NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE

Transcription

1 NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE INTERVENTIONAL PROCEDURES PROGRAMME Interventional procedure overview of insertion of endobronchial valves (with or without assessment for collateral ventilation) for lung volume reduction in emphysema Lung volume reduction with airway valves for emphysema Emphysema is one of a group of chronic lung diseases that form chronic obstructive pulmonary disease, or COPD. In emphysema, the walls of the smaller airways inside the lungs break down, creating abnormally large spaces.these air-filled spaces compress the healthy parts of the lungs, making it more difficult to breath. Breakdown of the walls of the airways can cause collateral ventilation. This is when air enters one part of the lung from another, instead of by the natural airway. Assessing collateral ventilation can help to identify the areas of the lung that would benefit from lung volume reduction. The aim of lung volume reduction is to limit airflow to the diseased parts of the lung so that as much air as possible can enter the healthy parts. In this procedure, a small flexible camera is inserted down the windpipe and small one-way valves are inserted into the airways (3 or 4 valves are usually inserted) to direct airflow to the healthy parts of the lung. Introduction The National Institute for Health and Clinical Excellence (NICE) has prepared this overview to help members of the Interventional Procedures Advisory Committee (IPAC) make recommendations about the safety and efficacy of an interventional procedure. It is based on a rapid review of the medical literature and specialist opinion. It should not be regarded as a definitive assessment of the procedure. Date prepared This overview was prepared in November Page 1 of 31

2 Procedure name Insertion of endobronchial valves (with or without assessment for collateral Specialist societies British Thoracic Society. Description Indications and current treatment Emphysema is a chronic lung disease in which the walls of the air sacs (alveoli) in the lungs weaken and disintegrate, leaving behind abnormally large air spaces that stay filled with air even when the patient breathes out. The most common symptoms of emphysema are shortness of breath (dyspnoea), coughing, fatigue and weight loss. Intercurrent illnesses (such as chest infections) often prompt exacerbations, which may need hospitalisation. Emphysema is one of a group of diseases that forms chronic obstructive pulmonary disease (COPD). Emphysema is usually smoking-related but may also be inherited. Treatment may include pulmonary rehabilitation advice (smoking cessation, patient and carer education, exercise training and breathing retraining) and the use of inhaled or oral medication, including bronchodilators and corticosteroid therapy. Oxygen therapy may also be indicated in more severe cases. Lung volume reduction surgery is an option for patients who experience breathlessness, and whose pulmonary function test results show severe obstruction and enlarged lungs. Such surgery can be done using an open approach (using a sternotomy or thoracotomy), or thoracoscopically (using video assisted thoracoscopy [VATS] or thoracotomy). Lung transplantation surgery may also be an option. Certain therapies may be particularly useful in regional lung disease. What the procedure involves The aim of insertion of endobronchial vales (with or without assessment for collateral is to reduce airflow to the lung segments that have poor function in order to improve gas exchange in healthier parts of the lung. It uses an endoscopic approach, which is less invasive than open or thoracoscopic lung volume reduction surgery. Page 2 of 31

3 The procedure is done using a bronchoscope with the patient under sedation or general anaesthesia. CT and perfusion scans are used to identify the lung segments that have poor function. A specially designed balloon catheter with a flow sensor may be used to assess collateral ventilation (when air enters a lobe of the lung from an adjacent lobe instead of by the natural airway) to identify the areas of the lung that would benefit from the procedure. Using a delivery catheter, a synthetic valve is put into the target location, expanded and fixed to the bronchial wall. The valve is designed to prevent air inflow during inspiration but allow air and mucus to exit during expiration. Several valves may be needed for different lung segments. Different devices are available for this procedure. Outcome measures St George s Respiratory Questionnaire (SGRQ) The SGRQ is designed to measure health impairment in patients with respiratory disease. Three component scores are calculated for the SGRQ: 1. Symptoms concerned with the effect of respiratory symptoms, their frequency and severity. 2. Activity concerned with activities that cause or are limited by breathlessness. 3. Impacts covers a range of aspects concerned with social functioning and psychological disturbances resulting from airways disease. A total score is also calculated, which summarises the impact of the disease on overall health status. Scores are expressed as a percentage of overall impairment in which 100 represents worst possible health status and 0 indicates best possible health status. Literature review Rapid review of literature The medical literature was searched to identify studies and reviews relevant to insertion of endobronchial valves (with or without assessment for collateral. Searches were conducted of the following databases, covering the period from their commencement to 8 October 2012: MEDLINE, PREMEDLINE, EMBASE, Cochrane Library and other databases. Trial registries and the Internet were also searched. No language restriction was applied to the searches (see appendix C for details of search strategy). Relevant published studies identified during consultation or resolution that are published after this date may also be considered for inclusion. Page 3 of 31

4 The following selection criteria (table 1) were applied to the abstracts identified by the literature search. Where selection criteria could not be determined from the abstracts the full paper was retrieved. Table 1 Inclusion criteria for identification of relevant studies Characteristic Publication type Patient Intervention/test Outcome Language Criteria Clinical studies were included. Emphasis was placed on identifying good quality studies. Abstracts were excluded where no clinical outcomes were reported, or where the paper was a review, editorial, or a laboratory or animal study. Conference abstracts were also excluded because of the difficulty of appraising study methodology, unless they reported specific adverse events that were not available in the published literature. Patients with emphysema. Insertion of endobronchial valves (with or without assessment for collateral ventilation). Articles were retrieved if the abstract contained information relevant to the safety and/or efficacy. Non-English-language articles were excluded unless they were thought to add substantively to the English-language evidence base. List of studies included in the overview This overview is based on 811 patients from 3 randomised controlled trials and 4 case series 1 7. Other studies that were considered to be relevant to the procedure, but were not included in the main extraction table (table 2), have been listed in appendix A. Page 4 of 31

5 Table 2 Summary of key efficacy and safety findings on insertion of endobronchial valves (with or without assessment for collateral Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Sciurba FC (2010) 1 Number of patients analysed: 321 (220 vs 101) Major adverse events at 90 days EBV Control p Randomised controlled trial (VENT trial US cohort) Primary outcomes (intention to treat analysis, change from baseline at 6 months) Death* n=214 2 n=87 0 value 1.00 EBV Control p (0.9%) USA (31 centres) Recruitment period: n=220 (95% CI) n=101 (95% CI) value COPD exacerbation (with 17 (7.9%) 1 (1.1%) 0.03 Study population: patients with advanced heterogeneous emphysema n=321 (220 endobronchial valves vs 101 standard medical care) Age: mean 65 years Sex: 57% (182/321) male Patient selection criteria: age years, FEV 1 of 15 45% of predicted value, total lung capacity >100% of predicted value, residual volume >150% of predicted value, body mass index 31.1 for men or 32.2 for women, PaCO 2 <50 mm Hg, PaO 2 >45 mm Hg, postrehabilitation 6-minute walk distance of at least 140 m. Exclusion criteria included a carbon monoxide diffusing capacity <20% of predicted value, presence of giant bullae or α 1-antitrypsin deficiency, previous thoracotomy, excessive sputum, severe FEV 1 Mean % change from baseline Mean change in value from baseline (ml) Mean % change in predicted value from baseline 4.3 (1.4 to 7.2) 34.5 (10.8 to 58.3) 1.0 (0.2 to 1.8) Distance on 6-minute walk test Median % change 2.5 from baseline (-1.1 to 6.1) Median change from 9.3 baseline (metres) (-0.5 to 19.1) -2.5 (-5.4 to 0.4) (-48.3 to -2.6) -0.9 (-1.7 to -0.1) -3.2 (-8.9 to 2.4) (-29.6 to 8.1) Page 5 of Secondary outcomes (intention to treat analysis, change from baseline at 6 months) EBV (95% CI) Control (95% CI) p value Mean change in SGRQ score from baseline -2.8 (-4.7 to -1.0) 0.6 (-1.8 to 3.0) 0.04 Scores on the SGRQ range from 0 to 100, with higher scores indicating a worse quality of life. The minimal clinically important difference is 4 points. The authors noted that greater heterogeneity of emphysema between lobes and intact interlobar fissures appeared to be associated with a greater likelihood of clinically important functional and physiological responses to EBV. hospitalisation) COPD exacerbation (without hospitalisation) 3 (1.4%) Pulmonary infection 4 (1.9%) Respiratory failure* 3 (1.4%) Pneumonia distal to 2 valve* (0.9%) Pneumonia not 5 distal to valve (2.3%) New or worsening 2 hypercapnia (0.9%) Hypoxaemia 3 (1.4%) Massive 1 haemoptysis* (0.5%) Any haemoptysis 12 (5.6%) Pneumothorax or 3 prolonged air leak (1.4%) (duration >7 days)* (2.3%) Empyema* * these 6 major complications were used to give a composite complication rate. Follow-up issues: Loss to follow-up at 12 months was 20.8% in the EBV group and 11.8% in the control group (p=0.04). Of the 220 patients assigned to EBV therapy, 6 patients did not receive it. Study design issues: Before randomisation, patients had 6 to 8 weeks of pulmonary rehabilitation and optimised medical management. Study sample size was calculated to provide a power of 90% to detect a mean improvement of 15% in the FEV 1 in the EBV group compared with the control group and a mean improvement in 17% in distance on the 6- minute walk test at 6 months.

6 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments pulmonary hypertension, active infection, or unstable cardiac conditions. Technique: the type of anaesthesia was decided according to local protocol; Zephyr endobronchial valves (Emphasys Medical [now Pulmonx]) were placed unilaterally in lobar, segmental or subsegmental bronchi to completely isolate the targeted lobe (mean number of valves per patient=3.8, range 1 9). Follow-up: 12 months Conflict of interest/source of funding: the study was funded by Pulmonx. Composite complication rate at 6 months: EBV=6.1% Control=1.2%, p=0.08 There were 6 deaths in the EBV group and none in the control group. Cause of death included respiratory failure not associated with the EBV procedure [n=3], cancer [n=1], ischaemic colitis [n=1], massive haemoptysis [n=1]. Composite complication rate at 12 months: EBV=10.3% Control=4.6%, p=0.17 Rate of pneumonia in the target lobe in the EBV group at 12 months=4.2% Implant-related events at 90 days: Valve expectoration, aspiration or migration=4.7% (10/214) Formation of bronchial granulation tissue=2.3% (5/214) Bronchial trauma=0.5% (1/214) In 12 months of follow-up, valves were removed in 31 patients between 1 and 377 days after insertion (reasons included migration [n=8], patient s request for an unspecified reason [n=7], pneumonia management [n=3], placement in incorrect lobe [n=3], COPD exacerbations [n=2], haemoptysis [n=1], other [n=7]). Page 6 of 31

7 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Herth F (2012) 2 Number of patients analysed: 171 (111 vs 60) Serious complications within 3 months (<97 days) of procedure Follow-up issues: 7% of patients in the Randomised controlled trial (VENT trial European cohort) Valves were not successfully placed in 9% of patients because of complex local anatomy. EBV n=111 Control n=60 p value EBV group and 3% in the control group were Death lost to follow-up. Germany, Austria, France, Ireland, Italy, Spain (23 centres) Outcome variables (intention to treat analysis, change from baseline at 6 months); data presented as mean±sd Respiratory failure (0.9%) 4 (1.7%) Study design issues: Before randomisation, Recruitment period: not reported EBV Control p (3.6%) (1.7%) patients had 6 to 8 n=111 n=60 value 24 hour ventilation weeks of pulmonary Study population: patients with FEV 1 absolute 7±20 0.5± (1.8%) (0%) rehabilitation and advanced heterogeneous percentage change Haemoptysis optimised medical emphysema 6-minute walk test 15±91 10± (5.4%) (1.7%) management. distance (metres) Pneumothorax or The sample size was n=171 (111 endobronchial Cycle ergometry 2±14-3± prolonged air leak (3.6%) (0%) smaller than the a valves vs 60 standard medical workload (watts) (duration >7 days) priori estimate care) SGRQ score -5±14 0.3± Empyema necessary to provide Pneumonia not a power of 90% to Age: mean 60 years Change in FEV 1 score at 12 months: distal to valve (3.6%) (1.7%) detect a mean Sex: 72% (123/171) male EBV=6±26% Pneumonia distal to improvement of 15% Control=-2±20%, p= valve (3.6%) in the FEV 1 and 17% Patient selection criteria: age COPD in 6-minute walk years, FEV 1 of 15 45% Change in clinical outcomes from baseline at 6 and exacerbation (with (11.7 (10.0%) distance. of predicted value, total lung 12 months according to fissure integrity (assessed by CT); hospitalisation) %) Study population capacity >100% of predicted data presented as mean±sd COPD issues: value, residual volume >150% Complete fissure Incomplete fissure exacerbation (27.9 (38.3%) A statistically of predicted value, body mass EBV control p EBV Control p value (without %) significantly higher index 31.1 for men or 32.2 for n=44 n=19 value n=67 n=40 hospitalisation) proportion of patients women, PaCO 2 <50 mm Hg, FEV 1 % 16±21 2± ±18-1± in the EBV group PaO 2 >45 mm Hg, postrehabilitation 6-minute walk 6-minute 11±34 19± ±36 2± ±29-2± ±23-2± Rates for valve expectoration: needed oxygen distance of at least 140 m. walk 13±35 10± ±30 0± days=7.2% (8/111) therapy (72%) compared with the Exclusion criteria included a distance Between 98 and 194 days=0.9% (1/111) control group (48%, carbon monoxide diffusing Cycle 4±14-3± ±14-3± Between 195 and 284 days=3.6% (4/111) workload 4±14-2± ±13-6± p=0.002). capacity <20% of predicted Between 285 and 386 days=0.9% (1/111) (watts) Other issues: value, presence of giant bullae SGRQ -6±15 3± ±14-1± The authors noted or α 1-antitrypsin deficiency, The rates for serious complications did not differ score 0±15 4± ±14-1± that there have been previous thoracotomy, significantly between patients who had EBV and subsequent technical excessive sputum, severe The top figure in each cell is the result at 6 months and the second patients in the control group. improvements to the pulmonary hypertension, active figure is the result at 12 months. EBV delivery system. infection, or unstable cardiac During the 1-year follow-up 6 patients died in the Page 7 of 31

8 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments conditions. Technique: the type of anaesthesia was decided according to local protocol; Zephyr endobronchial valves (Emphasys Medical [now Pulmonx]) were placed unilaterally in a single upper or lower lobe (mean number of valves per patient=3, range 1 5). Follow-up: 12 months Conflict of interest/source of funding: the study was funded by Pulmonx. In EBV-treated patients classified as having lobar occlusion (assessed by CT), clinical outcomes at both 6 and 12 months were better than outcomes for EBV-treated patients who were classified as not having lobar occlusion and those for controls. Clinically significant improvements were observed in patients with emphysema heterogeneity scores <15%, which were sustained through 12 months. EBV group (5%) and 4 patients died in the control group (7%). None of the deaths was deemed to be related to the study procedure. Page 8 of 31

9 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Ninane V (2012) 3 Number of patients analysed: 73 (37 vs 36) Procedural adverse events EBV Control p Randomised controlled trial Responders at 3-month follow-up (defined as having both a n=37 n=36 value 4-point improvement in SGRQ and a volume decrease in Any adverse event Belgium, Austria, Italy, Germany, Spain, UK, Canada, USA upper lobes with a compensatory volume increase of 7.5% in non-treated lobes measured by CT): EBV=22.9% (8/35) Any serious adverse event (16.2%) 0 (0%) (8.3%) 1 (2.8%) 0.49 Recruitment period: not reported Control=0% (0/35), p=0.005 Acidosis (2.7%) (0%) Study population: patients with from baseline at 3-month follow-up (5.4%) (5.6%) upper lobe predominant severe EBV Control p value COPD emphysema n=33 n=35 exacerbation (0%) (2.8%) Change in lung volume (assessed by CT) and SGRQ score Bronchospasm % change in -7.3±9 0.7±5.2 < Dyspnoea n=73 (37 bronchoscopy and upper lobes (5.4%) (0%) valve placement vs 36 (treated) Hypoxaemia bronchoscopy only) % change in 6.7± ± (2.7%) (2.8%) non-upper Nausea Age: Mean 61 years lobes (3.6%) (0%) Sex: 59% (43/73) male Patient selection criteria: patients with advanced upper lobe predominant emphysema. Technique: IBV Valves (Spiration Inc., USA) were placed through a flexible bronchoscope. One segment or sub-segment of the right upper lobe and the lingual segments of the left upper lobe were not treated to achieve incomplete occlusion of the upper lobes and to prevent lobar atelectasis. Mean number of valves placed=7.3. Follow-up: 6 months Conflict of interest/source of (untreated) Change in SGRQ score -4.3± ± Change in FEV 1 and 6-minute walk test EBV Control p value* Baseline 3 months Baseline 3 months FEV ± 0.90± 0.88± 0.87± (litres) minute walk test (metres) 337± ± ± ± * change from baseline, treatment versus control Patients who remained free of COPD exacerbations showed greater improvements than those who experienced an exacerbation. Adverse events during the 3-month blinded study period (after hospital discharge) EBV n=37 Control n=36 p value Any adverse event (40.5%) (25.0%) Any serious adverse event (18.9%) (11.1%) COPD exacerbation (29.7%) (22.2%) Bronchospasm (2.7%) (0%) Dyspnoea (8.1%) (0%) Cerebrovascular accident and death 1 (2.7%) 0 (0%) 1.00 Haemoptysis 1 (2.7%) Page 9 of 31 0 (0%) 1.00 Study design issues: Randomised, singleblinded shamcontrolled study. Randomisation occurred after bronchoscopic examination of the airways. Patients in the control group had a sham bronchoscopy and were followed up for 3 months. At 3 months, treatment assignment was unblinded and patients in the valve group were followed up for a further 3 months. Eligible patients in the control group could then receive bronchial valves and were followed for 3 months. A sample size of 100 patients was selected to estimate the difference in responder rates with a maximum standard error of 10%. Enrolment was stopped after 2 years for logistical reasons. Pulmonary rehabilitation was not needed.

10 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments funding: the study was funded by Spiration Inc. Deaths There was 1 death during the 3-month blinded period that was assessed as not related to the device (cerebrovascular accident). Beyond 3 months, 1 death was assessed as possibly related to the treatment (241 days after valve placement). A second death occurred 138 days after valve placement but this was assessed as unrelated to the device. Valve migration/erosion There were no reports of valve migration or erosion and there was no expectoration of valves. Study population issues: The2 groups were well matched in terms of demography, health status and physiological measurements. Page 10 of 31

11 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Herth FJ (2012) 4 Number of patients analysed: 80 Serious adverse events (defined as events needing hospitalisation, causing a serious (prospective) Presence of collateral ventilation assessed using the Chartis deterioration in health, or death) after EBV system=36.3% (29/80) procedure=20% (19/96) Germany, the Netherlands, Sweden (5 centres) Recruitment period: Study population: patients with late-stage heterogeneous emphysema n=96 Age: mean 63 years Sex: 49% (39/80) male Patient selection criteria: age 18 years. Exclusion criteria included active pulmonary infection, FEV 1<15 or >50% of predicted value. Technique: pre-designated target lobe was assessed using the Chartis Pulmonary Assessment System TM (Pulmonx) to determine presence or absence of collateral ventilation before EBV placement (in segmental or lobar bronchi). Follow-up: 30 days Conflict of interest/source of funding: study was sponsored by Pulmonx. Median target lobe volume reduction: No collateral ventilation group=752.7 ml (IQR to ) Collateral ventilation group=98.6 ml (IQR 26.8 to 232.1) p< (between groups) Of 51 patients without collateral ventilation, 36 had a target lobe volume reduction of 350 ml, giving a positive predictive value of 71%. 47% (17/36) of these patients had valves implanted into the lower lobes; 39% (14/36) of these patients were later re-classified as having a low degree of heterogeneity between lobes ( 15%). Of the 29 patients with collateral ventilation, 24 had a target lobe volume reduction <350 ml (i.e. non-responders), giving a negative predictive value of 83%. 75% of patients in this group were treated in the right upper lobe. Mean % increase in FEV 1 from baseline Improvement in distance for 6-minute walk test from baseline (metres) Mean reduction in SGRQ score from baseline NS=Not significant No collateral ventilation n=51 Collateral ventilation n=29 p value 16±22 1± NS 10 5 NS Pneumothorax=8.3% (8/96) (all patients were treated using standard procedures and recovered within 14 days) COPD exacerbation=7.3% (7/96) Pneumonia=1.0% (1/96) Dyspnoea=1.0% (1/96) Respiratory failure=1.0% (1/96) 2 patients died but neither of the deaths was determined to be related to the EBV procedure (1 patient died from an abdominal aortic aneurysm and 1 died after spinal surgery). The authors noted that no serious adverse events were attributed to using Chartis. This study used a proprietary system for assessing collateral ventilation. Follow-up issues: 16 patients were excluded from the final analysis (6 lacked HRCT, 8 were lost to follow-up and 3 died of unrelated causes). Demographic data were given for the 80 patients completing the study. Study design issues: The primary objective of the study was to evaluate the effectiveness of Chartis in selecting patients who would benefit from EBV therapy. Before the study, the investigators agreed that a 350 ml target lobe volume reduction would be considered significant. Short-term follow-up. There was no need for patients to have pulmonary rehabilitation or optimisation of medical treatment. Page 11 of 31

12 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Venuta F (2011) 5 Number of patients analysed: 40 Postoperative complications: Contralateral pneumothorax=2.5% (1/40) (prospective) Baseline 1 year 3 years 5 years n=9 p value (15 days after the procedure) Pneumonia in lobe adjacent to lobe in which Italy (single centre) n=40 n=33 n=18 valves were inserted=5.0% (2/40) Mild haemoptysis=2.5% (1/40) (3 years after Recruitment period: not reported procedure; patient needed anticoagulation after coronary artery revascularisation) Study population: patients with Small granulations in front of valve=5.0% heterogeneous emphysema (2/40) (no treatment was needed because the n=40 device was not obstructed and patients were asymptomatic) Age: mean 61 years Sex: 93% (37/40) male Patient selection criteria: age between 35 and 75 years, FEV 1<35%, residual volume >180%. Exclusion criteria included homogenous emphysema, currently smoking, presence of isolated bulla, PaCO 2>50 mmhg, diffusing capacity of the lung corrected for alveolar ventilation<20%, productive cough, small airway disease. Technique: unilateral treatment with Zephir valves (Pulmonx corporation). Mean number of valves per patient=3.6. Follow-up: median 32 months Conflict of interest/source of funding: none Mean FEV (litres) Mean minute walk test (metres) Mean < supplemental oxygen (litres/min) Mean MRC dyspnoea score < The improvements for FEV 1, 6-minute walk test and MRC score were statistically significant at all time points. The results for supplemental oxygen were statistically significant up to the third year. Mean survival=36±4.3 months 1-year survival=81.6%, 3-year survival=47.4%, 5-year survival=22.4% Outcomes for patients with visible fissures (37.5% [15/40]; diagnosed by HRCT): Baseline n=15 1 year n=15 3 years n=9 5 years n=6 Mean FEV 1 (litres) Mean 6-minute walk test (metres) Mean supplemental oxygen (litres/min) Mean MRC dyspnoea score patients had single lung transplantation and 1 patient had double lung transplantation at a mean of 6.3 months after valve placement; 2 patients died after the transplant with no valverelated complications and 1 was alive and well at the time of report. 1 patient died of respiratory failure on the waiting list for double lung transplantation 13 months after valve placement. 1 patient had lung volume reduction surgery 1 year after valve placement. 1 patient had the valve removed after 3 months. 40% (16/40) of patients died during follow-up (lung cancer, n=4; myocardial infarction with intractable arrhythmia, n=3; end-stage respiratory failure, n=7; post-transplant, n=2). Follow-up issues: 82.5% (33/40) of patients had a followup longer than 12 months. Study design issues: Heterogeneity was subjectively assessed by at least 2 members of the team using HRCT and lung perfusion scans. All patients received optimal medical therapy at the time of evaluation. The MRC dyspnoea scale ranges from 1 to 5, with higher scores indication more severe symptoms. Study population issues: Patients receiving bilateral treatment were excluded. Page 12 of 31

13 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Sterman DH (2010) 6 Number of patients analysed: 91 Procedure-related complications Pneumonia associated with valves within (prospective) Technical success=99.7% (609/611) airways 3 months of procedure=1.1% (1/91) Bacterial bronchitis associated with valves USA Health status changes within 3 months of procedure=1.1% (1/91) Recruitment period: month 3 months 6 months 12 months Bronchospasm (within 3 days of SGRQ total scores procedure)=8.8% (8/91) (1 was described as Study population: patients with n=80 n=78 n=67 n=53 serious and associated with respiratory failure severe emphysema Change -5.2± ± ± ±14.2 and myocardial infarction that began the n=91 compared evening after an uneventful procedure. The with p=0.001 p=0.011 p=0.001 p<0.001 patient had further episodes of bronchospasm baseline and the valves were removed on day 21. A Age: mean 65 years (range 42 Proportion 51.3% 52.6% 55.2% 56.6% second patient had valve removal on day 3 79), Sex: 56% male with 4 because the bronchospasm did not resolve). point Myocardial infarction on day 3=1.1%(1/91) improvement Injury to bronchi=3.3% (3/91) Transient hypercarbia=2.2% (2/91) (1 patient SF-36 physical function needed overnight ventilator support) Patient selection criteria: heterogeneous, upper-lobe predominant emphysema. Patients with a significant bronchospastic component to their emphysema, chronic bronchitis, or significant bronchiectasis were not included. Patients already accepted and listed for lung volume reduction surgery or lung transplantation were also excluded. Technique: The IBV valve (Spiration Inc., USA) was used for bilateral upper lobe placement. Mean number of valves per patient=6.7. Follow-up: 12 months Conflict of interest/source of funding: funded by Spiration Inc., USA. Change compared with baseline Proportion with 10 point change n=81 n=79 n=70 n=54 3.4± ± ± ±22.5 p=0.08 p=0.215 p=0.003 p= % 35.4% 42.9% 48.1% FEV 1 and 6-minute walk distance Baseline 6 months 12 months FEV 1 (litres) 0.87± ± ± minute walk distance (feet) 1108± ± ±303 Changes were not statistically significant Lung volume was decreased in 87% of patients at 6 months. Nonupper lobe volume increased in 93% of patients at 6 months. There were no occurrences of valve migration or erosion. Complications within 12 months Pneumothorax=12.1% (11/91) (5 were judged to be serious and definitely device-related. Pneumonia distal to valves=6.6% (6/91) Valve removal=17.6% (16/91) (between 97 and 358 days after device placement for pneumonia, bronchospasm, recurrent COPD exacerbations, or pneumothorax) Deaths (n=3, 3.3%) 1 patient died from tension pneumothorax 4 days after the procedure; 1 patient died on day 113 from respiratory failure and pneumonia; 1 patient died on day 33 related to respiratory failure and pneumonia after placing an endotracheal tube for surgical repair of a prolonged air leak. Page 13 of 31 Follow-up issues: No patients were lost to follow-up. During the 12-month study, 26 patients withdrew (10 associated with an adverse event). Study design issues: The primary outcome was safety (the rate of observed migration, erosion or infection associated with valves within the first 3 months after placement). Secondary outcomes were 15% increase in FEV 1, 15% increase in 6-minute walk distance and 4-point improvement on SGRQ.

14 Abbreviations used: ADO, age, dyspnoea, obstruction, CI, confidence interval; COPD, chronic obstructive pulmonary disease; CT, computed tomography; EBV, endobronchial valve; FEV 1, forced expiratory volume in 1 second; HRCT, high-resolution computed tomography scan; IBV, intrabronchial valve; IQR, interquartile range; MRC, Medical Research Council; PaCO 2, partial pressure of carbon dioxide; PaO 2, partial pressure of oxygen; SD, standard deviation; SGRQ, St George s Respiratory Questionnaire Study details Key efficacy findings Key safety findings Comments Hopkinson NS (2011) 7 Number of patients analysed: 19 Complications Pneumothorax, n=2 (1 occurred at day 1 and A radiologist blinded to clinical outcome assessed CT evidence of atelectasis 1 month after the procedure; 26.3% (5/19) of patients needed drainage; the other occurred at 4 weeks and resolved without intervention) UK had atelectasis. Distal lung infection needing drainage, n=1 (6 years after the procedure) Recruitment period: Ipsilateral empyema, n=1 (2 years after the procedure) Study population: patients with COPD n=19 Age: not reported for whole group Sex: 84% (16/19) male Patient selection criteria: significant dyspnoea despite optimum medical therapy (including pulmonary rehabilitation) and a heterogeneous pattern of disease. Technique: Emphasys EBV device used (Emphasys Medical, USA); all procedures were unilateral. Follow-up: 6 years Conflict of interest/source of funding: one of the authors has received research funding from Asthmatx, Olympus, Broncus and PneumRx. At 6 years, all 5 patients with atelectasis were still alive, whereas 57% (8/14) of patients without atelectasis had died (p=0.026). Four of the 8 deaths occurred within 3 years of the procedure (none were within 6 months). Death certificate data showed that 6 deaths were from respiratory failure, 1 was cardiovascular and 1 was due to lung cancer. In a stepwise regression model, atelectasis was retained as an independent correlate of survival at 6 years. 9 patients were defined as improvers in the original study on the basis of their exercise capacity. At 6 years, 2 of the improvers and 6 of the non-improvers had died (p=0.06). 3-year mortality rate for the whole study group=16% 3-year mortality rate for the non-atelectasis group=29% The authors noted that, according to the ADO score (age, dyspnoea, obstruction), predicted 3-year mortality was 31.1±10% in the nonatelectasis group and 32.2±15.1% in the atelectasis group. There were no pneumothoraces in the patients without atelectasis. Study design issues: Survival data was censored at 6 years. Primary analysis was survival at 6 years in individuals with or without atelectasis. All treatment response parameters were measured 1 month after the procedure. Study population issues: BMI was statistically significantly higher in the atelectasis group (28.4 versus 21.6, p<0.001). There were no significant differences with regard to spirometry, lung volumes, gas transfer, blood gas parameters, quality of life, exacerbation rates, or exercise capacity. Page 14 of 31

15 Efficacy Forced expiratory volume in 1 second (FEV 1 ) A randomised controlled trial of 321 patients treated by endobronchial valves or standard medical care (control group) reported a mean increase of 4% in FEV 1 in the endobronchial valve group compared with a decrease of 3% in the control group at 6-month follow up (p=0.005) 1. A randomised controlled trial of 171 patients treated by endobronchial valves or standard medical care (control group) reported a mean increase of 7% in the FEV 1 in the endobronchial valve group compared with 0.5% in the control group at 6-month follow up (p=0.067) 2. Patients with complete fissure (assessed by CT) treated by endobronchial valves (n=44) showed a mean improvement of 15% in the FEV 1 at 12 months compared with 0% in patients with incomplete fissure treated by endobronchial valves (n=67). A randomised controlled trial of 73 patients treated by endobronchial valves or bronchoscopy only (control group) reported no improvement in FEV 1 in either group at 3-month follow up (p=0.065 for change from baseline, treatment versus control) 3. A case series of 96 patients treated by endobronchial valves with assessment of collateral ventilation reported that patients with no collateral ventilation had a statistically significantly higher increase in FEV 1 from baseline than patients with collateral ventilation at 30-day follow up (16% versus 1%, p=0.0013) 4. A case series of 91 patients reported that there were no statistically significant changes in FEV 1 throughout the 12-month follow up 6. 6-minute walk distance The randomised controlled trial of 321 patients treated by endobronchial valves or standard medical care (control group) reported a median increase of 3% in the 6-minute walk test in the endobronchial valve group compared with a decrease of 3% in the control group at 6-month follow up (p=0.04) 1. The randomised controlled trial of 171 patients treated by endobronchial valves or standard medical care (control group) reported a mean increase of 15 metres from baseline in the 6-minute walk test in the endobronchial valve group compared with 10 metres in the control group at 6-month follow up (p=0.696) 2. The increase was greater in patients with complete fissure (assessed by CT) compared with patients with incomplete fissure (13 metres versus 5 metres, p value not stated). The randomised controlled trial of 73 patients treated by endobronchial valves or bronchoscopy only (control group) reported no statistically significant improvement in the 6-minute walk test in either group at 3-month follow up (p=0.410 for change from baseline, treatment versus control) 3. St George s Respiratory Questionnaire (SGRQ) scores The randomised controlled trial of 321 patients treated by endobronchial valves or standard medical care (control group) reported a mean decrease in the SGRQ Page 15 of 31

16 score of 2.8 points from baseline in the endobronchial valve group compared with an increase of 0.6 points in the control group at 6-month follow up (p=0.04) 1. Scores on the SGRQ range from 0 to 100, with higher scores indicating a worse quality of life. The minimal clinically important difference is 4 points. The randomised controlled trial of 171 patients treated by endobronchial valves or standard medical care (control group) reported a mean decrease in the SGRQ score of 5 points from baseline in the endobronchial valve group compared with an increase of 0.3 points in the control group at 6-month follow up (p=0.047) 2. The randomised controlled trial of 73 patients treated by endobronchial valves or bronchoscopy only (control group) reported decreases of 4.3 and 3.6 points respectively in the SGRQ score at 3-month follow up (p=0.837 for change from baseline, treatment versus control) 3. Safety Deaths Two patients died within 90 days and 4 patients died between 90 days and 6 months of endobronchial valve insertion in a randomised controlled trial of 321 patients treated by endobronchial valves or standard medical care (control group); there were no deaths in the control group over the same period (p=1.00) 1. Cause of death included massive haemoptysis (n=1), respiratory failure not associated with endobronchial valve insertion (n=3), ischaemic colitis (n=1) and cancer (n=1) (the paper does not specify the cause of deaths that occurred within 90 days). The death of 1 patient 241 days after valve placement, in a randomised controlled trial of 73 patients treated by endobronchial valves or bronchoscopy only (control group), was assessed as possibly related to the endobronchial valve treatment 3. One patient died from tension pneumothorax 4 days after valve insertion in a case series of 91 patients 6. Haemoptysis Haemoptysis within 90 days was reported in 6% (12/214) of patients treated by endobronchial vales compared with 0% (0/87) of patients treated by standard medical care in a randomised controlled trial of 321 patients (p=0.02) 1. One patient was described as having massive haemoptysis. Haemoptysis within 3 months was reported in 5% (6/111) of patients treated by endobronchial valve insertion and 2% (1/60) of patients treated by standard medical care in a randomised controlled trial of 171 patients 2. Pneumothorax Pneumothorax or prolonged air leak (duration longer than 7 days) within 90 days was reported in 1% (3/214) and 4% (4/111) of patients treated by endobronchial valve insertion and 0% (0/87, 0/60)) of patients treated by standard medical care in the randomised controlled trials of 321 and 171 patients respectively 1,2. Pneumothorax was reported in 8% (8/96) of patients in a case series of Page 16 of 31

17 96 patients (timing not stated); all patients recovered with 14 days after treatment 4. One patient had contralateral pneumothorax 15 days after the procedure in a case series of 40 patients 5. Pneumothorax within 12 months of valve insertion was reported in 12% (11/91) of patients in a case series of 91 patients; 5 of these were judged to be serious and definitely device-related 6. Pneumothorax was reported in 11% (2/19) of patients in a case series of 19 patients; 1 occurred at day 1 after the procedure and needed drainage, the other occurred at 4 weeks and resolved without intervention 7. Respiratory failure Respiratory failure within 3 months was reported in 1% (3/214) and 4% (4/111) of patients treated by endobronchial valve insertion and 0% (0/87) and 2% (1/60) of patients treated by standard medical care in the randomised controlled trials of 321 and 171 patients respectively (p=0.56 and 0.66) 1,2. One patient with respiratory failure was reported in a case series of 96 patients 4. Pneumonia distal to valve Pneumonia distal to the valve was reported in 1% (2/214) and 4% (4/111) of patients treated by endobronchial valve insertion in the randomised controlled trials of 321 patients and 171 patients respectively 1, 2. Pneumonia distal to valves was reported in 7% (6/91) of patients in a case series of 91 patients at a followup of 12 months 6. Valve expectoration, aspiration or migration Valve expectoration, aspiration or migration within 90 days was reported in 5% (10/214) of patients treated by endobronchial valve insertion in the randomised controlled trial of 321 patients 1. Valve expectoration occurred in 7% (8/111) of patients within 90 days of the procedure in the randomised controlled trial of 171 patients. A further 6 patients reported valve expectoration between 98 days and 386 days of follow-up 2. Bronchospasm Bronchospasm associated with the procedure was reported in 5% (2/37) of patients treated by endobronchial valve insertion and 6% (2/36) of patients in the control group treated by bronchoscopy alone (p=1.00) in a randomised controlled trial of 73 patients. One patient treated by endobronchial valve insertion had bronchospasm during a 3-month follow-up period (after hospital discharge) 3. Bronchospasm within 3 days of the procedure was reported in 9% (8/91) patients in a case series of 91 patients. One of these was described as serious and associated with respiratory failure and myocardial infarction that began the evening after the procedure; the patient had further episodes of bronchospasm and the valves were removed on day 21. A second patient had valve removal on day 3 because the bronchospasm did not resolve 6. Page 17 of 31

18 Exacerbation of COPD COPD exacerbation needing hospitalisation was reported in 8% (17/214) and 12% (13/111) of patients treated by endobronchial valve insertion compared with 1% (1/87) and 10% (6/60) of patients treated by standard medical care in 2 randomised controlled trials of 321 and 171 patients respectively (p=0.03 and 0.80) 1,2. COPD exacerbation during a 3-month follow-up period was reported in 30% (11/37) of patients treated by endobronchial valve insertion and 22% (8/36) of patients in the control group treated by bronchoscopy alone (p=0.60) in a randomised controlled trial of 73 patients 3. Other complications Injury to bronchi was reported in 3% (3/91) of patients in a case series of 91 patients (not further described) 6. In the same case series, 2% (2/91) of patients reported transient hypercarbia; 1 patient needed overnight ventilator support. Ipsilateral empyema was reported in 1 patient 2 years after the procedure in a case series of 19 patients 7. Validity and generalisability of the studies The treatment protocol varied between studies, in terms of the number of valves used and bilateral (as opposed to unilateral) treatment. One study used a proprietary system for assessing collateral ventilation 4. Only 1 of the 3 randomised controlled trials used a sham procedure as the control 3. The fact that there were improvements in the patients in the control group treated by a sham procedure suggests there may be some placebo effect. Some studies needed patients to have pulmonary rehabilitation and some did not this may influence the results of outcome measures such as the 6- minute walk test. Different studies used different kinds of endobronchial valves. Existing assessments of this procedure There were no published assessments from other organisations identified at the time of the literature search. Related NICE guidance Below is a list of NICE guidance related to this procedure. Appendix B gives details of the recommendations made in each piece of guidance listed. Page 18 of 31

19 Interventional procedures Bronchoscopic lung volume reduction with airway valves for advanced emphysema. NICE interventional procedures guidance 318 (2009). Available from Lung volume reduction surgery for advanced emphysema. NICE interventional procedures guidance 114 (2005). Available from Clinical guidelines Chronic obstructive pulmonary disease: Management of chronic obstructive pulmonary disease in adults in primary and secondary care (partial update). NICE clinical guideline 101 (2010). Available from Specialist Advisers opinions Specialist advice was sought from consultants who have been nominated or ratified by their Specialist Society or Royal College. The advice received is their individual opinion and does not represent the view of the society. Professor P Calverley, Dr N Hopkinson, Professor M Morgan, Dr M Munavvar, Mr R Page, Dr P Shah, Mr D Waller (British Thoracic Society) Four Specialist Advisers considered the procedure to be definitely novel and of uncertain safety and efficacy; 1 Specialist Adviser described it as the first in a new class of procedure; 1 Specialist Adviser described the procedure as established practice and no longer new. One Specialist Adviser noted that the technique is useful in selected patients but cannot be considered as standard or routine therapy. The comparator to this procedure would be lung volume reduction surgery. Anecdotal adverse events include pneumothorax, respiratory failure, death, haemoptysis, infection, pneumonia, and valve displacement. Adverse events reported in the literature include dyspnoea, bronchospasm, pneumothorax, respiratory failure, death, haemoptysis, valve migration, exacerbations, infection and pneumonia. Page 19 of 31