Surgical Management of Prolonged Air Leak in Patients with Underlying Emphysema

Prolonged air leak is one of the most common post-operative complications encountered after thoracic surgical operations involving mobilization or resection of lung parenchyma. Air leak typically manifests as persistent bubbling in a chest tube drainage system, but may also present with increasing subcutaneous emphysema or pneumothorax in a post-operative patient. No universal consensus exist as to the exact duration of air leak which constitutes a prolonged air leak, but it is generally regarded to exist if it is present for more than 5 days(14) or 7 days(2, 5-7) after initial surgery. It is an important complication that results in increased length of stay(8-15) and has been associated with other post-operative complications such as pneumonia(12, 14, 16), empyema(9, 10, 16) and ICU re-admission(12).


Introduction
Prolonged air leak is one of the most common post-operative complications encountered after thoracic surgical operations involving mobilization or resection of lung parenchyma.Air leak typically manifests as persistent bubbling in a chest tube drainage system, but may also present with increasing subcutaneous emphysema or pneumothorax in a post-operative patient.No universal consensus exist as to the exact duration of air leak which constitutes a prolonged air leak, but it is generally regarded to exist if it is present for more than 5 days(1-4) or 7 days(2, 5-7) after initial surgery.It is an important complication that results in increased length of stay (8)(9)(10)(11)(12)(13)(14)(15) and has been associated with other post-operative complications such as pneumonia (12,14,16), empyema (9,10,16) and ICU re-admission (12).
Patients with emphysema form a significant proportion of patients which will undergo thoracic surgical operations.Chronic smoking and emphysema predisposes an individual to developing a pneumothorax (17,18) or carcinoma of the lung (19,20) that may require surgical intervention for treatment.In addition, lung volume reduction surgery plays a role in the management of certain patients with advanced emphysema (21).Conversely, emphysema is regarded as a risk factor for developing prolonged air leak in cases where patients with emphysema require an operation (7).This is presumably because the underlying lung substrate in patients with emphysema is more easily injured during surgery and takes longer to heal.
The role of emphysema as a risk factor for prolonged air leak has been inferred from numerous surgical case series which reliably demonstrate that patients noted preoperatively to have emphysema will have a higher incidence of prolonged air leak.However, a major weakness of these studies, is that they are heterogenous in their definition of prolonged air leak, patient population (eg age, definition of impaired lung function), type of operation performed (eg video assisted vs open, chemical vs mechanical pleurodesis, type/extent of resection) and methods used to prevent air leak (eg use of pleural tenting), which limits the ability to compare between individual studies.In addition, several studies analyzing the specific risk factors for developing this complication have consistently shown that low FEV1 or FEV1/FVC will increase the risk of developing prolonged air leak after either pulmonary resection or lung volume reduction surgery (see below for details).Fig. 1.Severe subcutaneous emphysema in a patient with underlying emphysema with prolonged air-leak.
For surgical pleurodesis, several authors have described their experience in performing this operation on both primary spontaneous pneumothorax and secondary spontaneous pneumothorax (which mainly consist of patients with underlying emphysema).The reported incidence of prolonged air leak in patients with primary spontaneous pneumothorax undergoing surgical pleurodesis has been reported to range from 0-3.8%, while it has been reported to range from 7.1-29.1% for patients with secondary spontaneous pneumothorax.A similar trend is also demonstrable in patients undergoing pulmonary resection for carcinoma of the lung, with an incidence of prolonged air leak of 4.2-18.2% in patients without underlying emphysema, compared to 5.4-44% in patients with underlying emphysema.For lung volume reduction surgery, the incidence of prolonged air leak is much higher, ranging from 39-45.2%.This is expected, as the operation is conducted on both lungs, and usually on patients with more advanced underlying lung disease.This review will discuss the pathogenesis, risk factors, intra-operative and post-operative management strategies for prolonged air leak in patients with emphysema based on current available literature.In addition, we propose an algorithm for the management of prolonged air leak in this group of patients based on this discussion, and also define specific criteria for surgical intervention for prolonged air leak that we follow at our institution.Several recent reviews have previously discussed the problem of prolonged air leaks, but do not focus specifically on patients with emphysema (3,4) or neglect to discuss the utility of surgical intervention in greater detail (2,34) which we believe plays an important role for this challenging clinical problem, particularly in the small number (but no less important) of patients who are refractory to all other forms of therapy.

Pathogenesis and factors influencing incidence of prolonged air leak in patients with emphysema
Some degree of post-operative air leak is generally unavoidable in operations involving pulmonary resection or mobilization, usually reflective of an alveolo-pleural fistula arising from exposed alveoli, whereas more severe leaks suggest fistulas arising from larger, more proximal bronchi (5,7).The duration of the leak is related to the severity of the air leak as well as the time taken for the exposed parenchyma to heal, which occurs via an inflammatory reaction that results in granulation tissue formation and fibrin deposition (7).Moreover, this process is widely accepted to be facilitated by re-expansion of the lung to allow contact between the lung and parietal pleura.
Thus, it would follow that factors that would increase the risk of prolonged air leak include impaired wound healing (older age, more severe underlying emphysema), greater intraoperative surgical trauma (re-operations, extensive adhesions) and incomplete lung expansion post-operatively.This has been confirmed by a number of studies on patients undergoing pulmonary resection which have looked at specific factors that influence the incidence of prolonged air leak, summarized below.
Though no study looked specifically at risk factors for prolonged air leak in patients with emphysema undergoing pulmonary resection, DeCamp and colleagues (12)   Based on the above mentioned factors, methods geared to the prevention of prolonged air leaks aim to minimize intra-operative surgica l t r a u m a o r e n s u r e m o r e c o m p l e t e l u n g expansion.These approaches can be broadly divided into intra-operative and post-operative strategies.

General
The thoracic surgeon should ensure that lung tissue is handled as carefully as possible during dissection and manipulation to ensure minimal trauma, particularly in patients with emphysema, where the underlying lung is fragile.Any obvious parenchymal tears that are identified during surgery should be repaired meticulously.In addition, the remaining lung should be completely mobilized and decortication should be performed if necessary to aid maximal re-expansion of remaining lung tissue after pulmonary resection.

Fissureless technique for lobectomy
Conventional lobectomy involves dissection of lung parenchyma within the fissures by sharp or blunt dissection for exposure of the pulmonary artery that may result in air leaks.
The fissureless technique involves exposing the pulmonary artery without such dissection, only using staplers for division of lung parenchyma when it is required (40,41).
Although the efficacy of this technique has not been studied in patients with emphysema specifically, two previous studies on a general population of patients undergoing pulmonary resection have shown that this technique significantly decreases the incidence of prolonged air leak.Gomez-Caro and associates (42) demonstrated in a randomized prospective study of 63 patients undergoing either lobectomy or bilobectomy, that the incidence of prolonged air leak (>5 days) in patients whom a fissureless technique was employed was 3.2%, compared to 21.8% for those in whom conventional dissection was performed.A more recent retrospective case control study by Ng et al. (43) looking at 93 patients undergoing right upper lobectomy only, revealed similar results, with patients in the fissureless technique group having an incidence of prolonged air leak (>7 days) of 7.6%, compared to 22.2% in patients in the conventional lobectomy group.

No cut plication (non-resectional) technique for lung volume reduction surgery
For lung volume reduction surgery, an alternative technique involving no cut plication has been described by various authors as having lower rates of prolonged air leak while having short to intermediate term improvement in pulmonary function comparable to conventional lung volume reduction surgery (44)(45)(46)(47).With this alternative technique, lung tissue is folded up or pushed down onto itself before being stapled together instead of performing staple excision of lung tissue in traditional lung volume reduction surgery.
Swanson and colleagues reported that in their series of 50 procedures performed on 32 patients, the incidence of prolonged air leak (>7 days) was only 8.6% (44).In a series of 20 patients operated by Iwasaki and associates, they reported that no patient had an air leak beyond 5 days (45).The largest reported series of 66 patients at Tor Vergata University by Tacconi, Pompeo and Mineo, demonstrated an incidence of prolonged air leak (>7 days) of 18% in patients undergoing non-resectional lung volume reduction surgery under thoracic epidural anaesthesia, compared to 40% of patients in a control group undergoing conventional lung volume reduction surgery under general anaesthesia (48).
Moreover, Pompeo and colleagues at the Tor Vergata University also recently published a randomized control trial comparing 32 patients undergoing non-resectional lung volume reduction surgery with thoracic epidural anaesthesia against 31 patients undergoing conventional lung volume reduction surgery with general anaesthesia and found that the incidence of prolonged air leak in the former was 18.8% compared to 48.4% for the latter, while survival and improvement in post-operative pulmonary function were similar in both groups (49).The same group also compared the results of 41 patients undergoing nonresectional lung volume reduction surgery under thoracic epidural anaesthesia against 19 patients undergoing non-resectional lung volume reduction surgery under general anaesthesia, and found that the occurrence of prolonged air leak was similar between the two groups (12.1% vs 26.3%, p=0.26), which suggests that the type of lung volume reduction surgery rather than the type of anaesthesia was the main factor in determining risk of prolonged air leak (50).
The above published data indicate that this technique may potentially be superior to the traditional lung volume reduction surgical approach in terms of reducing morbidity from prolonged air leak.However, the long-term durability of pulmonary function improvement after plication is still not known, as the studies so far have only involved small numbers of patients and only limited follow-up, thus more research on this technique is required before its widespread adoption can be recommended.

Buttress material for staple lines
Another area of study in the intra-operative prevention of air leaks during thoracic surgery has been the use of buttress material for staple lines, which in theory would help reinforce the fragile staple lines and thus prevent air leak from these areas of weakness.A variety of buttress materials have been described for this purpose, both synthetic (eg polytetrafluoroethylene (51), polydioxanone( 52)) and biological [bovine pericardial strips (53)(54)(55)(56), bovine collagen(57), autologous parietal pleura (58)].However, only a few have been investigated in clinical practice, the most widely studied of which are bovine pericardial strips.
Unfortunately, the cost of using these are high (57), and the few small studies that have been performed on a general population of patients undergoing pulmonary resection have not shown a clear benefit (53,54).Several studies directed at emphysema patients specifically have been performed with more consistent results, but these are limited to those undergoing lung volume reduction surgery or bullectomy (55,56,58).On the other hand, an analysis of factors influencing post-operative air leak in patients undergoing lung volume reduction surgery in the National Emphsema Treatment Trial did not find that use of staple line buttressing (regardless of material) helpful in preventing or shortening duration of air leak (12).
In summary, current evidence suggest that the use of buttressing staple lines in patients with emphysema undergoing lung volume reduction surgery or bullectomy may be useful in reducing incidence of prolonged air leak, but its use in other operations, particularly pulmonary resection has not been demonstrated.
A table summarizing the results of the various studies mentioned above is presented below.

Pulmonary sealants
Pulmonary sealants have been the focus of a large amount of research in the area of intraoperative prevention air leaks, with over a dozen studies on various types of sealants including fibrin glue(59-62), PEG-based sealants(63-70) and coated collagen patches(71-73).However, as with studies on other strategies, these papers have generally not focused on patients with emphysema, and individually these studies each have small cohort sizes with very mixed patient populations as well as varying methods for reporting efficacy.
Moreover, the overall results of these studies so far have found no clear advantage in their routine use on all patients (74).Thus, the use of sealants should best be reserved for patients at highest risk for developing post-operative prolonged air leak (35,38), especially since rare complications, particularly empyema (63,67,75) may arise from the use of pulmonary sealants.Indeed, the studies which have focused on patients with emphysema have more consistently shown a significant reduction in the incidence of post-operative prolonged air leak and length of stay (62,73).Table 6.Studies comparing the utility of pulmonary sealants in preventing prolonged air leak.

Minimizing post-resectional spaces
Minimizing the potential space left behind after pulmonary resection allows for a more complete apposition of the lung surface with the parietal pleura to encourage the resolution of any post-operative air leak.Usually this can be accomplished with straightforward means such as the proper placement of chest tubes, division of the inferior pulmonary ligament and lysis of all adhesions at the conclusion of surgery or the use of adequate analgesia, chest physiotherapy or bronchoscopy to clear the airways of mucus and blood post-operatively to promote maximal re-expansion of the residual lung (7).In the event that the above mentioned methods are insufficient, several techniques have been described, including the creation of a pleural tent, creation of a pneumoperitoneum or deliberate diaphragmatic paralysis.
Again, interpretation of the results of studies on these methods to reduce post-resectional spaces is complicated by the heterogenous inclusion criteria and method of reporting outcomes in these studies.Furthermore, almost none have looked specifically at patients with emphysema, thus making it difficult to simply extrapolate the results of these studies to patients with emphysema.
Nonetheless, amongst the methods mentioned previously, pleural tenting has been the most widely studied technique for preventing prolonged air leak by minimizing post-resectional spaces.This involves stripping the parietal pleural over the apex, which is then resutured over the chest wall to produce an extrapleural space (7,77).It has been used as a means for controlling the size of the potential space post-pulmonary resection in the upper thoracic cavity, and thus has been predominantly studied in patients undergoing upper lobectomy.
In a retrospective review on risk factors for prolonged post-operative air leak, Brunelli and associates( 16) noted that patients with upper lobectomies who underwent a pleural tent had a significantly decreased duration of air leak compared to those who did not undergo a similar adjunctive procedure.Nevertheless, he later published a retrospective case matched analysis comparing patients with prolonged air leak after pulmonary resection and those without, which did not demonstrate that pleural tenting conferred any protective effect(9).DeCamp et al. (12) in reviewing the factors influencing air leak post-lung volume reduction surgery in patients from the National Emphysema Treatment Trial also did not find a significant decrease in incidence or duration of air leak in patients who underwent tenting compared to those who did not undergo tenting.
In addition, a number of randomized prospective studies have also been performed to assess its efficacy, and in general, the studies conducted on pleural tenting have shown an overall beneficial effect in terms of decreasing incidence of air leak, time to chest tube removal and length of stay.However, this procedure adds to operative time and may cause bleeding(35) though these were not shown to be significantly increased compared to controls in the studies below.
The Conversely, the creation a pneumoperitoneum has been utilized to minimize the postresectional space in the lower thoracic cavity.This has been described as both an intraoperative adjunct to prevent prolonged air leak (80) as well as a post-operative technique (81)(82)(83) to treat it.It can be accomplished through instillation of air into the peritoneal cavity by www.intechopen.coma variety of means, including under direct vision through a transdiaphragmatic opening made in the diaphragm during surgery (80), via insertion of a peritoneal dialysis catheter under local anesthesia (81), or with the aid of a Veres needle under local anesthesia (82,83) .
A small randomized prospective trial by Cerfolio and colleagues(80) studied 16 patients undergoing right middle and lower bilobectomy, dividing them into a group who underwent intra-operative pneumoperitoneum creation and a group who did not undergo this procedure.0/8 patients with an intra-operative pneumoperitoneum had air leak by POD3, compared to 4/8 patients who did not have an intra-operative pneumoperitoneum (p<0.001).Moreover, patients in the former group had a median hospitalization stay of 4 days compared to 6 days for patients in the latter group (p<0.001).Thus, this is an interesting technique, but conclusions on its efficacy are difficult to draw based on the limited data available.The results of postoperative pneumoperitoneum creation will be discussed later in the section on post-operative strategies for management of prolonged air leak.
Deliberate diaphragmatic paralysis is an alternative method used to decrease the potential space in the lower thoracic cavity to allow for more rapid resolution of air leak.Several means are available to achieve this, including infiltration of the phrenic nerve with local anesthetic, phrenic nerve crush or sectioning.The main drawback of diaphragmatic paralysis is the compromise in ventilatory function and cough mechanism.Thus, the use of para-phrenic local anesthetic has the advantage over phrenic nerve crush or sectioning, in that it only resulting in temporary paralysis, so that diaphragmatic function may recover after the effect of the local anesthetic wears off.A recent case report by Clavero and associates (84) explains how an epidural catheter can be placed in close proximity of the phrenic nerve through video-assisted thoracoscopic surgery or thoracotomy, so that the managing physician can dictate the exact duration of diaphragmatic paralysis required to resolve the air leak before reversing the effect of the local anesthetic infusion.However, no large studies specifically describing the use of diaphragmatic paralysis for preventing prolonged air leaks are available.

Post-operative strategies for management of prolonged air leak 4.1 Bronchoscopy and endobronchial techniques
Bronchoscopy plays an important role in the post-operative management of prolonged air leak.It can be used to clear the airways of mucus and blood to aid maximal re-expansion of the lung to promote resolution of air leaks.Furthermore, it should be performed in all patients with persistent air leak to exclude stump dehiscence, as its presence will often necessitate surgery to treat the problem.Should surgery be contraindicated for whatever reason, a large number of endobronchial approaches have been studied as an alternative therapeutic option for bronchopleural fistulas, including the use of glue (85,86), polidocanol (87), tetracycline (88), coils (89), surgicel (90), gelfoam (91), tracheobronchial stents (92), atrial septal defect closure devices (93) and even lasers (94).Unfortunately, experience with these techniques have been limited to mostly case reports and case series, with no controlled studies comparing the different methods or comparing them against surgical therapy.A recent systematic review of several of the larger case series by West et al. (95) showed that among 85 patients with post-pneumonectomy bronchopleural fistulas, endobronchial therapy (40 fibrin glue, 15 cyanoacrylate glue, 19 polidocanol, 6 lasers, 5 stents) succeeded in treating only 30% of them.Overall mortality was 40%, with many patients requiring multiple bronchoscopic procedures or additional surgical drainage.
In addition, the placement of endobronchial valves is a new technique that has emerged recently for the treatment of persistent air leak in patients with underlying lung disease such as emphysema that are not candidates for more extensive procedures such as surgery (96,97).Endobronchial one-wave valves inserted via bronchoscopy were initially developed as an investigational technique to treat emphysema by promoting atelectasis of emphysematous lungs distal to the valve, which would allow air to exit via the valve but not re-enter.They have now been used in selected patients with persistent air leaks, in hope that they accelerate closure of the leak by minimizing flow of air through the leak (98).
The procedure can be performed either under sedation or general anesthesia, using either a flexible or rigid bronchoscope.A balloon tipped catheter is used to provide selective bronchial occlusion to determine the segmental or subsegmental airway that results in the greatest decrease in air leak.The endobronchial valve is then inserted in these airways (98,99).The results of the two largest series on endobronchial valve placement are summarized below, and the overall conclusion is it is a promising mode of therapy particularly for patients with no other therapeutic options.

Bedside pleurodesis
Instillation of a sclerosing agent into the pleural space elicits an inflammatory reaction in the pleura that allows for the obliteration of the pleural space and resolution of an air leak.A variety of agents have been described for this purpose, including silver nitrate(100), quinacrine (101), minocycline (102), tetracycline (103), doxycycline (104), erythromycin (105), bleomycin (106), iodopovidone (107), talc powder (14) and autologous blood (108)(109)(110)(111). Be that as it may, contemporary literature has mainly focused on autologous blood for treatment of persistent post-operative air leaks, so the utility of the other agents for this clinical context are not as well known.Also, these studies were not limited to patients with emphysema, so their results may not be directly applicable for these patients with persistent post-operative air leaks.However, based on available published data, bedside pleurodesis is a reasonably efficacious modality of treatment with few adverse effects, so it is often used as first line therapy for patients with prolonged air leak, even in our own institution.
Several small observational studies have demonstrated the efficacy and safety of autologous blood in treating post-operative prolonged air leak (108)(109)(110).In these studies, patients with prolonged air leak (>5-10 days) after undergoing a variety of operations (lobectomy, wedge resection, bullectomy, lung volume reduction or decortication) were treated with 1-2 injections of autologous blood pleurodesis with resolution of air leak in all patients within 48 hours of therapy.No major complications occurred except for fever, pneumonia or prolonged pleural effusion in a minority of patients.
In addition, Shackcloth et al. (111) performed a randomized prospective study on 20 postlobectomy patients with prolonged air leak (>5 days) to evaluate autologous blood pleurodesis compared to controls.They showed that there was a statistically significant (p<0.001)reduction in median time to chest tube removal (6.5 vs 12 days) and hospital discharge (8 vs 13.5 days) with autologous blood pleurodesis.One patient in the pleurodesis arm however developed an empyema.
As for the other forms of chemical pleurodesis, Liberman and associates (14) reported their experience with 41 patients who underwent chemical pleurodesis (30 talc, 7 doxycycline, 1 doxycycline+talc, 1 bleomycin, 1 bleomycin+talc) for prolonged air leak (>5 days) after undergoing lobectomy or bilobectomy.Sclerosis was successful in 40 patients (97.6%), with the remaining one patient having to undergo a pectoralis major flap for persistent air leak despite talc pleurodesis.Also, one patient developed empyema after talc pleurodesis.
As indicated above, complications of bedside pleurodesis include mainly consist of fever, pain and empyema.In addition, the most feared complication of talc pleurodesis is a systemic inflammatory response to talc that can result in acute respiratory distress syndrome (112,113) particularly if the talc particle size is small (114).However, it has previously been found to be not associated with increased mortality in a meta-analysis of patients with malignant pleural effusion undergoing talc pleurodesis (115).

Post-operative creation of pneumoperitoneum
As mentioned previously, the creation of a pneumoperitoneum has been described as both an intra-operative as well as a post-operative method of controlling prolonged air leaks.This involves the instillation of air into the peritoneal cavity via insertion of a peritoneal dialysis catheter under local anesthesia (81), or with the aid of a Veres needle under local anesthesia (82,83).The creation of a pneumoperitoneum is often combined with a form of pleural sclerosis, such as talc (81,82) or autologous blood (83), to aid the resolution of air leak.Several potential disadvantages of this technique include the risk of insertion of peritoneal dialysis catheter / Veres needle (eg bleeding, injury to intra-abdominal viscera) and possible respiratory compromise from the creation of the pneumoperitoneum.
Not many studies have been performed to evaluate this modality of therapy, except for a few isolated case reports, so the technique has shown promise in treatment of some patients but has not been evaluated on a large scale basis.Handy and associates (81) reported the successful use of this technique to resolve a persistent air leak of more than 3 weeks duration in a patient with emphysema who underwent lung volume reduction surgery.De Giacomo and colleagues(82) described the use of post-operative pneumoperitoneum to manage persistent air leak (>5 days) in 14 patients who underwent pulmonary resection for lung cancer, with resolution of the air leak occurring within 4-12 (mean 8) days after the procedure.The most recent paper assessing this technique by Korasidis et al. (83) demonstrated that combined post-operative pneumoperitoneum and autologous blood patch was able to control prolonged air leak (>3 days) present in 39 patients who underwent pulmonary resection for lung cancer within 144 hours of therapy.No major complications with the technique were reported by any of the above studies.

Optimal chest tube management and outpatient chest tube management
Appropriate chest tube management has also been shown to influence the duration of postoperative air leak.With respect to chest tube suction, it may be viewed in one of two ways.Firstly, chest tube suction may promote pleural apposition to decrease duration of air leaks, or alternatively, suction may cause tension on suture lines to prolong air leaks.The experience in lung volume reduction surgery had previously demonstrated that duration of prolonged air leak was decreased by avoiding routine chest tube suction in these patients (116).
This was subsequently investigated in several randomized prospective studies to see if this also held true in patients undergoing other forms of thoracic operations.For patients undergoing apical pleurectomy following primary spontaneous pneumothorax, Ayed demonstrated that converting to water seal (no suction) after a period of initial active suction significantly decreased the risk of prolonged air leak and duration of chest tube drainage compared to active suction throughout (117).A similar benefit of converting to water seal after a period of initial active suction for patients undergoing pulmonary resection (lobectomy, segmentectomy or wedge resections) was demonstrated by two separate groups (118,119).However, a comparable study by Brunelli and associates (120) showed that water seal had no advantage over active suction when limited to a population of patients undergoing lobectomy.A follow-up study demonstrated that in patients undergoing lobectomy, alternate suction (at night) and water seal (during the day) was better than water seal alone (121).
A different approach was evaluated by Alphonso and colleagues, who studied a mixed cohort of patients undergoing a variety of operations (VATS as well as open lobectomy, wedge resections, lung biopsies or pneumothorax operations) and found that adopting water seal immediately after surgery showed no difference in air leak duration compared to active suction (122).
Whether these approaches are applicable to patients with underlying emphysema undergoing pulmonary resection or pleurodesis has yet to be conclusively demonstrated, but a strategy of minimizing duration of chest tube suction or alternating it with water seal would be prudent based on evidence available so far.In addition, it should be noted that patients on water seal, particularly those with large air leaks, should be monitored for evidence of increasing subcutaneous emphysema or enlarging pneumothorax, as these patients will need to be placed back on active suction to prevent clinical deterioration (118).An alternative strategy to prolonged air leaks is the use of Heimlich valves or portable chest drainage systems to allow for early discharge of patients who are otherwise ready to be discharged from hospital apart from their prolonged air leak.Heimlich valves are one way valves originally used for the outpatient management of a pneumothorax, and two studies have shown that they can be successfully used to discharge select patients with prolonged air leak early with relatively few complications (123,124).Portable chest tube drainage systems have an additional advantage over Heimlich valves in that they are able to handle fluid drainage in addition to air leak and can also be connected to active suction when required (125).In conclusion, outpatient chest tube management appears to be an acceptable approach that is fairly safe for managing most patients with prolonged air leak if they are reliable enough to handle their Heimlich valve or portable chest tube system on their own at home.As to which patients with prolonged air leak are suitable for discharge without suction, Cerfolio and colleagues (126) reported that they successfully discharged 199 post-pulmonary resection patients with a suctionless portable device safely without complications as long as there was no development of a new or enlarging pneumothorax or subcutaneous emphysema after converting the original chest tube suction to water seal.More importantly, most of these patients had their air leak resolve by the end of 2 weeks of outpatient chest tube therapy, and for the remaining 57 who still had air leak, the chest tube was safely removed if these patients were asymptomatic, had no increase in pneumothorax or new subcutaneous emphysema on the outpatient device.There were no complications except for the development of empyema in 3 of these 57 patients (5.7%), but these 3 patients were immunocompromised and were on chronic steroid therapy.

Re-operation
If all else fails, in cases of persistent air leak that is refractory to methods described above, re-operation can be considered to look for the source of air leak and perform therapeutic maneuvers.Often this can be accomplished with video assisted thoracoscopy, such as described by Suter and associates (127), who managed to identify the source of air leak thoracoscopically in 3 patients who had prolonged air leak after pulmonary resection.The air leaks were subsequently sealed with direct application of fibrin glue or pleurodesis with silver nitrate.
However, pat i ents with mas sive , s evere p rolonged air leaks, particularly those with a concomitant large pleural space problem, usually require a more extensive operation such as a thoracoplasty or muscle flap transposition via an open thoracotomy.Thoracoplasty, the reduction of thoracic cavity by removal of ribs, is rarely done as it results in thoracic deformity, restriction in shoulder mobility and decreased respiratory function (7,128).As such, muscular flap transpositions have become the preferred technique, and we have developed the combined latissimus dorsi-serratus anterior transposition flap for this purpose.We have previously described 5 patients who underwent this technique (two COPD patients with pneumothorax refractory to conservative management, one COPD patient with prolonged air leak post lung volume reduction surgery, two patients with bronchopleural fistula/empyemas), with resolution of air leak that allows the chest tubes to be removed within 5 days after surgery and no recurrence of air leak noted at 1 year followup (129).At our institution, our indications for surgical air leak repair with flap reconstruction are (1) severe air leaks (high leak rate or continuous leak despite application of chest tube suction), (2) persistent air leak exceeding 4 weeks despite conservative management (or beyond 1 week for patients with underlying lung disease such as COPD), and (3) significant pleural dead space defined radiologically by absence of pleural-pleural contact despite maximal reexpansion efforts (16,36,130).
The operation is performed via a muscle sparing posterolateral thoracotomy with a lazy-S incision extending from the axilla to the lumbar region.Then, the latissimus dorsi and the serratus anterior muscle flaps are raised, with care taken to ensure that the serratus anterior www.intechopen.comEmphysema 124 flap is sufficient to cover the intended axillary window (usually by raising muscle slips from the 2 nd to 4 th ribs) but sparing the lower slips of muscle that insert into the scapula to avoid scapular winging.Creation of the axillary window involves resection of the second to fourth ribs centered over the mid-axillary line which allows good exposure of the underlying lung for surgical treatment (eg suture repair of parenchymal tears, decortication) and allows the latissimus dorsi to the passed through without compressing its vascular pedicle.The latissimus dorsi is loosely anchored over the lung and a chest tube is inserted after a final check for air leak.The axillary window is then closed with the serratus anterior muscle flap and the skin incision is closed over a subcutaneous drain.We believe our technique has several distinct advantages, as firstly it offers direct visualization for repair of diseased lung parenchyma via an open thoracotomy.Secondly, the latissimus dorsi flap provides a large, well vascularised surface for the lung to adhere to for healing.Moreover, the large mass of the muscle eliminates any pleural dead space and facilitates subsequent controlled re-expansion of the lung with time.Finally, the serratus anterior flap compartmentalizes the pleural cavity from the large subcutaneous space created by the latissimus dorsi harvest to prevent seroma formation or spread of infection between compartments.Minimal functional disability occurs after these muscle harvests, and scapular winging is prevented by sparing the lower slips of the serratus anterior muscle and the long thoracic nerve.This is in contrast to other methods for reducing pleural dead space which may only be sufficient to deal with a small volume of space (pleural tenting, pneumoperitoneum), or reduces the patient's functional lung reserve (phrenic nerve paralysis, thoracoplasty).
Other muscular flaps that have been described in contemporary literature to eliminate potential pleural spaces (though these have been traditionally ascribed for managing empyema spaces rather than persistent air leaks) include isolated pectoralis major (14,131), latissimus dorsi (131,132), serratus anterior (131), rectus abdominis (131,133) and the trapezius flaps (131,134).However, we have found in our own experience that these flaps either lack the reach or necessary bulk in order to properly treat the large pleural space problems that we have encountered.Thus, we feel that this combination muscle flap technique is an important and useful tool in the thoracic surgeon's armamentarium in dealing with recalcitrant post-operative air leaks in a variety of situations, particularly in patients with a background of impaired respiratory function such as severe emphysema.
Fig. 4. Two months after the initial operation, this patient has good recovery of shoulder function.

Summary
In summary, prolonged air leak is a common problem for patients with emphysema undergoing thoracic surgery that is associated with significant morbidity.Clinicians involved in the surgical care of this group of patients should be aware of the various factors which can further increase the risk of this complication occurring and need to know the various measures that should be employed to prevent this problem, as well as the treatment options available should prolonged air leak occur even if preventive measures are taken.
Based on the review of best available evidence as discussed previously, we propose a suggested algorithm for the management of prolonged air leaks in patients with emphysema with gradual progression of therapy similar to what has been proposed by others (2)(3)(4) but that also takes into account criteria for surgical intervention as we have mentioned earlier.

Fig. 2 .
Fig. 2. (a)The latissimus dorsi and proximal slips of the serratus anterior are raised as pedicled flaps via a lazy S incision from mid-axillary line to the inferior limit of the latissimus dorsi.An axillary window is then created by resecting the 2 nd and 3 rd ribs superior to the serratus anterior.

Fig. 2 .
Fig. 2. (b) Latissimus dorsi and serratus anterior reflected to demonstrate the axillary window.The latissimus dorsi flap is then passed though the axillary window and laid over the lung to obliterate the pleural space and seal the air leak.

Fig. 2 .
Fig. 2. (c) Serratus anterior flap is rotated anteriorly over the latissimus dorsi flap to close the axillary window.Primary closure of the incision was then performed.

Fig. 3 .
Fig. 3. Pre-operative chest x-ray (left) showing a large potential pleural space with resulting persistent air leak in this patient who had underwent bilateral lung volume reduction surgery, and post-operative chest x-ray (right) showing effective re-expansion of the right lung after the placement of the latissimus dorsi flap.

Fig. 5 .
Fig. 5. Proposed algorithm for the management of prolonged air leaks in patients with emphysema.

Table 1 .
Studies comparing the incidence of prolonged air leak in patients with primary versus secondary spontaneous pneumothorax undergoing surgical pleurodesis.

Table 2 .
Studies comparing the incidence of prolonged air leak in patients with COPD versus those without COPD undergoing pulmonary resection.

Table 3 .
Studies reporting the incidence of prolonged air leak in patients undergoing lung volume reduction surgery.

Table 4 .
Studies analyzing risk factors for prolonged air leak in patients undergoing pulmonary resection.

Table 5 .
Studies comparing the utility of buttressing staple lines in preventing prolonged air leak.

Table 7 .
table below summarizes the results of the randomized prospective studies performed to evaluate this technique.Studies comparing the utility of pleural tenting in preventing prolonged air leak.

Table 8 .
Studies reporting the efficacy of endobronchial valve placement in the treatment of prolonged air leak.

Table 9 .
Studies comparing the utility of chest tube management strategies in preventing prolonged air leak.

Table 10 .
Studies reporting the use of Heimlich valves or portable chest tube systems in the outpatient treatment of prolonged air leak.