Prospective evaluation of nonstructural 1 enzyme-linked immunosorbent assay and rapid immunochromatographic tests to detect dengue virus in patients with acute febrile illness

Abstract

We prospectively evaluated the Bio-Rad nonstructural 1 (NS1) enzyme-linked immunosorbent assay (ELISA) and lateral flow immunochromatographic assay (LFIA) in comparison to an in-place reverse transcription–polymerase chain reaction for dengue diagnosis. Among 537 consecutive samples from patients with acute febrile disease, 264 (49.2%) tested positive in reverse transcription–polymerase chain reaction (RT-PCR), 156 (29.1%) in NS1-antigen (Ag) ELISA, and 125 (23.3%) in NS1-Ag LFIA. Compared to the RT-PCR status, the specificity was 100% for the NS1-Ag ELISA and LFIA, but their respective sensitivities were 61.2% [95% confidence interval (CI), 55.2−67.2] and 49.4% (95% CI, 43.2−55.6), with nadirs of 37.9% and 24.1% on day 6 of illness. The NS1-Ag ELISA and LFIA were positive, respectively, for 48.0% and 40.7% of the secondary infections versus 85.0% and 66.7% of the primary infections. For patients <5 years old, NS1-Ag ELISA and LFIA reached respective sensitivities of 100% and 90.5%. Reports of results of dengue NS1-Ag assays should specify that negativity does not preclude DENV infection, and require further investigations in the case of severe disease.

Introduction

Dengue has become the most common and widespread arthropod-borne viral disease as a consequence of the expansion of its principal vector, the Aedes aegypti mosquito. The disease threatens more than 2.5 billion people living in tropical and subtropical regions (WHO, 2009). Dengue is caused by 4 serologically related, but antigenically and genetically distinct, viruses designated DENV-1 to DENV-4, belonging to the Flaviviridae family (Martina et al., 2009). Classic dengue fever and its more severe forms, dengue hemorrhagic fever (DHF) or dengue shock syndrome (DSS), are major public health problems in terms of morbidity and mortality (Gubler, 1998). Severe dengue diseases associated with nonhemorrhagic manifestations, such as hepatitis or encephalitis, have also been reported (Rigau-Perez, 2006). Because dengue manifestations are similar to those of many other febrile syndromes, the availability of dengue-specific laboratory tests is useful for the differential diagnosis. Timely and accurate diagnosis of dengue infection during early phase of the illness is important for appropriate management of complications, development of physiopathologic studies, and enrollment of patients in clinical trials. It also provides early warning of dengue outbreaks, epidemiologic investigations, and optimization of vector-control measures.

Current diagnostic methods, based on serologic testing, virus isolation, and genome detection, all have limitations (Teles et al., 2005). Immunoglobulin M (IgM) enzyme-linked immunosorbent assays (ELISAs) are used by most laboratories, but the IgM response is absent early during the disease course and may remain undetectable in secondary dengue infection. A definitive serologic diagnosis requires testing of acute- and convalescent-phase blood samples, usually collected at least 7 days apart, to demonstrate a 4-fold IgG titer rise. Virus isolation from acute-phase sera is the most conclusive approach to diagnosis, but is useless for patient management, because several days are needed to obtain final results. Molecular methods based on the reverse transcription–polymerase chain reaction (RT-PCR) yield same or next-day detection of DENV in acute-phase serum or plasma. The 2-step nested RT-PCR protocol, originally reported by Lanciotti et al. (1992), is used worldwide in laboratories contributing to dengue clinical management and/or surveillance. Conventional RT-PCR is being progressively replaced by real-time PCR (Johnson et al., 2005). However, molecular diagnoses still require high technical skill and sophisticated instrumentation not available in most diagnostic laboratories.

A seemingly attractive alternative is the detection of DENV nonstructural protein 1 (NS1). NS1 is a glycoprotein highly conserved by all dengue serotypes and is expressed in either membrane-associated or secreted forms (Falconar, 1997, Flamand et al., 1999). Soluble NS1 circulates in the sera of patients during the early stage of DENV infection (Alcon et al., 2002, Young et al., 2000). Commercial ELISA (Bessoff et al., 2008, Bessoff et al., 2010, Blacksell et al., 2008, Chuansumrit et al., 2008, Dussart et al., 2006, Dussart et al., 2008, Hang et al., 2009, Huhtamo et al., 2009, Kumarasamy et al., 2007, Lapphra et al., 2008, McBride, 2009, Phuong et al., 2009, Ramirez et al., 2009) and an immunochromatographic assay (Chaiyaratana et al., 2009, Dussart et al., 2008, Hang et al., 2009, Ramirez et al., 2009, Shu et al., 2009, Tricou et al., 2010, Zainah et al., 2009) for the detection of DENV NS1 antigen (Ag) on human sera have recently been evaluated. Most of the evaluations were based on the retrospective use of laboratory-confirmed dengue and nondengue reference collections (Bessoff et al., 2008, Dussart et al., 2006, Dussart et al., 2008) or clinical panels (Chaiyaratana et al., 2009, Chuansumrit et al., 2008, Huhtamo et al., 2009, Kumarasamy et al., 2007, McBride, 2009, Phuong et al., 2009, Ramirez et al., 2009, Zainah et al., 2009). Only few studies were prospective (Blacksell et al., 2008, Hang et al., 2009, Lapphra et al., 2008) and data on the use of dengue NS1-testing in clinical settings remain scarce.

We prospectively evaluated the ELISA and immunochromatographic NS1-Ag kits from Bio-Rad Laboratories (Marnes-la-Coquette, France), and compared them to an existing in-place RT-PCR used for routine diagnosis of DENV infection in patients with acute febrile disease. The study was conducted in the context of a dengue outbreak, and the results were analyzed according to clinical and biological data.