Abstract
The reemergence of dengue virus (DENV) infection has created a requirement for improved laboratory diagnostic procedures. In this study, DENV genome detection in urine was evaluated as a diagnostic method. The DENV genome was detected by real-time reverse transcriptase PCR (RT-PCR) in urine and serum of dengue patients. The detection rate of DENV genome in urine was 25% (2/8) on disease days 0 to 3 and 32% (7/22) on days 4 to 5. The rate was 50% or higher on days 6 to 16, 52% (11/21) on days 6 to 7, 78% (7/9) on days 8 to 9, 80% (4/5) on days 10 to 11, 50% (2/4) on days 12 to 13, and 60% (3/5) on days 14 to 16. The last positive urine sample was on day 16. The detection rates in serum were highest on days 0 to 3 and were greater than 50% on days 0 to 7. Detection rates decreased thereafter, and the last positive detection was on day 11. These results indicate that the time frames for positive detection differ between urine and serum samples, whereby detection rates of 50% or higher are evident between days 6 to 16 for urine samples and days 0 to 7 for serum samples. Nucleotide sequences of PCR products were identical between urine and serum samples. The detection of DENV genome in urine samples by real-time RT-PCR is useful to confirm DENV infection, particularly after viremia disappears.
INTRODUCTION
Dengue virus (DENV) infections occur in most of the tropical and subtropical areas of the world. DENV infection with any of four serotypes leads to a broad spectrum of clinical symptoms and severity, including asymptomatic infection, dengue fever (DF), and fatal dengue hemorrhagic fever (DHF). DF/DHF is considered one of the most important reemerging infectious diseases (4). Physicians and pediatricians in countries in which these diseases are not endemic are often unfamiliar with the symptoms and unaware of the potential importation of patients with DF/DHF. As such, DF/DHF often may not be considered part of a differential diagnosis. Furthermore, laboratory diagnosis is hampered in areas where the disease is endemic because of the limited number of facilities with diagnostic capacity, and specimen collection in a proper time frame is not easy in areas were DF/DHF is endemic.
Several laboratory diagnostic techniques have been used for the confirmation of dengue virus infection: viral isolation, viral antigen detection, viral genome detection, and antibody (Ab) detection. IgM capture enzyme-linked immunosorbent assay (ELISA) and real-time reverse transcriptase PCR (RT-PCR) are commonly used (6, 8, 16). NS-1 antigen detection tests have also recently become commercially available (10); however, they cannot determine specific viral types. The antibody/antigen detection of DENV provides less information than the other detection assays, and the virus can be successfully isolated only during limited stages of infection. For detailed analyses, the detection of the DENV genome in serum samples by RT-PCR is widely used. A fluorogenic probe-based assay, which has a number of advantages over conventional RT-PCR, has recently been developed. It has the advantages of reduced turnaround time and a much lower risk of contamination compared to that of conventional RT-PCR (3). However, it is usually difficult to detect viral genomes after the development of antibodies against DENVs and the onset of defervescense (14). The use of urine samples for laboratory diagnostic testing has some advantages over the use of serum samples, such as ease of use and noninvasiveness. Our group and others have previously reported the detection of DENV genome in urine samples for a limited number of patients (11, 12). In the present study, we attempted to determine the usefulness of urine samples in the laboratory diagnosis of DENV infection. In the present study, we evaluated the usefulness of urine samples in the laboratory diagnosis of DENV infection by comparing real-time RT-PCR from serially collected urine and serum samples from confirmed DENV cases. We also compared RT-PCR for urine and serum to IgG and IgM ELISAs for serum and virus isolation from urine.
MATERIALS AND METHODS
Sample collection.
Serum and urine samples were collected from 53 dengue patients at clinics and hospitals in Japan from 2006 to 2008, and they were sent to the National Institute of Infectious Diseases (NIID) for laboratory diagnosis. The median age was 30 years with a range of 9 to 65 years. All patients had a history of visits to countries in which dengue is endemic before onset and had DENV genome detected by real-time RT-PCR or specific anti-dengue antibodies by ELISA.
Isolation of dengue viruses from urine samples.
Vero cells were used to isolate DENV from urine samples. The urine samples were filtered through 0.45-μm filters (Mix GS; Millipore). Urine samples (0.1 ml) were inoculated onto Vero cell monolayers in a 6-well cell culture plate (Corning Inc., NY) and incubated for 1 h. The cells were washed twice with phosphate-buffered saline without potassium and with 2% fetal calf serum (FCS) minimum essential medium (MEM) and then cultured at 37°C in 5% CO2 for 7 days. The presence of DENV in the culture fluids was checked by the real-time RT-PCR TaqMan method.
Real-time RT-PCR (TaqMan RT-PCR).
Dengue viral genomes in serum and urine samples were examined by real-time RT-PCR (TaqMan RT-PCR) assay as previously reported (5). Primer and probe sequences are provided in Table S2 in the supplemental material. Briefly, RNA was extracted from 200 μl of samples using a High Pure viral RNA kit (Roche Applied Science, Mannheim, Germany) and eluted by 50 μl RNase-free distilled water. Five microliters of RNA extracted from each sample was mixed with 100 pmol of each primer and 15 pmol of each probe in a 25-μl reaction volume using a TaqMan RT-PCR Ready-Mix kit (PE Applied Biosystems). The samples were amplified in an ABI Prism 7000 sequence detection system (PE Applied Biosystems). The real-time RT-PCR assay consisted of a 30-min RT step at 48°C and 45 cycles of PCR steps (95°C for 15 s and 57°C for 60 s).
Detection of DENV-specific IgM and IgG antibodies.
Anti-dengue virus IgM Abs in serum samples were detected by an IgM-capture ELISA kit (dengue virus IgM capture DxSelect; Focus Diagnosis, CA) according to the manufacturer's protocol. Anti-dengue virus IgG Abs were detected by ELISA using an IgG indirect ELISA kit (Dengue IgG indirect ELISA; Panbio Ltd., Queensland, Australia) according to the manufacturer's protocol. IgM and IgG Abs were determined to be positive when the IgM index and IgG index were equal to or greater than 1.1, respectively.
Sequence analysis of viral RNA.
RNA genomes were sequenced for serum and urine samples from selected patients after both samples were determined to be positive by real-time RT-PCR. To determine whether DENV genome detected in the urine sample originated from DENV in sera, the sequences of the E gene were analyzed for comparison. The E gene was used for analysis because it is the most frequently analyzed gene among DENV genomes (1). Primers used for RT-PCR for sequence analysis are provided in Table S1 in the supplemental material. Multiple pairs of primers were used to amplify and sequence the DENV genome. RNA genomes might be partially damaged in urine samples.
Ethical aspects.
This work was approved by the ethics committee of the NIID (application no. 98; 3 July 2006).
(Part of this paper was presented at the Second International Conference on Dengue and Dengue Haemorrhagic Fever International, Phuket, Thailand, October 2008.)
RESULTS
Detection of dengue viral genomes in urine samples.
Seventy-seven urine samples from 53 confirmed dengue patients were examined for the presence of DENV genome by real-time RT-PCR (Table 1). The results of real-time RT-PCR are presented based on disease days (Fig. 1), and the data were further analyzed by the sum of data from 2 consecutive days (Table 2). Disease days were defined based on the time of fever onset. The onset day was defined as day 0. DENV genome was detected as early as day 1 in one urine sample; however, before day 6 the rate of detection was low: 25% on disease days 0 to 3 and 32% on disease days 4 to 5. The positive detection rate reached 50% on disease days 6 to 7 and remained at 50% or higher until days 14 to 16. Positive samples were detected as late as day 16.
Table 1.
Results of real time RT-PCR of urine and serum samples and levels of serum IgM and IgG
| Patient no. | Disease day | Infecting serotype | RT-PCR result withb: |
Serum antibodya (index) |
||
|---|---|---|---|---|---|---|
| Urine | Serum | IgM | IgG | |||
| 1 | 7 | D1 | + | − | + (9.0) | + (2.0) |
| 8 | D1 | + | NT | NT | NT | |
| 14 | D1 | + | − | + (6.1) | + (2.3) | |
| 25 | D1 | − | − | + (4.1) | + (2.4) | |
| 2 | 5 | − | − | + (4.1) | − (0.4) | |
| 3 | 4 | D1 | − | + | + (2.0) | − (0.3) |
| 14 | D1 | − | NT | NT | + (3.5) | |
| 2 M | D1 | NT | NT | + (2.7) | + (1.9) | |
| 4 | 2 | D1 | NT | + | − (0.6) | − (0.3) |
| 7 | D1 | + | NT | + (6.3) | + (1.9) | |
| 5 | 4 | D1 | − | + | + (1.2) | + (2.7) |
| 6 | 6 | D1 | − | + | + (5.4) | + (2.6) |
| 22 | D1 | NT | − | + (5.9) | + (3.0) | |
| 7 | 7 | D1 | NT | NT | + (10.6) | + (1.4) |
| 9 | D1 | + | NT | + (17.0) | + (1.5) | |
| 8 | 5 | D1 | NT | + | − (0.9) | + (1.4) |
| 11 | D1 | + | NT | + (5.4) | + (3.9) | |
| 9 | 6 | D1 | + | + | + (8.5) | + (3.7) |
| 10 | 1 | D1 | + | + | − (0.8) | − (0.2) |
| 5 | D1 | + | NT | + (7.2) | + (1.0) | |
| 6 | D1 | + | NT | + (7.5) | + (1.8) | |
| 11 | 0 | D1 | NT | + | + (1.1) | − (0.1) |
| 6 | − | NT | NT | NT | ||
| 7 | D1 | + | + | + (6.4) | + (2.1) | |
| 8 | D1 | NT | + | + (8.1) | + (2.5) | |
| 10 | D1 | + | − | + (11.3) | + (3.2) | |
| 12 | 1 | D2 | NT | + | − (0.8) | + (1.2) |
| 4 | D2 | NT | + | + (1.2) | + (1.8) | |
| 6 | D2 | + | + | + (3.2) | + (2.2) | |
| 11 | D2 | + | + | + (4.1) | + (3.5) | |
| 13 | 3 | D2 | + | + | + (1.6) | + (4.2) |
| 6 | D2 | − | + | + (3.6) | + (6.6) | |
| 14 | 4 | D2 | + | + | + (1.0) | − (0.4) |
| 15 | 5 | D2 | − | + | + (2.5) | + (3.6) |
| 16 | 1 | D2 | NT | + | − (0.6) | + (1.0) |
| 3 | D2 | NT | NT | − (0.9) | + (1.3) | |
| 4 | − | NT | NT | NT | ||
| 5 | D2 | + | NT | − (0.9) | + (1.1) | |
| 9 | D2 | + | NT | NT | NT | |
| 12 | D2 | NT | NT | + (5.5) | + (3.5) | |
| 17 | 4 | D3 | + | + | + (2.8) | + (3.1) |
| 14 | − | − | + (8.6) | + (3.3) | ||
| 18 | 3 | D3 | NT | + | − (0.5) | − (0.6) |
| 5 | D3 | NT | + | + (1.8) | + (1.0) | |
| 8 | D3 | + | − | + (6.5) | + (2.2) | |
| 14 | D3 | + | − | + (5.4) | + (2.9) | |
| 19 | 4 | D3 | − | + | + (6.3) | + (1.1) |
| 20 | 4 | D3 | NT | + | + (1.9) | − (0.2) |
| 6 | D3 | + | + | + (8.4) | − (0.2) | |
| 21 | 5 | D3 | + | − | + (6.8) | + (2.5) |
| 22 | 3 | D3 | − | + | − (0.9) | + (3.1) |
| 23 | 0 | D3 | NT | + | − (0.6) | NT |
| 1 | D3 | NT | + | − (0.7) | − (0.3) | |
| 3 | D3 | − | + | + (2.7) | + (1.0) | |
| 24 | 9 | D3 | + | − | − (0.6) | NT |
| 25 | 3 | D3 | NT | + | + (1.1) | + (3.74) |
| 5 | NT | NT | + (4.6) | + (4.61) | ||
| 7 | D3 | + | NT | + (7.6) | + (4.7) | |
| 12 | D3 | + | NT | + (5.7) | + (4.7) | |
| 26 | 4 | D3 | NT | + | + (3.8) | + (1.7) |
| 5 | D3 | + | + | + (5.8) | + (2.3) | |
| 11 | NT | NT | + (9.6) | + (6.2) | ||
| 13 | D3 | + | − | + (10.6) | + (6.1) | |
| 27 | 6 | D3 | NT | + | + (2.4) | + (3.4) |
| 7 | D3 | + | NT | + (6.2) | + (4.1) | |
| 14 | NT | NT | + (9.0) | + (7.0) | ||
| 28 | 1 | D3 | − | + | − (0.6) | − (0.9) |
| 3 | − | NT | + (1.2) | + (1.1) | ||
| 5 | − | NT | + (6.4) | + (1.8) | ||
| 29 | 5 | D3 | + | + | + (6.2) | − (0.2) |
| 7 | NT | NT | + (12.0) | − (0.3) | ||
| 30 | 1 | D3 | NT | + | + (1.0) | − (0.1) |
| 16 | D3 | + | − | + (8.9) | + (2.2) | |
| 31 | 7 | D3 | + | + | + (6.8) | + (2.8) |
| 32 | 7 | D4 | − | − | + (6.8) | + (2.9) |
| 33 | 8 | D4 | − | + | + (8.3) | + (226.0) |
| 33 | − | NT | + (7.5) | + (3.4) | ||
| 34 | 3 | D4 | − | + | − (0.8) | + (1.0) |
| 5 | D4 | − | + | + (3.5) | + (2.1) | |
| 9 | D4 | + | − | + (8.7) | NT | |
| 11 | D4 | + | NT | NT | NT | |
| 35 | 1 | D4 | NT | + | + (1.1) | + (2.4) |
| 4 | NT | NT | + (1.5) | + (1.5) | ||
| 6 | D4 | + | NT | NT | NT | |
| 8 | D4 | + | NT | + (6.0) | + (3.0) | |
| 36 | 4 | D4 | − | + | − (0.8) | + (1.1) |
| 7 | NT | NT | + (1.9) | + (5.3) | ||
| 37 | 5 | D4 | − | + | + (3.1) | − (0.7) |
| 8 | NT | NT | + (4.3) | NT | ||
| 38 | 5 | − | NT | + (5.1) | + (1.9) | |
| 12 | − | NT | + (11.6) | + (3.0) | ||
| 39 | 4 | NT | − | + (3.0) | + (1.0) | |
| 9 | − | − | + (3.4) | + (1.4) | ||
| 40 | 7 | − | − | + (1.8) | + (1.4) | |
| 10 | NT | NT | + (3.6) | + (2.0) | ||
| 41 | 2 | NT | − | + (9.8) | + (1.6) | |
| 5 | − | NT | NT | NT | ||
| 42 | 4 | NT | + | + (4.4) | + (2.1) | |
| 6 | − | NT | + (11.4) | + (2.7) | ||
| 43 | 5 | − | − | + (5.6) | − (0.7) | |
| 44 | 5 | − | − | + (1.5) | + (4.5) | |
| 45 | 7 | − | − | + (13.1) | + (2.4) | |
| 46 | 4 | NT | − | + (11.8) | + (5.1) | |
| 5 | − | − | + (10.7) | + (5.3) | ||
| 6 | − | − | + (10.3) | + (5.0) | ||
| 47 | 7 | NT | − | + (5.0) | − (0.7) | |
| 10 | NT | − | + (5.3) | + (1.4) | ||
| 12 | − | NT | NT | NT | ||
| 18 | NT | − | + (8.9) | + (2.9) | ||
| 48 | 3 | − | + | − (0.9) | − (0.4) | |
| 49 | 5 | NT | − | + (4.7) | + (2.4) | |
| 10 | − | − | + (6.0) | + (4.0) | ||
| 50 | 21 | − | − | + (5.9) | + (6.2) | |
| 51 | 7 | − | + | + (14.6) | + (2.9) | |
| 52 | 10 | − | − | + (13.5) | − (0.5) | |
| 53 | 6 | − | − | + (3.7) | + (1.5) | |
IgM and IgG indexes were calculated by the following formula: index = optical density (OD) of samples/cutoff OD. A plus sign indicates positive IgM and IgG indices, and a minus sign indicates negative IgM and IgG indices.
NT, not tested.
Fig 1.
Number of DENV genome-positive and -negative urine samples determined by real-time RT-PCR based on disease days. Closed bar, positive; open bar, negative.
Table 2.
Positive rate of real-time RT-PCR with urine and serum samples and serum IgM and IgGa
| Disease days | No. (%) of samples positive by: |
||||
|---|---|---|---|---|---|
| RT-PCR |
Serum ELISA |
||||
| Urine | Serum | Serum and urine | IgM | IgG | |
| 0–1 | 1/2 (50) | 9/9 (100) | 9/9 (100) | 2/9 (22) | 2/8 (25) |
| 2–3 | 1/6 (17) | 8/9 (89) | 8/9 (89) | 5/11 (45) | 6/11 (55) |
| 4–5 | 7/22 (32) | 17/25 (68) | 20/32 (63) | 26/30 (87) | 21/31 (68) |
| 6–7 | 11/21 (52) | 9/16 (56) | 15/23 (65) | 25/25 (100) | 21/24 (88) |
| 8–9 | 7/9 (78) | 2/6 (33) | 9/10 (90) | 8/9 (89) | 6/6 (100) |
| 10–11 | 4/5 (80) | 1/4 (25) | 4/5 (80) | 8/8 (100) | 7/8 (88) |
| 12–13 | 2/4 (50) | 0/1 (0) | 2/4 (50) | 4/4 (100) | 4/4 (100) |
| 14–16 | 3/5 (60) | 0/4 (0) | 3/5 (60) | 5/5 (100) | 6/6 (100) |
| >16 | 0/3 (0) | 0/4 (0) | 0/5 (0) | 6/6 (100) | 6/6 (100) |
When samples of urine and blood were obtained from the same patient at the same time, the number of samples was considered to be 1.
Detection of viral genomes in serum samples.
Seventy-eight serum samples from the 53 patients were also examined by real-time RT-PCR. The results are presented based on disease days (Fig. 2), and the data were analyzed by the sum of data from 2 consecutive days (Table 2). The positive detection rate was high on early disease days: 100% on days 0 to 1, 89% on days 2 to 3, and 68% on days 4 to 5. The positive rate decreased thereafter, and no positive samples were detected after disease day 11.
Fig 2.
Number of DENV genome-positive and -negative serum samples determined by real-time RT-PCR based on disease days. Closed bar, positive; open bar, negative.
Comparison of real-time RT-PCR positive detection rates between urine and serum samples.
The results shown in Table 1 and the figures indicate that the time courses of positive detection rates differed between urine and serum samples. Positive rates of 50% or more were detected on disease days 6 to 16 for urine samples and days 0 to 7 for serum samples.
This tendency was clearly demonstrated using serial samples from patients 11 and 34. For patient 34, urine samples were negative on disease days 3 and 5 but positive on days 9 and 11, whereas serum samples were positive on days 3 and 5 but negative on day 9. For patient 11, urine samples were negative on day 6 but positive on days 7 and 10, whereas serum samples were positive on days 0 and 7 but negative on day 10. The relative index between a positive rate of urine samples and days after onset is 0.087. The relative index between a positive rate of serum samples and days after onset is 0.97. The P value is 0.00068, indicating significant differences between serum and urine sample detection rates over time (Table 2).
Relationship of DENV genome detection in urine and serum samples and serum IgM and IgG antibody responses.
IgM-positive rates were high (more than 87%) on and after days 4 to 5 (Table 2), and IgG-positive rates were high (more than 88%) on and after days 6 to 7. Real-time RT-PCR positive detection rates in serum declined once IgM and IgG were detected in serum samples, while real-time RT-PCR positive detection rates in urine samples remained at high levels, even after high IgM and IgG-positive rates in serum samples.
Comparison of nucleotide sequences of amplified DENV genomes between urine and serum samples.
Nucleotide sequences of amplified DENV genomes were compared between urine and serum samples from the same patients. Partial nucleotide sequences were identical between urine and serum samples for each of the 6 cases examined. Nucleotide sequencing showed that all four DENV serotypes were represented among the 53 patients.
DISCUSSION
The detection of DENV genome in serum samples by RT-PCR has been used widely for the confirmation of dengue virus infection. In the present study, the detection of DENV genome in urine samples was evaluated as an additional or alternative laboratory diagnostic test. Real-time RT-PCR was applied to the detection of DENV genome, and detection rates were compared between urine and serum samples. The nucleotide sequences of the PCR products on the E region were determined. The sequence results revealed that the nucleotide sequences of RT-PCR products were identical between urine and serum samples from respective patients, indicating that these represented the same infecting strain of DENV.
We previously reported a dengue case with the detection of DENV-1 genome in urine samples on disease days 7, 8, and 14, a time when the genome was not detected in serum samples (11), and another group has also reported the detection of DENV genome in urine samples (12). In those studies, the number of studied cases was limited to one and two cases, respectively. In the present study, we examined 77 urine and 78 serum samples from 53 confirmed dengue patients. The results indicate that real-time RT-PCR testing of urine samples is a useful diagnostic tool for DENV infection. It is of interest that DENV genome was detected at 50% or higher rates in urine samples even after disease day 7, when the positive detection rates of serum samples was lower than 50%. In particular, DENV genome was detected in 5 of 9 urine samples collected on disease days 12 to 16, when none of 5 serum samples was positive. For patients 30 and 34, DENV genome was detected in urine samples but not in serum samples after the onset of positive antibody detection in serum. The positive rates of the combined PCR results for urine and serum samples were calculated, and they were constantly at higher levels than the results for serum samples from days 6 to 16. The positive rates for urine samples started to increase on days 6 to 9. The positive rate was 67% (34 out of 51 cases) when only serum samples were tested, while it was 72% (38 out of 53 cases) when both serum and urine samples were tested. Furthermore, on day 6 and after, the positive rate was 41% (11 out of 27 cases) when serum samples were tested, while it was 56% (22 out of 39 cases) when both serum and urine samples were tested. The results suggest that real-time RT-PCR with urine samples is useful mainly as a supplemental test in the convalescent stages of DENV infection. Thus, it is recommended that both serum and urine samples be collected and examined in the convalescent stages of DENV infection. The RT-PCR method using urine samples has 3 advantages over serological assays: (i) infection can be determined using single samples, (ii) RT-PCR has high specificity, and (iii) serotypes can be determined. The serological diagnostic method needs 2 samples, at acute and convalescent stages, to confirm the latest infection. Serological assays often demonstrate cross-reactivity among flaviviruses, including DENV. The use of urine samples has advantages over the use of other samples for laboratory testing. Urine samples are easy to collect without invasive procedures, and they can be used for the dengue diagnosis of newborns, children, and patients with hemorrhagic symptoms. The isolation of infectious DENV was attempted from urine samples in which DENV genome was detected. Infectious DENV has not been isolated from any of the samples tested in this study. Further studies are needed to conclusively determine that there is no infectious DENV in urine samples. However, it is likely that DENV genome or noninfectious DENV virions, rather than infectious DENV, are present. The detection of dengue antibodies in urine samples has been reported (15). It is also possible that DENV-antibody complexes were detected by real-time RT-PCR in urine samples.
The detection of the genome of West Nile virus, another flavivirus, in urine samples from a patient with encephalitis has been reported; however, virus isolation from urine samples was also unsuccessful (13). Renal dysfunction is not common in dengue patients (9). The presence of DENV antigen in kidneys from dengue patients has been reported by several groups (2, 7). However, it has not clearly been determined whether DENV replication occurs in the kidney. Thus, the pathological mechanism corresponding to the presence of DENV genome in urine needs to be elucidated.
In conclusion, DENV genome was detected by real-time RT-PCR in urine samples in 24 of 53 confirmed dengue patients. It was possible to detect DENV genome in urine samples even after the appearance of antibodies in serum and the disappearance of viremia. Urine samples are readily obtainable, and the detection of DENV genome by real-time RT-PCR of urine samples is a useful laboratory diagnostic tool for DENV infection.
Supplementary Material
ACKNOWLEDGMENTS
This study was supported by grants (H20-shinkou-ippan-15 and H23-shinkou-ippan-10) for Research on Emerging and Reemerging Infectious Diseases from the Ministry of Health, Labor and Welfare, Japan.
None of the authors have a commercial or other association that might pose a conflict or interest.
Footnotes
Published ahead of print 21 March 2012
Supplemental material for this article may be found at http://jcm.asm.org/.
REFERENCES
- 1. Amarilla AA, et al. 2009. Genetic diversity of the E protein of dengue type 3 virus. Virol. J. 6:113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Basílio-de-Oliveira CA, Aquiar GR, Baldanza MS, Birth OM, Eyer Silva WA, Paes MV. 2005. Pathologic study of fatal case of dengue-3 virus infection in Rio de Janeiro, Brazil. Braz. J. Infect. Dis. 9:341–347 [DOI] [PubMed] [Google Scholar]
- 3. Callahan JD, et al. 2001. Development and evaluation of serotype- and group-specific fluorogenic reverse transcriptase PCR (TaqMan) assays for dengue virus. J. Clin. Microbiol. 39:4119–4124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Halstead SB. 2007. Dengue. Lancet 370:1644–1652 [DOI] [PubMed] [Google Scholar]
- 5. Ito M, et al. 2004. Development and evaluation of fluorogenic TaqMan reverse transcriptase PCR assays for detection of dengue virus type 1 to 4. J. Clin. Microbiol. 42:5935–5937 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Itoda I, et al. 2006. Clinical features of 62 imported cases of dengue fever in Japan. Am. J. Trop. Med. Hyg. 75:470–474 [PubMed] [Google Scholar]
- 7. Jessie K, Fong MY, Devi S, Law SK, Wong KT. 2004. Localization of dengue virus in naturally infected human tissues, by immunohistochemistry and in situ hybridization. J. Infect. Dis. 189:1411–1418 [DOI] [PubMed] [Google Scholar]
- 8. Johnson BW, Russell BJ, Lanciotti RS. 2005. Serotype-specific detection of dengue viruses in a fourplex real-time reverse transcriptase PCR assay. J. Clin. Microbiol. 10:4977–4983 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Karen LR, Kenneth LT. 2008. Viral encephalitis, p 2629–2633 In Fauci A, et al. (ed), Harrison's principles of internal medicine, 17th ed McGraw-Hill, New York, NY [Google Scholar]
- 10. Kumarasamy V, et al. 2007. Evaluation of a commercial dengue NS1 antigen-capture ELISA for laboratory diagnosis of acute dengue virus infection. J. Virol. Methods 140:75–79 [DOI] [PubMed] [Google Scholar]
- 11. Mizuno Y, et al. 2007. Confirmation of dengue virus infection by detection of dengue virus type 1 genome in urine and saliva but not in plasma. Trans. R. Soc. Trop. Med. Hyg. 101:738–739 [DOI] [PubMed] [Google Scholar]
- 12. Poloni TR, et al. 2010. Detection of dengue virus in saliva and urine by real time RT-PCR. Virol. J. 7:22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tonry JH, et al. 2005. West Nile virus detection in urine. Emerg. Infect. Dis. 11:1294–1296 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Vaughn DW, et al. 1997. Dengue in the early febrile phase: viremia and antibody responses. J. Infect. Dis. 176:322–330 [DOI] [PubMed] [Google Scholar]
- 15. Vázquez S, et al. 2007. Kinetics of antibodies in sera, saliva, and urine samples from adult patients with primary or secondary dengue 3 virus infections. Int. J. Infect. Dis. 11:256–262 [DOI] [PubMed] [Google Scholar]
- 16. World Health Organization 2009. Dengue: guidelines for diagnosis, treatment, prevention and control. WHO, Geneva, Switzerland: [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.