Thrombophilia testing: does it truly affect risk?

A tension exists for providers when determining whether thrombophilia testing should be completed after venous thromboembolism (VTE) depending on patients’ desire to understand underling risk factors and if the results would influence anticoagulation duration. In this issue of Blood Advances, Vrotniakaite-Bajerciene et al¹ completed a systematic review and meta-analysis evaluating the association between thrombophilia and recurrent VTE using 51 studies, including 7 secondary analyses of randomized controlled trials, 42 observational studies, and 2 conference proceedings. The incidence of recurrent VTE was identified after anticoagulation was discontinued, which is essential for clinical decision-making. The authors used the Grading of Recommendations Assessment, Development, and Evaluation approach for certainty of evidence, which provided a translation of the high statistical variability found in the study estimates. Estimates for any thrombophilia, heterozygous prothrombin 20210A gene mutation (PTM), protein C deficiency, and protein S deficiency had low or very low certainty of evidence, whereas heterozygous and homozygous factor V Leiden (FVL), compound heterozygous FVL/PTM, antithrombin deficiency, and APS had moderate or high certainty of evidence.

The American Society of Hematology (ASH) 2023 Thrombophilia Guidelines provide recommendations on which clinical populations to consider for thrombophilia testing.² As a part of this guideline process, a systematic review was performed through June 2018 to determine the prevalence of each thrombophilia and the relative risk (RR) for recurrent VTE among those with and without thrombophilia. Novel methodologies were used based on the rationale that testing would affect anticoagulation duration, which in turn would change the incidence of recurrent VTE. The absolute risk of events in people with and without thrombophilia was calculated using the prevalence of thrombophilia and the risk association between thrombophilia and VTE. People who did not undergo thrombophilia testing comprised the comparator group, and the absolute number of recurrent VTE events was calculated based on the duration of anticoagulation recommended by the ASH VTE Treatment Guidelines.³ The intervention group included those undergoing thrombophilia testing, for whom the duration of anticoagulation would change based on the presence or absence of thrombophilia. The proportion of people who tested positive for thrombophilia was simulated based on the prevalence of each thrombophilia. The pooled RR of recurrent VTE was then used to estimate the expected event rate. Finally, the effect of testing was determined by the increase or reduction in recurrent VTE outcomes resulting from testing and subsequent differences in duration of anticoagulation.

Accurate estimates of VTE recurrence risk in patients with thrombophilia were essential to the guideline process, and the current meta-analysis provides an important update (from inception through August 2025) given the limited studies evaluating this question. Unfortunately, as noted by the authors, most included studies were at moderate to high risk for bias. The table shows the risk estimates from the study by Vrotniakaite-Bajerciene et al compared to the estimates derived and used in the ASH Thrombophilia Guidelines.² The pooled risk estimates for VTE recurrence for any thrombophilia showed similar results (incidence rate ratio [IRR], 1.53 vs RR, 1.65) despite different methodologies and an updated literature search. Results were also consistent across “weaker” thrombophilias (heterozygous FVL and heterozygous PTM) and “stronger” thrombophilias (protein C, protein S, and antithrombin deficiency and APS), all of which had risk estimates that were within the confidence intervals of the other analysis. Protein S deficiency was the only evaluated thrombophilia that did not appear to be a significant risk factor for recurrent VTE, with an IRR of 1.0 in the current meta-analysis and 1.3 in the ASH Thrombophilia Guidelines, both of which had confidence intervals that crossed 1.

Comparison of incidence rate ratio and RR estimates for recurrent VTE in patients with inherited thrombophilia or APS

Thrombophilia	Vrotniakaite-Bajerciene et al		ASH 2023 Thrombophilia Guideline2		Agreement
	IRR	IRR 95% CI	RR	RR 95% CI
Any thrombophilia	1.53	1.34-1.74	1.65	1.28-2.47	Yes
Heterozygous FVL	1.37	1.19-1.57	1.36	1.19-1.57	Yes
Homozygous FVL	1.94	1.18-3.20	2.1	1.09-4.06	Yes
Heterozygous PTM	1.6	1.23-2.08	1.34	1.05-1.71	Yes
Compound heterozygous FVL/PTM	2.26	1.35-3.79	—	—	n/a
Antithrombin deficiency	1.75	1.43-2.14	2.07	1.50-2.87	Yes
Protein C deficiency	1.58	1.05-2.38	2.13	1.26-3.59	Yes
Protein S deficiency	1.03	0.72-1.48	1.3	0.87-1.94	Yes
APS	2.17	1.54-3.04	1.92	0.99-3.72	Yes

CI, confidence interval; n/a, not applicable.

The authors of this meta-analysis added further important information by evaluating the impact of thrombophilia in patients with unprovoked VTE, a population in which real-world testing is often performed.^4,5 Risk estimates were numerically higher for some thrombophilias (heterozygous FVL, antithrombin deficiency, and APS) and similar or slightly lower for the remainder, all of which were within the confidence interval of the overall analysis. When considering the clinical significance of these estimates, it is important to put the estimates into perspective, especially when comparing them to the magnitude of estimates of first VTE among various thrombophilias, for which the control group is comprised of those without thrombophilia who have not had thrombosis. In contrast, when evaluating the risk of recurrent VTE in patients who have had thrombosis, the control arm for these evaluations has a much higher risk, which comparatively serves to numerically decrease the RR estimate. The numerically smaller risk estimate in this setting may appear less meaningful because of the smaller magnitude but may actually contribute to a significant number of recurrent VTE cases. It is for this reason that the ASH Thrombophilia Guidelines evaluated absolute risk estimates in the evidence-to-decision frameworks that account for both the underlying prevalence of thrombophilia and the potential absolute risk of recurrent VTE.

Although much debate remains regarding the role of thrombophilia testing, the risk estimates from this updated systematic review largely corroborate previous findings and simultaneously demonstrate the ongoing lack of high-quality data. They also highlight and confirm the clinically unintuitive finding that “higher risk” thrombophilias do not appear to have remarkably higher VTE recurrence risks than “lower risk” thrombophilias. Clinically, many nuances of thrombophilia testing and its results remain unaddressed, and the role of a thrombosis expert is a vital component of comprehensive risk assessment and the shared decision-making process with patients. Optimally, future risk stratification tools would account for inherited thrombophilias among other known risk factors, recognizing the often multifactorial risk of VTE recurrence.

Conflict-of-interest disclosure: The authors declare no competing financial interests.