Fig 1.
Cohort-based and national surveillance analysis of CHIKV infection in Thailand.
(A) Map of Thailand indicating the five locations where samples were collected. Samples were obtained from individuals admitted to hospitals between 2020 and 2023 across five provinces: Bangkok (N = 841), Samut Prakan (N = 352), Samut Sakhon (N = 44), Chon Buri (Eastern region; N = 24), and Surin (Northeastern region; N = 3). The base map was adapted from Northern Thailand map 02.svg by Douglas Paul Perkins, Wikimedia Commons, licensed under CC BY 4.0 (https://commons.wikimedia.org/wiki/File:Northern_Thailand_map_02.svg) and modified by the authors. (B) Monthly counts of suspected and laboratory-confirmed CHIKV cases among study participants from March 2020 to December 2023. Suspected cases are represented by a line graph, while laboratory-confirmed cases are displayed as bar graphs, with exact counts shown above each bar. (C) Annual reported chikungunya cases and (D) heat map displaying the monthly distribution of reported cases from 2008 to 2023, both based on data from the Ministry of Public Health (MoPH), Thailand.
Fig 2.
Detection of CHIKV RNA and CHIKV-specific IgM/IgG antibodies.
(A) Venn diagram depicting diagnostic results for samples that tested positive by real-time RT-PCR (coral pink circle), IgM fluorescence immunoassay (FIA) (yellow circle), or IgG FIA (blue circle). CHIKV infection was confirmed in 638 individuals based on positive results from real-time RT-PCR and/or CHIKV-specific IgM antibody detection. (B) Positive cases detected by real-time RT-PCR and CHIKV-specific IgM/IgG FIA were categorized by days post-symptom onset. The heatmap displays the percentage of positive samples per assay per day after symptom onset. Below the heatmap, the corresponding number of positive samples and the total number of CHIKV-positive samples collected each day after symptom onset are shown. The bottom row, labeled “All suspected samples tested,” indicates the number of suspected CHIKV cases from which samples were collected each day after symptom onset.
Table 1.
Demographic and clinical characteristics of individuals according to CHIKV infection status (N = 1,264). The table summarizes the distribution of sex, age, and clinical symptoms among confirmed CHIKV and non-CHIKV cases, along with odds ratios (OR), confidence intervals (CI), and both unadjusted and FDR-adjusted p-values for group comparisons.
Table 2.
Clinical features by age group among CHIKV-infected patients (N = 638). The table shows the proportion of patients in each age group reporting fever, rash, arthralgia, myalgia, vomiting, conjunctivitis, headache, sore throat, fatigue, and diarrhea. P-values and FDR-adjusted p-values (false discovery rate) are provided for comparisons across age groups. Statistical significance was defined as FDR < 0.05.
Fig 3.
Correlation between CHIKV viral load, timing of symptom onset, and clinical outcomes (N = 454).
(A) The correlation between CHIKV viral load and timing of sample collection relative to symptom onset. (B) Analysis of the association between CHIKV viral load and different clinical feature groups. Each data point corresponds to an individual patient. (*; FDR-adjusted p value < 0.05).
Table 3.
Multivariate logistic regression identifying independent predictors of key clinical symptoms among RT-PCR–confirmed CHIKV cases (N = 454). For each symptom (fever, rash, arthralgia, myalgia, conjunctivitis), associations with age (per 10-year increase), sex (female vs male), log10 CHIKV viral load (copies/mL), and days from illness onset to sample collection were evaluated. Adjusted odds ratios (aORs), 95% confidence intervals (CIs), and p-values are shown.
Fig 4.
Phylogenetic tree analysis of CHIKV complete coding sequences.
A maximum-likelihood (ML) phylogenetic tree was constructed using the complete coding sequences of CHIKV strains from Thailand identified in this study, along with various strains obtained from the GenBank database. The tree was inferred using the GTR + F + I + G4 substitution model with 1000 ultrafast bootstrap replicates. Bootstrap values ≥80 are shown at major nodes. Red dots indicate CHIKV strains from Thailand identified in this study (GenBank accession numbers PQ637673–PQ637710). The left color strip represents the geographic regions of the isolates, and the right color strip indicates the viral genotypes.
Fig 5.
Molecular clock analysis of CHIKV ECSA-IOL.
(A) Temporal signal analysis showing the correlation between sequence sampling dates and their genetic distances from the root of the maximum likelihood phylogeny (R² = 0.967). (B) Maximum clade credibility (MCC) tree of the CHIKV ECSA-IOL genotype. Arrows indicate the time to the most recent common ancestor (tMRCA) with corresponding 95% highest posterior density (HPD) intervals and amino acid substitutions. Black nodes represent branches with posterior probability (PP) > 0.95, and blue node bars denote the 95% HPD intervals for node heights. Sequences from Thailand identified in this study (GenBank accession numbers PQ637673–PQ637710) are labeled in red. Previously published Thai strains are shown in black, and other global reference sequences are colored grey. Sequences are named following the format: accession number_country_collection year. Amino acid mutations specific to each isolate are shown next to the sequence tips in the MCC tree.