Purpose
To evaluate the association between platelet function and disc hemorrhage in patients with normal-tension glaucoma.
Design
Prospective, cross-sectional study.
Methods
Study involved a total of 315 subjects, including patients with normal-tension glaucoma and disc hemorrhage (n = 120), patients with normal-tension glaucoma without disc hemorrhage (n = 75), and healthy individuals (control group, n = 120). A detailed eye examination including visual field testing, color disc photography, optical coherence tomography scanning, and measurement of collagen/epinephrine closure time using a platelet function analyzer were performed for all subjects.
Results
The collagen/epinephrine closure time (s) as measured by the platelet function analyzer was approximately 14%–24% longer in the normal-tension glaucoma and disc hemorrhage group compared with the other groups (141.92 ± 53.44 [with normal-tension glaucoma and disc hemorrhage] vs 124.60 ± 46.72 [with normal-tension glaucoma without disc hemorrhage] vs 114.84 ± 34.84 [healthy individuals], 1-way analysis of variance test, P < .001). The activated partial thromboplastin time (s) value of the normal-tension glaucoma with disc hemorrhage group was also higher than the control group. Stepwise multiple logistic regression analysis revealed that only a longer collagen/epinephrine closure time (OR adjusted for age, sex, prothrombin time, activated partial thromboplastin time, diabetes mellitus, hypertension, hypotension, heart disease, hypothyroidism, migraine, stroke, hypercholesterolemia: 2.94; 95% CI: 1.40–6.17) was independently associated with disc hemorrhage. A similar trend was observed when platelet function was compared among the 3 groups with respect to age.
Conclusions
Our results suggest that platelet function is significantly associated with disc hemorrhage in patients with normal-tension glaucoma. Delayed absorption resulted from prolonged bleeding due to delayed platelet aggregation may have an effect on the detectability of disc hemorrhage in patients with normal-tension glaucoma.
Normal-tension glaucoma (NTG) is a form of open-angle glaucoma in which optic nerve damage and visual field (VF) defects are present despite normal intraocular pressure (IOP). The exact mechanism by which anatomic and functional damage is inflicted on patients with NTG remains unknown; however, 2 working theories have been proposed. These theories include the vascular theory (reduced blood flow to the optic nerve) and the mechanical theory (a relatively high IOP).
Disc hemorrhages in patients with glaucoma are typically splinter-shaped or flame-shaped and lie within the superficial nerve fiber layer at the border of the optic disc. These hemorrhages provide important information for the risk assessment and treatment of NTG, since disc hemorrhages are rarely found in the normal eye. The Ocular Hypertension Treatment Study demonstrated that disc hemorrhage is a significant risk factor for the development of primary open-angle glaucoma (POAG). In the Ocular Hypertension Treatment Study, eyes exhibiting disc hemorrhage were 6 times more likely to develop glaucoma than those without disc hemorrhage. Furthermore, disc hemorrhages in patients with glaucoma are considered to be a sign of active glaucoma and carry a negative prognostic value, since they are predictive of retinal nerve fiber defects and potential VF defects. Rasker and associates previously reported that VF deterioration occurs in 80% of all patients with NTG and disc hemorrhage and in 89% of all patients with POAG and disc hemorrhage after a mean follow-up period of 9 years. In contrast, the same study also reported VF deterioration in 32% of all patients with NTG or POAG without disc hemorrhage within the same time frame. The Early Manifest Glaucoma Trial also demonstrated a significant association between disc hemorrhage and progressive VF deterioration. Other studies have concluded that even a single disc hemorrhage can result in rim notching or loss, focal retinal nerve fiber layer defects, and corresponding local or global VF deterioration.
To date, many studies have attempted to identify the risk factors that contribute to disc hemorrhage. Several of these studies have reported that antiplatelet agents, including aspirin, are strongly associated with the development of disc hemorrhage in patients with NTG. These results suggest an association between platelet function and disc hemorrhage. However, few studies have explored the relationship between platelet function and disc hemorrhage in patients with NTG. We have found that platelet function as measured by whole blood levels of collagen/epinephrine closure time influences disc hemorrhages in patients with glaucoma in our retrospective study. In the present study, we sought to determine whether platelet function is associated with disc hemorrhage in South Korean patients with NTG prospectively.
Subjects and Methods
This was a prospective cross-sectional study. Subjects were recruited from the Department of Ophthalmology outpatient service at Kangbuk Samsung Hospital, Seoul, South Korea, from May 22, 2013 to May 21, 2014. Patients with NTG and disc hemorrhage, patients with NTG without disc hemorrhage, and healthy controls were recruited prospectively. This study was performed in accordance with the Declaration of Helsinki and was approved as prospective study by the Institutional Review Board and ethics committee of Kangbuk Samsung Hospital in Seoul, South Korea. Informed consent was obtained from all patients who participated in this study.
Disc hemorrhage was defined as an isolated flame-like or splinter-like hemorrhage on the optic disc or peripapillary area (within a radius equal to 1 disc diameter). Patients with disc swelling or other abnormal findings from anterior ischemic optic neuropathies, even in the presence of flame-like or splinter-like hemorrhages, were excluded from the study. In addition, if the hemorrhage was accompanied by multiple retinal hemorrhages typical of retinal vascular diseases such as diabetic retinopathy, hypertensive retinopathy, or vein occlusion, the hemorrhage was considered to be unrelated to glaucoma and the patient was excluded.
In total, 315 subjects were enrolled in this study, including 120 patients with NTG and disc hemorrhage, 75 patients with normal-tension glaucoma without disc hemorrhage, and 120 healthy individuals (control group; age- and sex-matched). The control group consisted of age- and sex-matched subjects who were scheduled for comprehensive blood laboratory testing, including assessment of platelet function, prior to cataract surgery, blepharoplasty, lacrimal surgery, or strabismus surgery. Age- and sex-matched healthy individuals who had blood laboratory testing in the same week as NTG and disc hemorrhage patients were recruited in order to minimize selection bias. All NTG without disc hemorrhage patients were recruited during the study period.
For inclusion in this study, all participants had to meet the following criteria: a logarithmic value of the minimal angle of resolution best-corrected visual acuity (logMAR BCVA) of 0.70 or better, a spherical equivalent within 6 diopters (D) with a cylinder within 3 D, the presence of an open angle as revealed by slit-lamp and gonioscopic examinations, and reliable VF test results. Subjects with any other ophthalmic disease that could have resulted in visual field abnormalities were excluded. Diagnosis of NTG was performed by a glaucoma specialist using the following criteria: presence of a typical glaucomatous optic disc abnormality, including rim thinning or notching in the inferior or superior temporal area of the optic nerve head; corresponding glaucomatous VF loss, including paracentral scotoma, arcuate scotoma, or a nasal step; a diurnal IOP measurement less than 21 mm Hg without medication; an open anterior chamber angle on gonioscopy; and no apparent secondary cause of glaucomatous optic neuropathy. Diurnal IOP was determined by measuring the IOP every 150 minutes from 9 AM to 5 PM at the first visit. Patient exclusion criteria for this study included active ocular disease, use of other ocular medications or therapies that might have a substantial effect on IOP, a history of ocular surgery, or the use of drugs that are known to alter platelet function, such as aspirin, clopidogrel, or nonsteroidal anti-inflammatory agents.
All subjects underwent a detailed ophthalmic examination that included complete medical and ophthalmic histories; BCVA testing; intraocular pressure measurements taken by Goldmann applanation tonometry; VF testing; color disc photography; optical coherence tomography (OCT) scanning; laboratory blood tests, including measurements of whole blood levels of collagen/epinephrine closure time for assessment of platelet function; prothrombin time testing; and activated partial thromboplastin time testing. Finally, liver function tests, including serum aspartate aminotransferase and alanine aminotransferase levels; renal function tests, including blood urea nitrogen and creatinine levels; and measurement of hematocrit and platelet counts were performed. In patients with disc hemorrhage, ophthalmic examinations and laboratory blood tests were completed at the time of disc hemorrhage detection. In the control group, ophthalmic examinations and laboratory blood tests were completed before the scheduled surgery.
Blood specimens were collected from the antecubital vein. To assess platelet function, collagen/epinephrine closure time was measured in all patients using a Platelet Function Analyzer-100 (Dade-Behring, Marburg, Germany). The Platelet Function Analyzer-100 system enables the measurement of platelet adhesion and aggregation, both of which are necessary for primary hemostasis. This instrument uses small membranes that are coated with either collagen and epinephrine or collagen and adenosine diphosphate. Anticoagulated whole blood is passed through the membranes at a high shear rate to simulate capillary in vivo hemostasis. Platelets adhere to and gradually occlude a small aperture in the center of each membrane. The time it takes for the blood to completely occlude the aperture is referred to as the closure time. The closure time reflects the platelet aggregation function. As platelet aggregation increases, the closure time shortens. Conversely, the longer closure time refers to decreased platelet aggregation function. However, closure time is also influenced by von Willebrand factor and hematocrit levels, which somewhat decreases the sensitivity and specificity of the method for detecting altered platelet function. The reference interval for collagen/epinephrine in 95% of South Korean adults has been reported to be 82–192 seconds. The sensitivity and specificity of the Platelet Function Analyzer-100 collagen/epinephrine have been shown to be 90.8% and 86.6%, and the positive predictive value and negative predictive value have been shown to be 81.8% and 93.4%, respectively. There are several studies showing a good reproducibility.
Methods
Baseline demographic information and clinical parameters were compared between groups using the independent sample t test or 1-way analysis of variance (ANOVA) with the Bonferroni post hoc test for continuous variables, and the χ 2 test or Fisher exact test for categorical variables.
Odds ratios (ORs) with 95% confidence intervals (95% CIs) were estimated using logistic regression models. Following univariate analyses, multivariate logistic regression analyses were performed to identify the independent risk factors for disc hemorrhage.
The study population was also divided into 3 groups based on age (20–39 years, 40–59 years, and 60 years or older) in order to perform subgroup analyses.
All data were analyzed using the SPSS statistical software system, version 21.0 (IBM SPSS, Inc, Chicago, Illinois, USA). P values < .05 were considered to be statistically significant. In cases of bilateral NTG and healthy control, only the right eye was considered.
Results
A total of 315 patients were enrolled in this study, 171 of whom were men and 144 of whom were women. The mean patient age was 56.20 ± 14.44 years (range, 20–88 years). Seventy patients (22.2%) had diabetes mellitus, and 60 (20.3%) had hypertension. None of the patients had a history of platelet or coagulation disorder, including disseminated intravascular coagulation, von Willebrand disease, congenital afibrinogenemia, Glanzmann thrombasthenia, or Bernard-Soulier syndrome.
The disc hemorrhage group and healthy control group were closely matched in terms of mean age and sex (independent sample t test, P = .939; χ 2 test, P = 1.000, respectively). The NTG with disc hemorrhage group and the healthy control group were composed of 61 men and 59 women, respectively. The NTG without disc hemorrhage group was composed of 49 men and 26 women. No significant differences among the 3 groups were observed regarding any of the demographic and clinical characteristics, including mean IOP ( Tables 1 and 2 ). The collagen/epinephrine closure time (s), as measured by the Platelet Function Analyzer-100, of the NTG with disc hemorrhage group was approximately 14%–24% longer than those of the other groups (141.92 ± 53.44 [with NTG and disc hemorrhage] vs 124.60 ± 46.72 [with NTG without disc hemorrhage] vs 114.84 ± 34.84 [healthy individuals], 1-way ANOVA test, P < .001). The prothrombin time (s) was approximately 2%–3% longer in patients with disc hemorrhage compared with NTG without disc hemorrhage and the healthy control group (10.73 ± 0.64 [with NTG and disc hemorrhage] vs 10.45 ± 0.64 [with NTG without disc hemorrhage] vs 10.56 ± 0.56 [healthy individuals], 1-way ANOVA test, P = .047). The activated partial thromboplastin time (s) was also approximately 2%–5% longer in patients with disc hemorrhage compared with NTG without disc hemorrhage and the healthy control group (32.47 ± 3.43 [with NTG and disc hemorrhage] vs 31.73 ± 3.47 [with NTG without disc hemorrhage] vs 30.81 ± 3.05 [healthy individuals], 1-way ANOVA test, P = .004) ( Table 2 ). Although only the collagen/epinephrine, prothrombin time, and activated partial thromboplastin time differ significantly among the 3 groups, all the test results in the NTG with disc hemorrhage group were worse than the normal group.
| NTG Patients With Disc Hemorrhage (N = 120) | NTG Patients Without Disc Hemorrhage (N = 75) | Healthy Individuals (N = 120) | P Value | |
|---|---|---|---|---|
| Sex | 61 male, 59 female | 49 male, 26 female | 61 male, 59 female | .089 |
| Mean age (y) | 56.38 ± 13.87 | 55.83 ± 15.31 | 56.24 ± 14.55 | .966 |
| Diabetes mellitus | 27 (22.5%) | 16 (21.3%) | 27 (22.5%) | .978 |
| Hypertension | 24 (20.0%) | 18 (24.0%) | 22 (18.3%) | .629 |
| Hypotension | 1 (0.8%) | 0 (0%) | 0 (0%) | 1.00 |
| Heart disease | 0 (0%) | 0 (0%) | 1 (0.8%) | 1.00 |
| Hypothyroidism | 0 (0%) | 0 (0%) | 1 (0.8%) | 1.00 |
| Migraine | 0 (0%) | 1 (1.3%) | 0 (0%) | 1.00 |
| Stroke | 1 (0.8%) | 2 (2.7%) | 0 (0%) | .251 |
| Hypercholesterolemia | 10 (8.3%) | 26 (34.7%) | 7 (5.8%) | .089 |
| IOP (mm Hg) a | 14.27 ± 2.93 | 14.33 ± 4.52 | 13.91 ± 3.46 | .659 |
| HVF MD (dB) | −5.34 ± 5.81 | −7.64 ± 7.47 | – | .083 |
| HVF PSD (dB) | 5.00 ± 3.02 | 4.91 ± 3.10 | – | .972 |
a In case of unilateral NTG, the affected eye was considered; in cases of bilateral NTG and healthy control, only the right eye was considered.
| Patients With NTG and Disc Hemorrhage (N = 120) | NTG Patients Without Disc Hemorrhage (N = 75) | Healthy Individuals (N = 120) | P Value | |
|---|---|---|---|---|
| Platelet count (× 10 3 /mm 3 ) | 234.88 ± 59.90 | 254.87 ± 59.35 | 246.50 ± 58.31 | .072 |
| Blood urea nitrogen (mg/dL) | 15.76 ± 5.21 | 15.71 ± 4.70 | 15.37 ± 4.66 | .808 |
| Creatinine (mg/dL) | 1.67 ± 7.27 | 0.88 ± 0.20 | 0.86 ± 0.23 | .420 |
| Aspartate transaminase (IU/L) | 24.46 ± 8.22 | 24.08 ± 7.96 | 23.25 ± 7.17 | .501 |
| Alanine transaminase (IU/L) | 23.64 ± 11.52 | 26.20 ± 14.89 | 23.12 ± 13.68 | .306 |
| Platelet Function Analyzer-100 results collagen/epinephrine closure time (s) | 141.92 ± 53.44 | 124.60 ± 46.72 | 114.84 ± 34.84 | <.001 a , b |
| Prothrombin time (s) | 10.73 ± 0.64 | 10.45 ± 0.64 | 10.56 ± 0.56 | .047 b |
| Activated partial thromboplastin time (s) | 32.47 ± 3.43 | 31.73 ± 3.47 | 30.81 ± 3.05 | .004 a |
a P < .05 for the difference between patients with NTG and disc hemorrhage and healthy individuals (1-way ANOVA with the Bonferroni post hoc test).
b P < .05 for the difference between patients with NTG and disc hemorrhage and patients with NTG without disc hemorrhage (1-way ANOVA with the Bonferroni post hoc test).
Subgroup analyses did not reveal any significant differences in patient demographics or laboratory findings, including platelet function, in patients aged 20–39 years ( Table 3 ). However, among patients aged 40–59 years, the collagen/epinephrine closure time (s) of the NTG with disc hemorrhage group was approximately 21%–26% longer than those of the other groups (144.31 ± 58.91 [with NTG and disc hemorrhage] vs 119.03 ± 41.33 [with NTG without disc hemorrhage] vs 114.61 ± 24.72 [healthy individuals], 1-way ANOVA test, P = .002). Among patients aged 60 years and older, the collagen/epinephrine closure time was approximately 10%–25% longer in patients with disc hemorrhage compared with other groups (139.04 ± 49.12 [with NTG and disc hemorrhage] vs 126.34 ± 54.65 [with NTG without disc hemorrhage] vs 111.48 ± 41.56 [healthy individuals], 1-way ANOVA test, P = .018) ( Table 3 ). No significant differences in any other demographic factors or laboratory findings were observed among patients 40 years and older.
| NTG Patients With Disc Hemorrhage (N = 120) | NTG Patients Without Disc Hemorrhage (N = 75) | Healthy Individuals (N = 120) | P Value | |
|---|---|---|---|---|
| Platelet Function Analyzer-100 results collagen/epinephrine closure time (s) | ||||
| 20–39 years | 142.81 ± 49.64 (n = 16) | 134.33 ± 37.50 (n = 12) | 126.13 ± 40.65 (n = 16) | .558 |
| 40–59 years | 144.31 ± 58.91 (n = 54) | 119.03 ± 41.33 (n =31) | 114.61 ± 24.72 (n = 54) | .002 a , b |
| 60+ years | 139.04 ± 49.12 (n = 50) | 126.34 ± 54.65 (n = 32) | 111.48 ± 41.56 (n = 50) | .018 a |
a P < .05 for the difference between patients with NTG and disc hemorrhage and healthy individuals (1-way ANOVA with the Bonferroni post hoc test).
b P < .05 for the difference between patients with NTG and disc hemorrhage and patients with NTG without disc hemorrhage (1-way ANOVA with the Bonferroni post hoc test).
Stepwise multiple logistic regression analysis revealed that only a longer collagen/epinephrine closure time (OR adjusted for age, sex, prothrombin time, activated partial thromboplastin time, diabetes mellitus, hypertension, hypotension, heart disease, hypothyroidism, migraine, stroke, hypercholesterolemia: 2.94; 95% CI: 1.40–6.17) was independently associated with disc hemorrhage in patients with NTG and disc hemorrhage in healthy individuals ( Table 4 ). A longer collagen/epinephrine closure time was also independently associated with disc hemorrhage in patients with NTG and disc hemorrhage and NTG without disc hemorrhage (OR adjusted for age, sex, prothrombin time, activated partial thromboplastin time, diabetes mellitus, hypertension, hypotension, heart disease, hypothyroidism, migraine, stroke, hypercholesterolemia: 2.72; 95% CI: 1.08–6.85) ( Table 5 ).
