Surgical Intervention for Primary Spontaneous Pneumothorax in Pediatric Population: When and Why?

Introduction

Spontaneous pneumothorax in pediatric patients is a relatively uncommon disease, with reported incidence of 3.4 per 100,000 children. ¹ Spontaneous pneumothorax is defined as collection of air in the pleural space without apparent traumatic or iatrogenic mechanism. It can be classified into primary spontaneous pneumothorax (PSP), which occurs in patients without underlying lung disease; or secondary spontaneous pneumothorax in those with preexisting lung conditions such as asthma, congenital cystic adenomatoid malformation, cystic fibrosis, or connective tissue disease such as Marfan syndrome or Ehlers–Danlos syndrome.

The management strategy varies in different centers due to dearth of evidence-based pediatric guidelines. Extrapolation of adult clinical data has not been found to be applicable in the management of pediatric population. The British Thoracic Society (BTS) guidelines suggested aspiration for large PSP in stable adult patients and showed equivalent success to the insertion of large-bore thoracostomy tube. ² However, much lower success rate of aspiration in the management of pediatric PSP and higher recurrence rate in children treated with conservative management compared to adult results have been reported.^3,4 Video-assisted thoracic surgery (VATS) has become increasingly popular nowadays in the surgical management of pediatric thoracic conditions such as lung resections, mechanical pleurodesis for pneumothorax, decortication for empyema, and repair of esophageal atresia and trachea-esophageal fistula.^5,6 In this study, we reviewed our experience of thoracoscopic management of PSP in children and adolescents and identified risk factors associated with postoperative air leakage, prolonged hospital stay, and recurrence.

Materials and Methods

We performed a retrospective analysis of all patients younger than 18 years who had PSP and underwent surgical management in our institution between April 2008 and March 2015. Neonates and children with secondary pneumothorax due to traumatic, iatrogenic causes or underlying pulmonary or systemic diseases were excluded. Large PSP is defined by either more than 3 cm of apical interpleural distance as suggested by the American College of Chest Physicians or more than 2 cm of interpleural distance at the level of the hilum by the BTS guidelines on presentation chest radiograph.^2,7 Pneumothorax volume is calculated with Collins method taken into account of interpleural distances at apex, midpoint over upper half and lower half of collapsed lung. ⁸ Demographic data, interventions, and surgical outcomes were analyzed. Postoperative persistent air leakage was defined as failed removal of thoracostomy tube beyond 4 days due to persistent pneumothorax or reaccumulation of pneumothorax after removal of thoracostomy tube requiring reinsertion in same hospitalization. Postoperative recurrence was defined as occurrence of new ipsilateral pneumothorax after prior VATS management of pneumothorax.

Statistical analysis was performed with SPSS 21.0 (SPSS, Chicago, IL). Values were reported as mean (± standard deviation) or median (interquartile range). Continuous variables were compared with Mann–Whitney U test. Discrete variables were compared using Pearson chi-square test. Univariate and multivariate analyses of risk factors for postoperative persistent air leak and recurrence were performed. Statistical significance was defined by P < .05.

Results

A total of 92 patients with 110 thoracoscopic surgery for PSP were identified. The majority of patients were male (95%, n = 91) and the median age was 17 years (range: 14–18). Fifteen patients (15.6%) had smoking history. Fifty-three episodes of pneumothorax occurred on left side (48.2%), 46 episodes on right side (41.8%), and 11 episodes of bilateral pneumothorax (10%). The most common presenting symptom was pleuritic chest pain (92.7%), followed by dyspnea (49.1%). Radiologically, 81 episodes were large pneumothorax (73.6%). The median pneumothorax volume was 47.4% (28.5–83.2) calculated with Collins method. Abnormal radiographic features such as pleural thickening, irregularities, and adhesion were seen on presentation chest radiograph in 21 episodes (19.1%). Computed tomography (CT) of thorax was performed in four episodes (3.6%), in which patients presented with recurrent pneumothorax, tension pneumothorax, worrisome of bronchopleural fistula, and atypical chest radiograph showing loculated lower zone pneumothorax.

On initial management, oxygen supplement was given to all patients, while needle aspiration was performed in 13 episodes of pneumothorax (11.8%, 13/110). Thoracostomy tube insertion was performed in 90 episodes (81.8%, 90/110). The indications for surgery were failed nonoperative management with persistent air leakage in 32.7% (36/110), recurrent ipsilateral pneumothorax in 36.4% (40/110), first contralateral pneumothorax in 14.5% (16/110), simultaneous bilateral pneumothorax in 10% (11/110), significant hemopneumothorax in 5.5% (6/110), and high-risk occupation requiring frequent air travel in 1 patient.

All surgeries were performed thoracoscopically with no conversion. The median operation time was 60 minutes (range: 30–165). Bullae were identified in 101 thoracoscopy (91.8%) with stapled bullectomy performed. Mechanical pleurodesis with pleurectomy was performed in all, while additional chemical pleurodesis was performed in 50 episodes (45.5%) at the end of the thoracoscopic operation.

Median postoperative thoracostomy tube drainage was 3 days and median hospital stay after operation was 4 days. Persistent postoperative air leakage occurred in 16 patients (14.5%) and was associated with prolonged hospital stay (P < .01, mean difference of 5.6 days). Majority were treated with reinsertion of thoracostomy tube and chemical pleurodesis using oxytetracycline (Oxylim^®). Operation within 7 days of symptoms onset was associated with less postoperative air leakage (P = .04).

The median follow-up was 56.5 months (32.8–80.3). Postoperative recurrence occurred in 19 patients (17.3%) at mean time of 11 months. Bilateral pneumothorax has significantly more postoperative recurrence (P = .042) than unilateral pneumothorax. Pneumothorax with abnormal radiographic features on presentation chest radiograph such as pleural thickening, irregularities, and adhesion was associated with more postoperative air leakage (P = .001) and recurrence (P = .001) (Table 1). There was no operative mortality. Regarding the 19 recurrence episodes, 10 of them had conservative management with oxygen and monitoring, 9 of them had thoracostomy tube inserted, in which 5 of them had additional chemical pleurodesis and 3 required redo VATS. The pathology identified on redo VATS included blebs over previous staple line and lower lobe apical bulla. No further recurrence was found on follow-up.

Discussion

The age-adjusted incidence of PSP was 7.4 to 18 per 100,000 males per year and 1.2 to 6 per 100,000 for females.^9,10 The exact pathophysiology underlying PSP remains unclear. Familial clusters of PSP have been reported and described as autosomal dominantly inherited with variable penetrance.^11,12 Rupture of subpleural bullae has been postulated as the cause of PSP. Subpleural bullae were found in CT of the thorax in 89% of patients with PSP^13,14 and in 76%–100% of patients with PSP during surgery.^15–17 Some centers suggested routine CT of thorax in patients with PSP to detect underlying lung pathology and contralateral bullae.^13,14,18 Incidental finding of contralateral bullae on CT scan was reported in 17%–20% of children with PSP. However, the risk of development of contralateral PSP in these cases is unknown.^19,20 Over the years, there have been controversies regarding synchronous bilateral VATS on initial hospitalization or delayed intervention for contralateral incidental bullae until development of pneumothorax. In our institution, bilateral VATS were performed for children with bilateral pneumothorax on chest radiograph on presentation. To avoid unnecessary radiation, CT of thorax was not routinely performed unless patients had atypical presentation or chest radiographs suspicious of alternate pathologies. On follow-up, only 4 patients (4.3%) developed contralateral pneumothorax, in which 2 of them required contralateral VATS bullectomy and pleurodesis.

Nonoperative management of pediatric PSP includes supplemental oxygen, needle aspiration, and thoracostomy tube insertion. Chadha, et al. observed an increased rate of resolution of pneumothorax sized less than 30% when treated with higher concentrations of inhaled oxygen in adult patients. However, those with large pneumothorax did not gain extra benefit with higher inhaled oxygen concentration. ²¹ Aspiration of large PSP alone in pediatric population was found to be ineffective, in contrary to adult data, with failure rate up to 60%.^20,22,23 Therefore, in many centers, thoracostomy tube insertion has replaced needle aspiration as the initial nonoperative management of large pediatric PSP and surgical referral recommended for those with ongoing air leakage for more than 4 days.^3,19,24,25 In our center, we do not routinely perform aspiration for large PSP nowadays. According to prior local studies in Hong Kong, Hui et al. found that the successful rate of thoracostomy tube in managing PSP was 47%, however, their study did not include option of aspiration. ²⁶ Lee et al. found that needle aspiration and thoracostomy tube insertion were successful in 78% and 67%, respectively. ²⁷ However, both of these two studies did not specify the severity of PSP or proportions of patients having large PSP.

On the contrary, VATS management of PSP includes stapled bullectomy, pleurectomy, and pleurodesis either mechanically alone or with addition of chemical pleurodesis. In our center, the indications of VATS include failed nonoperative management with persistent air leakage, recurrent ipsilateral pneumothorax, first contralateral pneumothorax, simultaneous bilateral pneumothorax, and significant hemopneumothorax. Bullae were identified in 91.8% of PSP as the culprit of disease. Persistent postoperative air leakage requiring prolonged thoracostomy tube placement or reinsertion was associated with prolonged hospital stay. Factors associated with persistent postoperative air leakage include delayed VATS more than 1 week from symptoms onset, bilateral pneumothorax, and those with abnormal radiographic features on presentation such as pleural thickening, irregularities, and adhesion. These patients may therefore benefit from prolonged postoperative thoracostomy tube drainage and chemical pleurodesis.

In our study, postoperative recurrence rate was 17.3%, which is comparable to the reported rate varying from 16% to 28% in other institutions.^{22,23,28–30} Parents should be advised on seeking early medical attention in case of presence of symptoms of recurrence, which mostly occurred in the first year after prior episode of pneumothorax. Large-size pneumothorax was found to be a significant factor influencing postoperative recurrence in pediatric population by Chiu, et al. ²⁸ However, in our study, we did not find the correlation between recurrence and size of pneumothorax. Risk factors for recurrence include bilateral pneumothorax and those with abnormal radiographic features among our patients.

We acknowledge the limitations of our study regarding its retrospective nature. Moreover, many of these patients had large PSP and were referred to us as a tertiary surgical center after failed nonoperative management. Those with small PSP managed conservatively in other hospitals might be missed with our database.

In conclusion, early thoracoscopic stapled bullectomy and mechanical pleurodesis after thoracostomy tube insertion could be offered as a primary option for management of large PSP in pediatric population since most of these patients would have underlying bullae as the culprit of the disease. Patients with delayed operation more than 1 week from symptoms onset, bilateral pneumothorax, or abnormal radiographic features have a higher risk of postoperative air leakage, which clinicians should be aware of.