Association between folate status, diabetes, antihypertensive medication and age-related cataracts in elderly Taiwanese

Abstract

Objective

To evaluate possible risk factors for cataract in elderly Taiwanese, and to investigate whether the relationship between age and cataract in older persons is modified by other cataract-associated risk factors.

Participants

A cross-sectional study of 661 males and 645 females aged ≥ 65 yrs was conducted as part of the Elderly Nutrition and Health Survey in Taiwan (1999–2000) (Elderly NAHSIT).

Methods

Self-reported cataracts were defined as any incidence of cataract that was diagnosed by a physician and treated by anticataractic drugs based on the medical history section of the Elderly NAHSIT. Potential risk factors for cataract were determined by multiple logistic regression analysis of data obtained from the health examination, blood biochemistry and interviewer-administered questionnaires.

Results

Results showed that the prevalence of self-reported cataract increased with age and was significantly higher in older women than in older men. Cataracts were associated with age, diabetes, antihypertensive medication and folate insufficiency in older men, and with age and antihypertensive medication in older women. Folate insufficiency remained associated with cataract in older men who had adequate vitamin B2, B6 and B12 status. Folate insufficiency was associated with cataract after adjustment for other risk factors in older men aged ≥ 75 yrs, while in older men aged 65–74 yrs, only diabetes and antihypertensive medication remained associated with cataract. In addition, age ≥ 75 yrs remained a risk factor for cataract in those without diabetes, not taking hypertensive medication and with normal folate status. Further analysis showed that the strength of the association between age ≥ 75 yrs in older men and cataracts was increased about 1.5-fold when combined with folate insufficiency (interaction p= 0.0198), and increased about 1.8-fold when combined with use of antihypertensive medication (interaction p = 0.0214).

Conclusion

Our results suggest that the combination of age ≥ 75 yrs in older men with either folate insufficiency or use of antihypertensive medication had an additive effect on the risk of cataract. Maintenance of good folate status should be emphasized to reduce the risk of cataract in the Taiwanese elderly, especially men.

Introduction

As in Western countries, Taiwan's elderly population has been growing rapidly in recent decades. This has led to increased attention to age-related diseases in Taiwan. Cataract is the most common condition affecting visual function in older people and nationwide population-based data has revealed that cataract is the leading cause of visual disability in elderly Taiwanese (1). Age is one of the most important risk factors for cataract (2, 3). Data from the Framingham Eye Study showed that the prevalence of cataract greatly increases with age (4, 5). Native Americans and people of African-American are at higher risk for developing cataracts than Caucasians (6, 7). High blood pressure is a risk factor for cataract in some studies (8, 9), but some results have been contradictory (10, 11). Diabetes has also been shown to be a risk factor for cataract (12, 13), and the Wisconsin study found that cataract was the commonest cause of severe visual loss in older onset diabetic subjects (14).

Oxidative damage is thought to play an important role in cataract development. It has been suggested that antioxidant vitamins might reduce the risk of cataract by protecting lens proteins from oxidative modification (15). Several studies have shown that elevated serum levels and/or intake of antioxidants such as vitamin A, C and E, diminish cataract risk (15, 16, 17, 18). However, other studies have found no correlation between plasma levels and/or intake of vitamin A, C, and E and cataract (19, 20, 21, 22). These conflicting results could be due to other factors or oxidants, besides these antioxidant vitamins, that effect oxidative status.

It is well known that hyperhomocysteinemia produces oxidative stress by generating reactive oxygen species spontaneously (23). Vitamin B2, B6, B12 and folate are involved in homocysteine (Hcy) metabolism and may be important determinants of plasma Hcy (24, 25, 26). In a previous study we found that folate, vitamin B2, B6 or B12 insufficiency was associated with hyperhomocysteinemia (27). The Blue Mountains Eye study found that long term use of vitamin B2, B12 and folate supplements was associated with reduced risk of cataract (28). Several studies have indicated that Hcy concentration is associated with cardiovascular disease and thus may affect the development of cataract or retinal vein occlusion (29). Hyperhomocysteinemia with low vitamin B12 and folate levels in older age groups could be an independent risk factor for senile cataract (30). However, a recent study found that age-related maculopathy was not related to serum Hcy, folate, or vitamin B12 in the NHANES population (31).

Prior studies have shown that age and diabetes are important risk factors for cataract. However, studies of the relationship between high blood pressure, vitamin A, C, E, B2, B6, B12, folate and cataract have been contradictory to date. We have conducted this study to evaluate possible risk factors for cataract in elderly Taiwanese and to investigate whether the relationship between age and cataracts in older persons is modified by other cataract-associated risk factors.

Subjects and Methods

Sampling subjects

The subjects in this study were sampled from the Taiwanese elderly population by a multi-staged, stratified and clustered probability sampling method in the Elderly NAHSIT, a nationwide survey conducted to investigate the nutrition and health status of elderly people in Taiwan. The sample consisted of 1306 subjects aged = 65 y, including 661 males and 645 females, who had complete data from a household interview, health examinations, and blood biochemical parameters.

Survey methodology

The 359 townships or city districts in Taiwan were classified into 13 strata according to the geographical location, ethnic characteristics and population density of townships, which were the primary sampling units. Three townships, and then two villages in each township, were selected from each of the thirteen strata by two consecutive processes of probability proportional to size (PPS) sampling. This resulted in the selection of a total of 78 villages. The survey consisted of two independent compounents: the household interview and the mobile health exam. The household interview was carried out by local part-time interviews who systematically collected cases over an average 12 month period. The local public health department nutritionist was responsible for supervising the household interview. The health exam was performed by a professional team who traveled to each locality to perform the examinations. The blood specimens were fractionated immediately after collection, serum/plasma specimens were aliquoted, frozen in a liquid nitrogen tank, and then delivered to Taipei. The samples were then stored in a freezer at -80°C at the Academia Sinica after which they were sent on to the relevant laboratories for further analysis. The overall response rate for the household interview was 55.2% and that for the health exam was 52.8%. The stratification principle and sampling procedure have been described in Pan et al's report (32).

Data collection

Data collection included household interview, health examinations, and blood biochemical parameters. The data items included sex, age, height, weight, systolic blood pressure (SBP), diastolic blood pressure (DBP), smoking history, alcohol intake, use of nutrient supplements, vitamin C intake, and blood biochemical parameters such as creatinine, fasting glucose, triglycerides, cholesterol, HDL-cholesterol, LDL-cholesterol, erythrocyte glutathione reductase activity coefficient (EGRAC), pyridoxal phosphate (PLP), cobalamin, vitamin A and E. Height and weight were used to calculate body mass index (weight in kg/height² in meters). We also calculated pulse pressure, which is the difference between SBP and DBP.

Biochemical Analyses

The plasma folate and Hcy concentrations were analyzed by our laboratory in the National Taiwan university. Plasma folate and Hcy were measured by an automated analyzer (IMMULITE 2000 analyzer, Diagnostic Products Corporation, LA, USA). A series of quality control tests were performed to evaluate assay precision. These tests confirmed both the between- and within-run consistency of this method. The Coefficient of Variation (CV) for the plasma folate and Hcy were 11% and 7.1%, respectively (27, 33).

Assessment of vitamin status and presence of chronic disease

The cut-off points for judging various B vitamins status was defined as following: insufficient folate status was defined as a plasma folate = 14 nmol/L (6 ng/mL) (34). Insufficient vitamin B2 status was defined as an EGRAC value of = 1.2, a functional test for vitamin B2 (35). Insufficient vitamin B6 status was defined as a PLP < 30 nmol/L (36), and insufficient vitamin B12 status was defined as a serum cobalamin = 258 pmol/L (37).

Hypertension was defined as a SBP = 140 mmHg or a DBP = 90 mmHg. Individuals currently using antihypertensive medication were also considered to have hypertension. Diabetes was defined as fasting glucose = 126 mg/dL or use of antidiabetic medications. Self-reported cataracts were defined as any incidence of cataract that was diagnosed by a physician and treated by anticataractic drugs based on the medical history section of the Elderly NAHSIT.

Statistical analyses

Statistical analyses were carried out using SAS statistical software (SAS/STAT Version 9.1, SAS Institute, Cary, NC). As the representativeness of each sample was not equivalent in the Elderly NAHSIT used a stratified, multistage probability design, sample weighting was required during the statistical analysis in order to obtain a nation-wide representative statistical estimates by SUDAAN, SAS-callable version 9.1. Weighting was carried out by the post-stratification method. Using the 13 geographical strata of the sample population, the sample was further divided into gender and the following age groups: 65-69, 70-74, 75-79 and 80 above. Student's t test was used for analysis of continuous variables, and the chi-squared test was used for analysis of categorical variables. Trends across age groups were evaluated using linear regression. Differences were considered significant if p < 0.05. Multiple logistic regression was used to evaluate possible risk factors for cataract in elderly Taiwanese. Model A included relevant risk factors for cataract (including age, DBP, pulse pressure, antihypertensive medication and diabetes mellitus); model B included relevant risk factors for cataract and Hcy; model C included relevant risk factors for cataract and folate; and finally, model D included relevant risk factors for cataract plus Hcy and folate. The data were stratified to investigate the AOR and 95% CIs of insufficiency in one B vitamins in relation to cataract. Further analyses were performed by stratifying the data to investigate whether the relationship between age and cataract in older persons was modified by other cataract-associated risk factors. The interaction term for age consisted of 2 categories: 65-74 y and = 75 y. A p-value < 0.05 was used to indicate a significant interaction.

Results

The prevalence of cataracts in elderly Taiwanese is shown in Table 1. The prevalence of cataract was 32.5% in males and 48.7% in females. A higher prevalence of cataract in females compared to males was noted in all age groups. In addition, a significant age trend (p for trend < 0.0001 for males, p for trend = 0.0412 for females) in the prevalence of cataract was found in both sexes.

Table 1.

Prevalence of self-reported cataract in the Taiwanese elderly by age and sex^a

Age group (y)	Male			Female			p for differenceb
	n	case	%	n	case	%
65-69	242	61	25.0	278	101	38.3	0.0065
70-74	234	70	33.0	201	92	47.4	0.0014
75-79	125	51	39.6	106	68	65.9	0.0008
>80	60	30	40.6	60	27	54.7	0.0217
Total	661	212	32.5	645	288	48.7	<0.0001
p for trend	<0.0001		0.0412

^a.All values are shown as number and percentage of participants;

^b.Calculated by the chi-squared test.

The characteristics of older participants with and without cataracts by sex are shown in Table 2. The mean ages of both males and females with cataract were significantly older than those without cataract. The pulse pressure, prevalence of diabetes, and antihypertensive medication usage were significantly higher in males with cataract. DBP and plasma folate were significantly lower in both males and females with cataract. This finding suggests that age, pulse pressure, diabetes, antihypertensive medication usage, DBP, and folate status may be important factors associated with cataract.

Table 2.

Characteristics of older persons with and without cataract^a

Characteristic	Male			Female
	Case (n=212)	Normal (n=449)	pb	Case (n=288)	Normal (n=357)	p
Age (y)	73.6±0.4	72.2±0.3	0.0029	73.3±0.4	71.7±0.3	0.0005
SBP (mmHg)	133±1	134±1	0.8188	139±1	140±1	0.4066
DBP (mmHg)	74.8±1.1	76.9±1.0	0.0206	74.5±1.1	76.7±1.0	0.0220
Pulse pressure (mmHg)	59.2±1.1	56.3±1.1	0.0493	64.4±1.1	63.8±1.0	0.6813
BMI (Kg/m2)	23.3±0.3	23.5±0.2	0.3236	24.0±0.3	24.2±0.2	0.4365
Cholesterol (mg/dL)	197±3	194±3	0.2735	213±3	215±3	0.5202
Triglycerides (mg/dL)	108±2	115±1	0.2824	140±2	139±1	0.9193
HDL-C (mg/dL)	52.8±1.1	51.5±1.3	0.3343	57.1±1.1	56.8±1.3	0.8295
LDL- C (mg/dL)	123±4	120±3	0.2106	128±4	130±3	0.3195
Blood glucose (mg/dL)	115±5	110±2	0.2334	114±5	119±2	0.1448
Creatinine (mg/dL)	1.13±0.1	1.11±0.2	0.6959	0.90±0.1	0.88±0.2	0.1310
Folate (nmol/L)	20.4±1.8	23.6±1.4	0.0459	27.2±1.4	30.0±1.8	0.0484
EGRAC	1.16±0.1	1.16±0.1	0.7715	1.14±0.1	1.15±0.1	0.4902
PLP (nmole/L)	50.1±5.3	52.0±3.6	0.6823	60.8±5.3	60.9±3.6	0.9782
VitB12(pmol/L)	601±138	585±173	0.2799	753±138	704±173	0.6427
Hcy (mol/L)	14.2±0.8	14.0±0.5	0.6503	10.8±0.8	11.2±0.5	0.3476
VitA (μg/mL)	0.78±0.1	0.82±0.1	0.2833	0.78±0.1	0.74±0.1	0.1013
VitE/triglycerides (mg/g)	12.1±0.9	11.6±0.5	0.4959	11.4±0.9	11.3±0.5	0.8404
VitC intake (mg/d)	149±8	148±18	0.1400	144±8	151±18	0.8603
Smoking every day (%)	36.6	32.1	0.5775	4.0	1.8	0.8423
Drinking > 1 time/week (%)	12.4	14.9	0.1540	0.4	1.5	0.3175
Nutrient supplements (%)	27.1	24.7	0.6452	32.8	30.2	0.1190
Hypertension (%)	54.1	51.2	0.1866	54.9	58.1	0.6695
Diabetic Mellitus (%)	21.7	16.1	0.0488	19.3	21.8	0.7192
Antihypertensive medication (%)	33.3	25.5	0.0038	30.5	26.0	0.2618

SBP = systolic blood pressure; DBP = diastolic blood pressure; EGRAC = Erythrocyte glutathione reductase activity; PLP = plasma pyridoxal phosphate; Hcy = Homocysteine;

^a.Mean±SE (all such values);

^b.Calculated by a t-test (continuous variables) or a chi-squared test (categorical variables).

Only variables with significant differences between subjects with and without cataract were included in multiple logistic regression models as possible risk factors for cataract (Table 3). Model D (including the variables age, DBP, pulse pressure, folate, Hcy, diabetic mellitus and use of antihypertensive medication) showed that age = 75 y (AOR = 2.6 95% CI [1.68, 3.96] for males, AOR = 2.5 95% CI [1.63, 3.75] for females) was significantly associated with cataract in both males and females, but age 70-74 y was only significantly associated with cataract in females (AOR = 1.5 95% CI [1.06, 2.25]. In addition, use of antihypertensive medication (AOR = 1.9 95% CI [1.15-2.99] for males, AOR = 1.5 95% CI [0.99-2.29] for females) was significantly associated with cataract in both sexes. Folate insufficiency (AOR = 1.5 95% CI [1.02-2.29]), and diabetes (AOR = 1.6 95% CI [1.08-2.50]) were only significantly associated with cataract in males.

Table 3.

Adjusted Odds Ratios (AOR) for cataract in older persons^a

Variables	Male
	Adjusted Odds ratio [95%CI]b
	Model A	Model B	Model C	Model D
Age (y)
65-69	1.0	1.0	1.0	1.0
70-74	1.2 [0.84, 1.89]	1.2 [0.84, 1.90]	1.2 [0.82, 1.87]	1.2 [0.83, 1.88]
=75	2.5 [1.64, 3.85]	2.5 [1.65, 3.85]	2.5 [1.66, 3.90]	2.6 [1.68, 3.96]
Diabetic mellitus
no	1.0	1.0	1.0	1.0
yes	1.6 [1.08, 2.50]	1.6 [1.08,2.50]	1.6 [1.08, 2.50]	1.6 [1.08,2.50]
Antihypertensive medication
no	1.0	1.0	1.0	1.0
yes	1.9 [1.18, 3.06]	1.9 [1.18, 3.07]	1.9 [1.15, 2.99]	1.9 [1.15,2.99]
Folate (nmol/L)
>14			1.0	1.0
=14			1.5 [1.01, 2.18]	1.5 [1.02, 2.29]

Variables	Female
	Adjusted Odds ratio [95%CI]
	Model A	Model B	Model C	Model D
Age (y)
65-69	1.0	1.0	1.0	1.0
70-74	1.5 [1.01, 2.24]	1.5 [1.06, 2.25]	1.5 [1.06, 2.24]	1.5 [1.06, 2.25]
=75	2.5 [1.63, 3.68]	2.5 [1.63, 3.16]	2.5 [1.63, 3.68]	2.5 [1.63, 3.75]
Antihypertensive medication
no	1.0	1.0	1.0	1.0
yes	1.5 [1.01, 2.28]	1.5 [0.99, 2.29]	1.5 [0.99, 2.29]	1.5 [0.99, 2.29]

^a.Only variables with significant effects are shown;

^b.Adjusted Odds ratios [95%CI] were determined by multiple logistic regression. The multiple logistic regression models were as follows: model A= relevant risk factors for cataract (age, diastolic blood pressure, pulse pressure, antihypertensive medication and diabetic mellitus) alone; model B= model A + homocysteine; model C= model A + folate; model D= model A + homocysteine +folate.

To investigate whether the relationship between folate and cataract in elderly Taiwanese was influenced by the status of other B vitamins, the associations between insufficiency in any of the B-vitamins and cataract are illustrated in Table 4, with subjects stratified on the basis of their B-vitamin status. After adjustment for age, DBP, pulse pressure, diabetes, and use of antihypertensive medication, we found that insufficiency in either vitamin B2, B6 or B12 was not significantly associated with cataract in older persons who do not have insufficiency in the other three B vitamins. Only folate insufficiency was significantly associated with cataract in older men with adequate vitamin B2, B6 and B12 status (AOR = 2.6 95% CI [1.30, 5.37]). These results confirm the protective effect of adequate folate status for cataract in older men.

Table 4.

Adjusted Odds Ratios (AOR) for insufficiency in B vitamins and cataract in older persons

Folate	B vitamin statusa., b.			Male
	Vitamin B2	Vitamin B6	Vitamin B12	Case/normal (n)	Adjusted Odds ratio [95%CI]c
				69/152	1.0
				22/19	2.6 [1.30, 5.37]
				15/46	0.6 [0.34, 1.33]
				38/64	1.3 [0.79, 2.23]
				4/30	0.8 [0.31, 1.44)]

Folate	B vitamin statusa., b.			Female
	Vitamin B2	Vitamin B6	Vitamin B12	Case/normal (n)	Adjusted Odds ratio [95%CI]c
				136/174	1.0
				20/15	1.7 [0.82, 3.46]
				42/48	1.1 [0.69, 1.80]
				41/41	1.2 [0.71, 1.92]
				8/9	1.0 [0.36, 2.66]

^a.The criteria for B-vitamin status were defined as follows: insufficient folate was defined as plasma folate = 14 nmol/L (6 ng/mL), insufficient vitamin B2 was defined as EGRAC value = 1.2, insufficient vitamin B6 was defined as plasma pyridoxal phosphate < 30 nmol/L, and insufficient vitamin B12 was defined as serum cobalamin = 258 pmol/L;

^b.“indicates B-vitamin adequacy;” indicates insufficiency;

^c.Adjusted Odds ratio [95%CI] were determined by multiple logistic regression, adjusted for age, diastolic blood pressure, pulse pressure, antihypertensive medication use and diabetes.

It is of note that the cataracts might occur at younger ages in women compared to men (Table 3). Therefore, we analyzed our data by stratification to investigate whether the relationship between age and cataract in elderly Taiwanese was influenced by sex, and other cataract-associated risk factors. As shown in Table 5, after adjustment for significant risk factors for cataract, folate insufficiency was not associated with cataract in either sex aged 65-74 yrs. However, diabetes (AOR = 1.8 95% CI [1.13, 2.87]) and use of antihypertensive medication (AOR = 1.8 95% CI [1.07, 3.14]) remained associated with cataract among older men aged 65-74 y. In older women aged 65-74 yrs, diabetes and folate insufficiency were not associated with cataract, but antihypertensive medication was associated (AOR = 1.6 95% CI [1.07, 2.57]). In addition, age = 75 yrs remained a risk factor for cataract in those without diabetes, not taking hypertensive medication and with adequate folate status in both older men and women. Further analysis showed that the odds ratio for cataract in older men with both age = 75 y and folate insufficiency (AOR = 3.4 95% CI [1.52, 7.53]) increased 1.5-fold compared to those aged = 75 y alone (AOR = 2.3 95% CI [1.54, 3.43]) (interaction p = 0.0198). The odds ratio for cataract in older men aged = 75 y and using antihypertensive medication (AOR = 4.2 95% CI [2.04, 8.52]) increased 1.8-fold compared to those aged = 75 y alone (AOR = 2.3 95% CI [1.49, 3.02]) (interaction p = 0.0214).

Table 5.

Adjusted Odds Ratio (AOR) for diabetes, folate status and antihypertensive medication and cataract in older persons by age and sex

Variables	Adjusted Odds ratio [95%CI]a, Male
	65~74 (y)	=75 (y)	p for interactionb
Diabetic mellitus
no	1.0	2.5 [1.66, 3.68]	0.1144
yes	1.8 [1.13, 2.87]	2.8 [1.19, 6.71]
Folate statusc
adequacy	1.0	2.3 [1.54, 3.43]	0.0198
insufficiency	1.5 [0.88, 2.38]	3.4 [1.52, 7.53]
Antihypertensive medication
no	1.0	2.3 [1.49, 3.02]	0.0214
yes	1.8 [1.07, 3.14]	4.2 [2.04, 8.52]

Variables	Adjusted Odds ratio [95%CI], Female
	65~74 (y)	=75 (y)	p for interaction
Diabetic mellitus
no	1.0	2.1 [1.37, 3.09]	0.2409
yes	1.1[0.70, 1.75]	2.1 [1.23, 3.26]
Folate status
adequacy	1.0	2.0 [1.38, 2.99]	0.3341
insufficiency	1.1 [0.59, 2.13]	2.0 [1.47, 2.96]
Antihypertensive medication
no	1.0	2.2 [1.38, 3.40]	0.0918
yes	1.6 [1.07, 2.57]	2.7 [1.76, 4.96]

^a.Adjusted Odds ratio [95%CI] were determined by multiple logistic regression, and adjusted for plasma folate, diastolic pressure, pulse pressure, antihypertensive medication use and diabetes;

^b.Significant interaction between folate status and age, antihypertensive medication and age, diabetes and age if p <0.05;

^c.The criterion for insufficient folate was defined as plasma folate = 14 nmol/L (6 ng/mL).

Discussion

Cataract is the leading cause of visual impairment in older adults in the world. Our results showed that the prevalence of cataract was higher in women than in men, which is consistent with previous studies (38, 39, 40, 41). This finding suggests that female sex could be a risk factor for cataract in the Taiwanese elderly. Recent studies have found that estrogens may protect against age-related cataracts by their antioxidant properties. A protective effect for nuclear cataract with current use of postmenopausal estrogens was shown in the Beaver Dam Eye Study (38). The cause of sex differences in cataract occurrence is not clear but could be related to the estrogen deficiency that occurs after menopause in women (38, 39). Another possible explanation for our finding is that reproductive exposures in women increase their risk of cataract (40).

In our study, the prevalence of cataract increased with age, and was particularly associated with age = 75 yrs in men and age = 70 yrs in women. This indicates that cataracts might occur at younger ages in women compared to men. In older persons without diabetes, not taking antihypertensive medication and with adequate folate status, age = 75 yrs still remained a risk factor for cataract. This finding supports previous studies demonstrating that age is the most important risk factor for cataracts (2, 3, 4, 5).

Further analysis of our data showed that DBP levels tended to decrease with age among men and women (p trend <0.05), while pulse pressure levels increased with age only in older men (p trend <0.0001). These results agree with reports that blood pressure is strongly related to aging (42). It is of interest to note that older persons with cataract had lower DBP and higher pulse pressure than those without cataract. However, in multiple logistic regression models, DBP and pulse pressure were not associated with cataract. The reason why this association was not found to be significant in our study could be because DBP and pulse pressure were strongly associated with age. Therefore, we hypothesize that the significant difference in DBP and pulse pressure between those with and without cataract is due to confounding between age-related changes in these factors and the increased risk of cataract with age. Therefore, DBP and pulse pressure do not have a direct relationship with risk of cataract.

Prior investigations have found that prolonged hyperglycemia in type 2 diabetes mellitus enhances the glycation of proteins and may be the cause of cataract formation (43, 44). Our data showed that cataract was associated with diabetes, as previously reported (12, 13, 14). It is notable that antihypertensive medication was associated with cataract (45, 46), although the role of antihypertensive medication on cataract has not yet been established. This suggests that further epidemiological and etiologic studies are needed to clarify the relationship between antihypertensive medication and cataract.

Animal models have shown that antioxidants can protect against cataract (47, 48), and some epidemiological studies have indicated a possible role for antioxidants in the prevention of cataract. However, prior epidemiological studies of the relationship between vitamin A, C, and E and cataract have been contradictory (15, 16, 17, 18, 19, 20, 21, 22). Our study did not find an association between serum vitamin A and E, vitamin C intake and cataract, suggesting that the role of vitamin A, C and E on cataract is yet to be confirmed.

In our study, Hcy, vitamin B2, B6 and B12 were not associated with cataract. This result was similar to that in the NHANES population (49), but differs from other studies (31, 50). Our study showed that elderly Taiwanese with cataract had a lower plasma folate level than those without cataract, and folate insufficiency was significantly associated with cataract. We found that folate insufficiency remained associated with cataract in older men who had adequate vitamin B2, B6 and B12 status. Previous studies have found that folate has a strong protective effect against cataractogenesis (28, 30, 51), though the biological mechanism for folate in cataractogenesis is yet to be established. Nevertheless, more attention should be paid to maintaining adequate folate status in the elderly.

In older men aged 65-74 yrs, folate insufficiency was not associated with cataract after adjustment for other risk factors. However, diabetes and/or antihypertensive medication were associated with cataract. Folate status had a more significant impact on older persons aged = 75 yrs. Moreover, the combination of age = 75 yrs with either folate insufficiency or use of antihypertensive medication increased dramatically the adjusted odds ratio for cataract in older men, but not in women, as compared to those aged = 75 yrs alone. These findings suggest that the association between age and cataract is influenced by folate status and use of antihypertensive medication. For older men aged = 75 yrs, either folate insufficiency or use of antihypertensive medication has an additive effect on the risk of cataract.

In summary, cataract is a multifactorial disease. Our results showed that cataract was not related to Hcy, vitamin A, C, E, B2, B6 or B12, but was associated with sex, age, diabetes, folate insufficiency and antihypertensive medication in the elderly Taiwanese population. Age is the most important risk factor for cataract. The combination of age = 75 yrs with either folate insufficiency, or antihypertensive medication had an additive effect on the risk of cataract. Although a causal relationship is unable to be established due to the cross-sectional nature of this study, maintenance of good folate status should be emphasized to reduce the risk of cataract in the Taiwanese elderly, especially in men.

Acknowledgments:

This study was supported by the Department of Health, in Taiwan (project no, DOH 94-TD-F-113-002). Data analyzed in this paper were collected by the research project “Elderly Nutrition and Health Survey in Taiwan (1999-2000) (Elderly NAHSIT)” carried out by the Institute of Biomedical Sciences of Academia Sinica. The Office of Nutrition Surveys, The Center for Survey Research of Academia Sinica was responsible for data distribution. The assistance provided by the institutes and aforementioned individuals is greatly appreciated.

Funding sources: Supported by the Department of Health, Taiwan, Republic of China (DOH 94-TD-F-113-002).