Abstract
Abstract
The application of thoracoscopic surgical techniques to pediatric solid tumors represents an important adjunctive tool for the surgical management of childhood cancer. Nearly four decades has passed since the introduction of minimally invasive chest surgery in children, and although the adoption of minimally invasive surgery in general pediatric surgical practice is better recognized, its role in pediatric oncology is still considered a developing field. As no consensus exists regarding the use of thoracoscopy for pediatric thoracic solid tumors, the purpose of this article is to review the current literature surrounding the use of thoracoscopic interventions in pediatric oncology and examine established indications, procedures, and technologic advances.
Introduction
T
P, prospective; R, retrospective.
Many children with pulmonary or mediastinal masses may initially require tissue for pathologic diagnosis and in some circumstances tumor resection. Although percutaneous biopsies may be performed with the assistance of image-guided technology, situations remain in which a surgical biopsy is required either via the thoracoscopic or the open surgical technique. A limited thoracotomy incision or potentially a traditional posterolateral thoracotomy, which requires a large incision, rib retraction, and possible division of the latissimus dorsi, 2 may be required to obtain adequate specimen. This morbid incision has led to devastating consequences in children such as shoulder elevation, winged scapula, 3 chest wall asymmetry, and scoliosis in up to 30% of neonates.4,5 Therefore, consideration for a thoracoscopic approach is certainly deserving in this special population.
The initial use of thoracoscopy in children was described in 1971, 6 and in 1995, Holcomb et al. 7 proposed indications for the use of MIS based on data collected from the Children's Cancer Group. Since then pediatric surgeons have used thoracoscopy to diagnose and resect benign, malignant, or metastatic intrathoracic tumors. 8 However, tumor extirpation remains controversial and continues to be limited.
The purpose of this article is to review current literature regarding the use of thoracoscopic surgery in pediatric oncology, including the established indications, procedures, and technologic advances.
Advantages of Using a Thoracoscopic Approach
Theoretical advantages of using thoracoscopy (Table 2) include access to a wide area through limited incisions, better visualization of thoracic and mediastinal structures, 9 and magnification of the local anatomy, which refines the assessment of the limits of resection 2 as seen in neurogenic tumors where greater precision allows for individualization and easier identification of nerve roots, 9 thereby potentially minimizing the risk of excessive stretch on the spinal roots. 2 Additionally, the use of thoracoscopy has led to a decreased intraoperative blood loss,9–11 decreased postoperative pain, reduced hospital length of stay (LOS),9–13 fewer chest tubes required and/or earlier removal,9,13,14 improved cosmetic result, faster return to normal activity, and reduced pulmonary adhesions.9,15–18
Although there are no randomized controlled trials dealing with thoracoscopic and open lung resection, several case series pertaining to non–oncologic-associated thoracoscopic surgeries have demonstrated data to support some of the proposed advantages of a minimally invasive approach. Children undergoing thoracoscopic resection of foregut duplications and congenital lung lesions experienced a decreased chest tube duration, LOS by at least 50% compared with open thoracotomy,13,19,20 less intraoperative blood loss, 11 and fewer postoperative complications. 19 However, a case-matched study could not demonstrate differences in the number of days of intravenous opioid use, cumulative dose of intravenous opioid, chest tube duration, chest tube output, days to discharge, or postoperative complication rate. 21
Children who underwent thoracoscopic neuroblastoma resection had a decreased LOS by up to 2 days,9,12 had less intraoperative blood loss, 9 and required a chest tube for up to 50% less time.9,12 Up to 42% of pediatric thoracoscopic surgeries did not even require chest tube drainage.22–24 Patients not requiring postoperative chest drainage may potentially have decreased postoperative pain 24 and the potential risk of tumor cell dissemination. 14 In children who underwent thoracoscopic resection of lung nodules, mean chest tube duration was 0.5 days, with LOS of 1.6 days, and all patients in their series with metastatic lung disease underwent additional chemotherapy and sometimes radiation therapy after resection. 25 Therefore, a rapid recovery time interval has been proposed to allow for additional adjunctive therapy to be started sooner than after open thoracotomy. 25
Limitations, Complications, and Contraindications of Thoracoscopy
Factors limiting the use of thoracoscopy in children with cancer (Table 3) are related to the surgeon, patient, pathology, and technology. Pediatric surgeons' experience with advanced thoracoscopic techniques may be limited. Although the ability to gain thoracoscopic experience has increased over time, the relatively low volume of pediatric oncologic cases encountered by general pediatric surgeons can negatively impact their decisions or confidence to use a thoracoscopic approach. Single-lung ventilation with double-lumen endotracheal tubes is unavailable for the smallest of patients, and, alternatively, bronchial blockers require highly skilled anesthesiologists for the fragile airway of small children. 26 Although thoracoscopy is feasible using only positive pressure carbon dioxide insufflation without single-lung ventilation, the resulting lung collapse is often suboptimal, thus hampering visualization of pulmonary nodules, decrease working space in the thorax, and increasing the risk of lung injury. The small working space within the thoracic cavity or mediastinum, the patient's ability to tolerate single-lung ventilation, and hemodynamic effects caused by pressure generated by pneumothorax are competing factors that make every case uniquely different.
Large masses can potentially impede safe accessibility and specimen delivery and contribute to the potential risk of intraoperative tumor spillage and port-site recurrence.27,28 Deep or subcentimeter pulmonary lesions are difficult to visualize, and because the tactile ability is lost in thoracoscopy, these pulmonary lesions may be missed. 29 Adhesions due to prior thoracotomy may limit the ability for a thoracoscopic approach. This is especially relevant in scenarios where multiple serial thoracic surgeries are expected over the course of the disease, such as with serial pulmonary metastasectomies for recurrent metastatic osteosarcoma.
Complications encountered in thoracoscopic surgery may include pneumothorax, bleeding, port-site recurrence, 28 port-site infection, 30 chylothorax, 2 Horner's syndrome,2,9,12,14 and tumor spillage.31,32 Although the most common complication after thoracoscopic resection of intrathoracic neurogenic tumors is Horner's syndrome,12,14 this may be a function of the location of these tumors in the upper mediastinum9,10 or a result from traction, thermal injury, or edema of the stellate ganglion. 12 However, no differences in complications rates were found in thoracoscopic resection of neuroblastoma. 12 Recurrence at the chest tube site has been reported after thoracoscopic resection of pulmonary metastasis from osteosarcoma, 28 although this is extremely rare.
Inadequate equipment, insufficient training, and experience are considered contraindications to performing advanced thoracoscopy. 14 Relative contraindications can be anatomical, such as difficult thoracic access due to large tumor size with consequently increased potential for tumor rupture or spill, and physiologic, such as abnormalities in cardiac output, difficulty tolerating single-lung ventilation or thoracic carbon dioxide insufflation, and coagulopathy. 14
Preoperative Considerations
Surgeon and patient
The safety threshold and the patient selection are important for consideration of any thoracoscopic approach. The surgeon's experience, patient size, detail regarding the potential pathology, location, and proximity to vital structures all impact the final decision of whether or not to pursue this minimally invasive approach. Additional anesthesia-related concerns play a pivotal role with regard to safely securing the airway and whether lung collapse is an option.
Anesthesia
Thoracoscopic surgery requires the ability to create enough working space within the hemithorax to safely visualize and perform the operative procedure. Impediments that lead to difficulty with anesthesia in infants and young children include one-lung ventilation (OLV), carbon dioxide insufflation, hypothermia, and the effect of lateral decubitus positioning.33,34 OLV methods in pediatric patients may be achieved with use of a double-lumen tube, uninvent tube, Arnt blocker, Fogarty catheter, and endobronchial intubation.34–37 In an attempt to create this working space, anesthesiologists are challenged to perform OLV to allow for lung collapse on the operative side. In order to achieve OLV, airway anatomy of the patient with detailed knowledge of the trachea size and of the size and length of the main bronchus to be selectively intubated is required. 38 OLV has been performed in children and infants, 39 in addition to a report of using a double-lumen tube in a 6-year-old girl. 38
The smaller tracheal and bronchial diameters in children and infants may prohibit the use of double-lumen tubes or bronchial blockers because there is a lack of double-lumen tubes and commercially available bronchial blockers. 37 Challenges with endobronchial isolation may include obstruction of the right upper lobe if the right lung is being selectively intubated, incomplete decompression of the operative lung, an inability to provide oxygen and continuous positive airway pressure to the isolated lung, and the potential for lung soiling.35,37,40 OLV currently remains a technically difficult and demanding task on behalf of the anesthesia team.
Key safety considerations in the decision to adopt a thoracoscopic surgical approach is heavily influenced by the risk of cardiopulmonary collapse under anesthesia. In children with anterior mediastinal masses due to Hodgkin's or non-Hodgkin's lymphoma or less commonly neuroblastoma or germ cell tumors, the pros and cons of proceeding with an anesthetic 41 need to be considered, especially in the presence of airway compression.42,43 General anesthesia should be avoided in patients with tracheal cross-sectional area or peak inspiratory flow rate less than 50% of predicted for age and sex, and those with less respiratory compromise can generally safely undergo general anesthesia. 44 Thoracic epidural anesthesia 45 or bilevel positive airway pressure management 46 has been used in children for open biopsies of anterior mediastinal masses. However, to our knowledge, similar attempts have not been reported for thoracoscopic surgery in children, highlighting the extreme risk of anesthesia in this population.
Localization and Biopsy
Thoracoscopic biopsy of intrathoracic lesions has been effectively used for mediastinal masses, most commonly neuroblastoma and lymphoma, and for pulmonary masses that are usually metastatic lesions or infiltrates.7,25,47 Approaches for localization include hook wire, methylene blue injection, and/or the utilization of endoscopic ultrasound. Preoperative, image-guided, lung nodule localization,48–53 and intraoperative ultrasound localization have been explored for lesions, especially when small and solitary. 54 A combination of computed tomography (CT)-guided microcoil localization and thoracoscopy 49 or CT-guided hook wire localization, methylene blue staining, and thoracoscopy was able to achieve a 97%–100% diagnostic yield for small subpleural lesions.48,49,51,52 CT-guided needle localization and thoracoscopy for 0.1-cm3 lesions were 100% diagnostic. 53 CT-guided needle localization with methylene blue staining and thoracoscopy was performed on small pulmonary nodules <1 cm or nodules up to 2 cm deep to the pleural surface with 92.3%–100% diagnostic yield.50,52 Minimally invasive thoracoscopic ultrasound has also been used to assist in localization of lesions between 0.3 and 2.9 cm. 54
Although these various preoperative localization techniques serve as adjunctive tools, several case series have reported using thoracoscopic biopsy without preoperative localization and demonstrated a 96.7%–100% histological and/or bacteriological diagnosis.22,25,30,31,47,55–60 Conversion to open thoracotomy ranged from 0% to 30%,22,25,30,31,51,55–59,61 and was primarily due to limited visibility,47,51,55,58 adhesion,55,58 bleeding,55,58 decreased intraoperative oxygen saturations, 55 and hypercarbia. 58 Therefore, thoracoscopy was largely successful, and some patients were able to begin adjuvant therapy earlier. 31
Applications
Mediastinal lesions
The most common thoracic neurogenic tumors in the pediatric age group originate in the sympathetic ganglia, nearly always in the posterior mediastinum, 12 accounting for up to one-third of all mediastinal tumors, and include neuroblastoma, ganglioneuroma, and ganglioneuroblastoma.2,10,12,14,32,62 Complete resection of intrathoracic neurogenic tumors has been reported in 70%–100% of cases for lesions up to 18 cm in size.2,9,10,12,14,32,47,62–64 In addition to neurogenic tumors, other mediastinal masses have been successfully resected thoracoscopically, including lymphoma, 64 teratoma,56,64 and thymoma. 64
Gross total thoracoscopic resection of neurogenic tumors has been performed,9,10,12,62 and when compared with the open thoracotomy group, they were significantly smaller, 12 were associated with up to 75% less intraoperative blood loss,9,10 had up to 19% shorter operative times, 10 had 33%–50% decreased time of thoracic drainage,9,12 and had up to 54% decreased hospital LOS.9,10,12 Stage 1 neuroblastomas that were less than 6 cm were resected, although there were 2 cases of intraoperative tumor spill. 32 Although patients were alive without evidence of disease up to 55 months postoperatively, 32 this demonstrates a significant fear of any oncologic surgery. During resection of a thoracic neurogenic tumor, grossly positive tumor margins were present at the neural foramina because of planned division without intention to resect the intraforaminal disease; despite this anticipated positive margin there was no difference between the thoracoscopic and open thoracotomy groups. 10 Conversion to open thoracotomy was under 45%9,12,14,25,32,47,63,64 and was primarily due to extensive disease, 64 bleeding, 47 large size, 2 and difficulty with single-lung ventilation. 10 Chest tube drainage was not required in up to 35% of patients.9,12,14
Keys to successful tumor resection identified included triangulated port placement, camera with zoom magnification, meticulous hemostasis, avoidance of heat dispersion, 14 use of endocatch bags,2,10,12,14,54,62,64 and enlargement of the trocar site.2,10,12,62,65 Although in some circumstances morcellation is possible and may facilitate thoracoscopic resection,2,14,62 this method of tumor removal should be used with extreme caution. Selective single-lung ventilation 10 in older children was performed with a double-lumen endotracheal tube,12,14,54 whereas in younger patients the mainstem bronchus was intubated with an uncuffed endotracheal tube or bronchial blocker,12,14,54,62,66 and small children who were not amenable to selective bronchial intubation underwent ipsilateral lung collapse with low-pressure carbon dioxide insufflation.2,14
Lung lesions
Osteosarcoma is the most common malignant bone tumor arising in children and adolescents. 67 Approximately 20% of patients will have synchronous metastatic pulmonary disease at diagnosis, and of those with nonmetastatic disease at diagnosis, 20%–25% will relapse and have metachronous metastatic pulmonary disease. 67 CT scans have demonstrated that they are unreliable with regard to underestimating the number of metastatic pulmonary nodules. 68 The most important prognostic indicator in metastatic osteosarcoma remains the complete surgical resection of disease. 69 Thoracoscopic resection in addition to digital palpation through a limited thoracotomy to provide confirmation of the presence of the nodule when it was not visualized has been performed with 100% diagnostic accuracy. 70 However, because complete resection of all disease is required for potential cure, it is advocated that metastasectomy be performed for metachronous pulmonary metastasis through the traditional posterolateral thoracotomy in order to allow thorough palpation. 67
Patients with osteosarcoma who presented with pulmonary nodules on CT were resected thoracoscopically in only 43% of cases and required conversion to open thoracotomy due to nonvisualization of the lesion, visualization of more lesions than predicted by CT, anatomic constraints due to apical metastasis with wide implantation, and pleural adhesions. 71 Select patients with metachronous metastatic pulmonary disease who presented with a solitary pulmonary nodule on CT scan greater than 2 months after completion of therapy were found to have no additional malignant lesions at the time of thoracotomy for metastasectomy. 67 Therefore, thoracoscopy may prove beneficial in this selected group of patients in order to spare this group an open thoracotomy and its associated complications when these patients are at risk for requiring repeated thoracic surgeries. 67
Future Developments
Single-incision surgery allows chest surgery to be performed through a single access site that admits multiple working instruments.72,73 Experiences with pediatric single-incision thoracoscopic surgery for tumors is extremely rare.74–78 Single-incision thoracoscopic surgery has been described by using multiple trocars or unsheathed instruments passed through a single small incision.78,79 Surgical resection using single-incision thoracoscopic surgery was performed for spontaneous pneumothorax, wedge biopsies, and even more rarely, apical extrapulmonary neuroblastoma. 78
Significant challenges include cost,76,80 steep learning curve,76,81,82 lack of triangulation, and close instrument proximity,75–77,82,83 which is even more difficult in smaller children. 76 Operative times may be longer because of a myriad of factors, including the learning curve of the surgeon, operating room staff, and familiarity with instrumentation.83,84
Robotic surgery eliminates tremor and allows three-dimensional vision with a magnified view and the use of articulated instrumentation. 26 Mediastinal masses are proposed as “the golden indication” for robotic resection, and successful resection has been reported of malignant tumors such as neuroblastoma and mediastinal germ cell tumor, tumors of intermediate malignant potential, including ganglioneuroblastoma and lipoblastoma, and benign tumors, including ganglioneuromas and mature teratomas. In two reports, no open or thoracoscopic conversions were required.26,85 Limitations such as robot costs compared with standard thoracoscopy are thought to be the single most limiting factor in the use of this technology. 85 Additional factors are the limited hemithorax space in patients less than 2.5 kg, instrumentation size and availability, and learning curve. 85 Although most of the published robotic surgery experience comes from a single-institution review, the authors have demonstrated the safe and effective application of robotic technology in selected pediatric surgical oncology cases.
Conclusions
With ultra-advanced technologies, rapidly improving thoracoscopic training, and adherence to the fundamental oncologic principles in surgical technique, pediatric surgeons from across the world are able to selectively customize treatments for thoracic and mediastinal tumors in neonates, infants, and children. Technical obstacles limiting the use of a minimally invasive approach include the loss of tactile ability, steep learning curve, need for familiarity with advanced thoracoscopic techniques, and instrument-to-patient size discrepancy. A significant oncologic concern remains whether malignant tumors can be safely removed in these children given the risk of potential iatrogenic tumor cell dissemination and metastasis.
Appropriate patient selection and correct surgical indications for advanced surgical procedures in children with cancer are necessary to minimize the risk of surgical complications. 86 The impacts of selection bias with regard to the literature and of underreporting of unfavorable outcomes must all be taken into consideration when thoroughly evaluating this approach to pediatric thoracic solid tumors. Although the lack of evidence has hindered rapid and widespread use of MIS for pediatric thoracic cancer, it appears to be an evolving and promising approach. We surmise that data generated from multicenter trials will support the creation of pediatric-specific guidelines for the use of thoracoscopy in pediatric cancer and show that MIS is as safe and effective as open conventional surgery.
Footnotes
Disclosure Statement
No competing financial interests exist.
