Related Experiment Video
Updated: Jun 6, 2026
Swabbing the Urban Environment - A Pipeline for Sampling and Detection of SARS-CoV-2 From Environmental Reservoirs
Published on: April 9, 2021
Sensitivity of gargle samples compared to swabs for SARS-CoV-2 detection with nucleic acid amplification testing: a
Iryna Kayda1, Kwana Lechiile2, Mai-Lei Woo Kinshella3,4
1School of Medicine, University of Western Australia, Perth, Australia.
Background:
Gargle sampling emerged as a novel method for diagnostic testing during the coronavirus disease 2019 (COVID-19) pandemic, yet uncertainty remained about its performance when compared with conventional sampling methods.
Objective:
To evaluate the performance of self-collected gargle samples compared to traditional healthcare worker (HCW)-collected upper respiratory tract swabs for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection with nucleic acid amplification testing (NAAT).
Methods:
We conducted a systematic review (PROSPERO registration: CRD42022312628) to (1) estimate sensitivity of gargle sampling, (2) estimate the difference in sensitivity between gargle and swab methods, and (3) understand how various testing contexts may impact gargle sensitivity. MEDLINE, EMBASE, Web of Science, Global Index Medicus, and preprint servers were searched. Studies reporting primary data and investigating COVID-19 diagnostic performance of self-collected gargle samples compared to HCW-collected swabs tested using NAAT were included. Quality assessment was performed, and random effects meta-analysis was conducted to estimate the pooled gargle sensitivity and mean difference in sensitivity between gargle and swab methods.
Results:
Searches identified 1453 results with 32 studies included. Meta-analysis pooled 34 gargle-swab comparisons. Gargle sensitivity was estimated to be 92.2% (95% confidence interval: 89.6% to 94.2%), and 3.3% (0.4% to 6.3%) less sensitive than swab collection. Gargle sensitivity was greater than 87.0% across diverse patient characteristics, settings, type or volume of gargle liquid, length of gargling time, wait time prior to gargling, and reference swab type used. Greatest sensitivities were observed when gargle sampling for 30 s or greater using 5-9 mL of saline. Gargle sensitivity of 93.0% (87.8% to 96.1%) was observed when there was no required wait time, and 90.1% (87.0% to 92.5%) when compared to combined nasopharyngeal and oropharyngeal swabs.
Conclusion:
Gargle sampling may be a sensitive, alternative method for SARS-CoV-2 detection across various testing contexts, and its implementation has potential to reduce barriers associated with HCW-collected swabs, that may be challenging to collect.
