Abstract / Summary
Abstract West Nile virus (WNV) diagnosis relies on serology and/or RNA detection in serum or cerebrospinal fluid (CSF). However, lumbar puncture may be contraindicated in some patients, and serological diagnosis can be limited by impaired antibody responses or persistent IgM positivity. We evaluated nasopharyngeal swabs (NS) as a minimally invasive biological matrix for Real-Time PCR detection of WNV RNA. In this retrospective, single-centre diagnostic accuracy study, reported according to the STARD 2015 statement, we analysed all patients with clinically suspected West Nile disease (WND) referred to the regional arbovirus reference laboratory (AOU Cagliari, Italy) between 2022 and 2025 who underwent both a nasopharyngeal swab (index test) and reference standard testing (n = 88). The reference standard for confirmed WND was detection of WNV RNA in serum and/or CSF and/or a WNV-specific antibody response. A pre-specified secondary analysis compared NS with RNA-confirmed WND. Diagnostic accuracy measures are reported with 95% confidence intervals (CI), and agreement was assessed using Cohen’s kappa. WND was confirmed in 53/88 patients (60.2%) and WNV RNA was detected in the NS of 25/88 (28.4%). Against the gold standard, NS showed a specificity of 97.1% (95% CI 85.5–99.5) and a positive predictive value of 96.0% (80.5–99.3), with a sensitivity of 45.3% (32.7–58.5), a negative predictive value of 54.0% (41.8–65.7), and a Cohen’s kappa of 0.374 (0.192–0.556). When the comparison was restricted to patients whose disease had been confirmed molecularly, sensitivity was 80.8% (62.1–91.5) and agreement was substantial (kappa 0.795, 0.639–0.951); only 3 of 27 patients confirmed by the antibody response alone had a positive NS. Sensitivity was almost identical in 2022–2024 and 2025 (44.4% vs 45.5%), whereas the prevalence of confirmed WND differed markedly (24.3% vs 86.3%; p < 0.001), which accounts for the lower negative predictive value during the outbreak period. Compared with the current gold standard for the laboratory confirmation of WND, NS WNV RNA detection is highly specific and has a high positive predictive value, but its sensitivity is limited and a negative result does not exclude disease. NS testing should therefore be regarded as an aid to diagnosis rather than as an alternative to serum and CSF testing. It may be particularly useful in selected circumstances in which the conventional approach is difficult to apply, such as when lumbar puncture is contraindicated by thrombocytopenia or coagulopathy, or when the antibody response is impaired.