Diet Quality of patients with chronic Chagas disease in a tertiary hospital : a case-control study

Introduction: Nutritional status has been implicated in the modulation of the immune response, possibly augmenting the pathogenesis of Chagas disease (Cd). We evaluated diet quality and nutritional status in adults and elderly patients with chronic Cd in a tertiary hospital. Methods: A case-control study of Cd patients was conducted, paired for gender, age, and co-morbidities with non-Cd patients. Anthropometric measurements and food frequency questionnaire was used, and diet quality was assessed by the Brazilian Healthy Eating Index-Revised (BHEI-R). The Estimated Average Requirement cut-off points were used to determine the dietary micronutrient adequacy. The Cd group was further grouped according to Los Andes classification. Results: The study participants were 67 ± 10 years old, 73.6% elderly and 63% female. The prevalence of overweight/obesity and abdominal fat was high in both groups; however, Cd group showed a lower prevalence of obesity and increased risk of disease according to waist circumference classification. There was no difference in BHEI-R score between groups (p=0.145). The Cd group had sodium and saturated fat intake above recommendations and low intake of unsaturated fat, vitamin D, E, selenium, magnesium, and dairy products; but higher intake of iron. According to Los Andes classification, group III presented lower intake of whole fruit and dietary fiber. Conclusions: Patients with Cd were overweight and the quality of their diet was unsatisfactory based on the recommended diet components for age and sex.


INTRODUCTION
Chagas disease (Cd) is caused by the parasite Trypanosoma (T.) cruzi, and is the third most important tropical infection in the world.Despite a substantial reduction in the number of T. cruzi infected individuals worldwide, Cd remains neglected by the media and politicians both nationally and internationally 1 .Cd is associated with poverty, marginalization, and social vulnerability.It has a major adverse impact on health, quality of life, and social economic development, particularly in lowincome and developing countries 2,3 .
The chronic phase of Cd has a variety of clinical presentations, usually beginning with the indeterminate form.About 30-40% of patients will develop lesions 10-30 years after infection on different organs; mainly the heart, the digestive system, or both; leading to the cardiac, digestive, and cardiodigestive forms of the chronic disease, according to former studies (contemporary natural history information is scarce) 4,5 .
Digestive lesions involving the esophagus and colon, or both occur in about 10-15% of chronically infected patients 6 .These abnormalities usually do not lead to a reduction in life expectancy and patients have a favorable prognosis, low morbidity, the same mortality as the general population, and they are capable of doing any type of activity 7 .Despite a good prognosis in patients with gastrointestinal dysfunction, epidemiological studies in endemic areas have shown that malnutrition can occur with the progression of the disease 8 .The effect of Cd in the digestive tract of affected individuals influences the dynamics of swallowing and can lead to changes in nutritional status 9,10 .Adequate nutrition can reduce the risk of colon infection, especially foods with antioxidant compounds 11 .However, nutritional imbalances affect the ability of inflammatory response to protect the host 12 leading to impairment of immune defenses, such as phagocytic function, cell-mediated immunity, and complement system, secretion of antibodies, cytokines production and function.Low levels of body antioxidant nutrients promote cell immunosuppression and may intensify the severity of infection and worsen its evolution 13 .
As such, there is a synergistic interaction between worsening nutritional status which contributes negatively to the development and evolution of infection, and the infection; leading to a worsening nutritional status 14 .Nevertheless, the investigation of Cd and food intake in human populations are rare.The evaluation of food and nutrients intake in Cd is important for the understanding of infection and for the formulation of strategies for prevention and control of Cd.The purpose of this study was to assess the diet quality and nutritional status in adults and elderly patients with chronic Cd in a tertiary hospital.

METHODS
The study was conducted between July 2015 and February 2016 at the Clementino Fraga Filho University Hospital (HUCFF/UFRJ), Rio de Janeiro, Brazil.A case-control approach was used.The Cd group was composed of patients that were out of the endemic area for over 20 years and, with an etiologic diagnosis of Cd (two different serological tests with the positive reaction to T. cruzi).The selection of patients for the non-infected group was performed from among the outpatients of HUCFF/UFRJ Cardiology Service matched by sex, age, and co-morbidities (hypertension, diabetes mellitus type 2, cerebrovascular accident, and dyslipidemia).Exclusion criteria were unable to accept meals orally, clinical suspicion or diagnosis of liver disease, oncology patients, those with neurological problems, or in the immediate postoperative period (up to 30 days post-surgery).

Ethical considerations
The study protocol was approved by the Research Ethics Committee of the Clementino Fraga Filho Hospital of the Federal University of Rio de Janeiro (CEP-HUCFF-UFRJ), under the protocol number CAAE 46502615.1.0000.5257,and all patients who agreed to participate signed a consent form.
Both the case and control groups underwent clinical assessment, anthropometric measurement [weight (kg), height (m) 15 and waist circumference (cm) 16 ], calculation of body mass index (BMI) 17,18 , and assessment of food intake and diet quality.Weight, height, and waist circumference were measured three times by a trained nutritionist.The classification of BMI considered a BMI ≤18.5 as underweight, BMI ≥18.5 and <25.0 as normal, BMI ≥25.0 and <30 as overweight, and ≥30.0 as obesity 17 .We used the recommended sex-specific cut-off points for waist circumference: 94cm (men) and 80cm (women) for increased risk, and 102cm (men) and 88 cm (women) for substantially increased risk.The waist circumference was considered as an indicator of disease risk for type 2 diabetes, hypertension, and CVD using cut off points determined by the World Health Organization (WHO) 16 .The assessment of food intake was carried out by the Food Consumption Frequency Questionnaire (FFQ) 19 previously validated for an adult population.A trained nutritionist applied the FFQ with a support instrument help (photographic material) 20 .The data was entered once by a researcher and reviewed by a second researcher.The calculation of the diet nutritional value was conducted by analyzing FFQ with Food Processor® program.
The diet quality was assessed by the revised Brazilian Healthy Eating Index (BHEI-R) 21 .This index is composed of 12 items that feature distinct aspects of a healthy diet.They are: Total fruit; Whole fruit; Total vegetables and legumes; Dark green and orange vegetables and legumes; Total grain; Whole grains; Milk; Meats, eggs and legumes; Oils; Saturated fat; Sodium; Solid fat, added sugar and alcohol.Each component was scored at 0, 5, 10 or 20 points, with intermediate values calculated in proportion to the foods or nutrients consumed.The maximum score of BHEI-R is 100 points and the higher the score the better the quality of the diet.The item Whole grains of BHEI-R was not used in this study as the FFQ does not distinguish between the types of grains consumed.Instead, ten points were awarded to three servings of grains to 1.000kcal as a criterion for the highest score in Total grains.Consequently, the BHEI-R in this study consisted of eleven component scores.Information on the energy value, saturated fat, monounsaturated fat (MUFA), polyunsaturated fat (PUFA), trans fat, sodium, and addition of sugar to each food consumed to calculate the BHEI-R were collected from the Brazilian Institute of Geography and Statistics tables 22 ; and for other foods, the Food Processor Plus® software (ESHA Research, USA), which consists of a complete food composition table developed by the US Department of Agriculture (United States Department of Agriculture) was used 23 .
The Estimated Average Requirement (EAR) cut-off points according to sex and age were used to determine the dietary micronutrient adequacy 24 .
Participants with Cd were divided into four groups according to the Los Andes classification: I-A, I-B, II, and III.Group I-A had normal electrocardiogram and echocardiogram -no heart involvement, group I-B included patients at an early stage of cardiac involvement, group II patients were at an advanced stage of cardiac involvement without heart failure (HF), and group III patients were at an advanced stage of cardiac involvement with HF 25 .
Quantitative variables were expressed as means and standard deviation or 95% confidence interval (CI) for continuous variables with normal distribution, and percentages for categorical variables.McNemar test was used to compare categorical variables by pairing non-Cd and Cd groups.The comparison of the continuous variables was done using paired Student t test.ANOVA test and Bonferroni post-test were used for comparing the Los Andes groups categorized into Cd group, while the Pearson's chi-square test was used for comparison of frequencies among Los Andes groups.Statistical Package for the Social Sciences (SPSS, version 20.0) software was used for analyses and p values <0.05 were considered significant.

RESULTS
Study participants were selected from a cohort of 158 Cd patients who were actively and regularly followed-up.Fifty-six patients declined participating in this study due to an incompatible work day (31), patient's (10), and family (15)  decisions; while 21 were excluded because they had one or more exclusion criteria.A final sample of 81 each of infected (with an etiological diagnosis of Cd) and non-infected (controls) from the HUCFF Cardiology Service agreed to participate in the study Table 1 presents the general characteristics of individuals.The mean age was 67 ± 10 years (range, 38 to 89 years), 63.0% of participants were female and 73.6% were elderly.There was a low prevalence of hypertension (24.7%), dyslipidemia (7.4%), diabetes (6.2%), and stroke (12.3%) as documented in medical records.Most of the participants were overweight in both groups, but the prevalence of obesity was lower in Cd group (p= 0.038).Waist circumference showed different distribution between Cd and non-infected groups.The substantially increased risk for diseases was lower in Cd group (p<0.001)(Table 1).There was no difference between the Los Andes groups in the general characteristics, BMI classification, and waist circumference classification.
The mean educational level in Cd group was 3.4 (± 3.2) years and family's income level was low (<2 salary/months).Most of the Cd patients migrated from Bahia in 24.5% of cases, followed by 22.5%, 14.3%, and 12.0% from Minas Gerais, Paraíba, and Pernambuco, respectively.All the patients had nutrition counselling.
The estimated energy intake was lower in Cd group (BHEI-R average score; 80.8±5.3 points) compared to the non-infected group (82.1±6.1 points) (p=0.154)(Table 2).Regarding BHEI-R components, Cd group showed a higher consumption of vegetables and sodium, and lower consumption of cereals, milk and dairy products, meat and saturated fats than the noninfected group.The Los Andes groups' analysis showed lower intake of whole fruit by patients at an advanced stage with HF compared to the groups with early cardiac involvement.Less than 50% of the groups presented with adequate intake of milk and dairy, saturated fat, and sodium, see Figure 1.
The intake of carbohydrates, lipids, and fibers were higher in the Cd group compared to the non-infected group (p<0.001)(Figure 2).It is noteworthy that there was a lower intake of trans fatty acids by the Cd group (p<0.0001) and low intake of MUFA and PUFA.There was a greater variation in the distribution of the dietary variables in the non-infected group.The dietary intake of vitamins showed lower intake of vitamins A, D, E, magnesium, and selenium in the Cd group.Although, vitamin E intake was less than the recommendation in both groups, values were even lower in the Cd group.A greater variation in the distribution of vitamins and minerals variables was evident in the non-infected group (Figure 2).
Table 3 shows the intake of macronutrients, dietary fiber, vitamins, and minerals according to Los Andes groups.There was a decrease in the dietary fiber intake of Los Andes groups II and III compared to IA and IB groups.

DISCUSSION
The present work evaluated the diet quality of Cd and non-infected groups by BHEI-R, as well as the intake of macronutrients, dietary fiber, and adequacy of micronutrients.
It is the first to date to evaluate food intake of patients with Cd, using a method that measures not only the energy and nutrients intake but also the quality of diet.The food intake profile of patients with Cd was different from non-infected patients.The Cd group showed a lower intake of energy, vitamins A, D, and E, magnesium, and selenium, lower BHEI-R score for total grain, milk, meats and sodium, and a higher BHEI-R score for saturated fat.Los Andes group III presented a lower intake of whole fruit and dietary fiber.Another important finding of this study was the high prevalence of overweight and abdominal fat in both groups, although Cd group showed a lower prevalence of obesity and increased risk of disease for type 2 diabetes, hypertension, and CVD according to waist circumference classification.
We observed the cohort effect when considering the mean age of patients.Similarly, Viotti et al 26 found Cd patients to be around 66 years, similar to our study.In studies carried out in the 1960s and 1970s [27][28][29] the mean age of patients with Cd was ≤25 years.The progressive increase in the mean age of patients over the years is referred to as the cohort effect 30 .In turn, studies from the 1990s tend to include adults ≥40 years and older (adult-elderly).Knowledge of the natural history of Cd anticipates impairment of organs with increasing age, due to the slowly progressing nature of Cd.However, in this study and in other urban series 26,31 , progression may already have occurred in study participants by virtue of their age and may explain the fewer co-morbidities reported by medical records in the Cd group.
The digestive form of Cd occurs in about 10-15% of chronically infected patients.Contemporary natural history information is scarce in relation to the digestive form.The HUCFF at Federal University of Rio de Janeiro/Brazil is a reference center dedicated to treatment and research of esophagopathy and colopathy in Cd, hence, a high esophagus and colon diseases are expected.The Cd and the cohort at HUCFF include patients with early involvement of the esophagus/colon according to Cabral et al. 32 However, this study was limited in that it was not possible to evaluate esophagus/colon disease in all patients in this study, because the most accurate diagnostic test uses the application of radioisotope and is not performed routinely.Digestive disease can affect the BMI and patients with digestive form of Cd receive nutritional counselling that can affect their diet quality.
This study showed that Cd patients had increased risk of disease considering the high BMI and high waist circumference.Similar findings were shown by Geraix 33 in 66 adult patients with positive serology for Cd, showing a high frequency of obesity (62%), as measured by BMI and increased risk of metabolic disease (55%), assessed by waist circumference.It is important to note that the cardiac form is the most severe and frequent manifestation of chronic Cd 34 , and the presence of obesity significantly increases morbidity and mortality from other diseases, such as hypertension, dyslipidemia, coronary artery disease, diseases biliary tract, osteoarticular diseases, type 2 diabetes mellitus, and some cancers 29,34,35  promotes the secretion of cytokines, leads to the initiation of pro-inflammatory events, and oxidative stress 36 .As Cd is an inflammatory disease, obesity could aggravate the evolution of the disease.According to some studies, adipocytes are an important target for infection by T. cruzi, functioning as host cell and reservoir in chronic Cd 37 .The Cd group presented a high prevalence of overweight/obesity, and it could worsen the course of the disease in these patients.
In the Cd group, it was possible to observe a smaller variation of consumption of BHEI-R components, this fact was a constant in the study.Possibly, the lower variation in food consumption in patients with Cd reflected the complications of these patients, such as dysphagia, leading to the impairment of the quality and amount of food.Other complications arising in these patients are impairment of the gastrointestinal tract, which affects food motility, and permanent reflux of partially digested food material.According to Torres 38 77% of Cd patients in their study reported discomfort during feeding.Possibly, this may explain why the Cd group obtained a better score for the fruit and vegetable groups, and a lower score for the cereals group because of adaptation by patients in their diet, resulting from difficulty in swallowing food, especially solid foods.
The results showed that the total score of BHEI-R did not differ among the groups.However, when evaluating the food components, significant differences were observed among the groups.As there was no study that evaluated dietary intake in Cd patients, we compared our findings to general studies that evaluated dietary intake and diet quality by BHEI-R.Since most patients in this study were older, a population-based study of 1,509 elderly 39 aimed at assessing quality diet of the elderly according to sociodemographic variables, behaviors related to health and morbidities was considered.Assumpção et al. 39 showed similar population characteristics that were found in the present study, a higher proportion of women (57%) and elderly (53.8%) and an average of BHEI-R score 62.4.In the Assumpção's study, a different method of dietary assessment, using a 24-hour recall, was used, while in this study, FFQ was used.According to the literature, other BHEI scores tend to be overestimated when the FFQ is used 40 .Also, Cd patients have clinical follow-up of many years, and are likely to have already received nutritional advise.The results of this study indicate worse scores on cereals, milk and dairy, meats and saturated fat in the Cd group.The consumption of fruits and vegetables in their usual diet can lead to lower cereal intake.The percentage of macronutrients intake showed that Cd group presented higher lipid consumption in diet, and the profile of this consumption was high in saturated fat, and lower in PUFA.Considering a population with low socioeconomic level, this profile of fat consumption could be related to the consumption of products rich in saturated fat and at a more affordable price, such as sausages, ultra-processed biscuits, and cakes and lower consumption of foods rich in unsaturated fat, such as oilseeds, vegetable and olive oils which are more expensive.Considering the cardiovascular impairment of these patients, the lipid profile is a concern for the dietary advise.The deleterious role of saturated fatty acid intake in glycolipid metabolism is well-established.Metabolic and epidemiological studies 41 have shown that saturated fatty acid raises the plasma concentration of total and low-density lipoprotein cholesterol (LDL-C), compared to PUFA.Margioris 42 reviewed the impact of consumption of saturated fats on inflammatory profile and showed that saturated fats increased inflammatory markers in the postprandial period, which does not occur with MUFA and PUFA.Fnu Nagajyoth 43 analyzed in vitro, the effect of the high-fat diet in the regulation of acute myocarditis caused by T. cruzi, and the effect on lipid metabolism in adipose tissue  and heart.They showed that persistence of the inflammatory infiltrate contributes to the chronic pathology in the heart 43 .
Dietary fiber intake was apparently satisfactory in Cd group, but there was a lower intake in the advanced stage of cardiac involvement.This could be associated with the lower intake of whole fruit.
It is important to point out that parasitic infection and micronutrient deficiencies coexist in developed countries, and usually have complex interactions that promote deleterious

deFIGURE 2 -
FIGURE 2 -Box plots representing macronutrients and micronutrients intake.The values in the boxes represent the median and interquartile intervals, the whiskers represent the 5 th and 95 th percentiles, and the outliers are plotted as individual values.Paired Student t test was performed.

de
Castilhos MP et al. -Diet quality of patients with Chagas disease

TABLE 1 :
. Also, it is known that obesity is associated with subclinical inflammation, General characteristics, body mass index and waist circumference classifications between non-infected and Chagas disease groups, and Los Andes groups.Paired Student t-test for comparison of means.p2: ANOVA test and Bonferroni post-test for comparison Los Andres Groups.a Represent the differences between Los Andes IA and IB vs. III for whole fruit (p<0,05).
IA: normal electrocardiogram and echocardiogram -no heart involvement; IB: early stage of cardiac involvement; II: advanced stage of cardiac involvement without heart failure; III: advanced stage of cardiac involvement with heart failure; BMI: body mass index.*Study matching variables.**The waist circumference cut off points: increased >94cm for men and >80cm for women; substantially increased >102 cm for men and >88cm for women 16 .p1: p value for comparison between non-infected and Chagas disease (Cd) groups; p2: p value for comparison between Los Andes Groups.a Paired Student t Test for means comparison of non-infected and Cd groups.b McNemar test for frequency comparison for paired non-infected and Cd groups.c ANOVA test and Bonferroni post-test for comparison Los Andres groups.d Pearson's chi-square test for comparison frequencies between Los Andes groups.n (%) or mean ± standard deviation.de Castilhos MP et al. -Diet quality of patients with Chagas disease ANOVA: analysis of variance; 95% CI: confidence interval 95%; Mean (95% CI).p1:

TABLE 2 :
Energy intake, Brazilian Healthy Eating Index-Revised total score and component scores between non-infected and Chagas disease groups, and Los Andes groups.
Data are expressed in mean [95% confidence interval (CI 95%)].IA: normal electrocardiogram and echocardiogram -no heart involvement; IB: early stage of cardiac involvement; II: advanced stage of cardiac involvement without heart failure; III: advanced stage of cardiac involvement with heart failure; TEI: total energy intake; EAR: estimated average requirement.p: ANOVA test and Bonferroni post-test for comparison Los Andres Groups.

TABLE 3 :
Intake of macronutrients, dietary fiber, vitamins and minerals according to Los Andes groups.