Interventional embolization for hemorrhagic pleural effusion in an infant with kaposiform lymphangiomatosis: a case report

Abstract

Background

Kaposiform lymphangiomatosis (KLA) is a rare, aggressive lymphatic anomaly that frequently involves the thorax and may be complicated by hemorrhagic pleural or pericardial effusion, thrombocytopenia, and coagulopathy. Acute respiratory compromise in infancy is uncommon and can be difficult to manage because surgery is often limited by infiltrative disease and bleeding risk.

Case presentation

A two-month-old male infant presented with a four-day history of progressive respiratory distress and reduced milk intake. Physical examination showed tachypnea and decreased breath sounds over the left hemithorax. Laboratory tests demonstrated anemia and thrombocytopenia. Contrast-enhanced computed tomography angiography revealed a large left pleural effusion with mediastinal shift, left lung atelectasis, and tortuous dilated vascular structures involving the left pericardium, diaphragm, and anterior thoracic wall. Emergency closed thoracic drainage yielded hemorrhagic pleural fluid. Angiography identified hypervascular lesions supplied by branches of the left internal thoracic artery and left phrenic artery, with active contrast extravasation from the phrenic arterial supply. Superselective embolization was performed using polyvinyl alcohol particles and gelatin sponge, followed by systemic methylprednisolone and sirolimus.

Outcome

Drainage output became clear after embolization. Six-month follow-up magnetic resonance imaging showed near-complete resolution of the lesion. During 5 years of follow-up, the platelet count remained above 100 × 10⁹/L and pleural effusion did not recur.

Conclusion

In selected patients with KLA and hemorrhagic pleural effusion, angiography can clarify the bleeding source and superselective arterial embolization may provide rapid hemorrhage control and respiratory stabilization, serving as a valuable adjunct to systemic therapy.

Introduction

Kaposiform lymphangiomatosis (KLA) is a rare and potentially life-threatening lymphatic anomaly characterized by infiltrative lymphatic malformation with kaposiform spindle-cell features. It most commonly affects children and young adults, but presentation in early infancy may be particularly severe because thoracic disease can rapidly compromise ventilation and circulation. The disorder is clinically distinct from generalized lymphatic anomaly and central conducting lymphatic anomaly, yet imaging overlap is substantial. Findings that support KLA include thoracic and mediastinal involvement, hemorrhagic pleural or pericardial effusions, infiltrative soft-tissue abnormality associated with abnormal tortuous vascular structures, and moderate thrombocytopenia (1).

Management is challenging. Patients may present with persistent effusions, anemia, thrombocytopenia, or coagulation abnormalities. Systemic therapy, particularly sirolimus, has become an important component of treatment, but response may be incomplete and acute hemorrhagic complications require immediate control. Surgical debulking is often impractical in infants because lesions may infiltrate the mediastinum, chest wall, diaphragm, and pericardium and because coagulopathy increases perioperative risk (2, 3).

The present case is unique because a two-month-old infant developed acute hemorrhagic pleural effusion with respiratory distress and imaging evidence of abnormal thoracic vasculature. Angiography demonstrated arterial supply and active extravasation, allowing superselective embolization as an emergency stabilizing intervention. The patient then achieved durable disease control with adjunctive corticosteroid and sirolimus therapy, with no recurrence during 5 years of follow-up.

Case description

A two-month-old male infant was brought to medical attention because of progressive respiratory distress and decreased milk intake for 4 days. The patient was de-identified for this report. The available clinical record did not document relevant antenatal, birth, family, or genetic history, and no prior intervention for the present thoracic episode was recorded before admission.

On examination, the infant had tachypnea and reduced breath sounds on the left side. Initial laboratory testing showed mild anemia, with hemoglobin of 95 g/L, and thrombocytopenia, with a platelet count of 56 × 10⁹/L. Baseline coagulation testing showed fibrinogen of 2.35 g/L, D-dimer of 0.82 mg/L FEU, prothrombin time of 12.4 s, and activated partial thromboplastin time of 36.8 s. These findings raised concern for bleeding into the pleural cavity in association with a complex vascular or lymphatic anomaly.

Contrast-enhanced computed tomography angiography (CTA) demonstrated a large left-sided pleural effusion causing marked compression of the left lung, left lung atelectasis, and mediastinal shift. Additional axial, sagittal, and coronal reformatted CT/CTA images demonstrated an infiltrative soft-tissue abnormality with associated tortuous and dilated vascular structures involving the left pericardium, diaphragm, and anterior inferior thoracic wall, thereby better showing the extent of disease. Three-dimensional reconstruction further showed two feeding arteries supplying the hypervascular lesion and an abnormal angiomatous vascular cluster. Contrast-enhanced CT examinations of the head, chest, and abdomen were performed to evaluate the disease extent. No definite extrathoracic manifestations of KLA were identified. In particular, abdominal CT showed no hepatosplenomegaly, intra-abdominal cystic lesions, splenic involvement, or retroperitoneal abnormalities. No obvious osseous abnormality was observed on the head, chest, or abdominal CT images.

Timeline

Table 1 summarizes the major clinical events, diagnostic findings, therapeutic interventions, and outcomes during the episode of care.

Table 1.

Timeline of the major clinical events, diagnostic findings, therapeutic interventions, and outcomes.

Time point	Clinical event	Relevant data/intervention	Outcome
Day 0	Presentation	Four-day history of progressive respiratory distress and decreased milk intake; tachypnea and decreased left breath sounds.	CTA and laboratory assessment performed.
Day 0	Initial assessment	Hemoglobin 95 g/L; platelet count 56 × 109/L; baseline coagulation profile was assessed. CTA showed large left pleural effusion, lung compression, mediastinal shift, and abnormal thoracic vascular structures.	KLA or another complex vascular/lymphatic anomaly complicated by hemorrhagic pleural effusion and thrombocytopenia was suspected.
Day 0	Emergency stabilization	Left-sided closed thoracic drainage was performed.	Hemorrhagic pleural fluid was drained; protein was 43.5 g/L, nucleated cell count was 1760 × 109/L, red blood cell count was 2.85 × 1012/L, triglyceride level was 0.42 mmol/L, and cytology showed no malignant or immature cells.
Day 1	Angiographic diagnosis	Angiography showed hypervascular lesions supplied by the left internal thoracic artery and left phrenic artery; active contrast extravasation was seen from the phrenic arterial supply.	Bleeding source and feeding vessels were identified.
Day 1	Intervention	Superselective embolization was performed under general anesthesia using 350-μm polyvinyl alcohol particles and gelatin sponge.	Drainage changed from hemorrhagic to clear fluid after embolization.
Post-procedure	Adjunctive systemic therapy	Methylprednisolone 12 mg every other day and sirolimus 0.25 mg every 12 h were administered. Oral compound sulfamethoxazole/TMP-SMX prophylaxis and clinical infection monitoring were initiated.	Therapy was used to reduce lymphangiogenic proliferation and maintain disease control, with prophylaxis and monitoring to reduce the risk of opportunistic infection.
6 months	Imaging follow-up	Chest MRI showed near-complete resolution of the lesion.	Radiologic response was favorable.
5 years	Long-term outcome	Serial follow-up showed platelet counts above 100 × 109/L and no recurrent pleural effusion.	Sustained clinical stability was achieved.

Diagnostic assessment

The principal diagnostic problem was to determine the cause of acute hemorrhagic pleural effusion in an infant with thrombocytopenia and abnormal thoracic vessels. The differential diagnosis included congenital or infantile vascular tumor, kaposiform hemangioendothelioma, KLA, generalized lymphatic anomaly, central conducting lymphatic anomaly, and other thoracic lymphatic malformations. Infection and simple chylous effusion were considered less likely because the drainage was hemorrhagic and CTA showed a hypervascular lesion with arterial feeders (4).

Emergency drainage confirmed hemorrhagic pleural effusion with elevated protein content. The pleural fluid findings, including the grossly hemorrhagic appearance, high red blood cell count, absence of malignant or immature cells, and low triglyceride level, supported hemorrhagic pleural effusion and did not suggest malignant effusion or simple chylous effusion. The combination of hemorrhagic effusion, thrombocytopenia, infiltrative thoracic vascular abnormality, and subsequent angiographic demonstration of abnormal hypervascular supply supported a diagnosis of KLA complicated by hemorrhagic pleural effusion and thrombocytopenia. The absence of definite extrathoracic involvement or obvious osseous abnormality on head, chest, and abdominal contrast-enhanced CT supported predominantly thoracic disease in this patient. Because the baseline coagulation profile did not show marked hypofibrinogenemia or overt consumptive coagulopathy, the diagnostic criteria for Kasabach-Merritt phenomenon were not met. Pathologic confirmation was not pursued during the acute episode because the patient was unstable and thrombocytopenic, and biopsy or surgery was considered to carry a substantial bleeding risk. NRAS mutation testing in lesional tissue or cell-free DNA from plasma or pleural effusion was also not performed. Therefore, the diagnosis was based on characteristic clinical, laboratory, imaging, angiographic, therapeutic-response, and long-term follow-up findings. The lack of histopathologic and molecular confirmation is acknowledged as an important diagnostic limitation (see Figure 1).

Figure 1.

Initial and follow-up imaging and drainage findings. (A) Emergency CT showed a large left-sided pleural effusion causing significant left lung compression and mediastinal shift. (B) Three-dimensional reconstruction showed two feeding arteries (red arrows) supplying the hypervascular lesion and a malformed vascular cluster (white arrow). (C) Follow-up chest MRI at 6 months after treatment showed near-complete resolution of the lesion. (D–F) Axial, sagittal, and coronal contrast-enhanced CT/CTA images demonstrated an infiltrative soft-tissue abnormality with tortuous and dilated vascular structures involving the left pericardium, diaphragm, and anterior inferior thoracic wall (red arrows), showing the extent of disease. (G,H) The initial drainage collection system showed a large volume of hemorrhagic pleural effusion. (I) After embolization, the drainage transitioned to clear fluid.

Therapeutic intervention

Emergency closed thoracic drainage was performed to relieve lung compression and mediastinal shift. The pleural drainage was grossly hemorrhagic, supporting ongoing bleeding into the pleural space. Pleural fluid analysis showed a protein level of 43.5 g/L, a nucleated cell count of 1760 × 10⁹/L, a red blood cell count of 2.85 × 10¹²/L, and a triglyceride level of 0.42 mmol/L. Cytological examination showed numerous erythrocytes and scattered inflammatory cells, without malignant cells. No immature cells were identified. Because continued drainage alone would not address the abnormal vascular supply, a multidisciplinary team elected to proceed with interventional angiography and embolization (5, 6).

Selective angiography demonstrated that the hypervascular lesion was supplied by branches of the left internal thoracic artery and the left phrenic artery. Angiography of the left phrenic artery also demonstrated active contrast extravasation, indicating ongoing bleeding. No discrete AVM-type vascular nidus, early draining vein, or rapid arteriovenous shunting was observed. The angiographic appearance was therefore interpreted as abnormal hypervascularity with active arterial bleeding rather than a typical high-flow arteriovenous malformation. Superselective catheterization was performed, followed by embolization using 350-μm polyvinyl alcohol (PVA) particles and gelatin sponge. PVA particles were selected because angiography showed diffuse abnormal hypervascularity with active contrast extravasation from the left phrenic arterial supply rather than a single discrete bleeding point or a typical high-flow arteriovenous malformation. Particle embolization was considered suitable for distal occlusion of the abnormal microvascular bed and reduction of hemorrhagic inflow. Gelatin sponge was additionally used to reinforce hemostasis. Coils were not used as the primary embolic agent because proximal mechanical occlusion alone might not have adequately embolized the distal hypervascular lesion and could have limited future endovascular access if repeat embolization became necessary. Other embolic agents, including liquid embolic materials and gelatin sponge alone, were considered; however, liquid agents were avoided because of the potential risk of non-target embolization in a small infant with thoracic collateral supply, and gelatin sponge alone was considered less durable. Therefore, combined PVA particle and gelatin sponge embolization was selected to balance effective hemostasis and procedural safety. The embolization procedure was performed under general anesthesia with continuous cardiorespiratory monitoring because the patient was a two-month-old infant with respiratory compromise caused by massive hemorrhagic pleural effusion and mediastinal shift. No blood transfusion was required during hospitalization or after embolization because the hemoglobin level remained clinically acceptable and no further hemorrhagic deterioration occurred (see Figure 2).

Figure 2.

Angiographic findings and embolization. (A) Angiography of the left internal thoracic artery demonstrated a hypervascular lesion in the left chest wall and diaphragm (white arrow). (B) Angiography of the left phrenic artery showed the same hypervascular lesion (white arrow) and an active contrast extravasation site indicating active bleeding (red arrow). (C,D) The feeding arteries were successfully embolized using polyvinyl alcohol particles and gelatin sponge.

On the first day after embolization, the pleural drainage output changed from hemorrhagic to clear fluid. The patient then received systemic methylprednisolone 12 mg every other day and sirolimus 0.25 mg every 12 h to suppress ongoing lymphatic proliferation and reduce the risk of recurrent effusion. Oral compound sulfamethoxazole, namely trimethoprim-sulfamethoxazole (TMP-SMX), was administered for Pneumocystis jirovecii pneumonia prophylaxis at a prophylactic dose equivalent to trimethoprim 5 mg/kg/day, given orally 3 days per week for 3 months during the early period of combined corticosteroid and sirolimus therapy. Clinical monitoring for infection, including fever, respiratory symptoms, mucositis, and other opportunistic infections, was performed during follow-up. Details of sirolimus trough levels and tapering schedule were not available in the source material and should be confirmed before final submission if required by the journal (7).

Follow-up and outcomes

The patient was monitored clinically, radiologically, and hematologically. Follow-up angiography was not performed because the patient showed rapid clinical stabilization after embolization, with transition of the pleural drainage from hemorrhagic to clear fluid and no evidence of recurrent bleeding. Given the invasive nature of angiography, repeat angiography was reserved for suspected recurrent hemorrhage or progressive vascularity. The follow-up strategy therefore consisted of clinical assessment, hematologic monitoring, and chest MRI. Six months after treatment, non-contrast chest MRI showed near-complete resolution of the lesion. Contrast-enhanced MRI was not performed because the patient had marked clinical improvement, non-contrast MRI demonstrated substantial lesion regression, and there was no clinical suspicion of recurrent active bleeding or disease progression. Avoiding unnecessary contrast administration and sedation was also considered appropriate in this infant. No recurrent respiratory compromise, recurrent hemorrhagic pleural effusion, Pneumocystis jirovecii pneumonia, or other opportunistic infection was documented.

During 5 years of follow-up, platelet counts remained stable above 100 × 10⁹/L. The pleural effusion did not recur, and the patient remained clinically stable. This durable response suggested that emergency embolization successfully controlled the acute hemorrhagic component, while systemic therapy contributed to longer-term stabilization of the lymphatic anomaly.

Discussion

KLA is an aggressive lymphatic anomaly with high morbidity and mortality, particularly when the thorax and mediastinum are involved. In the largest early clinicopathologic series, thoracic involvement was prominent, pleural or pericardial effusions were frequent, and reported five-year survival was approximately 51%. Imaging studies have also emphasized that hemorrhagic effusions, mediastinal infiltrative disease, and thrombocytopenia are important clues that favor KLA over other complex lymphatic anomalies (8, 9). KLA was differentiated from other vascular anomalies on the basis of the integrated clinical, laboratory, imaging, and angiographic findings. AVM was considered less likely because angiography did not show a compact nidus, early venous drainage, or rapid arteriovenous shunting. Venous malformation was also considered less likely because venous malformations are typically low-flow lesions and do not usually present with active arterial extravasation from a defined arterial supply. Kaposiform hemangioendothelioma was included in the differential diagnosis; however, the lesion was not a localized solid vascular tumor, and the baseline coagulation profile did not demonstrate marked hypofibrinogenemia or overt consumptive coagulopathy. Thus, the diagnostic criteria for Kasabach-Merritt phenomenon were not met. Taken together, the early-infantile presentation, hemorrhagic pleural effusion, thrombocytopenia, diffuse thoracic involvement, abnormal hypervascular supply, therapeutic response, and long-term follow-up stability supported the diagnosis of KLA.

We also revised the terminology regarding thrombocytopenia in this case. Although thrombocytopenia may occur in KLA and other complex vascular or lymphatic anomalies, Kasabach-Merritt phenomenon should be diagnosed only when characteristic thrombocytopenic consumptive coagulopathy is present, usually in association with kaposiform hemangioendothelioma or tufted angioma. Therefore, we avoided the term KMP in the revised manuscript and described the hematologic abnormality as thrombocytopenia associated with hemorrhagic pleural effusion (10).

KLA is an extremely rare and aggressive complex lymphatic anomaly. Its true incidence remains unknown because of its rarity and the small number of reported cases. It most often affects children and young adults, although severe presentations can occur in early infancy. The thorax and mediastinum are among the most frequently involved sites, and patients may present with pleural or pericardial effusions, respiratory compromise, thrombocytopenia, anemia, and coagulation abnormalities. Extrathoracic involvement, including bone, spleen, and abdominal organs, has also been reported in some patients. The pathophysiology of KLA is not fully understood. Histologically, KLA is characterized by abnormal dilated lymphatic channels with foci of kaposiform spindle-shaped endothelial cells and hemorrhagic components. These abnormal lymphatic and vascular channels may contribute to leakage of lymphatic or hemorrhagic fluid into the pleural or pericardial space. Thrombocytopenia and coagulopathy may result from platelet trapping, ongoing hemorrhage, and localized consumptive processes. Recent molecular studies have identified somatic activating alterations in the RAS-MAPK pathway, particularly NRAS p. Q61R, in patients with KLA, supporting the concept that dysregulated lymphangiogenic signaling contributes to disease development and progression.

The present case contributes several practical lessons. First, hemorrhagic pleural effusion in an infant with thrombocytopenia and abnormal thoracic vasculature should prompt consideration of a complex vascular or lymphatic anomaly rather than treatment as a simple effusion. Second, CTA and three-dimensional reconstruction can guide intervention by showing the distribution of abnormal vessels and potential feeding arteries. Third, angiography may have both diagnostic and therapeutic value when active bleeding is suspected.

Systemic therapy remains central to the management of KLA. Sirolimus, an inhibitor of the mammalian target of rapamycin pathway, is currently one of the most commonly used medical therapies for complicated lymphatic anomalies, including KLA, and may reduce effusions, improve hematologic abnormalities, and stabilize disease progression. Corticosteroids may be used as adjunctive therapy, particularly in patients with acute inflammatory, effusive, or hematologic manifestations. Vincristine has also been used in selected aggressive vascular anomalies and in some refractory or complicated cases, although evidence in KLA remains limited. More recently, MEK inhibitors, such as trametinib, have emerged as potential targeted therapies for selected patients with RAS-MAPK pathway alterations, especially NRAS-mutated or sirolimus-refractory disease. However, these therapies require careful patient selection, monitoring, and multidisciplinary decision-making. Endovascular treatment should be considered complementary to, rather than a replacement for, systemic therapy. Medical therapy is important for long-term disease control, but it may not provide immediate control of life-threatening hemorrhage. In the present infant, angiography demonstrated active extravasation from the left phrenic arterial supply, and superselective embolization directly targeted the bleeding source. The rapid transition from hemorrhagic to clear drainage after embolization suggests that arterial control was clinically meaningful. Thus, embolization served as an emergency stabilizing intervention and created a safer window for subsequent systemic therapy with corticosteroid and sirolimus (4).

Because sirolimus has immunosuppressive effects (11), infection prevention should be considered during treatment, especially in infants and in patients receiving concomitant corticosteroids. TMP-SMX is commonly used for Pneumocystis jirovecii pneumonia prophylaxis, although routine prophylaxis for all patients with vascular anomalies receiving sirolimus remains debated and may be individualized according to age, concomitant immunosuppression, lymphopenia, and other risk factors. In this patient, oral compound sulfamethoxazole/TMP-SMX prophylaxis was administered during the early period of combined corticosteroid and sirolimus therapy, and no opportunistic infection was observed during follow-up (12, 13).

Recent molecular studies have identified somatic activating NRAS variants, particularly NRAS p. Q61R (c.182A > G), as recurrent alterations in kaposiform lymphangiomatosis (14, 15). Detection of this mutation by droplet digital PCR or next-generation sequencing in lesional tissue, plasma cell-free DNA, or pleural effusion cell-free DNA may provide a useful minimally invasive adjunct to diagnosis. This approach may be especially valuable in unstable infants or thrombocytopenic patients in whom biopsy carries a substantial bleeding risk. Molecular confirmation may also help distinguish kaposiform lymphangiomatosis from other complex lymphatic anomalies and may have therapeutic implications, including consideration of targeted inhibition of the MAPK pathway in selected patients. In the present patient, NRAS mutation testing was not performed because no suitable stored plasma or pleural effusion sample was available for retrospective molecular analysis. Therefore, the diagnosis remains based on the characteristic clinical, laboratory, imaging, angiographic, therapeutic-response, and long-term follow-up findings. We acknowledge the absence of NRAS testing as an important limitation.

The major strength of this case is the long follow-up period. Durable stability over 5 years, including sustained platelet recovery and absence of recurrent pleural effusion, supports the potential value of combining acute endovascular control with longer-term medical therapy. The case also highlights the usefulness of multidisciplinary decision-making in rare pediatric vascular anomalies, especially when conventional surgical treatment is high risk.

This report has limitations. It describes a single patient and therefore cannot establish general efficacy or safety of embolization for all patients with KLA. The favorable outcome likely reflects combined management, so the independent contribution of embolization cannot be fully separated from corticosteroid and sirolimus therapy. Histopathologic confirmation was not obtained during the acute episode because biopsy was considered unsafe. NRAS mutation testing in tissue or cell-free DNA from plasma or pleural effusion was also not performed. Therefore, the diagnosis rests on clinical, imaging, angiographic, laboratory, therapeutic-response, and follow-up data, and the lack of molecular confirmation should be considered an important limitation. Finally, although baseline coagulation parameters were available, serial coagulation monitoring data were not included in the source material. Details about sirolimus blood concentrations, comprehensive adverse-event monitoring, and growth and developmental outcomes during long-term follow-up were also unavailable.

The key take-away lesson is that in selected infants with KLA complicated by acute hemorrhagic pleural effusion, superselective arterial embolization may be considered when imaging or angiography demonstrates abnormal arterial feeders or active extravasation. This approach should be viewed as an adjunctive emergency strategy to stabilize acute bleeding and respiratory compromise, while systemic therapy remains essential for longer-term disease control.

Funding Statement

The author(s) declared that financial support was not received for this work and/or its publication.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement

Written informed consent was obtained from the patient’s legal guardian for the publication of this case report and any potentially identifiable images or data included in this article.

Author contributions

JL: Conceptualization, Data curation, Investigation, Writing – original draft, Writing – review & editing. RW: Data curation, Investigation, Writing – original draft, Writing – review & editing. CL: Formal analysis, Resources, Writing – review & editing. LL: Project administration, Supervision, Writing – review & editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.