Grijalva Carlos G, Rolfes Melissa, Zhu Yuwei, Chappell James, Halasa Natasha, Kim Ahra, Reed Carrie, Fry Alicia, Talbot H Keipp
Vanderbilt University Medical Center, Nashville, Tennessee, USA.
Centers for Disease Control and Prevention, Atlanta, Georgia, USA.
Open Forum Infect Dis. 2021 Sep 25;8(11):ofab484. doi: 10.1093/ofid/ofab484. eCollection 2021 Nov.
Anterior nasal swabs (ANS) are established specimen collection methods for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection detection. While saliva (SA) specimens provide an alternative, few studies have comprehensively characterized the performance of SA specimens in longitudinal studies.
We compared SARS-CoV-2 detections between paired self-collected ANS and SA specimens from a household transmission study. Participants recorded symptoms and paired ANS and SA specimens daily for 14 days. Specimens were tested using RT-PCR. We calculated the proportion of detections identified by each specimen type among the detections from both types combined. We computed percent agreement and Kappa statistics to assess concordance in detections. We also computed estimates stratified by presence of symptoms and examined the influence of traditional and inactivating transport media on the performance of ANS.
We examined 2535 self-collected paired specimens from 216 participants. Among 1238 (49%) paired specimens with detections by either specimen type, ANS identified 77.1% (954; 95% CI, 74.6% to 79.3%) and SA 81.9% (1014; 95% CI, 79.7% to 84.0%), with a difference of 4.9% (95% CI, 1.4% to 8.5%). Overall agreement was 80.0%, and Kappa was 0.6 (95% CI, 0.5 to 0.6). Nevertheless, the difference in the proportion of detections identified by ANS and SA using traditional and inactivating transport media was 32.5% (95% CI, 26.8% to 38.0%) and -9.5% (95% CI, -13.7% to -5.2%), respectively. Among participants who remained asymptomatic, the difference in detections between SA and ANS was 51.2% (95% CI, 31.8% to 66.0%) and 26.1% (95% CI, 0% to 48.5%) using traditional and inactivating media, respectively.
Self-collected saliva specimens provide a noninvasive alternative to nasal swabs, especially to those collected in traditional transport media, for longitudinal field studies that aim to detect both symptomatic and asymptomatic SARS-CoV-2 infections.
前鼻拭子（ANS）是用于检测严重急性呼吸综合征冠状病毒2（SARS-CoV-2）感染的既定标本采集方法。虽然唾液（SA）标本提供了一种替代方法，但很少有研究在纵向研究中全面描述SA标本的性能。
我们比较了来自一项家庭传播研究的配对的自行采集的ANS和SA标本中SARS-CoV-2的检测情况。参与者连续14天每天记录症状以及配对的ANS和SA标本。使用逆转录聚合酶链反应（RT-PCR）对标本进行检测。我们计算了在两种类型标本合并检测出的阳性结果中，每种标本类型所检测出的阳性结果的比例。我们计算了一致性百分比和Kappa统计量以评估检测结果的一致性。我们还计算了按症状存在情况分层的估计值，并研究了传统运输介质和灭活运输介质对ANS性能的影响。
我们检查了来自216名参与者的2535对自行采集的配对标本。在1238对（49%）通过任一种标本类型检测出阳性的配对标本中，ANS检测出77.1%（954例；95%置信区间，74.6%至79.3%），SA检测出81.9%（1014例；95%置信区间，79.7%至84.0%），差异为4.9%（95%置信区间，1.4%至8.5%）。总体一致性为80.0%，Kappa值为0.6（95%置信区间，0.5至0.6）。然而，使用传统运输介质和灭活运输介质时，ANS和SA检测出的阳性结果比例差异分别为32.5%（95%置信区间，26.8%至38.0%）和 -9.5%（95%置信区间，-13.7%至 -5.2%）。在无症状的参与者中，使用传统运输介质和灭活运输介质时，SA和ANS检测出的阳性结果差异分别为51.2%（95%置信区间，31.8%至66.0%）和26.1%（95%置信区间，0%至48.5%）。
对于旨在检测有症状和无症状SARS-CoV-2感染的纵向现场研究，自行采集的唾液标本为鼻拭子提供了一种非侵入性替代方法，特别是对于那些在传统运输介质中采集的鼻拭子。
