ARTICLE IN PRESS doi:10.1510/icvts.2006.129676
Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607 www.icvts.org
Negative results - Thoracic general
Reoperations are common following VATS for spontaneous pneumothorax: study of risk factors Ingimar Ingolfsson, Erik Gyllstedt, Ramon Lillo-Gil, Andreas Pikwer, Per Jo ¨nsson, Tomas Gudbjartsson* Department of Cardiothoracic Surgery, Lund University Hospital, S 222 85, Lund, Sweden Received 26 January 2006; received in revised form 6 May 2006; accepted 3 July 2006
Abstract Objectives: To identify risk factors for reoperation following surgery for spontaneous pneumothorax. Material and methods: This is a retrospective clinical study conducted on 240 consecutive patients operated on for spontaneous pneumothorax in Lund University Hospital between January 1996 and December 2003. Patient information was gained from medical charts and operation reports. Logistic regression analysis was used to evaluate risk factors. Median follow-up was 54 months. Results: Video-assisted thoracoscopic surgery (VATS) was used in 93% of the patients, where bullectomy with or without pleurodesisypleurectomy was performed in most cases. Conversion to anterolateral thoracotomy was required in 6 cases (2.3%). Operative mortality (-30 days) was 0.8% (0% after VATS). Altogether 35 patients (13.7%) required a reoperation. Twenty patients were reoperated on within one month from surgery, most often due to prolonged air leakage (ns13, 5.1%) and bleeding (ns6, 2.3%). Fifteen patients (5.8%) were reoperated on because of recurrent pneumothorax, on average 17 months (range 1–39 months) after the primary operation. Significant independent risk factors for reoperation were; young age, operations performed for secondary pneumothorax due to emphysema and recurrent pneumothorax. The risk was also increased in patients where wedge resection was not performed. However, the type of pleurodesis or pleurectomy was not related to increased risk. Conclusions: A considerable number of patients require reoperation after VATS for pneumothorax. Most reoperations are due to prolonged air leakage or recurrent pneumothorax. The risk for reoperation is increased in younger patients, patients operated for recurrent or secondary pneumothorax, and in cases where pulmonary wedge resection is not performed. 䊚 2006 Published by European Association for Cardio-Thoracic Surgery. All rights reserved. Keywords: Pneumothorax; Recurrent pneumothorax; Surgical technique; Reoperations; Risk factors
1. Introduction Spontaneous pneumothorax (SP) is a relatively common condition affecting mostly young and otherwise healthy individuals. It is most often caused by a rupture of apical subpleural blebs (primary SP) but can also occur in patients with underlying lung diseases, typically COPD or emphysema (secondary SP). Conservative management with needle aspiration or more commonly chest tube drainage remains the standard treatment for the first episode of SP. However, recurrences are frequent and in a long-term perspective 40% or more of the patients treated conservatively eventually require a more definite treatment w1x. This includes instillation of pleural irritants (e.g. talc, tetracyclin) or more frequently surgery, including excision of leaking blebs and pleurodesis. Until a decade ago operations for pneumothorax were performed through thoracotomy. Today most of these operations are performed with a minimal invasive technique using video-assisted thoracoscopic surgery (VATS). In contrast to conservative treatment, recurrence rates following surgery are low, most often in the 1– 6% range w2–9x. On the other hand, complications are more *Corresponding author. Fax: q46 46 158635. E-mail address:
[email protected] (T. Gudbjartsson). 䊚 2006 Published by European Association for Cardio-Thoracic Surgery
serious following surgical procedures and some of the patients need a reoperation, with further risks of postoperative morbidity. To improve the long-term results of pneumothorax surgery and prevent recurrences, it is important to identify risk factors for reoperation. Some of these risk factors are well defined (underlying pulmonary disease) when others are not (type of surgery, age of the patient) w5,10–12x. We therefore studied the results of surgery for SP in our institution during an 8-year period. 2. Material and methods 2.1. Patient characteristics This is a retrospective clinical study conducted on 240 consecutive patients that underwent 256 first-time operations for spontaneous pneumothorax in Lund University Hospital between 1 January 1996 and 31 December 2003. Patient information was gained from medical charts and operation reports. The demographic distribution of the patients is shown in Table 1. There were 171 males and 69 females (MyF ratio 2.5:1). More than fifty percent of the patients (ns129)
ARTICLE IN PRESS I. Ingolfsson et al. / Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607 Table 1 Patient demographics (number of patients and percentages or standard deviation in parentheses) Males Females Age (years, mean"S.D.) Weight (kg) Length (cm) Body Mass Index Laterality of pneumothorax Right Left Bilateral Current smokers History of smoking Primary spontaneous pneumothorax (PSP) Secondary spontaneous pneumothorax (SSP)
171 (71.2) 69 (28.8) 42.0 ("20.8) 66.9 ("12.6) 176.9 ("9.3) 21.2 ("3.4) 141 (55.1) 110 (42.9) 5 (1.9) 76 (31.7) 130 (54.2) 173 (72.1) 67 (27.9)
were either former or current smokers whereas 27.9% (ns67) had a confirmed COPD or emphysema (secondary SP). 2.2. Surgical technique The surgery was performed by four senior surgeons. All procedures were performed with general anaesthesia and selective double lumen intubation. For VATS the patient was placed in a 908 lateral position and single-lung ventilation started before deflation of the lung with CO2 through a Veress-needle. Then a 10 mm trocar for the camera was placed in the sixth intercostal space in the mid-axillary line and a 5 mm trocar under the fourth rib, just below the scapula tip, for graspers and diathermy. The lung was carefully inspected. If blebs or bullae were found they were removed with a linear stapler (ETS-Flex 45, Ethicon Endo-Surgery Inc.) introduced through a 12 mm hole in the anterior fourth intercostal space. In most cases the procedure was combined with pleurodesis andyor pleurectomy (see results), the mode of treatment being decided by the individual surgeon. Two kinds of pleurodesis were performed; the parietal pleura was either rubbed with a pledget of sandpaper or wide-mesh polypropylene gauze to achieve inflammation (mechanical pleurodesis), or chemical pleuritis was induced with talc powder (dry asbestos free talcum). Partial pleurectomy was performed apically from the fifth intercostal space both anteriorly and posteriorly. At the end of the procedure one chest tube was placed and connected to an underwater seal suction with a negative pressure of 20 cm H2O. The thoracotomies were performed through an anterolateral incision through the 4th or 5th intercostal space with the patient in a 458 lateral position. Wedge resection, pleurectomy or pleurodesis was then performed in a similar fashion as described for the thoracoscopic procedure. 2.3. Postoperative care Patients were extubated in the operating room and observed for 4–6 h in a postoperative care unit. Daily chest X-rays were obtained for each patient. Chest tubes were usually removed after 2–3 days postoperatively. At that time the lungs were fully expanded, no air leakage was
603
present and pleural drainage was less than 100 ml per 24 h. Active physiotherapy was started on the first postoperative day and patients usually got patient-controlled analgesia or epidural analgesia if the procedure was performed through a thoracotomy. A control chest X-ray was done before discharge and 2–3 weeks postoperatively. A further follow-up was not conducted except in special circumstances where the patients had excessive postoperative pain or residual air. 2.4. Follow-up On follow-up only recurrent pneumothorax requiring redo surgery was registered. Median follow-up was 54 months (range 9–99 months), with end of follow-up on 31 March 2004. Prolonged air leakage was defined as chest tube air drainage for more than 7 days postoperatively. Most of these patients were reoperated on within 10 days from the primary operation, after conservative treatment had failed. 2.5. Statistics Patient information and intra- and postoperative data were collected on all patients. Continuous variables were checked using Student’s t-test and categorial variables with Chi-square test. Logistic regression analysis was used to identify significant prognostic factors for reoperation, where odds ratio (OR) was used as a measure of risk of reoperation. Univariate analysis was performed for different clinical factors, including age and gender. Each factor that was statistically significant by univariate analysis adjusted by age was included in the multivariate logistic regression model. Statistical significance was set at P-0.05 and 95% confidence intervals were used for odds ratios. 3. Results Altogether 256 operations were performed on 240 patients for spontaneous pneumothorax during the 8 years of the study. The indications for surgery are shown in Table 2. Most of the patients (58.6%) were operated on after experiencing one or more recurrences of pneumothorax. The remaining patients were operated on during their first episode of pneumothorax, usually due to persistent air leak or failure of re-expansion of the lung. VATS was the most common type of surgery (93.0%) (Table 3). Six of the VATS cases (2.3%) had to be converted to a thoracotomy, most often due to inadequate exposure or extensive pleuropulmonary adhesions. Eighteen patients (7.0%) were primarily operated on with anterolateral thoracotomy, all of them being elderly with advanced COPD. Wedge resection of blebs or bullae was performed in 226 (88.3%) of the cases (Table 3). In the remaining patients Table 2 Indications for surgery (number and percentage in parentheses) First episode First recurrence 2nd or 3rd recurrence 3rd ipsilateral recurrence First contralateral recurrence Synchronous bilateral
97 71 68 11 4 5
(37.9) (27.7) (26.6) (4.3) (1.6) (1.9)
ARTICLE IN PRESS 604
I. Ingolfsson et al. / Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607
(11.7%) no bullae were found and only pleurodesis andyor pleurectomy were performed. Altogether pleurodesis was performed in 191 cases (74.6%) and apical pleurectomy in 100 cases (39.1%). A combination of wedge resection and pleurodesis was done in 162 cases (63.3%) and wedge resection, pleurodesis and pleurectomy in 43 cases (16.8%). Talc (35.1%) or sandpaper (28.9%) were the most common mode of pleurodesis. The mean operative time was 63 min and the median length of stay was 4 days (range 1–54 days). Operative mortality (-30 days) was 0.8%, involving two patients operated with thoractomy. Both patients had advanced COPD and died of postoperative respiratory failure. There were no operative deaths after VATS (operative mortality 0%). Chest tubes could be withdrawn in 3 days (median). However, 29 patients (11.3%) had prolonged air leak that resulted in 13 early (-1 month) reoperations (5.1%), where missed blebs and staple leaks were found in 6 and 2 patients, respectively. Other early postoperative complications are summarized in Table 4. Postoperative, bleeding resulted in six reoperations (2.3%), the most common bleeding site being from the thoracic wall. Other complications were rare and most often minor, except for empyema that was found in two patients (0.8%), one of whom underwent evacuation and decortication.
Table 4 Early postoperative complications (number of patients and percentages in parentheses) Prolonged airleakage* Bleeding** Empyema*** Urinary tract infection Pleural effusion Pneumonia ARDS
29 6 2 1 1 1 1
(11.3) (2.3) (0.8) (0.4) (0.4) (0.4) (0.4)
* 13 patients reoperated; ** all patients reoperated for bleeding; *** one reoperated with decortication.
A total of 15 patients (5.8%) were reoperated because of recurrent pneumothorax, on average 17 months (range 1– 39 months) after the primary operation. Table 5 illustrates patient demographics, time of recurrence and the type of primary operations for the patients that were reoperated. No missed blebs were found. The late recurrences were treated with wedge resection in 5 patients (33.3%) and in 8 (53.3%) with talc pleurodesis. One patient was operated on with a middle lobectomy and another patient underwent pulmonary decortication. Seven of the redo operations were thoracotomies. Patients in the reoperation group were younger (25.5 vs. 42.9 years, P-0.01) and less often oper-
ARTICLE IN PRESS I. Ingolfsson et al. / Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607
605
Table 5 Patients reoperated for recurrent pneumothorax (late reoperations). The primary operation was VATS in all 15 cases Number
Year of operation
Agey Gender
Side
Indication for primary op.
Type of operation
Time to redo op (mo)
Type of reoperation
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1996 1996 1996 1997 1997 1998 1998 1998 1998 1999 1999 1999 1999 2000 2002
16yM 14yM 27yF 64yM 18yM 19yM 21yM 39yF 45yF 17yM 16yM 32yM 13yM 15yM 26yM
R L L R L R L L R RqL R R L L R
1st episode 2ndy3rd recurr. 1st recurr. 1st episode 1st recurr. 3rd recurr. 3 recurr. 1st recurr. 3 ips recurr. Bilat. recurr. 1st episode 1st recurr. 1st recurr. 2ndy3rd recurr. 2ndy3rd recurr.
WRqMP WRqMP MPqPE TPqPE PE WRqMP WRqPE TP WRqMPqPE WRqSP WRqPE WRqSP WRqPE WRqMP WRqPE
7 35 24 3 1 8 9 20 1.5 23 27 19 29 39 8
WRqTP SP (open) TP Lobectomy, middle (open) TPqPE SP (open) SP (open) SPqPE SPqdecort. (open) TPqPE (open) TPqPE WRqPE WRqSP WRqTPqSP WRqTPqPE (open)
WR s wedge resection; MP s Marlex䊛 -pleurodesis (polypropylene-net); TP s talc-pleurodesis; SP s sandpaper-pleurodesis; PE s pleurectomy (partial); Conv s VATS converted to thoracotomy; Open s thoracotomy.
ated on with pulmonary wedge resection, compared to patients that were not reoperated (73.3% vs. 89.9%, P-0.05). Risk factors for reoperation were evaluated with linear regression. Young patients (OR of advanced age per year 0.95, CI-95% 0.92–0.98, Ps0.001) and patients operated on for recurrent (OR 5.2, CI-95% 1.39–19.51, Ps0.015) or secondary pneumothorax with emphysema (OR 3.62, CI-95% 1.01–12.98, Ps0.048) had an increased risk of having redo surgery. The risk for reoperation was also increased for patients that were not operated on with a wedge resection during the primary surgical procedure (OR of wedge resection 0.23, CI-95% 0.07–0.74, Ps0.014). Genders, weight, length, BMI, laterality, pleurectomy and pleurodesis, including type of pleurodesis, were not significant predictive factors of recurrence. Risk factors were comparable for patients reoperated on for prolonged air leakage (early failure) or recurrent pneumothorax (late failure). 4. Discussion This study shows that a significant number of patients operated on for spontaneous pneumothorax require a reoperation because of early or late failure (10.9%). The late reoperation rate of 5.8% is in line with most recent reports, where it usually lies in the 3–6% range w3,5,8,10,13–18x. However, our 5.1% early reoperation rate is higher than that reported elsewere w3,5,8,10,17,18x and is of concern. In most instances it could be explained by technical failure, i.e. air leak from staple lines or missed blebsybullae. There are no obvious explanations for this high reoperation rate. Our patient material includes both elective and emergency cases, except that pneumothorax secondary to trauma (e.g. vehicle accidents) were excluded. Some of the previously mentioned studies only include patients with primary SP, and not patients with secondary SP (27.9% in the present study), where risks for recurrent pneumothorax and other complications are increased due to underlying lung disease w5,10–12x. Our surgical technique is similar to that described by others and the same is true for postoperative care, includ-
ing indications for chest tube removal w3,5,8–10,12,17,18x. All cases were performed by a team of four attending surgeons with many years of experience with operations for pneumothorax and specifically VATS. Furthermore, during this period residents or fellows did not perform pneumothorax surgery in our institution and the learning curve is therefore not a likely explanation for our results being different from others. The reoperation rate is affected by the surgeon9s indications for taking the patients back to the operating room. This is especially true for the patients with early failures. There were no formal criteria for a reoperation, however, patients with a significant and continuing air leakage for a week (or more) after surgery were usually reoperated, regardless of age. Surgery was often preferred over conservative measures, such as the application of talc slurry or tetracycline through the chest tubes, which is often painful for the patient and ineffective w19,20x. Risk factors for reoperations were similar for patients with early or late failure. This was especially true for patients not operated on with a pulmonary wedge resection. Most of these patients did not have visible blebs at the time of surgery, neither at the apex nor other parts of the lung. Therefore, only pleurodesis andyor limited pleurectomy was performed. Bullae are most often located at the apex of the lungs but can be difficult to identify during thoracoscopic surgery. Horio et al. found that patients operated on with VATS had higher recurrences compared to patients operated on with thoracotomy. They suggested that these differences could be explained by overlooked bullae in the VATS-group w21x. Compared to open surgery, air leakage tests are more technically challenging and difficult to interpret during VATS, and in the present series they were only performed selectively. Our results indicate that it is not adequate to only performing pleurectomy or pleurodesis in cases where blebs or bullae are not found. Other authors have also recommended wedge resection of the apex in all cases w11,18x, pointing out that the protective value of wedge resection does not only lie in the removal of blebs, but also in the creation of inflammation that leads to adhesions between the lungs and the thoracic wall.
ARTICLE IN PRESS 606
I. Ingolfsson et al. / Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607
Younger patients had an increased risk of reoperation compared to older individuals. The reason for this is not known, and to our knowledge this has not been reported in other studies. Treatment strategies were similar for this subgroup of patients, except that talc pleurodesis was less seldom used in patients under 40 years of age. Today VATS has replaced open pneumothorax surgery in most centres. VATS is considered to be a cost-effective procedure compared to conventional open surgery. It is also safe as demonstrated in the present study with zero mortality after 238 VATS procedures. Two patients died from respiratory failure after open surgery. This does not mean that open surgery is more dangerous than VATS, as the patients that underwent open surgery were older and had more often severe pulmonary disease. As for open surgery, complications after VATS are most often minor, and except for prolonged air leakage, postoperative bleedings are the most common early complications (2.3%). Only a handful of patients sustained serious complications, including two patients that had empyema. According to a number of studies, four of them randomised, postoperative pain and hospital stay after VATS is significantly reduced compared to open surgery while recurrence rates are only slightly higher w3,5,8,10,12–17,22x. However, other studies have reported much higher (up to 6–8 times) recurrence rates for VATS than for open surgery w19,23x. This is of concern and must be regarded as a shortcoming of the VATS approach. In general the rate of reoperation should be kept as low as possible as they are more technically challenging with a greater risk for thoracotomy and, therefore, added morbidity to the patient. The recurrence rates after VATS might also have to be taken into consideration, when the operation is performed on patients where recurrences must be avoided by all means, such as when the operation is performed on pilots and divers. Perhaps these patients should instead be offered conventional thoracotomy that only carries a 0.5–2% recurrence risk w24,25x. How can reoperations be prevented? Data from previous studies are often difficult to interpret, most of the studies being non-randomised with a limited number of patients. Furthermore, the differences in procedures and treatment strategies make direct comparison difficult. Our surgical approach must be regarded as relatively radical regarding the parietal pleura. Wedge resection and pleurectomy were performed on 35.9% of the patients, 63.3% underwent pleurodesis and 16.8% of the patients had both pleurectomy and pleurodesis in addition to the wedge resection. However, neither pleurectomy andyor pleurodesis (nor type of pleurodesis) in addition to wedge resection, affected the rate of recurrent pneumothorax in the logistic regression model. Our data, therefore, do not support this treatment strategy, at least not in all cases. This is important since the risk for postoperative complications can theoretically be increased by adding a potentially traumatic procedure, like pleurectomy, to the parietal pleura. Wheras previous studies have shown a slight advantage of pleurectomy over pleurodesis w25x, no such difference could be detected in the present series. A shortcoming of this study is the fact that it is not randomised. A certain selection bias regarding different treatment strategies and surgeons is therefore possible.
Still, we do not think that the non-randomised design of the present study significantly affects our results andyor conclusions. Cases were referred unselectively to each of the four operating surgeons, who thereafter decided treatment on an individual basis. Importantly, all of the four surgeons used a similar surgical technique and no differences in recurrence rates were found between the four surgeons (P-0.1). To detect minor differences in treatment, e.g. differences between different types of pleurodesis, a larger number of patients would have been of value. However, 240 patients must be regarded as a significant number of patients, and we think that if there are differences between treatments, these differences are most likely small. A prospective randomised multi-centre study is the only way to address these questions properly. 5. Conclusions Videothoracoscopic operations for pneumothorax are safe procedures and can be used for patients with both primary and secondary SP. However, there is a considerable risk of reoperations with more than one in ten patients needing further surgery due to recurrent pneumothorax or prolonged air leakage. This risk is increased in younger patients and patients operated on for secondary pneumothorax. Furthermore, pulmonary wedge resection seems to protect against recurrence, while pleurodesis andyor pleurectomy in addition to wedge resection, do not. Acknowledgements Helgi Sigvaldason PhD for statistical help, professor Jonas Magnusson MD, PhD for comments and Birgitta Sjo ¨gren for secretarial help. References w1x Beumann MH, Strange C. Treatment of spontaneous pneumothorax. Chest 1997;112:789–804. w2x Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J Med 2000;342:868–874. w3x Bernand PC, Regnard JF, Spaggiari L, Levi JF, Magdeleinat P, Guibert C, Levasseur P. Immediate and long-term results after surgical treatment of primary spontaneous pneumothorax by VATS. Ann Thorac Surg 1996;61:1641–1645. w4x Korner H, Andersen KS, Stangeland L, Ellingsen I, Engedal H. Surgery treatment of spontaneous pneumothorax by wedge resection without pleurodesis or pleurectomy. Eur J Cardiothorac Surg 1996;10:656–659. w5x Mouroux J, Elkaim D, Padovani B, Myx A, Pervin C, Rotomondo C, Chavaillon JM, Blaire B, Richelme H. Video-assisted thoracoscopic treatment of spontanous pneumothorax. Technique and results of one hundred cases. J Thorac Cardiovasc Surg 1996;112:385–391. w6x Crisci R, Coloni GF. Video-assisted thoracoscopic surgery versus thoracotomy for recurrent spontaneous pneumothorax: a comparison of results and costs. Eur J Cardiothorac Surg 1996;10:556–560. w7x Leo F, Pastorino U, Goldstraw P. Pleurectomy in primary pneumothorax: is extensive pleurectomy necessary? J Cardiovasc Surg 2000;41:633– 636. w8x Lang-Lazdunski L, Chapuis O, Bonnet PM, Pons F, Jancovici R. Videothoracoscopic bleb excision and pleural abrasion for the treatment of primary spontaneous pneumothorax: long-term results. Ann Thorac Surg 2003;75:960–965. w9x Ayed AK, Al-Din. The results of thoracoscopic surgery for primary spontaneous pneumothorax. Cheast 2000;118:235–238.
ARTICLE IN PRESS I. Ingolfsson et al. / Interactive CardioVascular and Thoracic Surgery 5 (2006) 602–607 w10x Passlick B, Born C, Ha ¨ussinger K, Thetter O. Efficiency of video-assisted thoracic surgery for primary and secondary spontaneous pneumothorax. Ann Thorac Surgery 1998;65:324–327. w11x Kruger M, Ermitsch M, Uschinsky K, Engelmann C. Results of videoassisted thoracoscopic surgery for pneumothorax. Zentralbl Chir 2003;128:645–651. w12x Waller DA, Forty J, Soni AK, Conacher ID, Morritt GN. Videothoracoscopic operation for secondary spontaneous pneumothorax. Ann Thorac Surg 1994;57:1612–1615. w13x Naunheim KS, Mack MJ, Hazelrigg SR, Ferguson MK, Ferson PF, Boley TM, Landrenau RJ. Safety and efficacy of video-assisted thoracic surgery techniques for treatment of spontaneous pneumothorax. J Thorac Cardiovasc Surg 1995;109:1198–1204. w14x Massard G, Thomas P, Wihlm JM. Minimally invasive management for first and recurrent pneumothorax. Ann Thorac Surg 1998;66:592–599. w15x Gebhard FT, Becker HP, Gerngross H, Bruckner UB. Reduced inflammatory response in minimal invasive surgery of pneumothorax. Arch Surgery 1996;131:1079–1082. w16x Ayed AK, Al-Din HJ. Video-assisted thoracoscopy versus thoracotomy for primary spontaneous pneumothorax: a randomized controlled trial. Med Principles Pract 2000;9:113–118. w17x Hatz RA, Kaps MF, Meimarakis G, Loehe F, Mu ¨ller C, Fu ¨rst H. Long-term results after video-assisted thoracoscopic surgery for first-time and recurrent spontaneous pneumothorax. Ann Thorac Surg 2000;70:253– 257.
607
w18x Czerny M, Andreas S, Fleck T, Hofmann W, Zimpfer D, Eckersberger F, Klepetko W, Wolner E, Mu ¨ller MR. Lung wedge resection improves outcome in stage I primary spontaneous pneumothorax. Ann Thorac Surg 2004;77:1802–1805. w19x Walter-Renard PB, Waughan LM, Sahn SA. Chemical pleurodesis for malignant pleural effusions. Ann Tntern Med 1994;120:56–64. w20x Kennedy L, Sahn SA. Talc pleurodesis for the treatment of pneumothorax and pleural effusion. Cheast 1994;106:1215–1222. w21x Horio H, Nomorio H, Fuyuno G, Kobayashi R, Suemasuk K. Limited axillary thoracotomy vs. video-assisted thoracoscopic surgery for spontaneous pneumothorax. Surg Endosc 1998;12:1155–1158. w22x Sekine Y, Miyata Y, Yamada K, Yamada H, Yasukawa T, Saitoh Y, Yoshica S, Fujisawa T. Video-assisted thoracoscopic surgery does not deteriorate postoperative pulmonary gasexchange in spontaneous pneumothorax patients. Eur J Cardiothorac Surg 1999;16:48–53. w23x Noppen N, Meysman M, d’Haese J, Monsieur I, Verhaeghe W, Schlesser M, Vincken W. Comparison of video-assisted thoracoscopic talcage for recurrent primary versus persistent secondary spontaneous pneumothorax. Eur Respir J 1997;10:412–416. w24x Deslauriers J, Beaulieu M, Despres JP, Lemieux M, Leblanc J, Desmeules M. Transaxillary pleurectomy for treatment of spontaneous pneumothorax. Ann Thorac Surg 1980;30:569–574. w25x Weeden D, Smith G. Surgical experience in the management of spontaneous pneumothorax, 1972–82. Thorax 1983;38:737–743.
