[Early indicators of dengue infection in children]

Effectiveness of a diagnostic algorithm for dengue based on an artificial neural network

Introduction

Dengue is a disease with a wide clinical spectrum. The early identification of dengue cases is crucial but challenging for health professionals; therefore, it is necessary to have effective diagnostic instruments to initiate timely care.

Objective

To evaluate the effectiveness of an algorithm based on an artificial neural network (ANN) to diagnose dengue in an endemic area.

Methods

A single-center case-control study was conducted in a secondary-care hospital in Ciudad Obregón, Sonora. An algorithm was built with the official operational definitions, which was called the "direct algorithm," and for the ANN algorithm, the brain.js library was used. The data analysis was performed with the diagnostic tests of sensitivity, specificity, positive predictive value (ppv), and negative predictive value (npv), with 95% confidence intervals and Cohen's kappa index.

Results

A total of 233 cases and 233 controls from 2022 were included. The ANN presented a sensitivity of 0.90 (95% CI [0.85, 0.94]), specificity of 0.82 (95% CI [0.77, 0.87]), npv of 0.91 (95% CI [0.87, 0.94]) and ppv of 0.81 (95% CI [0.76, 0.85]) and a kappa of 0.72. The direct algorithm had a sensitivity of 0.97 (95% CI [0.94, 0.99]), specificity of 0.96 (95% CI [0.92, 0.98]), npv 0.97 (95% CI [0.94, 0.98]), ppv 0.96 (95% CI [0.93, 0.98]) and kappa 0.93.

Conclusions

The direct algorithm performed better than the ANN in the diagnosis of dengue.

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.

Development of clinical algorithms for the diagnosis of dengue in Colombia

Introduction: Due to the increase in dengue incidence and mortality, its diagnosis is relevant for endemic countries. Clinical classifications and laboratory tests have a variable performance in clinical practice with a sensitivity level between 45% and 98%, and a specificity level between 4% and 98% partly due to the variety of contexts where they are applied. Objective: To develop clinical algorithms for the diagnosis of dengue in the Colombian context. Materials and methods: A cross-sectional study was conducted based on secondary sources. We constructed clinical diagnostic algorithms of dengue based on Bayesian methods combining symptoms, signs, and blood count parameters, and then we compared them in terms of diagnostic accuracy with gold standard tests. In addition, an external validation of the algorithm with greater accuracy and sensibility was performed comparing it with the WHO-1997 and the WHO-2009 clinical classifications, the Colombian guide for 2010, and the diagnostic scale recommended by the Ministerio de Salud y Protección Social of Colombia for 2013. Results: Four algorithms were generated, two for signs and symptoms, and two that included leukocytes (≤4,500/mm3) and/or platelets (≤160,000/mm3) counts. The most accurate algorithm included blood count parameters with a sensitivity of 76.5% (95%CI: 71.9-80.5) and a specificity of 46.0% (95%CI: 37.6-54.7). In the external validation we found a sensitivity of 11.1% (95%CI: 4.9-20.7) and a specificity of 91.9% (95%CI: 87.5-93.9). The scale of the Ministerio de Salud had a sensitivity of 76.4% (95%CI: 64.9-85.6) and a specificity of 38.0% (95%CI: 32.8-43.4). Conclusion: The inclusion of blood count parameters improved the sensitivity of diagnostics algorithms based on signs and symptoms. Clinical diagnosis of dengue remains a challenge for health research.

Burden of dengue among febrile patients at the time of chikungunya introduction in Piedecuesta, Colombia

OBJECTIVE

To estimate the age-specific incidence of symptomatic dengue and chikungunya in Colombia.

METHOD

A passive facility-based fever surveillance study was conducted among individuals with undifferentiated fever. Confirmatory diagnostics included serological and molecular tests in paired samples, and surveillance's underreporting was assessed using capture-recapture methods.

RESULTS

Of 839 febrile participants 686 completed the study. There were 33.2% (295/839) dengue infections (51% primary infections), and 35.9% (191/532) of negative dengue cases there were chikungunya cases. On average, dengue cases were younger (median = 18 years) than chikungunya cases (median = 25 years). Thrombocytopaenia and abdominal pain were the main dengue predictors, while presence of rash was the main predictor for chikungunya diagnosis. Underreporting of dengue was 31%; the estimated expansion factors indicate an underreporting rate of dengue cases of threefold for all cases and of almost sixfold for inpatients.

CONCLUSIONS

These findings highlight the ongoing coexistence of both arboviruses, a distinct clinical profile of each condition in the study area that could be used by clinicians to generate a differential diagnosis, and the presence of underreporting, mostly among hospitalised cases.