Buy Article(opens in new window) Permissions and Reprints(opens in new window)

Superficial vein thrombosis (SuVT) is a thrombus within the superficial veins of the upper or lower extremities, or occasionally non-extremity veins. Although SuVT is rarely accompanied by immediate, large-burden embolization with hemodynamic collapse, it commonly produces local discomfort, pain, swelling, and tenderness, and it may propagate proximally to involve the deep venous system, with associated risk of deep-vein thrombosis (DVT) and pulmonary embolism (PE).[1] [2] SuVT also serves as a clinical indicator of long-term venous thromboembolic risk, including recurrent SuVT, but also subsequent DVT and PE.[1] [3] Population-level estimates place the incidence at roughly 64 to 131 per 100,000 person-years.[1] These figures likely underestimate the true burden as milder cases are often underdiagnosed and underreported.

Management of SuVT is primarily symptomatic for many patients and includes warm compresses, compression wearables, and topical or systemic analgesics and anti-inflammatory agents; catheter-associated SuVT warrants removal of the offending device when feasible. Anticoagulation is reserved for higher-risk presentations, most commonly thrombus length > 5 cm or persistent symptoms despite conservative therapy. Therapeutic-intensity anticoagulation is generally considered only when the thrombus lies within approximately 3 cm of a deep venous junction. In other cases in which anticoagulation is considered, the evidence supports fondaparinux 2.5 mg once daily for 45 days as the standard regimen. Enoxaparin is an accepted alternative, although randomized evidence for enoxaparin is less robust.[1]

Direct oral anticoagulants (DOACs) are easier to use for patients and clinicians alike and have gained traction for the management of SuVT. However, the evidence base for DOACs for the management of SuVT has been relatively slim. One randomized trial compared the effectiveness of rivaroxaban 10 mg once daily versus fondaparinux, although the noninferiority margin was wide, and did not exclude a small difference (inferiority) for rivaroxaban compared with fondaparinux.[4] Another randomized trial also tested rivaroxaban 10 mg once daily compared with placebo, but had softer endpoints and was halted prematurely.[5]

It is in this context that Boccatonda et al. report the results of a systematic review and meta-analysis for the effectiveness and safety of DOACs for the treatment of SuVT in the current issue of Thrombosis and Haemostasis.[6] They pooled the results of six studies, including the above-mentioned two RCTs.[4] [5] Based on pooling that was performed, they reported reassurance that DOACs (namely rivaroxaban) were more effective than no treatment and comparable to fondaparinux, with an excellent safety profile.[6]

Further investigations into treatment regimens for SuVT are certainly timely and clinically needed. Prior meta-analyses had similarly looked into treatment options for SuVT.[7] However, in assessing such studies, many issues related to the treatment options for SuVT deserve further scrutiny. First, evidence base for the treatment or prevention of SuVT with agents such as apixaban, dabigatran, or edoxaban has remained very limited, including in this pooled analysis. Second, of the six included studies, only two were RCTs, and one of those had softer outcomes.[5] Third, the majority of included studies were observational studies, two of which lacked an apparent control group, making comparative estimates challenging to interpret. Fourth, it should be clarified that DOACs can be considered for various purposes in patients with SuVT. If patients already have an existing indication for full-intensity anticoagulation, it is plausible that SuVT occurs during a period of interruption (e.g., hospitalization) either in the lower extremities or upper-extremities associated with catheters. In such cases, resumption of therapeutic anticoagulation at discharge would be reasonable. Another scenario is secondary prevention of venous thromboembolic events, including but not limited to SuVT. In that setting, the paradigm for considering the use and type of anticoagulants depends on the nature of initial events and whether there are enduring risk factors, with emerging additional data.[8]

More importantly and on a broader note, ascertaining internal, external, and construct validity is incredibly important for generating reliable estimates in research studies.[9] To that end, the existing risk of bias tools are not well positioned to adequately alleviate the concerns for construct validity or internal validity (confounding) in most observational comparative effectiveness studies.[10] However, pooled analyses of comparative effectiveness data inclusive of observational studies do not frequently address these concerns ([Fig. 1]). In this sense, the readers may get a false sense of security for large numbers while the underlying methodological challenges are masked as a result of pooling and apparent reassurance for risk of bias assessment.

Fig. 1 Considerations for meta-analysis across various clinical study designs. * Unless the methodological quality, internal validity and construct validity for case and outcome ascertainment have been rigorously vetted for each individual observational comparative effectiveness study. In our experience, that level of rigor is often missing in critical appraisal of observational comparative effectiveness studies. Created in BioRender. Pfeferman, M. (2026) https://BioRender.com/9ky0jc7.

As for reporting, although observational studies can be designed to complement randomized data or provide insights where clinical trials are lacking, they are most often pragmatic and unable to tease out efficacy, but rather positioned to opine on effectiveness. We urge caution in the interchangeable use of efficacy (often preserved for high-quality double-blind trials) versus effectiveness.

In conclusion, we find systematic reviews and resultant meta-analyses important tools to broaden our understanding in clinical research. Meta-analyses of observational studies would be, in our opinion, best positioned to inform epidemiology, rather than comparative effectiveness questions, unless the methodological rigor of each and every included study has been deeply verified.

Received: 02 March 2026

Accepted: 02 March 2026

Accepted Manuscript online:
06 March 2026

Article published online:
16 March 2026

© 2026. Thieme. All rights reserved.

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany