This cross-sectional population study uses the National Health and Nutrition Examination Survey 2005-2008 data to explore the association between daily dietary polyunsaturated fatty acid consumption, including ω-3 fatty acids, and the prevalence of glaucoma in the US population.
Key Points
Question
Does an association exist between daily dietary intake of polyunsaturated fatty acids, especially ω-3 fatty acids, and the risk of glaucoma in the US population?
Findings
In a cross-sectional population study of 3865 participants, increased daily levels of eicosapentaenoic acid and docosahexaenoic acid intake were associated with significantly lower risks of glaucoma. Daily levels of total polyunsaturated fatty acid intake in the higher quartiles, however, were associated with significantly increased odds of having glaucoma.
Meaning
These findings indicate that longitudinal studies or randomized clinical trials assessing potential ω-3 fatty acid (eicosapentaenoic acid and docosahexaenoic acid) protection against glaucoma are warranted.
Abstract
Importance
Identifying whether an association exists between daily dietary polyunsaturated fatty acid (PUFA) consumption and the prevalence of glaucoma in the United States may provide modifiable dietary risk factors for the development of glaucoma.
Objective
To analyze the association between glaucoma and daily dietary intake of PUFAs, including ω-3 fatty acids, in the US population.
Design, Setting, and Participants
Data from 3865 participants in the National Health and Nutrition Examination Survey (NHANES) 2005-2008 database who were 40 years or older, had participated in the vision health and dietary intake questionnaires, and had available results from laboratory tests and eye examinations that included frequency-doubling technology visual field loss detection tests and optic disc photographs were included. Data collection was performed by NHANES from 2005 to 2006. Data for the present study were downloaded from their database May 1 to 30, 2017. Data analyses were performed from June 1 to October 1, 2017.
Exposures
Daily dietary intake of PUFAs, including ω-3 fatty acids.
Main Outcomes and Measures
Prevalence of glaucoma in the United States as defined using the Rotterdam criteria, which included a combination of optic cupping or asymmetry and visual field defect results.
Results
Of the 83 643 392 weighted survey participants included in this cross-sectional study, 43 660 327 (52.2%) were women and 3 076 410 (3.7%) met our criteria for having glaucoma. Compared with participants without glaucoma, those with glaucoma were older (mean [SE] age, 61.4 [0.8] vs 53.7 [0.4] years; P < .001). Increased levels of daily dietary intake of eicosapentaenoic acid (odds ratio [OR], 0.06; 95% CI, 0.00-0.73) and docosahexaenoic acid (OR, 0.06; 95% CI, 0.01-0.87) were associated with significantly lower odds of having glaucoma. However, participants with daily total dietary PUFA intake levels in the second (OR, 2.84; 95% CI, 1.39-5.79) and third (OR, 2.97; 95% CI, 1.08-8.15) quartiles showed significantly increased odds of meeting our criteria for a diagnosis of glaucoma.
Conclusions and Relevance
Increased daily dietary consumption levels of eicosapentaenoic acid and docosahexaenoic acid were associated with lower likelihood of glaucomatous optic neuropathy. However, consumption levels of total PUFAs in the higher quartiles were associated with a higher risk of glaucoma, which may have resulted from the relative intakes of ω-6 and ω-3 fatty acids and other confounding comorbidities. This study also hypothesizes that increasing the proportion of dietary ω-3 consumption levels while controlling overall daily PUFA intake may be protective against glaucoma. However, longitudinal studies or randomized clinical trials are needed to assess these hypotheses.
Introduction
Glaucoma is the leading cause of bilateral irreversible blindness worldwide. Known risk factors for glaucoma include family history, African American race/ethnicity, female sex, and older age. Currently, the only known modifiable risk factor for glaucoma is intraocular pressure, for which reduction serves as the target of medical and surgical therapies. Identification of other risk factors, especially those that are modifiable, may advance efforts to detect and treat glaucoma.
Dietary polyunsaturated fatty acids (PUFAs) are associated with serum lipoprotein levels and inflammatory status. The ω-3 and ω-6 fatty acids are subtypes of PUFAs and can only be obtained in the diet. The ω-3 fatty acids, with the first double bond on the third carbon position, include α-linolenic acid, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA) (Figure). Both subtypes of PUFAs are highly concentrated in the phospholipids of cellular membranes, especially in the brain, heart, retina, and testes, and are nutrients critical to human health and development.
Figure. Chemical Structures of ω-3 Fatty Acids α-Linolenic Acid, Eicosapentaenoic Acid, and Docosahexaenoic Acid.
A, α-Linolenic acid (18:3 ω-3) is a carboxylic acid with an 18-carbon (indicated as numbers below the structure) chain and 3 cis double bonds. B, Eicosapentaenoic acid (20:5 ω-3) is a carboxylic acid with a 20-carbon chain and 5 cis double bonds. C, Docosahexaenoic acid (22:6 ω-3) is a carboxylic acid with a 22-carbon chain and 6 cis double bonds. For all 3 structures, the first double bond is located at the third carbon (indicated by the numbers above the structures) from the ω end of the fatty acid chain.
Clinical trials have shown reduced levels of serum ω-3 fatty acid levels in patients with glaucoma compared with patients without glaucoma. Recent studies have shown a decreased incidence of glaucoma in participants with a high ω-6 to ω-3 intake ratio. The aim of our current study was to use the National Health and Nutrition Examination Survey (NHANES) database to assess whether an association existed between daily dietary intake of PUFAs and glaucoma in a sample of the US population.
Methods
Sample and Population
Data from the NHANES 2005-2008, a cross-sectional series of interviews and examinations of the US civilian, noninstitutionalized population, were used to study the association between daily dietary PUFA consumption and the risk of glaucoma. The Centers for Disease Control and Prevention administers NHANES to provide US health statistics of 10 000 persons per iteration. It uses a standardized sampling and weighting scheme to produce optimal estimates of disease prevalence in the US population. The survey includes interviews, physical examinations, and laboratory studies. Humphrey matrix frequency-doubling technology (FDT) perimetry testing (N30-5 FDT; Carl Zeiss Meditec) and optic disc photographs were also obtained from 2005 through 2008 as part of the eye examination. The dietary interview component of NHANES, What We Eat in America, was conducted as a partnership between the US Department of Agriculture (USDA) and the US Department of Health and Human Services. Data accumulation was performed by the National Center for Health Statistics with approval from their ethics review board. Because NHANES data are publicly available without participant identifiable data included, the present study received exemption from review by the institutional review board of the UCLA (University of California, Los Angeles). Informed consent was obtained by NHANES for all participants.
The present analysis included participants from the NHANES 2005-2008 who were 40 years or older, had completed the interview (including both vision health and dietary intake questionnaires), laboratory tests, and the eye examination portions of the study, and had available FDT visual field and optic disc photography results. Exclusion criteria included the following: (1) missing FDT visual field results or optic disc photographs or (2) documented alternative explanation for the cup-disc ratio (CDR) findings, such as dysplastic disc or marked anisometropia (cylindrical or spherical equivalents ≥2 diopters between adelphous eyes), or for a visual field defect from macular degeneration, retinal vascular disease, or cerebrovascular events. Data collection was performed by NHANES from 2005 to 2006. Data for the present study were downloaded from the NHANES database May 1 to 30, 2017. Data analyses were performed from June 1 to October 1, 2017.
Outcome Measures
The primary outcome was the prevalence of glaucoma as defined by the Rotterdam criteria, which uses a combination of optic nerve appearance and visual field defects. For NHANES, optic nerve appearance was assessed using optic nerve photographs, and glaucomatous visual field defects were assessed with FDT. A clinical diagnosis of glaucoma was defined as having 2 or more abnormal points (1 of which must be consistent between 2 consecutive tests) in at least 1 eye on the N30-5 FDT of 2 tests in the same eye along with a CDR in 1 eye or CDR asymmetry between eyes of at least 97.5% of the mean NHANES population. The Rotterdam criteria were chosen because of their ability to provide an objective clinical definition of glaucoma with available NHANES data, as reported in a recent study.
For NHANES, FDT examinations were conducted in the dark by trained technicians. Nineteen visual field locations were tested for each eye, and each location was tested until the participant responded. An eye was considered to have an abnormal visual field when at least 2 locations in both the first and second tests were below a 1% threshold level and at least 1 failed location was the same on both tests (2-2-1 algorithm). During each test, 3 false-positive tests and 3 blind spot tests were performed at random times as reliability checks. Participants were then classified as having normal, positive, insufficient, or unreliable FDT results. Those with insufficient or unreliable results were excluded from the study population. The CDR was determined with fundus photographs. The examination was conducted using two 45° nonmydriatic digital images of the retina captured by trained technicians. Trained graders at the University of Wisconsin assessed vertical CDR in each eye with digital images as they initially appeared on the back of the digital camera. Each image was reviewed by at least 3 of the 9 trained graders. If the CDR scores for at least 2 of 3 graders were within 0.1, the median value was assigned to the image. If there was a disagreement of at least 0.2 between any 2 graders, the image was re-reviewed in the presence of all graders and consensus was reached.
The primary exposure in this study was daily dietary PUFA consumption. During the dietary interview, participants underwent two 24-hour dietary recall interviews. Detailed information about the types and amounts of individual foods as well as amounts of energy (in calories) and nutritional components were recorded. The USDA Food and Nutrient Database for Dietary Studies, 3.0 (FNDDS 3.0) was used for processing the 2005 through 2008 intake values. The underlying nutrient values for FNDDS 3.0 were based on values in the USDA National Nutrient Database for Standard Reference, release 20, produced by the USDA Nutrient Data Laboratory (USDA Agricultural Research Service). Based on each participant’s reported daily food consumption, an estimate of the daily aggregates of food energy and 63 nutrients and food components, including saturated and monounsaturated fatty acids and PUFAs, were recorded for each interview day. The mean values between each of those from the 2 interview days were used in this study.
Statistical Analysis
Sampling for NHANES occurs in 4 stages starting with the primary sampling units and followed by segments within the primary sampling units (usually city blocks), households, and individuals. Based on this multistage probability sampling design, data are weighted and adjusted for nonresponse to produce weighted estimates meant to be representative of the US population. All data analyses conducted in the present study were based on weighted estimates with sample weights provided by NHANES. Data are reported as weighted and nationally representative estimates of means and frequencies unless otherwise stated.
Descriptive statistics were used to assess the baseline characteristics of the study population. Age, serum low-density lipoprotein cholesterol (LDL-C), and serum triglyceride levels were analyzed as continuous variables; sex, race/ethnicity, household income, and educational level as categorical variables. Dietary PUFA consumption was analyzed as a continuous variable as well as in quartiles. Dietary PUFA subtype consumption was analyzed as a continuous variable. Glaucoma was assessed as a binary outcome based on the Rotterdam criteria. The distribution or proportion of these variables was compared between participants with or without glaucoma (as defined herein) with design-adjusted Rao-Scott (Pearson-type) χ2 and Wald tests for categorical and continuous variables, respectively.
Logistic regression modeling was used to examine the association between daily dietary PUFA consumption and glaucoma with confounders, including fatty acid subtypes, age, sex, and race/ethnicity as well as income, educational, serum LDL-C, and triglyceride levels. All analyses were conducted with SAS, version 9.3 (SAS Institute Inc), and 2-sided P values <.05 were considered statistically significant.
Results
The NHANES 2005-2008 data yielded a total of 83 643 392 weighted and 3865 unweighted participants who were 40 years or older and who had participated in the interview, laboratory, and examination portions of the study. Of these, 43 660 327 (52.2%) were women, and 3 076 410 (3.7%) had glaucoma as defined herein. Compared with the group of participants without glaucoma, the group with glaucoma was composed of older individuals (mean [SE] age, 61.4 [0.8] vs 53.7 [0.4] years; P < .001) and had a higher percentage of non-Hispanic black individuals (non-Hispanic black individuals with glaucoma, 708 304 [23.0%] vs those without glaucoma, 7 221 260 [5.2%]; P < .001) (Table 1). The numbers of participants with more than a high school educational level (1 568 227 [51.0%] vs 46 273 522 [42.6%]; P = .17) and those with annual incomes greater than $75 000 (1 060 445 [37.2%] vs 3 029 423 [39.5%]; P = .27) were not significantly different between the 2 cohorts.
Table 1. Weighted and Unweighted Demographic Data in Participants With and Without Glaucoma.
| Characteristic | With Glaucoma (n = 3 076 410) | Without Glaucoma (n = 80 566 982) | P Value | ||
|---|---|---|---|---|---|
| No. (Weighted Frequency) | % of Total | No. (Weighted Frequency) | % of Total | ||
| Female | 91 (1 433 062) | 45.6 | 1856 (42 227 265) | 5.7 | .14 |
| Race/ethnicity | |||||
| Mexican | 37 (238 125) | 7.7 | 607 (4 590 974) | 3.4 | <.001 |
| Other Hispanic | 13 (101 132) | 3.3 | 277 (2 769 988) | 76.2 | |
| Non-Hispanic white | 70 (1 631 143) | 53.0 | 1879 (61 352 017) | 9.5 | |
| Non-Hispanic black | 77 (708 304) | 23.0 | 755 (7 221 260) | 5.2 | |
| Other | 15 (397 705) | 12.9 | 135 (4 192 744) | 42.6 | |
| Educational level | |||||
| High school or less | 121 (1 508 182) | 49.0 | 1892 (34 265 856) | 57.4 | .17 |
| More than high school | 91 (1 568 227) | 51.0 | 1760 (46 273 522) | 42.6 | |
| Annual income level, $ | |||||
| <50 000 | 94 (921 407) | 32.3 | 1363 (20 666 988) | 26.9 | .27 |
| 50000$75 000 | 50 (871 598) | 30.6 | 1078 (25 801 977) | 33.6 | |
| >75 000 | 49 (1 060 445) | 37.2 | 987 (3 0291 423) | 39.5 | |
As given in Table 2, compared with participants without glaucoma, those with glaucoma had significantly lower consumption levels of saturated fatty acids (mean [SE], 23.60 [0.98] vs 26.94 [0.39] g; P = .003) and monounsaturated fatty acids (mean [SE], 27.52 [1.24] vs 30.05 [0.41] g; P = .07). By contrast, no statistically significant difference in the mean daily dietary consumption of PUFAs was found between participants with and without glaucoma. The daily consumption of the different PUFA subtypes was also compared between the cohorts. Participants without glaucoma had a higher median (SE) intake of EPA (8.54 [1.2] vs 8.50 [0.55] mg; P = .002), DHA (34.4 [6.3] vs 33.8 [1.3] mg; P = .04), and docosapentaenoic acid (11.0 [1.4] vs 9.2 [0.5] mg; P = .02). Serum (mean [SE]) levels of total cholesterol (198.25 [3.55] vs 205.46 [0.82] mg/dL; P = .05), high-density lipoprotein cholesterol (54.40 [1.74] vs 53.82 [0.41] mg/dL; P = .73), LDL-C (113.20 [4.10] vs 121.10 [1.07] mg/dL; P = .06), and triglycerides (146.30 [10.85] vs 148.52 [3.72] mg/dL; P = .87) were not significantly different between participants with and without glaucoma. (To convert serum total cholesterol, high-density lipoprotein cholesterol, and LDL-C levels to millimoles per liter, multiply by 0.0259; serum triglyceride level to millimoles per liter, by 0.0113).
Table 2. Daily Dietary Fatty Acid Consumption and Clinical Profile in Participants With and Without Glaucoma.
| Variable | Median (SE) | P Value | |
|---|---|---|---|
| With Glaucoma | Without Glaucoma | ||
| Saturated fatty acid, g | 23.60 (0.98) | 26.94 (0.39) | .003 |
| Monounsaturated fatty acid, g | 27.52 (1.24) | 30.05 (0.41) | .07 |
| Polyunsaturated fatty acid, g | 17.22 (1.02) | 17.67 (0.22) | .67 |
| Total cholesterol, mg/dL | 198.25 (3.55) | 205.46 (0.82) | .051 |
| High-density lipoprotein cholesterol, mg/dL | 54.40 (1.74) | 53.82 (0.41) | .73 |
| Low-density lipoprotein cholesterol, mg/dL | 113.20 (4.10) | 121.10 (1.07) | .06 |
| Triglycerides, mg/dL | 146.30 (10.85) | 148.52 (3.72) | .87 |
| Octadecadienoic acid (PUFA 18:2), ga | 13.21 (0.77) | 13.85 (0.21) | .76 |
| Octadecatrienoic acid (PUFA 18:3), g | 1.25 (0.058) | 1.34 (0.016) | .26 |
| Octadecatetraenoic acid (PUFA 18:4), mg | 0 (0.33) | 0.26 (0.040) | .01 |
| Eicosatetraenoic acid (PUFA 20:4), mg | 11.0 (6.5) | 12.0 (2.3) | .54 |
| Eicosapentaenoic acid (PUFA 20:5), mg | 8.54 (1.2) | 8.50 (0.55) | .002 |
| Docosapentaenoic acid (PUFA 22:5), mg | 11.0 (1.4) | 9.2 (0.5) | .02 |
| Docosahexaenoic acid (PUFA 22:6), mg | 34.4 (6.3) | 33.8 (1.3) | .04 |
Abbreviation: PUFA, polyunsaturated fatty acid.
SI conversion factors: To convert total, high-density lipoprotein, and low-density lipoprotein cholesterol levels to millimoles per liter, multiply values by 0.0259; serum triglyceride level to millimoles per liter, by 0.0113.
For an explanation of PUFA entries, see the Figure and its legend.
The association between participant average daily dietary intake of PUFAs and glaucoma was analyzed with a logistic regression model. Age, sex, race/ethnicity (Mexican, Hispanic, non-Hispanic white, non-Hispanic black, and other individuals), socioeconomic status in the forms of income and educational levels, serum LDL-C and triglyceride levels, and mean daily dietary intake of saturated and monounsaturated fatty acids were all potential confounders and were included in the models. We found that PUFA as a continuous variable was not statistically associated with increased or decreased odds of having glaucoma (odds ratio [OR], 1.03; 95% CI, 0.97-1.09).
Analysis of the associations between mean daily dietary intake of the different subtypes of PUFAs was conducted with logistic regression models using the same set of confounders described earlier (Table 3). Daily dietary intake of EPA was associated with lower odds of having glaucoma (OR, 0.06; 95% CI, 0.00-0.73). Daily dietary intake of DHA was also associated with lower odds of having glaucoma (OR, 0.06; 95% CI, 0.01-0.87). Dietary intake of the other PUFA subtypes assessed in NHANES, including octadecadienoic acid (also known as PUFA 18:2; ie, an 18-carbon chain with 2 cis double bonds), octadecatrienoic acid (PUFA 18:3), octadecatetraenoic acid (PUFA 18:4), and eicosatetraenoic acid (PUFA 20:4), were not associated with either increased or decreased odds of having glaucoma (all P > .05). When the association between daily dietary PUFA intake and glaucoma was analyzed with the PUFA level in quartiles, after controlling for the same confounders, participants with a PUFA intake in the second (OR, 2.84; 95% CI, 1.39-5.79) and third (OR, 2.97; 95% CI, 1.08-8.15) quartiles showed increased odds of having glaucoma (Table 4).
Table 3. Logistic Regression Model Analyzing Association of Glaucoma With Daily Dietary Consumption of Eicosapentaenoic Acid and Docosahexaenoic Acida.
| Variable | Odds Ratio (95% Wald CI) |
|---|---|
| Eicosapentaenoic Acid (PUFA 20:5) | |
| Age | 1.08 (1.06-1.11)b |
| Sex | 1.19 (0.70-2.03) |
| Race/ethnicityc | |
| 2 vs 1 | 0.48 (0.10-2.32) |
| 3 vs 1 | 0.26 (0.10-0.66)b |
| 4 vs 1 | 1.39 (0.49-3.92) |
| 5 vs 1 | 1.77 (0.54-5.77) |
| Educational level 2 vs 1d | 0.99 (0.41-2.38) |
| Annual income levele | |
| 2 vs 1 | 0.63 (0.25-1.58) |
| 3 vs 1 | 1.14 (0.43-3.03) |
| Eicosapentaenoic acid intake | 0.06 (0.00-0.73)b |
| Triglycerides | 1.00 (0.99-1.01) |
| Low-density lipoprotein cholesterol | 1.00 (0.99-1.00) |
| Docosahexaenoic Acid (PUFA 22:6) | |
| Age | 1.08 (1.06-1.11)b |
| Sex | 1.19 (0.71-2.00) |
| Race/ethnicityc | |
| 2 vs 1 | 0.46 (0.10-2.17) |
| 3 vs 1 | 0.26 (0.10-0.66)b |
| 4 vs 1 | 1.32 (0.46-3.80) |
| 5 vs 1 | 1.86 (0.58-5.97) |
| Educational level 2 vs 1d | 0.99 (0.41-2.38) |
| Annual income levele | |
| 2 vs 1 | 0.62 (0.24-1.56) |
| 3 vs 1 | 1.13 (0.43-2.97) |
| Docosahexaenoic acid intake | 0.06 (0.01-0.87)a |
| Triglycerides | 1.00 (0.99-1.00) |
| Low-density lipoprotein cholesterol | 1.00 (0.99-1.00) |
Abbreviation: PUFA, polyunsaturated fatty acid.
For an explanation of PUFA entries, see the Figure and its legend.
Indicates statistically significant results.
Race/ethnicity 1 indicates Mexican; 2, Hispanic; 3, non-Hispanic white; 4, non-Hispanic black; and 5, other.
Educational level 1 indicates high school or lower; 2, more than high school.
Annual income level 1 indicates less than $50 000; 2, $50 000 to $75 000; and 3, more than $75 000.
Table 4. Logistic Regression Model Analyzing Association of Glaucoma With Daily Dietary Consumption of Polyunsaturated Fatty Acid by Quartiles.
| Variable | Odds Ratio (95% Wald CI) |
|---|---|
| Age | 1.09 (1.06-1.12)a |
| Sex | 1.35 (0.74-2.47) |
| Race/ethnicityb | |
| 2 vs 1 | 0.51 (0.11-2.30) |
| 3 vs 1 | 0.23 (0.09-0.59)a |
| 4 vs 1 | 1.23 (0.43-3.55) |
| 5 vs 1 | 1.74 (0.54-5.56) |
| Educational level 2 vs 1c | 0.98 (0.41-2.36) |
| Annual income leveld | |
| 2 vs 1 | 0.56 (0.22-1.42) |
| 3 vs 1 | 1.07 (0.42-2.69) |
| Polyunsaturated fatty acid | |
| Second quartile | 2.84 (1.39-5.79) |
| Third quartile | 2.97 (1.08-8.15) |
| Fourth quartile | 2.03 (0.74-5.52) |
| Triglycerides | 1.00 (0.99-1.00) |
| Low-density lipoprotein cholesterol | 1.00 (0.99-1.00) |
Indicates statistically significant results.
Race/ethnicity 1 indicates Mexican; 2, Hispanic; 3, non-Hispanic white; 4, non-Hispanic black; and 5, other.
Educational level 1 indicates high school or lower; 2, more than high school.
Annual income level 1 indicates less than $50 000; 2, $50 000 to $75 000; and 3, more than $75 000.
Discussion
Our results indicated that lower levels of EPA and DHA intake were associated with glaucoma, as defined herein. A similar association was found previously by Ren et al, who identified decreased EPA and DHA intake in participants with glaucoma compared with that in participants without glaucoma. They hypothesized that EPA and DHA modulate systemic microcirculation and ocular blood flow. Nguyen et al found that increasing dietary ω-3 intake in animal models reduced intraocular pressure with age, which was hypothesized but not yet proved to be a result of increased aqueous outflow and reduced ocular rigidity from production of specific types of prostaglandins and docosanoids. Vascular insufficiency in the forms of decreased blood flow and increased blood viscosity has been recognized as a potential risk factor in the pathogenesis of glaucoma. It is thought that ω-3 fatty acids may reduce blood viscosity by changing red blood cell membrane composition and plasma protein composition. Acar et al reported reduced DHA levels in red blood cells of patients with primary open-angle glaucoma, with a decline that started before the clinical signs of glaucoma. Those authors found a correlation between DHA deficit and visual field loss. Moreover, experimental studies have demonstrated functional and structural anomalies of the retina and optic nerve when DHA is lacking. Remé et al demonstrated that these fatty acids in ganglion cells are protective against toxic damage, although the exact mechanism for this action remains unclear. Cellini et al demonstrated improvement in blue-on-yellow perimetry results after a 3-month treatment of systemic ω-3 PUFA in patients with ocular hypertension.
Polyunsaturated fatty acids are critical nutrients to human health and development. Eicosapentaenoic acid incorporated into the phospholipids of cell membranes helps ensure an environment ideal for membrane protein function, membrane fluidity, and lipid raft formation. It also regulates transcription and cellular signaling and acts directly on inflammatory cells. Docosahexaenoic acid is important for brain development and cognition as well as for the prevention of neurodegenerative diseases. It is an important cellular membrane component that can prevent retinal cell structural degradation, decrease glial cell activation as a result of increased intraocular pressure, and protect retinal photoreceptors from apoptosis induced by oxidative stress. Polyunsaturated fatty acid modifications of the extracellular matrix of the trabecular meshwork may also modify the risk of glaucoma through oxidative states.
The odds of having glaucoma were nearly 3 times as high in participants whose daily dietary total PUFA consumption level was in the second and third quartiles compared with those whose intake was in the first quartile. Participants who had the highest quartile of daily PUFA intake also had odds of having glaucoma approximately twice as high as those whose intake was in the first quartile, although this result was not statistically significant. This finding, although apparently contradicting our finding that increased consumption of EPA and DHA was associated with glaucoma, suggests that even though certain subtypes of PUFAs may reduce the risk of glaucoma, excess PUFA consumption may not. Alternatively, the apparently discrepant finding may be secondarily associated with the ratio between ω-3 and ω-6 fatty acid intake. A previous study with a cohort of 17 000 patients showed an increased risk of glaucoma incidence in patients whose ω-3 to ω-6 intake is in the highest quintile compared with that for patients in the lowest quintile over an 8.2-year follow-up. Because not all ω-3 or ω-6 subtypes were identified in NHANES, the ratio between the daily intake of ω-3 and ω-6 fatty acids could not be calculated in our study. Future prospective studies investigating ω-3 and ω-6 PUFA subtypes are necessary to examine this hypothesis.
Limitations
Our study has some limitations due to the observational and cross-sectional nature of an NHANES-database study. Cross-sectional data prevented the investigation of temporal associations between dietary ω-3 fatty acid intake and the risk of glaucoma, which may hinder the interpretation of our findings because dietary intake may be either a risk factor for glaucoma or a consequence of lifestyle changes brought on by having glaucoma. Cross-sectional data also prevented the examination of an association of long-term dietary intake with glaucoma or of a potential causal association between them. Dietary intake was self-reported using a food frequency questionnaire, subjecting those data to recall bias and misclassification bias although these biases would not be expected to differ between participants with and without glaucoma. Our results may also be confounded by unmeasured factors, such as participant lifestyles, that may be related to the dietary intake.
Although the Rotterdam criteria we used to define glaucoma has been validated as a useful tool to estimate the prevalence of glaucoma in population studies, it also has limitations. Specifically, the FDT and optic nerve diagnostic criteria that were used to define glaucoma in this study are subject to misclassification, with potential 9% false-positive and 6% false-negative rates. Although FDT for visual field loss detection has good sensitivity, it is not very specific, especially in early-stage glaucoma, which increases the rate of false-positives. Similarly, CDR estimates do not discriminate well between glaucomatous and nonglaucomatous eyes. In addition, accurate optic nerve head assessment can be limited by grader skill and by optic nerve characteristics, such as peripapillary atrophy, myopic degeneration, and small, large, or tilted optic discs. Furthermore, it is generally accepted that there is discordance between structural optic nerve changes and functional deficits detected by perimtery. Thus, by the time a defect is noted on perimetry testing, up to half of the ganglion cells may already be lost. Besides these diagnostic challenges, a thorough ocular examination to differentiate subcategories of glaucoma was not conducted for inclusion in NHANES data, and the association between dietary fatty acid intake and glaucoma subtypes may differ. The NHANES 2007-2008 survey used the FNDDS 3.0 to calculate the nutrient level from typical daily portions of dietary intakes. This database has a food description component, a food portion and weight component, and a nutrient component in which food energy and 63 typical nutrients are recorded but which are far from inclusive of all possibilities. The reproducibility of using FNDDS 3.0 to calculate nutrient level is unknown, which may introduce some bias into the final calculation of the daily nutrient intake. Moreover, supplementation levels of various fatty acids, including PUFA, were not calculated in the dietary questionnaire. An analysis with combined daily consumption of both dietary and supplementary PUFA and its association with the prevalence of glaucoma was not available. Given the increasing numbers of people who consume dietary supplements of ω-3 or fish oil, this may be a confounding factor that was not controlled for in our logistic regression analysis.
Conclusions
This study found that increased levels of daily dietary consumption of the EPA and DHA ω-3 PUFAs were associated with lower likelihood of a diagnosis of glaucomatous optic neuropathy in the NHANES 2005-2008 population. However, consumption of total PUFAs (including different types of ω-6 and ω-3) in the higher quartiles was associated with a higher risk of glaucoma. Because a causal association between daily ω-3 fatty acid (or PUFA intake in general) and the risk of glaucoma could not be drawn in this observational study, additional longitudinal studies or randomized clinical trials are warranted to extend these findings. The role of the dietary consumption of ω-3 relative to ω-6 fatty acid levels also needs further elucidation.
References
- 1.Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol. 2006;90(3):262-267. doi: 10.1136/bjo.2005.081224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Calder PC. Omega-3 fatty acids and inflammatory processes. Nutrients. 2010;2(3):355-374. doi: 10.3390/nu2030355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nicolosi RJ. Dietary fat saturation effects on low-density-lipoprotein concentrations and metabolism in various animal models. Am J Clin Nutr. 1997;65(5):1617S-1627S. [DOI] [PubMed] [Google Scholar]
- 4.Ooi EM, Watts GF, Ng TW, Barrett PH. Effect of dietary fatty acids on human lipoprotein metabolism: a comprehensive update. Nutrients. 2015;7(6):4416-4425. doi: 10.3390/nu7064416 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Huang W-B, Fan Q, Zhang X-L. Cod liver oil: a potential protective supplement for human glaucoma. Int J Ophthalmol. 2011;4(6):648-651. doi: 10.3980/j.issn.2222-3959.2011.06.15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ren H, Magulike N, Ghebremeskel K, Crawford M. Primary open-angle glaucoma patients have reduced levels of blood docosahexaenoic and eicosapentaenoic acids. Prostaglandins Leukot Essent Fatty Acids. 2006;74(3):157-163. doi: 10.1016/j.plefa.2005.11.007 [DOI] [PubMed] [Google Scholar]
- 7.Kang JH, Pasquale LR, Willett WC, et al. . Dietary fat consumption and primary open-angle glaucoma. Am J Clin Nutr. 2004;79(5):755-764. [DOI] [PubMed] [Google Scholar]
- 8.Centers for Disease Control and Prevention, National Center for Health Statistics About the National Health and Nutrition Examination Survey. https://www.cdc.gov/nchs/nhanes/about_nhanes.htm. Accessed November 14, 2017.
- 9.Centers for Disease Control and Prevention, National Center for Health Statistics NCHS research ethics review board (ERB) approval. https://www.cdc.gov/nchs/nhanes/irba98.htm. Accessed November 14, 2017.
- 10.Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol. 2002;86(2):238-242. doi: 10.1136/bjo.86.2.238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wolfs RCW, Borger PH, Ramrattan RS, et al. . Changing views on open-angle glaucoma: definitions and prevalences–the Rotterdam Study. Invest Ophthalmol Vis Sci. 2000;41(11):3309-3321. [PubMed] [Google Scholar]
- 12.Quigley HA. Identification of glaucoma-related visual field abnormality with the screening protocol of frequency doubling technology. Am J Ophthalmol. 1998;125(6):819-829. doi: 10.1016/S0002-9394(98)00046-4 [DOI] [PubMed] [Google Scholar]
- 13.Shaikh Y, Yu F, Coleman AL. Burden of undetected and untreated glaucoma in the United States. Am J Ophthalmol. 2014;158(6):1121-1129.e1. doi: 10.1016/j.ajo.2014.08.023 [DOI] [PubMed] [Google Scholar]
- 14.Centers for Disease Control and Prevention National Health and Nutrition Examination Survey: 2005-2006 data documentation, codebook, and frequencies: ophthalmology-retinal imaging (OPXRET_D). https://wwwn.cdc.gov/nchs/nhanes/2005-2006/OPXRET_D.htm. Published December 2008. Last Revised June 2015. Accessed May 6, 2017.
- 15.Raper N, Perloff B, Ingwersen L, Steinfeldt L, Anand J. An overview of USDA’s Dietary Intake Data System. J Food Compost Anal. 2004;17(3-4):545-555. doi: 10.1016/j.jfca.2004.02.013 [DOI] [Google Scholar]
- 16.Centers for Disease Control and Prevention National Health and Nutrition Examination Survey: 2005-2006 data documentation, codebook, and frequencies. https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/DR1IFF_D.htm. Accessed November 14, 2017.
- 17.US Department of Agriculture The USDA food and nutrient database for dietary studies, 3.0–documentation and user guide. https://www.ars.usda.gov/ARSUserFiles/80400530/pdf/fndds/fndds3_doc.pdf. Accessed November 14, 2017.
- 18.Centers for Disease Control and Prevention National Health and Nutrition Examination Survey: 2005-2006 data documentation, codebook, and frequencies: demographic variables and sample weights. https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/DEMO_D.htm. Accessed November 14, 2017.
- 19.Centers for Disease Control and Prevention National Health and Nutrition Examination Survey: 2007-2008 data documentation, codebook, and frequencies: demographic variables and sample weights. https://wwwn.cdc.gov/Nchs/Nhanes/2007-2008/DEMO_E.htm. Accessed November 14, 2017.
- 20.Nguyen CTO, Bui BV, Sinclair AJ, Vingrys AJ. Dietary omega 3 fatty acids decrease intraocular pressure with age by increasing aqueous outflow. Invest Ophthalmol Vis Sci. 2007;48(2):756-762. doi: 10.1167/iovs.06-0585 [DOI] [PubMed] [Google Scholar]
- 21.Lands WE. Biochemistry and physiology of n-3 fatty acids. FASEB J. 1992;6(8):2530-2536. [DOI] [PubMed] [Google Scholar]
- 22.Yanagi M, Kawasaki R, Wang JJ, Wong TY, Crowston J, Kiuchi Y. Vascular risk factors in glaucoma: a review. Clin Exp Ophthalmol. 2011;39(3):252-258. doi: 10.1111/j.1442-9071.2010.02455.x [DOI] [PubMed] [Google Scholar]
- 23.Mueller BA, Talbert RL. Biological mechanisms and cardiovascular effects of omega-3 fatty acids. Clin Pharm. 1988;7(11):795-807. [PubMed] [Google Scholar]
- 24.Ernst E. Effects of n-3 fatty acids on blood rheology. J Intern Med Suppl. 1989;731:129-132. [DOI] [PubMed] [Google Scholar]
- 25.Acar N, Berdeaux O, Grégoire S, et al. . Lipid composition of the human eye: are red blood cells a good mirror of retinal and optic nerve fatty acids? PLoS One. 2012;7(4):e35102. doi: 10.1371/journal.pone.0035102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Remé CE, Grimm C, Hafezi F, Marti A, Wenzel A. Apoptotic cell death in retinal degenerations. Prog Retin Eye Res. 1998;17(4):443-464. doi: 10.1016/S1350-9462(98)00009-3 [DOI] [PubMed] [Google Scholar]
- 27.Cellini M, Rossi A, Moretti M. The use of polyunsaturated fatty acids in ocular hypertension: a study with blue-on-yellow perimetry. Acta Ophthalmol Scand Suppl. 1999;77(S229):54-55. doi: 10.1111/j.1600-0420.1999.tb01152.x [DOI] [Google Scholar]
- 28.Desmettre T, Rouland J-F. Hypothesis on the role of nutritional factors in ocular hypertension and glaucoma [in French]. J Fr Ophtalmol. 2005;28(3):312-316. [DOI] [PubMed] [Google Scholar]
- 29.Pérez de Arcelus M, Toledo E, Martínez-González MA, Sayón-Orea C, Gea A, Moreno-Montañés J. Omega 3:6 ratio intake and incidence of glaucoma: the SUN cohort. Clin Nutr. 2014;33(6):1041-1045. doi: 10.1016/j.clnu.2013.11.005 [DOI] [PubMed] [Google Scholar]
- 30.Caprioli J. Discrimination between normal and glaucomatous eyes. Invest Ophthalmol Vis Sci. 1992;33(1):153-159. [PubMed] [Google Scholar]
- 31.Quigley HA, Katz J, Derick RJ, Gilbert D, Sommer A. An evaluation of optic disc and nerve fiber layer examinations in monitoring progression of early glaucoma damage. Ophthalmology. 1992;99(1):19-28. doi: 10.1016/S0161-6420(92)32018-4 [DOI] [PubMed] [Google Scholar]
- 32.Heijl A, Leske MC, Hyman L, Yang Z, Bengtsson B; EMGT Group . Intraocular pressure reduction with a fixed treatment protocol in the Early Manifest Glaucoma Trial. Acta Ophthalmol. 2011;89(8):749-754. doi: 10.1111/j.1755-3768.2009.01852.x [DOI] [PubMed] [Google Scholar]
- 33.Harris WS. Are n-3 fatty acids still cardioprotective? Curr Opin Clin Nutr Metab Care. 2013;16(2):141-149. doi: 10.1097/MCO.0b013e32835bf380 [DOI] [PubMed] [Google Scholar]
- 34.Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, Zack DJ. Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. Invest Ophthalmol Vis Sci. 1995;36(5):774-786. [PubMed] [Google Scholar]
- 35.Sommer A, Katz J, Quigley HA, et al. . Clinically detectable nerve fiber atrophy precedes the onset of glaucomatous field loss. Arch Ophthalmol. 1991;109(1):77-83. [DOI] [PubMed] [Google Scholar]
- 36.Tuulonen A, Lehtola J, Airaksinen PJ. Nerve fiber layer defects with normal visual fields: do normal optic disc and normal visual field indicate absence of glaucomatous abnormality? Ophthalmology. 1993;100(5):587-597. doi: 10.1016/S0161-6420(93)31598-8 [DOI] [PubMed] [Google Scholar]
- 37.Caprioli J, Prum B, Zeyen T. Comparison of methods to evaluate the optic nerve head and nerve fiber layer for glaucomatous change. Am J Ophthalmol. 1996;121(6):659-667. doi: 10.1016/S0002-9394(14)70632-4 [DOI] [PubMed] [Google Scholar]