Performance of a Rapid Diagnostic Test for Influenza in a Tertiary Military Hospital, Philippines.

INTRODUCTION: It is important to evaluate the performance of existing rapid influenza diagnostic tests (RIDTs) and the factors that can affect performance especially when the circulation dynamics of influenza strains change such as the displacement and replacement of the circulating seasonal influenza strains. MATERIALS AND METHODS: Nasal swabs were collected from patients presenting at V Luna Medical Center, Armed Forces of the Philippines Health Service Command, with influenza-like illness (ILI) with one swab tested using Quickvue (QV) influenza A+B RIDT (Quidel) and the other swab tested using the ABI 7500 (Applied Biosystems) real-time reverse transcriptase-polymerase chain reaction. Sensitivity, specificity, positive predictive value, and negative predictive value were estimated. We identified clinical symptoms predictive of influenza subtype and evaluated the independence of QV sensitivity on (1) Cycle threshold (Ct) value, controlling for timing of collection; (2) timing of collection, controlling for Ct value; and (3) Ct value and timing of collection taken together. RESULTS: Between August 2011 and October 2016, patients presenting with ILI (n = 2333) underwent testing. Quickvue sensitivity across all subtypes was significantly correlated with lower Ct values (higher virus titers) (P <.001) and, except for flu A/H3 (P = .974), was also significantly associated with timing of specimen collection (P <.05). No statistically significant difference was noted in QV sensitivity for Flu A/H3 (P = .130), pandemic H1/N1 (P = .207), Flu A/H3 + pandemic H1/N1 (P = .341), and Flu B (P = .103) across different age groups but sensitivity of QV significantly differed (P <.001) across the different influenza subtypes. CONCLUSION: Overall specificity of QV was high across all flu subtypes, but overall sensitivity was low (Flu A/pdm H1) to moderate (Flu A/H3 and Flu B). The findings highlight the need to develop more sensitive influenza RDTs to detect circulating influenza strains and the use of the quadrivalent flu vaccine during the annual influenza vaccination.

A multi-country field validation of the FluChip-8G Insight Assay.

INTRODUCTION: It is critical to rapidly detect novel and non-seasonal influenza strains. Currently available assays have limited sensitivity in detecting novel influenza subtypes. We performed a multi-country field validation of the FluChip-8G Insight, an assay able to detect and characterize influenza A/B viruses and non-seasonal influenza viruses. MATERIALS AND METHODS: We evaluated the performance of the FluChip-8G Insight on nasal and throat swab clinical samples from Thailand, Philippines and Nepal. Influenza PCR positive and negative samples tested using the US CDC Human Influenza Dx Panel reference standard were selected for testing using the FluChip-8G Influenza Insight. RESULTS: A total of 909 specimens were included in the analysis. The overall sensitivity and specificity of the FluChip-8G Insight to detect combined influenza A+B was 86 % and 100%, respectively. PPV and NPV were estimated at 100 % (95 % CI 99-100) and 73 % (95 % CI 68-78), respectively. Sensitivity across all influenza subtypes was 100% for specimens with <20 and 20-25 Ct values, respectively, but as Ct values increased, sensitivity across all influenza subtypes decreased significantly (p < 0.001) for specimens with Ct values ≥32. CONCLUSION: The FluChip-8G Insight showed good precision and reproducibility among all 3 sites with robust identification of both influenza A and B targets with Ct values <32 and in the absence of co-infection. Positioning this platform in countries considered as hotspots for the emergence of novel/zoonotic influenza strains can increase the lead time in detecting and containing novel influenza strains with pandemic potential.

Multidrug Resistant Mycobacterium tuberculosis Among Military and Civilian Personnel seen at a Tertiary Military Hospital, Manila, Philippines (2015-2018).

INTRODUCTION: About one third of the world population is estimated to be infected with Mycobacterium tuberculosis (MTB), and this proportion is expected to be higher in countries with a high tuberculosis (TB) burden. The Philippines is both a high tuberculosis burden and a high multidrug resistant tuberculosis (MDR-TB) burden country. Though TB has been extensively described in the civilian population, there is limited data on TB in the military population. The objectives are: (1) To determine MTB/MDR-TB prevalence among military and civilian patients in the Philippines presenting with clinically suspected TB in a tertiary military hospital and (2) To determine performance of direct sputum smear microscopy (DSSM) using Ziehl-Neelsen (ZN) staining compared to Xpert MTB/RIF real-time reverse transcriptase polymerase chain reaction. MATERIALS AND METHODS: Sputum samples were obtained from patients, clinically suspected with TB, and/or with TB associated signs/symptoms. Sputum specimens were tested using DSSM with ZN staining and Xpert MTB/RIF assay (Cepheid, Sunnyvale, California) and patient demographic and clinical data were collected. RESULTS: From March 2015 to December 2018, a total of 795 (173 military personnel [164 active duty and 9 retired]; 618 civilians; and 4 with no data on military/civilian status) patients with TB associated symptoms or clinically suspected with TB were tested. Overall, MTB prevalence was 81/795 (10%). MTB prevalence among active duty and retired military personnel were 27/164 (16%) and 4/9 (44%), respectively while MTB prevalence for civilian patients was 50/618 (8%) (p value = 0.0003; OR = 2.48 [95% C.I. 1.5-4]). Among active and retired military personnel who tested positive for MTB, rifampin resistance was 4/27 (15%) and 1/4 (25%), respectively, while rifampin resistance for civilian patients was 9/50 (18%) (p value = 1; OR = 0.88 [95% C.I. 0.26-2.90]). For active duty military personnel, average MTB prevalence (based on Xpert MTB/RIF) covering years 2015-2018 was 21% and ranged from 13% to 35%, while average rifampin resistance among MTB positive active duty military personnel was 15% and ranged from 0% to 25%. Overall sensitivity and specificity of DSSM compared to Xpert MTB/RIF were 70% and 96%, respectively. Positive and negative predictive values of DSSM to accurately categorize MTB in symptomatic cases (with Xpert MTB/RIF as "true positive" reference) were 74% and 95%, respectively. Performance of DSSM varied according to MTB load detected by Xpert MTB/RIF with increasing DSSM sensitivity observed as the MTB load detected by Xpert MTB/RIF increased (p = 0.02). CONCLUSION: This report describes high MTB and MDR-TB prevalence rates among symptomatic military patients with military personnel having higher odds of MTB infection compared to the civilian patients in the study. Since DSSM (ZN) sensitivity greatly varied depending on MTB load, the Xpert MTB/RIF should be used as a first-line diagnostic tool to identify MTB and detect rifampin resistance among presumptive TB cases instead of DSSM (ZN) microscopy.

Evaluation of QuickVue influenza A+B rapid test for detection of pandemic influenza A/H1N1 2009.

BACKGROUND: A novel influenza A/H1N1 emerged in early 2009 and by June 2009 was declared a pandemic by WHO. Rapid influenza antigen detection tests have been used to diagnose seasonal influenza but have not been adequately evaluated for the pandemic strain among all age groups. In the Philippines, pandemic influenza A/H1N1 2009 was first detected in May 2009 and by July 2009, 3207 cases and 6 deaths were reported. OBJECTIVES: Using RT-PCR as the gold standard, clinical sensitivity/specificity of Quidel QuickVue (QV) influenza A+B was estimated across all age groups for pandemic influenza A/H1N1 using nasal swabs in a hospital setting. Effect of age, viral titers (Ct values), and timing of collection on QV sensitivity to detect pandemic influenza A/H1N1 2009 was also determined. STUDY DESIGN: Febrile patients with influenza-like illness (n=360) at the V. Luna General Hospital, Manila from 1 June to 31 August 2009 were included. Nasal swabs were tested using QV and RT-PCR. RESULTS: Of 360 nasal specimens 226 (63%) were positive for pandemic influenza A/H1N1. QV sensitivity was 63% (95% confidence interval (CI): 56-69%), specificity was 96% (95% CI: 91-99%), positive predictive value was 97% (CI: 93-99%), and negative predictive value was 57% (95% CI: 49-64%). Patient's age, fever severity, presenting symptoms or number of symptoms did not significantly affect QV sensitivity, however QV sensitivity was correlated with decreasing Ct values. CONCLUSION: QuickVue demonstrated moderate sensitivity for pandemic influenza A/H1N1 infection. There was a significant inverse association between Ct values and QV sensitivity for pandemic influenza A/H1N1.