aristotle university of thessaloniki faculty of health sciences school of medicine the potential role of

57 Pages·2017·0.67 MB·English

ARISTOTLE UNIVERSITY OF THESSALONIKI
FACULTY OF HEALTH SCIENCES
SCHOOL OF MEDICINE

THE POTENTIAL ROLE OF NUTRITION ON LENS PATHOLOGY: A SYSTEMATIC
REVIEW AND META-ANALYSIS

A thesis submitted in fulfilment
of the requirements for the degree of
Master of Science in Medical Research Methodology
By
Sideri Olympia

Thessaloniki, December 2017

Master’s Committee:
Advisor: Tsinopoulos Ioannis
Mataytsi Asimina
Tsaousis Konstantinos

Word count: 8,291 words
TABLE OF CONTENTS

ABSTRACT ....................................................................................... ERROR! BOOKMARK NOT DEFINED.
INTRODUCTION .............................................................................. ERROR! BOOKMARK NOT DEFINED.
METHODS .....................................................................................6ERROR! BOOKMARK NOT DEFINED.
RESULTS ......................................................................................... ERROR! BOOKMARK NOT DEFINED.
FIGURE 1........................................................................................... ERROR! BOOKMARK NOT DEFINED.
FIGURE 2........................................................................................... ERROR! BOOKMARK NOT DEFINED.
FIGURE 3........................................................................................... ERROR! BOOKMARK NOT DEFINED.
FIGURE 4........................................................................................... ERROR! BOOKMARK NOT DEFINED.
FIGURE 5........................................................................................... ERROR! BOOKMARK NOT DEFINED.
DISCUSSION .................................................................................... ERROR! BOOKMARK NOT DEFINED.
REFERENCES ................................................................................... ERROR! BOOKMARK NOT DEFINED.
CHARACTERISTICS OF STUDIES ....................................................... ERROR! BOOKMARK NOT DEFINED.
DATA AND ANALYSES ..................................................................... ERROR! BOOKMARK NOT DEFINED.
APPENDICES ................................................................................... ERROR! BOOKMARK NOT DEFINED.

2
ABBREVIATIONS

AHRQ: Agency for Healthcare Research and Quality
BCVA: Best Corrected Visual Acuity
BMI: Body Mass Index
BP: Blood Pressure
CDSR: The Cochrane Database of Systematic Reviews
CENTRAL: Central Register of Controlled Trials
CI: Confidence Intervals
FFQ: Food Frequency Questionnaire
HR: Hazard Ratio
HRT: Hormone Replacement Therapy
ICTRP: The WHO International Clinical Trials Registry Platform
IU: International Unit
IV: Inverse Variance
LILACS: Latin American and Caribbean Literature on Health Sciences
LOCS: Lens Opacification Classification System
mg: milligram
mRCT : the metaRegister of Controlled Trials
OR: Odds Ratio
PSC: Posterior Subcapsular
RR: Risk Ratio
RRR: Relative Risk Ratio
UV: Ultraviolet
VA: Visual Acuity
μg: microgram
μmol: micromole

3
ABSTRACT

Background: Cataract is a major cause of blindness today. Certain antioxidants are suspected to have
a protective effect on cataract, as oxidative stress is one of the main mechanisms of lens
opacification.
Objectives: We examine the role of certain antioxidants in cataract prevention.
Search methods: We searched the Cochrane Database of Systematic Reviews (CDSR), Central
Register of Controlled Trials (CENTRAL), MEDLINE, ScienceDirect, Latin American and Caribbean
Literature on Health Sciences (LILACS), the metaRegister of Controlled Trials (mRCT),
ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP), Agency for
Healthcare Research and Quality (AHRQ) and Open Grey (System for Information on Grey Literature
in Europe) and hand-searched relevant reference lists. Last search was performed on 15 October
2017.
Selection criteria: We included observational studies investigating the association between one or
more of the following micronutrients and cataract: Vitamin A, Vitamin C, Vitamin E, lutein,
zeaxanthin, α- and β-carotene.
Data collection and analysis: Two independent authors extracted data and assessed their quality. We
pooled results for cataract incidence for all types of cataract and independently for nuclear, cortical
and posterior subcapsular cataract. We did not perform sensitivity analysis.
Main results: Twenty-five studies were included in the qualitative and 24 in the quantitative part of
the study with 295,821 participants over 30 years old. Results from pooled analysis showed a
protective effect of antioxidants on cataract, but not all of them were statistically significant. For
Vitamin C OR=0.88, 95% CI [0.81, 0.97], for Vitamin E OR=0.84, 95% CI [0.70, 1.01], for Vitamin A
OR=0.90, 95% CI [0.80, 1.00], for alpha-carotene OR=0.92, 95% CI [0.85, 1.00], for beta-carotene
OR=0.89, 95% CI [0.83, 0.95], for lutein and zeaxanthin OR=0.92, 95% CI [0.85, 0.99].
Conclusions: Our study managed to show a relation between certain antioxidants and cataract
disease, but further studies, especially interventional, are needed to confirm the hypothesis.

4
INTRODUCTION
Rationale
Cataract is the leading cause of blindness worldwide. It is estimated that the prevalence of the
disease will rise by 50% until 2020 and 30.1 million people will be affected either in one or both
eyes, only in the United States (1). Today, one in two people over 75 years old have poor vision
because of cataract and the cost is high not only for patient’s quality of life but also for healthcare
systems all over the world. Previous studies have shown that a preventive policy that could delay the
onset of the disease by 10 years, it could also reduce the expenses related to cataract by five to six
billion USD (2).
The disease inflicts eye crystalline lens, causing a progressive opacification. Depending on the
affected area, there are three main types of cataract: Subcapsular, nuclear and cortical. Most
frequently, lens opacification is due to advanced age. Other cataract risk factors include diabetes, UV
radiation, smoking, hyperlipidemia and statins, obesity, alcohol, previous eye injury or surgery, use
of corticoids or hormone replacement treatment, myopia and family history.
The main symptom related to cataract is the progressive loss of visual acuity (VA). Today, the only
way of treatment is surgical, and it consists of the extraction of the affected lens and its substitution
with artificial lens. Undeniably, much progress has been done on the surgical procedure itself, which
is now easier, safer and painless for the patient. Nevertheless, clinicians still don’t have other means
of treatment available, e.g. eye drops, protective sunglasses, which is the case for other eye
diseases.
In terms of prevention, a couple of studies have been held the last few years examining a possible
relation between diet and lens opacification. The reason why such a relation would be interesting
lies on the fact that cataract formation is mainly due to the activation of oxidative mechanisms
resulting in the aggregation of damaged proteins on crystalline lens. Thus, nutrients with antioxidant
effect could possibly stop and even reverse the oxidative effect either by attenuating or by repairing
the damaged proteins. Furthermore, if such a relation exists for cataract, then the results could
easily be generalized to other degenerative diseases and we may even be in place to claim that
nutrition plays generally a very important role against aging of living organisms. Antioxidants that
are more frequently studied include Vitamin C, Vitamin E, tocopherols, carotenoids (β-carotene, α-
carotene, lutein, zeaxanthin, lycopene), Vitamin A and n-3 fatty acids.
Vitamin A plays a crucial role in formation of rhodopsin (3). Rhodopsin, the complex of opsin and
retinal (the aldehyde form of Vitamin A), is essential for the transmission of the visual signal. There
5
are two sources of Vitamin A, from animals (liver, cow’s milk, human milk) and from plants (orange,
red or green fruits and vegetables). Carotenoids are a group of micronutrients that can be either
converted into Vitamin A (pro-vitamin A carotenoids) or not. The first category includes α- and β-
carotene, that are usually found in fruits and vegetables, whereas the second category includes
lutein, zeaxanthin and lycopene, that are found in human milk and vegetables.
Lutein and zeaxanthin, normally found in green leafy vegetables, are lipid-based antioxidants and
they act protectively to the lens by absorbing light that normally peroxides lipids in the outer retina.
DHA (docosahexaenoic acid, C22:6 n-3) is a basic component of disc membranes, thus enabling
visual signal generation. Deficiency in this type of n-3 fatty acid reduces membranes’ fluidity and
causes problems throughout the whole visual path. DHA is normally found in fish, meat and eggs.
The same, Vitamin E protects polyunsaturated fatty acids (PUFA) of membranes from peroxidation.
Vitamin E is found mainly in meat, eggs and leafy vegetables.
Finally, Vitamin C, found mainly in fruits and vegetables but also in animal products, offers also
protection from oxidation. Nevertheless, there are studies supporting an adverse effect of Vitamin C
on crystalline lens. Excessive intake may contribute to protein modification and onset of cataract
genesis (4).

Objectives
The aim of this study is to present new evidence between a diet rich in antioxidants and prevention
of cataract, based on the latest observational studies findings. Although results from interventional
studies usually lead to less biased outcomes, the available trials in the bibliography on the topic are
very few and not sufficient to be combined in a meta-analysis. Apart from that, we will try to clarify
if there are specific doses to which these nutrients exert their optional effect. We will also try to
define whether the nutrient have the same effect on all types of cataract or not. The micronutrients
examined in the present study are Vitamin A, Vitamin C, Vitamin E and carotenoids. It is worth to say
that there isn’t any registered protocol of ongoing studies on the topic of interest now.

METHODS
Eligibility criteria
6
Types of participants
We included participants irrespectively of their age and origin, males and females, with the diagnosis
of age-related cataract in one or both eyes. We excluded studies with patients suffering from
cataract from causes others than age.

Types of interventions
We included case-control, cross-sectional studies and cohorts investigating the association between
one or more of the following micronutrients and cataract: Vitamin A, Vitamin C, Vitamin E, lutein,
zeaxanthin, α- and β-carotene. There were no restrictions regarding dosage, mode of administration,
frequency, duration and time of administration. Only published trials, written in English, concerning
Humans and publication date after 1996 were included.

Types of outcomes
Primary outcome
1. Incidence of cataract, as defined by the included studies
Secondary outcomes
1. Effect on each subtype of cataract (nuclear, cortical, subcortical)
2. Doses at which antioxidants exert their optimal effect

Information sources
We searched the Cochrane Database of Systematic Reviews (CDSR), Central Register of Controlled
Trials (CENTRAL), MEDLINE, ScienceDirect, Latin American and Caribbean Literature on Health
Sciences (LILACS), the metaRegister of Controlled Trials (mRCT), ClinicalTrials.gov and the WHO
International Clinical Trials Registry Platform (ICTRP). Regarding grey literature, we searched Agency
for Healthcare Research and Quality (AHRQ) and Open Grey (System for Information on Grey
Literature in Europe). Even though we chose to include only observational studies in our Review, we
also searched in databases mainly including clinical trials in order not to miss any studies with both
an observational and an interventional part. We also searched the reference list of recent Systematic
7
Reviews for any missing studies. All databases have been searched from 1997 until October 2017
and last search was performed on 15 October 2017.

Search
Search strategy of each database can be found in an appendix (appendix 1-7). The accuracy of the
search was evaluated by the Evidence Based Checklist for the Peer Review of Electronic Search
Strategies (5).

Study selection
The studies for inclusion were selected by two independent reviewers. Mendeley Desktop 1.17.9
was used for search results handling. The titles and abstracts of all relevant electronic search results
were examined. Duplicates were removed. For studies that met the inclusion criteria, the full-text
reports were obtained and examined. Full-text reports that were compatible with all inclusion
standards, together with data from other sources (grey literature), were examined again. Any
disagreements between reviewers were resolved by discussion. In case of multiple reports of the
same study, we decided to choose the highest quality paper, while for multiple papers of the same
study covering different follow-up periods we chose the paper with the longer follow-up period.

Data collection process
Data were extracted on excel data collection forms for non-interventional studies from two
independent review authors. Any disagreements were resolved by discussion. A pilot test of the
form was carried out firstly in few studies. The appropriate software (Review Manager 5.3) was used
for data management. When results from both per-protocol and intention-to-treat (ITT) analysis
were reported, we chose to include only results from the ITT analysis.

Data items
The following data were extracted:
• Study ID
• Methods: design; duration of study;
8
• Participants: population; setting; age; sex; country of origin; inclusion/exclusion criteria
• Outcomes: outcome name; time points measured; unit of measurement; scales: upper and
lower limits; dosage; subgroup analysis; imputation of missing data; number of missing
participants with reasons; power; statistical methods; covariates
• Others: study funding sources

Risk of bias in individual studies
Risk of bias assessment of included studies was based on the Newcastle-Ottawa quality assessment
scale (6). The domains evaluated were selection, comparability and either exposure (for case-control
studies) or outcome (for cohorts). Studies awarded with five or more stars were evaluated as high
quality or low risk studies. Two independent reviewers evaluated the included studies. Any
disagreements were resolved by discussion. The evaluation was done based on available, already
published data.

Summary measures
The Odds Ratios (OR), Risk Ratios (RR), Relative Risk Ratios (RRR), Hazard Ratios (HR) and adjusted-
OR or adjusted-RR were used for study comparison in the narrative part of the study. In the meta-
analysis we included only studies with all necessary statistical data available. We entered the above
data into RevMan software to extract the Inverse Variance (IV) Odds Ratio with 95% CI of each
outcome.

Unit of analysis issues
The unit of measurement was the individual. Nevertheless, in case that there were patients with
both eyes with cataract, we tried to confirm whether results were extracted separately for each eye
or not.

Synthesis of results
We assessed clinical heterogeneity of studies based on each study population characteristics,
setting, sample size and covariates. We assessed statistical heterogeneity using the Chi2 and I2
values. For I2>70% and/or important clinical heterogeneity the study results were presented only in
9
the narrative part of the study. It was decided that for important clinical heterogeneity we would
use the random-effects model, while for no heterogeneity or no more than three studies in the
meta-analysis we would use the fixed-effects model. Results would be also presented by forest plots
when more than two studies would be combined.

Risk of bias across studies
It was decided that we would create a funnel plot to exclude or confirm presence of publication bias
depending on the number of included studies for each outcome of interest. Egger’s test wasn’t
useful, as we didn’t expect to have more than 10 studies for most of the outcomes.

Additional analyses
Although we would like to perform sensitivity analysis, firstly analyzing all included studies and then
only low risk of bias studies, this was impossible due to small number of included studies in total.
Regarding subgroup analysis, we tried to extract results separately for nuclear, cortical and posterior
subcapsular cataract and synthesize them if data would be sufficient.

RESULTS
Study selection
We searched nine electronic databases and hand-searched references of previous studies for any
missing studies. MEDLINE search returned 979 results at first place. Search strategy was limited to
humans, excluding studies on animals and cell series, to articles published only in English and the
study period was from 1997 to 2017. We also excluded studies that deal with more complicated
cases of cataract. For example, patients with significant comorbidities that could influence cataract
onset or progress, especially high glucose levels or very high cholesterol levels compared to general
population. We also considered funding issues and were ready to exclude any studies that funding
was suspicious to change the study’s results. Of 979 studies retrieved, 179 studies remained after
title and abstract evaluation and of these, 59 studies were left after full-text examination. From
Cochrane database, 142 papers were retrieved at first place, of which 105 reviews and 33 trials, and
after title and abstract evaluation 50 studies were included. Thirteen studies were left after full-text
examination. Three studies were excluded due to lack of accessibility. ScienceDirect search retrieved
10