Comparison of clinical outcomes among patients with isolated axial vs muscular calf vein thrombosis: A systematic review and meta-analysis

Abstract

Objective

Thrombi in the axial calf veins have quite different anatomical and physiological characteristics from that in the muscular calf veins, but their treatment was usually addressed in the same manner. We performed a meta-analysis of randomized and cohort studies to compare clinical outcomes among patients with isolated axial vs muscular calf deep vein thrombosis (DVT).

Methods

Recurrent venous thromboembolism (VTE) was selected as the primary outcome. Resolution, proximal propagation of calf DVT, pulmonary embolism (PE), major bleeds, and clinically relevant non-major bleeds were separately analyzed as secondary outcomes. Data were pooled and compared with risk ratio (RR) and 95% confidence interval (CI).

Results

Thirteen studies, consisting of 4889 patients, met the inclusion criteria and were included for analysis. A greater rate of recurrent VTE (FE model: RR, 1.23; 95% CI, 1.00-1.53; I² = 29%), resolution (FE model: RR, 1.32; 95% CI, 1.01-1.72; I² = 31%), proximal propagation (FE model: RR, 1.63; 95% CI, 1.10-2.41; I² = 40%), and PE (FE model: RR, 2.79; 95% CI, 1.31-5.95; I² = 0%) in the axial group compared with the muscular group. There was no difference in the pooled estimates for major bleeds (FE model: RR, 1.09; 95% CI, 0.61-1.95; I² = 0%), and clinically relevant non-major bleeds (FE model: RR, 1.80; 95% CI, 0.93-3.48) in the axial and muscular arms.

Conclusions

Patients with calf DVT limited to muscular veins might have a lower rate of recurrent VTE, resolution, proximal propagation, and PE vs those with axial calf vein involvement and exhibited similar safety outcomes.

Graphical abstract

Article Highlights.

• •Type of Research: A clinical paper
• •Key Findings: A greater rate of recurrent venous thromboembolism (relative risk [RR], 1.23; 95% confidence interval [CI], 1.00-1.53; I² = 29%), resolution (RR, 1.32; 95% CI, 1.01-1.72; I² = 31%), proximal propagation (RR, 1.63; 95% CI, 1.10-2.41; I² = 40%), and pulmonary embolism (RR, 2.79; 95% CI, 1.31-5.95; I² = 0%) in the axial group compared with the muscular group.
• •Take Home Message: Patients with calf deep vein thrombosis limited to muscular veins might have a lower rate of recurrent venous thromboembolism, resolution, proximal propagation, and pulmonary embolism vs those with axial calf vein involvement.

Deep vein thrombosis (DVT), which is a common disease and could lead to fatal pulmonary embolism (PE), seriously harms worldwide population health.¹ Isolated calf DVTs, also known as distal DVTs, account for approximately one-half of all DVTs.² Owing to the uncertainty about the clinical features and outcomes of isolated calf DVT, the optimal management of patients with isolated calf DVT has not been clearly established.

In the past, isolated calf DVT was often studied as a single category.2, 3, 4, 5, 6, 7, 8, 9 With the improvement of diagnostic technology, the calf DVT can be further subdivided into axial calf DVTs (ie, peroneal, anterior tibial, and posterior tibial vein thrombosis) and muscular calf DVTs (ie, soleal and gastrocnemius vein thrombosis). Thrombi in the axial calf veins have quite different anatomical and physiological characteristics from that in the muscular calf veins, but their treatment were usually addressed in the same manner.¹⁰ The latest American College of Chest Physicians guidelines consider that isolated muscular calf DVTs have a lower risk of extension than thrombosis that involves the axial veins.¹¹ However, it remains undefined whether anticoagulation is equally applicable to all isolated calf DVTs or whether a distinction should be made between axial calf DVTs and muscular calf DVTs.

To the best of our knowledge, clinical features, outcomes and the optimal treatment of isolated axial or muscular calf DVTs have never been systematically evaluated. Therefore, we performed a meta-analysis of RCT and cohort studies to compare outcomes among patients with isolated axial vs muscular calf vein thrombosis.

Methods

Search strategy and study selection

The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines was followed to report this meta-analysis.¹² We performed a systematic review of the literature searching using the electronic databases Medline, Embase, Wanfang, and China National Knowledge Internet (CNKI) databases up to Match 2023. The literature search strategy included the following terms: ‘distal deep venous thrombosis [All Fields]’ OR ‘below-knee vein thrombosis [All Fields]’ OR ‘calf deep vein thrombosis [All Fields]’ OR ‘calf vein thrombosis [All Fields]’ OR ‘axial calf veins [All Fields]’ OR ‘axial distal deep vein thrombosis [All Fields]’ OR ‘calf crural vein thrombosis [All Fields]’ ‘deep calf paired vein thrombosis [All Fields]’ OR ‘deep calf vein thrombosis [All Fields]’ OR ‘calf muscle venous thrombosis [All Fields]’ OR ‘anterior tibial vein thrombosis [All Fields]’ OR ‘posterior tibial vein thrombosis [All Fields]’ OR ‘peroneal vein thrombosis [All Fields]’ OR ‘gastrocnemius vein thrombosis [All Fields]’ OR ‘intramuscular vein thrombosis [All Fields]’ OR ‘soleal vein thrombosis [All Fields]’.

To lessen the risk of potential selection bias, randomized controlled trials or cohort studies that compared outcomes among patients with isolated axial vs muscular calf DVTs was approached for study enrollment, irrespective of initial management. Patients diagnosed with calf DVTs by compression ultrasonography were enrolled in this research. Reviews, editorials, letters, case reports, and studies from 20 years earlier were excluded. Articles were limited to the English and Chinese languages.

All titles and abstracts were read for relevance by two independent authors (T.W. and C.J.W). Further, the articles that matched the inclusion criteria were enrolled to an independent full-text evaluation for final eligibility and data extraction. If the title and abstract do not provide sufficient information to make a judgment, the article was also subjected to a full text assessment. Discrepancies were resolved by consensus or opinion of a third reviewer (C.S.) if a consensus could not be reached.

This paper is a meta-analysis paper, and the approval of the Institutional Review Board is not required.

Data extraction and quality assessment

Two reviewers (T.W. and C.J.W.) independently extracted relevant data and transferred the information to a template in Microsoft Excel. The following items were extracted: design of study, publication year, group, number of included patients, baseline characteristics, method used for thrombosis diagnosis, type of anticoagulant, dose of anticoagulant, anticoagulation duration, follow-up duration, and outcomes (number of thrombus resolution, progression of calf DVT, recurrent calf DVT, proximal propagation of calf DVT, PE, major bleeding, and clinically relevant non-major bleeds). Patients with thrombus involving axial veins (posterior tibial vein, anterior tibial vein, peroneal vein, tibioperoneal trunk) were included in the axial calf DVT group,¹³ and those with thrombus limited to muscular (gastrocnemius, soleal) veins were placed in the muscular calf DVT group.

A modified version of Newcastle-Ottawa Quality Assessment Scale for cohort studies was used to assess the quality of each individual cohort study, and the Cochrane Risk of Bias tool was performed to evaluate the quality of each individual RCT. Heterogeneity was calculated for each summarized outcome using the Der Simonian-Laird method.

Outcomes

The primary composite outcome of this was the recurrence of VTE, including recurrent non-propagation of calf DVT, increased thrombus burden within the same named vessel, proximal propagation of calf DVT, and PE occurring during study period. Recurrent calf DVT was defined as an increase in the diameter of a previously thrombosed vein segment by >3 mm, or a new calf DVT in the contralateral leg. Proximal propagation of calf DVT was defined as extension of isolated calf DVT to popliteal, femoral, or iliac vein.

Resolution of calf DVT, proximal propagation of calf DVT, PE, major bleeds, and clinically relevant non-major bleeds were separately analyzed as secondary outcomes. Major bleeds and clinically relevant non-major bleeds were defined by International Society on Thrombosis and Haemostasis criteria.^14,15

Statistical analysis and risk of bias assessment

Review Manager 5 was used to pool individual study results in a weighted fashion and to calculate the measures of treatment effect. The risk ratio (RR) with 95% confidence intervals (CIs) was assessed for each study. A P-value less than .05 was considered to be statistically significant.

Publication bias was assessed using funnel plots. I² statistics were calculated to measure the amount of heterogeneity. In cases of no or low heterogeneity, a fixed-effect model was used for analysis. When moderate/significant heterogeneity (I² greater than 0.50), subgroup analyses were conducted to explore sources of heterogeneity. If heterogeneity still exists after repeat analysis, a random-effect model was used.

Results

Study population and qualitative analysis

The initial literature search retrieved 11,315 articles, and 9267 articles remained after removal of duplicates. After scanning of the title and/or abstract, we excluded 9072 studies and examined 195 full-text papers for eligibility. Next, we excluded an additional 182 studies because they were off the topic, they conducted more than 20 years ago, they utilized venography as a diagnosis method for calf DVT, they reported on duplicate data, or they were reviews, editorials, letters, or case reports. Finally, 13 studies,^13,16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 consisting of 4889 patients, met the inclusion criteria and were included for analysis. The PRISMA flow diagram was used to outline the selection process and reasons for exclusion in Fig 1.

Fig 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for search strategy.

Included studies

Baseline characteristics of included studies are shown in Table I and the Supplementary Table (online only). Of the 13 studies included, one¹⁶ was a randomized controlled trial (RCT), four^18,19,25,26 were prospective cohort studies, and eight^13,17,20, 21, 22, 23^,27 were retrospective cohort studies. Twelve^13,16, 17, 18, 19, 20, 21, 22, 23^,25, 26, 27 of the studies entirely included patients with isolated calf DVT, of which eight studies16, 17, 18, 19^,21,25, 26, 27 specifically included patients with symptomatic isolated calf DVT. In the other one study,²⁴ patients with isolated calf DVT were included as subgroups of a larger cohort of patients with hospital-acquired lower extremity DVT. Study size ranged from 40 to 1490 patients. The mean or median age of the patients and the prevalence of male gender varied from 44.5 to 65.0 years and 41.8% to 64.3%, respectively.

Table I.

Characteristics of included studies

Author	Year	Nation	Study design	Population	Thrombophilia, %	Symptomatic	No.	Treatment	Outcomes reported	Follow-up test method	Follow-up time frame, months
Ageno	2022	Italy	RCT	–	58.5	Yes	302	Rivaroxaban	Recurrent VTE	CUS	0, 0.75, 1.5, 3, 24
Donadini	2017	Italy	Retrospective cohort	Unclear	51.4	Yes	285	LMWH	Recurrent VTE, proximal DVT, isolated distal DVT, unusual site thrombosis	CUS + venography	6
Galanaud	2010	France	Prospective cohort	Unclear	90.2	Yes	612	86.5% ACVT, 76.7% MCVT treated with anticoagulants	Death, recurrent VTE, major bleeds	CUS	3
Galanaud	2014	France	Prospective cohort	Inpatients, outpatients	40.6	Yes	379	Discontinuation of anticoagulants	Recurrent VTE	CUS + venography	3, 12, 24, 36
Kim	2022	Korea	Retrospective cohort	Inpatients	>41.5	part	103	71.1% had received anticoagulation therapy.	Recurrent VTE, major bleeds	CUS	12
Krutman	2016	Brazil	Retrospective cohort	Unclear	>18.9	Yes	40	LMWH+OAC	Asymptomatic PE, major bleeds	Venography	/
Kuczmik	2020	USA	Retrospective cohort	Inpatients, outpatient	96.9	unclear	647	13.4% ACVT, 15.6% MCVT without anticoagulants.	recurrent VTE, symptomatic PE, major bleeds, clinical relevant non-major bleeds, death	CUS+ venography	24
Prahlad	2016	Australia	Retrospective cohort	Unclear	68.7	part	166	88% with therapeutic dose anticoagulation	Recurrent VTE	CUS	24
Qiu	2021	China	Retrospective	Inpatients	Unclear	unclear	1490	LMWH + VKA	PE	Venography	/
Sartori	2014	Italy	Prospective cohort	Outpatients	Unclear	Yes	90	LMWH	Symptomatic PE, proximal DVT, recurrent VTE	CUS + venography	3, 6, 12, 24
Sartori	2016	Italy	Prospective cohort	Outpatients	>29.1	Yes	172	LMWH	Clot recanalization, major bleeds	CUS	3
Sule	2009	Singapore	Retrospective cohort	Unclear	>28.6	Yes	45	LMWH + VKA	Complete resolution, progression of DVT, PE, death	CUS + venography	7
Utter	2020	USA	Retrospective cohort	Inpatients, outpatients	Unclear	Unclear	458	UFH, LMWH, VKA, a direct thrombin inhibitor	Proximal DVT or PE	CUS + venography	6

ACVT, Axial calf vein thrombosis; BMI, body mass index; CUS, compression ultrasonography; DVT, deep vein thrombosis; LMWH, low molecular weight heparin; MCVT, muscular calf vein thrombosis; OAC, oral anticoagulant; PE, pulmonary embolism; RCT, randomized controlled trial; UFH, unfractionated heparin; VKA, vitamin K antagonist; VTE, venous thromboembolism.

Data were analyzed descriptively according to the location of the calf DVT: axial calf DVT and muscular calf DVT. All included studies confined isolated clots in gastrocnemius or soleus veins as muscular calf DVT. However, five studies17, 18, 19, 20^,24 classified thrombosis confined only to tibial or fibular veins as axial calf DVT, six studies^{13,20,21,23,25,26} classified thrombosis of axial calf vein as axial calf DVT, including patients with both axial and muscular calf DVT, and the other two^16,27 had no clear definition.

Methodology

All studies used compression ultrasound scan of both lower limbs for initial diagnosis of muscular or axial calf DVT. However, studies differed in the duration, frequency, and method of follow-up. Across studies, follow-up duration varied between 3 months and 3 years. For confirmation of proximal propagation of calf DVT, surveillance whole-leg compression ultrasonography of both lower limb was used in five studies,^{16,18,20,23,26} and combined venography in six studies.^{13,17,19,22,25,27} Respecting the diagnosis of the outcome event, it was based on symptomatic plus asymptomatic (discovered by routine screening) events in six studies,^{13,16,22,23,26,27} and on reporting of symptomatic events in five studies.17, 18, 19, 20, 21 (Table I).

Primary outcome

Recurrence of VTE was defined as primary outcome. Twelve studies (1 RCT, 11 cohort studies),^13,16, 17, 18, 19, 20, 21, 22, 23, 24, 25^,27 including 4717 patients, reported on the rates of recurrent VTE for both the axial and muscular arms. The comprehensive analysis of these studies exhibited a greater rate of recurrent VTE in the axial group compared with the muscular group (FE model: RR, 1.23; 95% CI, 1.00-1.53; I² = 29%). In retrospective study designs subgroup, there appeared to be a significant difference in the rate of recurrent VTE among patients with isolated axial vs muscular calf DVT (FE model: RR, 1.36; 95% CI, 1.04-1.77; I² = 47%), but not in prospective study designs subgroup (FE model: RR, 1.03; 95% CI, 0.71-1.49; I² = 0%). There was no clear difference in the rate of recurrent VTE among patients with isolated axial vs muscular calf DVT in symptomatic calf DVT subgroup (FE model: RR, 1.08; 95% CI, 0.79-1.47; I² = 0%) and among patients with calf DVT limited to axial vein compared with those with muscular calf DVT (FE model: RR, 1.12; 95% CI, 0.74-1.71; I² = 0%) (Fig 2, A, B, C, D).

Fig 2.

Primary outcome in patients with axial calf deep vein thrombosis (DVT) or muscular calf DVT. A, Recurrent venous thromboembolism (VTE) in randomized controlled trial (RCT) or cohorts, separately. B, Recurrent VTE in prospective or retrospective study designs, separately. C, Recurrent VTE in symptomatic calf DVT subgroup. D, Recurrent VTE among patients with isolated axial calf DVT (thrombosis confined only to tibial or fibular veins) vs muscular calf DVT. E, Recurrent VTE for anticoagulation vs conservative. CI, Confidence interval.

Among 13 included studies, two studies^13,27 compared anticoagulation therapy with no anticoagulation. Both of these studies^13,27 were cohorts. Anticoagulants were given at therapeutic doses in the two studies. Unfractionated or low molecular weight heparin, or combinations of vitamin K antagonist plus heparins with international normalized ratio (INR) of 2 to 3 were administrated in two studies,^13,27 and a direct thrombin inhibitor was prescribed in one study.¹³ In patients with axial calf DVT, there was no clear difference in the rate of recurrent VTE between patients receiving anticoagulants and controls (FE model: RR, 0.71; 95% CI, 0.30-1.70), similar in patients with muscular calf DVT (FE model: RR, 0.50; 95% CI, 0.21-1.22). The combined summary effect from these studies showed a trend towards lower recurrent VTE in patients receiving anticoagulants as compared with controls (FE model: RR, 0.60; 95% CI, 0.32-1.12; I² = 0%) (Fig 2, E).

Considering studies differed in the duration of follow-up, the risk of recurrent thromboembolism at 3 months, 6 months, 1 year, and 2 years for the two study groups were evaluated. A greater rate of recurrent VTE in the axial group compared with the muscular group was found during 2-year follow up periods (FE model: RR, 1.48; 95% CI, 1.07-2.04) (Supplementary Fig, online only).

Secondary outcomes

Resolution of calf DVT, proximal propagation of calf DVT, PE, clinically relevant non-major bleeds, and major bleeds were separately analyzed as secondary outcomes.

Resolution of calf DVT

Two studies^26,27 reported on the rates of resolution of calf DVT in patients with isolated calf DVT and showed a higher rate of resolution in the axial group compared with the muscular group (FE model: RR, 1.32; 95% CI, 1.01-1.72; I² = 31%) (Fig 3, A).

Fig 3.

Secondary outcomes in patients with axial or muscular calf deep vein thrombosis (DVT). A, Resolution; B, proximal propagation; C, pulmonary embolism (PE); D, major bleeds; and E, clinical relevant non-major bleeds. CI, Confidence interval.

Proximal propagation

Proximal propagation was separately reported in seven studies^13,21, 22, 23, 24, 25^,27 of the 13 included studies. Of 2936 patients referred to proximal propagation, 177 proximal VTE events were analyzed. Thrombus in the axial veins were more likely to propagate proximally than thrombus in the muscular veins (FE model: RR, 1.63; 95% CI, 1.10-2.41; I² = 40%). No propagation was found in one of seven studies.²⁷ (Fig 3, B).

Pulmonary embolism

Four studies,^21,22,24,27 including 2222 patients, reported rates of PE. The rate of PE was different, ranging from 0% to 15.4% with a mean rate of 4.9%. None of the studies reported a death from PE. One study²² found significant differences in the rates of PE between the axial group and the muscular group. PE was more common for patients with axial calf DVT than those with muscular calf DVT (FE model: RR, 2.79; 95% CI, 1.31-5.95; I² = 0%) (Fig 3, C).

Bleeding outcomes

Outcome of major bleeds was reported in four studies^18,20,22,26 (Table I). For the four cohort studies, the major bleeds rate ranged from 0% to 6%. There was no difference in the pooled estimates for major bleeds in patients with axial calf DVT and muscular calf DVT (FE model: RR, 1.09; 95% CI, 0.61-1.95; I² = 0%). Clinically relevant non-major bleeds were reported in only one cohort study,²² which also demonstrated no statistical difference in clinically relevant non-major bleeding events in the axial and muscular arms (FE model: RR, 1.80; 95% CI, 0.93-3.48) (Fig 3, D, E).

Methodologic quality of included studies

Overall, included studies had moderate methodologic quality. The one RCT¹⁶ was a blinded, double-blind, placebo-controlled clinical trial from 28 centers (Fig 4). The 12 remaining cohort studies^13,17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 had a mean NOS score of 6 (range, 3-8). Demonstration that the comparability and adequacy of follow-up were the most frequent missing items (Table II).

Fig 4.

Cochrane Risk of Bias tool for randomized controlled trial. Green (+), Low risk of bias; yellow (?), unclear from the study.

Table II.

Newcastle-Ottawa Quality Assessment Scale for cohort studies

Study		Selection				Comparability	Outcome
		Representativeness of the exposed Cohort	Selection of the Nonexposed Cohort	Ascertainment of exposure	Absence of outcome at baseline		Assessment of outcome	Length of follow-up	Adequacy of follow-up	Total score
Donadini	2017	★	★	★	★	—	★	★	★	7
Galanaud	2010	★	★	★	★	★	★	★	★	8
Galanaud	2014	★	★	★	★	—	★	★	—	6
Kim	2022	★	★	★	★	—	★	★	★	7
Krutman	2016	★	★	★	—	★	★	—	—	5
Kuczmik	2020	★	★	★	★	★	★	★	★	8
Prahlad	2016	★	★	★	★	★	★	★	★	8
Qiu	2021	—	★	★	—	—	★	—	—	3
Sartori	2014	★	★	★	★	—	★	★	★	7
Sartori	2016	★	★	★	★	★	★	★	★	8
Sule	2009	★	★	—	★	—	★	★	—	5
Utter	2020	★	★	★	★	—	★	★	★	7

Selection: (1) Representativeness of the exposed cohort: if truly or somewhat representative of the average patient with deep venous thrombosis; (2) Selection of the nonexposed cohort: if drawn from the same community as the exposed cohort; (3) Ascertainment of exposure: if secure record or structural interview; (4) Absence of outcome at baseline: if demonstration that outcome of interest was not present at start of study.

Comparability: Comparability of cohorts on the basis of the study design or analysis: if study controls for comorbidities; additional if study controls for any additional risk factors (such as age or thrombophilia).

Outcome: (1) Assessment of outcomes: if independent blind assessment or record linkage; (2) Length of follow-up: if follow-up long enough for outcomes to occur; (3) Adequacy of follow-up of cohorts: if all subjects completed follow-up or subjects lost to follow-up unlikely to introduce bias (small number lost to follow-up).

Assessment for publication bias

Recurrence of VTE was the only outcome with 12 studies to create a funnel plot for evaluation of publication bias. On visual inspection, it appears that a bias is present for studies observing rates of recurrent VTE among patients with isolated axial vs muscular calf DVT (Fig 5).

Fig 5.

Funnel plot of studies reporting recurrence of venous thromboembolism (VTE). RCT, Randomized controlled trial; RR, risk ratio.

Discussion

Clinicians wondered whether anticoagulation apply equally to all calf DVTs or whether a distinction should be made between isolated calf DVT involving the axial or muscular venous segments. This meta-analysis showed that patients with calf DVT limited to muscular veins might have a lower rate of recurrent VTE, resolution, proximal propagation, and PE vs those with axial calf vein involvement. Moreover, our study found a trend towards lower recurrent VTE in patients receiving anticoagulants as compared with controls and exhibited similar safety outcomes. This information is important and may influence the clinical management of calf DVT.

Compared with proximal DVT, there are fewer RCTs on calf DVT (distal DVT), and those on calf DVT are often considered as a whole, without distinguishing or comparing axial and muscular thrombosis. Unfortunately, only one RCT was performed to compare outcomes of axial calf DVT with muscular calf DVT. Whereas isolated calf DVT was more frequent than proximal DVT,² and the therapy of calf DVT has been extensively disputed in prior studies without reaching clear consensus.¹¹ This dispute arises from insufficient understanding about the natural history and true risk of calf DVT.

Previous studies claimed isolated calf DVT to be a less harmful disease, which was associated with a low incidence of proximal propagation and PE. Some vascular ultrasound laboratories do not systematically scan muscular veins (limited to imaging the axial veins). At present, more and more studies have found that isolated calf DVT, even isolated muscular calf DVT, can lead to PE. Moreover, a staggering proportion of patients with acute PE have isolated calf DVT as unique source of emboli.²⁸ Our study shows that rates of PE in patients with isolated calf DVT range from 0% to 15.4% with a mean rate of 4.9%.

The treatment of isolated calf DVT has been an area of significant discussion in the literature. Guidelines from the American College of Chest Physicians suggest anticoagulant therapy only for patients with severe symptoms or those at increased risk of thrombus extension, defined by the thrombotic burden or by clinical factors and surveillance ultrasound scan.¹¹ In our study, there was no clear difference in the incidence of recurrent VTE between patients receiving anticoagulants and controls, whether in the axial group or the muscular group. This may be related to the relatively low probability of proximal propagation or PE occurrence in patients with calf DVT, resulting in less significant benefits. An in-depth understanding of the anatomy and pathophysiology enables more appropriate prevention and treatment of calf DVT.¹⁰ The peroneal and tibial veins have aligning arteries, whose venous return is mainly owing to the pulsation of the aligning artery, whereas the perfusion of the soleal and gastrocnemius vein mainly upon muscle pumping. When a thrombus is generated in the muscular vein, it asymptomatically progresses along the venous return to the proximal side as a free-floating thrombus and can become a huge embolic source.^10,28,29 The phenomenon can be demonstrated by medicolegal autopsy.³⁰ In study by Ro et al, >90% of autopsy cases of acute PE had a soleal vein thrombus.²⁸ A soleal vein thrombus is easy to form, then quickly dissolves or organizes and does not turn to a large embolus because of the many branches of the soleal vein. The anatomical structure is an important reason for primarily preventing lethal embolus formation. Perhaps this can explain why axial calf DVT is more often associated with PE than muscular DVT in our study. However, once a sequent thrombus of soleal DVT has formed in a thigh vein, which usually manifests as a free-floating thrombus without obvious clinical symptoms, it can easily break off the vascular wall with knee joint flexion, leading to a lethal embolus.¹⁰ The reported rate of such thrombus propagation is different, from 7.9% to 25%.^31,32 A soleal vein thrombus does not usually propagate to proximal, but propagated proximal thrombi are usually associated with a higher risk of fatal PE than primary proximal DVT. Although periodic ultrasound for calf DVT management is guideline-suggested, the possibility of incident PE must be kept in mind. Considering differences of the anatomical structure and pathophysiology between muscular and axial veins, the dilemma in the current treatment of isolated calf DVT may be due to not distinguishing them.

In our meta-analysis, anticoagulation appears to trend towards reduced recurrent VTE in patients with isolate axial calf DVT or muscular calf DVT; the effects of this therapy on the risk of proximal propagation or PE are less unclear. It is noteworthy that the included studies differ for type (heparins, warfarin, etc) and regimen (dosage and duration) of anticoagulants. The risk of bleeding needs to be balanced if anticoagulant therapy is initiated. In our meta-analysis, the major bleeds rate ranged from 0% to 6%. There was no difference in the pooled estimates for clinically relevant non-major bleeds or major bleeds in patients with axial calf DVT and muscular calf DVT. However, bleeding outcomes in patients with anticoagulation compared with conservative management was not discussed. Further, only a few studies assessed bleeding outcomes. This problem also exists in other reviews^9,33,34 and reduced our ability to draw a reliable assessment for the effect of anticoagulation in patients with calf DVTs.

There are some limitations in our study. First, it should be considered that a meta-analysis has inherent weaknesses due to the combination of heterogenous data sets. However, to minimize the heterogeneity of study populations due to the use of obsolete diagnostic techniques or treatment regimens, studies that were published before 2003 were excluded. In addition, our study only included patients diagnosed with calf DVT by compression ultrasound scan to reduce heterogeneity. Second, only one double-blind RCT has been included in our meta-analysis. Also, there was significant difference in the rate of recurrent VTE among patients with isolated axial vs muscular calf DVT in the retrospective study design subgroup, but not in the prospective study design subgroup. However, we included 12 cohort studies with 4889 patients, and can perform several sensitivity analyses. Third, five studies of the 13 included studies only reported symptomatic events, and asymptomatic recurrence, propagation, and PE may be missed. However, it is balanced between the axial and muscular groups. Fourth, only two studies^26,27 compared anticoagulation therapy with no anticoagulation, and we cannot perform subgroup analysis to assess whether the length of treatment impacts the rate of recurrent VTE. Moreover, only a few studies assessed resolution of calf DVT, PE, and bleeding outcomes. In light of the study limitations, large multi-center RCTs are needed to confirm our observations.

Conclusions

Patients with calf DVT limited to muscular veins might have a lower rate of recurrent VTE, resolution, proximal propagation, and PE vs those with axial calf vein involvement and exhibited similar safety outcomes.

Author Contributions

Conception and design: CW, CS, RG, TW

Analysis and interpretation: CW, CS, RG, TW

Data collection: CW, TW

Writing the article: CW, TW

Critical revision of the article: CW, CS, RG, TW

Final approval of the article: CW, CS, RG, TW

Statistical analysis: CW, TW

Obtained funding: Not applicable

Overall responsibility: TW

Disclosures

None.

Appendix

Supplementary Fig (online only).

Primary outcome in patients with axial calf deep vein thrombosis (DVT) or muscular calf DVT at 3 months, 6 months, 1 year, and 2 years follow-up, separately. CI, Confidence interval.

Supplementary Table (online only).

Supplementary characteristics of included studies

Author	Year	Risk factors considered	Symptoms, %
Ageno	2022	Malignancy, surgery, injury, prolonged bed rest, oral contraception, pregnancy, obesity	Swelling (55.5), erythema (5.5), pain (78.5)
Donadini	2017	Unprovoked, transient risk factor, previous history of VTE	Pain (82.9), edema (61.7), redness (22.1)
Galanaud	2010	History of VTE, malignancy, varicose, estrogen therapy obesity, immobilization, recent travel, surgery, infectious, pregnancy, postpartum	Swelling, dull or localized pain, and warmth of the lower limb
Galanaud	2014	Bed rest, recent surgery, plaster immobilization	Unclear
Kim	2022	Malignancy, recent trauma, surgery	Leg edema and/or pain (42.2)
Krutman	2016	Recent surgery, malignancy, recent trauma	All included patients reported leg pain and/or edema and had no initial complain of respiratory symptoms
Kuczmik	2020	Trauma, active malignancy	Unclear
Prahlad	2016	Surgery and/or immobility	Unclear
Qiu	2021	Unclear	Unclear
Sartori	2014	Obesity, malignancy, surgery, history of VTE estrogen-containing therapy	Pain (86.7) Edema (62.9) Redness or rash (14.8) Leg warmth (12.8)
Sartori	2016	Obesity, malignancy, surgery, mobility significantly reduced, trauma, history of VTE, estrogen-containing therapy,	Acute calf pain, Acute calf swelling, calf cramps, acute calf redness, acute calf warmth
Sule	2009	Malignancy, surgery, immobility	Unclear
Utter	2020	Unclear	Unclear

VTE, Venous thromboembolism.