A framework for best evidence approaches can improve the transparency of systematic reviews.

OBJECTIVE: Systematic reviewers often use a "best evidence" approach to address the key questions, but what is meant by "best" is often unclear. The goal of this project was to create a decision framework for "best evidence" approaches to increase transparency in systematic reviews. STUDY DESIGN AND SETTING: The project was separated into three areas: 1) inclusion criteria, 2) evidence prioritization strategies, and 3) evaluative approaches. This commentary focuses only on the second task. The full report is available on the Effective Healthcare Web site of the Agency for Healthcare Research and Quality. RESULTS: The four identified strategies were as follows: 1) Use only the single best study; 2) Use the best set of studies; 3) Same as 2, but also consider whether the evidence permits a conclusion; and 4) Same as 3, but also consider the overall strength of the evidence. Simpler strategies (such as #1) are less likely to produce false conclusions, but are also more likely to yield insufficient evidence (possibly because of imprecise data). CONCLUSION: Systematic reviewers routinely prioritize evidence in numerous ways. This document provides a conceptual construct to enhance the transparency of systematic reviewers' decisions.

Efficacy and safety of innovator versus generic drugs in patients with epilepsy: a systematic review.

Generic antiepileptic drugs achieve blood concentrations similar to those of innovator drugs in healthy volunteers, but their comparative effectiveness has not been well evaluated. Thus, we assessed the efficacy, tolerability, and safety of innovator versus generic antiepileptic drugs. We searched the MEDLINE database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Web of Science for studies that evaluated innovator and generic antiepileptic drugs in patients with epilepsy and reported data on prespecified outcomes. We extracted data on study design, interventions, quality criteria, study population, baseline characteristics, and outcomes. Compared with initiation of innovator antiepileptic drugs, initiation of generic antiepileptic drugs did not significantly alter seizure occurrence (relative risk [RR] 0.87, 95% confidence interval [CI] 0.64-1.18; strength of evidence: low) or frequency (standardized mean difference 0.03, 95% CI -0.08-0.14; strength of evidence: low), withdrawals due to lack of efficacy (RR 1.02, 95% CI 0.41-2.54; strength of evidence: low) or adverse events (RR 0.79, 95% CI 0.28-2.20; strength of evidence: low), pharmacokinetic concentrations (maximum, minimum, or area under the curve [strength of evidence: low]), or a myriad of adverse events (strength of evidence: low or insufficient) in clinical trials. In qualitatively evaluated observational studies, switching between forms of antiepileptic drug (innovator to generic, generic to generic) may increase the risk of hospitalization (strength of evidence: low), hospital stay duration (strength of evidence: low), and a composite end point of medical service utilization (strength of evidence: insufficient) but may not increase outpatient service utilization (strength of evidence: low). Data are limited predominantly to carbamazepine, phenytoin, and valproic acid. Clinical trials are limited by small sample size, short-term nature, and lack of specification of A-rated generic products (generics that the United States Food and Drug Administration has deemed bioequivalent to the innovator drug). Observational trials lack full accounting for confounders and have inherent limitations. With a low strength of evidence, it appears that initiating an innovator or generic antiepileptic drug will provide similar efficacy, tolerability, and safety but that switching from one form to the other may be associated with more hospitalizations and longer hospital stays.

Neurothrombectomy devices for the treatment of acute ischemic stroke: state of the evidence.

BACKGROUND: Acute ischemic strokes are associated with poor outcomes and high health care burden. Evidence exists evaluating the use of neurothrombectomy devices in patients receiving currently recommended treatments that may have limited efficacy. PURPOSE: To describe the state of the evidence supporting use of neurothrombectomy devices in the treatment of acute ischemic stroke. DATA SOURCES: MEDLINE, SCOPUS, the Cochrane Central Register of Controlled Trials, the Cochrane Database of Systematic Reviews, and Web of Science were searched, without language restrictions, from their inception through May 2010. The MEDLINE and Cochrane Central Register of Controlled Trials searches were updated through November 2010. STUDY SELECTION: Two independent investigators screened citations for human studies of any design or case series or case reports of patients with an acute ischemic stroke that evaluated a neurothrombectomy device and reported at least 1 clinical effectiveness outcome or harm. DATA EXTRACTION: Using standardized protocols, 2 independent investigators extracted information about study characteristics and outcomes, and a third reviewer resolved disagreement. DATA SYNTHESIS: 87 articles met eligibility criteria, including 18 prospective single-group studies, 7 noncomparative retrospective studies, and 62 case series or case reports. Two U.S. Food and Drug Administration (FDA)-cleared devices, the MERCI Retriever (Concentric Medical, Mountain View, California) (40%) and the Penumbra System (Penumbra, Alameda, California) (9%), represented a large portion of the available data. All prospective and retrospective studies provided data on successful recanalization with widely varying rates (43% to 78% with the MERCI Retriever and 83% to 100% with the Penumbra System). Rates of harms, including symptomatic (16 studies; 0% to 10% with the MERCI Retriever and 0% to 11% with the Penumbra System) or asymptomatic (13 studies; 28% to 43% and 1% to 30%, respectively) intracranial hemorrhage and vessel perforation or dissection (11 studies; 0% to 7% and 0% to 5%, respectively), also varied by device. Predictors of harm included older age, history of stroke, and higher baseline stroke severity scores, whereas successful recanalization was the sole predictor of good outcomes. LIMITATIONS: Most available data are from single-group, noncomparative studies. In addition, the patient population most likely to benefit from these devices is undetermined. CONCLUSION: Currently available neurothrombectomy devices offer intriguing treatment options in patients with acute ischemic stroke. Future trials should use a randomized design, with adequate power to show equivalency or noninferiority between competing strategies or devices, and strive to identify populations that are most likely to benefit from use of neurothrombectomy devices. PRIMARY FUNDING SOURCE: Agency for Healthcare Research and Quality.

Recombinant human growth hormone in the treatment of patients with cystic fibrosis.

CONTEXT: Recombinant human growth hormone (rhGH) improves growth in patients with growth hormone deficiency or idiopathic short stature. Its role in patients with cystic fibrosis (CF) is unclear. OBJECTIVE: To review the effectiveness of rhGH in the treatment of patients with CF. METHODS: Medline and the Cochrane Central Register of Controlled Trials were searched from the earliest date through April 2010. Randomized controlled trials, observational studies, systematic reviews/meta-analyses, or case reports were included if rhGH therapy was administered to patients with CF and data on prespecified harms, intermediate outcomes, or final health outcomes were reported. When applicable, end points were pooled by using a random-effects model. The overall body of evidence was graded for each outcome as insufficient, low, moderate, or high. RESULTS: Ten unique controlled trials (n = 312) and 8 observational studies (n = 58) were included. On quantitative synthesis of controlled trials, several markers of pulmonary function, anthropometrics, and bone mineralization were significantly improved versus control. Results of single-arm observational studies for the aforementioned outcomes were generally supportive of findings in clinical trials. There is insufficient evidence to determine the effect of rhGH on intravenous antibiotic use during therapy, pulmonary exacerbations, health-related quality-of-life, bone consequences, or total mortality, but moderate evidence suggests that rhGH therapy reduces the rate of hospitalization versus control. CONCLUSIONS: rhGH improved almost all intermediate measures of pulmonary function, height, and weight in patients with CF. Improvements in bone mineral content are also promising. However, with the exception of hospitalizations, the benefits on final health outcomes cannot be directly determined at this time.

The comparative efficacy of plant sterols and stanols on serum lipids: a systematic review and meta-analysis.

BACKGROUND: Plant sterols and stanols are plant steroids with a similar chemical structure and cellular function to human cholesterol, and are recommended as dietary modifiers of serum lipids. Plant sterols have a higher degree of absorption than plant stanols, suggesting differential efficacy between the two. DESIGN: A meta-analysis of randomized controlled trials was performed to summarize direct comparisons between the effect of plant sterols vs plant stanols on serum lipid levels in healthy patients and patients with hypercholesterolemia. METHODS: A systematic literature search of MEDLINE, EMBASE, Cochrane CENTRAL, and the Natural Medicines Comprehensive Database was conducted from January 1950 through January 2009. Trials were included in the analysis if they were randomized controlled trials evaluating the effect of plant sterols vs plant stanols in healthy patients or patients with hypercholesterolemia who reported efficacy data on total, low-density lipoprotein, and high-density lipoprotein cholesterols or triglycerides. The weighted mean difference (WMD) of the change from baseline (in mg/dL) with 95% confidence interval was calculated as the difference between the means in the plant sterol and plant stanol groups using a random-effects model. RESULTS: Fourteen studies (n=531 patients) met the inclusion criteria. Upon meta-analysis, the results showed that there is no statistically or clinically significant difference between plant sterols and plant stanols in their abilities to modify total cholesterol (WMD -1.11 mg/dL [-0.0286 mmol/L], 95% confidence interval [CI] -4.12 to 1.90, P=0.47), low-density lipoprotein cholesterol (WMD -0.35 mg/dL [-0.0091 mmol/L], 95% CI -2.98 to 2.28, P=0.79), high-density lipoprotein cholesterol (WMD -0.28 mg/dL [-0.00073 mmol/L], 95% CI -1.18 to 0.62, P=0.54), or triglycerides (WMD -1.80 mg/dL [-0.0203 mmol/L], 95% CI -6.80 to 3.21, P=0.48). CONCLUSIONS: Plant sterols and plant stanols do not have statistically or clinically relevant differing effects on total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, or triglyceride levels. The selection of plant sterols vs plant stanols should then be based on potential differences in safety parameters and further study is required to elucidate such differences.

Benefits and risks associated with beta-blocker prophylaxis in noncardiac surgery.

PURPOSE: The benefits and risks associated with use of beta-blocker prophylaxis in noncardiac surgery (NCS) are described. SUMMARY: Perioperative beta-blockade is recommended by the American College of Cardiology and the American Heart Association for use in patients already on beta-blockers or in high-risk patients undergoing NCS to reduce myocardial ischemia and myocardial infarction (MI); however, the recommendations are not as clear for patients undergoing intermediate- or low-risk NCS. Numerous trials have evaluated the effect of perioperative beta-blockers on MI, stroke, bradycardia, hypotension, overall mortality, and cardiovascular mortality in patients undergoing NCS. Several trials suggest that dosing, patient population, type of NCS, genetic polymorphisms, and type of anesthesia may be important in determining the benefits and risks of perioperative beta-blockade in these patients. In the meta-analyses evaluating beta-blockers in NCS, the balance of benefits to harms associated with aggressive perioperative beta-blocker therapy was not favorable. However, the largest, most recent trial drove the meta-analyses results and has some methodological caveats and limitations that must be considered. Recent meta-analyses have found that the use of beta-blockers reduces the rate of MI but increases the frequency of stroke, and these MIs and strokes can occur with differing severities. CONCLUSION: Perioperative use of beta-blockers in NCS can protect against postoperative MI but increases the risk of stroke, severe hypotension, and severe bradycardia. Although less common, the strokes are severe, and the troubling trend toward increasing cardiovascular and total mortality precludes the recommendation for their use in patients not previously treated with beta-blockers.

Differing effect of statins on insulin sensitivity in non-diabetics: a systematic review and meta-analysis.

BACKGROUND: To determine whether individual statins had differing effects on insulin sensitivity (IS) in patients without pre-existing diabetes mellitus. METHODS: A systematic literature search of MEDLINE, EMBASE and Cochrane CENTRAL was conducted through December 2008. Trials were included if they compared pravastatin, atorvastatin, rosuvastatin or simvastatin to placebo/control, excluded patients with diabetes, and reported data on insulin sensitivity/resistance. IS data was pooled and evaluated as standardized mean differences (SMDs) and 95% confidence interval (CI) using a random-effects model. RESULTS: 16 studies (n=1146) were included, with patients receiving pravastatin in three trials (n=164), atorvastatin in five trials (n=315), rosuvastatin in five trials (n=419), and simvastatin in five trials (n=369). When pooled as a class, statins had no significant impact on IS as compared with placebo/control [SMD -0.084 (95% CI -0.210 to 0.042); p=0.19]. Pravastatin was found to significantly improved IS [SMD 0.342 (95% CI 0.032-0.621); p=0.03], whereas simvastatin significantly worsened IS [SMD -0.321 (95% CI -0.526 to -0.117); p=0.03]. CONCLUSIONS: Statins do not appear to demonstrate a 'class effect' on IS in patients without diabetes. Differences between individual statins likely exist that may partially explain the findings of previously conducted meta-analyses examining the impact of statins on the development of diabetes.

Systematic review: comparative effectiveness of angiotensin-converting enzyme inhibitors or angiotensin II-receptor blockers for ischemic heart disease.

BACKGROUND: Patients with ischemic heart disease and preserved ventricular function experience considerable morbidity and mortality despite standard medical therapy. PURPOSE: To compare benefits and harms of using angiotensin-converting enzyme (ACE) inhibitors, angiotensin II-receptor blockers (ARBs), or combination therapy in adults with stable ischemic heart disease and preserved ventricular function. DATA SOURCES: MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews (earliest date, July 2009) were searched without language restrictions. STUDY SELECTION: Two independent investigators screened citations for trials of at least 6 months' duration that compared ACE inhibitors, ARBs, or combination therapy with placebo or active control and reported any of several clinical outcomes. DATA EXTRACTION: Using standardized protocols, 2 independent investigators extracted information about study characteristics and rated the quality and strength of evidence. Disagreement was resolved by consensus. DATA SYNTHESIS: 41 studies met eligibility criteria. Moderate- to high-strength evidence (7 trials; 32 559 participants) showed that ACE inhibitors reduce the relative risk (RR) for total mortality (RR, 0.87 [95% CI, 0.81 to 0.94]) and nonfatal myocardial infarction (RR, 0.83 [CI, 0.73 to 0.94]) but increase the RR for syncope (RR, 1.24 [CI, 1.02 to 1.52]) and cough (RR, 1.67 [CI, 1.22 to 2.29]) compared with placebo. Low-strength evidence (1 trial; 5926 participants) suggested that ARBs reduce the RR for the composite end point of cardiovascular mortality, nonfatal myocardial infarction, or stroke (RR, 0.88 [CI, 0.77 to 1.00]) but not for the individual components. Moderate-strength evidence (1 trial; 25 620 participants) showed similar effects on total mortality (RR, 1.07 [CI, 0.98 to 1.16]) and myocardial infarction (RR, 1.08 [CI, 0.94 to 1.23]) but an increased risk for discontinuations because of hypotension (P < 0.001) and syncope (P = 0.035) with combination therapy compared with ACE inhibitors alone. LIMITATIONS: Many studies either did not assess or did not report harms in a systematic manner. Many studies did not adequately report benefits or harms by various patient subgroups. CONCLUSION: Adding an ACE inhibitor to standard medical therapy improves outcomes, including reduced risk for mortality and myocardial infarctions, in some patients with stable ischemic heart disease and preserved ventricular function. Less evidence supports a benefit of ARB therapy, and combination therapy seems no better than ACE inhibitor therapy alone and increases harms. PRIMARY FUNDING SOURCE: Agency for Healthcare Research and Quality.

A clinician's perspective on rating the strength of evidence in a systematic review.

Systematic review is a means of reviewing clearly formulated questions by using an explicit methodology to minimize bias in the location, selection, critical evaluation, and synthesis of research evidence from existing studies. It is not enough, however, to simply know that the best evidence available was captured in a systematic review. Rather, health care decision makers also need to understand what the strength of that evidence is. Strong evidence of a therapy's benefits and harms facilitates sound judgment in clinical practice, compared with weak evidence. In the absence of an organized method, different clinicians may review the same data and differ on their impression of the strength of evidence but not understand why they differ. A wide variety of grading systems are available to rate the strength of evidence, but different organizations may weigh features or domains of these systems differently. Also, the published articles written on how to grade the strength of evidence are not likely to penetrate to the clinician's level and are not written so that practicing clinicians can understand them. In this article, we provide a better understanding of the key criteria or domains that should be considered when rating the strength of a body of evidence, why they are important, and the domains included in some of the validated and commonly used scales. This not only will enable clinicians and health care decision makers to personally grade the strength of evidence and be able to extend it to their clinical practices, but also will allow them to understand which domains are and which are not covered, and how different grading scales can provide different results and still be accurate based on the domains they include.

The impact of garlic on lipid parameters: a systematic review and meta-analysis.

In order to determine the impact of garlic on total cholesterol (TC), TAG levels, as well as LDL and HDL, and establish if any variables have an impact on the magnitude of this effect, a meta-analysis was conducted. A systematic literature search of MEDLINE, CINAHL and the Cochrane Database from the earliest possible date through to November 2007 was conducted to identify randomised, placebo-controlled trials of garlic that reported effects on TC, TAG concentrations, LDL or HDL. The weighted mean difference of the change from baseline (with 95 % CI) was calculated as the difference between the means in the garlic groups and the control groups using a random-effects model. Subgroup and sensitivity analyses were performed to determine the effects on type, brand and duration of garlic therapy as well as baseline TC and TAG levels, the use of dietary modification, and study quality on the meta-analysis's conclusions. Twenty-nine trials were included in the analysis. Upon meta-analysis garlic was found to significantly reduce TC ( - 0.19; 95 % CI - 0.33, - 0.06 mmol/l) and TAG ( - 0.11; 95 % CI - 0.19, - 0.06 mmol/l) but exhibited no significant effect on LDL or HDL. There was a moderate degree of statistical heterogeneity for the TC and TAG analyses. Garlic reduces TC to a modest extent, an effect driven mostly by the modest reductions in TAG, without appreciable LDL lowering or HDL elevation. Higher baseline line TC levels and the use of dietary modification may alter the effect of garlic on these parameters. Future studies should be conducted evaluating the impact of adjunctive garlic therapy with fibrates or statins on TAG concentrations.

Outcomes of perioperative beta-blockade in patients undergoing noncardiac surgery: a meta-analysis.

BACKGROUND: Several studies have evaluated the impact on myocardial infarction (MI), stroke, and overall mortality of perioperative beta-blocker use in patients undergoing noncardiac surgery (NCS). However, most studies did not have adequate sample size and statistical power and were therefore underpowered to adequately evaluate these endpoints. OBJECTIVE: To conduct a meta-analysis to determine the balance of benefits and harms associated with perioperative beta-blocker use in NCS. METHODS: A systematic literature search of MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials was conducted from January 1960 through February 2009. Manual reference search was performed to identify additional relevant trials. Randomized, double-blinded, placebo-controlled trials comparing the use of beta-blockers with placebo; using beta-blockers perioperatively in beta-blocker-naïve patients undergoing NCS; and evaluating endpoints of MI, stroke, or all-cause mortality were included. RESULTS: Six trials (N = 10,183) met our inclusion criteria. Perioperative beta-blocker use was associated with a significant reduction in patients' odds of developing MI (OR 0.74, 95% CI 0.61 to 0.89) but a significant increase in odds of developing stroke (OR 1.98, 95% CI 1.23 to 3.20) and also a nonsignificant increase in mortality (OR 1.21, 95% CI 0.98 to 1.49) versus placebo. Control-rate meta-regression determined that patients with highest baseline odds of stroke had decreased relative odds of having a stroke with a beta-blocker versus placebo (beta coefficient -0.97; 95% credible interval -1.04 to -0.90). CONCLUSIONS: When perioperative beta-blockers are used in NCS patients, there is a trade-off between reduction in MI and increase in stroke, with a troubling trend toward an increase in mortality. Patients with lower baseline odds of developing stroke appear to be at greater risk of beta-blocker-induced stroke.

Effect of non-ergot dopamine agonists on health-related quality of life of patients with restless legs syndrome.

BACKGROUND: Non-ergot dopamine agonists (NEDAs) have become the gold-standard agents for the treatment of restless legs syndrome (RLS). While the efficacy and safety of these drugs have been widely studied, their effect on patients' health-related quality of life (HRQoL) has not been fully elucidated. OBJECTIVE: To better define the usefulness of NEDAs by assessing their impact on HRQoL. METHODS: We performed a meta-analysis of randomized, double-blind, placebo-controlled trials evaluating the use of NEDAs in patients with RLS. A systematic literature search of MEDLINE, EMBASE, and Cochrane CENTRAL was performed from the earliest possible date through July 2008. Trials were included in the analysis if they evaluated NEDAs for the treatment of RLS and reported HRQoL using any RLS disease-specific HRQoL instrument. HRQoL data were pooled and evaluated using an inverse variance weighting approach as standardized mean differences (SMDs) and 95% confidence interval (CI). For trials reporting HRQoL data obtained using the Johns Hopkins RLS-QoL Questionnaire, adjusted mean difference data were pooled to calculate the weighted mean difference (WMD) and 95% CI. RESULTS: Seven trials (N = 1483) met all inclusion criteria. Patients with RLS taking NEDAs had significantly improved overall effect on HRQoL compared with those taking placebo (SMD 0.20; 95% CI 0.10 to 0.30; degree of inconsistency across studies [I(2)] = 0%). When analyzing trials using the Johns Hopkins RLS-QoL questionnaire, the results also showed improvement with NEDAs compared with placebo (WMD 4.72; 95% CI 2.96 to 6.47; I(2) = 0%). Study conclusions were unchanged upon sensitivity analysis. The number of trials for each NEDA was small, limiting the usefulness of between-agent comparisons. CONCLUSIONS: In patients with RLS, use of NEDAs showed improved HRQoL compared with placebo. Since pooled effect sizes observed in this meta-analysis appear to surpass accepted values for minimally important clinical differences, these improvements may be clinically relevant for the average studied patient. However, future studies evaluating long-term treatment of RLS with NEDAs are necessary, as are head-to-head comparative trials and economic assessments.

Intravenous magnesium sulfate enhances the ability of dofetilide to successfully cardiovert atrial fibrillation or flutter: results of the Dofetilide and Intravenous Magnesium Evaluation.

AIMS: A previous study found that the adjunctive use of intravenous magnesium sulfate with ibutilide could increase the odds of a patient chemically cardioverting from atrial fibrillation (AF) or flutter (AFL) to normal sinus rhythm (NSR) by 78%. Whether or not intravenous magnesium has the same effect on dofetilide's ability to chemically cardiovert patients from AF/AFL to NSR is not known. METHODS AND RESULTS: This was a retrospective cohort evaluation of consecutive eligible patients receiving dofetilide for chemical cardioversion of AF or AFL at a single institution. All AF or AFL patients received dofetilide according to the institution's standard protocol, which required patients to remain as an inpatient for a minimum of 3 days or 6 doses after the initiation of dofetilide therapy. Patients receiving any dose of intravenous magnesium starting on the same day as dofetilide constituted the treatment group. Controls received dofetilide, but no intravenous magnesium any time prior to chemical cardioversion. Patients underwent continuous electrocardiographic monitoring throughout their hospital admission. Multivariable logistic regression analysis was used to determine the impact of intravenous magnesium on dofetilide's efficacy. A total of 160 patients in persistent AF or AFL (mean age 66.6 +/- 11.0 years, 70.0% male, 30.0% in AF or AFL >15 days, 54.4% hypertension, 37.5% heart failure, 16.3% valvular disease, 16.3% previous myocardial infarction, and baseline serum magnesium levels 2.1 +/- 0.26 mg/dL) and receiving dofetilide (mean dose 428 +/- 118 microg/dose) were included in this analysis. The overall chemical cardioversion rate with dofetilide irrespective of adjunctive intravenous magnesium utilization was 41.9%. The concurrent administration of intravenous magnesium (n = 50) was associated with a 107% increased odds of successful chemical cardioversion [adjusted odds ratio: 2.07 (95% confidence intervals: 1.00-4.33)] compared with those who did not receive magnesium (n = 110). Only one case of torsade de pointes occurred in the no magnesium group during the index hospital admission. CONCLUSION: Concurrent use of intravenous magnesium is associated with an enhanced ability of dofetilide to successfully convert AF or AFL.

The effects of barley-derived soluble fiber on serum lipids.

PURPOSE: We wanted to determine the association between consumption of barley and changes in plasma lipids in healthy and hypercholesterolemic men and women. METHODS: A systematic literature search was conducted from the earliest possible date through January 2008. Trials were included in the analysis if they were randomized controlled trials of barley that reported efficacy data on at least 1 lipid endpoint. A DerSimonian and Laird random-effects model was used in calculating the weighted mean difference (WMD) and its 95% confidence interval (CI). Statistical heterogeneity was addressed using the I(2) statistic. Visual inspection of funnel plots, Egger's weighted regression statistics, and the trim and fill method were used to assess for publication bias. RESULTS: We found 8 trials (n = 391 patients) of 4 to 12 weeks' duration evaluating the lipid-reducing effects of barley. The use of barley significantly lowered total cholesterol (weighted mean difference [WMD], -13.38 mg/dL; 95% CI, -18.46 to -8.31 mg/dL), low-density lipoprotein (LDL) cholesterol (WMD, -10.02 mg/dL; 95% CI, -14.03 to -6.00 mg/dL) and triglycerides (WMD, -11.83 mg/dL; 95% CI, -20.12 to -3.55 mg/dL) but did not appear to significantly alter high-density lipoprotein (HDL) cholesterol (P=.07). CONCLUSION: Barley-derived beta-glucan appears to beneficially affect total cholesterol, LDL-cholesterol, and triglycerides, but not HDL-cholesterol.

Pharmacologic treatment of advanced Parkinson's disease: a meta-analysis of COMT inhibitors and MAO-B inhibitors.

OBJECTIVE: To perform a meta-analysis of randomized placebo-controlled trials evaluating catechol-O-methyltransferase (COMT) inhibitors or monoamine oxidase type B (MAO-B) inhibitors in addition to levodopa versus levodopa alone for the treatment of advanced Parkinson's disease (PD). METHODS: A systematic literature search was performed between 1990 and October 2007. The primary outcome measures assessed were the reduction in scores of Unified Parkinson's Disease Rating Scale (UPDRS) total, activities of daily living (ADL) and motor scores from baseline. Other efficacy and safety endpoints were also evaluated. RESULTS: A total of 13 trials (n=3775 subjects) were included in the meta-analysis. As compared to placebo, COMT and MAO-B inhibitor use resulted in greater improvement in UPDRS total score (weighted mean difference [WMD] -2.13, 95%CI -0.46 to -0.20; and WMD -5.03, 95%CI -7.38 to -2.68) ADL scores (WMD -0.99, 95%CI -1.56 to -0.43; and WMD -1.48, 95%CI -2.13 to -0.83) and motor scores (WMD -1.50, 95%CI -2.70 to -0.30; and WMD -3.19, 95%CI -4.57 to -1.80) as well as increase in "on" time, reduction in "off" time and decreased need in levodopa dose compared to placebo. Incidences of dyskinesia were significantly higher with the COMT and MAO-B inhibitors compared to placebo. CONCLUSION: The use of COMT or MAO-B inhibitors plus levodopa is superior to levodopa alone at reducing PD symptoms in patients with advanced PD. While combination therapies with COMT or MAO-B inhibitor plus levodopa seem especially useful amongst PD patients with wearing-off phenomenon, they are associated with more adverse events.

The effect of adding plant sterols or stanols to statin therapy in hypercholesterolemic patients: systematic review and meta-analysis.

OBJECTIVE: To characterize the effect of plant sterols/stanols on serum lipids in hypercholesterolemic patients on concurrent statin therapy, we conducted a meta-analysis of randomized controlled trials. METHODS: A systematic literature search of MEDLINE, EMBASE, Cochrane CENTRAL, and the Natural Medicines Comprehensive Database was conducted from the earliest possible date through May 2008. Trials were included in the analysis if they were randomized controlled trials evaluating the use of plant sterols/stanols in combination with statins in hypercholesterolemic patients that reported efficacy data on total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, or triglycerides. The weighted mean difference (WMD) of the change from baseline (in mg/dL) with 95% confidence interval (CI) was calculated as the difference between the mean in the plant sterol/stanol groups and the control groups, using a random-effects model. RESULTS: Eight studies (n = 306 patients) met the inclusion criteria. Upon meta-analysis, the use of plant sterols/stanols in combination with statin therapy significantly lowered total cholesterol (WMD, -14.01 mg/dL [95% CI, -18.66 to -9.37], p < 0.0001) and LDL cholesterol (WMD, -13.26 mg/dL [95% CI, -17.34 to -9.18], p < 0.0001) but not HDL cholesterol or triglycerides. CONCLUSIONS: Based upon the current literature, we can only say that plant sterols/stanols, when administered in addition to statins, favorably affect total and LDL cholesterol with 95% confidence. Randomized trials examining the impact of plant sterols/stanols in combinatation with statins on patient morbidity and mortality are needed.