• cataract;
  • HbA1c;
  • heart rate;
  • lens opacity;
  • metabolic control;
  • type 2 diabetes


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Purpose:  To evaluate the prevalence and risk factors of lens opacities in a geographically defined population of subjects with type 2 diabetes mellitus compared with a control population.

Methods:  Subjects in the community of Laxå with a diagnosis of type 2 diabetes mellitus (n = 275) and a control group (n = 256) participated in the study. Lens opacities were graded with Lens Opacities Classification System II in all participants. Lens Opacities Classification System score ≥2 was considered as significant lens opacity. Anthropometric and blood chemistry data were collected for all participants in connection with the eye examination. For the diabetic population, yearly updated information on glucose control, blood pressure and body mass index was available through medical records from diabetes diagnosis until the time of the eye examination.

Results:  The prevalence of significant cortical, posterior subcapsular and nuclear cataract was 65.5%, 42.5% and 48.0%, respectively, in the type 2 diabetes population in Laxå. In logistic regression analyses, all types of lens opacities were strongly associated with age (p < 0.0001). Cortical lens opacity was also associated with a diagnosis of diabetes (p < 0.0001), posterior subcapsular lens opacity with HbA1c (p < 0.0001) and nuclear lens opacity with female gender and higher heart rate (both p = 0.0004). In the diabetic population, all types of cataract were likewise strongly associated with age (p < 0.0001), posterior subcapsular cataract with HbA1c (p = 0.0032), nuclear cataract with female gender (p = 0.0002) and higher heart rate (p = 0.0008).

Conclusions:  Our study shows that cortical cataract is associated with diabetes mellitus, not necessarily defined by glucose control, whereas posterior subcapsular cataract is associated with glucose levels. Nuclear cataract is not associated with diabetes mellitus, but is more frequent in women and is also associated with higher heart rate.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Cataract is a leading cause of blindness worldwide and a major public health problem. Age and diabetes mellitus are known risk factors (Pollreisz & Schmidt-Erfurth 2010). Patients with diabetes tend to benefit less from cataract surgery than those without diabetes (Stanga et al. 1999; Zaczek et al. 1999; Tranos et al. 2004; Schmier et al. 2007). Ageing populations and the increasing prevalence of diabetes mellitus create an ever-increasing prevalence of cataracts and health care demands. To meet this challenge, it is necessary to know more about the association between diabetes and the different subtypes of lens opacities.

Thus, we aimed to study the epidemiology and risk factors of cataract subgroups in type 2 diabetes mellitus in a defined geographical area compared with an age- and gender-matched control population from the same geographical area.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Study area

The population of Laxå, a rural community in the central part of southern Sweden, has been the subject of several studies on epidemiological aspects of type 2 diabetes during more than 30 years from the beginning of the 1970s (Jansson et al. 2007). A local diabetes register created in 1972 at the primary health care centre in Laxå is continuously updated as new patients with diabetes are found.

Classification of the diabetes population

All inhabitants in Laxå diagnosed with type 2 diabetes before 1 January 1997 were invited to participate in the study. The diagnosis of diabetes was accepted if the 1985 World Health Organization (WHO) criteria (WHO-Study-Group 1985) were fulfilled. Patients with type 2 diabetes were identified, as those with a disease onset at the age of 30 years or after, who did not have ketonuria at diagnosis or did not need prompt insulin treatment. Patients under 30 years of age were also classified as type 2 diabetic subjects, if they were well controlled with diet or oral hypoglycaemic agents for several years and had no islet cell antibodies. Patients with type 1 diabetes were identified as those with onset before the age of 30 years and who needed insulin treatment within 1 year after diagnosis or when onset after 30 years of age a clinical history of glucosuria or ketonuria and prompt insulin treatment led to diagnosis. Patients were also classified as having type 1 diabetes if they had a more slowly developing insulin-dependent diabetes with islet cell antibodies or very low C-peptide secretion on glucagon stimulation. Analyses of islet cell antibodies were liberally undertaken in patients with diabetes in Laxå.

Control population

An age- and gender-matched control group from the population of the Laxå municipality was recruited to the study. The control group was found through the national population register where nondiabetic persons, living in Laxå, of the same sex and closest in age to each diabetic person were chosen. As a result of difficulties in finding enough matching control subjects for the very old diabetic persons, only one control person per diabetic subject was chosen. The research ethics committee at the Örebro County Council gave the study its approval.

Research methods

All participants – cross-sectional survey

An ophthalmologist examined all individuals during the time period from 23 July 1996 to 6 June 1998. Lens gradings were based on the Lens Opacities Classification System II (LOCS II) (Chylack et al. 1989) and performed with direct reference to photographic standards at the slit lamp, under dilatation. All levels of lens opacities were recorded. The presence of significant lens opacity was defined by a LOCS II score ≥2. The criteria did not consider whether there was another type of cataract also present in the same or contralateral eye. If there were lens opacities of the same type in both the eyes, the worst eye was considered. If only one eye could be judged, this has been considered to be the worst eye.

All participants of the study were asked to answer a questionnaire. Among the questions asked were smoking habits and if and why eye surgery had been performed. Data on type of diabetes treatment were recorded from the diabetic patient’s files at the primary health care centre in Laxå at the time of the cross-sectional examination. Systolic and diastolic blood pressure and heart rate were measured in a lying position after five minutes of rest. The mean of three measurements was registered. Height was measured without shoes and weight with light clothing on. Body mass index (BMI) was calculated as weight in kilogram divided by height in metre, squared (kg/m2). Blood samples were taken to estimate fasting blood glucose, HbA1c, cholesterol, triglycerides, HDL cholesterol, creatinine, uric acid and gamma glutamyl transferase and analysed at the laboratory of the nearby local county hospital of Karlskoga. Enzymatic methods were used to estimate cholesterol, triglycerides, HDL cholesterol and uric acid. Creatinine was analysed using the Jaffé method and gamma glutamyl transferase according to Szasz-Persijn. These analyses were made on a Hitachi 911 instrument. HbA1c was analysed with a low-pressure chromatography method on Glycomat mass transfer chromatography (MTC).

Diabetic population – longitudinal data

Data from the type 2 diabetic population were collected both from the time of diagnosis of the diabetes, over time and, as mentioned earlier, at the cross-sectional examination. The longitudinal information consisted of continuously collected measurements of fasting blood glucose, systolic and diastolic blood pressure and BMI values from the time of diabetes diagnosis until the cross-sectional examination. These data were assembled from the patient files at the health care centre in Laxå. Yearly mean values for these variables were summarized into a total mean value (total mean fasting blood glucose, total mean systolic blood pressure, total mean diastolic blood pressure and total mean BMI) from diabetes diagnosis until end of follow-up at the cross-sectional survey.

Participants in the study

A total of 275 subjects had a diagnosis of type 2 diabetes mellitus. They consisted of all, except one, known type 2 diabetic patients in the municipality of Laxå, representing a crude prevalence of 3.9%. They were all included in the study but did not participate in all parts; 33 patients with type 1 diabetes, aged 15 years or more, were not enrolled in this study.

The eligible control subjects numbered 275. No control subject was chosen for a 24-year-old diabetic person because he was first considered to have type 1 diabetes. The negative result of the islet antibody test that rendered him a type 2 diabetes diagnosis arrived after the examinations of the control subjects were completed. As 19 subjects, who at first had accepted to participate as controls, could not participate, most often because of an acute disease afflicting themselves or close members of their family, 256 control subjects were finally slit lamp examined for cataract. Thus, 531 residents of Laxå, all of Caucasian origin, were enrolled in the study.

Of the 531 subjects, anthropometric data were missing for 16 patients and 10 control subjects. Twenty-six patients in the diabetes group and 17 subjects in the control group left a questionnaire regarding smoking and eye surgery unanswered. Reasons for the missing data were mainly high age and concomitant severe disease such as dementia, terminal malignant disease or sequels of cerebrovascular disease.

Statistical analyses

The data were analysed using the sas software release 9.2 (SAS Institute, Cary, NC, USA). Patients with a lens removed were regarded to have had this done because of cataract surgery and were included when the analyses applied to any type of cataract, but not when the subtypes of cataract were analysed as this information was unknown.

Possible relationships between characteristics of the participants and outcome, defined as different types of lens opacities classified both as LOCS II ≥ 2 and as an increasing degree of cataract (LOCS II 0–6), were analysed in two steps. To find candidate determinants for the final analysis model, a set of screening bivariate analyses (Spearman) of possible relations to outcome was performed. p-values <0.05 were accepted as indicating statistical significance in the bivariate screening analyses.

In the final model (multivariate logistic regression with backward elimination of nonsignificant variables), the outcome was entered as a dependent variable. All significant determinants for the different types of cataracts from the screening analyses were entered as independent variables. The procedure provides odds ratios (OR), confidence intervals (CI) and Wald’s chi-square estimates. Wald’s chi-square is the test parameter on which the p-value is based. As a consequence, Wald’s chi-square may be used to rank the impact or the importance of the independent variables. All tests were two-tailed. In the final analysis model, p-values <0.005 and 99.5% CIs were used to account for multiple testing.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Characteristics of the study participants

The characteristics of the study participants are shown in Table 1. Mean age at the time of examination by ophthalmologist was 69 years in the diabetes group (range 24–91 years) and 70 years in the control group (range 36–93 years). The male:female ratio in the two groups was very similar, 1.20 in the diabetes group and 1.13 in the control group. Significant differences between the two groups consisted of lower total cholesterol and HDL cholesterol and higher fasting blood glucose, HbA1c, triglycerides, weight, BMI, systolic blood pressure and heart rate among the patients with diabetes.

Table 1.   Characteristics of the study participants.
 No diabetesDiabetesp-value (diabetes versus no diabetes
Female, n = 120Male, n = 136Female, n = 125Male, n = 150Female, n = 245Male, n = 286
Mean or %SDMean or %SDMean or %SDMean or %SDp-valuep-value
  1. p-values in bold denotes p < 0.05.

Age (year)71.012.869.012.470.312.0668.112.230.50570.5697
Diabetes duration (year)NANANANA8.96.319.47.71NANA
Total mean fasting blood glucose (mm)NANANANA8.11.967.81.88NANA
Total mean systolic blood pressure (mmHg)NANANANA158.612.58153.115.48NANA
Total mean diastolic blood pressure (mmHg)NANANANA87.24.8987.15.72NANA
Total mean body mass index (kg/m2)NANANANA30.05.7928.54.48NANA
Type of diabetes treatment
 Diet only (%)NANANANA36.00.4840.00.49NANA
 Oral hypoglycaemic agents (OHA) (%)NANANANA36.00.4835.30.48NANA
 Insulin with or without OHA (%)NANANANA28.00.4524.70.43NANA
Current smoker (%)12.40.3312.70.3313.50.3415.90.370.80320.4552
Fasting blood glucose (mm)4.70.574.60.547.72.518.13.04<0.0001<0.0001
HbA1c (%)4.50.584.20.576.41.306.51.57<0.0001<0.0001
Total cholesterol (mm)
HDL cholesterol (mm)1.50.391.30.381.30.401.20.43<0.00010.0009
Triglycerides (mm)1.60.751.<0.00010.0013
Creatinine (μm)89.621.1199.815.8690.026.42102.935.850.40220.7020
Uric acid (μm)312.786.39365.491.33367.1112.14365.396.41<0.00010.9594
Gamma glutamyl transferase (μm)0.50.570.70.630.70.641.01.29<0.00010.1189
Height (cm)160.15.97173.46.61160.26.33173.06.260.80550.8148
Weight (kg)68.011.2878.211.4076.316.5085.616.740.00020.0002
Body mass index (kg/m2)26.64.1926.03.2629.75.7828.54.84<0.0001<0.0001
Systolic blood pressure (mmHg)150.025.81142.920.24156.320.78150.320.370.01350.0011
Diastolic blood pressure (mmHg)81.310.5682.79.7582.87.3484.49.320.19610.0802
Heart rate (beats/min)71.69.7465.010.2073.912.8870.612.200.26990.0001

In the diabetes group, 38% had diet only as their diabetes treatment, while 36% had one or more oral hypoglycaemic agents and 26% insulin with or without hypoglycaemic agents at the cross-sectional survey. The mean diabetes duration was 9 years (range from less than half a year to 47 years). A total mean fasting blood glucose value of 7.9 mm, total mean systolic and diastolic blood pressure of 156 versus 87 mmHg and a total mean BMI of 29.2 kg/m2 represented the longitudinal data.

Twenty-three subjects in the diabetes population and 18 in the control population (p = 0.6) had one or both crystalline lenses removed. Table 2 shows prevalence and type of lens opacities stratified by age groups and gender. Male patients with type 2 diabetes had a significantly higher prevalence of cortical lens opacity (p = 0.0007), and female patients with type 2 diabetes posterior subcapsular lens opacity (p = 0.0220), compared with control subjects. We found no difference between the diabetes and the control group regarding nuclear lens opacity.

Table 2.   Number of subjects with different types of lens opacities among patients with type 2 diabetes and control subjects. Percentage in brackets.
Age (years)Type of lens opacityNo diabetes (%)Type 2 diabetes (%)p-value (diabetes versus no diabetes)
FemaleMaleFemaleMaleFemaleMaleBoth sexes
  1. * Includes subjects with lens removed.

<54 n = 14n = 23n = 13n = 23n = 27n = 46n = 73
 Cortical1 (7.14)1 (4.35)3 (23.08)5 (21.74)0.26110.08320.0371
Posterior subcapsular0 (0.00)0 (0.00)1 (7.69)1 (4.35)0.30860.32280.1501
Nuclear0 (0.00)0 (0.00)1 (7.69)1 (4.35)0.30860.32280.1501
Any type*1 (7.14)1 (4.35)3 (23.08)6 (26.09)0.26110.04090.0191
55–64 n = 22n = 22n = 24n = 30n = 46n = 52n = 98
 Cortical6 (27.27)4 (18.18)12 (50.00)18 (60.00)0.11980.00200.0008
Posterior subcapsular2 (9.09)1 (4.55)5 (20.83)8 (26.67)0.27920.03780.0214
Nuclear4 (18.18)3 (13.64)5 (20.83)2 (6.67)0.82570.40960.6822
Any type*10 (45.45)7 (31.82)13 (54.17)21 (70.00)0.56520.00570.0163
65–74 n = 33n = 46n = 36n = 51n = 69n = 97n = 166
 Cortical15 (45.45)24 (52.17)21 (58.33)35 (68.63)0.29160.09940.0515
Posterior subcapsular5 (15.15)14 (30.43)17 (47.22)17 (33.33)0.00380.76280.0382
Nuclear19 (57.58)18 (39.13)23 (63.89)16 (31.37)0.59780.42920.7969
Any type*24 (72.73)36 (78.26)32 (88.89)38 (74.51)0.08880.66840.4843
≥75 n = 51n = 45n = 52n = 46n = 103n = 91n = 194
 Cortical48 (94.12)32 (69.57)47 (90.38)39 (84.78)0.48400.11790.3835
Posterior subcapsular37 (72.55)24 (52.17)41 (78.85)27 (58.70)0.46100.61110.3910
Nuclear48 (94.12)39 (84.78)49 (94.23)35 (76.09)0.98070.19960.2925
Any type*51 (100.00)45 (100.00)52 (100.00)43 (93.48)NA0.08320.0849
All ages n = 120n = 136n = 125n = 150n = 245n = 286n = 531
 Cortical70 (58.33)61 (44.85)83 (66.40)97 (64.67)0.19390.00070.0008
Posterior subcapsular44 (36.67)39 (28.68)64 (51.20)53 (35.33)0.02200.23020.0161
Nuclear71 (59.17)60 (44.12)78 (62.40)54 (36.00)0.60600.16260.4660
Any type*86 (71.67)89 (65.44)100 (80.00)108 (72.00)0.12830.23300.0618

Table 3a,b present the results of the bivariate screening analyses between different types of lens opacities and determinants for the type 2 diabetes cohort and control cohort, respectively. No major differences were found when the bivariate analyses were made using all information of the grading in the LOCS II (data not shown).

Table 3.   Results of bivariate analyses between characteristics of the (a) type 2 diabetic subjects and different types of cataract (b) nondiabetic control subjects and different types of cataract.
 CorticalPosterior subcapsularNuclearAny type
Mean or %Mean or %p-valueMean or %Mean or %p-valueMean or %Mean or %p-valueMean or %Mean or %p-value
  1. p-values in bold denotes p < 0.05.

Age (year)61.473.2<0.000164.375.6<0.000162.176.6<0.000157.572.8<0.0001
Per cent female44.246.10.764438.654.70.007932.949.4<0.000137.348.10.1248
Diabetes duration (year)<0.00017.710.8<0.00017.09.90.0006
Longitudinal data
 Total mean fasting blood glucose (mm)7.638.060.18467.678.220.06657.728.120.11237.568.030.1547
 Total mean systolic blood pressure (mmHg)151.0157.90.0004152.5159.6<0.0001151.0160.5<0.0001149.6157.50.0002
 Total mean diastolic blood pressure (mmHg)87.387.10.743686.587.90.062587.486.90.357487.587.00.5296
 Total mean body mass index (kg/m2)30.428.70.005129.928.40.011130.428.10.000731.828.5<0.0001
Cross-sectional data
 Type of diabetes treatment
  Diet only (%)46.333.90.043946.826.50.000644.731.10.019550.734.10.0148
  Oral hypoglycaemic agents  (OHA) (%)33.736.70.624931.741.00.109135.735.60.992031.337.00.4007
  Insulin with or without  OHA (%)20.0029.40.090921.532.50.041119.633.30.009417.928.90.0771
 Current smoker (%)16.913.80.511217.011.80.254516.113.40.558214.315.10.8828
 Fasting blood glucose (mm)7.937.890.81457.688.210.09298.137.660.23297.667.980.7868
 HbA1c (%)6.396.480.36366.286.670.00756.546.340.52106.386.470.7336
 Total cholesterol (mm)5.605.850.07095.755.780.87555.755.780.77255.505.850.0275
 HDL cholesterol (mm)
 Triglycerides (mm)
 Creatinine (μm)89.9100.90.000893.0102.70.068193.1101.50.272189.399.60.0087
 Uric acid (μm)364.6366.90.8882357.0378.10.3133361.5371.10.6692351.6370.70.3854
 Gamma glutamyl transferase (μm)0.890.850.02800.830.910.79720.870.860.05540.810.880.3783
 Height (cm)168.7166.50.0459168.8165.20.0010168.9165.40.0007169.2166.60.0412
 Weight (kg)85.779.10.002084.776.7<0.000186.575.8<0.000190.278.6<0.0001
 Body mass index (kg/m2)<0.000131.528.2<0.0001
 Systolic blood pressure (mmHg)146.8156.20.0003148.6158.80.0001148.2158.2<0.0001145.1155.50.0002
 Diastolic blood pressure (mmHg)
 Heart rate (beats/min)
Age (year)62.576.9<0.000165.379.5<0.000161.078.4<0.000156.975.9<0.0001
Per cent female40.053.40.031443.953.00.174239.254.20.016142.049.10.2870
Current smoker (%)
Fasting blood glucose (mm)4.614.670.24824.644.650.92414.584.700.18314.594.670.1636
HbA1c (%)4.404.530.09564.454.500.52324.364.570.00454.414.490.2703
Total cholesterol (mm)
HDL cholesterol (mm)1.441.420.48071.451.390.48401.411.450.42271.421.430.9071
Triglycerides (mm)1.711.590.79141.641.680.14261.621.670.28251.601.670.1259
Creatinine (μm)
Uric acid (μm)327.4353.40.0432330.6362.20.0333325.8354.70.0219313.7353.00.0035
Gamma glutamyl transferase (μm)0.690.580.07730.650.590.09400.600.660.74370.650.620.4860
Height (cm)170.0164.4<0.0001168.2164.90.0091169.3165.10.0007169.9165.90.0015
Weight (kg)77.669.2<0.000175.568.9<0.000176.670.30.000277.771.40.0012
Body mass index (kg/m2)26.825.70.012226.725.30.006726.725.80.063126.925.90.1018
Systolic blood pressure (mmHg)141.5150.70.0002144.1150.60.1053138.9153.1<0.0001136.5150.6<0.0001
Diastolic blood pressure (mmHg)82.781.40.754083.279.40.014882.381.80.922581.882.10.5213
Heart rate (beats/min)66.369.90.014167.769.10.270865.670.50.000965.669.30.0352

When the significant variables for all study subjects were treated in a series of logistic regression analyses, age and a diagnosis of type 2 diabetes (both p < 0.0001) remained statistically significant for cortical lens opacity. For posterior subcapsular lens opacity, age and HbA1c (both p < 0.0001) and for nuclear lens opacity, age (p < 0.0001), female sex and heart rate (both p = 0.0004) remained statistically significant (Table 4). For any type of cataract, age (p < 0.0001) and HbA1c (p = 0.004) were statistically significant.

Table 4.   Multivariate regression analyses of the effects of significant characteristics of the whole study population/type 2 diabetes population on different types of lens opacities*.
 CorticalPosterior subcapsularNuclearAny type
OR CI (99.5%)χ2p-valueOR CI (99.5%)χ2p-valueOR CI (99.5%)χ2p-valueOR CI (99.5%)χ2p-value
  1. * According to Lens Opacities Classification System II score ≥2; χ= Wald chi-square estimate; OHA = oral hypoglycaemic agents.

Whole study population
 Age (years)1.11 (1.071.14)79.87<0.00011.12 (1.081.16)83.33<0.00011.21 (1.15–1.27)114.65<0.00011.18 (1.13–1.24)100.28<0.0001
 Diabetes (yes/no)2.62 (1.375.03)17.28<0.0001         
 HbA1c (%)   1.40 (1.121.75)18.44<0.0001   1.34 (1.01–1.78)8.260.0040
 Per cent female      2.54 (1.22–5.29)12.760.0004   
 Heart rate (beats/min)1.023 (0.996–1.05)5.540.0186   1.04 (1.01–1.08)12.770.00041.035 (1.00–1.07)7.750.0054
 Weight (kg)0.984 (0.96–1.01)4.750.0294      0.979 (0.96–1.01)5.230.0222
 Height (cm)   0.974 (0.94–1.01)4.620.0317      
Diabetes population
 Age (years)1.10 (1.051.15)37.42<0.00011.11 (1.051.16)36.35<0.00011.21 (1.121.31)50.05<0.00011.16 (1.091.23)46.26<0.0001
 HbA1c   1.40 (1.021.94)8.690.0032      
 Per cent female      3.40 (1.1210.31)9.540.0020   
 Heart rate (beats/min)      1.06 (1.011.11)11.210.0008   
 Height (cm)   0.962 (0.92–1.01)5.050.0246      
 Insulin treatment, with or without OHA (%)      3.06 (0.94–9.97)7.060.0079   
 Diastolic blood pressure, cross-sectional data, (mmHg)      0.95 (0.89–1.01)5.410.0200   
 Total mean fasting blood glucose, longitudinal data, (mm)         1.27 (0.92–1.75)4.420.0356

Restricting the logistic regression analyses to the diabetic population only, age (p < 0.0001) remained significant for cortical, age (p < 0.0001) and HbA1c (p = 0.0032) for posterior subcapsular, and age (p < 0.0001), female sex (p < 0.002) and a higher heart rate (p = 0.0008) for nuclear lens opacity (Table 4). Only age (p < 0.0001) was significant for any type of cataract among the type 2 diabetic subjects.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Age was, as expected, the strongest risk factor for all types of lens opacities both among patients with type 2 diabetes and a gender- and age-matched control group from the same geographical area. Besides age, in multivariate analyses, cortical lens opacity was significantly more common among patients type 2 diabetes. There was also a significant association between posterior subcapsular lens opacity and a higher HbA1c. Having nuclear lens opacity showed, however, no association with a diabetic state or glucose control, but was instead significantly related to female sex and surprisingly to a higher heart rate.

In the diabetic population, age was again the dominant significant determinant for all types of cataract. A significant association could also be seen between posterior subcapsular cataract and HbA1c. Nuclear cataract was significantly associated with female sex and higher heart rate.

Several studies have shown similar results, although the association between diabetes and different types of cataract is complex.

The Framingham study showed an increased prevalence of cataract in patients with diabetes (Klein et al. 1985). In the Wisconsin Epidemiologic Study of Diabetic Retinopathy, the risk of cataract surgery was higher for persons having type 2 diabetes using insulin (Klein et al. 1995a).

In the Beaver Dam Eye study, the incidence and progression of cortical and posterior subcapsular cataract were associated with diabetes. Further, HbA1c was associated with increased risk of nuclear and cortical cataract. In a further analysis in this study, longer duration of diabetes also increased the prevalence of cortical cataracts (Klein et al. 1995b, 1998).

In the cross-sectional Blue Mountains Eye Study, posterior subcapsular cataract, but not nuclear cataract, was significantly related to diabetes (Rowe et al. 2000). In a longitudinal study, cortical cataract was related to impaired fasting glucose (Saxena et al. 2004).

In the Barbados Eye study, the 4-year incidence of lens opacities was evaluated. The factors increasing the risk for cortical opacity were old age, female gender, low socioeconomic status and diabetes mellitus; for posterior subcapsular opacity were age and diabetes; and for nuclear opacity were age, female gender, leaner body mass and diabetes (Leske et al. 2002; Hennis et al. 2004).

The Los Angeles Latino Eye Study reported that all types of lens opacities increased with age and that nuclear opacity was more common in women (Varma & Torres 2004).

In the Reykjavik Eye Study, cortical lens opacity was related to age and outdoor exposure and nuclear lens opacity to age and to smoking. In this study, no association of lens opacities to diabetes was found, but the study included relatively few patients with diabetes (Sasaki et al. 2000; Katoh et al. 2001; Arnarsson et al. 2002).

The Tanjong Pagar Survey found diabetes to be related to cortical cataract, posterior subcapsular cataract, any cataract and cataract surgery (Foster et al. 2003).

In our study, the impact of diabetes on the prevalence of cataracts seems to be less important than in many of the studies cited earlier. Possible explanations of this could be the fact that our diabetic population was under good glucose control and most subjects had an early diagnosis of diabetes by opportunistic case finding. This probably led to less difference between the diabetic group and the control group concerning blood glucose levels. As our diabetic group also incorporated all elderly diabetic persons in the community, it might have augmented the impact of age on the occurrence of cataract and in doing so reducing the effect of other variables.

Although cortical and posterior subcapsular opacities showed similarities regarding the associations we found with a diagnosis of diabetes and metabolic control, there were also differences. For cortical cataract, it was having a diagnosis of diabetes, but not glucose control, that proved to be associated with our outcome variable, and for the diabetic population only, neither HbA1c nor fasting blood glucose measured cross-sectional or longitudinally was statistically significant. A possible interpretation of this finding can be that other conditions, tied to the diabetic state, than glucose control may play a role. With posterior subcapsular cataract, it was the other way round. Especially, not only among the whole study population but also among the diabetic population measures at the cross-sectional survey of glucose control were significantly associated with this kind of cataract. This result can indicate a role for glucose levels below the diabetes state in the development or worsening of posterior subcapsular cataract.

Finally, an unexpected result was seen for nuclear opacities. Others have shown that the associations were also found with age and female sex, but the strong and consistent association with a higher heart rate both among the whole study population and among the diabetic group is, to our knowledge, a new finding. If persistent in future studies, one may consider whether some age-related mechanism, other than chronological age, can be responsible or whether, perhaps, nuclear cataract and its relation to heart rate in adult life can be traced back to imperfect foetal development of certain organs, the so-called Barker hypothesis.

The strengths of our study are that we could examine and compare a complete type 2 diabetic population with a control group matched for age, gender and residency and that we also could use longitudinal data on continuously updated glucose, blood pressure and BMI values in the diabetic group.

On the other hand, our study is relatively small, and this may have prevented us from observing associations that would be significant in larger cohorts. Our findings are restricted to subjects of Caucasian origin.

Whether prevention and improved control of diabetes would reduce the burden of cataract remains to be demonstrated. Our data, together with others, indicate a possibility for such a scenario for cortical and posterior subcapsular cataract, but not for nuclear cataract. Future studies on the pathogenesis and epidemiology of cataracts must distinguish between the three types of lens opacities, as they clearly show different patterns.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study was supported by the Örebro County Council, Örebro, Sweden, The Research found of The National University Hospital, Reykjavik, Iceland and The Helga Jonsdottir and Sigurlidi Kristjansson Memorial Fund, Reykjavik, Iceland.


  1. Top of page
  2. Abstract.
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • Arnarsson A, Jonasson F, Sasaki H, Ono M, Jonsson V, Kojima M, Katoh N & Sasaki K (2002): Risk factors for nuclear lens opacification: the Reykjavik Eye Study. Dev Ophthalmol 35: 1220.
  • Chylack LT Jr, Leske MC, McCarthy D, Khu P, Kashiwagi T & Sperduto R (1989): Lens opacities classification system II (LOCS II). Arch Ophthalmol 107: 991997.
  • Foster PJ, Wong TY, Machin D, Johnson GJ & Seah SK (2003): Risk factors for nuclear, cortical and posterior subcapsular cataracts in the Chinese population of Singapore: the Tanjong Pagar Survey. Br J Ophthalmol 87: 11121120.
  • Hennis A, Wu SY, Nemesure B & Leske MC (2004): Risk factors for incident cortical and posterior subcapsular lens opacities in the Barbados Eye Studies. Arch Ophthalmol 122: 525530.
  • Jansson SP, Andersson DK & Svardsudd K (2007): Prevalence and incidence rate of diabetes mellitus in a Swedish community during 30 years of follow-up. Diabetologia 50: 703710.
  • Katoh N, Jonasson F, Sasaki H, Kojima M, Ono M, Takahashi N & Sasaki K (2001): Cortical lens opacification in Iceland. Risk factor analysis – Reykjavik Eye Study. Acta Ophthalmol Scand 79: 154159.
  • Klein BE, Klein R & Moss SE (1985): Prevalence of cataracts in a population-based study of persons with diabetes mellitus. Ophthalmology 92: 11911196.
  • Klein BE, Klein R & Moss SE (1995a): Incidence of cataract surgery in the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Am J Ophthalmol 119: 295300.
  • Klein BE, Klein R, Wang Q & Moss SE (1995b): Older-onset diabetes and lens opacities. The Beaver Dam Eye Study. Ophthalmic Epidemiol 2: 4955.
  • Klein BE, Klein R & Lee KE (1998): Diabetes, cardiovascular disease, selected cardiovascular disease risk factors, and the 5-year incidence of age-related cataract and progression of lens opacities: the Beaver Dam Eye Study. Am J Ophthalmol 126: 782790.
  • Leske MC, Wu SY, Nemesure B & Hennis A (2002): Risk factors for incident nuclear opacities. Ophthalmology 109: 13031308.
  • Pollreisz A & Schmidt-Erfurth U (2010): Diabetic cataract-pathogenesis, epidemiology and treatment. J Ophthalmol 2010: 608751.
  • Rowe NG, Mitchell PG, Cumming RG & Wans JJ (2000): Diabetes, fasting blood glucose and age-related cataract: the Blue Mountains Eye Study. Ophthalmic Epidemiol 7: 103114.
  • Sasaki H, Jonasson F, Kojima M, Katoh N, Ono M, Takahashi N & Sasaki K (2000): The Reykjavik Eye Study – prevalence of lens opacification with reference to identical Japanese studies. Ophthalmologica 214: 412420.
  • Saxena S, Mitchell P & Rochtchina E (2004): Five-year incidence of cataract in older persons with diabetes and pre-diabetes. Ophthalmic Epidemiol 11: 271277.
  • Schmier JK, Halpern MT, Covert DW & Matthews GP (2007): Evaluation of costs for cystoid macular edema among patients after cataract surgery. Retina 27: 621628.
  • Stanga PE, Boyd SR & Hamilton AM (1999): Ocular manifestations of diabetes mellitus. Curr Opin Ophthalmol 10: 483489.
  • Tranos PG, Wickremasinghe SS, Stangos NT, Topouzis F, Tsinopoulos I & Pavesio CE (2004): Macular edema. Surv Ophthalmol 49: 470490.
  • Varma R & Torres M (2004): Prevalence of lens opacities in Latinos: the Los Angeles Latino Eye Study. Ophthalmology 111: 14491456.
  • WHO Study Group (1985): Diabetes mellitus: technical report series 727. Geneva: World Health Organization.
  • Zaczek A, Olivestedt G & Zetterstrom C (1999): Visual outcome after phacoemulsification and IOL implantation in diabetic patients. Br J Ophthalmol 83: 10361041.