Lymphatic fluid is produced by the absorption of interstitial fluid generated from blood, which reaches capillaries through the distal lymphatic vessels. The distal lymphatic vessels become thicker as they merge, travel medially, and form a spiderweb-like lymphatic system by repeatedly crossing the lymph nodes. They complete the circulation of body fluids by returning them to veins through the thoracic duct [1]. Compared to the 17th century, when the physiology of the lymphatic system was studied vigorously, it currently does not receive sufficient medical attention despite its significant role. This is probably attributable to difficulties in surgical observation and manipulation of lymphatic vessels, let alone evaluation or imaging of their function, due to the minute size of the lymphatic vessels and the transparent characteristics of the circulating lymphatic fluid.

Pedal lymphangiography, developed by British vascular surgeon Kinmonth in 1952, was the first medical attempt to overcome this problem. He developed a method involving the surgical dissection of lymphatic vessels on the dorsal foot, injection of contrast agents, and subsequent observation through X-rays, leading to significant advancements in the diagnosis and treatment of various lymphatic diseases [2]. However, as the simple evaluation of lymph nodes through modern cross-sectional imaging methods such as ultrasonography and computed tomography (CT) became feasible, the technically demanding and time-consuming method of pedal lymphangiograhy quickly disappeared from the clinical field. Unfortunately, many of Kinmonth’s achievements in lymphatic disorders were cast aside along with the outdated technology. Therefore, pedal lymphangiography, owing to its difficulty, barely continued its existence and was limited to some specialized institutions.

Approximately 50 years later, when Constantine Cope, an interventional radiologist at the University of Pennsylvania, applied the Seldinger technique and embolization to the thoracic duct, a new field of study called lymphatic intervention emerged [3]. In addition, intranodal lymphangiography, which can be easily performed by any interventional radiologist, was introduced around 2010, providing an opportunity for lymphatic system intervention to be more available [4].

At the time when the present author began the lymphangiography procedure, the claim that Lipiodol, the agent injected during lymphangiography, might contribute to halting several lymphatic leakages became a topic of debate. Lipiodol has long been used in lymphangiography because it is not diluted in the lymphatic vessels and maintains the shape of a thickly contrasted oil droplet. In the meantime, the phenomenon that lymphatic leakage stopped after Lipiodol lymphangiography performed for diagnosis, was first reported in 2007 and since then, similar cases have been repeatedly reported by several researchers [5-10]. However, the mechanism through which Lipiodol prevents lymphatic leakage remains poorly understood. Most importantly, it was not possible to distinguish between spontaneous recovery and the impact of Lipiodol, since most of the reports were observations in a small number of patients without a control group. In addition, since it has been reported that the therapeutic effect of Lipiodol is rather low in massive lymphatic leakages that require active treatment, the development of a more effective treatment method is required, rather than merely expecting therapeutic effects that may not exist after performing Lipiodol lymphangiography.

Therefore, this study aimed to establish five hypotheses and analyze the diagnosis and treatment outcomes for each strategy. This will be achieved by comprehensively describing a series of verification processes for each hypothesis and presenting a systematic methodology for novel lymphatic intervention.

1) Hypothesis

Various embolization techniques using liquid embolic materials can prevent lymphatic leakage from damaged lymphatic vessels or nodes.

2) Background

As described in the introduction, despite some claims that injecting Lipiodol into the lymphatic system may treat lymphatic leakage, its effectiveness has been questioned in actual clinical practice. After a case report of treating chylous ascites by directly puncturing a lymphocele (a collection of lymphatic fluid leaked from a lymphatic vessel) with a needle and filling it with liquid embolic material in Japan in 2012 [12], similar trials were subsequently reported [13,14].

3) Methods

A retrospective observational study was conducted on 27 patients who underwent Lipiodol lymphangiography for postoperative lymphatic leakage at Asan Medical Center, Seoul National University Hospital, and National Cancer Center between August 2010 and January 2016.

A method presented in the existing literature report was used, by which leakage of lymphatic fluid from the damaged lymphatic vessels is blocked with a liquid embolic material (n-butyl cyanoacrylate [n-BCA], histoacryl glue) filling the lymphopseudoaneurysm that is immediately formed with surrounding tissue after the leakage of lymphatic fluid.

In the second method, after fluoroscopy-guided puncture of the lymph node located immediately before the lymphatic leakage site was confirmed by lymphangiography, the lymph node was embolized using the same liquid embolic material (Fig. 1).

Figure 1. A 61-year-old male patient underwent Lipiodol lymphangiography for a postoperative lymphocele after radical prostatectomy. (A) A volume rendering image from cone-beam computed tomography data of lymphangiography shows Lipiodol leaking into the lymphocele where a drainage catheter is installed through two bundles of lymphatic vessels (arrow and arrowhead) originating from a complex network of lymph nodes (dashed circle). (B) Only medial bundles of lymphatic vessels (arrowhead) were filled with glue (N-butyl cyanoacrylate: Lipiodol ratio=1:5) when the far most medial inguinal lymph node was accessed and embolized. (C) Glue was confirmed to reach the lateral bundles of lymphatic vessels (arrow) only after accessing the lymph node at a lower level in the digital subtraction angiography performed during glue injection. The drainage catheter could be removed eight days after the embolization.

4) Results

In the follow-up examination, spontaneous recovery without further embolization was achieved in 8 patients (50%) out of 16 patients after Lipiodol lymphangiography. In 16 patients, including 11 who underwent immediate embolization after lymphangiography and 5 who underwent delayed embolization, a total of 28 procedures of lymphopseudoaneurysm or lymph node embolization were performed. The technical, and clinical success rates were 89%, 94% in 16 patients who received embolization treatment. Overall clinical success rates including simple lymphangiography was 85% in 27 patients. Major complications were not observed.

5) Discussion

Lymphopseudoaneurysm embolization tends to be less effective when the surrounding tissues are not solid, and when the size of lymphopseudoaneurysm is large. Moreover, a large amount of embolic materials are required in these cases, thereby elevating the risk of foreign material infection [11]. To overcome these limitations, a new approach known as lymph node or interstitial lymphatic embolization has been introduced. This procedure involves embolizing the microlymph channels in the lymph node and leaking lymphatic vessels by injecting liquid embolic material, as opposed to a contrast agent, through the same route as intranodal lymphangiography. It was first described in 2016 in three studies, including the present study [11,15,16]. In particular, by selectively embolizing the closest upstream lymph node located just before the leakage site, lymphatic leakages can be treated while maximally preserving the normal lymphatic tissue. This approach bears similarities to arterial embolization procedures in which the bleeding arteries are selectively embolized.

Lymph node embolization can be performed at any site if the closest upstream lymph nodes are accessible. Most existing reports on the efficacy of lymph node embolization has been retrospective, lacking sufficient evidence. Further, well-designed prospective studies with large sample sizes are needed. However, the technical success of this procedure can be easily evaluated because the lymphatic vessels and nodes at the leakage site are permanently blocked by a fluoroscopically observable liquid embolus. If symptoms persist despite the presumable technical success, the following factors should be considered as potential causes of clinical failure: firstly, additional leakage sites might not be included in the scope of lymphangiography owing to the complexity of the lymphatic network, such as collateral passages; and secondly, there may be technically incomplete embolization. In such cases, additional lymphangiography and lymph node embolization can be performed to enhance clinical outcomes.

1) Retrospective observational study [17]

① Hypothesis

Lymph node embolization can be used to treat postoperative pelvic lymphoceles that are unresponsive to sclerotherapy.

② Background

Pelvic lymphocele is a common postoperative complication, reported in up to 20% of patients undergoing pelvic surgery accompanied by lymphadenectomy. Asymptomatic lymphoceles do not require treatment, but if symptoms are present, conservative measures are required with the insertion of drainage tubes. This approach can typically lead to the halting of lymphatic leakages in most cases within a few days. However, if simple drainage does not resolve this problem, sclerotherapy should be considered. Although sclerotherapy is a conventional treatment method, it is relatively safe and technically simple to perform. Nevertheless, in some cases of pelvic lymphocele, symptoms persist for a substantial duration despite conservative treatment or sclerotherapy.

③ Methods

This retrospective observational study included 40 lymphocele cases in 33 patients who underwent postoperative sclerotherapy or embolization for pelvic lymphocele at Seoul National University Hospital between 2009 and 2017. Lymph node embolization was performed for 24 lymphoceles in 19 patients, whereas sclerotherapy was performed for 16 lymphoceles in 14 patients.

④ Results

Lymph node embolization was associated with a higher clinical success rate than sclerotherapy (83.3% vs. 43.8%, respectively P=0.026). There were no statistically significant differences in the incidence of complications or the duration of treatment between the two groups. Based on these retrospective findings, the following prospective clinical trial was designed to demonstrate the advantage of lymph node embolization.

2) Randomized pilot study [18]

① Hypothesis

Lymph node embolization is more effective in treating postoperative pelvic lymphocele than sclerotherapy.

② Background

Based on the results of the aforementioned retrospective study, it was hypothesized that the clinical success rate of lymph node embolization is 80%, surpassing that of sclerotherapy (50%). A randomized clinical trial was conducted to test this hypothesis, targeting 88 patients for demonstration, with a statistical power of 0.80 and a significance level of 0.05.

③ Methods

Informed consent for participation in the clinical trial was obtained from patients who underwent percutaneous drainage for pelvic lymphocele following pelvic surgery at Seoul National University Hospital and National Cancer Center in August 2018. A 1:1 random assignment into the embolization or sclerotherapy treatment group was conducted. The outcomes of each treatment were compared and analyzed following the prescribed procedure.

④ Results

Owing to difficulties in patient recruitment at the participating hospitals, the trial was prematurely terminated after recruiting 11 patients. In the data analysis of these 11 patients, both treatment groups showed a 100% clinical recovery rate, with no difference in treatment efficacy. However, a greater number of procedures were required in the sclerotherapy group to achieve clinical success. Overall, both methods showed favorable outcomes and safety profiles for the treatment of postoperative pelvic lymphoceles.

3) A comprehensive review of retrospective observational studies and prospective clinical trials

Although it should be taken into account that the trial was terminated prematurely, the results of a randomized prospective clinical trial indicated that the treatment effect of embolization may not be inherently superior to that of sclerotherapy, which was inconsistent with the previous retrospective study [17,18]. There are several possible explanations for the discrepancy. Firstly, the possibility of detecting subtle differences might be diminished in this prospective trial because of the much smaller number of patients recruited than the initially targeted sample, resulting from early termination. Secondly, when considering that the clinical success rate of sclerotherapy for pelvic lymphocele in the prospective clinical trial was much higher than that in the retrospective study (100% vs. 43.8%), the difference could be partly attributed to the assumption that experienced operators were instructed to manage the drainage tube more faithfully and perform sclerotherapy according to a unified procedure. Unfortunately, in this retrospective study, refractory cases that did not respond favorably to drainage were selectively included; hence, the same refractory cases should have been included in the clinical trial. Unfortunately, due to challenges in recruiting patients and limitations in the study period, it was not feasible to apply a protocol with thorough confirmation. Consequently, it is assumed that several cases with simple pelvic lymphocele with a rapid recovery process, either with sclerotherapy or embolization, might have been included.

Despite the negative study outcomes in prospective study, it is believed that the conclusions obtained from the retrospective study remain valid. Specifically, in cases of refractory pelvic lymphocele, lymph node embolization might be helpful in the treatment of postoperative pelvic lymphocele. Since simple lymphoceles account for the majority of the cases and refractory pelvic lymphocele cannot be pre-determined, it could be currently suggested that a reasonable treatment strategy would involve applying embolization only when there is no therapeutic response to simple drainage or sclerotherapy, considering relatively high cost and requirements of high degree skills associated with embolization.

1) Hypothesis

The technical success rate of thoracic duct embolization can be improved with the assistance of a retrograde thoracic duct approach.

2) Background

Thoracic duct embolization is gradually emerging as the standard treatment for chylothorax after surgery, replacing prolonged conservative treatment or surgical thoracic duct ligation. However, the low technical success rate due to the difficulty in accessing the thoracic duct is a limiting factor for thoracic duct embolization. According to a recent systematic review, the treatment success rate of embolization in cases with successful thoracic duct cannulation was substantially high (94.2%); however, the success rate of thoracic duct cannulation itself remained relatively low at 63.1% [20].

3) Methods

A total of 45 patients who underwent Lipiodol lymphangiography prior to thoracic duct embolization for postoperative chylothorax, between February 2016 and November 2019 at Seoul National University Hospital, were retrospectively included. Transabdominal anterograde thoracic duct cannulation was attempted when feasible. In cases of failure, retrograde thoracic duct cannulation was attempted as a rescue (Fig. 2).

Figure 2. A 74-year-old male patient underwent Lipiodol lymphangiography for a persisting postoperative chylothorax (over 300 mL per day, lasting more than a week at medical treatment) after right upper lobectomy with mediastinal lymph node dissection. (A) Cisterna chyli is located behind the aorta, and the segment between the cisterna chyli and the upper part of the thoracic duct is a plexiform variation through which the catheter cannot pass. (B) A micro-guidewire was inserted into the thoracic duct retrogradely by fluoroscopically guided puncture of the cervical segment of the thoracic duct. (C) The lymphatic vessel leading to the right pulmonary hilum is selectively catheterized with a microcatheter, and a glue (N-butyl cyanoacrylate: Lipiodol ratio=1:3) was injected to embolize the vessel super-selectively while keeping the thoracic duct patent. The chylothorax stopped immediately, and the drainage tube could be removed three days after the procedure.

4) Results

Based on the lymphangiography results, thoracic duct cannulation was attempted in 40 patients, and the success rate of anterograde thoracic duct cannulation was 79% (31/39). In five of the eight patients with failed antegrade cannulation, retrograde thoracic duct cannulation was performed as a rescue remedy, and in one patient, retrograde thoracic duct cannulation was attempted from the beginning with an overall success rate of 93% (37/40). The chylothorax resolved within 2 weeks in 31 of 35 patients who underwent thoracic duct embolization. A major complication, bile peritonitis, occurred in one patient during transabdominal anterograde thoracic duct cannulation.

5) Discussion

Unlike conventional transabdominal antegrade thoracic duct cannulation, retrograde thoracic duct cannulation is a technique directed against the flow of lymph fluid in the thoracic duct. Since the first report by Mittleider et al. [21] in 2008, two major methods have been used for retrograde cannulation of the thoracic duct. The first method involves accessing the thoracic duct via a vein through the thoracic duct-vein junction, and the second method involves puncturing the thoracic duct near the neck under ultrasound guidance. The problem with the intravenous approach is the difficulty in predicting the success rate of the procedure in advance because of the abundant anatomical variation of duct-vein junction [22]. On the other hand, ultrasound-guided puncture of the thoracic duct in the neck poses difficulties because it is challenging to observe the thoracic duct using ultrasonography [23].

Most reports on retrograde thoracic duct access are retrospective studies that include a small number of patients, resulting in a low level of evidence for the success rate. However, it still holds significance as it offers an additional treatment option for patients with technical difficulties in transabdominal antegrade access to the thoracic duct, as those with severe obesity. Furthermore, it provides the technical basis for advanced lymphatic intervention, which is necessary for the treatment of diseases that cannot be resolved solely with general lymphatic intervention, such as refractory chylous ascites [19].

In 2021, the present author reported an enhanced overall technical success rate of thoracic duct access, reaching 93%, through the use of assistive retrograde thoracic duct access in European Radiology [19]. Since then, the success rate of 83% in anterograde access could be improved to 100% by incorporating retrograde access to the thoracic duct in 25 cases of traumatic chylothorax, by unifying the improved method of “fluoroscopy-guided” puncture of the thoracic duct near the neck [24].

1) Hypothesis

Non-traumatic chylothorax can be treated by understanding its pathology and addressing it accordingly.

2) Background

Unlike postoperative chylothorax, where the efficacy of thoracic duct embolization has been proven, the treatment success rate of thoracic duct embolization is much lower in cases of non-traumatic chylothorax unrelated to surgery. The treatment success rate drops to only 16% when leakage is not confirmed on lymphangiography, and in some cases, the symptoms may even worsen [26]. In fact, as thoracic duct embolization becomes more prevalent in Korea, there have been frequent cases of symptom exacerbation after performing thoracic duct embolization without a detailed examination, such as dynamic contrast-enhanced magnetic resonance lymphangiography (DCMRL), for non-traumatic chylothorax in institutions lacking expertise in lymphatic diseaser (Fig. 3). In this section, we aimed to subdivide non-traumatic chylothorax according to the pathophysiology shown in DCMRL and present a systematic approach.

Figure 3. (A) Non-traumatic chylothorax incidentally discovered on chest radiograph in a 29-year-old male. (B) A thoracic duct embolization was performed at another hospital, and the symptoms aggravated. (C) Dynamic contrast-enhanced Magnetic Resonance Lymphangiography performed at Seoul National University Hospital showed that the thoracic duct (dotted line) was occluded, and the contrast agent leaked into the thoracic cavity (arrowhead) through the collateral lymphatic vessel (arrows) in the retroperitoneal space. After selective lymphatic embolization of the collateral passage, the symptoms improved.

3) Methods

A total of 52 patients referred to the Hospital of the University of Pennsylvania for non-traumatic chylothorax between December 2014 and May 2020 were included. DCMRL was performed in all cases, and then classified into three different pathophysiologies: (1) abnormal pulmonary lymphatic perfusion originating only from the thoracic duct; (2) abnormal pulmonary lymphatic perfusion through the collateral branch passage of the retroperitoneal space (thoracic duct may or may not be affected); and (3) chylous ascites manifested as chylothorax through a defect in the diaphragm (Fig. 4).

Figure 4. Three different clinical scenarios of non-traumatic chylothorax in dynamic contrast-enhanced magnetic resonance lymphangiography (DCMRL). (A) A 41-year-old female with chylopericardium. Abnormal lymphatic perfusion of mediastinal structures is seen. Thoracic duct embolization successfully resolved the symptom. (B) A 14-year-old male with Noonan syndrome and related chylothorax. The parietal pleural lymphatic structure was enhanced through collateral lymphatic vessels in the body wall. Selective embolization of the collateral lymphatic vessels successfully resolved the symptom. (C) A 50-year-old female with idiopathic chylothorax. Computed tomography examination shows the presence of ascites. DCMRL shows a dilated and tortuous thoracic duct, but there is no clear sign of abnormal pulmonary perfusion. Even after thoracic duct embolization, the symptom was not resolved. The patient underwent pleurodesis to control the symptoms eventually.

4) Results

First or second type, abnormal pulmonary lymphatic perfusion originates from the thoracic duct or retroperitoneal space, manifested in 79% of the cases. For these cases, thoracic duct embolization or embolization of the collateral circulation in the retroperitoneal space was performed alone or in combination, resulting in successful treatment in 93% of cases. The remaining 21% of the cases were classified as the third type, in which ascites manifested as pleural effusion, and 75% resolved with treatment. In conclusion, the treatment outcomes of non-traumatic chylothorax can be improved by applying an algorithm based on the pathophysiology identified through DCMRL.

5) Discussion

Non-traumatic chylothorax presents with various manifestations, including idiopathic chylothorax, chylopericardium, neonatal chylothorax, and plastic bronchitis. It is referred to as pulmonary lymphatic perfusion syndrome because it is caused by the reflux of lymphatic fluid from the central lymphatic vessels toward the lungs. DCMRL has significantly contributed to enhancing the understanding of the diagnosis and pathology of pulmonary lymphatic perfusion syndrome by visualization. Considering that the basis of treatment for the disease is to block lymph reflux, this study demonstrated the successful performance of thoracic duct embolization or embolization of the collateral passage in the retroperitoneal space, depending on the location of the reflux [25].

In some cases, chylothorax occurs when chylous ascites passes through a defect in the diaphragm due to negative pressure in the thoracic cavity. In such cases, thoracic duct embolization may worsen the symptoms of chylothorax and ascites. Therefore, interventionists should be well acquainted with each pathophysiology in advance and should examine the existence of ascites or collateral branches other than the thoracic duct before performing thoracic duct embolization in patients with non-traumatic chylothorax.

1) Cases with no observed leakage from the lymphatic system of the retroperitoneal space

① Hypothesis

Chylous ascites can be treated by understanding its pathology and approaching it accordingly.

② Background

Unlike chylothorax, for which interventional procedures, including various lymphangiography and thoracic duct embolization, are effective for both diagnosis and treatment regardless of traumatic or non-traumatic conditions, the diagnosis and treatment success rates of chylous ascites are unsatisfactory. Some case reports have described the successful treatment of chylous ascites. However, it is very common in clinics for the leakage site not to be detected during lymphangiography. Even when the leakage site is found, establishing a clear treatment goal can often be challenging. According to the report by Nadolski et al. [27], only 55% of patients with chylous ascites had leakage sites detected by Lipiodol lymphangiography; among the remaining 45% of cases in which leakage was not confirmed, only 21% showed improvement in symptoms, either due to the effect of Lipiodol or spontaneous recovery.

③ Balloon-occluded retrograde transcatheter magnetic resonance lymphangiography [28]

In the current generalized imaging methods, a contrast agent is injected into the central lymphatic system through the inguinal lymph nodes, which constitute the lymphatic system of the lower extremities. Since the pathway joins the liver and mesenteric lymphatic systems only after reaching near the cisterna chyli, it is difficult to visualize the intraperitoneal lymphatic system without diffusion or reflux.

In the present case report, the leakage site of the mesenteric lymphatic vessel, which was not detected on general lymphangiography, was found only after a temporary occlusion of the thoracic duct with a balloon to allow the contrast agent to flow back into the mesenteric lymphatic system. The chylous ascites stopped only after super selective embolization of the lymphatic vessel at the leakage site (Fig. 5) [28].

Figure 5. A 52-year-old male with postoperative chylous ascites after partial gastrectomy and lymph node dissection for stomach cancer. (A, B) Lymphangiography showed no leakage of lymph fluid. (C) In balloon-occluded retrograde transcatheter magnetic resonance lymphangiography performed after retrograde thoracic duct cannulation and blockade of drainage of the thoracic duct with a balloon catheter, a contrast agent traveled through the cisterna chyli and then through the mesenteric lymphatic vessel (white arrowheads), which was a reversed flow (dashed arrow line) from the normal lymphatic physiology. It finally leaks from damaged lymphatic vessels and spreads into the abdominal cavity (white arrow). Adapted from the article of Lee et al. (J Vasc Interv Radiol 2021;32:617-619) [28] with original copyright holder’s permission.

④ Mesenteric lymphangiography and embolization [29]

For chylous ascites produced after extensive lymphadenectomy in the retroperitoneal space and pelvis, it is crucial to consider both clear lymph fluid originating from both lower extremities and chylous lymph fluid leakage from damaged lymphatic vessels near the cisterna chyli as potential sources of ascites. If the lymphatic vessels in the pelvis are completely resected, contrast agents may not reach the leakage site of chylous ascites on the upper side. Therefore, if chylous ascites persist despite performing inguinal lymph node lymphangiography and embolization for the treatment of lymphatic leakage from the lower extremities, as first reported by the present author, mesenteric lymphangiography can be performed after exposing the mesentery through laparotomy [29]. Although the safety of the procedure has not yet been sufficiently verified, there is a case report in which life-threatening massive chylous ascites were treated with extensive lymph node embolization of the mesenteric lymph nodes, and recovery occurred after temporary intestinal edema [30]. Based on successful early experience, mesenteric lymphangiography is performed after laparotomy, if necessary, in patients with chylous ascites that develop after surgery.

2) Cases with leakage observed from the lymphatic system of the retroperitoneal space [31]

① Hypothesis

Chylous lymphocele in the retroperitoneal space leading to chylous ascites can be treated by direct access to the retroperitoneal lymphocele.

② Background

Surgery involving the area around the aorta at the level of the renal hila is highly likely to result in the leakage of chylous lymph fluid because the mesenteric lymphatic system joins the retroperitoneal lymphatic system in this area. Because the retroperitoneal space is a potential space, the leaked lymph fluid forms a retroperitoneal lymphocele, surrounded by adjacent tissues. Tissue damage or increased pressure due to surgery may extend the lymphocele into the abdominal cavity, resulting in chylous ascites. In other words, chylous ascites caused by damage to the lymphatic system of the retroperitoneal space may manifest clinically as ascites (peritoneal fluid); however, it can be treated by directly accessing the causative retroperitoneal lymphocele and managing the lymphocele with or without embolization of the connected damaged lymphatic vessels.

③ Method

Six patients with chylous ascites that occurred after surgery of the perirenal space and were treated with embolization of lymphocele in the retroperitoneal space between January 2014 and October 2018 were examined. Among them, four were living donors of the left kidney, one underwent left partial nephrectomy, and the other underwent lymphadenectomy in the left perirenal space during prophylactic lymphadenectomy for testicular cancer. A drainage tube was directly inserted into the retroperitoneal lymphocele, and embolization using liquid embolic material was performed for the connected lymphopseudoaneurysm or the damaged lymphatic vessel through the route.

④ Results

Despite medical treatment, including fasting, the daily drainage volume of ascites was maintained at 300 mL or more (average 1,173 mL, range 305-2,800 mL). Intranodal Lipiodol lymphangiography was performed in 4 patients, and drainage tube installation and embolization were performed for retroperitoneal lymphoceles in all patients (Fig. 6). A liquid embolic material with Lipiodol: n-BCA ratio of 1:1 or 2:1 was used, with an average volume of 4.3 mL (3-6 mL). Leakage of ascites stopped within four days (average, two days), and the drainage tubes were removed in all patients. During the mean follow-up period of 37.3 months (range 21.1-48.4 months), there was no recurrence of ascites or procedure-related complications.

Figure 6. A three-dimensional reconstructed image (A) and a schematic diagram (B) for a drainage tube directly inserted into a lymphocele (asterisk) generated in the retroperitoneal space after damage to the lymphatic system around the aorta. (A) Damaged lymphatic vessels (black arrows) from the thoracic duct (T) and cisterna chyli to the back of the retroperitoneal space, the lymphatic vessels of the peri-aortic retroperitoneal space (white arrow) coming up from the bottom are all connected to the lymphocele (asterisk). (B) After chylous lymphatic fluid (large black arrow) originating from the cisterna chyli (CC) and clear lymph fluid (small black arrow) ascending from the legs and pelvis are all collected to a lymphocele (asterisk) and flow into the abdominal cavity along the defect in the wall of the lymphocele, it becomes chylous ascites (white arrows). Data from the article of Kwon et al. (Korean J Radiol 2021;22:376-383) [31].

⑤ Discussion

Appropriate installation of a drainage tube and monitoring of the drainage volume should be considered, even in cases of retroperitoneal lymphoceles found at a level higher than the pelvis. Unlike pelvic lymphoceles, retroperitoneal lymphoceles may form due to the leakage of chylous lymph fluid. While embolization of the upstream lymph node closest to the leakage site is effective for non-chylous pelvic lymphoceles, chylous lymphatic leakages often persist even after lymph node embolization. This is because conventional lymphangiography may only depict the path of non-chylous lymphatic fluid ascending from the lower extremities, omitting the path of chylous lymphatic fluid joined by the mesenteric lymphatic system in cases of chylous lymphocele. Likewise, when the rupture of chylous lymphocele in retroperitoneal space is manifested as chylous ascites, lymphatic leakage can be treated by inserting a drainage tube directly into the ruptured retroperitoneal lymphocele and selectively blocking the connected lymphopseudoaneurysm or damaged lymphatic vessels, as suggested in this study [31].

3) A comprehensive treatment strategy for chylous ascites

Even today, the diagnosis and treatment of chylous ascites remain challenging. The primary challenge stems from the fact that the lymphatic system of the abdomen is the site where the lymphatic system originating from intra-abdominal organs such as the liver, mesentery, and the lymphatic system originating from lower extremities and retroperitoneal space join. In conventional lymphatic system imaging, a contrast agent is injected through the lower extremity or inguinal region and travels through the retroperitoneal lymphatic system to the thoracic duct. Because the intraperitoneal lymphatic system is outside this route, false-negative results could be present despite the existence of leakage. Therefore, understanding the anatomical characteristics of the abdominal lymphatic system is crucial in the management of intra-abdominal lymphatic leakage, particularly in cases of chylous ascites.

DCMRL is the first step in diagnosing chylous ascites and abdominal lymphatic leakage. Compared to Lipiodol, magnetic resonance contrast agents can reproduce the actual physiological lymphatic flow more accurately because of their much lower viscosity and water solubility, and are even more useful because they have high sensitivity with a high signal-to-noise ratio and provide three-dimensional information about the leakage site [27]. In particular, because of the low viscosity and high diffusivity of the agents compared to Lipiodol, there are cases with leakage sites identified by DCMRL, which were negative on Lipiodol lymphangiography. However, lymphangiography using CT is more feasible than magnetic resonance imaging, which is relatively less accessible. This procedure is similar with other lymphangiographies in that the inguinal lymph node is used as the access route; however, it differs in that a water-soluble CT contrast agent is used instead of a magnetic resonance contrast agent. Despite the drawbacks of relatively low tissue contrast and reduced sensitivity to the contrast agent, it offers significant advantages in terms of accessibility, particularly when used with an angiography machine combined with multi-detector CT [32].

Leakage sites can be found using various types of lymphangiography in cases of damage to the lymphatic system in the retroperitoneal space. Damage to the mesenteric lymphatic system can be detected using the same method if it is sufficiently close to the retroperitoneal lymphatic system. In this case, the treatment method described in the previous section “2)” can be applied.

If the results of DCMRL are negative, there is a possibility that the leakage originates from the liver or mesenteric lymphatic system, far away from the retroperitoneal lymphatic system. Even if magnetic resonance lymphangiography is positive, if the main outflow originates upstream from the intra-abdominal lymphatic system, such as the mesentery, retroperitoneal lymphatic leakage may not respond to treatment. In this case, with the retrograde access to the thoracic duct, which is introduced in section “1)”, the direction of lymph fluid can be reversed with balloon occlusion. Liver or mesenteric lymphangiography, corresponding to the upstream of the intra-abdominal lymphatic system, can be performed to find the site of leakage, and if necessary, embolization can be attempted. If there are limitations to interventional treatment, a hybrid approach accompanied by laparotomy may be helpful.

In this study, a few valid hypotheses were established for the pathophysiology of multiple lymphatic diseases based on the interpretation of radiological diagnostic examinations. Various tools and methods of interventional radiology have been comprehensively applied for the diagnosis and treatment of lymphatic diseases that conform to these hypotheses. The following five hypotheses were verified by analyzing the treatment processes and outcomes: (1) various embolization techniques using liquid embolic materials can prevent the leakage of lymphatic fluid from damaged lymphatic vessels or nodes; (2) lymph node embolization can be used to treat postoperative pelvic lymphoceles that do not respond to sclerotherapy; (3) the technical success rate of thoracic duct embolization can be improved using a retrograde thoracic duct approach; (4) non-traumatic chylothorax can be successfully treated with an understanding of its pathogenesis and a systematic approach; and (5) chylous ascites can be treated with an understanding of its pathogenesis and a systematic approach.

This review article has been re-written based on the author’s thesis for the Philosophy of Doctor at Seoul National University College of Medicine. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government (MSIT) (No. 2023R1A2C2004638).

The author has nothing to disclose.