Discussion

Since the beginning of the pandemic, SARS-CoV-2 has undergone numerous mutations that have led to a succession of variants of concern. The omicron variant made its appearance in South Africa in October 2021.¹⁹ As a result of its high contagiousness, the omicron variant spread rapidly around the world becoming the predominant variant of concern at the end of January 2022.^{Reference Boscolo-Rizzo, Hummel, Hopkins, D'Alessandro, Menini and Dibattista20} Similarly, the characteristics of the general population have changed over time due to immunisation obtained with vaccination or with previous infection. It is therefore unsurprising that the clinical picture of Covid-19 has changed over time.^{Reference Maisa, Spaccaferri, Fournier, Schaeffer, Deniau and Rolland21}

During the pandemic, there was a gradual reduction in the prevalence of olfactory dysfunction,^{Reference Klimek, Hagemann, Hummel, Altundag, Hintschich and Stielow22,Reference Hintschich, Vielsmeier, Bohr, Hagemann and Klimek23} with a significant difference comparing the omicron variant with the first pandemic waves.^{Reference Boscolo-Rizzo, Tirelli, Meloni, Hopkins, Madeddu and De Vito11–Reference Cardoso, Rossi, Galliez, Faffe, Tanuri and Castiñeiras13} In this study, 25.5 per cent of patients self-reported a loss of smell during infection, and 33.3 per cent had threshold, discrimination and identification scores indicative of olfactory dysfunction. This prevalence is significantly lower than that recorded in the Sardinia region itself during the first pandemic waves, which exceeded 70 per cent.^{Reference Vaira, De Vito, Deiana, Pes, Giovanditto and Fiore24–Reference Vaira, Deiana, Fois, Pirina, Madeddu and De Vito26} However, impairment of olfactory function still affects more than one-third of patients and therefore remains an important symptom to be considered in order to suspect SARS-CoV-2 infection.

The reasons behind the reduced ability of the omicron variant to induce olfactory dysfunction are not yet fully understood. In fact, the omicron variant presents the D614 G mutation of the spike protein, which has been associated in the past with a higher incidence of olfactory dysfunction.^{Reference von Bartheld, Mathew and Butowt27} It is therefore possible that the lower olfactory dysfunction rate is due to the fact that the omicron variant diffuses less easily in the nasal mucus^{Reference Butowt, Bilinska and von Bartheld28} and that it uses transmembrane serine protease 2 less effectively as an access route into the olfactory epithelium cells.^{Reference Butowt, Bilinska and von Bartheld28,Reference Meng, Abdullahi, Ferreira, Goonawardane, Saito and Kimura29} On the other hand, it is possible that the more effective response in immunised subjects leads to a more rapid viral wash out before this can induce sufficient damage to cause clinically evident olfactory dysfunction.^{Reference Vaira, Lechien, Salzano, Maglitto, Saussez and De Riu30,Reference Sheikh-Mohamed, Isho, Chao, Zuo, Cohen and Lustig31} However, it is likely that both of these factors are implicated, because olfactory dysfunction has been found to be frequent with previous variants of concern, even in already immunised subjects.^{Reference Lechien, Chiesa-Estomba, Radulesco, Michel, Caira and Le Bon32,Reference Vaira, De Vito, Lechien, Chiesa-Estomba, Mayo-Yàñez and Calvo-Henrìquez33}

The very high number of patients with Covid-19 who self-reported persistent olfactory dysfunction represents a major challenge. A recent meta-analysis^{Reference Tan, Han, Zhao, Tan, Quah and Tan3} estimated that 5.6 per cent of patients self-reported a persistent olfactory disorder six months after Covid-19, with the prevalence rising to over 30 per cent when evaluated with psychophysical tests.^{Reference Boscolo-Rizzo, Hummel, Hopkins, Dibattista, Menini and Spinago34} Persistent olfactory dysfunction has a devastating effect on the quality of life of patients, who may isolate themselves socially, reduce their work productivity and withdraw from normal daily activities.^{Reference Vaira, Gessa, Deiana, Salzano, Maglitto and Lechien6,Reference Saniasiaya and Prepageran7,Reference Boesveldt and Parma35} For these reasons, persistent olfactory dysfunction represents an unprecedented health challenge.^{Reference Boscolo-Rizzo, Polesel and Vaira8,Reference Vaira, De Riu, Salzano, Maglitto, Boscolo-Rizzo and Lechien36–Reference Lechner, Liu, Counsell, Gillespie, Chandrasekharan and Ta38}

To the best of our knowledge, the recovery rate of olfactory function after omicron variant infection has not yet been investigated. The assessment of olfactory recovery should include psychophysical tests because patients often tend to overestimate recovery, especially if they present with complete anosmia at baseline,^{Reference Vaira, Salzano, Le Bon, Maglio, Petrocelli and Steffens1,Reference Boscolo-Rizzo, Hummel, Hopkins, Dibattista, Menini and Spinago34} and the follow up should be long enough to detect delayed recoveries that are significantly frequent in the first three months after infection.^{Reference Petrocelli, Cutrupi, Salzano, Maglitto, Salzano and Lechien39} In this study, the prevalence of persistent olfactory dysfunction was 11.3 per cent, with only one case of anosmia. The differences between the cases and the controls were not significant either for the threshold, discrimination and identification score or for the clinical classifications of olfactory function. In the pandemic waves caused by previous variants of concern, several authors found a six-month rate of complete anosmia of around 10 per cent, exceeding 25 per cent if we consider the hyposmic subjects.^{Reference Petrocelli, Cutrupi, Salzano, Maglitto, Salzano and Lechien39–Reference Boscolo-Rizzo, Menegaldo, Fabbris, Spinato, Borsetto and Vaira43}

The pathogenetic mechanism and risk factors underlying the persistence of olfactory dysfunction in some individuals have not been established with certainty.^{Reference Paderno, Schreiber, Grammatica, Raffetti, Tomasoni and Gualtieri44–Reference Amadu, Vaira, Lechien, Scaglione, Saba and Lampus46} It has been hypothesised that the damage to the olfactory epithelium during infection is so severe and extensive in some individuals that it causes the loss of basal cells, which are necessary for the epithelium to regenerate.^{Reference Dias de Melo, Lazarini, Levallois, Hautefort, Michel and Larrous47–Reference Vaira, Hopkins, Sandison, Manca, Machouchas and Turilli49} In this sense, the omicron variant has been shown not to cause such severe damage to the olfactory epithelium as previous variants of concern^{Reference Armando, Beythien, Kaiser, Allnoch, Heydemann and Rosiak50} and it does not appear to have neuroinvasive capabilities.^{Reference Armando, Beythien, Kaiser, Allnoch, Heydemann and Rosiak50,Reference Bauer, Rissmann, Benavides, Leijten, van Run and Begeman51} Moreover, a high concentration of nasal immunoglobulins was detected as a protective factor for the development of persistent olfactory dysfunction.^{Reference Saussez, Sharma, Thiriard, Olislagers, Vu Duc and Le Bon52,Reference Vaira, De Riu, Mayo-Yàñez, Gengler and Lechien53} Vaccinated subjects have been shown to have higher levels of nasal and salivary secretory immunoglobulins compared to post-infection immunised individuals, especially after the second dose.^{Reference Sheikh-Mohamed, Isho, Chao, Zuo, Cohen and Lustig31,Reference Guerrieri, Francavilla, Fiorelli, Nuccetelli, Passali and Coppeta54}

Finally, another reason for this improved long-term outcome may be related to the fact that all patients included in this study were initiated early on a therapeutic protocol of olfactory training associated with palmitoylethanolamide and luteolin supplementation.^{Reference Di Stadio, D'Ascanio, Vaira, Cantone, De Luca and Cingolani55} Several authors have demonstrated the effectiveness of olfactory training in promoting and speeding up the recovery of olfactory function after SARS-CoV-2 infection.^{Reference Altundag, Ylmaz and Kesimli56–Reference Asvapoositkul, Samuthopongtorn, Aeumjaturapat, Snidvongs, Chusakul and Seresirikachorn59} However, in our opinion, it is important to closely monitor patients with olfactory dysfunction during infection, starting them with olfactory training early for at least two reasons. Firstly, in this study, 5 per cent of patients still self-reported incomplete recovery at six months. Secondly, the 60-day re-evaluation of controls who developed infection in the follow-up period found that although threshold, discrimination and identification scores returned to levels that classified individuals as normosmic, median T values were significantly lower than before infection.^{Reference Vaira, Lechien, Salzano, Maglitto, Boscolo-Rizzo and Hopkins60} This olfactory alteration, although unconscious and therefore not affecting quality of life, can expose patients to environmental dangers.

This study has some limitations that need to be acknowledged. Firstly, although patient enrolment occurred when omicron variant circulation was greater than 98 per cent for more than two weeks, variant of concern determination was not performed and contamination may have occurred. Secondly, to reduce the risk of selection bias, patients were enrolled consecutively from lists provided by the Department of Prevention of the University Hospital of Sassari, and it was not initially disclosed that the test would involve smell. It is possible that patients with more severe symptoms were more likely to agree to evaluation, as well as individuals with self-perceived olfactory dysfunction at baseline agreeing to undergo re-evaluation at six months. Thirdly, the controls were not tested for SARS-CoV-2 infection at the time of olfactory evaluation. However, the individuals included in the control group were part of the hospital staff and then subjected to regular antigenic swabs and immunoglobulin assays, which were consistently negative. Fourthly, the six-month follow up is not long enough to rule out late recoveries that can occur spontaneously, even two years after infection.^{Reference Lee, Lee, Wee and Kim61}