Comparison of retinal vascular arcade trajectory between eyes with an idiopathic macular hole and the healthy fellow eye

We aimed to present a method to evaluate the retinal arcade vascular trajectory by measuring the distance between the retinal veins in eyes with idiopathic macular hole (IMH) and comparing this measurement with healthy fellow eyes. In this cross-sectional study 18 patients with unilateral IMH were enrolled. We used standard fundus photographs, which were obtained from each eye with a digital fundus camera using a 55-degree lens. The calculation of the retinal arcade vascular trajectories was done by drawing and measuring five vertical lines within the macular area. The mean age of the patients was 65 ± 9 years. The between-groups differences of each vertical line were not statistically significant, except for the differences between the most temporal line, which was greater in the diseased group (9388 vs. 8322 μm; P = 0.034). The ratio between the fifth (most temporal) and the third (fovea-center) vertical lines was greater than 1 (V-shape) in 72% of eyes with a macular hole, whereas it was less than one (U-shape) in 78% of control eyes (P = 0.003). We demonstrated that in eyes with an IMH the vascular arcade has more tendency to diverge on its path temporal to the fovea.


Introduction
Macular hole is a full-thickness defect of the retinal tissue involving the anatomic fovea, and primarily, the foveola; thereby causing significant visual loss and metamorphopsia [1]. Macular holes could be found in highly myopic eyes or following ocular trauma, but the great majority are idiopathic [1,2]. Idiopathic macular hole (IMH), with the approximate prevalence of 3.3 per 1000 people [3], usually affects healthy individuals with normal refractive errors in their sixth or seventh decades of life, and roughly two-thirds of the involved patients are females [3,4]. The risk of fellow eye involvement has been reported as 15.6% (range 3 to 22%) [5].
The primary cause of IMH is supposed to be originated from the abnormal attachment of the posterior hyaloid to the fovea during the process of posterior vitreous detachment, which exerts adequate force to detach and tear off the central fovea that eventually evolves into a full-thickness macular hole [1,2,6]. Nonetheless, the macular hole could be spontaneously closed, provided that the posterior vitreous has detached completely from the macula [1,2]. However, the tangential forces from the surrounding retina might prevent a fresh macular hole from successful closure. A typical vitrectomy for IMH usually includes induction of posterior vitreous detachment (if not already occurred) plus peeling of the internal limiting membrane of the retina for about 1-to 2-disk diameters [7][8][9]. The latter procedure would relax the retina and decrease the mentioned tangential forces to maximize the chance of macular hole closure.
Therefore, the tangential forces across the retina may have a role in the development of an IMH or its persistence after surgery. It is difficult to measure the retinal tangential force in vivo. The retinal vessels pass through the inner retina, and hence the trajectory of the arcade vessels may be correlated with retinal tangential forces at the macula. In this study, we aimed to present a method to evaluate the trajectory of arcade vessels and study its association with the development of IMH.

Subjects
In this cross-sectional study, patients with unilateral IMH (stage C 2) who were referred to Khalili hospital, Shiraz, Iran, from February 2017 to March 2019 were included. The diagnosis of IMH was confirmed by a retinal specialist (MHN) according to physical examination and OCT findings. We excluded eyes with concomitant glaucoma, myopic macular hole, diabetic retinopathy, retinal dystrophies, previous ocular trauma, retinal detachment, or any serious ophthalmic problem. The normal fellow eyes should also have no macular hole of any stage, vitreomacular traction, or epiretinal membrane. Finally, 18 patients were enrolled in the study.
Informed consent was obtained from all participants. The protocol of the study was approved by the Ethics Committee at Shiraz University of Medical Sciences and adhered to the principles of the Declaration of Helsinki.

Measurements
For assessing the trajectory of arcade vessels, we used fundus photographs which were obtained with an OCT device equipped with a digital fundus camera (Heidelberg Engineering, Heidelberg, Germany) using a 55-degree lens to create a wider field of the image. A single expert examiner took all images with the same camera. Then, the images were transferred to the Photoshop CS6 interface for measurements. The calculation of the arcade vessels' trajectories was performed by measuring the distance between the upper and the lower temporal retinal veins in five vertical lines. The first distance (D1) crossed the temporal border of the optic disk. The second line (D2) crossed the median distance from the center of the fovea to the temporal border of the optic disk. The third line (D3) crossed the center of the fovea. The fourth and fifth distances (D4 and 5) were mirror images of the second and the first lines according to the center of the macula, respectively. We also measured the distance between the fovea and the disk, and also the distances of the fovea from the superior and the inferior veins through the third vertical line (as described previously) (Fig. 1). The logic behind picking the veins (instead of arteries) for measurements was our initial observation of all images, in which the arcade veins showed less variability in branching than arteries. After each branching of the temporal veins, we chose the thicker vein as the main branch for measurement. To mask the IMH eyes from the examiner who measured the mentioned lines, the fovea of all fundus images was covered with a digital square shield prior to the analyses.

Statistical analysis
All analyses were performed using the SPSS software version 21. The normality of the data was assessed by the Kolmogorov-Smirnov test. Paired t test or Wilcoxon signed-rank test was used to compare the variables between fellow eyes (IMH vs. normal), when the data were distributed normally or not, respectively. The Spearman test was used to explore correlations between the macular holes features and the retinal vascular trajectory variables. The possible association between the stage or diameter of the macular hole and the vascular arcade shape was evaluated by the Chi-squared test. A P value of less than 0.05 was considered significant.

Results
Overall, data from both eyes of 18 patients with unilateral IMH were gathered and analyzed. The mean age of the enrolled cases was 65 years (SD, 9), and ten (55.6%) were female. There was no significant difference between the fellow eyes (IMH vs. normal) in laterality or axial length. The baseline characteristics of the patients and their eyes are summarized in Table 1.

Discussion
This study showed that the arcade veins tended to adopt a more divergent path in eyes with IMH than the control eyes. This divergence in vascular paths was particularly evident temporal to the fovea. As a result, eyes with IMH might have a more chance to assume a V-shape arcade in contrast to the normal eyes which usually take a U-shape path.
The results of our study are in line with Yoshihara et al. [10], who demonstrated that the eyes with IMH have a significantly wider retinal artery trajectory compared to the normal fellow eyes. Our study, however, has several differences from Yoshihara et al. in terms of the methods used to assess the arcade vessel trajectory. They picked retinal arteries, while we chose retinal veins. The logic behind choosing the veins over the arteries in our study was provided in the method section. Compared to Yoshihara et al., we also used a completely different and simpler way to measure retinal vascular paths. Moreover, their method mostly included the vascular path between the fovea and the disk, while we also analyzed the same distance temporal to the fovea. Anyway, with its distinct method, the present study confirmed the findings of Yoshihara et al. regarding a wider retinal vessel trajectory in eyes with IMH.
The fovea of the human eye is subjected to two main tractional forces from the posterior surface of the vitreous and tangential forces from the surrounding retina. The vitreous traction is supposed to be the most important cause of developing IMHs [1,11]. The fovea would be torn off if the mentioned forces overcome its tensile strength. For eyes with a persistent attachment of the vitreous to the edges of the macular hole, vitrectomy is mandatory to resolve the vertical (z-axis) traction from the vitreous, and this step is considered the most important phase of the operation in these eyes [12,13].  It is not thoroughly understood that why some macular holes with released vitreous traction are spontaneously closed, while others would persist and ultimately need surgical intervention. The flow of liquid vitreous through the macular hole and tangential forces from the surrounding retina are the major culprits [1,2]. A typical example of retinal tangential forces is the development of epiretinal membrane in the macular area [14]. A more subtle traction is introduced by the internal limiting membrane; hence, internal limiting membrane peeling is considered an essential part of modern macular hole surgery [7][8][9].
It is difficult to measure retinal tangential forces in vivo. Intuitively, the wider path of arcade vessels might be correlated with a greater tangential force at the fovea. These forces should probably exert their effect via the y-axis (Fig. 1B). In our study, we did not find any intergroup difference in the foveadisk distance, which could be correlated with tangential forces in the x-axis. These justifications, however, need to be verified by future studies. Fig. 3 Typical V-shape versus U-shape arcades in two cases with unilateral idiopathic macular hole. A, B the V-shape arcades in eyes with idiopathic macular hole. C, D the respective normal fellow eyes with the U-shape arcade vessels Fig. 4 Bar chart demonstrating the frequency of V-shape versus U-shape arcades in eyes with idiopathic macular hole versus control fellow eyes In addition, since the vitreous is usually attached firmly to retinal vessels, the larger distance between arcade vessels can generate stronger tractional forces (from the remaining vitreo-foveal attachment) on the fovea by inertia or momentary force during eye movements, leading to an increased chance for macular hole formation [10].
We found no significant difference between eyes with IMH and normal fellow eyes in vertical arcade vein distances and proportions nasal to the fovea, while we showed a significant difference in D5 and D5/D3, indicating a wider vascular path for eyes with macular hole temporal to the fovea. This finding might emphasize the importance of vertical tractional forces that are exerted temporal to the fovea rather than the nasal tractions. Considering the temporal venous paths, and based on the D5/D3 measurements, we introduced a classification (i.e., V-shape vs. U-shape), which might be clinically useful in a rapid distinction between eyes with probably more y-axis tangential traction (V-shape; and though a greater risk of MH formation) versus those with less traction (U-shape).
We found no significant association between the stage or diameter of macular holes and vessel trajectory variables. Indeed, full-thickness macular holes tend to progress in stage and diameter over time [15], and thus the duration of the macular hole is a major determinant (and though a major confounder) of its diameter and stage. Therefore, it would be difficult to provide a valid assessment for an independent association between macular hole features and vascular trajectory parameters.
The present study did not include cases with fullthickness or outer lamellar macular microholes (\ 150 lm). These cases generally have a favorable prognosis, and vitreomacular traction is involved in the pathogenesis of a significant subset of patients [16]. The role of arcade vessels trajectory in the development of macular microholes could be the subject of future studies.
We also did not enroll patients with secondary macular holes such as traumatic and myopic macular holes or macular holes on top of a pigment epithelial detachment [17]. In these conditions, the pathophysiology of the hole formation and contributing tractional forces probably differs from an IMH, and thus the findings of our study should not be extrapolated to these entities. This study is limited by its relatively small sample size. However, it seems acceptable from a statistical point of view, since we had matched controls (fellow eyes). Involving the fellow eye as control could be a power of the present study because it removes most of the confounding factors such as age, sex, and also most ocular biometric features, which are typically similar in fellow eyes. However, since the fellow eyes of IMHs are at greater risk of developing macular holes [18], the findings of this study should be confirmed by future studies with larger sample sizes that include normal patients as well. Finally, we used Photoshop CS6 interface to calculate retinal distances; therefore, the measured distances (in micrometers) are not exactly the actual values (and are somehow magnified). However, this issue affected all images in the same manner, and so it did not compromise the validity of statistical analyses or calculated proportions.

Conclusion
Our study showed that the arcade veins adopt a more divergent path (particularly temporal to the fovea) in eyes with IMH than their fellow eyes. From a clinical point of view, a V-shape arcade might be associated with an increased risk of macular hole formation compared with a U-shape vascular path. This association, however, does not prove any cause-and-effect relationship; such assumptions should be evaluated with appropriate prospective studies.
Ethical standards This study was conducted in accordance to the tenets of the Declaration of Helsinki and its protocol was approved by the ethics committee at Shiraz university of Medical Sciences (Approval ID: IR.SUMS.MED.REC.1398.167).
Human and animal rights Patients signed informed consent regarding publishing their data and photographs.
Informed consent Informed consent was obtained from all individual participants included in the study.
Consent to participate Informed consent was obtained from all participants.