Correlation between Adult Diabetic Cataracts and Red Blood Cell Aldose Reductase Levels

Abstract

purpose. To investigate the correlation between adult diabetic cataracts and levels of aldose reductase (AR) in red blood cells (RBCs).

methods. The study involved 337 eyes of 337 patients with diabetes. The extent and severity of lens opacity was assessed according to the Lens Opacities Classification System III (LOCS III). The AR levels within RBCs were determined with an ELISA. The relationship between the AR level in RBCs and the prevalence of nuclear cataract, cortical cataract, and posterior subcapsular cataract in patients with diabetes was examined.

results. There were no significant alterations in AR level in RBCs in patients with a diabetes duration of ≤10 years and patients <60 years of age. In each subgroup, a higher amount of AR levels in RBCs significantly correlated with the prevalence of posterior subcapsular cataracts. A significant association between cortical cataract and AR level in RBCs was also seen in a subgroup of patients younger than 60 years.

conclusions. AR emerges as an important factor affecting the onset of posterior subcapsular cataracts at the early stages of diabetes mellitus. This raises the possibility that AR inhibitors could play a useful role in treatment of adult diabetic cataract through its inhibition of AR activities.

It is well established that complications associated with diabetes mellitus such as diabetic cataract, diabetic retinopathy, and diabetic keratopathy lead to vision loss in patients with diabetes. It is believed that hyperglycemia leads to the stimulation of other factors that accelerate the progression of the diabetic complications. Kinoshita et al. ^{1 2} have demonstrated in animal studies that diabetic cataracts are initiated through the polyol osmotic theory, in which the intracellular accumulation of polyol via aldose reductase (AR) contributes to lens opacity. Furthermore, they have documented that these experimentally induced cataracts can be completely inhibited by aldose reductase inhibitors (ARIs). ^{3 4 5 6 7} Diabetic cataracts include both idiopathic and adult types. Idiopathic diabetic cataracts observed in juvenile patients, are characterized by a rapid swelling and liquefaction of lens cortical fibers that resemble diabetic cataract formation in animal models. Because of this similarity, it is conceivable that the intracellular accumulation of polyols that is mediated by AR may also play an essential role in the onset of idiopathic diabetic cataract. In contrast, adult diabetic cataracts progress gradually and are difficult to distinguish from senile cataracts. This suggests that the specific mechanism that initiated these diabetic cataracts cannot be clearly distinguished. We report the results of our investigation in which adult diabetic cataracts and AR levels in red blood cells (RBCs) were correlated. 

Methods

The study cohort included 337 consecutive patients with diabetes (182 men and 155 women) who visited the outpatient department of Fukui Medical University from March 1999 through June 2003. Research procedures were in accordance with the Declaration of Helsinki. Informed consent was obtained from all patients. The mean age was 63.3 ± 11.7 years and mean duration of diabetes was 12.9 ± 8.6 years. The mean HbA1c level in these patients was 7.79% ± 1.84%, whereas the mean AR level in the RBCs was 9.6 ± 3.8 ng/mgHb. With regard to the presence or absence of diabetic retinopathy, there were 116 eyes in which retinopathy was absent, 122 eyes with nonproliferative retinopathy, and 99 eyes with proliferative retinopathy. If bilateral cataracts were present, the eye with the more severe lens opacity was used. The extent and severity of lens opacity was assessed according to the Lens Opacities Classification System III (LOCS III). ⁸ Nuclear cataracts were classified into 6 grades based on slit lamp findings, and both cortical cataract and posterior subcapsular cataract were ranked in 5 grades according to slit lamp diaphanoscopic images. 

RBCs were isolated from 1 mL of whole blood collected at the same time cataracts were assessed. The AR levels in RBCs were determined with an ELISA (enzyme-linked immunosorbent assay) kit (MGC EIA; Mitsubishi Gas Chemical Co., Ltd., Osaka, Japan). The RBC fractions were washed with isotonic solution and then suspended in a volume equal to the original whole-blood volume, and this was then diluted 200-fold with 20 mM phosphate-buffered solution (pH 7.4). Details of the assay method have been reported by Tanimoto et al. ⁹ and the correlation between AR activity and AR amounts have been corroborated by Nishimura et al. ¹⁰  

Based on mean AR levels, the subjects were divided into the following three groups: the high-AR group, with AR levels of ≥12.1 ng/mg Hb; the middle-AR group, with AR levels ranging from 9.1 to 12.0 ng/mg Hb; and the low-AR group with AR of ≤9.0 ng/mg Hb. 

Correlations between the AR levels in RBCs and the patient’s age and the duration of diabetes mellitus were determined. Subsequently, the relationship between AR levels in RBCs and the prevalence of nuclear cataract, cortical cataract, and posterior subcapsular cataract in all patients with diabetes, and in patients with a diabetes duration period of ≤10 years and <60 years of age were conducted by examining the ratios of each grade of cataract in each AR group. In the present study, we investigated the morbidity of cataract in a diabetic population <60 years of age based on the Barbados Eye Study, in which the association of lens changes and diabetes was stronger among persons younger than 60 years than in those 60 years of age or older ¹¹ and the report of Jahn et al., ¹² in which the mean ages of patients with posterior subcapsular cataracts were 59 years in men and 65 years in women. We have previously reported a correlation between AR levels in RBCs and diabetic retinopathy. ¹³ In that study, higher AR levels in RBCs correlated with an increased morbidity of retinopathy in patients with diabetes with the duration period of 10 years or shorter. 

An ordinary least squares (OLS) regression test was performed, to investigate the relation between AR levels in RBCs and various variables, including age, duration of diabetes, and severity of nuclear, cortical, or posterior subcapsular cataract. 

Results

A total of 337 patients with diabetes were included in the study. The number of eyes classified by AR levels is presented in Table 1 . The distribution of patients with diabetes showing the relation between AR level in RBCs and age is illustrated in Figure 1 . A significant decrease in AR level with aging was seen in the analyses of all cases (P = 0.033; Fig. 1A ), but not in patients younger than 60 years (P = 0.285; Fig. 1B ). Next, we investigated the relation between AR level and the duration of diabetes; however, the difference was not statistically significant (Fig. 2A ; P = 0.093). Also, in the patients with duration of diabetes of ≤10 years, we found no significant association (Fig. 2B ; P = 0.493). 

Because it is well known that AR is an important causative factor of diabetic cataract, we investigated whether the level of AR in RBCs is related to the development of various types of cataract, including nuclear, cortical, and subcapsular. OLS regression analysis showed that the prevalence of subcapsular cataract correlated significantly with the amount of AR within the entire group of patients with diabetes (Table 2 ; P = 0.043). When analysis was performed in the patients aged <60 years, cortical and subcapsular cataract had significant association with AR level (Table 2 ; P = 0.044 and P = 0.006, respectively). When analyzed in the group of patients with a diabetic history of ≤10 or fewer years, only the prevalence of posterior subcapsular cataract had a strong correlation with AR level (Table 2 ; P = 0.006). 

In addition, we show the prevalence of posterior subcapsular cataract in each low-, middle-, and high- AR group in Tables 3(age, <60 years) and 4(duration of diabetes, ≤10 years). The ratio of P5 in the high-AR group was 18%, whereas that in other groups was 6% (Table 3) . The prevalence of cortical cataract in each level of AR is summarized in Table 5 . 

Discussion

To date, various surveys on select populations and anecdotal reports on patients undergoing cataract surgery indicate that patients with diabetes show a higher morbidity of adult cataracts compared with patients without diabetes. ^{11 12 14 15 16 17 18 19 20 21 22 23} Klein et al. ¹⁴ have documented that persons with diabetes are more predisposed to cortical and posterior subcapsular cataracts than are persons without diabetes. Elevated glycated hemoglobin values that are linked to diabetes also contribute to an increase in severity of nuclear and cortical cataract. Leske et al. ¹¹ have reported a higher presence of posterior subcapsular and cortical cataract in patients with diabetes than in patients without diabetes, whereas Jahn et al. ¹² and Skalka and Prchal ¹⁵ have reported that patients with diabetes demonstrate a higher prevalence of posterior subcapsular cataract. 

Animal studies demonstrate that the AR-mediated intracellular accumulation of polyols induce the collapse and liquefaction of lens fibers, which lead to the formation of lens opacities. ^{1 2} Furthermore, the onset of cataract in these animals is known to be completely inhibited by treatment with ARIs. ^{3 4} The idiopathic diabetic cataract observed in juvenile patients with diabetes is characterized by the rapid progression of cortical lens fiber swelling and liquefaction, which is similar to that observed in diabetic animal models. ²⁴  

In contrast, adult diabetic cataracts progress more gradually and are difficult to distinguish from senile cataract. This difference suggests that the mechanism for cataract formation is more difficult to predict. Immunohistochemical staining of normal human lenses with AR antibodies indicate that AR is present in the lens epithelial cells and lamellar lens fibers, particularly in the bow region. This indicates that AR is abundantly contained in epithelial cells and fibrocytes where active metabolism is being performed. Therefore, AR is likely to be involved in metabolism of the lens. ²⁵ In addition, Akagi et al. ²⁶ have documented in comparative studies of lenses from rats and humans that similar patterns of AR immunohistochemical staining occur. Increased staining was observed in the lenses from both diabetic rats and patients with diabetes, indicating that AR is associated with diabetic cataract formation. Chylack et al. ²⁷ using enucleated lenses have also reported that lenses from patients with diabetes show significantly higher AR levels than do lenses from patients without diabetes. Moreover, ARIs inhibited AR activity in human lenses, irrespective of whether they came from patients with or without diabetes. Inhibition was particularly remarkable in the lenses from patients with diabetes. Based on these findings, it may be postulated that AR is involved in the formation of diabetic cataracts. 

In the present study, possible correlations between the onset of cataract and AR levels in RBCs obtained from patients with diabetes have been investigated to determine whether AR is linked to the formation of adult diabetic cataracts. The ideal method for assessing the relationship between the onset of human cataract and AR would be direct measurement of AR levels within lens opacification. However, it is difficult to obtain essential number of enucleated human lens samples for such a study because the ultrasonic aspiration method, which destroys the lens during cataract surgery, is now the mainstay for cataract treatment. Alternatively, AR levels can be measured in blood samples that are easy to be obtained. We have demonstrated in an earlier study the positive relation between the prevalence of diabetic retinopathy and the AR level in RBCs. ¹³ We considered that AR level in RBCs may be proportional to that in other tissues such as lens, and we explored the relationship between AR level in RBCs and the prevalence of cataract in patients with diabetes. In fact, we demonstrated that the prevalence of posterior subcapsular cataract has the positive association with AR levels in RBCs in the entire group of patients with diabetes. However, we also found that there is a relationship between AR levels and age. There is also a relationship between age and increased prevalence of posterior subcapsular cataract. Thus, it should be noted that the increased prevalence of cataract related to AR level may be the result of increased age. However, in the present study, we demonstrated that younger age (<60 years) and short duration of diabetes (≤10 years) are not related to AR level in RBCs (Figs. 1A 2A ; Table 2 ). Therefore, we focused on these ranges in the early stage of diabetic mellitus and investigated the correlation between AR level and the prevalence of various type of cataract including nuclear, cortical, and posterior subcapsular cataract. The prevalence of posterior subcapsular cataract was highly associated with AR levels in each subgroup. Also, cortical cataract had weak correlation with AR level in the subgroup of younger patients (Table 2) . The morbidity of nuclear cataract was not correlated with AR levels in RBCs. It is reported that AR expresses in the opacity of cortical layer and posterior subcapsular region in human diabetic lens as well as the cataractous lens in diabetic animal models. From this localization pattern of AR in cataractous lenses, it is postulated that AR should not be directly implicated in the region of the lens nucleus. Our present study, indicative of no influence of AR on nuclear cataract, supports this conclusion. 

Based on our observed correlations between increased AR levels in RBCs and the presence of posterior subcapsular cataract in persons with diabetes who are younger than 60 years and with a short duration of diabetes, AR emerges as an important factor affecting the onset of posterior subcapsular cataracts in the early stages of diabetes mellitus. Furthermore, our results (Table 3)showed that 50% (41/82) of diabetic patients with grade P1 were in the low-AR group, whereas 54.5% (6/11) with P5 were in the high-AR group. This finding raises the possibility that the high levels of AR in RBCs may facilitate the development of cataract in patients with diabetes. It could be anticipated that in older patients with longer duration of diabetes the morbidity of posterior subcapsular cataract should become equivocally higher due to aging itself. Therefore, determining a potential correlation between AR levels in these older patients and subcapsular cataracts becomes more difficult. 

It is recognized that an increased glycated hemoglobin level is associated with increased risk of cataract in patients with diabetes. ^{14 20} Also, we showed that higher AR levels in RBCs were significantly associated with greater prevalence of cataract. In our understanding, the level of AR in RBCs is independent of the glycated hemoglobin level, since our previous study revealed that there were no correlation between the level of erythrocyte AR and HbA1c in patients with diabetes. ¹³ However, it should be noted that the AR-catalyzed pathway functions under the hyperglycemia implying the higher levels of HbA1c, though AR plays minor roles in glucose metabolism in normal glycemic state. Thus, it is considered that the high value of HbA1c which activates the polyol pathway is the primary cause of diabetic cataract, and the greater level of AR in RBCs is one of the risk factors of the progression of diabetic cataract. 

Our present study indicates that AR is associated with the onset of cortical and posterior subcapsular cataracts that occur during the early stages of diabetes mellitus. This raises the possibility that ARIs play a useful and important role clinically in preventing the progression of diabetic cataract. 

 Submitted for publication August 8, 2005; revised December 14, 2005, and February 1, 2006; accepted March 22, 2006.

 Disclosure: N. Oishi, None; S. Morikubo, None; Y. Takamura, None; E. Kubo, None; S. Tsuzuki, None; T. Tanimoto, None; Y. Akagi, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Diabetic Patients Classified into Each Group Based on AR Levels

AR (ng/mg Hb)	Total	Age <60 y	Duration ≤10 y
High (>12.0)	80	34	47
Moderate (>9.0, ≤12.0)	101	31	42
Low (≤9.0)	156	51	62
Total	337	116	151

 Data are the number in each group.

Figure 1.

 

Relationship between AR level in red blood cells and age (years). Statistical analysis was performed by the OLS regression method. (A) All patients with diabetes. (B) Patients with diabetes aged <60 years.

Figure 2.

 

Relationship between AR level in red blood cells and the duration of diabetes (years). Statistical analysis was performed by the OLS regression method. (A) All patients with diabetes; (B) patients with the diabetes duration period of 10 years or less.

Table 2.

 

Assessment of Factors Relevant to the AR Levels in Red Blood Cells

	Standardized Regression Coefficient	P
All cases
Age	−0.116	0.033*
Duration of diabetes	−0.094	0.093
Nuclear cataract	0.101	0.657
Cortical cataract	0.094	0.859
Posterior subcapsular cataract	0.11	0.043*
Age <60 years
Age	−0.1	0.285
Duration of diabetes	−0.114	0.232
Nuclear cataract	0.158	0.091
Cortical cataract	0.188	0.044*
Posterior subcapsular cataract	0.255	0.006*
Duration of diabetes ≤10 years
Age	−0.129	0.098
Duration of diabetes	−0.056	0.493
Nuclear cataract	0.105	0.186
Cortical cataract	0.055	0.483
Posterior subcapsular cataract	0.212	0.006*

 Statistical analysis was performed by OLS regression test.

*   P < 0.05.

Table 3.

 

Prevalence of Posterior Subcapsular Cataracts by Age

	P1	P2	P3	P4	P5
AR high (n = 34)	62% (21)	6% (2)	6% (2)	9% (3)	18% (6)
AR middle (n = 31)	65% (20)	13% (4)	6% (2)	10% (3)	6% (2)
AR low (n = 51)	80% (41)	8% (4)	2% (1)	4% (2)	6% (3)

 Data are the percentage (n) affected in each group. P1–P5, posterior subcapsular cataract, grades 1–5. Age <60 years; n = 116.

Table 4.

 

Prevalence of Posterior Subcapsular Cataracts by Duration of Diabetes

	P1	P2	P3	P4	P5
AR high (n = 47)	70% (33)	6% (3)	11% (5)	4% (2)	9% (4)
AR middle (n = 42)	74% (31)	2% (1)	7% (3)	7% (3)	10% (4)
AR low (n = 62)	87% (54)	3% (2)	3% (2)	5% (3)	2% (1)

 Data are as in Table 3 . P1–P5, posterior subcapsular cataract, grades 1–5. Duration ≤10 years; n = 151.

Table 5.

 

Prevalence of Cortical Cataracts

	C1	C2	C3	C4	C5
AR high (n = 34)	32% (11)	24% (8)	24% (8)	15% (5)	6% (2)
AR middle (n = 31)	26% (8)	32% (10)	26% (8)	13% (4)	3% (1)
AR low (n = 51)	39% (20)	25% (13)	29% (15)	6% (3)	0% (0)

 Data are as in Table 3 . C1–C5, cortical cataract, grades 1–5. Age <60 years; n = 116.