Acute-phase blood samples were collected from 69 patients with clinically suspected cases of DENV/CHIKV co-infection. Viral RNA was extracted from serum samples by using the MagNA Pure Compact Nucleic Acid Isolation System (Roche Diagnostics, Basel, Switzerland). Published primers and cycling conditions were used for the amplification of DENV (10) and CHIKV (11). cDNA was synthesized by using avian myeloblastosis virus reverse transcriptase (Promega Corp., Madison, WI, USA) and downstream consensus primer (D2) for DENV and random hexamers for CHIKV. DENV typing was performed by using second-round amplification with type-specific primers (10). The amplified products were visualized by electrophoresis on 2% agarose gels. Because samples were received during a dengue outbreak, ethical clearance was not required.

Partial nucleotide sequences of the envelope 1 (E1) gene (294 bp) of CHIKV were determined by using an automated 310 DNA sequencer (Applied Biosystems Inc., Foster City, CA, USA). Sequences were aligned, analyzed, subjected to homology search by BLAST analysis (www.ncbi.nlm.nih.gov/Education/BLASTinfo/information3.html), and submitted to GenBank (accession nos. EU727159–63 and EF539265). Phylogenetic analysis of CHIKV sequences (Table) was conducted by using ClustalW (www.ebi.ac.uk/Tools/clustalw2/index.html) and MEGA version 3.1 software (12), Kimura 2-parameter distances, and neighbor-joining algorithms.

Figure 1

Figure 1. Agarose gel electrophoresis showing chikungunya virus (A) and dengue virus (B) PCR products. A) Lane 1, 294-bp product specific for chikungunya virus; lane 2, 100-bp DNA marker. B) Lane 1, 100-bp...

Of 69 samples tested, DENV RNA was detected in 48 and CHIKV RNA in 17. Of the 17 CHIKV-positive samples, 6 were co-infected with DENVs. Three of the 6 samples from patients co-infected with CHIKV/DENV contained DENV-3; 1 contained DENV-4, and 2 contained 2 DENV serotypes (1 contained DENV-3 and DENV-4 and 1 contained DENV-3 and DENV-1) (Figure 1).

Of the 17 CHIKV-positive patients, 10 were male and 6 were female. Information regarding age, sex, and clinical features was not available for 1 patient. Thirteen samples were from adults (>12 years of age) and 3 were from children (<12 years of age).

Retrospective analysis of medical records identified clinical information for 6 patients co-infected with DENV and CHIKV. All 6 patients had fever, headache, joint pain, and low thrombocyte counts (<100,000/mm³). The patients with only CHIKV infection had fever, headache, and joint pain. Of the 6 patients with co-infections, 2 had dengue hemorrhagic fever with central nervous system (CNS) involvement. CNS involvement and hemorrhagic manifestations may be caused by DENVs because these manifestations are common in patients infected with DENV; CNS involvement has been documented in persons with DENV infections (13). In 2 patients with CNS involvement, 1 was infected with DENV-3 and 1 was infected with DENV-4. Of the 6 patients with co-infections, 5 fully recovered and 1 died.

Figure 2

Figure 2. Phylogenetic analysis of partial envelop 1 (E1) gene sequences (294 bp) of chikungunya virus strains from the 2006 dengue outbreak in Delhi, India. Neighbor-joining tree was constructed by using E1 gene...

Phylogenetic analysis of partial E1 gene sequences demonstrated that all CHIKV strains from Delhi grouped with isolates obtained during 2006 from southern India and islands in the Indian Ocean and belonged to the Central/East African genotype (Figure 2). This finding indicates that during 2006 similar strains were circulating throughout India. Isolates obtained in India during 1963–1973 clustered with isolates from Thailand (Thai 62–78) and formed a separate cluster in the Asian genotype.

For many years, it appeared that CHIKV had disappeared from India, but late in 2005 the virus reemerged on Reunion Island and in India (3). Confirmed cases of CHIKV infection have been reported from Delhi, Haryana, Uttar Pradesh, and Rajasthan provinces in northern India, although these states did not have large-scale epidemics (14).

DENV infections are endemic to northern India; in recent years, increasing trends of cocirculation of multiple DENV serotypes in Delhi suggest that DENVs are becoming hyperendemic to this region (8). During 2006, DENV and CHIKV were detected in Delhi (14). Because the clinical features of DENV and CHIKV are similar, CHIKV infections may go undiagnosed in DENV-endemic areas. In India, Ae. aegypti mosquitoes are primary vectors for DENV and CHIKV, and opportunities for co-infections in humans are increased by the feeding behavior of the mosquito (15), low socioeconomic conditions, and high population density.

We report co-infections with DENV and CHIKV in India after a long absence of the viruses in this region. It is difficult to comment on increased severity of illness in patients with DENV/CHIKV co-infections because the number of patients tested was small. Additional clinical information is needed to determine the influence of co-infections on clinical expression of dengue and chikungunya fever.

Our study indicates that co-infections with CHIKV and DENV occur in areas where these 2 viruses cocirculate. Concurrent infections may result in illness with overlapping signs and symptoms, making diagnosis and treatment difficult for physicians. Repeated outbreaks of dengue, recent activity of CHIKV, and CHIKV/DENV co-infections in the Delhi area suggest that the epidemiology of these viruses is changing in this region and that these viruses are becoming endemic to this region. Thus, in clinically suspected cases of dengue or chikungunya fever, it is advisable to test for both viruses in areas where they cocirculate.