New Technologies in Surgery, 2009; 1(1):

THE COMPLICATIONS OF VIDEOASSISTED THORACOSCOPIC SURGERY (VATS)

A. Imperatori, N. Rotolo, M. Gatti, C. Antonini, L. Dominioni Center for Thoracic Surgery, University of Insubria, Varese, Italy The authors have no proprietary interest in any aspect of this manuscript. Received: June 2009; Published: June 2009. Correspondence: A. Imperatori MD, Center for Thoracic Surgery, University of Insubria, Viale Borri, 57 - 21100 Varese, Italy. E-mail: andrea.imperatori@uninsubria.it; This study was supported by research grants from F.A.R. (Fondi di Ateneo per la Ricerca) by University of Insubria 20022004, Varese, Italy and from research grant by Fondazione Giancola per la ricerca sul cancro, Como, Italy.

Abstract
VATS has multiple indications, for diagnosis and treatment of many different thoracic diseases, the commonest being lung wedge resection, pleural and mediastinal biopsy, treatment of pneumothorax, pleurectomy. Moreover, in recent years a few surgeons have performed routinely major lung anatomic resections by the VATS approach, including segmentectomy, lobectomy and pneumonectomy. In our experience and in the reviewed literature, the most frequent complications after VATS procedures are: prolonged air leak, bleeding, infection, post-operative pain, port-site recurrence and the need to convert the access into thoracotomy. Their rate of occurrence is generally low and VATS procedures are considered safe and effective, as witnessed by the fact that they constitute about one third of all thoracic surgical procedures. Key words: Video-assisted thoracoscopic surgery, complications, bleeding, surgical site infection, air-leak, lung surgery. Abbreviations:VATS = Video-assisted thoracoscopic surgery;

Introduction
Video-assisted thoracoscopic surgery (VATS) procedures have proved to be safe and effective for treatment of a large variety of thoracic diseases, with low perioperative complication rates as reported in several large series of patients. Fifteen years after the beginning of the widespread application of VATS, this paper reviews the complications associated with VATS procedures

Overall morbidity and mortality after VATS

In three large studies published in 1996 the overall complication rates after a variety of VATS procedures were respectively 4%,(1) 4.3%(2) and 3.7%(3), with no reported deaths attributable to the VATS technique. Similarly, in the very large series reported by the Video Assisted Thoracic Surgery Group (55 surgeons from 40 institutions who performed general thoracic surgery as the majority of their practices)4, the complications observed in 1,358 patients undergoing VATS procedures were: air leak longer than 5 days in 3.2% of patients; atelectasis in 1.4%; arrhythmia in 1.3%; bleeding in 1.2% (one third required transfusion); pneumonia in 1.1%; prolonged ventilator dependence in 1.0%; empyema in 0.6%; wound infection in 0.4%. There were no intraoperative deaths, and the total postoperative mortality rate for the 1,358 patients was 2.0% (27 patients). Causes of deaths were sepsis in seven patients, respiratory insufficiency in four, progression of malignancy in three, pneumonia in two, renal failure in two; hepatic failure, cardiac arrest, pulmonary embolus, and stroke occurred each in one patient. In our center for thoracic surgery at the University of Insubria in Varese, after reviewing 721 consecutive VATS procedures performed from 1996 to 2005, we recorded the following complications: prolonged air leak in 4.7% of patients; pneumonia in 2.4%; bleeding (with blood transfusion) in 1.5%; wound infection in 0.9%. Conversion to thoracotomy was necessary to complete the procedure in 2.2% of cases; the perioperative mortality was nil. Table 1 summarizes the main complications observed in seven large series of VATS procedures and shows that the most prevalent are prolonged air leak and bleeding, while other complications are infrequent1-3,5-7. We will analyse separately the prevalence and feature of these complications of VATS , as related to the specific procedures.

Prolonged air leak

Prolonged air leak is the most prevalent complication of VATS and is more likely to occur after resection of lung nodules and bullae After VATS lung biopsy, moderate air leakage from parenchyma which has been sutured or biopsed or traumatized by instruments occurs frequently. The thermal injury of electrocautery may also predispose to prolonged air leak. As most patients can be discharged within 48 hours, persistent air leak should be avoided. Endoscopic stapling instruments of various sizes may be used to staple and resect leaking parenchyma. In the operating room, as the lung is inflated, staple lines should be visualized for bleeding and air leak. Even after complete control of air leakage, at the end of each VATS procedure it is advisable to place routinely a chest tube, under direct vision, to provide drainage of any subsequent air and fluid losses. As pointed out

by Adebonojo the definition of persistent air leak is arbitrary.8 Indeed, postoperative air leak has been reported with many different lengths of leakage (5 to 10 days), which makes comparison of results difficult. In different large series prolonged air leak was reported respectively in 3.2%4, 3.7%9, 4.7%10, 6.7%.3 In our experience air leak longer than 7 days after 721 VATS procedures occurred in 4.7%.

Bleeding
VATS usually offers an excellent visibility of the operative field, thus intrathoracic haemorrhage is rare. Bleeding occurs infrequently when an experienced surgeon performs the procedure; it is more likely to occur in patients with dense adhesions. Bleeding from one of the main pulmonary vessels during VATS, however, can be very dangerous because haemorrhage is sudden and catastrophic and access is limited. Indeed VATS lung resections of any type are contraindicated in patients with tumours near the pulmonary hilum; VATS should be limited to biopsy for diagnosis and a sponge stick should be available to apply pressure immediately for controlling haemorrhage. With bleeding temporarily controlled, a decision is immediately made as to whether thoracotomy is needed. Severe parenchymal haemorrhage may require open thoracotomy and lobectomy for control. Haemorrhage may also derive from trocar site access or from lung biopsy, but these are uncommon events, that are more likely to occur if the patient has a coagulation defect. Depending upon the circumstances, bleeding can be controlled by electro-coagulation, transfixed sutures, metal clips or with the linear stapler. Kaiser et al in 266 VATS reported postoperative bleeding (requiring transfusion) in 1.9% of cases (4 patients required thoracotomy).9 Walker et al reported a postoperative bleeding rate of 0.5%.10 In our series of 721 VATS procedures bleeding occurred in 1.5% (Table 1). Table 1. Postoperative complications after VATS procedures in different series (case-mix and procedures differ).

Authors Patients
Yim (2) Jancovici (3) Inderbitzi (5) Krasna (1) Kaiser (6) Hazelrigg (7) Dominioni 1,337 937 5,280 348 266 1,820 721

Prolonged Wound Bleeding Death air leak infection

21 (1.6%) 63 (6.7%) 93 (1.8%) 3 (0.9%) 10 (3.8%) 43 (3.2%) 34 (4.7%) 6 (0.4%) 18 (1.9%) 23 (0.4%) 3 (0.9%) 5 (1.9%) 21 (1.6%) 11 (1.5%) 13 (0.9%) n.d. n.d. n.d. 5 (1.9%) 21 (1.6%) 7 (0.9%) 1 (0.07%) 5 (0.5%) 16 (0.3%) nil n.d. nil nil

(1998-2005) n.d.=not determined

Infection
The literature data about infections complicating VATS procedures are scarce, in spite of the increased application of VATS in recent years. The few studies available refer to heterogeneous patient populations undergoing various VATS procedures, that cannot be easily compared. Thus the incidence of infections in the VATS setting is not well defined. Rovera et al reviewed 346 consecutive VATS procedures and reported about the risk factors affecting the incidence of surgical infection.11 Patients were divided into two groups: 139 patients undergoing lung wedge resection and 207 patients undergoing intrathoracic biopsy (pleura or mediastinal mass). The following risk factors for infection were assessed: haemoglobin concentration, haematocrit, serum albumin concentration, blood lymphocyte count, length of preoperative stay, duration of surgery, blood transfusion, age, co-morbidity and specifically chronic obstructive pulmonary disease (COPD, measured as FEV1 <70% of expected). Prophylactic antibiotics were given to more than 90% of patients in both series. The endpoint measures were the occurrence of postoperative infection within 30 days (surgical site infection, pneumonia, empyema) and the final patient outcome. The authors found that one or more postoperative infections occurred in 17/346 patients (4.9%, all of whom had received antibiotic prophylaxis). In the VATS lung resection group the total infection rate was 6.5%, and in the VATS biopsy group it was similar (6.3%). The incidence of surgical site infection was similar in two groups (2.8% vs. 1.0%; p = NS) and the incidence of pneumonia was also similar (2.8% vs. 3.4%, p = NS). All postoperative infections (surgical site, pneumonia, empyema) were treated successfully with antibiotic or antifungal therapy and surgical drainage as indicated. No postoperative deaths occurred. The infection rate in the 50 VATS patients with COPD was significantly higher than in controls (14% vs 4.9%; p <0.01). COPD was the only univariate predictor of postoperative infection. Indeed FEV1 <70% of predicted was observed much more frequently in infected patients than in non-infected patients (58% vs. 25%; p <0.05) (Table 2). Table 2. Postoperative infections in 346 patients undergoing VATS procedures at the Center for thoracic surgery, University of Insubria (1996-2002).(11)

Postoperative Infections P Infections Infections infections in 139 value in 207 in all 346 lung biopsies patients wedge of resections mediastinal mass or pleura

Surgical site infection no. (%) Pneumonia no. (%) Empyema no. (%) Total no. of infections 7 (5.0%) NS 10 (4.8%) 17a (4.9%) Total no. of pts with infections a Five patients developed two infections (two patients: surgical site infection + pneumonia; two patients: pneumonia + empyema; one patient: surgical site infection + empyema). 4 (2.8%) 4 (2.8%) 1 (0.7%) 9 (6.5%) NS NS NS NS 2 (1.0%) 7 (3.4%) 4 (2.0%) 13 (6.3%) 6 (1.7%) 11 (3%) 5 (1.4%) 22a (6.3%)

Post-operative pain
The surgical incisions of open thoracic operations are among the most painful. They cause reduced lung function, impairment of respiratory effort, atelectasis and ultimately hypoxemia. Moreover, pain causes less patient mobilization and increases the incidence of deep venous thrombosis and of pulmonary embolism. Several traumatic factors are known to contribute to acute pain after thoracic surgery: tissue trauma and drains, rib fracture and injury to intercostal nerves (due to retractor or trocar). The prominent causes of chronic postoperative pain are intercostal neuroma and healing rib fracture. Essentially, in open thoracic surgery pain results from the spreading of retractors.12 It has been experienced worldwide that patients have less pain after VATS procedures than after open thoracotomies. Walker compared the requirement for narcotic pain drugs in 83 VATS major pulmonary resection versus 110 patients who underwent thoracotomy during the same period and found that the VATS group required significantly less morphine (p<0.001).(10) In a randomised prospective trial of lobectomy in 67 patients (47 by VATS and 23 by muscle-sparing thoracotomy) Giudicelli et al reported that postoperative pain was significantly less (p<0.02) after VATS procedures.13 Landreneau et al prospectively evaluated daily narcotic requirements, hospital stay and visual analogue pain scale in 165 patients after muscle-sparing thoracotomy compared with 178 patients after VATS.14 The latter patients had less pain and greater shoulder strength in the first 6 months postoperatively, but after 1 year the difference disappeared.14 The severity and duration of pain after thoracotomy obviously depends on the degree of rib spreading and on the site and modality of thoracotomy incision. Indeed, Nomori et al who compared VATS lobectomy vs. an anterior limited thoracotomy approach, reported that patients in the VATS group experienced much less pain in the first postoperative week, but the difference was not significant subsequently.15 Landreneau et al addressed the problem of chronic pain after pulmonary resection by thoracotomy versus VATS and found that within the first year after thoracotomy there was a significantly grater increase in the pain reported, as compared to VATS. After 1 year, however, they observed no statistically significant difference in either reported pain level or shoulder dysfunction (VATS 22%; thoracotomy 29%), or narcotic usage (VATS 6%; thoracotomy 16%).14

Port site recurrence of tumour

Dissemination of thoracic tumours as a result of biopsy or thoracotomy for resection with curative intent is rare. It has been described that during VATS neoplastic cells can disseminate and implant, causing tumour recurrence at the port site. This complication has been reported following VATS biopsy of lymph nodes and after pulmonary wedge resection. Yim et al16 and Jancovici et al3 noted an increased risk of local recurrence at the thoracoscopy port sites when a retrieval device (glove or bag) was not used. They further noted that recurrence occurred more likely in operations for selected tumour types, namely mesothelioma, metastatic sarcoma and melanoma, or when malignant pleural effusion was present. Mesothelioma is well known to invade thoracic incisions of all types, including thoracotomy, needle biopsy tracks and chest drain sites; similarly, malignant pleural effusions have been shown to implant and cause recurrence in the thoracic wall at the drain sites. In the surgical management of early stage tumours, however, dissemination and chest wall implant of thoracic malignancy resulting for manipulation is a phenomenon rarely observed. After VATS lobectomy McKenna et al in 1998 reported a very low incidence of this complication: 1/298 patients of his series (0.3%) developed chest wall recurrence in the utility thoracotomy incision.17 Subsequently the same authors reviewed their further experience with VATS lobectomy including 1,100 cases and confirmed that the recurrence rate in the incision was very low (0.57%).18 Also in the experience of the Memorial Sloan-Kettering Cancer Center Thoracic Service the prevalence of port site tumour recurrence was rare: 1/374 patients (0.26%).19

Conversion to thoracotomy
In 321 patients undergoing VATS procedures for various indications Krasna et al reported the need to convert to thoracotomy in 8% of cases.1 Most commonly the conversion to thoracotomy was deemed necessary because of oncological reasons, such as centrally located tumours requiring vascular control or sleeve resection, or unexpected T3 tumours attached to the chest wall, diaphragm or superior vena cava. Abnormal hilar nodes with granulomatous or metastatic disease adherent to the superior pulmonary vein may be better evaluated and more safely resected with thoracotomy. In the experience of Krasna et al about 30% of thoracotomy conversions were for non-oncological reasons, such as pleural adhesions.1 In the series of the Memorial Sloan-Kettering Cancer Center Thoracic Service conversion to open thoracotomy because VATS was not technically adequate occurred in 44/410 patients (11%).19 In our experience of 721 VATS procedures, conversion to thoracotomy was done in 16 patients (2.2%), for one of the following reasons: inability to localize precisely the lesion to biopsy or resect; bleeding that was difficult to control; extensive pleural adhesions.

Complications of specific VATS procedures

Lung wedge resection The most serious intraoperative complication during VATS lung wedge resection is major bleeding from the lung parenchyma or from injured intercostal vessels. Massive bleeding may require immediate conversion to an open procedure to achieve haemostasis. Yim et al reported

bleeding as the third most common complication (after persistent air leak and superficial wound infection), occurring in 6/1,337 (0.4%) patients, 4 of whom required extension of the mini-thoracotomy for control, or re-operation by VATS.2 In the series of Jancovici3 the incidence of bleeding requiring conversion to thoracotomy after VATS lung wedge resection was 12.4% (116/937 patients) and it was 8% (25/321 patients) in the series of Krasna.1 The most common non-fatal complication of VATS lung resection is prolonged air leak, especially after wedge resection for bullous emphysema and for spontaneous pneumothorax. Downey et al reported air leak longer than 5 days in 3.2% of patients,4 while in the study of Jancovici et al air leak for longer than 5 days was seen in 6.7% of patients.3 Kaiser et al observed air leakage longer than 7 days in 3.7% of their patients.9 Yim et al reviewed 1,337 VATS procedures and found that the incidence of prolonged (longer than 10 days) air leak was 1.6%.2 In general most authors agree that air leak longer than 7 days should occur in less than 5% of VATS procedures2-4, 9, 18. Postoperative infection rate after VATS lung wedge resection is low, as discussed above. In our institutions experience after VATS wedge resection the wound infection rate was 2.8%, the incidence of pneumonia was 2.8% and that of empyema was 0.7%.11 When performing VATS wedge resection of small lung nodules, the difficulty in localizing exactly the nodule within the lung parenchyma may bring about a number of undesirable problems. The first and most frustrating is the inability to identify the nodule, which forces conversion to thoracotomy. Another complication is an incomplete resection of the nodule, the deepest part of which may be left behind, resected only in part. 20 There are many different techniques for the localization of small lung nodules to be resected by VATS: thoracoscopic inspection and probe-palpation, ultrasonography, hook-wire, finger palpation, methylene blue injection and lipiodol injection. In our experience the nodule identification by finger palpation is feasible in the vast majority (94%) of cases.20 Under guidance by the CT-scan images, one of the mini-access incisions is done at the chest wall projection of the nodule. If the lung nodule is superficial, it can be recognized at inspection of the pleural cavity; moreover, visceral pleural retraction or fibrous stigmata are often indirect signs of an underlying deep lung nodule, that can be exactly localized by finger palpation. With the finger palpation technique it is possible to localize most nodules, notably with some exceptions: nodules located close to the hilum or in the very central portion of a lobe (>2.5 cm distant from any visceral pleural surface) and nodules with non-solid consistency (pure bronchiolo-alveolar carcinoma)21, 22. In our institute from 1997 to 2005 we recorded 121 consecutive patients undergoing VATS wedge resection of lung nodules with diameter <3 cm. In all patients it was possible the mini-invasive approach and the localization by finger palpation. 90 of 121 lung nodules had a diameter <1,5 cm and were resected by VATS, 93% of which were localized by finger palpation. Very small nodules (diameter <0.7 cm; n=9) resulted to be benign lesion (89%); in the group of the small nodules (diameter from 0.7 cm to 1.5 cm; n=81) the 53% resulted to be lung cancer, primary or metastatic. In 6 patients (7%) was necessary the conversion to a mini-thoracotomy to performed wedge resection. In two cases for difficult identification of the nodule, in 3 cases for pleural adhesions and in one case for the position of the nodule (near the aorta). Only in one case there was an histological evidence of malignant infiltration of the resections limits. 22 Mortality after VATS lung wedge resection is very low in all series, ranging 0%-0.5%. The few cases of death reported were due to: massive bleeding, pulmonary embolism, septic shock,

ventricular arrhythmia and respiratory insufficiency 3, 11. We recently reviewed our institutional series of 254 lung wedge resections by VATS, performed in the years 1998-2005 and did not record any perioperative deaths; the complications observed are listed in Table 3. Table 3. Postoperative complications after 254 VATS lung wedge resection performed at the Center for thoracic surgery, University of Insubria (1998-2005). Complication Incidence Air leak longer than 7 days 3% Bleeding, with blood transfusion 0.8% Pneumonia 4% Wound infection 0.8% Conversion to thoracotomy 2% Perioperative mortality 0%

Lobectomy Major lung anatomic resections (lobectomy and pneumonectomy) are undoubtedly the most difficult operations to perform thoracoscopically. Several years after the first resections the initial concerns regarding the potential for intraoperative accidents have not subsided and have limited the development and acceptance of VATS anatomic lung resections. Many surgeons have done a few thoracoscopic resections, but only in few centers these operations are done routinely (Table 4) 18,23-27. Table 4. Morbidity and mortality in different series of VATS major lung resections

Conversion All Cancer Overall Overall to Authors Patient Patient postoperativ postoperativ thoracotom s s e mortality e morbidity y
Lewis(23) Yim(24) Kaseda(25) Walker(26) Roviaro(27) 100 266 145 179 294 92 n.d. 103 179 n.d. 1015 nil 0.4% 0.8% 1.8% 0.4% 0.8% 12% n.d. n.d. n.d 8.8% 15% nil 19.5% 11.7% 11.2% 22.8% 2.5%

McKenna(18 1100 ) n.d.=not determined

Among the asserted advantages of VATS major lung resections is the reduction of postoperative pain and of hospital stay; the cosmetic result is also improved. The increased technical difficulty and concern in the longer VATS procedure and the risk of intraoperative accidents, however, have limited the acceptance of VATS lobectomy by the surgical community. The oncological adequacy of videothoracoscopic major pulmonary resections is still controversial, although most authors admit to use the same oncological criteria employed for conventional surgery. Several reports of VATS lobectomy for cancer show very good 5-year survival. The results of some series are even superior to those obtained with conventional open surgery, but this may be due to patient selection, in the absence of randomization, as will be discussed below. Moreover, Roviaro et al have suggested that the superior outcome could be due to less immunosuppression from the lesser trauma of VATS, with consequent immunological advantage. 27 In their extensive experience of 301 patients undergoing VATS major lung resection, Roviaro et al reported the need to convert to open thoracotomy in 22.8% of cases, because of oncological problems or technical reasons. These authors reported no intraoperative deaths in their series and had 20 patients (8.8%) with postoperative complications that extended the hospital stay. 27 In the literature the overall postoperative morbidity reported for VATS major lung resection ranges 8.8%-15% 18, 23, 27 and the conversion to a standard thoracotomy ranges 0%-22.8% (table 4) 18,23-27. The predominant causes of conversion to thoracotomy are: bleeding, extensive adhesions, dense hilar adenopathy, local extension of disease and intraoperative cardiac complications. The most prevalent intraoperative complications were: injury to a blood vessel, stapling failure, lung injury and nerve injury.28 One of the most dreaded complications of VATS lobectomy is massive bleeding from major pulmonary vessels. Craig and Walker29 and Yim30 reported mechanical failure of vascular staplers during VATS lobectomy, that resulted in massive bleeding; this was controlled by pressing on the bleeder and by immediate conversion to thoracotomy. Yim et al, in their review, reported an overall surgical mortality ranging from 0% to 2% for VATS lobectomy, that compares favourably with the conventional technique.31 In evaluating the results of VATS major lung resections, it is important to point out that these VATS procedures are usually done in highly selected patients, mainly presenting with peripheral lung lesions (most of which are adenocarcinomas) of small size, in early stage of disease; in the 1,100-case series of McKenna et al, 73% of VATS anatomic lung resections were performed in stage I lung cancer.(18) For comparison, it should be noted that the proportion of open thoracotomy lobectomy performed for stage I lung cancer in unselected series is only about 40%. Mediastinal biopsy VATS biopsy of mediastinal masses (neurogenic tumours, lymph nodes, anterior mediastinal tumours) provides a convenient way, definitely less traumatic than biopsy by conventional thoracotomy or sternotomy, to obtain an adequate tissue sample for a definitive histological diagnosis. This is especially important if previous diagnostic exams, including fine needle aspiration, were unyielding. Moreover, some selected mediastinal masses, under favourable circumstances, can also be resected and definitively treated by VATS.27 The local complications of mediastinal biopsy are rare. They include the usual complications of VATS procedures, namely prolonged air leak and bleeding; the latter may be severe if superior vena

cava hypertension is present. In addition, a rare complication of mediastinal biopsy is chylothorax, caused by laceration of the main thoracic duct. The global incidence of chylothoraxis about 1% in all thoracic procedures. In approximately 50% of cases the treatment of chylothorax is conservative with pleural drainage, total parenteral nutrition and mediumchain triglyceride diet. High flow leak requires surgical closure (ligature or clips) of the thoracic duct, that can sometimes be performed by VATS.32 The review of personal experience with 46 mediastinal biopsies performed by VATS showed only one case of postoperative complication (pneumonia); only in one case it was necessary to convert to thoracotomy to control bleeding from the lymph node biopsy site in a patient with superior vena cava syndrome. Pleural biopsy The incidence of complications when samples of pleura are taken for biopsy by VATS is very low. The commonest complication is bleeding from an intercostal vessel or from the chest wall; the need to convert to open thoracotomy is very rare. Another important potential complication of VATS pleural biopsy is the dissemination of malignant disease in the chest wall incision. Especially pleural mesothelioma is well recognized to invade thoracic incisions of all types; accordingly, local radiotherapy on the chest wall access site is an established routine. In addition, many other histological types of malignant tumours present in pleural effusions are able to metastasise in the mini-thoracotomy and port site chest wall accesses.4 The incidence of surgical infection after VATS pleural biopsy has been studied by Rovera et al in 183 patients over 346 consecutive video-assisted thoracoscopic procedures.11 In 4.8% of patients a surgical infection developed (wound infection: 1%; pneumonia: 3.4%; empyema: 2%). Pleurectomy Effective treatment of pneumothorax can be achieved by several pleurodesis techniques, one of which is pleurectomy. Pleurectomy in fact is associated with the lowest rates of pneumothorax recurrence (<5%).33 In some patients the pleurodesis procedure is limited to pleural scarification by simple cauterization because of ongoing high-dosage anticoagulant treatment. Prolonged air leak, however, is not infrequent after VATS pleurectomy for pneumothorax: it is reported in 20% of cases.34 Indeed many patients undergoing pleurectomy are emphysematous, with fragile lung parenchyma. Other complications after VATS pleurectomy are uncommon; they include bleeding, persistent chest wall pain, surgical site infection and recurrence of pneumothorax. Some paresthesia at the skin level overlying the pleurectomy is common. Moreover, Roviaro et al reported a 4% incidence of haemothorax; in those patients thoracoscopy for haemostasis was redone.27 In spite of a blood loss of > 500 ml, the point of bleeding could not be located in any of those cases and the operation consisted of the removal of blood clots to clear the pleural space and to allow for pleurodesis to occur.27 In our series of 99 VATS pleurectomies for pleurodesis done in the years 1998-2005, the complications were: air leak for longer than 7 days: 13%; bleeding with blood transfusion: 3%; pneumonia: 2%; wound infection: 1%; conversion to thoracotomy: 1%. In 4/99 patients (4 %) we observed recurrence of pneumothorax; these recurrences were treated either by simple tube drainage or by talc pleurodesis.

References
1. Krasna MJ, Deshmukh S, McLaughlin JS. Complication of thoracoscopy. Ann Thorac Surg 1996;61:1066-9. 2. Yim APC, Liu HP. Complication and failures of video-assisted thoracic surgery: experience from two centers in Asia. Ann Thorac Surg 1996;61:538 -41. 3. Jancovici R, Lang-Lazdunski L, Pons F, Cador L, Dujon A, Dahan M, et al. Complication of video-assisted thoracic surgery: a five year experience. Ann Thorac Surg 1996;61:533-7. 4. Downey RJ. Complication after video-assisted thoracic surgery. Chest Surg Clin N Am 1998;8:907-915. 5. Inderbitzi RG, Grillet MP. Risk and hazard of videothoracoscopic surgery: a collective review. Eur J Cardiothorac Surg 1996;10:483-489. 6. Kaiser LR. Video-assisted thoracic surgery: current state of art. Ann Surg 1994;220:720734 7. Hazelrigg SR, Magee MJ, Cettindag IB. Video-assisted thoracic surgery for diagnosis of the solitary lung nodule. Chest Surg Clin N Am 1998;8:763-74. 8. Adebonojo SA. How prolonged is prolonged air leak? Ann Thorac Surg 1995;59:54950. 9. Kaiser LR, Bavaria JE. Complication of thoracoscopy. Ann Thorac Surg 1993;56:796-8. 10. Walker WS, Pugh GC, Craig SR, Carnochan FM. Continued experience with thoracoscopic major pulmonary resection. Int Surg 1996;81:255. 11. Rovera F, Imperatori A, Militello P, Morri A, Antonini C, Dionigi G et al. Infections in 346 consecutive video-assisted thoracoscopic procedures. Surgical Infection 2003;4:4551. 12. Hazelrigg SR, Cetindang IB, Fullerton J. Acute and chronic pain syndromes after thoracic surgery. Surg Clin N Am 2002;82:849-65 13. Giudicelli R, Thomas P, Lonjon T, Morati N, Ottomani R, Fuentes P et al. Videoassisted minithoracotomy versus muscle-sparing thoracotomy for performing lobectomy. Ann Thorac Surg 1994;58:712-7.

14. Landreneau RJ, Mack MJ, Hazelrigg SR, Naunheim K, Dowling RD, Ritter P, et al. Prevalence of chronic pain after pulmonary resection by thoracotomy or video-assisted thoracic surgery. J Cardiovasc Surg 1995;109:1255-6. 15. Nomori H, Horio H, Fuyuno G, Kobayashi R. Non-serratus-sparing antero-axillary thoracotomy with disconnection of anterior rib cartilage. Improvement in postoperative pulmonary function and pain in comparison to posterolateral thoracotomy. Chest 1997;111:572-6. 16. Yim APC. Port-side recurrence following video-assisted thoracoscopic surgery. Surg Endosc 1995;9:1133-35. 17. McKenna RL Jr, Fischel RJ, Wolf R, Wurning P. Video-assisted thoracic surgery (VATS) lobectomy for bronchogenic cancer. Semin Thorac Cardiovasc Surg 1998;10:321-5. 18. McKenna RJ, Houck W, Beeman Fuller C. Video-assisted thoracic surgery lobectomy: experience with 1100 cases. Ann Thorac Surg 2006;81:421-6. 19. Parekh K, Rusch V, Bains M, Downey R, Ginsberg R. VATS port site recurrence: a technique dependent problem. Ann Surg Oncology 2001;8:175-178. 20. Dominioni L, Imperatori A, Rovera F, Dionigi G, Vassallo F. The role of videothoracoscopy in the early diagnosis of lung cancer. Osp Ital Chir 2002;8:248-53. 21. Dominioni L. Chirurgia Toracica. In: R. Dionigi, editor. Chirurgia. Masson III Ed; 2002. p 857-936. 22. Rotolo N, Imperatori A, Dominioni L. Identificazione e trattamento dei piccoli noduli polmonari. Proceedings of IV Update in Thoracic Surgery 2-3 December 2005; Varese. 23. Lewis RJ, Caccavale RJ, Sisler GE, Bocage JP, Mackenzie JW. One hundred videoassisted thoracic surgical simultaneously stapled lobectomies without rib spreading. Ann Thorac Surg 1997;63:1415-1422. 24. Yim APC, Liu HP, Izzat MB, et al. Thoracoscopic major lung resection an Asian perspective. Semin Thorac Cardiovasc Surg 1998;10:326-331. 25. Kaseda S, Aoki T, Hangai N. Video-assisted thoracic surgery (VATS) lobectomy: the Japanese experience. Semin Thorac Cardiovasc Surg 1998;10:300-4

26. Walker WS, Codispoti M, Soon SY, Stamenkovic S, Carnochan F, Pugh G. Long-term outcome following VATS lobectomy for non-small cell bronchogenic carcinoma. Eur J Cardiothorac Surg 2003;23:397-402. 27. Roviaro GC, Varoli F, Vergani C, Maciocco M. State of the art in thoracoscopic surgery. A personal experience of 2000 videothoracoscopic procedures and an overview of the literature. Surg Endosc 2002;16:881-892. 28. Mc Kenna RJ. New approach to the minimally invasive treatment of lung cancer. Cancer J 2005;11:73-6. 29. Craig SR, Walker WS. Potential complications of vascular stapling in thoracoscopic major pulmonary resection. Ann Thorac Surg 1995;59:736-7. 30. Yim APC, Ho JKS. Malfunctioning of vascular staple cutter during thoracoscopic lobectomy. J Thorac Cardiovasc Surg 1995;109:1252. 31. Yim APC. VATS major pulmonary resection revisited-controversies, technique and results. Ann Thorac Surg 2002;74:615-23. 32. Fahimi H, Van Swieten HA. Current management of postoperative chylothorax. Ann Thorac Surg 2001;71:448-451. 33. Horio H, Nomori H, Kubayasmi R, Naruke T, Suemasu K. Impact of additional pleurodesis in video-assisted thoracoscopic bullectomy for primary spontaneous pneumothorax. Surg Endosc, 2002;16:630-4. 34. Bertrand Pl, Regard SF, Spaggiari L, Levi SF, Magdeleint P, Guibert L, Levasseur P. Immediate and long-term results after surgical treatment of primary spontaneous pneumothorax by VATS. Ann Thorac Surg, 1996;61:1641-5.