Retinal Vascular Changes with Non-Arteritis Anterior Ischemic Optic Neuropathy Assessed with IVAN Analysis

To assess the vascular changes associated with non-arteritis anterior ischemic optic neuropathy (NAION) by using integrative vessel analysis (IVAN) and risk factors associated with NAION. Methods We performed a case-control study of 120 patients, including 40 NAION patients during July 2015 to July 2019. Patients were divided into Normal, hypertension and NAION groups. IVAN was conducted to evaluate the caliber of retinal vessels. In addition, we also analyzed risk factors (age, gender and BMI) to identify their association with NAION.


Abstract Background
To assess the vascular changes associated with non-arteritis anterior ischemic optic neuropathy (NAION) by using integrative vessel analysis (IVAN) and risk factors associated with NAION.

Methods
We performed a case-control study of 120 patients, including 40 NAION patients during July 2015 to July 2019. Patients were divided into Normal, hypertension and NAION groups. IVAN was conducted to evaluate the caliber of retinal vessels. In addition, we also analyzed risk factors (age, gender and BMI) to identify their association with NAION.
Results 84 women and 36 men were involved in our study, the average age was 58 ± 9.98 years. CRAE, CRVE and AVR in the whole patients were 154.54 ± 21.53 µm, 252. 22

Conclusion
Retinal vascular caliber can be used as an indicator to evaluate the NAION. Our study suggested that an assessment of retinal vascular caliber may offer insights into the development of NAION.

Background
Non-arteritis optic neuropathy (NAION) is one of the most common disease that in uences the patients over 50 years old and causes irreversible visual loss [1]. Although NAION is a major cause of optic neuropathy except glaucoma, its pathogenesis, clinical characteristics and treatments still remain controversial [2]. Some researchers [1,3,4] considered a disk at risk (de ned as a crowded small opticnerve head with a small physiological cup) may be a major risk for NAION's development. In addition, poor optic nerve head blood supply, location of posterior ciliary artery watershed zones, ciliary artery disease etc. are some of the considered local risk factor of NAION [5,6]. Furthermore, several systemic risk factors were thought to be associated with NAION, including arterial hypertension, diabetes mellitus (DM), arterial hypotension, hypercholesterolemia, stroke, ischemic heart disease, tobacco use and so on [7][8][9].
Progressions in digital retinal photography and software have made it more convenient for us to measure the characteristics of vessels in retinal [10,11]. IVAN is a software designed by University of Wisconsin, which could measure retinal vessel caliber by using fundus photographs [12]. Since researchers considered NAION as a disease which results from vascular insu ciency [13,14], we assumed that vascular diameter is involved into pathophysiology of NAION.
By searching the database on the Internet, we found only few papers studied about risk factor of NAION.
Also, this is an innovative research that explore numerous risk factors of NAION in different groups. In this study, we measured retinal vessel caliber from fundus pictures by using IVAN to assess its relevance to the progression of NAION. Besides, we also value other factors to explore their in uences on the disease.

Study population
Tianjin Medical University Eye Hospital approved this study. All the patients involved in our study presented to Tianjin Medical University Eye Hospital service between July 2015 and July 2019.
We categorized the patients into three groups when they were in the outpatient clinic: normal group (Normal), hypertension group (Hyper) and NAION group (NAION). Grouping criteria are as below a) Patients without any systematic or retina diseases were included into our Normal group. This group comprised subjects with BCVA > 20/63, IOP < 21 mmHg, an open angle and normal optic disc appearance on fundus examination, there were no abnormities in anterior chamber. The degree of opacity of the lens should allow us to take clear photos. b) Hyper group consisted patients with hypertension (< 120 mmHg Systolic BP (SBP) and < 80 mmHg Diastolic BP (DBP)). The Inclusion criteria for this group was that patients should had diagnosis of hypertension from a physician and did not have any other systematic diseases, such as coronary heart disease, stroke, diabetes mellitus and so on. Besides, criteria for diagnosing NAION were 1) acute, painless vision loss in 2 weeks; 2) optic disc edema with or without super cial hemorrhage. I120 eyes from 120 patients (69 women, 51 men) were included in our research. Patients underwent a thoroughly ophthalmic evaluation, including best-corrected visual acuity (BCVA), intraocular pressure (IOP) with a noncontact pneumatic tonometer (Canon T-2; Canon, Tokyo), slip-lamp biomicroscopy and fundus examination. Other imaging examination including OCT, uorescein angiography (FA) and indocyanine green angiography were performed depended on the patients' situation. For NAION patients, we performed OCT, FA, indocyanine green angiography and visual elds test routinely. At last, medical history such as age, gender, BMI were also recorded to evaluate their relevance with NAION.

Retinal Fundus Pictures and Vessels' calculation
Fundus pictures were taken without dilating the pupils: a 30° color fundus picture of the eye centered on the disc were taken. We measured the caliber of vessels in patients' retina by using IVAN (University of Wisconsin, Madison) in fundus pictures from camera [15]. Brie y, the IVAN protocol required that vessel diameters should away from the six largest arterioles and six largest venules located in a zone, where 0.5 to 1.0 disc diameters from the disc margin. The measurement was performed by two ophthalmologists, who were masked to the characteristics of patients.

Ethics Statement
The institutional review board of Tianjin Medical University approved this clinical study protocol, and the protocols also adhered to Declaration of Helsinki. Besides, an oral informed consent was obtained from the patients after an explanation of the procedures and purpose of the examinations.

Statistical Analysis
All the statistical analyses in this study were conducted by using SPSS statistics version 19.0 (SPSS Inc., Chicago, IL). Our data were presented as mean ± standard deviations. We performed one-way analysis of variance (ANOVA) for comparisons of parametric variables, after that a post hoc comparisons using the Tukey HSD test was performed to compare difference between each group.
We performed the univariate analysis with the Spearman's rank correlation coe cient tests. The signi cance of the correlations between the CRAE and CRVO, were determined relative to the systematic factors. A P value of < 0.05 was considered statistically signi cant.

Results
The baseline characteristics of the patients were summarized in Table 1. We divided the patients into three groups: Normal group (Normal), hypertension group (Hyper) and NAION group (NAION). 120 patients (69 women and 51 men) were included in our study. The average age (± standard deviations) was 58 ± 9.98 years for the involved patients, and 57 ± 12 years, 56 ± 8.01years, 58 ± 11.4 years for Normal, Hyper, NAION group, respectively. Figure 1 showed representative pictures for each group. There were no difference of age and gender in three groups from baseline.

Analysis of Differences of Vessels Caliber between Groups.
A one-way between subjects' ANOVA was conducted to compare the retinal vessel caliber between 3 different groups (Fig. 3

Discussion
Totally 120 patients were involved in our study. Our research compared the caliber and their ratio in different groups of patients, NAION group has widest vessels' caliber. Also, we found negative correlation between the BMI value and CRAE, while positively correlation with CRVE, we assumed the obesity may have relationship with retinal vessels' calibers. Nowadays, several methods have been used to measure the retinal vascular caliber from the fundus images [16][17][18]. The IVAN method can be used as a "marker" for the systemic factors, such as age, hypertension, diabetes mellitus and physical activity [19].
When comparing CRAE, we found that CARE in NAION group was signi cantly increased than that in Normal group. The result was not showed in other papers previously. We assumed this artery changes may be caused by "hypoxia" status of the retinal environment in NAION. When the tissue is hypoxic, hypoxic vasodilation appears to increase the perfusion of blood to the tissue [20]. In response to decrease of ATP, K ATP channels of vascular smooth muscle cells opens, inducing the vasodilation [21].
While the result of comparing CRAE showed no difference between Normal and Hyper group, we can nd mean value of CRAE in Hyper group was smaller than Normal group. The narrowing trend had been con rmed by numerous studies [22,23]. Comparing with other systemic factors, retinal arteriolar narrowing has been reported to have the strongest relationship with the hypertension [24].
Comparing CRVE in 3 groups showed NAION group had the largest retina venous dimension. This result was consistent with previous study of NAION. Some researchers considered most of the NAION Patients have severe and acute optic disk edema, this abnormal changes may some degree of restriction of central retinal venous return, elevating the retinal venous pressure, thus leading the vascular dilation [25].
While the analysis showed no signi cantly difference between Hyper and Normal group, the mean value of CRVE in Hyper group was higher than it was in Normal group. This result was consistent with Jonas et al.'s study retinal vein diameters are associated with higher blood pressure [26].
The AVR result showed the ratio in NAION group was the smallest among three groups. This demonstrated the venous dilation was greater in NAION group than the other two groups. The result was also consistent with previous studies. In the Atherosclerosis Risk In Communities (ARIC) study, nonspeci c in ammatory factors were associated with smaller AVR [27]. Researchers considered the result was associated to arteriolar narrowing, while later it was con rmed to be related to venular dilation [28]. While Wong TY et al. found no association between nonspeci c in ammatory markers and vessel diameters in the Cardiovascular Health Study (CHS) [29].
Caliber of retina vessels in NAION group was greater than that in other groups, which may be associated with the ischemia state of the NAION patients [30]. For the unknown cause of the disease, presence of a small crowded optic disc and systemic comorbidities are the two most acceptable reasons for the disease [31]. Studies had showed that NAION patients have higher prevalence of "having absent or small cup", this unique structure may result in crowding of optic nerve bers and lead a restricted space in the optic disc [32,33]. Ischemia state stimulate the vessels dilate so the tissues could gain enough oxygen as this is a compensatory response.
Nevertheless, numerous researchers were trying to gure out the pathogenesis of NAION, to our knowledge, the mechanisms of NAION are still unknow. Some studies showed that a small cup to disc ratio was essential for the ANION development [34,35]. While other researchers considered that other factors such as optic nerve drusen and papilledema, which could lead to a "crowding optic nerve head", may be crucial to the disease [36]. Since the results of studies remain variable, we performed the analysis to nd association between disease and factors, the retinal vessels caliber changes in NAION.
Age, gender and BMI were discussed as risk factors of NAION in many studies. Melson  However, limitations of our study still existed. First, our study only included Chinese patients, more ethnic groups were needed to make the study more authentic. Second, our study has 120 involved patients, more patients were needed for the further study. At last, in hypertension group DM's duration and stages should be considered. We set the hypertension group to compare whether the caliber changes were primary or secondary to the disease, changes of blood vessels in hypertension patients were secondary to hypertension, we assumed part of caliber changes in NAION may have similar pathophysiology with hypertension.

Conclusions:
In summary, the result of our study indicated that retinal vascular caliber is related to BMI, obesity maybe a risk factor to the NAION patients. Arteriolar narrowing is related to hypertension, while venular dilation and higher AVR was associated NAION patients. Our result suggested an assessment of the CRAE, CRVE and AVR may be a marker for the NAION patients and more attention should be paid for the subclinical vascular abnormalities.  Representative IVAN measurement of vascular caliber. Arterioles are in red and the venules are in blue.
The measured area of retinal vascular caliber was standardized and de ned as the showed region from 0.5 to 2.0 disc diameters away from the disc margin. A, C: measurement of vascular caliber. B, D: calibers measured result in IVAN software.