Chikungunya and dengue virus infection among febrile children in North-Eastern Tanzania: prospective study

There are several unknown illnesses including chikungunya and dengue viruses that present with fever in children. Therefore there are many cases that are misdiagnosed. Consequently we performed a study to determine the clinical characteristics of dengue and chikungunya in order to assist clinicians in management. A total of 196 children with history of fever for ≤ 10 days were enrolled prospectively at Kilimanjaro Christian Medical Centre from September 2015 to May 2016. All cases were screened for chikungunya and dengue viruses by PCR as well as other febrile illnesses such as malaria, bacteria and HIV through other diagnostic method. We performed logistic regression to nd association between clinical symptoms and chikungunya infection.


Conclusion
This study reveals that chikungunya infection is common cause of febrile illnesses; however it does not have treatment. Therefore, we encourage chikungunya infection to be included in routine investigation as these children are receiving inappropriate treatments such as antibiotics and ant malarial which should be avoided in order to minimize over use of drugs and resource wastage.
Background Page 3/13 Chikungunya and dengue are arthropod-borne viruses transmitted between humans by the bite of infected female mosquitoes of the genus Aedes, particularly, Ae. Aegypti and Ae. Albopictus (1). At initial stage of infection dengue and chikungunya present with similar febrile illness, such as fever, rashes, joint pain, headache, fatigue, nausea, vomiting and body pain(2). Dengue and chikungunya share similar geographical location and vector (3). Furthermore it is also di cult to distinguish dengue and chikungunya clinically, their clinical symptoms mimics other febrile illnesses such as malaria and typhoid (4). Dengue virus has four serotypes (serotypes 1, -2, -3, and − 4), secondary infection with similar serotype confers lifelong immunity but reinfection with other serotype the immune protection becomes weak (5). Chikungunya was rst identi ed in Tanzania in 1952-3 (6). Since the discovery of chikungunya, several epidemics in various parts of the world such as west and central African countries have been reported (7,8). However, there are limited occurrence data and less awareness regarding dengue infection in Africa (9). Tanzania is among of the African countries that are endemic for arbovirus vectors (10,11).
Unlike dengue, chikungunya is considered to have established as an endemic infection in Tanzania, also it is considered that chikungunya spread from Zanzibar to other parts of the world (12). In 2010, 2012, 2013 and 2014 dengue outbreaks reported in Dar es Salaam,Tanzania (13). Therefore within that period some ndings revealed that the prevalence of dengue ranged from < 2% to > 50% (14). However dengue outbreak in Tanzania is considered sporadic epidemic most likely imported from elsewhere (11,15). Lack of public awareness on arboviral illnesses and lack of diagnostic capacities in health facilities bring confusion in differentiating causative agent of febrile conditions in Tanzania hospitals. Therefore we aimed to investigate the incidence of dengue and chikungunya infections among febrile children admitted at the Kilimanjaro Christian Medical Centre Hospital in Northern Tanzania.

Study design, area and population
A hospital-based cross sectional study was conducted at pediatric department at Kilimanjaro Christian Medical Centre (KCMC) which is located in Moshi urban district in Kilimanjaro region situated in Northern Tanzania. It serves ve regions in Northern part of Tanzania, namely, Arusha, Tanga, Manyara, Singida and Kilimanjaro. It has a bed capacity of 450. This study included children aged 6 months to 13 years, with history of fever 24 hours prior to admission from September 2015 to May 2016. We excluded Children without history of fever and those whose parents refused to consent on behalf.

Data Collection
Following admission, nger-prick blood samples of approximately 100 µl were collected from children and mixed with 900 µl of L2 buffer (120 g of guanidine thiocyanate in 100 ml of 0.1 M Tris-HCl, pH 6.4) and stored at -80 o C until further use. Informed consent from guardians was used to collect some demographic data such as age, sex, residence and clinical symptoms, provisional diagnosis; previous laboratory test and previous treatment given were recorded.

RNA Extraction
Ribonucleic acid (RNA) was extracted from serum samples using the Boom method (16). Brie y, 30 µl of silica was added and vortexed for 5 seconds followed by shaking for 10 min and centrifuged at 12,000 g for 15 seconds. Supernatant was removed and silica pellets was washed twice with 1 ml of L2 buffer (120 g of guanidine thiocyanate {GuSCN} in 100 ml of 0.1 M Tris-HCl, pH 6.4). Washed again twice with 1 ml of 70% ethanol and once with 1 ml of acetone. Silica pellets was dried at 56 °C for 10 minutes, RNA was eluted from the silica pellets in 50 µl of diethyl pyrocarbonate (DEPEC) treated water and incubated for 10 minutes at 56 °C. Sample was vortexed and centrifuged for 2 minutes at 12,000 g. 35 µl containing puri ed RNA was transferred to an eppendorf tube and stored at − 20 °C prior to use. Prior to cDNA synthesis, the extracted RNA were treated with DNase 1 to remove contamination. cDNA synthesis and Viral DNA detection cDNA was synthesized using Superscript® VILO™ cDNA synthesis kit (Invitrogen, life technologies, USA) according to manufacturer's instructions in a total volume of 20 µL containing 2 µL of 10X Superscript® Enzyme Mix, 4 µL of 5X VILOTM reaction Mix, 11 µL of DEPEC treated water (Ambion, USA) and 3 µL of extracted RNA. The reverse transcription programme involved incubation at 25 °C for 10 minutes, extension at 42 °C for 60 minutes and inactivation at 85 °C for 5 minutes. The resulting cDNA was stored in − 20 °C for further use for dengue and chikungunya polymerase chain reaction (PCR). Dengue PCR was conducted as previously described (17) with respect to dengue only and with minor modi cations (14) using TaqMan Gene expression master mix kit (Applied Biosystens, USA). Chikungunya virus investigation was performed using conventional PCR described by Reddy et al., (18). Primers and probes were purchased from Biolegio BV (Nijmegen, the Netherlands).

Data Analysis
Data were analyzed using IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp, Armonk, NY, USA).
Pearson's chi-square (χ 2) and Fisher's exact test (FE) were used to compare categorical data. Association of chikungunya acute infections and symptoms was analyses by logistic regression. A p-value of less than 0.05 was considered statistically signi cant.

Provision diagnosis and Chikungunya treatment
Due to the fact that laboratory screening for chikungunya was done retrospectively, chikungunya results were not available for clinical care. Majority of chikungunya positive 20 (87.0%) were treated with antibiotics (ceftriaxone) and none was prescribed with antipyretics or ant-pain (paracetamol). Among patients who were given anti-malarial treatment (artesunate), 52.4% were given wrong treatment, Table 5.

Discussion
Chikungunya and dengue fever have growing public health impact around the world and both are common infections in the tropical and sub-tropical countries. However, the cause of fever other than malaria and bacterial infection is not often reported. Current estimates suggest that DENV in sub Saharan Africa carries 16% of the annual world burden (13). The knowledge is still poor regarding arboviral febrile infection, because the clinical guidelines on management of febrile patient in low resource areas focus only on bacterial and malarial infection (19). Moreover there is lack of awareness among clinicians on concurrent of infection due to multiple agents (20). Non speci c clinical presentation, limited diagnostic capacities and weak surveillance system made the burden grow big and unnoticed (13). Chikungunya and Dengue infection become misdiagnosed with other febrile infections such as UTI, pneumonia, meningitis and malaria, it is also di cult to distinguish dengue and chikungunya clinically as well (2).
In this study we detected 11.7% of chikungunya positive cases among hospitalized febrile children. These ndings are similar to a pervious hospital study conducted in northern Tanzania, which reported 7.9% of chikungunya acute infection (21). These results suggest that northern Tanzania is endemic to chikungunya infection. This compel introduction of arboviral testing in the routine laboratory procedures. However absence of dengue infection in our study population and other African settings can possibly be due to lower OSBPL10 expression pro le in Africans which is protective against viral hemorrhagic fever and dengue shock syndrome (22). Transmission occurs mostly during epidemics.
Knowledge and awareness on febrile illness etiologies is important for proper disease management.
Adittionaly it minimizes drugs over use hence suppressing the emergence of drug resistance, to avoid resource westage because the cost of handling known fever is less than that of unknown fever. In Kenya reports reveals that there is high use rate (33.3%) of antmalarial treatments for patient having dengue infection (15), suggesting that there is still high under recognition of dengue infection. Morover dengue disease burden in sub saharan African among children is still hidden due to lack of national survaillance programs (23). A study conducted in Northern Tanzania among healthcare workers reported low level of knowledge regards to dengue and chikungunya that is why the really burden of arbovirus is not clear (24). Among children, 12.0% were provisionally diagnosed as having bacterial infection and 14.3% were provisionally diagnosed as malarial infection, but none of them with viral infection. Our study ndings suggest that chikungunya infection in children is under recognized and the burden is not clear. Incorrect diagnosis leaves the patient vulnerable to worsening of the underlying true cause of fever (25). More than 20% of children who were provisional diagnosed received wrong treatment. Over-treatment has contributed much on the global health challenge, including the emergence of drug resistance, unnecessary adverse drug effects and increased treatment costs (26)(27)(28).
None of children were dengue positive by real time PCR assay. To date there is no epidemiological study that shows the prevalence of dengue in northern part of Tanzania although two cases of dengue were detected at KCMC in 2014 (unpublished work). Similar results were obtained by a study (14) which involved patients presenting with malaria-like symptoms in a community cross-sectional study in northeastern Tanzania.

Conclusion
Chikungunya virus is common among febrile children in northern Tanzania. If symptoms like fever, rashes, vomiting and pain are commonly reported in the context of lack of diagnostic tools, there is strong possibility that chikungunya is unrecognized particularly in endemic settings. This warrants continuous diagnosis of chikungunya among febrile patients in Tanzania at improved healthcare.

Limitations
The number of con rmed bacterial infections was very low; the reason might be due to inadequate diagnostic tools to con rm wide range of bacterial species. In addition 153 (78.1%) children were not tested in the laboratory this might be because they were treated empirically and recovered in the ward. Most of patients were treated empirically due to lack of advanced diagnostic tools that may quickly yield result for fast treatment, especially for those who presented with severe febrile illnesses.

Availability of data and materials
The data included in this article will be provided upon request from the author.

Competing interests
The authors declare that they have no competing interests.

Funding N/A
Authors' contributions ATS: conceived the study, design and performed the study, participated in the eld, contributed to interpretation of results and drafted the manuscript. DCK: contributed to the overall study, interpretation of data and critical review of manuscript; KAA: participated in the eld and revised the manuscript; SM: contributed to overall study design and critical review of the manuscript, FWM: contributed to overall study design and critical review of the manuscript, MA: contributed to overall study design and critical review of the manuscript, GK: contributed to interpretation of results and critical review of the manuscript, RAK: contributed to overall study design, analysis of data and critical review of the manuscript. All authors read and approved the nal manuscript.