key: cord-1040459-yulwo7ux authors: Denis, F.; Septans, A.-L.; Periers, L.; Maillard, J.-M.; Legoff, F.; Gurden, H.; Moriniere, S. title: Olfactory training and visual stimulation assisted by web-application in patients with persistent olfactory dysfunction after SARS-CoV-2 infection: observational study. date: 2021-05-17 journal: nan DOI: 10.1101/2021.05.13.21257176 sha: e1311ab1f033d531fffd05c584427d27d7949334 doc_id: 1040459 cord_uid: yulwo7ux Background: Persistent olfactory dysfunction (OD) is a significant complication of SARS-CoV-2 infection. Olfactory training (OT) using aromatic oils are recommended to improve olfactory recovery, but quantitative data are missing. Objective: We aimed to quantify the benefit of OT associated with visual stimulation assisted by a dedicated web-application on patients with 1-month or more OD. Methods: We performed an observational real-life data-based study on a cohort of patients with at least 1-month persistent OD included between 1/30/21 and 3/26/2021. Analysis was performed after a 4-weeks mean time of OT and at least 500 patients assessable for primary outcome. Participants exposed themselves twice daily to odors from four high concentration oils and visual stimulation assisted by a dedicated web-application. Improvement was defined as a 2/10 points increase on self-assessed olfactory visual analogue scale. Results: 548 were assessable for primary outcome assessment. The mean baseline self-assessed olfactory score was 1.9/10 (SD 1.7) and increase to 4.6 (SD 2.8) beyond a mean time of olfactory training of 27.7 days (SD 17.2). Olfactory training was associated with at least 2-points increase in 64.2% (n=352). The rate of patients with improvement was higher in patients having trained for more than 28 days versus patients having trained for less than 28 days (72.2% vs 59.0% respectively, p=.002). The kinetic of improvement was 8 days faster in hyposmic than in anosmic patients (p<.001). The benefit was observed regardless of the duration of the OD. Conclusions: OT associated with visual stimulation assisted by a dedicated web-application was associated with significant improvement in olfaction, especially if OT duration was superior to 28 days. Anosmia is a frequent symptom of SARS-CoV-2 infection and its duration is usually less than two weeks before recovery. [1] [2] [3] However, at least 10% of patients will have persistent and chronic olfactory dysfunction such as diminished smell (hyposmia) or loss of smell (anosmia) which was shown to lead to decrease quality of life, depressive symptoms and nutrition issues. [4] [5] [6] One treatment option which is recommended against persistent olfactory dysfunction is the daily olfactory training using high concentration aromatic oils. [7] It showed significant results in postinfectious olfactory loss in a randomized controlled multicenter study. [8] In this trial, after 18 weeks olfactory training, function improved in 63% of patients having a duration of olfactory dysfunction of less than 12 months olfactory dysfunction using high concentration oils versus 19% in the control group using low concentration oils. Moreover, combination of visual stimulations to olfactory training may improve recovery results. [9] No data about olfactory training in persistent olfactory dysfunction are available in SARS-CoV-2 patients with persistent olfactory dysfunction, but most of patients having 30-days or more hyposmia or anosmia seems to have a low rate of spontaneous recovery. [4] In order to quantitatively study the time course of olfactory scores during olfactory training in real-life, we developed a web-application dedicated to olfactory training and visual stimulations as well as self-assessment and follow-up of olfactory scores. We assessed results in a real-life observational study. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The users were recruited via national media campaign in France including social media, radio and magazines between 1/30/2021 and 2/15/2021. This observational data-based study was approved by the French National Health Data Institute which reviews ethical conduct of human subject's research, data confidentiality, and safety. To participate, individuals were required to connect to the free covidanosmia.eu web-application and give electronic agreement. Respondents self-entered anonymously sociodemographic data, RT-PCR test results and diagnosis of SARS-CoV-2 olfactory dysfunctions. Patients were also asked to complete items about their co-morbidities, duration of olfactory symptoms and selfassessed intensity of olfactory dysfunction using subjective ratings with a visual analogue scale (0 no smell) to 10 (no smell alteration). [10] Patients were retained in the study analysis if they had a SARS-CoV-2 olfactory dysfunction, persistent from at least 1 month, with a reporting of at least 7 days of olfactory training and if their last olfactory function assessment on webapplication diary was available. Exclusion criteria were normosmia (visual scale score >7/10), other causes of olfactory dysfunction such as chronic rhinosinusitis, nasal polyposis, allergic or idiopathic rhinitis, post-traumatic olfactory loss, or other acute or chronic nasal diseases (e.g., acute viral infections), malignant tumors or/and oncology therapies (radiation, chemotherapy), and history of surgery on the nose or paranasal sinuses. Then, patients had to obtain the olfactory training kit from the application or from their pharmacist. Olfactory training was performed over a maximum period of 16 weeks. The webapplication provides videos, tutorials for the training as well as periodic encouragements. Participants exposed themselves twice daily to odors from four high concentration oils: phenyl ethyl alcohol: rose odor from geranium rosa, eucalyptol: eucalyptus odor, citronellal: lemon odor, and eugenol: cloves odor. These four odorants were chosen to represent primary odor categories claimed by Henning [11, 12] . They should sniff each odor for approximately 15 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 17, 2021. ; https://doi.org/10.1101/2021.05.13.21257176 doi: medRxiv preprint seconds blind and repeat this once with the name and a picture of the oil component on the screen by the application 30 seconds later (for example a picture of a lemon during lemon oil sniff)… Patients were asked to train in the morning and in the evening, resulting in a total of four expositions per day per odor. They were asked to keep a daily diary on the application where they rated overall olfactory abilities for each oil by subjective ratings with visual analogue scale. We assessed the rate of self-assessed improvement of overall olfactory function along training time from data collected anonymously by web-application diary of patients. Improvement was defined as a 2/10 points or more increase on olfactory visual analogue scale. Study analysis was performed when mean time of olfactory training of the population was at least 4-weeks and when at least 500 patients were assessable for primary outcome. Categorical variables were summarized using frequencies and percentages, Chi-square or Fisher's exact test were employed to make comparison. For quantitative variables be summarized with descriptive statistics, the following were presented: N, mean, standard deviation (SD), t-test was employed to compare group and the ANOVA test was allowed a comparison of more than two groups. The Kaplan-Meier methodology will be used to summarize time-to-event variables. Plots of Kaplan-Meier product limit estimates of time-to-event will be drawn, medians will be presented in addition to confidence intervals, set at 95 percent. To compare curves for two groups, the log-rank test will be employed. The level of statistical significance was 5% for all statistical tests (exploratory tests). To analyzed predictive factor of assessment, logistic regression was used in order to calculate the odds ratio, present with a confidence interval set at 95 percent. All statistical analyses were conducted with the SAS® System, Version 9.3 (SAS Institute Inc. Cary, NC, USA). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 1 . The mean baseline olfactory function of users registered on the web-application with less than 7 days olfactory training or without their last olfactory function assessment on web-application diary was higher than in the studied population (2.23 in 2824 patients and 1.9 in 548 patients, respectively (p Student test <.001). . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 17, 2021. ; . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 17, 2021. ; https://doi.org/10.1101/2021.05.13.21257176 doi: medRxiv preprint The mean baseline self-assessed olfactory score was 1.9/10 (SD 1.7) and increased to 4.6 (SD The rate of patients with improvement of at least 2/10 points on olfactory scale was higher in patients training more than 28 days versus patients training less than 28 days (72.2% vs 59.0% respectively, p=.002). Among patients with 28 days olfactory training or more and who have benefited from an improvement, the mean improvement was 4.4 (SD 2.0) on olfactory scale versus 3.8 points (SD 1.8) in patients with less than 28 days olfactory training (p Student test= .01). The mean improvement of self-assessed olfactory scale was similar whatever the anteriority of the olfactory dysfunction (p=0.7). Figure 3 . No predictive factors have been highlighted. Table 2. . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 17, 2021. ; https://doi.org/10.1101/2021.05.13.21257176 doi: medRxiv preprint .002 . CC-BY-NC-ND 4.0 International license It is made available under a 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 May 17, 2021. ; improvement as Reden et al. [20] More recently, Havervall et al reported that the duration of olfactory dysfunction after mild SARS-CoV-2 among seropositive health care workers was 14.6%, 10.8 and 9.0% 2, 4 and 8 months after infection, meaning that spontaneous recovery is low. [21] Spontaneous recovery after persistent olfactory dysfunction in SARS-CoV-2 patients is not well described. Vaira and al reported a mean of 1/10 point on the analogue subjective olfactory scale we used between 30 and 60 days in 138 patients without olfactory training and 20% patients presenting olfactory improvement. Our data suggest that improvement can be obtained tardily after 2-months training. In our study, olfactory training and visual stimulation allowed by a dedicated web-application was associated with 72.2% of patients with 2 points or more improvement after at least 28 days olfactory training and 4.4/10 points mean improvement. [4] In another study, Lechien et al reported that 15.3% and 4.7% of anosmic/hyposmic patients did not objectively recover olfaction at 60 days and 6 months, respectively. The comparison of our study with other studies is however limited because different olfactory tests and scale evaluations were used. [20] Our study had several limits. There was a lot of excluded patients. We could think that there is a selection bias because patient who don't feels improvement will more easily stop the training. It could be a confusion factor about the benefit statistically better if patients follow the training more than 28 days. The mean baseline olfactory function of users registered on the webapplication with less than 7 days olfactory training or without their last olfactory function assessment on web-application diary was higher than in the studied population (2.23 in 2824 patients and 1.9 in 548 patients, respectively (p Student test <.001). The distribution of the severity of patients suggests a higher severity of patient's olfactory dysfunction in the population retained in the analysis. This data suggests that the results of olfactory training could be higher in the whole population than in the studied population. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 17, 2021. ; https://doi.org/10.1101/2021.05.13.21257176 doi: medRxiv preprint There was no control group, therefore it remains unclear to which percentage spontaneous recovery distorts the results. The scale used to measure olfactory dysfunction and changes was subjective by using a self-assessment analogue scale, they were self-reported and data about olfactory assessment were not confirmed by physician and objective tests. However, the possibility to run olfactory training at home increased the number of recruited patients and triggered high levels of olfactory function recovery compared to spontaneous improvement. Olfactory training associated with visual stimulation assisted by a dedicated web-application is associated with significant olfactive improvement in persistent olfactory dysfunction following SARS-CoV-2 infection, especially if training duration is superior to 28 days. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted May 17, 2021. ; https://doi.org/10.1101/2021.05.13.21257176 doi: medRxiv preprint Smell and taste disorders in Spanish patients with mild COVID-19 Prevalence and Duration of Acute Loss of Smell or Taste in COVID-19 Patients Alterations in Smell or Taste in Mildly Symptomatic Outpatients With SARS-CoV-2 Infection Smell and taste recovery in coronavirus disease 2019 patients: a 60-day objective and prospective study Clinical and epidemiological characteristics of 1420 European patients with mild-to-moderate coronavirus disease 2019 Onset and duration of symptoms of loss of smell/taste in patients with COVID-19: A systematic review Olfactory Dysfunction in COVID-19: Diagnosis and Management Olfactory training is helpful in postinfectious olfactory loss: a randomized, controlled, multicenter study The nose smells what the eye sees: crossmodal visual facilitation of human olfactory perception Long term effects of olfactory training in patients with post-infectious olfactory loss Effects of Olfactory Training in Patients with Olfactory Loss Nonlinear response speedup in bimodal visualolfactory object identification Visual-Olfactory Interactions: Bimodal Facilitation and Impact on the Subjective Experience COVID-19 epidemic trends from a Self-Assessment Web Application in France: Observational Study Implementation of a Self-Triage Web Application for Suspected COVID-19 and Its Impact on Emergency Call Centers: Observational Study Epidemiological observation from a smartphone selfassessment web-application for suspected COVID-19 patients' triage: association of anosmia and outbreak extension in France Olfactory dysfunction Recovery of olfactory function following closed head injury or infections of the upper respiratory tract Prevalence and 6-month recovery of olfactory dysfunction: a multicentre study of 1363 COVID-19 patients After a mean time of olfactory training of 28 days, we observed 17.1% of patients (94/548)