Prevalence of malaria, typhoid, toxoplasmosis and rubella among febrile children in Cameroon

Fever is the most common symptom of infectious diseases in tropical countries. A wide range of treatable or preventable pathogens are known to cause fever in patients who present with malaria-like symptoms, but do not have malaria [1]. Malaria continues to be the major cause of mortality in Cameroon and accounts for 35–40 % of all deaths, with 50 % of morbidity among children under the age of five with a considerable drop in morbidity observed between 2008–2011 [2, 3]. Malaria diagnosis in Cameroon has been reported to be predominantly presumptive leading to over diagnosis and over prescription of antimalarials [4]. Consequently, at least 50 % of most febrile patients are considered to have malaria and are given treatment out of fear of missing life-threatening Plasmodium falciparum infections. Serious consideration of other etiologies may not occur unless there is no clinical response to antimalarials treatment.

Many malaria endemic countries including Cameroon have adopted the WHO evidence -based guidelines for effective case management of malaria and rapid diagnostic tests (RDTs) have been presented as a means to realize this key strategy [5]. The recent roll-out /scale up of RDTs for malaria has highlighted the decreasing proportion of malaria-attributable illness in endemic areas [5‐8] and the proportion of patients with non malarial febrile illnesses is likely to continue increasing with further adaptation of malaria RDTs [5‐8]. However, the lack of affordable diagnostic tests and knowledge of the prevalence of other infectious diseases means that febrile patients are not managed optimally. These factors may promote the development of antimalarial and antibiotic resistance and unnecessary morbidity and mortality.

Diagnosis of non-malaria febrile illness in resource limited settings remains challenging [9]. With malaria RDTs, current treatment algorithms leave healthcare providers at odds on how to treat non-malaria febrile cases when RDTs are negative. The importance of differential diagnosis of febrile illnesses has been reported in many malaria endemic countries [10‐21]. Typhoid fever is widely recognized as a major public health problem in most developing tropical countries and share similar symptoms with malaria [16], while toxoplasmosis is a cosmopolitan infection, which appears to be overshadowed in the tropics by other endemic diseases such as malaria and HIV [22].

With the successful contribution of RDTs as a key strategy in evidence-based malaria case management, RDTs, designed for peripheral health facilities have been developed or are being developed for other tropical infections including typhoid, toxoplasmosis, to improve patient and epidemiological surveillance. Most RDTs for non-malarial tropical infections are reported to rely on the detection of host antibodies against a single infectious agent and the sensitivity and specificity of host-antibody detection tests are both inherently limited [23]. However, point-of-care RDTs for typhoid, toxoplasmosis and rubella have been shown as useful tools for rapid diagnosis of these preventable infections in tropical countries [24‐27].

A number of infectious diseases such as typhoid fever, rickettsial and arboviral infections have been recognized among febrile adult patients attending health facilities in Cameroon [19‐21] while toxoplasmosis and rubella infections have mostly been reported amongst pregnant women [28‐30]. However, limited information is available on the prevalence of febrile illnesses in children. Moreover, the proportion of children with fevers has not changed despite the reported drop in malaria morbidity in Cameroon [3]. This study assessed the prevalence of commonly treatable or preventable febrile illnesses in children between 6 months and 15 years using rapid diagnostic tests at the point-of-care.

In this study, we determined the causes of commonly treatable or preventable febrile illnesses in children between 6 months and 15 years using rapid diagnostic tests at the point-of-care. All the 315 children in this study presented with fever and other symptoms that are common to malaria and other febrile illnesses like typhoid, toxoplasmosis and rubella thus confirming the need for a differential diagnosis of febrile illnesses as reported in many other malaria endemic countries [10‐21]. Self- medication in this study population was high (87 %) and medication taken by patients prior to health facility visit were mostly antimalarial (27.3 %), antibiotics (11.4 %) and antifungal (13.3 %). This could act as a potential source of resistance to antimalarials or antibiotics in subsequent treatment. As such, there is need for continuous education of the population on the dangers of self medication, as self- medication can lead to unnecessary use of antimicrobials and life- threatening adverse events as reported in other studies [34, 35].

Increased malaria prevention and control measures are dramatically reducing the malaria burden in many endemic countries [36]. In Cameroon, malaria hospital morbidity was reported to have declined from 40.8 % in 2008 to 27.5 % in 2012 [3]. However, findings from the study described herein showed a high prevalence of 56 %, 43 and 70 % by RDTs, microscopy and PCR respectively. P. falciparum still remains the most prevalent malaria parasite species. This shows that malaria prevalence is still high amongst children in Cameroon and remains the main cause of fevers in this population despite the introduction of current control efforts such as the use of insecticide treated bednets (ITNs). Though RDTs were more sensitive compared to microscopy, and they were significantly less sensitive compared to PCR (Table 4). This confirms previous reports in similar settings that PCR is still the most sensitive test for malaria diagnosis [37]. Though RDTs are currently being used for routine diagnosis of malaria in health facilities, continuous monitoring and evaluation of these tests are necessary to ensure they remain effective for an informed policy decision on malaria control. Incorporation of a cost effective PCR in routine malaria diagnosis could be necessary in the long-term for effective malaria diagnosis. Insecticide treated bednet (ITN) usage has been reported to be protective against malaria in malaria endemic countries including Cameroon [38]. Although ITNs usage in this study was relatively low (44.8 %) compared to ownership (79.7 %), participants who did not sleep under the bed net were more susceptible to malaria infection (Table 5). There is need for the National Malaria Control Programme(NMCP) to intensify the sensitization of the population on the use of this important malaria control tool.

Typhoid and malaria share social circumstances which are important for their transmission, and so individuals in areas endemic for both diseases are at substantial risk of contracting both these diseases, either concurrently or an acute infection superimposed on a chronic one [16, 39, 40]. In this study, typhoid (Salmonella typhi and S. paratyphi) IgM antibodies seroprevalence, as diagnosed by RDTs, was 4.4 % and these patients presented with malaria—like symptoms and were also positive for malaria. These results were similar to the reports of Nsutebu et al., [19] who observed a typhoid prevalence of 2.5 % in a similar setting in Yaoundé but contradicts the work of Ammah et al., [20] who reported a prevalence of 26 %. These findings also support the report of Nsutebu et al., [19] that typhoid fever is not as endemic in Cameroon as recently feared. All the typhoid cases recorded in this study occurred as co-infections with malaria. Risk factors such as the use of borehole or well as the main source of drinkable water was significantly associated with typhoid (p = 0.047). Public health measures such as improved personal hygiene and intensive community health education campaign [40] and the use of public water supply rather than wells could be useful for the reduction of typhoid prevalence in this population.

Toxoplasmosis infection in immunocompetent adults and children are usually asymptomatic or have symptoms that spontaneously resolve such as fever, malaise, and lymphadenopathy indicating a symptomless latent infection [41]. A report from the prevalence of toxoplasmosis in pregnant women in Cameroon showed a seroprevalence of toxoplasma IgG antibodies, IgG and IgM co-infection of 65 and 2.7 % respectively [28]. Febrile children in this study showed a seroprevalence of toxoplasmosis (T.gondii) IgM antibodies of 3.4 % and occurred as co-infection with malaria. Most of the toxoplasmosis infection observed in this study might have been congenital given the high prevalence previously reported in pregnant women [28] in a similar setting and the high prevalence (38.3 %) of past infections (IgG antibodies) observed in this study. These findings therefore, emphasize the need for a differential diagnosis of this preventable and treatable infection in febrile patients. Risk factors such as parents’ educational level and cats kept indoors have been reported to be associated with the acquisition of toxoplasmosis infection amongst primary school children [42]. However, no significant association was found between the raising of cats or dogs and toxoplasmosis IgM antibodies in this population. The lack of a significant association might have been due to a low prevalence of the disease, small sample size or the sensitivity/specificity of the RDTs used.

Rubella is a mild disease in children and adults, but can cause devastating problems if it infects the fetus, especially when the infection occurs during the first weeks of pregnancy [43, 44]. There is no specific treatment for rubella, but the disease is preventable by vaccination. A comprehensive vaccination program in most industrialized regions has reduced the incidence of the disease in these areas to low levels; vaccination is not carried out in many developing countries [45]. However, most children in Cameroon are vaccinated against rubella. The low seroprevalence of rubella IgM antibodies (1.3 %) observed in this study indicates the positive impact of the vaccination programmes. The 4 cases of rubella and malaria co-infection were from the same family whose parents reportedly did not believe in vaccines. Continuous community education may lead to complete elimination of this preventable febrile illness.

This study also has some limitations. First, the methods used in this study were unable to determine the cause of fever in approximately 30 % of the study population that were negative for all the four pathogens investigated. Investigations of other pathogens of viral and bacterial origin and biomarkers could provide additional useful diagnostic clues that may help to refine decision-making. Moreover, the RDTs for typhoid, toxoplasmosis and rubella maybe non-specific hence confirmation with more specific and sensitive tests such as cultures or PCR when available may be useful. Secondly, it is a health-facility-based study and it did not include patients within the community. Also, the influence of other factors such as the economic status of the family, education and the nutritional status of the children were not investigated.

Although the prevalence of typhoid, toxoplasmosis and rubella was low in this population, it will still be important to consider a differential diagnosis of these febrile illnesses so as to reduce overdiagnosis of malaria and overprescription of antimalarials.

Though malaria still remains the major cause of fever in children, using RDTs of other treatable febrile illnesses like typhoid and toxoplasmosis could facilitate the optimal management of febrile illnesses at the point- of- care. Results from RDTs also provide useful information to health-workers on how to manage febrile patients at the point of care either when malaria is negative or exists as a co-infection with other infectious diseases such as typhoid and toxoplasmosis