178.3 mg vs 137.1 mg, P 0.002) were noted. The authors opined that in an academic interventional pulmonology centre that practiced the apprenticeship model, trainee participation led to increased procedure time, higher doses of sedation and a trend towards increased complications, incorporation of bronchoscopy simulators and low-fidelity models into the training curriculum could reduce the burden of procedural learning on patients.[81-83]

Radial endobronchial ultrasound (pEBUS) has improved the diagnostic yield of transbronchial biopsy of peripheral pulmonary lesions. Two measurable components are diagnostic and visualization yields which are intimately related as the diagnostic yield is demonstrably higher if the lesion is visualized on pEBUS.[84] Tay JH et al. performed a retrospective analysis of 196 consecutive patients who underwent pEBUS. Size and distance of the lesion from the hilum and pleura were measured with CT scans. A definitive diagnosis was established in 56% of patients although the endobronchial ultrasound visualization yield was 79% indicative of diagnostic gap. pEBUS visualized lesions gave higher diagnostic yields (65%) than pEBUS-invisible lesions (20%; P 0.0001). Malignancy, lesions larger than 20 mm and within 50 mm from the hilum were factors associated with higher visualization yield. Clinicians should keep these in mind during the evaluation of patients with peripheral pulmonary lesions.[85]

A previous study found that the path to the pulmonary nodule (≤30 mm) was the most significant factor in influencing the diagnostic accuracy,[86] and navigation technology may aid in targeting the pulmonary nodule. Tamiya M et al. combined pEBUS with guide sheath (GS), thin bronchoscope and LungPoint (Broncus Medical, Inc. Mountain View, CA, USA), a virtual bronchoscopic navigation system for small pulmonary nodules.[87] Sixty-eight consecutive patients were recruited and LungPoint was used to identify the bronchus leading to the pulmonary nodule. Overall, diagnostic yield was 77.9%; 83.7% for malignancy and 68.0% for benign disease. Univariate and multivariate analyses showed that pEBUS localization was the most significant contributor to diagnostic yield especially if the probe was within the lesion (92%) compared with adjacent to the lesion (60%). In another study,[88] pEBUS-GS was performed first to localize peripheral pulmonary lesions, with as needed ENB if pEBUS-GS failed. Sixty patients were enrolled, average lesion size was 27 mm and mean pleural distance was 20 mm. pEBUS detected 75% of peripheral pulmonary lesions. Addition of ENB improved lesion visualization to 93%. However, diagnostic yields for pEBUS and combined pEBUS with ENB were not statistically different (43% vs 50%). Predictive factors for ENB were smaller peripheral pulmonary lesions and absent CT air bronchus sign. Although ENB improved localization, it was not adequate to achieve diagnosis, and improvements in sampling methods could bridge the gap between localization and diagnosis.[89]

Fruchter O et al. evaluated a novel sampling technique in a group of lung transplant recipients.[90] Bronchoscopic lung biopsies remain the gold standard for assessment of allograft rejection or infection. Forty patients underwent cryo-transbronchial biopsies and were matched against 40 controls with forceps-transbronchial biopsies. There were no major complications. The mean diameter of cryo-biopsy was 10 mm against 2 mm with forceps-transbronchial biopsies (P < 0.05). This increase in size and quality of cryo-biopsies translated to increase in alveolated tissue (65% vs 34%, P < 0.05) that enabled histological diagnoses of acute rejection, pneumonitis, diffuse alveolar damage and confident exclusion of acute rejection, infection or pneumonitis. In addition, fluoroscopy time was shorter in patients undergoing cryo-biopsy compared with controls (25 s vs 90 s, P < 0.05).

Fifty-one patients with both malignant and benign central airway obstruction underwent microdebrider bronchoscopy.[91] The lesions involved the trachea, main stem bronchi and bronchus intermedius with some patients having more than one. Satisfactory recanalization of central airway obstruction was achieved with microdebrider. There were no major complications: one patient developed pneumomediastinum while the other required stent removal, and two patients died within 30 days due to unrelated causes.

Retrospective chart review was conducted for 46 patients suspected of broncholithiasis.[92] Cough occurred in 46%, haemoptysis in 41% and purulent sputum in 24%. Cough was more common in intraluminal broncholiths. The broncholiths were classified based on CT and bronchoscopic findings as intraluminal, mixed with both intraluminal and extraluminal components and extraluminal. All intraluminal broncholiths were removed successfully via flexible or rigid bronchoscopy. However for mixed types, complete removal of the broncholiths was not accomplished with bronchoscopy and 40% required surgery. For those with extraluminal broncholiths, they were surgically resected for symptom control. No surgical morbidity or 30-day mortality was observed. Management of broncholithiasis depends on chest CT and bronchoscopic findings: intraluminal broncholiths can be removed using bronchoscopic techniques, while surgery is considered for symptomatic mixed or extraluminal broncholiths.

This bronchoscopic subsection ends with a case report of chronic silicosis diagnosed on endobronchial ultrasound-guided transbronchial needle aspiration of enlarged hilar lymph node when radiological features were inconclusive of silicosis.[93]


Clinical, radiographical and microbiological findings are considered important in determining NTM infection. The American Thoracic Society guidelines identify pulmonary nodules, cavities and bronchiectasis as acute infection. These criteria have been described for Mycobacterium avium complex, M. kansasii and M. abscessus.[94] CT abnormalities and how they relate to symptomatology and microbiological findings are less well known in other less common species of NTM. Maselli and Fernandez described CT features and correlated them with symptoms and microbiological disease in pulmonary M. xenopi.[95] Seventy consecutive immunocompetent patients with pulmonary M. xenopi isolation were classified according to American Thoracic Society diagnostic criteria as definite, possible and no disease. Mean age was 63 years, and 39% were women. Among patients with ‘definite disease’, CT showed nodules in 88%, cavities in 63% and bronchiectasis or tree-in-bud in 50%. All patients with ‘possible’ disease had nodules and 40% had bronchiectasis or tree-in-bud. For those with ‘no disease’, CT also showed nodules in 80%, bronchiectasis in 18% and tree-in-bud in 11%. Cavitation and nodules ≥ 5 mm satisfied microbiological criteria for disease, while bronchiectasis and nodules <5 mm were associated with symptoms of infection. The study by Marras TK et al. also sheds light on how CT abnormalities predict microbiological and clinical findings of M. xenopi infection in immunocompetent patients.[27] NTM infection should be considered in the diagnostic workup of patients with cavitary lesions affecting the upper lobes.

Chest tube insertion can be life saving but is a high-stake procedure. In fact, one study showed that 10% of clinically identified sites for pleural intervention would have led to organ perforation.[96] Ultrasound guidance is currently recommended for all pleural procedures.[97] Although ultrasound guidance has reduced the risk of organ perforation, intrapleural haemorrhage from intercostal artery trauma from chest tube insertion remains a significant concern, and such bleeding can be uncontrollable given the negative intrapleural pressure and inaccessibility to external compression. The intercostal artery is exposed within 6 cm of the spinal column as confirmed by cadaveric dissection, CT studies and thoracoscopic observation,[98, 99] and guidelines recommend mid-axillary chest tube placement within the ‘safe triangle’ where the rib protects the intercostal artery.[97] Anatomical variation can however occur. Salamonsen M et al. evaluated the sensitivity of thoracic ultrasound to detect the intercostal artery compared with CT.[100] Fifty patients underwent ultrasound examination to visualize the vessel at three positions labelled with radio-opaque fiducial markers before contrast enhanced CT. The vessel was unshielded by a rib according to CT in 114 of the 133 positions. The sensitivity, specificity and negative predictive values of portable ultrasound to image the vessel when it was within the intercostal space on CT were 0.86, 0.30 and 0.27, respectively. Time required for a pulmonologist to locate the vessel was 42 s and 18 s using the portable and high-end ultrasound respectively, and there was no difference in performance between two ultrasound systems. The authors concluded that portable ultrasound can be used to screen for the intercostal artery vessel quickly in a busy clinical practice, and recommended its use before any planned pleural procedure to minimize trauma and complications. However, the findings need to be further validated prospectively in a robustly designed study.[101]


C.W. has protected time supported by a Queensland Health Office of Health and Medical Research Fellowship. C.W. thanks Scott Bell, Felix Ratjen, Margaret Rosenfeld, Keith Grimwood and Peter Wark for suggested manuscripts to include in the CF review.