Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Oct 13:3:e1325.
doi: 10.7717/peerj.1325. eCollection 2015.

Clustering symptoms of non-severe malaria in semi-immune Amazonian patients

Antonio C Martins  1 , Felipe M Araújo  1 , Cássio B Braga  1 , Maria G S Guimarães  1 , Rudi Nogueira  1 , Rayanne A Arruda  1 , Lícia N Fernandes  2 , Livia R Correa  2 , Rosely Dos S Malafronte  2 , Oswaldo G Cruz  3 , Cláudia T Codeço  3 , Mônica da Silva-Nunes  1
Affiliations

Clustering symptoms of non-severe malaria in semi-immune Amazonian patients

Antonio C Martins et al. PeerJ. .

Abstract

Malaria is a disease that generates a broad spectrum of clinical features. The purpose of this study was to evaluate the clinical spectrum of malaria in semi-immune populations. Patients were recruited in Mâncio Lima, a city situated in the Brazilian Amazon region. The study included 171 malaria cases, which were diagnosed via the use of a thick blood smear and confirmed by molecular methods. A questionnaire addressing 19 common symptoms was administered to all patients. Multiple correspondence analysis and hierarchical cluster analysis were performed to identify clusters of symptoms, and logistic regression was used to identify factors associated with the occurrence of symptoms. The cluster analysis revealed five groups of symptoms: the first cluster, which included algic- and fever-related symptoms, occurred in up to 95.3% of the cases. The second cluster, which comprised gastric symptoms (nausea, abdominal pain, inappetence, and bitter mouth), occurred in frequencies that ranged between 35.1% and 42.7%, and at least one of these symptoms was observed in 71.9% of the subjects. All respiratory symptoms were clustered and occurred in 42.7% of the malaria cases, and diarrhea occurred in 9.9% of the cases. Symptoms constituting the fifth cluster were vomiting and pallor, with a 14.6% and 11.7% of prevalence, respectively. A higher parasitemia count (more than 300 parasites/mm(3)) was associated with the presence of fever, vomiting, dizziness, and weakness (P < 0.05). Arthralgia and myalgia were associated with patients over the age of 14 years (P < 0.001). Having experienced at least eight malaria episodes prior to the study was associated with a decreased risk of chills and fever and an increased risk of sore throat (P < 0.05). None of the symptoms showed an association with gender or with species of Plasmodium. The clinical spectrum of malaria in semi-immune individuals can have a broad range of symptoms, the frequency and intensity of which are associated with age, past exposure to malaria, and parasitemia. Understanding the full spectrum of nonsevere malaria is important in endemic areas to guide both passive and active case detection, for the diagnosis of malaria in travelers returning to non-endemic areas, and for the development of vaccines aimed to decrease symptom severity.

Keywords: Amazon; Clustering; Malaria; Symtoms.

PubMed Disclaimer

Conflict of interest statement

The authors declare there are no competing interests.

Figures

Figure 1
Figure 1. Frequency and intensity of each malaria-related symptom in Mâncio Lima (2012–2013).
Numbers on y axis are percentages. The shading pattern of each bar indicates the proportion of patients reporting a given symptom as absent, mild, moderate or severe. C1, Algic and fever-related symptoms; C2, Gastric symptoms; C3, Respiratory symptoms; C4, Vomiting and pallor; C5, Diarrhoea.
Figure 2
Figure 2. Joint distribution of the correspondence analysis dimensions for qualitative variables (parasitemia, number of previous malaria episodes and malaria species).
(A) Joint distribution of the correspondence analysis dimensions for qualitative variables (parasitemia, number of previous malaria episodes and malaria species). Dimension 1 and 2 contribute to explain 31.74% of the variance. The contribution of each symptom in dimension 1 and 2 is represented as a triangle, where ‘yes’ indicates the contribution of the presence of the symptom, and ‘no’ indicates the contribution of the absence of the symptom. Subjects are represented as blue dots, and their position in each quadrant demonstrate the referred symptom profile of each individual (near the presence of symptoms or near the absence of symptoms). (B) Joint distribution of the correspondence analysis dimensions for qualitative variables (parasitemia, number of previous malaria episodes and malaria species) with circles indicating groups of symptoms identified in the cluster analysis. C1, Algic and fever-related symptoms; C2, Gastric symptoms; C3, Respiratory symptoms; C4, Vomiting and pallor; C5, Diarrhoea.
Figure 3
Figure 3. Dendrogram of the coordinates of the first two dimensions of the correspondence analysis with clusters of groups.
C1, Algic and fever-related symptoms; C2, Gastric symptoms; C3, Respiratory symptoms; C4, Vomiting and pallor; C5, Diarrhoea.
Figure 4
Figure 4. Multiple correspondence analysis classified by Plasmodium species, parasitemia and number of previous malaria episodes.
The confidence ellipse delimitates the centroid around each variable. For Parasitemia, ellipses are separated, indicating a good distinction in symptom profile between low and high parasitemia. Patients with higher parasitemia (≥300 parasites/mm3) have more symptoms (pink ellipse is seen in the left-hand side of dimension 1) and those with lower parasitemia (<300 parasites/mm3) tend to have fewer symptoms (blue ellipse represented on the right-hand side of dimension 1). For Plasmodium species, ellipses are superimposed, indicating that the symptom profile shown in each type of malaria is very similar, and therefore malaria species cannot be differentiated using symptoms only. For number of previous episodes of malaria, there is little superimposition of the ellipses, indicating that it is possible to separate between those patients with more than eight previous malaria episodes, who generally experience fewer symptoms (pink ellipse is in the right side of Dimension 1, indicating less symptoms), except for symptoms sore throat, which is more pronounced in these patients (note that the pink ellipse is above the blue ellipse in Dimension 2, indicating that symptoms distributed along dimension 2 are differentiating patients with 8 or more episodes from those with less than 8 episodes).
Figure 5
Figure 5. Relationship between symptom intensity and parasitological features (number of previous malaria episodes, parasitemia and age) in Mâncio Lima (2012–2013).
Numbers on y axis are percentages. The shading pattern of each bar segment indicates the proportion of patients reporting a given symptom as absent, mild, moderate or severe. P value = Somers’ d test for ordinal variables.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Acre . State program of ecological-economical assessment of state of Acre. Phase II: synthesis document. 2nd edition. Rio Branco: Secretaria de Estado de Meio Ambiente e Recursos Naturais; 2010. Governo do Estado do Acre. 358 p.
    1. Anstey NM, Jacups SP, Cain T, Pearson T, Ziesing PJ, Fisher DA, Currie bJ, Marks PJ, Maguire GP. Pulmonary manifestations of uncomplicated falciparum and vivax malaria: cough, small airways obstruction, impaired gas transfer, and increased pulmonary phagocytic activity. Journal of Infectios Diseases. 2002;185:1326–1334. doi: 10.1086/339885. - DOI - PubMed
    1. Ansumana R, Jacobsen KH, Gbakima AA, Hodges MH, Lamin JM, Leski TA, Malanoski AP, Lin B, Bockarie MJ, Stenger DA. Presumptive self-diagnosis of malaria and other febrile illnesses in Sierra Leone. Pan African Medical Journal. 2013;15:34. doi: 10.11604/pamj.2013.15.34.2291. - DOI - PMC - PubMed
    1. Arévalo-Herrera M, Forero-Peña DA, Rubiano K, Gómez-Hincapie J, Martínez NL, Lopez-Perez M, Castellanos A, Céspedes N, Palacios R, Oñate JM, Herrera S. Plasmodium vivax sporozoite challenge in malaria-naïve and semi-immune Colombian volunteers. PLoS ONE. 2014;9(6):e1325. doi: 10.1371/journal.pone.0099754. - DOI - PMC - PubMed
    1. Carvalho MS, Struchiner CJ. Análise de correspondência: uma aplicação do método à avaliação de serviços de vacinação. Cadernos de Saúde Pública. 1992;8:287–301. doi: 10.1590/S0102-311X1992000300008. - DOI

Grants and funding

This study received financial support from UFAC (Brazil), FUNTAC (Edital FDCT 2011), CNPq (Edital Universal 2011) and FAPAC (Edital PPSUS 2013) as well as support from the UFAC Master’s Program in Public Health. Research fellowships were awarded by CNPq, UFAC and FAPAC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

LinkOut - more resources