Short-term and mid-term outcome of isolated symptomatic muscular calf vein thrombosis

Results 

During the study period, 131 isolated MCVTs were identified in 128 symptomatic patients, consisting of 78 women (60.9%) and 50 men (39.1%). The average age was 57.02 ± 15.36 years (median, 57; range, 20 to 87 years). Seventy-three thromboses (55.7%) involved the SoV, and 58 (44.3%) involved the MGV. Three patients presented with both SoV and MGV thrombosis. We did not observe any isolated LGV thrombosis. Thrombosis was located in the right calf in 69 patients (53.9%) and in the left calf in 59 (P = .47). In 30 (23.4%) of 128 patients, the MCVT was combined with superficial thrombophlebitis (ST) in the calf. Initial symptoms were isolated pain in the calf in 90 patients, isolated edema of the calf only in six patients, and pain combined with edema in 32 patients. Sixty patients (46.9%) reported that they had had a previous DVT, confirmed or not.

A triggering factor was identified in 85 patients (66.4%). The most common factors were previous surgery in 23, trauma in 16, and a long car trip in 7 or lengthy air travel in 6.

The risk factors for VTE were cancer in 7 patients (5.5%), in all cases known at the time of the MCVT diagnosis; thrombophilia in 19 (14.8%), including heterozygous factor V Leiden in 16, protein S deficiencies, in 2, protein C deficiency in 1, and APL antibodies in 1; both heterozygous factor V Leiden and protein S deficiency in 1; obesity in 29 (22.7%); and 20 (25.6%) women were receiving hormone replacement therapy (contraceptive in 13; hormonal therapy of menopause in 7). No additional cases of cancer were discovered during the follow-up. An abnormality of the deep venous system (iliac, femoral, or popliteal vein) was identified in 12 patients (9.4%): 10 with PTS, including reflux in 7, obstruction in 2, and reflux and obstruction in 1; and 2 with PDVI.

The distribution of patients according to number of thrombotic veins is listed in Table I, and the distribution of patients according to the diameter of the venous lumen filled by the clot is summarized in Table II.

Table I. Distribution of patients according to the number of thrombotic veins⁎

	Veins (n)	Patients, n (%)
		MGVT	SVT
	1	10(17.24)	41(56.16)
	2	25(43.1)	23(31.51)
	3	7(12.07)	7(9.59)
	4	15(25.86)	2(2.74)
	5	0	0
	6	1(1.72)	0
	Total	58(100)	73(100.00)

MGVT, Medial gastrocnemius vein thrombosis; SoVT, soleal vein thrombosis.

⁎ Every gastrocnemius or soleal vein identified individually by duplex ultrasound imaging was counted separately.

Table II. Distribution of patients according to the diameter of the vein lumen filled by the clot

	Diameter (mm)	Patients, n (%)
		MGVT	SVT
	5	4(6.90)	3(4.11)
	6	8(13.79)	19(26.03)
	7	12(20.69)	13(17.81)
	8	12(20.69)	13(17.81)
	9	7(12.07)	12(16.44)
	10	6(10.34)	5(6.85)
	11	6(10.34)	2(2.74)
	12	1(1.72)	3(4.11)
	13	0	3(4.11)
	14	0	3(4.11)
	15	1(1.72)	0
	17	1(1.72)	0
	Total	58(100)	73(100)

MGVT, Medial gastrocnemius vein thrombosis; SoVT, soleal vein thrombosis.

In the MGV thrombosis group, the thrombosis involved one to six veins, but 82.8% of patients presented several veins involved by the thrombosis. The diameter of the thrombosis measured in the transverse plane with DUS varied from 5 to 17 mm, and 58.6% presented a thrombosis in which the diameter measured ≥8 mm. The thrombosis was confined to the intramuscular segment of the vein in 35 patients (60.3%), an extension into the extramuscular gastrocnemius vein was observed in 23 (39.7%), and extension of from 1 to 3 cm into the popliteal vein but without complete occlusion of this vein was observed in nine (15.5%).

In the SoV thrombosis group, the thrombosis involved one to four veins, but 56% presented with a thrombosis limited to a single vein. The diameter varied from 5 to 14 mm, and 56.2% of patients presented with a thrombosis with a diameter of ≥8 mm.

Anticoagulation therapy at a curative dosage was prescribed for 1 month in 53 patients (41.4%), for 3 months in 59 (46.1%), and for ≥6 months in 13 (10.2%) because they presented with a PE or major risk factors for VTE. In three patients, the duration of AT was not known.

Nine PEs (7%), clinically suspected and confirmed with radiographs, which were complicated with MCVT, were observed in the baseline examination: six (10.3%) of 58 in the MGV group and three (4.1%) of 73 in the SoV group. The difference was not significant (P = .18). No patients with PE had PTS or PDVI. We did not observe a PE in the three patients presenting with both SoV and MGV thrombosis. None of those patients died or presented severe clinical signs or major abnormalities evidenced by laboratory methods.

PE was diagnosed with ventilation-perfusion lung scan in eight patients and with helical computed tomography imaging in one. The symptoms suggestive of PE were chest pain in five patients and dyspnea in four. In addition, one patient presented with hemoptysis. Regarding the features of the MCVTs, eight of nine patients presented with two to four veins involved by the thrombosis. In all patients, the diameter of the venous thrombosis was >8 mm. In the subgroup of the six patients with MGV thrombosis and PE, the thrombus was confined to the intramuscular segment of the vein in two patients, extended into the extramuscular gastrocnemius vein in two, and into the popliteal vein in two.

Follow-up of at least 1 month was achieved in 125 (97.7%) patients (127 MCVTs), at least 3 months in 120 (93.8%) patients (122 MCVTs), and at least 9 months in 110 (85.9%) patients (111 MCVTs). Beyond 9 months, 94 patients (73.4%) were followed up with a mean of 26.7 months. There was no difference in the distribution of the followed up patients in the SoV thrombosis and MGV thrombosis groups at the different times (P was .12, .66, .20 and .35 at 1, 3, 9, and beyond 9 months, respectively).

Three patients with SoV thrombosis died, but at the time of death, anticoagulation therapy had been discontinued. The cause of death was myocardial infarction in one patient, aortic aneurysm rupture in another, and cachexia in a patient with metastatic disease.

Two (1.56%) of 128 patients presented with a serious but nonfatal hemorrhagic event: a patient with gastric cancer required a blood transfusion for digestive bleeding, and a hematoma in the upper limb occurred after venipuncture in a patient whose INR was 4.8 (the fluindione dosage was reduced and the hematoma did not require any specific treatment).

Recanalization (Table III) was complete at 1, 3, and 9 months in 54.8%, 84.7%, and 96%, respectively, in all MCVTs, without a significant difference between the SoV thrombosis and MGV thrombosis groups at the different times of follow-up. Table III shows that the numbers of DUS performed at the three follow-up visits were lower than the numbers of followed up patients because some patients in whom prolonged anticoagulation therapy had been prescribed were not assessed by DUS at every visit.

Table III. Recanalization rate

	Time period	MGVT, % (n)	SVT, % (n)	P⁎
	1 month			0.62
	0	52.1(25)	56.7(38)
	1	47.9(23)	43.3(29)
	3 months			0.97
	0	84.9(45)	84.6(55)
	1	15.1(8)	15.4(10)
	9 months			1.00
	0	95.8(46)	96.1(49)
	1	4.2(2)	3.9(2)

MGVT, Medial gastrocnemius vein thrombosis; SoVT, soleal vein thrombosis; 0, complete recanalization (ie, no postthrombotic anatomic abnormality except for venous wall or valvular thickening); 1, incomplete recanalization.

⁎ Fisher exact test.

Twenty-nine symptomatic VTE recurrent events (Table IV) occurred in 24 patients (18.8%): none at 3 months, 11 between 3 and 9 months, and 18 between 9 and 36 months. One patient presented with three VTE recurrent events, and three patients with two events. There was no significant difference in the SoV thrombosis and MGV thrombosis groups comparing the numbers of patients with VTE recurrences or the numbers of VTE events. Two (8.3%) of 24 patients had PTS with reflux. We did not observe extension of the MCVT or VTE recurrence in patients treated with anticoagulant therapy. Six nonfatal PEs were observed, four (6.9%) of 58 in the MGV thrombosis group and two (2.7%) of 73 in the SoV thrombosis group (P = .40); as were 23 DVTs, including 19 MCVTs. Ten MCVTs (MGV, 5; SoV, 5) were observed in the muscle in which the initial MVCT occurred after its recanalization. In addition, 12 STs were identified during the follow-up.

Table IV. Venous thromboembolism recurrence events⁎

	Initial Thrombosis	9 months	9-36 months	No. total (%)
		MCVT(n=2)	MCVT(n=5)
	MGVT	Crural(n=1)	Crural(n=1)	14/58(24.1%)
		PE(n=1)	Proximal(n=1)
			PE(n=3)
		MCVT(n=5)	MCVT(n=7)
	SoVT	Crural(n=1)	PE(n=1)	15/73(20.5%)
		PE(n=1)

MGVT, Medial gastrocnemius vein thrombosis; SoVT, soleal vein thrombosis; MCVT, muscular calf vein thrombosis; PE, pulmonary embolism.

⁎ No thrombosis occurred at 1 or 3 months.

Discussion 

DUS is the imaging method of choice for the diagnosis of DVT in symptomatic patients, including calf vein thrombosis. DUS has been reported to have a diagnostic sensitivity of 94% to 100%, a specificity of 91% to 100%, a positive-predictive value of 80%, and a negative-predictive value of 94% for detection of crural vein thrombosis.4, 9, 12, 26, 27

Compared with contrast venography, DUS has an overall sensitivity, specificity, and accuracy of >87% for MCVT diagnosis.10, 12 Although MCVT is commonly seen in everyday practice, few publications address this subject and no treatment guidelines are available. We were not able to determine the precise prevalence of isolated MCVT diagnosed in our Vascular Medicine Clinic during the study period because the overall number of DVTs was not determined during the MVCT inclusion period. This prevalence of all lower limb DVTs varies in publications10, 19, 20, 21, 28 between 12.5% and 25%. Labropoulos et al10 detected 742 DVTs (14%) in 5250 patients referred to the vascular laboratory for clinical suspicion of DVT. An isolated calf DVT was detected in 282 limbs (33.8%) in 251. An isolated MCVT was found in 113 limbs (SoV, n = 57; MGV, n = 48; SoV plus VGM, n = 8), yielding a prevalence of 15% of all DVTs.

We observed a similar distribution between MGV and SoV thrombosis in agreement with some publications.10, 20 In other series,12, 29 SoV thrombosis was more common. In our series we observed that in 30 patients (23.4%), MCVT was combined with a ST. This association is not surprising: in a previous article30 we had identified that ST was frequently combined with DVT and particularly with MCVT.

In our series, pain was the most common symptom suggesting DVT in patients presenting with MCVT. Pain was often severe and disabling especially with MGV thrombosis. Isolated edema was rare, observed in 4.7% (6/128).

The natural history of MCVT is poorly known. MacDonald et al12 included 219 isolated MCVTs in 185 patients (SoV, 170; MGV, 42; SoV plus MGV, 7). Their aim was to establish the incidence of propagation of untreated isolated MCVT into the deep veins of the calf and thigh. Each patient was evaluated by DUS on the first day of presentation, with a repeat examination of the involved leg 5, 9, 14, 30, and 90 days later. At 3 months, 84 (38.4%) of 219 thromboses were excluded because the patients were lost to follow-up or for various other reasons (some of them had had been treated with anticoagulant therapy). During their study, 22 patients died, although none of these deaths seemed related to thromboembolic events. Postmortem examinations were not done, so the authors reported that death from thromboembolism could not definitively be ruled out. Within their 3-month follow-up, 22 (16.3%) of the 135 limbs with isolated MCVT had thrombus extension to the level of adjacent tibial or peroneal veins or higher, 20 (91%) of 22 of which occurred within 2 weeks of the initial DUS. Four (3%) of 135 extensions to the popliteal vein were identified.

In this series, recanalization of the MCVT seemed to be less satisfactory than in our series. Indeed the MacDonald et al study12 reported a complete recanalization at 1 and 3 months in 20.7% and 44.6% of MCVT, respectively, vs 54.8% and 84.7% in our series. One hypothesis to explain this difference is that the patients in our series were treated with anticoagulant therapy. However, we cannot exclude that the assessment criteria were different in the two studies given that we considered veins with simple wall thickening as completely recanalized. We did not assess the presence or absence of post-thrombotic reflux because patients with a history of DVT were not excluded.

No treatment guidelines currently exist. Schwarz et al31 conducted a prospective, nonrandomized study including 84 MCVTs. He investigated the outcome in two cohorts of consecutive patients. The first received compression therapy and heparin for 10 days at therapeutic doses, and the second received compression therapy alone. In the 52 patients who received heparin at therapeutic dosage, no progression to DVT occurred. A statistically significant higher rate of progression into the deep calf veins (25%) was shown in the 32 patients without anticoagulant therapy. The studies by MacDonald et al12 and Schwarz et al31 showed that the rate of extension of the thrombosis was high in patients not treated with anticoagulant therapy, and suggested that in most of the patients, the extension occurred early after the diagnosis.

The Seventh American College of Chest Physicians Conference on Antithrombotic and Thrombolytic Therapy32 recommended that, “patients with acute DVT require long-term anticoagulant treatment (…). This observation applies to patients with proximal vein thrombosis and also to patients with thrombosis confined to the deep veins of the calf.” However MCVT was not specifically referenced in this recommendation.

In our series, we did not observe extension or VTE recurrence during the anticoagulant therapy period.

Even though the clinical significance of extension of the thrombus into the deep veins is not clearly established, our results, reinforced by the results of MacDonad12 and Schwarz et al,31 indicate that MCVT should be treated with anticoagulant therapy. The issue is to determine the minimum duration of anticoagulant therapy to prevent the extension of the venous thrombosis with no major risk of bleeding. This risk depends on the duration of anticoagulant therapy. With an INR target range of between 2 and 3, the yearly risk of major bleeding is estimated at 3%.33 For this reason, in our opinion, long-term anticoagulant therapy as recommended in Chest32 seems excessive. In our series we observed only two nonfatal bleeding events; both occurred in frail patients: one had metastatic disease and the other was 79 years old. We suggest that anticoagulant therapy at a curative dosage, associated with compression therapy and walking, is necessary to prevent an extension of the thrombus into the deep veins, for at least 15 to 30 days and probably longer when a risk factor for VTE is present. Management without anticoagulant therapy could be used in patients with a major risk for bleeding, but this would require monitoring with repeated DUS examinations.

The association between MCVT and PE is controversial. Three publications have reported PE in patients with isolated MCVT with a prevalence of 15%, 37.5%, and 50% respectively.20, 21, 29

In our prospective series we have identified 7% of PEs, all of which were nonfatal and with no clinical or hemodynamic signs of seriousness. In designing this study we discussed whether to include patients with MCVT complicated by symptomatic PE at the initial examination. It is indeed impossible to identify the precise location of the venous thrombosis source of the PE. At the time of clot migration, was the venous thrombosis confined to a muscular vein, or was extension of the venous thrombosis into the deep venous system present? Probably both are possible; the migration of a thrombus from the muscle can cause a distal subsegmental PE. We decided to include patients with PE in our study insofar as the inclusion criterion was the identification of an isolated MCVT by DUS in suspected patients for DVT of the lower limbs.

In our series, all patients with PE, except one, presented with extended MCVTs (several veins involved by the thrombosis), and all had a large venous thrombosis (diameter >8 mm). These data confirm the result of Ohgi et al29 results, because in that series, all patients with PE presented a large SoV thrombosis measuring >7 mm.

It is difficult to estimate whether PE is more frequently associated with MGV thrombosis or with SoV thrombosis. In our series, the prevalence of PE seemed higher in patients with MGV thrombosis than in patients presenting SoV thrombosis, but the difference was not significant. Ohgi et al29 identified PE only in patients with SoV thrombosis. The distribution of PEs was similar in the two groups of MCVT in Guias et al20 publication.

As we have previously discussed, it is impossible to identify the precise origin of the PE. Are we entitled to speak of embolic risk of isolated MCVT? We can only note that in published reports, MCVT is associated with PE with a prevalence of 7% to 50%.

The patient profile with MCVT does not appear to be very different from the patient profile with any other DVT location. A triggering factor was identified in 66.8% of patients, the most frequent being surgery. The risk factors for VTE identified in our series were similar to those reported in other DVT publications. MCVT does not always appear as an isolated VTE event in the patient's life: 47% of patients reported that they had had a previous DVT (confirmed or not), and post-thrombotic anomalies (iliac, femoral, or popliteal vein) were identified in 8% of patients at the initial DUS examination. Furthermore, we observed a high rate of VTE recurrence (18.8%, with 6 PEs) with a mean follow-up of 26.7 months, and we cannot exclude that the real incidence of VTE recurrence was higher, because 27% of patients (34/128) were not followed up beyond 9 months. Consequently, and regarding the PE risk as well, MCVT should not be considered a minor venous thrombosis.

Schwarz et al34 reported one case of recurrent MCVT due to a venous aneurysm of the SoV. We also observed a feature of aneurysm of the MGV in two patients who presented a recurrent MGV thrombosis. The 12 STs observed during the follow-up were not taken into account as a VTE recurrence but suggest that some additional patients had a thrombotic profile. We did not assess the risk factors for VTE recurrence because we did not perform systematic screening at the initial examination.

Conclusion 

This study confirms that symptomatic PE is not rare (7%) at the initial diagnosis of isolated MCVT, as shown in previous publications. We did not observe any deaths related to venous thrombosis or anticoagulant therapy; a low rate (1.5%) of nonfatal bleeding, and no extension of the venous thrombosis or VTE recurrence at 3 months but a high 18.8% rate of VTE recurrence after 3 months during the follow-up period (mean, 26.7 months). These results underline the need for clarifying the treatment of symptomatic MCVT.

We suggest that, apart from patients who present with a major risk of bleeding, symptomatic MCVT requires anticoagulant therapy at curative dosage at least for a short duration of 15 to 30 days, and probably longer when a risk factor for VTE is present, to prevent extension of the thrombus into the deep veins. Only additional prospective and randomized studies with large sample sizes can provide high-grade recommendations. In addition, the high incidence of VTE recurrence underlines the need to follow such patients as well as all patients with any type of DVT and to prescribe the usual preventive measures.

Author Contributions