www.nature.com/eye
ARTICLE OPEN
Vitreomacular interface abnormalities in the Ghanaian African
Winfried M. Amoaku ]]]1✉, Laura Cushley2, Vittorio Silvestri3, Stephen Akafo4, Kwesi N. Amissah-Arthur4, Seth Lartey5, 
Courtney N. Hageman6, Christian M. Pappas6, William C. Hubbard6, Paul S. Bernstein ]]]6, Albert Vitale6, Megan Roberts7, Gianni Virgili3, 
Gregory S. Hageman6, Giuliana Silvestri7 and For the Ghana AMD Study Group*
© The Author(s) 2023
BACKGROUND/OBJECTIVE: Describe vitreomacular interface abnormalities (VMIA) using spectral-domain optical coherence 
tomography (SD-OCT), and correlations with age-related macular degeneration (AMD) grade in Ghanaian Africans.
SUBJECTS/METHODS: Prospective, cross-sectional study of adults aged ≥50 years recruited in Ghana AMD Study. Participant 
demographics, medical histories, ophthalmic examination, digital colour fundus photography (CFP) were obtained. High- 
resolution five-line raster OCT, Macular Cube 512 × 128 scans, and additional line scans in areas of clinical abnormality, were 
acquired. SD-OCT VMI features classified by International Vitreomacular Traction Study Group system and relationships to AMD 
grade were evaluated. Outcomes: VMIA prevalence, posterior vitreous detachment (PVD), vitreomacular adhesions (VMA), 
vitreomacular traction (VMT), epiretinal membranes (ERM), correlations with AMD grade.
RESULTS: The full Ghana AMD cohort included 718 participants; 624 participants (1248 eyes) aged ≥50 years (range = 50–101, 
mean = 68.8), 68.9% female were included in this analysis. CFP with OCT scans were available for 776 eyes (397 participants); 707 
(91.1%) had gradable CFP and OCT scans for both AMD and VMI grading forming the dataset for this report. PVD was absent in 504 
(71.3%); partial and complete PVD occurred in 16.7% and 12.0% respectively. PVD did not increase with age (p = 0.720). VMIA 
without traction and macular holes were observed in 12.2% of eyes; 87.8% had no abnormalities. VMIA was not significantly 
correlated with AMD grade (p = 0.819).
CONCLUSIONS: This provides the first assessment of VMIA in Ghanaian Africans. VMIA are common in Africans; PVD may be less 
common than in Caucasians. There was no significant association of AMD grade with VMIA.
Eye; https://doi.org/10.1038/s41433-023-02737-z
INTRODUCTION with different forms of epiretinal fibrocellular (myofibroblasts and 
Evidence to date suggests that age-related changes in the astrocytes) proliferation and collagen deposition. Chang et al. 
vitreous are likely to have significant impact on the pathogenesis have reported similar findings [19].
of macular diseases and function [1–8]. Adhesions between the Evaluation of the vitreoretinal interface (VMI) has been 
internal limiting membrane (ILM) and the vitreous become less significantly enhanced by the recent introduction of optical 
robust with age [7, 9, 10]. This, in addition to vitreous liquefaction coherence tomography (OCT) (especially high-resolution OCT), 
[5], results in posterior vitreous detachment (PVD) which usually into clinical practice [20–26]. It is non-invasive and considered a 
initiates at the macula, but may occur simultaneously at multiple valuable tool for evaluating the VMI, including vitreomacular 
sites [11–13]. PVD begins to occur at approximately 40 years of adhesions (VMA) or VMT detection and configuration analysis 
age in Caucasians, although the process may occur earlier in eyes [19]. An international systematic clinical grading system of VMI 
with high myopia, or other posterior segment diseases [1, 5]. It pathology based on OCT imaging was developed recently [2].
may develop over several years, rather than instantaneously as Several VMI studies in Caucasian populations exist and indicate 
one sudden event [12, 14]. It is further suggested that that variations in the strength of VMA lead to varying risks in PVD 
vitreomacular traction (VMT) results from partial PVD, with and potential complications [11, 12]. Similarly, the invariable 
peripheral separation in association with adherence in the macula association of VMT and epiretinal membranes (ERM) have been 
and optic nerve, of the vitreous [4, 6, 15]. VMT is thought to described [8, 27–29]. The Beaver Dam Eye Study (BDES) recently 
develop as a result of proliferation of fibroblasts on the surface of reported the population prevalence of VMI abnormalities (VMIA) 
persistent cortical vitreous attachments to the ILM [10, 16, 17]. in participants aged 63 years and older [29]. McKibbin et al. [30] 
This concept is supported by the clinicopathological study by reported on VMI grading on OCT in 25% of 8359 (2090) 
Gandorfer et al. [18] who reported two types of VMT syndrome participants with visual impairment (VI) (in at least one eye) in 
1Academic Ophthalmology, Mental Health & Clinical Neurosciences, University of Nottingham and University Hospitals, Nottingham, UK. 2Centre for Public Health, Queen’s 
University of Belfast, Belfast, UK. 3NICRN, Belfast Health & Social Care Trust, Belfast, UK. 4Unit of Ophthalmology, Department of Surgery, University of Ghana Medical School, 
Korle Bu, Accra, Ghana. 5Eye Unit, Eye Ear Nose and Throat Department, Komfo Anokye Teaching Hospital and Kwame Nkrumah University of Science and Technology, Kumasi, 
Ghana. 6Department of Ophthalmology & Visual Sciences, Moran Eye Center, Sharon Eccles Steele Center for Translational Medicine, University of Utah, Salt Lake City, UT, USA. 
7Ophthalmology Services, Eye and ENT Clinic, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA, UK. *A list of authors and their affiliations appears at the end of 
the paper. ✉email: wma@nottingham.ac.uk
Received: 24 March 2023 Revised: 5 September 2023 Accepted: 7 September 2023 
1234567890();,:
W.M. Amoaku et al.  
2
the UK Biobank study (aged 54–66 years), which included 374 SUBJECTS AND METHODS
(4.9%), 73 (4.4%), and 22 (5.3%) Black or Black British controls Study population
without VI, participants with VI without vitreoretinal interface Participants in this study were recruited as part of the larger ‘Ghana AMD 
abnormalities (VRIA), and those with VI and VRIA, respectively Study’, a prospective, cross-sectional study designed to investigate AMD 
[30]. in the Ghanaian population. The study was conducted in two teaching 
Other studies have implicated VMT in the causation or hospitals, Korle-Bu Hospital (KBTH), Accra, and Komfo Anokye Teaching 
prognosis of various diseases including age-related macular Hospital (KATH), Kumasi, during five different visits: March 2009 (Accra), April 2010 (Accra and Kumasi), February 2012 (Accra), April 2017 and 
degeneration (AMD) [31–37], and diabetic macular oedema January-February 2020 (Accra). The study enroled and consented 
(DMO) [14, 38–42]. A high incidence of VMT has been noted in Ghanaians aged 50 years and older. The research adhered to the Tenets 
treated eyes after photodynamic therapy [43]. More recently, VMA of the Declaration of Helsinki and the protocol was approved by the 
was suggested to be a prognostic factor in the visual outcomes institutional Ethics Review Boards of both KBTH (University of Ghana 
following intravitreal injections of anti-VEGFs [44–47]. Krebs et al. Medical School Protocol ID No: MS-Et/M.4-P./2008/2009) and KATH 
[27] reported that the spectral-domain OCT showed a high (Reference: KNUST CHRPE/32/10). Neither the study individuals nor the 
prevalence of type III neovascular AMD lesions in eyes with VMA public were involved in the design, conduct, reporting, or dissemination 
in the macula. A subsequent study by Waldstein et al. [48], plans of this study.
however, reported that VMAs do not influence the development 
of exudative AMD, or time to progression. Jackson et al. [49] Ghana AMD study
undertook a meta-analysis of vitreous attachments in eyes with Study details were explained to all study subjects with the aid of interpreters; participant information leaflets were provided and written 
AMD, DMO and RVO, and reported that eyes with nAMD were consent obtained. All patients enroled were of Ghanaian ethnicity. A 
twice as likely to develop VMA, and less likely to have PVD. questionnaire detailing the participant’s demographic details, ophthalmic 
However, Maggio et al. [6] concluded from their study, which and medical histories, was completed by local ophthalmic nursing staff. 
included 1067 eyes that there was no significant difference in Best-corrected Snellen visual acuity with glasses or pinhole was assessed, 
VMA prevalence in AMD eyes compared with age-matched intraocular pressures were measured with a Goldmann applanation 
controls, and further, that there was no difference in developing tonometry, and the pupils were dilated using tropicamide ophthalmic 1% 
de novo choroidal neovascularisation (CNV) in eyes with or and phenylephrine ophthalmic 2.5% solutions. Refraction was not 
without VMA. They further suggested that VMA may be a assessed. Each patient had slit lamp biomicroscopy and fundus 
consequence of CNV development. Other studies have described examination by a retinal specialist. Digital colour fundus photographs were captured using the Canon CR-DGI Non-Mydriatic Camera. Fields F1 
VMA abnormalities in eyes with DMO undergoing laser photo- (optic disc centred) and F2 (macula-centred) were captured in stereo. 
coagulation [50]. Peripheral images of any abnormalities noted on fundus examination 
More recently, the Handan study reported on the prevalence of were captured when necessary. A subset of patients was also imaged 
ERM in a rural Chinese adult population [51], and the using a Zeiss Cirrus HD-OCT, Model 400 (on the second and third visits) or 
MONTRACHET study reported a high prevalence of VMIA in Topcon 3D-OCT 2000 on the fourth study trip as the Cirrus was no longer 
French (Caucasian) participants aged 75 years and older [52]. functional. Retinal images were graded (AV, GS, PSB, WA) and the 
Similarly, a prevalence of 15.9% VMI disorders in a south participant’s eyes were classified according to their retinal status. AMD 
Netherlands (mainly Caucasian) population of 40–75 years olds, grading was independently performed by three masked investigators with 
which was related to age, sex and glucose metabolic status was retinal expertise (GS, GSH and WMA) based upon the Rotterdam AMD classification [56], as previously described in our manuscript on West 
reported in the Maastricht study [53]. African Crystalline Maculopathy [57]. When there was a disagreement, 
Previous reference to racial differences in the vitreoretinal panel adjudication involving all three investigators was used to derive 
interface (VRI) was made by Av-Shalom [54] who suggested that consensus. The presence of cataract and glaucoma were recorded in the 
degenerative vitreous changes were rare in Africans compared to grading information but was not used for classification for the purpose of 
Caucasians. To the best of our knowledge, only one report exists this report.
on VMI changes and potential associations with AMD in the SD-OCT imaging was available for the second, third and fourth study 
north-eastern African population, from Egypt [3], which found no trips, which provided the opportunity to assess OCT features of normal 
association between VMA and AMD [3]. The systematic review by and diseased eyes in this population. As such, participants recruited on 
Xiao et al. [55] confirmed the lack of studies on the VMI from the first and fifth trips were excluded from this analysis. The Cirrus HD- OCT 400 scanner was originally selected because of the advantage that 
Africa. As such, there are only limited data on the occurrence of the technique of obtaining images do not rely on the patient’s ability to 
VMT in the African. In particular, there are no studies on VMI from fixate, and that the data can be reviewed later and the position of the scan 
sub-Saharan Africa. The recent ultrastructural correlations of relocated to the centre of the fovea. This was important as we anticipated 
fibrocellular proliferations along the exposed inner retinal and that due to medial opacities in this population, scanning might be 
posterior vitreous surfaces correlating with pre-operative OCT difficult. The scan pattern selected was composed of high-resolution five- 
findings in eyes with VMT may have significant implications for line raster scans and the Macular Cube 512 × 128 scan. If any abnormality 
different populations [21]. As collagen formation and gliosis may was noted on clinical examination, additional scans were performed in the 
be different in the African compared to other races, an evaluation relevant areas. On the fourth trip, when the Cirrus was not functional, OCT 
of the vitreoretinal VRI in such populations may help to elucidate images were obtained with the Topcon 3D-OCT 2000. Macular thickness was measured using the Cirrus HD-OCT and the Topcon 3D-OCT 2000 
the role of the VMI in vitreoretinal disease pathogenesis. generated thickness maps using the Cirrus Macular Thickness Analysis and 
We hypothesized that VMIA occurs more frequently and is Topcon ImageNet tools.
more adhesive in the Ghanaian population in the age group ≥50 
years. We evaluated the VMI in all available OCTs of the macula OCT image grading
obtained from patients in a Ghanaian population as part of the All eyes with gradable scans were included in this study. Both eyes in any 
AMD study in Ghana (The Ghana AMD Study) which included individual, when available, were included as changes in the vitreoretinal 
participants with retinal diseases and controls. We investigated interface are not necessarily symmetrical in any single individual. In eyes 
any potential association of the VMI status with the AMD grade in where the OCT signal strength was reduced below four, or where fovea or 
each eye. optic disc were not visible, scans were considered ungradable and the 
The purpose of this report is to document changes at the VMI scans excluded from further analyses.
with spectral-domain (SD) OCT in this previously unreported The present report focuses on eyes with and without pathology with particular reference to eyes with AMD. Eyes that had gradable OCTs but 
Ghanaian African population and to investigate any association had confounding retinal vascular pathology in the macular area causing 
with AMD occurrence. disruption the inner retina, were excluded.
Eye
W.M. Amoaku et al.   
3
Table 1. Definitions of vitreomacular adhesion (VMA), traction (VMT) and macular hole (MH).
Classification Subclassification
Vitreomacular adhesion (VMA) Size: focal (<1500 mm) or broad (>1500 mm) Isolated or concurrent
Vitreomacular Traction (VMT) VMTFT Size: focal (<1500 mm) or VMTBT broad (>1500 mm); Isolated or concurrent
Full-thickness macular hole (FTMH) Size: small (<250 mm), medium (>250 to <400 mm), or large (>400 mm) 
Status of vitreous: with or without VMT 
Cause: primary or secondary
The International Vitreomacular Traction Study Classification System for Vitreomacular Adhesion, Traction, and Macular Hole [2].
VMTFT focal vitreomacular traction, VMTBT broad vitreomacular traction.
The VMI was assessed by a certified retinal grader (LNC) with and correlation using mixed models (linear or logistic as appropriate), with 
adjudicated by retinal experts (GS, GSH, WMA). OCT scans were assessed individuals as random effects. The software Stata 17.0 (StataCorp, College 
at different sittings using all the available OCTs as described by Ghazi et al. Station, TX) was used to calculate OR and 95% CIs.
[38], and the International Vitreomacular Traction Study Group (IVTS) 
classification as summarized in Table 1 [2]. A panel adjudication involving 
all three retinal experts was used to derive consensus where there was RESULTS
disagreement. VMA was defined as cortical vitreous detachment in the Total AMD study cohort demographics
perifoveal region, but with attachment within the 3 mm radius of the 
fovea (i.e. within the 3 ETDRS circles) without alteration within the foveal A total of 1436 eyes of 718 participants were examined during the 
contour (or underlying retinal tissue). If attachment of the hyaloid first four visits for the total cohort. The demographics of the total 
membrane over the fovea/macula was at a shallow angle, i.e. without a cohort were as follows: age range was 22–101 years, 65.9% were 
steep angle of attachment or focal distortions of the retina, VMA was female and 32.9% male, mean age was 65.0 years. The mean age 
further subdivided according to the width of attachment, as to whether of the females was 67.1 years (range 22–101), and of the males 
the attachment was narrow/focal based or broad based. 69.9 years (range 50–100).
VMT was diagnosed when there was evidence of perifoveal vitreous Of the total participants recruited, 624 participants (1248 eyes) 
detachment from the retinal surface within the 3 mm radius of the fovea, were aged over 50 years and were included in the analyses. 
where there were structural change in the fovea and a steep slope of the Participants aged 50 years and older had a mean age of 68.79 ± 
posterior hyaloid membrane to the inner macular surface. In addition, 
VMT was present when there was a sharp angulation of the hyaloid, and/ 10.21 (range 50–101) years and included 388 male eyes (31.1%) and 
or localized deformation of the retinal profile detectable at the adhesion 860 female eyes (68.9%). Colour fundus images were available on 
site of the posterior hyaloid membrane. VMT was divided into focal 1414 of the total 1436 eyes (98.5%), 92.6% of which were gradable. 
(VMTFT), and broad vitreomacular traction (VMTBT). The presence of focal OCT scans were available for 891 of the total 1436 eyes (62.0%). 
horizontal attachments of the posterior hyaloid membrane to the macula OCT scans from 36 eyes were ungradable due to media opacity. 
in more than one location was described as tangential or horizontal VMT Gradable OCT scans were available for 855 (59.5%) of eyes in the 
(VMTHT). Defects in the inner fovea with preservation of the outer foveal total cohort. Colour images and OCT scans were available for 776 
layers including photoreceptors is referred to as lamellar macular hole, eyes of 496 participants. Of the 776 eyes, 707 (91.1%) had gradable 
whilst loss of all foveal retinal layers was graded as full-thickness macular images and OCT scans for both AMD and VMI. These formed the 
hole.
The presence of associated macular distortions or oedema, ERM (focal dataset analyzed for this part of the project.
thickening of the hyaloid membrane on the surface of the retina) which 
manifested as focal hyper-reflective signal on the inner surface of the VMI analysis—cohort demographics
retina, or other associated findings were noted. Complete separation of The VMI dataset included 776 eyes, 505 female eyes (65.1%) and 
the hyaloid membrane from the macular area indicated by a clear visible 271 male eyes (34.9%). The mean age of the females in the VMI 
posterior vitreous cortex, which moved freely in the vitreous cavity was analysis cohort was 67.1 years (range 50–100), and of the males 
considered as a complete PVD. Care was taken to distinguishing this from 69.9 years (range 50–100). Gradable images and OCT scans for 
simulations of complete PVD by the presence of a liquid-filled pre-retinal both AMD and VMI were available in 707 (91.1%) eyes. There were 
bursa, in the presence of persistence of a pre-macular vitreous cortex four eyes with non-proliferative diabetic retinopathy (NPDR), one 
[11, 16, 58]. When the pre-macular bursa is visible, but the vitreous cortex 
is not detected, then PVD was absent. Persistent attachments to the optic of which was excluded due to the presence of cystoid macular 
disc margins were also assessed in eyes that had PVD. Eyes with oedema (CMO); two of the gradable eyes had signs of previous 
retinal vascular disease were classified as ‘concurrent’ as per IVTS branch retinal vein occlusion (BRVO), but none had macular 
classification [2]. oedema or retinal disruption on OCT and so were included.
The results are reported in terms of the presence or absence of PVD and The demographic details (age–sex distribution) of patients who 
whether this was complete or partial; the presence and type of VMIAs had OCTs were not different from the total study population 
including adhesions. These changes were correlated with AMD grade or (p = 0.57). In the cohort of 707 fully gradable eyes, 621 (87.8%) 
co-existing macular disease. were found to have VMIA.
Data were entered into IBM SPSS Version 25 predictive analytics 
software. The data were cleaned and a random 10% data entry recheck 
was carried out. Posterior Vitreous Detachment
PVD (partial or full) in this population was observed in 203 (28.7%) 
of 707 eyes, of which 118 (16.7%) had partial PVD and 85 (12.0%) 
Statistical analyses had complete PVD. No PVD was observed in 504 eyes (71.3%). The 
Descriptive statistics were used to summarize the occurrence of VMIA and occurrence of PVD was 40.1% (24% partial and 16.1% complete) 
associated diagnoses. All eyes in participants aged 50 and over which at age >65 years and in 34.2% (18.4% partial and 15.8% complete) 
were both gradable for both VMIA (absent, present) and AMD (none, early, in those aged >80 years. The presence of PVD did not increase 
GA, neovascular AMD) were examined for relationship between VMIA and 
AMD, together with odds ratios (OR) (and 95% confidence intervals) with age (p = 0.720 for linear trend).
measuring risk of VMIA (+ve) for stages of AMD relative to no AMD. Surface retinal disruption was present in 5.2% of eyes and an 
It should be noted that most participants provided two eyes in the ERM was present in 13.2% of eyes. CMO was present in 3.3% of 
study and those two eyes were not independent observations. Therefore, eyes. In eyes with no evidence of AMD (Stage 0 AMD) PVD was 
we conducted statistical analyses accounting for within-subject observed in 29 of 62 (46.8%) eyes. PVD occurred in five of 19 
Eye 
W.M. Amoaku et al.  
4
Table 2. Relationship between vitreomacular interface abnormalities (VMIA) and age-related macular degeneration (AMD).
No VMIA VMIA (any) ORa 95% CI
No AMD 63 438 1.00
Early AMD (Grades 1–3) 13 123 1.49 0.56–3.98
AMD Grade 4a (Geographic Atrophy) 3 21 1.43 0.18–11.6
AMD Grade 4b (Neovascular AMD) 7 39 0.82 0.22–3.02
aAccounting for within-subject correlation in a logistic mixed model.
(26.3%) with neovascular AMD (Grade 4b) compared with 33 of 64 in 62.2%, and total PVD was observed in only 8.6% of eyes in a 
(48.43%) eyes with non-neovascular AMD. In eyes with geographic cohort of 31–74 years [mean 52.3 yrs]. The onset of PVD (in the 
atrophy (AMD Grade 4a), the incidence of PVD was 25%. The Uchino series) began in subjects of less than 50 years who 
findings are summarized in Table 2. The presence of PVD either developed asymptomatic PVD localized to the perifoveal area and 
partial or complete did not correlate with AMD grade (p = 0.93). increased with age. Kumagai et al. [65] evaluated the VMI with 
OCT in eyes aged 22–95 years with various vitreoretinal diseases 
Vitreomacular interface abnormalities and reported vitreoretinal adhesions in 15% of healthy eyes 
VMIA was observed in 86 (12.2%) of 707 fully gradable eyes. VMA (average age 66.4 ± 12.7 years), as compared to 9% in eyes with 
without traction (VMAWOT) was noted in 81 (11.4%) of eyes, AMD (average age 68.0 ± 12.3 years). They also reported that 
VMTHT in seven (0.99%) eyes, VMTFT in 10 (1.4%) eyes, all of which vitreoretinal traction occurred in 2% of healthy eyes compared to 
were in eyes without any evidence of AMD or had Grade 1 AMD 1% in eyes with AMD. These differences were statistically 
only. No eyes with neovascular AMD had associated VMTFT. VMTBT significant. Kumagai et al. [65], however, did not describe any 
was uncommon, occurring in only 1 (0.1%) of eyes. Macular holes variations of these changes with age.
were observed in 4 (0.6%) of eyes, none of which had AMD. Overall, In this study in the Ghanaian population, the 40.1% occurrence 
there was no significant correlation between the presence or of PVD at age >65 years and in 34.2% in those aged >80 years is 
absence of VMI abnormalities and AMD grade (p = 0.85). significantly less than that reported from previous studies with 
CMO occurred in 3.1% of eyes. The oedema was associated ultrasonography in other populations. Larsson and Osterlin [62] 
with venous occlusions, exudative AMD, diabetes and ERM. ERM reported a complete PVD rate of 30% in those aged 65 years and 
was noticeable in 50 (7.7%) of eyes. The retinal interface was older, which is significantly higher than the 16.1% (95% CI 
disrupted in 4.7% of eyes. 12.3–20.7) found in the Ghanaian population in this study. Hayreh 
and Jonas [63] reported a total PVD rate of 60% in those aged ≥65 
Relationship of VMI abnormalities to AMD years, again, significantly different from that in our study 
Table 2 shows the relationship between VMIA and AMD, together population. Perichon et al. [61] reported a PVD rate in those 
with odds ratios (and 95% confidence intervals) measuring risk of aged 80 years and over in 75% of individuals which contrasts with 
VMIA (+ve) for stages of AMD relative to no AMD. There was no a rate of 34.2% (95% CI 24.5–45.4) in the Ghanaian population. 
evidence of an association between VMIA and AMD grading Explained differently, approximately 60% of eyes of patients >65 
overall (p = 0.819), as well as for any AMD grade vs no AMD. years, and 65% of >80 years in our study did not have any 
The foveal architecture was comparable to those reported in detectable PVD on OCT.
Caucasians. However, a gentler slope to the foveal pit that The Beijing Eye study which included persons aged 50–93 years 
differed significantly in width, slope, and volume, was observed in old, reported the population prevalence of incomplete PVD was 
the absence of any VMI abnormalities in this Ghanaian cohort. 60.5% of eyes, and was associated with younger age, and male 
gender [66]. The Handan Eye Study reported a PVD prevalence of 
2.7% associated with age in both women more than men in the 
DISCUSSION population of 30 years and older (mean age = 52 years) [67]. The 
It is known that, despite biomicroscopically visible PVD, VMA may BDES recently reported the population frequency of VMI changes 
persist. Similarly, biomicroscopy may not identify partial PVD [29]. The BMES reported findings, similar to those in the BDES [59], 
[11, 12, 23, 27, 59]. OCT has recently superseded ultrasonography whilst the NICOLA Study which included younger participants 
in evaluating the VMI has reported the SD-OCT was of higher (mean age = 62 years) reported lower prevalences of VMA (22.6%), 
precision in the evaluation of vitreoretinal adhesions, particularly VMT (0.5%), ERM (7.6%), and macular holes (0.3%) [68]. Similarly, the 
in exudative AMD [12, 13, 19, 27, 37, 60, 61]. Larsson and Oesterlin UK Biobank study reported a 3.6% prevalence of VMI changes (in 
[62] reported a 30% rate of complete PVD in individuals of 65 55–66 year olds), with increasing age, female sex and Asian 
years or older, whilst Hayreh and Jonas [63] reported that PVD ethnicity as significant associations [30]. The study by Zapata et al. 
may be found in approximately two-thirds of this age group. [69] included participants aged 45 years and older and reported a 
Perichon et al. [61] showed that PVD was not detected in 25% of lower prevalence of vitreoretinal interface abnormalities [70].
patients at 80 years or older; this was supported by a later study The Maastricht study (2008), using OCT, reported that VMIA 
[37]. There have been several recent reports on OCT studies of the were associated with older age, female sex and glucose metabolic 
VRI in various retinal diseases including DMO, and diabetic status [53]. The MONTRACHET study reported a high prevalence 
traction macular changes [14, 23, 25, 39, 40]. Koizumi et al. [64] of VMIA in participants aged 75 years and older, including VMA 
reported on the evaluation of the VRI with regards to VMT and (17.7%), VMT (1.4%), lamellar macular holes (1.0%), full-thickness 
ERM using SD-OCT [64]. However, only one study has been macular holes (0.2%), macular pseudoholes (0.4%), epiretinal 
reported from a north-eastern African population (Egypt), and membranes (38.9%), and macular cysts (5.8%).70 VMA were 
only in persons with different types of AMD [3]. This study is the associated with male sex, and negatively correlated with age 
first report on the VMI from a sub-Saharan African population and prior cataract surgery in that study, whilst older age and 
with or without AMD. previous cataract surgery were associated with ERM [71].
In an OCT study conducted by Uchino et al. [13], no PVD was It has been suggested that vitreous remnants on the retinal 
detected in 29.2%, whilst partial PVD of different grades occurred surface following PVD may predispose to epiretinal gliosis 
Eye
W.M. Amoaku et al.   
5
[10, 23, 60], and macular holes [16, 72]. In addition, worsening of ● Evaluation of the vitreoretinal interface (VMI) has been 
proliferative diabetic retinopathy [60] and diabetic epiretinal gliosis significantly enhanced by the introduction of OCT.
[16, 72], which may induce macular oedema with further reduction ● There are no studies on VMI from a sub-Saharan Africa 
in vision have been postulated as due to PVD. Sebag & Balazs [68] population.
and Kumagai et al. [65] have shown that there was residual foveal 
deformation in eyes with macular PVD in patients with contralateral 
macular hole suggesting that pre-existing strong VMA precedes What this study adds 
macular PVD in these eyes [69]. Doi et al. [70] previously described 
firm adhesions of the vitreous in the central macula [73]. As there ● This study provides the first assessment of VMIA in the 
was a high rate of vitreoretinal interface abnormalities in our Ghanaian African.
cohort, it is possible that some of the visual deficit in our study ● It suggests that VMIA, including adhesions with or without 
population may be corrected by removal of persistent vitreous traction, are common in Africans and that PVD may be less 
attachments and or ERMs. If such VMA or remnants are confirmed common than in Caucasians.
as increasing risk for reduced vision, pharmacologic vitreolysis, ● There was no association of AMD grade with VMIA.
which is now on the horizon, will be useful as treatment.
In our study, it is significant that tangential traction in the 
foveomacular zone was found mainly in eyes with PVD. This 
finding supports the concept that incomplete PVD predisposes to DATA AVAILABILITY
VMT. It would therefore become more important that in patients The data is available from the corresponding author upon reasonable request.
with symptoms of visual impairment where no other clinical 
findings are obvious that the persistence of VMA in the posterior 
pole, despite PVD elsewhere, should be looked for [11, 58]. REFERENCES
Previous studies have suggested that VMA may be important in 1. Chuo JY, Lee TY, Hollands H, Morris AH, Reyes RC, Rossiter JD, et al. Risk factors 
the development of neovascular AMD [31–37, 45]. By implication, for posterior vitreous detachment: a case-control study. Am J Ophthalmol. 
eyes (or populations) which have less PVD or more VMA should 2006;142:931–7.
have a higher incidence/prevalence of neovascular AMD. In the 2. Duker JS, Kaiser PK, Binder S, de Smet MD, Gaudric A, Reichel E, et al. The 
present study, there is no association of VMA or VMT and International Vitreomacular Traction Study Group classification of vitreomacular 
adhesion, traction, and macular hole. Ophthalmology. 2013;120:2611–9.
neovascular AMD or geographic atrophy. The less frequent 3. El-Hifnawy MA, Ibrahim HA, Gomaa AR, Elmasry MA. The vitreomacular interface 
occurrence of neovascular AMD in this population compared to in different types of age-related macular degeneration. Int J Ophthalmol. 
Caucasian populations would imply that VMA and VMT as such 2017;10:246–53.
are not important in the development and/or progression of 4. Jaffe N. Vitreous traction at the posterior pole of the fundus due to alterations in 
advanced AMD. This supports the studies by Waldstein et al. [48] the vitreous posterior. Trans Am Acad Ophthalmol Otolaryngol. 1967;71:642–52.
and Maggio et al. [6] which concluded that VMA had no 5. Kakehashi A, Kado M, Akiba J, Hirokawa H. Variations of posterior vitreous 
significant influence on the development or progression of detachment. Br J Ophthalmol. 1997;81:527–32.
neovascular AMD. It has also been suggested that the presence 6. Maggio E, Polito A, Guerriero M, Prigione G, Parolini B, Pertile G. Vitreomacular 
of VMT may affect the outcomes of treatment of neovascular adhesion and the risk of neovascular age-related macular degeneration. Oph-
thalmology. 2017;124:657–66.
AMD. Schmidt et al. [43] and Edwards et al. [74] surmised that 7. Sebag J, Hageman GS. Interfaces. Eur J Ophthalmol. 2000;10:1–3.
VMT does not predispose to, but rather may be protective of 8. Smiddy W. Vitreomacular traction syndrome. In: Yanoff N DJ, editor. Ophthal-
AMD. They explained this by a referral bias in which patients with mology, 2nd ed. St Louis: Mosby; 2004. p. 951–5.
AMD and VMT are not referred for VMT surgery. 9. Sebag J. Age-related differences in the human vitreoretinal interface. Arch 
Differences in the literature regarding occurrence of VMIA are Ophthalmol. 1991;109:966–71.
most likely due to the varied definitions adopted in the different 10. Sebag J. Anatomy and pathology of the vitreo-retinal interface. Eye. 
studies. However, irrespective of such differences, this study 1992;6:541–52.
confirms the value of OCT in evaluating the VMI. 11. Johnson MW. Perifoveal vitreous detachment and its macular complications. 
This study provides the first report on the VMI in the Sub- Trans Am Ophthalmol Soc. 2005;103:537–67.
12. Johnson MW. Posterior vitreous detachment: evolution and complications of its 
Saharan African. The results suggest that PVD may be less early stages. Am J Ophthalmol. 2010;149:371–82.e1.
common than in the Caucasian, and that vitreoretinal adhesions 13. Uchino E, Uemura A, Ohba N. Initial stages of posterior vitreous detachment in 
are common in Africans, and may be responsible for visual loss in healthy eyes of older persons evaluated by optical coherence tomography. Arch 
this population. The results of this study do not support a role of Ophthalmol. 2001;119:1475–9.
VMIA in AMD. 14. Gaucher D, Tadayoni R, Erginay A, Haouchine B, Gaudric A, Massin P. Optical 
The main limitation of this study is that the subjects may be coherence tomography assessment of the vitreoretinal relationship in diabetic 
self-selected as it is a hospital based, rather than a true population macular edema. Am J Ophthalmol. 2005;139:807–13.
cohort (as we did not sample a representative population), and 15. Kishi S, Demaria C, Shimizu K. Vitreous cortex remnants at the fovea after 
correlations of OCT findings with ultrasonography were not spontaneous vitreous detachment. Int Ophthalmol. 1986;9:253–60.
16. Kishi S, Hagimura N, Shimizu K. The role of the premacular liquefied pocket and 
undertaken. Furthermore, it did not specifically obtain best- premacular vitreous cortex in idiopathic macular hole development. Am J 
corrected vision, but rather the best habitual vision. It is therefore Ophthalmol. 1996;122:622–8.
impossible to fully attribute visual symptoms to VMI abnormal- 17. Sebag J. Age-related changes in human vitreous structure. Graefes Arch Clin Exp 
ities. As this study was based in Ghana, no data was obtained Ophthalmol. 1987;225:89–93.
from Caucasians simultaneously under similar study conditions. 18. Gandorfer A, Rohleder M, Kampik A. Epiretinal pathology of vitreomacular 
traction syndrome. Br J Ophthalmol. 2002;86:902–9.
19. Chang LK, Fine HF, Spaide RF, Koizumi H, Grossniklaus HE. Ultrastructural cor-
SUMMARY relation of spectral-domain optical coherence tomograpic findings in vitreo-
macular traction syndrome. Am J Ophthalmol. 2008;146:121–7.
What is known before 20. Gaudric A, Haouchine B, Massin P, Paques M, Blain P, Erginay A. Macular hole 
formation - new data provided by optical coherence tomography. Arch Oph-
thalmol. 1999;117:744–51.
● Age-related changes in the vitreous are likely to have 21. Ito YTH, Terasaki H, Suzuki T, Kojima T, Mori M, Ishikawa K, et al. Mapping 
significant impact on the pathogenesis of macular diseases posterior vitreous detachment by optical coherent tomography in eyes with 
and function. idiopathic macular hole. Am J Ophthalmol. 2003;135:351–5.
Eye 
W.M. Amoaku et al.  
6
22. de Boer JF, Cense B, Park BH, Pierce MC, Tearney GJ, Bouma BE. Improved signal- 47. Wu P, Lai C, Chen C, Kuo C. Optical coherence tomographic patterns in diabetic 
to-noise ratio in spectral-domain compared with time-domain optical coherence macula edema can predict the effects of intravitreal bewvacizumab injection as 
tomography. Opt Lett. 2003;28:2067–9. primary treatment. J Ocul Pharmacol Ther. 2012;28:59–64.
23. Gallemore RP, Jumper JM, McCuen BW 2nd, Jaffe GJ, Postel EA, Toth CA. Diag- 48. Waldstein SM, Sponer U, Simader C, Sacu S, Schmidt-Erfurth U. Inflence of 
nosis of vitreoretinal adhesions in macular disease with optical coherence vitreomacular adhesion on the development of exudative age-related macular 
tomography. Retina. 2000;20:115–20. degeneration: 4-Year results of a longitudinal study. Retina. 2012;32:424–33.
24. Wilkins JR, Puliafito CA, Hee MR, Duker JS, Reichel E, Coker JG, et al. Character- 49. Jackson T, Nicod E, Angelis A, Grimaccia F, Prevost A, Simpson A, et al. 
ization of epiretinal membranes using optical coherence tomography. Oph- Vitreous attachment in age-related macular degeneration, diabetic macular 
thalmol Ophthalmol. 1996;103:2142–51. edema, and retinal vein occlusion: a systematic review and meta-analysis. Retina. 
25. Mirza RG, Johnson MW, Jampol LM. Optical coherence tomography use in 2013;33:1099–108.
evaluation of the vitreoretinal interface: a review. Surv Ophthalmol. 50. Khan MA, Haller JA. Clinical management of vitreomacular traction. Curr Opin 
2007;52:397–421. Ophthalmol. 2015;26:143–8.
26. Schmidt-Erfurth U, Leitgeb RA, Michels S, Povazay B, Sacu S, Hermann B, et al. 51. Duan XR, Liang YB, Friedman DS, Sun LP, Wei WB, Wang JJ, et al. Prevalence and 
Three-dimensional ultra-high-resolution optical coherence tomography of associations of epiretinal membranes in a rural Chinese adult population: the 
macular diseases. Invest Ophthalmol Vis Sci. 2005;46:3393–402. Handan Eye Study. Invest Ophthalmol Vis Sci. 2009;50:2018–23.
27. Krebs I, Glittenberg C, Zeiler F, Binder S. Spectral domain optical coherence 52. Creuzot-Garcher C, Binquet C, Daniel S, Bretillon L, Acar N, de Lazzer A, et al. The 
tomography for higher precision in the evaluation of vitreoretinal adhesions in Montrachet Study: study design, methodology and analysis of visual acuity and 
exudative age-related macular degeneration. Br J Ophthalmol. 2011;95:1415–8. refractive errors in an elderly population. Acta Ophthalmol. 2016;94:e90–7.
28. Chatziralli I, Theodossiadis G, Grigoropoulos V, Datseris I, Chatzirallis A, Theo- 53. Liesenborghs I, De Clerck EEB, Berendschot T, Goezinne F, Schram MT, Henry 
dossiadis P. Comparison of macular findings due to vitreomacular traction alone RMA, et al. Prevalence of optical coherence tomography detected vitreomacular 
or in association with epiretinal membrane. Eur J Ophthalmol. 2017;27:86–92. interface disorders: The Maastricht Study. Acta Ophthalmol. 2018;96:729–36.
29. Meuer SM, Myers CE, Klein BEK, Swift MK, Huang Y, Gangaputra S, et al. The 54. Av-Shalom A, Berson D, Landau L, Michaelson IC, Zauberman H. The vitreo- 
epidemiology of vitreoretinal interface abnormalities as detected by spectral- retinopathy associated with retinal detachment among Africans. Am J Oph-
domain optical coherence tomography. The Beaver Dam Eye Study. Ophthal- thalmol. 1967;4:387–91.
mology. 2015;122:787–95. 55. Xiao W, Chen X, Yan W, Zhu Z, He M. Prevalence and risk factors of epiretinal 
30. McKibbin M, Farragher T, Shickle D. Vitreoretinal interface abnormalities in middle- membranes: a systematic review and meta-analysis of population-based studies. 
aged adults with visual impairment in the UK Biobank study: prevalence, impact on BMJ Open. 2017;7:e014644.
visual acuity and associations. BMJ Open Ophthalmol. 2017;1:e000057. 56. Klaver CC, Assink JJ, van Leeuwen R, Wolfs RC, Vingerling JR, Stijnen T, et al. 
31. Krebs I, Brannath W, Glittenberg C, Zeiler F, Sebag J, Binder S. Posterior vitreo- Incidence and progression rates of age-related maculopathy: the Rotterdam 
macular adhesion: a potential risk factor for exudative age-related macular Study. Invest Ophthalmol Vis Sci. 2001;42:2237–41.
degeneration? Am J Ophthalmol. 2007;144:741–6. 57. Amoaku WM, Sampalli A, Silvestri V, Cushley LN, Akafo S, Amissah-Arthur KN, 
32. Lee SJ, Lee CS, Koh HJ. Posterior vitreomacular adhesion and risk of exudative et al. Characterization of West African crystalline macular dystrophy in the 
age-related macular degeneration: paired eye study. Am J Ophthalmol. Ghanaian population. Ophthalmol Retin. 2022;6:723–31.
2009;147:621–6.e1. 58. Spaide RF. Measurement of the posterior precortical vitreous pocket in fellow 
33. Mojana F, Cheng L, Bartsch DU, Silva GA, Kozak I, Nigam N, et al. The role of eyes with posterior vitreous detachment and macular holes. Retina. 
abnormal vitreomacular adhesion in age-related macular degeneration: spectral 2003;23:481–5.
optical coherence tomography and surgical results. Am J Ophthalmol. 59. Liew G, Nguyen H, Ho IV, White AJ, Burlutsky G, Gopinath B, et al. Prevalence of 
2008;146:218–27. vitreoretinal interface disorders in an Australian population: The Blue Mountains 
34. Ondes F, Yilmaz G, Acar MA, Unlu N, Kocaoglan H, Arsan AK. Role of the vitreous Eye Study. Opthalmol Sci. 2021;1:100019.
in age-related macular degeneration. Jpn J Ophthalmol. 2000;44:91–3. 60. Chu TG, Lopez PF, Cano MR, Freeman WR, Lean JS, Liggett PE, et al. Posterior 
35. Quaranta-El Maftouhi M, Mauget-Faysse M. Anomalous vitreoretinal adhesions in vitreoschisis. An echographic finding in proliferative diabetic retinopathy. Oph-
patients with exudative age-related macular degeneration: an OCT study. Eur J thalmology. 1996;103:315–22.
Ophthalmol. 2006;16:134–7. 61. Perichon JY, Brasseur G, Uzzan J. Etude e´chographique du de´collement poste- 
36. Robinson CD, Krebs I, Binder S, Barbazetto IA, Kotsolis AI, Yannuzzi LA, et al. ´rieur du vitre´ chez l’emme´trope [Ultrasonic study of posterior vitreous 
Vitreomacular adhesion in active and end-stage age-related macular degen- detachment in emmetropic eyes]. J Fr Ophtalmol. 1993;16:538–44.
eration. Am J Ophthalmol. 2009;148:79–82 e2. 62. Larsson L, Osterlin S. Posterior Vitreous Detachment - a combined clinical and 
37. Weber-Krause B, Eckardt U. Incidence of posterior vitreous detachment in eyes physicochemical study. Graef Arch Clin Exp. 1985;223:92–5.
with and without age-related macular degeneration. An ultrasonic study. Oph- 63. Hayreh SS, Jonas JB. Posterior vitreous detachment: clinical correlations. Oph-
thalmologe. 1996;93:660–5. thalmologica. 2004;218:333–43.
38. Ghazi NG, Ciralsky JB, Shah SM, Campochiaro PA, Haller JA. Optical coherence 64. Koizumi H, Spaide RF, Fisher YL, Freund KB, Klancnik JM, Yannuzzi LA. Three- 
tomography findings in persistent diabetic macular edema: The vitreomacular dimensional evaluation of vitreomacular traction and epiretinal membrane 
interface. Am J Ophthalmol. 2007;144:747–54. using spectral-domain optical coherence tomography. Am J Ophthalmol. 
39. Imai M, Iijima H, Hanada N. Optical coherence tomography of tractional macular 2008;145:509–17.
elevations in eyes with proliferative diabetic retinopathy. Am J Ophthalmol. 65. Kumagai K, Hangai M, Larson E, Ogino N. Vitreoretinal interface and foveal 
2001;132:458–61. deformation in asymptomatic fellow eyes of patients with unilateral macular 
40. Kang SW, Park CY, Ham DI. The correlation between fluorescein angiographic holes. Ophthalmology. 2011;118:1638–44.
and optical coherence tomographic features in clinically significant diabetic 66. Schumann RG, Schaumberger MM, Rohleder M, Haritoglou C, Kampik A, 
macular edema. Am J Ophthalmol. 2004;137:313–22. Gandorfer A. Ultrastructure of the vitreomacular interface in full-thickness idio-
41. Otani T, Kishi S, Maruyama Y. Patterns of diabetic macular edema with optical pathic macular holes: a consecutive analysis of 100 cases. Am J Ophthalmol. 
coherent tomography. Am J Ophthalmol. 1999;127:688–93. 2006;141:1112–9.
42. Panozzo G, Parolini B, Gusson E, Mercanti A, Pinackatt S, Bertoldo G, et al. Diabetic 67. Capeans C, Lorenzo J, Santos L, Suarez A, Copena MJ, Blanco MJ, et al. Com-
macular edema: an OCT-based classification. Semin Ophthalmol. 2004;19:13–20. parative study of incomplete posterior vitreous detachment as a risk factor for 
43. Schmidt JC, Mennel S, Horle S, Meyer CH. High incidence of vitreomacular proliferative vitreoretinopathy. Graef Arch Clin Exp. 1998;236:481–5.
traction in recurrent choroidal neovascularisation after repeated photodynamic 68. Sebag J, Balazs EA. Pathogenesis of Cystoid Macular Edema - an anatomic 
therapy. Br J Ophthalmol. 2006;90:1361–2. consideration of vitreoretinal adhesions. Surv Ophthalmol. 1984;28:493–8.
44. Ciulla TA, Ying GS, Maguire MG, Martin DF, Jaffe GJ, Grunwald JE, et al. Influence of 69. Zapata MA, Figueroa S, Esteban Gonzalez E, Huguet C, Giralt J, Gallego Pinazo R, 
the vitreomacular interface on treatment outcomes in the comparison of Age-related et al. Prevalence of vitreoretinal interface abnormalities on spectral-domain OCT 
Macular Degeneration Treatments Trials. Ophthalmology. 2015;122:1203–11. in healthy participants over 45 years of age. Ophthalmol Retina. 2017;1:249–54.
45. Lee SJ, Koh HJ. Effects of vitreomacular adhesion on anti-vascular endothelial 70. Doi N, Uemura A, Nakao K, Sakamoto T. Vitreomacular adhesion and the defect 
growth factor treatment for exudative age-related macular degeneration. Oph- in posterior vitreous cortex visualised by triamcinolone-assisted vitrectomy. 
thalmology. 2011;118:101–10. Retina. 2005;25:742–5.
46. Mayr-Sponer U, Waldstein SM, Kundi M, Ritter M, Golbaz I, Heiling U, et al. Influence 71. Shao L, Xu L, You QS, Wang YX, Chen CX, Yang H, et al. Prevalence and asso-
of the vitreomacular interface on outcomes of ranibizumab therapy in neovascular ciations of incomplete posterior vitreous detachment in adult Chinese: the 
age-related macular degeneration. Ophthalmology. 2013;120:2620–9. Beijing Eye Study. PLoS ONE. 2013;8:e58498.
Eye
W.M. Amoaku et al.   
7
72. Quinn NB, Steel DH, Chakravarthy U, Peto T, Hamill B, Muldrew A, et al. COMPETING INTERESTS
Assessment of the vitreomacular interface using high-resolution OCT in a WMA: Consultant for Abbvie, Alimera, Allergan Inc, Bayer, Bioeq, Novartis, Pfizer, 
population-based cohort study of older adults. Ophthalmol Retina. Roche, Siloton and Thrombogenics, and has undertaken research sponsored by 
2020;4:801–13. Allergan, Novartis, and Pfizer. He has received speaker fees and travel grants from 
73. Ben Ghezala ISA, Gabrielle PH, Bouche-Pillon J, Binquet C, Delcourt C, Daien V, Abbvie, Allergan, Bausch and Lomb, Bayer, Novartis, Pfizer, and Roche. WMA is a 
et al. Epidemiology of vitreomacular interface abnormalities using macular member of the Eye editorial board. PSB: Consultant for Alcon, Genentech, Kalsec, 
spectral-domain optical coherence tomography in an elderly population (The Kemin Health, Science Based Health, and has received Speaker fees from DSM, 
MONTRACHET Study). Retina. 2021;41:60–7. Kemin, and Sequenom and travel sponsorship from Koyo Mercantile. GSH: Scientific 
74. Edwards AO, Lee SJ, Fridley BL, Tosakulwong N. Density of common complex co-founder, CSO and shareholder for Perceive Biotherapeutics Inc.; Inventor/Co- 
ocular traits in the aging eye: analysis of secondary traits in genome-wide inventor of patents owned by the University of Iowa and the University of Utah. GS: 
association studies. PLoS ONE. 2008;3:e2510. Consultant for Allergan, Bayer and Roche. AV: Consultant for Bausch and Lomb and 
ACIONT. SA, KNAA, LC, CNH, WH, SL, MR, CP, VS and GV: None.
ACKNOWLEDGEMENTS ADDITIONAL INFORMATION
We are extremely grateful to the faculty and staff members of University of Ghana 
Medical School, Korle Bu and the Komfo Anokye Teaching Hospital and Kwame Correspondence and requests for materials should be addressed to Winfried M. 
Nkrumah University of Science and Technology for their superb technical assistance Amoaku .
and provision of space for this study. Statistical advice/analysis was kindly provided 
by Michael Stevenson, Department of Statistics, The Queen’s University of Belfast. Reprints and permission information is available at http://www.nature.com/ 
reprints
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims 
AUTHOR CONTRIBUTIONS in published maps and institutional affiliations.
WMA, GSH, GS: conception, design, data acquisition, data analysis, report and 
manuscript writing. LS: data acquisition, image analysis, report writing. VS: Image 
acquisition and analysis, report writing. KNK-A, SA and SL: planning, ethics 
submission, participant recruitment, report writing. CNH: data acquisition, report 
writing; CMP and WCH: data acquisition, analysis, report writing. AV and PB: data Open Access This article is licensed under a Creative Commons 
acquisition, and report writing. MR: image analysis, report writing. GV: statistical Attribution 4.0 International License, which permits use, sharing, 
analysis, report writing. adaptation, distribution and reproduction in any medium or format, as long as you give 
appropriate credit to the original author(s) and the source, provide a link to the Creative 
Commons licence, and indicate if changes were made. The images or other third party 
material in this article are included in the article’s Creative Commons licence, unless 
FUNDING indicated otherwise in a credit line to the material. If material is not included in the 
This work was supported in part by the National Institutes of Health (R01 EY014800 article’s Creative Commons licence and your intended use is not permitted by statutory 
[GSH], R24 EY017404 [GSH]), charitable donations made to the Sharon Eccles Steele regulation or exceeds the permitted use, you will need to obtain permission directly 
Center for Translational Medicine (GSH), an unrestricted grant from Research to from the copyright holder. To view a copy of this licence, visit http:// 
Prevent Blindness, New York, NY, to the Department of Ophthalmology & Visual creativecommons.org/licenses/by/4.0/.
Sciences, University of Utah, and the donation of an OCT instrument by John 
Cunningham, Belfast. The sponsor or funding organization had no role in the design 
or conduct of this research. © The Author(s) 2023
FOR THE GHANA AMD STUDY GROUP
Edem K. Ahiabor4, Stephen Akafo4, Amos Atkins5, Winfried M. Amoaku ]]]1✉, Paul S. Bernstein ]]]6, Jessica Feilmeier6, Michael Feilmeier6, 
Courtney N. Hageman6, Gregory S. Hageman6, Lisa S. Hancox8, William C. Hubbard6, Seth Lartey5, Sheri L. McCormick9, Norma Miller6, 
Lisa R. Nichols6, Christian M. Pappas6, Giuliana Silvestri7, Vittorio Silvestri3, Albert Vitale6, Laura N. Cushley2, Cóilin P. Ferrin3, 
Stacie Matthews6, Susie Choi6 and Christopher Ricks6
8Department of Microbiology & Immunology, University of Iowa, Iowa city, IA, USA. 9Department of Ophthalmology, University of Iowa, Iowa city, IA, USA.
Eye