Growing evidence indicates that formation of a fibrin clot composed of compact fiber networks, resistant to fibrinolysis, is associated with thrombotic events. Reduced fibrin clot permeability and impaired fibrinolysis have been reported in patients with a history of myocardial infarction [1], ischemic stroke [2] and venous thromboembolism (VTE) [3].

It is estimated that 16 million people worldwide suffer from retinal vein occlusion (RVO) [4], involving branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO). The pathogenesis of RVO is not fully understood [5]. Proven risk factors for RVO are age, arterial hypertension and glaucoma. Data on links between prothrombotic state and RVO are inconsistent [5]. A role of known thrombophilic factors in RVO appears minor [5]. To our knowledge, there has been no published report on fibrin clot characteristics in RVO. Given the association between atherosclerosis risk factors and prothrombotic fibrin clot phenotype, we hypothesize that abnormal properties of a fibrin clot can be observed in patients who experienced RVO, and therefore contribute to the occurrence of this thrombotic complication.

Of the 120 screened white adult patients in whom first-ever RVO was diagnosed during the preceding 6 to 36 months, 59 individuals were included in this case–control study from March 2009 to May 2010 in Krakow. The diagnosis of RVO was based on typical findings in the fundus of the eye documented by digital photography, fluorescein angiography and optical coherence tomography. Inclusion criteria were age between 18 and 75 years, and absence of chronic and acute disorders known to affect clot properties. Exclusion criteria were: age above 75 years (n = 26); C-reactive protein (CRP) > 10 mg L−1 (n = 2); cancer (n = 2); hepatic or renal dysfunction (n = 0); anticoagulant therapy (n = 1); acute myocardial infarction or stroke within the preceding 6 months (n = 4); previous VTE (n = 4); and diabetes (n = 22). One hundred and eighteen subjects matched for age, sex, body mass index (BMI), ischemic cardiovascular events and smoking from the general population of the south of Poland, described previously [6], served as ‘disease’ controls. The same exclusion criteria were used for controls. Individuals with abnormal findings during ophthalmologic evaluation were excluded from the controls. The University Bioethical Committee approved the study and patients provided written, informed consent.

Fasting blood samples were drawn from the antecubital vein with the minimal stasis in the morning. Lipid profile, glucose, platelet count and creatinine were measured using routine laboratory techniques. Total homocysteine (tHcy) was assessed using high pressure liquid chromatography as described [7]. Immunoenzymatic assays were used to determine tissue plasminogen activator (tPA) and plasminogen activator inhibitor-1 (PAI-1) antigens. CRP and fibrinogen were measured by nephelometry (Dade Behring, Marburg, Germany).Thrombophilia screening was performed as described [6].

Clot variables were determined in citrated plasma at least in duplicate by technicians blinded to the origin of the samples (intra-assay and inter-assay coefficients of variation, 5%–8%), as described in detail [3].

  • 1
     Fibrin clot permeation using a pressure-driven system, with calculation of a permeation coefficient (Ks), which indicates the pore size. Lower Ks values indicate reduced permeability.
  • 2
     The lag phase of the turbidity curve, which reflects the time required for initial protofibril formation and maximum absorbance at 405 nm at the plateau phase (ΔAbsmax), indicating the number of protofibrils per fiber.
  • 3
     Clot lysis time, defined as the time required for a 50% decrease in clot turbidity at 405 nm (t50%), induced by 1 μg mL−1 recombinant tissue plasminogen activator (rtPA; Boerhinger Ingelheim, Ingelheim, Germany).
  • 4
     Maximum rate of increase in D-dimer levels (American Diagnostica, Greenwich, CT, USA) and maximum D-dimer concentrations measured every 30 min in a buffer containing 0.2 μmol L−1 rtPA (Boerhinger Ingelheim) percolating through fibrin clots prepared as in the permeation evaluation.

Data are given as mean ± SD or median (interquartile range), or as otherwise stated. Inter-group comparisons were analyzed by Fischer exact, chi-squared, t-Student and Mann–Whitney tests as appropriate. Correlations were assessed by the tau-B Kendall test. To determine whether any fibrin clot variables are independent correlates for RVO, a forward multivariate logistic regression for matched pairs data using univariate variables with a P-value of < 0.05 was performed. A value of P < 0.05 was considered significant.

The study was powered to have an 80% chance of detecting a 10% difference in Ks using a P-value of 0.05 and Ks obtained for thrombosis patients [3]. In order to demonstrate such a difference or greater, 30 subjects were required in each group.

A mean time from the diagnosis to enrollment was 14.4 ± 9.9 months. There were 33 BRVO patients (56%) (intrapapillary BRVO, n = 11; extrapapillary BRVO, n = 22) and 26 CRVO (44%) patients (non-ischemic CRVO, n = 18; ischemic CRVO, n = 8). The RVO patients and controls did not differ with regard to demographics, risk factors, medication and basic laboratory tests (Table 1).

Table 1.   Characteristics of patients with retinal vein occlusion (RVO) and controls
 RVO patients (n = 59)Controls (n = 118)P-value
  1. Data are shown as median (interquartile range) or mean ± SD. CRP, C-reactive protein; PAI-1, plasminogen activator inhibitor-1; RVO, retinal vein occlusion; tPA, tissue plasminogen activator.

Men, n (%)33 (55.9)62 (52.5)0.67
Age, years55 (51–63)57 (52–67)0.29
BMI (kg m−2)28.3 (20.6–36.3)27.9 (21.9–37.4)0.36
Hypertension, n (%)38 (64.4)70 (59.3)0.51
Previous stroke, n (%)3 (5.1)8 (6.8)0.91
Previous myocardial infarction, n (%)4 (6.8)6 (5.1)0.91
Current smokers, n (%)12 (20.3)24 (20.3)1.0
 Aspirin, n (%)21 (35.6)39 (33.1)0.74
 Statins, n (%)15 (25.4)29 (24.6)0.9
 Angiotensin-converting enzyme inhibitors, n (%)25 (42.4)59 (50)0.39
Laboratory parameters
 Platelets (×103 μL−1)223.76 ± 42.05225.7 ± 34.20.46
 Glucose (mmol L−1)5.1 (4.9–5.6)5.0 (4.6–5.6)0.29
 Creatinine (μmol L−1)68.1 (60.0–81.1)69.5 (61.8–78.6)0.53
 Fibrinogen (g L−1)3.5 (3.2–3.9)3.3 (2.8–3.8)0.14
 CRP (mg L−1)1.87 (1.22–2.33)1.93 (1.41–2.78)0.24
 tPA (ng mL−1)17.0 (11.0–20.3)14.3 (10.6–19.1)0.11
 PAI-1 (ng mL−1)19.5 (12.3–27.9)20.5 (14.1–27.6)0.45
Fibrin clot variables
 Ks (10−9 cm2)6.9 (5.9–8.0)8.7 (6.9–9.8)0.0003
 ΔAbmax (405 nm)0.86 (0.80–0.94)0.78 (0.72–0.85)0.02
 Lag phase (s)41.42 ± 4.5944.8 ± 4.510.007
 t50% (min)9.3 (8.5–10.2)7.9 (7.2–8.9)< 0.0001
 D–Dmax (mg L−1)4.04 (3.48–4.55)3.62 (3.34–3.98)< 0.0001
 D–Drate (mg L−1 min−1)0.071 (0.063–0.074)0.072 (0.066–0.079)0.34

RVO patients were characterized by altered plasma fibrin clot properties compared with controls. RVO patients had lower Ks, longer t50%, shorter lag phase indicating faster fibrin formation, and greater ΔAbmax, indicating thicker fibrin fibers. D–Dmax indicating thrombotic mass available for fibrinolytic agents was higher in the RVO group (Table 1). After exclusion of seven RVO patients with a history of stroke or myocardial infarction and 14 with the same history from the control group, the intergroup differences remained significant (all P < 0.01). Thrombophilia screening showed similar frequencies in the RVO and control groups (antiphospholipid syndrome [3 vs. 7], factor V Leiden [6 vs. 9], prothrombin mutation F2 20210 G>A [1 vs. 2], deficiencies of antithrombin [0 vs. 0], protein C [0 vs. 0] and protein S [1 vs. 0], respectively; all P > 0.05). Thrombophilia was not associated with altered clot phenotype (data not shown).

Interestingly, in RVO patients CRP was positively associated with ΔAbmax, t50% and D–Dmax (all are from 0.36 to 0.55, P < 0.05). None of the fibrin clot parameters showed associations with time from the RVO occurrence, PAI-1, tPA, lipid variables, tHcy and glucose in RVO patients (data not shown). Fibrin clot properties did not differ between patients with diagnosed BRVO or CRVO, with or without neovascularization and macular edema, or non-ischemic or ischemic CRVO, indicating the lack of correlations between the severity of RVO and fibrin clot properties (data not shown). All fibrin clot variables except D–Drate were significantly related to RVO after univariate analysis (all P < 0.05). After multivariate analysis Ks and ΔAbmax are the only independent correlates for RVO (odds ratio [OR] 1.22, 95% confidence interval [CI] 1.04–1.46, and OR 1.18, 95% CI 1.03–1.22, both P < 0.01).

To our knowledge, this study is the first to show that fibrin clot characteristics are unfavorably altered in patients with a history of RVO. We demonstrated that RVO is associated with faster formation of significantly less permeable and poorly lysable dense plasma fibrin clots. The strongest independent links were shown for clot permeability and maximum absorbancy on turbidimetry. One might speculate that abnormal clot properties, similar to those observed in VTE and atherosclerotic vascular disease [1–3], predispose to RVO. The differences in fibrin variables between groups, however, cannot be attributed to cardiovascular risk factors or thrombophilic factors because they were similarly distributed in the RVO and control groups. Interestingly, we demonstrated the association between elevated CRP and some clot parameters, particularly clot lysis time, as previously demonstrated using the same methodology in subjects at high cardiovascular risk [8,9]. It has been demonstrated that CRP binds to fibrinogen and fibrin [10], although the molecular mechanisms underlying such fibrin(ogen) modification are unknown. The precise mechanisms underlying the intergroup differences in fibrin properties between the RVO and well-matched control groups remain to be established.

The study has limitations. The size of the population studied is limited and our analysis was performed at a single time point. We did not determine genetic polymorphisms reportedly affecting clot structure [3]. Finally, it remains unclear whether such a fibrin clot phenotype is a marker of RVO, or its risk factor. A longitudinal study performed on subjects with known plasma fibrin clot characteristics, who did not experience RVO at the time of enrollment, is needed to assess the future risk of developing such a disease.

In conclusion, we demonstrated that unfavorably altered fibrin clot properties, including reduced susceptibility to lysis, occur in RVO patients and might contribute to the hyperviscosity reported in this disease [11]. Our observations suggest a previously unknown role of alterations in fibrin formation and degradation in the pathogenesis of RVO. It is tempting to hypothesize that RVO via altered fibrin clot features might also be a marker of an increased risk of arterial and venous thrombosis. This issue merits further investigations.


  1. Top of page
  2. Acknowledgements
  3. Disclosure of Conflict of Interests
  4. References

This work was supported by grants from Jagiellonian University Medical College no. K/ZBW/000573 (to I. Karska-Basta) and no. K/ZDS/000565 (to A. Undas).

Disclosure of Conflict of Interests

  1. Top of page
  2. Acknowledgements
  3. Disclosure of Conflict of Interests
  4. References

The authors state that they have no conflict of interest.


  1. Top of page
  2. Acknowledgements
  3. Disclosure of Conflict of Interests
  4. References