Does metformin reduce the risk of cancer in obesity and diabetes? A systematic review and meta-analysis

AIM

To evaluate the effect of metformin on cancer incidence in subjects with overweight/obesity and/or prediabetes/diabetes.

MATERIALS AND METHODS

We searched MEDLINE, Embase and CENTRAL for randomized controlled trials (RCTs) in adults with overweight/obesity and/or prediabetes/diabetes that compared metformin to other interventions for ≥24 weeks. Independent reviewers selected and extracted data including population and intervention characteristics and new diagnoses of cancer. We used the RoB 2.0 risk-of-bias tool and the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) framework to assess risk of bias and certainty of evidence.

RESULTS

From 14 895 records after removal of duplicates, 27 trials were included, providing a total of 10 717 subjects in the metformin group and 10 003 in the control group, with 170 and 208 new cases of cancer, respectively. Using a random-effects model, the relative risk was 1.07 (95% confidence interval 0.87-1.31), with similar results in subgroup analyses by study duration or effect of control intervention on weight. Risk of bias in most studies was low, and no evidence of publication bias was found. Trial sequential analysis provided evidence that the cumulative sample size was large enough to exclude a significant effect of metformin on cancer incidence.

CONCLUSIONS

Metformin did not reduce cancer incidence in RCTs involving subjects with overweight/obesity and/or prediabetes/diabetes.

Long-term lifestyle intervention can reduce the development of type 2 diabetes mellitus in subjects with prediabetes: A systematic review and meta-analysis

AIMS

To evaluate the effectiveness of intensive lifestyle intervention (ILI) on the risk of type 2 diabetes (T2D) in prediabetes (PD).

METHODS

We searched the Cochrane Central, Embase, MEDLINE, and Web of Science (until February 2024) to include RCTs of adults with PD, comparing ILI vs. general advice on the incidence of T2D. Two authors extracted the data, applied the Cochrane Risk of Bias (RoB) 2.0 tool and the GRADE framework. Meta-analysis was performed using random effects models, estimating relative risk (RR) and the 95%CI.

RESULTS

Fifteen studies (n = 8,563, 46.7 % female, 53.3 ± 8.7 years, BMI 26.7 ± 5.4 Kg/m2) were included. ILI reduced T2D risk by 22 % when compared with general advice (RR 0.78; 95 %CI 0.72-0.85; I2 = 40 %; low certainty of evidence). Most studies had high risk of bias or raised some concerns. Sensitivity analysis showed that studies with mostly female populations and those using the WHO 1985 criteria for T2D had lower risk of the disease and that the longer the follow-up, the lower the protection.

CONCLUSION

ILI can prevent T2D in subjects with PD. Healthcare teams should aim for structured ILI to maintain long-term lifestyle improvements.

The II Brazilian Guidelines for the pharmacological treatment of patients hospitalized with COVID-19 Joint Guidelines of the Associação Brasileira de Medicina de Emergência, Associação de Medicina Intensiva Brasileira, Associação Médica Brasileira, Sociedade Brasileira de Angiologia e Cirurgia Vascular, Sociedade Brasileira de Infectologia, Sociedade Brasileira de Pneumologia e Tisiologia and Sociedade Brasileira de Reumatologia

OBJECTIVE

To update the recommendations to support decisions regarding the pharmacological treatment of patients hospitalized with COVID-19 in Brazil.

METHODS

Experts, including representatives of the Ministry of Health and methodologists, created this guideline. The method used for the rapid development of guidelines was based on the adoption and/or adaptation of existing international guidelines (GRADE ADOLOPMENT) and supported by the e-COVID-19 RecMap platform. The quality of the evidence and the preparation of the recommendations followed the GRADE method.

RESULTS

Twenty-one recommendations were generated, including strong recommendations for the use of corticosteroids in patients using supplemental oxygen and conditional recommendations for the use of tocilizumab and baricitinib for patients on supplemental oxygen or on noninvasive ventilation and anticoagulants to prevent thromboembolism. Due to suspension of use authorization, it was not possible to make recommendations regarding the use of casirivimab + imdevimab. Strong recommendations against the use of azithromycin in patients without suspected bacterial infection, hydroxychloroquine, convalescent plasma, colchicine, and lopinavir + ritonavir and conditional recommendations against the use of ivermectin and remdesivir were made.

CONCLUSION

New recommendations for the treatment of hospitalized patients with COVID-19 were generated, such as those for tocilizumab and baricitinib. Corticosteroids and prophylaxis for thromboembolism are still recommended, the latter with conditional recommendation. Several drugs were considered ineffective and should not be used to provide the best treatment according to the principles of evidence-based medicine and to promote resource economy.

Cancer Outcomes Among Prediabetes and Type 2 Diabetes Populations With Dietary and Physical Activity-based Lifestyle Interventions

CONTEXT

People with type 2 diabetes (T2D) have higher risks of cancer incidence and death.

OBJECTIVE

We aimed to evaluate the relationship between dietary and physical activity-based lifestyle intervention and cancer outcomes among prediabetes and T2D populations.

METHODS

We searched for randomized controlled trials with at least 24 months of lifestyle interventions in prediabetes or T2D populations. Data were extracted by pairs of reviewers and discrepancies were resolved by consensus. Descriptive syntheses were performed, and the risk of bias was assessed. Relative risks (RRs) and 95% CIs were estimated using a pairwise meta-analysis with both a random-effects model and a general linear mixed model (GLMM). Certainty of evidence was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation framework, and trial sequential analysis (TSA) was conducted to assess if current information is enough for definitive conclusions. Subgroup analysis was performed by glycemic status.

RESULTS

Six clinical trials were included. Among 12 841 participants, the combined RR for cancer mortality comparing lifestyle interventions with usual care was 0.94 (95% CI, 0.81-1.10 using GLMM and 0.82-1.09 using random-effects model). Most studies had a low risk of bias, and the certainty of evidence was moderate. TSA showed that the cumulative Z curve reached futility boundary while total number did not reach detection boundary.

CONCLUSION

Based on the limited data available, dietary and physical activity-based lifestyle interventions had no superiority to usual care on reducing cancer risk in populations with prediabetes and T2D. Lifestyle interventions focused on cancer outcomes should be tested to better explore their effects.

Coconut oil: an overview of cardiometabolic effects and the public health burden of misinformation

Recent data from meta-analyses of randomized clinical trials (RCTs) suggest that dietary intake of coconut oil, rich in saturated fatty acids, does not result in cardiometabolic benefits, nor in improvements in anthropometric, lipid, glycemic, and subclinical inflammation parameters. Nevertheless, its consumption has surged in recent years all over the world, a phenomenon which can possibly be explained by an increasing belief among health professionals that this oil is as healthy as, or perhaps even healthier than, other oils, in addition to social network misinformation spread. The objective of this review is to present nutritional and epidemiological aspects related to coconut oil, its relationship with metabolic and cardiovascular health, as well as possible hypotheses to explain its high rate of consumption, in spite of the most recent data regarding its actual effects.

Sodium-glucose cotransporter-2 inhibitors and cancer outcomes: A systematic review and meta-analysis of randomized controlled trials

AIMS

Concerns regarding breast and bladder cancer risk with Sodium-glucose cotransporter-2 (SGLT2) inhibitors remain controversial and its effect on cancer mortality is unknown. We aim to evaluate the association between SGLT2 inhibitors and the risk of cancer outcomes.

METHODS

We searched PubMed, Embase and CENTRAL up to June 20th, 2022, for randomized controlled trials of SGLT2 inhibitors in adults, with a minimum follow-up of 48 weeks. Researchers extracted study-level data and assessed within-study risk of bias with the RoB 2.0 tool and quality of evidence with GRADE. We performed meta-analyses summarizing the relative risks (RRs) of cancer outcomes.

RESULTS

Seventy-six trials encompassing 116,375 participants were selected. Overall risk of bias was low. SGLT2 inhibitors did not reduce/increase the overall risk of cancer (RR, 1.03; 95% confidence interval [CI], 0.96-1.10) and cancer mortality (RR, 0.99; 95% CI, 0.85-1.16). SGLT2 inhibitors likely result in little to no difference in the risk of breast (RR, 1.01; 95% CI 0.77-1.32) and bladder cancers (RR, 0.93; 95% CI 0.71-1.21). Trial sequential analysis provided evidence that the sample size was sufficient to avoid missing alternative results.

CONCLUSIONS

SGLT2 inhibitors are not associated with an increased risk of cancer outcomes, providing reassuring data regarding previous safety concerns.

Long-term Effect of Lifestyle Interventions on the Cardiovascular and All-Cause Mortality of Subjects With Prediabetes and Type 2 Diabetes: A Systematic Review and Meta-analysis

BACKGROUND

Lifestyle interventions improve the metabolic control of individuals with hyperglycemia.

PURPOSE

We aimed to determine the effect of lifestyle interventions on cardiovascular and all-cause mortality in this population.

DATA SOURCES

Searches were made through MEDLINE, Cochrane CENTRAL, Embase, and Web of Science (no date/language restriction, until 15 May 2022).

STUDY SELECTION

We included randomized clinical trials (RCTs) of subjects with prediabetes and type 2 diabetes, comparing intensive lifestyle interventions with usual care, with a minimum of 2 years of active intervention.

DATA EXTRACTION

Data from the 11 RCTs selected were extracted in duplicate. A frequentist and arm-based meta-analysis was performed with random-effects models to estimate relative risk (RR) for mortality, and heterogeneity was assessed through I2 metrics. A generalized linear mixed model (GLMM) was used to confirm the findings.

DATA SYNTHESIS

Lifestyle interventions were not superior to usual care in reducing cardiovascular (RR 0.99; 95% CI 0.79-1.23) or all-cause (RR 0.93; 95% CI 0.85-1.03) mortality. Subgroup, sensitivity, and meta-regression analyses showed no influence of type of intervention, mean follow-up, age, glycemic status, geographical location, risk of bias, or weight change. All of these results were confirmed with the GLMM. Most studies had a low risk of bias according to the RoB 2.0 tool and the certainty of evidence was moderate for both outcomes.

LIMITATIONS

Most studies had a low risk of bias according to the RoB 2.0 tool, and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach resulted in moderate certainty of evidence for both outcomes. Differences in lifestyle programs and in usual care between the studies should be considered in the interpretation of our results.

CONCLUSIONS

Intensive lifestyle interventions implemented so far did not show superiority to usual care in reducing cardiovascular or all-cause mortality for subjects with prediabetes and type 2 diabetes.

The effects of coconut oil on the cardiometabolic profile: a systematic review and meta-analysis of randomized clinical trials

Background

Despite having a 92% concentration of saturated fatty acid composition, leading to an apparently unfavorable lipid profile, body weight and glycemic effect, coconut oil is consumed worldwide. Thus, we conducted an updated systematic review and meta-analysis of randomized clinical trials (RCTs) to analyze the effect of coconut oil intake on different cardiometabolic outcomes.

Methods

We searched Medline, Embase, and LILACS for RCTs conducted prior to April 2022. We included RCTs that compared effects of coconut oil intake with other substances on anthropometric and metabolic profiles in adults published in all languages, and excluded non-randomized trials and short follow-up studies. Risk of bias was assessed with the RoB 2 tool and certainty of evidence with GRADE. Where possible, we performed meta-analyses using a random-effects model.

Results

We included seven studies in the meta-analysis (n = 515; 50% females, follow up from 4 weeks to 2 years). The amount of coconut oil consumed varied and is expressed differently among studies: 12 to 30 ml of coconut oil/day (n = 5), as part of the amount of SFAs or total daily consumed fat (n = 1), a variation of 6 to 54.4 g/day (n = 5), or as part of the total caloric energy intake (15 to 21%) (n = 6). Coconut oil intake did not significantly decrease body weight (MD -0.24 kg, 95% CI -0.83 kg to 0.34 kg), waist circumference (MD -0.64 cm, 95% CI -1.69 cm to 0.41 cm), and % body fat (-0.10%, 95% CI -0.56% to 0.36%), low-density lipoprotein cholesterol (LDL-C) (MD -1.67 mg/dL, 95% CI -6.93 to 3.59 mg/dL), and triglyceride (TG) levels (MD -0.24 mg/dL, 95% CI -5.52 to 5.04 mg/dL). However, coconut oil intake was associated with a small increase in high-density lipoprotein cholesterol (HDL-C) (MD 3.28 mg/dL, 95% CI 0.66 to 5.90 mg/dL). Overall risk of bias was high, and certainty of evidence was very-low. Study limitations include the heterogeneity of intervention methods, in addition to small samples and short follow-ups, which undermine the effects of dietary intervention in metabolic parameters.

Conclusions

Coconut oil intake revealed no clinically relevant improvement in lipid profile and body composition compared to other oils/fats. Strategies to advise the public on the consumption of other oils, not coconut oil, due to proven cardiometabolic benefits should be implemented.

Registration

PROSPERO CRD42018081461.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-022-01685-z.

Misinformation in nutrition through the case of coconut oil: An online before-and-after study

BACKGROUND AND AIMS

Despite recent scientific evidence indicating absence of cardiometabolic benefit resulting from coconut oil intake, its consumption has increased in recent years, which can be attributed to a promotion of its use on social networks. We evaluated the patterns, reasons and beliefs related to coconut oil consumption and its perceived benefits in an online survey of a population in southern Brazil.

METHODS AND RESULTS

We conducted a before-and-after study using an 11-item online questionnaire that evaluated coconut oil consumption. In the same survey, participants who consumed coconut oil received an intervention to increase literacy about the health effects of coconut oil intake. We obtained 3160 valid responses. Among participants who consumed coconut oil (59.1%), 82.5% considered it healthy and 65.4% used it at least once a month. 81.2% coconut oil consumers did not observe any health improvements. After being exposed to the conclusions of a meta-analysis showing that coconut oil does not show superior health benefits when compared to other oils and fats, 73.5% of those who considered coconut oil healthy did not change their opinion. Among individuals who did not consume coconut oil, 47.6% considered it expensive and 11.6% deemed it unhealthy.

CONCLUSIONS

Coconut oil consumption is motivated by the responders' own beliefs in its supposed health benefits, despite what scientific research demonstrates. This highlights the difficulty in deconstructing inappropriate concepts of healthy diets that are disseminated in society.

Meta-analysis of prevalence: I2 statistic and how to deal with heterogeneity

Over the last decade there has been a ten-fold increase in the number of published systematic reviews of prevalence. In meta-analyses of prevalence, the summary estimate represents an average prevalence from included studies. This estimate is truly informative only if there is no substantial heterogeneity among the different contexts being pooled. In systematic reviews, heterogeneity is usually explored with I-squared statistic (I² ), but this statistic does not directly inform us about the distribution of effects and frequently systematic reviewers and readers misinterpret this result. In a sample of 134 meta-analyses of prevalence, the median I² was 96.9% (IQR 90.5 to 98.7). We observed larger I² in meta-analysis with higher number of studies and extreme pooled estimates (defined as <10% or > 90%). Studies with high I² values were more likely to have conducted a sensitivity analysis, including subgroup analysis. Only 3 (2%) systematic reviews reported prediction intervals. Meta-analyses of prevalence often present high I² values. However, the number of studies included in the meta-analysis and the point estimate can be associated with the I² value, and a high I² value is not always synonymous with high heterogeneity. In meta-analyses of prevalence, I² statistics may not be discriminative and should be interpreted with caution, avoiding arbitrary thresholds. To discuss heterogeneity, reviewers should focus on the description of the expected range of estimates, which can be done using prediction intervals and planned sensitivity analysis. This article is protected by copyright. All rights reserved.

Brazilian Guidelines for the pharmacological treatment of patients hospitalized with COVID-19

Objective

Several therapies are being used or proposed for COVID-19, and many lack appropriate evaluations of their effectiveness and safety. The purpose of this document is to develop recommendations to support decisions regarding the pharmacological treatment of patients hospitalized with COVID-19 in Brazil.

Methods

A group of 27 experts, including representatives of the Ministry of Health and methodologists, created this guideline. The method used for the rapid development of guidelines was based on the adoption and/or adaptation of existing international guidelines (GRADE ADOLOPMENT) and supported by the e-COVID-19 RecMap platform. The quality of the evidence and the preparation of the recommendations followed the GRADE method.

Results

Sixteen recommendations were generated. They include strong recommendations for the use of corticosteroids in patients using supplemental oxygen, the use of anticoagulants at prophylactic doses to prevent thromboembolism and the nonuse of antibiotics in patients without suspected bacterial infection. It was not possible to make a recommendation regarding the use of tocilizumab in patients hospitalized with COVID-19 using oxygen due to uncertainties regarding the availability of and access to the drug. Strong recommendations against the use of hydroxychloroquine, convalescent plasma, colchicine, lopinavir + ritonavir and antibiotics in patients without suspected bacterial infection and also conditional recommendations against the use of casirivimab + imdevimab, ivermectin and rendesivir were made.

Conclusion

To date, few therapies have proven effective in the treatment of hospitalized patients with COVID-19, and only corticosteroids and prophylaxis for thromboembolism are recommended. Several drugs were considered ineffective and should not be used to provide the best treatment according to the principles of evidence-based medicine and promote economical resource use.

Brazilian Guidelines for the pharmacological treatment of patients hospitalized with COVID-19: Joint guideline of Associação Brasileira de Medicina de Emergência, Associação de Medicina Intensiva Brasileira, Associação Médica Brasileira, Sociedade Brasileira de Angiologia e Cirurgia Vascular, Sociedade Brasileira de Infectologia, Sociedade Brasileira de Pneumologia e Tisiologia, Sociedade Brasileira de Reumatologia

OBJECTIVE

Several therapies are being used or proposed for COVID-19, and many lack appropriate evaluations of their effectiveness and safety. The purpose of this document is to develop recommendations to support decisions regarding the pharmacological treatment of patients hospitalized with COVID-19 in Brazil.

METHODS

A group of 27 experts, including representatives of the Ministry of Health and methodologists, created this guideline. The method used for the rapid development of guidelines was based on the adoption and/or adaptation of existing international guidelines (GRADE ADOLOPMENT) and supported by the e-COVID-19 RecMap platform. The quality of the evidence and the preparation of the recommendations followed the GRADE method.

RESULTS

Sixteen recommendations were generated. They include strong recommendations for the use of corticosteroids in patients using supplemental oxygen, the use of anticoagulants at prophylactic doses to prevent thromboembolism and the nonuse of antibiotics in patients without suspected bacterial infection. It was not possible to make a recommendation regarding the use of tocilizumab in patients hospitalized with COVID-19 using oxygen due to uncertainties regarding the availability of and access to the drug. Strong recommendations against the use of hydroxychloroquine, convalescent plasma, colchicine, lopinavir + ritonavir and antibiotics in patients without suspected bacterial infection and also conditional recommendations against the use of casirivimab + imdevimab, ivermectin and rendesivir were made.

CONCLUSION

To date, few therapies have proven effective in the treatment of hospitalized patients with COVID-19, and only corticosteroids and prophylaxis for thromboembolism are recommended. Several drugs were considered ineffective and should not be used to provide the best treatment according to the principles of evidence-based medicine and promote economical resource use.

Conducting proportional meta-analysis in different types of systematic reviews: a guide for synthesisers of evidence

BACKGROUND

Single group data present unique challenges for synthesises of evidence. Proportional meta-analysis is becoming an increasingly common technique employed for the synthesis of single group data. Proportional meta-analysis shares many similarities with the conduct and reporting of comparative, or pairwise, meta-analysis. While robust and comprehensive methods exist detailing how researchers can conduct a meta-analysis that compares two (or more) groups against a common intervention, there is a scarcity of methodological guidance available to assist synthesisers of evidence in the conduct, interpretation, and importance of proportional meta-analysis in systematic reviews.

MAIN BODY

This paper presents an overview targeted to synthesisers of evidence and systematic review authors that details the methods, importance, and interpretation of a proportional meta-analysis. We provide worked examples of how proportional meta-analyses have been conducted in research syntheses previously and consider the methods, statistical considerations, and presentation of this technique.

CONCLUSION

This overview is designed to serve as practical guidance for synthesisers of evidence in the conduct of proportional meta-analyses.

Do GLP-1 Receptor Agonists Increase the Risk of Breast Cancer? A Systematic Review and Meta-analysis

CONTEXT

Risk of cancer is a major concern in the development of drugs for the treatment of obesity and diabetes. In randomized controlled trials (RCTs) of the Liraglutide Clinical Development Program, subjects treated with a glucagon-like peptide-1 receptor agonist (GLP-1RA) had a higher absolute number of breast cancer events.

OBJECTIVE

To assess whether patients treated with GLP-1RAs had a higher risk of breast neoplasms.

DATA SOURCES

We searched MEDLINE, Embase, Web of Science, and CENTRAL from July 31, 2019 to February 8, 2020.

STUDY SELECTION

Reviewers assessed abstracts and full-text articles for RCTs of GLP-1RAs in adults with excessive weight and/or diabetes and a minimum follow-up of 24 weeks.

DATA EXTRACTION

Researchers extracted study-level data and assessed within-study risk of bias with the RoB 2.0 tool and quality of evidence with Grading of Recommendations Assessment, Development and Evaluation (GRADE).

DATA SYNTHESIS

We included 52 trials, of which 50 reported breast cancer events and 11 reported benign breast neoplasms. Overall methodological quality was high. Among 48 267 subjects treated with GLP-1RAs, 130 developed breast cancer compared with 107 of 40 755 controls (relative risk [RR], 0.98; 95% confidence interval [CI], 0.76-1.26). Subset analyses according to follow-up, participant/investigator blinding, and type of GLP-1RA did not reveal any differences. The risk of benign breast neoplasms also did not differ between groups (RR, 0.99; 95% CI, 0.48-2.01). Trial sequential analysis provided evidence that the sample size was sufficient to avoid missing alternative results.

CONCLUSIONS

Treatment with GLP-1RAs for obesity and diabetes does not increase the risk of breast neoplasms.

Is lorcaserin really associated with increased risk of cancer? A systematic review and meta-analysis

The US Food and Drug Administration (FDA) reported in February 2020 an increased risk of cancer with lorcaserin in the follow-up of the CAMELLIA-TIMI 61 trial. This systematic review and meta-analysis addresses whether lorcaserin is associated with higher incidence of cancer compared with other interventions or no treatment. We searched MEDLINE, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) for randomized controlled trials that compared lorcaserin with other interventions or no treatment in adults. We performed descriptive synthesis of all included studies and conducted meta-analysis of trials that reported new cases of cancer. From 11 trials, comprising 21,299 individuals, four studies were included in the meta-analysis and reported 476 cases of cancer in 10,342 subjects in the lorcaserin group and 438 among 9429 individuals randomized to placebo (relative risk [RR]: 1.08; 95% confidence interval [95% CI]: 0.96-1.23). The result was heavily influenced by the CAMELLIA-TIMI 61 trial. In this study, the lorcaserin group had a higher risk of lung and pancreatic but not colon cancer. Overall risk of bias was low, and quality of evidence was moderate. The current evidence does not confirm the increased risk of cancer with lorcaserin but suggests a trend in this direction, with a greater incidence of some subtypes such as lung and pancreas.

Brazilian guidelines for the management of brain-dead potential organ donors. The task force of the AMIB, ABTO, BRICNet, and the General Coordination of the National Transplant System

OBJECTIVE

To contribute to updating the recommendations for brain-dead potential organ donor management.

METHOD

A group of 27 experts, including intensivists, transplant coordinators, transplant surgeons, and epidemiologists, joined a task force formed by the General Coordination Office of the National Transplant System/Brazilian Ministry of Health (CGSNT-MS), the Brazilian Association of Intensive Care Medicine (AMIB), the Brazilian Association of Organ Transplantation (ABTO), and the Brazilian Research in Intensive Care Network (BRICNet). The questions were developed within the scope of the 2011 Brazilian Guidelines for Management of Adult Potential Multiple-Organ Deceased Donors. The topics were divided into mechanical ventilation, hemodynamic support, endocrine-metabolic management, infection, body temperature, blood transfusion, and use of checklists. The outcomes considered for decision-making were cardiac arrest, number of organs recovered or transplanted per donor, and graft function/survival. Rapid systematic reviews were conducted, and the quality of evidence of the recommendations was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system. Two expert panels were held in November 2016 and February 2017 to classify the recommendations. A systematic review update was performed in June 2020, and the recommendations were reviewed through a Delphi process with the panelists between June and July 2020.

RESULTS

A total of 19 recommendations were drawn from the expert panel. Of these, 7 were classified as strong (lung-protective ventilation strategy, vasopressors and combining arginine vasopressin to control blood pressure, antidiuretic hormones to control polyuria, serum potassium and magnesium control, and antibiotic use), 11 as weak (alveolar recruitment maneuvers, low-dose dopamine, low-dose corticosteroids, thyroid hormones, glycemic and serum sodium control, nutritional support, body temperature control or hypothermia, red blood cell transfusion, and goal-directed protocols), and 1 was considered a good clinical practice (volemic expansion).

CONCLUSION

Despite the agreement among panel members on most recommendations, the grade of recommendation was mostly weak. The observed lack of robust evidence on the topic highlights the importance of the present guideline to improve the management of brain-dead potential organ donors.

High-dose versus low-dose angiotensin converting enzyme inhibitors in heart failure: systematic review and meta-analysis

OBJECTIVE

To systematically review evidence comparing the effect of low-dose versus high-dose ACE inhibitors (ACEIs) on all-cause and cardiovascular mortality and hospitalisation, functional capacity and side effects in patients with heart failure (HF).

METHODS

We searched PubMed, Embase, Cochrane CENTRAL and LILACS up to January 2019. We included randomised controlled trials (RCTs) comparing low-dose versus high-dose ACEIs in adults with HF with reduced left ventricular ejection fraction (HFrEF). Study selection and data extraction were performed by two independent reviewers. Risk of bias was assessed with RoB 2.0, and quality of evidence with Grading of Recommendations Assessment, Development and Evaluation (GRADE). We conducted random effects meta-analysis and trial sequential analysis.

RESULTS

We included eight RCTs (5829 patients with HF). In comparison with low-dose ACEIs, high-dose ACEIs showed a non-significant effect on all-cause mortality (8 RCTs, n=5828, relative risk (RR) 0.95, 95% CI 0.88 to 1.02; moderate quality of evidence), cardiovascular mortality (6 RCTs, n=4048, RR 0.93, 95% CI 0.85 to 1.01; moderate quality of evidence), all-cause hospitalisation (5 RCTs, n=5394, RR 0.95, 95% CI 0.82 to 1.10; moderate quality of evidence) and cardiovascular hospitalisation (4 RCTs, n=5242, RR 0.98, 95% CI 0.83 to 1.17; low quality of evidence). High-dose ACEI increased functional capacity (4 studies, n=555, standardised mean difference 0.38, 95% CI 0.20 to 0.55; low quality of evidence) and the risk of hypotension (4 RCTs, n=3783, RR 1.64, 95% CI 1.30 to 2.05; moderate quality of evidence). High-dose ACEI had no effect on dizziness (3 RCTs, n=4994, RR 1.37, 95% CI 0.97 to 1.93; low quality of evidence), but decreased the risk of cough (4 RCTs, n=5146, RR 0.85, 95% CI 0.73 to 0.98; moderate quality of evidence).

CONCLUSIONS

The magnitude of benefit of using high dose versus low to intermediate doses of ACEIs might be less than traditionally suggested in clinical guidelines. These findings might help clinicians address the complex task of HF management in a more rational and timely fashion, saving efforts to implement strategies with the greatest net clinical benefit.

Quality assessment of prevalence studies: a systematic review

OBJECTIVES

The objective of the study is to identify items and domains applicable for the quality assessment of prevalence studies.

STUDY DESIGN AND SETTING

We searched databases and the gray literature to identify tools or guides about the quality assessment of prevalence studies. After study selection, we abstracted questions applicable for prevalence studies and classified into at least one of the following domains: 'population and setting', 'condition measurement', 'statistics', and 'other'. PROSPERO registration:CRD42018088437.

RESULTS

We included 30 tools: eight (26.7%) specifically designed to appraise prevalence studies and 22 (73.3%) adaptable for this purpose. We identified 12 unique items in the domain "population and setting", 16 in the domain "condition measurement", and 14 in the domain "statistics". Of those, 25 (59.5%) were identified in the eight specific tools. Regarding the domain "other", we identified 77 unique items, mainly related to manuscript writing and reporting (n = 48, 62.3%); of those, 24 (31.2%) were identified in the eight specific tools and 53 (68.8%) in the additional 22 nonspecific tools.

CONCLUSION

We provide a comprehensive set of items classified by domains that can guide the appraisal of prevalence studies, conduction of primary prevalence studies, and update or development of tools to evaluate prevalence studies.

Clinical practice guidelines in Brazil - developing a national programme

In Brazil, governmental and non-governmental organisations develop practice guidelines (PGs) in order to optimise patient care. Although important improvements have been made over the past years, many of these documents still lack transparency and methodological rigour. In order to conduct a critical analysis and define future steps in PG development in Brazil, we carried out a structured assessment of strengths, weaknesses, opportunities and threats (SWOT analysis) for the development of a national guideline programme. Participants consisted of academia, methodologists, medical societies and healthcare system representatives. In summary, the PG development process has improved in Brazil and current investments in methodological research and capacity-building are ongoing. Despite the centralised processes for public PGs, standardised procedures for their development are not well established and human resources are insufficient in number and capacity to develop the amount of trustworthy documents needed. Brazil’s capacity could be strengthened and initial efforts have been made such as the adoption of standards proposed by world-renowned institutions in PG development and enhancement of the involvement of key stakeholders. Further steps involve the alignment between health technology assessment and PG processes for synergy and the development of a national network to promote the interaction between groups involved in the development of PGs. The lessons learned from this paper could be used to foster debate on guideline development, especially for countries facing similar threats on this topic.

Guidelines for the pharmacological treatment of COVID-19. The task-force/consensus guideline of the Brazilian Association of Intensive Care Medicine, the Brazilian Society of Infectious Diseases and the Brazilian Society of Pulmonology and Tisiology

INTRODUCTION

Different therapies are currently used, considered, or proposed for the treatment of COVID-19; for many of those therapies, no appropriate assessment of effectiveness and safety was performed. This document aims to provide scientifically available evidence-based information in a transparent interpretation, to subsidize decisions related to the pharmacological therapy of COVID-19 in Brazil.

METHODS

A group of 27 experts and methodologists integrated a task-force formed by professionals from the Brazilian Association of Intensive Care Medicine (Associação de Medicina Intensiva Brasileira - AMIB), the Brazilian Society of Infectious Diseases (Sociedad Brasileira de Infectologia - SBI) and the Brazilian Society of Pulmonology and Tisiology (Sociedade Brasileira de Pneumologia e Tisiologia - SBPT). Rapid systematic reviews, updated on April 28, 2020, were conducted. The assessment of the quality of evidence and the development of recommendations followed the GRADE system. The recommendations were written on May 5, 8, and 13, 2020.

RESULTS

Eleven recommendations were issued based on low or very-low level evidence. We do not recommend the routine use of hydroxychloroquine, chloroquine, azithromycin, lopinavir/ritonavir, corticosteroids, or tocilizumab for the treatment of COVID-19. Prophylactic heparin should be used in hospitalized patients, however, no anticoagulation should be provided for patients without a specific clinical indication. Antibiotics and oseltamivir should only be considered for patients with suspected bacterial or influenza coinfection, respectively.

CONCLUSION

So far no pharmacological intervention was proven effective and safe to warrant its use in the routine treatment of COVID-19 patients; therefore such patients should ideally be treated in the context of clinical trials. The recommendations herein provided will be revised continuously aiming to capture newly generated evidence.

How are systematic reviews of prevalence conducted? A methodological study

Background

There is a notable lack of methodological and reporting guidance for systematic reviews of prevalence data. This information void has the potential to result in reviews that are inconsistent and inadequate to inform healthcare policy and decision making. The aim of this meta-epidemiological study is to describe the methodology of recently published prevalence systematic reviews.

Methods

We searched MEDLINE (via PubMed) from February 2017 to February 2018 for systematic reviews of prevalence studies. We included systematic reviews assessing the prevalence of any clinical condition using patients as the unit of measurement and we summarized data related to reporting and methodology of the reviews.

Results

A total of 235 systematic reviews of prevalence were analyzed. The median number of authors was 5 (interquartile range [IQR] 4–7), the median number of databases searched was 4 (3–6) and the median number of studies included in each review was 24 (IQR 15–41.5). Search strategies were presented for 68% of reviews. Forty five percent of reviews received external funding, and 24% did not provide funding information. Twenty three percent of included reviews had published or registered the systematic review protocol. Reporting guidelines were used in 72% of reviews. The quality of included studies was assessed in 80% of reviews. Nine reviews assessed the overall quality of evidence (4 using GRADE). Meta-analysis was conducted in 65% of reviews; 1% used Bayesian methods. Random effect meta-analysis was used in 94% of reviews; among them, 75% did not report the variance estimator used. Among the reviews with meta-analysis, 70% did not report how data was transformed; 59% percent conducted subgroup analysis, 38% conducted meta-regression and 2% estimated prediction interval; I² was estimated in 95% of analysis. Publication bias was examined in 48%. The most common software used was STATA (55%).

Conclusions

Our results indicate that there are significant inconsistencies regarding how these reviews are conducted. Many of these differences arose in the assessment of methodological quality and the formal synthesis of comparable data. This variability indicates the need for clearer reporting standards and consensus on methodological guidance for systematic reviews of prevalence data.

Prevalence of human papillomavirus (HPV) in Brazil: A systematic review and meta-analysis

OBJECTIVES

This study aimed to estimate the prevalence of genital, anal and oral HPV infection in Brazil through systematic review and meta-analysis.

METHODS

We searched EMBASE, LILACS, MEDLINE, Web of Science and SciELO from inception to December 2018. Original research articles that assessed the prevalence of genital (i.e., cervical, penile), anal and oral HPV infection in Brazil were selected in pairs by independent authors. No sex, age, HPV vaccination, language or date restrictions were applied. HPV prevalence was estimated and stratified according to risk factors population and by geographic area throughout the country. The study prevalence was pooled using a random effects model. Analysis was performed using R (version 3.5.2), packages meta version 4.9-4 and metaphor 2.0-0. This review is registered on PROSPERO under protocol number CRD42016032751.

RESULTS

We identified 3,351 references. After the screening process, 139 of them were eligible for this systematic review (57,513 total participants). Prevalence of cervical HPV was 25.41% (95% CI 22.71-28.32). Additionally, prevalence was 36.21% (95% CI 23.40, 51.33) in the penile region, 25.68% (95%CI 14.64, 41.04) in the anal region, and 11.89% (95%CI 6.26, 21.43) in the oral region. Subgroup analysis showed prevalence in each anatomic site was higher in high-risk populations.

CONCLUSION

The prevalence of HPV is high in the Brazilian population and varies by population risk and anatomic body site, with lower rates in the oral cavity compared to that in the cervical, penile and anal region. Studies on HPV have primarily been developed to evaluate infection and cancer in the cervical region. There is a profound lack of HPV data in many geographic regions of Brazil and for different anatomic sites.

Isosorbide and nifedipine for Chagas' megaesophagus: A systematic review and meta-analysis

BACKGROUND

Chagas disease is a neglected tropical disease. About 6 to 8 million people are chronically infected and 10% to 15% develop irreversible gastrointestinal disorders, including megaesophagus. Treatment focuses on improving symptoms, and isosorbide and nifedipine may be used for this purpose.

METHODOLOGY

We conducted a systematic review to evaluate the effectiveness of pharmacological treatment for Chagas' megaesophagus. We searched MEDLINE, Embase and LILACS databases up to January 2018. We included both observational studies and RCTs evaluating the effects of isosorbide or nifedipine in adult patients with Chagas' megaesophagus. Two reviewers screened titles and abstracts, selected eligible studies and extracted data. We assessed the risk of bias using NIH 'Quality Assessment Tool for Before-After (Pre-Post) Studies with No Control Group' and RoB 2.0 tool. Overall quality of evidence was assessed using GRADE.

PRINCIPAL FINDINGS

We included eight studies (four crossover RCTs, four before-after studies). Three studies evaluated the effect of isosorbide on lower esophageal sphincter pressure (LESP), showing a significant reduction (mean difference -10.52mmHg, 95%CI -13.57 to-7.47, very low quality of evidence). Three studies reported the effect of isosorbide on esophageal emptying, showing a decrease in esophageal retention rates (mean difference -22.16%, 95%CI -29.94 to -14.38, low quality of evidence). In one study, patients on isosorbide reported improvement in the frequency and severity of dysphagia (moderate quality of evidence). Studies evaluating nifedipine observed a decrease in LESP but no effect on esophageal emptying (very low and low quality of evidence, respectively). Isosorbide had a higher incidence of headache as a side effect than nifedipine.

CONCLUSIONS

Although limited, available evidence shows that both isosorbide and nifedipine are effective in reducing esophageal symptoms. Isosorbide appears to be more effective, and its use is supported by a larger number of studies; nifedipine, however, appears to have a better tolerability profile.

TRIAL REGISTRATION

PROSPERO CRD42017055143. ClinicalTrials.gov CRD42017055143.

Amiodarone for arrhythmia in patients with Chagas disease: A systematic review and individual patient data meta-analysis

Background

Chagas disease is a neglected chronic condition caused by Trypanosoma cruzi, with high prevalence and burden in Latin America. Ventricular arrhythmias are common in patients with Chagas cardiomyopathy, and amiodarone has been widely used for this purpose. The aim of our study was to assess the effect of amiodarone in patients with Chagas cardiomyopathy.

Methodology

We searched MEDLINE, Embase and LILACS up to January 2018. Data from randomized and observational studies evaluating amiodarone use in Chagas cardiomyopathy were included. Two reviewers selected the studies, extracted data and assessed risk of bias. Overall quality of evidence was accessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE).

Principal findings

We included 9 studies (3 before-after studies, 5 case series and 1 randomized controlled trial). Two studies with a total of 38 patients had the full dataset, allowing individual patient data (IPD) analysis. In 24-hour Holter, amiodarone reduced the number of ventricular tachycardia episodes in 99.9% (95%CI 99.8%-100%), ventricular premature beats in 93.1% (95%CI 82%-97.4%) and the incidence of ventricular couplets in 79% (RR 0.21, 95%CI 0.11–0.39). Studies not included in the IPD analysis showed a reduction of ventricular premature beats (5 studies), ventricular tachycardia (6 studies) and ventricular couplets (1 study). We pooled the incidence of adverse side effects with random effects meta-analysis; amiodarone was associated with corneal microdeposits (61.1%, 95%CI 19.0–91.3, 5 studies), gastrointestinal events (16.1%, 95%CI 6.61–34.2, 3 studies), sinus bradycardia (12.7%, 95%CI 3.71–35.5, 6 studies), dermatological events (10.6%, 95%CI 4.77–21.9, 3 studies) and drug discontinuation (7.68%, 95%CI 4.17–13.7, 5 studies). Quality of evidence ranged from moderate to very low.

Conclusions

Amiodarone is effective in reducing ventricular arrhythmias, but there is no evidence for hard endpoints (sudden death, hospitalization). Although our findings support the use of amiodarone, it is important to balance the potential benefits and harms at the individual level for decision-making.

Chagas disease is a chronic neglected tropical disease, with high prevalence and burden in Latin America. About 30% of chronically infected patients develop Chagas cardiomyopathy. Ventricular arrhythmias are common in patients with Chagas cardiomyopathy and treatment approaches include medications, resynchronization therapy, and implantable cardioverter defibrillator. Studies published from 1980 to 1990 have evaluated the effect of amiodarone. According to our systematic review and individual patient meta-analysis, amiodarone reduced ventricular tachycardia, ventricular premature beats and incidence of ventricular couplets. Although the strong evidence of clinical benefit with arrhythmia reduction, this information should be interpreted with caution, since arrhythmia is a surrogate outcome and since its clinical impact on death and hospitalization reduction over time is not clear. Little information was identified related to hard endpoints. Regarding side effects, our systematic review observed that amiodarone was associated with corneal microdeposits, gastrointestinal events, sinus bradycardia, dermatological events, pneumonitis, hypothyroidism and drug discontinuation. The currently available evidence shows that amiodarone seems to be an effective antiarrhythmic drug for patients with Chagas disease, especially in settings where an implantable cardioverter defibrillator is not available or affordable, but that a balance between potential benefits and harms at the individual level is needed.

Lifestyle factors, cardiovascular disease and all-cause mortality in middle-aged and elderly women: a systematic review and meta-analysis

Cardiovascular disease (CVD) risk factors, incidence and death increases from around the time of menopause comparing to women in reproductive age. A healthy lifestyle can prevent CVD, but it is unclear which lifestyle factors may help maintain and improve cardiovascular health for women after menopausal transition. We conducted a systematic review and meta-analysis of prospective cohort studies to evaluate the association between modifiable lifestyle factors (specifically smoking, physical activity, alcohol intake, and obesity), with CVD and mortality in middle-aged and elderly women. Pubmed, Embase, among other databases and reference lists were searched until February 29th, 2016. Study specific relative risks (RR) were meta-analyzed using random effect models. We included 59 studies involving 5,358,902 women. Comparing current versus never smokers, pooled RR were 3.12 (95% CI 2.15-4.52) for CHD incidence, 2.09 (95% CI 1.51-2.89) for stroke incidence, 2.76 (95% CI 1.62-4.71) for CVD mortality and 2.22 (95% CI 1.92-2.57) for all-cause mortality. Physical activity was associated with a decreased risk of 0.74 (95% CI 0.67-0.80) for overall CVD, 0.71 (95% CI 0.67-0.75) for CHD, 0.77 (95% CI 0.70-0.85) for stroke, 0.70 (95% CI 0.58-0.84) for CVD mortality and 0.71 (95% CI 0.65-0.78) for all-cause mortality. Comparing moderate drinkers versus non-drinkers, the RR was 0.72 (95% CI 0.56-0.91) for CHD, 0.63 (95% CI 0.57-0.71) for CVD mortality and 0.80 (95% CI 0.76-0.84) for all-cause mortality. For women with BMI 30-35 kg/m² the risk was 1.67 (95% CI 1.24-2.25) for CHD and 2.3 (95% CI 1.56-3.40) for CVD mortality, compared to normal weight. Each 5 kg/m² increase in BMI was associated with 24% (95% CI 16-33%) higher risk for all-cause mortality. This meta-analysis suggests that physical activity and moderate alcohol intake were associated with a reduced risk for CVD and mortality. Smoking and higher BMI were associated with an increased risk of these endpoints. Adherence to a healthy lifestyle may substantially lower the burden of CVD and reduce the risk of mortality among middle-aged and elderly women. However, this review highlights important gaps, as lack of standardized methods in assessing lifestyle factors and lack of accurate information on menopause status, which should be addressed by future studies in order to understand the role of menopause on the association between lifestyle factors and cardiovascular events.

Prevalence of papillomavirus in Brazil: a systematic review protocol

INTRODUCTION

Human papillomavirus (HPV) infection is a cause of premalignant and malignant cancer in the lower genital and digestive tracts. In Brazil, there have been no prevalence studies that included a nationwide sample, and the prevalence of HPV has not been determined in many regions.

METHODS

We will search the EMBASE, LILACS, MEDLINE, Web of Science and SciELO databases and previously published review articles to identify original research articles assessing HPV prevalence of the perineal (cervical, penile and anal) and oral areas. No exclusion criteria related to language or publication date will apply. 2 reviewers will independently screen for eligibility and perform data extraction. Discrepancies will be resolved through consensus; the opinion of a third reviewer will be sought as necessary. Relevant measures and data about study and population characteristics will be extracted from the included studies. Where possible, study prevalence will be pooled using a random-effects meta-analysis. The methodological quality of the studies will be assessed using an adapted version of the NIH 'Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies'. The overall quality of evidence will be assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE).

ETHICS AND DISSEMINATION

We expect to estimate the prevalence of perineal and oral HPV infection in the general population as well as the prevalence of HPV infection in individuals with premalignant and malignant lesions in Brazil and its 5 geographic regions. This systematic review does not require ethical approval.

TRIAL REGISTRATION NUMBER

CRD42016032751.

The Passo Fundo Cohort Study: design of a population-based observational study of women in premenopause, menopausal transition, and postmenopause

BACKGROUND

The Passo Fundo Cohort Study (PFS) is a population-based longitudinal observational study of pre-, peri-, and postmenopausal women that has been ongoing since 1995 in Passo Fundo, a city in southern Brazil. This paper describes the rationale and design of the PFS and summarizes objectives and procedures that have been updated during follow-up.

METHODS/DESIGN

Women in the PFS have been followed for a variety of diseases that are frequent in menopause. Sampling was conducted in 154 randomly selected census divisions (geographical subdivisions of the city as defined by the Brazilian Institute of Geography and Statistics). One block in each census division was chosen by lot and two women were randomly selected for interview in each block. The first cycle, conducted between 1995 and 1997, included a representative sample of 298 women aged 35 to 55 years. In the second cycle, conducted between 2001 and 2002, additional participants were enrolled based on the same sampling strategy used in 1995, for a final sample of 358 women. In 2010, a third follow-up was initiated, when all 358 participants or their relatives were located. Participants completed a standardized questionnaire on demographic and socioeconomic characteristics. They also answered questions about lifestyle, medical and reproductive characteristics, sexual life, hormone therapy and mental aspects by using validated instruments. Physical activity was assessed and anthropometric measurements, blood sampling and pelvic ultrasound examination were performed. In the third cycle, bone mineral density by dual-energy X-ray absorptiometry and abdominal fat and coronary artery calcium score by computed tomography were also determined.

DISCUSSION

The study findings provide relevant information to evaluate the association between menopausal status, female aging and the risk of cardiovascular diseases, and bone health aspects in a representative sample of women from southern Brazil.

Physical activity in climacteric women: comparison between self-reporting and pedometer

OBJECTIVE

To compare two methods of assessing physical activity in pre-, peri- and postmenopausal women.

METHODS

Cross-sectional study nested in a cohort of pre-, peri- and postmenopausal women in a city in Southern Brazil. The participants completed a questionnaire that included sociodemographic and clinical data. Physical activity was assessed using a digital pedometer and the International Physical Activity Questionnaire, short version. The participants were classified into strata of physical activity according to the instrument used. For statistical analysis, the Spearman correlation test, Kappa index, concordance coefficient and Bland-Altman plots were used.

RESULTS

The concordance (k = 0110; p = 0.007) and the correlation (rho = 0.136, p = 0.02) between the International Physical Activity Questionnaire, short version, and pedometer were weak. In Bland-Altman plots, it was observed that differences deviate from zero value whether the physical activity is minimal or more intense. Comparing the two methods, the frequency of inactive women is higher when assessed by pedometer than by the International Physical Activity Questionnaire--short version, and the opposite occurs in active women.

CONCLUSIONS

Agreement between the methods was weak. Although easy to use, Physical Activity Questionnaire--short version overestimates physical activity compared with assessment by pedometer.

Causes of death and associated risk factors among climacteric women from Southern Brazil: a population based-study

BACKGROUND

Aging and menopause are particular cardiovascular risk factors for women, due to estrogen deprivation at the time of menopause. Studies show that diabetes mellitus (DM), smoking, hypertension, high body mass index (BMI), and serum lipids are associated with increased risk of cardiovascular disease (CVD), the main cause of female mortality in Brazil. The aim of this study was to assess the mortality rate, causes of death and associated risk factors in a cohort of women from Brazil.

METHODS

A longitudinal population-based study of menopausal status is currently underway in a city in South Brazil. In 2010, a third follow-up of this population was performed to assess cardiovascular risk and mortality rate between 1995 and 2011. For this analysis, 358 participants were studied. At baseline, participants had completed a standardized questionnaire including demographic, lifestyle, medical and reproductive characteristics. In addition to the contacts with relatives, mortality data were obtained through review of medical records in all city hospitals and the Center for Health Information (NIS/RS-SES). Multivariate-adjusted hazard risk (HR) and 95% confidence intervals (CI95%) were estimated using Cox proportional hazards regression. Survival curves were estimated using the Kaplan-Meier curve.

RESULTS

There were 17 (4.7%) deaths from all causes during the study period. Seven (41.2%) deaths were caused by CVD, including four cases of stroke and three cases of myocardial infarction. Six (35.3%) deaths were due to cancer, and four (23.5%) were due to other reasons. In the age and smoking-adjusted multivariate models, diabetes (HR 6.645, 95% CI: 1.938-22.79, p=0.003), alcohol intake (HR 1.228, 95% CI: 1.014-1.487, p=0.035) and postmenopausal status (HR=6.216, 95% CI: 0.963-40.143, p=0.055) were associated with all-cause mortality. A significant association was found between abdominal obesity (WHR≥0.85) and mortality even after the adjustment for BMI (HR=9.229, 95% IC: 2.083-41.504, p=0.003).

CONCLUSION

CVD was an important cause of mortality in this cohort and DM and/or central adiposity were associated with all-cause mortality. Lifestyle and dietary factors seem to be related to risk of mortality in middle-aged women.