Terapia nutricional nas alterações metabólicas em pessoas vivendo com HIV / aids Nutritional therapy in metabolic changes in individuals with HIV / AIDS

MÉTODOS: Revisão sistemática de literatura no PubMed, Lilacs e Cochrane, entre 1996 e 2010, do tipo ensaio clínico, controlado, randomizado, crossover, adultos, vivendo com HIV/aids em uso de terapia antirretroviral e sem doenças oportunistas. A intervenção de interesse foi suplementação nutricional via oral e/ou mudança de estilo de vida por tratamento dietoterápico específi co: dislipidemia, resistência insulínica, lipodistrofia e hipertensão arterial sistêmica. A escala de Jadad foi utilizada para classifi cação qualitativa dos artigos.


INTRODUCTION
The World Health Organization (WHO) estimated there were 33.4 million people infected with HIV in 2009, of which 2.7 million new cases, and 2 million were aids-related deaths.a The availability of highly active antiretroviral therapy (HAART) has had a major impact on HIV/AIDS mortality and morbidity in recent years.However, HAART has been associated with metabolic adverse events characterized by dyslipidemia, body composition changes / lipodystrophy, insulin resistance / glucose intolerance and hypertension.Living with HIV/AIDS has become a condition similar to other non-communicable chronic diseases requiring lifestyle changes and drug management for prevention of cardiovascular events, among others.b The WHO recommends that nutrition interventions should be part of all HIV/AIDS control and treatment programs because adequate diet and nutrition can improve adherence to antiretroviral therapy (ART) and its effectiveness, as well as help tackling metabolic abnormalities in people living with HIV/ AIDS.c However, there is no consensus on the effect of nutrition therapy in people living with HIV/AIDS (PLWHA) on ART.
The present study aimed to assess the effect of nutrition therapy on improving metabolic changes caused by the use of ART.

METHODS
Systematic literature review using a search engine especially designed by the authors but not recorded in specifi c databases for systematic review.The following aspects were predefi ned: topic; inclusion criteria; search and selection strategies; quality assessment; data collection and analysis form; and presentation and interpretation of study results.The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 12 recommendations were used as an aid tool.There were selected cross-over, randomized, controlled, blinded or open, clinical trials published between 1996 and November 2010, conducted in both female and male adults living with HIV/AIDS on ART and with no manifestations of opportunistic diseases.The lower data limit for published studies was set based on the year of introduction of ART.Studies in English, Spanish, Portuguese and French were selected.
Interventions of interest were nutrition therapies aiming to improve metabolic abnormalities through oral supplementation or specifi c dietary therapies plus lifestyle changes.Oral supplementation regardless of dosage, dosing interval and duration of treatment / intervention and interventions tested against placebo or no intervention were assessed.The following outcomes of these interventions were studied: dyslipidemia, insulin resistance or glucose intolerance; lipodystrophy / body composition changes; and hypertension.Outcome measures in mmol/L, g/L, and mg/dL were converted into mg/dL according to the International System of Units to make them consistent and comparable.
Subjects requiring enteral or parenteral nutrition were not included in the study because they had different nutritional needs and required different approaches than those used for metabolic changes and they were likely to have opportunistic infections.
The following was reviewed and analyzed in the articles selected: author; year of publication; country; follow-up duration; methods; type of population; sample size; type of intervention; comparability between intervention and control groups at baseline; quality of outcome measurements; outcome results in intervention and control groups; treatment adherence; and adverse effects.The Jadad scale 9 was used to independently and blindly assess methodological quality of studies.Two researchers assigned scores (zero to fi ve) to the studies selected based on the following criteria: method of randomization (sequences and subject randomization criteria); use of blinding (subjects and investigators); and description of lost-to-follow-up proportions.
Two independent blinded researchers conducted the process of searching and selecting articles to avoid selection bias.When there was no agreement whether an article should be included or not in the study, this decision was made by a third researcher.The authors were not contacted at all.The literature search was performed in PubMed, LiLACS, and Cochrane electronic databases (The Cochrane Central Register of Controlled Trials -CENTRAL) using a combination of three actions:

-[(life and style and intervention or life and style and supplementations or nutritional and support or nutrition and therapy or exposure and dietetics or dietary and intake or food or food and consumption or eating or energy and intake or nutrient or diet)] AND [(lipodystrophy or hiv and associated and lipodystrophy and syndrome or body and fat and distribution or body and fat and redistribution or lipoatrophy or lipohypertrophy or body and composition or hypertension or insulin and resistance or glucose and intolerance or diabetes and melittus or dyslipidemias or hypertriglyceridemia or hyperlipidemia or hypercholesterolemia or metabolic and syndrome or metabolic and abnormalities)] AND [(antiretroviral and therapy, and highly and active or haart)] AND Species=Humans.
Search fi lters ("humans" and "all adults") were used to increase specifi city.References in review and consensus articles and in articles retrieved using the search strategy were searched manually to ensure the inclusion of all articles relevant to the topic.No contact was made with medical researchers to check for ongoing research.

RESULTS AND DISCUSSION
A total of 385 non-duplicate articles were retrieved.Seven were included after eligibility criteria were applied (Figure).
The articles retrieved while searching for insulin resistance were the same ones retrieved while searching for glucose intolerance.
Of the seven studies included, four were conducted in the United States, 8,17,18 two in Canada, 1,2 one in France, 16 and one in Australia. 4They were published from 2005 to 2009 (Table 1).
Diet, diet and exercise, diet and supplementation and supplementation only were the most common nutrition interventions used.Two studies 8,18 used diet as study intervention.One 8 followed the National Cholesterol Education Program-Adult Treatment Panel Guidelines (NCEP III) 4,d and the other one 18 had no reference.Diet was used in an attempt to equate the intervention group (IG) and controls before implementing the intervention being tested. 4,16Food supplements included chromium nicotinate and omega 3. Omega 3 was tested in fi ve studies, with doses ranging from 3 g to 9 g/day (Table 1).
The number of subjects included in each group (IG and controls) ranged from fi ve to 62.All studies but one analyzed data from those who completed the study (completers).Truchis et al 16 (2007) also included an intention-to-treat analysis.Lost-to-follow-up was reported in six studies. 1,2,4,8,16,17The follow-up period ranged from 13 weeks to six months (Table 1).
Two were randomized, controlled and double blind studies, 1,4 two were randomized cross-over, 2,16 and three were randomized controlled trials. 8,17,18(Table 1).Intraevaluator and interevaluator reliability as well as measures of association were not reported in the articles.Dyslipidemia was reported in all studies, 1,2,4,8,16-18 lipodystrophy / body composition changes in two, 1,8 and insulin resistance / glucose intolerance 1 and hypertension 8 in one (Table 1).
The quality assessment of studies using the Jadad scale 9 showed that all had good methodological quality: two scored fi ve, 1,16 one score four, 4 and four score three. 2,8,17,18Most studies did not use blinding.The three double-blind studies 1,4,16 used supplementation as nutrition intervention, including placebo and blinding; one of them 4 did not provide information on the method of randomization.
The effects of each intervention by the outcomes dyslipidemia, body composition changes / lipodystrophy, insulin resistance / glucose intolerance and hypertension are presented in Tables 2, 3 and 4.

Effects on dyslipidemia
.4,16-18 Omega-3 fatty acids can promote a reduction in plasma triglycerides by decreasing hepatic synthesis of VLDLcholesterol in the general population and may have other cardiovascular effects.At high doses (4 to 10 g/ day), they can reduce triglycerides and slightly increase HDL-cholesterol, but may increase LDL-cholesterol.Omega-3 can be administered as adjuvant therapy in patients with hypertriglyceridemia, 6 but its effects on PLWHA on ART is unknown.
Woods et al 18 (2009) conducted an intervention study using more specific dietary supplementation with omega-3 (3 g/day).They found more significant reductions in triglycerides in IG than controls.HDL-cholesterol signifi cantly increased in IG at the end of the study compared to baseline, but there was no signifi cant difference between IG and controls.Total cholesterol in IG was signifi cantly lower compared to baseline and was lower than controls at week 3 of a 13-week follow-up.No signifi cant differences were seen in LDL-cholesterol.
Supplementation with omega-3 (3 g/day) was also used in Baril et al study. 2 There was a signifi cant reduction in triglycerides in IG compared to controls in the fi rst phase of the study, but no signifi cant reduction compared to baseline.Total cholesterol was reduced in IG.There was no significant change in either LDL-cholesterol or HDL-cholesterol.
Dietary changes were not evaluated, although all subjects were instructed not to change their diet and level of physical activity.A limitation of this study is its open design, though the authors claimed the study scenario matched a real-life one where patients and physicians are aware of their treatment.
Truchis et al 16 (2007) conducted a randomized study in patients with hypertriglyceridemia following a 4-week diet as proposed by the American Heart Association. 10IG received omega-3 (6 g/day).There was a signifi cant reduction in triglycerides in IG compared to controls and also a signifi cant reduction in the fi rst phase of the study compared to baseline in the intention-to-treat analysis and completers.There was also seen a reduction in triglycerides in controls in the open phase of the study compared to the last double-blind phase.There was no statistical difference for total cholesterol and HDL-cholesterol.
Carter et al 4 (2006), in an attempt to equate adult male patients, provided them with dietary counseling following the National Heart Foundation guidelines 13 for six weeks and then they were randomized to receive omega-3 (9 g/day).A signifi cant reduction in triglycerides was seen in IG compared to controls.There was no signifi cant difference in total cholesterol between the two groups.Although it was a randomized, controlled, double-blind study, the sample size was small.
Wohl et al 17 (2005) assessed supplementation with omega-3 (3 g/day) for 16 weeks.All subjects (IG and controls) were instructed to reduce their consumption of trans fatty acids, increase their fi ber intake and engage in a daily aerobic exercise program.There was a significant mean reduction in triglycerides in IG compared to baseline at week 4 to week 16.At week 16, there was signifi cant reduction in triglycerides in IG compared to controls.LDL-cholesterol signifi cantly increased in IG compared to baseline.Total cholesterol and HDL-cholesterol did not change signifi cantly in either group.7][18] This reduction was not dependent on the supplementation dosing (3, 6 or 9 g/day), dieting and exercising.The only study using a supplementation dose of 9 g/day of omega-3 showed the greatest reductions in triglycerides. 4 While it evidences a positive effect of supplementation on triglycerides, the amount of supplementation prescribed (9 g/day) may hinder adherence to treatment.Thus, the ability of patients to tolerate it in high doses in the long run remains unknown.
Patients on lipid lowering drugs were included in the sample but no analysis was carried out to differentiate the effect of the drug from the nutrition intervention, which may have been a major bias.The same was seen in two other studies. 2,17ere is no evidence supporting an effect of specifi c diets or dietary counseling and exercise on hypertriglyceridemia in PLWHA on ART.Wohl study 17 (2005) investigated nutritional counseling, recommendations on physical activity, and supplementation with omega-3, and found the greatest reduction in triglycerides among studies investigating the use of omega-3 at a dose of 3 g/day.A set of actions including a specifi c diet, daily exercise, and supplementation with omega-3 may be the more suitable approach for treating hypertriglyceridemia in these patients.
7][18] Two studies 2,18 found a signifi cant reduction.One of them 2 found a reduction compared to baseline.Omega-3 may reduce total cholesterol, but the results are inconsistent.Of three studies 2,17,18 evaluating LDL-cholesterol, two 2,18 did not fi nd any signifi cant reductions and one 17 found a signifi cant increase compared to baseline.
7][18] A study 18 showed a signifi cant increase in HDL compared to baseline.In this study, in addition to omega-3 a specifi c diet low in total and saturated fat, high in fi ber and omega-3 and with low glycemic index was prescribed as part of the intervention. 18gh levels of triglycerides was an eligibility criterion in some studies; in one study, these criteria included high levels of triglycerides or abnormal quantitative insulin sensitivity check index. 18Except for one study, 4 all studies did not fi nd any signifi cant differences between the intervention group and controls at baseline.Two 4,16 of fi ve studies 2,4,[16][17][18] assessing the effect of supplementation with omega-3 on dyslipidemia included placebo in controls.These same two studies provided nutritional counseling before supplementation in an attempt to equate the intervention and control groups.In another study, subjects (IG and controls) were provided with counseling on diet and exercise throughout follow-up. 17ere is no evidence to indicate that any treatment has signifi cant effects on reducing total cholesterol and LDL-cholesterol and increasing HDL-cholesterol.Further studies are needed, but it has been suggested that a customized food plan with specifi c recommendations on dyslipidemia is the most effective approach to reduce total cholesterol and increase HDL-cholesterol because it has no unfavorable, deleterious effects.
Supplementation with chromium nicotinate 400 mg/ day was evaluated in a 16-week study with signifi cant reduction in triglycerides in IG compared to baseline. 1 Chromium is an essential nutrient of glucose and lipid metabolism as it enhances insulin action. 5e authors pointed to their study limitations including a heterogeneous sample of patients regarding to metabolic parameters, duration of HIV infection, duration of HAART and other drug therapies, as well as inaccuracies in evaluating dietary chromium intake.The population studied included predominantly male subjects, which may have been a bias in the study.
There were not evaluated any differences in the effect of the intervention between IG and controls at the end of the study.Statistical analyses were performed only in each individual group at baseline and at the end of the study and no intention-to-treat analysis was performed.
Supplementation with chromium showed signifi cant results, but they were not quite remarkable when compared to the reductions seen with omega-3.A single study investigating supplementation with chromium is not enough to support whether chromium should be used or not.
One intervention study included lifestyle changes with no supplementation. 8This study comprised patients with metabolic syndrome were randomly distributed between IG and controls.The IG received counseling on healthy diet and physical activity throughout the study.Sociodemographic and clinical characteristics of subjects in both groups were compared at baseline and no signifi cant differences were found.After 16 weeks of follow-up, there were no signifi cant effects on triglycerides, total cholesterol, HDL-cholesterol and LDL-cholesterol.
The study limitations were small sample size and short follow-up, which may have affected the results.Unlike studies that evaluated the effect of supplementation with omega-3 on dyslipidemia, this study 8 did not include high levels of triglycerides as an eligibility criterion.

Effect on body composition and lipodystrophy
Lipodystrophy was fi rst reported in 1998, two years after the introduction of ART. 3 Early controlled studies investigating how to minimize metabolic abnormalities caused by HAART dated back in 2006.Lipodystrophy is characterized by changes in body fat distribution including central lipohypertrophy, peripheral lipoatrophy and mixed lipodystrophy. 3,7,11pplementation with chromium nicotinate significantly reduced total body fat and measures of body fat in the trunk, arms and legs compared to baseline. 1 There was an increase in total lean mass; however, a reduction in the percent of fat in the arms and legs did not result in any improvement in lipodystrophy.There was no signifi cant reduction in waist circumference (WC).
Although IG and controls had similar physical activity profi les, changes in physical activity can affect body fat outcomes.This aspect was not investigated at the end of a 16-week follow-up.There is no clear evidence supporting the use of chromium nicotinate to treat lipodystrophy caused by ART.
Fitch et al 8 (2006) conducted a 6-month follow-up study with a lifestyle intervention including diet and exercise.Subjects in the IG had a signifi cant reduction in WC and lipodystrophy compared to controls.They showed signifi cant lifestyle changes compared to baseline while controls did not show signifi cant changes.
The analyses between IG and controls at baseline and at the end of the study and the analysis of differences between IG and controls at the end of the study showed better results in IG than controls.One of the limitations of this study is the use of a wide variety of ARTs.
Based on the findings of the present review, it is suggested lifestyle changes including diet and exercise to reduce lipodystrophy and WC.

Effect on insulin resistance
Insulin resistance can be caused by weight gain, changes in body fat distribution and use of retroviral protease inhibitors.As only a single study evaluated insulin resistance no recommendations can be made on nutrition interventions and/or supplementation to manage insulin resistance.

Effect on hypertension
An intervention study carried out by Fitch et al 8 (2006) evaluated blood pressure in PLWHA on ART.The intervention was lifestyle change including a specifi c diet for metabolic syndrome control and physical activity.
There was a signifi cant reduction in systolic blood pressure in IG compared to controls.
The Sixth Brazilian Guidelines on Hypertension recommend that hypertension treatment include lifestyle changes. 15This recommendation seems applicable to PLWHA on ART as lifestyle changes showed favorable results, even if only in systolic blood pressure.However, no further conclusions can be draw as only a single study was reviewed.

FINAL CONSIDERATIONS
One of the limitations of the present review is that only a small number of clinical trials were selected to assess the effect of nutrition therapy in metabolic changes caused by the use of HAART.In addition, eligibility criteria for individuals with metabolic changes were inconsistent across the studies reviewed.The studies used different interventions and doses, making it difficult a comparison between studies.Some studies did not meet key requirements for producing high-quality scientifi c evidence, especially blinding.However, it is well-known that blinding can be diffi cult to achieve in clinical trials including nutrition interventions with a diet plan or counseling on lifestyle changes.
There are little evidence of nutrition therapy for managing metabolic abnormalities in PLWHA on ART.
The results are controversial and scientifi c information available is inconclusive.
There is strong evidence showing reduction in triglycerides after supplementation with omega-3.The prescription of a specifi c diet plus supplementation with omega-3 was the single intervention that resulted in increased levels of HDL-cholesterol, but this intervention was evaluated in a single study and requires further investigation.
No inferences can be made regarding the best intervention for reducing total cholesterol and LDL-cholesterol.Lifestyle changes including physical activity and healthy eating habits seem to help improving lipodystrophy and hypertension.Insulin resistance may be managed with supplementation with chromium nicotinate.
Randomized controlled studies are needed to further elucidate the effects of nutrition therapy on dyslipidemia, lipodystrophy, insulin resistance, and hypertension in PLWHA on ART.The scientifi c evidence available supports current recommendations for treating these abnormalities in non-specifi c population groups.Further research can help develop more reliable treatments for these patients and reduce the risk of cardiovascular morbidity and mortality and increase these patients' life expectancy.

dFigure .
Figure.Flowchart of identifi cation, retrieval, eligibility, and selection of clinical studies for review.
c 35% lipid diet (less than 10% saturated fat, more than 10% polyunsaturated fat with emphasis on the consumption of omega-3 plus more than 20% of monounsaturated fat, 25 g to 35 g of soluble and insoluble fi ber).

Table 2 .
Results of clinical trials of nutritional intervention for dyslipidemia.2005-2009.
a All subjects received the intervention in the open phase of the study.TG: triglycerides; IG: intervention group; CG: control group; TC: total cholesterol; HDL: high-density lipoprotein; LDL: low-density lipoprotein; NS: not signifi cant; NA: not assessed.

Table 3 .
Results of clinical trials of nutritional intervention on body composition changes and lipodystrophy.2005-2009.

Table 4 .
Results of clinical trials with dietary intervention for insulin resistance and hypertension.2005-2009.