ORIGINAL ARTICLE  
Vitreo-macular Interface Abnormalities in Diabetic  
and Non-Diabetic Patients Using Optical Coherence  
Tomography  
Uzma Hamza1, Waqas Asghar2, Qasim Lateef Chaudhry3, Muhammad Hassaan Ali4, Sana Jahangir5  
1-5Department of Ophthalmology, Allama Iqbal Medical College, Jinnah Hospital, Lahore  
ABSTRACT  
Purpose: To study the frequency of vitreomacular interface abnormalities (VIAs) in diabetic and non-diabetic  
patients presenting in a tertiary care hospital.  
Study Design: Comparative cross-sectional study.  
Place and Duration of Study: Jinnah hospital, Lahore from May 2013 to June 2016.  
Methods: The frequency of vitreomacular interface abnormalities (VIAs) was assessed among 278 patients, who  
presented in outpatient department of our hospital. Patients were categorized into diabetic and non-diabetic  
groups on the basis of hemoglobinHbA1c. Patients with altered macular reflex on slit lamp examination underwent  
spectral domain (SD) optical coherence tomography (OCT) of macula to determine VIAs.  
Results: There were 278 patients in the study with mean age 59.7 ± 11.7(range: 40 – 65) years and male to  
female ratio of 1:1.06. Prevalence of VIAs was observed to be higher among diabetic patients than non-diabetics  
in all age groups (p-value < 0.05). Overall frequency of different VIAs was found to be 10.7% for epiretinal  
membrane, 6.4% for posterior vitreous detachment, 6.1% for macular edema/macular cyst, 4.3% for  
vitreomacular traction, 1.8% for full thickness macular holes and 0.71% for partial thickness macular holes.  
Macular edema/macular cystwas the most common. VIA was more commonly observed in diabetic patients  
(17.2%). Except for ERM, all lesions of VIAs were significantly more prevalent in females as compared to males.  
Conclusion: VIAs are found in significantly larger number in diabetics compared to non-diabetic patients.  
Female gender with advancing age is associated with a higher frequency of VIAs.  
Key Words: Vitreomacular interface abnormalities, optical coherence tomography, epiretinal membrane,  
vitreomacular traction.  
How to Cite this Article: Hamza U, Asghar W, Chaudhry QL, Ali MH, Jahangir S. Frequency of Vitreomacular  
Interface Abnormalities in Diabetic and Non-Diabetic Patients Using Optical Coherence Tomography. Pak J  
Ophthalmol. 2020; 36 (3): 282-286.  
Doi: 10.36351/pjo.v36i3.1018  
retinopathy1,2,3. Apart from triggering diabetic macular  
edema, these lesions contribute to the development of  
advanced stages of diabetic retinopathy4,5,6. VIAs  
include epiretinal membrane (ERM), partial thickness  
macular hole (PTMH), full thickness macular hole  
(FTMH), vitreomacular traction (VMT), macular cyst  
or macular hole (MC/MH) and posterior vitreous  
detachment (PVD). Symptoms vary from mild  
metamorphopsia to severe visual deterioration.  
Lesions like PTMH and FTMH always result in visual  
INTRODUCTION  
Vitreomacular interface abnormalities (VIAs) are most  
commonly seen in patients with diabetic  
Correspondence to: Waqas Asghar  
Eye Department, Jinnah Hospital, Lahore  
Email: waqasasghar2008@gmail.com  
Received: March 14, 2020  
Accepted: May 4, 2020  
Revised: May 4, 2020  
282  
Pakistan Journal of Ophthalmology, 2020, Vol. 36 (3): 282-286  
Vitreomacular Interface Abnormalities in Diabetic and Non-Diabetic Patients Using Optical Coherence Tomography  
deterioration, therefore the techniques that can  
Patients with altered macular reflex on slit lamp  
bio-microscopy were referred for OCT test. Macular  
scans were acquired using standard 6×6 mm protocol  
on Cirrus HD-OCT 500 by Zeiss, USA. Presence of  
any VIA was recorded and categorized into ERM,  
PTMH, FTMH, PVD, VMT and MC/MH. Record of  
each patient including demographic data, ocular and  
OCT findings were recorded in a pre-designed  
proforma.  
diagnose their precursor lesions are very useful in  
clinical practice7.  
After the advent of ocular coherence tomography  
(OCT), VIAs have attracted significant clinical  
attention. Virgili et al have shown the value of  
Spectral-domain OCT (SD-OCT) for excellent  
visualization of VIAs which could potentially be  
missed on direct ophthalmoscopy or slit lamp  
biomicroscopy8. SD-OCT provides higher resolution  
and greater scanning speed than the time domain  
(TD)-OCT. Duker et al showed that SD-OCT has  
enabled ophthalmologists to visualize and monitor the  
vitreomacular interface with better accuracy and  
repeatability9.  
Data was analyzed using Statistical Package for  
Social Sciences (SPSS, IBM Statistics, Chicago, IL,  
USA version 23.0). Mean ± SD was calculated for  
numerical variables like age and duration of diabetes  
mellitus whereas frequencies and percentages were  
calculated for qualitative variables like gender and  
various VIAs. The statistical significance of  
differences between various numerical and qualitative  
variables was calculated using t-test and chi-square  
test respectively. The p-value < 0.05 was considered  
statistically significant.  
The rationale of the study was to find the reason  
for unexpected visual loss in patients with diabetic  
retinopathy. The objective was to compare frequency  
and pattern of various VIAs in diabetic and non-  
diabetic patients in our local population.  
RESULTS  
MATERIAL AND METHODS  
The study included 278 patients with mean age of 59.7  
± 11.7 (range: 40 – 65) years (Table 1). There were  
135 males and 143 females in the study (Table 2).  
Among 278 patients, 151 were diabetics and 127 were  
non diabetics. The mean duration of diabetes mellitus  
was 12.3 ± 5.2 years with 55 patients diagnosed with  
diabetes mellitus within last 5 years and 96 patients  
had diabetes for more than 5 years.  
This comparative cross-sectional study was conducted  
at Department of Ophthalmology, Jinnah Hospital,  
Lahore, Pakistan from May 2016 to June 2019. The  
study was conducted after approval from Ethical  
Review Board of the same institution and adhered to  
the principles of ethical medical practice as laid down  
in Declaration of Helsinki 2011. Patients were  
recruited from outpatient department of Jinnah  
Hospital after obtaining informed written consent.  
Prevalence of VIAs was observed to be higher  
among diabetic patients (66 patients) than non-  
Patients of both genders and above 40 years of age  
were included in the study and divided into two  
groups: diabetics and non-diabetics on the basis of  
hemoglobin HbA1c levels. The diabetes mellitus was  
defined as HbA1c 6.2%. Patients with history of  
vitreoretinal surgery and retinal vascular disorders like  
retinal vein occlusion were excluded from the study.  
After taking detailed ophthalmic history, detailed  
ophthalmic examination was performed which  
included assessment of unaided and best corrected  
visual acuity, pupillary examination, anterior segment  
examination using slit lamp biomicroscope and  
intraocular pressure measurement using applanation  
tonometer. The pupils were pharmacologically dilated  
using 1% tropicamide and 1% cyclopentolate eye  
drops. Dilated fundus examination was performed  
using slit lamp biomicroscope with 90D and 66D  
lenses.  
Table 1: Distribution of Patients in Different Age Groups.  
Diabetic  
Non-Diabetic  
Total  
n
70  
86  
122  
278  
Age (Years)  
n
33  
46  
72  
151  
n
37  
40  
50  
127  
40-54  
55-64  
>65  
Total  
n: Number of patients  
Table 2: Gender Distribution of Study Population.  
Diabetic  
Non-Diabetic  
Total  
n
135  
143  
278  
P-value  
n
81  
70  
n
54  
73  
Female  
Male  
Total  
0.071  
151  
127  
n: Number of patients  
Pakistan Journal of Ophthalmology, 2020, Vol. 36 (3): 282-286  
283  
Uzma Hamza, et al  
Table 3: Various vitreomacular interface abnormalities seen in diabetic and non-diabetic patients in different age groups.  
Diabetic Patients  
(n)  
Non-Diabetics  
(n)  
Total  
(n)  
VIA  
45-54  
55-64  
≥65  
8
0
Total  
18  
1
3
7
26  
11  
66  
45-54  
55-64  
≥65  
Total  
10  
1
ERM  
4
1
0
1
7
6
0
2
0
0
1
2
3
8
3
1
1
2
2
5
0
1
2
2
28  
2
5
12  
32  
19  
98  
PTMH  
FTMH  
VMT  
MC/ME  
PVD  
2
1
2
5
6
8
3
3
8
4
11  
4
3
16  
2
3
TOTAL  
23  
27  
11  
13  
32  
n: Number of patients  
diabetics (32 patients) in all age groups (Table 3). The  
frequency of VIAs increased with age (Table 3).  
Overall frequency of different VIAs was found to be  
10.7% for epiretinal membrane, 6.4% for posterior  
vitreous detachment, 6.1% for macular edema/macular  
cyst, 4.3% for vitreomacular traction, 1.8% for full  
thickness macular holes and 0.71% for partial  
thickness macular holes. Macular edema/ macular cyst  
were the most common. VIAs were more commonly  
observed in diabetic patients (17.2%). Except for  
ERM, all lesions of VIAs were significantly more  
prevalent in females (Table 4).  
prevalent in female patients. The prevalence of VIAs  
increased with advancing age of the patients.  
OCT provides high resolution cross-sectional  
scans of retina that is used to identify pathological  
changes at vitreoretinal interface7. There are different  
conventional methods for assessment of retinal  
pathologies which include slit lamp bio-microscopy,  
indirect ophthalmoscopy, fluorescein angiography and  
fundus stereo-photography. SD-OCT is  
a new  
modality that allows excellent visualization of  
vitreomacular interface, thus enabling us to study the  
vitreomacular abnormalities with high precision8. The  
pathophysiology of most of the VIAs is based on  
changes in vitreous with age. With advancing age,  
vitreous liquifies and collapses, thus causing complete  
or incomplete posterior vitreous detachment.  
Incomplete posterior vitreous detachment is associated  
with abnormal vitreomacular adhesions, which can  
become symptomatic and can lead to the development  
of VIAs such as vitreomacular traction and an  
operculum9. Similarly, epiretinal membrane can lead  
to development of partial or full thickness macular  
hole and macular edema or cyst9,10. The symptoms of  
the patients can vary from metamorphopsia to severe  
visual deterioration. Furthermore, VIAs not only  
trigger other retinal pathologies like myopic tractional  
maculopathy but also contribute to the development of  
Table 4: Frequency of Various Vitreomacular Interface  
Abnormalities in Males and Females.  
VIAs  
ERM  
PTMH  
FTMH  
VMT  
MC/ME  
PVD  
Male  
20  
0
1
4
14  
8
47  
Female  
Total  
28  
2
8
2
4
5
8
12  
32  
19  
98  
18  
11  
51  
VIAs  
ERM: Epiretinal membrane  
PTMH: Partial thickness macular hole  
FTMH: Full thickness macular hole  
VMT: Vitreomacular traction  
MC/ ME: Macular cyst/ macular edema  
PVD: Posterior vitreous detachment  
VIAs: Vitreoretinal interface abnormalities  
severe diabetic retinopathy11,12  
.
Unlike the current study, which utilized SD-OCT  
for classification of various VIAs, previous studies  
have reported prevalence of various VIAs on the basis  
of clinical diagnosis made on clinical examination  
and/or grading of fundusphotograph11,12,13,14. However,  
Beaver Dam Eye Study, Handan Eye Study and  
Maastricht Study used OCT imaging to report high  
resolution images of vitreoretinal interface16,17,18. The  
Beaver Dam and Handan Eye studies did not compare  
prevalence of VIAs in diabetic and non-diabetic  
DISCUSSION  
Our study found frequency of various VIAs on SD-  
OCT in diabetic and non-diabetic patients who  
presented to us with altered macular reflex. Overall,  
the commonest VIA was ERM in this study. However,  
macular edema and macular cysts were the commonest  
VIAs in diabetic patients. We also observed that,  
except for ERM, all VIAs were significantly more  
284  
Pakistan Journal of Ophthalmology, 2020, Vol. 36 (3): 282-286  
Vitreomacular Interface Abnormalities in Diabetic and Non-Diabetic Patients Using Optical Coherence Tomography  
patients16,17. Maastricht study calculated the prevalence  
gender with advancing age is associated with a higher  
frequency of VIAs. Optical coherence tomography  
proved to be a viable tool for the detection of various  
vitreomacular interface abnormalities.  
of all VIAs and stratified them according to the age,  
sex and diabetics status18.  
In our study we observed prevalence of ERM to be  
10.7%. This prevalence was higher in diabetic patients  
(6.4%) as compared with non-diabetics (3.5%). The  
prevalence of ERM was reported to be 6.1% and 3.4%  
in Maastricht and Handan studies respectively16,18. The  
Beaver study reported much higher prevalence of  
ERM (34.1%)17. All studies confirmed that the  
frequency of ERM increased with age. Our results are  
consistent with Maastricht study as we also found  
significantly higher prevalence of ERM in diabetics  
versus non-diabetics.  
Ethical Approval  
The study was approved by the Institutional review  
board/Ethical review board.  
Conflict of Interest  
Authors declared no conflict of interest.  
Authors’ Designation and Contribution  
The frequency of vitreomacular traction was found  
to be 4.3% in this study which is in accordance with  
the results of the Maastricht study18 (7.0%) but differ  
from the findings of Beaver Dam Study17 (26%). An  
earlier study reported prevalence of VMT to be 23.9%  
in patients with diabetic macular edema which is  
Uzma Hamza; Assistant Professor: Study design, data  
collection, Critical analysis, Statistical analysis,  
Manuscript writing.  
Waqas Asghar; Medical Officer: Data collection,  
Critical analysis, Statistical analysis, Manuscript  
writing.  
significantly higher than our finding (4.3%)15,19,20,21  
.
Qasim Lateef Chaudhry; Associate Professor:  
Concept, Design, Statistical analysis, final review.  
This implies that patients with diabetic macular edema  
have higher chances of developing vitreomacular  
tractions and should undergo OCT testing to check for  
macular pathology early in the course of the disease.  
Muhammad Hassaan Ali; Senior Registrar: Data  
collection, Statistical analysis, final review  
The frequency of macular hole in our study was  
found to be 1.79% with females affected 4 times more  
than males (2.69% versus 0.69%). Results of an earlier  
study showed prevalence of macular hole to be 0.5%.  
Similarly, we found prevalence of lamellar hole to be  
0.71%, which is consistent with results of the  
Maastricht Study (0.9%) but less than the findings of  
Sana Jahangir; Vitreoretina fellow: Data collection,  
analysis, final review.  
REFERENCES  
1. Tabish SA. Is diabetes becoming the biggest epidemic  
of the twenty-first century? Int J Health Sci. 2007; 1  
(2): 5-8.  
2. Copete S, Martí-Rodrigo P, Muñiz-Vidal R, Pastor-  
Idoate S, Rigo J, Figueroa MS et al. Preoperative  
vitreomacular interface abnormalities on Spectral  
Domain Optical Coherence Tomography as risk factor  
for pseudophakic cystoid macular edema. Retina. 2019;  
39 (11): 2225-32.  
3. Chang CK, Cheng CK, Bai CH, Peng CH, Hu CC.  
Development of vitreomacular interface abnormality in  
patients with diabetic macular edema. Taiwan J  
Ophthalmol. 2012; 2 (3): 93-8.  
4. Aziz-ul-Hasan Aamir, Zia Ul-Haq, Saeed A Mahar,  
Faisal Masood Qureshi, Ibrar Ahmad. Diabetes  
Prevalence Survey of Pakistan (DPS-PAK): prevalence  
of type 2 diabetes mellitus and prediabetes using  
HbA1c: a population-based survey from Pakistan.BMJ  
Open. 2019; 9 (2): e025300.  
the Beaver Dam study (3.6%)17,18  
.
The frequency of macular edema, macular cyst  
and posterior vitreous detachment were found to be  
significantly higher in diabetic patients when  
compared to non-diabetics (p-value < 0.05). This  
shows that the suspicion for diagnosing VIAs should  
be kept high in diabetic patients and where needed,  
OCT imaging should be done to acquire high  
resolution images of the vitreomacular interface for  
early diagnosis of various macular pathologies.  
The limitation of this study is the small number of  
patients and a larger study is required to be done to  
confirm the results in the general population.  
CONCLUSION  
VIAs were found in significantly larger number in  
diabetics compared to non-diabetic patients and female  
5. Mumtaz SN, Fahim MF, Arslan M, Shaikh SA, Kazi  
U, Memon MS. Prevalence of diabetic retinopathy in  
Pakistan; A systematic review. Pak J Med Sci. 2018; 34  
Pakistan Journal of Ophthalmology, 2020, Vol. 36 (3): 282-286  
285  
Uzma Hamza, et al  
(2): 493–500.  
15. McCannel CA, Ensminger JL, Diehl NN, Hodge DN.  
Population-based incidence of macular holes.  
Ophthalmology. 2009; 116 (7): 1366-9.  
16. Duan XR, Liang YB, Friedman DS, Sun LP, Wei  
WB, Wang JJ, et al. Prevalence and associations of  
6. Mikhail M, Stewart S, Seow F, Hogg R, Lois N.  
Vitreomacular interface abnormalities in patients with  
diabetic macular oedema and their implications on the  
response to anti-VEGF therapy. Graefes Arch Clin Exp  
Ophthalmology. 2018; 256 (8): 1411-8.  
epiretinal membranes in  
a rural Chinese adult  
7. Mirza RG, Johnson MW, Jampol LM. Optical  
coherence tomography use in evaluation of the  
vitreoretinal interface: a review. Surv Ophthalmol.  
2007; 52 (4): 397-421.  
8. Virgili G, Menchini F, Dimastrogiovanni AF,  
Rapizzi E, Menchini U, Bandello F, Chiodini RG.  
Optical coherence tomography versus stereoscopic  
fundus photography or biomicroscopy for diagnosing  
diabetic macular edema: a systematic review. Invest  
Ophthalmol Vis Sci. 2007; 48 (11): 4963-73.  
9. Duker JS, Kaiser PK, Binder S, de Smet MD,  
Gaudric A, Reichel E, Sadda SR, Sebag J, Spaide  
RF, Stalmans P. The International Vitreomacular  
Traction Study Group classification of vitreomacular  
adhesion, traction, and macular hole. Ophthalmology.  
2013; 120 (12): 2611-9.  
10. Syed YY, Dhillon S. Ocriplasmin: a review of its use in  
patients with symptomatic vitreomacular adhesion.  
Drugs. 2013; 73 (14): 1617-25.  
population: the Handan Eye Study. Invest Ophthalmol  
Vis Sci. 2009; 50 (5): 2018-23.  
17. Meuer SM, Myers CE, Klein BE, Swift MK, Huang  
Y, Gangaputra S et al. The epidemiology of  
vitreoretinal interface abnormalities as detected by  
spectral-domain optical coherence tomography: the  
beaver dam eye study. Ophthalmology. 2015; 122 (4):  
787-95.  
18. Liesenborghs I, De Clerck EE, Berendschot TT,  
Goezinne F, Schram MT, Henry RMA, et al.  
Prevalence of optical coherence tomography detected  
vitreomacular interface disorders: The Maastricht  
Study. Actaophthalmologica. 2018; 96 (7): 729-36.  
19. Fatima N, Islam QU, Shafique M. Frequency of  
vitreomacular traction in diabetic macular edema on  
optical coherence tomography. Pak Armed Forces Med  
J. 2017; 67 (1): 47-50.  
20. Crosson JN, Swain TA, Clark ME, Huisingh CE,  
Mc Gwin Jr G, Owsley C, Curcio CA. Retinal  
Pathologic Features on OCT among Eyes of Older  
Adults Judged Healthy by Color Fundus Photography.  
Ophthalmol Retina. 2019; 3 (8): 670-680.  
21. Sun P, Tandias RM, Yu G, Arroyo JG. Spectral  
Domain Optical Coherence Tomography findings and  
visual outcome after treatment for vitreomacular  
traction. Retina. 2019; 39 (6): 1054-60.  
11. Klein R, Klein BE, Moss SE. Visual impairment in  
diabetes. Ophthalmology. 1984; 91 (1): 1-9.  
12. Klein R, Klein BE, Wang Q, Moss SE. The  
epidemiology of epiretinal membranes. Transactions of  
the American Ophthalmological Society.1994; 92: 403.  
13. Mitchell P, Smith W, Chey T, Wang JJ, Chang A.  
Prevalence and associations of epiretinal membranes:  
The Blue Mountains Eye Study, Australia.  
Ophthalmology. 1997; 104 (6): 1033-40.  
14. Sen P, Bhargava A, Vijaya L, George R. Prevalence  
of idiopathic macular hole in adult rural and urban  
south Indian population. Clin Exp Ophthal. 2008; 36  
(3): 257-60.  
22. Brinkhues S1, Dukers-Muijrers NHTM, Hoebe  
CJPA, van der Kallen CJH, Koster A, Henry RMA,  
et al. Social Network Characteristics Are Associated  
With Type 2 Diabetes Complications: The Maastricht  
Study. Diabetes Care, 2018 Aug; 41 (8): 1654-1662.  
Doi: 10.2337/dc17-2144. Epub 2018 Jun 15.  
.…….  
286  
Pakistan Journal of Ophthalmology, 2020, Vol. 36 (3): 282-286