Pulmonary glue embolism mimicking thrombotic pulmonary embolism after varicocele embolization: a case report

Abstract

Background

Pulmonary glue/iodized oil embolism is an uncommon form of non-thrombotic pulmonary embolism caused by migration of cyanoacrylate-based embolic agents. Although most cases have been reported after endoscopic treatment of gastric varices, cases after varicocele embolization are exceptionally rare, and longitudinal data on clinically stable patients remain limited.

Case presentation

We report a 35-year-old man who developed acute dyspnea, chest pain, and palpitations immediately after varicocele embolization, mimicking thrombotic pulmonary embolism. Computed tomography pulmonary angiography demonstrated bilateral radiopaque intravascular material within the pulmonary arterial circulation, consistent with migrated glue/iodized oil material, with associated right-dominant consolidation, ground-glass opacities, and minimal pleural effusion. The patient remained hemodynamically stable, with no sustained hypoxemia, right ventricular dysfunction, or pulmonary hypertension, and was managed with individualized conservative/supportive treatment without surgical or endovascular intervention. Follow-up CT at 1 month showed marked regression of parenchymal abnormalities and an apparent partial decrease in the visible radiopaque embolic material. At 6 months, pulmonary function, including DLCO, normalized, echocardiography showed no pulmonary hypertension, and CT findings remained stable. At 1 year, the patient was asymptomatic, and chest radiography showed no clinically significant residual or new parenchymal or pleural abnormality.

Conclusions

This case adds to the limited literature on pulmonary glue/iodized oil embolism after varicocele embolization by documenting a hemodynamically stable course with favorable longitudinal clinical, radiological, and functional follow-up. In carefully selected patients without right ventricular dysfunction or clinical deterioration, conservative/supportive management may be reasonable, but treatment decisions should remain individualized and multidisciplinary.

Background

Pulmonary glue/iodized oil embolism is a rare form of non-thrombotic pulmonary embolism resulting from the migration of cyanoacrylate-based embolic agents into the pulmonary circulation. It has been most frequently reported following endoscopic treatment of gastric varices [1], whereas cases related to varicocele embolization or other pelvic venous embolization procedures remain exceedingly uncommon [2, 3].

The clinical presentation of pulmonary glue embolism is variable, ranging from asymptomatic incidental findings to severe respiratory compromise. However, current knowledge is based on isolated case reports, and the natural history and optimal management of this condition remain poorly defined.

Here, we report one of the rare cases of pulmonary glue embolism after varicocele embolization, characterized by a favorable longitudinal course despite bilateral radiological involvement, providing rare longitudinal follow-up data and highlighting the potential role of non-invasive conservative/supportive management without surgical or endovascular intervention in carefully selected, clinically stable patients.

Case presentation

A 35-year-old man with no chronic diseases, a non-smoker, and no history of tuberculosis or thromboembolism underwent varicocele embolization at an external center. Detailed procedural documentation and external-center records were not accessible to the author, and the patient was also unable to obtain detailed procedural information. The available hospitalization records did not include any external procedural document containing these technical details. The treating physician at the external center verbally confirmed that a mixture of N-butyl cyanoacrylate (NBCA, glue) and iodized oil (Lipiodol) had been used; however, the treated venous territory, vascular access route, NBCA-to-iodized oil ratio, total volume injected, use of coils or other embolic agents, and whether intraprocedural migration occurred could not be verified. Therefore, no assumptions or reconstructed procedural details were added regarding these unavailable variables. Immediately after the procedure, he developed acute dyspnea, chest pain, and palpitations and was admitted to our hospital.

On admission, he was normotensive, with a blood pressure of 143/74 mmHg, heart rate of 96/min, respiratory rate of 18/min, temperature of 37 °C, and oxygen saturation of 95–97% on room air. Chest examination revealed rhonchi, and lower-extremity examination showed no clinical signs of deep venous thrombosis. Electrocardiography showed sinus rhythm. Arterial blood gas analysis was not performed. Laboratory tests showed an elevated C-reactive protein (CRP) level of 195 mg/L, whereas procalcitonin was low at 0.09 ng/mL. Additional laboratory findings included leukocytosis, with a white blood cell count of 16.44 × 10⁹/L, neutrophil count of 12.7 × 10⁹/L, lymphocyte count of 1.65 × 10⁹/L, and platelet count of 357 × 10⁹/L. The activated partial thromboplastin time was 28.6 s, and antithrombin III activity was 94.8%. Fibrinogen was not available. Admission D-dimer, troponin, and NT-proBNP values were not available and were therefore not used for acute risk stratification.

Pulmonary CT angiography performed on the day of presentation demonstrated bilateral radiopaque/hyperdense intravascular material within the pulmonary arterial circulation, more prominent on the right side, with nodular and tubular/branching morphology compatible with migrated glue/iodized oil material (Fig. 1). Associated parenchymal findings included right-dominant lower-lobe consolidation, scattered ground-glass opacities, and minimal right pleural effusion. No conventional thrombotic pulmonary embolism, such as low-attenuation intraluminal filling defects within contrast-opacified pulmonary arteries, was identified. The radiopaque intravascular appearance, immediate temporal relationship with varicocele embolization, and verbal confirmation of glue/iodized oil use supported the diagnosis of non-thrombotic pulmonary glue/iodized oil embolism rather than conventional thrombotic pulmonary embolism. Formal radiological burden quantification by the number of involved segmental or subsegmental arteries was not available; therefore, the embolic burden was described qualitatively according to laterality, central predominance, and associated parenchymal findings.

Fig. 1.

Initial computed tomography pulmonary angiography findings after varicocele embolization. Axial mediastinal-window images demonstrate hyperdense/radiopaque embolic material within central and hilar pulmonary arterial branches (A), with additional multifocal radiopaque embolic deposits in more distal pulmonary arterial branches, predominantly on the right side (B). Axial lung-window image shows right lower-lobe consolidation with surrounding ground-glass opacity and minimal pleural effusion, findings considered compatible with pulmonary infarction and inflammatory reaction rather than microbiologically proven pneumonia (C)

Initial transthoracic echocardiography demonstrated preserved left ventricular ejection fraction of 60%, no regional wall motion abnormality, no D-shaped septum, no echocardiographic evidence of right ventricular strain, trace mitral and tricuspid regurgitation, estimated systolic pulmonary artery pressure of 18 + 3 mmHg, and TAPSE of 24 mm. Because admission NT-proBNP and troponin values were unavailable and because the event represented non-thrombotic glue/iodized oil embolism rather than conventional thrombotic pulmonary embolism, formal ESC risk stratification could not be fully applied; however, the available clinical and echocardiographic findings did not support an intermediate-high-risk profile [4]. Lower-extremity venous Doppler ultrasonography was not performed because CT pulmonary angiography did not demonstrate findings compatible with thrombotic pulmonary embolism and because the patient had no clinical signs or symptoms of lower-extremity deep venous thrombosis.

The patient was evaluated by a multidisciplinary team, including pulmonology, cardiovascular surgery, thoracic surgery, infectious diseases, and interventional radiology. No indication for surgical or endovascular intervention was identified because the patient remained hemodynamically stable, had no echocardiographic evidence of right ventricular strain, and showed no clinical deterioration.

Empirical antibiotic therapy was initiated because infection could not be confidently excluded at admission in the presence of marked CRP elevation, leukocytosis, and pulmonary consolidations. The patient received intravenous antibiotic therapy for 7 days. Intravenous moxifloxacin was started at admission and continued for 7 days; after Infectious Diseases consultation, intravenous cefepime was added for 6 days to broaden antibacterial coverage.

During hospitalization, blood cultures obtained at admission showed no growth, the respiratory viral panel was negative, and sputum culture could not be obtained because the patient was unable to produce a sputum sample. Procalcitonin remained low. At discharge, despite the absence of microbiological confirmation, a short 1-week oral antibiotic regimen with cefixime and moxifloxacin was prescribed because concomitant infection could not be completely excluded.

Given the absence of microbiological confirmation, persistently low procalcitonin, and the radiological pattern, the consolidations were interpreted primarily as pulmonary infarction and/or a foreign-body inflammatory reaction related to the embolic glue/iodized oil material. Therefore, the subsequent radiological and inflammatory improvement was interpreted cautiously and was not attributed solely to antibiotic therapy.

Because the embolic material represented non-thrombotic glue/iodized oil embolism, full-dose therapeutic anticoagulation for conventional thrombotic pulmonary embolism was not considered necessary. Enoxaparin was administered at a prophylactic dose of 40 mg/0.4 mL subcutaneously once daily during hospitalization and was continued after discharge for the first 3 months as secondary thrombosis prevention around persistent intravascular foreign material. This approach was individualized and extrapolated rather than guideline-proven. The patient remained hemodynamically stable throughout hospitalization, with no sustained hypoxemia and no need for advanced respiratory support or intensive care. He was discharged in stable clinical condition on prophylactic-dose enoxaparin, together with the short 1-week oral antibiotic regimen.

At the 1-month follow-up, CT imaging demonstrated marked regression of the parenchymal consolidations and ground-glass opacities, with an apparent partial decrease in the visible radiopaque glue/iodized oil embolic material (Fig. 2). Clinically, the patient reported intermittent chest pain, dyspnea, and tachycardia/palpitations, but no clinical deterioration was observed. Pulmonary function testing revealed preserved spirometric values (FEV₁/FVC 93.8; FEV₁ 146% predicted; FVC 128% predicted) with diffusion capacity at the lower limit of normal (DLCO 82%). These findings suggested a limited functional impact despite the initial bilateral radiological involvement.

Fig. 2.

One-month follow-up CT comparison after pulmonary glue/iodized oil embolism. Initial axial mediastinal-window CT image shows radiopaque intravascular embolic material within the pulmonary arterial branches (A), whereas the corresponding 1-month follow-up axial image demonstrates residual but less conspicuous visible radiopaque embolic material (B). Initial axial lung-window image shows right lower-lobe consolidation with surrounding ground-glass opacity and minimal pleural effusion (C), while the 1-month follow-up lung-window image demonstrates marked regression of the parenchymal consolidations and ground-glass opacities, with only minimal residual peripheral opacity (D)

Because residual intravascular foreign material persisted radiologically at the 1-month follow-up, a multidisciplinary council recommended initiation of acetylsalicylic acid 100 mg/day from the beginning of the second month. This decision was individualized and extrapolated, aiming to reduce the theoretical risk of platelet activation or secondary thrombus formation around the persistent intravascular foreign material.

Follow-up D-dimer, troponin, and NT-proBNP values at 4 months were within reassuring ranges; however, these values were not used to retrospectively define the acute ESC risk category. At the 6-month follow-up, CT findings were similar to those observed at 1 month, with persistent residual radiopaque glue/iodized oil material and no progression of parenchymal abnormalities.

Pulmonary function testing showed normalization of pulmonary function parameters (FEV₁ 150% predicted, FVC 135% predicted, DLCO 99% predicted). Transthoracic echocardiography confirmed preserved biventricular function (ejection fraction, 65%) and no evidence of pulmonary hypertension. Based on repeated multidisciplinary evaluation, acetylsalicylic acid therapy was discontinued.

By the 1-year follow-up, the patient remained clinically stable and asymptomatic, without recurrent chest pain, dyspnea, palpitations, or evidence of late cardiopulmonary complications. Follow-up posteroanterior chest radiography demonstrated no clinically significant residual or new parenchymal or pleural abnormality (Fig. 3). Overall, the clinical course remained stable, with progressive radiological and functional improvement throughout follow-up.

Fig. 3.

Long-term chest radiographic follow-up. Admission posteroanterior chest radiograph shows right lower-zone parenchymal opacity and blunting of the right costophrenic angle, corresponding to the CT-documented right lower-lobe consolidation and minimal pleural effusion (A). One-year follow-up posteroanterior chest radiograph demonstrates marked interval radiographic improvement, with no clinically significant residual or new parenchymal or pleural abnormality (B)

Discussion and conclusions

Pulmonary glue embolism is an uncommon form of non-thrombotic pulmonary embolism caused by migration of cyanoacrylate-based embolic material into the pulmonary arterial circulation. It has been most frequently reported after endoscopic cyanoacrylate injection for gastric varices, whereas cases occurring after varicocele embolization are exceptionally rare, with only isolated reports in the broader context of pelvic venous or varicocele-related embolization procedures [1–3, 5]. Varicocele embolization is generally considered an effective and safe endovascular treatment; however, rare non-target embolic complications may occur, particularly when liquid embolic agents are used [6, 7].

Only a few reports have described pulmonary glue/iodized oil embolism after pelvic venous embolization, and the clinical course in these reports differs from the present case. Alonso et al. reported pulmonary embolism of N-butyl cyanoacrylate–iodized oil material after treatment of pelvic varicose veins, with metallic-density foreign material in subsegmental pulmonary arterial branches on CT; because the patient was symptomatic, therapeutic low-molecular-weight heparin was prescribed for 3 months, and clinical–radiological follow-up was performed [2]. Longueville et al. described a more severe presentation after radiological embolization of pelvic varices: glue migration was detected during the procedure, the patient developed respiratory failure with oxygen requirement up to 6 L/min, and follow-up CT angiography demonstrated associated thrombotic pulmonary embolism, requiring therapeutic anticoagulation, antibiotics, corticosteroids, and oxygen therapy [3]. In contrast, the present case occurred after varicocele embolization, was characterized by same-day symptom onset, hemodynamic stability, absence of echocardiographic right ventricular strain or pulmonary hypertension, no need for intensive care, advanced respiratory support, or surgical/endovascular intervention, and favorable longitudinal clinical, radiological, and functional recovery. This comparison suggests that pulmonary glue/iodized oil embolism after pelvic venous or varicocele-related embolization may encompass a broad clinical spectrum, and that management should be individualized according to hemodynamic status, respiratory compromise, right ventricular function, thrombotic complications, and follow-up evolution.

The present case is notable because pulmonary glue/iodized oil embolism occurred immediately after varicocele embolization and mimicked conventional thrombotic pulmonary embolism clinically and radiologically. Despite bilateral radiopaque intravascular embolic material and associated parenchymal abnormalities, the patient remained hemodynamically stable, had no echocardiographic evidence of right ventricular strain or pulmonary hypertension, and did not require intensive care, advanced respiratory support, or surgical/endovascular intervention. Serial follow-up demonstrated gradual symptomatic improvement, regression of parenchymal abnormalities, apparent partial reduction in the conspicuity of visible radiopaque embolic material, normalization of pulmonary function including DLCO, and absence of pulmonary hypertension at 6 months. These findings provide longitudinal clinical, radiological, and functional follow-up data in a rarely reported complication after varicocele embolization.

Radiological differentiation from conventional thrombotic pulmonary embolism is central to diagnosis. Thrombotic pulmonary embolism typically appears on CT pulmonary angiography as low-attenuation intraluminal filling defects within contrast-opacified pulmonary arteries. In contrast, glue/iodized oil embolism is suggested by radiopaque or hyperdense intravascular material, often with nodular, tubular, or branching morphology, reflecting the iodized oil-containing embolic mixture [2, 3]. Pulmonary cement embolism may also present as very high-attenuation linear or branching material along the pulmonary arterial tree, most commonly after vertebroplasty or kyphoplasty [8]. In the present case, the radiopaque intravascular morphology, immediate temporal relationship with varicocele embolization, absence of conventional thrombotic filling defects, and verbal confirmation of glue/iodized oil use supported the diagnosis of non-thrombotic pulmonary glue/iodized oil embolism rather than conventional thrombotic pulmonary embolism.

The inflammatory and parenchymal findings also require careful interpretation. The patient had marked CRP elevation, leukocytosis, and pulmonary consolidations; therefore, empirical antibiotics were initiated because infection could not initially be excluded. However, procalcitonin remained low, blood cultures showed no growth, the respiratory viral panel was negative, and sputum culture could not be obtained. In this context, the pulmonary consolidations were interpreted primarily as pulmonary infarction and/or a foreign-body inflammatory reaction related to embolic glue/iodized oil material rather than microbiologically proven pneumonia. The subsequent radiological and inflammatory improvement should therefore be interpreted cautiously and cannot be attributed solely to antibiotic therapy.

Management of pulmonary glue/iodized oil embolism is not standardized. Although the ESC pulmonary embolism risk-stratification framework is useful for conventional thrombotic pulmonary embolism, it has not been validated for non-thrombotic pulmonary glue/iodized oil embolism [4]. Therefore, in the present case, ESC-based risk assessment could only be used as a clinical reference framework rather than as a formally applicable classification system. Unlike conventional thrombotic pulmonary embolism, the embolic material is not primarily thrombus; therefore, therapeutic-dose anticoagulation cannot be assumed to be beneficial in all cases. In the present patient, full-dose anticoagulation for thrombotic pulmonary embolism was not considered necessary. Prophylactic-dose enoxaparin and later acetylsalicylic acid were used as individualized, multidisciplinary, extrapolated measures to reduce the theoretical risk of secondary thrombus formation around persistent intravascular foreign material. This approach was not guideline-proven, and the optimal antithrombotic or antiplatelet strategy for pulmonary glue/iodized oil embolism remains undefined. Because this is a single observational case, no causal inference can be made regarding whether prophylactic-dose enoxaparin, acetylsalicylic acid, or empirical antibiotics contributed to the favorable outcome.

The potential role of catheter-based intervention should also be individualized. In postoperative thrombotic pulmonary embolism, catheter-based reperfusion strategies have been evaluated in selected intermediate-high-risk patients; however, such evidence cannot be directly extrapolated to non-thrombotic pulmonary glue/iodized oil embolism [9]. In patients with hemodynamic compromise, progressive right ventricular dysfunction, or large accessible central embolic material, catheter-directed retrieval or aspiration may be considered after multidisciplinary evaluation. However, polymerized cyanoacrylate may be firm, adherent, or difficult to aspirate, and attempted retrieval may carry risks of fragmentation, distal embolization, vascular injury, or incomplete removal. In the present case, surgical or endovascular intervention was not pursued because the patient remained hemodynamically stable, had no echocardiographic evidence of right ventricular strain, and improved with non-invasive conservative/supportive management.

Prevention remains an important aspect of liquid embolic procedures. Because the index varicocele embolization was performed at an external center and procedural details were unavailable, the precise technical mechanism of migration could not be determined in this case. Nevertheless, general preventive principles include careful pre-procedural assessment of venous anatomy and collateral outflow, appropriate selection and preparation of the embolic agent, controlled injection under continuous fluoroscopic monitoring, stable catheter positioning, avoidance of excessive injection volume or pressure, and immediate cessation of injection if non-target migration is suspected. These measures are particularly relevant when using NBCA–iodized oil mixtures, because the NBCA-to-iodized oil ratio may influence radiopacity, viscosity, and polymerization behavior [7]. Several liquid embolic agents are available, each with distinct advantages and limitations. Cyanoacrylate-based glues, particularly N-butyl cyanoacrylate, provide rapid and durable embolization and can be mixed with iodized oil to improve radiopacity and modify polymerization time [7]. However, they require technical expertise and may be associated with catheter adhesion, non-target embolization, or migration if injection dynamics, dilution, or vascular anatomy are unfavorable. Other liquid embolic agents, such as ethylene-vinyl alcohol copolymer-based agents, may allow more controlled injection but require specific preparation and may not be suitable for all vascular territories or indications. Therefore, embolic agent selection should be individualized according to vascular anatomy, flow characteristics, operator experience, and the intended embolization target.

Potential follow-up complications after pulmonary glue/iodized oil embolism include persistent or recurrent respiratory symptoms, pulmonary infarction, secondary infection or inflammatory consolidation, foreign-body reaction, pulmonary hypertension, and chronic thromboembolic-like physiology. In the present case, intermittent chest pain, dyspnea, and tachycardia/palpitations persisted during the first month but gradually resolved. Importantly, by 6 months, the patient was clinically asymptomatic, pulmonary function had normalized, and echocardiography showed no evidence of pulmonary hypertension. The absence of pulmonary hypertension or chronic cardiopulmonary impairment at 6 months is an important strength of this longitudinal follow-up.

The main strength of this case is the documented longitudinal clinical, radiological, functional, and echocardiographic follow-up after a rare pulmonary glue/iodized oil embolism following varicocele embolization. The immediate temporal relationship with the procedure, the characteristic radiopaque intravascular CT findings, the absence of conventional thrombotic pulmonary arterial filling defects, and the favorable 6-month and 1-year follow-up findings support the educational value of the case despite the absence of complete procedural documentation from the external center.

This case has limitations. The index embolization was performed at an external center, and detailed procedural documentation and external-center records were not accessible to the author. The patient was re-contacted to request any available procedural report, discharge document, or interventional radiology procedure note. The patient confirmed that he did not have such a report in his possession, and no external procedural document containing the requested technical details was available in the hospitalization file. Therefore, the treated venous territory, vascular access route, NBCA-to-iodized oil ratio, total injected volume, use of coils or other embolic agents, and intraprocedural migration status could not be verified. To avoid reporting unverified procedural details, no assumptions or reconstructed information regarding these procedural variables were included. This limits the reproducibility of the procedural aspects of the case and prevents a definitive assessment of the exact technical mechanism of pulmonary migration. Formal radiological burden quantification was also not available, and the embolic burden was described qualitatively. In addition, admission NT-proBNP and troponin values were unavailable, limiting formal application of ESC risk stratification, which is designed for thrombotic pulmonary embolism and has not been validated for glue/iodized oil embolism. Finally, because this is a single case, no general management recommendation can be made.

In conclusion, pulmonary glue/iodized oil embolism after varicocele embolization is exceptionally rare and may mimic conventional thrombotic pulmonary embolism. This case suggests that non-invasive conservative/supportive management may be reasonable in carefully selected hemodynamically stable patients without right ventricular dysfunction or clinical deterioration, even when bilateral radiological involvement is present. However, treatment decisions should remain individualized and multidisciplinary, taking into account hemodynamic status, symptoms, embolic burden, right ventricular function, follow-up trajectory, and available local interventional expertise.

Abbreviations

CRP

C-reactive protein

CT

Computed tomography

DLCO

Diffusing capacity of the lung for carbon monoxide

ESC

European Society of Cardiology

FEV₁

Forced expiratory volume in one second

FVC

Forced vital capacity

NBCA

N-butyl cyanoacrylate

NT-proBNP

N-terminal pro-B-type natriuretic peptide

TAPSE

Tricuspid annular plane systolic excursion

Authors' contributions

İÇ conceptualized the study, collected the data, prepared the figures, and wrote the manuscript. The author reviewed and approved the final manuscript.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Data availability

All data generated or analyzed during this case report are included in this published article. Additional anonymized clinical details are available from the corresponding author upon reasonable request, provided that patient confidentiality is maintained.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Başakşehir Çam and Sakura City Hospital Observational Research Ethics Committee. Written informed consent was obtained from the patient.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and accompanying images.

Competing interests

The authors declare no competing interests.