Large chest wall resections are essential for a variety of

Titanium Plates and Dualmesh: A Modern Combination for Reconstructing Very Large Chest Wall Defects Jean Philippe Berthet, MD, Ludovic Canaud, MD, Thomas D Annoville, MD, Pierre Alric, MD, PhD, and Charles-Henri Marty-Ane, MD, PhD Department of Thoracic Surgery, Hospital Arnaud de Villeneuve, Montpellier, France GENERAL THORACIC Background. The reconstruction of large full-thickness chest wall defects after resection of T3/T4 non-small cell lung carcinomas or primary chest wall tumors presents a technical challenge for thoracic surgeons and plays a central role in determining postoperative morbidity. The objective is to evaluate our results in chest wall reconstruction using a combination of expanded polytetrafluoroethylene (eptfe) mesh and titanium plates. Methods. Since 2006, 19 patients underwent reconstruction for wide chest wall defects using a combination of eptfe mesh and titanium plates. The chest wall reconstruction was achieved by using a layer of 2-mm thickness eptfe shaped to match the chest wall defect and sewed under maximum tension. The eptfe is placed close to the lung and fixed onto the bony framework and onto the titanium plate, which is inserted on the ribs. Results. Seventeen patients underwent a complete R0 resection with the removal of 3 to 9 ribs (mean, 4.8 ribs), including the sternum in 7 cases. Reconstruction required 1 to 4 horizontal titanium bars (mean, 1.7 bars). In 1 patient, a vertical titanium device was implanted for a large posterolateral defect. There were 2 cases of infection, which required explantation of the osteosynthesis system in 1 patient. One patient had partial skin necrosis that required prompt debridement. One patient had a major complication in the form of respiratory failure. Conclusions. Our experience and initial results show that titanium rib osteosynthesis in combination with Dualmesh can easily and safely be used in a one-stage procedure for major chest wall defects. (Ann Thorac Surg 2011;91:1709 16) 2011 by The Society of Thoracic Surgeons Large chest wall resections are essential for a variety of conditions, such as primary or secondary chest wall tumors and T3/T4 non-small cell lung cancer (NSCLC). The classical management of chest wall lesions involves first, large disease-free margins with or without en-bloc lung resection and skin or soft tissue resection; and second, skeletal reconstruction, with or without soft tissue coverage. It is this second aspect that we address here. When the defect is large, complications after chest wall reconstruction are common and range between 46% and 69% [1, 2], a rate that includes a 27% rate of respiratory morbidity [1]. The primary goals of skeletal reconstruction are to prevent paradoxical chest wall movement resulting in respiratory failure, and to protect the lungs, heart, and major vessels from injury and infection. As a secondary objective, skeletal reconstruction aims to prevent the incarceration of the scapula (posterior defects) and to maintain cosmetic integrity. Numerous synthetic materials such as polytetrafluoroethylene (PTFE), polypropylene with or without methylmethacrylat, and autogenous materials (such as muscle flaps and bone grafts) have been applied in chest wall Accepted for publication Feb 4, 2011. Address correspondence to Dr Berthet, MD, Department of Thoracic Surgery, Hospital Arnaud de Villeneuve, 191 avenue Doyen Gaston Giraud, 34090 Montpellier, France; e-mail: jeanphilippe.berthet@gmail. com or jp-berthet@chu-montpellier.fr. reconstruction. The material is usually chosen on the basis of the location and extent of the defects, together with the surgeon s experience. Synthetic material implantation is most often associated with a muscle or musculocutaneous flap in full-thickness resections. The following is a review of our experience in the combined use of titanium plates and expanded PTFE (eptfe [Dualmesh 2 mm; W.L. Gore & Assoc, Flagstaff, AZ]) in chest wall reconstructions after very large chest wall resections. The titanium implants were STRATOS (Strasbourg thoracic osteosynthesis system [MedXpert GmbH, Heitersheim, Germany]) for horizontal osteosynthesis (Fig 1), and the vertical expandable prosthetic titanium rib (VEPTR) system (Synthes Spine Company, West Chester, PA) for vertical osteosynthesis. Material and Methods From October 2006 to January 2010, 19 patients underwent a large chest wall resection. We employed the combination of Dualmesh and titanium plates as our procedure of choice for large thoracic wall defect reconstructions. This study was approved by our Ethics Committee, and individual consent was obtained in each cases. In all patients, before resection, we systematically performed computed tomography scans of the thorax, 2011 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2011.02.014

GENERAL THORACIC 1710 BERTHET ET AL Ann Thorac Surg INNOVATIVE CHEST WALL RECONSTRUCTION 2011;91:1709 16 Fig 1. Strasbourg thoracic osteosynthesis system (STRATOS). (A) Rib clips and connecting bars. (B) Reinforced joints between rib clips and bars. (C, D) Different methods to implant the STRATOS device with or without bridging the defect. abdomen, and brain; and none of these patients had extrathoracic metastases at the time of evaluation. For posterior tumors, magnetic resonance imaging of the entire spinal cord was performed to rule out any abnormalities. To plan the therapeutic approach for T3-T4 NSCLC or chest wall tumor, a multidisciplinary approach involving respiratory oncology physicians, radiologists, radiotherapists, thoracic surgeons, spinal surgeons, and plastic surgeons was employed, and the multidisciplinary committee needed to approve the treatment plan. The topographic location of the disease was defined as follows: anterior chest wall defects were defined as being located between the sternum and the anterior axillary line, lateral defects were those between the anterior and posterior axillary lines, and posterior defects were those between the spine and the posterior axillary line. The surgical approach depended on both the chest wall parts involved and the associated lung requiring resection: for posterior resections, thoracotomy incisions were made between the medial border of the scapula and the midline spine proximally and gently curved anteriorly in an inverted J shape. The level depended on the associated lung or vertebral resection required. Anterosuperior tumors with posterior extensions were resected through a combined anterior and posterior approach. Sternal tumors were resected through a direct anterior approach. En bloc resection was performed in all cases by the thoracic surgeon, with an orthopedic surgeon as required. The disease-free resection margins were established during operation on the basis of frozen tissue section analysis. Large chest wall defects were defined as any resection involving more than three ribs or a combined resection of three ribs (at least) and the sternum. s of fewer than three ribs and those located in the posterior chest wall beneath the scapula were not included in this study. In these cases, ribs approximation and soft-tissue coverage was the standard management for the thoracic defect. Chest wall resection included the ribs, parietal muscles, and when required, one or more of the following: sternum, clavicles, vertebral body (Fig 2), subclavian vessels, superior vena cava, lung (wedge, lobectomy, or pneumonectomy), diaphragm, pericardium, soft tissues, and skin. The chest wall reconstruction used the combination of titanium plates and Dualmesh. The first step in chest wall reconstruction was achieved by using a layer of 2-mm thickness eptfe shaped to match the chest wall defect. The Dualmesh is placed close to the lung and fixed onto the bony framework and onto the titanium plate, which is inserted on the ribs. The Dualmesh, 2-mm thick, was anchored to the inner corners and edges of the defect with multiple pericostal nonabsorbable sutures, to sepa- Fig 2. En-bloc resection of the right upper lobe, ribs, and vertebral bodies.

Ann Thorac Surg BERTHET ET AL 2011;91:1709 16 INNOVATIVE CHEST WALL RECONSTRUCTION rate the remaining lung or mediastinum from the superficial tissues. It is usually sewn around the defect under maximum tension, to reestablish the skeletal continuity of the chest (Fig 3). To reestablish the rigidity of the thoracic wall, we used the STRATOS to bridge the defect. It consists of titanium rib clips available in different angles and of connecting bars available in different lengths. After the ribs were isolated lateral to the resection margin, the rib clips and connecting bars were molded to match the shape of the defect. One complete implant consists of two rib clips and one connecting bar (Fig 1). Stability was obtained without rebuilding each rib and required one or more implants, depending on the location and size of the defect and the surgeon s experience. In cases involving associated sternal resection or when the disease-free ribs were too short lateral to the disease margins, the STRATOS bars were inserted onto the opposite ribs (Fig 4). In cases involving posterior defects, we used angled rib clips and performed horizontal osteosynthesis without taking into account the level of posterior and anterior segments of disease-free ribs (Fig 1). In cases involving very large posterior defects without the opportunity to lock the STRATOS clip posteriorly, we performed a vertical rib osteosynthesis with a VEPTR device. The VEPTR device is an expandable metal rod that is curved to fit the back of the chest. It is placed in a vertical position between the disease-free ribs (Fig 5). We hung the eptfe from the titanium bars with multiple nonabsorbable sutures (1.0 or 0.0) anchoring the patch to the plates. Subsequently, absorbable sutures were used for soft tissue reconstruction, and in cases of fullthickness resection, flaps of skin, muscle (pectoralis major, latissimus dorsi), or the greater omentum were used. The surgical and oncologic data were obtained from operation notes, including the indication, the type of procedure, the location, and the extent of the chest wall 1711 Fig 4. Postoperative computed tomography scan: three-dimensional reconstruction after implantation of Strasbourg thoracic osteosynthesis system (STRATOS)/Dualmesh (3 bars). lesion; pathology reports, including the histologic diagnosis, the size of the chest wall defect, and the number of resected ribs; and postoperative care reports of perioperative complications. Data concerning aesthetic and functional results, mortality, and hospital stay were collected. Pulmonary function tests were routinely done preoperatively to assess the patient s ability to withstand operation. A postoperative function test was performed at postoperative month 6. A restriction was defined as a GENERAL THORACIC Fig 3. Dualmesh is sewn around the defect under maximum tension. Fig 5. Postoperative computed tomography scan: three-dimensional reconstruction after implantation of vertical expandable prosthetic titanium rib (VEPTR)/Dualmesh.

GENERAL THORACIC 1712 BERTHET ET AL Ann Thorac Surg INNOVATIVE CHEST WALL RECONSTRUCTION 2011;91:1709 16 recorded postoperative total lung capacity less than 85% of the preoperative value. Clinical (TNM) and pathologic (ptnm) stages were assigned in agreement with the new International System for Staging Lung Cancer, as developed by the American Joint Committee on Cancer and the Union Internationale Contre le Cancer. Survival was calculated from the date of surgery until death or the date of last follow-up without recurrence. No patient was lost in the follow-up. Survival was estimated by the Kaplan- Meier product limit method. Results Between 2006 and 2010, 19 patients (10 women) with a median age of 58 years (range, 34 to 72 years) with chest wall invasion from either primary or secondary chest wall tumors or T3/T4 NSCLC were treated at our institution. Histology, neoadjuvant therapy, resection quality, and tumor classification are reported in Table 1. There were 11 NSCLC patients, all of whom underwent neoadjuvant chemotherapy and preoperative radiation therapy. The tumor was classified as T4 in 8 cases, thereby meaning vertebral body, subclavian vessel, or superior vena cava involvement. We found no mediastinal lymph node involvement. The other patients presented with five sarcomas, a localized mesothelioma, and the recurrence of one thyroid carcinoma, one breast tumor, and one deep basal cell carcinoma. The resection status was complete (R0) in 89.5%. Bone s The type of chest wall resection is summarized in Table 2. The thoracic defect was located posteriorly in 1 case, in the posterolateral region in 6 cases, the anterolateral region in 5, the anterior part in 5, and the defect was total in 2 cases. We performed chest wall resections, with a mean number of 4.8 ribs resected (range, 3 to 9 ribs), and additional sternum resections in 7 cases. The mean defect area was 198 cm 2 (range, 95 to 400 cm 2 ). Vertebral bodies were involved in 5 patients. Both anterior and posterior osteosynthesis of the vertebral column was required in 4 cases. The mean volume of the en-bloc resection was 1,110 cm 3 (range, 450 to 2,800 cm 3 ) and was calculated from the pathologist s data to give an approximation of the tumor size. Lung and Vessel s was extended to the lung in all but 1 patient. We performed one right pneumonectomy, 10 upper lobectomies (including one completion pneumonectomy), and seven wedge resections. Subclavian vessel management in superior sulcus tumors needed reconstruction using a 10-mm ringed eptfe in 2 patients (two veins and one artery) and a resection-anastomosis in another (both vessels), as described in Table 2. The superior vena cava was replaced in patient 17, who had mediastinal and cervical extension of a sternal sarcoma. Table 1. Pathologic Status Patient No. Histology Neoadjuvant Therapy Status TNM 1 Malignant CR R0 mesothelioma 2 NSCLC CR R0 T3N0M0 3 NSCLC CR R1 T4N1M0 4 NSCLC CR R0 T3N0M0 5 NSCLC CR R0 T4N0M0 6 Chondrosarcoma C R0 7 Thyroid carcinoma RT R0 8 NSCLC CR R0 T4N0M0 9 NSCLC CR R0 T4N0M0 10 Chondrosarcoma CR R0 T4N0M0 NSCLC 11 NSCLC CR R0 T3N0M0 12 Sarcoma R0 13 NSCLC CR R0 T4N0M0 14 Breast Tumor C R0 15 Sarcoma CR R0 16 NSCLC CR R0 T4N1M0 17 Sarcoma R1 18 Recurrent basal cell R0 epithelioma 19 NSCLC CR R0 T4N0M0 C chemotherapy; CR chemoradiation therapy; NSCLC non-small cell lung cancer; RT radiation therapy. Chest Wall Reconstruction Dualmesh was used in each case, 2 cm greater than the defect area (Table 2). As a horizontal rib osteosynthesis device to stiffen the prosthetic patch, only one titanium implant (STRATOS) was used in 10 cases, 2 bars in 5 cases, 3 bars in 2 cases, and 4 bars in 1 case. A mean of 1.76 titanium plates (range 1 to 4 plates) were implanted. In our experience, one plate is sufficient for the reconstruction of 2.75 ribs. In patient 1, the upper bars crossed the resected portion of the sternum, and the lower bar was positioned so as to bypass the sternum because there was no way to attach it to the left side of the sternum. The two bars implanted in patient 18 were also distally implanted onto the contralateral ribs without enlarging the thoracic approach, thanks to the use of the molded plates. For vertical rib osteosynthesis, only one VEPTR bar was implanted into the final patient in the series (Patient 19). In this case, there was no posterior segment of a healthy rib in the five resected levels sufficient to lock a STRATOS implant. We performed a vertical osteosynthesis that locked on to the middle arch of the second and eighth rib. A muscular flap was added in 9 cases by rotating either the pectoralis major or the latissimus dorsi. A flap of the greater omentum was used in 2 patients because of insufficient soft tissue cover, and in a third patient as a result of local infection. A concurrent reconstruction of the diaphragm and the pericardium was performed in 4 cases by using Vicryl (Ethicon, Somerville, NJ) or 1-mm thickness PTFE mesh (Table 3).

Ann Thorac Surg BERTHET ET AL 2011;91:1709 16 INNOVATIVE CHEST WALL RECONSTRUCTION Table 2. Type of Chest Wall Patient No. Topography of Defect Lung No. of Resected Ribs Sternum Muscle Skin Others Volume (cm 3 ) 1 Ant LUL 3 Subtotal SCA, SCV 520 130 2 Ant/Lat RUL 4 PM SCV 910 130 3 Ant/Lat P 5 Partial SCA, SCV 1,050 150 4 Post/Lat RUL 5 572 143 5 Post WR 6 PV Yes VB 575 105 6 Ant WR 4 Subtotal PM Yes 572 115 7 Post/Lat WR 3 D 585 120 8 Ant/Lat/Post RUL 4 Sc, PM VB 1,462 210 9 Post/Lat RUL 4 VB 1,638 350 10 Ant/Lat RUL 4 LD Yes Surrenalectomy 975 300 11 Ant/Lat LUL 3 Sc 450 95 12 Ant/Lat/Post 9 LD 2,800 280 13 Post/Lat RUL 4 Sc VB 964 108 14 Ant WR 4 Partial D, PM Yes Pericardial 1,575 225 15 Ant WR 7 Total Yes Pericardial 1,000 200 16 Post/Lat CP 6 LD Esophageal 1,008 126 Musculosae 17 Ant WR 6 Partial PM Yes SVC 1,650 400 18 Ant/Lat WR 6 Partial PM, D Yes Pericardial 1,800 360 19 Post/Lat RUL 5 Sc VB 990 220 Minimum Maximum Mean 3 9 4.84 450 2,800 1,110 1713 Area (cm 2 ) 95 400 198 GENERAL THORACIC Ant anterior; CP completion pneumonectomy; D diaphragm; Lat lateral; LD latissimus dorsi; LUL left upper lobectomy; P pneumonectomy; PM pectoralis major; Post posterior; PV paravertebral muscles; RUL right upper lobectomy; Sc scalen muscle; SCA subclavian artery; SCV subclavian vein; SVC superior vena cava; VB vertebral body; WR wedge resection. Complications Seventeen patients experienced no major complications, and we observed no perioperative complications. We also found no spinal cord injury; however, two vertebral body resections were associated with a temporary dural leak. Graft patency was excellent (with only one episode of subclavian vein graft occlusion at 3 months). Patients 3 and 4 experienced a delayed rupture of the titanium bar (STRATOS) at the joint. In patient 4, the ruptured bar was removed at 6 months to avoid it migrating into the soft tissues or mediastinum (Table 3). Patient 18 had partial skin necrosis that required prompt debridement and primary closure. Follow-up was then uneventful. In that patient, the initial coverage of the thoracic osteosynthesis material was performed using a large skin flap because the local muscle flaps were no longer fit for use. Patients 3 and 17 had infections in the osteosynthesis system. In patient 3, temporary removal of the STRATOS bars and the Dualmesh was required at postoperative day 10. After intensive cleaning of the surgical area, a STRATOS bar was reinserted together with a Vicryl mesh and patch of the greater omentum to stabilize the thoracic wall. There was 1 early death resulting from a massive pulmonary embolism (patient 7). One patient died after 68 days in the intensive care unit because of a late postoperative bacterial pneumonia, followed by respiratory failure. After excluding the seventh and third patients (early death on postoperative day 7, and an intensive care unit stay of more than 30 days), median in-hospital stay was 10.8 days (range, 7 to 18), including a median stay of 3.3 days in intensive care. Pulmonary function testing was performed in 14 of the 17 surviving patients. Pulmonary function revealed no significant difference between preoperative and postoperative forced expiratory volume in 1 second measurements in patients with lobectomy or wedge resections. There was no restriction pattern. No chest wall deformity was observed in any patient. Survival without recurrent disease is shown in Figure 6. Comment Surgery represents the cornerstone of treatment in primary chest wall tumor and NSCLC with parietal involvement but no lymph node involvement (N0). The main objective of chest wall resection is to perform a radical R0 en-bloc resection. The main basis of chest wall reconstruction is summarized in the introduction according to the literature [3 5]. The importance of radical surgery is well known and has been demonstrated by many authors [1 3, 5]. Since Downey and colleagues [6] stated that an incomplete resection, even if it leaves only microscopic disease, will not cure the patient, the essential goal of surgery should be an oncologic en-bloc resection of the whole tumor and adjacent structures to leave disease-

GENERAL THORACIC 1714 BERTHET ET AL Ann Thorac Surg INNOVATIVE CHEST WALL RECONSTRUCTION 2011;91:1709 16 Table 3. Type of Reconstruction Patient No. Mesh No. STRATOS Barres No. VEPTR Muscle Flap Material Infection Material Rupture Material Removal 1 DM 2 PM mobilization 2 DM 1 PM mobilization 3 DM 1 PM mobilization GO Yes Yes Yes 4 DM 1 Yes Yes 5 DM 1 6 DM 1 PM mobilization 7 DM 1 8 DM 1 PM mobilization 9 DM 2 10 DM V 2 GO 11 DM 1 12 DM 4 LD mobilization 13 DM 2 14 DM 1 15 DM V 3 PM mobilization 16 GT V 3 17 DM 1 PM mobilization Yes 18 DM V 2 Skin advancement 19 DM 1 LD mobilization DM Dualmesh 2-mm thickness; GO great omentum; LD latissimus dorsi; PM pectoralis major; STRATOS Strasbourg osteosynthesis system; V Vicryl mesh; VEPTR vertical expandable prosthetic titanium rib. free margins. It is also well established that survival after major chest wall resection for NSCLC also depends upon the extent of lymph node involvement [7]. For primary chest wall tumor, tumor grade has been reported as the most important prognostic factor, together with adequate surgical resection [8, 9]. In our series, we systematically checked the resection margins by examining the soft tissues intraoperatively. In bone, we decided to respect the 3-cm limit of macroscopic healthy tissue anteriorly and posteriorly wherever possible. The intercostal spaces located below and above macroscopic limits of the disease were always removed. For posterior tumors, disease-free margins of 3 cm cannot be performed and lead to the resection of the transverse process and the vertebral insertion of the ribs. A Fig 6. Survival without recurrent disease (Kaplan-Meier product limit method). larger vertebral resection was carried out in cases of RMN with obvious signs of invasion of the vertebral column. This strategy allowed us to obtain a R0 resection rate of 89.4% in this series of very large chest wall tumors (mean volume 1,110 cm 3, mean area 198 cm 2, mean number of resected ribs 4.84). Weyant and associates [1] presented the results of chest wall resection and reconstruction from a large series. The median defect size was 80 cm 2 (range, 2.7 to 1,200 cm 2 ), and the median number of resected ribs was 3 (range, 1 to 8 ribs). Postoperatively, 3.8% of patients died and respiratory failure occurred in 3.1%. According to a multivariate analysis, the size of the chest wall defect was the most significant predictor of complications. Several reports [10] suggest that stabilizing the chest wall through the use of prostheses decreases the need for prolonged mechanical ventilation. The thorax is a tridimensional structure with the vertebral column and the sternum composing the anterior and posterior pillars. The musculoskeletal thorax expands the lungs through increasing the volume of thorax by contracting the diaphragm and expanding the rib cage. When this respiratory musculoskeletal pump cannot provide passive circumferential support for the lungs, as witnessed sometimes in young patients, then thoracic insufficiency syndrome develops, namely, an inability of the thorax to support normal respiration and lung growth [11]. It follows that very large chest wall resections, with or without lung resection, can cause acute thoracic insufficiency syndrome because lung expansion and thoracic volume are interdependent. The thoracic mechanism of respiration may be compromised in the early postoper-

Ann Thorac Surg BERTHET ET AL 2011;91:1709 16 INNOVATIVE CHEST WALL RECONSTRUCTION ative period because of the reduction in thoracic volume available for lungs, which may induce an acute restrictive lung disease (Fig 7). For Pfannschmidt and colleagues [12], chest wall resections that included lung resections did not significantly impair postoperative pulmonary function any more than chest wall resections without concomitant lung resection. An immediate, single-step thoracic reconstruction is necessary so that the respiratory pump can continue to function, with a normal, stable volume and some ability to change that volume [11]. The core principle of the reconstructive technique is to restore the maximum thorax volume possible in a single step, thus avoiding the bidimensional thoracic deformity responsible for acute thoracic insufficiency syndrome. Furthermore, it should also allow for immediate extubation. Lardinois [10] asserts that the reconstruction of the chest wall plays a crucial role in determining postoperative morbidity and mortality. The type and the importance of chest wall reconstruction has been described by historical reports and reviewed recently, and there are still controversies [2] as to which chest wall lesions should be reconstructed. According to Gonfiotti and coworkers [13], the basic rule is that defects smaller than 5 cm in size in any location and those as large as 10 cm posteriorly do not require reconstruction. Our experience corroborates this rule, but we would add that the level of the defect, the number of ribs resected (more than the length of the resection), and the degree of sternal resection also influence the indications for reconstruction. The resection of other tissues (lung, main vessels, pulmonary apex, vertebral body, and diaphragm) and any associated disease (heart failure, contralateral lung disease) should also be taken into account. The choice of prosthetic materials is usually based on the habits of the surgical team. Deschamps and associates [2] found no correlation between the choice of prosthetic material and the complication rate. Nevertheless, the choice of material follows rules and is based on consensus [14]. Large anterolateral chest wall defects and complete sternectomies require rigid reconstruction. In Fig 7. Postoperative computed tomography scan: two-dimensional reconstruction using a single polytetrafluoroethylene mesh after anterolateral chest wall resection. 1715 contrast, nonrigid prostheses may be sufficient to reconstruct subtotal sternectomies and posterior defects. Since 1972 [15], a sandwich of two layers of Marlex mesh with a filler of methylmethacrylat has been considered a technique that satisfies the requirements for rigidity, protection, and chest wall shaping. Many reports underline the drawbacks of the methylmethacrylat sandwich, citing technical difficulties, the time consumed during surgery, and the need for experienced hands to manipulate the methylmethacrylat sandwich. Moreover, the use of methylmethacrylat is associated with complications, such as surrounding dead parietal space, early fractures, infection, extrusion, and perioperative heat damage. From 1995 to 2006, in our practice, we use a single layer of 2-mm thickness PTFE combined with muscle flaps when rigid reconstruction was required. The single use of the PTFE mesh in very large chest wall defects provides bidimensional reconstruction. Early and long-term results of this thoracic bidimensional reconstruction frequently show chest walls distorted by curve rotation, which causes volume depletion and transient paradoxic respiration in 26% of our patients [16]. Since 2006, the author used titanium bars combined with eptfe as modern rigid implants to reconstruct full thickness anterolateral or large posterior chest wall defects. Ours is the most important series using a combination of titanium and eptfe to repair very large chest wall defects. Metal plates are considered to have insufficient affinity for tissues, and its responsibility for lung injury is discussed [17]. The association of Dualmesh that is placed under the titanium plates carries out the protection of the lungs and the affinity to tissues. Dualmesh consists totally of eptfe and has two distinct surfaces: the smoother surface is designed for minimal tissue attachment, and the patterned, indented surface is designed for active tissue incorporation. In contrast with experimental studies [18], we did not notice midterm shrinkage of the Dualmesh. The most common downside after reconstruction using synthetic implants is the development of a seroma. Deschamps and coworkers [2] demonstrated that the use of PTFE to reconstruct the chest wall did not reduce the rate of local complications. In our study, no seromas developed, and the two local complications were probably due to an infectious process that arose from the surrounding superficial tissues. Although our study is a retrospective study and has a limited number of cases, we make a favorable assessment of the biocompatibility of the Dualmesh combined with titanium plates. The use of numerous chest drains with negative pressure, the roughened macroporous layer of Dualmesh, and the fixing of the dual mesh to the titanium plates seem to prevent fluid collection and facilitate a high degree of adhesion between the Dualmesh and the surrounding tissues. The use of titanium bars has been previously described for sternal reconstruction after sternectomy [19] or dehiscence of the sternum. Compared with other implants, titanium is corrosion free, chemically inert, and quickly and precisely adaptable to the GENERAL THORACIC

GENERAL THORACIC 1716 BERTHET ET AL Ann Thorac Surg INNOVATIVE CHEST WALL RECONSTRUCTION 2011;91:1709 16 shape of the thoracic wall. Titanium can be imaged safely with both computed tomography scan and magnetic resonance imaging, and therefore, it does not affect the oncologic follow-up of the patient. In our study, most full thickness defects were reconstructed through horizontal rib osteosynthesis. For this, we used STRATOS, according to the technique described by Coonar and colleagues [19] in a case report about a large anterior defect. We encountered no difficulty to adapt STRATOS to many anatomic situations, and the shape of the pliers allowed us to minimize the skin incision. A large posterolateral full-thickness defect with a combined resection of four vertebral bodies was treated by the implantation of VEPTR. This vertical ribs osteosynthesis system is usually used for skeletally immature patients with a diagnosis of constrictive chest wall syndrome. In most cases, Campbell and associates [11] used the VEPTR in very young patients with severely deformed ribcages, where the expansion of the device may later keep up with growth of the patient. In our experience, the most interesting aspect of the VEPTR is the verticality of this thoracic osteosynthesis system. This characteristic permits stable thoracic wall repairs even in cases of very large posterior chest wall defects associated with vertebral resection (without posterior segment of disease-free ribs). During posterior resection, surgeons must keep the rib above and below intact to firmly attach the VEPTR. The superior VEPTR device encircles the first disease-free rib. According to Campbell and colleagues [11], it should be placed no more superiorly than the second rib to avoid impinging the brachial plexus; they noted particular attention to skin and soft tissue coverage, as the device has a high profile. According to those researchers, several clinical situations are not compatible with VEPTR implantation: absence of disease-free ribs for proximal attachment, body weight less than the 25th percentile, and active pulmonary infection. Living tissue is particularly necessary to help combat infection and fill the dead space between the prosthesis and the surrounding tissue after stabilization of the chest wall [4]. The main local flaps are pectoralis major, latissimus dorsi, rectus abdominis, and omentum. In our procedures, the pectoralis muscles flaps were approximated in the midline to cover both the Dualmesh and STRATOS as the commonest technique for covering anterior defects. We believe that the low rate of postoperative wound complication results from our immediate and simultaneous soft tissue reconstruction, and furthermore, that the low rate of respiratory distress is due to our use of skeletal reconstruction using titanium bars and eptfe. In conclusion, our experience and initial results shows that rib osteosynthesis combined with Dualmesh can be used easily and safely in a one-stage procedure, with low morbidity and an apparent improvement in some aspects of early postoperative respiratory function. For large posterior resections, horizontal osteosynthesis is not possible, and vertical rib osteosynthesis should be considered. This technique simplifies reconstruction when compared with previous techniques that involve solid reconstruction. Further studies and long-term follow-up is necessary to more accurately determine the long-term outcomes. References 1. Weyant MJ, Bains MS, Venkatraman E, et al. Results of chest wall resection and reconstruction with and without rigid prosthesis. Ann Thorac Surg 2006;81:279 85. 2. Deschamps C, Tirnaksiz BM, Darbandi R, et al. Early and long-term results of prosthetic chest wall reconstruction. J Thorac Cardiovasc Surg 1999;117:588 91. 3. Mansour KA, Thourani VH, Losken A, et al. Chest wall resection and reconstruction: a 25-year experience. Ann Thorac Surg 2002;73:1720 5. 4. Chapelier A, Macchiarini P, Rietjens M, et al. Chest wall reconstruction following resection of large primary malignant tumors. Eur J Cardiothorac Surg 1994;8:351. 5. Martini N, Huvos A, Burt M, et al. Predictors of survival in malignant tumors of the sternum. J Thorac Cardiovasc Surg 1996;111:96 106. 6. Downey RJ, Martini N, Rusch VW, Bains MS, Korst RJ, Ginsberg RJ. Extent of chest wall invasion and survival in patients with lung cancer. Ann Thorac Surg 1999;68:188 93. 7. Martini N. Mediastinal lymph node dissection for lung cancer: the Memorial experience. Chest Surg Clin North Am 1995;5:189 203. 8. Arnold PG, Pairolero PC. Chest wall reconstruction: an account of 500 consecutive patients. Plast Reconstr Surg 1996;98:804 10. 9. Widhe B, Bauer HC, for the Scandinavian Sarcoma Group. Surgical treatment is decisive for outcome in chondrosarcoma of the chest wall. J Thorac Cardiovasc Surg 2009;137: 610 4. 10. Lardinois D, Muller M, Furrer M, et al. Functional assessment of chest wall integrity after MM reconstruction. Ann Thorac Surg 2000;69:919-23. 11. Campbell RM, Smith MD, Mayes TC, et al. The effect of opening wedge thoracostomy on thoracic insufficiency syndrome associated with fused ribs and congenital scoliosis. J Bone Joint Surg Am 2003:1615 24. 12. Pfannschmidt J, Geisbüsch P, Muley T, Dienemann H, Hoffmann H. Surgical treatment of primary soft tissue sarcomas involving the chest: experience. Thorac Cardiovasc Surg 2006;54:182 7. 13. Gonfiotti A, Santinia PF, Campanacci D, et al. Malignant primary chest-wall tumours: techniques of reconstruction and survival. Eur J Cardiothorac Surg 2010;38:39 45. 14. Chapelier AR, Missana MC, Couturaud B, et al. Sternal resection and reconstruction for primary malignant tumors. Ann Thorac Surg 2004;77:1001 7. 15. McCormack PM. Use of prosthetic materials in chest-wall reconstruction. Assets and liabilities. Surg Clin North Am 1989;69:965 76. 16. Berthet JP, Vidal R, Alric P, Marty-Ané CH. T3 Non small cell lung cancer with chest wall invasion: surgical strategy. J Chirurg Thorac Cardiovasc 2007;11:65 128. 17. Briccoli A, Manfrini M, Rocca M, Lari S, Giacomini S, Mercuri M. Sternal reconstruction with synthetic mesh and metallic plates for high grade tumours of the chest wall. Eur J Surg 2002;168:494 9. 18. Schoenmaeckers EJ, van der Valk SB, van den Hout HW, Raymakers JF, Rakic S. Computed tomographic measurements of mesh shrinkage after laparoscopic ventral incisional hernia repair. Surg Endosc 2009;23:1620 3. 19. Coonar A, Qureshi N, Smith I, Wells FC, Reisberg E, Wihlm JM. A novel titanium rib bridge system for chest wall reconstruction. Ann Thorac Surg 2009;87:e46 8.