Download Diabetes, Hyperglycemia, and Central Corneal Thickness

Diabetes, Hyperglycemia, and Central
Corneal Thickness
The Singapore Malay Eye Study
Daniel H. W. Su, MMed(Ophth), FRCS(Ed),1 Tien Y. Wong, PhD, FRCS(Ed),1,2,3 Wan-Ling Wong, BSc,1
Seang-Mei Saw, PhD,1,2,4 Donald T. H. Tan, FRCS(Ed),1,2 Sunny Y. Shen, MMed(Ophth), MRCS(Ed),1,2
Seng-Chee Loon, FRCS(Ed),5 Paul J. Foster, PhD, FRCS(Ed),6 Tin Aung, PhD, FRCS(Ed),1,2 for the
Singapore Malay Eye Study Group
Purpose: To examine the relationship of diabetes and hyperglycemia with central corneal thickness (CCT) in
Malay adults in Singapore.
Design: Population-based cross-sectional study.
Participants: Three thousand two hundred eighty Malay adults ages 40 – 80 years living in Singapore.
Methods: The study population was selected using an age-stratified random sampling procedure of Malay
40- to 80-year-olds living in the southwestern part of Singapore. Participants had a standardized interview,
examination, and ocular imaging at a centralized study clinic. Central corneal thickness was measured with an
ultrasound pachymeter, and nonfasting serum glucose and glycosylated hemoglobin (Hb A1C) was obtained from
all participants. Diabetes was defined as having nonfasting glucose levels of ⱖ200 mg/dl (11.1 mmol/l), a
self-report of diabetic medication use, or physician diagnosis of diabetes.
Main Outcome Measures: Central corneal thickness.
Results: Of the 3280 (78.7% response) participants, data on CCT were available on 3239 right eyes. Central
corneal thickness was normally distributed, with a mean of 541.2 ␮m. There were 748 persons with diabetes
(23.0%). After controlling for age and gender, central corneas were significantly thicker in persons with diabetes
than in those without diabetes (547.2 ␮m vs. 539.3 ␮m, P⬍0.001) and, in the total population, with higher serum
glucose (539.6, 540.2, 541.3, and 544.4, comparing increasing glucose quartiles; P ⫽ 0.023) and higher Hb A1C
(537.8, 541.0, 541.4, and 545.5, comparing increasing Hb A1C quartiles; P⬍0.001) levels. In multiple linear
regression models adjusting for age, intraocular pressure (IOP), body mass index, and axial length, persons with
diabetes had, on average, central corneas 6.50 ␮m thicker than those of persons without diabetes.
Conclusions: This population-based study among Malays showed that diabetes and hyperglycemia are
associated with thicker central corneas, independent of age and IOP levels. These findings may have implications
for understanding the relationship between diabetes and glaucoma. Ophthalmology 2008;115:964 –968 © 2008
by the American Academy of Ophthalmology.
The measurement of central corneal thickness (CCT) aids in
the clinical assessment of glaucoma.1 Increasing evidence
suggests that CCT not only influences intraocular pressure
Originally received: April 5, 2007.
Final revision: July 27, 2007.
Accepted: August 10, 2007.
Available online: October 26, 2007.
Manuscript no. 2007-472.
1
Singapore National Eye Centre & Singapore Eye Research Institute,
Singapore.
2
Department of Ophthalmology, Yong Loo Lin School of Medicine,
National University of Singapore, Singapore.
3
Centre for Eye Research Australia, University of Melbourne, Melbourne,
Australia.
4
Department of Community, Occupational, and Family Medicine, Yong
Loo Lin School of Medicine, National University of Singapore, Singapore.
5
National University Hospital, Singapore.
964
© 2008 by the American Academy of Ophthalmology
Published by Elsevier Inc.
(IOP) levels, but also predicts the risk of glaucomatous optic
neuropathy. Data from the Ocular Hypertension Treatment
Study and European Glaucoma Prevention Study demonstrated that CCT was a predictor for the development of
primary open-angle glaucoma (POAG).2,3 In both studies,
subjects with thinner corneas were more likely to develop
POAG, independent of age and other factors. A recent study
6
Institute of Ophthalmology, University College London, London, United
Kingdom.
Funded by the National Medical Research Council, Singapore (grant no.
0796/2003), and Biomedical Research Council, Singapore (grant no. 501/
1/25-5), with support from the Singapore Prospective Study Program and
Singapore Tissue Network, A*STAR.
Correspondence to A/Prof Tin Aung, PhD, Glaucoma Department, Singapore
National Eye Centre, 11 Third Hospital Avenue, Singapore 168751. E-mail:
[email protected].
ISSN 0161-6420/08/$–see front matter
doi:10.1016/j.ophtha.2007.08.021
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.
Su et al 䡠 Diabetes and Central Corneal Thickness in Singaporean Malays
among primary angle-closure glaucoma subjects revealed
similar results.4 Conversely, thinner corneas that result from
laser refractive surgery are known to cause inaccurately low
estimates of IOP.5,6
Persons with diabetes are thought to be at higher risk of
glaucoma. The Blue Mountains Eye Study and other studies
showed that persons with diabetes are at increased risk of
developing glaucoma.7,8 However, the prospective Rotterdam Study and 2 other longitudinal studies did not find a
significant relationship between diabetes and open-angle
glaucoma.9 –11 Additionally, the mechanism in which diabetes and hyperglycemia predispose a person to glaucoma is
uncertain.12 One hypothesis is that chronic hyperglycemia
influences corneal thickness and IOP measurements. However, there have been a few studies that have directly
examined the relationship of diabetes and hyperglycemia
with CCT. In a small clinic-based study (n ⫽ 81), persons
with diabetes had thicker central corneas than persons without diabetes.13 In addition, those with proliferative retinopathy (n ⫽ 23) had greater CCT values than those without
proliferative retinopathy. Another study described differences of corneal thickness and corneal endothelial morphology in patients with diabetes as compared with age-matched
healthy controls.14 In that study of 200 persons with diabetes and 100 controls, those with diabetes were found to have
thicker corneas. Data from a randomized clinical trial, the
European Glaucoma Prevention Study, showed that participants with diabetes had thicker central corneas than persons
without diabetes.15
To our best knowledge, there are no population-based
studies on the relationship of diabetes or hyperglycemia
with CCT, which is the subject of this current analysis.
Materials and Methods
Study Population
The Singapore Malay Eye Study was a population-based crosssectional study of 3280 (78.7% response rate) Malay subjects ages
40 to 80 in Singapore. The study methodology has been described
previously.16 The sampling frame consisted of all Malays ages 40
to 80 living in 15 residential districts across the southwestern part
of Singapore. From an initial list of 16 069 Malay names provided
by the Ministry of Home Affairs, an age-stratified random sampling procedure was used to select 5600 names (1400 people from
each decade of 40 – 49, 50 –59, 60 – 69, and 70 – 80 years). Of the
5600 initially identified, 4168 participants (74.4%) were determined to be eligible to participate based on the inclusion criteria
mentioned earlier.16
Of these, 3280 (78.7%) were examined in the clinic, and the
remaining 888 (21.3%) were classified as nonparticipants. Nonparticipants were older but did not differ by gender (Table 1
[available at http://aaojournal.org]).
Approval for the study protocol was granted by the hospital’s
institutional review board, and the study was conducted in accordance with the Declaration of Helsinki. Written informed consent
was obtained from all subjects before enrollment.
Study Measurements
All participants underwent a standardized interview, examination,
and ocular imaging at a centralized study clinic.16 Five CCT
measurements were obtained from each eye with an ultrasound
pachymeter (Advent, Mentor O & O, Norwell, MA) and the
median reading was taken. As there was good correlation between
the measurements in both eyes, only the readings from the right
eye were used for analysis. A Goldmann applanation tonometer
(Haag-Streit, Koeniz, Switzerland) was used to obtain one reading
of IOP from each eye, and the reading of the right eye was used for
analysis, as there was no statistical difference between the readings
of the left and right eyes. Noncontact partial coherence laser
interferometry (IOL Master V3.01, Carl Zeiss Meditec AG, Jena,
Germany) was used to measure axial length. Each participant had
height and weight measurements, and these were used to determine
the body mass index (BMI). Systolic and diastolic blood pressures
(BPs) were taken also with an automated sphygmomanometer.
Nonfasting blood samples were drawn from all subjects to determine serum glucose and glycosylated hemoglobin (Hb A1C).
Diabetes mellitus was defined in this study as having nonfasting glucose levels ⱖ 200 mg/dl (11.1 mmol/l) or having a physician diagnosis of diabetes and using diabetic medications.17
During the interview, subjects were asked if they were smokers
and if they had smoked cigarettes in the past. Participants were
classified as current, past, or never smokers. The questions related
to education level were structured to match the national education
program and have been described previously in other epidemiological studies in Singapore.18
Statistical Analysis
Statistical analysis was performed using SPSS (version 11.5, SPSS
Inc., Chicago, IL). Proportions were compared using the chisquare test and means compared using the t test. Analysis of
covariance models were used to estimate mean CCT adjusted for
covariates. Multiple linear regression models were developed to
assess the change in CCT by presence/absence of diabetes,
1-mmol/l increase in serum glucose, and 1% increase in Hb A1C.
Models were initially adjusted for age and gender and then further
for IOP, BMI, and axial length.
Results
A total of 3280 subjects were recruited, giving a response rate of
78.7% for the study. Central corneal thickness was recorded in the
right eye of 3239 subjects; this report will describe only this
subgroup of subjects who had CCT measurements.
In this population, CCT was normally distributed, with a mean
of 541.2 ␮m. There was no statistical difference in mean CCT
between males (540.7 ␮m) and females (541.6 ␮m) (P ⫽ 0.415).
Central corneal thickness was found to decrease with age
(P⬍0.001) at an average of 5.13 ␮m a decade.
There were 748 persons with diabetes (23.0%). Table 2 shows
the characteristics of the study population, by diabetes status. In
general, persons with diabetes were older, more likely men, and
had received fewer years of formal education. They also had
higher BMI and systolic BP but lower serum cholesterol levels. As
expected, persons with diabetes had higher serum glucose, Hb
A1C, and serum creatinine values.
In univariate analysis, CCT was associated with higher IOP
measurements (533.1, 539.4, 544.3, and 551.5 comparing increasing IOP quartiles; P⬍0.001). Central corneal thickness was not
related to systolic or diastolic BP or height (P ⫽ 0.730, P ⫽ 0.989,
and P ⫽ 0.282, respectively).
Table 3 shows that, after controlling for age and gender, persons with diabetes had significantly thicker central corneas than
persons without diabetes (P⬍0.001). Central corneal thickness
965
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.
Ophthalmology Volume 115, Number 6, June 2008
Table 2. Participant Characteristics by Diabetes Status
Age, years (⫾ SD)
Men [n (%)]
Education [n (%)]
No formal education
Less than elementary
Elementary
High school
College/university
Smoking status [n (%)]
Current smokers
Past smokers
Never smoked
Systolic blood pressure [mmHg (⫾ SD)]
Diastolic blood pressure [mmHg (⫾ SD)]
BMI [kg/m2 (⫾ SD)]
Serum glucose [mmol/l (⫾ SD)]
Hb A1C [% (⫾ SD)]
Total cholesterol [mmol/l (⫾ SD)]
HDL cholesterol [mmol/l (⫾ SD)]
LDL cholesterol [mmol/l (⫾ SD)]
Serum creatinine [␮mol/l (⫾ SD)]
Diabetes
(N ⴝ 748)
No Diabetes
(N ⴝ 2491)
62.59 (9.36)
328 (43.9)
57.50 (11.21)
1238 (49.7)
214 (28.6)
93 (12.4)
333 (44.5)
81 (10.8)
25 (3.3)
454 (18.2)
208 (8.3)
1128 (45.3)
510 (20.5)
186 (7.5)
96 (12.8)
157 (21.0)
495 (66.2)
154.3 (23.3)
79.2 (10.8)
27.5 (4.8)
11.1 (5.4)
8.4 (2.0)
5.48 (1.28)
1.28 (0.31)
3.34 (1.04)
105.4 (84.2)
562 (22.6)
433 (17.4)
1496 (60.0)
144.8 (23.4)
79.9 (11.3)
26.0 (5.2)
5.5 (1.3)
5.9 (0.6)
5.67 (1.13)
1.38 (0.34)
3.61 (0.99)
90.0 (41.7)
P Value
⬍0.001
0.005
⬍0.001
⬍0.001
⬍0.001
0.11
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001
BMI ⫽ body mass index; HDL ⫽ high-density lipoprotein; LDL ⫽ low-density lipoprotein; SD ⫽ standard deviation.
was also greater with higher serum glucose levels (P ⫽ 0.023) and
higher Hb A1C levels (P⬍0.001).
In multiple linear regression models adjusting for age, IOP,
BMI, and axial length, persons with diabetes had, on average,
central corneas 6.50 ␮m thicker than those of persons without
diabetes (Table 4). Similarly, for each 1-mmol/l increase in serum
glucose levels, there was a 0.37-␮m increase in CCT, and for each
1% increase in serum Hb A1C level, there was a 1.07-␮m increase
in CCT. This association was essentially similar in men and
women.
Discussion
This population-based study demonstrates a relationship
between diabetes or glycemic levels and CCT, independent
of age, IOP, and other factors. On average, persons with
diabetes had central corneas 6.5 ␮m thicker than those of
persons without diabetes mellitus, and mean CCT was positively related with increasing levels of serum glucose and
Hb A1C. This association was essentially similar in men and
women. Serum glucose levels were related to CCT in ageand gender-adjusted models only in univariate analysis.
However, the relationship between Hb A1C and CCT persisted even after multivariate adjustment. This suggests that
variations in glucose levels of at least up to 3 months
probably affect CCT to a greater extent than short-term
fluctuations of glucose levels.
The association between diabetes and CCT is potentially
important, as diabetes is a common condition in most countries, with an estimated 20.8 million (7.0% of the popula-
Table 3. Mean (Standard Error) of Central Corneal Thickness, Right Eye
Diabetes
No
Yes
Serum glucose (mmol/l)
First quartile
Second quartile
Third quartile
Fourth quartile
Serum hemoglobin A1C (%)
First quartile
Second quartile
Third quartile
Fourth quartile
N
Age and Gender
Adjusted
2491
748
539.3 (0.7)
547.2 (1.2)
⬍0.001
539.8 (0.7)
546.4 (1.3)
⬍0.001
849
719
782
765
539.6 (1.1)
540.2 (1.2)
541.3 (1.2)
544.4 (1.2)
0.023
540.4 (1.2)
541.0 (1.3)
541.4 (1.2)
543.3 (1.2)
0.40
898
693
810
768
537.8 (1.1)
541.0 (1.3)
541.4 (1.2)
545.5 (1.2)
⬍0.001
539.2 (1.1)
541.5 (1.3)
541.3 (1.2)
544.2 (1.2)
0.04
P Value
Multivariate
Adjusted*
P Value
*Analysis of covariance, adjusted for age, gender, intraocular pressure, body mass index, and axial length.
966
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.
Su et al 䡠 Diabetes and Central Corneal Thickness in Singaporean Malays
Table 4. Multiple Linear Regression of Corneal Thickness,
Right Eye
Characteristics
All*
Diabetes, yes versus no
Serum glucose, per 1-mmol/l increase
Serum hemoglobin A1C, per 1% increase
Men†
Diabetes, yes versus no
Serum glucose, per 1-mmol/l increase
Serum hemoglobin A1C, per 1% increase
Women†
Diabetes, yes versus no
Serum glucose, per 1-mmol/l increase
Serum hemoglobin A1C, per 1% increase
␤ Coefficient
P Value
6.550
0.370
1.074
⬍0.001
0.03
0.01
5.453
0.328
1.030
0.01
0.17
0.08
6.965
0.364
0.974
0.001
0.11
0.06
Each row represents a separate linear regression model.
*Linear regression models adjusted for age, gender, intraocular pressure,
body mass index, and axial length.
†
Linear regression models adjusted for age, intraocular pressure, body mass
index, and axial length.
tion) having diabetes in the United States alone.19 A study
found that persons with diabetes had central corneas thicker
than those of healthy controls.14 Subjects who had diabetes
for more than 10 years also had central corneas thicker than
those of persons with diabetes for ⬍10 years’ duration.
Another study found that persons with diabetes had thicker
central corneas than those without diabetes, but there was no
relationship between duration of diabetes and CCT.13 Both
these studies were limited in being hospital based and
having small samples. The European Glaucoma Prevention
Study recently showed that persons with diabetes had
thicker central corneas than persons without diabetes (588
vs. 571 ␮m). However, this study was a randomized clinical
trial, and the number of subjects with diabetes was small.15
The CCT measurements in this small white population were
greater than those in our Malay population, and there was a
greater difference in CCT between persons with diabetes
and persons without diabetes (17 vs. 6.5 ␮m). The basis for
the association we found between diabetes and CCT is
unknown, but we can speculate on several mechanisms.
Hyperglycemia may cause corneal endothelial dysfunction
with resultant stromal hydration and swelling of the cornea.
Indeed, abnormalities of corneal endothelial morphology
such as polymorphism, polymegatheism, decrease in percentage of hexagonal cells, higher coefficient of variation,
and increased CCT have been detected on specular microscopy in persons with diabetes.20,21 Older diabetics also
demonstrated a significant decrease in cell density in the
fourth and fifth decades compared with age-matched controls.21 In experiments comparing persons with type 1 diabetes, normoglycemic cystic fibrosis patients, hyperglycemic cystic fibrosis patients, and age-matched controls (n ⫽
31), persons with diabetes were found to have normal CCT
but significantly increased corneal endothelial permeability
and pump rates compared with controls. Similar increases in
endothelial permeability and pump rate were seen in the 2
groups of cystic fibrosis patients. Central corneal thickness
was greatest in hyperglycemic cystic fibrosis patients, fol-
lowed by normoglycemic cystic fibrosis patients and then
persons with diabetes and controls, suggesting that hyperglycemia exacerbated the corneal endothelial dysfunction.22
A limitation of our study was that the findings are crosssectional and that subjects had only one blood sample taken
for serum glucose and Hb A1C measurement. Although we
found a positive correlation between these 2 factors and
mean CCT, we could not demonstrate that a rise or fall in
serum glucose or Hb A1C would be accompanied by a
corresponding change in CCT in an individual patient.
Thus, we are unable to confirm a causal relationship. Only
a cohort study in which subjects have repeated CCT and
blood glucose measurements would be able to reveal the
exact relationship between CCT and blood glucose fluctuations. Information obtained from other studies may also
show if other factors play a part in this relationship.
In conclusion, our study shows that persons with diabetes
mellitus or higher glycosylated hemoglobin levels have
greater CCT, independent of age, gender, or IOP levels.
These findings suggest that CCT measurements may be
affected by chronic hyperglycemia and, together with future
research findings, may aid in understanding the pathophysiological processes in which diabetes influences the risk of
glaucoma.
References
1. Shih CY, Graff Zivin JS, Trokel SL, Tsai JC. Clinical significance of central corneal thickness in the management of
glaucoma. Arch Ophthalmol 2004;122:1270 –5.
2. Gordon MO, Beiser JA, Brandt JD, et al. The Ocular Hypertension Treatment Study: baseline factors that predict the
onset of primary open-angle glaucoma. Arch Ophthalmol
2002;120:714 –20.
3. European Glaucoma Prevention Study (EGPS) Group. Predictive factors for open-angle glaucoma among patients with
ocular hypertension in the European Glaucoma Prevention
Study. Ophthalmology 2007;114:3–9.
4. Hong S, Kim CY, Seong GJ, Hong YJ. Central corneal thickness and visual field progression in patients with chronic
primary angle-closure glaucoma with low intraocular pressure. Am J Ophthalmol 2007;143:362–3.
5. Kaufmann C, Bachmann LM, Thiel MA. Intraocular pressure
measurements using dynamic contour tonometry after laser in
situ keratomileusis. Invest Ophthalmol Vis Sci 2003;44:
3790 – 4.
6. Svedberg H, Chen E, Hamberg-Nystrom H. Changes in corneal thickness and curvature after different excimer laser
photorefractive procedures and their impact on intraocular
pressure measurements. Graefes Arch Clin Exp Ophthalmol
2005;243:1218 –20.
7. Nakamura M, Kanamori A, Negi A. Diabetes mellitus as a risk
factor for glaucomatous optic neuropathy. Ophthalmologica
2005;219:1–10.
8. Mitchell P, Smith W, Chey T, Healey PR. Open-angle glaucoma and diabetes: the Blue Mountains Eye Study, Australia.
Ophthalmology 1997;104:712– 8.
9. de Voogd S, Ikram MK, Wolfs RC, et al. Is diabetes mellitus
a risk factor for open-angle glaucoma? The Rotterdam Study
Ophthalmology 2006;113:1827–31.
10. Le A, Mukesh BN, McCarty CA, Taylor HR. Risk factors
associated with the incidence of open angle glaucoma: the
967
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.
Ophthalmology Volume 115, Number 6, June 2008
11.
12.
13.
14.
15.
16.
Visual Impairment Project. Invest Ophthalmol Vis Sci 2003;
44:3783–9.
Ellis JD, Evans JM, Ruta DA, et al. Glaucoma incidence in an
unselected cohort of diabetic patients: is diabetes mellitus risk
factor for glaucoma? Br J Ophthalmol 2000;84:1218 –24.
Becker B. Diabetes mellitus and primary open-angle glaucoma. The XXVII Edward Jackson Memorial Lecture. Am J
Ophthalmol 1971;1:1–16.
Busted N, Olsen T, Schmitz O. Clinical observations on the
corneal thickness and the corneal endothelium in diabetes
mellitus. Br J Ophthalmol 1981;65:687–90.
Lee JS, Oum BS, Choi HY, et al. Differences in corneal
thickness and corneal endothelium related to duration in diabetes. Eye 2006;20:315– 8.
European Glaucoma Prevention Study Group. Central corneal
thickness in the European Glaucoma Prevention Study. Ophthalmology 2007;114:454 –9.
Foong AW, Saw SM, Loo JL, et al. Rationale and methodology for a population-based study of eye diseases in Malay
people: the Singapore Malay Eye Study (SiMES). Ophthalmic
Epidemiol 2007;14:25–35.
17. American Diabetes Association. Diagnosis and classification
of diabetes mellitus. Diabetes Care 2007;30(suppl):S42–S47.
18. Wong TY, Foster PJ, Hee J, et al. Prevalence and risk factors
for refractive errors in adult Chinese in Singapore. Invest
Ophthalmol Vis Sci 2000;41:2486 –94.
19. Centers for Disease Control and Prevention. National diabetes
fact sheet: general information and national estimates on diabetes in the United States, 2003. Atlanta: U.S. Department of
Health and Human Services, Centers for Disease Control and
Prevention; 2004. Available at: http://www.cdc.gov/diabetes/
pubs/factsheet.htm. Accessed August 6, 2007.
20. Larsson LI, Bourne WM, Pach JM, Brubaker RF. Structure
and function of the corneal endothelium in diabetes mellitus
type I and type II. Arch Ophthalmol 1996;114:9 –14.
21. Schultz RO, Matsuda M, Yee RW, et al. Corneal endothelial
changes in type I and type II diabetes mellitus. Am J Ophthalmol 1984;98:401–10.
22. Lass JH, Spurney RV, Dutt RM, et al. A morphologic and
fluorophotometric analysis of the corneal endothelium in type I
diabetes mellitus and cystic fibrosis. Am J Ophthalmol 1985;
100:783– 8.
968
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.
Su et al 䡠 Diabetes and Central Corneal Thickness in Singaporean Malays
Table 1. Comparison of Participants and Nonparticipants
in the Singapore Malay Eye Study
Age group (yrs)
40–49
50–59
60–69
70–80
Gender
Male
Female
Participants
(N ⴝ 3280)
[N (%)]
Nonparticipants
(N ⴝ 888)
[N (%)]
814 (24.8)
957 (29.2)
780 (23.8)
729 (22.2)
154 (17.4)
244 (27.5)
233 (26.2)
257 (28.9)
1576 (48.0)
1704 (52.0)
423 (47.6)
465 (52.4)
P Value
⬍0.001
0.82
968.e1
FOR HKMA CME MEMBER USE ONLY. DO NOT REPRODUCE OR DISTRIBUTE.