Management of Pneumothorax in Lymphangioleiomyomatosis* Effects on Recurrence and Lung Transplantation Complications

Transcription

1 Original Research LYMPHANGIOLEIOMYOMATOSIS Management of Pneumothorax in Lymphangioleiomyomatosis* Effects on Recurrence and Lung Transplantation Complications Khalid F. Almoosa, MD; Jay H. Ryu, MD; Jose Mendez, MD; J. Terrill Huggins, MD; Lisa R. Young, MD; Eugene J. Sullivan, MD; Janet Maurer, MD; Francis X. McCormack, MD; and Steven A. Sahn, MD Study objectives: Pneumothorax is a common complication of lymphangioleiomyomatosis (LAM), and the optimal approach to its treatment and prevention is unknown. Chemical or surgical pleurodesis are often required to prevent recurrence. However, their efficacy in LAM is unclear, and whether they contribute to perioperative complications during lung transplantation is uncertain. Setting: The LAM Foundation database of registered patients. Design: A questionnaire was sent to all registered patients who had at least one pneumothorax to determine rates and patterns of recurrence and efficacy of interventions. A second questionnaire was sent to registered LAM patients who received a lung transplant. Patients or participants: Of 395 registered patients, 260 patients (66%) reported at least one pneumothorax during their lifetime, 193 of whom (74%) completed the questionnaire. Of the 85 lung transplant patients who were sent a separate questionnaire, 80 patients (94%) responded. Interventions: None. Measurements and results: Of the 193 respondents to the pneumothorax questionnaire, data on 676 episodes of pneumothorax were collected. Eighty-two percent (158 of 193 patients) had their first pneumothorax prior to a diagnosis of LAM. One hundred forty patients (73%) had at least one additional pneumothorax, either an ipsilateral recurrence (99 of 140 patients, 71%) or a contralateral pneumothorax (104 of 140 patients, 74%). Recurrence rates were 66% after conservative therapy, 27% after chemical pleurodesis, and 32% after surgery. In patients who had undergone lung transplantation, prior chemical or surgical pleurodesis was performed in 45 of 80 patients (56%). Fourteen of 80 patients (18%) reported pleural-related postoperative bleeding, 13 of whom (93%) had prior pleurodesis. Conclusions: Chemical pleurodesis or surgery are equally effective and better than conservative therapy in preventing recurrence of pneumothorax in LAM. Due to the high recurrence rate, either procedure should be considered for the initial pneumothorax in these patients. However, both contribute to increased perioperative bleeding following lung transplantation, with no effect on length of hospital stay. (CHEST 2006; 129: ) Key words: lung transplantation; lymphangioleiomyomatosis; pleurodesis; pneumothorax Abbreviations: BSP bilateral spontaneous pneumothorax; LAM lymphangioleiomyomatosis; SSP secondary spontaneous pneumothorax; TSC tuberous sclerosis complex Lymphangioleiomyomatosis (LAM) is a rare and frequently fatal lung disease that almost exclusively affects women, usually in their child-bearing years. LAM is associated with mutations in tuberous sclerosis complex (TSC) genes 1 and 2 and is characterized by abnormal proliferation of smoothmuscle like cells that infiltrate the pulmonary interstitium and obstruct airways, lymphatics, and blood vessels. 1 3 This process leads to thin-walled cyst formation in the pulmonary parenchyma and progressive decline in lung function. Patients commonly present with progressive dyspnea that can be con Original Research

2 fused with asthma or COPD. However, the development of recurrent pneumothorax or chylothorax, hallmarks of LAM that distinguish it from more common obstructive lung diseases, often leads to the correct diagnosis. 1 3 These complications contribute substantially to the morbidity associated with LAM. In LAM, the lifetime incidence of pneumothorax ranges from 39 to 81% Kitaichi and coworkers 7 reported pneumothorax in 18 of 46 LAM patients (39%). The rate of pneumothorax was 60% in a series of 49 patients from the United Kingdom 6 ; the ipsilateral recurrence rate was 67% in pneumothoraces managed conservatively, compared to 18% when treated surgically. Urban and colleagues 10 reported that 47 of 69 patients (68%) had at least one pneumothorax during the course of their illness; most (52%) occurred before a diagnosis of LAM was established. Chu and coworkers 4 reported pneumothorax in 24 of 35 LAM patients (69%), 19 of whom (79%) had undergone either pleurodesis or pleurectomy. Pleural interventions such as pleurodesis and pleurectomy are often employed to manage recurrent pneumothorax in patients with LAM. Furthermore, many patients eventually require lung transplantation, and survival in these patients is similar to that of other chronic lung disease populations undergoing transplantation but is associated with a higher rate of perioperative bleeding. 12,13 However, there is no consensus regarding optimal pneumothorax management, and the effects of pleural symphysis on the success and morbidity of lung transplantation are incompletely understood. The purpose of this study was to accrue and analyze data on the management of pneumothorax from a large population of LAM patients, and use this data to evaluate the efficacy of pleural interventions in preventing recurrence as well as its adverse effects on lung transplantation. *From the Division of Pulmonary and Critical Care Medicine (Drs. Almoosa, Young, and McCormack), University of Cincinnati, Cincinnati, OH; Division of Pulmonary and Critical Care Medicine (Drs. Ryu and Mendez), Mayo Clinic College of Medicine, Rochester, MN; Division of Pulmonary and Critical Care Medicine (Drs. Huggins and Sahn), Medical University of South Carolina, Charleston, SC; US Food and Drug Administration (Dr. Sullivan), Rockville, MD; and Cigna Health Care (Dr. Maurer), West Granby, CT. The views presented in this article do not necessarily reflect those of the US Food and Drug Administration. Financial support was provided by the LAM Foundation. Manuscript received June 6, 2005; revision accepted November 15, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Khalid Almoosa, MD, Division of Pulmonary and Critical Care, University of Cincinnati, 231 Albert Sabin Way, 6053 MSB, Cincinnati, OH ; khalid. almoosa@uc.edu Methods and Materials Data on pleural complications of LAM were collected using the LAM Patient Database developed by the LAM Foundation (Cincinnati, OH). Registration with the LAM Foundation is open to any LAM patient worldwide. On registration, each new patient is sent a demographic and clinical data questionnaire (Enrollment Questionnaire) with a consent form. The LAM Pleural Disease Consensus Group was commissioned by the LAM Foundation to address the management of pneumothorax in LAM. The LAM Pleural Disease Consensus Group met three times from November 2002 to September 2003 to analyze the questionnaire-derived data. Patients who answered yes to the question Have you ever had a collapsed lung? on the enrollment questionnaire were sent a secondary questionnaire by mail (Pneumothorax Questionnaire; online supplement Appendix 1) that focused on radiographically confirmed pneumothorax. This questionnaire specifically inquired into the number of pneumothoraces and their time frames, associated symptoms and activity, hospitalizations, and management. Prior to distributing the survey, 15 telephone interviews using test questionnaires were completed to reduce ambiguity and assess the patients understanding of the questions, as no formal cognitive testing was done. Amendments to the questionnaire based on the responses were made accordingly. Responses were collected on patients who had signed the consent, and the data were entered into a database and analyzed. Nonresponders were sent the questionnaire a second time. The study protocol and consent form were approved by the Institutional Review Board of the University of Cincinnati. Due to potential errors associated with recalling details of past episodes of pneumothorax, some questions were written so that answers were grouped into broad categories. For example, the question How long was a chest tube in place? offered the options: 3 days, 3 to 7 days, 7 to 14 days, and 14 days. Similarly, the question When did you have your pneumothorax? requested a year and a season, rather than a specific month or day. Furthermore, because retrospective distinction between a persistent and recurrent pneumothorax can be problematic, all pneumothorax reported within the same season of a given year were considered as single pneumothorax. If a pneumothorax was treated with more than one intervention, the most aggressive intervention was listed as the treatment of record for that event. A separate questionnaire was sent by mail to registered patients who underwent lung transplantation (Transplant Questionnaire; online supplement Appendix 2). Data on pleural complications of surgery and previous pleurodesis procedures were collected and analyzed. To examine the reproducibility of the responses, a duplicate questionnaire was sent to a random sample of 20 patients, and the initial and repeat responses were compared. Although formal validity measures were not performed, the survey did appear to have face validity. Statistical software (version 9.1.3; SAS Institute; Cary, NC) was used to compare data using the 2 test, Student t test, sign test, Wilcoxon signed-rank test, and logistic regression model; p 0.05 was considered statistically significant. Results Initial Pneumothorax Of 395 patients who completed the LAM Foundation enrollment questionnaire, 260 patients reported at least one spontaneous pneumothorax during their lifetime (incidence, 66%), and 200 of 260 patients (77%) indicated that they had had subsequent pneumothora- CHEST / 129 / 5/ MAY,

3 Table 1 Clinical Presentation, Interventions, and Hospital Days for Pneumothorax in LAM* Variables Data Total respondents 193 Total episodes of pneumothorax (confirmed by chest 676 radiograph) Patients with pneumothorax as presenting symptom of 158 (82) LAM, No. (%) Average number of pneumothoraces per patient 3.5 Average number of pneumothoraces prior to diagnosis 2.2 of LAM Symptoms associated with pneumothorax, % of patients Dyspnea 69 Chest pain 59 Felt pop 14 Activity during pneumothoraces episode, % of patients Rest or minimal activity 81 Active 10 Unknown 9 Total interventions (excluding observation) 962 Interventions per pneumothorax 1.4 Interventions per patient 5.0 Hospitalization Total days 5,553 Average days per pneumothorax 8 Average days per patient for all pneumothoraces 29 *Data are presented as No. unless otherwise indicated. Intervention: either aspiration, tube thoracostomy, chemical pleurodesis, or surgery. ces. All patients were women, and the average age of the respondents was 45 years. The pneumothorax questionnaire was sent to the 260 patients with at least one pneumothorax, and 193 patients (74%) responded (Table 1). In the majority (136 of 193 patients, 71%), the first pneumothorax occurred in the third or fourth decade; the initial pneumothorax was less common in the second, fifth, or sixth decades (Fig 1). The mean age at the time of the first pneumothorax was 34.9 years (range, 18 to 63 years). Those with only one pneumothorax during their lifetime (53 of 193 patients, 27%) had a mean age of 37.8 years at the time of the pneumothorax; those with recurrence (140 of 193 patients, 73%) had a mean age of 33.8 years at the time of the initial pneumothorax (p 0.20). In the 140 patients who had recurrent pneumothoraces, there were a total of 623 pneumothoraces (4.5 pneumothoraces per patient). Eighty-two percent (158 of 193 patients) had their first pneumothorax before the diagnosis of LAM was made (Fig 1). The initial pneumothorax was equally likely to occur on either side (right side, 95 patients; left side, 90 patients; p 0.26), and simultaneous bilateral pneumothoraces occurred in 8 of 193 patients (4%) as the initial manifestation of LAM. Twenty-two of 193 LAM patients (11%) had a diagnosis of TSC. Seventy-one pneumothoraces occurred in this group, an average of 3.2 pneumothoraces per patient. More than 90% of the TSC patients reported that their first pneumothorax occurred before a diagnosis of LAM was established. Most patients (179 of 193 patients, 93%) had symptoms associated with the onset of pneumothorax, with dyspnea (69%) and chest pain (59%) being most common (Table 1). Most patients reported that episodes (545 of 676 episodes, 81%) occurred at rest or with minimal activity. The majority of patients were hospitalized during these episodes, with an average of 8 days of hospitalization per episode. Recurrence and Contralateral Pneumothoraces Recurrences and contralateral pneumothoraces were common (Fig 2). Only 53 of 193 respondents (27%) had a single pneumothorax during their lifetime. Figure 3 demonstrates the breakdown of recurrent and contralateral pneumothoraces in the group with more than one pneumothorax. For our pneumothorax questionnaire analyses, we chose to focus only on the initial pneumothorax, contralateral pneumothorax, and the first recurrence on each side. Twenty-eight percent (54 of 193 patients) had more than the maximum of four pneumothoraces per patient that were characterized in this analysis. Fifteen of 193 respondents (8%) reported 21 episodes of bilateral simultaneous pneumothorax (BSP). BSP occurred once in 10 patients, twice in 4 patients, and three times in 1 patient. The average age of onset of the first pneumothorax in patients who had BSP was 37.5 years, nearly 3 years older than the mean for the rest of the pneumothorax group, although the difference did not reach statistical significance. Management of Pneumothorax The clinical management of pneumothorax in this cohort was examined. Of the 676 episodes of pneumothorax reported, 117 episodes (17%) were observed without intervention. A total of 962 interventions were performed for the remaining 559 episodes of pneumothorax: simple aspiration 39 episodes, small-caliber chest tube with Heimlich valve in 42 episodes, tube thoracostomy in 481 episodes, chemical pleurodesis in 209 episodes, and surgery in 191 episodes. The average number of interventions per pneumothorax was 1.4. Of those who required a tube thoracostomy, 269 of 481 patients (56%) had a chest tube for 7 days. Talc was the most common sclerosing agent used for 34% (71 of 209 patients) who had chemical pleurodesis. Mechanical pleurodesis was the most frequent (81 episodes) surgical intervention, followed by bullectomy (56 episodes) and pleurectomy (54 episodes). An analysis was performed to assess the success of interventions for pneumothorax (Fig 4). Only ipsilateral pneumothoraces were included. Of 301 initial pneumo Original Research

4 Figure 1. Age at initial pneumothorax compared to age at diagnosis of LAM. The initial thorax developed in most women between the ages of 20 and 49 years, usually before the diagnosis of LAM was established. thoraces, the majority (154 pneumothoraces, 51%) were treated conservatively (with either simple aspiration or tube thoracostomy but without chemical or surgical pleurodesis), while the rest were managed surgically or with chemical pleurodesis. Recurrence was common in the conservative treatment group (66%) but was reduced by more than half by chemical (27%) or surgical (32%) pleurodesis (p 0.01 for both compared to conservative treatment). Of the 147 recurrent pneumothoraces, the majority (56%) were treated more aggressively with chemical pleurodesis or surgery. The reproducibility of the self-reported data were assessed by sending a duplicate questionnaire to a subset of patients several months after completion of the first questionnaire. Comparison of the data showed that the vast majority of responses were identical or similar (see online supplement Appendix 3). Effects of Treatment of Pneumothorax on Subsequent Lung Transplantation Of the 395 LAM Foundation-registered patients who completed the enrollment questionnaire, 85 patients (22%) reported having undergone lung Figure 2. Recurrence rate of pneumothorax (PTX) in LAM. The majority of LAM patients had multiple pneumothoraces during their lifetime, averaging 4.5 pneumothoraces for those who had recurrence. In some cases, 10 pneumothoraces occurred. CHEST / 129 / 5/ MAY,

5 Figure 3. Incidence and recurrence of ipsilateral and contralateral pneumothorax in LAM. The LAM patient enrollment questionnaire identified 260 of 395 patients with a history of pneumothorax. These patients were sent a second (pneumothorax) questionnaire. Most registered LAM patients had at least one pneumothorax, with the majority of recurrences developing within 2 years (yrs). See Figure 2 legend for expansion of abbreviation. transplantation. The 85 patients were sent a second transplant/pleural disease questionnaire (Transplant Questionnaire); 80 responses were received for 81 transplants. Details for these patients are shown in Table 2. Three patients had died, all 2 years after transplantation. The vital status of the five nonre- Figure 4. Effectiveness of the treatment of pneumothorax in patients with LAM. Each hemithorax was considered separately. The highest recurrence rates of pneumothorax occurred with conservative therapy, but surgical and chemical pleurodesis procedures also frequently failed. See Figure 2 legend for expansion of abbreviation Original Research

6 Table 2 Lung Transplantation in LAM* Variables Data Respondents 80 Lung transplants 81 Left lung transplant 13 (16) Right lung transplant 25 (31) Bilateral lung transplant 43 (53) Patients with prior pleural procedures 45 (56) Left side only 5 (11) Right side only 6 (13) Bilateral 34 (76) Pleural procedures performed 1 yr before surgery 37 (82) Patients with pleural-related complications 14 (18) Prior pleural procedures 13 (93) Required surgical therapy for complication 7 (50) *Data are presented as No. or No. (%). One patient had a retransplant. sponders is unknown. In 45 of 80 transplant patients (56%), pleural procedures (chemical or surgical pleurodesis) were performed before transplantation for either a pneumothorax or a chylothorax. In 12 of 45 patients (27%), the side of the pleural procedure dictated the side that received the transplant. Fourteen of 80 patients (18%) reported pleural-related bleeding complications perioperatively; all but 1 patient (13 of 14 patients, 93%) had undergone previous pleural procedures (Fig 5; p 0.01). Of the 45 patients who had prior pleural procedures, the risk of perioperative bleeding was 29% (13 of 45 patients), compared to 3% in patients with no prior pleurodesis (p 0.01). Of the 13 pleural-related bleeding complications that occurred in patients with prior pleural procedures, 9 complications (69%) occurred in patients with bilateral lung transplants and 8 of these complications (89%) occurred in patients with bilateral pleural procedures. One bleeding event occurred in a single, left-lung transplant patient who had bilateral pleural procedures, and three events occurred in patients with right lung transplants, two of whom had a bilateral procedure and one of whom had a left pleural procedure. One half (7 of 14 complications) required a return to surgery for bleeding, most (6 of 7 complications; 86%) in women with bilateral pleural procedures. Also, of the 34 patients who had undergone bilateral pleurodesis, 19 patients (56%) received a bilateral lung transplant. The average length of stay was days (mean SEM) for the group who underwent prior pleurodesis, compared to days for the pleurodesis-naïve group (Fig 5; p 0.4). centimeters in diameter, often with striking involvement of the pleural surface. 3,5 Pneumothorax can occur from either direct cyst rupture into the pleural space or indirectly through alveolar wall disruption, followed by leak of air into the lung interstitium, mediastinum, and pleural space. Pneumothorax will develop in most patients with LAM during the course of their illness, and most will require a pleurodesis procedure. Physicians who care for patients with LAM are frequently confronted with problematic pleural management issues. Current practice has generally been to postpone definitive preventative measures because pleural interventions are thought to complicate lung transplantation. 12,13 Our data indicate that LAM patients with an initial pneumothorax are typically destined to have two or more additional pneumothoraces. Our conclusion is that conservative approaches (not involving chemical or surgical pleurodesis) to the initial pneumothorax in LAM seldom prevent recurrence and lead to considerable morbidity, strongly suggesting that pleural symphysis should be the goal following the initial event. Secondary spontaneous pneumothorax (SSP) occurs in a variety of pulmonary diseases and is particularly common in COPD, cystic fibrosis, and Pneumocystis carinii pneumonia Our data indicate that the incidence of SSP in LAM is greater than in all other chronic pulmonary disorders. In fact, pneumothorax is often the sentinel event that leads to the diagnosis of LAM. 2,20 The 66% prevalence and 73% recurrence rates for pneumothorax in LAM are consistent with data from previous smaller studies. 5 8,10 The initial pneumothorax is most likely to occur in the fourth decade, followed by the third and second decades. The majority of episodes of pneumothorax are symptomatic, with dyspnea and chest pain being the most Discussion The gross appearance of the lung in LAM reveals numerous cysts ranging from a few millimeters to Figure 5. Morbidity of prior pleural procedures for lung transplantation. Lung transplantation patients with prior pleural procedures had increased bleeding complications but no difference in hospital stay or perioperative mortality. Comparisons are prior pleural procedure vs no prior procedure. CHEST / 129 / 5/ MAY,

7 common complaints and most requiring hospitalization. Hospitalization is costly, both in use of resources for medical care and loss of productivity; and early definitive therapy has the potential to reduce recurrences and optimize health-care utilization. In addition, appropriate management reduces the risk of associated life-threatening complications, such as tension pneumothorax and respiratory failure. The majority of pneumothoraces occurred during rest or minimal activity. There was no side predilection. BSP occurred in 8% of patients and was the presenting feature of LAM in one half of those individuals. This life-threatening manifestation of LAM should immediately suggest the diagnosis in a young woman, as very few other lung disorders commonly present with or are associated with BSP. The ratio of patients with TSC LAM to sporadic LAM in this cohort was low (1:8) but quite similar to the ratio of patients with TSC LAM to sporadic LAM in the LAM Foundation database (1:7). The prevalence of pneumothorax in the groups with TSC LAM and sporadic LAM was nearly identical. Pneumothorax recurrence in LAM is very common; approximately 73% of patients will have another event. Approximately one half the time, the recurrence is ipsilateral. The interval between the initial pneumothorax and ipsilateral recurrence was 1.8 years, 50% shorter than the interval between the initial pneumothorax and contralateral pneumothorax. The second ipsilateral recurrence interval was 1.3 years. Collectively, these data indicate that the odds that a LAM patient will not have a pneumothorax are only 30%. Once a pneumothorax occurs, the chances that the patient will have a subsequent pneumothorax on either side are high. Treatment A primary goal of this study was to examine the efficacy of current therapies for pneumothorax in LAM. Due to the rarity of LAM, most physicians have limited, if any, experience in treating pneumothorax in LAM. Furthermore, as our results indicate, 80% of patients have their first pneumothorax before the diagnosis of LAM is established and have two or more pneumothoraces before the diagnosis is confirmed. This is true even in TSC patients, in whom the clinical context should certainly suggest LAM in any affected female with pneumothorax. Given that pneumothorax precedes the diagnosis in most cases, it is not surprising that the initial pneumothorax is typically treated conservatively (Fig 4). However, this approach is usually ineffective, resulting in a 66% recurrence rate. With recurrence, surgical intervention was the preferred approach, employed in 127 of 193 patients (66%), compared to 96 of 193 patients (50%) who underwent chemical pleurodesis through a chest tube (data not shown). The efficacy of these two treatments was similar: 27% recurrence rate for chemical pleurodesis and 32% recurrence rate for surgery. This result was surprising since surgical interventions are generally associated with a lower recurrence rate of SSP (approximately 5%) than chemical pleurodesis (approximately 15%). 21,22 One report 11 indicated that recurrence after surgical pleurodesis was 0% following an initial pneumothorax in patients with Langerhans cell histiocytosis, a disease that is commonly considered in the differential of LAM. It is unclear why the recurrence rate after surgical treatment is higher in LAM than is observed in other diseases. One possibility is that the remarkable profusion of blebs on the surface of the LAM lung may limit apposition and fusion of the visceral and parietal pleural surfaces following abrasion of the lung surface or talc instillation. Another possibility may be that the success of surgical pleurodesis in LAM is critically dependent on the experience of the operator (eg, performance by thoracic surgeon rather than a general surgeon). Exploring the mechanism of pleurodesis failure in LAM is clearly an important future research direction. The average lifetime hospital burden for each patient in the study was approximately 1 month. With average daily hospital charges of $2,500, the total pneumothorax-related hospital expenditure for each patient was estimated to be $75,000, and for all patients in the study was approximately $14 million. Effect of Pneumothorax Therapy on Lung Transplantation A prominent concern among LAM patients considered for lung transplantation is that pleural interventions will increase the risk of perioperative adverse events, including excessive bleeding and death. Thirtyone percent (14 of 45 patients) with prior pleurodesis had perioperative bleeding, and 50% (7 of 14 patients) of those required a return to the operating room for control of hemorrhage. Of the 45 patients who had prior pleural procedures, 34 patients (76%) had had bilateral pleural procedures and received two lungs. Only 3% (1 of 35 patients) with pleurodesis-naïve hemithoraces had bleeding complications at transplantation. Prior pleural manipulation did not significantly increase the length of hospital stay and did not result in perioperative death. We acknowledge that this study may underestimate the impact of pleural procedures on the mortality associated with transplantation, since only survivors were polled. Furthermore, it is unknown how many patients, if any, were rejected for lung transplantation based on a history of pleural procedures. Nevertheless, our data suggest that prior pleural interventions for recurrent pneumothorax does not 1280 Original Research

8 preclude successful lung transplantation and that, in general, bleeding complications are manageable. Limitations of this Study This survey was designed to assess the incidence, recurrence, and management of pneumothoraces in a population at risk. Several limitations are inherent in retrospective, observational studies such as this, especially the reliance on patient recall. Furthermore, limited patient understanding of specific clinical features and treatment options may have led to inaccurate reporting. However, although standard validity measures were not performed, our validation questionnaires indicated that the responses were acceptably consistent. In addition, our perception, as well as that of the LAM Foundation staff, was that this cohort was well informed and very knowledgeable about their disease. Another possible limitation is that some of the questions on the questionnaire were not mutually exclusive on retrospective reevaluation. However, reanalysis of the data indicates that this limitation did not significantly affect the results. There was also no formal cognitive testing performed. Instead, 15 test cases were completed on the telephone to refine the questions and reduce ambiguity. The limitations of the transplant questionnaire were similar to the pneumothorax questionnaire. Patients were asked to recall perioperative events and complications that may be inaccurate and biased according to the patients knowledge and understanding of the situation. Also, the incidence of pneumothorax may be underestimated because some of the patients who never had a pneumothorax at the time of the survey may have subsequently had one. A follow-up questionnaire for that particular group would be useful to reduce this bias. Finally, due to the rarity of LAM, the interventions employed may reflect different practice patterns among physicians of different specialties worldwide. This suggests that the level of expertise and experience of the operator with pleural procedures may play a critical role in the effectiveness of the interventions employed. Conclusion Pneumothorax is a common complication of LAM and often recurs prior to establishing a definitive diagnosis of LAM. Due to the morbidity and cost associated with multiple recurrences, we recommend early, definitive intervention, preferably at the time of the initial pneumothorax. Although pleurodesis may be associated with an increased risk of perioperative bleeding with lung transplantation, our data suggest that complications are manageable and do not preclude successful transplantation. ACKNOWLEDGMENT: We are grateful to Sue Byrnes, Meg Coutinho, and Nancy Reynolds, of the LAM Foundation for their assistance and support in the data collection and preparation of this article. References 1 Kalassian KG, Doyle R, Kao P, et al. Lymphangioleiomyomatosis: new insights. Am J Respir Crit Care Med 1997; 155: Sullivan EJ. Lymphangioleiomyomatosis: a review. Chest 1998; 114: Tattersfield AE, Harrison TW. Step 3 of the asthma guidelines. Thorax 1999; 54: Chu SC, Horiba K, Usuki J, et al. Comprehensive evaluation of 35 patients with lymphangioleiomyomatosis. Chest 1999; 115: Corrin B, Liebow AA, Friedman PJ. Pulmonary lymphangiomyomatosis: a review. Am J Pathol 1975; 79: Johnson SR, Tattersfield AE. Clinical experience of lymphangioleiomyomatosis in the UK. Thorax 2000; 55: Kitaichi M, Nishimura K, Itoh H, et al. Pulmonary lymphangioleiomyomatosis: a report of 46 patients including a clinicopathologic study of prognostic factors. Am J Respir Crit Care Med 1995; 151: Oh YM, Mo EK, Jang SH, et al. Pulmonary lymphangioleiomyomatosis in Korea. Thorax 1999; 54: Taylor JR, Ryu J, Colby TV, et al. Lymphangioleiomyomatosis: clinical course in 32 patients. N Engl J Med 1990; 323: Urban T, Lazor R, Lacronique J, et al. Pulmonary lymphangioleiomyomatosis: a study of 69 patients; Groupe d Etudes et de Recherche sur les Maladies Orphelines Pulmonaires (GER- M O P). Medicine (Baltimore) 1999; 78: Mendez JL, Nadrous HF, Vassallo R, et al. Pneumothorax in pulmonary Langerhans cell histiocytosis. Chest 2004; 125: Boehler A, Speich R, Russi EW, et al. Lung transplantation for lymphangioleiomyomatosis. N Engl J Med 1996; 335: Pechet TT, Meyers BF, Guthrie TJ, et al. Lung transplantation for lymphangioleiomyomatosis. J Heart Lung Transplant 2004; 23: Flume PA. Pneumothorax in cystic fibrosis. Chest 2003; 123: Mitchell-Heggs PF. Spontaneous pneumothorax in cystic fibrosis [letter]. Thorax 1970; 25: Penketh AR, Knight RK, Hodson ME, et al. Management of pneumothorax in adults with cystic fibrosis. Thorax 1982; 37: Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000; 342: Sassoon CS. The etiology and treatment of spontaneous pneumothorax. Curr Opin Pulm Med 1995; 1: Yoshida S, Yamagishi T, Fukutake K, et al. Thoracotomy as treatment for pneumothorax associated with Pneumocystic carinii pneumonia in a patient with hemophilia A and the acquired immunodeficiency syndrome [in Japanese]. Nihon Kyobu Shikkan Gakkai Zasshi 1995; 33: Glassberg MK. Lymphangioleiomyomatosis. Clin Chest Med 2004; 25: Hatz RA, Kaps MF, Meimarakis G, et al. Long-term results after video-assisted thoracoscopic surgery for first-time and recurrent spontaneous pneumothorax. Ann Thorac Surg 2000; 70: Massard G, Thomas P, Wihlm JM. Minimally invasive management for first and recurrent pneumothorax. Ann Thorac Surg 1998; 66: CHEST / 129 / 5/ MAY,