Introduction 

Retinal vascular occlusions are among the most common causes of severe vision loss, with retinal vein occlusion (RVO) being the second most common retinal vascular disorder after diabetic retinopathy.[1],[2] They occur mainly in older patients and are more prevalent in men than women.[3],[4],[5] In 2015, the global prevalence of any RVO, branch RVO (BRVO), and central RVO (CRVO) in people aged 30–89 years was 0.77%, 0.64%, and 0.13%, translating to a total of 28.06 million, 23.38 million and 4.67 million affected individuals, respectively.[6] RVO is classified according to anatomic location and degree of retinal ischemia.[7] Based on anatomic location, there are two primary categories: BRVO and CRVO.[8] Hemi-CRVO (HCRVO) is a terminology used for a particular variant of CRVO where only one of the two trunks of the central retinal vein within the optic nerve is occluded.[9] CRVO and HCRVO can be further divided into nonischemic and ischemic types based on fluorescein angiography (FA) findings. Nonischemic HCRVO and CRVO appear as ischemic zones that are < 10 disc areas on FA, while the ischemic variants have ischemic zones of more than ten disc areas of the retinal surface.[8] BRVO is divided into two different types: major BRVO, when one of the major branch retinal veins is occluded, and macular BRVO, when one of the macular venules is occluded, and these two types have different fundus changes.[7],[8] CRVO is four times less common than BRVO, but patients with CRVO have a worse prognosis.[8]

Retinal artery occlusion (RAO) is a rare event.[10] It consists of two major types: central RAO (CRAO) and BRAO and each one is made up of multiple clinical entities. CRAO consists of four distinct clinical entities – non-arteritic CRAO, nonarteritic CRAO with cilioretinal artery sparing, arteritic CRAO associated with giant cell arteritis (GCA) and transient nonarteritic CRAO. Similarly, BRAO comprises permanent BRAO, transient BRAO and cilioretinal artery occlusion (CLRAO), and the latter further consists of three distinct clinical entities – nonarteritic CLRAO alone, nonarteritic CLRAO associated with CRVO, and arteritic CLRAO associated with GCA.[11] The incidence of CRAO is approximately 1 in 100,000 outpatient visits, with <2% presenting with bilateral involvement.[12]

There have been several studies on retinal vascular occlusions in sub-Saharan Africa (SSA), but the majority have been hospital-based retrospective studies on RVO.[13],[14],[15],[16],[17],[18] Only one study considered the prevalence and pattern of both RVO and RAO.[18] The purpose of this study, therefore, was to determine the prevalence and pattern of retinal vascular occlusions and the associated systemic risk factors among patients attending the retina subspecialty clinics and outpatient general ophthalmology clinics in four hospitals in SSA. A larger study such as this will provide more information on the prevalence and pattern of the disease in the region.

   Materials and Methods 

This was a multicenter, hospital-based, cross-sectional study carried out from January to December 2018 in four general ophthalmology and retina clinics in Nigeria. The clinics were situated in three public hospitals and one private eye hospital and located in different regions of Nigeria, with one each in the north-central and south-west regions, and two in the south-south region. There was a retina specialist in each clinic who examined the patients and ensured the accuracy and timely reporting of data.

Informed consent was obtained from each patient in the clinic before enrolment in the study. The patient's biodata were then obtained, and a history of any systemic disorder documented. Each patient seen in the retina clinic underwent a comprehensive eye examination, including Snellen visual acuity (VA) testing with refraction, anterior segment examination using a slit lamp, intraocular pressure measurement, dilated fundus biomicroscopy using a +90D or +78D lens, and binocular indirect ophthalmoscopy (BIO). Categorization of VA was done using the International Classification of Diseases 11 (2018) categorization of presenting distance vision impairment as indicated as follows: normal vision: ≥6/12, mild visual impairment <6/12–6/18, moderate visual impairment: <6/18–6/60, severe visual impairment: <6/60–3/60, and blindness: <3/60 to no perception of light.[19]

Diagnosis of retinal vascular occlusion was made from clinical examination. However, additional ocular investigations, including optical coherence tomography, fundus photography, and fundus FA were performed in centers where these were available. A BRVO was characterized by the presence of a localized area of retinal thickening or hemorrhages along retinal vessels with abnormal caliber or a congested vein at an arteriovenous crossing site. An old BRVO was defined by any combination of the following characteristics: a localized area of the sheathing of the retinal blood vessels, and arteriovenous collaterals. A CRVO was defined by hyperemia of the optic nerve head with scattered superficial and deep retinal hemorrhages in the four quadrants, retinal edema, and venous dilatation. Old occlusions were detected through the presence of arteriovenous collaterals, neovascularization at the optic disc, opticociliary shunts, or pan-retinal photocoagulation without signs of diabetic retinopathy in the other eye. The criteria for detection of HRVO were the same as for BRVO except that the area involved comprised either the upper or the lower half of the retina.[13],[20] CRAO was defined by retinal opacification, especially of the posterior pole with or without a cherry red spot, while BRAO was identified by retinal opacification in the distribution of the affected vessel only.[21]

Ethical approval was obtained from the Ethical Committee of the Jos University Teaching Hospital, and the tenets of the Helsinki Declaration were adhered to. Data from each of the four collaborating clinics were entered into an excel spreadsheet and transmitted at the end of each month to a central data collection point, where collation and analysis were done with the Statistical Package for the Social Sciences (SPSS) software version 22.0 (IBM Corp. Armonk, NY, USA). Frequencies, means, and standard deviations (SDs) were determined, and tests of significance were done using Pearson's Chi-squared test. A value of P < 0.05 was considered statistically significant.

   Results 

A total of 8614 new patients were seen over the 1-year study period. A retinal diagnosis was made in 876 (10.1%) patients comprising 455 (51.9%) males and 421 (48.1%) females. Ninety eyes of 81 patients had retinal vascular occlusion giving a disease prevalence of 0.9%. These patients were made up of 43 (53.1%) males and 38 (46.9%) females. There was no significant association between gender and retinal vascular occlusion (P = 0.084). Eighty-one eyes of 72 (88.9%) patients had RVO, while 9 eyes of 9 (11.1%) patients had RAO. Nine (12.5%) patients with RVO had bilateral involvement. The overall prevalence for RVO was, therefore, 0.8% and for RAO 0.1%. CRVO accounted for the majority of RVOs, 47 (58%), while CRAO was the most common type of RAO occurring in 7 (77.8%) eyes [Figure 1]. The overall prevalence for the various vascular occlusions were CRVO 0.55%, BRVO 0.33%, HRVO 0.05%, CRAO 0.08%, and BRAO 0.02%. As a proportion of retinal diseases, retinal vascular occlusion constituted 9.2%, with RVO accounting for 8.2% and RAO for 1%.

The mean age of patients with RVO was 59.5 years (range: 30–85 years, SD 12.4), while patients with RAO had a mean age of 52.4 years (range; 31–60 years, SD 9.0). [Table 1] shows the age distribution of the patients. Age distribution was statistically significant in the diagnosis of retinal vascular occlusion, with the majority of patients with RAO (77.8%) and RVO (57%) being between 50 and 69 years of age compared with 44.8% for patients with the nonretinal vascular occlusive disease (P < 0.0001).

Patients with RVO comprised 39 (54.2%) males and 33 (45.8%) females, while those with RAO were made up of 4 (44.4%) males and 5 (55.6%) females. There were 41 (50.6%) left eyes and 40 (49.4%) right eyes for RVO and 4 (44.4%) right eyes and 5 (55.6%) left eyes for RAO.

Sixty-one (75.3%) patients had at least one systemic condition [Table 2] and [Table 3]. The most common associated systemic conditions were hypertension and diabetes mellitus (DM). While 67.9% of patients with retinal vascular occlusions had either hypertension or DM or both, only 46.3% of patients with other nonretinal vascular occlusive diseases had these systemic conditions. This difference was statistically significant using the Chi-squared test with a P < 0.0001.

Click here to view

Click here to view

The mean duration of symptoms was 18.5 months (range 1–192, SD 36) for patients with RVO and 6 months (range 1–46, SD 15) for patients with RAO. Fifty-one eyes (56.7%) were blind (VA <3/60), while only 8 (8.9%) eyes had normal vision (VA ≥6/12) on presentation [Table 4]. Eyes with CRAO had the highest proportion of blind eyes, 6 (85.7%), followed by CRVO, 31 (66%).

   Discussion 

The majority of studies on retinal vascular occlusions in SSA have been on only RVOs, so this current study, being a multicenter, cross-sectional study with a larger sample size, provides vital information on both the prevalence and pattern of RVOs and RAOs in the region. The prevalence of retinal vascular occlusions of 0.9% in our study is slightly lower than the 1.4% reported by Adenuga et al., in a similar study in Jos, north-central Nigeria.[18] As a proportion of retinal diseases, however, retinal vascular occlusions constituted 9.2% in this study. This is at variance with lower figures of 4.3% and 1.7% reported in earlier studies in the southern part of Nigeria.[22],[23],[24] Our results therefore suggest that retinal vascular occlusions are an increasing cause of retinal disease in the region.

The overall prevalence of 0.8% for RVO in this study contrasts with 0.03% and 1.1% reported in Cotonou, the Republic of Benin, and north-central Nigeria, respectively.[17],[18] The much lower prevalence obtained in Cotonou is probably as a result of the larger sample size of the study, which included all patients seen over the 2½ year study period. The sample population in this study was made up of only new patients. For RAO, the prevalence of 0.1% compares favorably with 0.2% reported in north-central Nigeria.[18] Population-based studies on RVO or RAO from SSA are lacking; however, the prevalence of RVO in this study is within 0.4% and 1.1% reported in previous population-based studies in Europe, Asia, and America.[20],[25],[26] The study by Cheung et al. in America also showed no major racial/ethnic differences in the prevalence of RVO among the various ethnic groups, which included African Americans, Caucasians, Chinese, and Hispanics.[26] The authors concluded that previously observed variations in prevalence estimates were likely related to other differences between these studies, including study design, participant characteristics, and study methodology. We, therefore, believe that with the large sample size of our study, our findings may be representative of the prevalence of retinal vascular occlusions in the general population in SSA. Most population-based studies on RAO are on the incidence of the disease. In a predominantly white population in America, the annual age-and sex-adjusted annual incidence for CRAO was 1.90/100,000, while in Korea, the incidence rate for CRAO among men and women was 2.15 and 1.47/100,000 person-years, respectively.[4],[5] A recent study in Germany reported higher values with the standardized incidence rate for CRAO as 2.7/100,000 person-years and other RAO as 4.5/100,000 person-years.[27]

CRVO was the most common type of RVO in this study which is consistent with previous hospital-based studies in the region.[13],[14],[15],[16],[17],[18] Two hospital-based studies in Asia also reported similar findings.[28],[29] Population-based studies in Europe, America, and Asia, however, reported BRVO as the most common type of RVO.[20],[25],[26] It has been suggested that this difference could be because patients with CRVO, having more profound visual symptoms are more likely to attend the eye clinic ahead of those with BRVO, hence the higher proportion of CRVO in hospital-based studies in Africa.[30] Several hospital-based studies in Asia, however, also reported BRVO as being the most common type of RVO.[31],[32],[33] This, therefore, suggests a different pattern of presentation of RVO in SSA, and other factors such as genetic makeup, environment, and diet may be responsible for this.

In agreement with previous reports, both RVO and RAO were associated with advancing age in this study.[4],[5],[20],[25],[33] Functional and structural changes in the retinal vessels as a function of aging contribute to both RAO and RVO pathogenesis.[25] The mean age of patients with RVO in this study (59.5 years) compares favorably with most studies in Nigeria and the Republic of Benin, with a range of 54–58 years.[13],[15],[16],[17],[18] Slightly higher mean ages of 63 years were reported in Korea and Germany.[20],[25] In contrast to our finding, however, a retrospective study on the characteristics of CRVO in African Americans reported a much higher mean age of 68 years.[34] Similarly, for RAO, the mean age of 52.4 years in our study contrasts with higher mean ages of 74 years in America, 61 years in South Korea, and 66 years in Europe.[4],[5],[35] Despite this wide variation which may be as a result of the few cases of RAO in this current report, our findings suggest that retinal vascular occlusions tend to occur at an earlier age in SSA. Additional studies will, however, be necessary in order to determine the incidence of RAO as well as the demographic and clinical profile of patients with the condition in the region.

The retinal vascular occlusive disease is associated with significant systemic pathologies. Our results confirm the known association of hypertension and DM with both RVO and RAO, consistent with previous studies.[4],[25],[33],[36] Hypertension which was the most common systemic association in this report, is the most common vascular risk factor for both RVO and RAO.[6],[35] Hypertension and DM are associated with an increased risk of systemic arteriosclerosis, which is generally considered the main pathophysiological component of developing RVO.[3] Arterial occlusive disease of the retina is the result of either arteriosclerotic thrombosis, vasculitis, embolic impaction, vasospasm, or systemic hypotension.[21] Over 75% of patients with CRAO suffer from a generalized atheromatous disease, which is frequently associated with hypertension or DM, or both.[21] The severity of hypertension and DM have also been shown to have an impact on the risk of developing CRVO. Stem et al. showed that subjects with advanced hypertension had a 92% increased risk of developing CRVO compared to 36% for those with noncomplicated hypertension while subjects with end-organ damage from DM had a 53% increased risk for developing CRVO compared with no increased risk for those with no end-organ damage.[37] Hypertension in blacks tends to be more severe and develops at an earlier age.[38] The overall mortality due to hypertension as well as DM and its consequences, are also more likely in blacks.[38],[39] This may explain the occurrence of retinal vascular occlusions at an earlier age in our cohort of patients.

Most eyes with RAO in this study were blind, in agreement with the earlier report from north-central Nigeria.[18] Acute CRAO presents with a sudden, painless, and profound drop in vision, usually with an initial Snellen acuity of counting fingers or worse, unless a cilioretinal artery is present, in which case central vision may be preserved.[21] The proportion of blind eyes with RVO in our study falls within a wide range of 46%–79% from previous studies in Nigeria.[14],[16],[17],[18] This wide variation may be due to differences in the proportion of the types as well as the ischemic status of the RVOs seen. In agreement with these studies, CRVO, which carries the worst prognosis, accounted for the majority of blind eyes with RVO in this report. Macular involvement by ischemia or edema is usually responsible for visual loss in acute RVO.[8] We did not analyze the complications of retinal vascular occlusions in this report, as these will be addressed in another study.

Late presentation was a feature in our patients, and from earlier studies, this appears to be the pattern in patients with retinal vascular occlusions in SSA. In agreement with our findings, studies in the southern part of Nigeria observed late presentation in patients with RVO.[14],[15] The mean duration of symptoms in our patients with RVO (18.5 months) is similar to 21 months reported in Benin in south south Nigeria but higher than 7.2 months in southwest Nigeria.[14],[21] In contrast to the pattern in our region, patients with retinal vascular occlusive disease in developed countries tend to present to the hospital much earlier. In Korea, the mean duration of symptoms for patients with CRVO and BRVO was 29 days and 35 days, respectively.[33] In America,[4] patients with CRAO presented on average within 4 days of onset of symptoms in contrast to 6 months for patients with RAO in our study. This implies that the impact of retinal vascular occlusions on vision and quality of life is likely to be worse in SSA compared to other regions of the world. A recent study by Kiew et al., in North Carolina, also observed that African Americans with CRVO presented with more severe visual impairment and a greater treatment burden in the 1st year of follow-up compared with other races.[34]

Our study has some strengths, and these include the large sample size, its multicenter nature spread across different regions of Nigeria as well as the method of retinal examination. Carrying out a dilated fundus biomicroscopy and BIO for all the patients would have reduced the chances of missing vascular occlusions in the retinal periphery.

There are a few limitations to this study. Being a hospital-based study, the results may not apply to the general population. However, with the large sample size and comparable prevalence with other population-based studies, we expect that our findings would likely be representative of what obtains in the general population in SSA. Furthermore, the presence and visual impact of co-existing nonretinal diseases such as glaucoma and cataract, which could have a profound effect on vision, were not taken into consideration when analyzing the patients' VA.

   Conclusion 

Retinal vascular occlusions are an increasing cause of retinal disease in SSA, with the prevalence of RVO comparable to other parts of the world. There was a significant association with hypertension, DM and increasing age, though occurring at an earlier age in the region compared to other parts of the world. CRVO was the most common type seen; however, further studies will be required to establish the incidence of RAO as well as the demographic and clinical profile of patients with the disorder in SSA.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

   References