Endobronchial valves for patients with heterogeneous emphysema and without interlobar collateral ventilation: open label treatment following the BeLieVeR-HIFi study

Outcomes in early trials of bronchoscopic lung volume reduction using endobronchial valves for the treatment of patients with advanced emphysema were inconsistent. However improvements in patient selection with focus on excluding those with interlobar collateral ventilation and homogeneous emphysema resulted in significant benefits in the BeLieVeR-HIFi study compared with sham treated controls. In this manuscript we present data from the control patients in the BeLieVeR-HIFi study who went on to have open label endobronchial valve treatment after completion of the clinical trial (n=12), combined with data from those in the treatment arm who did not have collateral ventilation (n=19). Three months after treatment FEV1 increased by 27.3 (36.4)%, residual volume reduced by 0.49 (0.76) L, the 6 min walk distance increased by 32.6 (68.7) m and the St George Respiratory Questionnaire for COPD score improved by 8.2 (20.2) points. These data extend the evidence for endobronchial valve placement in appropriately selected patients with COPD. Trial registration number: ISRCTN04761234; Results.


Panel S1: METHODS
The BeLieVeR-HIFi study was a 1:1 randomised, parallel group, double-blind sham bronchoscopy controlled trial of unilateral, endobronchial valve placement (Zephyr valves, PulmonX, California) in patients with heterogeneous emphysema and intact interlobar fissures. The study was approved by the London -Bentham Research Ethics Committee (REC No. 11/LO/1608), the sponsor was Imperial College, London. There was a trial steering group and an independent data monitoring committee. All patients provided written informed consent and were recruited between March 2012 and September 2013. The procedures were performed using moderate sedation with Midazolam and Alfentanyl in the bronchoscopy suite at the Royal Brompton Hospital. Endobronchial valves were placed to occlude all segmental bronchi leading to the target lobe Patients with optimally medically managed and stable severe or very severe COPD were included (forced expiratory volume in one second (FEV1) <50% predicted, residual volume (RV) >150% predicted, 6 minute walk distance (6MWD) <450m, and significant breathlessness (MRC dyspnoea score >3)). Their CT scans demonstrated heterogeneous emphysema with a defined target lobe, and intact adjacent interlobar fissures. Assessments were performed at baseline and 3 months after bronchoscopy, by an assessment team blind to treatment allocation. Assessments included complete lung function testing, endurance cycle ergometry with metabolic measurements at 70% of their maximal workload determined on an initial incremental test, a 6MWD test, and quality of life questionnaires (SGRQc [1]

Open Label group
Measurements of collateral ventilation using the Chartis TM (PulmonX, California) balloon catheter system were made in all participants during the initial sham bronchoscopy. At the 3 months end of study visit, all treatment options available to the control group patients were discussed. Only patients where collateral ventilation was not detected (CV -ve) were offered open label valve placement. Treatment was performed within 2 weeks of the end of study visit.
Of 25 control patients, 14 had BLVR treatment and of the remaining 11, two had died, four had positive collateral ventilation during Chartis assessment (of these two had no further treatment, one had LVRS and one was enrolled into another bronchoscopic LVR trial), and of the five without collateral ventilation on Chartis assessment two preferred LVRS, one had recurrent infections and much airway secretions on initial bronchoscopy and thus valves were avoided in favour of LVRS, one was felt to have a very high risk of pneumothorax due to the presence of ipsilateral paraseptal bullae in the non-target lobe, and one was found to have airway anatomy which did not support endobronchial valve placement.

Statistical Analysis
The primary endpoint in this follow-up study is the difference in the percentage change in FEV1 Non-normally distributed numeric data are presented as median (IQR). Statistical analysis for differences between baseline values (at the final control visit) and 3 months following endobronchial valve treatment was assessed by Wilcoxon signed rank tests. A p value <0·05 was taken to indicate statistical significance. For responder analyses minimum clinically important differences (MCID) were pre-specified as a 15% increase for FEV1, 350mls reduction for the residual volume [4], 4 points decrease for the SGRQc[1 5], 2 points decrease for the CAT [3 6] an increase of 105 seconds for endurance cycle Tlim [7] and an increase of 26m for the 6MWD [8].

Panel S2: Adverse events
There was one death secondary to a pneumothorax directly attributable to treatment with endobronchial valves. Changes to our admission and follow-up protocols were subsequently

Panel S3: Limitations of the study
Limitations of this extension study include the small sample size and open-label and unblinded nature of the treatment and follow-up. Furthermore, analysis does not account for missing data from patient who died. However the primary and most secondary outcomes were objective rather than subjective. All treatments were undertaken at a single highly specialised centre with extensive experience of treating patients with end stage emphysema using LVRS and a variety of bronchoscopic lung volume reduction techniques, as part of clinical trials and in routine clinical care. The set-up at our centre includes a pre-existing multidisciplinary team that meets on a weekly basis involving dedicated thoracic radiologists, thoracic surgeons, pulmonologists and physiotherapists. It may not be possible to achieve similar outcomes at other centres without this infrastructure or clinical experience