key: cord-0946475-hwehf4mr authors: Fougere, Y.; Schwob, J. M.; Miauton, A.; Hoegger, F.; Opota, O.; Jaton, K.; Brouillet, R.; Greub, G.; Genton, B.; Gehri, M.; Taddeo, I.; D'Acremont, V.; Asner, S. A. title: Performance of RT-PCR on saliva specimens compared to nasopharyngeal swabs for the detection of SARS-CoV-2 in children: A prospective comparative clinical trial date: 2021-03-01 journal: nan DOI: 10.1101/2021.02.27.21252571 sha: cb40d99687c436f868a7f31155f146bfb1418a15 doc_id: 946475 cord_uid: hwehf4mr Background: Saliva RT-PCR is an attractive alternative for the detection of SARS-CoV-2 in adults with much less known in children. Methods: Children and adolescents with symptoms suggestive of COVID-19 were prospectively enrolled in a comparative clinical trial of saliva and nasopharyngeal (NP) RT-PCR between November and December 2020. Detection rates and sensitivities of saliva and NP RT-PCR were compared. Participants with discordant NP and saliva RT-PCR results including viral load (VL) were also analyzed. Result: Out of 405 patients enrolled, 397 patients had two tests performed. Mean age was 12.7 years (range 1.2-17.9). Detection rates were 22.9% (95%CI 18.8-27.1%) by saliva RT-PCR, 25.4% (21.2-29.7%) by NP RT-PCR, and 26.7% (22.4-31.1%) by any test. The sensitivity of saliva was 85.2% (78.2-92.1%) when using NP as the gold standard; in contrast, when saliva was considered the gold standard, the sensitivity of NP was 94.5% (89.8-99.2%).For a NP RT-PCR VL threshold of [≥]103 and [≥]104 copies/ml, sensitivity of saliva increases to 88.7% and 95.2% respectively. Sensitivity of saliva and NP swabs was respectively 89.5% and 95.3% in patient with symptoms less than 4 days (p=0.249) and 70.0% and 95.0% in those with symptoms [≥] 4 to 7 days (p=0.096). The 15 patients who had an isolated positive NP RT-PCR were significantly younger (p=0.034), had a lower NP VL (median 5.6x103 vs 3.9x107, p<0.001), and were not able to drool saliva at the end of the sampling (p=0.002). VLs were significantly lower with saliva PCR than with NP RT-PCR (median 8.7 cp/ml x104; IQR 1.2x104-5.2x105; vs median 4.0x107cp/ml; IQR 8.6x105-1.x108; p<0.001). Conclusion: Saliva PCR shows diagnostic performances close to NP RT-PCR for SARS-CoV2 detection in most symptomatic outpatient children and adolescents. statistically different in post-hoc analyses between the age groups of 0-6 years and 191 ≥ 6-12 years with a higher detection rate among children of ≥ 6-12 years of age (3.2% 192 vs 30.7 %, p=0.004) (Figure 1 ) 193 Diagnostic test performance (sensitivity, specificity) of NP RT-PCR and saliva Using NP as the gold standard, the sensitivity of saliva was 85.2% 92.1%); in contrast, when saliva was considered the gold standard, the sensitivity of 197 NP was 94.5% (95%CI 89.8-99.2%) (p=0.058). The sensitivity of saliva RT-PCR was 198 dependent on NP VLs and was maximal with a viral load of 10 6 cp/ml. (Table 2 and 199 age, 4 patients were detected positive from NP swabs, whereas only one child was 205 documented positive from saliva. As a result, the reported sensitivity was only 25% 206 for saliva PCR and 100% for NP swabs in this subgroup. The reported sensitivity was 207 significantly different between age groups for saliva samples (p=0.0001) but not for 208 NP swabs (p=0.320). Yet, in post-hoc analyses, sensitivity for saliva remained 209 statistically different only between the 0-6 year subgroup compared to the one 210 including children of 12 years of age and onwards (25% vs 89,9%, p=0.003). When 211 stratified by the duration of symptoms, the respective sensitivity of saliva and NP 212 swabs was 89.5% and 95.3% in patient with symptom duration < 4 days (95%CI -213 14.8 to 3.2%, p=0.249) and 70.0% and 95.0% in those with symptom duration 214 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 (95%CI -52.2 to 2.2%, p=0.096) ≥ 4 to 7 days. Only 3 patients had symptoms above 215 7 days and all were tested negative from both samples. 216 Figures 3A and 3B display the distribution of VLs and CT by analyzed specimen. VLs 218 documented from saliva were significantly lower compared to those reported from 219 paired NP swabs (median 8.7 cp/ml x10 4 ; IQR 1.2x10 4 -5.2x10 5 ; vs median 220 4.0x10 7 cp/ml; IQR 8.6x10 5 -1.x10 8 ; p<0.001, 95CI: -4.5x10 2 to -7.7x101). VLs 221 measured from saliva increased significantly with age but not with the duration of 222 symptoms ( Figure 4A /B). VLs measured from NP were not affected by age. (Figure 223 4A). 224 Table 3 displays the characteristics of the 15 children with NP swabs only positive 226 compared to the 86 documented positive from both NP swabs and saliva samples. 227 Children with isolated SARS-CoV-2 positive NP swabs were significantly younger 228 (p=0.034) and with significantly lower NP VLs (median 5.6x10 3 vs 3.9x10 7 , p<0.001) 229 compared to those with both saliva and NP positive specimens. In addition, they 230 were not able to drool at the end of saliva collection (p=0.002). Variables such as the 231 duration of symptoms or the person who performed the procedure did not affect the 232 above findings. The 5 patients who had isolated positive saliva samples were all 233 males, had a median age of 11.1 y (10.3-14.6), and presented with median VLs of 234 4.0x10 3 cp/ml (1.1x10 3 -8.8x10 3 ). 235 236 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 Discussion 237 The overall detection rate for SARS-CoV-2 between saliva and NP specimens by PCR was comparable. However, a significantly lower sensitivity was reported from 239 saliva specimens compared to NP swabs when compared to any positive test. Yet, 240 the sensitivity of saliva increased with VLs and was equivalent to NP swabs when 241 using a VL threshold of ≥ 10 4 copies/ml. The procedure of saliva collection and 242 younger age affected the detection of SARS-CoV-2 in saliva. In contrast, the duration 243 of symptoms over 7 days of onset did not significantly affect the sensitivity of SARS-244 CoV-2 detection in saliva. 245 With more than twice as many children included 8 , this study is by far, the largest 246 cohort reporting on the performance of saliva specimens in children. As supported by 247 another study including children 8 and other studies conducted in adults 10,19 , a 248 detection rates slightly lower of SARS-COV-2 were reported from saliva as compared 249 to NP specimens, yet still supporting the use of saliva as an alternative to NP swabs 250 in children. Moreover, our findings are in line with the RADICO study conducted in 251 symptomatic adults, that used the same saliva collection approach 10 . In contrast, 252 other studies indicated a poor concordance between NP and saliva specimens 9,20,21 , 253 likely as a result of using different sampling procedures that did not necessarily 254 include saliva drooling. In addition, the inclusion of adults and children with a variable 255 duration of symptoms upon testing and asymptomatic patients possibly limited their 256 The sensitivity of RT-PCR assays reported from saliva was lower compared to NP 258 swabs and the gold standard test, albeit only reaching statistical difference when 259 compared to any positive test. This difference in sensitivity is likely explained by a 2 260 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101/2021.02.27.21252571 doi: medRxiv preprint log lower VL detection in saliva compared to NP samples as evidenced elsewhere 261 9,10,22 . Furthermore, SARS-CoV-2 detection in saliva is dependent on viral loads and 262 reaches an equivalent sensitivity to NP swabs for NP VL thresholds of 10 4 copies/ml. 263 Recent studies 23-25 suggest that no cultivable viable virus is detected from patients 264 with VLs under the threshold of 10 4 copies/ml. Altogether ours and others findings 265 suggest 8 that children detected SARS-CoV-2 negative from saliva samples 266 presented VLs under the threshold of 10 4 copies/ml from their NP and were thereby 267 potentially less contagious. The use of saliva as an alternative to NP for SARS-CoV-268 2 detection in children would thus limit quarantine measures to most contagious 269 children 26 . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101/2021.02.27.21252571 doi: medRxiv preprint Furthermore, drooling into the tube also raises biosafety concerns. The edge of the 286 tube can be contaminated thereby questionning the use of precautionary measures. 287 Duration of symptoms over 7 days of onset did not significantly affect the sensitivity 288 of SARS-CoV-2 detection in saliva as already supported in other adult 19 and 289 paediatrics 9 studies. 290 Strengths of the current study include the largest pediatric sample size collected so 291 far in addition to the detailed prospectively collected information. Limitations are 292 predominantly related to the inclusion of outpatients and not hospitalized nor 293 asymptomatic children, which might affect the generalizability of our findings. 294 Furthermore, young children were under-represented and only a few of them were 295 detected SARS-CoV2 positive, thus limiting extrapolation to this age-group. In 296 addition, our study was conducted during a high prevalence of SARS-CoV-2 (up to 297 30%), thereby affecting our positive predicted values. Finally, our study was 298 conducted before the introduction of 501Y mutants in Switzerland, which currently 299 represent 8 to 15% of the analyzed samples. As such, we were not able to compare 300 the performance of mutant typing in saliva vs NP. Yet, recent evidence supports that 301 the lower VLs detected from saliva limit mutant typing analyses 27 . 302 In conclusion, saliva is a reliable alternative for SARS-CoV-2 detection among 304 symptomatic children. Saliva collection being a non-invasive easy procedure will 305 facilitate large-scale screening in children and thus provide more evidence on the 306 impact of SARS-CoV-2 in children. 307 308 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. and Anne Tabard-Fougère for her help in the statistical analyses and figure 317 elaboration. 318 None 320 336 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101/2021.02.27.21252571 doi: medRxiv preprint M a r i n h o L C N , S i l v a D N d e A , e t a l . S a l i v a a s a p o s s i b l e t o o l f o r t h e S A R S -380 C o V -2 d e t e c t i o n : A r e v i e w . T r a v e l M e d I n f e c t D i s . 2 0 2 0 ; 3 8 : 1 -1 2 . 381 2 1 . C z u m b e l L M , K i s s S , F a r k a s N , e t a l . S a l i v a a s a C a n d i d a t e f o r C O V I D -1 9 D i a g n o s t i c 382 T e s t i n g : A M e t a -A n a l y s i (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Viral loads (copies/ml) • Positive: NP+Saliva Positive: only NP Positive: only saliva slope=−0 Symptom duration (days) All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101/2021.02.27.21252571 doi: medRxiv preprint 416 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10.1101/2021.02.27.21252571 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 1, 2021. ; https://doi.org/10. 1101 /2021