SciELO - Scientific Electronic Library Online

vol.54 issue5Is muscle biopsy still necessary for the diagnosis of mitochondrial disease?Evaluation of acute toxicity of β-lapachone associated with chitosan as a cytoprotective agent author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Jornal Brasileiro de Patologia e Medicina Laboratorial

Print version ISSN 1676-2444On-line version ISSN 1678-4774

J. Bras. Patol. Med. Lab. vol.54 no.5 Rio de Janeiro Sept./Oct. 2018 


Original article

Aspartate aminotransferase to platelet ratio index (APRI) for differentiation of primary and secondary infection by dengue virus

Aplicação do índice de relação aspartato aminotransferase sobre plaquetas (APRI) na diferenciação de infecção primária e secundária pelo vírus da dengue

Suellen R. Martins1 

Melina B. Pinheiro2 

Luci Maria S. Dusse1 

Ana Paula L. Mota1 

Patrícia N. Alpoim1 

1Universidade Federal de Minas Gerais (UFMG), Minas Gerais, Brazil

2Universidade Federal de São João del-Rei (UFSJ), Minas Gerais, Brazil



Dengue virus (DENV) infection has been considered a major public health problem in tropical countries. The unavailability of serologic testing in public health centers might adversely impact patients' outcome.


This study aimed to evaluate the accuracy of mean platelet volume (MPV) and aspartate aminotransferase (AST) to platelet ratio index (APRI) as laboratory markers of DENV infection that could be used to differentiate primary and secondary infections.


We assessed laboratory results from 503 patients with positive rapid test for DENV infection.


Severe thrombocytopenia and increased liver involvement were observed in patients with DENV heterotypic secondary infection. Our data suggest that APRI was able to distinguish patients with primary and secondary infection (p = 0.006) with a relevant sensitivity (75%), specificity (76%) and a cut-off of 1.06. A total of 80 out of 105 (76%) patients with primary DENV infection had APRI ≤ 1.06, and 12 (75%) with secondary DENV infection had APRI > 1.06. On the other hand, MPV did not show significance in the differentiation of types of infection, coming up with poor area under the receiver operating characteristic (ROC) curve (0.61).


APRI seems to be a powerful tool for early identification of DENV secondary infection cases in health centers.

Key words: dengue; aspartate aminotransferases; diagnosis; platelet count



A infecção pelo vírus da dengue (DENV) é considerada um grande problema de saúde pública nos países tropicais. A indisponibilidade de testes sorológicos em centros de saúde pública pode afetar negativamente o prognóstico do paciente.


Este estudo teve como objetivo avaliar a precisão do volume médio de plaquetas (MPV) e o índice da relação de aspartato aminotransferase (AST) sobre plaquetas (APRI) como marcadores laboratoriais de infecção por DENV, que poderiam ser utilizados para diferenciar infecções primárias e secundárias.


Foram avaliados os resultados laboratoriais de 503 pacientes com teste rápido positivo para infecção por DENV.


Foram observadas trombocitopenia grave e disfunção hepática em pacientes com infecção secundária heterogênea por DENV. Nossos dados sugerem que o APRI foi capaz de distinguir os pacientes com infecção primária e secundária (p = 0, 006), com relevante sensibilidade (75%) e especificidade (76%) e corte de 1, 06. Um total de 80 de 105 (76%) pacientes com infecção primária por DENV tinha APRI ≤ 1, 06; e 12 (75%) com infecção secundária por DENV, APRI > 1, 06. Por outro lado, o MPV não mostrou significância na diferenciação de tipos de infecção, apresentando baixo valor da área sob a curva de característica de operação do receptor (ROC) (0, 61).


APRI parece ser uma ferramenta poderosa para identificação precoce de casos de infecção secundária de DENV em centros de saúde.

Unitermos: dengue; aspartato aminotransferases; diagnóstico; contagem de plaquetas


Dengue virus (DENV) infection is a global public health problem and a human vector-borne viral disease. According to the World Health Organization (WHO), cases across the Western Pacific, South-East Asia and Americas reached over 3.2 millions in 2015. Brazil had the highest number of infection episodes in 2015, with most cases being DENV 1 infection. That figure was approximately three times higher than that in 2014(1-3). Besides, the total cases that require hospitalization increase each year, and many of them evolve to death(4).

Dengue is an arboviral infection represented by four distinct serotypes (DENV 1 to 4) and transmitted by Aedes ssp., mainly Aedes aegypti, broadly found in tropical and subtropical regions(1). The immune response arising from one serotype can confer life-long protection against that same serotype, but not against the others, leading to the possibility of secondary infections. Clinically, dengue infection may vary from asymptomatic infection to severe dengue clinical presentation and even death(5,6). The more severe types of dengue infection, known as dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS), are commonly caused by secondary infection, when an individual with a previous history of dengue is subsequently infected with a second virus serotype. DHF/DSS is characterized by coagulopathy, increased vascular fragility, and loss of fluid due to capillary permeability that may progress to hypovolemic shock(7).

The DENV nonstructural glycoprotein 1 (NS1) plays an important role in replication, pathogenesis and modulation of immune response, although its functions are not completely understood. The NS1 antigen is released abundantly in circulation, and its presence is detected in blood from the 1st to the 9th day of infection(8). Anti-NS1 antibodies [immunoglobulin class M (IgM) type] may develop cross-reaction with platelets in dengue infection and, consequently, contribute to a transient thrombocytopenia. Furthermore, it was shown that infection by DENV in hematopoietic cells may impair megakaryocytopoiesis in mice(9). Serum levels of NS1 antigen provide a preliminary indication of DENV and represent a new approach to the diagnosis of acute dengue infection. However, its role as an early predictor of severe dengue infection remains unclear.

The viral culture assay and viral ribonucleic acid (RNA) detection by reverse transcriptase polymerase chain reaction (RT-PCR) are considered more specific tests to confirm infection, as they detect viral dengue particles. However, these tests require highly trained staff and sophisticated equipment, besides being time consuming and costly, what limits their routine applicability in health units of developing countries(10,11). For this reason, cheaper and simpler tests are urgently needed so they can be available in health services of such countries.

Alternative laboratory tests have been used in screening and monitoring the evolution of viral infection, such as tourniquet test, blood count, alanine aminotransferase (ALT) and aspartate aminotransferase (AST)(12). These tests are cheaper and widely available in health services when compared with the gold standard diagnostic tests.

The rapid test for DENV infection has been largely used in Brazilian health units as an alternative diagnostic test. It is based on a qualitative and differential diagnosis with detection of IgM and IgG DENV antibody in human serum or plasma, enabling the detection of primary and secondary infection, respectively(13).

Once DHF/DSS are associated with severe thrombocytopenia, bleeding and increased vascular permeability, the decreasing platelet counts have been found to predict the severity of the disease. Platelet count is also used as a recovery parameter for these conditions(14). Mean platelet volume (MPV) is a geometric measure that assesses the size of platelets circulating in the plasma. It is often used to indicate platelet activation and, indirectly, bone marrow activity(15,16). Not much work has been done on this parameter related to DENV infection. AST to platelet ratio index (APRI) includes AST and platelet count as variables. It is calculated as (AST/upper limit of normal range)/platelet count (109/l) x 100. This index was firstly proposed to predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C(16). Once viral replication can affect liver function, and DENV infection is associated with thrombocytopenia, APRI test could be useful for early infection diagnosis. No previous studies have been performed that relate APRI and DENV infection.

Despite the several studies investigating DENV and its multiple sites of infection, such as endothelial cells, liver cells, bone marrow, spleen, kidneys and lungs, the pathophysiology of this disease is still not fully understood(17-19). Due to variation of clinical manifestations in DENV infections and the multifactorial nature of the host response to the pathogen(1), it becomes necessary to find new biomarkers that can indicate the infection severity and the involvement of other organs. Such biomarkers could guide better management approach and result in better outcomes for patients. In addition, due to the increasing number of cases of DENV infection and the low availability of gold standard diagnostic tests in health units of developing countries, cheaper and simpler tests are urgently needed. In this context, we aim to evaluate the accuracy of MPV and APRI as laboratory markers of DENV infection that could be used to differentiate primary and secondary infections with a better cost-benefit than current gold standard tests.


Ethical aspects

This study was approved by the Ethics Committee of Universidade Federal de Minas Gerais (UFMG), Brazil (CAAE 61330316.6.0000.5149), and it was conducted in accordance with the Helsinki Declaration. The research protocol did not interfere with any medical recommendations or prescriptions.

Study design

We assessed laboratory results from 503 patients with positive rapid test for DENV infection. These patients were treated in outpatient units of a reference hospital in the metropolitan area of Belo Horizonte (MG), Brazil, from January to July 2015.

Rapid tests for dengue were performed with an immunochromatographic kit (Dengue fever test duo – Bioeasy® Diagnostica Ltda. – Belo Horizonte, Brazil), according to the manufacturer's instructions to measure NS1 antigen levels (sensitivity: 92.8%, specificity: 98.8%) and IgM/IgG levels (sensitivity: 99.4%, specificity 93%). These biomarkers are commonly used in clinical practices at health units, according to the Technical Guides on dengue of the Brazilian Ministry of Health(13).

Considering the results of these rapid tests, all participants were distributed into two groups: group I – patients with DENV primary infection (n = 473), and group II – patients with DENV heterotypic secondary infection (n = 30).

Inclusion criteria

• Group I (primary test infection): positive for NS1 antigen or positive NS1 and IgM DENV.

• Group II (heterotypic secondary infection): positive test for NS1 antigen and IgG DENV or positive NS1, IgM and IgG DENV.

Non-inclusion criteria

Negative immunochromatographic tests for the three parameters (NS1, IgM and IgG DENV).

Positive tests of IgM or/and IgG without the presence of NS1 antigen. Positive test, but no request for additional biochemical or hematological analysis.

Laboratory parameters assays

Laboratory parameters as red blood cells, hematocrit, total leukocyte count, absolute neutrophil count, absolute lymphocyte count, platelets, MPV, AST and ALT were assessed and compared between the two studied groups. APRI was calculated as (AST/upper limit of normal range)/platelet count (109/l) x 100.

Biochemical parameters (AST and ALT) were measured in the integrated device Abbott Architect C4000i1000SR- Illinois, USA, by turbidimetric and colorimetric methods. Hematologic parameters analyses (red blood cells, hematocrit, total leukocyte count, absolute neutrophil count, absolute lymphocyte count, platelets, MPV) were performed on the XS-1000iTM- Sysmex Corporation, by flow cytometry and impedance.

Statistical analysis

Statistical analyses were made using SPSS software (version 19.0). Data normality was tested with Shapiro-Wilk test. Nonparametric variables (AST, ALT, APRI, leukocyte, neutrophil, lymphocyte and age) were presented as median and 25th-75th percentiles (interquartile range). Data on biomarkers obtained for each group were compared by Mann-Whitney method. Student t test was used to compare parametric continuous variables [red blood cell (RBCs), hematocrit, MPV, platelets], shown as mean and standard deviation. The comparison of categorical variable (sex) was performed using the Pearson's chi-square test. The MPV and APRI values were assessed by the area under the receiver operating characteristic (ROC) curves. The best model derived from the training set then was applied to the validation set to test for accuracy by measuring the areas under the ROC curves (AUC). Values of p < 0.05 were considered statistically significant.


The two groups showed similar age and sex distribution (p > 0.05; Table). Therefore, potential differences of variables studied between the groups are likely to reflect the different nature of DENV infection. For hematological tests (RBCs, hematocrit, total leukocyte count, absolute neutrophil count, absolute lymphocyte count, platelets, MPV), a higher count of absolute neutrophil and platelet was found (p = 0.004 for both parameters; Table) in group I than in group II. On the other hand, absolute lymphocytes count and MPV were lower in group I than in group II (p < 0.001 and p = 0.042, respectively; Table).

TABLE Comparison of demographic and laboratorial variables between the groups 

Variables Group I – primary DENV infection (n = 473) Group I – secondary DENV infection (n = 30) p
Agea (years) 36 (33) 45 (32) 0.112
Sex1 (%)
Male 209 (44%) 14 (47%) 0.791
Female 264 (56%) 16 (53%)
Laboratory parameters
no. RBCsb × 106/mm3 4.69 (0.51) 4.8 (0.57) 0.312
Hematocritb (%) 39.7 (4.1) 40.5 (4) 0.29
Global leukocytea/mm3 3,985 (2,588) 4,260 (1,755) 0.988
Neutrophils absolutea value/mm3 2,357 (2,265) 1,607 (1,01) 0.004*
Lymphocytes absolutea value/mm3 791 (522) 1,288 (1,093) < 0.001*
Platelet countb × 103/mm3 163 (74) 127 (95) 0.004*
MPVb 10.6 (1.3) 11 (1.75) 0.042*
AST U/la 35 (28) 65 (111) 0.005*
LT U/la 30 (28) 70 (76) 0.008*
A APRIa 0.63 (0.6) 1.87 (2.17) 0.006*

RBCs: red blood cells; MPV: mean platelet volume; AST: aspartate aminotransferase; ALT: alanine aminotransferase; APRI: AST to platelet ratio index; DENV: dengue virus;

1Chisquare Pearson asymptotic test;

anon-parametric data expressed as median (interquartile range) (Mann-Whitney);

bnormally distributed data are presented as mean ± standard deviation (t-student);

*p < 0.05.

For biochemical parameters, higher aminotransferases (AST and ALT) serum activity and APRI index were found in secondary DENV infection compared to primary (p = 0.005; p = 0.008 and p = 0.006, respectively). Only 117 patients had aminotransferases results, being 105 with primary infection (group I) and 12 with secondary infection (group II).

ROC curves of MPV and APRI for discriminate primary and secondary DENV infection were plotted in the Figure. The AUC was 0.613 (p = 0.042) and 0.744 (p = 0.744), respectively.

FIGURE ROC curves of MPV and APRI in the prediction of significant discrimination of primary and secondary DENV infections. An AUC of 1 is characteristic of an ideal test, whereas an AUC of 0.5 or less indicates a test of no diagnostic valueROC: receiver operating characteristic; MPV: mean platelet volume; APRI: aspartate aminotransferase to platelet ratio index; DENV: dengue virus; AUC: area under the curve. 


Dengue fever is a growing public health concern in most tropical countries. Brazil is one of the countries with the highest number of cases reported in the latest years with predominance of DENV 1 infection(1-3). The present study showed a small number of patients in group II, composed of patients with secondary infections, what can be explained by the epidemiological DENV infection profile in Brazil in recent years. The predominance of DENV 1 serotype, which is not related to the severe form, and the low incidence of secondary infection by other serotypes justify the reduced size of group II.

Our data show that the number of circulating platelets was lower in group II. It is known that thrombocytopenia by DENV infection can be caused by cross-reactivity of antibodies to platelets(20,21) by infection of hematopoietic cells and rich circulation of environment cytokines(22,23). In addition, viral replication in the liver, which disturbs thrombopoietin synthesis, can contribute to the thrombocytopenia observed in group II(19).

The markers of liver function showed a higher AST and ALT activity in group II versus group I. Elevated liver enzymes suggest that the liver is affected, but the role of hepatic damage in coagulopathy and disease severity remains to be established(19) Therefore, AST and ALT monitoring during DENV infection seems to be essential for the prognosis of liver infection.

As our data showed a good relationship between change in AST activity and platelet count in secondary DENV infections, the analysis of APRI by ROC curve was carried out. The first study developed and published using APRI as a marker of fibrosis and cirrhosis in patients with hepatitis C longed for a biomarker that easily identified fibrosis and cirrhosis with good significance in this condition(24). Wai et al. (2003)(16) developed a formula based on the good relationship between AST and platelets related to abnormal liver (fibrosis/cirrhosis).

Our findings showed that APRI was able to distinguish patients with primary and secondary infection (p = 0.006) with a relevant sensitivity (75%) and specificity (76%) and a cut-off of 1.06. A total of 80 out of 105 (76%) patients with primary DENV infection had APRI ≤ 1.06, and nine out of 12 (75%) patients with secondary DENV infection had APRI > 1.06. We did an exhaustive literature search looking for APRI results in DENV infection, however nothing was found. Our data strongly suggest that APRI can have other applications, besides fibrosis and cirrhosis in patients with hepatitis C. It can help discriminate primary or secondary DENV infection, since patients with secondary dengue showed a higher degree of hepatic changes (APRI median: 1.84) in comparison to patients with primary dengue (APRI median: 0.6).

The accuracy measured by AUC of the ROC curve for MPV was 0.613. Since the rough guide for classifying the accuracy of a diagnostic test considers an AUC between 0.6 and 0.7 poor, we could not choose a cut-off point for MPV. Recently, the results of Sharma and Yadav (2015)(25) research corroborated ours. They showed no correlation of MPV with severity, serology and treatment outcome in dengue fever(25).

Anthropometric data, as age and gender, were collected from the patients with DENV infection, and no differences were found between the two groups. Although we did not have access to clinical data, it is known that MPV is not altered by smoking, obesity, and alcohol abuse, as shown by Maluf et al. (2015)(26) in a Brazilian population. These authors reported influence of ethnicity in MPV, but differences related to ethnicity are hard to interpret in our population, probably because of the five centuries of high miscegenation of Europeans, Africans, and Native Americans(27).

Regarding the blood count parameters, no differences were found in hematocrit and global leukocytes between the groups. It is important to note that a hematocrit rise can indicate a need for hydration and possible increase in capillary permeability, while leukopenia and neutropenia suggest high opsonization of leukocytes(21). It is known that such changes are observed in greater intensity in the critical phase of DENV infection. By contrast, patients usually seek health care in the early stage of the infection, the febrile period; therefore, it is unlikely that significant changes of these parameters can be identified in this first consultation.

It should be highlighted that dengue fasting tests (NS1, IgM and IgG) are not currently available in basic health units in developing countries such as Brazil. However, conventional tests such as blood count and AST and ALT activity are more easily acquired and available in centers that perform basic hematology and biochemistry tests. Accordingly, the proposed use of APRI as a biomarker to classify dengue infection type (primary or secondary) may be applicable in these places. It is recommended that the results should be evaluated along with the previous health records and current patient clinical presentation to support medical decisions.


Our findings suggest that APRI could help distinguishing patients with primary and secondary infection. Since the APRI calculation only requires the results of simple, cheap and widely distributed tests (AST and platelet count), APRI may help improve DENV infection diagnosis in developing countries.


1 Bhatt S, Gething PW, Brady OJ, et al. The global distribution and burden of dengue. Nature. 2013; 496(7446): 504-7. [ Links ]

2 Khetarpal N, Khanna I. Dengue fever: causes, complications, and vaccine strategies. J Immunol Res. 2016; 2016: 6803098. [ Links ]

3 Marinho PE, Bretas de Oliveira D, Candiani TM, et al. Meningitis associated with simultaneous infection by multiple dengue virus serotypes in children, Brazil. Emerg Infect Dis. 2017; 23(1): 115-8. [ Links ]

4 Shepard DS, Coudeville L, Halasa YA, Zambrano B, Dayan GH. Economic impact of dengue illness in the Americas. Am J Trop Med Hyg. 2011; 84(2): 200-7. [ Links ]

5 Dhanoa A, Hassan SS, Ngim CF, et al. Impact of dengue virus (DENV) co-infection on clinical manifestations, disease severity and laboratory parameters. BMC Infect Dis. 2016; 16(1): 406. [ Links ]

6 Siddiqui O, Chakravarti A, Abhishek KS. Dengue: lessons of an outbreak. J Clin Diagn Res. 2016; 10(6): Dc01-4. [ Links ]

7 Guzman MG, Alvarez M, Halstead SB. Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: a historical perspective and role of antibody-dependent enhancement of infection. Arch Virol. 2013; 158(7): 1445-59. [ Links ]

8 Avirutnan P, Punyadee N, Noisakran S, et al. Vascular leakage in severe dengue virus infections: a potential role for the nonstructural viral protein NS1 and complement. J Infect Dis. 2006; 193(8): 1078-88. [ Links ]

9 Sun DS, King CC, Huang HS, et al. Antiplatelet autoantibodies elicited by dengue virus non-structural protein 1 cause thrombocytopenia and mortality in mice. J Thromb Haemost. 2007; 5(11): 2291-9. [ Links ]

10 Dussart P, Petit L, Labeau B, et al. Evaluation of two new commercial tests for the diagnosis of acute dengue virus infection using NS1 antigen detection in human serum. PLoS Negl Trop Dis. 2008; 2(8): e280. [ Links ]

11 Oliveira de Paula S, Malta Lima D, Clotteau M, Pires Neto Rd Rda J, Lopes da Fonseca BA. Improved detection of dengue-1 virus from IgMpositive serum samples using C6/36 cell cultures in association with RTPCR. Intervirology. 2003; 46(4): 227-31. [ Links ]

12 Souza LJ, Alves JG, Nogueira RM, et al. Aminotransferase changes and acute hepatitis in patients with dengue fever: analysis of 1,585 cases. Braz J Infect Dis. 2004; 8(2): 156-63. [ Links ]

13 Ministério da Saúde. Secretaria de Vigilância em Saúde. Departamento de vigilância das doenças transmissíveis. Dengue: diagnóstico e manejo clínico: adulto e criança. 5 ed. Brasília (DF): Ministério da Saúde; 2016. 58p. [ Links ]

14 Jayashree K, Manasa GC, Pallavi P, Manjunath GV. Evaluation of platelets as predictive parameters in dengue fever. Indian J Hematol Blood Transfus. 2011; 27(3): 127-30. [ Links ]

15 Eminler AT, Uslan MI, Ayyildiz T, et al. Mean platelet volume is an important predictor of hepatitis C but not hepatitis B liver damage. J Res Med Sci. 2015; 20(9): 865-70. [ Links ]

16 Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology. 2003; 38(2): 518-26. [ Links ]

17 Durbin AP, Vargas MJ, Wanionek K, et al. Phenotyping of peripheral blood mononuclear cells during acute dengue illness demonstrates infection and increased activation of monocytes in severe cases compared to classic dengue fever. Virology. 2008; 376(2): 429-35. [ Links ]

18 Jessie K, Fong MY, Devi S, Lam SK, Wong KT. Localization of dengue virus in naturally infected human tissues, by immunohistochemistry and in situ hybridization. J Infect Dis. 2004; 189(8): 1411-8. [ Links ]

19 Martina BE, Koraka P, Osterhaus AD. Dengue virus pathogenesis: an integrated view. Clin Microbiol Rev. 2009; 22(4): 564-81. [ Links ]

20 Mannucci PM, Capoferri C, Canciani MT. Plasma levels of von Willebrand factor regulate ADAMTS-13, its major cleaving protease. Br J Haematol. 2004; 126(2): 213-8. [ Links ]

21 Robert M, Sepandj F, Liston RM, Dooley KC. Random proteincreatinine ratio for the quantitation of proteinuria in pregnancy. Obstet Gynecol. 1997; 90(6): 893-5. [ Links ]

22 Nachman RL, Rafii S. Platelets, petechiae, and preservation of the vascular wall. N Engl J Med. 2008; 359(12): 1261-70. [ Links ]

23 Pearson MG, Pinker GD. ABO blood groups and toxaemia of pregnancy. Br Med J. 1956; 1(4970): 777-8. [ Links ]

24 Faraci FM, Brian Jr JE, Heistad DD. Response of cerebral blood vessels to an endogenous inhibitor of nitric oxide synthase. Am J Physiol. 1995; 269(5 Pt 2): H1522-7. [ Links ]

25 Sharma K, Yadav A. Association of mean platelet volume with severity, serology & treatment outcome in dengue fever: prognostic utility. J Clin Diagn Res. 2015; 9(11): Ec01-3. [ Links ]

26 Maluf CB, Barreto SM, Vidigal PG. Standardization and reference intervals of platelet volume indices: insight from the Brazilian longitudinal study of adult health (ELSA-BRASIL). Platelets. 2015; 26(5): 413-20. [ Links ]

27 Pena SDJ, Bastos-Rodrigues L, Pimenta JR, Bydlowski SP. DNA tests probe the genomic ancestry of Brazilians. Braz J Med Biol Res. 2009; 42(10): 870-6. [ Links ]

Received: February 28, 2018; Revised: September 06, 2018; Accepted: September 27, 2018; pub: October 20, 2018

CORRESPONDING AUTHOR: Patrícia Nessralla Alpoim, Faculdade de Farmácia; Universidade Federal de Minas Gerais, Avenida Antônio Carlos, 6627; Pampulha. CEP: 31270-901; Belo Horizonte-MG, Brasil, Phone: +55 (31) 3409-7492. e-mail:


All authors affirm the inexistence of conflicts of interest in this study.

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.