Subsegmental pulmonary embolism: yet another case for being a medical conservative

Abstract

Abstract Background The advent of multi-detector computed tomographic pulmonary angiography has allowed better assessment of the peripheral pulmonary arteries, thereby increasing the incidence of pulmonary embolism (PE). Even though most patients with PE are treated with anticoagulation, its value in the subsegmental setting (SSPE) has not yet been confirmed. Purpose To perform a meta-analysis aimed at ascertaining the extent to which anticoagulation results in a net positive effect in patients with SSPE. Methods We systematically searched MEDLINE, Embase, Web of Science, Cochrane Library and Google Scholar, from inception to March 2021, for controlled studies addressing the effect of anticoagulation on SSPE patients. Specifically, venous thromboembolism (VTE) recurrence served as the primary efficacy endpoint, whereas clinically significant bleeding represented the primary safety outcome. Furthermore, major bleeding, PE-related and all-cause mortality were also studied, as secondary endpoints. All anticoagulation strategies, namely oral or parenteral, met inclusion criteria. Study-specific odds ratios (ORs) were pooled, under a random-effects model. Results 1 cross-sectional, 8 retrospective and 4 prospective non-randomized studies, encompassing 82, 641 and 157 patients, respectively, were regarded as eligible for quantitative evaluation. 667 patients (75.8%) were allocated to the anticoagulation arm. The absolute number of events for each outcome may be reported as follows: primary efficacy endpoint, 5; primary safety endpoint, 60; major bleeding, 38; PE-related mortality, 0; all-cause mortality, 25. 7 studies reported their respective outcomes under a prespecified 3-month follow-up period, while only 1 featured cancer patients as its entire sample. As for the primary efficacy endpoint, and despite the surprising adjudication of all its 5 events to the anticoagulated patients, their relative overrepresentation (371 vs. 143 patients) stemmed a non-significant tendency towards a decrease in VTE recurrence in this arm (OR 0.59, 95% CI 0.09–3.81, P 0.58, i2 0%). On the other hand, anticoagulation was associated with a significant increase in clinically significant hemorrhages (OR 2.89, 95% CI 1.07–7.80, P 0.04, i2 0%) and a non-significant propensity towards an increment in major bleeding (OR 2.44, 95% CI 0.79–7.59, P 0.12, i2 0%). Lastly, and even though no events of PE-related mortality were reported, anticoagulation was linked with a meaningful reduction in all-cause mortality (OR 0.31, 95% CI 0.11–0.82, P 0.02, i2 0%). Conclusion Currently available evidence underpins marginal efficacy and safety concerns regarding the use of anticoagulation in SSPE patients, who are expected to experience very low to none PE-related mortality. The association of anticoagulation with lower all-cause death may be attributable to selection bias. Randomized controlled trials are, however, still needed to fully validate this hypothesis. Funding Acknowledgement Type of funding sources: None.