Introduction
In recent months, the prognostic value of olfactory disturbances in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been widely debated.Reference Vaira, Hopkins, Petrocelli, Lechien, Soma and Giovanditto1–Reference Lechien, Ducarme, Place, Chiesa-Estomba, Khalife and De Riu6 Most authors report a higher prevalence of olfactory dysfunction in mild forms of coronavirus disease 2019 (Covid-19),Reference Yan, Faraji, Prajapati, Ostrander and DeConde4–Reference Lechien, Ducarme, Place, Chiesa-Estomba, Khalife and De Riu6 proposing that this is the result of an enhanced immune response in the upper respiratory tract.Reference Le Bon and Horoi7
Recently, several studies have highlighted the prognostic value of some serum inflammatory markers, such as D-dimer,Reference Li, Zhao, Wei, Chen, Wang and Jia8,Reference Zhang, Yan, Fan, Liu, Liu and Liu9 C-reactive protein (CRP),Reference Liu, Li, Xu, Wu, Luo and Zhu10 procalcitonin,Reference Lippi and Plebani11 ferritin,Reference Vargas-Vargas and Cortes-Rojo12 lactate dehydrogenase (LDH)Reference Zhang, Lee, Ang, Leo and Young13 and neutrophil-to-lymphocyte ratio.Reference Yang, Liu, Tao and Li14,Reference Zinellu, Scano, Masotto, De Riu, Vaira and Carru15 Specifically, D-dimer is elevated in the microangiopathy and hypercoagulability states of the most severe cases of Covid-19.Reference Li, Zhao, Wei, Chen, Wang and Jia8,Reference Zhang, Yan, Fan, Liu, Liu and Liu9 Lactate dehydrogenase, released by cells after damage to their membrane, has an immunosuppressive action and inhibits cytolytic cells, thus weakening the immune response against the virus.Reference Zhang, Lee, Ang, Leo and Young13 Procalcitonin is a marker of bacterial superinfection that may contribute to complications of Covid-19. C-reactive protein, ferritin and neutrophil-to-lymphocyte ratio are instead markers of acute inflammation, and are particularly elevated during the cytokine storm typical of the most severe forms of Covid-19.Reference Lippi and Plebani11,Reference Vargas-Vargas and Cortes-Rojo12,Reference Yang, Liu, Tao and Li14,Reference Zinellu, Scano, Masotto, De Riu, Vaira and Carru15 All of these markers have been associated with a higher rate of intensive care unit admission, acute respiratory distress syndrome and mortality.Reference Li, Zhao, Wei, Chen, Wang and Jia8–Reference Zinellu, Scano, Masotto, De Riu, Vaira and Carru15
In order to determine the prognostic power of olfactory disturbances in Covid-19 patients, it would be useful to establish whether there are correlations between chemosensitive dysfunction, disease severity and these already validated serological markers. This study therefore aimed to analyse the correlations between the olfactory scores determined by psychophysical tests and the serum levels of D-dimer, CRP, ferritin, LDH, procalcitonin and neutrophil-to-lymphocyte ratio in patients affected by Covid-19 and admitted to the coronavirus disease departments of the University Hospital of Sassari. The secondary objective was to establish whether any of these markers correlated with the persistence of olfactory dysfunction 60 days after onset.
Materials and methods
This cohort study was conducted in the coronavirus disease departments of the University Hospital of Sassari (namely the Infectious and Tropical Diseases, Pneumology, Onco-COVID and Neuro-COVID operative units).
The criteria for patient inclusion in the study were as follows: adults aged over 18 years, rhino-pharyngeal swab positive for SARS-CoV-2 infection, Covid-19 symptoms present for less than 10 days, and patient acceptance for participation in the study. The exclusion criteria were: uncooperative patients, assisted ventilation, psychiatric or neurological disorders, previous surgery or radiotherapy in the oral and nasal cavities, pre-existing self-reported smell and taste alterations, history of head trauma, allergic rhinitis, and chronic rhinosinusitis. Moreover, patients were excluded if they presented with underlying conditions that could alter serum levels of: procalcitonin (e.g. lung or thyroid cancer), CRP (e.g. known acute and chronic inflammatory conditions, Crohn's disease, acute myocardial infarction), ferritin (e.g. alcohol abuse, iron metabolism diseases, iron therapy), D-dimer (e.g. pregnancy, history of thromboembolism, severe liver failure), LDH (e.g. acute kidney, liver or pancreatic disease, haemolytic anaemia) or the neutrophil-to-lymphocyte ratio (e.g. haematological diseases with alteration of the granulocyte or lymphocyte line, advanced stage neoplasms).
The study was conducted in accordance with the ethical standards of the institutional research committee, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
The following clinical and epidemiological information was collected for all patients: age, gender and Covid-19 symptoms. All patients were followed up clinically until the rhino-pharyngeal swab results were negative. The overall clinical severity of Covid-19 was classified according to Tian et al.Reference Tian, Hu, Lou, Chen, Kang and Xiang16 as mild, moderate, severe or critical.
Psychophysical olfactory evaluation was performed with the Connecticut Chemosensory Clinical Research Center test. This test is a validated, widely used and easy-to-perform psychophysical test. The methodology, the scoring system and its application in Covid-19 patients have been extensively described in previous studies.Reference Vaira, Deiana, Fois, Pirina, Madeddu and De Vito17–Reference Vaira, Salzano, Petrocelli, Deiana, Salzano and De Riu20 The test includes the assessment of the olfactory threshold using solutions with increasing concentrations of N-butyl acid and an identification task for common odorants. The olfactory score thus obtained allows a clinical classification of olfactory function according to five categories: normal function (scores of 90 or 100), mild hyposmia (scores of 70 or 80), moderate hyposmia (scores of 50 or 60), severe hyposmia (scores of 20, 30 or 40) or anosmia (scores of 0 or 10).
Within 24 hours before or after the olfactory test, plasma levels of D-dimer, CRP, ferritin, LDH, procalcitonin and neutrophil-to-lymphocyte ratio were determined from a peripheral blood sample taken from each patient. The olfactory function of each patient was re-evaluated with the Connecticut Chemosensory Clinical Research Center test 60 days after the first evaluation.
Statistical analyses were performed with SPSS 26.0 software (IBM, Armonk, New York, USA). Categorical variables are reported in numerals and percentages of the total. Descriptive statistics for quantitative variables are given as means ± standard deviations or medians (interquartile ranges). The Kruskal–Wallis test was performed to evaluate the statistical significance of differences in olfactory scores and assessed inflammatory markers between clinical severity subgroups. Post-hoc analysis with the Mann–Whitney U test was used to evaluate the significance of the differences between each subgroup of clinical severity of Covid-19 for those markers that presented significant p-values on the Kruskal–Wallis test. The correlations between olfactory scores and D-dimer, CRP, ferritin, LDH, procalcitonin and neutrophil-to-lymphocyte ratio levels were assessed with the Spearman rank correlation co-efficient. The level of statistical significance was set at p > 0.05, with a 95 per cent confidence interval.
Results
Following application of the inclusion and exclusion criteria, 77 patients were included in this study. Table 1 summarises the epidemiological and clinical features of the patients.
Table 1. General and clinical features of study population
*n = 77; †n = 73. CI = confidence interval; SD = standard deviation; Covid-19 = coronavirus disease 2019; NV = normative value; CRP = C-reactive protein; PCT = procalcitonin; LDH = lactate dehydrogenase
The severity of Covid-19 was classified as mild in 34 patients, moderate in 26 and severe in 17. On the psychophysical tests, 57 patients (74 per cent) presented with olfactory dysfunction, including 11 with mild hyposmia (14.3 per cent), 19 with moderate hyposmia (24.7 per cent), 13 with severe hyposmia (16.9 per cent), and 14 with anosmia (18.1 per cent) (Table 1). The median olfactory score was 60 (interquartile range, 30–90). No significant differences in olfactory scores between the Covid-19 severity subgroups were reported (Table 2). In contrast, D-dimer, procalcitonin, ferritin and neutrophil-to-lymphocyte ratio demonstrated significantly higher serum levels in patients with severe Covid-19 (Tables 2 and 3).
Table 2. Olfactory scores and inflammatory marker levels according to Covid-19 clinical severityReference Tian, Hu, Lou, Chen, Kang and Xiang16
Data represent medians (interquartile ranges), unless indicated otherwise. *n = 34; †n = 26; ‡n = 17. Covid-19 = coronavirus disease 2019; CCCRC = Connecticut Chemosensory Clinical Research Center test; CRP = C-reactive protein; PCT = procalcitonin; LDH = lactate dehydrogenase; NLR = neutrophil-to-lymphocyte ratio
Table 3. Post-hoc analysis results
Covid-19 = coronavirus disease 2019; PCT = procalcitonin; NLR = neutrophil-to-lymphocyte ratio
The correlations between baseline olfactory scores and serum levels of inflammatory markers were weak and non-significant for all the indexes evaluated (Table 4).
Table 4. Correlation analysis results
CCCRC = Connecticut Chemosensory Clinical Research Center test; CRP = C-reactive protein; PCT = procalcitonin; LDH = lactate dehydrogenase; NLR = neutrophil-to-lymphocyte ratio
Seventy-three patients underwent an olfactory assessment with the Connecticut Chemosensory Clinical Research Center test 60 days after the first assessment. This second evaluation revealed a median olfactory score of 90 (interquartile range, 70–100). At this observation time, 65.8 per cent of patients had normal olfactory function. The remaining 34.2 per cent of patients had olfactory dysfunction, including 10 with mild hyposmia (13.7 per cent), 6 with moderate hyposmia (8.2 per cent), 6 with severe hyposmia (8.2 per cent) and 3 with anosmia (4.1 per cent) (Table 1).
There were no significant correlations between the markers assessed and the olfactory scores at two months (Table 4).
Discussion
To our knowledge, only two studies have previously analysed correlations between olfactory function and one or more serological markers of inflammation in Covid-19 patients,Reference Talavera, Garcìa-Azorìn, Martìnez-Pìas, Trigo, Hernàndez-Pérez and Valle-Penacoba21,Reference Benkirane, Heikel, Laamiri, Bouziani, Lahmam and Al-Jawaldeh22 and reported different results. Benkirane et al.Reference Benkirane, Heikel, Laamiri, Bouziani, Lahmam and Al-Jawaldeh22 evaluated the correlations between self-reported olfactory loss and different clinical and laboratory markers in 108 Covid-19 patients. The authors found no significant correlations between loss of smell and serum levels of CRP, ferritin and LDH. Only D-dimer was significantly associated with the presence of olfactory dysfunction. In another study, by Talavera et al.,Reference Talavera, Garcìa-Azorìn, Martìnez-Pìas, Trigo, Hernàndez-Pérez and Valle-Penacoba21 performed with the same methodological setting, contrasting results were found, with significant correlations detected between the absence of self-reported olfactory disturbance and high levels of CRP and D-dimer and, even if not significantly, with all the other negative prognostic markers. On the basis of these findings, the authors attributed a positive prognostic value to olfactory disturbances during Covid-19.
In another study, Elibol and BaranReference Elibol and Baran23 analysed the relationship between D-dimer and ferritin and Covid-19 related gustatory dysfunctions. As noted by Benkirane et al.Reference Benkirane, Heikel, Laamiri, Bouziani, Lahmam and Al-Jawaldeh22 for olfactory disorders, the authors found a significant correlation between high levels of D-dimer and the presence of ageusia. It was not possible to find any study investigating correlations between systemic inflammatory markers and other post-viral olfactory dysfunctions.
The major limitation of the previously published studies is that the assessment of smell was not conducted using quantitative tests. It is now well known that self-reported smell loss alone significantly underestimates the real prevalence of Covid-19 related olfactory dysfunction in infected individuals.Reference Mazzatenta, Neri, D'ardes, De Luca, Marinari and Porreca24,Reference Hannum, Ramirez, Lipson, Herriman, Toskala and Lin25 Moreover, by reducing olfactory dysfunction to a dichotomous variable, it is not possible to perform a statistical analysis based on direct correlations between continuous variables, which is certainly more accurate. These limitations reduce the reliability of the results, and this possible bias is not acceptable if we think that these prognostic studies may influence people's behaviours or the implementation of public health measures.Reference Hopkins, Vaira and De Riu26
A further strength of our study is that the assessment of smell and laboratory markers occurred in the first 10 days of clinical onset, when the recovery of olfactory dysfunction has not generally yet begun and Covid-19 has not reached its maximum clinical severity.Reference Vaira, Hopkins, Petrocelli, Lechin, Chiesa-Estomba and Salzano27–Reference Lechien, Journe, Hans, Chiesa-Estomba, Mustin and Beckers30 The severity of Covid-19 was then monitored throughout the duration of the infection. In this way, it was possible to evaluate the prognostic power of all the variables taken into consideration. Regarding laboratory markers, D-dimer, procalcitonin, ferritin and neutrophil-to-lymphocyte ratio have shown a significant and directly proportional correlation with severe forms of Covid-19 (Tables 2 and 3). The prognostic value of these indices has been pointed out by other authors in recent months.Reference Li, Zhao, Wei, Chen, Wang and Jia8,Reference Zhang, Yan, Fan, Liu, Liu and Liu9,Reference Lippi and Plebani11,Reference Vargas-Vargas and Cortes-Rojo12,Reference Yang, Liu, Tao and Li14,Reference Zinellu, Scano, Masotto, De Riu, Vaira and Carru15,Reference Deiana, Azara, Dettori, Delogu, Vargiu and Gessa31
Seventy-four per cent of the study patients had olfactory dysfunction at the time of evaluation. This high prevalence is influenced by the fact that the test was performed at a very early stage of the disease and is in line with findings at our centre for patients in the first wave of the pandemic.Reference Vaira, Deiana, Fois, Pirina, Madeddu and De Vito17,Reference Vaira, Hopkins, Salzano, Petrocelli, Melis and Cucurullo18,Reference Vaira, Salzano, Petrocelli, Deiana, Salzano and De Riu20
Olfactory scores proved unreliable as prognostic markers. In fact, there were no significant differences in the median scores between the subgroups of clinical severity of Covid-19, nor significant correlations with any of the laboratory indices analysed. This finding is evidently in contrast with that previously reported by Talavera et al.Reference Talavera, Garcìa-Azorìn, Martìnez-Pìas, Trigo, Hernàndez-Pérez and Valle-Penacoba21 and, as regards the D-dimer, by Benkirane et al.Reference Benkirane, Heikel, Laamiri, Bouziani, Lahmam and Al-Jawaldeh22 This study conflicts with others which suggest that the development of olfactory loss may predict a less severe course of disease and the avoidance of hospitalisation.Reference Yan, Faraji, Prajapati, Ostrander and DeConde4,Reference Foster, Jauregui, Tajudeen, Bishehsari and Mahdavina32 It should be noted that, in contrast to the current study, all of these studies were performed retrospectively and relied on the self-reporting of olfactory loss. The findings are therefore subject to recall bias that likely differs according to disease severity: patients with severe disease may simply neglect transient olfactory dysfunction in the setting of severe respiratory compromise.
• Olfactory scores proved unreliable as prognostic markers
• There were no significant differences in median scores between the coronavirus disease 2019 (Covid-19) clinical severity subgroups
• In addition, there were no significant correlations with any of the laboratory indices analysed
• The correlations between olfactory scores and serum inflammatory markers were weak and non-significant at baseline and 60 days, for all indexes evaluated
• There were no correlations between the severity and duration of olfactory disturbance and serum levels of inflammatory markers
• This suggests that the pathogenetic mechanisms underlying smell loss in Covid-19 patients are related to local rather than systemic inflammatory factors
Furthermore, the levels of systemic inflammatory markers during the acute phase of infection were not correlated with the persistence of olfactory dysfunction at two months. The first 6- and 12-month follow-up studies are finding a significant prevalence of severe, long-lasting olfactory disturbances in Covid-19 patients.Reference Vaira, Hopkins, Petrocelli, Lechin, Chiesa-Estomba and Salzano27–Reference Lechien, Journe, Hans, Chiesa-Estomba, Mustin and Beckers30,Reference Hopkins, Surda, Vaira, Lechien, Safarian and Saussez33–Reference Petrocelli, Cutrupi, Salzano, Maglitto, Salzano and Lechien35 This means that a large number of patients will seek assistance for the treatment of this disabling long-term morbidity. In the near future, it will be crucial to identify whether there are epidemiological, clinical or laboratory risk factors for the development of persistent disorders. In this way, it would be possible to establish which patients should be subjected to specific therapy to prevent the persistence of olfactory dysfunction.Reference Vaira, Hopkins, Petrocelli, Lechien, Cutrupi and Salzano19,Reference Lechien, Hoch, Vaira and Saussez36,Reference Huart, Philpott, Altundag, Fjaeldstad, Frasnelli and Gane37
Finally, the results of this study may provide some indication about the pathogenesis of Covid-19 related olfactory dysfunctions, which may be related to local rather than systemic inflammatory factors.Reference Kirschenbaum, Imbach, Ulrich, Rushing, Keller and Reimann38–Reference Bilinska and Butowt41 Future studies will be needed to determine if there are any correlations between the severity of olfactory dysfunction and the levels of inflammatory cytokines in the nasal mucus.
This study has several limitations. The number of patients is too small to draw definitive conclusions. However, the number of patients included was sufficient to show the prognostic value of D-dimer and other markers, suggesting that a type 2 statistical error is less likely. In addition, the 60-day follow up is still too short to detect some delayed functional recoveries that can occur even longer than 6 months after clinical onset.Reference Vaira, Hopkins, Petrocelli, Lechin, Chiesa-Estomba and Salzano27–Reference Lechien, Journe, Hans, Chiesa-Estomba, Mustin and Beckers30,Reference Hopkins, Surda, Vaira, Lechien, Safarian and Saussez33,Reference Boscolo-Rizzo, Guida, Polesel, Marcuzzo, Antonucci and Capriotti34
Conclusion
Olfactory disturbances appear to have a weak prognostic value in predicting the severity of Covid-19 compared to other markers such as D-dimer, ferritin, procalcitonin and neutrophil-to-lymphocyte ratio.
Competing interests
None declared
