Age-related macular degeneration (AMD) is now the leading cause of blindness in developed countries. AMD-related choroidal neovascularization (CNV) or geographic atrophy (GA), in the United States, is expected to increase by 50% by 2020 [1
]. Effective treatments, for both early and late AMD, are presently lacking. Major efforts have been made in order to detect the pathogenetic mechanisms of AMD, but the exact etiology of AMD is still unclear [2
]. Previous epidemiological studies showed that tobacco smoking was the only consistent causative factor and that other risk factors, such as alcohol consumption and cardiovascular diseases, are inconsistent for AMD incidence or progression [3
]. The detections of the potential modifiable factors for AMD incidence would provide better strategies for primary prevention in the future.
As oxidative stress is one of the key pathogenetic factors in the development of AMD, use of antioxidant supplements has been regarded as an effective management strategy of AMD. Antioxidant supplement consumption, including polyunsaturated fatty acids (PUFAs) intake, has been postulated to be a protective factor of AMD [4
]. Evidence from cross-sectional [5
] and cohort studies [6
] demonstrated a significant association between n
-3 fatty acid consumption and reduced risk of late AMD. In a study of an elderly French population, high concentrations of plasma n
-3 fatty acids were associated with a decreased risk of late AMD [7
]. As we know, the main dietary source of PUFAs is oily fish (e.g., mackerel, tuna, salmon, sardines, and herring) [8
], and fish consumption has been reported to be associated with a reduced risk of different types of cancers, diabetes, and several other diseases [9
]. Based on cross-sectional [11
], case-controlled [13
], and cohort studies [14
], fish intake was reported to be associated with a lower risk of AMD. However, there were also a few studies that demonstrated no effect of fish intake on AMD risk. The Eye Disease Case Control Study (EDCC) found no effect of fish intake on incidence for neovascular AMD [15
]. In addition, a retrospective analysis of 1968 participants found that fish intake was not associated with AMD incidence compared to less frequent fish consumption [16
Meta-analyses, which are a useful statistical tool, could pool the relevant, but independent, studies together and, thus, come to a more powerful conclusion. Meta-analysis was also used in the detection of potential risk factors for AMD. For instance, based on a combination of five prospective cohort studies, Chong et al. found that heavy alcohol consumption was associated with an increased risk of early AMD [17
]. For these reasons, a meta-analysis and systematic review of the association between fish intake and risk of AMD may help to clarify this issue. The aim of the current meta-analysis was to quantitatively evaluate findings from observational studies on the association between fish consumption and AMD incidence.
A total of 4202 cases with 128,988 individuals from eight cohort studies were identified in this meta-analysis. All the included studies demonstrated a relatively high methodological quality. The findings of this meta-analysis indicated that fish consumption was associated with a reduced risk of AMD. Meanwhile, subgroup analysis by AMD stages showed that fish consumption could reduce the risk of both early and late AMD. When stratified by the follow-up duration, fish consumption was a protective factor of AMD, in both over and less than a 10-year follow-up duration. We also detected an association between different types of fish and risk of AMD. Advanced subgroup analysis showed that dark meat fish and tuna fish showed a protective effect on AMD. In addition, obvious evidence of a statistically significant dose-response relationship between fish intake and AMD risk was detected.
It was reported that inflammation and oxidative stress were key pathologic processes in the development of AMD [31
]. Those two examples have long been regarded as potential targets of pharma-projects and primary prevention. PUFAs, which are usually acquired from seafood intake, have been reported to modify the inflammatory reactions and oxidative stress in several diseases [32
]. It is natural to presume that additional supplementation of PUFAs would lead to a prevention in the incidence and progression of AMD. Previous epidemiological studies and clinical trials have shown that PUFA supplementation could reduce the risk of AMD [33
]. However, there were also two studies that reported an increased risk of AMD with higher n
-3 PUFA consumption [5
]. A meta-analysis on the association between n
-3 PUFA intake and AMD risk showed that higher n
-3 PUFA intake could reduce AMD risk [36
]. Plasma n
-3 PUFA, a nutritional biomarker of n
-3 PUFA status, was reported to be associated with the incidence of AMD. In a population-based study on nutrition and age-related eye diseases, performed in 963 residents of Bordeaux (France) aged ≥73 years [7
], it was found that high concentrations of plasma n
-3 PUFAs are associated with a decreased risk for late AMD.
Fish, especially tuna fish, is the main source of PUFAs, and higher fish consumption can increase the concentrations of n
-3 PUFA in blood [37
]. In this study, we found that fish consumption could reduce the risk of AMD, and a dose-response effect of fish intake on the incidence of AMD was detected. This result was very consistent with the results from several previous studies. In this meta-analysis, only prospective cohort studies were included. Certainly, case-control studies and cross-sectional studies can provide clues of the related factors of diseases, however, the evidence from cohort studies avoid these types of potential selection biases. The meta-analysis of cohort studies can provide evidence with a higher quality. In the US Twin Study, increased intake of fish reduced risk of AMD, particularly for two or more servings per week [38
]. Another multicenter eye disease case-control study involving five US clinical ophthalmology centers showed interesting results; compared with age and sex matched controls, higher fish consumption tended to reduce risk of AMD when the diet was low in linoleic acid. In another case-control study, with 437 advanced AMD cases and 259 unrelated controls, risk of AMD incidence was found to be 51% lower in the highest quartile of fish intake compared to the lowest quartile (OR = 0.49, 95% CI, 0.26–0.90) [39
]. Considering that most case-control showed a significant association between fish intake and reduced AMD incidence, it could be presumed that the exclusion of case-control studies in this current meta-analysis would not lead to a significant change in the main outcome.
In this meta-analysis, we found that it was tuna fish but not other types of fish that could reduce the risk of AMD. This was also found in the Nurses’ Health Study (NHS) and the Health Professionals Follow-up Study, a significant inverse association was found only in tuna group. The pooled RR of participants who ate canned tuna more than once 4 times per week was significantly lower (RR, 0.61; 95% CI: 0.45, 0.83) [28
]. Tuna fish is rich in PUFAs and it is usually consumed because of its low price. Tuna oil, which is from the muscles of deep sea tuna, contains high concentration of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). A previous cross-sectional study showed that both DHA and EPA consumption was associated with a reduced risk of neovascular AMD [40
], while only DHA but not EPA was found to associated with AMD risk in NHS. High DHA content is detected in both brain and retina. Therefore a constant supply of DHA was required for normal regeneration of photoreceptor outer segments and thus produced protective effect in degenerative diseases such as AMD. Increasing evidence showed that the function of DHA on the photoreceptor and retinal degenerative diseases was quite important. DHA is very important on normal conduction in retinal light stimulation. Exogenous DHA helps to keep the fluidity retinal cell membrane. EPA can reduce blood viscosity, dissolve excess fat in the blood vessel wall and reduce blood fat, prevent and improve the effect of cardiovascular. Besides, EPA could help the normal function of DHA in the retina. Moreover, in the consumption of tuna salad, essential fatty acid in tuna salad was mixed vegetable fat and might thus produce more powerful protection in the AMD incidence. Additionally, some other components of tuna fish might affect the incidence or progression of AMD. As we know, different risks modified by fish intake were associated with processing methods. As we see, baked, broiled, fried, and smoked fish intake was not associated with a risk of AMD. We hypothesized that baking, broiling, frying, and smoking processing methods might be harmful to the beneficial materials in fish. In addition, the baked, broiled, fried, and smoked fish processing methods might produce harmful effects for AMD development. Besides, because more significant effect of DHA was detected in mechanism of action and epidemiological features, high DHA/EPA ratio in tuna might explain its particularly stronger inverse association with AMD. Considering few study focused the contribution of DHA/EPA ratio in the AMD progressing, advanced epidemiological studies and experimental studies were required. However, it should be noted that the amount of the publications included in the fish subtype meta-analysis was small and the results in this meta-analysis need to be further confirmed by advanced well-designed study.
Several previous trials were conducted in order to explore the effects of PUFA supplementation on the prevention of AMD. The Nutritional AMD Treatment 2 Study was conducted to evaluate the efficacy of DHA-enriched oral supplementation in preventing exudative AMD [41
]. The study was a randomized, placebo-controlled, double-blind, parallel, comparative study, and a total of 263 patients with early AMD lesions and a visual acuity better than 0.4 logarithm of minimum angle of resolution units were included. In wet AMD cases, DHA-enriched supplementation for three years had no significant protective effect on choroidal neovascularization (CNV) incidence in the second eye, as did the placebo. The Age-Related Eye Disease Study 2 (AREDS2) was a multicenter, placebo-controlled RCT in 2006–2012. A total of 4203 participants who were at risk for AMD progression were included in the clinical trial and therapeutic effects of different treatment protocols were compared [42
]. It was reported that supplementation of lutein + zeaxanthin, DHA + EPA, or both, failed to further reduce the risk of progression to late AMD. The evidence from the RCTs showed the PUFA supplementation might be not associated with the incidence or progression of AMD. However, the conclusion that fish consumption could reduce the incidence of AMD may not be influenced. Fish is a kind of food with complex components and we cannot exclude the possibility that some other components in fish may also contribute to the association. It should be noted that tuna fish, especially tinned tuna, is an important source of meso-zeaxanthin. Meso-zeaxanthin supplementation has been shown to improve macular pigment optical density in both AMD patients and healthy subjects in a dose-response relationship [43
]. In the Meso-zeaxanthin Ocular Supplementation Trial (MOST), it was found that a significant increase in macular pigment from baseline was observed in the meso-zeaxanthin treated group [44
]. A previous meta-analysis regarding RCTs showed that n
-3 PUFA supplementation in people with AMD does not increase the progression or development of AMD [45
]. As reported in the SELECT Trial, it was found that men in the highest quartile n
-3 PUFA level had an increased risk for prostate cancer [46
]. It was observed that n
-3 PUFA supplementation might produce certain harmful effects on chronic inflammation, and a possible explanation for this relates to the fact that polyunsaturated fatty acids act as a substrate for reactive oxygen damage. Dark meat fish, which was the richest source of (docosahexaenoic acid) DHA and (eicosapentaenoic acid) EPA, was associated with reduced AMD risk in this meta-analysis. Thus, additional well-designed studies are required for the detection of the protective effects of anti-oxidants in early AMD.
There are several strengths in this current meta-analysis: (1) A relative comprehensive literature search strategy was used in the search for related publications. We searched databases, including the key words “life style” OR “dietary factor” to detect all available studies; (2) Only prospective cohort studies were included in this meta-analysis, and all included studies demonstrate a relatively high quality. Thus, no significant selection bias influences the conclusion of this study. Robust conclusions were proven through detailed sensitivity analysis and, thus, it suggests that the conclusions of this study are quite credible; (3) A dose-response analysis was conducted and we detected a dose-response relationship between fish intake and AMD risk. The advanced analyses using available data could provide a better understanding of the effect of fish consumption on the risk of AMD.
As with any meta-analysis of observational studies, our study has several limitations. Firstly, the amount of included studies was small. Even through a comprehensive literature search was conducted, only eight studies were included in this meta-analysis. This limited the dependability of subgroup analysis, as only a few studies were included. Secondly, most studies did not provide data stratified by some important confounding factors, such as tobacco smoking and family history. Although all the RR values of the included studies were adjusted by key factors, the influence of these factors should not be ignored. These points all indicate the requirement of additional well-designed studies in the future.