Ⅰ. Introduction
Amblyopia is a preventable eye disorder where visual development in the brain is not sufficient to obtain best-corrected vision in one or two eyes due to a functional imbalance between the eyes. It is caused by stimulus-deprivation, strabismus, and refractive errors during infant and early childhood and may persist throughout life if proper treatment is not given. Its prevalence is approximately 2% to 3% in the general population.1-4) Treatment involves restoration of vision in the amblyopic eye, and is traditionally executed by eye care professionals until the critical period ends, usually up to 8 years of age.5) However, a recent clinical study using functional magnetic resonance imaging (fMRI) demonstrated that the visual cortex was reorganized in 50-year-old man in response to loss of cortical inputs caused by a stroke.6) Additionally, recent human studies7-11) have showed visual improvement in the amblyopic eye at the age beyond the critical period, a certain stage of development which the visual system would be most readily acquired by environmental stimuli. Animal studies12-20) have also been anatomically and physiologically underpinning the findings of human studies, indicating that there are some possibilities to recover amblyopia after the end of the critical period.
Environmental enrichment (EE) has been highlighted as an alternative model for amblyopia study. EE is a condition that basically promotes motor activities, sensory experience, social interaction and nutritional support. Their process stimulates the brain by encouraging social, physical, sensory and nutritional activities in their complex surroundings.21,22) Many researchers have started paying attention to the visual cortex since Wiesel and Hubel23,24) have proposed the visual cortex plasticity that is the capacity of the visual system to be responsive to the environmental stimuli, and there has been also interest in studies of EE in the visual cortex. Morphological studies have supported the notion of synaptic plasticity in the visual cortex, induced by EE during the critical periods, with provision of evidences such as increased extent of dendritic spine density and branching,25,26) and of neurotrophins. 27-30) On the ground of the outcomes of the amblyopia studies introduced above, it is therefore worth investigating the effects of EE on visual cortex plasticity after the end of the critical period.
This systematic review critically appraises whether the adult visual system can be anatomically, physiologically, behaviorally re-developed in response to the exposure of EE, and whether adult models of EE can be adequate to apply to other species of animals and humans.
Ⅱ. Methods
This systematic review adopted PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist.31)
1. Eligibility criteria for considering studies for the review
Studies describing any anatomical, physiological and behavioral effects of environmental enrichment on the visual cortex after the critical period in comparison with those of standard laboratory environment were selected. This systematic review focused at a rat model of EE as the most frequently used for EE studies. Articles written in other languages than English, conference proceedings, clinical studies, in vitro studies and reviews were excluded. Since adult-like vision of rats is nearly fully developed by the age of postnatal day 40,32) this review determined rats after postnatal day 40 as the target population.
2. Search methods for identifying studies
Our research question was “How did the visual system of adults with amblyopia respond to the exposure of EE?” The following electronic bibliographic databases were systematically searched by two authors (DL and SK): PubMed [1946-July 2014], Medline [1960-July 2014], EMBASE [1974-July 2014], Web of Science [1945- July 2014], and CINAHL [1960-July 2014]. The following search strategy was applied in Pubmed: ('environment* enrich*' OR 'enrich*' OR 'environment [MeSH Major Topic]' OR 'housing [MeSH Major Topic]' OR 'motor activity [MeSH Major Topic]' OR 'social environment [MeSH Major Topic]') AND ('plastic*' OR 'neuronal plasticity [MeSH Major Topic]' OR 'cortic* plasticit*' OR 'brain plasticit*' OR 'ocular dominance plasticit*' OR 'synaptic plasticit*' OR 'neuroplasticit*' OR 'long-term potentiat*' OR 'long-term synaptic depress*' OR 'experience dependent plasticit*') AND ('visual cortex [MeSH Major Topic]' OR 'visual cort*' OR 'primary visual cort*' OR 'ocular dominanc*' OR 'striate cort*'). All the key terms except MeSH Major Topic were probed in both title and abstract at each step of search. The same strategy except MeSH major topic was applied in Medline, CINAHL, Web of Science, and EMBASE.
3. Study selection
Amongst the studies initially collected, duplicate studies were removed by scanning titles and abstracts. Three authors (DL, SK and MJ) independently assessed the titles and abstracts in accordance with the research question and eligible criteria. When some studies were relevant to this systematic review, their full-text articles were retrieved and reassessed by the authors. Each full-text article selected was examined by the authors independently and then the eligibility of the systematic review was discussed. Additional studies were included after determining the eligibility of the full-text articles identified from the reference lists of the included studies.
4. Data collection and risk of bias assessment
In each included study, the following items were extracted for critical appraisal: authors, publication year, species of animal, strain, number and age of animals used, sex, duration of EE, conditions of housing, and types of assessment. Methodological appraisal was independently performed by three authors, and if the authors failed to reach a consensus, a third party intervened and resolved it by majority decision. The following parameters, adopted from Krauth, Woodruff and Bero,33) were appraised: randomization of allocation, description of housing conditions, blinded assessment of outcome, sample size calculation, incompletion of data, and conflict of interest statement. Each parameter was graded as “high risk”, “low risk” or “unclear”. When any parameters were graded as “unclear”, the authors directly contacted the corresponding authors for clarification.
Ⅲ. Results
1. Selection of studies
A total of 499 studies were initially identified from the five databases (72 from Medline, 3 from CINAHL, 224 from Web of Science, 95 from PubMed, and 106 from EMBASE) (Fig. 1). After removal of 107 duplicates, 392 studies were potentially eligible for this systematic review, and were retrieved for title and abstract review. Full-text review for the 49 remaining studies were retrieved and full-text review for four additional studies from the reference list of the 49 retrieved studies were also performed. In total, 15 studies were included for this systematic review and 39 studies were excluded. The reasons for exclusion are detailed in Fig. 1.
2. Characteristics of the included studies
Descriptive data of the 15 included studies including authors, year, species, strain, total number and age of animals, sex and duration of EE are summarized in Table 1. Long-Evans hooded rats were the most favorable animals used, and the total number of animals varied from 9 to 118. Amongst the included studies, only 2 studies used different age groups so as to investigate the relationship between EE and aging. Six out of fifteen studies did not state the sex of the animals used but the rest clearly disclosed it. Duration of EE varied from 14 to 365 days. The types of assessment used, not shown in Table 1, were behavioral assessment (1 of 15 studies), histological assessment (6 of 15 studies) and mixed assessment (8 of 15 studies). As shown in Table 2, cage size, housing density and cage contents for the EE groups were relatively similar to each other. Conversely, the size of cage and housing density for the control groups were definitely smaller than those for the counterparts (Table 3) but more details about the housing conditions for the control groups were not sufficiently provided in the half of the studies.
3. Risk of bias assessment
The included studies had strengths and weaknesses in methodology. For the strengths, all the included studies were less likely to be biased by providing clear descriptions of housing for EE, test procedures and animal conditions. When evaluating the conflict of interest statement, we found that only a few studies directly declared it. We decided that any statements of financial support from government or non-profit organization were considered as “low risk” and consequently, 13 studies were free from publication bias.
In contrast to the strengths, many studies were still likely to be biased. None of the included studies specifically stated sample size calculation. Some studies omitted report randomization of allocation of the animals used (11 studies), and reported with incomplete data such as the number of animals in the different groups (4 studies), sex (6 studies) and age (2 studies).
4. Evidence from the included studies
All the findings from the selected studies are summarized by anatomical, physiological and behavioral categories in Table 4.
1) Anatomical effects of EE
Several studies reported that experiencedependent changes through environmental enrichment result in morphological alterations in the visual cortex of adult rats. Compared to the control groups, rats exposed to EE had some growth in the number of spines and dendritic branches, and the length of dendrites and dendritic branches, which are markers for the post-synaptic site of synapses.34,35) However, sex differences were still present.34) EE was also likely to promote synaptic plasticity by increasing the rate of synapse formation, accompanied by a numerical increment of neurons in the visual cortex.36,37) A decrease of perineuronal nets (PNNs), as a role of synaptic stabilization, after EE exposure has been suggested as an opportunity to reopen visual cortex plasticity.38)
Furthermore, studies for neurotransmitters supported environmental enrichment reactivating visual cortex plasticity even after the critical period. EE declined a level of gamma amino butyric acid (GABA), an inhibitory neurotransmitter, in the visual cortex leading to initiation of ocular dominance plasticity by reducing intracortical inhibition.17,39,40) Specifically, glutamate decarboxylase 67 positive (GAD67+) interneurons related to lessening GABAergic inhibition and the density of PNNs were significantly reduced by EE.19,38) Conversely, serotonin and dopamine became more active in the visual cortex of EE adult rats, promoting heterosynaptic plasticity that synaptic strength at a postsynaptic cell is enhanced by stimulated neurons as well as other non-stimulated neighboring neurons.39,41) Nevertheless, other studies showed that excitatory mechanisms related to glutamate and [3H]d-aspartic acid were less likely to be involved in adult visual cortex plasticity.39,40)
According to immunohistochemistry studies, neurotrophins are also likely to be concerned in adult visual cortex plasticity. EE brought about both additional cells and expression levels for nerve growth factor (NGF) in adult visual cortex for promoting neuronal survival.27,30,42,43) For neuronal survival and differentiation, both the extra cells and expression levels of brain-derived neurotrophic factor (BDNF) and neurotropic factor-3 (NT-3) were also detected in that area.17,19,30,39) Notwithstanding the manifestation of such neurotrophins being positively correlated with adult visual cortex plasticity, the mechanism of cortical plasticity following EE exposure still remains unclear.
2) Physiological effects of EE
While it has been widely accepted that visual acuity (VA) develops and completes only during the critical period, environmental enrichment can promote visual improvement even afterwards. Sale et al.17) and Tognini et al.19) found that VA in an adult eye with initially reduced vision caused by monocular deprivation became almost similar to that in the other eye with normal vision by EE for at least two weeks by penalizing the normal eye. As evident from Baroncelli et al.40) vision improvement was primarily derived from exposure to various visual stimuli and physical activities rather than grouping.
As well as vision improvement, ocular dominance for binocularity may still be plastic even after the critical period. It is a functional ability of ocular dominance columns in the striate cortex to opt visual stimuli from one eye or the other, and might be fixed at the end of the critical period. A number of studies measured contralateral/ ipsilateral visually evoked potential ratio (C/I VEP ratio) in ocular dominance columns after the critical period by means of both monocular deprivation during the critical period and EE after the end of the critical period, and consistently found that preference to respond to the visual stimuli in ocular dominance columns changed from the initially non-deprived eye towards the deprived eye.17,38,39) In particular, analogous to the changes of VA, ocular dominance was more likely to be affected by both rich physical and visual circumstances.40)
3) Behavioral effects of EE
Converging findings for the functional effects of environmental enrichment from several studies showed that behavioral vision is liable to be improved even after a critical period. According to studies using a visual water-box task for measuring behavioral VA, adult rats exposed to EE could respond to higher frequencies of visual stimuli, which was originally unresponsive, in the eye with initially lower VA.17,19) With a Morris water maze task for behavioral cognitive assessment, performance to escape to the platform was enhanced in adult EE rats, indicating that their spatio-visual learning ability became more sensitive to respond.30,43) Furthermore, adult rats with reduced vision in one eye showed some improvement of stereoscopic sense in a visual cliff task by exposure of EE.12)
Ⅳ. Discussion
1. Strength and limitations of the included studies
The included studies have contributed to the development of reliability and transparency. Because visual cortex plasticity after the critical period has not been fully understood, such experimental designs used in the included studies help extend knowledge of the research topic. By means of good description of housing conditions in the included studies, a good reliability of the study design can be guaranteed and more feasible outcome can be also obtained in future studies. Furthermore, declaration of no conflict of interest in most of the included studies can assure that research outcomes are ethically proven.
However, the main weaknesses of the studies are in reporting research methodology. As mentioned above, several studies did not directly state randomization (79%), blinding (64%) or complete descriptive data of the animals used (i.e. the number of animals per group, sex or age; 29%). Whether or not these were intentionally omitted for convenience, such omission may undermine the research outcomes or mislead readers. For instance, unless randomization and blinding are applied and reported, more positive outcomes are prone to be reported.44) Additionally, a lack of information about the number of animals per group was more likely to overestimate the effect size of intervention.45) To prevent such adverse effects, better reporting to precisely provide these important elements is necessary.
2. Applicability to other species
EE applied to other animals after the end of the critical period of visual cortex deserves attention. Since Wiesel and Hubel23,24) firstly discovered functional visual processing in the visual cortex of both mice and cats, many studies of visual sciences including EE have preferentially proposed rodent models. Additionally, recent studies have supported this trend by providing evidence that the visual cortex in mice shows similar properties of neurons for both the orientation and spatial frequency compared to other higher species such as monkeys.46) It was also found that the genetic properties of the visual cortex in mice are analogous to those in human beings.47) Following this trend, several studies attempted to prove it with other species. For instance, a few EE studies with young mice showed anatomical, physiological and behavioral changes similar to those found in the selected studies for this review,29,48,49) and a recent work conducted by Greifzu et al.50) examines these effects in adult mice as well. In addition to mice, young cats also show some functional visual development induced by EE.51,52) Reflecting on these findings, it is worth applying EE to adult mice and cats in order to prove possibilities of reactivation of visual cortex plasticity after critical period, which are shown in the selected studies with adult rats.
Regardless of the age of animals used, it remains unclear how the findings of studies related to EE can be applied to human beings. The first barrier is that most animals in these studies have nocturnal lifestyles dissimilar to human beings, indicating that other senses instead of vision are selectively more developed. For example, as the visual acuity of mice is tenfold worse than that of humans,53) different visual acuity may act on visual development and processing in different ways. Apart from visual acuity, the level of BDNF responsible for synaptic plasticity is more elevated when the rodents are exposed to daylight than when they are in darkness.54) Therefore, its mechanism may be hard to directly apply to diurnal animals including humans. Additionally, development of binocularity in the visual cortex can be considerably different between such nocturnal animals and humans, because the smaller and simpler ocular dominance columns of cats and mice are less binocularly-driven than humans.55,56) Namely, less developed binocularity in these animals may be difficult to reflect the effects of EE on humans whose binocular vision is more advanced. Larger and higher animal models therefore need to be considered for EE before and the critical periods.
3. Limitations of this systematic review
With the objective of discussing the potentiality of the reopening of plasticity in adult amblyopia reacted by the exposure of EE, this review focused on rats due to the limited amount of studies. In contrast to the studies for EE during the critical periods, other animal models had been rarely reported so that the target population of this systematic review needed to converge towards the rat model. In addition, while this systematic review attempted to separately provide anatomical, physiological and behavioral effects of EE, there are still gaps to explain how much each enrichment contributes to these effects. To overcome these weaknesses, further investigations with other species of animals are essential to appraise the validity of the effects of EE in the adult brain, and especially each category of enrichment needs to be analyzed individually.
Ⅴ. Conclusions
The outcomes from the included studies substantiate the possibility of EE effects on adult visual cortices but significant gaps to apply them to other species of animals including humans still remain. The potential effects of EE may extend into adulthood as shown in the included studies, but we should avoid direct interpretation of them as no studies have attempted to differentiate the influences of possible confounders such as stress, sensory stimulation and motor activities. While the findings from this review do not guarantee analogous outcomes in other species, this review strived to offer methodological guidelines for the future studies of EE into adulthood. We can therefore suggest that better reporting and preclinical studies with other species of animals are an essential alternative and still far outweigh the necessity for clinical trials of EE studies on adult amblyopia.