Berries, tea, grapes and wine lower diabetes risk

A new study in healthy women suggests that consuming high levels of flavonoids, including compounds found in berries, tea, grapes, and wine, could potentially lower the risk of type 2 diabetes.

The study, published in the February issue of the Journal of Nutrition, indicates that greater intake of these dietary compounds is associated with lower insulin resistance and better blood glucose regulation. The researchers, led by Amy Jennings, PhD, from the department of nutrition, University of East Anglia, Norwich, United Kingdom, say their study is one of the first to examine consumption of different flavonoid subclasses and insulin resistance.

"We found that those who consumed plenty of anthocyanins and flavones had lower insulin resistance. So what we are seeing is that people who eat foods rich in these 2 compounds — such as berries, herbs, red grapes, wine — are less likely to develop the disease," said senior author Aedin Cassidy, PhD, also from the department of nutrition, University of East Anglia, in a statement.

Researchers also found that those who ate the most anthocyanins were least likely to suffer chronic inflammation, which is associated with diabetes, obesity, cardiovascular disease, and cancer. And those who consumed the most flavone compounds had improved levels of adiponectin, which helps regulate a number of metabolic processes, including glucose levels, Dr. Cassidy noted.

Importantly, the difference between the highest and lowest intakes of foods containing these compounds was small, consisting of just one portion of grapes or berries or a couple of oranges, say the authors. Also, the effects on insulin that were associated with high consumption of such foods was equivalent to those observed for other lifestyle factors, such as an hour's walk a day or low-fat diet for a year, they noted.

Nevertheless, Dr. Cassidy said it is not yet know exactly how much of one of these compounds is necessary to potentially reduce the risk of type 2 diabetes. "Dose–response trials are needed to ascertain optimal intakes for the potential reduction of type 2 diabetes risk," she and her colleagues stress.

One of the First Large Human Studies of Flavonoid Subclasses
Researchers note that a previous prospective study, published last year in the American Journal of Clinical Nutrition suggested a 15% reduction type 2 diabetes risk by comparing the highest and lowest quintiles of anthocyanin intake. However, the researchers emphasize that their current study is one of the first large-scale human trials to examine all subclasses of these powerful bioactive compounds to see how they might affect insulin resistance, blood glucose regulation, and inflammation.

The cross-sectional study was conducted in almost 2000 women aged 18 to 76 years from the Twins UK registry. Women who had high glucose levels were excluded. Participants completed a 131-item food-frequency questionnaire, from which flavonoid intakes were estimated using a United States Department of Agriculture database.

The researchers looked at the self-reported intake of 6 subclasses of flavonoids: flavanones, anthocyanins, flavan-3-ols, polymeric flavonoids, flavanols, and flavones.

In multivariable analyses, higher anthocyanin and flavone intakes were associated with significantly lower peripheral insulin resistance (homeostasis model assessment of insulin resistance; quintile 5 [Q5] to Q1 = 20.1, P-trend = .04 for anthocyanins and flavones), as a result of a decrease in insulin concentrations (Q5–Q1 = 20.7 mU/mL, P-trend = .02 anthocyanins; Q5–Q1 = 20.5 mU/mL, P-trend = .02 flavones).

Tea was the main source of overall flavonoid intake, with 4 foods contributing more than 10% of anthocyanin intake (grapes, pears, berries, and wine) and 3 foods making up more than 10% of flavone consumption (oranges, wine, and peppers).

Higher anthocyanin intake was also associated with lower C-reactive protein (hs-CRP) levels (Q5–Q1 = 20.3 mg/L, P-trend = .04), whereas those in the highest quintile of flavone intake had improved adiponectin levels (Q5–Q1 = 0.7 mg/L, P-trend = .01).

Higher intakes of both anthocyanins and flavones were associated with improvements in insulin resistance and hs-CRP, the researchers note.

No significant associations were observed for total or other flavonoid subclasses.

Findings Are Clinically Relevant, Easy to Achieve
Although these findings are from cross-sectional data and require confirmation, they are clinically relevant because of the 0.7-mU/mL difference in insulin observed between the top and bottom quintiles of anthocyanin intake, the researchers note.

The difference in anthocyanin intake between the top and bottom quintiles was 35 mg, which can be readily incorporated into the diet by consuming approximately one portion of grapes (78 g) or berries, such as strawberries (105 g), raspberries (90 g), blueberries (21 g), or blackberries (39 g).

Similarly, the difference in flavones between the top and bottom quintiles was 3.6 mg, equivalent to that found in approximately 2.5 oranges.

These results "are of public-health importance because the intakes associated with these findings are easily achievable through the habitual diet" and make a significant contribution to the knowledge base needed to refine the current fruit and vegetable dietary recommendations, the authors conclude.

Chinese herbal formula reduces progression to diabetes by 32%

A combination of 10 Chinese medicinal herbs in a capsule, known as Tianqi, reduced progression to type 2 diabetes in people with impaired glucose tolerance (IGT) in a randomized controlled trial in China.

This is the first study to show that a Chinese herbal medicine can "reduce the progression of prediabetes to diabetes," says study author Chun-Su Yuan, MD, PhD, from the Tang Center for Herbal Medicine Research at the University of Chicago, Illinois. Tianqi "could provide a new option for diabetes management, using herbal medicine alone or as an adjuvant to currently used therapies," he noted.

The results are published online January 16 in the Journal of Clinical Endocrinology & Metabolism, and Dr. Yuan said robust data such as these are needed to help promote the use of Chinese herbal medicine by physicians in different countries.

Most significantly, the researchers found that taking Tianqi reduced the risk for diabetes by almost a third (32.1%) compared with placebo, after adjustment for age and gender.

A Role for Chinese Medicine in Diabetes Prevention?
The findings show that the Chinese herbal medicine was comparable to some pharmaceuticals in reducing progression to type 2 diabetes, say the researchers. For example, the results seen with Tianqi were similar to those found with acarbose, at 25%, and metformin, at 31%.

"Although no direct comparison has been made between Tianqi and antidiabetic prescription drugs, our data indicate that this Chinese herbal medicine had similar effects to metformin," reported Dr. Yuan.

Asked to comment on whether diabetes prevention was regularly practiced in the United States, he remarked that unacceptable adverse effects limited regular use of conventional therapies in the prevention of type 2 diabetes, with reports showing that long-term administration of acarbose or metformin had often been associated with unfavorable gastrointestinal events.

Around 79 million individuals in the United States aged over 20 years have prediabetes, a state in which blood glucose levels are higher than normal but do not meet the diagnostic criteria for type 2 diabetes.

Dr. Yuan added that their data also show that after a period of cessation of the Tianqi treatment, the preventive effects on type 2 diabetes development remained significant. "Moreover, the safety profile of this herbal medicine is very good without obvious adverse effects," he commented.

Chinese Herb Combination and Study Design
The Chinese medicine comprises several herbs that have been shown to lower blood glucose levels after meals. The Tianqi capsule is manufactured by Heilongjiang Baoquan Pharmaceutical and consists of 10 Chinese herbal medicines: Astragali Radix, Coptidis Rhizoma, Trichosanthis Radix, Ligustri Lucidi Fructus, Dendrobii Caulis, Ginseng Radix, Lycii Cortex, Ecliptae Herba, Galla Chinensis, and Corni Fructus. The quality of these herbs and decoction preparation was in accordance with the Chinese Pharmacopoeia, the researchers note.

Dr. Yuan said the key herb in the combination was Huanglian (Coptidis Rhizoma). "The critical component of this herb is berberine, which has been reported to have good antidiabetic effects," he said in an interview. "Huanglian has been used traditionally in Chinese medicine in treating diabetic symptoms."

A total of 420 participants with IGT recruited from 11 research sites in China underwent double-blind randomization to receive either Tianqi (n=210) or a placebo (n=210) for 12 months. Participants had IGT with a 2-hour plasma glucose concentration of 7.8–11.1 mmol/L after a 75-g oral glucose tolerance test and fasting plasma glucose greater than 7.0 mmol/L.

Oral glucose tolerance tests were conducted every 3 months to assess the development of diabetes or restoration to normal glucose tolerance. In addition, all participants received similar lifestyle education.

The primary end point was the conversion of IGT to type 2 diabetes; body weight, body mass index, and adverse effects were monitored.

Need For More Controlled Trials of Chinese Medicinal Herbs
"Although the results of the present study need to be confirmed in future larger clinical trials, Tianqi holds promising potential as an effective and practical means to prevent type 2 diabetes, particularly in places in which herbal medicines are culturally accepted and widely used," say the authors.

They note that treating diabetes with Chinese herbal medicines is popular in China, particularly in rural areas.

"Our encouraging data should initiate further studies, both in China and in the West, to evaluate the role of Chinese herbal medicine in preventing and treating diabetes," said Dr. Yuan. "We are currently conducting several studies in this field."

High Blood Glucose Linked to Dementia Even Without Diabetes

Higher blood glucose levels, shy of the diabetic range, increase the risk for dementia in adults, new data from the longitudinal Adult Changes in Thought (ACT) study suggest.

"We considered blood glucose levels far into the normal (nondiabetic) range, and even there found an association between higher glucose levels and dementia risk," said Paul K. Crane, MD, associate professor of medicine at the University of Washington School of Medicine, Seattle, and affiliate investigator at the Group Health Research Institute in Seattle.

He said the results suggest that the "clinical determination of diabetes/not diabetes may miss important associations still there for people who are categorized as not having diabetes."

Diabetes Already a Risk Factor
Although diabetes is a risk factor for dementia, it's been unclear whether higher glucose levels, short of diabetes, also increase the risk for dementia. "Most studies that have investigated associations between glucose metabolism and the risk of dementia have focused on diabetes itself, and they have yielded inconsistent results," the authors note.

For their study, Dr. Crane and colleagues had access to 35,264 clinical measurements of glucose levels and 10,208 measurements of glycated hemoglobin from 839 men and 1228 women without dementia at baseline. Their mean age at baseline was 76 years.

During a median follow-up of 6.8 years, 524 participants developed dementia, including 74 with diabetes and 450 without.

Among participants without diabetes, higher average glucose levels within the preceding 5 years were associated with increased risk for dementia (P = .01). At a glucose level of 115 mg/dL, as compared with 100 mg/dL, the adjusted hazard ratio for dementia was 1.18 (95% confidence interval [CI], 1.04 - 1.33).

Higher average glucose levels were also related to an increased risk for dementia in those with diabetes (P = .002). At a glucose level of 190 mg/dL, as compared with 160 mg/dL, the adjusted hazard ratio for dementia risk was 1.40 (95% CI, 1.12 - 1.76). The findings were consistent across of variety of sensitivity analyses, the authors say.

The data suggest that higher levels of glucose may have "deleterious effects on the aging brain," the authors say.

The ACT study is "unique in being linked to clinical data from a healthcare delivery system along with research-quality dementia diagnoses," said Dr. Crane.

The findings are based on an average of 17 blood glucose measurements per person, "very rich data," senior author Eric B. Larson, MD, from Group Health Research Institute and University of Washington, added in a statement.

"This work is increasingly relevant because of the worldwide epidemics of dementia, obesity and diabetes," Dr. Crane said.

Modest Effect
Last month in a publication in JAMA Neurology, data from the Baltimore Longitudinal Study of Aging failed to show a link between glucose intolerance, diabetes or insulin resistance Alzheimer's disease or amyloid-β accumulation in the brain.

Richard J. O'Brien, MD, PhD, chair, Department of Neurology, Johns Hopkins Bayview Medical Center, Baltimore, Maryland, was an author on that study. Asked for his thoughts on the ACT data, Dr. O'Brien said it should be noted that "the effect is modest but they have a much larger sample size and we could have missed a small effect like that."

He said it should also be noted that the outcomes of the 2 studies are different. "Their outcome is dementia — they have no way of knowing about the specific Alzheimer's component to that outcome. It is certainly possible that vascular disease related to diabetes is the important factor increasing the rate of dementia. Dementia is often multifactorial — with Alzheimer's and vascular disease being the two biggest contributors," he noted. "The vascular component is difficult to detect clinically because it is often asymptomatic — only pathology can detect it."

Dr. Crane and colleagues say the underlying mechanisms of the association between elevated glucose levels and dementia need to be clarified in future studies.

"This was an observational study, and clinical practice really should not change based on the results of this study," Dr. Crane cautioned. "We do not know whether interventions designed to lower blood glucose would also lower dementia risk. That important question requires a different study design.

"However, from other research, including research from the ACT study, we know that exercise is associated with reduced dementia risk," he added. "Exercises such as walking or aerobic exercises in swimming pools are well tolerated and fun."

Chocolate reduces stroke risk in men

More evidence supports a link between chocolate consumption and reduced stroke risk.

A new study shows that men who consume the most chocolate have a 17% lower risk for stroke than those who consume the least. A meta-analysis included in the study showed an overall 19% decreased risk for stroke for the highest consumers of chocolate — male and female — compared with those who ate the least.

Although results "suggest that chocolate consumption is inversely associated with risk of stroke," further research is needed to confirm these findings before any recommendations about chocolate consumption can be given, according to the authors, led by Susanna C. Larsson, PhD, of the Division of Nutritional Epidemiology, National Institute for Health and Welfare, in Helsinki, Finland.

Chocolate Categories
The study included 37,103 men from the Cohort of Swedish Men, a prospective study that began in 1997, when all men aged 45 to 79 years residing in 2 counties in central Sweden were asked to provide detailed information on diet and other lifestyle factors. Researchers assessed chocolate consumption using a self-administered food frequency questionnaire that included 96 foods and beverages.

For chocolate consumption, respondents could choose from 8 categories, ranging from never, to eating this confection 3 or more times per day. To determine consumption in grams, researchers multiplied the frequency of consumption by 4 age-specific portion sizes to arrive at quartiles of chocolate consumption: less than 12.0 g/week (0); 12.0 to 19.5 g/week (12.5); 19.8 to 51.3 g/week (38.4); and at least 51.8 g/week (62.9).

By linking the study population to the Swedish Hospital Discharge Register, researchers identified incident cases of stroke. They classified stroke events as cerebral infarction, intracerebral hemorrhage, subarachnoid hemorrhage, and unspecified stroke.

From 1998 to 2008, 1995 cases of first stroke were reported, including 1511 cerebral infarctions, 321 hemorrhagic strokes, and 163 unspecified strokes. Compared with men who had the lowest chocolate consumption, those who ate the most chocolate were younger on average but were less likely to be current smokers. They tended to be slightly leaner, and to consume more alcohol, red meat, fruits, and vegetables, but less fish.

Adjusted Risk
After adjustment for several factors, including age, education, smoking status, body mass index, physical activity, aspirin use, history of hypertension, and atrial fibrillation, high chocolate consumption was associated with a statistically significant lower risk for total stroke.

Compared with men in the lowest quartile of consumption (0 g/week), those in the highest category (62.9 g/week) had a reduced risk for stroke of 17%. Results were similar for cerebral infarction and for hemorrhagic stroke.

Age-standardized incidence rates of stroke were 85 per 100,000 person-years for those consuming the least chocolate, and 73 per 100,000 person-years for those in the highest quartile.

In an analysis stratified by history of hypertension, an inverse relation between chocolate consumption and risk for total stroke was observed in men without hypertension, but not in men with a history of hypertension.

It is possible that the blood pressure–lowering effect of chocolate consumption helps explain the association in men without a history of hypertension; such men may have had normal blood pressure at baseline as the result of treatment for hypertension, say the authors.

The researchers also carried out a meta-analysis that included results from the current study as well as from 4 other prospective studies assessing the association between chocolate consumption and stroke risk. In this analysis, which included 4260 stroke cases, the overall RR for stroke for the highest vs lowest category of chocolate consumption was 0.81, or a 19% reduction in risk.

Chocolate is a rich source of flavonoids that may protect against cardiovascular disease through antioxidant, antiplatelet, and anti-inflammatory effects. Flavonoids may also lower blood pressure, increase high-density lipoprotein (HDL) cholesterol, and improve endothelial function. Chocolate contains antioxidant-rich caffeine, albeit in low amounts.

One of the limitations of the study is that chocolate consumption was self-reported and was measured only at a single time point. Another is that information on the type of chocolate consumed was not available. As a result, any association with dark chocolate, previously linked to health benefits, could not be examined, but in general, about 90% of chocolate consumed in Sweden is milk chocolate, containing about 30% cocoa solids, they write.

The authors stress also that chocolate is high in sugar, saturated fat, and calories and should be consumed in moderation.

Confirms Observations
Reached for a comment, Gustavo Saposnik, MD, director of the Stroke Outcomes Research Center, and associate professor and clinician scientist in the Departments of Medicine (Neurology) and Health Policy, Management, and Evaluation, at St. Michael's Hospital, University of Toronto, in Ontario, Canada, said the present study confirms previous observations, including those of his own systematic review, that eating chocolate is associated with lower risk for stroke.

Results of his and his colleagues' review, presented at the American Academy of Neurology Meeting in 2010 showed that in one study, a 22% reduction in stroke risk was noted for participants who had a single serving of chocolate per week, and in another, a 46% reduction in stroke mortality resulted from weekly consumption of flavonoids in 50 g of chocolate vs no consumption. However, the number of studies looking at the relationship between chocolate consumption and stroke risk, and included in the analysis, was small.

Dr. Saposnik found it interesting that in the current study, even very low chocolate consumption (median, 62.9 g/week) was associated with a 17% stroke reduction.

He noted that the study was limited by the use of questionnaires that were not specifically intended to target the association between chocolate and risk for stroke. This, he said, could introduce potential risk for misclassification. As well, Dr. Saposnik said that the study provided limited information about the composition of the chocolate consumed.

For Dr. Saposnik, the strength of the study was the meta-analysis, which included results from 5 studies.

Cocoa, Even With Few Flavonoids, Boosts Cognition

Drinking cocoa, whether rich in flavonoids or not, appears to boost the effect of blood flow on neuronal activity in the brain, known as neurovascular coupling (NVC).

A new study shows not only that drinking flavonoid-rich or flavonoid-poor cocoa improves NVC but also that higher NVC is associated with better cognitive performance and greater cerebral white matter structural integrity in elderly patients with vascular risk factors.

As researchers search for ways to detect dementia at the earliest possible stage, the study results could pave the way for using NVC as a biomarker for vascular function in those at high risk for dementia, said lead author Farzaneh A. Sorond, MD, PhD, Department of Neurology, Stroke Division, Brigham and Women's Hospital, Boston, Massachusetts.

"Our study shows that NVC is modifiable and can be enhanced with cocoa consumption," said Dr. Sorond.

The study is published online August 7 in Neurology.

Tight Correlation
The double-blind proof-of-concept study included 60 community-dwelling participants, mean age 72.9 years. About 90% of the participants were hypertensive, but with well-controlled blood pressure, and half had diabetes mellitus type 2 with reasonably good control. Three quarters were overweight or obese.

Participants were randomly assigned to 2 cups a day of cocoa rich in flavonoids (609 mg per serving) or cocoa with little flavonoids (13 mg per serving). Diets were adjusted to incorporate the cocoa, each cup of which contained 100 calories. Participants were also asked to abstain from eating chocolate.

Researchers measured cerebral blood flow in these participants using transcranial Doppler ultrasonography. Among other things, they documented changes in the middle cerebral artery and blood flow velocity at rest and in response to cognitive tasks (NVC).

The study showed that NVC was tightly correlated with cognition; scores for Trail making Test B, a test of executive function, were significantly better in those with intact NVC (89 seconds vs 167 seconds; P = .002). Participants with intact NVC also had significantly better performance on the 2-Back Task, a test for both attention and memory (82% vs 75%; P = .02).

"The higher you increase your blood flow during a cognitive task, the better your cognitive performance," commented Dr. Sorond, adding that this is something that has never been shown before.

NVC was also correlated with cerebral white matter structural integrity. Higher NVC was associated with overall less white matter macro- and micro-structural damage. In general, those with intact NVC had a greater volume of normal white matter and smaller volume of white matter hyperintensities, higher fractional anisotropy, and lower mean diffusivity in the normal white matter and WMH.

Therapeutic Target
These results suggest that NVC could be an important therapeutic target. But before NVC can be considered a biomarker, it has to be shown to be changeable, and the clinical importance of the modification must be shown.

To that end, the study authors opted to use cocoa. They could have chosen many other potential modifiers but chose cocoa because the literature has shown the beneficial effects of cocoa on brain health and also because it's something that many people enjoy, said Dr. Sorond.

The study found that blood pressure, blood flow, and change in NVC were not significantly different between the 2 cocoa groups. In the combined cocoa groups, 30-day blood pressures were not significantly different from baseline (P > .5).

In contrast, response to cocoa differed significantly depending on NVC status. Cocoa had a significant effect on NVC in those with impaired (<5%) coupling at baseline. Of those with impaired NVC, 89% responded to 30 days of cocoa consumption and increased NVC compared with only 36% of those with intact NVC (P = .0002). In those with impaired baseline coupling, cocoa consumption was associated with an 8.3% (P < .0001) increase in NVC at 30 days.

Gut Bacteria Affect Colorectal Cancer Risk

The gut microbiome of patients with colorectal cancer (CRC) was less diverse than that of matched patients without cancer, and the presence of some taxa was associated with increased CRC risk, according to research published online December 6 in the Journal of the National Cancer Institute.

"Because of the potentially modifiable nature of the gut bacteria, our findings may have implications for CRC prevention," write Jiyoung Ahn, PhD, assistant professor of population health and a member of the NYU Cancer Institute, New York University School of Medicine, New York City, and colleagues.

The results came from an analysis of fecal bacterial DNA from 47 patients with CRC and 94 control participants matched to the CRC group by sex and body mass index. The investigators amplified 794,217 16S rRNA genes and then classified the sequences taxonomically.

They note that this is the first study to compare the gut microbiomes of people with and without CRC that included multiple comparisons of bacteria while controlling for possible confounders.

The researchers found decreased microbiome community diversity in patients with CRC , compared with that of healthy participants(P= .02). In an analysis by taxa, patients with CRC had lower relative abundances of Clostridia, at 68.6% compared with 77.8% in people without CRC. In contrast, patients with CRC carried a higher relative abundance of Fusobacterium (31.9% vs 11.7% for control patients).

A higher relative abundance of Fusobacterium was associated with increased CRC risk (multivariable odds ratio [OR], 4.11; 95% confidence interval [CI], 1.62 - 10.47), after adjusting for age, sex, body mass index, race, smoking, and sequencing batch.

Actinobacteria Atopobium (OR, 14.36; 95% CI, 2.78 - 74.30; P < .001) and the Bacteriodetes Porphyromonas(OR, 5.17; 95% CI, 1.75 - 15.25; P = .001) were also associated with CRC risk. The Gram-positiveAtopobium is associated with Crohn's disease and is reported to inhibit colon cancer apoptosis in vitro.Polyphyromonas, which is often found in the mouth and gastrointestinal tract, is associated with periodontal disease.

Patients with CRC tended to have more Bacteroidetes phylum bacteria (16.2% relative abundance vs 9.9% for control participants) and fewer Firmicutes (74.0% for patients with CRC compared with 80.3% for control participants). The depletion of Firmicutes was highest for the class Clostridia (68.6% for patients with CRC vs 77.8% for control participants; P = .005; false discovery rate–adjusted P < .05). Among the depleted taxa was Coprococcus in the Clostridia family Lachnospiracea. Coprococcus is responsible for the efficient fermentation of dietary fiber and other complex carbohydrates into butyrate, a major colonic metabolite that may inhibit colonic inflammation and carcinogenesis, the authors write.

In an additional assay using quantitative polymerase chain reaction for Porphyromonas and Fusobacterium, the association between CRC and these taxa remained significant. In that study, the OR for CRC with the presence of Porphyromonas was 1.44 (P = .05), and the OR for CRC with the presence of Fusobacteriumwas 1.44 (P = 0.01).

Daily Multivitamin Lowers Risk for Cataracts

Taking a multivitamin supplement daily may stave off development of cataracts in middle-aged and older men, according to an article published online November 21 in Ophthalmology. However, there appeared to be no association between daily vitamins and visually significant age-related macular degeneration (AMD).

William G. Christen, ScD, from the Division of Cardiovascular Disease, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, and colleagues analyzed results of the Physicians' Health Study (PHS) II, in which cataract and AMD were secondary endpoints. PHS II results for cardiovascular and cancer risk were published earlier.

The eye disease component of PHS II involved 11,497 male physicians randomly assigned to take a daily multivitamin (Centrum Silver, Pfizer) (n = 5736) or placebo (n = 5761) in the cataract part of the study, and 14,233 men randomly assigned to a multivitamin (n = 7111) or placebo (n = 7122) in the AMD part of the study. Study participants reported no cataract or AMD incidence at baseline.

The physicians completed annual questionnaires between 1997 and 2011 and provided information about pill-taking adherence, possible adverse events, risk factors, and occurrence of any new cataract or AMD cases. Researchers followed the participants for a mean of 11.2 years, confirming the occurrence of 1817 cataract and 1337 cataract extractions, along with 538 cases of AMD. Of the AMD cases, 281 were visually significant and 144 were advanced.

The researchers found a 9% lower risk for cataract among participants in the multivitamin group compared with the placebo group (hazard ratio [HR], 0.91; 95% confidence interval [CI], 0.83 - 0.99; P = .04). The benefit was greater in older men.

They found no statistically significant difference for visually significant AMD, although there was a trend for increased risk among the multivitamin group (152 vs 129 cases; HR, 1.19; 95% CI, 0.94 - 1.50; P = .15). However, the increased risk associated with vitamin use reached statistical significance when the investigators analyzed any type of AMD, with or without visual loss (294 vs 244 cases; HR, 1.22; 95% CI, 1.03 - 1.44; P = .02).

Small Percentage, Large Impact
"Cataract is a very common source of ocular morbidity in US adults. If our findings are real, that is, if multivitamins really do reduce the risk of cataract, even by a modest 10%, this rather small reduction would nonetheless have a large public health impact," Dr. Christen wrote.

Emily Y. Chew, MD, deputy clinical director at the National Eye Institute in Bethesda, Maryland, agrees. "It's akin to cardiovascular disease. Cardiovascular disease is very common, so even a 10% reduction is a huge relief of the burden of disease," she said. "Cataract is the most common surgery that Medicare pays for, and it's going to get worse. Anything we can do to reduce it is a major feat."

Dr. Chew was a leader in another prominent research project, Age-Related Eye Disease Study, or AREDS, which found similar results for multivitamin use. "We found that Centrum was protective for cataracts, just like they did here, but not for macular degeneration," she said. She is now a leader on the AREDS2.

The new data "at least suggests to the general clinician or the general ophthalmologist, maybe even the primary care physician, that the notion of taking a multivitamin is probably something that we should encourage in more patients than not," Matthew J. Welch, MD, a vitreoretinal surgeon at the Loyola University of Chicago Stritch School of Medicine in Maywood, Illinois, said. "You're probably doing yourself some good — you're definitely not doing yourself any harm — by taking a multivitamin, and maybe that does stave off your cataract progression."

Why Not AMD?
Researchers still have to tease out why no significant benefit was shown for AMD in PHS II.

"A true lack of effect is one possibility," Dr. Christen said. "Another possibility is that our finding of no significant benefit or harm may simply be due to the play of chance. Other possibilities include inadequate dosages or inadequate follow-time. It may also be that a different combination of nutrients is required."

The newly published results differ from the results of AREDS, in which participants taking high-dose antioxidant combination of vitamin C, vitamin E, β-carotene, and zinc showed a slower progression to AMD. But in PHS II, the doses were only recommended daily allowances, and AREDS participants were higher-risk individuals.

One issue with the PHS II study, according to Dr. Welch, is that it was designed primarily for finding cardiovascular and cancer risks and that it was based on self-reported information from physicians.

"You're dealing with a group of individuals that view their time as highly valuable and aren't necessarily going to be running to the doctor every time they notice something," he explained. "I think probably if this had been designed specifically to evaluate the eye, they may have found more of a difference [for AMD]. There may be more cataracts and AMD found if individuals were systematically screened."

Dr. Welch said he thinks unknown benefits from dietary supplements probably exist: "It's just that it's hard to tease out. It's not as easy as it is in a drug trial."

Depression Linked to Low Zinc Levels in Blood

People who are depressed have lower concentrations of zinc in their peripheral blood compared with nondepressed individuals, a meta-analysis suggests.

"The pathophysiological relationships between zinc status and depression, and the potential benefits of zinc supplementation in depressed patients, warrant further investigation," write Walter Swardfager, PhD, from the Sunnybrook Research Institute, University of Toronto, Canada, and colleagues.

Their findings were published in the December 15 issue of Biological Psychiatry.

Link to Zinc Deficiency
"A growing body of evidence demonstrates that experimental zinc deficiency can induce depressive-like behavior in animals, which can be effectively reversed by zinc supplementation," the authors write. Furthermore, preliminary clinical trials have suggested that zinc added to antidepressant therapy may produce more rapid or more effective improvement in depressive symptoms.

Many, although not all, studies that have measured zinc concentrations of peripheral blood in depressed and nondepressed individuals during the past several decades have suggested that depression might be associated with lower zinc concentrations in various population samples, the researchers note.

The aim of this meta-analysis was to determine whether the clinical evidence collectively supports lower zinc concentrations in depressed patients compared with nondepressed individuals.

The analysis included 17 studies that measured peripheral blood–zinc concentrations in 1643 depressed patients and 804 control participants. Ten of these studies reported on psychiatric inpatients, and 7 reported on community samples. Of the participants, 34.4% were male; the mean age was 37.7 years.

The researchers found that mean peripheral blood–zinc concentrations were lower by approximately 1.85 µmol/L in depressed individuals compared with control participants (95% confidence interval [CI], -2.51 to -1.19; P < .00001).

More severe depression was associated with greater differences in zinc levels between depressed and control participants (P = .026).

Further, the authors report that effect sizes were larger in studies of inpatients (weighted mean difference [WMD], -2.543; 95% CI, -3.522 to -1.564; P < .0001) vs community samples (WMD, -.943; 95% CI, -1.563 to -.323; P = .003) and in studies with higher methodologic quality (WMD, -2.354; 95% CI, -2.901 to -1.807; P < .0001).

Biologically Plausible
"Although association studies cannot determine the direction of causation, a causal association between zinc status and depression is biologically plausible," the authors write.

"Zinc has antioxidant properties, helps to maintain endocrine homeostasis and immune function, and plays multiple roles in regulating the hippocampal and cortical glutamatergic circuits that subserve affective regulation and cognitive function. Thus, changes in zinc homeostasis might compromise neuroplasticity and contribute to long-term neuropsychological and psychiatric decline," they write.

Zinc also plays an important role in immune function. Lower serum levels of zinc have been associated with disturbances in fatty acid metabolism and in serum lipid levels, which might affect brain function and vascular health. Lower zinc levels have also been associated cardiovascular disease, a common comorbidity of major depressive disorder (MDD), the authors note.

The investigators point out limitations of their study. The quality and risk of bias "were uneven among studies included in this meta-analysis," they write.

Not all studies reported demographic data "sufficiently to be included in investigations of heterogeneity, the use of antidepressants and other concomitant medications were not consistently reported, and data on diet and alcohol use were often not reported," the authors write.

Limitations notwithstanding, the authors conclude that their results suggest that depression is associated with reduced concentrations of zinc in peripheral blood.

"The findings suggest the need to further investigate potential roles of zinc in the pathophysiology of depression, the potential utility of zinc and related biomarkers in monitoring MDD and its clinical sequelae, and potential benefits of zinc supplementation in MDD patients," they conclude.

The study was supported by the Ontario Mental Health Foundation. Dr. Swardfager was supported by fellowships from the Heart and Stroke Foundation Center for Stroke Recovery and the Toronto Rehabilitation Institute. Dr. Swardfager and the other authors report no relevant financial relationships.

Vitamin E May Slow Functional Decline in Mild Alzheimer's

For patients with mild to moderate Alzheimer disease (AD), 2000 IU of vitamin E daily slows functional decline and reduces caregiver burden, according to results of a large randomized controlled trial.

"Previous guidelines had recommended vitamin E for those with moderately severe dementia, but hesitation grew based on several unrelated studies," said Mary Sano, PhD, director, Alzheimer Disease Research Center, and professor, Department of Psychiatry, Mount Sinai School of Medicine in New York.

"Our study indicates that vitamin E could be recommended to improve functional outcomes, such as activities of daily living, for all levels of Alzheimer's disease starting with mild stages and gives good confidence that it is safe," she said.

No benefit was seen in this trial, however, with memantine or the combination of memantine and vitamin E.

The study is published in the January 1, 2014, issue of JAMA.

"Meaningful" Clinical Benefit
The study, conducted at 14 Veterans Affairs medical centers, assessed the effect of vitamin E alone or with memantine on functional decline in 613 patients (97% men) with mild to moderate AD. At baseline, all of them were receiving an acetylcholinesterase inhibitor — most often donepezil (65%) and galantamine (32%).

By random assignment, 152 patients received 2000 IU of α-tocopherol per day, 155 received 20 mg of memantine per day, 154 received the combination of vitamin E and memantine, and 152 received placebo. They were followed for a mean of 2.3 years.

A total of 256 patients (42%) did not complete the trial. The most common reasons were death (128 patients [50%]) and withdrawal of consent (77 patients [30%]). Data from 561 patients were analyzed (140 in the vitamin E group, 142 in the memantine group, 139 in the combination group, and 140 in the placebo group).

Baseline characteristics were similar in the 4 groups. The mean Alzheimer's Disease Cooperative Study/Activities of Daily Living (ADCS-ADL) Inventory score and Mini-Mental State Examination total score at baseline were 56.8 and 21.0, respectively.

During follow-up, patients taking vitamin E alone had significantly slower functional decline than those taking placebo, experiencing 3.15 units less decline on the ADCS-ADL Inventory.

The annual rate of decline in ADLs was reduced by 19% (P = .03) with vitamin E compared with placebo. This treatment effect translates into a "clinically meaningful" delay in progression of functional decline in the vitamin E group of 6.2 months (95% confidence interval [CI], 5.4 - 7.4 months) compared with placebo, the investigators say.

In addition, on the Caregiver Activity Scale, caregiver time increased least in the vitamin E group compared with the other 3 groups. The delay in functional decline coupled with the reduced caregiver burden seen with vitamin E 2000 IU/d could have a "major effect on informal and direct medical care costs," the investigators say.

They note that the magnitude of functional decline in the placebo group (roughly 3 units more on the ADCS-ADL Inventory than in the vitamin E group) "could translate into either the complete loss of being able to dress or bath independently, for example, or losing independence on any 3 different ADLs. Because vitamin E is inexpensive, it is likely these benefits are cost-effective as alpha tocopherol improves functional outcomes and decreases caregiver burden," they write.

Neither memantine nor the combination of vitamin E and memantine showed clinical benefit in the study, the authors report.

"The lack of effectiveness of memantine in the current study is consistent with the negative findings reported in previous studies of AD patients with mild dementia," the investigators say. The findings "reinforce current VA memantine treatment guidelines that restrict the use of memantine to patients with moderately severe AD," they add.

The "Best" in AD Trials
The results for memantine are "not encouraging," Denis A. Evans, MD, Martha Clare Morris, ScD, and Kumar Bharat Rajan, PhD, from Rush University Medical Center in Chicago, Illinois, note in an accompanying editorial.

Memantine is approved by the US Food and Drug Administration for use in moderate to severe AD, they point out, and use in individuals with milder AD "may be widespread despite little evidence suggesting the agent is beneficial at this level of disease severity." This study of memantine in mild to moderate AD "does not provide any new data to support its use" in this group, the editorialists say.

All-cause mortality and safety data from the study showed a difference only in the serious adverse event of "infections or infestations" with greater frequencies in the memantine (31 events in 23 patients) and vitamin E/memantine combination groups (44 events in 31 patients) compared with placebo (13 events in 11 patients).

Mortality did not increase significantly with vitamin E, a finding that contrasts with a 2005 meta-analysis of vitamin E (Ann Intern Med. 2005;142:37-46), which showed that high-dose vitamin E (400 IU/d or greater) may increase the risk for all-cause mortality, the investigators point out.

In their editorial, Dr. Evans and colleagues say many features of this trial "reflect the best in trials of AD therapy, especially its size, duration, and separation from commercial motivation. However, as with almost all previous AD trials, the therapeutic effect seen was modest and more relevant to AD symptoms and consequences than to reversal of the disease process."

"Considering the difficulties inherent in trying to treat rather than prevent very high-prevalence diseases and the limitations thus far of the therapeutic efforts for people with AD, shifting to more emphasis on prevention seems warranted," they conclude.

The study was supported by the Veterans Affairs Cooperative Studies Program, Office of Research and Development, Clinical Science R&D. Forest Research Institute, a Division of Forest Laboratories, donated the memantine and matching placebo tablets. DSM Nutritional Products donated the DL-α-tocopheryl acetate oil and funding for the purchase of the soybean oil from Arista Industries. A complete list of author disclosures is listed with the original article.