Plasma fatty acids and primary open-angle glaucoma in the elderly: the Montrachet population-based study

Background To compare plasma fatty acids (FAs) between participants with primary open-angle glaucoma (POAG) and participants without neuropathy in an elderly population and to investigate specific FAs pattern in POAG. Methods We conducted a population-based study in participants older than 75 years. Participants underwent a comprehensive eye examination with optic nerve photographs, visual field test and optic nerve OCT with RNFL thickness measurement. Glaucomatous status was defined according to the International Society for Epidemiologic and Geographical Ophthalmology classification. Lipids were extracted from plasma and FAs methylesters prepared and analyzed by gas chromatography-mass spectrometry. Results Among the 1153 participants of the Montrachet study 810 were retained for analysis and 68 had POAG. The mean age was 82.11 ± 3.67. In multivariable analysis FAs levels were not different between POAG participants and controls (P = 0.078). A FAs pattern characterized by high negative weight of gamma-linoleic acid, eicosapentaenoic acid polyunsaturated FAs (PUFAs), Cis-7 hexadecenoic acid monounsaturated FAs (MUFAs) and high positive weight of eicosadienoic acid, docosatetraenoic acid, docosapentaenoic n-6, alpha linoleic acid PUFAs, eicosenoic acid MUFAs, margaric acid and behenic acid saturated FAs was positively associated with POAG. After adjustment for major confounders, individuals in the upper tertile of FAs pattern scores compared with those in the lower tertile were more likely to present POAG (OR = 3.09 [95% CI 1.29–7.40] P = 0.013). Conclusions We found no significant difference regarding isolated plasma FAs between participants with POAG and participants without neuropathy in elderly but specific FAs pattern might be associated with POAG.


Introduction
Glaucoma is the most common optic neuropathy leading to irreversible blindness worldwide [1]. It is a progressive neuropathy with multiple risk factors characterized by the destruction of retinal ganglion cells with a corresponding visual field loss. Glaucoma will affect 111.8 million people worldwide by 2040 with primary open-angle glaucoma (POAG) as the most common form [2]. Despite extensive research, its pathogenesis is still controverted. Established risk factors are age, elevated intraocular pressure (IOP), ethnic background and family history of glaucoma [3]. If elevated IOP is recognized as a major modifiable risk factor for the progression of glaucoma [4], literature shows that glaucoma can also develop under normal IOP conditions [5]. Other factors such as vascular disorder could be involved: decreased ocular blood flow could lead to ischemia and reperfusion damage of the optic nerve [6][7][8][9]. Inflammation and oxidative stress leading to microglial and complement activation are also suspected to play a role in the pathogenesis of glaucoma [10,11].
Polyunsaturated fatty acids (PUFAs), are essential molecules of our organism. They are constituents of cell membrane, precursors of hormones such as prostaglandins and steroids and participate in the regulation of genes [12]. Linoleic acid (LA) and α-linoleic acid (ALA) are called essentials PUFAs because they cannot be synthetized and have to be ingested in the diet. They are respectively the precursors of the omega 3 and omega 6 family of PUFAs and share the same enzymatic pathway for their production [12]. An imbalance in the input or production of these PUFAs was associated with the presence of systemic [13][14][15] and retinal diseases [16] such as age-related maculopathy [17], diabetic retinopathy [18]. In glaucoma, PUFAs have been suspected to play a role in the pathogenesis through their inflammatory and vascular regulation activity [19]. Previous studies investigated the link between some plasma and red blood cells membrane (RBCm) fatty acids (FAs) and the presence of glaucoma, notably docosahexaenoic acid (DHA), ALA, and eicosapentaenoic acid (EPA) [19][20][21]. Results of these case-control studies converge towards a reduction of omega 3, especially of EPA and DHA in glaucomatous participants. Similar analysis in larger population studies are necessary to confirm these results.
The purpose of our study was to compare the total plasma FAs level between participants with POAG and participants without neuropathy in an elderly population-based study and to investigate specific FAs pattern in POAG.

Study design
The Montrachet (Maculopathy Optic Nerve and nuTRition neurovAsCular and HEarT diseases) populationbased study was conducted from October 2009 to March 2013, in Dijon with volunteers, aged over 75 years. It was an ancillary study of the Three-City (3C) populationbased study. The methodology of the Montrachet and the 3C studies has previously been described [22,23]. All participants gave their informed consent and the study followed the tenets of the Declaration of Helsinki. The study has been approved by the Ethical Committee and registered as 2009-A00448-49. We followed the STROBE (Strengthening the Reporting of OBsevational studies in Epidemiology) statements [24] according to the EQUATOR (Enhancing the QUAlity and Transparency Of health Research) guidelines [25].

Eye examination
Participants underwent a complete examination in the Department of Ophthalmology, in Dijon University Hospital, France. Clinical, treatment, lifestyle and demographic data as well as eating habits were collected by a self-questionnaire. The eye examination included best corrected visual acuity, slit-lamp examination, IOP measurement by air tonometry (Tonoref II, Nidek, Aichi, Japan), central corneal thickness by ultrasonic contact pachymeter (DGH 500, DGH Technology, Exton, PA, USA), axial length measurement, macular and optic disc photographs with a fundus camera (TRC NW6S, Topcon, Tokyo, Japan) after pupil dilation and a visual field for screening (Frequency-Doubling Technology, Carl Zeiss Meditec, Dublin, CA, USA). An optic nerve head spectral-domain optical coherence tomography (SD-OCT) (Spectralis, Heidelberg Engineering Co., Heidelberg, Germany) was performed, after pupil dilation with retinal nerve fiber layer (RNFL) thickness measurement around a 3.5-mm-diameter circle. The diagnosis of glaucoma was determined from photographs of optical discs interpreted by two trained ophthalmologists, blinded for clinical and RNFL SD-OCT thickness data. In case of discrepancy, a glaucoma specialist adjudicated. Persons identified as glaucomatous were further examined with gonioscopy and visual field by the Swedish Humphrey Interactive Threshold Algorithm (SITA) 24-2 (Carl Zeiss Meditec, Dublin, CA, USA). Following these examinations, they were classified into three levels of evidence according to the ISGEO (International Society for Geographical and Epidemiological Ophthalmology) classification [26]. The OCT results were therefore not taken into account (Foster et al., 2002).The 97.5 and 99.5 percentiles for the vertical cup-to-disc-ratio found in our population defined the limits of the three levels of evidence at 0.7 and 0.8 respectively [27]. In this analysis, only cases of primary open-angle glaucoma were taken into account. Severity was determined with mean deviation (MD) of Humphrey's visual field according to the classification published by Hodapp et al. [28] An abnormal visual field test was defined when the Glaucoma Hemifield Test (GHT) was out of normal limits or at the limit [29,30]. We included only one eye per participants and we chose the eye with the most severe POAG in case of bilateral optic neuropathy.

Blood sampling
Blood samples were collected from fasted volunteers for plasma lipids and FAs analysis [31]. Lipids extracted from plasma were stored under inert gas then transmethylated with boron trifloride in methanol [32]. Finally, fatty acid methyl esters were isolated with hexane and analyzed by gas chromatography using a Hewlett Packard Model 5890 (Palo Alto, CA, USA) with a CPSI L-88 column (100mx0,25 mm i.d., fim thickness 0,20 μm; Varian, Les Ulis, France) equipped with a flame ionization detector. The carrier gas used was hydrogen. Sample concentration were determined by comparison to commercial and synthetic standards. The analysis focused on 25 FAs reported as a percentage of total FAs using the EZChrom Elite software (Agilent Technologies, Massy, France).

Exclusion criteria
We excluded secondary and angle-closure glaucoma and all participants with missing data on FAs or glaucomatous status.

Statistical analysis
Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range) when appropriate and categorical variables as number and percentages (n, %). Bivariate comparisons were performed with Student test or ANOVA or Kruskal Wallis tests for continuous variables and Chi-squared or Fisher exact tests for categorical variables when appropriate. In the first step of the analysis, associations between individual plasma FAs and POAG were evaluated by a multivariable logistic regression analysis. In the second step of the analysis, in order to identify the FAs patterns (included saturated, monounsaturated and polyunsaturated FAs) associated with the presence of POAG, we used the partial leastsquares (PLS) regression method. Before identifying FAs patterns, all FAs variable were centered and reduced at baseline. We retained the PLS component that was significantly associated with the presence of POAG. Identified PUFA pattern was constructed further to a score by weighting FAs concentration with factor weight values. To interpret the FA pattern, we kept those FAs with absolute values of weights ≥0.20. Individual factor scores were then categorized into tertiles. The lowest tertile of FAs pattern scores was defined as the reference group. Models were adjusted for age and sex, lipid-lowering drugs use and axial length. For all analysis, results were considered significant when P < 0.05. Analysis was performed using SAS software (version 9.4; SAS institute Inc.; Cary, NC, USA).

Results
Among the 1153 participants of the Montrachet study, 810 were included in this ancillary analysis with 68 participants with POAG ( Fig. 1).
Non-participants were statistically older than participants at the inclusion, 82.59 ± 3.89 and 82.11 ± 3.67 respectively (P = 0.046) ( Table 1). Clinical and demographic characteristics of POAG participants and participants without neuropathy are presented in Table 2. POAG participants were older (P < 0.003), had a higher axial length (P < 0.006) and higher vertical cup disc ratio (P < 0.001). The mean percentage of PUFAs in both groups was presented in Table 3. The univariate analysis showed that POAG participants had a significantly lower level of EPA (P = 0.032). There was no significant association for others FAs (including DHA) as well as for total omega 3 and omega 6. The omega 6 / omega 3 ratio was high in both groups, 7.89 ± 2.15 and 7.97 ± 1.56, respectively (P = 0.775). After adjustment for age, sex, axial length and lipidlowering drug intakes there was no significant association between EPA and POAG (P = 0.058) ( Table 4).
A plasma FAs pattern associated with the presence of POAG has been identified. This pattern explained 10.0% of the total variance in the original set of FAs. The pattern presented high negative weight of gamma-linoleic acid (GLA) and EPA PUFAs and Cis-7 hexadecenoic acid monounsaturated FAs (MUFAs). On the contrary, this pattern presented high positive weight of eicosadienoic acid, docosatetraenoic acid, docosapentaenoic n-6, ALA PUFAs, eicosenoic acid MUFAs and margaric acid and behenic acid saturated FAs (SFA) ( Table 5).

Discussion
In this population-based study we compared plasma FAs levels in participants with POAG and in participants without neuropathy by means of two methods: FAs individually and covariance analysis. We did not find any difference regarding plasma FAs level in POAG participants compared to participants without neuropathy after multivariable analysis.
Our results are not in line with previous studies which found an association between the level of plasma and RBCm FAs and glaucoma [19][20][21]33]. Ren et al. reported a decrease of EPA, DHA, ALA, and total omega 3 in glaucoma with a linear association with severity of POAG [20]. Yu et al. found a decrease of EPA only in severe NTG [21]. Acar et al. found that DHA level in RBCm was decreased in POAG patients in the preclinical stage [19]. At the opposite, Yuki did not find any association between NTG and FAs [33]. Contrary to our study, these studies were conducted with univariate statistical analysis and they included a smaller selected population. Moreover our population based study analyzed older age participants with a mean age of 82 years. We found a high ratio of omega 6 / omega 3 in the two groups. According to the nutritional recommendations this ratio should be comprised between 2 and 3 in order to reduce the risk of cardiovascular and neurodegenerative diseases [17]. Moreover, according to the nutritional AFSSA recommendation the linoleic acid (LA) / ALA ratio should be around 5 [34]. In this study both groups are 6 times superior to this recommendation [34]. This is in line with our previous findings showing that Montrachet population was in very good health.
We identified a specific plasma FAs pattern significantly associated with POAG. After adjustment for major confounders, individuals in the upper tertile of FAs pattern scores compared with those in the lower tertile were more likely to present POAG. This FAs parttern seems to be associated with POAG in our old population. The analysis of a FAs pattern might be more appropriate to investigate the pathogenesis of glaucoma rather than considering FAs individually.
The involvement of FAs in the pathogenesis of glaucoma could intervene at three different levels. First, FAs have an anti-inflammatory and neuroprotective effect in the retina by preventing the apoptosis of retinal cells and by regulating glia [16]. They are also major components of cell membranes and therefore have a key role in the fluidity and the aggregability of RBC [35,36]. An imbalance in their composition could lead to decreased ocular blood flow and optic nerve perfusion pressure as it is found in glaucomatous patients [37]. Finally, they are known to be the precursors of prostaglandins, well known for these hypotonizing and anti-inflammatory effects.  [38][39][40]. In humans, no strong association have been established between PUFAs intakes and glaucoma. Renard suggested in a recent observational study on a French glaucomatous population that omega 3 fatty acids intake have a protective effect against POAG [41]. A recent observational populationbased study suggest that an increased DHA and EPA and global omega 3 intake decrease the risk of glaucoma only if global PUFAs level intake is controlled [42]. A randomized controlled trial conducted by Garcia Medina did not find any improvement in glaucoma with antioxidants (with and without omega 3) intake [43]. Moreover, Kang found that a globally high omega 3 / omega 6 ratio increases the risk of glaucoma [44].
As no strong evidence of specifics FAs implication was demonstrated in our study, the benefit of supplementation in some FAs remains debated.
We acknowledge several limitations to this study. First, participants of Montrachet study are Caucasian, voluntary, elderly and from an urban population. These volunteers follow a healthy lifestyle. We could not extrapolate our results to another group. Second, we included uneven number of participants in the two groups because there were not many subjects diagnosed with glaucoma and therefore limits the interpretation of the results. Furthermore, the low number of participants with primary open-angle glaucoma could have reduced the   [45,46].
In conclusion, there is no significant difference regarding plasma FAs level in multivariate analysis when FAs are analyzed isolated. The global study of FAs highlighted a specific FAs pattern associated with POAG. The clinical relevance of this pattern warrants further studies. substantively revised it; AS, analysis and interpretation of data; PHG, revision of the draft; FB, revision of the draft; NA, conception and design of the work; CCG, conception and revision of the work. Each authors have approved the submitted version and have agreed both to be personally accountable for the author's own contributions and to ensure that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and the resolution documented in the literature.

Funding
Supported by an interregional grant (Programme Hospitalier de Recherche Clinique) and the Regional Council of Burgundy; by INRA, CNRS, Université de Bourgogne, Regional Council of Burgundy France (PARI Agrale 1), FEDER (European Funding for Regional Economic Development); and a French Government grant managed by the French National Research Agency (ANR) under the "Investissements d'Avenir" program, ANR-11-LABX-0021-01-LipSTIC Labex.

Availability of data and materials
The data that support the findings of this study are available from the 3C Committee but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of 3C Committee (E3C.U708@inserm.fr or louis.arnould@chudijon.fr).

Declarations
Ethics approval and consent to participate The study has been approved by the Ethical Committee of the 3 Cities Study and registered as 2009-A00448-49. Written informed consent was obtained from the participants or from legal guardians of the participants if they are blind and the study followed the tenets of the Declaration of Helsinki.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no non-financial competing interests.