Video-assisted thoracoscopic surgery (VATS) lobectomy in children is technically more challenging than in adults for the following reasons. First, maintaining successful single-lung ventilation in children is often difficult. The smaller diameter and the short length of the airways demand specific modalities and technique to achieve adequate one-lung anesthesia. Even a slight position change of the patient or traction of the lung in the operating field can cause displacement of the endotracheal tube, which causes single-lung–ventilation failure.1 Second, the relatively small thoracic cavities and the narrow intercostal spaces of children make it difficult to apply large-sized conventional endoscopic surgical devices that are designed for adult patients.2
For these reasons, there have been only small numbers of reports for VATS lobectomy in the pediatric population. The aim of this study was to evaluate the safety and the efficacy of VATS lobectomy in children by comparing the clinical outcomes of a VATS group and a conventional thoracotomy group. In addition, we tried to evaluate the factors related to thoracotomy conversion from the failed VATS approach.
From May 2005 to October 2010, a retrospective review was done for 65 pediatric patients who underwent lobectomies in Seoul National University Children’s Hospital. Thirty-seven patients underwent VATS exploration for lobectomy attempt, and the other 28 patients underwent lobectomy by conventional thoracotomy approach. Surgeries were performed by three surgeons. Two surgeons with more than 300 accumulated cases of VATS lobectomy experience in adults preferred to apply the VATS, but the remaining one surgeon performed lobectomies exclusively by thoracotomy approach. Three incisions were made in an inverted triangular fashion for camera and instruments, with the lowest apex of the triangle being the incision for a camera port (Fig. 1). For the camera, 30-degree–angled fiber-optic thoracoscope that has a diameter of 5 or 10 mm was used, depending on the size of the patient. The incision made on the fourth or the fifth intercostal space along the anterior axillary line was used for utility purpose, including endoscopic stapler passage. This utility incision was 1.5 to 3 cm to accommodate the endoscopic stapler. Various endoscopic instruments and conventional instruments were used. For small vessels, a Harmonic scalpel (Ethicon Endo-surgery, Inc., Cincinnati, OH USA) was used for vessel sealing and division. For major pulmonary vessels, the lung parenchyma and the bronchus, 45-mm Endo-GIA staples (Covidien, Mansfield, MA USA) were used. Incomplete fissures were dissected either with energy devices or by applying staples.
For single-lung ventilation, three kinds of modalities were used: double-lumen endotracheal tube; single-lumen endotracheal tube with an endobronchial blocker; and direct endobronchial intubation into the left main bronchus using a conventional single-lumen endotracheal tube, which was used for right-sided lung procedures. Selection of the single-lung ventilation modality was decided on the basis of the size of the patient’s airway. If the patient had an airway large enough to accommodate the double-lumen endotracheal tube, double-lumen endotracheal intubation was preferred. When the patient had a small-sized airway that could not accommodate a double-lumen endotracheal tube, either the single-lumen endotracheal tube with a blocker or direct endobronchial intubation was used. In cases of poor thoracoscopic view due to inadequate deflation of the lung, CO2 gas insufflation was used. In cases of pulmonary sequestration, division of the aberrant systemic arterial branch that originates from the aorta was achieved by applying vascular clips on the proximal site first and then dividing the distal site with a vascular stapler (Fig. 2) to prevent risk of bleeding from the stapling site.
Preoperative clinical variables and perioperative outcomes were analyzed. For statistical analysis of the clinical data of the two groups, SPSS 18.0 (IBM SPSS Statistics, USA) was used. Student t test and Fisher exact test were used for the analyses.
The patients’ characteristics are listed in Table 1. The median age and body weight at the time of operation were higher in the VATS group than in the thoracotomy group. These were statistically different (P < 0.001 and P < 0.001). However, there was no difference in sex (P = 1.000), history of pneumonia (P = 0.804), location of the lesion (P = 0.185), and the diagnoses of lung disease (P = 0.194) between the two groups. In both groups, congenital cystic adenomatoid malformation (CCAM) was the most common diagnosis, which occupied 60% (22/37 cases) and 75% (21/28 cases) in the VATS and the thoracotomy group, respectively. There was a difference in single-lung–ventilation modality type between the two groups (P < 0.001). In the VATS group, single-lumen endotracheal tube with a blocker was the most common modality, which was used in 65% (24/37) of the cases. In the thoracotomy group, direct endobronchial intubation was the most common modality used in 50% (14/28) of the cases. There were three single-lung–ventilation failures (8%) in the VATS group, and all of them were converted to thoracotomy.
There was no in-hospital mortality in both groups. The mean chest tube indwelling days and the mean postoperative hospital days were not statistically different between the two groups (P = 0.198 and P = 0.179). Incidence of postoperative morbidities were 22% (8/37) and 29% (8/28) in the VATS group and the thoracotomy group respectively, which showed no statistical difference (P = 0.570). There was one case of bleeding, which required reoperation in the VATS group. When we re-explored the patient thoracoscopically, the site of the bleeding was found at the trochar insertion site, which was managed easily (Table 2).
In the aspect of efficacy, the success rate of the VATS lobectomy without converting to thoracotomy was 78% (29/37 cases). Eight cases were converted to thoracotomy. Three conversions were due to single-lung–ventilation failure, four were due to severe adhesion in the pleural cavity or anatomical variations, and the remaining one case was due to injury of the pulmonary vein during the procedure.
With accumulation of experience, the annual average operation time has decreased as years passed by, which started from 242 minutes in 2005 to 147 minutes in 2006, 161 minutes in 2007, 140 minutes in 2008, 170 minutes in 2009, and 119 minutes in 2010.
Univariate analysis was performed to evaluate the risk factors related to thoracotomy conversion from the VATS approach (Table 3). Age, sex, body weight, type of single-lung–ventilation modality, and location of the lesion were not significantly related to thoracotomy conversion. However, single-lung–ventilation failure (P = 0.007) and history of pneumonia (P = 0.001) were statistically significant risk factors of thoracotomy conversion.
The most common diseases requiring surgical resection of the lung for treatment in the pediatric population are congenital lung diseases such as CCAM and pulmonary sequestration, which are more frequently diagnosed by prenatal ultrasound examinations.3 Patients with these diseases experience recurrent upper respiratory tract infection and pneumonia more frequently than healthy children do. In our study, 49% of the VATS group and 54% of the thoracotomy group had history of pneumonia. These congenital lesions also have the potential risk for later malignant transformation. For these reasons, surgical resection of the lesion has been the most preferred treatment.4,5 The approach to the lesion by thoracotomy has been considered conventional, but as the VATS approach gained popularity in adults, the VATS approach has also been increasingly applied to the pediatric population, with the aid of advancing technologies and support from anesthesiologists. In our earlier experience, we performed surgical treatment when the diagnosis was made or when symptoms were present. Recently, however, the prenatal sonographic surveillance is getting popular, and small-sized, asymptomatic lesions are frequently detected even before birth. As a consequence, we needed to define the indications for appropriate timing for the surgery. We have previously demonstrated that older age and emergency operation were related to the risk for postoperative complications.6 On the basis of those results, we did not delay our surgical treatment of congenital lung disease and recommended to perform surgeries even in infants in an elective schedule. However, because of technical challenges in performing the VATS lobectomy in smaller patients, we opted to delay operation until at least 1 year of age if the patient was asymptomatic. This explains why the patients were older in the VATS group.
Achievement of effective single-lung ventilation is crucial for successful pediatric VATS lobectomy. Indeed, our results demonstrated that it was the most significant factor related to thoracotomy conversion. There were three cases (8%) of single-lung–ventilation failure in the VATS group, which led to conversion to thoracotomy. However, there were no anesthesia-related complications such as bronchial or tracheal injuries. Currently, the smallest available double-lumen endotracheal tube is 26F or 28F, depending on the manufacturer. As a consequence, many other alternative modalities have been developed to achieve successful one-lung ventilation in the pediatric population.7–9 We most commonly used single-lumen intubation with an endobronchial balloon blocker for small children, which was placed by fiber-optic bronchoscopic guidance.
Preoperative history of pneumonia was the second risk factor related to thoracotomy conversion. It may be due to inflammation and adhesion in a small thoracic cage. Our previous study revealed increasing episodes of pneumonia and minor complications when the operation was performed in older age.6
This series of results shows that VATS lobectomy in children is safe and effective, with its clinical outcomes comparable with those of the conventional thoracotomy approach. However, there were significant episodes of thoracotomy conversions. In the univariate analysis, we found that maintaining successful single-lung ventilation is crucial for successful VATS lobectomy in children. We also found that we have to be prepared for possible thoracotomy conversion if children have history of pneumonia. An anatomical variation could be another reason. In one case, a 5-year-old male patient weighing 15.5 kg who was diagnosed with left upper lobe CCAM, we found additional extralobar sequestration between the anterior and the apicoposterior segment of the left upper lobe, with the feeding artery originating from the proximal descending thoracic aorta. In addition, there were three distinct anomalous pulmonary vein branches draining the left upper lobe, which were difficult to handle thoracoscopically. In that case, the feeding artery was not clearly visible from the preoperative contrast-enhanced computed tomographic scan with a thickness of 5 mm. In another converted case, a 5-year-old female patient weighing 18.5 kg had left lower lobe CCAM. The left upper pulmonary vein was injured during left lower pulmonary vein division using an endoscopic stapler. The injured left upper pulmonary vein was successfully repaired after thoracotomy conversion.
We found the VATS lobectomy to be technically feasible and easier in relatively older and larger patients. As a consequence, for asymptomatic patients, we recommend to defer surgery until the first year of age. It may be reasonable because there is an article reporting a spontaneous resolution of the lesion during infancy.5 However, if the patients have symptoms or experience repetitive episodes of pneumonia, we recommend going ahead with surgery. Once one-lung ventilation is achieved, we recommend attempting the VATS approach as the primary approach. We found that the operation time became shorter and the conversion rates decreased after the learning curve. We may be able to apply this technique for younger and smaller patients in the near future, as our experience accumulates. Indeed, several pioneering centers reported successful VATS lobectomy in infants10 weighing less than 10 kg.
To conclude, VATS lobectomy in children is a safe and effective treatment modality, with results comparable with those of conventional lobectomy. In the univariate analysis, failure of single-lung–ventilation and history of pneumonia were two factors related to thoracotomy conversion.
1. Neustein SM. The use of bronchial blockers for providing one-lung ventilation. J Cardiothorac Vasc Anesth
. 2009; 23: 860–868.
2. Rothenberg SS. Thoracoscopic lung resection in children. J Pediatr Surg
. 2000; 35: 271–274.
3. dell’Agnola C, Tadini B, Mosca F, Colnaghi M, Wesley J. Advantages of prenatal diagnosis and early surgery for congenital cystic disease of the lung. J Perinat Med
. 1996; 24: 621–631.
4. Budorick NE, Pretorius DH, Leopold GR, Stamm ER. Spontaneous improvement of intrathoracic masses diagnosed in utero. J Ultrasound Med
. 1992; 11: 653–662.
5. MacGillivray TE, Harrison MR, Goldstein RB, Adzick NS. Disappearing fetal lung lesions. J Pediatr Surg
. 1993; 28: 1321–1324.
6. Kim YT, Kim JS, Park JD, Kang CH, Sung SW, Kim JH. Treatment of congenital cystic adenomatoid malformation—does resection in the early postnatal period increase surgical risk? Eur J Cardiothorac Surg
. 2005; 27: 658–661.
7. Camci E, Tuğrul M, Tuğrul ST, Sentürk M, Akpir K. Techniques and complications of one-lung ventilation in children with suppurative lung disease: experience in 15 cases. J Cardiothorac Vasc Anesth
. 2001; 15: 341–345.
8. Hammer GB, Manos SJ, Smith BM, Skarsgard ED, Brodsky JB. Single-lung ventilation in pediatric patients. Anesthesiology
. 1996; 84: 1503–1506.
9. Tobias JD. Anaesthetic implications of thoracoscopic surgery in children. Paediatr Anaesth
. 1999; 9: 103–110.
10. Rothenberg SS, Kuenzler KA, Middlesworth W, et al.. Thoracoscopic lobectomy in infants less than 10 kg with prenatally diagnosed cystic lung disease. J Laparoendosc Adv Surg Tech A
. 2011; 21: 181–184.
In adults, the video-assisted approach to thoracic disease (VATS) is not only accepted; in many situations, it is considered standard of care. For children, however, it has been rarely reported and rarely studied; as the authors state, it is technically demanding and few surgeons have the skill sets to perform a safe procedure.
Although this article has a low case volume and has a great deal of bias in its methods and presents no standards for safety and efficacy in the patient population, the authors are exploring the potential of VATS lobectomy in children, and, from this, there is a great deal to learn. As one would expect with the available technology, VATS is best performed in the larger and older child. The youngest patient in the thoracotomy group was 5 days old; we presume that that was the child who was 3.4 kg; yet, the youngest in the VATS group was 6 months and was 8.6 kg. The anatomical lobe to be resected and the indication were not a deterrent for VATS. There was a difference in the technique of single-lung ventilation; the patients who underwent thoracotomy were more likely to undergo selective bronchial intubation, likely being smaller than their VATS counterparts. The authors provide us with an additional “pearl,” which is, not surprisingly, that the inability to achieve single-lung ventilation and the presence of pneumonia increased the risk for VATS conversion to thoracotomy. These data have to be taken in context with the fact that there was no normal distribution in the VATS patient population because they seemed to be older and of larger size than the thoracotomy group.
The technique they describe and the demonstration that their surgical times have decreased over time should offer the potential to standardize their technique, allowing for a possible prospective trial. There is yet a great deal to learn from the use of VATS in this population. Because the case volume is low, it would be advisable and significantly more powerful to see this progress to a multi-institutional trial. We look forward to hearing more from this group in the future.