Why is dengue rapid test Better?

INTRODUCTION

Globally, over 2.5 billion people are estimated to be at risk of dengue virus (DENV) infection with an estimated 96 million symptomatic cases of dengue occurring annually (1). Despite ongoing research efforts, there are no sustainable vector control approaches or effective antiviral drugs to prevent or treat dengue, and vaccines have only recently been licensed by few countries and with suboptimal efficacy (2, 3, 4, 5). However, improvements in patient clinical management has been shown to reduce mortality among patients with severe dengue from 5% to <0.5% (6'9).

Accu-Tell supply professional and honest service.

Delayed dengue case identification often occurs in areas with limited diagnostics or surveillance resources, especially where dengue outbreaks are episodic or have been under-recognized (e.g., Africa and Oceania) (10, 11). These factors decrease health care provider awareness to include dengue in the differential diagnosis with other acute febrile illnesses (AFIs) such as malaria, leptospirosis, and influenza (12). Lastly, minimal or no laboratory infrastructure to conduct standard dengue diagnostic assays or perform them in a timely manner limits their utility for case management (13).

Laboratory diagnosis of dengue can be achieved with a single serum specimen obtained during the febrile phase of the illness by testing for DENV analytes (e.g., nucleic acid, nonstructural protein 1 [NS1], and anti-DENV IgM) (14). DENV viremia occurs for up to 7 days after the onset of fever, and anti-DENV IgM begins to appear around 3 days after fever onset (15, 16). Although detection of DENV nucleic acid by real-time reverse transcriptase PCR (rRT-PCR) is the most sensitive and specific means to detect DENV viremia (17), immunoassays to detect DENV NS1 antigen provide acceptable levels of detection sensitivity and specificity (18, 19). Immunoassays with good sensitivities and specificities to detect anti-DENV IgM are also widely available (20). However, both of these diagnostic approaches are instrument dependent and require facilities capable of performing complex diagnostic tests.

The availability of dengue rapid diagnostic tests (RDTs) has the potential to change the current situation in resource-limited areas and improve dengue clinical management. We evaluated an RDT that detected both DENV NS1 antigen and anti-DENV IgM for its ability to provide accurate information for detecting dengue from outbreaks as the main cause of febrile illness in areas without ongoing laboratory testing.

RESULTS

Patient demographics and overall RDT performance.

The mean age of suspected dengue patients included in this study by country was between 15 and 28 years old (range, 0 to 99 years). There was relatively equal proportions of male and female patients (Table 1). The overall sensitivity of the RDT NS1 test for all four sites was 65.9% (CI95 = 62.2 to 69.4), with a specificity of 80.9% (CI95 = 77.8 to 83.8) and an accuracy of 73.4%. The overall sensitivity of the anti-DENV IgM RDT was 69.9% (CI95 = 62.0 to 77.0), with a specificity of 76.9% (CI95 = 74.5 to 79.1) and an accuracy of 76.2%.

RDT sensitivity varied by DENV serotype and RDT analyte.

RDT NS1 sensitivity was highest 92.9% (CI95 = 76.5 to 99.1) for specimens collected during the DENV-1 Angola outbreak, although this estimate was relatively unstable with wide confidence intervals due to the small sample size (n = 46). DENV-1 had the lowest specificity (22.2%, CI95 = 6.4 to 47.6), and the accuracy was 65.2% (Fig. 1; see also Table S1 in the supplemental material). The next highest NS1sensitivity was 84.4% (CI95 = 75.3 to 91.2) for the DENV-3 Fiji specimens (n = 302), with a specificity of 78.2% (CI95 = 71.8 to 83.7) and an accuracy of 80.1% (Fig. 1; see also Table S5 in the supplemental material). A lower sensitivity was observed for NS1 for the DENV-4 RMI specimens (n = 796), with a sensitivity of 66.8% (CI95 = 61.9 to 71.3), a specificity of 79.9% (CI95 = 74.3 to 84.7), and an accuracy of 71.8% (Fig. 1; see also Table S2 in the supplemental material). The lowest sensitivity for NS1 was DENV-2 Yap specimens (n = 534), with a sensitivity of 49.7% (CI95 = 42 to 57.4), a specificity of 89.0% (CI95 = 84.2 to 92.7), and an accuracy of 72.1% (Fig. 1; see also Table S6 in the supplemental material).

The sensitivity of RDT anti-DENV IgM was highest for the DENV-1 Angola outbreak specimens at 91.7% (CI95 = 61.5 to 99.8), with a specificity of 85.3% (CI95 = 68.9 to 95.1) and an accuracy of 87.2% (Fig. 1A-B, Table S1 in the supplemental material). The next highest sensitivity was for DENV-4 RMI specimens at 80.0% (CI95 = 61.4 to 92.3), with a specificity of 92.2% (CI95 = 88.9 to 94.8) and an accuracy of 91.3% (Fig. 1; see also Table S2 in the supplemental material). DENV-2 Yap specimens had a sensitivity of 56.6% (CI95 = 42.3 to 70.2), with a specificity of 93.1% (CI95 = 91.4 to 95.9) and an accuracy of 90.2% (Fig. 1; see also Table S6 in the supplemental material). The lowest sensitivity was observed for the DENV-3 Fiji specimens at 55.3% (CI95 = 38.3 to 71.4), with a specificity of 98.5% (CI95 = 96.2 to 99.6) and an accuracy of 93.2% (Fig. 1; see also Table S5 in the supplemental material).

Combination test results for RDT NS1 and anti-DENV IgM.

The combination test result from either NS1 and/or the anti-DENV IgM was assessed with respect to classification of suspected dengue cases as laboratory positive. In this analysis we assumed all anti-DENV IgMs were due to the DENV serotype circulating for each outbreak. To calculate the additive effect of the two analytes (anti-DENV IgM and NS1), any positive by anti-DENV IgM and/or NS1 from the RDT that correlated to a positive by anti-DENV IgM in the ELISA and/or positive by rRT-PCR was considered a positive result. Conversely, a negative specimen was negative by both rRT-PCR and anti-DENV IgM. For DENV-1 specimens there was a minimal increase in sensitivity to 94.6% (CI95 = 81.8 to 99.3) compared to 91.7% for NS1 and 92.9% for anti-DENV IgM alone (Fig. 1). The combined sensitivity for NS1 among the DENV-2 patients was similar (56.4%, CI95 = 49.3 to 63.3) to that of the anti-DENV IgM analyte alone (56.6%) since most the NS1 positive specimens were also positive for anti-DENV IgM. For DENV-3, the combination of tests yielded a sensitivity of 85.5% (CI95 = 78.3 to 91) compared to NS1 alone at 84.4% or anti-DENV IgM alone at 55.3%. The DENV3 sensitivity was much lower for anti-DENV IgM alone due to the high number of false-negative results. Lastly, for DENV-4, the combination of tests yielded a sensitivity of 66.1% (CI95 = 61.7 to 70.3), which was similar to NS1 alone (66.8%). The DENV-4 serotype had higher sensitivity for anti-DENV IgM (80.0%). Since there were only 24 true positives in the anti-DENV IgM analysis for DENV-4, the addition of IgM for the combination of tests had limited influence in the test sensitivity. Overall, the difference in sensitivity of the combination of tests compared to NS1 alone was not significantly different (P > 0.05) for all DENV serotypes.

RDT sensitivity and timing of specimen collection.

RDT sensitivity by DPO of specimen collection could only be analyzed for RMI and Yap specimens due to limited information from the other countries. For NS1, the lowest sensitivity was observed on day 0 (RMI = 53.5%) and day 5 (RMI = 50.0%; Yap = 42.0%) (Fig. 2; see also Tables S3 and S7 in the supplemental material). The sensitivity for RMI specimens increased incrementally from days 1 to 3 and then decreased from days 4 to 5; however, the sensitivity of the Yap specimens had a range from 42.0 to 53.0 throughout the first 5 days (see Table S7 in the supplemental material). Conversely, in RMI the lowest anti-DENV IgM was on days 0 and 1 (0 and 28.6%, respectively), and the highest sensitivity occurred on day 5 (83.3%) (see Table S4 in the supplemental material). For specimens collected after day 5, there was also a decrease in sensitivity (47.1%), which is attributed to the wide range of DPOs (DPO 6 to 45) for this data point. For Yap, there were insufficient anti-DENV IgM-positive specimens for analysis.

The company is the world’s best dengue rapid test supplier. We are your one-stop shop for all needs. Our staff are highly-specialized and will help you find the product you need.

The anti-DENV IgG titer is negatively correlated with RDT NS1 positivity.

Finally, we measured the correlation between anti-DENV IgG titers and the proportion positivity of NS1 in the true positive serum specimens as determined by rRT-PCR. Of the 147 specimens tested by anti-DENV IgG ELISA, 136 were determined to be positive by anti-DENV IgG, indicating a secondary infection status, and 11 were determined to be negative for anti-DENV IgG, indicating a primary infection status. The results indicated that increasing anti-DENV IgG titer was negatively correlated with the detection of NS1 (Fig. 3).

DISCUSSION

This study examined the utility of an RDT that has been previously evaluated analytically (24'27) to identify and monitor suspected dengue epidemics in areas without routine diagnostics and/or established dengue surveillance. Although this RDT has been previously used for outbreak response, its performance in this context has never been evaluated. We were able to define the analytical characteristics of the RDT because all specimens were subsequently tested by reference diagnostic tests. Although the RDT only had intermediate sensitivity for NS1 (65.9%) and generally high specificity (80.9%) compared to reference tests, its widespread use among persons presenting with AFI correctly identified these outbreak as being caused primarily by dengue. The identification of a dengue outbreak triggered a public health response focused on implementation of timely clinical case management; the only effective dengue prevention tool available today (4'7).

The results of this study indicated that analyte specific sensitivity varied by DENV serotype, as well as the timing of specimen collection relative to illness onset. Anti-DENV IgM was most sensitive at 5 days after onset of illness, whereas NS1 was most sensitive at 3 days. An unexpected result in this study was the low specificity (22.2%) of DENV1 observed in Angola specimens. We originally attributed this to the small sample size; however, other variables could have contributed to this finding, including operator error or degradation of DENV RNA after the transport of specimens from Angola to San Juan, Puerto Rico. Although it was anticipated that the additive effect of both NS1 and IgM test results should have improved diagnostic accuracy, this effect was minimal compared to a single test result. This was probably due to the time of illness when the patient was tested with respect to illness onset, which in each of the respective outbreaks occurred within the peak sensitivity of one of the two analytes.

Most studies of dengue RDTs have been retrospective and used archived specimens to determine test sensitivity and specificity (24'26, 28). In those studies, the overall sensitivity varied from 48.5 to 72.4%, and the specificity varied from 88.8 to 100%. In these studies, the effect of combining the analyte results improved overall sensitivity from 48.5% to 92.9% (25). This differed from our study, where we observed a modest increase in sensitivity by combining RDT results, which may be due to the methods used to confirm a dengue case. Specifically, we compared RDT results to reference testing for respective analytes, whereas previous studies compared RDT results to confirmed dengue cases that were defined as anti-DENV IgM seroconversion in paired specimens.

There have been multiple prospective studies of this RDT with varied sensitivities (29'31). These studies differed from our prospective study because specimens were obtained primarily from hospitalized suspected dengue cases with a relatively small sample size. Andries et al. observed a sensitivity of 58.4% and a specificity of 98.3% from a clinical cohort of 57 hospitalized pediatric dengue patients with DENV-1 infection in Cambodia (30). A similar prospective study in Singapore of 46 primarily adult inpatients reported an observed sensitivity for DENV-2 of 81.6% and a specificity of 98% (31). Our finding that test performance varied by DENV serotype was similar to what has been observed previously in retrospective evaluations of both NS1 and IgM RDT's. However, in our study we observed a low rate of NS1 detection in DENV-2 samples, suggesting a lower binding affinity of the monoclonal antibody used for NS1 antigen capture. Similarly, differences in anti-DENV IgM by serotype also implies serotype-specific differences in viral antigen binding affinities. Moreover, the presence of anti-DENV IgG altered the performance of the NS1 test, which is presumed due to immune complexes that reduce NS1 availability generally observed in secondary DENV infections (32). Many of the commercially available NS1 RDTs have been known to have reduced sensitivity in secondary DENV infections compared to primary DENV infections (24).

This study had several limitations. First, not all sites received formal training and, although some sites received guidance materials in the form of pocket guides and posters, differences of sensitivity and/or specificity could be influenced by differences in the proficiency of the staff performing the test. Furthermore, the same individual was the tester and reader, with no additional readers for the test. Despite this nonstandardization, we believe that these data reflect a real-world experience in resource-limited settings. Second, since this study was conducted during outbreak response settings, there were unequal sample sizes by DENV serotype especially for Angola specimens where there was limited test availability. Also, we observed an unusually low specificity for the Angola site (22%) attributed to reading the test at the appropriate time. Third, this study did not address the issues of antibody cross-reactivity generally observed among flavivirus, although all outbreaks were confirmed to be primarily due to DENV. Lastly, not all sites collected DPO or disease severity information which limited the final analysis, including that the RDT could not be evaluated by dengue disease severity or primary and secondary DENV infection status.

Most RDTs are prone to operator error, an important factor that contributes to variance in test results for both sensitivity and specificity (20). This RDT test is dependent on visual assessment of the result and hence inherently subjective. Operator error is especially problematic if there are poor lighting conditions or if the test is read before or after the time specified in the package insert, resulting in incorrect results. This study had multiple operators in multiple countries and, although we aimed to minimize operator error by training laboratory technicians and providing visual guidelines on test interpretation in the form of pocket guides or posters, we could not provide ongoing monitoring of quality assurance. Hence, these errors contribute to low sensitivity and especially low specificity because reading the test after the designated time period can result in false-positive results.

Dengue RDTs are an important diagnostic tool and are valuable for surveillance in resource-limited regions since they require fewer laboratory infrastructure requirements. RDTs also allow for improved surveillance by decreasing specimen testing turnaround times for rapid identification of DENV transmission and outbreaks. It also enabled public health officials to identify risk factors for DENV infection, design appropriate control measures, and initiate activities to improve clinical case management. In order to better understand the utility of this RDT in nonoutbreak conditions, we believe that prospective, large-scale clinical studies of dengue RDT's with a minimum performance profile of the one used in this study should be conducted to determine their utility in dengue case identification and management in clinic settings, since these results were not captured in this study. This study design determined that the RDTs were a rapid method to confirming a dengue case in resource limited regions and allowed for a rapid, more-focused outbreak response, including prevention methods such as community outreach, mosquito prevention/control, and clinician awareness. Nevertheless, there is a need to improve RDT performance in order to increase test sensitivity and achieve equal sensitivity across all DENV serotypes.

Availability

There is one dengue NS1 test cleared by the U.S. Food and Drug Administration. There are other NS1 tests available as commercial diagnostic kits. Some public health and commercial laboratories use these tests.

Want more information on Dengue Combo Rapid Test Cassette? Feel free to contact us.