Year in review 2015: Lung cancer, pleural diseases, respiratory infections, bronchiectasis and tuberculosis, bronchoscopic intervention and imaging

Kazuhisa Takahashi Approximately 1.5 million new lung cancer cases were diagnosed globally in 2014, and 80% were non-small cell lung cancer (NSCLC). Five-year survival for NSCLC remains dismal at 15% due to delayed diagnosis and ineffective treatment; however, if NSCLC is diagnosed early, the 5-year survival rate increases to 80%. Fong and van Zandwijk edited a series of reviews that focused on recent advances in diagnosis and treatment of lung cancer.1 Wang et al.2 reported a meta-analysis on micro-RNA (miRNA) as a non-invasive biomarker for NSCLC. Pooled sensitivity and specificity of miRNA for NSCLC were 0.75 and 0.79, respectively. Pooled positive likelihood ratio was 3.6, negative likelihood ratio was 0.32 and the diagnostic odds ratio was 12. The area under the curve (AUC) was 0.84. Interestingly, miRNA was more accurate for Caucasians than for Asians, and multiple miRNA assays were more accurate than single assay (AUC: 0.89, sensitivity: 0.83, specificity: 0.82 vs AUC: 0.81, sensitivity: 0.77 and specificity: 0.71). Serum miRNA assays were found to be more accurate than those from sputum, which led them to conclude that multiple serum miRNA assays might be a potential biomarker for NSCLC because of its relatively high diagnostic accuracy. Although fluoro-deoxy-glucose positron emission tomography/computed tomography (FDG-PET/CT) is widely used for staging of lung cancer, there are no accepted criteria for classifying thoracic nodes as malignant. Nguyen et al.3 proposed to use the maximum standardized uptake value (SUVmax) of the lymph node (SUVmaxLN), the ratio SUVmaxLN/SUVmax of the primary lung malignancy, the ratio SUVmaxLN/SUVaverage of the liver, the ratio SUVmaxLN/SUVmax of the liver and the ratio SUVmaxLN/SUVmax of the blood pool as criteria and validated this model by comparing against cytology results obtained from endobronchial ultrasound (EBUS) lymph node aspirations as well as expert visual interpretation by nuclear medicine physician. The investigators found that FDG-PET/CT-based evaluations of hilar and mediastinal lymph node metastases were useful and highly accurate but did not exceed expert visual interpretation. Implementation of evidence-based guidelines for lung cancer staging depends on expertise and resources, and FDG-PET/CT may not be available in developing countries. Some authors called for regional respiratory societies such as the Asian Pacific Society of Respirology (APSR) to take the lead in formulating minimum standards and encouraging cost-effective staging strategies such as incorporating EBUS, with national respiratory bodies.4, 5 Lau and coworkers also highlighted that not all clinically important diagnostic and staging technologies are accepted as standard of care in all Asian countries because of financial and reimbursement constraints. Several authors have stressed the importance of multidisciplinary team (MDT) approach to lung cancer management. Prabhakar et al.6 emphasized the role of the respiratory physician, while Campbell et al.7 reported an increase in lung cancer patients receiving radiotherapy when managed by MDT (66% vs 33%, P < 0.001). Kidane et al.8 debated about sublobar resection versus standard lobectomy for patients with stage I NSCLC and expounded on the role of surgery for T4 NSCLC with spinal invasion. Molecular genetics have led to the recognition of distinct subsets of NSCLC and targeted therapies. Routine tumour genotyping for epidermal growth factor receptor mutations and anaplastic lymphoma kinase rearrangement in stage 3 or 4 lung adenocarcinoma has relegated chemotherapy to second choice given only if there is no drug for driver oncogene. Obtaining tissue for epidermal growth factor receptor mutations and anaplastic lymphoma kinase rearrangement is an essential step towards personalized therapy for patients with advanced NSCLC. Comprehensive genomic study is able to detect multiple driver genes of which 75% are drug targetable within a short time, and new monoclonal antibodies that inhibit immune checkpoint pathways show promise in NSCLC.9 Nakanishi reported on the adoption of modern therapeutic approaches and research for NSCLC in Japan,10 which has witnessed more collaborative style clinical trials since the 1990s. José M Porcel Lymphangioleiomyomatosis and thoracic endometriosis are listed among the causes of secondary spontaneous pneumothorax. The former is a rare cystic lung disease affecting young and middle-aged women presenting with recurrent pneumothorax or chylothorax. In a Japanese database of 280 hospitalized patients with lymphangioleiomyomatosis, whose mean age was 41.7 years, 38% of hospitalizations were due to pneumothoraces, of which 30% were recurrent.11 Incidence of pneumothorax was 28%, chylothorax 4.6% and pleural effusion 2.5% over 33 months. Thoracic endometriosis can present as catamenial or non-catamenial pneumothorax, depending on its onset in relation to menstrual cycle. In a series of 150 women with thoracic endometriosis-related pneumothorax, 37% were catamenial, and all except one affected the right lung.12 Although all patients had endometrial deposits on the diaphragm, a significantly higher proportion in the catamenial group (62% vs 44%) also had pleural deposits, thereby suggesting that ectopic endometriosis may influence the timing of pneumothorax episodes. Lung cancer accounts for more than one-third of all malignant pleural effusions. In a comprehensive review of 556 consecutive lung cancer patients, it was found that about 40% developed pleural effusions during the course of the disease, of which half were too small to be tapped (minimal effusions) and did not progress so as to require a pleural intervention.13 The pleural effusions were malignant in 83% of 113 patients who were subjected to a diagnostic thoracentesis. Eighty-four percent of these malignant effusions needed palliative procedures, either therapeutic thoracenteses, pleurodesis or tunnelled pleural catheters. Survival was shortest among those with malignant pleural effusions (5.5 months) compared against those without pleural effusions (12.65 months) and minimal effusions (7.5 months). About 70% of cell blocks prepared from pleural fluid of 30 lung adenocarcinomas stained positive for thyroid transcription factor-1 and napsin A but negative in remaining non-lung cancer and benign effusions.14 Although there was no correlation between expression and concentration of these markers as measured by enzyme-linked immunosorbent assay, two (33%) of six double negative (thyroid transcription factor-1 and napsin A) stained lung adenocarcinomas had pleural fluid napsin A levels above 3.32 µg/mL, a 100% specific threshold for malignancy. Napsin A in the pleural fluid supernatants of patients with negative immunocytological examination warrants further evaluation. Mesothelioma is an aggressive malignancy with limited treatment options. Among targeted therapies are agents that inhibit mammalian target of rapamycin. Using a syngeneic murine mesothelioma model, temsirolimus (mammalian target of rapamycin inhibitor) curtailed mesothelioma growth by inducing tumour cell apoptosis, inhibiting tumour angiogenesis and modulating host immune response.15 Whether temsirolimus combined with standard chemotherapy (cisplatin with pemetrexed) would improve overall survival of mesothelioma patients has yet to be explored. Pleural fluid cultures identify pathogens in 60% of empyemas. Bacteriology differs between community-acquired and hospital-acquired pleural infections. Klebsiella oxytoca, which is primarily a pathogen in long-term care facilities and hospital settings, has been reported for the first time as a causative agent in 19 patients with empyema.16 The cases were nosocomial in nearly 90%, polymicrobial in 70% and two-thirds had a history of cancer. All isolates were resistant to amoxicillin/ampicillin.16 Obtaining a definitive diagnosis of tuberculous pleuritis may be challenging due to the paucibacillary nature of the disease and the limitations of conventional microbiological tests. Therefore, many pleural fluid biomarkers have been explored, and adenosine deaminase (ADA) is most robust. The diagnostic value of cytokines, chemokines and effector molecules of T-lymphocytes was assessed in 95 patients with lymphocyte-predominant exudates (35 tuberculosis (TB), 46 malignant and 14 others).17 Interferon-g ≥75 pg/mL and ADA ≥40 IU/mL were found to be highly discriminatory, with AUC of 0.905 and 0.855, respectively. Adding decoy receptor-3 and soluble tumour necrosis factor receptor-1 improved the surprisingly low sensitivities of interferon-g (66%) and ADA (40%) in isolation to 83%, at the expense of 10% reduction in specificity. Nevertheless, testing these four biomarkers simultaneously is not justified, based on the reported success with ADA. In systemic lupus erythematosus, a disease charaterized by multiorgan involvement, the prevalence of pulmonary manifestations was found to be 33% among 184 patients from Western Saudi Arabia.18 The most common was pleural effusion that affected 50% of cases and was bilateral in 60%. Although lupus pleuritis accounted for more than half of pleural effusions, TB, pneumonia and transudates should be considered, especially when the disease is not active.19 Solitary fibrous tumours of the pleura (SFTP) are rare neoplasms that arise from the mesenchymal cells beneath the mesothelium. A retrospective study of 80 patients who underwent surgical resection for SFTP showed that 59% had respiratory complaints, constitutional symptoms and/or hypertrophic pulmonary osteoarthropathy.20 About 80% arose from visceral pleura, 83% were pedunculated and 65% were classified as benign. Twenty percent of SFTP had pleural effusions, more common encountered in malignant than benign SFTP (40% vs 6%). Ten-year survival for benign and malignant SFTP was 88% and 80%, respectively. Surgical resection with clear margins was achieved in 93%, and recurrence was 7%. Marcos I Restrepo, Chi Chiu Leung, Yuanlin Song Three publications addressed the importance of identifying a pathogen in different populations of patients with pneumonia. Shi et al.21 performed a systematic review of the literature to determine the incidence of viral infection in childhood community-acquired pneumonia (CAP) using polymerase chain reaction or real-time polymerase chain reaction techniques. The authors found that respiratory viruses are widely detected in childhood CAP, but the incidence of viral infection is affected by significant heterogeneity among studies. The three most common viruses causing CAP in childhood were rhinovirus, respiratory syncytial virus and bocavirus. In addition, more than half of viral infections are probably concurrent with bacterial infections. These data suggest that there is a need to standardize the techniques to collect samples for respiratory viral pathogens in order to appropriately determine the incidence of these pathogens in children with CAP. Etherton-Beer et al.22 examined the prevalence of resistant Gram-negative bacteria in the oropharynx of 124 residents of 11 long-term care facilities. The authors found only one extended-spectrum b-lactamase producing Escherichia coli isolate representing 0.8% of multi-drug-resistant pathogens in this population. From the data, it suggested that empiric coverage against multi-drug-resistant bacteria might not be warranted in long-term care facility residents. Controversy surrounds the role of bronchoscopy in the identification of pathogens and its impact on clinical outcomes of mechanically ventilated patients. Patients who had an early bronchoscopy within 24 h after intubation had lower intensive care unit (ICU) and 90-day mortality regardless of the initial empiric antibiotic therapy. In addition, patients who survived the event and had an early bronchoscopy were more likely to have shorter duration of mechanical ventilation and length of ICU stay, respectively. The data suggest that bronchoscopic evaluation might improve clinical outcomes among patients with aspiration pneumonia that require mechanical ventilation.23 Previous studies suggested that chronic obstructive pulmonary disease (COPD) led to worse clinical outcome when they developed CAP. Jiang et al.24 performed a systematic review that included cohort and case–control studies composing of 257 958 patients and concluded that COPD was not a risk factor for increased mortality, longer length of hospital stay, admission to the ICU and need for mechanical ventilation in patients who developed CAP. The study by Sibila and colleagues 25 suggested that patients with Pseudomonas aeruginosa CAP had lower 30-day mortality when empiric antibiotics that had antipseudomonas activity were initiated within 48 h of presentation. The authors concluded that there was a need to better characterize risk factors associated with P. aeruginosa CAP in order to institute appropriate antimicrobial therapy to improve their clinical outcomes. Sanz et al.26 hypothesized that partial oxygen concentration/inspired oxygen fraction (PaO2/FiO2) from percutaneous oxygen saturation (SpO2) by the Ellis or Rice equations might correlate with PaO2/FiO2 determined by arterial blood gases. They concluded that the Ellis equation was more accurate than the Rice equation for estimating PaO2/FiO2, especially at higher levels of PaO2/FiO2 ratio, and accurate enough to assess initial oxygenation in patients with CAP. The authors recommended arterial blood gas analysis of oxygenation at the threshold for mechanical ventilation because there was misclassification using Ellis equation among those who had higher PaO2/FiO2 levels. Smith et al.27 performed a systematic review of non-pharmaceutical interventions to prevent transmission of influenza virus in adults and found seven articles with important design flaws. Diligent handwashing and dental hygiene were useful, but other interventions were not fully assessed. Marras and colleagues28 used a Delphi method to survey North American experts regarding clinical aspects related to the care of patients with non-tuberculous mycobacteria (NTM). The survey yielded extensive variability and underscored the need for more research. The editorial29 opined that the survey raised relevant questions regarding care of these patients and served as motivation for research into understanding of NTM disease, its evolution, progression, treatment responsiveness and prognosis. A retrospective cohort study showed that patients with pulmonary NTM had obstructive ventilatory defects and after 12 months of antimycobacterial therapy, there was no remarkable improvement in the pulmonary function tests. Small improvements were observed among patients with asthma, but these findings need to be evaluated further due to small numbers.30 Numata and colleagues31 evaluated 296 patients with suspected Mycobacterium avium complex pulmonary disease and found that anti-glycopeptidolipid-core IgA antibody test had diagnostic sensitivity of 59% and specificity of 98%. Malignancy was associated with false negative results, and the authors concluded that anti-glycopeptidolipid-core IgA antibody should be applied with caution in these patients. Bishwakarma et al. reviewed the epidemiological link between TB and cigarette/biomass smoke exposure.32 While a fairly large number of studies significantly associate both direct and second-hand smoke exposure with tuberculous infection, disease and/or mortality, fewer link biomass fuel smoke exposure and TB. These observational studies, as usual, might not be able to control all confounders and fell short of establishing a direct causal link. In Hong Kong, smoking was also found to adversely affect baseline disease severity, bacteriological response, treatment outcome and relapse in TB in the setting of a territory-wide TB control programme.33 Complex sociobiological interactions are likely involved at multiple stages of the pathogenesis. Further mechanistic studies are desirable to elucidate the possible links, and population-based randomized controlled trials may also be needed to provide convincing evidences for full integration of control and smoking into national TB control 50% of pulmonary TB patients bronchoscopy were found to have endobronchial in a with duration and of TB as than had at and in were risk factors for factors that have to the high observed prevalence of endobronchial TB, results suggest a need for early diagnostic bronchoscopy in patients with pulmonary TB accounted for ICU in and nearly two-thirds was 3 months after nosocomial transmission The nosocomial was related to symptoms and and observed therapy short was observed among isolates of TB in Hong with most of transmission the diagnostic of would be useful to infection control in treatment of TB infection if are to TB by In a study among of TB patients in Hong Kong, test in TB over the 5 of The more specific TB in probably by in that between and was was an for and while age and were for assays TB in The duration of by months of in with to on the risk of research is needed to limitations by available diagnostic and treatment for this highly but guidelines for are which can to outcomes observed with this disease it is in COPD risk factors may and From two studies, most was accounted for about and in the immune about an important role in the of A study the role of and in of patients with and without bacteria levels were higher in those whose yielded P. aeruginosa versus those without bacterial which might to potential of P. aeruginosa with and coworkers bacteria of fluid obtained from patients with and found when compared against those without of was significantly with isolation of in the fluid samples using NTM was also from the of patients with a disease associated with These studies suggest an role of NTM in the of which might be by by that lung on Further is therapy and can be recommended in clinical obstructive pulmonary disease patients who developed higher mortality compared with COPD which suggested that chronic led to and from recurrent A study comparing and tissue found and during of and be potential and might provide against and and were related to the and of as well as in in and have been for patients with of ventilatory however, the is not high given limited clinical Although can use of and and had been to increase of and be In the for the review series in opined that as the to new technologies and there is an need for multidisciplinary approaches for the diagnosis and treatment of pulmonary that require interactions among experts of different From an patients should not the of which the of and evidence-based such as and other and coworkers the for endobronchial aspiration among with bronchoscopy but clinical and h with All a underwent review and by evaluation of their baseline with two test They were assessed after 25 clinical cases with the two test was significant in of with improvement in after clinical The study can to for and A systematic review and meta-analysis were to diagnostic of for in patients with detected on In 14 studies of patients with a prevalence of the diagnostic of was 0.79, sensitivity and specificity lymph node were evaluated for and using were for patients and applied to threshold only benign lymph nodes and malignant cases to that analysis was not accurate in malignant from benign in to study the of endobronchial ultrasound of pulmonary in an to that benign from malignant were pulmonary in the and in the classified 75% of pulmonary that analysis of may in with a high negative value for is a new that the to real-time and coworkers compared of and pulmonary obtained during bronchoscopy against of of the of NSCLC of and with increased which also with cancer These during suggest malignant lung reduction treatment was to be and for patients with in the short and colleagues reported data of patients who underwent lung reduction or were observed on and the improved over years, improvements in and and at 3 years, 50% of the patients improvements in and is a useful for diagnostic evaluation of pleural but with can be challenging when the pleura is and was no significant with or in all patients. and obtained pleural tissue for was than vs with better and tissue and vs Although there was no improvement in diagnostic when was combined with was to be and its use would require further