http://orcid.org/0000-0002-7980-281X
Figures
Abstract
Background
Previous results regarding the prevalence of Factor V Leiden (FVL) in patients with cerebral venous thrombosis (CVT) varied remarkably. Therefore, we performed a meta-analysis to evaluate the potential association between FVL and CVT.
Methods
The PubMed, Embase, Cochrane Central Register of Controlled Trials, and Web of Science were searched for relevant case-control studies. The data were pooled for analysis of the association between FVL and CVT with a random-effects model. Subgroup and sensitivity analyses were performed to evaluate the robustness of the results. Publication bias was assessed by funnel plot and the Egger’s test.
Results
A total of 39 case-control studies including 1807 cases of CVT and 7699 control cases were included. FVL was more common in patients with CVT (195/1822) than in controls(369/7795), [odds ratio(OR) = 2.70, 95% confidence interval(CI) 2.16–3.38, P < 0.00001]. Results of subgroup analyses indicated that the association between FVL and CVT varied according to geographic regions. The associations between FVL and CVT were significant in studies from Germany (OR = 2.42, 95%CI 1.70–3.45, P < 0.00001), Brazil(OR = 2.82, 95%CI 1.24–6.42, P = 0.01), France (OR = 5.44, 95%CI 1.35–21.92, P = 0.02), Iran (OR = 6.61, 95%CI 1.83–23.93, P = 0.004), and Tunisia (OR = 6.54, 95%CI 2.89–14.82, P < 0.00001), but not in those from Italy (OR = 1.49, 95%CI 0.59–3.78, P = 0.40) or the US (OR = 2.37, 95%CI 0.59–9.42, P = 0.22).
Citation: Li X, Cui L, Li Y, Zhu L, Wang C, Liu J, et al. (2018) Prevalence and geographical variation of Factor V Leiden in patients with cerebral venous thrombosis: A meta-analysis. PLoS ONE 13(8): e0203309. https://doi.org/10.1371/journal.pone.0203309
Editor: Stephan Meckel, Universitatsklinikum Freiburg, GERMANY
Received: April 28, 2018; Accepted: August 18, 2018; Published: August 29, 2018
Copyright: © 2018 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was supported by The National Natural Science Foundation of China (81271294). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Cerebral venous thrombosis (CVT) has been recognized as a rare disease that is primarily defined as thrombosis in cerebral veins, with a reported annual incidence of 3 to 4 cases per 2 million in adults and 7 cases per 1 million in neonates[1]. Clinically, CVT has been suggested to be a rare cerebrovascular disease that accounts for 0.5% of all cerebral strokes[2]. With the development of novel sensitive imaging techniques and increasing awareness of clinicians for the disease, the prevalence of CVT, as reported in recent studies, may be higher than previously estimated[3]. Although the exact cause of CVT remains to be determined, some risk factors for the disease have been proposed, such as surgery, trauma, immobilization, oral contraceptive pills, pregnancy, puerperium, and others. Moreover, recent evidence suggests that genetic factors related to thrombosis, such as Factor V Leiden (FVL), prothrombin G20210A[4], and methylenetetrahydrofolate reductase variant C677T may also be involved in the pathogenesis of CVT.
Accumulating evidence suggests that FVL is an important genetic determinant of venous thromboembolism(VTE)[5]. Similarly, some studies have indicated FVL plays a considerable role in the occurrence of CVT. However, the results of these studies varied significantly according to different geographical regions. In previously published meta-analyses, Marjot et al[6] included 767 CVT cases and 4020 controls and concluded that FVL was significantly associated with CVT. This was further confirmed in a similar study by Lauw et al[7], which included 919 CVT cases and 3168 healthy controls. However, a few case-control studies have been published since these meta-analyses, and whether geographical factors contribute to the heterogeneity regarding the association between FVL and CVT remains to be determined. Therefore, in this updated meta-analysis, we aimed to evaluate the overall association between FVL and CVT and also explored whether this association differed according to the geographical origin of the studies.
Methods
Search strategy
We performed this meta-analysis according to the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) working group (S1 File)[8] and Meta-analysis on Genetic Association Studies(S1 Table). We systematically searched electronic databases including PubMed, Web of Science, Embase, and Cochrane Center Register of Controlled Trials (CENTRAL) hosted by the Cochrane Library for relevant studies, and the last search was performed on Nov 10, 2017. No restrictions of language or the region of the study were applied. The search terms were (Factor V Leiden OR Factor V OR 1691A OR 1691) AND (cerebral venous thrombosis OR cerebral vein thrombosis OR cerebral sinus thrombosis OR intracranial venous thrombosis OR intracranial vein thrombosis OR intracranial sinus thrombosis). The detailed search strategy was available (S2 File). The search was supplemented by comprehensive manual search of the references of the related studies or reviews, conference abstracts, letters to editor, or key review articles.
Inclusion and exclusion criteria
Studies were included if they met the following eligibility criteria: (1) case-control studies; (2) enrollment of patients with cerebral venous or sinus or cortical vein thrombosis, which was objectively confirmed by accepted imaging methods (computed tomography, magnetic resonance imaging, magnetic resonance venography, magnetic resonance black-blood thrombus imaging, or angiography) or objectively documented by medical records; (3) control participants were non-genetically related subjects without a history of venous thromboembolism; (4) both cases and controls underwent genetic testing for FVL; and (5) the frequencies of FVG1691A for both CVT cases and controls were reported in original studies or could be calculated.
Studies were excluded from the current meta-analysis if they met the following criteria: (1) CVT was neither confirmed by imaging methods or documented by medical files; (2) genetic testing for FVL was not performed for the control group; or (3) data of interest were not available. If two studies with considerable overlapped (>50%) population were retrieved, the one most recently published study or the study of highest quality was included to avoid publication bias[7].
Data extraction
Two authors (Xinyuan Li and Li Cui) independently assessed the eligible studies and extracted data using standard abstraction forms. Disagreements were resolved by reaching a consensus with a third author. In addition, we attempted to contact the corresponding author to request missing data via E-mail. The following information for each study was collected: the first author’s name, year of publication, number of participants, country, study design, age, sex, diagnostic methods of CVT, source of control, and FVL genotype frequencies in case and control groups.
Quality assessment
Two authors (Xinyuan Li and Yunbo Li) independently evaluated the quality of the included studies according to the Newcastle-Ottawa Scale (NOS) for observational studies[9], and disagreements were resolved by reaching a consensus with a third author(Shaokuan Fang). The NOS consists of three main parts: selection (4 stars), comparability (2 stars), and outcome (3 stars) and ranges from 0–9 stars[10]. Thus, the quality of each study was determined on a scale from 0–9 points. Studies with 7–9 points were regarded as “high quality”, with 4–6 points as “moderate quality”, and with less than 4 points as “low quality”[11].
Statistical analysis
The primary outcome of the present study was defined as the odds ratio(OR) for the prevalence of FVL in patients with CVT as compared with that in healthy controls. We used the Review Manager 5.3 (Cochrane Collaboration, London, UK) and Stata (Version 12.0) software for statistical analyses. The significance of the pooled estimates was determined by Z statistic with 95% confidence interval (CI), and statistical significance was set at a two-tailed P<0.05. Data were pooled when there were at least three case-control studies. ORs were combined using the Mantel-Haenszel method and a random-effects model that takes into account the variability among the included studies. Study heterogeneity was evaluated by the I2 statistic: a value of<50% indicated low to moderate heterogeneity, ≥50% indicated substantial heterogeneity. In general, significant heterogeneity was defined as I2≥50% and P<0.05. The genotypes of FVL mutation (G1691A) consisted of normal homozygote (G/G), mutated homozygote (A/A), and heterozygote (G/A). We regarded homozygote (A/A) and heterozygote (G/A) carriers of FVL as one group due to the rarity of homozygote carriers (A/A). Thus, a dominant model was applied. We performed a subgroup analysis based on geographical regions of the studies to identify whether this factor contributed to the heterogeneity among the included studies. We were unable to conduct other subgroup analyses (race/ethnicity, education level, family income, and disease severity) owing to studies not providing sufficient stratified data. We used a sensitivity analysis to evaluate the robustness of the results by omitting one study at a time. The publication bias was assessed by the funnel plot and the Egger’s test [12–13].
Results
Database search
A total of 651 articles were initially retrieved from the electronic database search (290 from PubMed, 240 from Web of Science, 86 from Embase, and 26 from CENTRAL). Then 369 irrelevant articles, 124 duplicate papers, and 63 reviews were excluded based on screening of the title and abstracts. Among the remaining 47 studies, characteristics of the eight studies[14–21] excluded from analysis were presented(S2 Table). Seven were excluded due to the overlapped population of already included studies, and the other one was excluded because no FVL was detected in cases or controls. Thus, 39 studies(S3 File) were included in the current meta-analysis. Fig 1 illustrated the flowchart of database search and study identification and Table 1 summarized the characteristics of the included studies.
FVL, Factor V Leiden; CVT, cerebral venous thrombosis.
Study characteristics
Overall, 39 case-control studies were included in the current analysis, with 1822 cases of CVT and 7795 controls. The included studies were published between 1996 and 2017. Among 39 studies, 14 studies did not report specific genotypes of FVL, 4 studies reported all genotypes including homozygous G1691A polymorphism, and the remaining 21 studies only reported heterozygous polymorphism. Thus, we considered homozygote (A/A) and heterozygote (G/A) carriers of FVL as one group using a dominant model. Additionally, we failed to detect the Hardy-Weinberg equilibrium(HWE) of controls.
Quality assessment
The quality of included 39 case-control studies assessed by NOS was shown in Table 2. The scores from 4–9 were regarded as moderate and high quality.
Prevalence of FVL in CVT patients versus controls
Upon combining the data from all available studies, FVL/G1691A was found in 195/1822 CVT cases and 369/7795 controls (OR = 2.70, 95%CI 2.16–3.38, P<0.00001) (Fig 2).
CI, confidence interval; CVT, cerebral venous thrombosis.
Subsequently, we performed a subgroup analysis based on region of the studies. The studies were performed in countries including Italy (7 studies), Germany (9 studies), France (2 studies), Brazil (4 studies), Iran (3 studies), Tunisia (2 studies), the US (2 studies), the UK (1 study), Switzerland (1 study), Argentina (1 study), Israel (1 study), India (1 study), Spain (1 study), Turkey (1 study), the Netherlands (1 study), Estonia (1 study), and Mexico (1 study). We performed subgroup analysis according to countries where the at least two studies were performed (Fig 3). Briefly, in Italy, FVL was found in 24/316 CVT cases and 111/2415 controls, and the prevalence was not significantly different (OR = 1.49, 95%CI 0.59–3.78, P = 0.40, I2 = 44%). In Germany, FVL was found in 96/724 CVT cases and 139/2139 controls(OR = 2.42, 95%CI 1.70–3.45, P<0.00001, I2 = 14%). In Brazil, FVL was found in 11/154 CVT cases and 19/719 controls (OR = 2.82, 95%CI 1.24–6.42, P = 0.01, I2 = 0%). In France, FVL was found in 9/73 CVT cases and 4/157 controls (OR = 5.44, 95%CI 1.35–21.92, P = 0.02, I2 = 17%). In Iran, FVL was found in 10/121 CVT cases and 3/201 controls (OR = 6.61, 95%CI 1.83–23.93, P = 0.004, I2 = 0%). In the two studies from Tunisia, the prevalence of FVL was higher in CVT cases than in controls (OR = 6.54, 95%CI 2.89–14.82, P<0.00001, I2 = 0%). In the US, the prevalence of FVL did not significantly differ between the groups (OR = 2.37, 95%CI 0.59–9.42, P = 0.22, I2 = 0%).
CI, confidence interval; CVT, cerebral venous thrombosis.
We also performed a subgroup analysis based on CVT ascertainment (Fig 4) to find the source of heterogeneity and found that the association was not significant between FVL and documented CVT (OR = 0.91, 95%CI 0.39–2.16, P = 0.83) and the prevalence of FVL was higher in cases of CVT diagnosed by specific radiological methods than in healthy controls (OR = 2.86, 95%CI 2.27–3.60, P<0.00001).
CVT,cerebral venous thrombosis.
Publication bias
There was no possibility of publication bias as indicated by the funnel plot and results of Egger’s test (P = 0.135) (Fig 5).
Sensitivity analysis
The results of sensitivity analysis for the prevalence of FVL indicated that the results were not significantly affected by excluding any one of the studies (Fig 6), suggesting the robustness of the results.
FVL, Factor V Leiden; CVT, cerebral venous thrombosis.
Discussion
Although FVL has been established as the most common inherited risk factor for VTE[5,22], it has been confirmed that the prevalence of FVL in patients with deep vein thrombosis (DVT) and pulmonary embolism varies significantly among different regions and ethnicities[23–26]. However, whether the association between FVL and CVT is similar remains to be determined. To assess this issue, we performed a meta-analysis of 39 case-control studies including 1822 cases of CVT and 7795 controls and found that FVL was more common in patients with CVT than in health controls. However, the association between FVL and CVT varied depending on the geographic origin of the studies. To the best of our knowledge, this is the largest comprehensive meta-analysis for the overall prevalence of FVL mutation in patients with CVT. This is also the first evaluation of geographical variation of FVL in patients with CVT.
In general, we found that FVL was significantly associated with CVT, which was consistent with the results of previous reviews[6–7]. The current meta-analysis confirmed an overall significant association between FVL and CVT, but this association was highly dependent on the geography of patients. The association was significant between FVL and CVT in studies conducted in Germany, Brazil, France, Iran, and Tunisia, but not in those from Italy or the US. These results were consistent with previous reports that FVL is a contributory risk factor for the development of VTE[27–28]. However, it should be noted that the association between FVL and CVT was not statistically significant in Italy, which was inconsistent with the previous literature suggesting the prevalence of FVL in VTE was high in the Italian population[28]. Therefore, the potential association between FVL and CVT in studies from the above countries should be confirmed in future studies. Moreover, although the results of subgroup analyses were statistically insignificant in some studies, we found the results of subgroup analyses were generally similar to the overall prevalence of FVL. In addition, the results of subgroup analysis based on the CVT ascertainment showed that the prevalence of FVL was higher in studies in which CVT was diagnosed by specific radiological methods, but not in studies that included documented CVT. These analyses further confirmed the robustness of the results.
Our analysis had some strength as compared to previous reviews. In order to make our findings more robustness and credible, we applied several methods. First, in addition to electronic databases searches, we also conducted comprehensive manual-search of the reference lists of relevant reviews and tried our best to include all eligible studies. To date, our meta-analysis has the largest sample size to confirm the prevalence of FVL in CVT. In addition, this is also the first meta-analysis for geographical variation of FVL in patients with CVT and suggests that the prevalence of FVL varies from country to country. Second, all the included studies were of moderate to high quality as evaluated by NOS, which made the extracted data more reliable. Third, we performed sensitivity and subgroup analyses to identify the source of heterogeneity and ensure the stability of the results. Finally, we attempted to minimize publication bias by manually searching for abstracts and letters, and excluding studies with considerable population overlap (>50%), and no significant publication bias was detected by funnel plot or the Egger’s test.
However, there were some limitations that should be noted when interpreting the results. First, the balance of baseline characteristics of CVT cases and controls varied in the included studies. Not all studies matched participants for age and sex, which likely introduced a certain degree of heterogeneity. We attempted to overcome this limitation by performing subgroup analysis and the results were generally similar to the overall results. Second, we were unable to extract data for adults or children separately due to a broad age range of population in most of the studies. Therefore, our analysis failed to adequately control for age and included adults and children of all ancestries. However, we conducted a sensitivity analysis for the overall results, which showed good stability. Third, we did not distinguish between provoked and unprovoked thrombotic factors for CVT as we were unable to obtain these data. Finally, our subgroup analysis according to countries did not take into account different ethnicities within a country, which may make the interpretation of the results difficult.
In conclusion, our meta-analysis confirmed a significant association between FVL and CVT. However, this association was highly dependent on the geography of origin. More studies are warranted to evaluate the prevalence of FVL in unprovoked or provoked CVT. Moreover, the potential role of FVL as a predictor of recurrent CVT also deserves further evaluation.
Supporting information
S3 File. References of 39 studies included in the meta-analysis.
https://doi.org/10.1371/journal.pone.0203309.s003
(DOC)
S1 Table. Meta-analysis on genetic association studies form.
https://doi.org/10.1371/journal.pone.0203309.s004
(DOC)
S2 Table. Characteristics of 8 studies excluded from the meta-analysis.
https://doi.org/10.1371/journal.pone.0203309.s005
(DOC)
Acknowledgments
We would like to thank Changgui Kou of Department of Epidemiology and Biostatistics, School of Public Health of Jilin University for his scientific suggestions for the betterment of the manuscript. We would like to thank Xinmin Li of Department of Psychiatry, University of Alberta, Canada for supporting our work.
References
- 1. Martinelli I, Bucciarelli P, Passamonti SM, Battaglioli T, Previtali E, Mannucci PM. Long-term evaluation of the risk of recurrence after cerebral sinus-venous thrombosis. Circulation. 2010;121(25):2740–2746. https://doi.org/10.1161/CIRCULATIONAHA.109.927046 pmid:20547928
- 2. Bousser MG, Ferro JM. Cerebral venous thrombosis: an update. Lancet Neurol. 2007;6(2):162–170. https://doi.org/10.1016/S1474-4422(07)70029-7 pmid:17239803
- 3. Cantu C, Alonso E, Jara A, Martínez L, Ríos C, Fernández Mde L, et al. Hyperhomocysteinemia, low folate and vitamin B12 concentrations, and methylene tetrahydrofolate reductase mutation in cerebral venous thrombosis. Stroke. 2004;35(8):1790–1794. https://doi.org/10.1161/01.STR.0000132570.24618.78 pmid:15192249
- 4. Gonzalez JV, Barboza AG, Vazquez FJ, Gándara E. Prevalence and Geographical Variation of Prothrombin G20210A Mutation in Patients withCerebral Vein Thrombosis:A Systematic Review and Meta-Analysis. PLoS One. 2016;11(3):e0151607. https://doi.org/10.1371/journal.pone.0151607 pmid:27031503
- 5. Crous-Bou M, Harrington LB, Kabrhel C. Environmental and Genetic Risk Factors Associated with Venous Thromboembolism. Semin Thromb Hemost. 2016;42(8):808–820. https://doi.org/10.1055/s-0036-1592333 pmid:27764878
- 6. Marjot T, Yadav S, Hasan N, Bentley P, Sharma P. Genes associated with adult cerebral venous thrombosis. Stroke. 2011;42(4):913–918. https://doi.org/10.1161/STROKEAHA.110.602672 pmid:21350198
- 7. Lauw MN, Barco S, Coutinho JM, Middeldorp S. Cerebral venous thrombosis and thrombophilia: a systematic review and meta-analysis. Semin Thromb Hemost. 2013;39(8):913–927. https://doi.org/10.1055/s-0033-1357504 pmid:24129682
- 8. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. https://doi.org/10.1136/bmj.b2535 pmid:19622551
- 9.
Wells GA, Shea B, O'Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analysis. Ottawa, Canada: Department of Epidemiology and Community Medicine, University of Ottawa; 2010. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.
- 10. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603–605. https://doi.org/10.1007/S10654-010-9491-Z pmid:20652370
- 11. Zhu B, Wu X, Wu B, Pei D, Zhang L, Wei L.The relationship between diabetes and colorectal cancer prognosis: A meta-analysis based on the cohort studies. PLoS One. 2017;12(4):e0176068. https://doi.org/10.1371/journal.pone.0176068 pmid:28423026
- 12. Egger M, Davey-smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–634. pmid:9310563
- 13. Mistry EA, Mistry AM, Nakawah MO, Chitale RV, James RF, Volpi JJ, et al. Mechanical Thrombectomy Outcomes With and Without Intravenous Thrombolysis in StrokePatients: A Meta-Analysis. Stroke. 2017;48(9):2450–2456. https://doi.org/10.1161/STROKEAHA.117.017320 pmid:28747462
- 14. Le Cam-Duchez V, Bagan-Triquenot A, Ménard JF, Mihout B, Borg JY. Association of the protein C promoter CG haplotype and the factor II G20210A mutation is a riskfactor for cerebral venous thrombosis. Blood Coagul Fibrinolysis. 2005;16(7):495–500. pmid:16175009
- 15. Lichy C, Kloss M, Reismann P, Genius J, Grau A, Reuner K. No evidence for plasminogen activator inhibitor 1 4G/4G genotype as risk factor for cerebralvenous thrombosis. J Neurol. 2007;254(8):1124–1125. https://doi.org/10.1007/s00415-006-0417-3 pmid:17593467
- 16. Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, Mannucci PM. High risk of cerebral-vein thrombosis in carriers of a prothrombin-gene mutation and in users of oral contraceptives. N Engl J Med. 1998;338(25):1793–1797. https://doi.org/10.1056/NEJM199806183382502 pmid:9632445
- 17. Martinelli I, Bucciarelli P, De Stefano V, Passamonti SM, Menegatti M, Tormene D, et al. Effect of prothrombin 19911 A>G polymorphism on the risk of cerebral sinus-venous thrombosis. Eur J Neurol. 2010;17(12):1482–1485. https://doi.org/10.1111/j.1468-1331.2010.03068.x pmid:20482605
- 18. Dindagur N, Kruthika-Vinod TP, Christopher R. Factor V gene A4070G mutation and the risk of cerebral veno-sinus thrombosis occurring during puerperium. Thromb Res. 2007;119(4):497–500. https://doi.org/10.1016/j.thromres.2006.04.008 pmid:16781765
- 19. Tufano A, Coppola A, Varricchione N, De Simone C, Cirillo F, Palmieri NM, et al. Predisposing factors in patients with early-onset cerebral vein thrombosis. Thromb Res. 2005;115(5):439–440. https://doi.org/10.1016/j.thromres.2004.11.002 pmid:15733979
- 20. Martinelli I, Landi G, Merati G, Cella R, Tosetto A, Mannucci PM. Factor V gene mutation is a risk factor for cerebral venous thrombosis. Thromb Haemost. 1996;75(3):393–394. pmid:8701395
- 21. Meng Q, Pu C. Cerebral venous thrombosis and factor V Leiden mutation. Zhonghua Yi Xue Za Zhi. 2002;82(1):47–49. pmid:11953127
- 22. Cinier G, Öz A, Tekkesin AI, Hayıroğlu Mİ, Keskin M, Avsar Ş. A young male patient with multiple thromboembolism associated with factor V Leiden mutation. Int Heart J. 2016; 57(5):654–656. https://doi.org/10.1536/ihj.16-004 pmid:27581673
- 23. Pomero F, Ageno W, Serraino C, Borretta V, Gianni M, Fenoglio L, et al. The role of inherited thrombophilia in patients with isolated pulmonary embolism: a systematic review and a meta-analysis of the literature. Thromb Res. 2014;134(1):84–89. https://doi.org/10.1016/j.thromres.2014.04.012 pmid:24837252
- 24. Ekim M, Ekim H, Yılmaz YK. The prevalence of Factor V Leiden, prothrombin G20210A, MTHFR C677T and MTHFR A1298C mutations in healthy Turkish population. Hippokratia. 2015;19(4): 309–313. pmid:27688694
- 25. White RH, Keenan CR. Effects of race and ethnicity on the incidence of venous thromboembolism. Thromb Res. 2009;123 Suppl 4:S11–7. https://doi.org/10.1016/S0049-3848(09)70136-7 pmid:19303496
- 26. Zakai NA, McClure LA. Racial differences in venous thromboembolism. J Thromb Haemost. 2011;9(10):1877–1882. https://doi.org/10.1111/j.1538-7836.2011.04443.x pmid:21797965
- 27. de Paula Sabino A, Guimarães DA, Ribeiro DD, Paiva SG, Sant'Ana Dusse LM, das Graças Carvalho M, et al. Increased Factor V Leiden frequency is associated with venous thrombotic events among young Brazilian patients. J Thromb Thrombolysis. 2007;24(3):261–266. https://doi.org/10.1007/s11239-007-0024-x pmid:17401546
- 28. Hotoleanu C. Genetic Risk Factors in Venous Thromboembolism. Adv Exp Med Biol. 2017;906:253–272. https://doi.org/10.1007/5584_2016_120 pmid:27638626