Accuracy of dengue clinical diagnosis with and without NS1 antigen rapid test: Comparison between human and Bayesian network model decision

Integration of population-level data sources into an individual-level clinical prediction model for dengue virus test positivity

The differentiation of dengue virus (DENV) infection, a major cause of acute febrile illness in tropical regions, from other etiologies, may help prioritize laboratory testing and limit the inappropriate use of antibiotics. While traditional clinical prediction models focus on individual patient-level parameters, we hypothesize that for infectious diseases, population-level data sources may improve predictive ability. To create a clinical prediction model that integrates patient-extrinsic data for identifying DENV among febrile patients presenting to a hospital in Thailand, we fit random forest classifiers combining clinical data with climate and population-level epidemiologic data. In cross-validation, compared to a parsimonious model with the top clinical predictors, a model with the addition of climate data, reconstructed susceptibility estimates, force of infection estimates, and a recent case clustering metric significantly improved model performance.

Cross-reactivity of SARS-CoV-2 with other pathogens, especially dengue virus: A historical perspective

Dengue is a vector-borne viral disease caused by a Flavivirus whereas the COVID-19 pandemic was caused by a highly contagious virus, SARS-CoV-2 belonging to the family Coronaviridae. However, COVID-19 severity was observably less in dengue-endemic countries and vice versa especially during the active years of the pandemic (2019-2021). We observed that dengue virus (DENV) antibodies (Abs) could cross-react with SARS-CoV-2 spike antigen. This resulted in SARS-CoV-2 false positivity by rapid Ab test kits. DENV Abs binding to SARS-CoV-2 receptor-binding domain (and the reverse scenario), as revealed by docking studies further validated DENV and SARS-CoV-2 cross-reactivity. Finally, SARS-CoV-2 Abs were found to cross-neutralize DENV1 and DENV2 in virus neutralization test (VNT). Abs to other pathogens like Plasmodium were also cross-reactive but non-neutralizing for SARS-CoV-2. Here, we analyze the existing data on SARS-CoV-2 cross-reactivity with other pathogens, especially dengue to assess its impact on health (cross-protection?) and differential sero-diagnosis/surveillance.

Development and validation of a clinical rule for the diagnosis of chikungunya fever in a dengue-endemic area

Rio de Janeiro is a dengue-endemic city that experienced Zika and chikungunya epidemics between 2015 and 2019. Differential diagnosis is crucial for indicating adequate treatment and assessing prognosis and risk of death. This study aims to derive and validate a clinical rule for diagnosing chikungunya based on 3,214 suspected cases consecutively treated at primary and secondary health units of the sentinel surveillance system (up to 7 days from onset of symptoms) in Rio de Janeiro, Brazil. Of the total sample, 624 were chikungunya, 88 Zika, 51 dengue, and 2,451 were negative for all these arboviruses according to real-time polymerase chain reaction (RT-qPCR). The derived rule included fever (1 point), exanthema (1 point), myalgia (2 points), arthralgia or arthritis (2 points), and joint edema (2 points), providing an AUC (area under the receiver operator curve) = 0.695 (95% CI: 0.662-0.725). Scores of 4 points or more (validation sample) showed 74.3% sensitivity (69.0% - 79.2%) and 51.5% specificity (48.8% - 54.3%). Adding more symptoms improved the specificity at the expense of a lower sensitivity compared to definitions proposed by government agencies based on fever alone (European Center for Disease Control) or in combination with arthralgia (World Health Organization) or arthritis (Pan American Health Organization, Brazilian Ministry of Health). The proposed clinical rule offers a rapid, low-cost, easy-to-apply strategy to differentiate chikungunya fever from other arbovirus infections during epidemics.

Typhoid Fever among Patients Diagnosed with Dengue in a Tertiary Care Centre: A Descriptive Cross-sectional Study

Introduction

Dengue and typhoid fever are different entities with overlapping signs and symptoms which are indistinguishable and there have been few reports of co-infections from endemic areas. The resemblance of symptoms makes accurate clinical diagnosis and treatment difficult. Both are major health problems mainly during monsoon and co-infection, if not timely diagnosed and treated can be fatal. The aim of this study was to find out the prevalence of typhoid fever among patients diagnosed with dengue at a tertiary care centre.

Methods

A descriptive cross-sectional study was done among patients of age >15 years with dengue fever attending the medicine outpatient department in a tertiary care centre from 1 July 2021 to 30 June 2022. Ethical approval was taken from the Institutional Review Committee (Reference number: 466/2020). Convenience sampling was used. Patients with other risk factors for febrile illness were excluded from the study. Point estimate and 90% Confidence Interval were calculated.

Results

Among 95 dengue cases, typhoid fever was observed in 18 (18.95%) (12.36-25.54, 90% Confidence Interval). The mean age of presentation was 35±9 years with a male to female ratio of 0.8:1. Fever was the most common presentation with a mean temperature of 100.8±2.1°F.

Conclusions

The prevalence of typhoid fever among dengue-positive cases was higher as compared to other studies done in similar settings.

Keywords

dengue; fever; typhoid fever.

Mysterious dengue-like virus: A case report from Pakistan

Recently, in Pakistan, several cases of mysterious dengue-like illness are being reported, which has concerned the authorities and requires prompt action. We present a case of a 52-year-old female patient presenting with a history of continuous fever, documented up to 104 F, for 5 days. The symptoms were associated with headache, nausea, retro-orbital headache, arthralgia, and myalgia. Currently, to the best of our knowledge, this is the first reported case in the literature for the endemic mysterious virus and may serve as the groundwork for future studies.

Utilization of Artificial Intelligence in Disease Prevention: Diagnosis, Treatment, and Implications for the Healthcare Workforce

Artificial intelligence (AI) has been described as one of the extremely effective and promising scientific tools available to mankind. AI and its associated innovations are becoming more popular in industry and culture, and they are starting to show up in healthcare. Numerous facets of healthcare, as well as regulatory procedures within providers, payers, and pharmaceutical companies, may be transformed by these innovations. As a result, the purpose of this review is to identify the potential machine learning applications in the field of infectious diseases and the general healthcare system. The literature on this topic was extracted from various databases, such as Google, Google Scholar, Pubmed, Scopus, and Web of Science. The articles having important information were selected for this review. The most challenging task for AI in such healthcare sectors is to sustain its adoption in daily clinical practice, regardless of whether the programs are scalable enough to be useful. Based on the summarized data, it has been concluded that AI can assist healthcare staff in expanding their knowledge, allowing them to spend more time providing direct patient care and reducing weariness. Overall, we might conclude that the future of "conventional medicine" is closer than we realize, with patients seeing a computer first and subsequently a doctor.

Machine learning for emerging infectious disease field responses

Emerging infectious diseases (EIDs), including the latest COVID-19 pandemic, have emerged and raised global public health crises in recent decades. Without existing protective immunity, an EID may spread rapidly and cause mass casualties in a very short time. Therefore, it is imperative to identify cases with risk of disease progression for the optimized allocation of medical resources in case medical facilities are overwhelmed with a flood of patients. This study has aimed to cope with this challenge from the aspect of preventive medicine by exploiting machine learning technologies. The study has been based on 83,227 hospital admissions with influenza-like illness and we analysed the risk effects of 19 comorbidities along with age and gender for severe illness or mortality risk. The experimental results revealed that the decision rules derived from the machine learning based prediction models can provide valuable guidelines for the healthcare policy makers to develop an effective vaccination strategy. Furthermore, in case the healthcare facilities are overwhelmed by patients with EID, which frequently occurred in the recent COVID-19 pandemic, the frontline physicians can incorporate the proposed prediction models to triage patients suffering minor symptoms without laboratory tests, which may become scarce during an EID disaster. In conclusion, our study has demonstrated an effective approach to exploit machine learning technologies to cope with the challenges faced during the outbreak of an EID.

Zika Virus Non-Structural Protein 1 Antigen-Capture Immunoassay

Infection with Zika virus (ZIKV), a member of the Flavivirus genus of the Flaviviridae family, typically results in mild self-limited illness, but severe neurological disease occurs in a limited subset of patients. In contrast, serious outcomes commonly occur in pregnancy that affect the developing fetus, including microcephaly and other major birth defects. The genetic similarity of ZIKV to other widespread flaviviruses, such as dengue virus (DENV), presents a challenge to the development of specific ZIKV diagnostic assays. Nonstructural protein 1 (NS1) is established for use in immunodiagnostic assays for flaviviruses. To address the cross-reactivity of ZIKV NS1 with proteins from other flaviviruses we used site-directed mutagenesis to modify putative epitopes. Goat polyclonal antibodies to variant ZIKV NS1 were affinity-purified to remove antibodies binding to the closely related NS1 protein of DENV. An antigen-capture ELISA configured with the affinity-purified polyclonal antibody showed a linear dynamic range between approximately 500 and 30 ng/mL, with a limit of detection of between 1.95 and 7.8 ng/mL. NS1 proteins from DENV, yellow fever virus, St. Louis encephalitis virus and West Nile virus showed significantly reduced reactivity in the ZIKV antigen-capture ELISA. Refinement of approaches similar to those employed here could lead to development of ZIKV-specific immunoassays suitable for use in areas where infections with related flaviviruses are common.

Laboratory Findings in Patients with Probable Dengue Diagnosis from an Endemic Area in Colombia in 2018

As demonstrated with the novel coronavirus pandemic, rapid and accurate diagnosis is key to determine the clinical characteristic of a disease and to improve vaccine development. Once the infected person is identified, hematological findings may be used to predict disease outcome and offer the correct treatment. Rapid and accurate diagnosis and clinical parameters are pivotal to track infections during clinical trials and set protection status. This is also applicable for re-emerging diseases like dengue fever, which causes outbreaks in Asia and Latin America every 4 to 5 years. Some areas in the US are also endemic for the transmission of dengue virus (DENV), the causal agent of dengue fever. However, significant number of DENV infections in rural areas are diagnosed solely by clinical and hematological findings because of the lack of availability of ELISA or PCR-based tests or the infrastructure to implement them in the near future. Rapid diagnostic tests (RDT) are a less sensitive, yet they represent a timely way of detecting DENV infections. The purpose of this study was to determine whether there is an association between hematological findings and the probability for an NS1-based DENV RDT to detect the DENV NS1 antigen. We also aimed to describe the hematological parameters that are associated with the diagnosis through each test.

The role of clinical signs and spirometry in the diagnosis of obstructive airway diseases: a systematic analysis adapted to general practice settings

Background

In general practice (GP), the diagnosis of obstructive airway diseases much relies on diagnostic questions, in view of the limited availability of lung function. We systematically assessed the relative importance of such questions for diagnosing asthma and chronic obstructive pulmonary disease (COPD), either without or with information from spirometry.

Methods

We used data obtained in a pulmonary practice to ensure the validity of diagnoses and assessments. Subjects with a diagnosis of COPD (n=260), or asthma (n=433), or other respiratory diseases (n=230), and subjects without respiratory diseases (n=364, controls) were included. The diagnostic questions comprised eight items, covering smoking history, self-attributed allergic rhinitis, dyspnea, cough, phlegm and wheeze. Optionally standard parameters of the flow-volume-curve were included. Decision trees for the diagnosis of COPD and asthma were constructed, moreover a probabilistic diagnostic network based on the results of path analyses describing the relationship between variables.

Results

In the decision trees, age, sex, current smoking, wheezing, dyspnea upon mild exertion, self-attributed allergic rhinitis, phlegm, forced expiratory volume in one second (FEV₁), and expiratory flow rates were relevant, depending on the diagnostic comparison, while cough, dyspnea upon strong exertion and ex-smoker status were not relevant. In contrast, the probabilistic network for the diagnosis of COPD and asthma versus controls incorporated all diagnostic questions, i.e., dyspnea upon mild or strong exertion, current smoking, ex-smoking, wheezing, cough and phlegm but from spirometry only FEV₁. Depending on the individual pattern, the probability for COPD could raise from 25% to 81%, while the diagnostic gain for asthma was lower.

Conclusions

The study developed simple diagnostic algorithms for asthma and COPD that take into account the relative importance of clinical signs and history, as well as spirometric data if available. The diagnostic accuracy was especially high for COPD. These algorithms may be helpful as a starting point in the standardisation of diagnostic strategies in GP practices.

Trial registration

The study is registered under DRKS00013935 at German Clinical Trials Register (DRKS, Date of registration 01/03/2018).

A comprehensive scoping review of Bayesian networks in healthcare: Past, present and future

No comprehensive review of Bayesian networks (BNs) in healthcare has been published in the past, making it difficult to organize the research contributions in the present and identify challenges and neglected areas that need to be addressed in the future. This unique and novel scoping review of BNs in healthcare provides an analytical framework for comprehensively characterizing the domain and its current state. A literature search of health and health informatics literature databases using relevant keywords found 3810 articles that were reduced to 123. This was after screening out those presenting Bayesian statistics, meta-analysis or neural networks, as opposed to BNs and those describing the predictive performance of multiple machine learning algorithms, of which BNs were simply one type. Using the novel analytical framework, we show that: (1) BNs in healthcare are not used to their full potential; (2) a generic BN development process is lacking; (3) limitations exist in the way BNs in healthcare are presented in the literature, which impacts understanding, consensus towards systematic methodologies, practice and adoption; and (4) a gap exists between having an accurate BN and a useful BN that impacts clinical practice. This review highlights several neglected issues, such as restricted aims of BNs, ad hoc BN development methods, and the lack of BN adoption in practice and reveals to researchers and clinicians the need to address these problems. To map the way forward, the paper proposes future research directions and makes recommendations regarding BN development methods and adoption in practice.

Bayesian networks in healthcare: What is preventing their adoption?

There has been much research effort expended toward the use of Bayesian networks (BNs) in medical decision-making. However, because of the gap between developing an accurate BN and demonstrating its clinical usefulness, this has not resulted in any widespread BN adoption in clinical practice. This paper investigates this problem with the aim of finding an explanation and ways to address the problem through a comprehensive literature review of articles describing BNs in healthcare. Based on the literature collection that has been systematically narrowed down from 3810 to 116 most relevant articles, this paper analyses the benefits, barriers and facilitating factors (BBF) for implementing BN-based systems in healthcare using the ITPOSMO-BBF framework. A key finding is that works in the literature rarely consider barriers and even when these were identified they were not connected to facilitating factors. The main finding is that the barriers can be grouped into: (1) data inadequacies; (2) clinicians' resistance to new technologies; (3) lack of clinical credibility; (4) failure to demonstrate clinical impact; (5) absence of an acceptable predictive performance; and (6) absence of evidence for model's generalisability. The facilitating factors can be grouped into: (1) data collection improvements; (2) software and technological improvements; (3) having interpretable and easy to use BN-based systems; (4) clinical involvement in the development or review of the model; (5) investigation of model's clinical impact; (6) internal validation of the model's performance; and (7) external validation of the model. These groupings form a strong basis for a generic framework that could be used for formulating strategies for ensuring BN-based clinical decision-support system adoption in frontline care settings. The output of this review is expected to enhance the dialogue among researchers by providing a deeper understanding for the neglected issue of BN adoption in practice and promoting efforts for implementing BN-based systems.

Surveillance and Epidemiology of Dengue in Medellín, Colombia from 2009 to 2017

Dengue is the most prevalent arthropod-borne viral disease in humans, primarily transmitted by the Aedes aegypti mosquito. We conducted a descriptive analysis of dengue cases from 2009 to 2017 in Medellín, Colombia, using data available from the Secretariat of Health. We analyzed the burden of outbreak years on the healthcare system, risk of cases exhibiting severe illness, potential disease surveillance problems, gender and age as risk factors, and spatiotemporal patterns of disease occurrence. Our data consisted of 50,083 cases, separated based on whether they were diagnostic test negative, diagnostic test positive (primarily IgM ELISA), clinically confirmed, epidemiologically linked, or probable. We used dengue incidence to analyze epidemiological trends between our study years, related to human movement patterns, between gender and age-groups, and spatiotemporally. We used risk to analyze the severity of dengue cases between the study years. We identified human movement could contributed to dengue spread, and male individuals (incidence rate: 0.86; 95% CI: 0.76-0.96) and individuals younger than 15 years (incidence rate: 1.24; 95% CI: 1.13-1.34) have higher incidence of dengue and located critical parts of the city where dengue incidence was high. Analysis was limited by participant diagnostic information, data concerning circulating strains, and a lack of phylogenetic information. Understanding the characteristics of dengue is a fundamental part of improving the health outcomes of at-risk populations. This analysis will be useful to support studies and initiatives to counteract dengue and provide context to the surveillance data collected by the health authorities in Medellín.

Comparing machine learning with case-control models to identify confirmed dengue cases

In recent decades, the global incidence of dengue has increased. Affected countries have responded with more effective surveillance strategies to detect outbreaks early, monitor the trends, and implement prevention and control measures. We have applied newly developed machine learning approaches to identify laboratory-confirmed dengue cases from 4,894 emergency department patients with dengue-like illness (DLI) who received laboratory tests. Among them, 60.11% (2942 cases) were confirmed to have dengue. Using just four input variables [age, body temperature, white blood cells counts (WBCs) and platelets], not only the state-of-the-art deep neural network (DNN) prediction models but also the conventional decision tree (DT) and logistic regression (LR) models delivered performances with receiver operating characteristic (ROC) curves areas under curves (AUCs) of the ranging from 83.75% to 85.87% [for DT, DNN and LR: 84.60% ± 0.03%, 85.87% ± 0.54%, 83.75% ± 0.17%, respectively]. Subgroup analyses found all the models were very sensitive particularly in the pre-epidemic period. Pre-peak sensitivities (<35 weeks) were 92.6%, 92.9%, and 93.1% in DT, DNN, and LR respectively. Adjusted odds ratios examined with LR for low WBCs [≤ 3.2 (x103/μL)], fever (≥38°C), low platelet counts [< 100 (x103/μL)], and elderly (≥ 65 years) were 5.17 [95% confidence interval (CI): 3.96-6.76], 3.17 [95%CI: 2.74-3.66], 3.10 [95%CI: 2.44-3.94], and 1.77 [95%CI: 1.50-2.10], respectively. Our prediction models can readily be used in resource-poor countries where viral/serologic tests are inconvenient and can also be applied for real-time syndromic surveillance to monitor trends of dengue cases and even be integrated with mosquito/environment surveillance for early warning and immediate prevention/control measures. In other words, a local community hospital/clinic with an instrument of complete blood counts (including platelets) can provide a sentinel screening during outbreaks. In conclusion, the machine learning approach can facilitate medical and public health efforts to minimize the health threat of dengue epidemics. However, laboratory confirmation remains the primary goal of surveillance and outbreak investigation.

Diagnosis of COVID-19 in a Dengue-Endemic Area

Emergence of SARS-CoV-2 in dengue virus (DENV)-endemic areas complicates the diagnosis of both infections. COVID-19 cases may be misdiagnosed as dengue, particularly when relying on DENV IgM, which can remain positive months after infection. To estimate the extent of this problem, we evaluated sera from 42 confirmed COVID-19 patients for evidence of DENV infection. No cases of SARS-CoV-2 and DENV coinfection were identified. However, recent DENV infection, indicated by the presence of DENV IgM and/or high level of IgG antibodies, was found in seven patients. Dengue virus IgM and/or high IgG titer should not exclude COVID-19. SARS-CoV-2 reverse transcription polymerase chain reaction (RT-PCR) testing is appropriate when dengue nonstructural protein 1 (NS1) or RT-PCR is negative. Given the possibility of coinfection, testing for both DENV and SARS-CoV-2 is merited in the setting of the current pandemic.

Development and Performance of Dengue Diagnostic Clinical Algorithms in Colombia

Diagnosing dengue in endemic areas remains problematic because of the low specificity of the symptoms and lack of accurate diagnostic tests. This study aimed to develop and prospectively validate, under routine care, dengue diagnostic clinical algorithms. The study was carried out in two phases. First, diagnostic algorithms were developed using a database of 1,130 dengue and 918 non-dengue patients, expert opinion, and literature review. Algorithms with > 70% sensitivity were prospectively validated in a single-group quasi-experimental trial with an adaptive Bayesian design. In the first phase, the algorithms that were developed with the continuous Bayes formula and included leukocytes and platelet counts, in addition to selected signs and symptoms, showed the highest sensitivities (> 80%). In the second phase, the algorithms were applied on admission to 1,039 consecutive febrile subjects in three endemic areas in Colombia of whom 25 were laboratory-confirmed dengue, 307 non-dengue, 514 probable dengue, and 193 undetermined. Including parameters of the hemogram consistently improved specificity without affecting sensitivity. In the final analysis, considering only confirmed dengue and non-dengue cases, an algorithm with a sensitivity and specificity of 65.4% (95% credibility interval 50-83) and 40.1% (34.7-45.7) was identified. All tested algorithms had likelihood ratios close to 1, and hence, they are not useful to confirm or rule out dengue in endemic areas. The findings support the use of hemograms to aid dengue diagnosis and highlight the challenges of clinical diagnosis of dengue.

Evaluating the validity of dengue clinical-epidemiological criteria for diagnosis in patients residing in a Brazilian endemic area

BACKGROUND

We evaluated the validity of clinical diagnosis compared with laboratory diagnosis of dengue in a retrospective sample of patients in São José do Rio Preto, Brazil.

METHODS

Our sample included 148 299 clinically (56.3%) or laboratory-diagnosed (43.7%) dengue cases. We compared the sensitivity, specificity, positive and negative predictive value (PPV and NPV) of dengue patients' demographic and clinical characteristics with laboratory-based diagnosis. We used logistic regressions to estimate the correlation between clinical and laboratory diagnosis of dengue and a full set of dengue signs and symptoms.

RESULTS

We found substantial variability in sensitivity and specificity of signs and symptoms ranging from 0.8-81.1 and 21.5-99.6, respectively. Thrombocytopenia exhibited the highest PPV (92.0) and lowest NPV (42.2) and was the only symptom showing agreement with laboratory-confirmed dengue (φ = 0.38). The presence of exanthema and thrombocytopenia led to a greater likelihood of concordant clinical and laboratory diagnoses (exanthema: OR: 4.23; 95% CI: 2.09 to 8.57; thrombocytopenia: OR: 4.02; 95% CI: 1.32 to 12.27).

CONCLUSIONS

We found substantial variation in sensitivity, specificity, PPV and NPV of dengue signs and symptoms. For accuracy, clinical and laboratory diagnosis of dengue should be performed concurrently. When laboratory tests are not available, we suggest focusing on the clinical manifestations most associated with dengue.

MCP-1 LEVELS AND ATYPICAL LYMPHOCYTES IN EARLY FEVER OF DENGUE VIRUS INFECTION WITH NON-STRUCTURAL PROTEIN 1 (NS-1) ANTIGEN TEST IN dr DARSONO HOSPITAL, PACITAN

Dengue infection caused by DENV and transmitted by mosquitoes Aedes aegypti and Aedes albopictus is a major health problem in the world, including Indonesia. Clinical manifestations of dengue infection are very widely, from asymptomatic until dengue shock syndrome (DSS). DENV will attack macrophages and dendritic cells (DC) and replicate them. Monocytes are macrophages in the blood (±10% leukocytes). Macrophages produce cytokines and chemokines such as monocyte chemotactic protein-1 (MCP-1)/CCL2. The monocytes that are infected with DENV will express MCP-1, which will increase the permeability of vascular endothelial cells so that they have a risk of developing DHF/DSS. Macrophages and DC secrete NS1 proteins, which are the co-factors that are needed for viral replication and can be detected in the early phase of fever. The increased MCP-1 levels in dengue infection followed by an increase in the number of atypical lymphocytes indicate the arrival of macrophages and monocytes to the site of inflammation which triggers proliferation rather than lymphocytes. This is an observational analytical study with a cross-sectional design to determine the MCP-1 level in dengue infection patients with 1st until the 4th day of fever and the presence of atypical lymphocytes. Dengue infection was determined by rapid tests NS1 positive or negative and MCP-1 levels were measured using by ELISA sandwich method.MCP-1 level of sixty patients dengue infection NS-1 rapid positive or negative with 2nd until 4rt fever were significantly higher than healthy subjects (420.263±158,496vs29, 475±23.443;p=0.000), but there was no significant difference in subjects with DF, DHF or DSS (436,47±225,59 vs422,77±170,55vs 448,50±117,39; p =0.844). Atypicallymphosite differs significantly in healthy  subjects than subjects infected with DENV an average of 2% (p= 0,000). In conclusion, this shows the arrival of macrophages and monocytes to the site of inflammation, which triggers the proliferation of lymphocytes.

Pitfalls in acute febrile illness diagnosis: Interobserver agreement of signs and symptoms during a dengue outbreak

AIMS AND OBJECTIVES

To compare and evaluate interobserver (nurses and physicians) agreement for dengue clinical signs and symptoms, including the World Health Organization diagnostic algorithm.

BACKGROUND

Agreement of clinical history defines the capacity of the examiner to measure a given clinical parameter in a reproducible and consistent manner, which is prerequisite for diagnosis validity. Nurses play a major role in the triage and care of dengue patients in many countries.

STUDY DESIGN

This is a sub-study on interobserver agreement performed as part of a cross-sectional diagnostic accuracy study for acute febrile illness (AFI) using the checklist STARD.

METHODS

A previously validated semi-structured sign and symptom standardised questionnaire for AFI was independently administered to 374 patients by physician and nurse pairs. The interobserver agreement was estimated using kappa statistics.

RESULTS

For a set of 27 signs and symptoms, we found six interobserver discrepancies (examiner detected red eyes, lethargy, exanthema, dyspnoea, bleeding and myalgia) as identified by regular and moderate kappa indexes. Four signs (patient observed red eyes, cough, diarrhoea and vomiting) and one symptom (earache) had near-perfect agreement. Most signs and symptoms showed substantial agreement. The WHO (Dengue guidelines for diagnosis, treatment, prevention and control: new edition, World Health Organization, 2009) clinical criteria for dengue comprise a group of symptoms known as "pains and aches." Interobserver agreement for abdominal pain, retro-orbital pain and arthralgia exceed that found for headache and myalgia.

CONCLUSIONS

During a dengue outbreak, the interobserver agreement for most of the signs and symptoms used to assess AFI was substantial.

RELEVANCE TO CLINICAL PRACTICE

This result suggests good potential applicability of the tool by health professionals following training. A well-trained health professional is qualified to apply the standardised questionnaire to evaluate suspected dengue cases during outbreaks.

Severity Index for Suspected Arbovirus (SISA): Machine learning for accurate prediction of hospitalization in subjects suspected of arboviral infection

BACKGROUND

Dengue, chikungunya, and Zika are arboviruses of major global health concern. Decisions regarding the clinical management of suspected arboviral infection are challenging in resource-limited settings, particularly when deciding on patient hospitalization. The objective of this study was to determine if hospitalization of individuals with suspected arboviral infections could be predicted using subject intake data.

METHODOLOGY/PRINCIPAL FINDINGS

Two prediction models were developed using data from a surveillance study in Machala, a city in southern coastal Ecuador with a high burden of arboviral infections. Data were obtained from subjects who presented at sentinel medical centers with suspected arboviral infection (November 2013 to September 2017). The first prediction model-called the Severity Index for Suspected Arbovirus (SISA)-used only demographic and symptom data. The second prediction model-called the Severity Index for Suspected Arbovirus with Laboratory (SISAL)-incorporated laboratory data. These models were selected by comparing the prediction ability of seven machine learning algorithms; the area under the receiver operating characteristic curve from the prediction of a test dataset was used to select the final algorithm for each model. After eliminating those with missing data, the SISA dataset had 534 subjects, and the SISAL dataset had 98 subjects. For SISA, the best prediction algorithm was the generalized boosting model, with an AUC of 0.91. For SISAL, the best prediction algorithm was the elastic net with an AUC of 0.94. A sensitivity analysis revealed that SISA and SISAL are not directly comparable to one another.

CONCLUSIONS/SIGNIFICANCE

Both SISA and SISAL were able to predict arbovirus hospitalization with a high degree of accuracy in our dataset. These algorithms will need to be tested and validated on new data from future patients. Machine learning is a powerful prediction tool and provides an excellent option for new management tools and clinical assessment of arboviral infection.

Etiologies of Acute Undifferentiated Febrile Illness in Bangkok, Thailand

Acute undifferentiated febrile illness (AUFI) has been a diagnostic dilemma in the tropics. Without accurate point-of-care tests, information on local pathogens and clinical parameters is essential for presumptive diagnosis. A prospective hospital-based study was conducted at the Bangkok Hospital for Tropical Diseases from 2013 to 2015 to determine common etiologies of AUFI. A total of 397 adult AUFI cases, excluding malaria by blood smear, were enrolled. Rapid diagnostic tests for tropical infections were performed on admission, and acute and convalescent samples were tested to confirm the diagnosis. Etiologies could be identified in 271 (68.3%) cases. Dengue was the most common cause, with 157 cases (39.6%), followed by murine typhus (20 cases; 5.0%), leptospirosis (16 cases; 4.0%), influenza (14 cases; 3.5%), and bacteremia (six cases; 1.5%). Concurrent infection by at least two pathogens was reported in 37 cases (9.3%). Furthermore, characteristics of dengue and bacterial infections (including leptospirosis and rickettsioses) were compared to facilitate dengue triage, initiate early antibiotic treatment, and minimize unnecessary use of antibiotics. In conclusion, dengue was the most common pathogen for AUFI in urban Thailand. However, murine typhus and leptospirosis were not uncommon. Empirical antibiotic treatment using doxycycline or azithromycin might be more appropriate, but cost-benefit studies are required. Physicians should recognize common causes of AUFI in their localities and use clinical and laboratory clues for provisional diagnosis to provide appropriate treatment while awaiting laboratory confirmation.

A fully automated sample-to-answer PCR system for easy and sensitive detection of dengue virus in human serum and mosquitos

BACKGROUND

The insulated isothermal PCR (iiPCR) technology enables consistent PCR amplification and detection in a simple heating device. A pan-dengue virus (DENV) RT-iiPCR, targeting the 5' untranslated region, was validated previously on the semi-automated POCKIT combo system (involving separate devices for nucleic acid extraction and PCR amplification/detection) to offer performance comparable to a laboratory real-time PCR. Working on the same technologies, a compact automated sample-in-answer-out system (POCKIT Central Nucleic Acid Analyser) has been available commercially for iiPCR, minimizing human error risks and allowing easy molecular bio-detection near points of need. Here, we evaluated the analytical and clinical performance of the pan-DENV RT-iiPCR on the fully automated system by comparison to those on the semi-automated system.

METHODOLOGY/PRINCIPAL FINDINGS

Testing sera containing serial diluted DENV-1, -2, -3, or -4 cell culture stock, the pan-DENV RT-iiPCR system had similar 100% detection endpoints on the two systems; i.e. at 1, 10, 1 and 10 PFU/ml, respectively, on the fully automated system, and at 10, 1, 10 and 10 PFU/ml, respectively, on the semi-automated system. Furthermore, both fully automated and semi-automated PCR system can detect all four DENV serotypes in mosquitos. Clinical performance of the reagent on the two systems was evaluated by testing 60 human serum samples. Both systems detected the same 40 samples (ten DENV-1, -2, -3, and -4 positive each) and did not detect the other 20; 100% agreement (κ = 1) was found between the two systems.

CONCLUSIONS/SIGNIFICANCE

With performance comparable to a previously validated system, the fully-automated PCR system allows applications of the pan-DENV reagent as a useful tool near points of need to facilitate easy, fast and effective detection of dengue virus and help mitigate versatile public health challenges in the control and management of dengue disease.

Minor groove binder modification of widely used TaqMan hydrolysis probe for detection of dengue virus reduces risk of false-negative real-time PCR results for serotype 4

Dengue is a vector-transmitted viral infection that is a significant cause of morbidity and mortality in humans worldwide, with over 50 million apparent cases and a fatality rate of 2.5 % of 0.5 million severe cases per annum in recent years. Four serotypes are currently co-circulating. Diagnosis of infection may be by polymerase chain reaction, serology or rapid antigen test for NS1. Both pan-serotype and serotype-specific genome detection assays have been described, however, achieving adequate sensitivity with pan-serotype assays has been challenging. Indeed, as we show here, inspection of components and cycling parameters of a pan-serotype RT-qPCR assay in use in laboratories worldwide revealed insufficient probe stability to accommodate potential nucleotide mismatches, resulting in false-negatives. A minor-groove binder (MGB)-modified version of the probe was designed and its performance compared with that of the original probe in 32 samples. Eight of the samples were undetected by the original probe but detected by the MGB modified probe and six out of seven of these that could be serotyped belonged to serotype 4. Sequencing of the region targeted by the probe in these samples revealed two mismatches which were also universally present in all other serotype 4 sequences in a public database. We therefore recommend adoption of this MGB modification in order to reduce the risk of false-negative results, especially with dengue serotype 4 infections.