Neurodegeneration of the retina in type 1 diabetic patients

Abstract

Introduction:
The degeneration of retinal neurons and glial cells has recently been postulated in the pathogenesis of diabetic retinopathy. Optical coherence tomography (OCT) allows to perform qualitative and quantitative measurements of retinal thickness (RT) with identification of individual retinal layers.

Objectives:
We compared RT, retinal nerve fiber layer (RNFL) thickness, and ganglion cell layer (GCL) thickness obtained by OCT in type 1 diabetic patients with and without clinically diagnosed retinopathy.

Patients and methods:
The study included 77 consecutive patients with type 1 diabetes (39 men, 38 women; median age, 35 years [interquartile range (IQR), 29-42]; median disease duration, 10 years [IQR, 9-14]) and 31 age- and sex-matched controls. We measured RT in the fovea, parafovea, and perifovea, as well as RNFL and GCL thickness. We divided diabetic patients into 2 subgroups, i.e., those with diabetic retinopathy and without retinopathy.

Results:
We observed thicker perifoveal retina (P = 0.05), mean RNFL (P = 0.002), inferior RNFL (P <0.0001), and superior and inferior GCL (P = 0.05 and P = 0.04, respectively) in diabetic subjects compared with the control group. We detected retinopathy in 23 diabetic patients (29%). Compared with patients without retinopathy, subjects with retinopathy had thinner parafoveal retina (P = 0.05), mean RNFL (P = 0.002), inferior and nasal RNFL (P = 0.002, P = 0.03), superior (P = 0.05) and inferior GCL (P = 0.006). Significant correlations were found between duration of diabetes and nasal RNFL thickness (r = -0.32, P = 0.004) and parafoveal RT (r = -0.47, P <0.001).

Conclusions:
The results might suggest the loss of intraretinal neural tissue in type 1 diabetic patients with retinopathy. Neurodegeneration in diabetic retinopathy is closly associated with disease duration.

Full text

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ
2012; 122 (10)
464
INTRODUCTION Diabetic retinopathy is still
the leading cause of blindness in the working‑age
population of the Western countries.1 Early rec‑
ognition of changes in the retina in diabetic sub‑
jects is essential in the prevention of vision loss.2
Degeneration of retinal neurons and glial cells has
been postulated recently in the pathogenesis of
diabetic retinopathy.
3‑5
ese abnormalities were
described even before the development of mi‑
croaneurysms.6 However, for years apoptosis of
the neural tissue could be assessed only by fun‑
dus photography.
5
Recently, optical coherence to‑
mography (OCT) has been introduced into clin‑
ical practice as the most sensitive and objective
method to visualize the retina.7,8 First, OCT was
applied to detect macular edema in diabetic pa‑
tients.9 en, it allowed to perform quantitative
and qualitative measurements of retinal thick‑
ness (RT) and volume with the identification of
individual retinal layers. is method provided
significant clinical and pathological data in sev‑
eral studies of different retinal conditions.1 0,11
However, the information concerning the clini‑
cal value of measuring RT in type 1 diabetic pa‑
tients is still limited. Moreover, the usefulness
of this procedure in the diagnosis of retinopa‑
thy has not been fully elucidated.
Correspondence to:
Aleksandra Araszkiewicz, MD,
PhD, Katedra i Klinika Chorób
Wewnętrznych i Diabetologii,
Uniwersytet Medyczny
im. K. Marcinkowskiego w Poznaniu,
ul. Mickiewicza 2,
60-834 Poznań, Poland,
phone/fax: +48‑61‑847‑45‑79,
e‑mail: olaaraszkiewicz@interia.pl
Received: July 20, 2012.
Revision accepted: August 22, 2012.
Published online: August 22, 2012.
Conflict of interest: none declared.
Pol Arch Med Wewn. 2012;
122 (10): 464‑470
Copyright by Medycyna Praktyczna,
Kraków 2012
ABSTRACT
INTRODUCTION
The degeneration of retinal neurons and glial cells has recently been postulated in the pa-
thogenesis of diabetic retinopathy. Optical coherence tomography (OCT) allows to perform qualitative
and quantitative measurements of retinal thickness (RT) with identification of individual retinal layers.
OBJECTIVES We compared RT, retinal nerve fiber layer (RNFL) thickness, and ganglion cell layer (GCL)
thickness obtained by OCT in type 1 diabetic patients with and without clinically diagnosed retinopa-
thy.
PATIENTS AND METHODS
The study included 77 consecutive patients with type 1 diabetes (39 men, 38 women;
median age, 35 years [interquartile range (IQR), 29–42]; median disease duration, 10 years [IQR, 9–14]) and
31 age- and sex-matched controls. We measured RT in the fovea, parafovea, and perifovea, as well as RNFL
and GCL thickness. We divided diabetic patients into 2 subgroups, i.e., those with diabetic retinopathy and
without retinopathy.
RESULTS
We observed thicker perifoveal retina (P = 0.05), mean RNFL (P = 0.002), inferior RNFL
(P <0.0001), and superior and inferior GCL (P = 0.05 and P = 0.04, respectively) in diabetic subjects
compared with the control group. We detected retinopathy in 23 diabetic patients (29%). Compared with
patients without retinopathy, subjects with retinopathy had thinner parafoveal retina (P = 0.05), mean
RNFL (P = 0.002), inferior and nasal RNFL (P = 0.002, P = 0.03), superior (P = 0.05) and inferior GCL
(P = 0.006). Significant correlations were found between duration of diabetes and nasal RNFL thickness
(r = –0.32, P = 0.004) and parafoveal RT (r = –0.47, P <0.001).
CONCLUSIONS
The results might suggest the loss of intraretinal neural tissue in type 1 diabetic patients with
retinopathy. Neurodegeneration in diabetic retinopathy is closly associated with disease duration.
KEY WORDS
diabetic retinopathy,
neurodegeneration,
optical coherence
tomography, retinal
thickness, type 1
diabetes
ORIGINAL ARTICLE
Neurodegeneration of the retina in type 1
diabetic patients
Aleksandra Araszkiewicz
1
, Dorota Zozulińska‑Ziółkiewicz
1
,
Mikołaj Meller
2
, Jadwiga Bernardczyk‑Meller
2
, Stanisław Piłaciński
1
,
Anita Rogowicz‑Frontczak
1
, Dariusz Naskręt
1
, Bogna Wierusz‑Wysocka
1
1 Department of Internal Medicine and Diabetology, Poznan University of Medical Sciences, Poznań, Poland
2 OCU SERVICE Private Ophthalmology Unit, Poznań, Poland

ORIGINAL ARTICLE Neurodegeneration of the retina in type 1 diabetic patients 465
(Bio‑Rad Laboratories, Hercules, California, Unit‑
ed States). e clinical characteristics of the study
group are presented in TABLE 1.
During ophthalmological examination, pa‑
tients were asked about previous ocular diseas‑
es, visual symptoms, and ophthalmic treatment.
Best‑corrected visual acuity was measured using
the standard methods. e pupils were dilated
on both eyes using 1% tropicamide and 10% phe‑
nylephrine eye drops. Fundus examinations were
performed using ophthalmoscopy as well as slit
lamp and indirect Volk lens. Subsequently, using
a 45‑degree digital camera VISUCAM (Zeiss, Ger‑
many), 2 fundus photographs were taken of each
eye, one centered on the fovea and the other on
the optic disc. e evaluation of opthalmoscopy
results and fundus photographs was performed
for the entire group by the same ophthalmologist
with experience in diabetic retinopathy.
Diabetic retinopathy was graded according to
the classification of the American Academy of
Ophthalmology as no retinopathy, mild nonprolif‑
erative, moderate nonproliferative, severe nonpro‑
liferative, and proliferative retinopathy.12 We di‑
vided diabetic patients into 2 subgroups: with di‑
abetic retinopathy and without retinopathy ac‑
cording to clinical examination and fundus photo‑
graphs. One microaneurysm was enough to assign
a patient to the subgroup with retinopathy.
We performed OCT in the study group with
the RTVue model version 3.5 (Optovue Inc., Can‑
ada). RTVue OCT is a modern equipment with ul‑
tra‑high speed (26,000 A‑scn/s), high scan resolu‑
tion (5 µm), and high transverse scan resolution
(15 µm). As Biallosterski et al.7 showed a signifi‑
cant correlation between RT in the left and right
eye of the study population, we chose the right
eyes of the patients and performed 5 scans in
each.
7
ere were 3 retinal scan patterns: high‑
‑definition B‑scan, horizontal and vertical scans,
and dense thickness/elevation maps. We mea‑
sured RT: in the fovea (the central circle of 1 mm
in diameter), in the parafovea (the circle of 1
mm in the inner diameter and 3 mm in the out‑
er diameter), and in the perifovea (the circle of
3 mm in the inner diameter and 6 mm in the out‑
er diameter). ere were 2 glaucoma scan pat‑
terns: 3.45 mm RNFL thickness and GCL thick‑
ness. e measurements of RNFL thickness were
obtained along a 360° path for 4 quadrants of
the optic disc and as a mean value. e quadrants
were defined as follows: superior (46°–135°), in‑
ferior (226°–315°), temporal (316°–45°), and na‑
sal (136°–225°).
Statistical analysis
Statistical analysis was per‑
formed using the Statistica PL version 8.0 (Stat
Soft Inc., Tulsa, United States). e results of
continuous variables are shown as median val‑
ues and IQR or the number and percentage of
patients. Comparisons between the subgroups
with and without retinopathy were performed
using the t test for normally distributed vari‑
ables, the Mann‑Whitney U test for continuous
e aim of the study was to compare RT, reti‑
nal nerve fiber layer (RNFL) thickness, and gan‑
glion cell layer (GCL) thickness measured by OCT
in type 1 diabetic patients with and without retin‑
opathy. Moreover, we assessed the potential cor‑
relations between RT and metabolic parameters
as well as duration of the disease.
PATI ENTS AND M ETHO DS We recruited 77 consec‑
utive patients with type 1 diabetes (39 men and 38
women), hospitalized in the Department of Inter‑
nal Medicine and Diabetology in Poznań, Poland.
e patients were admitted to the hospital for edu‑
cation, adjustment of appropriate insulin dose, and
assessment of late diabetic complications. e me‑
dian age of the patients was 35 years (interquartile
range [IQR], 29–42); median disease duration was
10 years (IQR, 9–14). All subjects were informed
about the aim of the study and gave their written
consent. e study was conducted according to
the guidelines of the Helsinki Declaration and was
approved by the local Ethics Committee.
e exclusion criteria were as follows: prolifer‑
ative retinopathy after laser treatment, myopia
bigger than 2 diopters, and diabetic clinical signs
of macular edema observed on ophthalmoscopy.
A total of 31 age‑ and sex‑matched controls were
recruited from the staff members of the ophthal‑
mology clinic and their families. Controls had
no history of refractive errors, diabetes, or oth‑
er chronic diseases.
e participants completed a standardized
questionnaire including data on sex, age, medi‑
cal history, duration of diabetes, treatment, smok‑
ing status, and blood glucose self‑control. All pa‑
tients underwent complete physical examination
with anthropometric and blood pressure measure‑
ments. Blood samples were collected in a fasting
state after a period of rest, with minimal occlu‑
sion of the vein using the S‑Monovette blood col‑
lection system. Hemoglobin A
1c
(HbA
1c
) was mea‑
sured using high‑performance liquid chromatog‑
raphy with the Variant Hemoglobin A1c Program
TABLE 1 Clinical characteristics of the study group
Characteristics Diabetic patients,
n = 77
Healthy subjects,
n = 31
P
age, y 35 (29–42) 46 (25–50) 0.33
sex, male/female 39/38 13/18 0.52
duration of diabetes, y 10 (9–14) – –
smoking 20 (26) 8 (26) 0.99
SBP, mmHg 120 (110–130) 120 (110–128) 0.82
DBP, mmHg 76 (70–80) 75 (70–80) 0.81
HbA1c, % 7.9 (7.0–9.2) – –
retinopathy 23 (29) – –
nephropathy 14 (18) – –
neuropathy 15 (19) – –
Data are presented as median (IQR) or number (%) of patients.
Abbreviations: DBP – diastolic blood pressure, HbA1c – hemoglobin A1c, IQR – inter-
quartile range, SBP – systolic blood pressure

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ
2012; 122 (10)
466
RT, RNFL thickness, and GCL thickness in diabet‑
ic patients are shown in TABLE 3. Compared with
patients without retinopathy, subjects with retin‑
opathy had thinner parafoveal retina (P = 0.05),
mean RNFL (P = 0.002), inferior and nasal RNFL
(P = 0.002, P = 0.03), superior (P = 0.05) and in‑
ferior GCL (P = 0.006).
We also compared diabetic subjects without
retinopathy with the control group (
TABLE 3
). Sub‑
jects without retinopathy had thicker perifove‑
al retina (P = 0.048), mean RNFL (P <0.0001),
inferior RNFL (P <0.0001), as well as superi‑
or and inferior GCL (P = 0.007 and P = 0.003,
respectively).
Associations between clinical characteristics and
retinal thickness, retinal nerve fiber layer thickness,
and ganglion cell layer thickness RT, RNFL thick‑
ness, and GCL thickness in diabetic patients were
variables with skewed distributions, and the Fish‑
er’s exact test for categorical data. Pearson’s cor‑
relation coefficients were calculated to assess
the association between the continuous vari‑
ables. All tests were performed with the signif‑
icance level of 0.05 (two‑sided).
RESULTS Comparison between diabetic subjects
and controls
We observed thicker perifoveal ret‑
ina (P = 0.05), mean RNFL (P = 0.002), inferior
RNFL (P <0.0001), and superior and inferior GCL
(P = 0.05 and P = 0.04, respectively) in diabetic
subjects compared with controls (TABLE 2).
Com parison between type 1 d iab etic pa tients with and
without retinopathy We detected retinopathy in
23 diabetic patients (29%); 19 patients had mild
nonproliferative retinopathy and 4 subjects mod‑
erate nonproliferative retinopathy. e results of
TABLE 2 Comparison of retinal thickness, retinal nerve fiber layer thickness, and ganglion cell layer thickness between
diabetic patients and healthy subjects
Diabetic patients,
n = 77
Healthy subjects,
n = 31
P
RT
foveal 270 (254–288) 271 (254–280) 0.38
parafoveal 333 (321–341) 328 (321–333) 0.15
perifoveal 310 (301–317) 305 (296–309) 0.05
RNFL
mean 112.9 (105.8–119.5) 105.9 (97.7–113.1) 0.002
superior 132.7 (123.2–145.2) 132.0 (114.7–140.7) 0.30
inferior 156.2 (143.0–165.5) 135.7 (125.7–144.7) <0.0001
temporal 83.6 (77.7–90.7) 79.7 (74.0–89.2) 0.35
nasal 78.0 (66.7–85.5) 77.7 (73.2–85.7) 0.50
GCL superior 102.1 (95.9–106.9) 99.1 (93.7–101.4) 0.05
inferior 102.8 (98.1–108.2) 100.1 (95.2–103.6) 0.04
Data are presented as median (IQR), µm.
Abbreviations: GCL – ganglion cell layer, RNFL – retinal nerve fiber layer, RT – retinal thickness, others – see TABLE 1
TABLE 3 Comparison of retinal thickness, retinal nerve fiber layer thickness, and ganglion cell layer thickness between healthy subjects, diabetic
patients without retinopathy, and diabetic patients with retinopathy
Healthy subjects,
n = 31
Diabetes without
retinopathy,
n = 54
P
aDiabetes with retinopathy,
n = 23
P
b
RT
foveal 271 (254–280) 271 (258–290) 0.22 261 (250–285) 0.25
parafoveal 328 (321–333) 335 (322–342) 0.054 329 (313–336) 0.05
perifoveal 305 (296–309) 310 (301–321) 0.048 310 (298–316) 0.52
RNFL
mean 105.9 (97.7–113.1) 115.2 (108.8–120.8) <0.0001 109.4 (101.1–115.1) 0.002
superior 132.0 (114.7–140.7) 133.7 (126.0–149.6) 0.11 131.5 (118.2–141.7) 0.08
inferior 135.7 (125.7–144.7) 162.5 (149.0–167.7) <0.0001 148.2 (138.0–161.0) 0.002
temporal 79.7 (74.0–89.2) 85.0 (79.1–91.5) 0.25 80.8 (72.5–89.5) 0.21
nasal 77.7 (73.2–85.7) 79.2 (72.2–87.7) 0.69 72.0 (64.7–81.5) 0.03
GCL superior 99.1 (93.7–101.4) 102.5 (96.6–107.1) 0.007 97.6 (92.1–104.2) 0.05
inferior 100.1 (95.2–103.6) 104.8 (99.7–109.6) 0.003 98.8 (93.9–104.6) 0.006
Data are presented as median (IQR), µm.
a differences between diabetic patients without retinopathy and healthy subjects
b differences between diabetic patients with retinopathy and without retinopathy
Abbreviations: see TABLES 1 and 2

ORIGINAL ARTICLE Neurodegeneration of the retina in type 1 diabetic patients 467
neurodegeneration in diabetic retina. In diabet‑
ic patients, increased retinal vascular permeabil‑
ity related to hyperglycemia leads to the leakage
of serum proteins and lipids into the intraretinal
space.
13
is may result in higher values observed
on OCT in diabetic patients compared with con‑
trols, as observed in our study. However, in sub‑
jects with recognized retinopathy, we also noted
significant thinning of the retina. Several studies
conducted in experimental animal models have
recently indicated that neuroglial tissue loss may
occur at the early stages of diabetic retinopathy
and even precede vascular changes.8,14 ,15 It has also
been postulated that diabetic retinopathy should
be considered as a disease that involves vascu‑
lar pathology and retinal neurodegeneration.
8
OCT is a noninvasive and sensitive method that
might help identify the thinning of particular ret‑
inal layers. Segmentation of the intraretinal lay‑
ers obtained by OCT could lead to an earlier de‑
tection of diabetic retinal damage and facilitate
the understanding of its pathogenesis. Cabrera
et al.8 showed reduced RNFL and GCL thickness
in diabetic patients with mild retinopathy com‑
pared with subjects without retinopathy. However,
the study group of Cabrera et al.8 was not homo‑
geneous – there was a wide range of patients’ age.
ere have been studies on type 2 diabetic sub‑
jects that revealed RNFL defect in patients with
early diabetic retinopathy.
3
However, the studies
using OCT in type 2 diabetes focused mostly on
the assessment of macular edema.
16‑18
e knowl‑
edge on clinical usefulness of RT measurement in
type 1 diabetes is still limited. However, we would
like to emphasize that the group of type 1 diabet‑
ic subjects seems to be much more homogenous
than that of type 2 diabetic patients with well‑
‑defined onset of the disease. Ciresi et al.19 found
no difference between type 1 diabetic patients
with and without diabetic retinopathy and
the control group for all OCT parameters. e au‑
thors suggested that retinopathy without mac‑
ular edema in type 1 diabetic patients cannot be
detected with OCT.
19
On the other hand, Bial‑
losterski et al.
7
showed significantly decreased
pericentral RT in patients with retinopathy com‑
pared with controls. Similarly, Van Dijk et al.
20
compared type 1 diabetic subjects with retinop‑
athy with the control group and showed thinning
of the total retina. We detected the thinning of
the retina and of particular neuroglial layers in
type 1 diabetic subjects with retinopathy com‑
pared with those without retinopathy.
Interestingly, we observed negative correla‑
tions between all studied OCT parameters and
the duration of the disease. e results are con‑
sistent with those reported by Asefzadeh et al.21
(who, however, investigated type 2 diabetes) and
those reported by Biallosterski et al.7 (whose
study involved type 1 diabetic patients). Simi‑
larly, in the study by Chihara et al.,3 the risk fac‑
tors for nerve fiber layer thinning were the de‑
gree of diabetic retinopathy, high systolic blood
pressure, and patient’s age, but not HbA
1c
levels.
3
analyzed along with various clinical parameters.
A significant correlations were found between
the duration of diabetes and nasal RNFL thick‑
ness (r = –0.32, P = 0.004) and parafoveal RT
(r = –0.47, P <0.001) (FIGURES 1 and 2).
ere were no significant correlations between
RT, RNFL thickness, and GCL thickness and gly‑
cemic control of diabetes.
DISCUSSION
e main finding of the study is
that RT measured by OCT is higher in type 1 di‑
abetic patients compared with controls. Interest‑
ingly, however, the retina becomes thinner if di‑
abetic retinopathy is present. e results of our
study show that OCT could help identify early
changes in the neural layers of the retina in di‑
abetic patients. e measurement of RNFL and
GCL thickness could serve as the early sign of
40
50
60
70
90
80
100
110
120
30
duration of diabetes (years)
nasal RNFL (mm)
0 5 10 15 20 25 30 35
240
260
280
300
340
320
360
380
220
duration of diabetes (years)
parafoveal RT (mm)
0 5 10 15 20 25 30 35
FIGURE 2 Correlation between parafoveal retinal thickness (RT) of the right eye and
duration of diabetes; Pearson’s r = –0.47, P <0.001
FIGURE 1 Correlation between retinal nerve fiber layer (RNFL) thickness in the nasal
quadrant of the right eye and duration of diabetes; Pearson’s r = –0.32, P = 0.004

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ
2012; 122 (10)
468
before the clinical onset visible on fundus exam‑
ination.30 Moreover, we cannot exclude that oth‑
er mechanisms leading to increased permeabili‑
ty of the endothelium could be stimulated early
in diabetes. e formation of advanced glycation
end‑products and oxidative stress associated with
hyperglycemia lead to the thickening of the capil‑
lary basement membrane and pericyte loss, pre‑
ceding clinically visible retinopathy.31,32
e study has several limitations. First, we ob‑
served thinner retina in diabetic patients with
retinopathy only in some quadrants and layers.
Although the results of OCT seem to be repro‑
ducible, the future studies are needed with more
than 1 OCT measurement in the same group of
patients. Second, based on our results, the pro‑
cess of retinal thinning seems to be selective and
limited. e nature of this process in type 1 dia‑
betes requires further research.
In conclusion, the results might suggest
the loss of intraretinal neural tissue in type 1 di‑
abetic patients with retinopathy. Neurodegener‑
ation in diabetic retinopathy is strongly associat‑
ed with disease duration. OCT might be valuable
in the assessment of diabetic retinopathy.
REFERENCES
Cheung N, Wong TY. Diabetic retinopathy and systemic vascular compli-
1
cations. Prog Retin Eye Res. 2008; 27: 161-176.
Photocoagulation for diabetic macular edema. Early Treatment Diabet-
2
ic Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy
Study Research Group. Arch Ophthalmol. 1985; 103: 1796-1806.
Chihara E, Matsuoka T, Ogura Y, Matsumura M. Retinal nerve fiber lay-3
er defect as an early manifestation of diabetic retinopathy. Ophthalmology.
1993; 100: 1147-1151.
Gardner TW, Antonetti DA, Barber AJ, et al. Diabetic retinopathy: more 4
than meets the eye. Surv Ophthalmol. 2002; 47: S253-S262.
Peng PH, Lin HS, Lin S. Nerve fibre layer thinning in patients with pre-5
clinical retinopathy. Can J Ophthalmol. 2009; 44: 417-422.
Park SH, Park JW, Park SJ, et al. Apoptotic death of photoreceptors
6
in the streptozotocin-induced diabetic rat retina. Diabetologia. 2003; 46:
1260-1268.
Biallosterski C, van Velthoven M, Michels R, et al. Decreased optical co-
7
herence tomography-measured pericentral retinal thickness in patients with
diabetes mellitus type 1 with minimal diabetic retinopathy. Br J Ophthalmol.
2007; 91: 1135-1138.
Cabrera DeBuc D, Somfai GM. Early detection of retinal thickness chang-
8
es in diabetes using Optical Coherence Tomography. Med Sci Monit. 2010;
16: MT15-M21.
Ceklic L, Maár N, Neubauer AS. Optical coherence tomography fast ver-9
sus regular macular thickness mapping in diabetic retinopathy. Ophthalmic
Res. 2008; 40: 235-240.
Bressler NM, Edwards AR, Antoszyk AN, et al.; Diabetic Retinopathy 10
Clinical Research Network. Retinal thickness on Stratus optical coherence
tomography in people with diabetes and minimal or no diabetic retinopathy.
Am J Ophthalmol. 2008; 145: 894-901.
Yeung L, Lima VC, Garcia P, et al. Correlation between spectral domain
11
optical coherence tomography findings and fluorescein angiography patterns
in diabetic macular edema. Ophthalmology. 2009; 116: 1158-1167.
Wilkinson CP, Ferris FL 3rd, Klein RE, et al.; Global Diabetic Retinopa-12
thy Project Group. Proposed international clinical diabetic retinopathy and di-
abetic macular edema disease severity scales. Ophthalmology. 2003; 110:
1677-1182.
Ehrlich R, Harris A, Ciulla TA, et al. Diabetic macular oedema: physical,
13
physiological and molecular factors contribute to this pathological process.
Acta Ophthalmol. 2010; 88: 279-291.
Barber AJ. A new view of diabetic retinopathy: a neurodegenerative 14
disease of the eye. Prog Neuropsychopharmacol Biol Psychiatry. 2003; 27:
283-290.
Martin PM, Roon P, Van Ells TK, et al. Death of retinal neurons in strep -15
tozotocin-induced diabetic mice. Invest Ophthalmol Vis Sci. 2004; 45:
3330-3336.
It seems that degeneration of the neurons and
ganglion cells is a gradual process, which pccurs
over time. In the present study, we did not find
any correlations between RT, RNFL thickness,
and GCL thickness and glycemic control of dia‑
betes. However, it is possible that HbA1c reflects
only the mean values of glycemia from the last
3 months without showing the fluctuations of gly‑
cemia, and it is just not the perfect determinant
of good metabolic control.
22
Moreover, the acti‑
vation of adenosine monophosphate‑activated
protein kinase and metabolic stress in diabet‑
ic patients probably occurs as a result of hyper‑
glycemia, hypoglycemia, and hypoxia.23 ere is
strong evidence that the combination of high
metabolic demand and minimal vascular supply
may limit the ability of intraretinal neural tis‑
sue to adapt to the metabolic stress of diabetes.24
ese aspects may partially explain the patho‑
genesis of neurodegeneration as an additional
component to microvascular pathomechanism
of diabetic retinopathy as well as the lack of cor‑
relation between HbA1c and OCT parameters ob‑
served in our study.
e results of RT measurements in diabetic pa‑
tients without retinopathy compared with healthy
subjects presented in the literature are inconsis‑
tent. A number of studies have been conduct‑
ed in type 2 diabetes and the results cannot be
compared to type 1 diabetes. Asefzadeh et al.
21
performed a study in type 2 diabetic patients
and found no significant differences between
RT in controls, in subjects with mild retinopathy,
and those without retinopathy.
21
Van Dijk et al.
20
showed no statistically significant differences in
RT between diabetic patients without retinopa‑
thy and the control group.
20
Similarly to our study,
Biallosterski et al.
7
divided patients into 3 sub‑
groups, but they only found a difference in RT
in the pericentral ring between diabetic patients
with retinopathy and the control group.
7
Howev‑
er, there are some data suggesting the thinning of
the retina even in type 1 diabetic subjects with‑
out retinopathy. Lopes de Faria et al.
25
showed
significant nerve fiber loss in some segments of
the retina only in 12 patients without retinopa‑
thy compared with controls.25 e inconsistent
data might result from the early functional and
hemodynamic changes in the retina observed as
a result of hyperglycemia. ere is evidence of vas‑
cular dysfunction and abnormal autoregulation
of retinal circulation in diabetes that can lead to
retinal hyperperfusion.
26
is high level of reti‑
nal perfusion is assumed to induce endothelial
damage and increased permeability of the capil‑
laries due to increased shear stress.
27, 28
is hy‑
pothesis is in line with the observation that sys‑
temic hypertension increases the frequency and
rate of progression of diabetic retinopathy.
29
is
might result in the accumulation of extracellular
fluid and retinal thickening observed in OCT in
the group without retinopathy. It has been shown
that the total retinal blood flow and blood veloci‑
ty are increased in early diabetic retinopathy even

ORIGINAL ARTICLE Neurodegeneration of the retina in type 1 diabetic patients 469
Lattanzio R, Brancato R, Pierro L et al. Macular thickness measured by
16
optical coherence tomography (OCT) in diabetic patients. Eur J Ophthalmol.
2002; 12: 482-487.
Goebel W, Kretzchmar-Gross T. Retinal thickness in diabetic retinopa-17
thy: a study using optical coherence tomography (OCT). Retina. 2002; 22:
759-767.
Yang CS, Cheng CY, Lee FL, et al. Quantitative assessment of retinal
18
thickness in diabetic patients with and without clinically significant mac-
ular edema using optical coherence tomography. Acta Ophthalmol Scand.
2001; 79: 266-270.
Ciresi A, Amato MC, Morreale D, et al. OCT is not useful for detec-
19
tion of minimal diabetic retinopathy in type 1 diabetes. Acta Diabetol. 2010;
47: 259-263.
Van Dijk HW, Kok PH, Garvin M, et al. Selective loss of inner retinal lay-
20
er thickness in type 1 diabetic patients with minimal diabetic retinopathy.
Invest Ophthalmol Vis Sci. 2009; 50: 3404-3409.
Asefzadeh B, Fisch BM, Parenteau CE, Cavallerano AA. Macular thick-21
ness and systemic markers for diabetes in individuals with no or mild diabet-
ic retinopathy. Clin Experiment Ophthalmol. 2008; 36: 455-463.
Homa K, Majkowska L. Difficulties in interpreting HbA(1c) results.
22
Pol Arch Med Wewn. 2010; 120: 148-154.
Hardie DG, Hawley SA, Scott JW. AMP-activated protein kinase – de-23
velopment of the energy sensor concept. J Physiol. 2009; 574: 7-15.
Antonetti DA, Barber AJ, Bronson SK, et al.; JDRF Diabetic Retinopa-24
thy Center Group. Diabetic retinopathy: seeing beyond glucose-induced mi-
crovascular disease. Diabetes. 2006; 55: 2401-2411.
Lopes de Faria JM, Russ H, Costa VP. Retinal nerve fibre layer loss in 25
patients with type 1 diabetes mellitus without retinopathy. Br J Ophthalmol.
2002; 86: 725-728.
Pemp B, Polska E, Garhofer G, et al. Retinal blood flow in type 1 dia-26
betic patients with no or mild diabetic retinopathy during euglycemic clamp.
Diabetes Care. 2010; 33: 2038-2042.
Schmetterer L, Wolzt M. Ocular blood flow and associated functional 27
deviations in diabetic retinopathy. Diabetologia. 1999; 42: 387-405.
Kohner EM, Patel V, Rassam SM. Role of blood flow and impaired au-28
toregulation in the pathogenesis of diabetic retinopathy. Diabetes. 1995; 44:
603-607.
Knowler WC, Bennett PH, Ballintine EJ. Increased incidence of retinop-
29
athy in diabetics with elevated blood pressure. A six-year follow-up study in
Pima Indians. N Engl J Med. 1980; 302: 645-650.
Grunwald JE, DuPont J, Riva CE. Retinal haemodynamics in patients 30
with early diabetes mellitus. Br J Ophthalmol. 1996; 80: 327-331.
Arden GB, Sivaprasad S. Hypoxia and oxidative stress in the causation
31
of diabetic retinopathy. Curr Diabetes Rev. 2011; 7: 291-304.
Samborski P, Naskręt D, Araszkiewicz A, et al. Assessment of skin
32
autofluorescence as a marker of advanced glycation end product accumula-
tion in type 1 diabetes. Pol Arch Med Wewn. 2011; 121: 67-72.

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ
2012; 122 (10)
470
Adres do korespondencji:
dr med. Aleksandra Araszkiewicz,
Katedra i Klinika Chorób
Wewnętrznych i Diabetologii,
Uniwersytet Medyczny
im. K. Marcinkowskiego,
ul. Mickiewicza 2, 60-834 Poznań,
tel./fax: 61‑847‑45‑79,
e‑mail: olaaraszkiewicz@interia.pl
Praca wpłynęła: 20.07.2012.
Przyjęta do druku: 22.08.2012.
Publikacja online: 22.08.2012.
Nie zgłoszono sprzeczności
interesów.
Pol Arch Med Wewn. 2012;
122 (10): 464‑470
Copyright by Medycyna Praktyczna,
Kraków 2012
STRESZCZENIE
WPROWADZENIE Ostatnio podkreśla się rolę zwyrodnienia neuronów siatkówki oraz komórek glejowych
w patogenezie retinopatii cukrzycowej. Za pomocą optycznej koherentnej tomografii (optical coherence
tomography – OCT) można wykonywać jakościowe i ilościowe pomiary grubości siatkówki (retinal
thickness – RT) z identyfikacją poszczególnych warstw siatkówki.
CELE
Porównanie RT, grubości warstwy włókien nerwowych siatkówki (retinal nerve fiber layer – RNFL)
oraz warstwy komórek zwojowych (ganglion cell layer – GCL) mierzonych za pomocą OCT u chorych
na cukrzycę typu 1 bez retinopatii oraz z klinicznie rozpoznaną retinopatią.
PACJENCI I METODY
Do badania włączono 77 kolejnych chorych na cukrzycę typu 1 (39 mężczyzn,
38 kobiet; mediana wieku – 35 lat [rozstęp międzykwartylowy (interquartile range – IQR): 29–42]; mediana
czasu trwania cukrzycy – 10 lat [IQR: 9–14]) oraz 31 osób zdrowych dobranych pod względem wieku
i płci do grupy badanej. Zmierzono RT w centrum dołka, okołodołkowo i pozadołkowo, a także grubość
RNFL i GCL. Podzielono chorych na cukrzycę na 2 podgrupy: z retinopatią cukrzycową i bez retinopatii.
WYNIKI Stwierdzono grubszą siatkówkę okołodołkowo (p = 0,05), średnią RNFL (p = 0,002), RNFL
w dolnym kwadrancie (p <0,0001) oraz GCL w górnym i dolnym kwadrancie (p = 0,05; p = 0,04) u cho-
rych na cukrzycę w porównaniu z grupą kontrolną. Retinopatię rozpoznano u 23 pacjentów z cukrzycą
(29%). W porównaniu z osobami bez retinopatii, u pacjentów z retinopatią stwierdzono cieńszą siatkówkę
okołodołkowo (p = 0,05), mniejszą średnią grubość RNFL (p = 0,002), mniejszą grubość RNFL w dolnym
i nosowym kwadrancie (p = 0,002, p = 0,03) oraz grubość GCL w górnym (p = 0,05) i dolnym kwadrancie
(p = 0,006). Stwierdzono istotne statystycznie korelacje między czasem trwania cukrzycy a grubością
RNFL w kwadrancie nosowym (r = –0,32; p = 0,004) i RT okołodołkowo (r = –0,47; p <0,001).
WNIOSKI
Wyniki mogą sugerować utratę komórek tkanki nerwowej w obrębie siatkówki u chorych
na cukrzycę typu 1 z retinopatią cukrzycową. Neurodegeneracja występująca w retinopatii cukrzycowej
jest ściśle związana z czasem trwania choroby.
Słowa kluczowe
cukrzyca typu 1,
grubość siatkówki,
neurodegeneracja,
optyczna koherentna
tomografia,
retinopatia
cukrzycowa
aRTYkuł oRYGINalNY
Neurodegeneracja siatkówki u chorych
na cukrzycę typu 1
Aleksandra Araszkiewicz
1
, Dorota Zozulińska‑Ziółkiewicz
1
,
Mikołaj Meller
2
, Jadwiga Bernardczyk‑Meller
2
, Stanisław Piłaciński
1
,
Anita Rogowicz‑Frontczak
1
, Dariusz Naskręt
1
, Bogna Wierusz‑Wysocka
1
1 Katedra i Klinika Chorób Wewnętrznych i Diabetologii, Uniwersytet Medyczny im. K. Marcinkowskiego, Poznań
2 Specjalistyczny Okulistyczny Niepubliczny Zakład Opieki Zdrowotnej „OCU SERVICE” Poznań