Chinese Medical Journal 2013;126(23):4448-4452:10.3760/cma.j.issn.0366-6999.20131401
Differential diagnosis of isolated calf muscle vein thrombosis and gastrocnemius hematoma by high-frequency ultrasound

SU Li-ya, GUO Fa-jin, XU Guang, HAN Xiu-jie, SUN Chang-kun, ZHANG Zheng and JING Qing-hong

high-frequency ultrasound; calf; venous thrombosis; hematoma
Background Differential diagnosis of isolated calf muscle vein thrombosis (ICMVT) and gastrocnemius hematoma is essential for early identification of deep vein thrombosis (DVT). This study aimed to investigate the diagnostic value of high-frequency color Doppler ultrasound for differential diagnosis of ICMVT and gastrocnemius hematoma.
Methods A retrospective case series of 35 ICMVT (M:F, 21:14; mean age (64.5±10.6) years) and 23 gastrocnemius hematoma (M:F, 16:7; mean age (75.4±11.8) years) patients with bilateral/unilateral lower limb pain was conducted between January 2006 and September 2012. Characteristics and the morphology of high-frequency color Doppler ultrasonography of the lower limb deep vein, great saphenous vein, calf muscles, skin, and soft tissue were examined.
Results ICMVT hypoechoic signals were characterized by long, tube-like masses on longitudinal sections and oval masses on transverse sections, with apparent muscle thrombosis boundaries, distal and proximal venous connections, and, often, lower limb DVT. Gastrocnemius hematoma hypoechoic signals were characterized by large volumes, enhanced posterior hematoma echo, hyperechoic muscle boundaries, no hematoma blood flow, and no DVT, and clear differences in trauma/exercise- and oral anticoagulant-induced hematomas were readily apparent. According to the measurement, the ratio of long diameter/transverse diameter (D/T) in ICMVT patients was about less than 2.0, whereas in gastrocnemius hematoma patients the ratio was more than 2.0. Early stage isoechoic and hypoechoic signals were detected with gradually increasing ovular anechoic areas. Partial muscle fibers in the hematoma due to muscle fractures were apparent.
Conclusion High-frequency color Doppler ultrasound was found to be a sensitive and reliable method for differential diagnosis of ICMVT and gastrocnemius hematoma due to trauma and exercise or prolonged oral anticoagulant use.
Deep vein thrombosis (DVT) of the legs can result in symptoms that range from mild nonspecific pain, swelling, and redness due to engorged superficial veins to life-threatening pulmonary embolism caused by thrombus embolization to the lungs.1Diagnosing DVT and identifying patients who may be at risk for future DVT development are challenging in clinical settings because DVT, hematoma, ruptured popliteal cyst, infection, lymph edema, and sarcoma have many similar signs and symptoms, including unilateral calf swelling and pain, but require very different treatments.2Therefore, improved methods for timely and accurate differential diagnosis are urgently needed to improve prognosis in DVT patients.
A thrombus often forms in the distal regions of the calf muscle veins without DVT,3known as isolated calf muscle vein thrombosis (ICMVT).4In fact, ultrasound studies have revealed that in up to 33% of all DVT patients the problems originate from the calf veins due to ICMVT.5Thus, early diagnosis of ICMVT is essential for treatment of DVT and good prognosis.
A notable number of ICMVT patients, however, are misdiagnosed as gastrocnemius hematoma. Despite diagnosis of lower limb DVT by high-frequency color Doppler ultrasound has been extensively reported,4the characteristic similarities and differences between ICMVT and gastrocnemius hematoma using this method are scarcely reported. Thus, the current study examined the morphological and ultrasonic characteristics of ICMVT and gastrocnemius hematoma in a series of patients treated at a single facility.
Study design
A retrospective case series of ICMVT patients (n=35; M:F, 21:14; mean age (64.5±10.6) years with range of 38–88 years) and gastrocnemius hematoma patients (n=23; M:F, 16:7; mean age (75.4±11.8) years with range of 46–90 years) admitted to the emergency department or inpatient department of the Beijing Hospital (China) between January 2006 and September 2012 was examined.
All the patients included in this study were adults (age >18 years), and initially presented with bilateral or unilateral lower limb swelling and discomfort, with the major complaint of pain, walking difficulty, or petechiae in the leg; their diagnosis was ICMVT or gastrocnemius hematoma using high-frequency color Doppler ultrasound. Later on, five cases received surgery to remove focus, four cases received vein puncture and noncondensing blood was extracted, and four cases were also diagnosed by Magnatic Resonance Imaging (MRI) with fresh hemorrhage. Diagnoses were also confirmed by follow-up outcomes of ultrasonography; all these cases had complete ultrasound image records and complete clinical information.
Examinations were performed on the lower limb vein as previously described.6Briefly, patients were positioned at prone or lateral posture with legs in flexion or elevated, and the popliteal, femoral, great saphenous, lesser saphenous, tibial, and peroneal veins and the calf muscle venous plexus were examined by color and pulsed wave Doppler ultrasonography. A longitudinal section was used when the veins were intermittently exposed at the cross section. Peroneal and posterior tibial veins were examined along with the gastrocnemius and soleus muscles. Patients were examined by high-frequency color Doppler ultrasound at the time of admission (2 hours–7 days after symptom onset) using a Philips IU22 xMatrix Ultrasound System with wide view imaging functionality (Royal Philips Electronics, MA, USA), LOGIQ 9 (General Electric Company, CT, USA), and ESAOTE DU6 color Doppler ultrasonic diagnostic apparatus (Genova, Italy) with a probe frequency of 5.0–7.5 MHz for patients with normal Body Mass Index (BMI) and 3.5–5.0 MHz for obese patients or patients with apparent swelling. The lower limb deep vein, great saphenous vein, calf muscles (the gastrocnemius and soleus muscles), and skin and soft tissues were examined on the affected side. The opposite limb was checked as reference.
Data collection
From ultrasonographic data, abnormal echoes characteristic in the calf muscle were recorded, including those of the lesion envelope, relationship with surrounding tissues, connection with blood vessels, echo enhancement posterior to lesions, and blood flow signals. Furthermore, morphological characteristics of lesion size, boundary, internal echogenicity, and flow features were recorded using archived ultrasonic images and recorded data.
Diagnostic criteria for DVT
The diagnostic criteria for DVT applied in this study were similar to those published by Wells et al.7Briefly, DVT was described as widened vein diameter with poor lumen acoustics upon color Doppler or pulsed Doppler ultrasound examinations; low or no echo from the vessel; disappearance of blood flow when vessels were extruded transversely, without lumen deflation; disappearance or weakening of the spectrum phase of the distal vein; and disappearance or weakening of the Valsalva reaction. Simple unilateral calf muscle vein thrombosis, tibial DVT, fibula DVT, and mixed DVT were distinguished.
Follow-up outcomes
Post-treatment venous obstruction and alteration, venous connections, and symptom alleviation were recorded. Major adverse events, surgeries, and deaths were also reported.
Statistical analysis
All data were recorded as mean ± standard deviation (SD) for continuous data or percentages for categorical variables and analyzed using SPSS version 18.0 (SPSS, Inc., Chicago, IL, USA). Between-group and within-group data were compared using χ2tests for categorical variables or independent samplest-tests for continuous data, respectively.P<0.05 was considered as statistically significant.
Patient demographic and clinical characteristics
Of the 35 ICMVT and 23 gastrocnemius hematoma patients, unilateral right limb symptoms were observed in 10 (28.6%) and eight (34.8%) patients, respectively, and unilateral left limb symptoms were observed only in 23 (65.7%) and 15 (65.2%) patients, respectively. Bilateral symptoms were observed in two (5.7%) ICMVT patients and 0 (0%) gastrocnemius hematoma patients. A total of 20 (57.1%) ICMVT patients exhibited a single thrombosis, while 15 (48.6%) exhibited multiple thrombi (Table 1). Of ICMVT patients, 17 (48.6%) underwent the limb surgery or exhibited bone fracture, six (17.1%) had experienced limb trauma or sprain requiring bed rest history, nine (25.7%) underwent abdominal or pelvic surgery, and three (8.6%) were idiopathic. Of gastrocnemius hematoma patients, nine (39.1%) were prescribed long-term anticoagulants (warfarin or enteric-coated aspirin) after cardiac valve replacement and lower limb DVT history, one (4.3%) was currently treated with Plavix for multiple myeloma and long-term bed rest, nine (39.1%) had suffered trauma in the affected limb, two (5.7%) exhibited history of severe movement dysfunction, eight (22.9%) had ecchymosis, and two (8.7%) were idiopathic. No significant differences in gender, age, affected side, or idiopathic presentations were observed between groups (P>0.05); however, the side where there was occurrence of surgeryor trauma or use of oral medication varied significantly (P<0.05, Table 2).
DVT occurrence
In the ICMVT group, 21 (60.0%) patients were diagnosed with DVT, with five (14.3%) tibial DVT patients, one (2.9%) fibial DVT patient, and 14 (40.0%) mixed DVT patients. The two ICMVT patients with bilateral lesions were diagnosed with left tibial DVT or left mixed DVT+right tibial/fibula DVT. While no patients with gastrocnemius hematoma were diagnosed with DVT, one gastrocnemius hematoma patient with myeloma was initially misdiagnosed as DVT. Emergency ultrasonography indicated poor filling of the posterior tibial and peroneal veins. Subsequent administration of anticoagulant therapy resulted in exacerbation of the condition of the affected limb, with symptoms gradually alleviated after terminating anticoagulant therapy.
ICMVT high-frequency color Doppler ultrasound features
ICMVT was characterized by a round or oval homogeneous hypoechoic signals on the longitudinal section of the gastrocnemius and soleus muscles that exhibited 2.1–9.8 cm
long, tube-like or dendritic structures. In the transverse sections, round hypoechoic signals with a maximum diameter of 0.41–1.48 cm were apparent. Notably, hypoechoic areas were not compressed, and no internal blood flow signal was visualized. Intact gastrocnemius and soleus muscle fibers were apparent, and thrombosis muscle boundaries were clearly delineated (Figure 1, Table 3).
Gastrocnemius hematoma high-frequency color Doppler ultrasound features
Characteristic features of gastrocnemius hematoma in high-frequency color Doppler ultrasonography varied by hematoma stages and causes (trauma/exercise or oral anticoagulant induced). Gastrocnemius muscle rupture due to exercise or trauma (11/23, 47.8%) exhibited hypoechoic signals with few irregular anechoic areas located within the muscle layers, with some nonintact muscle fibers and hematoma-occupied defects. These high-tension hematomas were ovular on longitudinal sections (maximum diameter, 3.4–5.6 cm) and round on transverse sections (transverse diameter, 2.4–4.3 cm; anteroposterior diameter, 1.9–4.5 cm, Figure 2). Spontaneous hematomas caused by long-term oral anticoagulant treatment (10/23, 43.5%) exhibited hypoechoic lesions with irregular anechoic areas. In these low-tension lesions, the major lesion was anechoic mixed with hyperechoic strips, sometimes divided into several parts and mixed with cloudy echo or irregular heterogeneous hypoechoic signals. Hematomas were oval or fusiform on longitudinal sections (maximum diameter, 4.8–13.7 cm) and ovular on transverse sections (anteroposterior diameter, 3.7–5.1 cm). Surrounding muscles were extruded by hematomas, and heterogeneous and hyperechoic signals were observed between hematomas and muscles. The echo signal posterior to the lesion was enhanced, and the blood flow signal was not detected within the lumen (Figure 3, Table 3).
Treatment outcomes
In both tibial and mixed DVT patients, ICMVT patients experienced posterior tibial venous thrombosis (5/35, 14.3%) or pulmonary venous thrombosis (7/35, 20%). All other ICMVT patients exhibited proximally and distally tapering veins. All symptoms were gradually alleviated by thrombolysis and anticoagulant treatments. High-frequency color Doppler ultrasound for 5–14 post-treatments revealed 11 completely recanalization veins, 16 partially recanalization veins, seven unaltered veins, and three veins progressing. A total of eight (34.8%) patients with gastrocnemius hematomas were diagnosed as large hematomas (>2 cm in diameter), in which the surrounding blood vessels were compressed by the lesion and blood flow slowed down. Notably, normal flow was still observed by pressing on the distal extremities. Hematomas were surgically removed in four (17.4%) patients who underwent ultrasound-guided decompression. No deaths were reported in either group.
A thrombus often forms in the distal regions of the calf muscle veins without DVT,3known as ICMVT.4Within a single week, these thrombi can spread to the proximal veins in up to 25% of patients if untreated, often leading to deep calf vein thrombosis.6-8Venous thromboembolism, encompassing both deep vein thrombosis (DVT) and pulmonary embolism, is a major cause of morbidity and mortality in hospitalized patients or patients undergoing surgery for various conditions, particularly elderly or bedridden.9A notable number of ICMVT patients, however, were misdiagnosed as gastrocnemius hematoma, a condition generally induced by trauma or athletic activity that is generally responsive to conservative treatments and rarely develops into DVT.10Thus, in many patients inappropriate treatments that exacerbate the ICMVT increase the chance of DVT occurrence, and prolonged symptoms are administered.2While proper treatment for ICMVT is still highly debated,8the important first step to limit DVT occurrence and resultant poor prognosis in those patients is to establish a method for a clear differential diagnosis of ICMVT and gastrocnemius hematoma. The current study expands the ability of high-frequency color Doppler ultrasound to distinguish ICMVT from gastrocnemius hematoma. In previous studies, the ultrasonic characteristics of ICMVT11,12and gastrocnemius hematoma13have been described consistently with the present study.
From data of the inducement of the two diseases, ICMVT mainly resulted from bed rest history (32/35, 91.4%); however, gastrocnemius hematoma resulted not only from bed rest history (11/23, 47.8%) but also from oral anticoagulation (10/23, 43.5%, Table 2). Furthermore, Antignani and Aluigi4recently suggested high-frequency color Doppler ultrasound as the diagnostic gold standard for indicating DVT. Follow-up data of ICMVT patients in the current study revealed lesion signals completely conform to evolution of thrombus (Table 4). Meanwhile, it is well known that early period of hematoma exhibits hypoechoic-isoechoic lesions with irregular boundaries, and as the time elapses, the major lesion was anechoic mixed with hyperechoic strips, as a consequence of thrombolysis and fibrin exudation.14The current data also conform to evolution of hematoma (Table 4). So, high-frequency color Doppler ultrasoundcan also be considered to be gold standard for indicating ICMVT and gastrocnemius hematoma. Moreover, each type of lesion has unique ultrasonic characteristics, including muscle hematoma boundary, echo signal, posterior echo signal, blood flow, dimensions, and morphology, that can be broadly used to distinguish those patients in clinical settings.
The current study revealed the differences in morphological and ultrasonic characteristics between ICMVT and gastrocnemius hematoma (Table 3).
Most importantly, lesions of ICMVT (within 7 days after symptom onset) were characterized by hypoechoic–isoechoic and anechoic signals, about 83.8% (31/37) and 16.2% (6/37), and no mixed echo signals were detected. However, lesions of gastrocnemius hematomas were characterized by hypoechoic–isoechoic and mixed choic signals, about 78.3% (18/23) and 21.7% (5/23), and no anechoic signals were detected (Table 4). According to the measurement, the ratio of long diameter/transverse diameter (D/T) in ICMVT patients was about less than 2.0, whereas in gastrocnemius hematomas patients the value of D/T was about more than 2.0 (P<0.05). The current study further revealed, unlike other studies, that spontaneous gastrocnemius hematomas in the patients undergoing anticoagulant therapy exhibit different and unique ultrasonographic features, of which the muscle fiber is integrated while the muscle remains obviously extruded. Thus, these injuries are readily observed based on the hyperechoic signals around the hematoma, which, pending further study, may be useful in developing broadly applicable clinical guidelines for differential diagnosis of ICMVT and gastrocnemius hematoma due to either trauma or oral anticoagulant administration.
Furthermore, the presence of large gastrocnemius hematomas can obstruct distal venous reflux, thus slowing blood flow detected by routine ultrasound and complicating accurate diagnosis.15Therefore, the application of high-frequency color Doppler ultrasound is suggested to improve diagnostic sensitivity.4Additionally, careful application of pressure and examination of resultant blood flow may help to discriminate between these conditions, according to the current findings. These techniques may be useful additions to current techniques for distinguishing venous thrombosis and hematoma, therefore avoiding potential misdiagnoses and missed diagnoses. In addition, high-frequency color Doppler ultrasound is relatively inexpensive, simple, and noninvasive compared with MRI and CT examinations.16So, previously published work and the current findings indicate that high-frequency color Doppler ultrasound of the calf muscle examination should be included as a routine method for inspection of the venous function in the lower limb.
Recently, high-frequency color Doppler ultrasound has become the gold standard for venography of the lower limbs and a critical tool in venous thromboembolism prevention;4however, few guidelines exist that detail the differential diagnosis of conditions with high potential for DVT and other conditions. The current study indicates that ICMVT and gastrocnemius hematoma can be clearly distinguished using a set of ultrasonographic features, potentially providing a valuable tool for use in clinical DVT diagnosis and prevention.
  • 1.

    Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep vein thrombosis. Lancet 2012; 379: 1835-1846.

  • 2.

    Akar S, Manisali M, Birlik M, Onen F, Akkoc N. A case with recurrent calf pain and swelling: recurrent spontaneous calf haematoma . Rheumatol Int 2002; 21: 247-249.

  • 3.

    Palareti G, Schellong S. Isolated distal deep vein thrombosis: what we know and what we are doing. J Thromb Haemost 2012; 10: 11-19.

  • 4.

    Antignani PL, Aluigi L. The calf vein thrombosis. Rev Vasc Med 2013; 1: 1-4.

  • 5.

    Mattos MA, Melendres G, Sumner DS, Hood DB, Barkmeier LD,Hodgson KJ, et al. Prevalence and distribution of calf vein thrombosis in patients with symptomatic deep venous thrombosis: a color-flow duplex study. J Vasc Surg 1996; 24: 738-744.

  • 6.

    Zhang XR, He W, Tang H, Liang XN, Xu LQ, Dai KD. Imaging diagnosis of deep venous thrombosis in the calf. Chin J Med Imaging Technol 2004; 2: 1064-1067

  • 7.

    Wells PS, Owen C, Doiucette S, Fergusson D, Tran H. Review: Does this patient have deep vein thrombosis? JAMA 2006; 295: 199-207.

  • 8.

    Labropoulos N, Webb KM, Kang SS, Mansour MA, Filliung DR,Size GP, et al. Pattern and distribution of isolated calf deep vein thrombosis. J Vasc Surg 1999; 30: 787-793.

  • 9.

    Schwarz T, Buschmann L, Beyer J, Halbritter K, Rastan A, Schellong S. Therapy of isolated calf muscle vein thrombosis: a randomized, controlled study. J Vasc Surg 2010; 52: 1246-1250.

  • 10.

    Shah JR, Shah BR, Shah AB. Pictorial essay: Ultrasonography in “tennis leg”. Indian J Radiol Imaging 2010; 20: 269-273.

  • 11.

    Brain S,Fries RB,Bhagwan S. Clinical characteristics of patients with isolated calf vein thrombosis in a large teaching hospital. Int J Vasc Med 2011; 2011: 1-4.

  • 12.

    Kazmers A, Groehn H, Meeker C. Acute calf vein thrombosis: outcomes and implications. Am Surg 1999; 65: 1124-1127.

  • 13.

    Flecca D, Tomei A, Mavazzolo N, Matinelli M. US evaluation and diagnosis of rupture of the medial head of the gastrocnemius (tennis leg). J Ultrasound 2007; 10: 194-198.

  • 14.

    Su LY, Xu G, Peng Y. The value of color Doppler ultrasound in the diagnosis of soft tissue layer hematoma. Chin Physician Med 2009; 32: 24-26.

  • 15.

    Yang P, Wu QH, Lin Y, Zhang YA, Zhang JH. Diagnostic value of noninvasive vascular test in calf haematoma and deep vein thrombosis. J Cardiovasc Pulm Dis 2002; 21: 209-210.

  • 16.

    Li ZY, Zhang JT, Feng J, Xia R, Li QS, Liu WZ, et al. Diagnosis value of ultrasound on calf muscle venous plexus thrombosis. J Ultrasound Clin Med 2009; 11: 478-480.

(Received May 28, 2013)
Edited by PAN Cheng
Table 1.Clinical and demographic information of included patients
Protocol ICMVT
χ2ort Pvalues
Gender (M:F) 21:14 16:7 0.55 0.46
Age (years; mean±SD
66.5±10.6 (38–88) 68.3±11.8 (46–90) 0.58 0.56
Affected side (n(%))     0.62
Left 23 (65.7) 15 (65.2)
Right 10 (28.6) 8 (34.8)
Bilateral 2 (5.7) 0 (0)
Thrombi (n(%))    
Single 20 (57.1)    
Multiple 15 (48.6)    
Table 2.The inducement of ICMVT and gastrocnemius hematoma
(n=35) (n(%))
hematoma (n=23) (n(%))
Surgery or trauma 32 (91.4) 11 (47.8)
Oral mediation 0 (0) 10 (43.5) 0.00
Idiopathic 3 (8.6) 2 (8.7)
Table 3.Summary of ultrasonic characteristics of ICMVT and gastrocnemius hematoma
ICMVT (n=35)
Gastrocnemius hematoma (n=23)
Trauma/exercise induced Oral anticoagulant induced
Longitudinal morphology Round or ovular; tube-like or dendritic structures Ovular Ovular or fusiform
Longitudinal length 2.1–9.8 cm maximum diameter 3.4–5.6 cm maximum diameter 4.8–13.7 cm maximum diameter
Transverse morphology Rounded Rounded Eounded
Transverse length 0.41–1.48 cm maximum diameter 2.4–4.3 cm transverse diameter; 1.9–4.5 cm anteroposterior diameter 3.7–5.1 cm anteroposterior diameter
Deflation by local pressure No Yes Yes
Internal blood flow signal No Slowed Not detected
Intact gastrocnemius and soleus
muscle fibers
Yes Some nonintact muscle fibers Some nonintact muscle fibers
Thrombosis muscle boundaries Clearly apparent Highly obscured Obscured
Echo signal Hypoechoic signal Hypoechoic signals with few irregular anechoic regions Irregular anechoic regions with several intervals mixed with flocculation or irregular hypoechoic signals
Posterior area echo signal Normal Enhanced Enhanced
Defects None Some hematoma-occupied defects None
Tension None High tension Low tension

view in a new window
Figure 1. High-frequency color Doppler ultrasound of acute muscular venous thrombosis. 1A: Dendritic hypoechoic signal on the longitudinal section. 1B: Round hypoechoic signal on the transverse section.
Figure 2. High-frequency color Doppler ultrasound of gastrocnemius hematoma following trauma or exercise. Post-symptom occurrence at Day 3 showing apparent anechoic signals with mixed irregular hypoechoic signals and partially nonintact muscle fibers with defects occupied by hematomas (2A). Post-symptom occurrence at 7 hours showing an oval isoechoic signal in the posterior of gastrocnemius muscle (2B). The hematoma exhibited high tension, with a hyperechoic boundary and enhanced signal from the posterior of the hematoma. Nonintact muscle fibers were observed in the gastrocnemius muscle.
Figure 3. High-frequency color Doppler ultrasound of gastrocnemius hematoma induced by oral anticoagulant use. Post-symptom occurrence at Day 1 showing hypoechoic lesions with irregular anechoic regions (3A). Post-symptom occurrence at Day 6 showing soft tissues exhibiting a spindle shape with irregular hypoechoic signals (3B). Enhanced signals were observed from the posterior of the hematoma. The surrounding muscle was extruded, and an enhanced hyperechoic signal was observed between the muscle and hematoma.
Table 4. Follow-up data of included patients (n)
Items Anechoic Hypoechoic-isoechoic Mixed echo
First check -up 6 31 0
After 3–7 days 0 37 0
Gastrocnemius hematoma      
First check-up 0 18 5
After 3–7 days 0 3 20
Fisher’s Eexactttest,P=0.01;χ2=19.71, df=1,P=0.00.