Choroidal thickness in asymp­tomatic patients with carotid artery stenosis

Summary

Aim: To measure the choroidal thickness (CT) with enhanced-depth imag­­ing optic coherence tomography (EDI-OCT) in patients with internal carotid artery (ICA) stenosis and to investigate the relationship between the CT and ICA stenosis.

Material and methods: We included 36 eyes of 25 asymp­tomatic patients with 50% or higher ICA stenosis and 36 eyes of 21 healthy controls in the study. The CT was measured with EDI-OCT from a total of 6 points in both groups. The results were compared statistical­ly between the groups.

Results: There were no significant dif­ferences between patients with asymp­tomatic ICA stenosis and non-stenotic healthy individuals at subfoveal CT (P = 0.085), at 500 μm nasal to the fovea (P = 0.076), at 1,000 μm nasal to the fovea (P = 0.052), at 500 μm temporal to the fovea (P = 0.182), at 1,000 μm temporal to the fovea (P = 0.115) and at 1,500 μm temporal to the fovea (P = 0.174). Additional­ly, no significant dif­ference was observed in CT values measured from 6 points between the stenotic side and the non-stenotic side in 14 patients with unilateral ICA stenosis (P > 0.05 for all points).

Conclusion: The CT may not alter in asymp­tomatic ICA stenosis compared with healthy non-stenotic individuals. However, more studies are needed to cor­roborate our findings.

Keywords:

internal carotid artery – Stenosis – choroid – optic coherence tomography – thickness

Introduction

Carotid artery dis­ease (CAD) is a major cause of morbidity and mortality that usual­ly develops due to atherosclerosis. It is characterized by internal carotid artery (ICA) stenosis or occlusion that can lead to cerebral or retinal ischaemia. CAD-related eye findings include amaurosis fugax, ischaemic optic neuropathy, ocular ischaemic syndrome (OIS), retinal embolism, retinal and iris neovascularization, venous stasis retinopathy and fundus cholesterol (Hol­lenhorst) plaques [1– 3]. These ocular findings are thought to occur due to thromboembolic or haemodynamic mechanisms. The ICA becomes progres­sively nar­rower due to thromboembolic mechanism in atherosclerosis. The arterial thrombi on the atheromatous plaque break off and cause occlusion and ischaemia in the distal ves­sels. On the other hand, inadequate perfusion due to chronic ICA stenosis or occlusion may lead to retinal ischaemia due to the haemodynamic mechanism [1,2,4].

Orbital vascularization is mainly from the ophthalmic artery (OA), which is the first major branch of the ICA. The OA is the ori­gin of the central retinal artery, posterior ciliary artery (PCA), lacrimal artery and muscular artery branches within the orbit. These branches make anastomoses, especial­ly with the branches of the maxil­lary artery and other external system arteries. A small part of the orbital blood flow is provided by the orbital branch of the middle meningeal artery and the infraorbital artery originat­­ing from the external carotid artery [5,6].

The choroid, located between the retina and the sclera, has one of the highest blood flow rates in the human body, receiv­­ing 70% of all the blood flow of the eye [7,8]. This vascular structure consists mainly of the posterior ciliary branches of the OA and the plexus of these branches [7]. Short PCAs supply the posterior choroid and peripapil­lary region while the anterior parts of the choroid are supplied by the long PCAs and anterior ciliary arteries, which are the terminal branches of the muscular artery that also derives from the OA [9]. Therefore, the choroid receives approximately 85% of the OA blood [5,8]. Accord­­ing to these anatomical vascular relations, ICA stenosis may af­fect the choroidal perfusion due to hypoperfusion in branches of the OA, which originates from the ICA.

Until recently, the information available on choroidal thicknes­s (CT) was based on histopathological studies conducted on cadavers. However, now it is pos­sible to obtain in vivo sections of the choroid using the enhanced depth imag­­ing optic coherence tomography (EDI-OCT), which was developed recently [10– 12]. The aim of our study was to evaluate the CT in patients with asymp­tomatic ICA stenosis and to investigate the relationship between the degree of stenosis and the CT.

We hypothesized that develop­­ing ischaemia due to ICA stenosis might af­fect the perfusion of the choroid and alter the CT even in the asymp­tomatic stage of stenosis.

Material and Methods

In total of 36 eyes of 25 patients with ICA stenosis and 36 eyes of 21 healthy individuals with a similar mean age were included in the study.

Subjects who refer­red to Kirsehir Ahi Evran University Research and Train­­ing Hospital Ophthalmology Department with no previous history of stroke or transient ischaemic attack (TIA) underwent carotid duplex US. A detailed ophthalmological examination includ­­ing visual acuity, intraocular pres­sure (IOP) measurement, anterior segment and fundus examination was performed in all subjects.

Inclusion Criteria

Subjects with a cor­rected distance visual acuity of 20/ 25 or above, – 1.5 to +1.5 dioptries of spherical refractive er­ror and with an IOP of 21 m­m Hg or less were included in the study. The study group consisted of asymp­tomatic patients with out history of stroke or TIA who had > 50% ICA stenosis at carotid US imaging. Age-matched subjects without history of stroke or TIA and with normal carotid Doppler imag­­ing and ophthalmological examination were included in the control group.

Exclusion Criteria

The presence of refractive er­ror > ±1.5 di­o-ptres, choroidal neovascularization or any other macular/ retinal dis­eases that might af­fect the vision, intraocular inflam­mation and/ or infection, or a history of any type of intraocular surgery, trauma, serious eye dis­ease (corneal dis­ease, glaucoma, serious cataract), TIA or stroke and any systemic dis­ease that might af­fect the eye (such as diabetes mel­litus, arterial hypertension, vasculitis), smok­­ing and alcohol use, cof­fee addiction or the use of vasoactive drugs were excluded

Internal Carotid Artery Imaging

Colour Doppler sonographic scan­n­­ing was performed by an Aplio 500 apparatus and a 4– 11-MHz linear ar­ray transducer (Toshiba, Tokyo, Japan). Plaque images were documented in the B-mode, colour mode and colour mode with a pulsed wave spectrum, report­­ing peak systolic velocity and end-diastolic velocity and plaque areas on longitudinal and transverse scans at the point of maximum stenosis for of­fline analysis and quantification (Fig. 1).

Optic Coherence Tomography Protocol and Choroidal Thickness Measurement

The EDI-OCT method has been described previously [13] and integrated as software into spectral domain OCT (SD-OCT) devices. We used a Heidelberg SD-OCT (Heidelberg Engineering, Heidelberg, Germany) and software version 6.3.3.0. The instrument contained an 870-nm wavelength superluminescent diode. It could acquire 40,000 A-scans per second at an axial resolution of 7 μm and a transverse resolution of 14 μm. We obtained two high-quality horizontal single line scans through the fovea within a 1 × 30 degree foveal area and averaged 100 scans for each section. The automatic real-time averag­­ing mode was used to maximize the signal-to-noise ratio and to ensure high-quality images. The CT was accepted as the distance between the outer reflective retinal pigment epithelium layer and the in­ner choroid-sclera border and measured manual­ly us­­ing the caliper tool. Measurements were made horizontal­ly across the fovea at 500 μm intervals. The measurements were performed from a total of 6 points: subfoveal (SF), 500 μ (N1) and 1,000 µ (N2) nasal to the fovea, and 500 μ (T1), 1,000 μ (T2) and 1,500 μ (T3) temporal to the fovea (Fig. 2).

Choroidal thickness measurements were performed at the same time (10:00– 12:00) every day by two ophthalmologists (CO, AK). The mean value of the measurements was calculated and recorded. In the event of a discrepancy, another measurement was performed by both ophthalmologists.

Statistical Analysis

The SPSS 20.0.0 (IBM, Armonk, NY, USA) software program was used in the analysis of the data. The measured data were described as the arithmetic mean ± standard deviation whereas the categorical data were described as percentages (%). Normal distribution of the measured data was evaluated with the Kolmogorov-Smirnov test. Student’s t-test was used to compare the groups if the data were normal­ly distributed. The Man­n-Whitney U test was used if the data were not normal­ly distributed. The relationship between the OCT values and the severity of the carotid stenosis was evaluated us­­ing the Spearman cor­relation analysis.

A post-hoc power analysis was performed us­­ing G Power 3.1.9.2 software for Windows (Heinrich-Heine-Universität Düs­seldorf, Düs­seldorf, Germany). Accord­­ing to this test, our study had 99.9% power within 0.05 alpha [13].

Sample Size Calculation

Since similar studies were published, to be able to determine the adequate sample size, accord­­ing to the groups in the previous study by Wang et al [13] we tried to predict the ef­fect size us­­ing the descriptive statistics in the “central choroidal thicknes­s” variable, which was the same term with subfoveal thicknes­s. We calculated a total of 54 subjects, 27 subjects in each group, to be able to statistical­ly detect 17.70 units of dif­ference between the groups in terms of SF CT under the conditions of 80% power and 5% type I er­ror.

Results

The study group (group 1) included 36 eyes (16 right, 20 left) of 25 patients and the control group (group 2), 36 eyes (19 right, 17 left) of 21 subjects. The mean age was 69.32 ± 9.27 (49– 80) years in group 1 and 70.52 ± 9.03 (51– 83) years in group 2; there was no significant dif­ference between the mean ages of both groups (P = 0.691). The study group included 5 women and 20 men and the control group 5 women and 16 men (P = 0.755). Eleven patients had bilateral and 14 patients had unilateral ICA stenosis. The degree of ICA stenosis was 50– 70% in 23 patients and ≥ 70% in 13 patients. Two patients had total occlusion. The total mean degree of ICA stenosis was 64.4% (50– 100) (Tab. 1).

The SF CT was 219 μm in group 1 and 242.9 µm in group 2 (P = 0.085). Also, measurements of the CT from extrafoveal points were not significantly dif­ferent between both groups (Tab. 2).

The relationship between the degree of stenosis and the CT was also investigated and no statistical­ly significant cor­relation was found (SF [P = 0.589], N1 [P = 0.424], N2 [P = 0.288], T1 [P = 0.345], T2 [P = 0.611], T3 [P = 0.916]).

In 14 patients with unilateral ICA stenosis, no significant dif­ference was observed in the CT values measured from 6 points between the stenotic side and the non-stenotic side (P > 0.05, for all points) (Tab. 3).

Discus­sion

In the present study, we evaluated the CT in asymp­tomatic ICA stenosis us­­ing EDI-OCT. This study showed that there was no significant dif­ference in CT between patients with asymp­tomatic ICA stenosis and healthy individuals without ICA stenosis who were age- and gender-matched. Additional­ly, in patients with unilateral stenosis, no significant dif­ference was observed in the CT at the side with ICA stenosis compared to the non-stenotic side.

The choroid is one of the most vasculari­zed tis­sues in the body and receives the highest blood volume among the ocular tis­sues [8]. With the advances in OCT technology and updated software in recent years, evaluation of the choroid has become pos­sible [10– 14]. The relationship between the CT and gender, age and circadian rhythm has been investigated in certain studies on factors af­fect­­ing CT and haemodynamics. CT has been reported to be higher in males and in young people. It has also been suggested to be higher at night and to be related to systolic blood pres­sure changes [15– 17].

In our study, we hypothesized that ICA stenosis might af­fect the choroidal perfusion and CT due to vascular mechanism. Several studies showed that the choroid was sensitive to blood pres­sure changes and was af­fected by blood flow and perfusion pres­sure [6]. Congestion-related SF CT increase has been reported in a case with carotid cavernous fistula (CCF) with the value decreas­­ing after treatment with embolization [18]. The SF CT of a 47-year-old female dia­gnosed with CCF was found to be significantly higher on the fistula side in a similar case report. Once the fistula was embolized through the OA, the asym­metry disappeared and the subfoveal CT became equal on both sides. The authors reported that OCT could be used as an auxiliary test in the dia­gnosis of CCF [19]. Lareyre et al retrospectively evaluated SF CT changes fol­low­­ing carotid endarterectomy (CEA) in patients with severe chronic carotid stenosis. They found that SF CT increased bilateral­ly and more prominently on the ipsilateral side fol­low­­ing CEA [20]. The cor­relation between ocular pulse amplitude (OPA), subfoveal CT and ICA Doppler US findings was investigated in another study. OPA was found to provide useful information on intraocular blood flow and to be an indirect indication of choroidal perfusion. The results were reported to indicate a moderately positive cor­relation between OPA and SF CT [21]. Despite these reports [17– 21] that have investigated changes in ocular blood flow and perfusion, there is still limited information in the literature regarding the CT of patients with ICA stenosis. Kang et al reported that the CT as measured with OCT was lower in the eye on the stenotic side than on the healthy side in 3 cases with severe ICA stenosis [22]. Unlike this previous report, in our study, no statistical­ly significant dif­ference was observed between the CT values on the stenotic and the non-ste­notic side in unilateral ICA stenosis. In a cros­s-sectional study performed by Sayin et al, a decrease in the CT central and paracentral to the fovea was shown in patients with ICA stenosis, but no significant cor­relation could be shown between the degree of stenosis and the CT value [23]. Similarly to the literature, in our study there was no statistical­ly significant cor­relation between the degree of ICA stenosis and the CT.

Recent studies have investigated the relationship between ICA stenosis and OIS. The CT and choroidal volume values were compared with the af­fected and healthy eyes of 19 OIS patients in one study and the SF CT and choroidal volume were shown to have decreased in these eyes. The interpretation was that ICA stenosis decreases choroidal circulation [24].

In a retrospectively designed study performed by Wang et al, the CT values were lower in patients with severe ICA stenosis and they suggested that choroidal thin­n­­ing might occur before retinal changes in OIS patients and evaluation of the CT may therefore be useful in choos­­ing the optimal therapeutic tim­­ing for patients with ICA stenosis [13]. About 30% of patients with symp­tomatic ICA occlusion were reported to have asymp­tomatic retinal vascular changes and 1.5% of these were to become symp­tomatic within 1 year in another study [25]. The degree of stenosis, presence of col­lateral ves­sels, duration of CAD, presence of bilateral or unilateral stenosis and, presence of systemic vascular dis­ease determines the severity of OIS [26]. A comparison of patients with a mean stenosis of 74 and 47.5% showed SF CT values of 231.9 μm and 216.2 μm, resp., in a study evaluat­­ing the relationship between ICA stenosis and SF CT in the elderly population. The authors stated that a compensatory increase in SF CT could be seen with an ICA stenosis of 70% and higher [27].

Our study has several limitations. First, the degree of ICA stenosis was less than 70% (mean 64.4%) in the majority of our patients. Second, the sample size of the study was relatively smal­l. The lack of a statistical significance could be explained by the low degree of ICA stenosis and a small sample size. Third, the axial length was not measured in our patients. However, we excluded patients with myopia and hypermetropia with a spherical equivalent of ±1.5 dioptries or higher to minimize the ef­fect of the axial length. Final­ly, despite the advances in OCT technology, the choroidal borders are still defined by manual measurement, which is also encountered as a limitation in all studies in this field.

Conclusion

The present study showed that there was no dif­ference in the CT of asymp­tomatic ICA stenosis patients compared with non-stenotic individuals. However, more studies with a larger sample size are needed to evaluate the ef­fects of ICA stenosis and cor­roborate these findings.

Ethical Principles

The study was approved by the Local Ethics Committee and followed the principles of Helsinki Declaration of 1975 (as revised in 2004 and 2008). The participants were informed about the study and written consent was obtained from all subjects.

Disclosures

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

The Editorial Board declares that the manu­script met the ICMJE “uniform requirements” for biomedical papers.

Accepted for review: 22. 5. 2019

Accepted for print: 18. 11. 2019

Çağlar Öktem, MD

Department of Ophthalmology

Alaaddin Keykubat University

Alanya Education and Research Hospital

074 00 Antalya

Turkey

e-mail: cglroktm@gmail.com