• View in gallery

    Geographic destination of travelers included in the study with malaria and confirmed arbovirus infection from 2013 to 2016.

  • View in gallery

    Distribution of white blood cell and platelet counts on admission in travelers with malaria and arbovirus infection.

  • View in gallery

    Platelet distribution in malaria and non-malaria cases (A) and diagnostic performance (area under the receiver operating characteristic [ROC] curve) for platelet counts to discriminate malaria and non-malaria cases (B). Dotted line indicates the cutoff value for thrombocytopenia.

  • View in gallery

    Leukocyte distribution in arbovirus and non-arbovirus cases (A) and diagnostic performance (area under the receiver operating characteristic [ROC] curve) for leukocyte counts to discriminate arbovirus and non-arbovirus infection cases (B).

  • 1.

    Bottieau E, Clerinx J, Van Den Enden E, Van Esbroeck M, Colebunders R, Van Gompel A, Van Den Ende J, 2007. Fever after a stay in the tropics: diagnostic predictors of the leading tropical conditions. Medicine (Baltimore) 86: 1825.

    • Search Google Scholar
    • Export Citation
  • 2.

    Freedman DO, Weild LH, Kozarsky PE, Fisk T, Robins R, von Sonnenburg F, Keystone JS, Pandey P, Cetron MS; GeoSentinel Surveillance Network, 2006. Spectrum of disease and relation to place of exposure among ill returned travellers. N Engl J Med 354: 119130.

    • Search Google Scholar
    • Export Citation
  • 3.

    Kutsuna S, Hayakawa K, Kato Y, Fujiya Y, Mawatari M, Takeshita N, Kanagawa S, Ohmagari N, 2015. Comparison of clinical characteristics and laboratory findings of malaria, dengue, and enteric fever in returning travelers : 8-year experience at a referral center in Tokyo, Japan. J Infect Chemother 21: 272276.

    • Search Google Scholar
    • Export Citation
  • 4.

    Wickramasinghe SN, Abdalla SH, 2000. Blood and bone marrow changes in malaria. Baillieres Best Pract Res Clin Haematol 13: 277299.

  • 5.

    Ladhani S, Lowe B, Cole AO, Kowuondo K, Newton CR, 2002. Changes in white blood cells and platelets in children with falciparum malaria: relationship to disease outcome. Br J Haematol 119: 839847.

    • Search Google Scholar
    • Export Citation
  • 6.

    Kelton JG, Keystone J, Moore J, Denomme G, Tozman E, Glynn M, Neame PB, Gauldie J, Jensen J, 1983. Immune-mediated thrombocytopenia of malaria. J Clin Invest 71: 832836.

    • Search Google Scholar
    • Export Citation
  • 7.

    Potts JA, Rothman AL, 2009. Clinical and laboratory features that distinguish dengue from other febrile illnesses in endemic populations. Trop Med Int Health 13: 13281340.

    • Search Google Scholar
    • Export Citation
  • 8.

    Daumas RP, Passos SR, Oliveira RV, Nogueira RM, Georg I, Marzochi KB, Brasil P, 2013. Clinical and laboratory features that discriminate dengue from other febrile illnesses: a diagnostic accuracy study in Rio de Janeiro, Brazil. BMC Infect Dis 13: 77.

    • Search Google Scholar
    • Export Citation
  • 9.

    Musso D, Gubler DJ, 2016. Zika virus. Nature 11: 1020.

  • 10.

    Lee VJ, Chow A, Zheng X, Carrasco LR, Cook AR, Lye DC, Ng LC, Leo YS, 2012. Simple clinical and laboratory predictors of chikungunya versus dengue infections in adults. PLoS Negl Trop Dis 6: e1786.

    • Search Google Scholar
    • Export Citation
  • 11.

    Chraïbi S, Najioullah F, Bourdin C, Pegliasco J, Deligny C, Résière D, Meniane JC, 2016. Two cases of thrombocytopenic purpura at onset of Zika virus infection. J Clin Virol 83: 6162.

    • Search Google Scholar
    • Export Citation
  • 12.

    Steffen R, Hill DR, DuPont HL, 2015. Traveler’s diarrhea: a clinical review. JAMA 313: 71.

  • 13.

    Okhuysen PC, 2013. Traveler’s diarrhea due to intestinal protozoa. Clin Infect Dis 33: 110114.

  • 14.

    Azmatullah A, Qamar FN, Thaver D, Zaidi AK, Buhatta ZA, 2015. Systematic review of the global epidemiology, clinical and laboratory profile of enteric fever. J Glob Health 5: 020407.

    • Search Google Scholar
    • Export Citation
  • 15.

    Zboromyrska Y, Hurtado JC, Salvador P, Alvarez-Martínez MJ, Valls ME, Mas J, Marcos MA, Gascón J, Vila J, 2014. Aetiology of traveller’s diarrhoea: evaluation of a multiplex PCR tool to detect different enteropathogens. Clin Microbiol Infect 20: O753O759.

    • Search Google Scholar
    • Export Citation
  • 16.

    Johnson BW, Russell BJ, Lanciotti RS, 2005. Serotype-specific detection of dengue viruses in a fourplex real-time reverse transcriptase PCR assay. J Clin Microbiol 43: 49774983.

    • Search Google Scholar
    • Export Citation
  • 17.

    Aranda KR, Fagundes-Neto U, Scaletsky IC, 2004. Evaluation of multiplex PCRs for diagnosis of infection with diarrheagenic Escherichia coli and Shigella spp. J Clin Microbiol 42: 58495853.

    • Search Google Scholar
    • Export Citation
  • 18.

    Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM, Bloomfield CD, Cazzola M, Vardiman JW, 2016. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood 127: 23912406.

    • Search Google Scholar
    • Export Citation
  • 19.

    Thachil J, 2017. Platelets and infections in the resource-limited countries with a focus on malaria and viral haemorrhagic fevers. Br J Haematol 177: 960970.

    • Search Google Scholar
    • Export Citation
  • 20.

    Grynberg P, Fernandes Fontes CJ, Braga EM, 2007. Association between particular polymorphic residues on apical membrane antigen 1 (AMA-1) and platelet levels in patients with vivax malaria. Clin Microbiol Infect 13: 10891094.

    • Search Google Scholar
    • Export Citation
  • 21.

    Leal-Santos FA, Silva SB, Crepaldi NP, Nery AF, Martin TO, Alves-Junior ER, Fontes CJ, 2013. Altered platelet indices as potential markers of severe and complicated malaria caused by Plasmodium vivax: a cross-sectional descriptive study. Malar J 12: 462.

    • Search Google Scholar
    • Export Citation
  • 22.

    Gachot B, Ringwald P, 1998. Severe malaria [article in French]. Rev Prat 48: 273278.

  • 23.

    Berens-Riha N et al. 2014. Evidence for significant influence of host immunity on changes in differential blood count during malaria. Malar J 13: 155.

    • Search Google Scholar
    • Export Citation
  • 24.

    Lacerda MV, Mourão MP, Coelho HC, Santos JB, 2011. Thrombocytopenia in malaria: who cares? Mem Inst Oswaldo Cruz 106 (Suppl 1): 5263.

  • 25.

    Lampah DA et al. 2015. Severe malarial thrombocytopenia: a risk factor for mortality in Papua, Indonesia. J Infect Dis 211: 623634.

  • 26.

    Khan SJ, Abbass Y, Marwat MA, 2012. Thrombocytopenia as an indicator of malaria in adult population. Malar Res Treat 2012: 405981.

  • 27.

    Kochar DK et al. 2010. Thrombocytopenia in Plasmodium falciparum, Plasmodium vivax and mixed infection malaria: a study from Bikaner (northwestern India). Platelets 21: 623627.

    • Search Google Scholar
    • Export Citation
  • 28.

    Snow RW, Guerra CA, Noor AM, Myint HY, Simon I, 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434: 214217.

    • Search Google Scholar
    • Export Citation
  • 29.

    Jelinek T et al. 2002. Imported falciparum malaria in Europe: sentinel surveillance data from the European network on surveillance of imported infectious diseases. Clin Infect Dis 34: 572576.

    • Search Google Scholar
    • Export Citation
  • 30.

    Pistone T, Diallo A, Mechain M, Receveur MC, Malvy D, 2014. Epidemiology of imported malaria give support to the hypothesis of “long-term” semi-immunity to malaria in sub-Saharan African migrants living in France. Travel Med Infect Dis 12: 4853.

    • Search Google Scholar
    • Export Citation
  • 31.

    Possas C, 2016. Zika: what we do and do not know based on the experiences of Brazil. Epidemiol Health 38: e2016023.

  • 32.

    Duijster JW et al. Dutch ZIKV Study Team, 2016. Zika virus infection in 18 travellers returning from Surinam and the Dominican Republic, The Netherlands, November 2015–March 2016. Infection 44: 797802.

    • Search Google Scholar
    • Export Citation
  • 33.

    Meltzer E, Leshem E, Lustig Y, Gottesman G, Schwartz E, 2016. The clinical spectrum of Zika virus in returning travelers. Am J Med 129: 11261130.

  • 34.

    Leder K et al. GeoSentinel Surveillance Network, 2013. GeoSentinel surveillance of illness in returned travelers, 2007–2011. Ann Intern Med 158: 456468.

    • Search Google Scholar
    • Export Citation
  • 35.

    Harvey K, Esposito DH, Han P, Kozarsky P, Freedman DO, Plier DA, Sotir MJ; Centers for Disease Control and Prevention (CDC), 2013. Surveillance for travel-related disease—GeoSentinel Surveillance System, United States, 1997–2011. MMWR Surveill Summ 62: 123.

    • Search Google Scholar
    • Export Citation
 
 
 

 

 
 
 

 

 

 

 

 

 

Diagnostic Value of Platelet and Leukocyte Counts in the Differential Diagnosis of Fever in the Returning Traveler

View More View Less
  • 1 Department of Clinical Microbiology, Hospital Clinic, Barcelona, Spain;
  • | 2 Department of Tropical Medicine and International Health, Hospital Clinic, Barcelona, Spain;
  • | 3 ISGlobal Barcelona Institute for Global Health, Barcelona, Spain

Malaria, arbovirus infection and travelers’ diarrhea are among the most common etiologies of fever after a stay in the tropics. Because the initial symptoms of these diseases often overlap, the differential diagnostic remains a challenge. The aim of this study was to establish the effectiveness of platelet and leukocyte counts in the differential diagnosis of fever in the returning traveler. Between 2013 and 2016, patients with a clinical suspicion of malaria, who had thick blood smears performed were retrospectively included. The microbiological etiology of each episode was established based on molecular detection in the case of arbovirus infection, the detection of pathogens in stool samples for diarrhea and other gastrointestinal symptoms and the thick and thin blood smear results for malaria. A total of 1,218 episodes were included. Malaria, arbovirus infection, and diarrhea and other gastrointestinal symptoms caused 102 (8.4%), 68 (5.6%), and 72 (5.9%) episodes, respectively. The median platelet counts in malaria episodes were 89 × 109/L and thrombocytopenia (< 150,000 × 109 platelets/L) yielded a 98% negative predictive value to predict malaria. The median leukocyte counts in arbovirus infection episodes were 3.19 × 109/L and leucopenia (< 4 × 109 leukocytes/L) yielded a 97.9% negative predictive value to predict arbovirus infections. Platelet and leukocyte counts were not significantly altered in episodes caused by diarrhea and other gastrointestinal symptoms. Initial platelet and leukocyte counts might be useful for the clinical differential diagnosis of fever in the returning traveler. Although these results are insufficient to establish a diagnosis, they should be considered in the initial clinical assessment.

INTRODUCTION

Fever is one of the main reasons to attend emergency departments in travelers returning from tropical areas. Malaria, arbovirus infection, and travelers’ diarrhea are among the most common etiologies of fever after a stay in the tropics.13 The initial clinical presentation for all these diseases is frequently unspecific and might include fever, arthralgia, headache, myalgia, and gastrointestinal symptoms. In consequence, the differential diagnosis based on clinical symptoms remains a challenge. Because of the increasing number of people travelling abroad and the difference in severity, outcome, and treatment of each condition, early diagnosis is crucial for the successful management of the patient, especially for life-threatening conditions such as malaria and severe dengue. Diagnostic predictors such as abnormalities in hematological parameters may be useful because they can suggest the etiology before the definitive microbiological diagnosis.

Acute malaria is often associated with mild or moderate thrombocytopenia in nonimmune adults and in children from malaria-endemic areas, and is a sensitive but nonspecific indicator of infection with malaria parasites. Profound thrombocytopenia is unusual. Malarial thrombocytopenia is rarely associated with hemorrhagic manifestations or disseminated intravascular coagulation either in nonimmune adults or children in endemic areas.46

Arbovirus infections are commonly accompanied by thrombocytopenia and also by leucopenia. Potts and Rotham7 reviewed differences in laboratory indicators between patients with dengue and other febrile illnesses (measles, typhoid fever, leptospirosis, and severe acute respiratory syndrome) in endemic areas and found that leukocyte and platelet counts were significantly lower among patients with dengue. Moreover, in endemic areas leukocyte parameters have been used for diagnostic models of dengue in combination with clinical symptoms.8 The same modifications in hematological parameters have been reported for Zika,9 chikungunya, and other arbovirus infections. These parameters are sensitive for arbovirus infection, but they lack specificity, as they vary among different arbovirus infections10 and they heighten in more severe cases.7,11

Travelers’ diarrhea and other gastrointestinal symptoms after a stay in a tropical or subtropical area can be caused by bacterial, viral, and parasitic pathogens. Laboratory indicators vary depending on the microbiological agent and its virulence. Few studies document abnormalities in hematological parameters as an indicator for travelers’ diarrhea, except for enteric fever, which is commonly associated with thrombocytopenia and either leucopenia or leukocytosis.1215

The aim of this is study was to establish the diagnostic effectiveness of leukocyte and platelet counts in the differential diagnosis of patients returning from tropical or subtropical areas with a clinical suspicion of malaria.

MATERIALS AND METHODS

Patients.

The study was conducted at the Hospital Clinic i Provincial de Barcelona (an 800-bed teaching hospital in Barcelona, Spain). Following current clinical guidelines, screening for malaria in all febrile patients returning from tropical and subtropical areas is performed. This protocol includes fever measured using a thermometer and also a history of fever reported by the patient. Cases for this study were selected based on the thick and thin blood smear tests for malaria requested by the attending physician. All thick blood smears performed in the microbiology and parasitology department between 2013 and 2016 were retrospectively analyzed. Thick blood smears performed as a control of the evolution in already diagnosed patients were excluded, and only the first thick blood smear of each episode was analyzed. Date of analysis, last visited country, identified Plasmodium species, and parasite density were collected for each episode and patient. In addition, the results from the different diagnostic methods to detect arbovirus infections and data on bacterial and parasitic intestinal pathogens were collected retrospectively. Finally, platelet and leukocyte counts at the beginning of each malaria-suspected episode (±3 days difference from the date of performance of the first thick and thin blood smear) were collected retrospectively.

Diagnostic procedures.

The diagnosis of malaria was defined by the presence of Plasmodium trophozoites in the thick and thin blood smears. In addition, a rapid diagnostic test (BinaxNOW Malaria, Alere, Ireland) was performed in some cases depending on the last visited country and the chemoprophylaxis taken by the patient. In cases where the Plasmodium species could not be established solely by microscopy, a commercial nested polymerase chain reaction (PCR) with specific primers for Plasmodium falciparum, Plasmodium vivax, Plasmodium malariae, and Plasmodium ovale was performed (BIOMALAR Gel Form Kit; BioTools, Madrid, Spain). PCR products were visualized in a 2% agarose gel.

Confirmed arbovirus infections were defined based on a positive specific real-time reverse transcription polymerase chain reaction (RT-PCR) for dengue, chikungunya, or Zika in serum or urine samples within the malaria suspected episode. An in-house multiplex real-time PCR was used for the molecular diagnosis of the four dengue serotypes using EXPRESS One-Step SuperScript qRT-PCR (Thermo Fisher, Madrid, Spain).16 For the molecular diagnosis of chikungunya and Zika, two different commercial real-time RT-PCR were performed: RealStar chikungunya RT-PCR (Altona Diagnostics, Hamburg, Germany) and RealStar Zika RT-PCR (Altona Diagnostics).

The diagnosis of travelers’ diarrhea or presence of other intestinal pathogens was established with the detection of viral, bacterial, or parasitic pathogens in a stool sample collected during the same episode as the thick blood smear.

The routine methods for the detection of bacterial pathogens were cultures of the stool sample in Blood agar (Oxoid®; Thermo Fisher), MacConkey agar (Becton Dickinson®, Heidelberg, Germany), CCDA agar (Becton Dickinson®) for Campylobacter isolation, Salmonella Shigella (SS) agar (Becton Dickinson®) for Shigella and Salmonella isolation, Cefsulodin-Irgasan-Novobiocin (CIN) agar (Becton Dickinson®) for Yersinia isolation, and Rappaport-Vassiliadis Salmonella Enrichment Broth (Becton Dickinson®) for the recovery of Salmonella and latter planting onto SS agar. The isolated bacteria were identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (Bruker, Bremen, Germany), and the Salmonella/Shigella identified were serotyped by agglutination with commercial antisera (Bio-Rad®, Marnes-la-Coquette, France). An in-house multiplex PCR15,17 was used for the detection of virulence genes of diarrheagenic Escherichia coli targeted to CVD432 probe of enteroaggregative E. coli and the lt and st genes of enterotoxigenic E. coli. PCR products were visualized in a 2% agarose gel.

Parasitic pathogens were detected by direct microscopic observation from fresh and concentrated stool samples with the merthiolate formalin ether method and the modified Kinyoun acid-fast stain from the concentrated samples for the detection of Cryptosporidium and Cyclospora cayetanensis.

The detection of rotavirus and adenovirus was performed with a rapid immunochromatographic test (bioNexia®; Rota-Adeno, Marcy-l’Etoile, France) from stool samples.

The different diagnostic techniques were performed based on diagnostic suspicion and clinical criteria. We should emphasize that there was no clinical request for Norovirus detection in the group of patients included in the study.

Statistical analysis.

Clinical, epidemiologic, and laboratory data were collected on standardized forms and entered in a password protected database. Univariate and multiple logistic regression models were fitted for all clinical variables available using etiology as dependent variables. Interaction of independent variables was checked using the likelihood ratio test.

The area under the receiver operating characteristic (ROC) curves was used to compare the sensitivity and specificity of selected markers. Cutoff values were chosen based on the highest sensitivity and specificity to predict outcome using the “roctab/detail” function (Stata 11.0; Stata Corporation, College Station, TX). When the diagnostic performance was assessed for more than one variable, the estimates were derived from a logistic regression model using the selected markers or clinical features as independent variables and the condition to diagnose as the dependent variable. These analyses were carried out using the “lroc,” “lstat,” and “roctab/graph” functions in Stata. Data were analyzed with STATA 11.

RESULTS

Characteristics of the study population and geographic distribution.

During the study period, 1,218 episodes belonging to 1,185 patients returning from tropical or subtropical areas with a clinical suspicion of malaria were included in the study, and 46.8% were female. The most commonly visited area was sub-Saharan Africa (47.5%), followed by Southern Asia/Pacific (31.3%), Latin America (19.5%), and North Africa/Middle East (1.6%).

Among patients diagnosed with malaria, 85.3% had returned from sub-Saharan Africa; the single country with most malaria cases was Equatorial Guinea. Sub-Saharan Africa was the origin of 93% of patients diagnosed with P. falciparum malaria and all patients diagnosed with P. malariae and P. ovale malaria. Patients diagnosed with P. vivax malaria had arrived from South Asia/Pacific (55.6%) and Latin America (44.4%). Patients with arbovirus infection were mostly diagnosed after a stay in Southern Asia/Pacific (52.9%) and Latin America (44.1%), with Thailand and Indonesia as the most frequently visited countries. The distribution of patients with travelers’ diarrhea and parasitic intestinal pathogens was more heterogeneous, although the most frequently visited region was sub-Saharan Africa (41.7%), and the most frequent visited country was India. Figure 1 shows the geographical distribution of the cases diagnosed with malaria and arbovirus infection.

Figure 1.
Figure 1.

Geographic destination of travelers included in the study with malaria and confirmed arbovirus infection from 2013 to 2016.

Citation: The American Journal of Tropical Medicine and Hygiene 100, 2; 10.4269/ajtmh.18-0736

Confirmed etiologic diagnosis.

A confirmed etiologic diagnosis of arbovirus infection, malaria, or travelers’ diarrhea and other intestinal pathogens was found in 238 (19.5%) cases. Malaria was responsible for 102 episodes (8.4%), arbovirus infection for 68 episodes (5.6%), and traveler’s diarrhea and other gastrointestinal symptoms for 72 episodes (5.9%) (Table 1). Four patients were diagnosed with more than one tropical infection: three cases of malaria and an intestinal pathogen and one case of arbovirus infection and an intestinal pathogen.

Table 1

Etiology of infection in travelers presenting with acute febrile illness with a suspected diagnosis of malaria

AetiologyNumber (%)N%
Unknown979 (80.4)*
Arbovirus68 (5.6)*
 Dengue5073.5
 Chikungunya1217.7
 Zika68.8
Malaria102 (8.4)*
Plasmodium falciparum8684.3
Plasmodium vivax98.8
Plasmodium ovale43.9
Plasmodium malariae22.0
Plasmodium spp.11.0
Traveller’s diarrhea40 (3.3)*
 EAEC2152.5
 ETEC1025.0
 ETEC + Shigella flexneri12.5
Campylobacter jejuni37.5
Salmonella typhimurium25.0
Salmonella spp.12.5
Vibrio spp.12.5
Shigella sonnei12.5
Intestinal parasites32 (2.6)*
Blastocystis hominis1031.3
Cyclospora cayetanensis13.1
Entamoeba histolytica/dispar39.4
E. histolytica/dispar + Iodamoeba butschlii13.1
E. histolytica/dispar + Chilomastix mesnili13.1
E. histolytica/dispar + Giardia lamblia13.1
G. lamblia1134.4
G. lamblia + C. mesnili13.1
G. lamblia + C. mesnili + Cryptosporidium sp.13.1
Hymenolepis nana13.1
Schistosoma intercalatum + Trichuris trichiura13.1

EAEC = enteroaggereagtive Escherichia coli; ETEC = enterotoxigenic E. coli.

* Percentages may not total 100 because patients may have more than one diagnosis.

Among the malaria-confirmed cases, the most common species was P. falciparum, followed by P. vivax, P. ovale, and P. malariae. In one case, neither microscopic nor molecular methods could determine the Plasmodium species. The median parasitemia was 0.6% (IQR: 0.05–3.6). The most frequently found arbovirus was dengue, followed by chikungunya and Zika.

Regarding gastrointestinal symptoms, 40 cases were caused by a bacterial pathogen and 32 by a parasitic pathogen. However, we did not perform molecular or serological investigations to look for potential viruses associated with gastrointestinal (GI) symptoms. Diarrheagenic E. coli was the most commonly identified bacterial agent and Giardia lamblia and Blastocystis hominis were the most commonly identified parasitic agents. The latter probably not being the cause of infection. Table 1 shows the etiology of all intestinal pathogens identified.

Leukocyte and platelet counts may help establish fever etiology.

Figure 2 shows a scatter plot with the distribution of platelets and leukocytes counts according to the different etiologies, and Table 2 shows the median leukocyte and platelet counts according to the malaria and arbovirus etiologies. The four episodes with more than one etiology were classified in the malaria and arbovirus infection group because it is known that these diseases cause more abnormalities in platelet and leukocyte counts. Although some overlap was found, malaria presented a major trend toward thrombocytopenia without affecting leukocyte counts. In the cases of arbovirus, both leucopenia and thrombocytopenia were observed.

Figure 2.
Figure 2.

Distribution of white blood cell and platelet counts on admission in travelers with malaria and arbovirus infection.

Citation: The American Journal of Tropical Medicine and Hygiene 100, 2; 10.4269/ajtmh.18-0736

Table 2

Platelet and leukocyte counts according to the etiology of fever in the returning traveler

Etiology (N)Platelet counts (109/L)Leukocyte counts (109/L)
MedianIQRMedianIQR
Malaria (102)8952–1354.753.73–6.00
Arbovirus infection (68)154120–19533.192.24–4.18
Travelers’ diarrhea bacteria (40)257204–3087.285.39–8.63
Intestinal parasites (32)209146–2646.374.89–9.31
Unknown (979)215175–2576.775.12–9.21

Malaria.

Figure 3A shows the platelet distribution among patients with and without malaria. Platelet counts were significantly lower (P < 0.001) in patients with malaria compared with patients without malaria. The ROC curve analysis for platelet count and malaria yielded an area under the curve of 0.89 (95% CI: 0.86–0.93, P < 0.05) (Figure 4B). Applying a cutoff value of 150,000 × 109/L (thrombocytopenia), the sensitivity, specificity, negative and positive predictive values were 82.4%, 82.9%, 98.1%, and 30.6%, respectively. There was a higher tendency toward thrombocytopenia in episodes caused by P. falciparum than for the other Plasmodium species, but these differences were not statistically significant. A weak but significant inverse correlation between parasitemia levels and platelet counts was observed (r = −0.31, P < 0.01). Eighteen patients diagnosed with malaria did not have thrombocytopenia. In 16 cases, it was caused by P. falciparum in patients returning from sub-Saharan Africa, one case was caused by P. vivax in a patient returning from Pakistan, and one case was caused by P. ovale in a patient returning from Cameroon. Among the P. falciparum cases, three patients were travelers of sub-Saharan Africa origin living in a nonendemic area and visiting family or relatives in their country of origin, five patients had undergone previous episodes of malaria in previous travels to endemic areas, one patient was already under treatment when the platelet count was performed, and one patient was splenectomized.

Figure 3.
Figure 3.

Platelet distribution in malaria and non-malaria cases (A) and diagnostic performance (area under the receiver operating characteristic [ROC] curve) for platelet counts to discriminate malaria and non-malaria cases (B). Dotted line indicates the cutoff value for thrombocytopenia.

Citation: The American Journal of Tropical Medicine and Hygiene 100, 2; 10.4269/ajtmh.18-0736

Figure 4.
Figure 4.

Leukocyte distribution in arbovirus and non-arbovirus cases (A) and diagnostic performance (area under the receiver operating characteristic [ROC] curve) for leukocyte counts to discriminate arbovirus and non-arbovirus infection cases (B).

Citation: The American Journal of Tropical Medicine and Hygiene 100, 2; 10.4269/ajtmh.18-0736

Arbovirus infection.

Both platelet and leukocyte counts in patients with arbovirus infection were significantly lower (P < 0.001) compared with patients without arbovirus infection (Table 2, Figure 4A). The ROC curve analysis for platelet and leukocyte counts and arbovirus infection yielded an area under the curve of 0.71 for platelet counts and 0.88 for leukocyte counts (Figure 4B). Applying a cutoff value of 4.00 × 109 leukocytes/L (leucopenia), the sensitivity, specificity, negative and positive predictive values were 72.1%, 87.3%, 98.1%, and 25.1%, respectively. The median platelet counts for patients with dengue, chikungunya, and Zika were 143, 223, and 155 (×109/L), respectively. The median leukocyte counts for patients with dengue, chikungunya, and Zika were 2.68, 3.24, and 3.55 (×109/L), respectively. The platelet and leukocyte levels were significantly lower in episodes caused by dengue (P = 0.002 and P = 0.004, respectively) compared with episodes caused by chikungunya.

The ROC curve for leukocyte counts, including only arbovirus infection and malaria cases, yielded an area under the curve of 0.78, and leucopenia had sensitivity, specificity, negative and positive predictive values of 72.1%, 59.8%, 78.4%, and 67.7%, respectively, to differentiate arbovirus infection from malaria.

Travelers’ diarrhea and other intestinal pathogens.

No hematological parameter was significantly altered for patients, where a parasitic intestinal pathogen was identified. Platelet levels among patients with diarrhea of bacterial etiology were significantly higher (P < 0.001) compared with patients without this etiology. However, none showed a platelet count greater than 450,000 × 109/L (thrombocythemia18). Leukocyte levels were not significantly altered in patients with travelers’ diarrhea of bacterial etiology.

DISCUSSION

In this study, we have shown that an initial laboratory screening using hematological parameters such as platelet and leukocyte counts might assist the clinical differential diagnosis in returning travelers. Indeed, we show that in the case of malaria, platelet counts under 150,000 × 109/L (thrombocytopenia) have sensitivity and specificity values of 82.4%, and 82.9%, respectively, and a negative predictive value of 98.1%. Although this diagnostic performance is not sufficient to establish a diagnosis, the clinician can use its high negative predictive value to consider diagnoses other than malaria in an initial assessment. Moreover, we show an inverse correlation between parasitemia levels and platelet count, indicating that levels of thrombocytopenia can be used as a predictive factor of severity in malaria. Few publications directly correlate parasitemia with lower platelet counts, and this correlation has mainly been presented in P. vivax malaria.1921 Thrombocytopenia in the absence of significant bleeding is not considered a clinical manifestation for severe malaria.22 However, some studies correlate the levels of thrombocytopenia with malaria severity.2325 Our results show a tendency for P. falciparum malaria to have lower platelet counts compared with other malaria species. These results agree with previous studies,25,26 although other publications27 show that P. vivax infection presents higher rates of thrombocytopenia. Our study does not have enough cases of non-falciparum malaria to verify if thrombocytopenia is a discriminating feature for a particular type of malaria.

Among patients diagnosed with P. falciparum malaria without thrombocytopenia, we show a high rate of patients with previous episodes of malaria (8/16). It is known that the repeated exposure to malaria parasites is associated with semi-immunity against malaria and a reduced risk of severe malaria.28 Some studies have shown that malaria severity is lower in patients of sub-Saharan Africa origin living in non-endemic areas and visiting family or relatives in their country of origin.29,30 In addition, different hematological patterns have been observed in semi-immune patients compared with nonimmune patients, including a lower rate of severe thrombocytopenia.23,30 These findings could explain the lack of thrombocytopenia in patients previously exposed to malaria.

We have shown that low platelet counts are not specific to malaria because they are also associated with arbovirus infections. However, in arbovirus infections, we usually find concomitant leucopenia.7 In this study, we observed statistically lower leukocyte and platelet levels associated with dengue, chikungunya, and Zika. In our sample, leukocyte levels show a better diagnostic performance than platelets because leukocyte levels are more specifically altered in arbovirus infections.

Leukocyte counts of 4.00 × 109 leukocytes/L or less (leucopenia) have sensitivity and specificity values of 72.1% and 87.3%, respectively, and a negative predictive value of 98.1% for the detection of arbovirus infection. In addition, leucopenia showed a negative predictive value of 78.4% to differentiate malaria from arbovirus infection.

In agreement with previous studies,10 our results show that leucopenia and specially thrombocytopenia are more severe in dengue compared with chikungunya cases. No statistical significance in hematological parameters is shown for episodes caused by Zika due to the low number of included cases, since we only started testing for Zika virus in 2016, after the large 2015 Brazil outbreak.31 However, in agreement with previously reported cases, Zika cases in our study presented mild leucopenia.32,33

No hematological parameter was significantly altered in episodes of travelers’ diarrhea or where parasitic intestinal pathogens were identified, probably because of the lack of invasiveness of parasitic and bacterial agents causing gastrointestinal symptoms. Blastocystis hominis or Hymenolepis nana were identified in some episodes. Probably, in these cases the focus of fever was other than gastrointestinal but a parasitic study was additionally demanded by the clinician because of gastrointestinal symptoms or other epidemiological characteristics.

In the cases included in our study, the country visited did not provide a major insight into the etiology. Most cases of P. falciparum malaria came from sub-Saharan Africa, particularly from Equatorial Guinea. However, because Equatorial Guinea is a former Spanish colony, and therefore a preferred destination, we do not consider it to reflect the severity of malaria in this country. Most cases of arbovirus infection arrived from Southern Asia/Pacific and Latin America. Our results show that the distribution of travelers’ diarrhea and other gastrointestinal symptoms is more heterogeneous among tropical areas. These findings are in agreement with previously reported epidemiological information.34

This study has several limitations. First, underlying conditions that could potentially alter leukocyte or platelet counts were not systematically investigated. Second, some of the selected cases may not correspond to acute febrile illness but rather to reported fevers. Finally, diagnoses other than malaria and arbovirus infection were not analyzed. These limitations would explain the large number of episodes of unknown etiology included in the study, although previous studies show a 25% rate of undiagnosed episodes of fever after travelling to the tropics.3,35 However, platelet and leukocyte counts were able to set apart malaria and arbovirus infections from other conditions. These indicators are inexpensive, easily available, and can provide useful information in the differential diagnosis of returning travelers presenting with fever.

In conclusion, this study provides quantitative evidence for the potential role of platelet and leukocyte counts as diagnostic predictors of malaria and arbovirus infection, two of the most common causes of fever in the returning traveler and which may rapidly evolve into severe diseases. These parameters can be used as an initial guidance to accelerate the definitive diagnosis and administer the appropriate treatment.

REFERENCES

  • 1.

    Bottieau E, Clerinx J, Van Den Enden E, Van Esbroeck M, Colebunders R, Van Gompel A, Van Den Ende J, 2007. Fever after a stay in the tropics: diagnostic predictors of the leading tropical conditions. Medicine (Baltimore) 86: 1825.

    • Search Google Scholar
    • Export Citation
  • 2.

    Freedman DO, Weild LH, Kozarsky PE, Fisk T, Robins R, von Sonnenburg F, Keystone JS, Pandey P, Cetron MS; GeoSentinel Surveillance Network, 2006. Spectrum of disease and relation to place of exposure among ill returned travellers. N Engl J Med 354: 119130.

    • Search Google Scholar
    • Export Citation
  • 3.

    Kutsuna S, Hayakawa K, Kato Y, Fujiya Y, Mawatari M, Takeshita N, Kanagawa S, Ohmagari N, 2015. Comparison of clinical characteristics and laboratory findings of malaria, dengue, and enteric fever in returning travelers : 8-year experience at a referral center in Tokyo, Japan. J Infect Chemother 21: 272276.

    • Search Google Scholar
    • Export Citation
  • 4.

    Wickramasinghe SN, Abdalla SH, 2000. Blood and bone marrow changes in malaria. Baillieres Best Pract Res Clin Haematol 13: 277299.

  • 5.

    Ladhani S, Lowe B, Cole AO, Kowuondo K, Newton CR, 2002. Changes in white blood cells and platelets in children with falciparum malaria: relationship to disease outcome. Br J Haematol 119: 839847.

    • Search Google Scholar
    • Export Citation
  • 6.

    Kelton JG, Keystone J, Moore J, Denomme G, Tozman E, Glynn M, Neame PB, Gauldie J, Jensen J, 1983. Immune-mediated thrombocytopenia of malaria. J Clin Invest 71: 832836.

    • Search Google Scholar
    • Export Citation
  • 7.

    Potts JA, Rothman AL, 2009. Clinical and laboratory features that distinguish dengue from other febrile illnesses in endemic populations. Trop Med Int Health 13: 13281340.

    • Search Google Scholar
    • Export Citation
  • 8.

    Daumas RP, Passos SR, Oliveira RV, Nogueira RM, Georg I, Marzochi KB, Brasil P, 2013. Clinical and laboratory features that discriminate dengue from other febrile illnesses: a diagnostic accuracy study in Rio de Janeiro, Brazil. BMC Infect Dis 13: 77.

    • Search Google Scholar
    • Export Citation
  • 9.

    Musso D, Gubler DJ, 2016. Zika virus. Nature 11: 1020.

  • 10.

    Lee VJ, Chow A, Zheng X, Carrasco LR, Cook AR, Lye DC, Ng LC, Leo YS, 2012. Simple clinical and laboratory predictors of chikungunya versus dengue infections in adults. PLoS Negl Trop Dis 6: e1786.

    • Search Google Scholar
    • Export Citation
  • 11.

    Chraïbi S, Najioullah F, Bourdin C, Pegliasco J, Deligny C, Résière D, Meniane JC, 2016. Two cases of thrombocytopenic purpura at onset of Zika virus infection. J Clin Virol 83: 6162.

    • Search Google Scholar
    • Export Citation
  • 12.

    Steffen R, Hill DR, DuPont HL, 2015. Traveler’s diarrhea: a clinical review. JAMA 313: 71.

  • 13.

    Okhuysen PC, 2013. Traveler’s diarrhea due to intestinal protozoa. Clin Infect Dis 33: 110114.

  • 14.

    Azmatullah A, Qamar FN, Thaver D, Zaidi AK, Buhatta ZA, 2015. Systematic review of the global epidemiology, clinical and laboratory profile of enteric fever. J Glob Health 5: 020407.

    • Search Google Scholar
    • Export Citation
  • 15.

    Zboromyrska Y, Hurtado JC, Salvador P, Alvarez-Martínez MJ, Valls ME, Mas J, Marcos MA, Gascón J, Vila J, 2014. Aetiology of traveller’s diarrhoea: evaluation of a multiplex PCR tool to detect different enteropathogens. Clin Microbiol Infect 20: O753O759.

    • Search Google Scholar
    • Export Citation
  • 16.

    Johnson BW, Russell BJ, Lanciotti RS, 2005. Serotype-specific detection of dengue viruses in a fourplex real-time reverse transcriptase PCR assay. J Clin Microbiol 43: 49774983.

    • Search Google Scholar
    • Export Citation
  • 17.

    Aranda KR, Fagundes-Neto U, Scaletsky IC, 2004. Evaluation of multiplex PCRs for diagnosis of infection with diarrheagenic Escherichia coli and Shigella spp. J Clin Microbiol 42: 58495853.

    • Search Google Scholar
    • Export Citation
  • 18.

    Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM, Bloomfield CD, Cazzola M, Vardiman JW, 2016. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood 127: 23912406.

    • Search Google Scholar
    • Export Citation
  • 19.

    Thachil J, 2017. Platelets and infections in the resource-limited countries with a focus on malaria and viral haemorrhagic fevers. Br J Haematol 177: 960970.

    • Search Google Scholar
    • Export Citation
  • 20.

    Grynberg P, Fernandes Fontes CJ, Braga EM, 2007. Association between particular polymorphic residues on apical membrane antigen 1 (AMA-1) and platelet levels in patients with vivax malaria. Clin Microbiol Infect 13: 10891094.

    • Search Google Scholar
    • Export Citation
  • 21.

    Leal-Santos FA, Silva SB, Crepaldi NP, Nery AF, Martin TO, Alves-Junior ER, Fontes CJ, 2013. Altered platelet indices as potential markers of severe and complicated malaria caused by Plasmodium vivax: a cross-sectional descriptive study. Malar J 12: 462.

    • Search Google Scholar
    • Export Citation
  • 22.

    Gachot B, Ringwald P, 1998. Severe malaria [article in French]. Rev Prat 48: 273278.

  • 23.

    Berens-Riha N et al. 2014. Evidence for significant influence of host immunity on changes in differential blood count during malaria. Malar J 13: 155.

    • Search Google Scholar
    • Export Citation
  • 24.

    Lacerda MV, Mourão MP, Coelho HC, Santos JB, 2011. Thrombocytopenia in malaria: who cares? Mem Inst Oswaldo Cruz 106 (Suppl 1): 5263.

  • 25.

    Lampah DA et al. 2015. Severe malarial thrombocytopenia: a risk factor for mortality in Papua, Indonesia. J Infect Dis 211: 623634.

  • 26.

    Khan SJ, Abbass Y, Marwat MA, 2012. Thrombocytopenia as an indicator of malaria in adult population. Malar Res Treat 2012: 405981.

  • 27.

    Kochar DK et al. 2010. Thrombocytopenia in Plasmodium falciparum, Plasmodium vivax and mixed infection malaria: a study from Bikaner (northwestern India). Platelets 21: 623627.

    • Search Google Scholar
    • Export Citation
  • 28.

    Snow RW, Guerra CA, Noor AM, Myint HY, Simon I, 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434: 214217.

    • Search Google Scholar
    • Export Citation
  • 29.

    Jelinek T et al. 2002. Imported falciparum malaria in Europe: sentinel surveillance data from the European network on surveillance of imported infectious diseases. Clin Infect Dis 34: 572576.

    • Search Google Scholar
    • Export Citation
  • 30.

    Pistone T, Diallo A, Mechain M, Receveur MC, Malvy D, 2014. Epidemiology of imported malaria give support to the hypothesis of “long-term” semi-immunity to malaria in sub-Saharan African migrants living in France. Travel Med Infect Dis 12: 4853.

    • Search Google Scholar
    • Export Citation
  • 31.

    Possas C, 2016. Zika: what we do and do not know based on the experiences of Brazil. Epidemiol Health 38: e2016023.

  • 32.

    Duijster JW et al. Dutch ZIKV Study Team, 2016. Zika virus infection in 18 travellers returning from Surinam and the Dominican Republic, The Netherlands, November 2015–March 2016. Infection 44: 797802.

    • Search Google Scholar
    • Export Citation
  • 33.

    Meltzer E, Leshem E, Lustig Y, Gottesman G, Schwartz E, 2016. The clinical spectrum of Zika virus in returning travelers. Am J Med 129: 11261130.

  • 34.

    Leder K et al. GeoSentinel Surveillance Network, 2013. GeoSentinel surveillance of illness in returned travelers, 2007–2011. Ann Intern Med 158: 456468.

    • Search Google Scholar
    • Export Citation
  • 35.

    Harvey K, Esposito DH, Han P, Kozarsky P, Freedman DO, Plier DA, Sotir MJ; Centers for Disease Control and Prevention (CDC), 2013. Surveillance for travel-related disease—GeoSentinel Surveillance System, United States, 1997–2011. MMWR Surveill Summ 62: 123.

    • Search Google Scholar
    • Export Citation

Author Notes

Address correspondence to Elisa Rubio, Department of Clinical Microbiology, Hospital Clínic de Barcelona, Villarroel, 170, Barcelona 08036, Spain. E-mail: elrubio@clinic.cat

Authors’ addresses: Elisa Rubio, Izaskun Alejo-Cancho, Cristian Aylagas, Roser Ferré, Ma Rosa Albarracín, Verónica Gonzalo, Josep Barrachina, Míriam José Álvarez-Martínez, Maria Eugenia Valls, Jordi Mas, Jordi Vila, Miguel J. Martínez, and Climent Casals-Pascual, Department of Clinical Microbiology, Hospital Clinic de Barcelona, Barcelona, Spain, E-mails: elrubio@clinic.cat, alejo@clinic.cat, aylagas@clinic.cat, mrferre@clinic.cat, ralbarra@clinic.cat, vgonzal1@clinic.cat, barrachina@clinic.cat, malvarez@clinic.cat, mevalls@clinic.cat, jmas@clinic.cat, jvila@clinic.cat, myoldi@clinic.cat, and ccasals@clinic.cat. Daniel Camprubí and Irene Losada, Department of Tropical Medicine and International Health, Hospital Clinic de Barcelona, Barcelona, Spain, E-mails: dcamprub@clinic.cat and ilosada@clinic.cat.

Save