Clinical Briefs

With Comments from Russell H. Greenfield, MD

Dr. Greenfield is Clinical Assistant Professor, School of Medicine, University of North Carolina, Chapel Hill, NC; and Visiting Assistant Professor, University of Arizona, College of Medicine, Tucson, AZ.

Pills in Peds: Supplement Use in Children

Source: Picciano MF, et al: Dietary supplement use among infants, children, and adolescents in the United States, 1999-2002. Arch Pediatr Adolesc Med. 2007;161:978-985.

Goal: To describe the extent of dietary supplement (DS) use in a nationally representative sample of US children.

Study Design: Analysis of data from the 1999-2002 National Health and Nutrition Examination Survey (NHANES, an annual cross-sectional, population-based examination survey), representing the latest estimates of DS use available.

Subjects: Children from birth to 18 years who participated in NHANES (n = 10,136).

Methods: Parents of children younger than age 16 years, or participants aged 16-18 years, were asked about their use of DS in the 30 days prior by trained interviewers. If DS use was confirmed, additional questions regarding duration, dosage, and frequency of use were asked and specific supplement data (including name and manufacturer) recorded. Data from the 1999-2000 and 2001-2002 NHANES data releases were combined, and children were categorized by age into one of 5 groups (< 1 year, 1-3 years, 4-8 years, 9-13 years, and 14-18 years).

Results: For the time period 1999-2002, 31.8% of children used DS, with no significant gender differences. The highest rate of use was in 4-8-years-olds (48.5%), with a higher rate of use in 14-18-year-olds (25.7%) than in infants (11.9%). DS use was also higher in non-Hispanic white and Mexican American families, among children with lower body mass index, in families with higher levels of income, in smoke-free homes, and where screen time (television, computer, video games) was limited. The most common type of agent used was a multivitamin and mineral supplement (18.3%). Prevalence of DS use increased from low levels during infancy and early childhood to their highest levels at age 5 years, followed by a steady decline to 20% among 15-year-olds, and a bump in prevalence of use among older teens (28%). Use of botanical supplements in this sample of children was but 0.8%, with very few children under age 3 years being exposed to herbal medicine. Iron was used by 19% of children, and fluoride by 3.0%. The majority of children (83.9%) took only one DS.

Conclusion: Nearly one-third of American children take dietary supplements on a regular basis.

Study strengths: Multivariate analysis data taken directly from participants in their homes and from supplement labels.

Study weaknesses: Cross-sectional design.

Of note: The majority of US adults take dietary supplements (57% of women and 47% of men reported using a DS in the past 30 days in the 1999-2000 NHANES); DS use has remained stable or decreased in children (use has declined especially in young children), while it has increased in adults; it is possible that DS use is higher among children with lower BMI because parents feel they are more in need of nutrients.

We knew that: The American Academy of Pediatrics recommends nutritional supplementation for children at risk for deficiency states (to prevent rickets, for example); the prevalence of DS use in adults increases linearly from 19-65 years of age, with highest levels in those > 65.

Comments: Picciano and colleagues of this well done study plead their case that in future evaluations of the nutritional status of children, DS use be taken into serious consideration. Indeed, this can now only be considered a necessity. Over 30% of American children consistently get some of their daily nutrients from DS. It is important, if not ironic, to note that the highest rates of DS use in children can be found among those whose families have the highest incomes, which should translate into exposure to the healthiest diets. Indeed, it is clearly preferable that children get their nutrients from healthy fare, but for many, that is simply not possible. Yet the children who might benefit most from DS, often do not have access to them, another inequity that should be addressed. For those practitioners who care for children, it is important to ask about the use of DS and to review the contents of DS with families in order to optimize potential benefits and minimize potential untoward effects.

What to do with this article: Keep a hard copy in your file cabinet.

Eye C? Dietary Antioxidants and ARMD

Source: Chong, E W-T, et al: Dietary antioxidants and primary prevention of age related macular degeneration: systematic review and meta-analysis. BMJ. 2007 doi:10.1136/bmj.39350.500428.47

Goal: To determine whether or not a range of dietary antioxidants are effective for the primary prevention of age-related macular degeneration (ARMD).

Study Design: Systematic review and meta-analysis.

Source material: Seven databases without limits on language or year of publication, and references from reviews and articles. Only randomized clinical trials (RCTs) and prospective cohort studies (PCS) were considered for inclusion in the analysis.

Methods: Two reviewers independently searched the databases and selected the studies using standardized criteria. Disagreement between the reviewers was resolved through discussion with senior investigators. Out of 4,192 abstracts screened, a total of 12 studies (9 PCS with data on 149,203 people including 1,878 incident cases of ARMD, and 3 RCTs) met selection criteria, and data extraction and study quality evaluation were independently reviewed, also using standardized criteria. Results of meta-analysis were pooled quantitatively.

Results: All trials were performed in the United States or other Western countries. Participants in the cohort studies were generally over age 49 years, but 2 trials included subjects in their early forties, and subjects were followed for a mean of 9 years. Pooled results from PCS showed that dietary intake of vitamins A, C and E, as well as zinc, lutein, zeaxanthin, alpha carotene, beta carotene, beta cryptoxanthin, and lycopene, has little or no benefit with respect to primary prevention of ARMD. Results from the 3 RCTs that followed people for 4-12 years did not support the use of vitamin E or beta carotene for the primary prevention of ARMD.

Conclusion: Sufficient evidence is lacking for a primary preventive effect of dietary antioxidants, including dietary supplements, against ARMD. The only widely accepted modifiable risk factor for the primary prevention of ARMD remains smoking.

Study strengths: Authors of published trials were contacted if salient data were missing; authors are frank regarding shortcomings of results.

Study weaknesses: One of 9 PCS, and one of the 3 RCTs, were published in abstract only; relatively short periods of follow-up, problems inherent in the use of food frequency questionnaires FFQ (for example, in the PCS trials the FFQ was administered but once, at baseline); few RCTs; mainly well-nourished subjects; as the authors state, meta-analyses of observational data offer the potential for even bias than meta-analyses of RCTs; unable to evaluate potential antioxidant synergism.

Of note: The AREDS trial, a secondary prevention study, showed that patients with intermediate forms of ARMD treated with high-dose zinc, beta carotene, vitamin C, and vitamin E had a 28% reduction in risk of disease progression compared with placebo; the only consistently reported risk factors for ARMD include older age, genetic markers, and cigarette smoking; in this review there was significant heterogeneity among the PCS with respect to specific types of antioxidants under investigation (not all studies contributed to the meta-analysis of each antioxidant), but little heterogeneity beyond that; the reviewers specifically excluded studies of people with early ARMD (secondary prevention trials); some of the diet data in the PCS were obtained prior to 1988.

We knew that: ARMD is the leading cause of significant central vision loss in people over age 50 years living in industrialized nations; early ARMD is characterized by changes in retinal pigmentation and the presence of drusen (yellow deposits), while with late ARMD, there can be new vessel formation that results in sub-retinal bleeding ("wet" ARMD) or macular atrophy ("dry" ARMD); treatment is available for a small proportion of people with "wet" ARMD, but no effective treatment yet exists for "dry" ARMD; the retina is highly susceptible to oxidative damage due to constant light exposure and high oxygen concentrations; preventive treatment against ARMD with antioxidants is plausible because they (especially carotenoids) may help reduce photo-oxidative damage from blue light.

Comments: As Chong and colleagues point out, there is a common perception that a diet rich in antioxidants can protect against ARMD. The results of this meta-analysis made its way into the lay media and dampened the hopes of many people; however, a thorough review of the data reminds us that there's more to research than headlines.

Even the best of meta-analyses are plagued with problems that make interpretation, let alone evaluation, of clinical management difficult at best. Chong et al are to be commended for their attempt to clarify a challenging issue, but perhaps even more so for being forthright about the limitations of their work. Beyond what is printed in the way of limitations, however, there are other problems, not the least of which is duration of follow-up in the studies analyzed. The majority of people enrolled in the trials were over 50, but some had not yet reached age 45. Consider that the average follow-up was 9 years in the PCS and 4-6 years in the RCTs. This is quite problematic, as the average age at which the diagnosis of ARMD is made is over age 65 years. In addition, because ARMD develops and progresses slowly, and may be essentially asymptomatic early on, it is very difficult to identify unless people seek regular ophthalmologic evaluation.

The data supporting antioxidant therapy for established ARMD seem quite strong. Results of this trial, focusing on primary prevention of ARMD, were not positive, but the methodology employed negates the findings for practical purposes. There remains promise that antioxidant therapy may help prevent ARMD, but this study does not contribute significantly to the discourse. Better-designed, prospective trials are needed.

What to do with this article: Remember that you read the abstract.

A Fit Mind: Exercise and Depression

Source: Blumenthal JA, et al: Exercise and pharmacotherapy in the treatment of major depressive disorder. Psychosom Med. 2007;69:587-596.

Goal: To evaluate the effects of home or supervised group aerobic exercise on major depressive disorder (MDD) as compared to use of sertraline (Zoloft) or placebo.

Study Design: Prospective, randomized, parallel group, controlled study performed over 16 weeks.

Subjects: Adults over age 40 years (mean age = 53 years) with MDD (n = 202, 153 women).

Methods: Potential subjects were screened using the Beck Depression Inventory II (BDI). Those who scored 12 or above on the BDI and subsequently met diagnostic criteria for MDD, were recruited through public media. A structured clinical interview for depression was performed by a licensed clinical psychologist, and the Hamilton Depression Rating Scale (HAM-D) employed at baseline and at 16 weeks. Participants were randomly assigned in roughly equal proportion to one of 4 groups: supervised exercise in a group setting (3 supervised sessions per week); home-based exercise (initial home visit to establish training routine, visit from exercise physiologist at end of months 1 and 2, and telephone calls at weeks 1 to 4 and biweekly thereafter); sertraline 50-200 mg daily (starting dose of 50 mg/day with increases contingent upon response and absence of side effects); or placebo pill. The BDI was administered by phone weekly for the first 4 weeks, then biweekly until the end of the trial. Clinic visits occurred at baseline and then at weeks, 2, 4, 8, 12, and 16. Graded treadmill exercise testing was conducted before treatment and again at study's end to document subjects' fitness levels. The primary endpoint of interest was remission of MDD (no diagnosis of MDD and a HAM-D score < 8).

Results: At trial's end, a total of 41% of participants no longer met criteria for MDD and had a HAM-D score < 8. Those who received active treatments had higher remission rates than those in the placebo group: supervised exercise 45%; home-based exercise 40%; medication 47%; and placebo 31% (P = .057). The adjusted odds ratio for remission was 2.6 when comparing active treatment with placebo. All treatment groups had lower HAM-D scores at the end of the study, but scores were not significantly different between placebo and active groups. Aerobic capacity improved more in the group exercise arm than for those exercising at home, with both groups' aerobic capacity improving more than either "pill" group. Dropout rates were as follows: 14% sertraline. 29% placebo group, 6% home exercise, and 20% group exercise. Results suggest that treatment affected subjects in similar fashion regardless of severity of depression.

Conclusion: Exercise is generally as effective as antidepressant medication, and both are more effective than placebo, for the treatment of MDD in middle-aged and older adults.

Study strengths: Comparison of home-based vs group exercise to try to control for effects of non-specific social support; eligibility criteria and close follow-up established to ensure subject safety in the event they received placebo; intention-to-treat analysis; blinding of psychologists to treatment condition.

Study weaknesses: Significant amount of interaction with staff even in home study group; compliance with medication or placebo evaluated by pill count; limited generalizability (subjects were mainly Caucasian women over age 40 years with mild depression, and all were volunteers); attrition rates.

Of note: MDD is the 4th leading cause of disability-adjusted life-years worldwide; findings of prior studies of exercise for people with MDD have largely been compromised due to methodological flaws and confounding factors; Blumenthal and colleagues note that they chose sertraline as standard treatment for MDD for purposes of comparison due to its "widespread usage and proven efficacy, tolerability, and low toxicity" the fitness prescription was the same for both exercise groups, the only difference being the setting; for inclusion in the study, participants had to have been relatively sedentary prior to randomization; only the research pharmacist was aware of which people were using sertraline vs. placebo; if difficulties with the exercise prescription were being experienced by participants, an additional 2 home visits by staff was permitted; 40% of subjects had a history of recurrent depression; the majority of participants in the trial enrolled with the hope they would receive exercise treatment for treatment of their depression.

We knew that: The lifetime prevalence of MDD is 15-20%; it is estimated that MDD will be second only to cardiovascular disease as a worldwide cause of illness burden by the year 2020; in studies of MDD, the placebo group often experiences marked improvement (response rates of 30-50%), especially during the first week of treatment; no single treatment is effective for everyone with MDD; simple reduction in the symptoms of MDD does not result in as good a prognosis or as improved levels of daily functioning as does remission; HAM-D scores < 8 are associated with low relapse rates; MDD remission rates range from 35-40% in most psychopharmacology trials; exercise might help ameliorate MDD through a wide variety of mechanisms including effects on neurotransmitters and enhanced self-efficacy.

Comments: The results of this trial should give us reason to smile - a prescribed fitness regimen may be as effective as prescription aids for the treatment of MDD. People should be dancing in the streets (literally) when they learn of this...except that methodological shortcomings do muddy the waters a bit.

The researchers tried hard to limit interaction with the home-based exercisers, but there was still quite a bit of contact between subjects and staff that may have non-specifically benefited participants through social support. In addition, the results are not easily generalizable, as the subjects were mainly middle-aged white women, but even more importantly because they were all volunteers (the majority of who were hoping to be randomized to an exercise group).

The study's flaws are not so great that its conclusions can be eschewed. The independent effects of exercise appear to be comparable to those of antidepressant medication, and better than placebo, for the treatment of MDD in select instances. Of additional interest is the finding that exercise was equally effective for the treatment of MDD in both fitness groups, even though the supervised group developed better aerobic conditioning. These are data that can be built upon with further study that hopefully includes a broader swath of patients reflecting a variety of ages and patient circumstances. Until that time, however, fitness activities can, and perhaps should, be recommended as part of a multidisciplinary approach to management of MDD.

What to do with this article: Keep a hard copy in your file cabinet.