Color Doppler imaging of the retrobulbar vessels in diabetic retinopathy

Abstract

Diabetes is a metabolic disease characterized by elevated blood glucose level due to impaired insulin secretion and activity. Chronic hyperglycemia leads to functional disorders of numerous organs and to their damage. Vascular lesions belong to the most common late complications of diabetes. Microangiopathic lesions can be found in the eyeball, kidneys and nervous system. Macroangiopathy is associated with coronary and peripheral vessels. Diabetic retinopathy is the most common microangiopathic complication characterized by closure of slight retinal blood vessels and their permeability. Despite intensive research, the pathomechanism that leads to the development and progression of diabetic retinopathy is not fully understood. The examinations used in assessing diabetic retinopathy usually involve imaging of the vessels in the eyeball and the retina. Therefore, the examinations include: fluorescein angiography, optical coherence tomography of the retina, B-mode ultrasound imaging, perimetry and digital retinal photography. There are many papers that discuss the correlations between retrobulbar circulation alterations and progression of diabetic retinopathy based on Doppler sonography. Color Doppler imaging is a non-invasive method enabling measurements of blood flow velocities in small vessels of the eyeball. The most frequently assessed vessels include: the ophthalmic artery, which is the first branch of the internal carotid artery, as well as the central retinal vein and artery, and the posterior ciliary arteries. The analysis of hemodynamic alterations in the retrobulbar vessels may deliver important information concerning circulation in diabetes and help to answer the question whether there is a relation between the progression of diabetic retinopathy and the changes observed in blood flow in the vessels of the eyeball. This paper presents the overview of literature regarding studies on blood flow in the vessels of the eyeball in patients with diabetic retinopathy.

Introduction

Diabetes is a metabolic disease characterized by chronic hyperglycemia and disorders in the carbohydrate, lipid and protein metabolism due to impaired insulin secretion and/or activity. The disease is caused by multiple factors, including genetic and environmental ones as well as those related to lifestyle. The basic classification distinguishes between type I diabetes – insulin-dependent disease which is a consequence of immunological damage to the pancreatic cells and occurs mainly in young patients, and type II diabetes – insulin-resistant disease which occurs in the majority of other patients and is associated with impaired insulin activity and secretion. Other types of diabetes are rarely encountered⁽¹⁾. Epidemiological studies unequivocally indicate a constant increase in the incidence of diabetes, which is associated with ageing population, longer life expectancy of diabetic patients as well as demographic and economic development⁽²⁾. According to WHO analyses, in the period from 2000 to 2030, the number of patients will have doubled (from 170 million to approximately 366 million). The data concerning the population in Poland also confirm a constant increase in the incidence of both type I and type II diabetes. Type II diabetes develops in over 5% of the population. It is predicted that by the year 2025, there will be a four-fold increase in the incidence of type I diabetes among the pediatric population^(3–5).

The complications related to diabetes include acute, life-threatening metabolic disturbances and chronic complications which predominantly involve slight and large blood vessels and the nervous system. The risk of vascular complications increases with the duration of the disease. Diabetic retinopathy is the most common microangiopathic complication, specific for type I and II diabetes.

The features or retinopathy, which is the prime cause of blindness in patients aged 20–74, are observed nearly in all patients with type I diabetes and in over 60% of patients with type II diabetes who have suffered from this disease for 20 years. Genetic factors and metabolic disturbances are responsible for the progression of this complication. The increase in the glucose level and the impossibility of obtaining physiological metabolism of the retina lead to microangiopathy which develops in the retinal capillary vessels, arterioles and veins. The mechanisms underlying vascular alterations are not fully understood. Biochemical processes that take place in the cells are associated with, among others, oxidative stress, kinase C activation and production of factors that stimulate growth of blood vessels, so-called angiogenic factors. The basement membrane becomes thickened and epithelial cells are damaged which results in the closure of slight vessels and increase in their permeability. Alterations in the vascular wall and blood rheological factors lead to microaneurysms, extravasation, effusion, edema, formation of pathological vessels and proliferation of capillaries. Impaired vision is the consequence of these changes and it is clinically observable as macular edema, extravasation of blood into the vitreous humor and/or retina, tractional retinal detachment or secondary glaucoma. The following methods are used to assess diabetic retinopathy: ophthalmoscopic evaluation of the ocular fundus, examination through a slit lamp with the application of lenses, retinal photography, and supplemental examinations, such as: fluorescein angiography, optical coherence tomography of the retina and B-mode ultrasound imaging^{(6, 7)}.

Doppler sonography in diabetic retinopathy – literature overview

In order to evaluate blood flow velocity in ocular and retrobulbar vessels, Doppler imaging techniques are applied. They enable qualitative and quantitative assessment of the ophthalmic artery (OA) and its branches. The most frequently assessed vessels include: the OA, the central retinal vein and artery (CRV, CRA), and the posterior ciliary arteries (PCAs). During examinations, vascular morphology is assessed and blood flow parameters are evaluated, such as: peak systolic velocity (PSV), end-diastolic velocity (EDV), mean velocity (MV), resistance index (RI) and pulsatility index (PI). The measurements are taken with the use of a linear or sector probe with the frequency of over 7 MHz by applying it to the closed eyelid of a patient in the supine position. Pressure to the eyeball should be avoided. It is recommended to maintain uniform conditions of examinations, such as: lighting in the room, size of the gate and a constant site for measurements in subsequent examinations^(8–12).

The literature mentions the presence of hemodynamic alterations in the vessels of the eyeball observed in diabetic patients. Some authors, however, report different results. This may be a consequence of assessing vascular changes in various stages of diabetes and non-uniform criteria of patient selection. This diversity concerns the types of diabetes, its duration, the stadium of diabetic retinopathy, concomitant diseases, coexistent ocular lesions and tested parameters. The essential assumption in these observations is the assessment of flow parameters and their relevance in the prognosis of the disease progression and its monitoring.

Many authors refer to the studies of Goebel et al.⁽¹³⁾ who were ones of the first to demonstrate a high sensitivity of this method in monitoring hemodynamic changes in diabetic retinopathy. They observed a significant decrease in PSV and EDV in the CRA as compared to healthy individuals. The study group compared patients with simple, preproliferative and proliferative types of retinopathy. A decrease in blood flow velocity in the retrobulbar vessels was observed in all groups, which correlated with the progression of diabetic retinopathy. Unambiguous conclusions were not drawn only in patients with simple retinopathy. The authors did not notice any significant difference in flow parameters in the OA and PCAs. The resistance index (RI) was not calculated.

Güven et al.⁽¹⁴⁾ reached the same conclusions. The authors noted a decrease in PSV in the CRA in patients with non-proliferative and proliferative retinopathy as well as in patients after retinal laser panphotocoagulation. In this case, the RI index was not calculated either and no significant differences in the OA and PCAs flows were noted.

Kawagishi et al.⁽¹⁵⁾ confirmed the decrease in PSV and EDV in the CRA. Moreover, changes in the CRA were observed before clinical features of retinopathy appeared. The RI was elevated and correlated with increased blood glucose level.

Mendivil et al.⁽¹⁶⁾ confirmed a significant decrease in PSV and EDV in the CRA and OA in patients with proliferative retinopathy compared to healthy individuals. No velocity disturbances were observed in the PCAs. The same author⁽¹⁷⁾ also studied the influence of laser panphotocoagulation on blood flow velocity and demonstrated that retinal photocoagulation caused a significant decrease in the velocity in the OA, CRA and CRV. These values did not change in a significant way within a year of follow-up.

Evans et al.⁽¹⁸⁾ conducted their studies in hyperoxic conditions in healthy individuals and in diabetic patients without or with minimal changes in the ocular fundus. Both groups were examined during regular breathing and subsequently, after oxygen saturation. In hyperoxic conditions, the end-diastolic velocity and resistance index (RI) changed in a significant way in healthy individuals – EDV decreased and RI increased. By contrast, the diabetic patients examined under hyperoxic conditions demonstrated increased EDV in the CRA and decreased RI compared to the control group. The healthy subjects showed no statistically significant changes in the OA.

Arai et al.⁽¹⁹⁾ emphasized t he role of t he R I i ndex i n t he OA and CRA. According to the authors, it is a more sensitive parameter for predicting the intensity of diabetic retinopathy than flow velocity measurements in these vessels. Greater vascular resistance may occur prior to retinopathy. The authors claim that measurements in the CRA are more relevant for hemodynamic alteration assessment in diabetes than measurements in the OA. The PSV and EDV values in the CRA were considerably lower both when diabetic lesions were found in the fundus and when there were no such lesions. The differences observed depended on the intensiveness of these lesions. Perhaps the CRA is more sensitive to autoregulation disorders which may develop even before any changes are observed in the fundus. No differences in PSV in the OA were noted between the groups.

MacKinnon et al.⁽²⁰⁾ assessed velocities in the OA and CRA as well as the RI index in three groups of patients: without diabetic lesions in the ocular fundus or with simple retinopathy, with preproliferative or proliferative lesions and in healthy controls. In comparison with the control group, the RI index was considerably higher in the OA and lower in the CRA. In the group of patients with preproliferative and proliferative retinopathy, PSV and EDV in the CRA were significantly lower than in patients without lesions or with simple retinopathy. In the OA, however, no difference in velocities was noted between the examined groups.

Ino-ue et al.⁽²¹⁾ analyzed changes in blood flow in the OA in persons with ocular ischemic syndrome and with simple as well as proliferative retinopathy. The patients with diabetic lesions presented lower EDV values and higher PI values as compared with the control group, and patients with ocular ischemic syndrome and diabetes presented significantly lower systolic and diastolic velocities and higher pulsatility indices. No significant differences between the groups were observed and no relation was found between the vascular flow velocity and progression of diabetic retinopathy.

Dimitrova et al.⁽²²⁾ compared the results of flow measurements in patients with simple retinopathy following 21 months from the primary examination and assessed the influence of the disease progression on hemodynamic changes. After this period of time, no changes in the flow parameters were observed in the CRA and ciliary arteries. The authors suggest that together with the progression of diabetic retinopathy, initial lesions develop in the CRV.

Krepler et al.⁽²³⁾ assessed the influence of vitrectomy conducted in patients with diabetic retinopathy on the ocular perfusion. After the surgery, the mean flow velocity in the CRA and PCAs decreased in a considerable way, and the RI value increased in the CRA and remained the same in the PCAs. The results indicate that the vitrectomy procedure may cause a decrease in the ocular blood flow.

Sullu et al.⁽²⁴⁾ conducted similar studies and assessed hemodynamics of the vessels in the eyeball in the first and sixth months after the procedure. A considerable decrease of the RI index in the CRA was observed after the procedure.

Gracner⁽²⁵⁾ assessed the parameters in the OA, CRA and PCAs on various stages of diabetic lesion development and demonstrated a significant increase in PSV in the OA with advanced preproliferative and proliferative lesions as well as a decrease in PSV and EDV in the CRA as compared to the control group. The author also noted a decrease in EDV in the PCAs. The RI index in the OA and CRA did not change in a significant way, but increased in the PCAs. The author suggested that the results may be useful in predicting the progression of the disease.

Kraśnicki et al.⁽²⁶⁾ assessed flow velocities in the vessels of the eyeball in patients with type II diabetes in a group without lesions in the ocular fundus and in a group with the presence of non-proliferative lesions. The authors observed considerably lower PSV and EDV values in the OA in all patients compared to healthy controls. The RI index did not change in a statistically significant way. In the CRA, PSV and EDV were considerably lower only in the group with more intensive diabetic lesions. The RI index was significantly higher in patients with more advanced lesions. In the PCAs, the significant difference was observed only in the PSV in the group of patients with non-proliferative retinopathy. The authors emphasized the role of Doppler sonography and the influence of decreased flow parameters in the CRA and ciliary arteries on the development of retinopathy.

When assessing the vessels of the eyeball in pediatric patients with type I diabetes but without retinopathy in the fundus, Yilmaz Ovali et al.⁽²⁷⁾ detected alterations in the blood flow in the OA and CRA as compared with healthy peers. EDV in the OA was significantly higher and the RI index was lower in patients with diabetes lasting for more than 5 years. The RI index in the CRA increased in patients who presented higher levels of microalbuminuria. Moreover, the examination revealed early lesions in the OA that were not as significant as reported by other researchers^{(15, 19, 20)}.

In the group of young patients with type I diabetes, Modrzejewska et al.⁽²⁸⁾ assessed hemodynamic parameters of the vessels in the eyeball and their relation to the lipid level in blood. The subjects had initial vascular lesions in the ocular fundus. The analysis revealed a decrease in blood flow parameters which occurred when the vascular lesions in the ocular fundus were at their initial stage. The following values were decreased: PSV, MV and RI in the CRA as well as PSV, MV and PI in the temporal posterior ciliary arteries (TPCAs). The variations in total cholesterol level (HDL), LDL cholesterol and apolipoproteins (ApoB) were associated with vascular resistance alterations, mainly in the OA. The results suggest that the influence of lipid disorders on the risk of retinal ischemia in young patients with type I diabetes may be significant.

Dimitrova et al.⁽²⁹⁾ analyzed the relation between diabetic retinopathy and arterial hypertension on the basis of flow analyses in the central retinal vein and artery in four groups of patients. PSV and EDV velocities were the highest in the group of patients without retinopathy and hypertension, and in patients without retinopathy but with arterial hypertension. The Pourcelot index in the CRV was considerably lower in both groups. A significant relation was observed between arterial hypertension, stadium of diabetic retinopathy and PSV in the CRA.

Lockhart et al.⁽³⁰⁾ used novel techniques of spectral analysis for a quantitative assessment of blood flow changes in the vessels of the eyeball and carotid arteries which were evaluated with the use of color Doppler techniques. The aim of the study was to answer the question whether it is possible to detect subclinical circulation disorders in patients with uncomplicated type I diabetes. Having applied these methods, the authors identified measurable differences in flow velocities prior to the development of explicit retinopathy.

Finally, Pemp et al.⁽³¹⁾ analyzed the relation between blood flow in the CRA measured in a color Doppler examination and blood flow in the CRA assessed with the use of a laser Doppler flowmeter with respect to the caliber of the vessel. The subjects included patients with type I diabetes with mild retinopathy and without it, and were compared with healthy individuals. The patients with diabetes presented greater diameter of the CRA, but no significant differences in blood flow were observed. The authors recommend caution in interpreting results since assessing the vessels of the eyeball with the use of color Doppler sonography does not provide information on their diameters.

Conclusions

Diabetes is still a diagnostic and therapeutic challenge in various aspects. Doppler sonography is a non-invasive method which may be a valuable supplementation in assessing hemodynamic alterations in the retrobulbar vessels in diabetic retinopathy. However, the results of analyses concerning flow parameters in these vessels vary. At present, it is difficult to unambiguously assess the relation between blood flow in the ophthalmic, central retinal and posterior ciliary arteries with respect to the stadium of the progression of diabetic lesions in the ocular fundus. This inconsistency may result from assessing pathological lesions in non-uniform study groups – in terms of the type of diabetes, its duration, stadium of retinopathy and local factors. It needs to be emphasized that some authors found decreased velocities in the vessels of the eyeball, particularly in the central retinal artery, in the initial stage of the disease, whereas another group of authors demonstrated increased flow velocities and resistance indices, but observed a decrease in these values when diabetes lasted for a longer period of time. This might be associated with the severity of retinopathy. Since numerous publications on retrobulbar circulation in patients with diabetes report discrepant results, it is necessary to conduct further studies and verify them. The conclusions might create a better perspective in predicting the progression of diabetes, in indicating directions of treatment and perhaps in preventing vascular complications that occur in the course of this common disease.

Conflict of interest

Authors do not report any financial or personal links with other persons or organizations, which might affect negatively the content of this publication and/or claim authorship rights to this publication.