- Email: [email protected]
Antioxidant vitamins and nuclear opacities
Antioxidant
Vitamins
The Longitudinal
and Nuclear
Opacities
Study of Cataract
M. Cristina Leske, MD, MPH,’ Leo T. Chyhck, Jr., MD,2 Qimei He, PhD,’ Sub-Yuh WU, MA,’ Elinor Schoenfeld, PhII,’ Judith Friend, MA,’ John Wolfe, P/ID,~ the Longitudinal Study of Cataract GroupJa2 Objective: The association of antioxidant nutrients and risk of nuclear opacification was evaluated in the Longitudinal Study of Cataract. Design: Nutritional data were collected at baseline on the 764 participants, which included assessment of dietary intake, use of vitamin supplements, and plasma levels of vitamin E. Ophthalmologic and other data were collected at baseline and at yearly follow-up visits, including lens photographs, which were graded using the Lens Opacities Classification System III protocol. Main Outcome Measures: Analyses examined whether the nutritional factors at baseline were related to increases in nuclear opacification at follow-up. The MULCOX2 approach, an extension of the Cox regression model, was used. Results are presented as relative risks (RRs) and 95% confidence intervals. Intervention: Intervention was not applicable. Results: The risk of nuclear opacification at follow-up was decreased in regular users of multivitamin supplements (RR = 0.69; 0.48-0.99) vitamin E supplements (RR = 0.43; 0.19-0.99), and in persons with higher plasma levels of vitamin E (RR = 0.58; 0.36-0.94). Conclusions: In regular users of multivitamin supplements, the risk of nuclear opacification was reduced by one third; in regular users of vitamin E supplements and persons with higher plasma levels of vitamin E, the risk was reduced by approximately half. These results are similar to those obtained in our earlier case-control study. Because these data are based on observational studies only, the results are suggestive but inconclusive. The possible effect of nutritional supplements on the lens requires confirmation by ongoing clinical trials. Ophthalmology
1998; 105:83 l-836
The importance of age-related cataract as a major cause of worldwide blindness’ hasled to many efforts to identify risk factors, particularly those that are potentially modifiable. Laboratory researchon the lens suggeststhe possible role of antioxidants, such as some vitamins, in cataractogenesis.2m5 In the Lens Opacities Case-Control Study (Locs),6,7 we reported that all types of opacities were less likely in regular users of multivitamin supplements (odds ratio [OR] = 0.63; 95% confidence interval [CI], 0.47-0.84) and in persons with higher dietary intake of individual antioxidant vitamins or higher levels of a dietary antioxidant index (OR = 0.40; 0.21-0.78).(’ In addition, higher plasma levels of vitamin E (a-tocopherol) were associatedwith a decreasedrisk of nuclear opacities (OR = 0.44; 0.21-0.90).7 In other reports, a decreased cataract risk also has been associated with the intake of some dietary or supplemental vitamins*-” and higher
plasmalevels of vitamins, particularly those with antioxidant potential.‘4-‘7 However, a temporal relationship between vitamin intake and development of lens opacities has not been established clearly. The aim of this longitudinal study was to determine the relationship between antioxidant nutrients at baseline and subsequentincreasesin nuclear opacification at follow-up. In contrast to previous studies, data on lens changes were collected at yearly intervals. In addition, results were based on masked, standardized gradings of lens photographs, which is a design feature to avoid misclassification and biases, and enhancesthe interpretation of study results. Data on associationswith other types of opacities will be presented separately.
Origmally received: July 22, 1997. Revision accepted: December 1, 1997. ’ University Medical Center at Stony Brook, Stony Brook, New York. *Harvard Medical School, Brigham and Women’s Hospital and the Center for Ophthalmic Research, Boston, Massachusetts. Supported by National Eye Institute Grant ROI EY0829 1. Reprint requests to M. Cristina Leske, MD, MPH, Department of Preventive Medicine, University Medical Center at Stony Brook, Stony Brook. NY 11794-8036.
Data Collection
Methods
The LongitudinalStudy of Cataract(LSC, 1989- 1993)wasan epidemiologicstudy of the natural history of lens opacities, conductedwith funding from the National Eye Institute. This study aimed to measure longitudinal
changes in nuclear, corti-
cal, andposteriorsubcapsular opacitiesin a clinic-basedpopulation, as well as to assess risk factors for cataract growth.“-*” For efficiency, the LSC baseline data were obtained using lens
photographsand risk factor data collectedaspart of an earlier
831
Ophthalmology
Volume
study (i.e., LOCS, 1985- 1988).6.7 However, the lens classification data from the case-control study were not applicable to LSC because each study required a different classification system for its own goals. To achieve its sample size (which allowed sufficiently precise measurements of progression), LSC recruited 764 of the case-control study participants and regraded their photographs. Informed consent had been obtained from all participants. The LSC methods and other results were presented elsewhere. “-“’ In brief, LSC follow-up visits began in September 1989 and continued at an average of l-year intervals over a 4year period. The median length of follow-up was 4.8 years (mean t standard deviation: 4.6 + 1.6 years), with 80% completing three or more visits. All visits followed a standardized protocol, which included a comprehensive ophthalmologic examination, standardized color slit-lamp (Carl Zeiss, Oberkochen, Germany), retroillumination (Neitz CTR, Neitz Instruments, Co., Torrance, CA), and Scheimpflug photographs (Topcon SL-45, Topcon America Corp., Paramus, NJ) of the lens.‘” Although the original case-control study used a simple lens classification system designed for its purposes,2’~22 the LSC used the Lens Opacities Classification System III (LOCS III),” which provides more grading categories to assess change. This system uses photographic standards to grade cataract severity on a decimal scale, which ranges from a low of 0.1 to a maximum of 6.9 for nuclear opalescence scores. Photographs from the baseline and the follow-up visits were graded according to the LOCS III protocoLz3 Gradings were completed by the same two independent masked observers, and discrepancies were resolved by consensus. Nuclear photographs were graded separately from other photographs. To avoid drift in gradings, all photographs for all visits of each participant were arranged randomly in the same carrousel and graded at the same session. These grading sessions were conducted over a 6-month period at the end of the study. Quality control of the gradings was monitored throughout these sessions. The 95% tolerance limits for intra- and intergrader variation ranged from 20.4 to 20.6 in the LOCS III nuclear grading scale.”
Data Analyses Of the 764 LSC participants, 744 had data on nuclear changes. Data from these participants were analyzed to examine the relationship between antioxidant nutrients at baseline (case-control study visit) and subsequent increases in nuclear opacification, as measured by changes in LOCS III nuclear scores at followup. To allow for minor fluctuations in scores due to grader or photograph variation between visits, an increase in nuclear opacities at follow-up was considered present only when the change in LOCS III nuclear scores exceeded a threshold of 0.7. This threshold was based on the highest variation observed in the 95% tolerance limits for nuclear gradings, as derived both from quality control results of this study and from previous LOCS III reproducibility evaluations2” For eyes with nuclear gradings for two visits only, increases in nuclear opacities at follow-up simply were defined according to these criteria, that is, whenever the nuclear score at the follow-up visit increased at least 0.7 compared to the nuclear score at baseline. However, more than four fifths of the eyes had nuclear gradings for three or more visits. For these eyes, an increase in nuclear opacities was based on confirmed increases at follow-up, which used all available nuclear scores. The definition of confirmed increases was an increase in nuclear scores of at least 0.7 from baseline to last visit, plus a similar increase in nuclear scores between any two LSC visits, which was either followed by another in-
832
105, Number 5, May
1998
crease in scores or by no change in scores at subsequent visits. For eyes with cataract surgery, an increase in nuclear opacities also was considered present if the nuclear opacities were the most severe cataract type (i.e., the nuclear score was proportionally higher than scores for cortical and posterior subcapsular opacities) at the last visit before surgery. The status of nuclear opacity can vary between fellow eyes of the same patient. Of the 698 persons with nuclear opacities in both eyes, 159 had increases in nuclear opacities in at least 1 eye (50 in left eyes only, 45 in right eyes only, and 64 in both eyes). The correlation between the event times of the 2 eyes for these 64 persons was 0.51 (P value, 0.0001). To use the information collected from each eye in this longitudinal study fully, the MULCOX224,2” approach was used in the analysis. This approach is an extension of the Cox regression model and allows eye-specific analyses while accounting for correlation between fellow eyes. An increase in nuclear opacities, as defined above, was the event endpoint. The time to reach such an endpoint was defined as the midpoint between the followup visit in which the increase in nuclear opacities was first observed and its preceding visit. Associations with baseline antioxidant nutrients were evaluated individually in separate models that adjusted for potential confounding variables of age, gender, race, education, current smoking status, and coexisting cortical and posterior subcapsular opacities at baseline” (defined as LOCS III cortical or posterior subcapsular score 22.0). Baseline nutritional variables evaluated included dietary intake of vitamins C, E, riboflavin, and carotene, as assessed with the food frequency questionnaire described and evaluated by Block et al’“-‘*; use of vitamin supplements (i.e., multivitamins and vitamins C and E on a regular basis) (at least once a week for at least 1 year); and plasma level of vitamin E (a-tocopherol).’ For each dietary nutrient and for plasma level of vitamin E, the nutrient was grouped as low if below the lowest quintile for the study population, high if above the highest quintile, and medium otherwise. A dietary antioxidant index, which was defined by quintiles of dietary vitamins C, E, riboflavin, and carotene (the sum of LYcarotene, P-carotene, lutein, cryptoxanthin, other xanthins, and lycopene),h also was evaluated. The results of the MULCOX2 analyses are presented as relative risks (RRs) and their 95% confidence intervals.
Results Table 1 presents the characteristics of the 744 participants at baseline. The median age was 65 years and approximately half had nuclear opacities (LOCS III nuclear score ~2.0); one fourth used multivitamins and approximately two fifths used some kind of vitamin supplements. During follow-up, 177 or 24% of the participants had increases in nuclear opacities in at least 1 eye. Table 2 compares the use of multivitamin supplements and plasma levels of vitamin E between these 177 participants and the other 567 who had no increases in nuclear opacities in either eye at follow-up. In these comparisons, patients who had increased severity of nuclear opacities at follow-up were lessfrequent users of multivitamin supplements than others (21% vs. 26.1%). These differences in multivitamin use were consistent for every demographic subcategory, except nonwhites, the results of which could be related to the relatively small size of that group. In addition to using fewer supplements, the group with increased nuclear opacities had lower plasma levels of vitamin E. Figure 1 compares increases in nuclear opacities at follow-
Leske et al * Antioxidant Table
1. Characteristics
of Participants
at Basehne
AS
Mean IT SD (yrs) Median (yrs) 40-49 yrs (%) 50-59 yrs (%) 60-69 yrs (%) 70+ yrs (%) Males (%) White race (%) Education (< 12 yrs) (%) Nonprofessional occupation (%) Nuclear opacities m at least one eye (%) Co-existing opacities (%) Cortml KC Current smoking (%) Nutritional supplements Multivitamins Vitamin C Vitamin E Any vitamms Plasma vttamin E (mg/dl) Mean t SD Median PSC = posterior
supcapsular
(N
Vitamins
Percent of persons M& use of vitamm E supplements.
1.48 2 0.57 1.38
Table 2. Use of Multivitamin Supplements and Plasma Levels Vitamin E at Baseline in Persons with anal without Increases in Nuclear Opactttes at Follow-up
opacities
? No increase in nuclear
Any Increase* (n = 177)
No Increaset (n = 567)
21.0
26.1
24.6 19.3
28.2 23.9
15.6 25.3
19.5 31.6
20.7 25.0
27.6 18.2
23 5 20.4
31.2 24.3
15.8 21.9 (n = 171) 1.44 +- 0 59 1.33 at follow-up
opacities
\
in either
opacities
hy tune and
251
I
Users of multivitamin supplements
18.3 27.5 (n = 527) 1.49 + 0.57 I .40
eye.
increased nuclear
Table 3 presents the results of the MULCOX2 analyses to evaluate the effects of nutritional factors on nuclear opacifications at follow-up. Regular use of multivitamin supplements (RR = 0.69; 0.48-0.99), vitamin E supplements (RR = 0.43; 0.19-0.99), and higher levels of plasma vitamin E (RR = 0.58; 0.36-0.94) were associated with a lower risk. No associations were found for other vitamin supplements or dietary vitamins, although all RRs were under unity. Table 4 presents results of additional analyses considering duration of supplement use at follow-up. In these analyses, persons who used multivitamins regularly at baseline and continued to use the supplements for at least one follow-up visit also had a significantly decreased risk of progression compared with those who never used multivitamin supplements (RR = 0.55; 0.36-0.85). Dose-response analyses that categorized the supplement-use variables into “none,” “users for less than 5 years, ” and “users for at least 5 years” also suggested negative associations of nuclear opacification with increasing duration of multivitamin use (P value of 0.07) and vitamin E (P value of 0.04). Table 4 also presents results of similar analyses based on plasma levels of vitamin E, which also showed dose-response effects (P value of 0.04).
1
2
3
4
5
opacities
hy time and
Years
m at least one eye.
at follow-up
I-
.-t c .=
Figure 1.
of
in nuclear
Users of vitamin E supplements \
Years 24.9 10.6 7.3 40.4
up between users and nonusers of vitamin E supplements. The percent of nonusers experiencing increases in nuclear opacities is more than twice as high as in users of vitamin E supplements. Similar curves for users and nonusers of multivitamins are presented in Figure 2.
* Increase
Non-users of vitamin E supplements \,
$
25.5 8.1 16.4
cataract.
Multivitamm supplements (%) Overall use By demographic characteristics Age <65 yrs ~65 yrs Gender Male Female Race White Nonwhite Occupation Professional Nonprofessional Education <12 yrs 212 yrs Plasma vitamin E (mg/dl) Mean + SD Median
Opacities
25r
= 744)
63.8 + 8.0 65.0 5.8 19.4 52.7 22.2 44.9 86.6 14.7 25.3 47.0
and Nuclear
Figure 2. Percent use of multivitamm
of persons with increased nuclear supplements.
833
Volume 105, Number
Ophthalmology
Table 3. Relative Risks of Increases of Nuclear Opacification According
to Antioxidant
the 744 Participants
Nutrient
Status
at Baseline
(Based on MULCOXZ
Models)
RR Use of Use of Use of Plasma High Dietary High
multivitamm supplements vitamin C supplements wtamm E supplements level of vitamin E versus low? antioxidant index versus low$
RR = relative
risk; CI = confidence
among
95%
CI
0.69 0.80 0.43
0.48-0.99* 0.49- 1.32 0.19-0.99*
0.58
0.36-0.94*
0.71
0.43-1.18
interval.
* P < 0.05. t Categories mg/dl.
based on quintile
$ Categories based on quintde riboflavin, and carotene.
values: high,
>1.788
values of dietary
mg/dl;
vitamin
low, cl.078 C, vitamm
E,
Discussion This longitudinal study determined whether baseline antioxidant nutrients were related to subsequent changes in nuclear opacities at follow-up. Regular usersof multivitamins or vitamin E supplements at baseline had a decreasedrisk of nuclear opacification at follow-up. Furthermore, higher plasma levels of vitamin E at baseline also were associated with a lower risk of nuclear opacities (Table 3). These findings are similar to the multivitamin and vitamin E findings of our earlier case-control study,6,7although LSC used a different lens classification system and had a smaller sample size. Several studies have examined the relationship between multivitamin supplements, vitamin E, and the risk of cataract or lens opacities, but comparisonsamong studies are limited because of their different methodologies. Although our earlier case-control study6*’shared several design features with parallel case-control studies in Indiaz9and Italy,30 the results of the three studiesare difficult to compare,“’ especially those concerning intake of nutritional supplementsand vitamin E. The Indian study had different inclusion and exclusion criteria (e.g., it used its own lens classification system and required visual acuity loss for all cases). It also excluded users of nutritional supplements, whereas the Italian study had few supplement users, a reason that might explain the lack of a significant association. Furthermore, the dietary intake data differed among studies (e.g., the Indian study and the Italian study did not assessdietary vitamin E or total carotenoids), so that associationswith those dietary antioxidants could not be evaluated. Given these differences among the three studies, the lack of consistent results for these variables is not surprising. A more recent casecontrol study in Italy, which did assessdietary vitamin E, found that cataract extraction was associated with a lower vitamin E intake.r3 In a nutritional intervention trial in China, the use of multivitamins and mineral supplements was associated
834
5, May 1998
with a decreasedprevalence of nuclear opacities.” In the Physicians’ Health Study, men who used multivitamin supplementsexperienced a decreasedrisk of cataract extraction after 5 years of follow-up.” Associations between a decreased cataract risk and nutritional supplements, such as vitamin C,s,9vitamin E,8 or multivitamins,” were reported in other studies. In another report, higher plasma levels of vitamin E also were associatedwith a reduced frequency of nuclear opacities at a follow-up examination.14 The same study, however, found no associations of plasmabeta-carotene, ascorbic acid, and overall antioxidant statusto the risk of nuclear opacities.14An increased risk of cataract in persons with low levels of an antioxidant index, which included plasmalevels of ascorbic acid, vitamin E, and carotenoids, also was reported in a smaller study.15A Finnish study reported associationsof plasma levels of vitamin E (a-tocopherol) and cataract extraction after a median follow-up of 15 yearsI Another Finnish study reported the association between low plasma vitamin E and progression of early cortical cataract.” In contrast to the LSC, none of these studies, except the latter, were basedon multiple follow-up examinations with lens assessment.The longitudinal design of the LSC also addressedmajor possible sources of biases and misclassification. The nutritional data were collected at baseline using standardized methods, and the lens data were based on maskedphotographic classification of opacity type and severity. Our findings of a lower risk of nuclear opacities in regular users of multivitamins or vitamin E supplements and in personswith higher levels of plasma vitamin E thus provide evidence to support these relationships. Because LSC was an observational study, however, appropriate caution must be exercised in evaluating the results, especially becausethe data from all studies are not entirely consistent. Interpretation of our findings must consider alternative explanations, such as failure to adjust for confounding variables. Although the analyses included a number of
Table
4.
Dose-Response
Results
for Increases
m Nuclear
Opacities According to Use of Multivitamm Supplements, Vitamin E Supplements, and Plasma Level of Vitamin E RR (95% Multivitamin supplements Nonusers Users <5 yrs Users 25 yrs Dose-response Vitamin E supplements Nonusers Users <5 yrs Users 25 yrs Dose-response Plasma vitamm E Low qumtile Median 3 qumtiles High quintile Dose-response RR = relative
risk; CI = confidence
CI)
1 .oo 0.68 (0.42-1.11) 0.69 (0.42-1.11)
1 .oo 0.81 (0.32-2.06) 0.26 (0.06- 1.06)
1.0 0.72 (0.49-1.05) 0.58 (0.36-0.94)
Interval.
P
0.12 0.13 0.07
0.66 0.06 0.04
0.09 0.03 0.04
Leske et al * Antioxidant potential confounders, the possibility
Vitamins
of missing an un-
known confounder can never be dismissed. For example, the use of multivitamin supplements may be a marker for demographic, socioeconomic, or “lifestyle” related factors, which may be the true determinants of risk. This explanation seems unlikely, however, because persons who progressed had a consistently lower intake of vitamin supplements, regardless of age, gender, education, and occupation (Table 2). The association with antioxidant vitamin status also is supported by the biochemical data, because persons with higher plasma vitamin E had a lower risk of nuclear opacities. The MULCOX2 models used in the analyses were based on data from both eyes of a person. For completeness, we conducted separate analyses on data for left and right eyes in Cox regression analyses. The results from these analyses are consistent with those from the MULCOX2 models, with the expected loss of statistical power. A spurious association between nutritional factors and nuclear opacification would result if losses to follow-up were related, not only to nuclear opacification but also to supplement use or plasma levels of vitamin E.32 For example, our results could be caused by bias if losses to follow-up were more frequent among persons who used
supplements and also had increases in opacities than in others. To examine this possible source of biases, chisquare tests evaluated whether losses to follow-up
varied
among persons with different levels of individual nutrient variables. Additionally, log-linear models were used to detect possible three-way
nuclear opacification,
interactions
among increases in
losses to follow-up, and nutritional
factors. No statistically significant results were found in these analyses. Although these negative results do not necessarily imply the absence of biases, our evaluations suggest that losses to follow-up were less likely to lead to large biases when estimating the RRs, In conclusion, nuclear opacities are the most frequent
type reported by cataract prevalence studies in the United States and Europe.33-36 In contrast to cortical opacities, nuclear opacities often cause visual loss that requires cataract surgery. Given the high prevalence of nuclear opacities, the identification
of risk factors amenable to modifi-
cation or intervention has high priority. Our results suggest that use of nutritional supplements and of higher plasma vitamin E levels are related to a decreased risk of progression of nuclear opacities. These conclusions are
strengthened by the longitudinal design of the study and the careful lens assessment. Because of the observational nature of the study and the possibility of biases, our results are suggestive, but not conclusive, and should be interpreted cautiously. For example, observational studies suggested a protective effect of vitamin E and beta-carotene on lung cancer,37,38 whereas clinical trials found an opposite effect or no effect.39-4’ To assess the effect of vitamin supplementation on the development of nuclear opacities,
it is necessary to obtain conclusive evidence from ongoing clinical trials, such as from the National Eye Institute’s Age-Related Eye Disease Study. Currently, we emphasize that there is no firm evidence to support a recommendation to use nutritional supplements for cataract prevention.
and Nuclear
Opacities
References 1. Thylefors B. The role of international ophthalmology in blindness prevention. Am J Ophthalmol 1995; 119:229-30. 2. Bunce GE, Kinoshita J, Horwitz J. Nutritional factors in cataract. Annu Rev Nutr 1990; 10:233-54. 3. Lens and Cataract Panel. A report of the National Advisory Eye Council. In: National Advisory Eye Council (U.S.) Vision Research: A National Plan: 1994- 1998. A report of the National Advisory Eye Council. Washington, DC: US. Dept. of Health and Human Services, Public Health Service, National Institutes of Health, 1993 (NIH Publication; 93-3186). 4. Maisel H, ed. The Ocular Lens: Structure, Function, and Pathology. New York: Dekker, 1985. 5. Senile cataract and vitamin nutrition. Nutr Rev 1989; 47:326-8. 6. Leske MC, Chylack LT Jr, Wu SY. The Lens Opacities Case-Control Study. Risk factors for cataract. Arch Ophthalmol 1991; 109:244-51. 7. Leske MC, Wu SY, Hyman L, et al. Biochemical factors in the lens opacities. Case-control study. The Lens Opacities Case-Control Study Group. Arch Ophthalmol 1995; 113:1113-9. 8. Robertson JM, Donner AP, Trevithick JR. Vitamin E intake and risk of cataracts in humans. Ann NY Acad Sci 1989;570:372-82. 9. Hankinson SE, Stampfer MJ, Seddon JM, et al. Nutrient intake and cataract extraction in women: a prospective study. BMJ 1992;305:335-9. 10. Sperduto RD, Hu TS, Milton RC, et al. The Linxian cataract studies. Two nutrition intervention trials. Arch Ophthalmol 1993; 111:1246-53. 11. Seddon JM, Christen WG, Manson JE, et al. The use of vitamin supplements and the risk of cataract among US male physicians, Am J Public Health 1994;84:788-92. 12. Mares-Perlman JA, Klein BE, Klein R, Ritter L. Relation between lens opacities and vitamin and mineral supplement use. Ophthalmology 1994; 101:315-25. 13. Tavani A, Negri E, La Vecchia C. Food and nutrient intake and risk of cataract. Ann Epidemiol 1996;6:41-6. 14. Vitale S, West S, Hallfrisch J, et al. Plasma antioxidants and risk of cortical and nuclear cataract. Epidemiology 1993;4:195-203. 15. Knekt P, Heliovaara M, Rissanen A, et al. Serum antioxidant vitamins and risk of cataract. BMJ 1992;305: 13924. 6. Jacques PF, Chylack LT Jr, McGandy RB, Hartz SC. Antioxidant status in persons with and without senile cataract. Arch Ophthalmol 1988; 106:337-40. 7. Rouhiainen P, Rouhiainen H, Salonen JT. Association between low plasma vitamin E concentration and progression of early cortical lens opacities. Am J Epidemiol 1996; 144:496-500. 8. Leske MC, Chylack LT Jr, Wu SY, et al. Incidence and progression of nuclear opacities in the Longitudinal Study of Cataract. Ophthalmology 1996; 103:705-12. 9. Leske MC, Chylack LT, He Q, et al. Incidence and progression of cortical and posterior subcapsular opacities: the Longitudinal Study of Cataract. Ophthalmology. 1997; 104: 1985-93. 20. Leske MC, Chylack LT, He Q, et al. Risk factors for nuclear opalescence in a longitudinal study. Am J Epidemiol 1998;147:36-41.
83.5
Ophthalmology
Volume 105, Number
21. Chylack LT Jr, Leske MC, Sperduto R, et al. Lens opacities classification system. Arch Ophthalmol 1988; 106:330-4. 22. Leske MC, Chylack LT Jr, Sperduto R, et al. Evaluation of a Lens Opacities Classification System. Arch Ophthalmol 1988; 106:327-9. 23. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. Arch Ophthalmol 1993; Ill:83 l-6. 24. Lin DY. MULCOX2: a general computer program for the Cox regression analysis of multivariate failure time data. Comput Methods Programs Biomed 1993;40:279-93. 25. Wei LJ, Lin DY, Weissfeld L. Regression analysis of multivariate incomplete failure time data by modeling marginal distributions. JASA 1989;84:1065-71. 26. Cummings SR, Block G, McHenry K, Baron RB. Evaluation of two food frequency methods of measuring dietary calcium intake. Am J Epidemiol 1987; 126:796-802. 27. Block G, Dresser CM, Hartman AM, Carroll MD. Nutrient sources in the American diet: quantitative data from the NHANES II survey. I: vitamins and minerals. Am J Epidemiol 198.5; 122:13-26. 28. Block G, Dresser CM, Hartman AM, Carroll MD. Nutrient sources in the American diet: quantitative data from the NHANES II survey. II. Macronutrients and fat. Am J Epidemiol 1985; 122:27-40. 29. Mohan M, Sperduto RD, Angra SK, et al. India-US casecontrol study of age-related cataracts. India-US Case Control Study Group [published erratum appears in Arch Ophthalmol 1989; 107:1288]. Arch Ophthalmol 1989; 107:6706. 30. The Italian-American Cataract Study Group. Risk factors for age-related cortical, nuclear, and posterior subcapsular cataracts. Am J Epidemiol 1991; 133541-53. 31. Schoenfeld ER, Leske MC, Wu SY. Recent epidemiologic studies on nutrition and cataract in India, Italy and the United States. J Am Co11 Nutr 1993; 12:521-6.
836
5, May 1998
32. Kelsey JL, Thompson WD, Evans AS. Methods in Observational Epidemiology. New York Oxford University Press, 1986; 10 I (Monographs in epidemiology and biostatistics; v. IO). 33. Klein BE, Klein R, Linton KL. Prevalence of age-related lens opacities in a population. The Beaver Dam Eye Study. Ophthalmology 1992;99:546-52. 34. Sperduto R, Hiller R. The prevalence of nuclear, cortical, and posterior subcapsular lens opacities in a general population sample. Ophthalmology 1984;91:815-8. 35. Giuffre G, Giammanco R, Di Pace F, Ponte F. Casteldaccia eye study: prevalence of cataract in the adult and elderly population of a Mediterranean town. Int Ophthalmol 19945; l&363-71. 36. Hirvela H, Luukinen H, Laatikainen L. Prevalence and risk factors of lens opacities in the elderly in Finland. A population-based study. Ophthalmology 1995; 102: 108- 17. 37. Peto R, Doll R, Buckley JD, et al. Can dietary beta-carotene materially reduce human cancer rates? Nature 1981; 290:201-8. 38. Committee on Diet, Nutrition, and Cancer, Assembly of Life Sciences, National Research Council. Diet, nutrition, and cancer. Washington, DC: National Academy Press, 1982. 39. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (ATC). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994;330: 1029-35. 40. Hennekens CH, Buring JE, Manson JE, et al. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med 1996; 334: 1 145-9. 41. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996; 334: 1150-5.
Vitamins
The Longitudinal
and Nuclear
Opacities
Study of Cataract
M. Cristina Leske, MD, MPH,’ Leo T. Chyhck, Jr., MD,2 Qimei He, PhD,’ Sub-Yuh WU, MA,’ Elinor Schoenfeld, PhII,’ Judith Friend, MA,’ John Wolfe, P/ID,~ the Longitudinal Study of Cataract GroupJa2 Objective: The association of antioxidant nutrients and risk of nuclear opacification was evaluated in the Longitudinal Study of Cataract. Design: Nutritional data were collected at baseline on the 764 participants, which included assessment of dietary intake, use of vitamin supplements, and plasma levels of vitamin E. Ophthalmologic and other data were collected at baseline and at yearly follow-up visits, including lens photographs, which were graded using the Lens Opacities Classification System III protocol. Main Outcome Measures: Analyses examined whether the nutritional factors at baseline were related to increases in nuclear opacification at follow-up. The MULCOX2 approach, an extension of the Cox regression model, was used. Results are presented as relative risks (RRs) and 95% confidence intervals. Intervention: Intervention was not applicable. Results: The risk of nuclear opacification at follow-up was decreased in regular users of multivitamin supplements (RR = 0.69; 0.48-0.99) vitamin E supplements (RR = 0.43; 0.19-0.99), and in persons with higher plasma levels of vitamin E (RR = 0.58; 0.36-0.94). Conclusions: In regular users of multivitamin supplements, the risk of nuclear opacification was reduced by one third; in regular users of vitamin E supplements and persons with higher plasma levels of vitamin E, the risk was reduced by approximately half. These results are similar to those obtained in our earlier case-control study. Because these data are based on observational studies only, the results are suggestive but inconclusive. The possible effect of nutritional supplements on the lens requires confirmation by ongoing clinical trials. Ophthalmology
1998; 105:83 l-836
The importance of age-related cataract as a major cause of worldwide blindness’ hasled to many efforts to identify risk factors, particularly those that are potentially modifiable. Laboratory researchon the lens suggeststhe possible role of antioxidants, such as some vitamins, in cataractogenesis.2m5 In the Lens Opacities Case-Control Study (Locs),6,7 we reported that all types of opacities were less likely in regular users of multivitamin supplements (odds ratio [OR] = 0.63; 95% confidence interval [CI], 0.47-0.84) and in persons with higher dietary intake of individual antioxidant vitamins or higher levels of a dietary antioxidant index (OR = 0.40; 0.21-0.78).(’ In addition, higher plasma levels of vitamin E (a-tocopherol) were associatedwith a decreasedrisk of nuclear opacities (OR = 0.44; 0.21-0.90).7 In other reports, a decreased cataract risk also has been associated with the intake of some dietary or supplemental vitamins*-” and higher
plasmalevels of vitamins, particularly those with antioxidant potential.‘4-‘7 However, a temporal relationship between vitamin intake and development of lens opacities has not been established clearly. The aim of this longitudinal study was to determine the relationship between antioxidant nutrients at baseline and subsequentincreasesin nuclear opacification at follow-up. In contrast to previous studies, data on lens changes were collected at yearly intervals. In addition, results were based on masked, standardized gradings of lens photographs, which is a design feature to avoid misclassification and biases, and enhancesthe interpretation of study results. Data on associationswith other types of opacities will be presented separately.
Origmally received: July 22, 1997. Revision accepted: December 1, 1997. ’ University Medical Center at Stony Brook, Stony Brook, New York. *Harvard Medical School, Brigham and Women’s Hospital and the Center for Ophthalmic Research, Boston, Massachusetts. Supported by National Eye Institute Grant ROI EY0829 1. Reprint requests to M. Cristina Leske, MD, MPH, Department of Preventive Medicine, University Medical Center at Stony Brook, Stony Brook. NY 11794-8036.
Data Collection
Methods
The LongitudinalStudy of Cataract(LSC, 1989- 1993)wasan epidemiologicstudy of the natural history of lens opacities, conductedwith funding from the National Eye Institute. This study aimed to measure longitudinal
changes in nuclear, corti-
cal, andposteriorsubcapsular opacitiesin a clinic-basedpopulation, as well as to assess risk factors for cataract growth.“-*” For efficiency, the LSC baseline data were obtained using lens
photographsand risk factor data collectedaspart of an earlier
831
Ophthalmology
Volume
study (i.e., LOCS, 1985- 1988).6.7 However, the lens classification data from the case-control study were not applicable to LSC because each study required a different classification system for its own goals. To achieve its sample size (which allowed sufficiently precise measurements of progression), LSC recruited 764 of the case-control study participants and regraded their photographs. Informed consent had been obtained from all participants. The LSC methods and other results were presented elsewhere. “-“’ In brief, LSC follow-up visits began in September 1989 and continued at an average of l-year intervals over a 4year period. The median length of follow-up was 4.8 years (mean t standard deviation: 4.6 + 1.6 years), with 80% completing three or more visits. All visits followed a standardized protocol, which included a comprehensive ophthalmologic examination, standardized color slit-lamp (Carl Zeiss, Oberkochen, Germany), retroillumination (Neitz CTR, Neitz Instruments, Co., Torrance, CA), and Scheimpflug photographs (Topcon SL-45, Topcon America Corp., Paramus, NJ) of the lens.‘” Although the original case-control study used a simple lens classification system designed for its purposes,2’~22 the LSC used the Lens Opacities Classification System III (LOCS III),” which provides more grading categories to assess change. This system uses photographic standards to grade cataract severity on a decimal scale, which ranges from a low of 0.1 to a maximum of 6.9 for nuclear opalescence scores. Photographs from the baseline and the follow-up visits were graded according to the LOCS III protocoLz3 Gradings were completed by the same two independent masked observers, and discrepancies were resolved by consensus. Nuclear photographs were graded separately from other photographs. To avoid drift in gradings, all photographs for all visits of each participant were arranged randomly in the same carrousel and graded at the same session. These grading sessions were conducted over a 6-month period at the end of the study. Quality control of the gradings was monitored throughout these sessions. The 95% tolerance limits for intra- and intergrader variation ranged from 20.4 to 20.6 in the LOCS III nuclear grading scale.”
Data Analyses Of the 764 LSC participants, 744 had data on nuclear changes. Data from these participants were analyzed to examine the relationship between antioxidant nutrients at baseline (case-control study visit) and subsequent increases in nuclear opacification, as measured by changes in LOCS III nuclear scores at followup. To allow for minor fluctuations in scores due to grader or photograph variation between visits, an increase in nuclear opacities at follow-up was considered present only when the change in LOCS III nuclear scores exceeded a threshold of 0.7. This threshold was based on the highest variation observed in the 95% tolerance limits for nuclear gradings, as derived both from quality control results of this study and from previous LOCS III reproducibility evaluations2” For eyes with nuclear gradings for two visits only, increases in nuclear opacities at follow-up simply were defined according to these criteria, that is, whenever the nuclear score at the follow-up visit increased at least 0.7 compared to the nuclear score at baseline. However, more than four fifths of the eyes had nuclear gradings for three or more visits. For these eyes, an increase in nuclear opacities was based on confirmed increases at follow-up, which used all available nuclear scores. The definition of confirmed increases was an increase in nuclear scores of at least 0.7 from baseline to last visit, plus a similar increase in nuclear scores between any two LSC visits, which was either followed by another in-
832
105, Number 5, May
1998
crease in scores or by no change in scores at subsequent visits. For eyes with cataract surgery, an increase in nuclear opacities also was considered present if the nuclear opacities were the most severe cataract type (i.e., the nuclear score was proportionally higher than scores for cortical and posterior subcapsular opacities) at the last visit before surgery. The status of nuclear opacity can vary between fellow eyes of the same patient. Of the 698 persons with nuclear opacities in both eyes, 159 had increases in nuclear opacities in at least 1 eye (50 in left eyes only, 45 in right eyes only, and 64 in both eyes). The correlation between the event times of the 2 eyes for these 64 persons was 0.51 (P value, 0.0001). To use the information collected from each eye in this longitudinal study fully, the MULCOX224,2” approach was used in the analysis. This approach is an extension of the Cox regression model and allows eye-specific analyses while accounting for correlation between fellow eyes. An increase in nuclear opacities, as defined above, was the event endpoint. The time to reach such an endpoint was defined as the midpoint between the followup visit in which the increase in nuclear opacities was first observed and its preceding visit. Associations with baseline antioxidant nutrients were evaluated individually in separate models that adjusted for potential confounding variables of age, gender, race, education, current smoking status, and coexisting cortical and posterior subcapsular opacities at baseline” (defined as LOCS III cortical or posterior subcapsular score 22.0). Baseline nutritional variables evaluated included dietary intake of vitamins C, E, riboflavin, and carotene, as assessed with the food frequency questionnaire described and evaluated by Block et al’“-‘*; use of vitamin supplements (i.e., multivitamins and vitamins C and E on a regular basis) (at least once a week for at least 1 year); and plasma level of vitamin E (a-tocopherol).’ For each dietary nutrient and for plasma level of vitamin E, the nutrient was grouped as low if below the lowest quintile for the study population, high if above the highest quintile, and medium otherwise. A dietary antioxidant index, which was defined by quintiles of dietary vitamins C, E, riboflavin, and carotene (the sum of LYcarotene, P-carotene, lutein, cryptoxanthin, other xanthins, and lycopene),h also was evaluated. The results of the MULCOX2 analyses are presented as relative risks (RRs) and their 95% confidence intervals.
Results Table 1 presents the characteristics of the 744 participants at baseline. The median age was 65 years and approximately half had nuclear opacities (LOCS III nuclear score ~2.0); one fourth used multivitamins and approximately two fifths used some kind of vitamin supplements. During follow-up, 177 or 24% of the participants had increases in nuclear opacities in at least 1 eye. Table 2 compares the use of multivitamin supplements and plasma levels of vitamin E between these 177 participants and the other 567 who had no increases in nuclear opacities in either eye at follow-up. In these comparisons, patients who had increased severity of nuclear opacities at follow-up were lessfrequent users of multivitamin supplements than others (21% vs. 26.1%). These differences in multivitamin use were consistent for every demographic subcategory, except nonwhites, the results of which could be related to the relatively small size of that group. In addition to using fewer supplements, the group with increased nuclear opacities had lower plasma levels of vitamin E. Figure 1 compares increases in nuclear opacities at follow-
Leske et al * Antioxidant Table
1. Characteristics
of Participants
at Basehne
AS
Mean IT SD (yrs) Median (yrs) 40-49 yrs (%) 50-59 yrs (%) 60-69 yrs (%) 70+ yrs (%) Males (%) White race (%) Education (< 12 yrs) (%) Nonprofessional occupation (%) Nuclear opacities m at least one eye (%) Co-existing opacities (%) Cortml KC Current smoking (%) Nutritional supplements Multivitamins Vitamin C Vitamin E Any vitamms Plasma vttamin E (mg/dl) Mean t SD Median PSC = posterior
supcapsular
(N
Vitamins
Percent of persons M& use of vitamm E supplements.
1.48 2 0.57 1.38
Table 2. Use of Multivitamin Supplements and Plasma Levels Vitamin E at Baseline in Persons with anal without Increases in Nuclear Opactttes at Follow-up
opacities
? No increase in nuclear
Any Increase* (n = 177)
No Increaset (n = 567)
21.0
26.1
24.6 19.3
28.2 23.9
15.6 25.3
19.5 31.6
20.7 25.0
27.6 18.2
23 5 20.4
31.2 24.3
15.8 21.9 (n = 171) 1.44 +- 0 59 1.33 at follow-up
opacities
\
in either
opacities
hy tune and
251
I
Users of multivitamin supplements
18.3 27.5 (n = 527) 1.49 + 0.57 I .40
eye.
increased nuclear
Table 3 presents the results of the MULCOX2 analyses to evaluate the effects of nutritional factors on nuclear opacifications at follow-up. Regular use of multivitamin supplements (RR = 0.69; 0.48-0.99), vitamin E supplements (RR = 0.43; 0.19-0.99), and higher levels of plasma vitamin E (RR = 0.58; 0.36-0.94) were associated with a lower risk. No associations were found for other vitamin supplements or dietary vitamins, although all RRs were under unity. Table 4 presents results of additional analyses considering duration of supplement use at follow-up. In these analyses, persons who used multivitamins regularly at baseline and continued to use the supplements for at least one follow-up visit also had a significantly decreased risk of progression compared with those who never used multivitamin supplements (RR = 0.55; 0.36-0.85). Dose-response analyses that categorized the supplement-use variables into “none,” “users for less than 5 years, ” and “users for at least 5 years” also suggested negative associations of nuclear opacification with increasing duration of multivitamin use (P value of 0.07) and vitamin E (P value of 0.04). Table 4 also presents results of similar analyses based on plasma levels of vitamin E, which also showed dose-response effects (P value of 0.04).
1
2
3
4
5
opacities
hy time and
Years
m at least one eye.
at follow-up
I-
.-t c .=
Figure 1.
of
in nuclear
Users of vitamin E supplements \
Years 24.9 10.6 7.3 40.4
up between users and nonusers of vitamin E supplements. The percent of nonusers experiencing increases in nuclear opacities is more than twice as high as in users of vitamin E supplements. Similar curves for users and nonusers of multivitamins are presented in Figure 2.
* Increase
Non-users of vitamin E supplements \,
$
25.5 8.1 16.4
cataract.
Multivitamm supplements (%) Overall use By demographic characteristics Age <65 yrs ~65 yrs Gender Male Female Race White Nonwhite Occupation Professional Nonprofessional Education <12 yrs 212 yrs Plasma vitamin E (mg/dl) Mean + SD Median
Opacities
25r
= 744)
63.8 + 8.0 65.0 5.8 19.4 52.7 22.2 44.9 86.6 14.7 25.3 47.0
and Nuclear
Figure 2. Percent use of multivitamm
of persons with increased nuclear supplements.
833
Volume 105, Number
Ophthalmology
Table 3. Relative Risks of Increases of Nuclear Opacification According
to Antioxidant
the 744 Participants
Nutrient
Status
at Baseline
(Based on MULCOXZ
Models)
RR Use of Use of Use of Plasma High Dietary High
multivitamm supplements vitamin C supplements wtamm E supplements level of vitamin E versus low? antioxidant index versus low$
RR = relative
risk; CI = confidence
among
95%
CI
0.69 0.80 0.43
0.48-0.99* 0.49- 1.32 0.19-0.99*
0.58
0.36-0.94*
0.71
0.43-1.18
interval.
* P < 0.05. t Categories mg/dl.
based on quintile
$ Categories based on quintde riboflavin, and carotene.
values: high,
>1.788
values of dietary
mg/dl;
vitamin
low, cl.078 C, vitamm
E,
Discussion This longitudinal study determined whether baseline antioxidant nutrients were related to subsequent changes in nuclear opacities at follow-up. Regular usersof multivitamins or vitamin E supplements at baseline had a decreasedrisk of nuclear opacification at follow-up. Furthermore, higher plasma levels of vitamin E at baseline also were associated with a lower risk of nuclear opacities (Table 3). These findings are similar to the multivitamin and vitamin E findings of our earlier case-control study,6,7although LSC used a different lens classification system and had a smaller sample size. Several studies have examined the relationship between multivitamin supplements, vitamin E, and the risk of cataract or lens opacities, but comparisonsamong studies are limited because of their different methodologies. Although our earlier case-control study6*’shared several design features with parallel case-control studies in Indiaz9and Italy,30 the results of the three studiesare difficult to compare,“’ especially those concerning intake of nutritional supplementsand vitamin E. The Indian study had different inclusion and exclusion criteria (e.g., it used its own lens classification system and required visual acuity loss for all cases). It also excluded users of nutritional supplements, whereas the Italian study had few supplement users, a reason that might explain the lack of a significant association. Furthermore, the dietary intake data differed among studies (e.g., the Indian study and the Italian study did not assessdietary vitamin E or total carotenoids), so that associationswith those dietary antioxidants could not be evaluated. Given these differences among the three studies, the lack of consistent results for these variables is not surprising. A more recent casecontrol study in Italy, which did assessdietary vitamin E, found that cataract extraction was associated with a lower vitamin E intake.r3 In a nutritional intervention trial in China, the use of multivitamins and mineral supplements was associated
834
5, May 1998
with a decreasedprevalence of nuclear opacities.” In the Physicians’ Health Study, men who used multivitamin supplementsexperienced a decreasedrisk of cataract extraction after 5 years of follow-up.” Associations between a decreased cataract risk and nutritional supplements, such as vitamin C,s,9vitamin E,8 or multivitamins,” were reported in other studies. In another report, higher plasma levels of vitamin E also were associatedwith a reduced frequency of nuclear opacities at a follow-up examination.14 The same study, however, found no associations of plasmabeta-carotene, ascorbic acid, and overall antioxidant statusto the risk of nuclear opacities.14An increased risk of cataract in persons with low levels of an antioxidant index, which included plasmalevels of ascorbic acid, vitamin E, and carotenoids, also was reported in a smaller study.15A Finnish study reported associationsof plasma levels of vitamin E (a-tocopherol) and cataract extraction after a median follow-up of 15 yearsI Another Finnish study reported the association between low plasma vitamin E and progression of early cortical cataract.” In contrast to the LSC, none of these studies, except the latter, were basedon multiple follow-up examinations with lens assessment.The longitudinal design of the LSC also addressedmajor possible sources of biases and misclassification. The nutritional data were collected at baseline using standardized methods, and the lens data were based on maskedphotographic classification of opacity type and severity. Our findings of a lower risk of nuclear opacities in regular users of multivitamins or vitamin E supplements and in personswith higher levels of plasma vitamin E thus provide evidence to support these relationships. Because LSC was an observational study, however, appropriate caution must be exercised in evaluating the results, especially becausethe data from all studies are not entirely consistent. Interpretation of our findings must consider alternative explanations, such as failure to adjust for confounding variables. Although the analyses included a number of
Table
4.
Dose-Response
Results
for Increases
m Nuclear
Opacities According to Use of Multivitamm Supplements, Vitamin E Supplements, and Plasma Level of Vitamin E RR (95% Multivitamin supplements Nonusers Users <5 yrs Users 25 yrs Dose-response Vitamin E supplements Nonusers Users <5 yrs Users 25 yrs Dose-response Plasma vitamm E Low qumtile Median 3 qumtiles High quintile Dose-response RR = relative
risk; CI = confidence
CI)
1 .oo 0.68 (0.42-1.11) 0.69 (0.42-1.11)
1 .oo 0.81 (0.32-2.06) 0.26 (0.06- 1.06)
1.0 0.72 (0.49-1.05) 0.58 (0.36-0.94)
Interval.
P
0.12 0.13 0.07
0.66 0.06 0.04
0.09 0.03 0.04
Leske et al * Antioxidant potential confounders, the possibility
Vitamins
of missing an un-
known confounder can never be dismissed. For example, the use of multivitamin supplements may be a marker for demographic, socioeconomic, or “lifestyle” related factors, which may be the true determinants of risk. This explanation seems unlikely, however, because persons who progressed had a consistently lower intake of vitamin supplements, regardless of age, gender, education, and occupation (Table 2). The association with antioxidant vitamin status also is supported by the biochemical data, because persons with higher plasma vitamin E had a lower risk of nuclear opacities. The MULCOX2 models used in the analyses were based on data from both eyes of a person. For completeness, we conducted separate analyses on data for left and right eyes in Cox regression analyses. The results from these analyses are consistent with those from the MULCOX2 models, with the expected loss of statistical power. A spurious association between nutritional factors and nuclear opacification would result if losses to follow-up were related, not only to nuclear opacification but also to supplement use or plasma levels of vitamin E.32 For example, our results could be caused by bias if losses to follow-up were more frequent among persons who used
supplements and also had increases in opacities than in others. To examine this possible source of biases, chisquare tests evaluated whether losses to follow-up
varied
among persons with different levels of individual nutrient variables. Additionally, log-linear models were used to detect possible three-way
nuclear opacification,
interactions
among increases in
losses to follow-up, and nutritional
factors. No statistically significant results were found in these analyses. Although these negative results do not necessarily imply the absence of biases, our evaluations suggest that losses to follow-up were less likely to lead to large biases when estimating the RRs, In conclusion, nuclear opacities are the most frequent
type reported by cataract prevalence studies in the United States and Europe.33-36 In contrast to cortical opacities, nuclear opacities often cause visual loss that requires cataract surgery. Given the high prevalence of nuclear opacities, the identification
of risk factors amenable to modifi-
cation or intervention has high priority. Our results suggest that use of nutritional supplements and of higher plasma vitamin E levels are related to a decreased risk of progression of nuclear opacities. These conclusions are
strengthened by the longitudinal design of the study and the careful lens assessment. Because of the observational nature of the study and the possibility of biases, our results are suggestive, but not conclusive, and should be interpreted cautiously. For example, observational studies suggested a protective effect of vitamin E and beta-carotene on lung cancer,37,38 whereas clinical trials found an opposite effect or no effect.39-4’ To assess the effect of vitamin supplementation on the development of nuclear opacities,
it is necessary to obtain conclusive evidence from ongoing clinical trials, such as from the National Eye Institute’s Age-Related Eye Disease Study. Currently, we emphasize that there is no firm evidence to support a recommendation to use nutritional supplements for cataract prevention.
and Nuclear
Opacities
References 1. Thylefors B. The role of international ophthalmology in blindness prevention. Am J Ophthalmol 1995; 119:229-30. 2. Bunce GE, Kinoshita J, Horwitz J. Nutritional factors in cataract. Annu Rev Nutr 1990; 10:233-54. 3. Lens and Cataract Panel. A report of the National Advisory Eye Council. In: National Advisory Eye Council (U.S.) Vision Research: A National Plan: 1994- 1998. A report of the National Advisory Eye Council. Washington, DC: US. Dept. of Health and Human Services, Public Health Service, National Institutes of Health, 1993 (NIH Publication; 93-3186). 4. Maisel H, ed. The Ocular Lens: Structure, Function, and Pathology. New York: Dekker, 1985. 5. Senile cataract and vitamin nutrition. Nutr Rev 1989; 47:326-8. 6. Leske MC, Chylack LT Jr, Wu SY. The Lens Opacities Case-Control Study. Risk factors for cataract. Arch Ophthalmol 1991; 109:244-51. 7. Leske MC, Wu SY, Hyman L, et al. Biochemical factors in the lens opacities. Case-control study. The Lens Opacities Case-Control Study Group. Arch Ophthalmol 1995; 113:1113-9. 8. Robertson JM, Donner AP, Trevithick JR. Vitamin E intake and risk of cataracts in humans. Ann NY Acad Sci 1989;570:372-82. 9. Hankinson SE, Stampfer MJ, Seddon JM, et al. Nutrient intake and cataract extraction in women: a prospective study. BMJ 1992;305:335-9. 10. Sperduto RD, Hu TS, Milton RC, et al. The Linxian cataract studies. Two nutrition intervention trials. Arch Ophthalmol 1993; 111:1246-53. 11. Seddon JM, Christen WG, Manson JE, et al. The use of vitamin supplements and the risk of cataract among US male physicians, Am J Public Health 1994;84:788-92. 12. Mares-Perlman JA, Klein BE, Klein R, Ritter L. Relation between lens opacities and vitamin and mineral supplement use. Ophthalmology 1994; 101:315-25. 13. Tavani A, Negri E, La Vecchia C. Food and nutrient intake and risk of cataract. Ann Epidemiol 1996;6:41-6. 14. Vitale S, West S, Hallfrisch J, et al. Plasma antioxidants and risk of cortical and nuclear cataract. Epidemiology 1993;4:195-203. 15. Knekt P, Heliovaara M, Rissanen A, et al. Serum antioxidant vitamins and risk of cataract. BMJ 1992;305: 13924. 6. Jacques PF, Chylack LT Jr, McGandy RB, Hartz SC. Antioxidant status in persons with and without senile cataract. Arch Ophthalmol 1988; 106:337-40. 7. Rouhiainen P, Rouhiainen H, Salonen JT. Association between low plasma vitamin E concentration and progression of early cortical lens opacities. Am J Epidemiol 1996; 144:496-500. 8. Leske MC, Chylack LT Jr, Wu SY, et al. Incidence and progression of nuclear opacities in the Longitudinal Study of Cataract. Ophthalmology 1996; 103:705-12. 9. Leske MC, Chylack LT, He Q, et al. Incidence and progression of cortical and posterior subcapsular opacities: the Longitudinal Study of Cataract. Ophthalmology. 1997; 104: 1985-93. 20. Leske MC, Chylack LT, He Q, et al. Risk factors for nuclear opalescence in a longitudinal study. Am J Epidemiol 1998;147:36-41.
83.5
Ophthalmology
Volume 105, Number
21. Chylack LT Jr, Leske MC, Sperduto R, et al. Lens opacities classification system. Arch Ophthalmol 1988; 106:330-4. 22. Leske MC, Chylack LT Jr, Sperduto R, et al. Evaluation of a Lens Opacities Classification System. Arch Ophthalmol 1988; 106:327-9. 23. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. Arch Ophthalmol 1993; Ill:83 l-6. 24. Lin DY. MULCOX2: a general computer program for the Cox regression analysis of multivariate failure time data. Comput Methods Programs Biomed 1993;40:279-93. 25. Wei LJ, Lin DY, Weissfeld L. Regression analysis of multivariate incomplete failure time data by modeling marginal distributions. JASA 1989;84:1065-71. 26. Cummings SR, Block G, McHenry K, Baron RB. Evaluation of two food frequency methods of measuring dietary calcium intake. Am J Epidemiol 1987; 126:796-802. 27. Block G, Dresser CM, Hartman AM, Carroll MD. Nutrient sources in the American diet: quantitative data from the NHANES II survey. I: vitamins and minerals. Am J Epidemiol 198.5; 122:13-26. 28. Block G, Dresser CM, Hartman AM, Carroll MD. Nutrient sources in the American diet: quantitative data from the NHANES II survey. II. Macronutrients and fat. Am J Epidemiol 1985; 122:27-40. 29. Mohan M, Sperduto RD, Angra SK, et al. India-US casecontrol study of age-related cataracts. India-US Case Control Study Group [published erratum appears in Arch Ophthalmol 1989; 107:1288]. Arch Ophthalmol 1989; 107:6706. 30. The Italian-American Cataract Study Group. Risk factors for age-related cortical, nuclear, and posterior subcapsular cataracts. Am J Epidemiol 1991; 133541-53. 31. Schoenfeld ER, Leske MC, Wu SY. Recent epidemiologic studies on nutrition and cataract in India, Italy and the United States. J Am Co11 Nutr 1993; 12:521-6.
836
5, May 1998
32. Kelsey JL, Thompson WD, Evans AS. Methods in Observational Epidemiology. New York Oxford University Press, 1986; 10 I (Monographs in epidemiology and biostatistics; v. IO). 33. Klein BE, Klein R, Linton KL. Prevalence of age-related lens opacities in a population. The Beaver Dam Eye Study. Ophthalmology 1992;99:546-52. 34. Sperduto R, Hiller R. The prevalence of nuclear, cortical, and posterior subcapsular lens opacities in a general population sample. Ophthalmology 1984;91:815-8. 35. Giuffre G, Giammanco R, Di Pace F, Ponte F. Casteldaccia eye study: prevalence of cataract in the adult and elderly population of a Mediterranean town. Int Ophthalmol 19945; l&363-71. 36. Hirvela H, Luukinen H, Laatikainen L. Prevalence and risk factors of lens opacities in the elderly in Finland. A population-based study. Ophthalmology 1995; 102: 108- 17. 37. Peto R, Doll R, Buckley JD, et al. Can dietary beta-carotene materially reduce human cancer rates? Nature 1981; 290:201-8. 38. Committee on Diet, Nutrition, and Cancer, Assembly of Life Sciences, National Research Council. Diet, nutrition, and cancer. Washington, DC: National Academy Press, 1982. 39. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (ATC). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994;330: 1029-35. 40. Hennekens CH, Buring JE, Manson JE, et al. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med 1996; 334: 1 145-9. 41. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996; 334: 1150-5.