Year in Review 2016: Respiratory Infections, Acute Respiratory Distress Syndrome, Pleural Diseases, Lung Cancer and Interventional Pulmonology
Marcos I. Restrepo,James D. Chalmers,Yuanlin Song,Christopher Mallow,Justin Hewlett,Fabien Maldonado,Lonny Yarmus
DOI: https://doi.org/10.1111/resp.13016
IF: 6.9
2017-01-01
Respirology
Abstract:Marcos I. Restrepo and James D. Chalmers Respirology has made important contributions related to respiratory infections over the past year. In this review, we focus on bronchiectasis, pneumonia, tuberculosis and non-tuberculous mycobacteria (NTM) infections as the most relevant topics in the area of respiratory diseases. Park et al. reported a study of 155 patients with bronchiectasis to identify the predictors of radiological progression of the disease.1 Treatment in bronchiectasis aims to reduce inflammation and airway infection to prevent further lung damage.2 Therefore, identifying the drivers of poor outcome is important. In this analysis, older patients, those with lower BMI and patients infected with Pseudomonas aeruginosa or NTM were more likely to show radiological progression, measured using the Bhalla score.1 In multivariable analysis, only P. aeruginosa and BMI were statistically significant. The results are supported by existing literature. P. aeruginosa colonization and BMI are both incorporated into the bronchiectasis severity index, a validated prediction tool, and P. aeruginosa in particular is strongly associated with other indicators of progression such as quality of life (QOL), forced expiratory volume in 1 s (FEV1), exacerbation frequency and mortality.3, 4 However, the most important step in the treatment of bronchiectasis is identifying the underlying cause, as many causes such as NTM infection, rheumatoid arthritis, primary ciliary dyskinesia, immunodeficiency or allergic bronchopulmonary aspergillosis require specific treatments.5, 6 Gao et al.7 conducted a systematic review into the underlying causes of bronchiectasis identified in 56 studies (n = 8608).7 The study was limited by the high variability in testing across different studies, and the variable definitions used, but was able to suggest that 18.3% of bronchiectasis patients have an aetiology with a specific treatment.7 Among the different respiratory infection papers published in Respirology, the vast majority were in the area of pneumonia as evidence of this important healthcare priority. We now summarize the most relevant contributions to the journal, focusing on the mechanism of disease secondary to infection, the controversial issue of microbiological diagnosis, biomarkers, treatment, health-related outcomes and complications of patients with pneumonia. Tang et al.8 showed that neutrophils from patients with asthma release C-X-C ligand 8 (CXCL8), neutrophil elastase (NE) and matrix metalloproteinase-9 (MMP-9) in response to viral surrogates. In addition, Toll-like receptor (TLR7/8) dysregulation may play a role in neutrophilic inflammation in viral-induced exacerbations. This evidence suggest that neutrophils carry important immunological properties to directly detect and respond to both viral and bacterial pathogens.8 In addition, out of the three respiratory viruses including rhinovirus, influenza and respiratory syncytial virus (RSV), only RSV was able to activate neutrophils.8 The neutrophil activation and inflammation observed in asthma may be the result of the complex interactions between infectious pathogens and airway epithelium and smooth muscle.8 The authors also suggest that it may not be only the neutrophil activation that plays the leading role during exacerbations, but also the resolution or regulation of neutrophilic inflammation following infection.8 In an Editorial, Vlahos and Bozinovski9 suggest that neutrophils do indeed play an important role in the immune response of patients with asthma. The presence of overactivation of neutrophils during exacerbations increases asthma disease severity and emphasizes the different mechanisms by which neutrophils may contribute to the exacerbation period or the resolution period.9 Appropriate antimicrobial therapy and antimicrobial stewardship require knowledge of the underlying causative pathogen of community-acquired pneumonia (CAP). Several studies published in Respirology attempted to address the issue about the aetiology in patients with CAP. Sahuquillo-Arce et al. used a large Spanish data set to examine the impact of age on the aetiology of CAP.10 The study found that older patients were more likely to have Gram-negative infection, particularly with Haemophilus influenzae and enteric Gram-negative organisms. Co-morbidities were as, or more important than age, with diabetes being associated with pneumococcal and Staphylococcus aureus pneumonia and COPD associated with H. influenzae.10 The study confirms previous reports and points to the increasing problem of the ageing population changing the microbiology of CAP towards an increase in Gram-negative organisms.10 Along with an ageing population, there is also an expansion of patients receiving immunosuppressive drugs, such as patients receiving chemotherapy for malignancy.11 Guidol et al.12 studied 1723 patients with bacteraemia which included 795 patients with neutropenia and underlying malignancy. The most frequent cause was P. aeruginosa and the mortality rate was 46.2%. 12.8% of Gram-negative organisms identified were multidrug resistant.12 This study again emphasizes the need to consider unusual or resistant organisms in patients with immunosuppression.12 Metersky et al.13 presented data from the large USA Veterans Hospital database from 2002 to 2012, and found that a small proportion of patients with risk factors for healthcare-associated pneumonia (HCAP) had Pseudomonas pneumonia (1.9%) and methicillin-resistant Staphylococcus aureus (MRSA) pneumonia (1.0%). In order to stratify who had one pathogen or the other, the authors suggest that MRSA pneumonia patients were more likely to be males, elderly (age >74 years), diabetics, have COPD, recent nursing home or hospital stay, recent exposure to fluoroquinolone or antibiotics treating Gram-positive organisms and severe pneumonia. In addition, patients with Pseudomonas pneumonia were more likely to have a prior hospitalization, immunosuppression, COPD, hemiplegia, recent exposure to inhaled corticosteroids, beta-lactam/cephalosporin/carbapenem antibiotics, antibiotics against Gram-positive organisms, ‘other antibiotics’ and severe pneumonia requiring intensive care unit (ICU) admission, vasopressors or use of invasive ventilation within 48 h of hospital admission.13 These results suggest that the appropriate selection of antibiotics may be driven by identifying the characteristics of patients with HCAP due to Pseudomonas or MRSA. In an elegant Editorial, Waterer14 suggest that the key aspect of antibiotic utilization is to know whether the pathogen causing pneumonia is present at the time of evaluation, rather than identifying whether the patients have HCAP or not. In addition, Dr Waterer14 stresses the point that these pathogen-specific risk factors had a positive or negative predictive value sufficient to reliably determine empiric antibiotic therapy as also recently pointed out in the literature.15 Cillóniz et al.16 presented an elegant review manuscript emphasizing the importance and clinical relevance of polymicrobial infections as the cause of CAP. The authors suggest that polymicrobial infection is an understudied and growing entity with distinct inflammatory, host response and disease-related characteristics that differ from other patients with CAP.16 Rapid identification of pathogens is the best solution to this problem. Diagnosis within hours using PCR or similar methods is not yet fully established, but matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALTI-TOF MS) is now in use in many centres to give a more rapid identification of pathogens. Mok et al.17 examined the impact of MALTI-TOF on their practice in severe pneumonia, showing that identification of pathogens from bronchoalveolar lavage in severe CAP, hospital-acquired pneumonia and ventilator-associated pneumonia was more rapid with MALTI-TOF resulting in earlier de-escalation of antibiotics.17 Specific pathogens can usually not be identified from their presenting features alone. Two papers reported on the characteristics of less frequently identified pathogens in CAP. An organism increasingly identified in hospital-acquired pneumonia is Acinetobacter baumannii and its related pathogens. It is frequently multidrug resistant and is a challenge globally. Özvatan et al.18 reported on 356 adult patients from Turkey of whom the large majority had ventilator-associated pneumonia. Mortality was high (53.1% at 30 days) and reduced by combination antibiotic therapy. Adenovirus is an unusual pathogen in immunocompetent CAP patients, but Yoon et al.19 reported a large case series of 91 patients with adenovirus pneumonia compared with 55 patients without adenovirus detected. Not surprisingly, there were no clinical features that could accurately identify an adenoviral pneumonia. The authors showed that adenovirus can be associated with the broad spectrum of disease, from mild CAP to CAP in the ICU.19 Monocytopenia was the only factor found to be clearly different between adenoviral and non-adenoviral CAP.19 Legionnaires disease is a relatively common cause of CAP, accounting for 1–10% of patients outside of epidemics.20, 21 While Legionella pneumophila is most frequently encountered worldwide, Isenman et al.22 provided a timely reminder that Legionella longbeachae is the predominant cause in some regions including parts of Australasia.22 The disease presents similarly to pneumophila, with a middle age to elderly male predominance, diarrhoea, myalgia, hyponatraemia and abnormal liver function tests—the classical features of Legionnaires disease, all common. Admission to the ICU was common at 25% but outcomes were relatively good if appropriate antibiotics were used.22 Procalcitonin (PCT) is thought to be a useful marker when trying to decide whether to withhold antibiotics from low-risk patients with respiratory infections. Ito et al. tested whether PCT could be used to predict mortality. Ito et al. 23 found that C-reactive protein, PCT and confusion, urea > 7 mmol/L, respiratory rate ≥ 30 breaths/min, low blood pressure (systolic < 90 mm Hg or diastolic ≤ 60 mm Hg) and age ≥ 65 years (CURB-65) were predictive of 30-day mortality, and PCT was additive to CURB-65. In addition to markers of systemic inflammation, the journal presented data on markers of coagulation in pneumonia with the latest in series from Cangemi et al.24 They showed in 104 patients that multiple markers of thrombosis were altered in acute pneumonia with evidence of prothrombotic state. This may contribute to the excess risk of cardiovascular disease that is consistently reported in patients with CAP.25 Another study published by Liu et al.26 presented a systematic review and meta-analysis that evaluated the prognostic value of PCT in patients with pneumonia. The authors identified 21 studies with 6007 patients in which high PCT levels were associated with an increased risk of death among patients with low severity of illness score according to a low CURB-65 score and critically ill patients.26 In addition, this study questions the use of the PCT cut-off value of 0.5 ng/mL due to the low sensitivity and the inability to recognize patients at risk of death.26 Finally, the study suggested that there are no differences in the PCT performance characteristics between patients with CAP and ventilator-associated pneumonia.26 Beta-lactam plus macrolides have become the standard of care for the management of patients with CAP. Horita et al.27 performed a systematic review and meta-analysis evaluating whether beta-lactam plus macrolide antibiotics lead to better survival than beta-lactam alone in patients with CAP. The authors included 14 studies, two open-label randomized controlled trials (RCTs) comprising 1975 patients, one non-RCT interventional study comprising 1011 patients and 11 observational studies comprising 33 332 patients.27 Horita et al.27 found that CAP patients treated with beta-lactam plus macrolide compared with beta-lactam alone had a lower all-cause death with substantial heterogeneity. However, in the subgroup of patients with severe CAP, the use of combination therapy had a protective effect for mortality compared with beta-lactam antibiotic therapy alone.27 Therefore, despite the suggested beneficial effect of combination therapy with macrolides observed mainly in the severe CAP group, data are limited by the number of observational studies identified in this systematic review.27 Jefferies et al.28 performed a double-blind RCT to assess the clinical efficacy of paracetamol as anti-pyretic treatment for adult patients with influenza infection. The original concern was derived from animal models that suggested a higher mortality when infected with influenza.28 Among 80 adult patients with influenza-like symptoms and positive influenza rapid antigen test, the use of 1-g paracetamol four times a day for 5 days was not associated with viral shedding, temperature change, time to resolution of influenza illness or improved clinical symptoms compared with the placebo-treated group.28 A following elegant Editorial by Peters29 highlights the importance of challenging the collective therapeutic intuition and the conventional general wisdom of policy and guidelines regarding therapies such as paracetamol and suggest that similar studies should be performed with aspirin or other anti-inflammatory medications. Quality of care, such as timely antibiotic administration, resuscitation and oxygen therapy, determine outcome from CAP.30 Many countries have reduced availability of staff and resources at weekends compared with weekdays leading to concerns of a ‘weekend effect’ of increased mortality from common diseases. Uematsu et al.31 investigated this in a nationwide Japanese database and showed that from 23 532 patients with severe CAP, weekend admission was associated with a 10% increase in mortality even after adjustment for confounders. A clue to poorer quality of care was suggested by a lower frequency of microbiological tests, but the data set was not sufficiently detailed to dig deeper into other markers of quality of care.31 These data are similar in many countries but are controversial because of the different characteristics of patients admitted at weekends and therefore the risk of residual confounding. Kim et al.32 evaluated the predictors of prolonged stay in patients with CAP and complicated parapneumonic effusions in Korean tertiary cohort of patients. The authors performed a retrospective cohort study in 158 CAP patients with complicated parapneumonic effusions and identified that patients tended to stay at least 18 days in the hospital and require at least 10 days of chest tube placement.32 The most common pathogen was Streptococcus viridans and 54% of the patients received intrapleural fibrinolytics.32 A prolonged hospital stay was independently associated with fever, lower oxygen level, haemoglobin level, neutrophilia, pathogen identification and ineffective pleural drainage, suggesting a possible beneficial effect of an early thoracotomy, but further studies are need to test this observation.32 During the past year, there were two interesting contributions to the journal in the area of tuberculosis. The first corresponds to the work presented from Hong Kong on HIV-infected patients by Leung et al.33 The authors assess the diagnostic characteristics of tuberculin skin test (TST), T-Spot-TB (T-Spot) and QuantiFERON-TB Gold-In Tube (QFT) to determine latent tuberculosis (LTBI) in this high-risk group of patients. They identified 110 HIV-infected patients without previous history of LTBI, with 75% of them receiving anti-retroviral therapy and a median CD4 count (414/μL) with low median viral load (<75/μL).33 The results showed almost complete discordance between positive baseline TST (5.5%), T-Spot (5.6%) and QFT (4.9%). There were no correlations between CD4 count and viral load with positive LTBI testing.33 Test conversion rates were higher for QFT (10.7%) and TST (8.9%), and lower in T-Spot test (5.0%) with no correlation with CD4 count or viral load. More than half of the positive T-Spot and QFT results reverted to negative during the follow-up period.33 None of these tests picked up the single case of culture-confirmed tuberculosis observed after 798 person-years of follow-up.33 The authors conclude that the major discordant results found in TST, T-Spot and QFT tests limit the ability of these tests to identify LTBI patients in high-risk populations.33 In addition, it suggests that programmes directed at screening HIV-infected patients in intermediate tuberculosis burden areas may lack accuracy and requires further assessment.33 Regarding novel treatments for the management drug-resistant tuberculosis, Bai et al.34 tested ‘curcumin’ as a direct anti-TB and immune response stimulator. Curcumin is a spice of bright yellow colour of turmeric, derived from the root of the perennial herb Curcuma longa.34 Curcumin is recognized as a potent inducer of apoptosis and linked to the mechanism of how macrophages kill intracellularly. The authors used an in vitro human macrophage infection model to test whether curcumin affects Mycobacterium tuberculosis (MTB) survival.34 The authors found that MTB clearance was enhanced in THP-1 human monocytes and primary human alveolar macrophages.34 They also found that curcumin was an inducer of caspase-3-dependent apoptosis and autophagy and mediates these anti-MTB cellular functions, in part, via inhibition of nuclear factor-kappa B (NFκB) activation.34 Therefore, curcumin seems to present immunomodulatory properties against MTB infection in human macrophages, but further animal and translational studies are encouraged in order to treat patients at risk for drug resistance development.34 Several manuscripts addressed the clinical characteristics, the clinical outcomes and the treatment of NTM pulmonary infections. The classic patients with NTM is described as being a middle-aged female with low BMI and Mycobacterium avium infection. In 2016, the United States Bronchiectasis Registry has recently published their first results showing 63% of patients with bronchiectasis in the USA have a history of NTM isolation or disease, the majority of whom were females and had M. avium infection.35 Nishimura et al.36 assessed the prior conceived association that body habitus is associated with M. avium complex (MAC) pulmonary disease. The classic literature regarding MAC has linked middle-aged white women with thin body habitus with the diagnosis of MAC lung disease. The authors found that among 1033 healthy subjects, the body habitus was not associated with MAC lung disease.36 Differences may relate to environmental, genetic or microbiological differences between Asia and the USA, or differences in the included study populations. Yeung et al.37 reported a systematic review to determine the health-related QOL, co-morbidities and mortality in pulmonary NTM infections. The authors identified 17 studies, mainly from Taiwan (n = 5) and the USA (n = 4).37 This systematic review concludes that there is high clinical heterogeneity and important bias limiting the ability to define trends, despite the suggested increased health burden from respiratory diseases and increased mortality associated with pulmonary NTM disease.37 In the February issue of Respirology, Ellender et al.38 reported a retrospective study of 45 patients treated with i.v. amikacin for difficult NTM. Patients have a mixture of MAC, M. abscessus and one individual with M. fortuitum and received i.v. amikacin as an add-on to multidrug regimens.38 Reflecting the difficulties in treatment in this patient population, sputum conversion was achieved in only 38% but the treatment was well tolerated. Yuanlin Song Ventilator-induced lung injury (VILI) is a major drawback of mechanical ventilation when used improperly in patients with acute respiratory distress syndrome (ARDS). The exact mechanisms leading to VILI have not been completely elucidated but volume trauma, barotraumas and biotrauma have been proposed as likely causes.39 Extensive studies using a low tidal volume strategy showed improved survival benefit in ARDS patients,40 and recent recalculation of lung mechanics using large clinical trial data suggested that low driving pressure is the main contributor to the survival benefit.41 Given the fact that, in the majority of cases, driving pressure by itself could not be directly correlated to VILI if breathing rate was not included, it has been proposed that lung strain and energy load may collectively account for VILI rather than various single components.42 Identification of novel key factors of particular signalling pathways may provide potential therapeutic targets against ARDS. In an animal model, high-volume ventilation or cell culture model cyclic stretch induced degradation of p120-catenin, an adhesion molecule that plays an important role in epithelium integrity. In dissecting the mechanism, inhibition of protein kinase Cα (PKCα) could successfully block c-Src kinase activation and p120-catenin degradation in VILI model, suggesting that inhibition of PKCα may have potential application in reducing VILI.43 ARDS has been considered a heterogeneous disease with hyper-inflammation in the lungs and systemic circulation. Oxidation has been considered a strong risk factor for ARDS development due to endothelium and epithelium damage through reactive oxygen species (ROS) activity. It is therefore proposed that N-acetylcysteine may attenuate acute lung injury (ALI) through free radicals eradication. However, in a human lung resection study which simulated ALI, perioperative administration of N-acetylcysteine did not prevent local and systemic inflammation after lung resection,44 in which plasma IL-6, 8-isoprostane and ischaemia-modified albumin were almost identical between pre- and post-operation. Statin has also been considered an anti-inflammation medicine and has potential application in ALI/ARDS; however, a meta-analysis summarizing 13 studies including 12 145 patients did not show an improvement in clinical outcomes following statin administration in high-risk ALI/ARDS patients.45 According to these two studies, the anti-inflammation and antioxidation action of these two molecules may not be of benefit in ALI/ARDS patients. Sepsis is the leading risk factor for ALI/ARDS, and it is interesting to know whether the infection site (e.g. lung, abdominal cavity, urinary tract, etc) affects the outcome prediction in patients with ALI. In a group of sepsis patients (624) who developed ALI (251), there was a strong correlation between sepsis location and mortality in ALI patients but not in non-ALI sepsis patients. Sepsis due to lung infection contributed largely to ALI development compared with other infection sites, and morbidity of ALI was associated with higher mortality in sepsis from pulmonary infection sites, but not in abdominal sepsis.46 This evidence suggests that lung infection and lung infection-induced sepsis is a leading cause of ALI with high mortality and morbidity, and deserves a high level of attention in research and clinical practice. Activin-A belongs to the transforming growth factor-beta (TGF-beta) family of molecules, with a potential role in cell proliferation and division. In a study enrolling 138 clinically suspected sepsis patients, high serum Activin-A was associated with sepsis severity, acute physiology and chronic health evaluation II (APACHE-II) score and sequential organ failure assessment (SOFA) score,47 suggesting that Activin-A may be potentially useful in sepsis outcome prediction. The weaning process is challenging when the patient has delirium. A study enrolling 393 mechanical ventilation patients showed that mental status such as delirium significantly delayed the weaning process using the spontaneous breathing test (SBT) method.48 When confounding factors were balanced, delirium was still associated with difficult but not prolonged weaning. Aspiration in non-invasive ventilation is a serious complication and needs to be avoided as much as possible. A study using a healthy young and a healthy elder population investigated aspiration rate in different ventilation modes (bi-level positive airway pressure (BiPAP), continuous positive airway pressure (CPAP) and control) and showed a high probability of aspiration after swallow during BiPAP use49 compared with CPAP and control. This suggests that swallow with the detection of non-inspiratory flow may increase the risk of aspiration through triggering inspiratory support in BiPAP mode, which has been widely used in clinical practice. In general, various factors were associated with ARDS patients’ prognosis. Improved understanding of risk factors in each patient will enable more accurate prognostication and lead to better outcomes in ALI/ARDS. Christopher Mallow and Lonny Yarmus Both ultra-low-dose computed tomography (LDCT) and dynamic contrast-enhanced MRI (DCE-MRI) were evaluated in the past year. Murray et al.50 evaluated the use of LDCT on asbestos-exposed population screening. Of the 906 patients who underwent LDCT for annual screening, they identified 104 indeterminate nodules in 77 participants over a 12-month period. They found no correlation between the number of nodules and cumulative asbestos exposure, the pack-years of smoking or time since quitting. Parietal pleural plaques were identified in 64% of participants, in which 88.3% were calcified. Visceral pleural plaques were identified in 11.7% of patients, diffuse pleural thickening in 7.9% and diaphragmatic plaques in 56.4% of patients. A total of 162 (17.8%) subjects were recalled for further investigation due to indeterminate nodules or other incidental findings. Of those patients recalled, 8 patients were subsequently identified as having histologically confirmed lung cancer. DCE-MRI was evaluated by Karaman et al.51 to evaluate its use in identifying characters of cystic-cavitary lung lesions that may aid in distinguishing malignant versus benign lesions in comparison to CT alone. This was a prospective study of 36 patients with cystic and/or cavitary lung lesions, of whom 11 were found to have malignant lesions. Inner wall irregularity was more sensitive in malignant lesions on DCE-MRI (82% vs 64%), but specificity was higher with the CT (96% vs 88%). Due to the DCE-MRI's ability to better distinguish tissue from fluid, biopsy site was changed in two patients, and cancelled in three patients who had indeterminate lesions on CT. Malignant lesions had a greater increase in contrast enhancement in the early phase and had a higher maximal peak (P = 0.001). Although these findings were promising, the small study size requires that further investigation be performed before formal recommendations can be made in regards to its use in screening. Guisier et al.52 evaluated the adequacy of molecular analysis of peripheral non-small cell lung cancer (NSCLC) sampled by radial endobronchial ultrasound (r-EBUS). They performed 362 r-EBUS procedures of which 194 cancers were identified, 113 were non-squamous NSCLC and 88 had sufficient material for molecular analysis. The mean nodule diameter was 28 mm, and all 88 samples were analysed for epidermal growth factor receptor (EGFR), KRAS, anaplastic lymphoma kinase (ALK), human epidermal growth factor receptor 2 (HER2), phosphoinositide 3-kinase (PI3K) and BRAF. Upper and middle lobe biopsies, as well as at least three biopsies were predictive factors of molecular analysis feasibility. The authors concluded that multigene molecular analysis can be performed in 79% of non-squamous NSCLC, and that the sensitivity of molecular analysis, although limited due to repeat sampling in only 22 patients, is approximately 95%. Prognostic indicators were assessed both using circulating tumour cell (CTC) counts and the presence of COPD and emphysema.53, 54 Zhang et al.53 investigated whether patients with NSCLC had detectable CTCs that could provide prognostic information. Forty-six patients were enrolled and CTCs were measured at baseline in all patients. They found the presence of CTCs in 40 patients, with a range of CTCs from zero to 80 cells/3.2 mL (mean = 6.11 cells/3.2 mL). They found that the baseline CTC count >8 (unfavourable group) was a predictor of a shorter progression-free survival (PFS) and overall survival (OS). The median OS in the favourable group (CTC < 8) was significantly longer than in the unfavourable group (21.3 months vs 9.0 months; P = 0.026). The median PFS was also longer in the favourable group (7.4 months vs 5.3 months; P = 0.018). These remained significant in multivariate analysis. They did not find a correlation between CTC count and tumour size after cycles of chemotherapy. A meta-analysis of studies that evaluated COPD and/or emphysema and survival in patients with lung cancer was done by Gao et al.54 They found that the presence of COPD was associated with poorer OS in patients with COPD versus without COPD (hazard ratio (HR) = 1.17, P < 0.001), although there was high heterogeneity amongst the studies (I2 = 74%, P < 0.001). Six of the studies also evaluated disease-free survival (DFS) and found that COPD was also associated with worse DFS (HR = 1.52, P = 0.03) and high heterogeneity (I2 = 78%, P < 0.001). The presence of emphysema was associated with poorer OS (HR = 1.66, P < 0.001). The overall conclusion was that there is a strong association between both COPD and emphysema and survival. Numan et al.55 prospectively analysed a cohort of patients with pulmonary metastases to decipher which surgical option, VATS versus thoracotomy, led to superior QOL. A total of 100 patients were assigned either VATS or thoracotomy, with their Short form survey 36 (SF-36) and Brief Pain Inventory scores assessed preoperat