Viruses | Free Full-Text | Detection of Chikungunya Virus RNA in Oral Fluid and Urine: An Alternative Approach to Diagnosis?
1. Introduction
Chikungunya virus (CHIKV) is an alphavirus of the Togaviridae family. It was first detected near the border between Tanzania and Mozambique in 1952 [1,2]. In urban areas, Aedes aegypti and Ae. albopictus mosquitoes are the main vectors [3,4]. The autochthonous transmission of CHIKV in the Americas was first identified in the Caribbean region in 2013 [5]. Since then, CHIKV has spread across the American continents, causing multiple outbreaks and becoming a public health challenge in these regions of the world [6,7,8,9,10,11,12,13,14,15,16,17]. Brazil, where the first cases of chikungunya were detected in 2014, has been the most affected country in the Americas, with more than 1.6 million cases reported [16].
In humans, CHIKV infection can evolve asymptomatically or cause acute disease manifested by fever, headache, fatigue, myalgia, skin rash, joint swelling and especially arthralgia, which can be severe and last for months to years [14,18]. The diagnosis of chikungunya during its acute phase is based on clinical criteria, epidemiological data and, preferably, laboratory techniques. Reverse transcription–polymerase chain reaction (RT-PCR) is commonly used to detect CHIKV RNA in serum or plasma samples obtained up to seven days after the onset of symptoms [19,20,21]. Then, serological tests, such as enzyme-linked immunosorbent assay (ELISA), are the most frequently used to detect anti-CHIKV IgM antibodies seven days after the onset of symptoms or to detect antibody seroconversion between paired blood samples collected in the acute and convalescent phases of illness [21,22]. Other diagnostic approaches, such as viral isolation, hemagglutination inhibition assay and plaque reduction neutralization tests, are used less often and restricted mainly to research laboratories. Because CHIKV serological tests may cross-react with other alphaviruses, and most patients with chikungunya seek healthcare within the first three days after the onset of symptoms [23], serum RT-PCR is generally accepted as the gold-standard method for the definitive diagnosis of CHIKV infection.
However, obtaining serum from pediatric patients, especially neonates, can be complex. In these cases, the use of non-invasive biological samples is desirable. Testing easy-to-obtain samples would also facilitate the diagnostic investigation of febrile patients during periods of outbreaks when healthcare facilities are overcrowded or as part of the surveillance monitoring of virus transmission. Case reports [24,25] and small case series [26,27] have suggested that saliva and urine can offer viable means to detect CHIKV RNA through RT-PCR. However, only two main studies have compared the positivity rate of RT-PCR in the saliva, urine and serum of laboratory-confirmed chikungunya patients. These studies found superior serum performance (80.3% to 86.1% positivity) and a wide range of positivity in saliva (30% to 58.3%) and urine (8.3% to 23%) samples [28,29]. Because of the paucity of data on the potential usefulness of biological samples other than serum for diagnosing chikungunya, we investigated whether oral fluid (OF) and urine could serve as alternative specimens for diagnosing CHIKV infection via RT-qPCR.
4. Discussion
Our study on the utility of different biological samples for investigating CHIKV infection using RT-qPCR confirmed the findings of prior studies that serum should be the gold-standard sample used for diagnostic testing [28,29]. However, in this study, we found that 47% of the patients with acute chikungunya diagnosed via RT-qPCR in serum had detectable CHIKV RNA in OF (56% positivity when OF was collected within the first two days after symptom onset). Thus, in situations in which blood collection is not possible, OF can be used as an alternative sample, but a negative OF RT-qPCR result should not discount the likelihood of a CHIKV infection. In contrast, the frequency of which the urine samples were positive in RT-qPCR was too low (5%) to be useful, and this type of sample should not be used in CHIKV diagnosis.
We also found that 17% of the patients with acute chikungunya confirmed solely through CHIKV IgM seroconversion had detectable CHIKV RNA in OF but not in urine. Thus, the frequency of RT-qPCR positivity in OF was approximately three times lower among chikungunya patients in the group with IgM seroconversion alone compared to that with the virus confirmed through serum RT-qPCR. This finding is not surprising, as patients with detectable CHIKV RNA in serum are more likely to also have detectable CHIKV RNA in other biological samples than patients without CHIKV RNA in serum. Nonetheless, we did detect CHIKV RNA in OF from serum RT-qPCR-negative patients, though the observed frequency was too low to guide routine testing of both serum and OF to increase diagnostic capacity. Further studies with increased numbers of chikungunya patients should be carried out to better assess any potential gain from testing paired serum and OF during the investigation of CHIKV infection.
None of the OF or urine samples from the chikungunya patients confirmed only via CHIKV IgM in the acute-phase serum or from the control groups were RT-qPCR-positive. The failure to detect CHIKV RNA in OF and urine of patients with CHIKV infection confirmed only through the presence of CHIKV IgM in the acute-phase sample was in line with the low yield of RT-qPCR in OF and urine of patients who had a diagnosis of CHIKV infection solely based on IgM seroconversion. This finding indicates that the detection of CHIKV RNA in OF and urine is less likely when CHIKV RNA is not detected in serum. In addition, patients with negative RT-qPCR and positive IgM in the acute-phase serum are more likely to represent a group with longer disease duration compared to those who are RT-qPCR-positive or exhibit IgM seroconversion, as we observed in our study. This may also have hampered the detection of CHIKV RNA in OF and urine.
Regardless of the chikungunya group to which the study patients belonged, samples from all patients with a positive RT-qPCR result from serum, OF or urine were collected within the first five days after the onset of symptoms. We found that the measured Ct values in serum and OF tended to increase from day 0 to day 5 after symptom onset, though a trend could not be evaluated in the urine samples because only one urine sample was positive. This finding was expected in serum samples, as Ct values in positive samples are inversely correlated with the viremia level, and CHIKV viremia declines over time [19,33]. However, a cohort study found the persistence of CHIKV RNA in the serum and saliva of patients for up to 60 days and in the urine for up to 95 days after the onset of the disease [29], indicating that molecular diagnosis may be attempted in samples collected during the post-acute phase of the illness. Because we tested only acute-phase samples, it was not possible to investigate the frequency of RT-qPCR positivity after the first week of symptom onset. Nevertheless, the findings of both the previously mentioned cohort study and our study were similar regarding increasing Ct values over time and lower Ct values in the serum samples than in the OF and urine samples. Combined, these results indicate that the detection of CHIKV RNA is more likely to occur in serum than in OF and urine samples and in samples collected in the acute phase than in samples collected in the post-acute or chronic phase of the illness.
Several reports have shown that CHIKV RNA can be detected in body fluids other than serum, such as saliva [26,28,29], urine [24,28,29,34], sperm [24,29], vaginal secretions [29], placenta or amniotic fluid [35], breast milk [36], synovial liquid [37] and cerebrospinal fluid [38,39]. Infectious CHIKV has also been detected in the saliva of mice, monkeys and humans [27], raising concerns about the potential for non-vector-borne transmission [40]. However, that viable and replicating viruses can be identified in saliva does not mean that the amount present can mediate direct person-to-person viral transmission. Additional studies are needed to elucidate whether transmission of CHIKV through non-vector-borne routes is indeed possible and to inform whether the presence of CHIKV genetic material in these body fluids can serve as a marker of the risk of direct transmission.
Our study has both limitations and strengths. This study, which included samples of 51 patients with laboratory evidence of CHIKV infection, is one of the most extensive in comparing the ability of RT-qPCR to detect CHIKV RNA in paired serum, OF and urine. However, the number of cases per group based on the confirmation criteria was small. Nonetheless, identifying and collecting paired biological samples from patients with chikungunya in the first week of symptoms is challenging because it is difficult to predict when and where a CHIKV outbreak will occur. Therefore, obtaining paired acute-phase samples from 51 patients with confirmed chikungunya can be considered a strength of our study. Furthermore, although the OF and urine samples tested did not show visual signs of blood, we cannot rule out the possibility of blood contamination due to a mucosal lesion. Nevertheless, none of the patients with CHIKV RNA detected in the OF samples reported gingival bleeding, and only one reported having oral ulcers. The only patient who had a positive urine sample did not report hematuria. In an additional limitation, the frequency of detection of CHIKV RNA in the patients’ acute-phase urine and OF samples, as well as the Ct of detection, may have been influenced by the chosen RNA extraction and RT-qPCR methods or by storing the samples at −80 °C instead of using fresh samples. However, the implementation of the same protocols for RNA extraction and RT-qPCR in all the tested samples ensured the consistency of our comparative study. Finally, in terms of strengths, our study differs from the other two main studies that investigated the usefulness of a set of biological samples for diagnosing CHIKV infection via RT-PCR [28,29] because it is the first to show that OF can be RT-qPCR-positive in patients without positive RT-qPCR in serum (but with IgM seroconversion). It also differs from the previous studies by including two control groups, one comprising patients with RT-PCR-confirmed dengue and one with non-arbovirus acute febrile illness, thus supporting the high specificity of CHIKV qRT-PCR in non-serum samples.
In summary, our results confirm that serum is the best sample for RT-qPCR-based CHIKV diagnosis during acute disease, especially when collected in the first five days after the initial onset of symptoms. However, when serum cannot be obtained, or the laboratory detection of CHIKV is employed as part of surveillance efforts to monitor virus transmission trends among suspected patients, rather than for case diagnosis and management, testing OF may be attractive as a non-invasive alternative sample. However, while OF may prove helpful for surveillance or diagnosis in specific situations, a negative result should not be used to rule out a CHIKV infection. Our findings showed that the sensitivity of RT-qPCR performed in OF was about 50% that of the same assay performed in serum RT-qPCR. Nevertheless, because we found cases in which CHIKV RNA was detected in OF but not in serum, additional studies should be conducted to determine whether the parallel testing of serum and OF increases the capacity of case confirmation, to justify the routinization of the parallel testing of these two samples.
