101: Endoscopic Approaches for Treatment of Emphysema

Treatment of patients with chronic obstructive pulmonary disease (COPD) traditionally has been the task of the internal medicine physician. Global Initiative for Chronic Obstructive Lung Disease recommendations for treatment of COPD include the use of bronchodilators, anti-inflammatory agents, oxygen therapy, aids to assist with smoking cessation, and pulmonary rehabilitation.¹ While pharmacotherapy improves lung function, symptoms, quality of life, and exacerbation rates in COPD, no medical therapies have been shown to improve emphysema, where the primary abnormality is destruction of alveoli with loss of elastic recoil, and subsequent lung hyperinflation. The National Emphysema Treatment Trial (NETT), a large multicenter randomized clinical trial to evaluate the effectiveness of lung volume reduction surgery (LVRS) for the treatment of emphysema, suggested that surgical lung volume reduction, which directly addresses the problem of lung hyperinflation through resection of the most damaged tissue, should be considered for selected patients with emphysema. The findings of this trial, while applicable to a defined subset of COPD patients with advanced upper lobe predominant (ULP) emphysema and reduced exercise capacity, clearly indicate that LVRS can affect lung physiology, symptoms, and even mortality for this disease.²

LVRS alters respiratory physiology in several ways—accordingly, posttreatment improvement is multifactorial.^3–10 As originally proposed by Brantigan and Mueller in the 1950s¹¹ and convincingly demonstrated by Fessler and Permutt,⁴ LVRS partially normalizes the mechanical relationship between the hyperinflated emphysematous lung and the surrounding chest wall by increasing the vital capacity and isovolume transpulmonary recoil pressures. Moreover, by reducing the overall size of the hyperinflated lung, LVRS produces space within the less compliant chest cavity for the remaining lung to expand and function.

While this “resizing” process appears to be the primary mechanism responsible for improvements after lung reduction, other factors play a role. Increased recoil pressures cause an increase in airway conductance in a subset of patients, presumably by raising airway isovolume transmural pressures and increasing airway dimensions.^12,13 The reduction in lung size after LVRS normalizes diaphragmatic and chest wall dimensions and improves ventilatory capacity by shortening the operating length over which the respiratory muscles contract.^8–10 In a smaller number of patients, temporary improvements in oxygenation have been observed as a result of local changes in lung impedance that act to normalize ventilation/perfusion matching. LVRS also may improve dynamic lung mechanics by eliminating lung zones with the longest expiratory time constants, not only reducing the tendency for gas trapping and dynamic hyperinflation during exercise but also increasing the inspiratory capacity.¹⁴ Emerging evidence further indicates that the benefits of LVRS go beyond primary respiratory effects; LVRS may in fact improve the compromised cardiac function that is a common comorbidity in this population.^15–17

Although LVRS provides a treatment option for many patients with advanced emphysema, it is a major procedure performed in a sick population and is associated with substantial morbidity and mortality. Procedural (90-day) mortality was 5.5% in NETT. In that trial, serious complications were observed in 59% of patients (i.e., respiratory failure, prolonged air leak, pneumonia, and cardiac morbidity).¹⁸ Furthermore, when expressed in terms of quality-adjusted life-years, LVRS is more expensive than other currently accepted surgical interventions that improve quality of life for individuals with end-stage disease, such as coronary artery bypass grafting, cardiac transplantation, and lung transplantation.¹⁹ Due to the risks and costs associated with LVRS, fewer than 200 procedures are estimated to be performed annually in the United States,²⁰ and as a result, a great deal of effort has been spent to devise safer, less invasive, and less costly alternatives. Several different approaches have been and are being tested in clinical trials in the United States and elsewhere. Initial results suggest that the physiologic basis for symptomatic improvement after “nonsurgical lung volume reduction” may not be the same for each of these new methods and may, in fact, be distinct from the effect of LVRS itself. In this chapter, we summarize the technology, methodology, published experiences, and limitations of each approach (Table 101-1).

The work of Fessler and colleagues⁴ and Ingenito and colleagues⁶ has shown that lung volume reduction therapy improves respiratory function in emphysema primarily by reducing the size of the hyperinflated lung within the rigid chest cavity. Thus, any process that eliminates areas of hyperinflated lung could potentially achieve the same effect as surgical resection. A variety of nonsurgical techniques have been developed in an attempt to accomplish this including primarily, one-way valves, endobronchial coils, tissue sealants, thermal airway ablation, and airway bypass.

One-way Valve Devices

Lung volume reduction, in principle, could be accomplished by placing a device in a proximal airway, thereby impeding distal gas flow. Theoretically, gas “trapped” beyond the obstructing device would eventually be absorbed and the lung would collapse. Endobronchial plugs and blockers were the original method developed to promote resorption atelectasis.^21,22 However, the high rate of postobstructive pneumonia, pneumothoraces, and device migration led to their abandonment. One-way valves are an evolution of this concept. They are deployed in the proximal airway through a flexible bronchoscope. Once positioned, these devices are designed to block air from entering the target area during inhalation, while allowing gas to escape during exhalation, leading to volume reduction by promoting progressive deflation and atelectasis in distal emphysematous lung. These valves also allow drainage of mucus, reducing the potential for postobstructive pneumonia.

Two one-way valve systems have been developed, both intended primarily for treatment of heterogeneous ULP emphysema: the endobronchial valve (EBV) and the intrabronchial valve (IBV). The EBV, manufactured by Pulmonx (formerly Emphasys Medical), is designed with a nitinol skeleton and a silicone body with a “duckbill” valve on the proximal end. Originally deployed over a guide wire, the most recent version – the Zephyr® EBV (Fig. 101-1) – is deployed through the working channel of a bronchoscope under direct vision. The deployment catheter also functions as a sizing mechanism so that the valve which best fits the bronchus can be chosen. Several studies,^23–26 including a randomized controlled trial,²⁷ have investigated the use of these valves in patients with severe heterogeneous emphysema.

In a retrospective analysis from a prospective multicenter registry, Wan et al.²⁶ reported the experience of the first 98 patients treated with EBVs (Emphasys EBV). In the registry, four valves were delivered on average per patient using several different treatment strategies—unilateral (predominant approach) versus bilateral, lobar versus nonlobar exclusion, upper lobe (most common) versus lower lobe. There were modest but statistically significant improvements in residual volume (RV: −350 ± 970 mL, −4.9 ± 17.4%, p = 0.025), forced expiratory volume in 1 second (FEV₁: +60 ± 210 mL, 10.7 ± 26.2%, p = 0.007), forced vital capacity (FVC: +120 ± 470 mL, 9.0 ± 23.9%, p = 0.024), and 6-minute walk distance (6MWD: +36.9 ± 90 m, p <0.001) at 90-day follow-up. Patients with lobar exclusion and unilateral treatment had the greatest benefit. Eight patients (8.2%; one death) had serious complications, the majority of which were pneumothoraces thought secondary to lung volume changes rather than iatrogenic injury. Thirty patients had other complications including COPD exacerbations and pneumonia in nontreated lobes. Importantly, postobstructive pneumonia was not seen as with the earlier generation endobronchial plug devices.

The Endobronchial Valve for Emphysema Palliation Trial (VENT) ²⁷ was a multicenter, prospective, randomized controlled trial designed to evaluate the safety and efficacy of unilateral EBV therapy with the newer Zephyr® valves (deployed through an internal bronchoscope channel rather than over guidewire). Three hundred and twenty-one patients were randomized to EBV therapy (n = 220) or optimal medical care (control, n = 101). Patients in the EBV group underwent unilateral and unilobar treatment with the aim of completely isolating the target—most diseased lobe (upper in 76.6% of patients). Again, four valves, on average, were placed per patient. At 6 months, there was a 34.5 mL increase (95% CI 10.8 to 58.3) in FEV₁ in the EBV group compared with a 25.4 mL decrease (95% CI −48.3 to −2.6) in the control group (p = 0.002 for between-group difference). The 6MWD increased by 9.3 m (95% CI −0.5 to 19.1) in the EBV group and decreased by 10.7 m (95% CI –29.6 to 8.1) in the control group (p = 0.02 for between-group difference). Functional outcomes were not as good as those in the multicenter registry. There were also modest (not clinically significant) improvements in disease-specific quality of life (measured by the St. George's Respiratory Questionnaire, SGRQ) and dyspnea (measured by the Modified Medical Research Council scale, mMRC). Further analysis revealed greatest benefit in patients with computed tomographic (CT) complete fissures (i.e., absent interlobar collaterals), complete lobar isolation, and greater emphysema heterogeneity and paralleled findings in the European VENT cohort.²⁸ At 6 months, there was a trend toward more major complications in the EBV versus control group (6.1% vs. 1.2%, respectively, p = 0.08), though this was less apparent between 6 and 12 months (4.7% vs. 4.6%). The most common adverse events (AEs) in the EBV group included postobstructive pneumonia, hemoptysis, and pneumothorax. Severe COPD exacerbations were significantly more common in the EBV group than the control group during the first 90 days, but there was no difference in severe exacerbations after 90 days. At 6 months, 67/194 (34.5%) patients with CT imaging were found to have evidence of valve malposition.

A smaller, longitudinal, single-center study²⁹ of 40 patients with a median follow-up of 32 months (up to 5 years) suggested a lasting benefit of EBV treatment. While 40% of patients died during the follow-up, no deaths were procedure-related and this proportion can be contrasted with a projected mortality rate in excess of 40% at 5 years in patients with similar COPD severity.³⁰ There were statistically and clinically significant improvements in FEV₁, 6MWD, and dyspnea score that persisted out to 5 years.

The second type of one-way valve – Spiration IBV® Valve (Olympus Corp., formerly Spiration, Inc.; Fig. 101-2) – has an umbrella design in which an elastomeric covering is stretched over a nitinol wire frame that anchors the device in place. Air (and mucus) can escape from the lung around the edges of the flexible covering as the umbrella-shaped frame partially collapses, but is prevented from flowing in the forward direction. As with the EBV, the IBV can be deployed through the working channel of a flexible bronchoscope under direct visualization and is designed for placement in segmental or subsegmental bronchi. A calibrated balloon is used to determine the valve with the best fit for the target airway.

A number of studies have shown the ability of IBVs to improve perceived quality of life though associated improvements in anatomic lung inflation patterns are variably affected and physiological measures remain unchanged. Moreover, it remains unclear as to the percentage of patients who actually receive benefit or the persistence of treatment. In an initial safety evaluation, Wood et al.³¹ reported the results of a multicenter, prospective, open-enrollment cohort study of 30 patients with severe to very severe airflow obstruction, hyperinflation, and ULP emphysema who underwent bilateral upper lobe treatment with IBVs. A mean of 6.5 valves were placed per patient. The procedure was well tolerated with few reported in-hospital or 30-day AEs and no late complications attributed to the valves. While the trial was not designed to assess effectiveness, there was a measurable improvement in SGRQ (mean change from baseline to 6 months: −6.8 ± 14.3 points, p = 0.05 with 52% of patients exceeding the minimal clinically important difference [MCID] of at least a 4-point decrease in SGRQ³²) with no significant change in physiologic or exercise outcomes. In another study of 57 subjects with ULP emphysema who had paired CT assessments both prior to treatment and either 1, 3, or 6 months after bilateral upper lobe IBV treatment, Coxson and colleagues³³ further associated such changes in SGRQ with changes in regional lung volume and speculated that the improved disease-specific quality of life might be due to increased ventilation and perfusion of the untreated/less-diseased nonupper lobes.

A larger, multicenter, prospective, open-enrollment case series³⁴ of 91 patients with ULP emphysema who underwent bilateral IBV treatment (mean of 6.7 valves placed per patient) also had similar findings with >50% of patients demonstrating a clinically significant improvement in SGRQ at 1, 3, 6, and 12 months following device placement (no change in FEV₁ or 6MWD). In this larger study, seven patients had device-related serious adverse events (SAEs) in the first 3 months. Pneumothorax was the most common complication particularly when all segments within a lobe were occluded and 16 patients required valve removal (44 valves total; for pneumonia, bronchospasm, recurrent COPD exacerbations, or pneumothorax). There were no occurrences of valve migration or erosion.

In two subsequent multicenter, sham bronchoscopy-controlled trials of IBV treatment (first 73 patients, reported; ³⁵ second 277 patients, completed) SGRQ again improved, though there were fewer responders. In the first study, 73 patients with ULP emphysema were randomized to IBV placement (n = 37) versus sham bronchoscopy (n = 36). To avoid complete occlusion of the upper lobes, lobar atelectasis, and potential pneumothorax, one segment (or subsegment) of the right upper lobe and the lingula were left untreated. A mean number of 7.3 valves were placed per patient in the treatment group. At 3 months, there were 8 (24%) responders (composite endpoint: ≥4-point improvement in SGRQ and lobar volume shift measured by CT with a decrease in upper lobe volume and a volume increase of ≥7.5% in nontreated lobes) in the treatment group versus none in the sham group (p = 0.002). In the treatment group, upper lobe volume decreased by 7.3 ± 9.0% and lower lobe volume increased by 6.7 ± 14.5%; these volume shifts persisted at 6 months. At 3 months, both groups had statistically significant mean improvements in SGRQ compared to baseline (treatment: −4.3 ± 16.2 points, control: −3.6 ± 10.7 points; there was no difference between groups, p = 0.8), though mean change in SGRQ only exceed the MCID in the treatment group. At 6 months, treated patients had further improvement in SGRQ (mean change from baseline: −10.9 ± 18.2 points). There were no significant improvements in pulmonary function tests, dyspnea, or exercise capacity. There were also no differences in procedural AEs or hospital length of stay between groups, indicating that as a whole, IBV treatment appeared safe, though effective in only a subset of patients.

In the second larger study (IBV treatment, n = 142 vs. sham bronchoscopy, n = 135; reported as abstracts^36–38), the responder rate was again low overall (responder defined as having both an improvement in SGRQ of ≥4 points and regional lung volume changes on CT [decreased volume of the upper lobes and ≥10% increase in nonupper lobes] at 6 months) though significantly higher for the treatment group than the control group (5.1% vs. 0.8%; treatment control difference 5.0%, 95% CI 0.1%–9.5%). As individual metrics, 19.2% of treated patients had a CT lobe volume response and 32.2% had a SGRQ response, though in this study (in contrast to Coxson et al.³³), the two measures were not correlated. Further, there was no significant difference in mean SGRQ response between the groups, and, as in the prior randomized controlled trial (RCT), there was no difference in lung function or exercise capacity between the groups at 6 months. In this larger study, there were significantly more SAEs in the treatment group (27%) compared with the control group (13%) including more deaths in treatment group (six vs. one, borderline significant), though deaths were not thought device-related. Also, while pneumothorax occurred in three out of the first 37 patients, early algorithmic changes to leave an untreated right upper lobe airway eliminated this AE.

Heterogeneity in trial results and patient response may reflect underlying and unaccounted for heterogeneities in the COPD population as well as the multifactorial mechanisms of action of treatments. While it was initially postulated that bronchial valves would lead to improvement by causing lobar atelectasis,²³ the presence of collateral ventilation between treated and untreated lobes through incomplete lobar fissures confounds regional collapse and treatment response. Subsequent studies have suggested other mechanisms for improvement. Hopkinson et al.³⁹ evaluated 19 patients before and 4 weeks after unilateral EBV placement. Only five developed visible atelectasis on CT, four of whom had improvements in exercise capacity. However, one-third of patients with no atelectasis on CT also had improved exercise capacity which was attributed to reduced dynamic hyperinflation (end-expiratory lung volume at isotime was reduced in patients with and without atelectasis). Collateral resistance is likely important in this phenomenon⁴⁰—in patients with relatively low resistance collateral channels, the region distal to a valve may still hyperinflate under certain conditions such as exercise; as resistance increases, this dynamic hyperinflation of distal lung tissue decreases as ventilation is directed toward more normal lung units; with even higher collateral resistance, atelectasis occurs and lung mechanics improve through the same mechanisms as LVRS. Another potential mechanism for improvement after valve placement is interlobar ventilatory shifts and improved ventilation perfusion matching as proposed by Coxson et al.³³

The limited improvements observed in patients with significant collateral ventilation (CV) have led to the development of new techniques for determining the amount of CV in a given segment or subsegment such as the Chartis® Pulmonary Assessment System (Pulmonx). This specialized catheter with a balloon at the distal tip is inserted via the working channel of a bronchoscope. Once inflated, the balloon blocks the airway permitting air to only flow out from the target compartment through the catheter's central lumen allowing airway flow and pressure to be measured. As a result, airway resistance and compartmental CV can be calculated. In an observational study,⁴¹ 80 patients were evaluated for CV with this system prior to unilateral placement of Zephyr® EBVs (either upper or lower lobe). Fifty-one subjects had no evidence of CV. Of those without CV, 71% had a significant improvement in target lobe volume reduction (TLVR, the primary outcome measure) at 30 days as compared to only 17% of those with CV. Moreover, of TLVR responders without CV, 39% had a low degree of heterogeneity between target and nontarget lobes. Clinical responses were better in those without than with CV and were even better in those without CV who achieved TLVR. While an observational study, these findings suggest absence of CV may be more important in predicting response to valve-based therapies than emphysema heterogeneity, at the same time expanding and refining the population who may benefit from such therapies. Additional prospective studies evaluating the effects of Zephyr® EBVs in patients evaluated pretreatment with the Chartis® system are ongoing.

Coils

Lung volume reduction coils (LVRC) are thought to work by (1) compressing diseased tissue, allowing expansion of better-functioning areas of lung; (2) retensioning adjacent lung tissue so that the lung contracts more efficiently during breathing; and (3) tethering open small airways to prevent air trapping. The RePneu® LVRC (Fig. 101-3), manufactured by PneumRx, Inc., is made from nitinol. It is delivered bronchoscopically into subsegmental airways and regains its predetermined shape upon deployment. For placement, a guidewire is advanced into the airway under fluoroscopic guidance and a catheter is passed over the wire and aligned with its distal tip. The airway length is measured with radiopaque markers to choose the appropriate coil size. The guidewire is then removed and a straightened coil is pushed through the catheter into position. Finally, the catheter is removed and the coil regains its original shape (coiling up the surrounding airway and tensioning the adjacent parenchyma).

In a pilot study, Slebos and colleagues⁴² evaluated the safety and effectiveness of LVRCs in 16 patients with heterogeneous emphysema. Four patients were treated unilaterally in a single procedure and 12 patients were treated bilaterally in two sequential procedures (28 total procedures performed). A median of 10 coils were placed per lung. AEs possibly related to the device or the procedure within 30 days of treatment included pneumothorax (1), pneumonia (2), COPD exacerbation (6), chest pain (4), and mild (<5 mL) self-limited hemoptysis (21). Between 30 days and 6 months, the main AEs were pneumonia (3) and COPD exacerbations (14). There were no life-threatening events. At 6 months, there were significant improvements in SGRQ, the primary efficacy endpoint (−14.9 ± 12.1 points, p <0.001; 79% of patients exceeded the MCID), FEV₁ (+14.9 ± 17.0%, p = 0.004; 64% exceeded the MCID⁴³), FVC (+13.4 ± 12.9%, p = 0.002), RV (−11.4 ± 9.0%, p <0.001), and 6MWD (+84.4 ± 73.4 m, p <0.001; 86% exceeded the MCID⁴⁴). Outcomes were better in patients who received bilateral treatment.

In a small prospective randomized control trial (RESET)⁴⁵ at three centers in the United Kingdom, 47 patients with severe emphysema (ULP, lower lobe predominant [LLP], or homogeneous emphysema) were randomized to LVRC treatment (n = 23) or optimal medical care (control, n = 24). The majority of treated patients underwent treatment of the contralateral lung at 1 month (n = 21). Coils were placed in segmental airways of the most affected lobe or lobes (on average, 18.5 coils per bilaterally treated patient). Improvement in SGRQ was significantly greater in the treated group (−8.11 points [95% CI −13.83 to −2.39] vs. 0.25 [95% CI −5.58 to 6.07] in the control group; p = 0.04 for between-group difference) with 65% of treated patients having an MCID versus only 22% of control patients (p = 0.01). At 90 days, mean improvements in FEV₁ and RV were significantly greater in the treated group (14.2 vs. 3.6%, p = 0.03; and −0.51 L vs. −0.20 L, p = 0.03; respectively). 6MWD increased by an average of 51.2 m in the treated group but decreased by 12.4 m in the control group (p <0.001 for between-group difference) with 74% of treated patients exceeding the MCID. In the first 30 days, there were more SAEs in the treatment group (two COPD exacerbations, two pneumothoraces, and two lower respiratory tract infections) than the control group (one COPD exacerbation). There was no between-group difference in SAEs from 30 to 90 days, and there were no deaths in either group.

While awaiting more definitive trials, the LVRC system has already demonstrated a number of attractive features, despite its need for fluoroscopy. Firstly, LVRC is unaffected by collateral ventilation—a major consideration with one-way valve placement. Perhaps, as a result, coil placement (at least in RESET) appears to be effective in a broad population with severe emphysema, not only in patients with heterogeneous disease but also those with homogeneous disease (where collateral ventilation is thought to be more prevalent⁴⁶). Like the valves, the coils can be removed or repositioned. A large randomized placebo-controlled trial (RENEW) is currently ongoing in the United States and Europe and plans to randomize 315 patients with severe emphysema and hyperinflation (RV ≥225%) to bilateral LVRC treatment versus optimal medical therapy. The primary outcome is functional (mean absolute change in 6MWD from baseline to 12 months) with subjective SGRQ changes at 12 months as a secondary outcome. Furthermore, treated patients will have long term 5-year follow-up with analyses stratified according to emphysema phenotype (heterogeneous or homogeneous).

Tissue Sealant

Like the one-way valves, tissue sealants are designed to directly reduce lung volume by collapsing and sealing damaged areas of hyperinflated lung. The site and mechanism of action are fundamentally different from the valves, however. This system acts at an alveolar rather than an airway level, thus overcoming the issue of collateral ventilation. Unlike valves and coils, this treatment is not reversible—it is intended to produce a permanent change in tissue configuration similar to LVRS.

The initial tissue sealant, developed by Aeris Therapeutics, Inc., used a series of biologically active reagents to promote collapse and scar formation in damaged regions of lung. Phase 1 and 2 trials^47–49 showed biologic lung volume reduction was safe and provided physiologic benefits in appropriately selected patients. Subsequently, Aeris developed the AeriSeal® System, a synthetic polymer foam sealant, which is delivered as follows: the treatment catheter is positioned 2 to 4 cm beyond the end of the bronchoscope under direct visualization; sealant components (5 mL) and air (15 mL) are mixed to generate 20 mL of liquid foam, which is rapidly injected through the catheter while maintaining wedge position; immediately post injection, the 30 mL of air is delivered to push the foam peripherally; the bronchoscope is left in wedge position for 1 minute to allow for complete polymerization of the sealant.

Three open-label, single-arm multicenter trials of the AeriSeal® System have been conducted in Europe and Israel. In the first,⁵⁰ 25 patients with heterogeneous ULP emphysema underwent unilateral unilobar treatment (upper lobe or superior segment lower lobe) of up to 6 subsegments over one to two treatment sessions. There were no serious procedural complications. Three patients had spillage of the material into central airways, but it was easily suctioned out. All patients experienced an inflammatory flu-like reaction beginning 8 to 24 hours following treatment that was generally mild and responsive to supportive care; this reaction was more significant in patients treated at four sites versus two to three sites in a single session. Short-term side effects were also more frequent among patients who received treatment at adjacent subsegments. There were eight treatment-related severe COPD exacerbations. Late treatment-related complications (>90 days post treatment) were not seen. There was no evidence of treatment-related mediastinal or pleural pathology. At 3 months, mean change in the ratio of residual volume to total lung capacity (RV/TLC, the primary endpoint) was −3.4 ± 9.2%, p = 0.09 (−4.7 ± 9.5%, p = 0.04 at 6 months) and the percentage of patients with clinically meaningful improvements in FEV₁, FVC, Medical Research Council Dyspnea score (MRCD)⁵¹, 6MWD, and SGRQ was 41%, 41%, 33%, 38%, and 50%, respectively. Similar results were seen at 6 months.

The second study⁵² included 56 patients with ULP (n = 19), LLP (n = 7), or homogeneous emphysema (n = 30). To reduce posttreatment acute inflammation, patients received steroids and antibiotics periprocedure. In addition, AeriSeal® System therapy was limited to nonadjacent subsegments. Patients received an initial treatment at two subsegments in one lobe and were eligible for repeat treatment after 12 weeks at 2 or 3 additional subsegments in the contralateral lung (39/56 subjects underwent a second treatment session). The modified protocol reduced acute and subacute side effects substantially (posttreatment inflammation decreased >60% and incidence of all-cause severe COPD exacerbations in the first 90 days decreased from 44% to 9%). Patients with a diffusing capacity for carbon monoxide (DLCO) between 20% and 60% predicted who were treated in the upper lobes (either ULP or homogeneous with reduced perfusion to the upper lobes) had significantly improved lung function, gas trapping, and respiratory-related quality of life at 3 months. For unclear reasons, patients treated in the lower lobes did not improve.

Based on these results, a third prospective study⁵³ was performed, incorporating both modified protocol and patient selection criteria based on the earlier findings. Twenty patients with a baseline DLCO between 20% and 60% predicted and ULP (n = 10) or homogeneous disease with decreased upper lobe perfusion (n = 10) received bilateral upper lobe treatment (two nonadjacent subsegments/upper lobe) with the AeriSeal® System. At 3 months, there was a significant reduction in upper lobe lung volume on CT (−895 ± 484 mL, p < 0.001 = primary endpoint) that was durable out to 1 year. Treatment was also associated with significant improvements in FEV₁ (265 ± 248 mL, 28.9 ± 30.6%, p = 0.003), FVC (251 ± 405 mL, 11.8 ± 19.4%, p = 0.033), RV (−639 ± 894 mL, −10.9 ± 18.2%, p = 0.036), dyspnea [∆MRCD −1 (−1 to 1), p = 0.011], and SGRQ (−9.1 ± 12.9 points, p = 0.016) at 3 months that also persisted to 1 year. TLVR as well as clinical improvements were greater in patients with heterogeneous ULP disease (Fig. 101-4). There was one serious procedural complication and seven all-cause significant respiratory AEs over 17 patient-years of follow-up.

ASPIRE, a multicenter randomized controlled trial of AeriSeal® System treatment plus optimal medical therapy compared to optimal medical therapy alone in patients with advanced ULP heterogeneous emphysema, is currently underway in the US, Europe, and Israel. Approximately 300 patients will be randomized 3:2 to treatment versus medical therapy. The primary endpoint is the mean change from baseline in postbronchodilator FEV₁ at 12 months; other objective as well as subjective secondary measures include change in CT upper lobe volume and 6MWD as well as the proportion of patients achieving a MCID in SGRQ and MRCD. Long-term follow-up will continue through 5 years in treated patients.

Bronchial Thermal Vapor Ablation

Like a tissue sealant, bronchial thermal vapor ablation (BTVA) causes lung volume reduction by producing irreversible fibrosis and scarring of the target area and is insensitive to collateral ventilation. The InterVapor® System (Uptake Medical Corp) includes three components: an application that provides preferred treatment locations and times based on the patient's CT, a vapor generator, and a vapor catheter. The vapor generator is an electronically controlled pressure vessel that generates and delivers precise amounts of energy (heated vapor) through the vapor catheter (flexible shaft with occlusion balloon at the distal end) into a targeted lung segment. The vapor catheter is introduced through the bronchoscope to the airway of the lung segment selected for treatment. An occlusion balloon is inflated (to protect other airways from the heated vapor), and the vapor dose is delivered. The heated vapor induces a thermal reaction in targeted areas of lung leading to a localized inflammatory response followed by fibrosis of airways and parenchyma.

Snell and colleagues⁵⁴ reported the results of two open-label, single-arm efficacy and safety clinical studies with a total of 44 patients with ULP emphysema who received unilateral upper lobe BTVA. At 6 months, the average volume loss in the treated lobe (measured by CT) was 715.5 ± 99.4 mL (p <0.001), a 48% reduction in lobar volume (Fig. 101-5). The mean increase in FEV₁ at 6 months (coprimary efficacy endpoint) was 140.8 ± 26.3 mL, 17% (p < 0.001) with 55% of patients achieving the MCID. SGRQ (coprimary efficacy endpoint) improved by 14 ± 2.4 points (p < 0.001) with 73% of patients exceeding the MCID. There were also significant improvements in RV (−406 ± 112.9 mL, p < 0.001), 6MWD (46.5 ± 10.6 m, p < 0.001), and mMRC dyspnea score (−0.9 ± 0.17 points, p < 0.001 with 63% of patients exceeding the MCID⁵¹). There were no AEs during the procedure. In the first 30 days, there were 11 SAEs, and 29 over 6 months including one death. The majority of SAEs at both time points were respiratory-related. A posttreatment inflammatory reaction (elevated inflammatory markers ± symptoms) peaked within 2 to 4 weeks and resolved within 8 to 12 weeks.

Herth and colleagues⁵⁵ followed the above cohort out to 12 months. Changes from baseline in treated upper lobe and ipsilateral lower lobe volumes were similar; however, improvements in physiologic and clinical measures were smaller in magnitude than those seen at 6 months (∆FEV₁ 86.2 ± 173.8 mL (10%), p < 0.05, 46% achieved a MCID compared with baseline; ∆SGRQ −11 ± 14 units, p = 0.05, 68% met MCID from baseline). Responses were greater in patients with higher versus lower heterogeneity indices. Incidence of SAEs diminished over time with only 10 between 6 and 12 months.

Airway Bypass

Instead of attempting to collapse damaged regions of lung, airway bypass creates extra-anatomic passages between damaged, collaterally ventilated lung parenchyma and the central airways. This approach is designed to bypass small, collapsible, high-resistance airways in damaged emphysematous lung by creating low-impedance pathways into the central airways, resulting in more effective emptying and improved respiratory mechanics. While this procedure should theoretically be useful in heterogeneous or homogeneous emphysema with collateral ventilation, trials have focused on homogeneous disease where collateral ventilation is prevalent. The procedure is performed by passing a flexible bronchoscope into an area of known emphysema and using a Doppler probe to identify an area free of blood vessels. Next the bronchial wall is pierced with a transbronchial needle and dilating balloon. Finally, a stent is placed to expand and maintain the new passage between the airway and adjacent lung tissue (Fig. 101-6). The Exhale® drug-eluting stent (Broncus Technologies, Inc.) is composed of stainless steel and silicone and contains paclitaxel, which is intended to inhibit fibrotic responses and improve long-term passage patency (in previous animal studies, non-drug-eluting stents became occluded within 1 week of placement).^56,57

In a multicenter pilot study,⁵⁸ 35 patients received bilateral upper lobe airway bypass with the Exhale® Emphysema Treatment System. Ninety-four percent of patients had homogeneous emphysema. A median of eight stents were placed per patient. At 1-month follow-up, there were statistically significant improvements in RV, TLC, FVC, FEV₁, mMRC, 6MWD, and SGRQ. But, at 6 months, only improvements in RV and mMRC remained significant (−400 mL [−6.6%], p = 0.04 and −0.5 points, p = 0.025, respectively). A retrospective analysis suggested the degree of pretreatment hyperinflation may predict which patients achieve the best results, as patients with an RV/TLC ratio above the median of 0.67 had greater benefit (though results still only significant for RV (−870 mL [−14.1%], p = 0.022) and mMRC [−0.5 points, p = 0.035] at 6 months). For less hyperinflated patients, no outcome measures were significant at 6 months. In a subset of patients, bronchoscopic exam at 6 months revealed 18/26 (69%) of the stents were patent. There were three intraoperative SAEs including one death due to major bleeding into the airway and two episodes of pneumomediastinum/subcutaneous emphysema. Postoperatively, 22 patients experienced SAEs, most commonly COPD exacerbations and infection. Despite antibiotic prophylaxis, five patients had respiratory infections in the first week.

Exhale® airway stents for emphysema (EASE)⁵⁹ was a multicenter, double-blind, randomized controlled trial in which 315 patients with severe hyperinflation (RV/TLC ≥0.65) and homogeneous emphysema were randomized to airway bypass (208) or sham bronchoscopy (107). Patients were followed for 12 months. A mean of 4.7 stents were placed per patient. While there were early improvements in the airway bypass group, the benefits declined by 1 month. At 6 months, there was no difference between arms in the coprimary efficacy endpoint (FVC increase ≥12% and decrease in mMRC score by ≥1 point). Stent expectoration (clinically reported by 11.5% of participants) and passage occlusion (secondary to mucus, granulation tissue, and lack of adequate, maintained airflow) may have contributed to lack of efficacy. In the airway bypass group, the 7-day rate of SAEs was 3.4% versus 0% in sham group. At 6 months, there was no difference in SAE rates between the groups; however, there were more moderate COPD exacerbations in the bypass group.

Lung volume reduction surgery removes emphysematous, hyperinflated lung tissue, thereby improving the mechanical relationship between the lung and the surrounding chest wall and allowing the remaining more normal lung to expand and function better. Beyond smoking cessation and oxygen therapy, LVRS is the only treatment that definitively prolongs survival in patients with emphysema. However, given the substantial morbidity and mortality associated with major thoracic surgery in an already sick population, less invasive bronchoscopic strategies capable of achieving similar benefits have been sought. Although none of these technologies (one-way valves, coils, tissue sealant, bronchial thermal vapor ablation, and airway bypass procedures) are FDA-approved to date, all are approved for clinical use outside the United States. Despite diverse mechanisms of action, all procedures can be performed via flexible bronchoscopy (typically under moderate sedation rather than general anesthesia) and are associated with substantially shorter inpatient stays than LVRS and lower procedural mortality. In addition, nontraditional LVRS candidates (i.e., those with nonULP or homogeneous emphysema) or patients with significant comorbidities may, in the future, be eligible for bronchoscopic lung volume reduction. While differences in endpoints make comparisons between studies somewhat difficult and most studies to date have had relatively short follow-up compared with 5-year follow-up of NETT patients,⁶⁰evidence to date suggests differential efficacy and safety between types of bronchoscopic lung volume reduction. Furthermore, it is clear that certain modalities work best in certain patients and improved mechanistic understanding and therapeutic targeting may well be a way to improve outcomes beyond device modifications alone. As trials continue, it is likely that some of these methods will prove clinically useful, helping to reduce the tremendous suffering and burdens associated with advanced emphysema.