Methodological quality of the review: Low confidence
Authors: Patel S, Mathan JJ, Vaghefi E, Braakhuis AJ
Region: Italy, Japan, China, Korea
Equity focus: None specified
Review type: Other review
Quantitative synthesis method: Meta-analysis
Qualitative synthesis method: Not applicable
Glaucoma is a leading cause of irreversible blindness worldwide. A major symptom of this pathology is the loss to the visual field in a peripheral to central pattern. Flavonoids are polyphenol compounds sourced from plants, commonly found in green tea, red wine and cocoa, and they have neuroprotective and antioxidant characteristics proposed to be advantageous within the context of glaucoma. Currently, the literature presents conflicting evidence regarding the effect of flavonoids on patients with glaucoma and ocular hypertension.
To synthesise evidence to achieve a precise, statistically convincing consensus for the future of glaucoma management in relation to the use of flavonoids.
In total, the authors included six studies in the meta-analysis and a further two in the systematic review. Six trials were parallel-arm trials and two employed a randomized cross-over design. All studies gave patients the flavonoid compound in the form of an oral tablet that they swallowed. Two studies were conducted in Italy, four in China, one in Korea and one in Japan.
Based on the risk of bias assessment, the authors noted that some studies did not provide sufficient detail on sequence generation, allocation concealment or blinding of assessment.
Three studies contained data for ocular blood flow. The flavonoid intervention showed a statistically significant reduction in the resistive index of the ophthalmic artery, although it did not directly demonstrate that this improved blood flow of the optic nerve head.
Overall, the authors found the mean difference among participants with severe baseline deterioration was 2.75 dB (1.77, 3.73), compared to 1.33 dB across all studies. However, no difference was noted when the same subgroup analysis was performed for intraocular pressure (IOP) or blood pressure (BP). Additionally, the duration of the majority of the studies was six months or less, with only three studies greater than six months in length. In an independent meta-analysis of short-term trials, the authors report that it was clear that the inclusion of long-term trials had no effect on their results for visual field (test statistic 2.02 dB; 1.22, 2.82). The authors noted that there were not enough long-term trials with data on MD to perform a separate meta-analysis. When short-term studies were analysed independently for IOP outcome, there was no longer any trend towards favouring the flavonoid and the summary measure crossed the line of no effect. In meta-analysis of the long-term studies on IOP, there was a greater trend towards favouring the flavonoid, but the data still crossed the line of no effect (−0.85 mmHg; −1.97, 0.27).
Data on IOP from different flavonoids (EBHM, GBE) were analysed separately from group results to determine the effect of flavonoid type. The summary measure still did not cross the line of no effect, and favoured the flavonoid for visual field, and there was also no change in the IOP meta-analyses.
The authors overall conclusions were that the meta-analysis on visual field showed that flavonoids had a statistically significant effect on maintaining or improving deterioration scores despite the short four-week time frame of some trials.
The authors conducted a search on EMBASE, AMED, CINHAL and CENTRAL. Two reviewers performed the literature search independently. Inclusion criteria consisted of patients aged 16 years or over; patients suffering from glaucoma and or/ocular hypertension; dietary interventions longer than one month; a placebo group as the control intervention; and randomized-controlled trials or randomized-controlled crossover trial study design.
Primary outcome measures were visual field mean deterioration and intraocular pressure.
The authors used the Cochrane collaboration tool to evaluate the risk if bias in the include studies. Data extraction was conducted by two reviewers independently, discrepancies were resolved by discussion.
The authors conducted a meta-analysis of included studies, where p<0.05 was considered statistically significant. Continuous data from studies included were expressed as mean difference and standard deviation with 95% confidence interval. The authors calculated statistical heterogeneity using the Higgins I2 statistic. Mean differences were combined using the fixed effect models and generic inverse method to obtain the weighted mean difference. The authors also conducted a sub-group analysis to analyse the influence if each study characteristics on the meta-analysis. These characteristics were length of intervention, type of flavonoid consumed, baseline visual deterioration and diabetic participants versus non-diabetic.
The authors note that despite limited data, they conclude that flavonoids are capable of improving ocular blood flow.
The authors did not discuss applicability of findings to low- and middle-income countries.
Summary of quality assessment:
Overall, low confidence was attributed in the conclusions about the effects of this review as important limitations were identified. Although the authors used appropriate methods to statistically analyse findings of included studies and screen and extract data of included studies, the authors did not conduct a thorough search of the literature to ensure that all relevant studies were included in the review. In addition, it is not clear from the review if the authors avoided language bias and if no date restrictions were applied.
Patel S, Mathan JJ, Vaghefi E, Braakhuis AJ (2015) The effect of flavonoids on visual function in patients with glaucoma or ocular hypertension: a systematic review and meta-analysis. Graefes Arch Clin Exp Ophthalmol. 2015 Nov;253(11):1841-50.