Search

Background Pathogens causing acute fever, with the exception of malaria, remain largely unidentified in sub-Saharan Africa, given the local unavailability of diagnostic tests and the broad differential diagnosis. Methodology/Principal Findings We conducted a cross-sectional study including outpatient acute febrile syndromes in both children and adults, between November 2015 and June 2016 in Kinshasa, Democratic Republic of Congo. Serological and molecular diagnostic tests for arboviral infections were performed on blood, including PCR and NS1-RDT for acute, and ELISA and IFAT for past infections. Conclusions/Significance Among 342 patients, aged 2 to 68 years, 45.3% tested positive on malaria Rapid Diagnostic Test. However, 87.7% received antimalarial and 64.3% antibacterial treatment. Further investigation among 235 fever cases revealed 19 (8.1%) acute dengue and 2 (0.9%) acute chikungunya infections, with an important proportion of participants already exposed to flaviviridae (possibly dengue) and alphaviridae (possibly chikungunya) in the past, namely 30.2 % and 26.4% respectively. We found no evidence of exposure to Zika nor yellow fever virus.

Background Pathogens causing acute fever, with the exception of malaria, remain largely unidentified in sub-Saharan Africa, given the local unavailability of diagnostic tests and the broad differential diagnosis. Methodology We conducted a cross-sectional study including outpatient acute undifferentiated fever in both children and adults, between November 2015 and June 2016 in Kinshasa, Democratic Republic of Congo. Serological and molecular diagnostic tests for selected arboviral infections were performed on blood, including PCR, NS1-RDT, ELISA and IFA for acute, and ELISA and IFA for past infections. Results Investigation among 342 patients, aged 2 to 68 years (mean age of 21 years), with acute undifferentiated fever (having no clear focus of infection) revealed 19 (8.1%) acute dengue–caused by DENV-1 and/or DENV-2 –and 2 (0.9%) acute chikungunya infections. Furthermore, 30.2% and 26.4% of participants had been infected in the past with dengue and chikungunya, respectively. We found no evidence of acute Zika nor yellow fever virus infections. 45.3% of patients tested positive on malaria Rapid Diagnostic Test, 87.7% received antimalarial treatment and 64.3% received antibacterial treatment. Discussion Chikungunya outbreaks have been reported in the study area in the past, so the high seroprevalence is not surprising. However, scarce evidence exists on dengue transmission in Kinshasa and based on our data, circulation is more important than previously reported. Furthermore, our study shows that the prescription of antibiotics, both antibacterial and antimalarial drugs, is rampant. Studies like this one, elucidating the causes of acute fever, may lead to a more considerate and rigorous use of antibiotics. This will not only stem the ever-increasing problem of antimicrobial resistance, but will–ultimately and hopefully–improve the clinical care of outpatients in low-resource settings. Trial registration ClinicalTrials.gov NCT02656862., Author summary Malaria remains one of the most important causes of fever in sub-Saharan Africa. However, its share is declining, since the diagnosis and treatment of malaria have improved significantly over the years. Hence leading to an increase in the number of patients presenting with non-malarial fever. Often, obvious clinical signs and symptoms like cough or diarrhea are absent, probing the question: “What causes the fever?” Previous studies have shown that the burden of arboviral infections–like dengue and chikungunya–in sub-Saharan Africa is underestimated, which is why we screened for four common arboviral infections in patients presenting with ‘undifferentiated fever’ at an outpatient clinic in suburban Kinshasa, Democratic Republic of Congo. Among the patients tested, we found that one in ten presented with an acute arboviral infection and that almost one in three patients had been infected in the past. These findings suggest that clinicians should think about arboviral infections more often, thereby refraining from the prescription of antibiotics, a practice increasingly problematic given the global rise of antimicrobial resistance.

Newborn screening programs for severe metabolic disorders using tandem mass spectrometry are widely used. Medium-Chain Acyl-CoA dehydrogenase deficiency (MCADD) is the most prevalent mitochondrial fatty acid oxidation defect (1:15,000 newborns) and it has been proven that early detection of this metabolic disease decreases mortality and improves the outcome. In previous studies, data mining methods on derivatized tandem MS datasets have shown high classification accuracies. However, no machine learning methods currently have been applied to datasets based on non-derivatized screening methods. A dataset with 44,159 blood samples was collected using a non-derivatized screening method as part of a systematic newborn screening by the PCMA screening center (Belgium). Twelve MCADD cases were present in this partially MCADD-enriched dataset. We extended three data mining methods, namely C4.5 decision trees, logistic regression and ridge logistic regression, with a parameter and threshold optimization method and evaluated their applicability as a diagnostic support tool. Within a stratified cross-validation setting, a grid search was performed for each model for a wide range of model parameters, included variables and classification thresholds. The best performing model used ridge logistic regression and achieved a sensitivity of 100%, a specificity of 99.987% and a positive predictive value of 32% (recalibrated for a real population), obtained in a stratified cross-validation setting. These results were further validated on an independent test set. Using a method that combines ridge logistic regression with variable selection and threshold optimization, a significantly improved performance was achieved compared to the current state-of-the-art for derivatized data, while retaining more interpretability and requiring less variables. The results indicate the potential value of data mining methods as a diagnostic support tool.