O9.2 Lung volume reduction surgery and bronchoscopic interventions

Lung volume reduction surgery (LVRS) involves resection of the most severely affected areas of emphysematous, non-bullous lung (Cooper 1995). This can improve lung elastic recoil and diaphragmatic function (Geddes 2000). The National Emphysema Treatment Trial was a large randomised multicentre study which investigated the effectiveness and cost-benefit of this procedure (NETT 1999). A total of 1,218 patients with severe emphysema underwent pulmonary rehabilitation and were then randomised to LVRS or continued medical therapy. Pulmonary rehabilitation plays an important role in preparing patients for interventions such as lung volume reduction (Ries 2005). There was no overall survival advantage of surgery, but after 24 months there was significant improvement in exercise capacity in the surgical group. Patients allocated to LVRS took significantly longer (median 2 versus 1 year) than those who continued medical therapy to reach a composite endpoint of death or meaningful deterioration in disease related quality of life (Benzo 2009). Among patients with predominantly upper lobe emphysema and impaired exercise capacity, mortality was significantly lower in the surgical than the medical group. However, high risk patients with diffuse emphysema and well preserved exercise capacity are poor candidates for surgery because of increased mortality and negligible functional gain (Fishman 2003) [evidence level II]. A 2016 Cochrane Review on lung volume reduction surgery was very heavily influenced by data from the NETT study (van Agteren 2016) [evidence level I]. The authors concluded that short-term mortality was higher for LVRS (odds ratio (OR) 6.16, 95% CI 3.22 to 11.7 than for control, but long-term mortality favoured LVRS (OR 0.76, 95% CI 0.61 to 0.95) level 1. 96% of the patients contributing to the long term mortality data is from patients enrolled in the NETT study. The authors made note of high post-operative complications, especially persistent air leak and pneumonia.

A variety of nonsurgical techniques are currently under investigation. These include endobronchial one-way valves, self-activating coils, targeted destruction of emphysematous tissue, bypass tract airway stenting and transpleural ventilation.  Of these techniques, only valves are in regular clinical use in Australia.

Several randomised controlled trials examining endobronchial valves have been reported (Sciurba 2010, Herth 2012, Davey 2015, Klooster 2015, Wood 2014, Valipour 2016Klooster 2017, Kemp 2017, Criner 2018).  The Valipour, Kemp and Criner trials are multi-centre.  All trials recruited highly selected COPD patients with severe obstruction and gas trapping and excluded patients with significant hypercapnia (PaCO2 > 50mmHg) and poor mobility (6-minute walk distance (6MWD) < 100m). Patients in the trials by Sciurba, Herth, Davey and Criner underwent pulmonary rehabilitation prior to randomisation. Only the Davey and Wood trials used a sham placebo bronchoscopy. Klooster, Valipour, Kemp and Criner excluded patients without intact interlobar fissures on CT chest and with collateral ventilation detected at bronchoscopy by the Chartis system. Valipour specifically recruited patients with homogeneous rather than upper lobe predominant emphysema.

All but the Wood trial reported improvements in FEV1. Data on improvement in exercise outcomes was generally positive (Herth, Davey, Klooster, Valipour, Kemp and Criner). Data on adverse events was significant. The Sciurba and Wood trials reported high hospital admissions for COPD exacerbations, Klooster reported an 18% pneumothorax rate, Valipour reported a 26% pneumothorax rate and Kemp reported that 29% of patients experienced a pneumothorax and that one patient died as a consequence.  Criner reported a 27% pneumothorax rate and a 3% 45-day mortality rate (Criner 2018).  By 12 months, 22% of patients in the Klooster trial required permanent valve removal (Klooster 2017).

The Klooster, Valipour, Kemp and Criner trials were the most thorough in excluding patients with collateral ventilation using both the intact lobar fissure on CT chest and the Chartis system during bronchoscopy. This may explain why these trials had the most impressive results with regards to lung function, quality of life and exercise improvements. The trials where the control arm used sham placebo bronchoscopy reported no difference between the study arms with respect to improvement in quality of life. Valves cannot yet be recommended as routine therapy and patients need to be informed of the short-term mortality and complication rates. Davey recognises the high complication rates and appropriately calls for future trials to compare valve placement with surgical lung volume reduction (Davey 2015).

In a meta-analysis by van Geffen et al, data from randomised controlled trials across all modalities of lung volume reduction (surgical and endobronchial) were analysed (van Geffen 2019).  The mean differences compared with the control were an increase in FEV1 of 15∙87% (95% CI 12∙27 to 19∙47), improvement in 6-minute walk test (6MWT) of 43∙28 m (31∙36 to 55∙21), and reduction in the St George’s Respiratory Questionnaire (SGRQ) of 9∙39 points (–10∙92 to –7∙86).  The authors note a high risk of bias due to lack of blinding.  The odds ratio for a severe adverse event, which included mortality, was 6∙21 (95% CI 4∙02 to 9∙58) following intervention.

van Geffen then performed a meta-analysis of endobronchial lung volume reduction surgery (van Geffen 2019).  6 trials were included in the analysis of endobronchial valves (620 participants) and 3 trials were included in the analysis of endobronchial coils (458 participants).  The authors reported improvements in lung function, 6-minute walk distance and symptom scores with both modalities.  The odds ratio for an adverse event for trials examining endobronchial valves was (9∙58, 95% CI 5∙56 to 16∙50).  The most frequent adverse events with endobronchial valve treatment were pneumothorax (1∙4 to 25%) and COPD exacerbations (4 to 20%).  The odds ratio for an adverse event for trials examining coils was (OR 8∙73, 95% CI 2∙69 to 28∙32; figure 3A). The most common adverse events were pneumonia (5 to 20%),  COPD exacerbations (7 to 28%) and pneumothorax (5 to 10%).  There was no difference in early mortality between valves/ coils and control, however the confidence intervals were very wide.  There is concern regarding the lack of sham bronchoscopy and/or unclear status of blinding in some studies that may cause a risk of bias (van Agteren 2017).

A small (n=45) non-blinded randomised controlled trial comparing lung volume reduction coils to standard care showed significant improvement in quality of life, lung function and 6-minute walk distance at 90 days (Shah 2013). Of the 23 patients receiving the intervention, two sustained a pneumothorax. 22 patients from the control arm went on to receive endobronchial coils, with significant improvement in quality of life, 6-minute walk distance (and to a lesser degree lung function) at day 180 and 360 (Zoumot 2015). Deslee has reported on the largest randomised controlled trial (non-blinded) to date (Deslee 2014). 100 highly selected patients were randomised to usual care or bilateral coil placement. All patients had undergone pulmonary rehabilitation. The pre-specified primary end point of percentage of patients achieving a 54m improvement in 6-minute walk distance at 6 months was met (36% versus 18% p=0.03). However at 12-month follow up there was no significant difference in mean improvement in 6-minute walk distance. There were sustained improvements in spirometry and quality of life at 12 months. Pneumonia rates were far higher in the coil group (18% versus 4% p=0.03) but pneumothorax rates were similar between the two groups. Cost effectiveness analysis found the treatment to be prohibitively expensive (12-month incremental cost effectiveness ratio was approximately $1,000,000 per QALY).

van Agteren performed a meta-analysis of endobronchial lung volume reduction surgery (van Agteren 2017) [evidence level I].  Results from 14 trials comprising almost 2,000 participants were analysed.  The authors concluded that evidence for short-term (up to one year) improvements in disease status were most evident for studies testing endobronchial valves (five studies) and coils (three studies), including improvements in lung function and quality of life.  The authors note a significant increase in adverse events. The odds ratio for an adverse event reported for trials examining endobronchial valves was 5.85 (95% CI 2.16-15.84) and the overall odds ratio for an adverse event amongst all endobronchial lung volume reduction techniques was 3.00 (95% CI 2.04-4.43). Pneumothorax rates of over 20% were reported in several endobronchial valve trials.  It is important to note the authors’ concerns regarding the lack of sham bronchoscopy and/or unclear status of blinding in some studies that may cause a risk of bias (van Agteren 2017).

Herth et al (Herth 2016b) performed an open label, multi centre, randomised controlled trial of staged, single lobe segmental steam thermal ablation on 70 patients with severe COPD and hyperinflation. All patients had undergone pulmonary rehabilitation and had a 6-minute walk distance over 140m. Patients with incomplete fissures and collateral ventilation were not excluded. At six months, there was a significant improvement in lung function and quality of life but not 6-minute walk distance. 24% of patients undergoing steam thermal ablation experienced a COPD exacerbation compared with 4% of controls. This procedure is not available in Australasia and its precise role is not yet clear, but as further long-term data emerge this may be an option for patients with severe COPD and hyperinflation with collateral ventilation.

A non-blinded randomised controlled trial comparing endobronchial lung volume reduction using Emphysematous Lung Sealant (ELS) was terminated early due to loss of funding prior to the 12 month pre-specified endpoints (Come 2015). Limited data at 6 months showed significant improvements in spirometry, 6MWD, QoL and dyspnoea. However, the complication rate was unacceptably high with increased hospitalisations (44% versus 17%) and serious adverse events with two deaths in the intervention arm and no deaths in the control arm [evidence level II].

In summary, endobronchial valves may be appropriate in highly selected patients with severe COPD and hyperinflation if collateral ventilation can be excluded (intact fissure on imaging and Chartis negative during bronchoscopy).  This therapy should only be considered in high volume specialised centres (Herth 2017).  All patients being considered for lung volume reduction surgery and bronchoscopic interventions should be referred for pulmonary rehabilitation and discussed by an expert panel that includes a radiologist, respiratory physician, interventional pulmonologist and thoracic surgeon (Herth 2017).