9 1 Jan.18, 16 www. IJO.

cn
8629 8629-82210956 ijopress

Clinical Research

Macular edema after cataract surgery in diabetic eyes

evaluated by optical coherence tomography

1
Department of Ophthalmology, Peking University Third KEYWORDS: macular edema; diabetes; optical coherence
Hospital, Beijing 100191, China
tomography; cataract surgery
2
Department of Ophthalmology, Liangxiang Hospital, Beijing DOI:10.18240/ijo.2016.01.14
102401, China
Correspondence to: Lin Zhao. Department of Chen XY, Song WJ, Cai HY, Zhao L. Macular edema after cataract
Ophthalmology, Peking University Third Hospital, No. 49, surgery in diabetic eyes evaluated by optical coherence tomography.
Hua Yuan North Road, Hai Dian District, Beijing 100191, 2016;9(1):81-85
China. [email protected]
Received: 2014-12-30 Accepted: 2015-04-23 INTRODUCTION

M
acular edema (ME) is one of the most common causes
Abstract of visual loss after uncomplicated cataract surgery
nowadays[1-3]. A higher incidence of ME after cataract surgery
AIM: To assess quantitative changes of the macula in
is reported to occur in eyes with diabetic retinopathy (DR),
diabetic eyes after cataract surgery using optical
and worsening of ME often occur after surgery in eyes with
coherence tomography (OCT) and to estimate the
incidence of development or worsening of macular
pre-operative diabetic macular edema (DME) [1,4-5]. Several
edema (ME) in diabetic eyes with or without pre-existing studies made attempts to identify the risk factors of
ME. post-operative ME in diabetic eyes, though the exact cause of
this phenomenon is still undetermined. Baker [6]

METHODS: In this prospective, observational study, 92 indicated that pre-operative macular status and history of
eyes of 60 diabetic patients who underwent cataract
DME treatment might be associated with the increased risk.
surgery were evaluated before surgery and 1, 3mo after
Despite these efforts, the accurate prediction of
surgery using OCT. Macular thickness was measured
with OCT at nine macular subfields defined by the 9
post-operative macular status before surgery is still no easy
zones early treatment of diabetic retinopathy study task [7-8]. However, with the availability of the optical
(ETDRS), as well as total macular volume obtained by coherence tomography (OCT), we can obtain qualitative and
OCT at 1, 3mo after surgery were compared with baseline quantitative parameters of macula better than ever and
features obtained before surgery. In addition, the explore the relationship of macular status before and after
incidence of development or worsening of ME was cataract surgery in diabetic patients[9-12].
analyzed in diabetic eyes with or without pre-existing ME. In this study, we assessed the changes of macular thickness in
RESULTS: The central subfield mean thickness diabetic patients after cataract surgery using OCT and also
increased 21.0 m and 25.5 m at 1, 3mo follow -up,
examined the influence of preoperative laser treatment and
respectively ( <0.01). The average thickness of inner severity of DR on macular thickness [7]. We also analyzed the
ring and outer ring increased 14.2 m and 9.5 m at 1mo, incidence of both central-involved ME and non-central-
18.2 m and 12.9 m at 3mo. Central -involved ME involved ME following cataract surgery regarding
developed in 12 eyes at 3mo, including 4 eyes with pre- preoperative status of macula [1]. Through our study, we aimed
existing central -involved and 8 eyes with pre -existing to assess the quantitative changes of macula in diabetic eyes
non -central involved ME. Pre -existing diabetic macular after cataract surgery and help to estimate the incidence of
edema (DME) was significantly associated with central- ME after surgery in diabetic eyes.
involved ME development ( <0.001). SUBJECTS AND METHODS
CONCLUSION: A statistically significant increase could The study was conducted with the approval of ethics
be detected in the central subfield as well as perifoveal committee of Peking University Third Hospital in China and
and parafoveal sectors though the increase was mild. in accordance with ethical requirements of Declaration of
And eyes with pre -operative DME prior to cataract Helsinki. Patients with diabetes mellitus who underwent
surgery are at higher risk for developing central-involved phacoemulsification with intraocular lens insertion at our
ME. hospital between January 2012 and July 2013 were enrolled
81
Diabetic macular edema after cataract surgery
Table 1 Baseline and outcome DM E definitions
Baseline DM E category Definition
No central DME and no non-central CSMT <310 m, all ISF thickness <356 m, and all OSF thickness 303 m
DME
No central DME and non-central DME CSMT <310 m, and 1 ISF thickness 356 m or OSF thickness 303 m
Central DME and no non-central DM E CSMT310 m, all ISF thickness <356 m, and all OSF thickness 303m
Central DME and non-central DME CSMT310 m, and 1 ISF thickness 356 m or OSF thickness303 m
DME at follow-up

New development or progression 1) CSM T 310 m and CSM T increased 1 logOCT unit from baseline; 2) CSMT increased
central-involved M E 2 logOCT units from baseline.
New development or progression
1) 1 ISF thickness 356m and the corresponding ISF thickness increased 1 logOCT unit
non-central-involved M E from baseline, or 1 OSF thickness 303 m and the corresponding OSF thickness increased
1 logOCT unit from baseline; 2) 1 ISF thickness increased 2 logOCT units from baseline
or 1 OSF thickness increased 2 logOCT units from baseline.
DME: Diabetic macular edema; ME: M acular edema; CSMT: Central subfield macular thickness; ISF: Inner subfields; OSF: Outer subfields;
OCT: Optical coherence tomography.

in this study. Patients with pre-operative DME were also well as absolute change of macular volume to analyze the
included in the study. Patients were excluded from the study development or worsening of ME. The definition of
if they had any of the following conditions: prior or central-involved and non-central involved ME at baseline and
concomitant surgery such as vitreoretinal surgery or follow-up were shown in Table 1, as modified from criteria
glaucoma surgery in the operated eye, intraoperative offered by DR clinical research network (DRCR.net) [1]. In
complications, a history of ocular and systemic conditions addition, the influence of prior laser treatment and
associated with potentially irreversible significant vision loss, pre-operative ME was also analyzed.
presence of any retinal or choroidal disease other than Statistical analysis was performed by SPSS ver. 13.0 (SPSS
diabetes that could affect retinal thickness and history of Inc., Chicago, IL, USA). Descriptive statistics, including
treatment for DME or proliferative DR (PDR) within six mean and standard deviation were calculated for each group.
months prior to surgery. Group comparisons were performed using the student -test.
Cataract surgeries were performed by an experienced Fisher's exact, two-tailed test was used for analysis of data on
surgeon. A clear corneal incision was made, and continuous a nominal scale. A -value less than 0.05 was considered
curvilinear capsulorrhexis was performed. For cataract statistically significant.
surgery, phacoemulsification equipment (Legacy; Alcon RESULTS
Laboratories Inc., Fort Worth, TX, USA) was used. The In this prospective study, 104 eyes of 70 diabetic patients
nucleus was divided and phacoemulsification and aspiration were enrolled initially. And 4 eyes of 4 patients were
were performed after cortical aspiration, an acrylic foldable excluded because of loss to follow up at 1mo after surgery.
intraocular lens was inserted in the capsular bag. Also 8 eyes of 6 patients were excluded with low quality of
Preoperative and postoperative examinations including pre-operative OCT image for subsequent macular thickness
best-corrected visual acuity using standard visual acuity analysis because of advanced cataract. As a result, 92 eyes of
chart, which was translated into a logarithm of the minimal 60 diabetic patients were evaluated in the final study.
angle of resolution (logMAR) scale for analysis purposes, The 60 subjects in this series were 32 males and 28 females
slit-lamp examination, fundus photography were checked at with mean age of 70.2 9.1y. Mean preoperative best-
every follow-up visit. All patients were followed at 1wk, 1 corrected visual acuity in logMAR units was 0.570.47 and
and 3mo after surgery. OCT was performed at 1 and 3mo mean preoperative CSMT was 246.6 30.8 m. The mean
follow-up, using commercially available equipment (Cirrus preoperative thicknesses of inner and outer four quadrants as
HD-OCT, Carl Zeiss Meditec, Dublin, CA, USA). After well as macular volume are shown in Table 2.
pupil dilation, a macular thickness map was generated. This According to definition of DME at baseline, twenty eyes
standard OCT protocol uses six radially oriented scan lines to presented with DME before surgery. In these twenty eyes,
produce a topographical map of the macula and the mean two eyes had central-involved DME without non-central
value of 128 thickness values obtained in the central subfield, involved DME. Four eyes presented with both
which is circular area of diameter 1 mm centered around the central-involved and non-central involved DME. Another
center point [central subfield mean thickness (CSMT)], was fourteen eyes had only non-central involved DME. The
measured. The thicknesses of inner and outer four quadrants remaining 72 eyes showed no sign of pre-operative ME
of the nine early treatment of DR study (ETDRS) grid as well according to the criteria in Table 1.
as total macular volume were also measured. We used the DR severity was also assessed on clinical examination on a
absolute change of thickness, relative change of thickness as scale that included following: no DR, mild to moderate
82
 9 1 Jan.18, 16 www. IJO. cn
8629 8629-82210956 ijopress

Table 2 Demographics Table 3 Absolute change in thickness of CSMT, average thickness of inner and
outer ring at 1, 3mo after surgery compared with baseline
Demographics Data
Measurements 1mo P 3mo P
Eyes/patients 92/60
CSMT (m) 21.0 (5.4-36.6) 0.009 25.5 (12.1-39.0) 0.007
Sex (M:F) 32:28 Average inner ring
14.2 (7.6-20.8) 0.005 18.2 (9.5-26.6) 0.008
Age (a) 70.29.1 thickness (m)
Average outer ring
Preoperative BCVA (logMAR) 0.570.47 9.5 (5.2-13.7) 0.009 12.9 (7.9-18.0) 0.015
thickness (m)
3
Baseline CSMT (m) 246.630.8 Macular volume (mm ) 0.2 (0.1-0.8) 0.003 0.4 (0.1-0.9) 0.007
Baseline average inner ring thickness (m) 314.023.5 CSMT: Central subfield macular thickness. P<0.01.
Baseline average outer ring thickness (m) 276.524.3 Table 4 CSMT before surgery and at 1, 3mo follow-up regarding
3
Baseline macular volume (mm ) 10.50.6 pre-operative macular status
Pre-op macular status Pre-operation 1mo follow-up 3mo follow-up
DR stage
Central DME (+) 328.0 446.6 452.2
No DR 72/92 Non-central DME (-) (P) - 0.015 0.039
Mild to moderate NPDR 8/92 Central DME (+) 331.3 423.4 421.3
Severe NPDR or PDR 12/92 Non-central DME (+) (P) - 0.037 0.023
Central DME(-) 243.5 275.7 282.1
History of photocoagulation Non-central DME (+) (P) - 0.008 0.017
Yes 4/92 Central DME(-) 240.2 252.3 256.6
No 88/92 Non-central DME (-) (P) - 0.428 0.342

BCVA: Best corrected visual acuity; CSMT: Central subfield macular

thickness; DR: Diabetic retinopathy; NPDR: Non-proliferative diabetic At 1-month follow-up, 15 eyes met the criteria of new
retinopathy; PDR: Proliferative diabetic retinopathy.
development or worsening of central-involved DME,
including 4 eyes with pre-existing central-involved ME.
non-PDR (NPDR), severe NPDR or PDR. The DR severity
Fifteen eyes met the criteria of new development or
was identified according to International Clinical DR Disease
worsening of non-central-involved DME, including 6 eyes
Severity Scale published in 2002. During pre-operative
with pre-existing non-central-involved ME. Respective
examinations, seventy-two eyes included in the study did not
values at 3mo were 20 eyes for central-involved DME
show any sign of DR during preoperative examination.
progression and 19 eyes for non-central involved DME
Another eight eyes had mild to moderate NPDR while twelve
progression. The details of these incidences can be seen in
eyes had severe NPDR or PDR.
During the follow-up at 1wk, anterior segment examination Table 5.
was unremarkable except mild anterior segment inflammation Severity of DR was not significantly correlated to ME ( =
in six eyes (6.5% ). And this inflammation subsided at 0.46). Since the laser treatment could affect macular
1-month follow-up. Mean best corrected visual acuity thickness and confuse the interpretation of the results, we
(logMAR) improved from 0.57 to 0.35 at 1mo and 0.26 at 3mo. re-evaluated the correlation between severity of DR and
There was a significant increase of central subfield thickness changes of macular thickness after patients who had prior
and average inner ring and outer ring thickness at 1, 3mo laser treatment were excluded. Still no statistical difference in
compared with baseline measurements ( <0.01). Mean macular thickness was revealed through severity of DR after
central subfield foveal thickness (CSFT) increased from exclusion of these patients ( =0.39).
246.6 m to 267.6 m at 1-month follow-up and to 272.1 m DISCUSSION
at 3mo. The average inner ring thickness increased from This prospective, observational study demonstrates that in
314.0 m preoperatively to 328.2 m (post-op 1mo) to diabetic eyes, macular thickness of the central subfield as
332.2 m (post-op 3mo). And the average outer ring well as the inner and outer rings increase statistically up to
thickness increased from 276.5 m preoperatively to 286.0 m three months after surgery, though the mean increase may be
(post-op 1mo) to 289.4 m (post-op 3mo). However, the too small to reveal any clinical significance. In addition,
absolute changes in thickness were mild in all measurements pre-operative macular status is associated with progression or
as shown in Table 3. In eyes with central-involved and no worsening of ME after surgery. In this study, eyes with
non-central involved DME at 1mo, the mean change of pre-operative DME had a higher mean CSMT change than
CSMT was 118.6 m, which was much higher than the eyes without at 1, 3mo after surgery, and eyes with
average change of all eyes. In eyes with both pre-operative pre-operative central-involved DME had the highest mean
central-involved and non-central involved DME, the mean change of CSMT. Besides, eyes with pre-operative DME
change of CSMT was 92.4 m. In eyes with pre-operative suffered a higher incidence of central or non-central involved
non-central involved DME at 1mo, the mean change of ME progression.
CSMT was 32.2 m ( <0.01) while mean change of CSMT One unique feature of our study was the definition of ME. In
of remaining eyes was 12.1 m. Similarly, the mean changes the pre-OCT era, ME was diagnosed with fundus
of CSMT in eyes at 3mo were shown in Table 4. photography or fluorescein angiography. With the availability
83
Diabetic macular edema after cataract surgery

Table 5 Incidence of development or worsening of ME at 1, 3mo regarding pre-operative macula status

Central-involved ME Non-central-involved ME
Measurements progression progression
1mo 3mo 1mo 3mo
Pre-op macular status
Central DME and no non-central DME 1/2 1/2 1/2 1/2
Central DME and non-central DME 3/4 4/4 3/4 4/4
No central DME and non-central DME 6/14 8/14 6/14 6/14
No central DME and no non-central DME 5/72 7/72 5/72 8/72
The incidence is shown in absolute number.

of OCT, clinicians attempted to detect more subtle macular after uncomplicated cataract surgery. Similar results have
changes using qualitative and quantitative measurements been reported by Biro and Balla [17], which indicated an
obtained by OCT. However, there was still no consensus around 5% percent increase in CPT and perifoveal retinal
when it comes to the OCT definition of ME [13-15]. We adapted thickness in eyes of non-diabetic patients. Though the extent
and modified the concepts provided by DRCR.net and of increase was pretty much the same extent as the accuracy
defined ME based on macular thickness. And we categorized of OCT measurement, the definite trend of increase in
ME into four subgroups in regard to involvement of central thickness in all quadrants makes the probability of
subfield and we evaluated progression or worsening of ME measurement variation unlikely. However, the clinical
based on the log change in OCT CSFT from baseline, which meaning of this increase in macular thickness remains to be
takes into account baseline thickness and requires OCT explored, as most patients with this subclinical increase
change beyond variability of the OCT machine itself. maintain good vision after surgery[18-19].
The incidence of ME after cataract surgery was 21/92 One limitation of our study is that we did not differentiate
(22.8% ) in our study, similar to results from reports of DME from Irvine-Gass cystoid macular edema (CME). As
post-surgical ME in diabetic eyes, which confirmed the previous studies revealed, both types of ME can occur in
feasibility of our definition of ME. In a single center study of diabetic eyes undergoing cataract surgery. However, in our
50 eyes, Kim [16]
reported an incidence rate of 22% study, we did not differentiate these two types. It is difficult
(95% CI, 13%-35%) for DME exacerbation (defined as 30%
to tell these two forms apart just assessing OCT
increase in OCT center-point thickness compared with
measurements, obtaining parallel results of fundus
pre-surgical OCT) 1mo after cataract surgery.
fluorescence angiography may help to differentiate as
The further analysis based on pre-operative macular status
petaloid accumulation of fluorescein around the fovea with
revealed that eyes with central-involved ME had a higher
staining of the optic disc characteristic of Irvine-Gass CME[20].
CSMT increase, and still suffered central-involved ME after
Follow-up time of only three months may be another
surgery. And these eyes also revealed a higher incidence in
limitation of study. Though studies suggested that more than
developing non-central involved ME, though the clinical
60% ME occur at 1-month follow-up, there are reports that
significance remains to be explored. In addition, among eyes
ME occur at 6mo or even 1y after surgery. Longer follow-up
without central-involved ME, eyes with non-central involved
in the future is needed to derive a more accurate incidence of
ME before surgery had a higher chance of developing
ME in diabetic eyes after cataract surgery.
central-involved ME after surgery. The result was consistent
This study shows an increase in central and perifoveal
with reports of DRCR.net[1].
However, the incidence of progression of ME in this very thickness in diabetic eyes after cataract surgery regardless of
group was almost 50% in our study, which was much higher stages of DR. Our study also shows that pre-operative
than incidence of 10% (10 of 97 eyes) in Baker 's [6] macular status is a risk factor for post-operative ME.
study. The unusually high incidence may be contributed to Clinicians should continue to maintain vigilance in diabetic
small sample included in our study, as only 14 eyes were patients after cataract extraction, and obtaining an OCT
included in this group. In addition, previous study suggested before surgery can establish baseline measurements as well
that history of photocoagulation correlated with the incidence as help assess the risk of ME development. Further studies
of ME after surgery [7]. However, this correlation was not with large sample size are needed to assess the risk of DME
observed in our study, possibly due to the small number of progression after cataract surgery.
eyes who had photocoagulation before surgery. ACKNOWLEDGEMENTS
Our study also found a mild increase of the central subfield Conflicts of Interest: Chen XY, None; Song WJ, None;
and inner and outer ring retinal thickness in diabetic eyes Cai HY, None; Zhao L, None.
84
 9 1 Jan.18, 16 www. IJO. cn
8629 8629-82210956 ijopress
REFERENCES coherence tomography (OCT) for detection of macular oedema in patients
1 Kim SJ, Equi R, Bressler NM. Analysis of macular edema after cataract with diabetic retinopathy. 2011;7:CD008081.
surgery in patients with diabetes using optical coherence tomography. 12 Otani T, Kishi S, Maruyama Y. Patterns of diabetic macular edema with
2007;114(5):881-889. optical coherence tomography. 1999;127(6):688-693.
2 Diabetic retinopathy clinical research network, Browning DJ, Glassman 13 Browning DJ, Altaweel MM, Bressler NM, Bressler SB, Scott IU;
AR, Aiello LP, Relationship between optical coherence Diabetic retinopathy clinical research network. Diabetic macular edema:
tomography-measured central retinal thickness and visual acuity in what is focal and what is diffuse? 2008;146(5):649-655,
diabetic macular edema. 2007;114(3):525-536. 655.e1-6.
3 Mentes J, Erakgun T, Afrashi F, Kerci G. Incidence of cystoid macular 14 Ferris FL 3rd, Miller KM, Glassman AR, Beck RW; Diabetic
edema after uncomplicated phacoemulsification. 2003; retinopathy clinical research network. A proposed method of logarithmic
217(6):408-412. transformation of optical coherence tomography data for use in clinical
4 Schimel AM, Fisher YL, Flynn HW Jr. Optical coherence tomography in research. 2010;117(8):1512-1516.
the diagnosis and management of diabetic macular edema: time-domain 15 Diabetic retinopathy clinical research network, Krzystolik MG,
versus spectral-domain. 2011;42 Suppl: Strauber SF, Aiello LP, Beck RW, Berger BB, Bressler NM, Browning DJ,
S41-55. Chambers RB, Danis RP, Davis MD, Glassman AR, Gonzalez VH,
5 Baskin DE. Optical coherence tomography in diabetic macular edema. Greenberg PB, Gross JG, Kim JE, Kollman C. Reproducibility of macular
2010;21(3):172-177. thickness and volume using Zeiss optical coherence tomography in
6 Diabetic retinopathy clinical research network authors/writing committee, patients with diabetic macular edema. 2007;114 (8):
Baker CW, Almukhtar T, Bressler NM, Glassman AR, Grover S, Kim SJ, 1520-1525.
Murtha TJ, Rauser ME, Stockdale C. Macular edema after cataract surgery 16 Kim BY, Smith SD, Kaiser PK. Optical coherence tomographic patterns
in eyes without preoperative central-involved diabetic macular edema. of diabetic macular edema. 2006;142(3):405-412.
2013;131(7):870-879. 17 Biro Z, Balla Z. OCT measurements on the foveal and perifoveal retinal
7 Photocoagulation for diabetic macular edema. Early treatment diabetic thickness on diabetic patients after phacoemulsification and IOL
retinopathy study report number 1. Early treatment diabetic retinopathy implantation. 2010;24(4):639-647.
study research group. 1985;103(12):1796-1806. 18 Eriksson U, Alm A, Bjarnhall G, Granstam E, Matsson AW. Macular
8 Arevalo JF. Diabetic macular edema: changing treatment paradigms. edema and visual outcome following cataract surgery in patients with
2014;25(6):502-507. diabetic retinopathy and controls.
9 Browning DJ, McOwen MD, Bowen RM Jr, O'Marah TL. Comparison of 2011;249(3):349-359.
the clinical diagnosis of diabetic macular edema with diagnosis by optical 19 Sourdille P, Santiago PY. Optical coherence tomography of macular
coherence tomography. 2004;111(4):712-715. thickness after cataract surgery. 1999;25 (2):
10 Romero-Aroca P, Fernandez-Ballart J, Almena-Garcia M, Mendez- 256-261.
Marn I, Salvat-Serra M, Buil-Calvo JA. Nonproliferative diabetic 20 Kang SW, Park CY, Ham DI. The correlation between fluorescein
retinopathy and macular edema progression after phacoemulsification: angiographic and optical coherence tomographic features in clinically
prospective study. 2006;32(9):1438-1444. significant diabetic macular edema. 2004;137 (2):
11 Virgili G, Menchini F, Murro V, Peluso E, Rosa F, Casazza G. Optical 313-322.

85