Intrabronchial valves for persistent air-leaks: what’s the verdict?

KEYWORDS:

• Intrabronchial valve
• persistent air leak
• alveolar-pleural fistula
• bronchopleural fistula
• pneumothorax

Bronchopleural fistula and alveolar-pleural fistula represent the abnormal passage of air from the bronchi and alveoli directly into the pleural space. Although there is no standard definition, air leaks persisting 3–7 days following thoracic surgery are considered abnormal and often termed a persistent air leak (PAL) [1]. No such definition for nonsurgical PAL currently exists, but has often been extrapolated to utilize the same time frame.

The multiple causes of a PAL can generally be grouped into several categories: postoperative, iatrogenic, and non-iatrogenic (primary spontaneous pneumothorax and secondary spontaneous pneumothorax). Postoperative air leaks, often termed ‘surgical air leaks’, can develop in the setting of any thoracic surgery, often leading to increased hospital lengths of stay and cost [1–3]. Iatrogenic air leaks can be precipitated by nonsurgical procedures such as bronchoscopic lung biopsies, transthoracic lung biopsies, pleural procedures, or numerous other etiologies. Non-iatrogenic causes can include infection, barotrauma, and spontaneous pneumothoraces. Whether iatrogenic or non-iatrogenic in etiology, a PAL represents a clinical dilemma, with little guidance in management and even less guidance in evidence-based interventions.

Despite the lack of data, multiple treatment options exist for PAL in the setting of active chest tube drainage with recommendations often from the surgical literature [1,4,5]. Typical management ranges from watchful waiting with continued pleural drainage (with or without an ambulatory Heimlich valve) to pleurodesis (surgical versus nonsurgical). Primary spontaneous pneumothoraces rarely result in PAL; however, secondary spontaneous pneumothoraces can often result in high rates of PAL and are frequently treated with more definitive pleural interventions [6]. Bronchoscopic treatment modalities for PAL have recently developed after the realization that occluding the affected bronchi can retard airflow and allow for closure with subsequent healing of the fistula. Although several valves have been designed and utilized for this purpose, including the Watanabe spigot (Novatech, Grasse, France) and early versions of an endobronchial valve (Emphasys Medical, Redwood City, CA), they are not available for use in the United States [1].

A one-way intrabronchial valve (IBV), created by Spiration Inc. (Olympus Respiratory, Redmond, WA, USA), was initially designed for bronchoscopic lung volume reduction [7]. However, after multiple United States clinical trials were unsuccessful in establishing clinical efficacy for lung volume reduction, it became apparent that this IBV may be potentially helpful in the management of PAL. In 2008, a Humanitarian Device Exemption (HDE) from the Food and Drug Administration approved IBV placement for postsurgical PAL under compassionate use [8]. The HDE is generally designed to bring new technology to clinical use when/if there is a limited population affected (<4000 patients per year) and companies could not otherwise bring the technology to the market. As part of this approval process, data on efficacy is not actually required, but rather that ‘… the device does not pose an unreasonable or significant risk of illness or injury, and that the probable benefit to health outweighs the risk …’. Since its HDE approval for use in postsurgical PAL, additional data has been published in both postsurgical and nonsurgical PAL. Furthermore, two separate reviews have provided overall guidance for the procedure, including potential indications, patient selection, and billing recommendations [9,10].

One of the first case series on IBV use consisted of 7 patients undergoing 8 procedures for PAL lasting for a median of 4 weeks [11]. Patients included within this study had evidence of PAL (half were postsurgical and half were spontaneous pneumothorax) and failed to resolve with standard surgical and nonsurgical therapies. The majority of procedures were identified as successful with cessation of the leak in 6 of 8 procedures after a mean of 5.2 days. No particular complications were noted within the trial related to IBV placement; however, two patients in the series did not have valves placed after initial bronchoscopy as the PAL was not isolated during balloon occlusion. Although the main limitations of this study were the lack of a control group and its small size, it provided some much needed information as one of the first series utilizing IBV for PAL.

A few prospective studies do exist, but are weighted toward surgical PAL and relatively small in population size. A single-center study of 16 patients (mainly nonsurgical) identified 13 patients undergoing IBV placement (both Spiration and Emphasys valve systems), with 10 classified as ‘responders.’ This response was defined by chest tube removal without further intervention, which occurred at a median time of 7.6 days after IBV placement. No major complications were reported in this series. Many patients were identified as poor surgical candidates due to severe comorbidities or previous interventions (radical pleurectomy for mesothelioma—3 patients, empyema with tube thoracostomy and/or surgical decortication—4 patients) [12]. A single-center, prospective study evaluating IBV placement for PAL after anatomical surgical lung resection identified 277 lung resection surgeries in which 10 patients were diagnosed with PAL. Nine of the 10 (90%) underwent successful balloon occlusion and IBV placement. Median PAL cessation was 2 days after IBV, with median chest tube removal of 4 days. They noted 6 patients had near complete resolution within 24 h, whereas the remaining 3 had a continued leak, but to a lesser extent. No major complications were reported [13].

Three larger retrospective studies have offered more IBV data, including off-label usage. Another single-center retrospective trial reported on 21 patients with PAL, ongoing pleural drainage, and an inability to tolerate water seal or ambulatory drainage. All patients underwent IBV implantation, with eight on-label and 13 off-label indications. On-label use resulted in median time to discharge after IBV placement of 3 days; however, half the patients were discharged with pleural drainage in situ (overall median pleural drainage time = 13.5 days). Off-label use for spontaneous pneumothorax (n = 10) and iatrogenic pneumothorax (n = 1) (excluding the two pneumonectomy patients undergoing IBV implantation) resulted in median pleural drainage of 7 days (range 0–83), but longer post-IBV placement hospital stays (mean of 25 days with a broad range), presumably secondary to multiple comorbid conditions [14]. This study was limited again by its single-center design and small sample size. A multicenter, retrospective review of the Zephyr valve (currently unavailable in the United States) identified 40 patients with PAL, and is the largest to date study, utilizing that valve system for PAL. The majority of patients were nonsurgical (80%), with multiple comorbidities including cancer (30%), chronic obstructive pulmonary disease (30%), and pneumonia (7.5%). The median time from valve implantation to chest tube removal was 7.5 (IQR, 3–29) days. Adverse events were reported in six patients ranging from valve expectoration to pneumonia [15]. More recently, the largest and only multicenter retrospective study in patients with PAL undergoing placement of the Spiration IBV (the only available IBV in the United States) was reported. A total of 112 potential patients were identified with 75 patients (67%) undergoing IBV implantation. Following IBV placement, the mean time to PAL resolution was 4 days (range 0–101), with 56% resolving within 1 day or less. However, PAL was still present more than 1 week after IBV placement in 37% of patients. The least successful group appeared to be PAL from secondary spontaneous pneumothorax, with only 58% having PAL resolution after IBV placement (mean resolution time was 15.6 days). There were two complications reported in the population: an empyema after IBV placement in one patient and the development of a contralateral pneumothorax in one patient during bronchoscopy [16].

Despite the dearth of data, potential advantages for IBV use clearly exist. IBV placement, when compared with alternative thoracoscopic interventions, appears to offer a less invasive option. Valves are relatively easy to place with likely overall low risk of procedural complications. Multiple studies and case series have documented good safety profiles; however, publication bias for positive results of IBV implantation will likely exist, thus making it difficult for a likely balanced interpretation of current IBV usage. Disadvantages of IBV placement for PAL relate mainly to the lack of controlled studies that demonstrate efficacy or cost-effectiveness. There are numerous case reports exhibiting success; however, it also appears that unsuccessful IBV utilization may occur in up to one-third of patients considered for implantation. Additional disadvantages include the need for two bronchoscopic procedures (implantation and retrieval) and that balloon occlusion is unsuccessful in 10–20% of bronchoscopies, therefore subjecting those patients to a procedure without a resulting intervention.

The future of IBV placement for PAL remains unclear. The currently reported data suggests efficacy and a tolerable safety profile in on-label use for postsurgical PAL. However, more data is needed to better define specific comparative outcomes such as rate of complete and partial resolution of the PAL, time to removal of chest tube, length of hospitalization, need for surgery, and cost-effectiveness. It would also be helpful to determine factors leading to successful balloon occlusion necessary for subsequent valve placement so as to limit unnecessary bronchoscopies.

The potential ease of bronchoscopic placement, avoidance of surgical intervention, and desire to ‘do something’ is obviously attractive to physicians caring for these complex patients as many are often quite ill with a number of comorbid conditions. However, an unproven, easy procedure should not trump time-tested management of prolonged pleural drainage and/or a more invasive procedure such as pleurodesis without further data supporting its role. A prospective trial attempting to compare IBV placement versus standard of care is currently registered and recruiting patients [17]. However, its current study design may likely produce more questions than answers at its completion. Unfortunately, it appears the jury may still be out on the use of IBV in PAL and likely will be for some time.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.