Introduction

Laennec [1] first described pneumomediastinum as the presence of free air in the mediastinum. Its clinical symptoms are cough, dyspnoea, chest pain, and acute respiratory dysfunction, which may occur in severe cases. Tension pneumomediastinum (TPM), one of its severe complications, has been reported in only a few cases [2]. Pneumomediastinum, associated with severe hypoxia, tachycardia, metabolic acidosis, and high ventilation pressures, is characterized by clinically significant tension in the mediastinum [3]. Chest radiography shows that the cardiac silhouette appears flattened, indicating cardiac compression caused by tension pneumomediastinum, called the earth-heart sign [4], which is a typical imaging manifestation of tension pneumomediastinum. In addition to a flattened anterior cardiac contour, computed tomography (CT) scans can sometimes also show compression of the aorta, vena cava, or tracheobronchial tree. If left untreated, it may develop into a life-threatening condition such as cardiac tamponade [5]. Traditional treatments include tracheotomy [2], sternotomy [6], and mediastinotomy [7]. New therapies include parasternal approach drainage [1, 8,9,10] and posterior mediastinal drainage [11]. We reviewed the clinical data of 19 children with tension pneumomediastinum who underwent CT imaging-guided parasternal approach drainage and summarized our experience.

Patients and methods

The present study adhered to the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of our hospital. Additionally, written informed consent was obtained from the parents of the patients.

Patients

From June 2018 to February 2023, 27 children with tension pneumomediastinum were admitted to our hospital. The selection criteria were patients who accepted CT imaging-guided parasternal approach drainage. The exclusion criteria were as follows: (1) Patients received surgical drainage treatment in other hospitals. (2) Patients received traditional surgical approaches, including tracheotomy, sternotomy, and mediastinotomy. (3) Patients with incomplete clinical data and those lost to follow-up. Finally, 19 patients were included (Fig. 1). The indications for drainage were as follows [4]: (1) The patient needed mechanical ventilation due to acute respiratory deterioration and hemodynamic instability that was not improved after general treatment. (2) Imaging suggested significant compression of the heart, aorta, vena cava, or tracheobronchial tree.

Fig. 1
figure 1

CONSORT flow diagram of participants

Full size image

Surgical technique

The same paediatric thoracic surgical treatment group performed all procedures. The CT machine type was RevolutionCT; the average radiation dose of a single chest CT scan was approximately 1 mSv, and the radiation time was approximately 3 s. The puncture site and path were planned according to CT imaging and 3D reconstruction, combined with bony markers. The depth (D) of pneumomediastinum, the spacing (S) between the puncture point and the midline of sternum, and the angle (A) of needle insertion were measured based on CT imaging, with the central emphysema area as the puncture site, and the puncture site (P) was marked on the skin surface (Fig. 2).

Fig. 2
figure 2

2a: Plan the puncture path using preoperative chest computed tomography. Extensive mediastinal tension emphysema compressing the mediastinal vessels and both lungs, the trachea is pushed posteriorly. Extensive lateral subcutaneous emphysema. 2b: Skin surface markers of the puncture site. A: The angle of needle insertion; D: The depth of pneumomediastinum; S: The spacing between the puncture site and the sternum midline. P: The puncture site

Full size image

Sedation method: Midazolam was slowly infused intravenously at an initial dose of 0.05–0.1 mg/kg and then continuously infused intravenously at a dose of 1–2 µg/kg/min to maintain sedation. After sufficiently sedating the patient, 1% lidocaine was used for local anaesthesia. A modified Seldinger technique was used [12]. Mediastinal puncture was performed using a 20-gauge needle attached to a 5-mL syringe to avoid significant damage. While applying negative pressure, insert the needle along the selected puncture site along the upper margin of the rib. Stop inserting the needle when air or fluid is obtained. Fix the needle in position and remove the syringe. Insert the guidewire along the needle core until it is deep enough into the mediastinum. Insert the dilator over the guide wire and carefully twist it to dilate the skin, muscles, and pleura. Once dilation is complete, remove the dilator, keeping the guidewire in place. Thread the caudal fibre catheter over the guidewire. After successful catheterization, fix the catheter to the skin around the puncture site with sutures, and connect the catheter to a sealed negative pressure water seal bottle. For drain size, we prefer a small chest tube (8–14 Fr) recommended by the Danish Pulmonary Society (DLS) [13]. In general, 6–8 Fr pigtail catheters were used for patients weighing < 10 kg, and 10 Fr catheters were used for patients weighing > 10 kg and adjusted to the situation.

Postoperative bedside X-ray was used to determine catheter location and evaluate efficacy. The patient’s radiation exposure did not increase. Three of them underwent chest CT because of pneumonia progression, and a satisfactory catheter location was seen from their CT images (Fig. 3). The response to treatment was measured by comparing images and ventilatory parameters before and after catheter drainage and by assessing each patient’s clinical outcome.

Fig. 3
figure 3

3a: Postoperative chest CT showed a satisfactory location of the pigtail catheter in the anterior mediastinum. (→). 3b: The place to insert the pigtail catheter.(▲)

Full size image

Statistical analysis

Continuous data are presented as the mean ± standard deviation and range, and the categorical variables are presented as frequencies (%). Clinical parameters are shown in Table 1. SPSS (Windows version 26.0 IBM Co, Armonk, NY, USA) was used for all statistical analyses.

Table 1 Demographical and clinical characteristics of the patients
Full size table

Results

Medical history and diagnosis

A total of 19 children (10 males and 9 females) with a mean age of 3.1 ± 3.4 (0.1–11) years and weight of 15 ± 9.1 (2.7–38) kg were recruited in this study. Primary diseases: Seven patients (36.8%) had pneumonia and presented severe coughing and shortness of breath. Two patients had COVID-19, 2 patients had bronchial asthma, 4 patients had a foreign body in the trachea or oesophagus, and 2 patients had blunt chest trauma; 1 patient had a case occurring secondary to thoracoscopic segmentectomy, and 1 was receiving chemotherapy. Fifteen patients (78.9%) required ventilator support, and the other 4 patients (21.9%) required only oxygen. All patients required treatment with broad-spectrum antibiotics. All patients were diagnosed by chest CT scan; 11 additionally underwent flexible bronchoscopy, and 4 underwent oesophagoscopy. Our data showed that 52.9% of patients with tension pneumomediastinum had associated pneumothorax, and 82.4% had associated subcutaneous emphysema (Table 1).

Management of drainage tube

The mean procedure time was 11.8 ± 2.4 min (range 8 to 15 min), the mean drainage time was 6.7 ± 3.4 days (range 2 to 13 days), and the mean hospitalization time after the operation was 12.5 ± 4.6 days. Parasternal approach drainage was performed in all patients, with 9 performed on the left and 10 on the right. Of the 19 drains, 12 were 10 Fr (63.2%), 6 were 6–8 Fr (31.6%), and only one was 4 Fr. Ten patients (52.6%) had unilateral or bilateral pneumothorax, of whom 4 underwent additional thoracentesis drainage.

Evaluate of the efficacy of parasternal approach drainage

We recorded preoperative clinical parameters, including heart rate (HR), blood pressure (BP), blood oxygen saturation (SPO2), and fraction of inspiration O2 (FIO2), which all improved significantly after surgery. The ventilator setting decreased from the preoperative period, including positive end-expiratory pressure (PEEP) and peak inspiratory pressure (PIP). There were no severe complications, such as haemothorax, empyema, or damage to the lung parenchyma or mediastinal structure. Drainage was effective in all patients as assessed by comparing images and ventilatory parameters, and no additional surgical treatment was needed. There was no recurrence during the follow-up of more than 2 months. (Table 2)

Table 2 Pneumomediastinum associated symptoms and clinical outcomes
Full size table

Discussion

Pathophysiology of pneumomediastinum

Pneumomediastinum may occur spontaneously or as a result of other underlying conditions. Spontaneous pneumomediastinum occurs when air from ruptured alveoli travels through the pulmonary interstitium and bronchovascular sheaths towards the pulmonary hila and into the mediastinum. This phenomenon was first described by Macklin et al. in 1944 and is known as the Macklin effect [14]. Conversely, secondary pneumomediastinum is commonly associated with trauma, such as injuries to the oesophagus or bronchi. When there is a sudden increase in mediastinal pressure, the pleura in the mediastinum may rupture, resulting in pneumothorax. The air then spreads to the neck and chest through the loose subcutaneous layer, causing extensive subcutaneous pneumatosis. As a result, patients with tension pneumomediastinum often present with concurrent pneumothorax and subcutaneous emphysema. Our data show that 52.9% of patients presented with pneumothorax, and 82.4% had subcutaneous emphysema.

Pneumomediastinum associated with COVID-19

Pneumomediastinum is a complication of COVID-19 that has been reported in children [15, 16]. The mechanisms would be alveolar injury secondary to viral infection combined with the rupture of the alveolar wall due to the increased pressure resulting from cough. Mechanical ventilation and air pressure injury are also associated with the development of pneumomediastinum [17]. COVID-19 usually presents as a clinically uncomplicated process in children. In recent data, 8% of children with severe acute respiratory distress syndrome require intensive care, only 1% of whom may require ECMO [18]. Once associated with pneumomediastinum, pneumothorax, and subcutaneous emphysema, the lengths of ICU and hospital stay increase, and the mortality rate also increases [19]. Early drainage can decrease the associated complication and mortality rates [15]. Two cases of COVID-19 causing acute respiratory distress syndrome (ARDS) and TPM have been reported. Both patients had severe ARDS with secondary pulmonary infection requiring invasive ventilation and moderate PEEP. Both required vasopressors to maintain hemodynamics and allow mediastinal drainage, and one patient received extracorporeal membrane oxygenation. Fortunately, both patients recovered and were discharged from the hospital.

The treatment focuses on cause identification and surgical drainage

Many conditions can lead to alveolar ruptures, such as bronchial asthma, severe coughing or vomiting, and other activities associated with Valsalva modulation [14]. The main risk factor for developing pneumomediastinum in hospitalized children is mechanical ventilation, and inappropriate airway pressure may lead to alveolar rupture. Pneumomediastinum may be an ominous sign of damage to mediastinal organs, including the trachea or oesophagus, and treatment should focus on identifying the cause [20, 21]. In our cases, 4 patients had a history of foreign bodies in the trachea or oesophagus, 2 had blunt chest trauma, 1 had lobectomy, and one received chemotherapy. Treatment of patients with tension pneumomediastinum should be started before cardiac tamponade develops [5]. In our case, a patient received foreign body removal by gastroscopy. Because the symptoms of pneumomediastinum were not obvious, the patient was not diagnosed early nor examined until the blood pressure decreased on the second day after gastroscopy. We suggest that the possibility of pneumomediastinum should be considered in patients with the risk factors mentioned above. Traditional surgical approaches include tracheotomy [4], sternotomy [6], and mediastinotomy [7]. However, parasternal approach mediastinal puncture and drainage is a good option for patients at higher risk for severe respiratory and cardiovascular disease [1, 8,9,10].

CT is an essential means for diagnosing and treating pneumomediastinum

CT is the gold standard for diagnosing pneumomediastinum because of its higher sensitivity and specificity than X-ray [10]. In addition to showing flattening of the anterior cardiac contour, CT scans can sometimes also show compression of the aorta, vena cava, or tracheobronchial tree [4]. CT is essential to identify the source of pneumomediastinum, thus avoiding unnecessary invasive procedures. There are few reports of mediastinal puncture guided by ultrasound and fluoroscopy [5]. The parasternal long-axis and short-axis views of the heart were difficult to ascertain due to an air artefact that seemed to coincide with respiratory variation; therefore, ultrasound is not recommended [22]. In contrast, CT imaging guidance allows for more precise localization and planning of the safest path to avoid damage to blood vessels, lung parenchyma, and other vital structures. Our patients were successfully catheterized under CT imaging guidance, and no severe complications occurred.

Advantages of pigtail catheters: safe, minimally invasive, and effective

The pigtail catheter, known as the Fuhrman catheter [23], is a spiral, single-lumen polyurethane catheter with a size ranging from 5 to 12 Fr. Unlike traditional chest tubes, pigtail catheters are not placed via blunt chest wall dissection but through a modified Seldinger technique [11]. The technique is simple and easy to master, with an average operating time of 11.8 ± 2.4 min. A fine needle was used to puncture the mediastinum, which prevented the mediastinal organs from severe damage. The head of the catheter is automatically bent into a pigtail shape so that the stimulation to the mediastinal organs is slight. We used a suture to fix the catheter, which can effectively reduce the incidence of catheter detachment caused by traction. A catheter with the appropriate calibre is usually selected according to age, weight, and disease state. We prefer to place a small chest tube (8–14 Fr) as recommended by the Danish Pulmonary Society (DLS) [13]. In general, 6–8 Fr pigtail catheters were used for patients weighing < 10 kg, and 10 Fr catheters were used for patients weighing > 10 kg and adjusted to the situation. Of 19 drains, 12 were 10 Fr (63.2%), 6 were 6–8 Fr (31.6%) and only one was 4 Fr. It is necessary to have multiple side holes to avoid mediastinal connective tissue blocking the side holes [8]. Roberts, J. S. et al. [24] reported a 20% rate of drain complications, including drainage failure, dislocation or kinks, loss of ventilation fluid, empyema, and drain disconnections. However, no related complications were observed in our study.

Limitations of the study

Despite the success of 19 cases, this study has many limitations. Although pneumomediastinum occurs mainly in the anterior mediastinum, the approach is inappropriate for posterior pneumomediastinum drainage. It was a single-centre, small-sample study, retrospective analysis, and lacked a control group. Therefore, more cases are needed in the future to confirm the current findings using prospective, comparative studies.

Conclusion

CT imaging-guided parasternal approach drainage is safe and feasible for children with tension pneumomediastinum.