Image Characteristics As Predictors for Thoracoscopic Anatomic Lung

Home , Sputum culture

Image Characteristics as Predictors for Thoracoscopic Anatomic Lung Resection in Patients With Pulmonary Tuberculosis Yi-Ting Yen, MD, Ming-Ho Wu, MD, Lili Cheng, MD, Yi-Sheng Liu, MD, Sheng-Hsiang Lin, PhD, Jung-Der Wang, MD, PhD, and Yau-Lin Tseng, MD, PhD Division of Thoracic Surgery, Department of Surgery, and Department of Diagnostic Radiology, National Cheng Kung University Hospital; Institute of Clinical Medicine, and Department of Public Health, College of Medicine, National Cheng Kung University, EEA THORACIC GENERAL Tainan; Division of Thoracic Surgery, Department of Surgery, E-Da Hospital, Kaohsiung, Taiwan

Background. A variety of complications occur in pa- Results. Fifty patients were enrolled; 21 given VATS tients with pulmonary tuberculosis. The feasibility of a and 29 given a thoracotomy. The VATS group had thoracoscopic approach to anatomic lung resection for significantly lower gradings in pleural thickening, peri- the complications of mycobacterial infection has not bronchial lymph node calcification, tuberculoma, cavity, ؍ been well evaluated. and aspergilloma than did the thoracotomy group (p Methods. We retrospectively analyzed chest computed 0.000, 0.015, 0.001, 0.023, and 0.022, respectively). Mean tomography (CT) scans of patients who underwent ana- operative time, blood loss, and complication rate were tomic lung resections without additional procedures for not significantly different, but the mean hospital stay was ؍ tuberculosis between January 2007 and September 2009. significantly shorter (10.00 days versus 14.96 days, p Image characteristics on chest CT scans were classified as 0.048) in the VATS group. bullae, pleural thickening, peribronchial lymph node Conclusions. Multiple cavities, multiple aspergillomas, calcification, tuberculoma, cavity, aspergilloma, atelecta- multilobar tuberculoma, extensive pleural thickening, sis, and bronchiectasis, and graded according to the and peribronchial lymph node calcification preclude number of the lesions and degree of lobar involvement. VATS. It is reasonable to attempt a thoracoscopic ap- Patients were divided into two groups, video-assisted proach in patients without these preoperative image thoracoscopic surgery (VATS) and thoracotomy for ana- characteristics. tomic lung surgery, according to the eventual operative procedure. The variables between these two groups were compared using the Student t test; the image character- (Ann Thorac Surg 2011;92:290–6) istics were compared using a ␹2 test. © 2011 by The Society of Thoracic Surgeons

espite improvements in medication, a variety of with the thoracoscopic approach. A thoracotomy was D complications occur in patients with pulmonary done if thoracoscopic adhesiolysis or a hilar dissection tuberculosis (TB) [1], including cavitary lesions, aspergil- was intraoperatively evaluated as difﬁcult. loma, segmental or lobar atelectasis, and bronchiectasis. We hypothesize that the image characteristics on a Some other complications, such as bullous formation, chest computed tomography (CT) scan affect the feasibil- tuberculoma, and pleural thickening, are also caused by ity of thoracoscopic anatomic lung resection. We there- infection. Surgical resection is pivotal for treating pulmo- fore retrospectively analyzed the CT scans of all the nary TB; eg, managing hemoptysis and salvaging pa- patients who underwent anatomic lung resections for tients with limited responsiveness to chemotherapy [2– pulmonary TB, and evaluated whether they had unfavor- 6]. However, the feasibility of a thoracoscopic approach able lesions that precluded thoracoscopic anatomic lung to anatomic lung resection for the complications of my- resections. cobacterial infection has not been well evaluated [7]. Since January 2007, VATS has become the preferred method for anatomic lung resection in our institution, not Material and Methods only for patients with malignancies, but also for those with complications of pulmonary TB. All the anatomic Patient Enrollment lung resections for pulmonary TB are initially attempted We retrospectively reviewed the records of 65 patients with complications of pulmonary TB who underwent Accepted for publication Feb 11, 2011. surgical intervention between January 2007 and Septem- ber 2009. All the surgeries were done by Dr Tseng or Dr Address correspondence to Dr Tseng, Division of Thoracic Surgery, Department of Surgery, National Cheng Kung University Hospital, 138 Wu, both experienced in TB lung surgery [8], and all were Sheng-Li Rd, Tainan 704, Taiwan; e-mail: [email protected]. assisted by Dr Yen. Patients who underwent anatomic

© 2011 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2011.02.039 Ann Thorac Surg YEN ET AL 291 2011;92:290–6 THORACOSCOPIC LUNG RESECTION FOR PULMONARY TB

Table 1. Deﬁnition and Grading of Image Characteristics on Chest Computed Tomographic Scan in Patients Who Underwent Anatomic Lung Resection for Pulmonary Tuberculosis

Image Characteristics Grading Deﬁnition

Bulla 0 No such lesion 1 Thin-walled air-containing space noted in only one lobe on the operated side 2 Thin-walled air-containing space noted in more than one lobe on the operated side Pleural thickening 0 No pleural thickening 1 Pleural thickening in less than four 10-mm-thick sections or involving less than half the circumference on the operated side 2 Pleural thickening in at least four 10-mm-thick sections, involving more than half the circumference on the operated side Peribronchial lymph node 0 No such lesion calcification 1 Single area of peribronchial lymph node calcification on the operated side 2 Multiple areas of peribronchial lymph node calcification on the operated side Tuberculoma 0 No such lesion 1 Well-defined nodule or nodule with calcification under TB context in only one lobe on the operated side 2 Well-defined nodule or nodule with calcification under TB context in more than one lobe on the operated side GENERAL THORACIC Cavity 0 No such lesion 1 Single thick-walled air-containing space on the operated side 2 Multiple thick-walled air-containing spaces on the operated side Aspergilloma 0 No such lesion 1 Single lung lesion with “ball-in-hole” appearance noted on the operated side 2 Multiple lung lesions with “ball-in-hole” appearance noted on the operated side Atelectasis 0 No such lesion 1 Decrease of lung volume in only one lobe on the operated side 2 Decrease of lung volume in more than one lobe on the operated side Bronchiectasis 0 No such lesion 1 Dilated bronchial tree in only one lobe on the operated side 2 Dilated bronchial tree in more than one lobe on the operated side lung resections were enrolled, including pneumonec- node calcification, tuberculoma, cavity, aspergilloma, at- tomy, bilobectomy, lobectomy with or without a wedge electasis, and bronchiectasis, with the radiologic criteria resection, and segmentectomy with or without a wedge listed in Table 1. Pleural thickness was measured from resection. Patients who underwent a wedge resection the mediastinal window images. All the image character- only, bronchoscopy for blood clot evacuation, concomi- istics were graded according to the number of lesions and tant decortications or thoracoplasty, or bilateral surgeries degree of lobar involvement (Table 1; Figs 1 and 2). were excluded. Patients newly diagnosed with pulmonary TB or operated on because of treatment failure were Statistical Analysis given antimycobacterial medications for 6 additional Patients were also divided into two groups; those who months; those who had been operated on because of were successfully treated with thoracoscopic anatomic multidrug-resistant TB (MDRTB) were given antimyco- lung surgeries (VATS) and those who were converted to bacterial medications for at least 24 months after the thoracotomy for anatomic lung surgeries. The age, oper- sputum culture had become negative for TB. Informed ative time, blood loss, and hospital stay between these consent was waived because the study was retrospective, two groups were compared with the Student t test, while and the review of medical records was approved by the the image characteristics were compared using a ␹2 test. Institutional Review Board of National Cheng Kung Uni- versity Hospital (ER-99-079). Results Classification and Grading of Image Characteristics Fifty patients were finally included. Twenty-one of them The preoperative chest CT scans were reviewed by two were treated with VATS, and 29 of them were converted board-certified radiologists specialized in thoracic im- to thoracotomy. The patients’ age, gender, surgical indi- ages, who were blinded to the patients’ identities and cations, procedures performed, operative time, blood ultimate operative procedures. Abnormal findings iden- loss, hospital stay, and complications are listed in Table 2. tified on the chest CT scans were interpreted and classi- The percentage of male patients was significantly higher fied as bulla, pleural thickening, peribronchial lymph in the thoracotomy group. The mean operative time and 292 YEN ET AL Ann Thorac Surg THORACOSCOPIC LUNG RESECTION FOR PULMONARY TB 2011;92:290–6

Fig 1. Image characteristics that preclude thoracoscopic anatomic lung resection, showing (A) grade 2 pleural thickening (arrow) and grade 2 aspergilloma (arrowhead), and (B) grade 2 peribronchial lymph node calci- ﬁcation (arrow). EEA THORACIC GENERAL

amount of blood loss of the VATS group were not group. The bleeding from the bronchial artery in the significantly different from those of the thoracotomy VATS group occurred postoperatively, which was con- group. The mean hospital stay of the VATS group, trolled with the thoracoscopic approach. The bleeding in however, was significantly shorter than that of the tho- the thoracotomy group also occurred postoperatively and racotomy group. There were significant differences in the was managed with thoracotomy. All 3 patients with gradings of image characteristics between the VATS and bronchopleural fistula were eventually successfully thoracotomy groups in terms of pleural thickening, peri- treated with thoracoplasty and suture of the broncho- bronchial lymph node calcification, tuberculoma, cavity, pleural fistula. One patient in the thoracotomy group and aspergilloma (Table 3). The gradings of other image with a persistent air leak was found to have an infection characteristics, such as atelectasis, bronchiectasis, and in the residual pleural space 6 months later. Decortica- bulla, were not significantly different between the VATS tion and thoracoplasty were performed to eliminate the and thoracotomy groups. infection and obliterate the residual space. Five of the 21 patients in the VATS group and 15 of the 29 in the thoracotomy group had a positive preoperative sputum culture. One patient in the VATS group and 2 in the thoracotomy group had positive sputum cultures Comment during the immediate postoperative period (day 7). All It has been strongly suggested in several studies [9–12] the patients were followed up for 7 to 33 months (mean, that the thoracoscopic approach has superior results such 18.2 months). During the follow-up period, the sputum cultures of all treatment failure patients were negative for as less wound pain, fewer pulmonary complications, and, mycobacteria. hence, shorter hospital stays than the thoracotomy ap- There was, in the thoracotomy group, one surgical proach in lung cancer surgery. Unlike lung cancer sur- mortality, which was due to exsanguinous bleeding dur- gery, however, the surgeon will always intraoperatively ing a pneumonectomy and resultant multiple organ fail- encounter inflammation-induced pleural space adhesion, ure. Two patients in the VATS group and 8 in the incomplete lobar fissures, and adhesive or calcified peri- thoracotomy group had complications (Table 4). Al- bronchial and perivascular lymph nodes with pulmonary though the complication rate was lower in the VATS TB. The impact of these common findings of TB on the group, the difference was not significant. One case of surgical approach, especially for VATS, has not been bleeding and one of prolonged air leak comprised the elucidated. Weber and colleagues [13] and Yim and complications in the VATS group, while 3 cases of bron- colleagues [14, 15] concluded that the thoracoscopic ap- chopleural fistula and 3 of prolonged air leak constituted proach provided a safe, effective diagnostic modality the majority of the complications in the thoracotomy (pleural biopsy, wedge resections) and therapeutic mo-

Fig 2. Image characteristics that are thoracoscopically approach- able, showing (A) grade 1 cavity (arrow), and (B) grade 1 pleural thickening (arrow) and grade 1 aspergilloma (arrowhead). Ann Thorac Surg YEN ET AL 293 2011;92:290–6 THORACOSCOPIC LUNG RESECTION FOR PULMONARY TB

Table 2. Demographics, Indications, Procedures, and Comparisons of Patients Who Underwent Video-Assisted Thoracoscopic Anatomic Lung Resection, and Who Were Converted to Thoracotomy Anatomic Lung Resection

VATS (n ϭ 21) Thoracotomy (n ϭ 29) p Value

Age (mean Ϯ SD) 48.57 Ϯ 10.07 51.09 Ϯ 16.89 0.455 Gender (M, %) 13 (61.9%) 26 (89.7%) 0.036a Indications 0.144 Hemoptysis 9 9 Treatment failure 10 20 Pneumothorax / bronchopleural fistula 1 0 RML syndrome 1 0 Procedures 0.068 Pneumonectomy 0 3 Bilobectomy 1 2 Lobectomy 19 21 Bisegmentectomy 1 7 Segmentectomy 0 1 Wedge resection 7 3 Preoperative sputum culture (positive: negative) 5:16 15:14 0.079 Ϯ Ϯ Ϯ GENERAL THORACIC Preoperative FEV1 (mean SD, l) 2.65 0.81 2.37 0.57 0.375 Preoperative FVC (mean Ϯ SD, l) 3.39 Ϯ 0.99 3.28 Ϯ 0.59 0.772 Ϯ Ϯ Ϯ Preoperative FEV1/FVC (mean SD, %) 79.08 8.61 72.30 11.12 0.122 Operative time (mean Ϯ SD, minute) 260.67 Ϯ 71.93 228.31 Ϯ 111.79 0.251 Blood loss (mean Ϯ SD, mL) 810.71 Ϯ 628.72 854.48 Ϯ 2230.66 0.931 Hospital stay (mean Ϯ SD, day) 10.00 Ϯ 7.79 14.96 Ϯ 8.96 0.048a Complications (%) 2 (9.52%) 9 (31.03%) 0.162 a Significant difference: p Ͻ 0.05. ϭ ϭ ϭ ϭ ϭ FEV1 forced expiratory volume in one second; FVC forced vital capacity; RML right mediolateral; SD standard deviation; VATS video-assisted thoracoscopic surgery. dality (decortications, drainage of empyema) for selected tissue [18]. We previously reported [19] that the presence patients with pulmonary tuberculosis but might not be of severe pleural adhesion and multiple cavities signifi- suitable for anatomic lung resection. cantly increased the surgical risk during pulmonary re- In this study, we retrospectively compared the differ- section by thoracotomy for TB. These image findings may ences of clinical parameters and findings on CT scans in render the thoracoscopic mobilization of the lung diffi- patients who underwent VATS and thoracotomy. We cult. Although VATS provides better vision of the pleural found that VATS was feasible for TB patients. We also space than does thoracotomy, and adhesion and incom- found that there was a gender difference between the plete fissures have been regarded as only relative con- VATS and thoracotomy groups. Although the incidence traindications for VATS [20], the lack of tactile feedback of patients with TB is lower in females than in males, TB of the thoracoscopic approach limits its use in adhesioly- is still the leading cause of death of women worldwide sis in severe pleural and interlobar adhesions [21]. The [16]. It has also been reported that female patients with presence of multiple tuberculomas in more than one lobe TB tend to have a lower default from treatment rate, a indicates repeated chronic infections and inflammatory lower retreatment rate, and a higher cure rate [17]. Our changes, which may result in dense adhesion in the results support this finding because our female patients fissure and interlobar area. Granulomatous involvement who underwent anatomic lung resection had a higher and the resultant calcification may cause the lymph incidence of success with thoracoscopic lobectomy. nodes to adhere markedly to the vascular and bronchial We also investigated the CT scan findings that pre- structure, making the dissection hazardous with the cluded thoracoscopic anatomic lung resection for TB thoracoscopic approach. These consequences are always patients. Our results demonstrated that extensive pleural combined with distortion of the lobar anatomy. The lobe thickening, multiple cavities, multiple aspergillomas, can be resected only using the thoracotomy approach, as multilobar tuberculomas, and calcified peribronchial the literature suggests [13]. On the contrary, a tubercu- lymph nodes limited using VATS for TB. The signifi- loma limited to one lobe often manifests as a solitary cantly shorter hospital stay in the VATS group was very pulmonary nodule and can be resected using the thora- likely due to the less complex parenchymal status. The coscopic approach for diagnostic purposes [14, 15, 22]. outcome of treatment is compromised if pulmonary tu- This is compatible with our findings that the thoraco- berculosis is diagnosed at a late stage and irreversible scopic approach is feasible for patients with a unilobar morphologic changes are already evident in the lung tuberculoma. 294 YEN ET AL Ann Thorac Surg THORACOSCOPIC LUNG RESECTION FOR PULMONARY TB 2011;92:290–6

Table 3. Comparisons of Image Characteristics and Grading aspergillomas. There was no surgical complication in in Patients Who Underwent Thoracoscopic Anatomic Lung patients with aspergilloma treated with the thoracoscopic Resection, and Who Were Converted to Thoracotomy approach. Anatomic Lung Resection Resection of cavitary lesions or destroyed lobes, which VATS Thoracotomy p harbor large numbers of bacilli, has been advocated [2, 5, Image Characteristic Grading (n ϭ 21) (n ϭ 29) Value 29] to improve the efﬁcacy of medical treatment for patients with MDRTB or open TB (TB in which tubercle Bullae 0 18 26 0.686 bacilli are being discharged from the body). Although the 13 3 number of cases of open TB and MDRTB were limited in 20 0 this study, they do not appear to contraindicate thoraco- a Pleural thickening 0 6 0 0.000 scopic anatomic lung resection. The most commonly 11311

EEA THORACIC GENERAL encountered complication in the thoracotomy group was 22 18 a bronchopleural fistula, known to be associated with a a Peribronchial lymph 0 18 15 0.015 positive preoperative sputum culture [2]. Two of the 3 node calcification bronchopleural fistulas occurred in patients who had a 12 7 positive preoperative sputum culture. Although some of 21 7 the literature [2, 29, 30] suggest that the bronchial stump a Tuberculoma 0 11 7 0.001 of those patients with open TB or MDRTB should be 19 6 reinforced with a viable intercostal muscle, pleural flaps, 21 16 or a pericardial fat pad, we did not perform any rein- Cavity 0 8 3 0.023a forcement in either group. There were no bronchopleural 11017 fistulas in the VATS group even though 5 of them had a 23 9 preoperative sputum culture positive for mycobacteria. Aspergilloma 0 17 16 0.022a Closure of the bronchial stump by stapler seemed suc- 14 7 cessful in the VATS group. This could be the result of 20 6 easier preservation of peribronchial soft tissue in the Atelectasis 0 19 24 0.356 VATS group. However, the patient number with positive 12 4 preoperative sputum culture in the thoracotomy group 20 1 was 3 times more than that in the VATS group. Because Bronchiectasis 0 10 16 0.567 the patient number was small in the VATS group, we 18 10 believe it is not safe to close the bronchial stump without 23 3 a muscle flap when the preoperative TB status is active. The thoracoscopic approach may not be appropriate for a Significant difference: p Ͻ 0.05. these patients. Further investigation is mandatory to VATS ϭ video-assisted thoracoscopic surgery. evaluate the real efficacy and safety of stapled bronchial closure. Our study does have some limitations because of its Surgical resection has been recommended as the treat- retrospective nature and small number of patients. We ment for pulmonary aspergilloma because of the risk of started using VATS at the beginning of 2007 and it has massive hemoptysis [23, 24]. Since Belcher and Plummer become the routine procedure for anatomic lung resec- [25] classified pulmonary aspergilloma into simple and tion since July of 2007. Retrospectively, before the learn- complex types, it has been reported [26] that patients ing curve of VATS was overcome, some suitable patients with simple forms are likely to present simpler and better might have been excluded. With increased experience postoperative recovery. Operations on complex aspergil- performing VATS, the surgeon’s tolerance and persever- lomas are often technically challenging because of the ance for performing dissection over adhesive and chron- dense fibrosis around the cavity, the obliteration of pleural space and fissures, induration of the hilar structure, the enlarged and tortuous bronchial arteries, and the diseased pulmonary parenchyma surrounding the Table 4. Complications of Patients Who Underwent Thoracoscopic Anatomic Lung Resection, and Who Were lesion. The patients in our study had all been affected by Converted to Thoracotomy for Anatomic Lung Resection pulmonary tuberculosis, and those with aspergilloma were diagnosed as having complex ones. It has been VATS Thoracotomy ϭ ϭ reported [26, 27] that if patients are meticulously selected, Complications (n 21) (n 29) a good long-term outcome may be expected for the Postoperative pneumonia 0 1 complex form. It is in cases where the disease has Prolonged air leak 1 3 reached an advanced stage and surgery is the only means Bronchopleural fistula 0 3 of controlling the symptoms that surgical mortality has Bleeding 1 2 proved to be extremely high [28]. Our results showed that Total 2 9 in patients with complex aspergilloma, VATS could be readily applied to those with single but not multiple VATS ϭ video-assisted thoracoscopic surgery. Ann Thorac Surg YEN ET AL 295 2011;92:290–6 THORACOSCOPIC LUNG RESECTION FOR PULMONARY TB

ically inflammatory areas may gradually increase and 13. Weber A, Stammberger U, Inci I, et al. Thoracoscopic lobec- more TB patients will benefit from VATS. tomy for benign disease--a single centre study on 64 cases. In conclusion, VATS can be safely performed in metic- Eur J Cardiothorac Surg 2001;20:443–8. 14. Yim AP. The role of video-assisted thoracoscopic surgery in ulously selected patients with TB or MDRTB. Except for the management of pulmonary tuberculosis. Chest 1996;110: those with multiple cavities, multilobar tuberculoma, 829–32. extensive pleural thickening, and peribronchial lymph 15. Yim AP, Izzat MB, Lee TW. Thoracoscopic surgery for node calcification on a chest CT scan, VATS can be used pulmonary tuberculosis. World J Surg 1999;23:1114–7. on patients with complications of pulmonary TB. 16. Borgdorff MW, Nagelkerke NJ, Dye C, Nunn P. Gender and tuberculosis: a comparison of prevalence surveys with noti- fication data to explore sex differences in case detection. Int We sincerely thank Mr Bill Franke for the revisions of grammar, J Tuberc Lung Dis 2000;4:123–32. sentence construction, word choice, and syntax. 17. Jimenez-Corona ME, García-García L, DeRiemer K, et al. Gender differentials of pulmonary tuberculosis transmission and reactivation in an endemic area. Thorax 2006;61:348–53. 18. Perelman MI, Strelzov VP. Surgery for pulmonary tubercu- References losis. World J Surg 1997;21:457–67. 19. Wu MH, Chang JM, Haung TM, et al. Computed tomo- 1. Kim HY, Song KS, Goo M, Lee JS, Lee KS, Lim TH. Thoracic graphic assessment of the surgical risks associated with sequelae and complications of tuberculosis. Radiographics fibrocavernous pulmonary tuberculosis. Surg Today 2004;34: 2001;21:839–60. 204–8. 2. Shiraishi Y, Nakajima Y, Katsuragi N, Kurai M, Takahashi N. 20. Patterson GA, Pearson FG, Cooper JD, et al, eds. Pearson’s Resectional surgery combined with chemotherapy remains thoracic and esophageal surgery. Philadelphia, PA: Chur-

the treatment of choice for multidrug-resistant tuberculosis. chill Livingstone; 2008:970–83. GENERAL THORACIC J Thorac Cardiovasc Surg 2004;128:523–8. 21. Zhao H, Bu L, Yang F, Li J, Li Y, Wang J. Video-assisted 3. Erdogan A, Yegin A, Gürses G, Demircan A. Surgical man- thoracoscopic surgery lobectomy for lung cancer: the learn- agement of tuberculosis-related hemoptysis. Ann Thorac ing curve. World J Surg 2010;34:2368–72. Surg 2005;79:299–302. 22. Rieger R, Woisetschläger R, Schinko H, Wayand W. Thora- 4. Takeda S, Maeda H, Hayakawa M, Sawabata N, Maekura R. coscopic wedge resection of peripheral lung lesions. Thorac Current surgical intervention for pulmonary tuberculosis. Cardiovasc Surg 1993;41:152–5. Ann Thorac Surg 2005;79:959–63. 23. Massard G, Roeslin N, Wihlm JM, Dumont P, Witz JP, 5. Kir A, Inci I, Torun T, Atasalihi A, Tahaoglu K. Adjuvant Morand G. Pleuropulmonary aspergilloma: clinical spec- resectional surgery improves cure rates in multidrug- resistant tuberculosis. J Thorac Cardiovasc Surg 2006;131: trum and results of surgical treatment. Ann Thorac Surg 693–6. 1992;54:1159–64. 6. Sihoe AD, Shiraishi Y, Yew WW. The current role of thoracic 24. Pecora DV, Toll MW. Pulmonary resection for localized surgery in tuberculosis management. Respirology 2009;14: aspergillosis. N Engl J Med 1960;263:785–7. 954–68. 25. Belcher JR, Plummer NS. Surgery in broncho-pulmonary 7. Kim K, Kim HK, Park JS, et al. Video-assisted thoracic aspergillosis. Br J Dis Chest 1960;54:335–41. surgery lobectomy: single institutional experience with 704 26. Kim YT, Kang MC, Sung SW, Kim JH. Good long-term cases. Ann Thorac Surg 2010;89:S2118–22. outcomes after surgical treatment of simple and complex 8. Wu MH, Lin MY, Tseng YL, Lai WW. Results of surgical pulmonary aspergilloma. Ann Thorac Surg 2005;79:294–8. treatment of 107 patients with complications of pulmonary 27. Akbari JG, Varma PK, Neema PK, Menon MU, Neelakand- tuberculosis. Respirology 1996;1:283–9. han KS. Clinical proﬁle and surgical outcome for pulmonary 9. Landreneau RJ, Hazelrigg SR, Mack MJ, et al. Postoperative aspergilloma: a single center experience. Ann Thorac Surg pain-related morbidity: video-assisted thoracic surgery ver- 2005;80:1067–72. sus thoracotomy. Ann Thorac Surg 1993;56:1285–9. 28. Henderson RD, Deslaurier J, Ritcey EL, Delarue NC, Pearson 10. Kaseda S, Aoki T, Hangai N, Shimizu K. Better pulmonary FG. Surgery in pulmonary aspergillosis. J Thorac Cardiovasc function and prognosis with video-assisted thoracic surgery Surg 1975;70:1088–94. than with thoracotomy. Ann Thorac Surg 2000;70:1644–6. 29. Pomerantz BJ, Cleveland JC Jr, Olson HK, Pomerantz M. 11. Li WW, Lee TW, Lam SS, et al. Quality of life following lung Pulmonary resection for multi-drug resistant tuberculosis. cancer resection: video-assisted thoracic surgery vs thoracot- J Thorac Cardiovasc Surg 2001;121:448–53. omy. Chest 2002;122:584–9. 30. Mohsen T, Zeid AA, Haj-Yahia S. Lobectomy or pneumo- 12. Nomori H, Horio H, Naruke T, Suemasu K. Posterolateral nectomy for multidrug-resistant pulmonary tuberculosis can thoracotomy is behind limited thoracotomy and thoraco- be performed with acceptable morbidity and mortality: a scopic surgery in terms of postoperative pulmonary function seven-year review of a single institution’s experience. J Tho- and walking capacity. Eur J Cardiothorac Surg 2002;21:155–6. rac Cardiovasc Surg 2007;134:194–8.

INVITED COMMENTARY

The article by Yen and colleagues [1] delineates imaging omy included marked pleural thickening, peribronchial characteristics portending successful anatomic pulmo- lymph node calciﬁcation, multilobar tuberculomas, mul- nary resection by video-assisted thoracic surgery (VATS), tiple cavities, and multiple aspirgillomas. as well as features militating against a minimally invasive Surgical treatment of tuberculosis (TB) and its compli- approach. Their series, accumulated over a relatively cations can be formidable. Completion of VATS anatomic short time, includes 50 patients undergoing an initial resection in 42% of the authors’ cases represents an VATS approach, with complete anatomic resection impressive advance. Their report adds to an increasing achieved in 21 of these patients. Findings on computed recent global experience with open anatomic pulmonary tomography forecasting ultimate conversion to thoracot- resection and nonresectional VATS approaches for a