点击显示 收起
1 From the Jean Mayer USDA Human Nutrition Research Center on Aging (AT, PFJ, and GR) and the Gerald J and Dorothy R Friedman School of Nutrition Science and Policy (AT and PFJ), Tufts University, Boston; The Channing Laboratory, the Department of Medicine (SEH and WCW), Harvard Medical School (LTC), Boston; the Departments of Nutrition (WCW) and Epidemiology (SEH and WCW), Harvard School of Public Health, Boston; and the Center for Ophthalmic Research (PMK, JF, WT, JKW, NP, and LTC), Brigham and Women's Hospital (SEH and WCW), Boston.
2 Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the US Department of Agriculture. 3 Supported by the US Department of Agriculture (58-1950-9-001), the National Research Initiative Competitive Grant Program (98-01023 and 92-37200-7704), the Brigham Surgical Group, NIH grants CA40356 and EY09611, and grants from the Florida Department of Citrus and Roche Vitamins. 4 Address reprint requests to A Taylor, USDA HNRC on Aging, Tufts University, Laboratory for Nutrition and Vision Research, 711 Washington Street, Boston, MA 02111. E-mail: ataylor{at}hnrc.tufs.edu.
ABSTRACT
Background: Proper nutrition appears to protect against cataracts. Few studies have related nutrition to the odds of developing cortical or posterior subcapsular (PSC) cataracts.
Objective: We assessed the relation between usual nutrient intakes and age-related cortical and PSC lens opacities.
Design: We studied 492 nondiabetic women aged 53–73 y from the Nurses' Health Study cohort who were without previously diagnosed cataracts. Usual nutrient intake was calculated as the average intake from 5 food-frequency questionnaires collected over a 13–15-y period before the eye examination. Duration of vitamin supplement use was determined from 7 questionnaires collected during this same period. We defined cortical opacities as grade 0.5 and subcapsular opacities as grade 0.3 of the Lens Opacities Classification System III.
Results: Some lenses had more than one opacity. No nutrient measure was related to prevalence of opacities in the full sample, but significant interactions were seen between age and vitamin C intake (P = 0.02) for odds of cortical opacities and between smoking status and folate (P = 0.02), -carotene (P = 0.02), ß-carotene (P = 0.005), and total carotenoids (P = 0.02) for odds of PSC opacities. For women aged <60 y, a vitamin C intake 362 mg/d was associated with a 57% lower odds ratio (0.43; 95% CI: 0.2, 0.93) of developing a cortical cataract than was an intake <140 mg/d, and use of vitamin C supplements for 10 y was associated with a 60% lower odds ratio (0.40; 0.18, 0.87) than was no vitamin C supplement use. Prevalence of PSC opacities was related to total carotenoid intake in women who never smoked (P = 0.02).
Conclusions: Our results support a role for vitamin C in diminishing the risk of cortical cataracts in women aged <60 y and for carotenoids in diminishing the risk of PSC cataracts in women who have never smoked.
Key Words: Cortical lens opacity • posterior subcapsular lens opacity • Nurses' Health Study • Nutrition and Vision Project • vitamins • carotenoids • women
INTRODUCTION
Clouding of the eye lens is called cataract. Approximately 45% of the elderly >75 y of age have vision-impairing cataracts, and the proportion of those affected by some form of cataract is much higher. Removal of cataracts is the most frequently performed surgical procedure among the elderly, and costs associated with this dysfunction account for the largest line item in the Medicare budget (1). There are 3 metabolically distinguishable zones of the lens: the epithelium, the cortex, and the lens nucleus or core (2). Epithelial cells are found just under the collagenous capsule that surrounds the lens. These are the most recently formed cells and they are the most metabolically active. Some of these cells divide to form lens fiber cells. It is in these cells that the major gene products of the lens, the crystallins, are elaborated. The outer layers of such fiber cells comprise the cortex. These cells also undergo extensive changes as they denucleate, age, and get compressed as new cells form exteriorly. Buried under the cortical cells are the oldest lens cells, called nuclear or core cells. Thus, there is a gradient, with the most recently elaborated proteins in the epithelium and the oldest proteins, which were elaborated during embryonic stages, in the nuclear cells. Posterior subcapsular (PSC) opacities are primarily due to aberrations in the outermost layers of the lens (2). Cortical opacities involve inner and outer cortical tissue. Many cataracts involve the cortex. Nuclear opacities are found in the central and oldest zone, which is metabolically quiescent. Because of these metabolic distinctions, some investigators think that opacification in these 3 zones has different etiologies, and most epidemiologic studies treat the 3 zones separately.
With aging, lens constituents sustain extensive photooxidative damage. The long-lived lens proteins and the enzymes that might otherwise recognize and remove these proteins are themselves damaged with aging. It has been hypothesized that this results in the accumulation and precipitation of proteins in cataracts (3–7). This hypothesis is supported by data from our vitamin C and cataract study (8) and in results regarding the risk of nuclear cataracts from the baseline phase of our Nutrition and Vision Project (NVP) (9). These studies show that persons with higher intakes of antioxidants such as vitamin C or lutein/zeaxanthin have a markedly diminished prevalence of early nuclear opacities. Other epidemiologic studies also indicate possible protective roles for nutritional antioxidants in the risk of nuclear cataracts or cataract extraction (3, 8, 10–32).
The literature that relates nutrient intake or plasma antioxidant concentrations with cortical (12, 13, 17, 18, 20, 33–35) or PSC opacities (12, 13, 18, 21, 30, 32) is much more limited. In the present study, we examined the lenses of women who had been recruited for the Nurses' Health Study (NHS) (36). Our objective was to assess the relation between the womens' usual nutrient intakes and subsequently identified, age-related cortical and PSC lens opacities. All the nutrient intake data were gathered before the eye examinations were obtained as part of the NVP. Unlike most earlier studies, most of the opacities in the NVP were of very early grades and could be considered predictive of changes later in life (35).
SUBJECTS AND METHODS
Subjects
In 1976, 121700 female nurses aged 30–55 y who resided in 11 US states completed a mailed questionnaire on known and suspected risk factors for cancer and heart disease. These women formed the NHS cohort (36). Every 2 y since 1976, these women have been contacted by mail to update information on risk factors and disease status.
Details regarding recruitment and participation were described previously (9). With a goal of enrolling 600 women in the NVP, we identified 1442 NHS cohort members aged 54–73 y in 1993. All of the women resided in the Boston area, were free of diagnosed cancer other than nonmelanoma skin cancer, had complete dietary data, and had both lenses intact. We received positive responses from 730 women with this one mailing, and 603 (42% of the original number identified) of these volunteers were ultimately scheduled and examined as part of the NVP between April 1993 and August 1995. Informed consent was obtained from all study participants, and all procedures were approved by the Human Investigations Review Committee at the New England Medical Center (Boston) and the Human Research Committee at the Brigham and Women's Hospital.
Assessment of nutrient intake
A 61-item, semiquantitative food-frequency questionnaire was initially incorporated into the NHS biennial questionnaire in 1980 (36). The food-frequency questionnaire covered usual intakes over the previous year and included 9 possible response categories, which ranged from "never or less than once per month" to "6 or more times per day." In addition, the 1980 biennial questionnaire also collected information on vitamin supplement use in 1980 and the duration of vitamin supplement use before 1980. In 1984, 1986, and 1990, revised and expanded versions of the food-frequency questionnaire were included in the biennial questionnaire. Every biennial questionnaire since 1980 has included questions on vitamin supplement use. The present version of the food-frequency questionnaire includes 130 food items and details of vitamin and mineral supplement use that collectively account for >90% of the total absolute intake of the 70 nutrients measured by the questionnaire (37). The food-frequency questionnaire has been extensively validated in relation to both long-term diet records (36, 37) and biochemical markers of nutrient status (38–40). In addition to the food-frequency questionnaires collected routinely as part of the NHS, we administered an additional food-frequency questionnaire that included questions on vitamin supplement use as part of the NVP.
We estimated both dietary nutrient intake (from foods) and total nutrient intake (from foods and supplements). We used the data from women who completed 5 food-frequency questionnaires collected between 1980 and 1993–1995 to calculate the average nutrient intake for each participant.
We used 7 reports of vitamin supplement use from 1980 through 1993–1995 to categorize women by the duration of their use of supplements containing vitamins C and E and of multivitamin supplements. We assigned 2 y of supplement use to the duration-of-use variable for each report of supplement use between 1980 and 1990. For women reporting supplement use on the food-frequency questionnaire collected as part of the NVP, we added the interval between 1990 and the date of the eye examination to the duration-of-use variable. Finally, we added the reported duration of use before 1980 to the duration-of-use variable. We assumed that a woman who started or stopped using supplements during the interval between questionnaires did so halfway through the period. Women with incomplete supplement data were included in supplement analyses if they did not have missing data from consecutive questionnaires and if their reported use was the same before and after the period for which the data were missing. In this case, we assumed that supplement use was similar at the intermediate (missing) time period.
Assessment of lens status
All NVP participants received a detailed eye examination with the use of standardized techniques. The examination included an ocular history, a medical history, the Bailey Lovie test of visual acuity and manifest refraction, an external ocular examination, applanation tonometry, contrast sensitivity function and glare testing, and a slitlamp examination of the anterior segment. The latter included an assessment of the anterior chamber to determine the risk of angle-closure glaucoma. Measurement of intraocular pressure was required to determine whether it was safe to complete the eye examination, including dilation. Before a slitlamp examination of the lens was performed, the pupils were dilated to a minimum of 6 mm with phenylephrine and tropicamide. The posterior segment was examined with the use of direct and indirect ophthalmoscopy. The examiner had no knowledge of the nutrient status of any of the participants.
Each participant had 2 types of photographs taken for assessing the severity of lens opacities. Two digital retroillumination black-and-white images were taken with a Nidek EAS 1000 camera (Hiroishi, Japan), one with an anterior image focus (focused on the pupillary plane) to assess the degree of cortical opacification and one with a posterior focus (focused on the posterior lens capsule) to assess the degree of cortical and PSC opacification. The Lens Opacities Classification System III (LOCS III) was used for the present analysis to measure the degree of lenticular opacification (41, 42). This system assesses cortical and PSC opacity, with possible grades ranging from 0.1 to 5.9. To do the grading, each of 2 individual graders wrote down a score for the photograph or image on a score sheet. The graders then compared their scores and arrived at a consensus score, which was recorded on the third form. Consensus scores typically fell between the 2 individual scores, but were not true averages. All images were graded in several sessions within a 2-mo period after all photographs and images were obtained. Because of the difficulty in assessing certain features of the PSC region of the lens with the use of images, grading of the extent of PSC opacification was also done in vivo at the slitlamp by using LOCS III. We used this in vivo measurement to grade PSC opacification in the present analyses. We considered eyes to have opacities if the LOCS III cortical opalescence grade was 0.5 or the PSC grade was 0.3. These thresholds represented early stages of opacification and were not associated with symptoms such as reduced vision.
Defining nonnutritional variables
Data on known or suspected nonnutritional determinants of cataract risk were obtained from the biennial Nurses' Health Study questionnaires obtained from 1980 through 1990. For the present analyses, we considered the history of diabetes and hypertension (yes or no) as reported on the 1990 or previous questionnaires, cigarette pack-years (ie, number of packs of cigarettes smoked per day x number of years of smoking) smoked through 1990, reported summertime sunlight exposure (8 h/wk) as reported on the 1980 questionnaire, alcohol use calculated on the basis of the average from 5 food-frequency questionnaires collected between 1980 and 1993–1995, and height and weight as reported on the 1980 questionnaire. The latter 2 measures were used to calculate body mass index (weight in kg/height2 in meters).
Statistical methods
We estimated the odds ratios relating the prevalence of lens opacities to the average nutrient intake and the duration of vitamin supplement use from logistic regression with the SAS GENMOD procedure (43). This procedure allowed the individual eyes to be the unit of observation and required that information for both eyes be available. This generalized, estimating-equation approach to estimate logistic regression models adjusts the SEs of the model variables for the correlated data resulting from repeated measurements on the same individual.
To examine the relation between usual nutrient intakes and the prevalence of lens opacities, the primary independent variable in the logistic regression models was the average nutrient intake for each individual calculated on the basis of data from the 5 food-frequency questionnaires. These intake variables were classified into quintile categories (Table 1), which were modeled with indicator variables; the women in the lowest quintile category served as the reference category. To assess trends across quintile categories, we assigned the median intake of each quintile category to everyone with intakes in the category and then included this quintile median variable as a continuous factor in the logistic regression models. The P for trend was the resulting P value for the associated logistic regression coefficient. To assess the independence of the associations between the measures of intake and the prevalence of opacities, we performed a backward-selection procedure that included the nutrients that were individually associated with the prevalence of opacities. Those nutrients that remained significantly (P < 0.05) associated with opacities after the removal of nutrient variables starting with the least significant were considered to be independently associated with the prevalence of opacities.
View this table:
TABLE 1 . Nutrient quintile values used to define intake and plasma nutrient categories
We assessed the possibility that relations between cortical or PSC opacities and nutrient intake were modified by age and smoking. The interaction term for age consisted of 2 categories: <60 and 60 y of age. The interaction term for smokers was constructed by grouping current smokers with former smokers and comparing the combined group with never smokers. We used a P value <0.05 to signify a significant interaction.
To examine the relation between the duration of vitamin supplement use and the prevalence of lens opacities, categories of duration of vitamin supplement use were modeled with indicator variables; those women who reported no vitamin supplement use served as the reference category. A test for a trend of prevalence across supplement duration categories was performed by assigning each participant the median value of their duration category and entering this continuous variable into the logistic regression model. The P trend was the resulting P value for the associated logistic regression coefficient.
Odds ratios for the prevalence of lens opacities for both the intake quintile categories and the supplement duration categories were calculated as the antilogarithm of the logistic regression coefficient for each of the categories. All odds ratios were adjusted for age and other potential confounders, including history of hypertension through 1990, cigarette pack-years smoked through 1990, reported summertime sunlight exposure in 1980, usual alcohol use calculated on the basis of data from the 5 questionnaires, and body mass index in 1980.
RESULTS
Participant characteristics
To avoid the possibility that prior knowledge of lens opacification might influence nutrient intake, we excluded 76 of the 603 women examined as part of the NVP because they reported a history of cataracts at their study eye examination. In addition, we excluded 9 women who had a confirmed diagnosis of diabetes by 1990; 19 women who had incomplete, questionable, or missing lens data; and 7 women for whom information was missing regarding the covariates that were used in these analyses. The 492 remaining women were included in these analyses. Of the 984 eyes available for analyses, 440 (44.7%) had clear lenses, 253 (25.7%) had nuclear opacities (ie, LOCS III grade 2.5), 336 (34.1%) had cortical opacities, and 127 (12.9%) had PSC opacities. According to our definitions, pure cortical opacities, pure nuclear opacities, and pure PSC opacities were present in 37.0%, 23.1%, and 6.3% of eyes with opacities, respectively. One-third of the eyes had opacities in more than one location.
As shown in Table 2, the women with either cortical or PSC opacities were older and had a significantly lower alcohol intake than did the women without these opacities. The women with cortical opacities had slightly higher concentrations of plasma reduced vitamin C than did the women without cortical or PSC opacities. Smoking history, sunlight exposure, blood pressure, intake of vitamins C and E and multivitamins, and concentrations of plasma -tocopherol and total carotenoids were indistinguishable between the 3 groups. The women with PSC opacities had higher body mass indexes than did the women without cortical or PSC opacities.
View this table:
TABLE 2 . Characteristics of participants categorized by the presence or absence of cortical or posterior subcapsular (PSC) opacities1
Nutrient intake and the prevalence of opacities
The relations between the prevalence of cortical or PSC lens opacities and usual intake of nutrients (from both food and supplements), including vitamins C and E, riboflavin, folate, and total and individual carotenoids, measured over a 13–15-y period are shown in Tables 3 and 4, respectively. For the entire cohort, there were no significant trends between any of the nutrients and either cortical (Table 3) or PSC (Table 4) opacities, and none of the odds ratios relating the highest quintile category to the lowest category were significantly different from 1. These results differ from those for nuclear opacities, which showed that, before mutual adjustment for different intakes, elevated intakes of vitamins C and E, lutein, zeaxanthin, riboflavin, folate, and ß-carotene were related to decreased odds of nuclear opacities (9).
View this table:
TABLE 3 . Relation between the prevalence of cortical lens opacities and total nutrient intake from foods and supplements1
View this table:
TABLE 4 . Relation between the prevalence of posterior subcapsular lens opacities and total nutrient intake from foods and supplements1
We observed a significant interaction between age and vitamin C intake for cortical opacities (P = 0.02). Among those aged <60 y, vitamin C intake was inversely associated with the prevalence of cortical opacities (Table 5). The odds ratio in the highest quintile category compared with the lowest was 0.43 (95% CI: 0.20, 0.93), but the odds ratio in the fourth quintile category was 2.19 (95% CI: 1.10, 4.37). Intake of vitamin C was unrelated to the prevalence of opacities among those aged 60 y. These associations were unaffected by adjustment for the presence of nuclear opacities.
View this table:
TABLE 5 . Relation between the prevalence of cortical lens opacities and quintile category of total vitamin C intake from foods and supplements by age1
For PSC opacities, significant interactions were also observed between smoking (coded as never or ever) and intakes of folate (P = 0.02), -carotene (P = 0.02), ß-carotene (P = 0.005), and total carotenoids (P = 0.03). Thus, among women who never smoked, the prevalence of PSC opacities was inversely related to folate, total carotenoid, -carotene, and ß-carotene intakes (Table 6). The odds ratio in the highest folate intake quintile category compared with the lowest was 0.26 (95% CI: 0.09, 0.77). Never smokers with intakes in the fourth quintile category also had reduced odds, but there was no association between the prevalence of cataracts and folate intake among present and former smokers. For never smokers, the odds ratios in the highest intake quintile categories compared with the lowest categories were 0.19 (95% CI: 0.05, 0.68) for total carotenoids, 0.29 (95% CI: 0.08, 1.05) for -carotene, and 0.28 (95% CI: 0.08, 0.96) for ß-carotene. However, after mutual adjustment for folate, -carotene, ß-carotene, and total carotenoids, only total carotenoids remained associated with the odds of PSC opacities. As we adjusted for other nutrients, folate and total carotenoids were the last 2 nutrients remaining in the model. Both nutrients were similarly associated with the odds of PSC opacities, but the P value for total carotenoids was lower. Therefore, we removed folate from the model. In addition, these associations were not present among current or former smokers. There was no significant interaction between smoking and the other carotenoids.
View this table:
TABLE 6 . Relation between the prevalence of posterior subcapsular lens opacities and quintile category of total nutrient intake from foods and supplements by smoking status1
Vitamin supplement use and the prevalence of opacities
The duration of use of vitamin supplements was unrelated to the prevalence of either cortical or PSC opacities in the entire cohort (Table 7). However, as with total vitamin C intake, we observed an age interaction for duration of vitamin C supplement use (P = 0.003) (Table 8). There was a significant inverse trend between the duration of vitamin C supplement use and the prevalence of cortical opacification in women aged <60 y but not in those aged 60 y. The odds ratio for the younger women who consumed vitamin C supplements for 10 y compared with the younger women who never used supplements was 0.40 (95% CI: 0.18, 0.87). These associations were not affected by adjustment for the presence of nuclear opacities.
View this table:
TABLE 7 . Relation between the prevalence of cortical or posterior subcapsular (PSC) lens opacities and duration of vitamin supplement use1
View this table:
TABLE 8 . Relation between the prevalence of cortical lens opacities and duration of vitamin C supplement use by age1
Plasma nutrients and the prevalence of opacities
There were no significant associations between the prevalence of either cortical or PSC opacification and plasma concentrations of vitamins C and E and total carotenoids (Table 9). In addition, there were no significant interactions between the plasma vitamin concentrations and either age or smoking.
View this table:
TABLE 9 . Relation between the prevalence of cortical or posterior subcapsular (PSC) lens opacities and plasma nutrient concentrations1
DISCUSSION
In the NVP, we related nutrient intake to the odds of cortical and PSC cataracts in 492 women. None of the nutrient measures were related to the prevalence of opacities in the full sample (Tables 3 and 4), but significant interactions were seen between age and vitamin C intake for cortical opacities (Table 5), between the duration of vitamin C supplement use and cortical cataracts (Table 8), and between smoking status and folate (P = 0.02), -carotene (P = 0.02), ß-carotene (P = 0.005), and total carotenoids (P = 0.02) for PSC opacities (Table 6). Because we considered >100 separate interactions, we must consider the possibility that these associations are chance observations. However, most of the associations are consistent with, or can be rationalized by, prior information, and they provide valuable guides for future studies (3).
Cortical opacities
For women aged <60 y, vitamin C intakes 362 mg/d were associated with 57% lower odds of cortical cataracts than were intakes <140 mg/d, and use of vitamin C supplements for 10 y was associated with 60% lower odds of cortical cataracts than was no use. The cutoff for the referent (lowest) group for vitamin C intake was 2 times the recommended dietary allowance. This is interesting because recommended dietary allowances are considered to be significantly greater than the vitamin C intakes required to prevent vitamin C–deficiency-related disease.
However, 2 observations make these data appear confusing. First, the women with intakes in the fourth quintile category (ie, between 241 and 362 mg/d) experienced a higher prevalence of cortical opacities than did the women in the lowest intake category. Taken together, the data from the fourth and fifth quintile categories may suggest that there is no association between a vitamin C intake >241 mg/d and the odds of cortical opacities for women aged <60 y. Second, the observation of a higher risk of cortical cataracts in women aged 60 y who took vitamin C supplements for 5–9 y than in the other groups also makes it difficult to extract a consistent message from these data.
These concerns notwithstanding, the inverse correlation between vitamin C intake and the odds of cortical opacities in women aged <60 y is consistent with other data. First, in addition to the lower odds of cataracts for women aged <60 y with vitamin C intakes 362 mg/d and for users of vitamin C supplements for 10 y, there were also lower odds, albeit not significant, for women aged <60 y who used vitamin C supplements for 5–9 y. Many of the women with vitamin C intakes 362 mg/d used vitamin C supplements. Second, there is no apparent reason to think that intakes in the fourth quintile, which should be similar to intakes that are required to saturate eye tissues with vitamin C (48), would be procataractogenic and higher intakes (ie, the fifth quintile) would not. Third, another study suggests inverse relations between vitamin C intake and diminished risks of cataracts. Chasan-Taber et al (30) found a 28% diminished risk of cataract extraction in women aged <60 y who used vitamin C supplements for 10 y. These authors also evaluated the risk of reported (in the medical records obtained) cortical cataracts among all long-term vitamin C supplement users and observed a 49% decreased risk (NS). Fourth, Jacques and Chylack (12) showed 70% lower odds of cortical cataracts in persons with high total vitamin C intake, although this study did not consider potential interactions with age. Finally, a protective role for vitamin C is also consistent with observations in this NHS cohort and several other cohorts of decreased odds of nuclear opalescence in persons with higher vitamin C intakes from diet and supplements or with higher concentrations of vitamin C in blood (3, 8, 9, 12). The latter is reasonable support for this data because cataractogenic stress in the inner cortex and in the nucleus, both of which are metabolically quiescent, is likely to share many similarities.
Of the limited information available, not all studies support a protective role for vitamin C. Vitale et al (17) failed to observe any association between either plasma vitamin C or vitamin C intake and cortical opacification, even when stratified by age group. Leske et al (13) observed no association between cortical cataracts and total vitamin C intake, and Mares-Perlman et al (18) showed an increased prevalence of cortical opacities in persons who consumed supplements that contained vitamin C.
The benefit provided by long-term use of supplements is consistent with mechanistic considerations regarding cataract formation because cataract-related damage to the very long-lived lens proteins accrues after extended periods of time (3, 7, 49). At this time it is not clear why we observed an age interaction with cortical opacities, whereas age did not alter relations between odds of nuclear opacity and nutrient intake.
Posterior subcapsular opacities
Even fewer studies related nutrient intake or plasma concentrations to odds of PSC cataracts (11–13, 18, 33). In the present study, we noted the novel observation that the odds of PSC opacities were 74%, 71%, 72%, and 81% lower in never smokers with the highest intakes of folate, -carotene, ß-carotene, and total carotenoids, respectively, than in those with the lowest intakes (Table 6). After mutual adjustment for these nutrients, total carotenoid intake remained associated most strongly with PSC opacities among never smokers, although the relation with folate was only slightly weaker. This seems reasonable because diets that are high in carotenoids are generally also high in folate. In this data set, the correlation between carotenoid and folate intakes was 0.42 (9). Moreover, indications of protection by - and ß-carotene would need to be rationalized because we did not find measurable concentrations of - and ß-carotene when we analyzed fat-soluble antioxidants in human or beef lens samples (50, 51) (also see below). Instead, lutein and zeaxanthin, which were not associated with PSC opacities in the present study, were the most concentrated carotenoids in the lens.
A role for carotenoids can be rationalized because these compounds have roles either in forming components of membranes or in maintaining membrane integrity and facilitating regulated transport (3). As noted earlier, cells in the outermost lens tissues are affected in PSC cataracts. These epithelial cells are metabolically more active than cortical or nuclear cells and might require the highest level of membrane integrity for function. Thus, stresses that result in membrane and transport dysfunction may play greater roles in the etiology of this type of opacity.
The finding that the odds of PSC opacities are lower in never smokers, but not in smokers, with high carotenoid intakes may be related to the known effect of smoking in depressing carotenoid and other antioxidant concentrations. Perhaps the body stores of carotenoids required to achieve a decreased risk of PSC opacities are difficult to achieve in smokers. These data are consistent with data from Jacques and Chylack (12), who found that the odds of PSC opacities among those with high total plasma carotenoid concentrations were 80% lower than the odds among those with low plasma carotenoid concentrations. However, some other studies indicate that lutein and zeaxanthin intakes, but not carotene intakes, are related to the odds of nuclear cataract (3, 52). We did not see a relation between lutein and zeaxanthin intake and the odds of PSC opacities. These data emphasize the need for further investigation.
Summary
This study has some advantages over traditional retrospective studies: the nutrient intake data were collected prospectively and well before any of the women knew of their lens status (because we excluded women with diagnosed cataracts). Thus, prior knowledge of cataracts could not alter the relations we observed. Furthermore, most of the women had very early opacities and so should not have experienced any visual symptoms. Use of low-grade opacities should have predictive value, since early-grade opacities progress to more mature cataracts.
In the overall sample we observed no associations between antioxidant nutrient intake and either cortical or PSC opacities, but we noted significant and provocative associations in selected subgroups. The inverse association observed in the present study between vitamin C intake and the prevalence of cortical opacities in women aged <60 y and the fact that significantly decreased odds of cortical opacities were found only with a duration of vitamin C supplement use 10 y provide added support for a protective role for vitamin C against the formation of lens opacities. We also found that for PSC opacities there were decreased odds for nonsmokers with higher intakes of carotenoids. These data add more weight to the accumulating evidence that antioxidant nutrients can be exploited to alter the rates of development of these major (but less studied) forms of age-related opacities and provide indirect evidence that smoking attenuates the putative benefits of antioxidants.
ACKNOWLEDGMENTS
We acknowledge the invaluable assistance of project staff and the many others whose efforts supported this project. In particular, we thank Laura Bury, Rosaline Bowen, Esther Epstein, Mini Balaram, Sheila Crosby, Karen Corsano, Kate Saunders, Suzen Moeller, and Thomas Nowell. We are indebted to the nurses who participated in the study for their continuing contributions and cooperation. Finally, we acknowledge the support of Frank E Speizer, overall principal investigator for the NHS.
REFERENCES