Ten year experience of bioabsorbable mesh support in pectus excavatum repair

Transcription

1 The British Association of Plastic Surgeons (2004) 57, Ten year experience of bioabsorbable mesh support in pectus excavatum repair L. Luzzi a, *, L. Voltolini a, J. Zacharias b, A. Campione a, C. Ghiribelli a, M. Di Bisceglie a, G. Gotti a a Thoracic Surgery Unit, University Hospital of Siena, Viale Bracci 14, Siena, Italy b Cardiothoracic Centre, Freeman Hospital, Newcastle upon Tyne, UK Received 28 April 2003; accepted 25 May 2004 KEYWORDS Sternum; Chest wall; Remodeling; quality of life Summary Introduction. We reviewed 10 years experience in the treatment of this deformity using a retro-sternal bioabsorbable mesh in place of a metallic device to lift and stabilise the sternum. Moreover, the mesh supports the thoracic and upper abdominal wall reconstruction. Materials and methods. From January 1990 to December 2000, in our Thoracic Surgery Unit, 65 patients with PE were assessed for surgical repair, mean age 16 ^ 3.5 years, fronto sagittal thoracic index (FSTI) 0.21, ranging from Twentythree of them underwent surgical correction after initial assessment, 22 were deferred and sent to physiotherapy. At a subsequent assessment, five of the patients sent to physiotherapy were deemed to require surgery. Results. Of the 28 patients who underwent surgery, 2 (10%) presented a mild recurrence of PE after 1 year (0.30, FSTI. 0.34), meanwhile all other patients maintained a FSTI For all patients the improvement in FSTI was statistically significant, p ¼ 0:001: Patients satisfaction after 24 months was thus shared: excellent 18 patients (65%), good seven patients (25%), fair one patient (3.5%) and poor two patients (7%). No major complications were observed in preoperative period. Patients mobilisation was soon achieved thanks to the postoperative pain control and the absence of retro-sternal metallic support. Conclusions. The introduction of bioabsorbable mesh in the Robicsek technique is a safe procedure related to a high percentage of success. The high tolerance of the material reduces the inflammatory reaction. Moreover, the procedure prevents patients from having complications caused by retro-sternal device dislodgment, avoiding a second intervention for device reposition and reducing the postoperative chest pain achieving an early patient mobilisation. In the end a complete reconstruction of the upper abdomen wall has been produced. Q 2004 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. *Corresponding author. Tel.: þ address: luzzi.luca@virgilio.it Pectus excavatum (PE) is the most frequent chest deformity in children affecting 0.4 1% of live births, with a male to female ratio of 4:1. 1 The percentage of symptomatic patients is low and very few of them are referred for surgical correction. 2 S /$ - see front matter Q 2004 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. doi: /j.bjps

2 734 L. Luzzi et al. Many surgical papers advocate the benefits of PE correction in symptomatic patients or in patients with severe chest deformity. 3 However, the evidence that suggests improvement in lung function post operatively is controversial, in fact in a recent study a reduction in pulmonary volume at long term follow-up has been found. 4 Despite this a great number of treated patients report an improvement of symptoms with psychological benefits due to the aesthetic results. In a large majority of patients psychological benefits due to correction of the malformation seems to be the real end point of surgical repair. 5 During the last 15 years some centres have proposed new minimally invasive techniques in the effort to minimise the surgical and postoperative discomfort. 6,7 Since 1990, we have introduced in our surgical practice a modified Robicsek technique 3 for PE repair. We use a bioabsorbable mesh made of Polyglactin 910 (Ethicon inc.) to support our sternal osteotomy in place of a sub sternal bar. The objective of this technique is to reduce postoperative pain, minimise potential complications, avoid abdominal muscle weakness and prevent recurrences. We have reviewed our practice over the last 10 years of PE correction to look for potential benefits of this approach. Materials and methods From January 1990 to December 2000, 65 patients with PE were referred to the Thoracic Surgery Unit of the University Hospital of Siena. Twenty-three of them underwent surgical correction after an initial assessment. Twenty-two were deferred and were sent for physiotherapy to improve pectoral muscle bulk. At a subsequent assessment 20 were evaluated after physiotherapy and five patients were deemed to require surgery. The criteria used during the first and second assessment are shown in Table 1. Preoperative work up The initial assessment in patients consisted of a physical examination and an interview to evaluate the psychological distress due to the PE. The surgeon with the help of a psychologist conducted the interviews. The distress was graded as: severe, if there was a limitation in social activity associated with depression or affective disorders; moderate, if there is only limitation in social activities and mild, if the malformation is present but accepted by patients. All patients received a standard chest X- ray to assess the Fronto-Sagittal Thoracic Index (FSTI). 8 This is defined as the ratio between the shortest distance from sternum to the vertebra (on a lateral chest roentgenogram) and the transverse thoracic diameter obtained from a radiological image at the same level. 9 A normal value of FSTI is considered over Patients affected by skeletal anomalies (Marfan s Syndrome and connective tissue disorder) were excluded from surgery because of the high risk of recurrence. 10 Pulmonary preoperative measurements consisted of lung volume and pulmonary function. Lung volume measurement included vital capacity (VC), total lung capacity (TLC), residual volume (RV), functional residual capacity (FRC) and RV/TLC ratio. Pulmonary functional measurement were made up of maximal voluntary ventilation (MVV), forced vital capacity (FVC), forced expiratory volume in one second (FEV 1 ) and maximal expiratory flow in 75 and 50% vital capacity. In all patients, an echocardiogram was performed to evaluate the presence of congenital cardiac defects or compressions. In recent patients, we performed a spiral CT scan with 3D reconstruction. This new technique permits us to discover minimal asymmetries in the deformity, the point of major sternal depression and the number and extension of deformed cartilages. 11 We can then plan where to perform the sternal osteotomy and how many cartilages have to be removed (see Figs. 1 and 2). Patients having their surgery deferred underwent 3 months of physiotherapy. About the timing of surgical correction, we agree with Fonkalsrud 12 who considers the optimal age for repair in the range of years. We know that only a small number of patients with PE will be symptomatic after 10 years. However, it is very difficult to predict which of them will Table 1 Criteria used for the first and second patients assessment First assessment Second assessment (1) Presence of respiratory and/or cardiac symptoms (1) The degree of malformation despite physiotherapy (2) FSTI (Fronto-Sagittal Thoracic Index) grade (2) The persistence of psychological distress (severe, moderate or mild) (3) The grade of psychological distress (severe, moderate or (3) The persistence of symptoms mild) (4) The age (more than 10 years)

3 Ten year experience of bioabsorbable mesh support in pectus excavatum repair 735 Figure 2 3D CT scan at 6 months from surgical correction. Figure 1 The arrow shows the point chosen for the sternum osteotomy based on the sternum 3D image by the CT scan. became symptomatic. 14 Performing corrections in children older than 10 years old, we reduce the risk of unsafe surgical procedures. For this reason, all patients less then 10 years old were re-evaluated after 6 months of physiotherapy and most of them regularly reviewed until 10. Operative technique We perform a median thoracic skin incision in men and a transverse sub mammary incision in women. Skin flaps are mobilised with electrocautery. The pectoralis major are detached from the sternum with a portion of pectoralis minor and elevated. The costal cartilage is exposed and the deformed cartilages ranging from the 3rd to 8th were removed following a sub perichondral dissection. A wedge osteotomy of the sternum at the point of posterior angulation was performed and the osteotomy was repaired using heavy single stitches. Often the Xiphoid process is so deformed that it is impossible to lift it up despite the sternal wedge osteotomy, therefore, we usually remove this during the procedure. A substernal plane is created and the mesh inserted under the sternum. In this series, we used a Vicryl mesh produced By Ethicon LTD (UK) in Poliglactil 910. It has been tested for a tensile resistance of Kg/cm 2 and a traction of 45 Kg over 10 cm, after 2 weeks the mesh maintain 80% of its initial tensile resistance. The mesh is stitched to the intercostal muscles, the pericondrium and, sometimes, to the cartilage stump as described by Robicsek. 13 The inferior margin of the pectoralis muscle is also stitched to the mesh using bioabsorbable sutures. This process makes the mesh taut while the pectoralis major contributes to lift up the sternum. The rectus muscles are joined to the mesh, in this way the tension produced by muscle contraction is not applied directly to the lower portion of the sternum (see Fig. 3). A retrosternal drain is left after the procedure. If the pleura is entered an intercostal drain is inserted. All patients were extubated after surgery before leaving the theatre. An early patient mobilisation was soon achieved. This objective is gained by the minimal postoperative pain due to theabsenceoftheretro-sternalsupport.inthe Nuss procedure, the high force needed to lift up the sternum (varying from 20 to 40 lbs) causes an intense postoperative chest pain which

4 736 L. Luzzi et al. Results Sixty-five patients were referred for preoperative assessment, 52 of them were male and 13 female, average age 16 ^ 3,5 years. Nineteen reported respiratory symptoms (see Table 2) with associated palpitation in four cases. During the preoperative work-up, one patient presented an asymptomatic right bundle branch block (RBB) picked up on the ECG. A further four patients had right axis deviation, two of them had associated respiratory problems. The mean FSTI value was 0.21 ranging from 0.15 to 0.33 (v.n ) this value is referred to all 65 patients. At the end of preoperative work up 28 patients were considered suitable for surgical correction. They represented only 43% of all patients referred. Perioperative results Figure 3 The picture shows the technique for PE correction by positioning the reabsorbable mesh. compromises the patients mobilisation exposing them to pulmonary atelectasis. Normally patients need high doses of analgesic or an epidural analgesia, which produce immobilisation in bed. Despite the reduced postoperative pain given by the Ravitch technique, the patient is prohibited from performing any sport or rehabilitation activity for at least 6 weeks after surgery to avoid device dislodgement. We chose not to perform our technique in cases of patients with PE associated to Marfan s syndrome in which the use of a retro-sternal device is mandatory in preventing early recurrence as reported by Sherer et al. 10 Follow up All surgical patients were seen at 6 weeks, 12 months and 24 months after correction. Routine follow up consisted of chest X-ray, physical examination and respiratory function tests. The results of surgical treatment were evaluated based on: (1) patients satisfaction in a scale of excellent, good, fair and poor. (2) Objective resolution of symptoms. (3) The improving of postoperative FSTI. The positioning of the bioabsorbable mesh was possible in all patients without intra-operative technical problems. No relapses or patient reaction to the mesh were observed during the perioperative period. Early postoperative complications were: blood loss greater than 500 cc/24 h in six patients, pulmonary atelectasis in three patients, mild postoperative pain in two patients. Patients were discharged after 5 days ^ 2. No patients needed a scar revision or a readmission after discharge. Postoperative follow-up More than 90% of patients repeated lung volume evaluation and lung function tests after surgery. The follow up after 24 months showed a mild improvement of lung volume and lung function but it was not statistically significant (see Table 3). This improvement in most cases can be due to the physiologic growth of these young patients in 24 months from the correction. Patients who presented ECG changes such as RBB or right axis deviation maintained the same alteration despite correction. The FSTI was significantly greater than before with a mean value of 0.35 ranging from 0.34 to 0.38 ðp ¼ 0:001Þ: We had a mild recurrence of PE Table 2 Respiratory symptoms reported during the first assessment Symptoms Decrease in exercise tolerance 8 Easy fatigability 3 Inability to take a deep breaths 5 Shortness of breath 3 Number of patients

5 Ten year experience of bioabsorbable mesh support in pectus excavatum repair 737 Table 3 No statistical differences have been found during the follow-up at 24 months from surgery in lung volumes and function after 12 months in two patients (0.30, FSTI, 0.34) with consequent low patient satisfaction. The reintervention was not proposed because both of them were no longer symptomatic. In one of the two patients, the recurrence was due to an asymmetrical inclination of about 158 in the right chest. Every patient began postoperative sport activity very soon because there was no risk of sternal device dislodging. In the first month after surgery no patients developed abdominal muscle relapse or abdominal weakness. Despite the high percentage of success, 90%, and the two mild recurrences, patient satisfaction was divided as such: excellent results in 18 patients (64.2%), good in seven patients (25%), fair in one patients (3.5%), poor two patients (7%). Excellent and good are considered satisfactory. After correction no symptoms were recorded from patients that were preoperatively symptomatic. Discussion Preoperative Postoperative (24 months later) TLC ^ ^ Ns RV ^ ^ Ns FEV ^ ^ Ns FRC ^ ^ Ns p Value Despite the success reported in the literature by the surgical correction of PE, the optimal age for repair and the surgical procedure most suitable to achieve the best result is still under discussion. 14 Many procedures have been proposed during the years and each presents different advantages and disadvantages. One of the first described is the Brown technique 15 later modified by Ravitch 16 and Welch. 17 These concern the resection of the costal cartilages in their deformed portion associated with a sternal osteotomy and the positioning of a retrosternal metallic support. Despite the high percentage of success reported, especially in resolution of symptoms, the use of retrosternal steel support presents some disadvantages. In the literature there are many adverse consequences caused by device dislodgement: migration into the peritoneal cavity, 18 laceration of the phrenic artery and haemorrhage, 19 migration through the pericardium causing tamponade, 20 damage of cardiac valve, 21 cardiac chamber perforation, 22 and formation of vascular thrombus with systemic embolic events. To avoid such complications less invasive procedures have been proposed. One of these is the use of autologous tissue, silastic, or other prosthetic materials to fill the PE deformity. However, this procedure is associated with an high number of migration causing further cosmetic deformity. 14 Thereafter is only a cosmetic intervention which does not effect cardiac and respiratory symptoms. During the last 10 years, Nuss has proposed a new minimally invasive technique 23 which avoids any cartilage resections and sternal osteotomy achieving a good aesthetical and functional result. This is possible by placing a convex steel bar under the sternum through bilateral thoracic incisions and then turning the bar over to elevate the deformed sternum and costal cartilages until the desired position. Nuss has recently reviewed his series of 329 corrections to assess the percentage of postoperative complications. 24 He found 2.4% of pericarditis, 2.6% of wound infections (40% of which extended to the device), 8.8% of bar dislodgment and one case of oversensitivity to the device. The author concludes that his technique provide an excellent long-term result with a low rate of complications. However, his results are difficult to reproduce by other centres because of the low number of patients treated for PE every year, which prolong the time needed to complete the learning curve. Despite this, the principal disadvantages of the Nuss procedure is the length of time that the bar must be left in the patient to achieve a stable correction. It is ascertained that each patient should retain the steel support for at least 2 years but, depending on the age of the patient and on the severity of the deformity, a time longer than 3 years is common. The long duration of the device left in patients affects a more complex second operation, and takes more time then in the Ravitch technique. At the end of the 80 s, we set up a procedure for PE repair. We tried to achieve the same aesthetical and functional results of traditional techniques decreasing the postoperative pain, avoiding the retro-sternal device with its complication and the second surgical procedure, and secure an early patient mobilisation after surgery. The first procedure, which avoided the retro-sternal device was proposed by Robicsek in 1978, who used a Marlex mash in PE repair. 13 This technique was initially reserved for the treatment of recurrent PE or in case of a very deep malformation. The feasibility of procedure was well documented in literature. However, the use of a nonbioabsorbable mesh caused an important fibrous reaction associated with an improvement of chest wall rigidity and the felling of heaviness, as described for inguinal hernioplasty. The good results in reconstruction of chest wall and in hernioplasty with complete and

6 738 L. Luzzi et al. partial bioabsorbable mesh pushed us to try this option modifying the original Robicsek technique. Fonkalsrud 25 has recently measured the force required to elevate the sternum in patients affected by PE, this force varied between 32 and 41 lbs, depending on the age. This force falls dramatically under 1.5 lbs after surgical correction. The low force required to maintain the sternum in the correct position is the key of the faster consolidation of the sternocondroplasty and permit the Vicryl mesh to maintain the reconstruction. Other authors have proposed new procedure for PE correction avoiding retro-sternal metallic support using bioabsorbable materials. Patrick et al. 26 presented seven cases of patients treated by positioning of a retrosternal mesh of Polyglactin 910 (Ethicon inc.). However, they do not specify in which way they fix the mesh to the rectus muscles and pectoralis major. Therefore, after a follow-up of 12 months, they reported one recurrence with an incidence of 15%. In 2002, Gurkok et al. 27 described a new technique, adopted in 33 patients, using a bioabsorbable copolymer plaque and polymer screws for fixing the sternum at the level of wedge osteotomy. They did not observe PE recurrences but the follow up reported is too short and the Equipment used is expensive, about $1000 for each procedure. These two procedures like ours try to achieve a stable lift of the sternum at the osteotomy level. We consider this point the fulcrum of repair. When we stitch the rectus muscles to the inferior margin of the pectorals major, after cartilage removal, a small triangular area often remains uncovered, especially after removing the xiphoid process. To cover this lack a high-tension suture is required. This can increase the downward pull of the rectus muscles on the lower end of the sternum causing an instability that could lead to a failure at the level of the osteotomy resulting in a high probability of PE recurrence. This problem is overcome by positioning the bioabsorbable mesh behind the sternum fixing it to the intercostal bundles, the perichondrium and the bottom of the petoralis major which contributes to push the sternal tip upwards, rather than compressing it (Figs. 4 and 5). There are many advantages in avoiding the retro-sternal device. At first patients can start physiotherapy and sport activity sooner after the operation, secondly the complications due to the stainless steel strut dislodgement are avoided. Before 1990, when the Ravitch technique was still used, some patients developed an unappealing protruding of the upper abdomen after correction. This unsightly complication has been recently reported also by Scoenmakers et al. 28 They Figure 4 Introduction of the mesh under the pectoralis major before the suture. showed an incidence of abdominal muscle weakness after surgery of 56% against 19% seen during the preoperative assessment. However, this is reversible after 18 months of active physiotherapy in 90% of patients. This complication could probably be due to a weakness of rectus muscles after the disconnection from the lower margin of the anterior chest wall and fixing to the pectoralis major. After the introduction of the mesh technique no patients treated showed a protruding abdomen. In our technique the rectus muscles are stitched to the mesh that simulate the inferior thoracic cage permitting a better anatomical reconstruction of the chest and upper abdomen wall providing a better aesthetic result. The percentage of patient satisfaction after correction was 90% and of PE recurrence, 7%. The two recurrences observed were mild with a FSTI more than They did not need a second correction because the surgical treatment had lead to a symptom resolution. The data about postoperative pulmonary function did not show any significant improvement over that normally predicted. 29 Six weeks after surgery no patients previously symptomatic reported the persistence of any symptoms. 30 However, the persistence of Figure 5 The fixing of mesh under the sternum and the anterior chest cage. In this way, we a obtain the complete reconstruction also of the upper abdomen.

7 Ten year experience of bioabsorbable mesh support in pectus excavatum repair 739 cardiac ECG abnormalities proves the irreversibility of some of the anatomical and physiological changes that occur during chest structures adaptation to the PE. Morshuis et al. 4 have recently analysed lung function tests before surgery and at long-term follow up in 152 patients affected by PE. They found a significant unexpected reduction in all pulmonary volumes after surgery. The cause of this postoperative restrictive pattern is probably due to the chest wall rigidity as result of the reparative process after surgery. 31 It is well known that one of the complications of PE repair during early childhood is acquired Jeune s syndrome, a remodelling of the chest wall resulting in a severe respiratory restrictive syndrome. 32 We did not see a reduction in lung function in any of our patients. It could be possible that the short half-life of Polyglactin 910 minimises the modification of chest wall elasticity, however, this hypothesis still needs to be verified. The use of bioabsorbable mesh in our modified Robicsek technique is a safe procedure correlated to high percentage of success. Our technique saves patients from the complications caused by retrosternal device dislodgment, avoiding the second intervention for device remotion and reducing the postoperative chest pain achieving an early patient mobilisation. Therefore, the bioabsorbable mesh seems to reduce the fibroid reaction that we usually have with nonbioabsorbable materials like Marlex preserving the elasticity of thoracic cage. Contrary to Nuss technique, our PE correction can be applied in teenagers and young adults also affected by asymmetrical malformation. Furthermore, the postoperative pain is lower due to the less force needed to maintain the correction compared to Nuss technique. No postoperative protruding abdomen were observed with an improvement of aesthetical success reported by patients. References 1. Shambergher RC, General thoracic surgery, vol. 1. Philadelphia: Lippincott/Williams and Wilkins; p Shambergher RC. Cardiopulmonary effect of anterior chest wall deformity. Chest Surg Clin N Am 2000;10(2): Robicsek F. Surgical treatment of pectus excavatum. Chest Surg Clin N Am 2000;10: Morshuis W, Folgering H, Barentsz J, et al. Pulmonary function before surgery for pectus excavatum and at long term follow up. Chest 1994;105(6): Einsiedel E, Clausner A. Funnel chest. Psychological and psychosomatic aspects in children, youngsters, and young adults. J Cardiovasc Surg (Torino) 1999;40(5): Hebra A, Swolvwland B, Egbert M, Tagge EP, Georgeson K, Othersen HB, Nuss D. Outcome analysis of minimal invasive repair of pectus excavatum: review of 251 cases. J Pediatr Surg 2000;35(2): Experience and modification update for the minimally invasive Nuss technique for pectus excavatum repair in 303 patientscroitoru DP, Kelly RE, Goretsky MJ, Lawson ML, Swoveland B, Nuss D. J Pediatr Surg 2002;37(3): Backer OG, Brunner S, Larsen V. Radiologic evaluation of funnel chest. Acta Radiol (Stockh) 1961;55: Ohno K, Nakahira M, Takeuchi S, Shiokawa C, Moriuchi T, et al. Indications for surgical treatment of funnel chest by chest radiograph. Pediatr Surg Int 2001;17: Sherer, et al. Surgical management of children and young adults with Marphan syndrome and pectus excavatum. J Pediatr Surg 1988;(23): Pretorius ES, Haller JA, Fishman EK. Spiral CT with 3D reconstruction in children requiring reoperate for failure of chest growth after pectus excavatum surgery. Preliminary. Clin Imaging 1998;22(2): Fonkalsrud EW, DeUgarte D, Choi E. Repair of pectus excavatum and carinatum deformities in 116 adults. Ann Surg 2002;236: Robicsek F. Marlex mesh support for the correction of very severe and recurrent pectus excavatum. Ann Thorac Surg 1978;26: Eric W, Fonkalsrud EW. Current management of pectus excavatum. World J Surg 2003;27: Brown AL. Pectus excavatum (funnel chest): anatomic basis; surgical treatment of the incipient stage in infancy and correlation with deformity in the developed stage. J Thorac Surg 1939;9: Ravitch MM. Operative technique of pectus excavatum repair. Ann Surg 1949;129: Welch KJ. Satisfactory surgical correction of pectus excavatum deformity in childhood: a limited opportunity. J Thorac Surg 1958;36: Stefani A, Morandi U, Lodi R. Migration of pectus exacavatum correction metal support into the abdomen. Eur J Cardiothorac Surg 1998;14: Paret G, Taustein I, Vardi A, Yellin A, Dekel B, Barzylai Z. Laceration of phrenyc artery: a life threatening complication after repair of pectus excavatum. J Cardiovasc Surg 1996; 37: Elami A, Liberman Y. Hemopericardium: a late complication after repair of pectus excavatum. J Cardiovasc Surg 1991; 32: Pircova A, Sekarski-Hunkler N, Jeanrenaud X, Ruchat P, Sadeghi H, Frey P. Cardiac perforation after surgical repair of pectus excavatum. J Pediatr Surg 1995;30: Dalrimple-Hay MJ, Claver A, Lea RE, Monro JL. Migration of pectus excavatum correction bar into the left ventricle. Eur J Cardiothorac Surg 1997;12: Nuss D, Kelly Jr RE, Croitoru DP, et al. A ten-year review of a minimally invasive technique for the correction of pectus excavatum. J Pediatr Surg 1998;33: Nuss D, Croitoru DP, Kelly Jr RE, Goretsky MJ, Nuss KJ, Gustin TS. Review and discussion of the complications of minimally invasive pectus excavatum repair. Eur J Pediatr Surg 2002; 12: Fonkalsrud EW, Reemtsen B. Force required to elevate the sternum of pectus excavatum patients. J Am Coll Surg 2002; 195: Patrick B, Trip HF. Bioabsorbable wave technique for repair of pectus excavatum. J Thorac Cardiovasc Surg 2000;119(1): Gürkök S, Genç O, Dakak M, Balkanli K. The use of bioabsorbable material in correction of pectus deforities. Eur J Cardiothorac Surg 2001;19:711 2.

8 740 L. Luzzi et al. 28. Schoenmakers M, Gulmans V, Bax N, Helders P. Physiotherapy as an adjuvant to the surgical treatment of anterior chest wall deformities: a necessity? A prospective descriptive study in 21 patients. J Pediatr Surg 2000;35(10) Wyn SR, Driscoll DJ, Ostrom NK, Staats BA, O Connel EJ, Mottram CD, Telander RL. Exercise cardiorespiratory function in adolescents with pectus excavatum. Observation before and after operation. J Thorac Cardiovasc Surg 1990; 99(1): Quigley PM, Haller JA, Jelus KL, Loughlin GM, Marcus CL. Cardiorespiratory function before and after corrective surgery in pectus excavatum. J Pediatr 1996;129(6): Deveroux L, Ivanoff I, Rochette F, Demedts M. Mechanism of pulmonary function changes after surgical correction for funnel chest. Eur Respir J 1988;1: Haller JA, Colombani PM. Chest wall constriction after too extensive and too early operation for pectus excavatum. Ann Thorac Surg 1996;61(6):