Methods: We performed a retrospective, cross-sectional analysis of medical records from patients presenting with dengue or chikungunya to health centers in Guayas, Ecuador from October 2014 to January 2015. We included patients with dengue fever confirmed by IgM and chikungunya confirmed by PCR. We compared percent with specific clinical symptoms by chi-square analysis and mean laboratory values by t-test. Additionally, multivariate logistic regression was performed to identify statistically significant clinical predictors of dengue or chikungunya diagnosis.
Results: A total of 52 patients were included, of whom 26 were confirmed as dengue (50%) and 26 (50%) confirmed as chikugunya. Both groups consisted of predominantly women and the average age was similar between groups. Using multivariate logistic regression we found patients presenting with dengue had an increased odds of presenting with headache (OR 6.57, 95% CI 1.37 to 31.31 p=0.01), myalgia (OR 5.5, 95% CI 1.31 to 22.92 p=0.01), nausea (OR 6.35, 95% CI 1.08 to 37.23 p=0.01), and leukopenia (OR 3.6, 95% CI 0.97 to 13.34 p=0.04). While patients with chikungunya had an increased odds of presenting with chills (OR 13.23, 95% CI 1.24 to 140.81 p=0.01) and granulocytosis > 70% granulocytes (OR 7.5, 95% CI 1.70 to 33.03 p=0.00).
Conclusion: We demonstrate the difficulty in distinguishing DENV from CHIKV infection. From a clinical perspective consistent with other literature, our results indicate that leukopenia and thrombocytopenia in a dengue endemic area better supports a diagnosis of dengue as opposed to chikungunya. Contrary to findings elsewhere, in our American cohort we did not find that CHIKV was more commonly associated with arthralgias than DENV. Larger cohort studies are needed to better define the clinical patterns of CHIKV and DENV in the Americas.
M. López Garcia,
R. Litardo Salinas, None
W. A. Qualls, None
A. Y. H. Chang, None
D. Vásquez Cedeño, None