The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of Beijing Shijitan Hospital Affiliated to Capital Medical University Hospital, and informed consent was waived because the study was a retrospective review. Between January 2009 and December 2021, 295 consecutive patients with suspected lymphatic PB and chylothorax admitted to Shijitan Hospital were retrospectively analysed.

The inclusion and exclusion criteria are shown in Figure 1. Patients without CTL data (n = 3) were excluded. The diagnostic criteria for lymphatic PB are the presence of bronchial casts and the abnormal development of lymphatic vessels. Chylothorax patients met at least one of the following criteria: (1) Had a positive chylous test in the effusion; (2) had a triglyceride level > 1.25 mmol/L in the effusion; (3) had a DLG showing contrast medium in the pleural cavity; and (4) had nuclear imaging showing contrast medium entry into the pleural space[18,19]. Patients who met the following criteria were selected for the study: Patients who were clinically diagnosed with lymphatic PB and primary chylothorax. The exclusion criteria were as follows: (1) Had chylothorax or PB caused by secondary factors (tumour, infection, trauma, surgery); (2) had chylous ascites; and (3) did not undergo DLG or CTL. The clinical and CTL data of 71 patients (20 with lymphatic PB, 41 with primary chylothorax, and 10 with lymphatic PB with primary chylothorax) were collected retrospectively.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Flow diagram of study cohort. DLG: Direct lymphangiography; PB: Plastic bronchitis; CTL: Computed tomography lymphangiography.

The main demographic and clinical variables of the patients in the three groups are shown in Table 1.

In our study, 61 of 71 patients underwent thoracic duct outlet surgery. Twenty patients in group A, 33 patients in group B, and 8 patients in group C underwent this operation.

Before the lower extremity DLG, consent was obtained from the patients or their guardians. DLG was performed using a GE Innova 2000-IQ DSA machine (AXIOM, Siemens, Erlangen, Germany). The patient was placed in the supine position. The surgeon punctured the skin between the first and second toes of the healthy or less oedematous side of the foot. Methylene blue dye (2.5% Patent Blue V dye; 1 mL, Guerbet Laboratories, Aulnay-Sous-Bois, France) and 1 mL of 2% lidocaine (1:1) were injected into the skin. Subsequently, the superficial lymphatic vessels were examined under a microscope and injected via fine needle puncture with 8-20 mL of superfluid contrast agent (Lipiodol; Guerbet Laboratories, Villepinte, France). The reflex of the contrast agent (lipiodol) along the lymphatic vessels can be observed dynamically. Depending on the patient’s age and the development of lymphatic vessels, the injection time was approximately 2 h. In normal people, lipiodol cannot enter tissues or lesions through the lymphatic vessel wall and can only be removed through the thoracic duct. The DLG lasted approximately 5-6 h.

All patients underwent CTL within 30 min-2 h after DLG. CT (Somatom Sensation Cardiac 16, Siemens Healthcare; Brilliance iCT, Philips Medical Health Care, Best, the Netherlands) scans were performed bilaterally from the inferior border of the thyroid cartilage in the neck to the lower border of the lungs. Patients were asked to hold their breath after deep inhalation during the scan, and a CT scan was performed at the end of inhalation. After the scan, the patients breathed freely. The scanning parameters were as follows: Tube voltage of 120 kVp, collimation of 1 mm, pitch of 1, and tube current of 110 mAs. After scanning, the raw data were transferred to the workstation for three-dimensional reconstruction with a layer thickness of 2 mm and a layer spacing of 2 mm.

The patient was placed in the supine position and under general anaesthesia. The patient's shoulders are padded with pillows so that the head is tilted slightly to the right. After disinfection of the neck area with iodine, a sterile towel, a medium sheet, and a small opening sheet were laid out. A transverse skin incision is made from the left sternoclavicular joint to the left external jugular vein between the body projection lines, approximately one finger above the clavicle. Under the microscope (12.5 /8-10 magnification), the subcutaneous layer, the broad cervical muscle, and the left sternocleidomastoid muscle were cut with an electric knife. The cut edge was treated with electrocoagulation to stop the bleeding. The left jugular vein angle (located under the clavicle) was revealed by posterior dissection, and the thoracic duct was explored and released from the obstruction. The wound surface was checked for obvious blood and fluid spillage, and sodium hyaluronate was injected locally to prevent adhesions. The gauze and instruments were counted, and silk sutures were interrupted to close the sternocleidomastoid dissection, the cervical brevis layer, and the subcutaneous layer. The skin incision was glued with tissue glue, and the wound was covered with a sterile dressing. At the end of the operation, the tracheal tube and gastric tube were removed upon awakening from anaesthesia.

All 71 patients underwent DLG successfully, and the quality of the CTL images met the criteria for disease assessment. All CTL images were double-blind reviewed by two radiologists (diagnostic radiologists with over 15 years of experience).

After the contrast agent (lipiodol) has been injected into the superficial lymphatic vessels of the unilateral foot, the lipiodol follows the lymphatic fluid up to the lumbar trunk and converges into the cisterna chyli, then into the thoracic duct, and finally into the blood at the angle of the left jugular vein. During this process, the distribution of lipiodol deposits in any other part of the body, such as the diaphragm, peritracheal region, pericardium, pleura, and hilum, is considered abnormal. Such deposits represent dilated lymphatic vessel hyperplasia, reflux, or fistula.

CTL and conventional chest CT were performed to assess the following features: (1) Abnormal distribution of lipiodol in the peritracheal region, pleura, pericardium, hilum, and thoracic duct outlet; and (2) ground-glass opacity, atelectasis, interlobular septal thickening, bronchovascular bundle thickening, pleural effusion, and mediastinal and extrapleural soft tissue changes.

Statistical analysis was performed by using software (SPSS, version 26.0; IBM, Armonk, NY, United States). Normally distributed quantitative data are expressed as the mean ± SD, and nonnormally distributed continuous variables are expressed as the median (25^th-75^th percentiles). Normally distributed and nonnormally distributed variables were evaluated by using one-way analysis of variance and the Kruskal-Wallis H test, respectively. The chi-square test or Fisher's exact test was used to compare the differences in abnormal CTL lipiodol deposition and abnormal changes in the lungs among the three groups. The Bonferroni correction was used for pairwise comparisons when the overall difference was statistically significant. Statistical significance was established at P values less than 0.05 for overall and pairwise adjusted tests of significance.

The data of 71 patients (mean age: 35 ± 18 years; age range: 1-73 years; 32 men) with a diagnosis of lymphatic PB (n = 20), primary chylothorax (n = 41), or lymphatic PB with primary chylothorax (n = 10) were reviewed. For further characteristics of the patients, see Table 1.

All DLG examinations (100%) were technically successful at visualizing anatomic variations or lymph reflux.

In 1 patient in group A, the lipiodol did not continue upward at the first lumbar vertebra. In the remaining 19 patients, reflux was observed in 3 patients in the left bronchomediastinal lymphatic trunk, 5 patients in the bilateral bronchomediastinal lymphatic trunk, 3 patients in the left subclavian lymphatic trunk, 1 patient in the right subclavian lymphatic trunk, 3 patients in the bilateral subclavian lymphatic trunk, 2 patients in the left jugular lymphatic trunk and 4 patients in the bilateral jugular lymphatic trunk.

In group B, 4 patients did not experience continued upward movement of lipiodol at the level of the 10^th thoracic vertebra to the first lumbar vertebra. Among the remaining 37 patients, 7 had reflux in the left bronchomediastinal lymphatic trunk, 15 had reflux in the left subclavian lymphatic trunk, 1 had reflux in the bilateral subclavian lymphatic trunk, 14 had reflux in the left jugular lymphatic trunk, and 2 had reflux in the bilateral jugular lymphatic trunk. During lymphangiography, lipiodol leakage into the pleural cavity was observed in 2 patients.

In group C, 1 patient experienced lipiodol obstruction at the level of the second lumbar vertebra, and 1 patient experienced lipiodol obstruction at the level of the sixth thoracic vertebra. In the remaining 8 patients, reflux was observed in 2 patients with a left bronchomediastinal lymphatic trunk, 5 patients with a left subclavian lymphatic trunk, 4 patients with a left jugular lymphatic trunk, and 1 patient with a bilateral jugular lymphatic trunk.

The abnormal deposition of lipiodol in patients in groups A, B and C was as follows: Terminal thoracic duct in 14, 33 and 8 patients (P = 0.674); peritracheal region in 18, 15 and 8 patients (P < 0.001); pleura in 6, 33 and 9 patients (P < 0.001); pericardium in 6, 6 and 4 patients (P = 0.134); and hilum in 16, 11 and 7 patients (P < 0.001) (Figures 2-4).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 A 34-year-old woman with primary chylothorax. A: The axial computed tomography lymphangiography (CTL) shows abnormal contrast medium accumulation in the thoracic duct outlet (long white arrow); B: A nonenhanced axial CT scan shows massive bilateral pleural effusions (black asterisk). Compressive atelectasis can be seen in both lungs (white asterisk); C: The CTL in the same plane as B shows abnormal contrast medium accumulation in the left pleura (black arrowhead), bilateral lung tissue (white arrowhead).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 A 16-year-old woman with primary chylothorax. A: The contrast-enhanced thoracic axial mediastinal window shows diffuse mediastinal soft-tissue infiltration (long white arrow), right pleural effusion (short white arrow), left pleural thickening (white arrowhead); B: The computed tomography lymphangiography in the same plane as A shows abnormal contrast medium scattered in surrounding mediastinum (short black arrow).

Open in New Tab   Full Size Figure   Download Figure

Figure 4 A 28-year-old man with plastic bronchitis. The axial omputed tomography lymphangiography shows an abnormal contrast medium surrounding the right main bronchus (long white arrow), right hilar lung (short white arrow).

The abnormalities on conventional chest CT in the three groups were as follows: Ground-glass opacity in 19, 18 and 8 patients (P < 0.001); atelectasis in 4, 26 and 7 patients (P = 0.003); interlobular septal thickening in 12, 11 and 3 patients (P = 0.037); bronchovascular bundle thickening in 14, 6 and 4 patients (P < 0.001); mediastinal confined speckle or tubular shadow in 14, 14, and 7 patients (P = 0.012); diffuse mediastinal changes in 6, 5, and 3 patients (P = 0.146); cystic hypodense foci in the axilla in 2, 6, and 2 patients (P = 0.796); and cystic hypodense foci in the chest wall in 0, 2, and 2 patients (P > 0.999) (Figure 5).

Open in New Tab   Full Size Figure   Download Figure

Figure 5 A 41-year-old woman with plastic bronchitis and primary chylothorax. A and B: The thoracic axial lung window shows interlobular septal thickening (long black arrow), bronchovascular bundles thickening (short black arrow), diffuse ground-glass opacities in pulmonary parenchyma (black asterisk).

The proportions of imaging features in the three groups of patients are detailed in Table 2.

There were significant differences between groups A and B in the following presentations: Lipiodol deposition in the peritracheal region (P < 0.001), lipiodol deposition in the pleura (P < 0.001), lipiodol deposition in the hilum (P < 0.001), ground-glass opacity (P < 0.001), atelectasis (P < 0.001), interlobular septal thickening (P = 0.036), and bronchovascular bundle thickening (P < 0.001). These significant differences were also detected between groups A and C in terms of the following manifestations: Lipiodol deposition on the pleura (P = 0.006) and atelectasis (P = 0.045).

The pairwise comparison of patients between the three groups is shown in Table 2.

In our study, 61 of 71 patients underwent thoracic duct outlet surgery. Twenty patients in group A, 33 patients in group B, and 8 patients in group C underwent this operation. Among them, 12 patients in group A underwent chest CT reexamination 3-19 d after the operation, and the signs were alleviated in 10 patients, similar in 1 patient, and aggravated in 1 patient. In group B, 8 patients were reexamined with chest CT 4-10 d after the operation, and the chest CT signs were alleviated in 3 patients, similar in 1 patient, and aggravated in 4 patients. Twenty-seven patients in group C were reexamined via chest CT 5-9 d after the operation; 13 patients were relieved, 9 patients were similar, and 5 patients were worse than before.