Key-words: Benghazi - central macular thickness - diabetic retinopathy - Libya - optical coherence
tomography
Introduction
The prevalence of diabetes in Middle East and North Africa is 9.1% and this is expected
to reach 11.4% by the year 2040.[[1 ]] Benghazi is the second-largest city in Libya with 541,104 inhabitants according
to the Bureau of Statistics and Census Libya on 2012,[[2 ]] and with a prevalence of diabetes mellitus (DM) equal to 14.1%.[[3 ]] Diabetic retinopathy (DR) is a major cause of vision loss in working-age group
with a global prevalence of 35%.[[4 ]] Roaeid and Kadiki,[[5 ]] reported a prevalence of 30.6% of DR among patients with diabetes attending Benghazi
Diabetic Centre that in their opinion was underestimated because it was based only
on the fundus examination using direct ophthalmoscopy.
Diabetic macular edema (DME) is the most common cause of decreased vision in DR.[[6 ]]
Detecting DME with a slit-lamp biomicroscopic examination needs equipment and skills;
and the use of fluorescein angiography (FA) is considered effective, but it is an
invasive method for diagnosis.[[7 ]] Optical coherence tomography (OCT) is an objective technique that can measure retinal
morphologic characteristic which help improving diagnosis, treatment and the ultimate
visual outcomes in patients with diabetes.[[8 ]]
The aim of this study was to evaluate the role of OCT macula in the screening for
DME in patients with diabetes by comparing it to clinical funduscopy in addition to
assessing the relationship between central macular thickness (CMT) as measured by
OCT macula and the visual acuity of patient.
Subjects and Methods
A retrospective study was used. The records of ophthalmology patients attended Sidi
Hussein Health Center, Benghazi, Libya, between January 2014 and December 2017 were
reviewed.
Most of the patients were diabetics referred by the physicians in Sidi Hussein Health
Center for an ophthalmological check-up at the outpatient department (OPD) of the
center as a routine examination for diabetics attending the center, and fewer numbers
of patients were known cases of DR attending the OPD for follow-up.
Inclusion criteria
Patients with an established diagnosis of type 2 DM with or without DR.
Exclusion criteria
Exclusion criteria were eyes with significant media opacities that can result in poor
OCT signal, high refractive errors, vascular or pathological changes other than DR
and patients with previous intravitreal injection or any retinal surgery or laser
photocoagulation.
Data collection tools
A complete data record was extracted including age, gender, duration of DM (from the
time of diagnosis to the time of examination), their glycosylated hemoglobin level,
anti-diabetic treatment used, presence or absence of systemic hypertension, best-corrected
Snellen visual acuity recorded as logarithm of the minimum angle of resolution (log
MAR), and slit-lamp biomicroscopy fundus examination record of the stage of DR by
the same experienced ophthalmologist.
Fundus examination was done through dilated Pupil using 1% tropicamide with the help
of noncontact fundus lens (+90-diopter lens) and classified according to the criteria
of the Early Treatment Diabetic Retinopathy Study (ETDRS) Research Group protocol[[9 ]] as follow:
No diabetic retinopathy (NDR), nonproliferative diabetic retinopathy with clinically
significant macular edema (NPDR with CSME), NPDR without CSME, PDR without CSME and
CSME.
Macular edema was considered to be clinically significant as defined by the ETDRS
protocol.[[10 ]]
Although the use of contact lens biomicroscopic examination is considered more accurate
in evaluation of DR in clinical practice, the noncontact +90 diopter lens was chosen
here because it is used more often in most of the OPDs.[[11 ]]
OCT scans images were performed by (RS-3000 Advance– NIDEK); which is a spectral-domain
(SD) OCT; in a spoke-like pattern through a dilated pupil (1% tropicamide eye drops),
OCT setting: MACULA RADIAL 12 (6.0 mm {1024}) according to the manufacturer's protocol
as described in the user's manual.[[12 ]]
CMT/foveal thickness was defined as the retinal thickness in central 1 mm area of
the ETDRS layout and it was automatically calculated by the instrument software. Therefore,
we reviewed the existing literature in which foveal thickness was measured by OCT
in healthy controls without diabetes,[[13 ]],[[14 ]],[[15 ]],[[16 ]] and analysis suggested that 200 μm cutoff for the upper level of normal foveal
thickness in healthy nondiabetic adults; and so, an OCT foveal thickness of 200 μm
or less was defined as normal, 201–300 μm as mild thickening, 301–400 μm as moderate
thickening, and >400 μm as severe thickening.[[8 ]]
Ethical issues
The study was conducted according to the principles of the World Medical Association
Declaration of Helsinki. A formal approval was obtained from the Department of Health
Services Benghazi authorities.
Statistical analysis
Data were presented as frequencies and mean ± standard deviation (SD). Statistical
analyses were performed using Statistical Package for the Social Science (Windows
version 17.0; SPSS Inc., Chicago, IL, USA).
Chi-square test was used to analyze the statistical differences between two categorical
variables. P ≤ 0.05 was considered statistically significant.
Results
The records of 558 patients attended the ophthalmology clinic of Sidi Hussein Health
center, Benghazi, between January 2014 and December 2017 were reviewed.
Only 393 patients (786 eyes) fulfill the criteria's and were included in this study.
Patients excluded were about 165, out of them 89 patients (54%) were having media
opacity, 21 patients (13%) with another eye pathology (e.g., glaucoma, high refractive
errors, and age-related macular degeneration) and 55 patients (33%) were having previous
intravitreal injection, retinal surgery, or laser photocoagulation.
[[Table 1 ]] shows the categorical variable presentation, while [[Table 2 ]] shows statistical presentation of the cases with mean ± SD.
Table 1: Categorical variable case characteristics in 786 eyes
Table 2: Continuous case characteristics in 786 eyes
Subjective examination by slit-lamp biomicroscopic fundus examination with +90 D lens
showed that 619 eyes (80%) were having NDR changes and the remaining 167 eyes (20%)
were having different levels of DR [[Table 1 ]].
On the other hand, OCT measurement showed that 537 eyes (82%) out of 653 eyes with
mild macular edema (CMT = 201–300 μm) were diagnosed as NDR when examined biomicroscopically,
this disagreement between the subjective fundus examination and OCT measurements was
statistically significant (P < 0.0001) [[Figure 1 ]].
Figure 1: Central macular thickness interval in µm measured by optical coherence tomography
in relation to type of diabetic retinopathy. *P < 0.0001 NDR: No diabetic retinopathy,
NPDR: Nonproliferative diabetic retinopathy, PDR: Proliferative diabetic retinopathy,
CSME: Clinically significant macular edema
About 555 eyes (85%) out of 653 eyes with mild macular edema (CMT = 201–300 μm), were
having good visual acuity (log MAR 0.3 and less), which was highly significant (P
= 0.0001) [[Figure 2 ]]
Figure 2: Central macular thickness interval in µm measured by optical coherence tomography
in relation to VA (visual acuity) intervals. *P = 0.0001
Nearly 10 eyes (71%) out of 14 eyes with severe macular edema (CMT >400 μm) were having
the duration of diabetes of 10–20 years, and this was statistically significant (P
= 0.027) [[Figure 3 ]].
Figure 3: Central macular thickness interval in µm measured by optical coherence tomography
in relation to duration intervals in years *P = 0.027
There was a significant difference regarding CMT between males and females, 11 eyes
(78.5%) out of 14 eyes with severe macular edema (CMT > 400 μm) were male whereas
45 eyes (75%) out of 60 eyes with normal macular thickness (CMT = 0–200 μm) were females
(P = 0.002) [[Figure 4 ]].
Figure 4: Central macular thickness interval in µm measured by optical coherence tomography
in relation to gender *P = 0.002
Discussion
DR is one of the main causes of visual impairment and blindness.[[17 ]]
Macular edema is the most important sign of visual loss in type 2 diabetes if undetected
or untreated.[[18 ]],[[19 ]] Moreover, it can affect visual acuity at any stage of DR.[[1 ]]
OCT is believed to be superior to other diagnostic methods such as biomicroscopy,
FA, ultrasound, and retinal thickness analysis in the way that it quantifies, has
a higher sensitivity and resolution in the discovery of macular edema and subretinal
fluid and its ability to quantitatively monitor responses to various types of the
treatment of DME.[[20 ]],[[21 ]],[[22 ]],[[23 ]],[[24 ]],[[25 ]]
In the current study, there was disagreement between OCT measures and subjective fundus
examination; while objectively there were 537 eyes (82%) with mild macular edema (CMT
= 201–300 μm) measured by OCT; these were diagnosed by noncontact lens fundus examination
as NDR. Hence, the increased macular thickness was not detected in a large number
of cases when done subjectively. Many investigators reported similar results, Brown
et al. found that there was a poor agreement between contact lens fundus examination
and OCT when foveal thickness was mildly increased (201–300 μm).[[8 ]] Shahidi et al. in a study done by the retinal thickness analyzer, reported that
noncontact slit-lamp biomicroscopy could not detect the increase in retinal thickness
when it was 1.5 times the normal thickness.[[26 ]] Abrar et al. reported the increase in macular thickening with progressing stages
of DR without evidence of any CSME.[[27 ]] Oshima et al. in their study showed that the retinal thickness in patients with
diabetes without CSME was significantly greater than nondiabetic eyes.[[28 ]] Lattanzio et al. in their study concluded that macular thickness was greater in
diabetics than controls and advised the use of OCT for early detection of macular
edema.[[14 ]]
In the current study, it was found that 555 eyes (85%) having mild macular edema (CMT
= 201–300 μm) retained good visual acuity (log Mar 0.3 and less). Therefore, the mild
increase in macular thickness was not reflected by visual deterioration which means
that OCT detected macular thickness before it could affect the vision and this is
seriously important to consider when monitoring the patients.
Sánchez-Tocino et al. stated that the methods used to evaluate macular thickening
are relatively insensitive both to early changes in retinal thickness and in those
cases in which vision has not been affected yet.[[13 ]] Arora et al., although showed a strong relation between macular thickness and visual
acuity they suggested that a broad range of visual acuity are possible for the degree
of changes in macular edema.[[29 ]] In a large study by DRCR.net, they concluded that central retinal thickness accounts
for about 27% of the variation in visual acuity and this is consistent with the results
in this study.[[30 ]]
In this study, there was a statistically significant relation between CMT and duration
of DM; since 10 eyes (71%) with severe macular edema (CMT > 400 μm) were having a
duration of DM between (10–20) years. Although this contrasts with a previous study
done by Elzarrug et al., in Benghazi/Libya, who showed no relation between CSME and
duration of DM but that study was done subjectively by biomicroscopic fundus examination
and without the use of OCT.[[31 ]] Moreover, our results are in consistent with other's studies.[[32 ]],[[33 ]]
It was found that there was a significant relation with gender; 11 eyes (78.5%) with
severe macular edema (CMT > 400 μm) were male, whereas 45 eyes (75%) with normal macular
thickness (CMT = 0–200 μm) were female. and this contrasts with a study done previously
in the United States by means of fundus photographs that shows no differences in DME
prevalence by sex.[[34 ]]
Brown et al. suggested the term subclinical foveal edema to describe eyes with mild
macular thickness (201–300 μm) detected by objective imaging methods, because they
fail to detect it by contact lens biomicroscopy,[[8 ]] and similarly, in our study, these cases were neither detected subjectively nor
affecting the visual acuity of the patients. Researchers showed that OCT is more sensible
for the detection of early stage of edema that still not evident with biomicroscopy
or FA.[[17 ]],[[24 ]]
In their study, Browning et al. analyzed the errors in clinical diagnosis of DME and
showed that it is less sensitive than OCT and suggest that the shift to early diagnosis
by OCT would lead to earlier intervention and treatment.[[25 ]] Without the use of pupil dilatation; Hirano et al., diagnosed fovea threatening
DME using SD-OCT with a sensitivity of 100%,[[35 ]] and Medina et al. considered SD-OCT as a useful tool to detect and to measure DME
in the nonmydriatic pupil.[[36 ]]
In this study, OCT detected the increase in macular thickness in 82% of the eyes that
shows absence of biomicroscopic evidence of ME thus confirming that OCT is superior
to noncontact slit-lamp biomicroscopy in detecting and quantifying the increase in
macular thickness in diabetics.
Conclusions
OCT is a noninvasive method that takes few seconds to be done without exhausting and
irritating the patient by strong light. It can detect and quantify early macular edema
before affecting the vision of patients or becoming clinically apparent. It is strongly
recommended that this technique should be the technique of choice for detecting and
screening of DME in Libyan patients.
Limitation of the study
There was no control group, and we did not find any previous studies that measure
the normal macular thickness in Libyan population, so we recommend researcher to evaluate
this point.
