A retrospective study on the diagnostic value of ultrasonography for primary vascular leiomyosarcoma misdiagnosed as deep vein thrombosis

Introduction

Deep vein thrombosis (DVT) is a common vascular disorder, primarily diagnosed through clinical examination and imaging studies. Among imaging modalities, ultrasonography stands out as the diagnostic tool of choice, because of its noninvasive, convenient, and repeatable nature.^1–3 However, certain primary vascular tumors closely mimic DVT in both clinical presentation and ultrasound findings, often leading to misdiagnosis.^1–3 Primary vascular tumors, including angiosarcoma and hemangioendothelioma, are rare neoplasms of vascular origin whose early diagnosis is crucial for improved patient prognosis. Angiosarcomas account for approximately 1% of all soft tissue sarcomas, with an incidence of 1.1–2.0 cases per million population and a median age at diagnosis of 65–70 years. Among them, vascular leiomyosarcoma arising from major limb vessels has a poor prognosis, with a reported 5-year survival rate of only 25%.^4,5

The present study aimed to delineate the diagnostic role of ultrasonography in differentiating primary vascular tumors from DVT, identify factors contributing to misdiagnosis, and propose strategies to enhance diagnostic accuracy. We hypothesized that the meticulous analysis of ultrasound image characteristics, consideration of clinical manifestations, and consideration of advanced ultrasound techniques, could improve diagnostic accuracy and reduce misdiagnosis rates for primary vascular tumors.

Methods

This single-center retrospective study analyzed data from patients initially misdiagnosed with DVT but later confirmed to have primary vascular leiomyosarcoma who were treated at our hospital from January 2018 to December 2024. The inclusion criteria were (1) initial diagnosis of DVT, (2) final pathological diagnosis of primary vascular leiomyosarcoma, and (3) availability of complete ultrasound data. Patients were excluded if they had (1) secondary vascular tumors, (2) incomplete ultrasound data, or (3) concurrent DVT. Demographic and clinical data, including age, gender, symptoms, signs, and laboratory results, were collected. Ultrasound images of all the included patients were reviewed, with a focus on tumor location, size, echogenicity, margin definition, vascular infiltration, and adjacent tissue involvement. All statistical analyses were performed using SPSS 22.0, with continuous data expressed as mean ± standard deviation (SD) and categorical data as frequency and percentage.

Results

A total of five patients were included in the study, of whom three were male and two were female, with an age range of 61-73 years. The main symptoms were lower limb swelling (100%), pain (86.7%), and increased skin temperature (73.3%). All patients underwent ultrasound examination, with an initial diagnosis of DVT before a final pathological confirmation of primary vascular leiomyosarcoma. Ultrasound image analysis revealed that all tumors were located in the deep veins of the lower limbs, with a mean (SD) maximum diameter of 16 mm.Of the tumors examined, 2 (40%) exhibited heterogeneous echogenicity, and 3 (60%) showed hypoechogenicity. Clear boundaries were observed in 3 cases (60%), while the remaining 1 case (20%) had unclear boundaries. Signs of vascular infiltration and surrounding tissue invasion were present in 2 (40%) and 2 (40%) cases, respectively. Notably, these characteristic features were often overlooked or misdiagnosed during the initial examination (Figures 1(a)-(e); Supplemental Video 1).OPEN IN VIEWER

Discussion

This study found that primary vascular tumors had characteristic ultrasound features, including heterogeneous echogenicity, vascular infiltration, and surrounding tissue invasion. Despite differing from the typical ultrasound characteristics of DVT (e.g., venous dilation and intraluminal thrombus), these features were often overlooked or misinterpreted during the initial examination, leading to misdiagnosis. Historically, the diagnosis and treatment of vascular tumors have been challenging, and conventional ultrasonography has shown limited accuracy in diagnosing intravascular leiomyomatosis, ⁶ underscoring the need for advanced imaging techniques. Research has shown that contrast-enhanced ultrasonography can improve diagnostic accuracy through enhanced visualization of fine blood flow within tumors, with an accuracy rate of 83% for intravascular leiomyomatosis. ⁷

Below, we outline key ultrasound criteria for differentiating vascular tumors from venous thrombosis. (1)The presence of arterial blood flow signals within a mass is suggestive of a tumor, as acute venous thrombi typically lack internal flow. Partial recanalization of a thrombus may show color flow, and in some cases, arterial-like signals can be seen as part of the physiological mechanism of venous recanalization. Therefore, waveform analysis and correlation with grayscale features remain essential. To assess for arterial flow within a suspected thrombus or tumor, color flow Doppler should be applied, and both color and PW Doppler settings should be optimized—particularly by lowering the scale and wall filters—to detect slow arterial flow that may otherwise be missed. The absence of venous compressibility is a well-established criterion for diagnosing DVT, with reported sensitivity and specificity over 95% for proximal veins. However, intravascular tumors can also result in non-compressibility, so this single feature alone cannot reliably distinguish between thrombus and tumor. ⁸ Furthermore, contrast-enhanced ultrasonography is considered an adjunct to color Doppler flow imaging, enabling real-time assessment of tissue perfusion and enhancing diagnostic accuracy (2) Vascular tumors are associated with progressive enlargement of the affected vessel, with the mass diameter increasing as the disease advances. In contrast, although acute thrombosis can cause vessel dilation, the vessel diameter remains typically normal or only slightly reduced in subacute thrombosis and gradually decreases or becomes thinner in chronic venous thrombosis. (3) The presence of tumor metastasis or vascular invasion is another distinguishing feature, strongly indicating neoplasia rather than DVT. On the other hand, lower limb venous thrombosis causes stasis near the venous valves, potentially as a result of reduced calf muscle pump function. Furthermore, the anatomical structure of the lumen and adjacent tissues in venous thrombosis remains poorly defined on imaging.

The limitations of this study include its small sample size and single-center retrospective design, which may have introduced selection bias. Future larger-scale, multicenter prospective studies are warranted to further validate our findings and refine the proposed diagnostic criteria.

Supplemental Material