Alcohol Sniff Test (AST): An Important Tool for Screening Post-Viral Olfactory Loss in Acute Flu-Like Dysfunction

Abstract

Olfactory disorders have a significant impact on patients' quality of life but are often underestimated in clinical practice. Upper respiratory tract infections (URTIs) are a common cause of olfactory loss. While most cases of olfactory loss due to URTIs are conductive and reversible, post-viral olfactory dysfunction (PVOD) persists despite symptom improvement. PVOD is attributed to damage to the olfactory epithelium and nerves or central olfactory pathway lesions. The Alcohol Sniff Test (AST) has been proposed as a tool to assess olfactory function in the acute phase and aid in differentiating PVOD from conductive disorders. This study aims to evaluate the effectiveness of the AST as a predictor of post-viral olfactory loss in patients with flu-like syndrome. An observational cross-sectional study was conducted among employees with flu-like syndrome at a tertiary hospital. Three groups were formed: flu-like syndrome with conductive disorder without COVID-19 (PVOD-), flu-like syndrome with neurosensory and/or central disorder due to COVID-19 (PVOD +), and an asymptomatic control group. The Alcohol Sniff Test was performed to assess olfactory function. Statistical analysis was conducted to evaluate the AST's performance. For a cut off of 10 cm, 88.57% of PVOD + patients and 60.53% of PVOD – patients showed AST alteration, respectively (p = 0.013, OR = 5.05, 95% CI [1.48–17.25]). There was a statistically significant difference in the mean distance between the PVOD + group (4.35 ± 4.1 cm) and the control group (20 ± 4.33 cm) (p < 0.05). This relationship was also observed between the PVOD + and PVOD- groups (9 cm ± 7.5) (p < 0.05) and between the PVOD- and control groups (p < 0.05). For a cut off of 10 cm, the AST showed a sensitivity of 88% and specificity of 41%, resulting in an Odds Ratio of 9.7 (95% CI 3.3–28.1) (p < 0.001) and a Positive Predictive Value of 69.4% for PVOD. PVOD, including cases associated with COVID-19, is a prevalent cause of olfactory loss. The Alcohol Sniff Test demonstrated promising results in identifying PVOD in patients with flu-like syndrome. The test's simplicity and accessibility make it a valuable tool for early screening and identifying individuals who may benefit from prompt treatment. The Alcohol Sniff Test (AST) shows potential as an effective tool for screening post-viral olfactory loss in patients with flu-like syndrome. It can aid in early identification of PVOD cases and facilitate timely interventions to reduce the likelihood of persistent hyposmia.

Introduction

Olfactory disorders are often underestimated in clinical practice and significantly impair the quality of life of patients, with social and emotional impacts [1]. They affect a large number of individuals, with estimated prevalence ranging from 2.7% to 76.8% [2].

Upper respiratory tract infections (URTIs) are one of the most common causes of olfactory loss, accounting for approximately 11% of all cases [3]. In most individuals, this alteration is conductive (conductive disorder), secondary to mucosal edema and increased nasal secretion, and once the symptoms improve, olfactory function returns to normal [4, 5]. However, in post-viral olfactory disfunction (PVOD), this loss persists despite the improvement of nasal obstruction, likely due to damage to the olfactory epithelium and nerves (neurosensorial disorder) and/or central olfactory pathway lesions (central disorder) [4–6]. Quint et al. analyzed a series of 120 patients with non-conductive olfactory disorders and found that 42.5% of them had a previous upper respiratory infection as an etiological factor [8].

In a study conducted in Japan in 1997, the main viruses identified in patients with PVOD were Rhinovirus, Coronavirus, Parainfluenza, and Epstein Barr virus [9]. More recently, in 2020, SARS-CoV-2 was also identified as a causative agent of PVOD, especially in patients infected with the ancestral strain in cases of mild flu-like syndrome [10].

The diagnosis of PVOD can be made when there is no improvement in olfactory function after an episode of URTI, in the absence of other pathologies that justify this alteration. However, there is no definitive test that clearly indicates that olfactory loss is of post-viral origin. Additionally, there is no test that can be performed in the acute phase to differentiate conductive disorders from PVOD. Furthermore, approximately 40% of patients with PVOD may continue to experience residual hyposmia for at least 3–6 months after the initial episode of URTI [8].

There are different treatment options for PVOD described in the literature [11]. Several studies suggest that the use of systemic and/or nasal steroids in the acute phase could decrease local inflammatory processes, reduce epithelial/olfactory nerve damage, and thus minimize the possibility of persistent olfactory loss [11–13].

First described by Davidson et al. in 1997 [14], the Alcohol Sniff Test (AST) is an accessible tool that could be used to test the olfactory function of patients in the acute phase of flu-like syndrome and allow for better screening of PVOD. This could lead to improved selection of individuals who could benefit from early treatment for PVOD and reduce the possibility of residual or persistent hyposmia.

Objective

The objective of this study is to evaluate the effectiveness of the Alcohol Sniff Test (AST) as a predictor of post-viral olfactory loss in patients with flu-like syndrome.

Methods

Study Design

An observational cross-sectional study was conducted between the months of September and December 2020 among the employees of a tertiary hospital presenting with flu-like syndrome.

Study Population

The study was approved by the ethics committee under the number 37266520.0.0000.5463, and all participants signed an informed consent form.

The individuals were selected through non-probabilistic judgment sampling. For the model of olfactory loss due to conductive disorder, patients with flu-like syndrome without COVID-19 were included, while those without COVID-19 were used as a model for post-viral olfactory disfunction [15].

Thus, three groups were formed: flu-like syndrome with conductive disorder without COVID-19 (PVOD-), flu-like syndrome with neurosensory and/or central disorder due to COVID-19 (PVOD +), and a third asymptomatic group.

The inclusion criteria for patients with flu-like syndrome were age between 18 and 65 years, flu-like syndrome under investigation for COVID-1 and presence of flu-like symptoms up to 10 days. The exclusion criteria were history of olfactory loss, history of nasal surgery, history of dementia or other neurological diseases, history of smoking, history of allergic rhinitis and/or chronic rhinosinusitis, history of craniofacial radiotherapy. Patients with flu-like syndrome without COVID-19 who experienced persistent olfactory loss after the acute phase were also excluded from the study.

For the control group, asymptomatic healthcare professionals between 18 and 65 years old, with no history of prior or active COVID-19 infection, were included. The exclusion criteria were the same as those used for patients with flu-like symptoms.

Patients with flu-like syndrome, after initial assessment by occupational physicians, underwent consultation with an otolaryngologist. General information such as age and gender were collected, along with questioning about olfactory changes and the presence of nasal symptoms (obstruction and/or rhinorrhea). Subsequently, they underwent the Alcohol Sniff Test (AST) and were referred for RT-PCR testing for SARS-CoV-2, which was collected between the third and 10th day of symptoms. Nasal endoscopy was not performed as it was prohibited at that time, in accordance with the guidelines of the Brazilian Association of Otolaryngology and Head and Neck Surgery. For the control group, general information regarding age and gender was collected, and questioning about olfactory changes was conducted. After the interview, they were also referred for the AST.

Olfactory Evaluation—Alcohol Sniff Test (AST)

Olfactory evaluations were performed using the Alcohol Sniff Test (AST), according to the initial description by Davidson et al [14]. The test was conducted using a sachet containing a tissue soaked in 70% isopropyl alcohol and a 30-cm ruler, following these steps:

1. With the participant seated, the sachet was opened, exposing approximately 0.5 cm of its contents;

2. The sachet was then brought close to the patient's nose twice to familiarize them with the smell of alcohol;

3. The individual was asked to close their eyes and mouth, place their hands on their legs, and breathe normally, avoiding deep or forced inhalations;

4. The ruler was positioned with the zero mark at the right nasal vestibule of the individual, and they were instructed to inform the examiner when they detected the smell of alcohol. The alcohol sachet was positioned at the 30-cm mark and moved one centimeter closer to the nose with each exhalation until the odor was detected;

5. The distance at which the participant reported the olfactory sensation was recorded, and the process was repeated three more times, resulting in four different distance measurements;

6. The average distance of these values was calculated, and the final result for each patient was recorded.

Statistical Analysis

The variables considered for this study were age, gender, olfactory alteration, presence of nasal symptoms (obstruction/rhinorrhea), and the average distances obtained after olfactory perception of the alcohol sachet. Categorical variables were described using frequencies, and numerical variables were described using means and standard deviations (SD). A significance level of α = 0.05 (5%) was used for this study. Statistical analysis was conducted using SigmaPlot v. 12.0 software (Systat, 2020). The normal distribution of quantitative variables was assessed using the Shapiro–Wilk and Mann–Whitney methods. For quantitative variables, ANOVA and Student's t-tests were used. Fischer's correlation and Chi-square (X2) tests were chosen for qualitative variables. Data regarding the validation of the AST were included in a Receiver Operating Characteristic (ROC) curve and subjected to Odds Ratio analysis.

The sample size calculation for this cross-sectional study was based on the work of Kemper RR et al. [16], using Epi Info software with a 95% confidence level, 80% power, and 5% margin of error. Using the Kelsey, Fleiss, and Fleiss methods with continuity correction, the calculated total number of individuals was 69, 76, and 90, respectively.

Results

A total of 103 individuals were studied: 35 (33.98%) with PVOD + , 38 (36.89%) with conductive disorder (PVOD-), and 30 asymptomatic individuals (29.12%). The three groups showed similar epidemiological distributions regarding age and gender variables. Patients with PVOD + reported significantly more olfactory loss than those with PVOD- (p < 0.05). There was no significant difference in the presence of rhinorrhea and/or nasal congestion between the PVOD + and PVOD- groups (p > 0.05) (Table 1).

Table 1.

Distribution of study variables

	PVOD + (n = 35)	PVOD—(n = 38)	Controls (n = 30)	p Value
Age*	45.2 ± 6.9	46 ± 8.4	38 ± 3.8	p > 0.05
Gender (M/F)**	25.7%/74.3%	7.9%/92.1%	23.3%/76.7%	p > 0.05
Report of olfactory alteration **	57.1%	26.3%	–	p < 0.05
Sintomas nasais**	48.7%	55.26%	–	p > 0.05

n = number of individuals, M = male gender, F = female gender, PVOD = post-viral olfactory dysfunction, * = values in mean and standard deviation, ** = values in percentages

For a threshold of 20.63 cm, 97.14% and 86.85% of patients showed AST alteration in the PVOD + and PVOD- groups, respectively (p > 0.05). For a cut off of 10 cm, 88.57% of PVOD + patients and 60.53% of PVOD- patients showed AST alteration, respectively (p = 0.013, OR = 5.05, 95% CI [1.48–17.25]).

The overall mean distance of the AST test was 10 ± 8.2 cm. There was a statistically significant difference in the mean distance between the PVOD + group (4.35 ± 4.1 cm) and the control group (20 ± 4.33 cm) (p < 0.05). This relationship was also observed between the PVOD + and PVOD- groups (9 cm ± 7.5) (p < 0.05) and between the PVOD- and control groups (p < 0.05) (Fig. 1).

Fig. 1.

Measurement of olfactory perception distances in the alcohol sniff test in 35 patients with PVOD, 38 patients with PVOD-, and 30 controls. Results are expressed in Box Plot, which represents the median, lower quartile, upper quartile, maximum value (Max), and minimum value (Min). Statistical analysis was performed using the Kruskal–Wallis test. The level of statistical significance was set at α = 0.05

The ROC curve for the AST test yielded an area under the curve of 0.83 (95% CI 0.74–0.91; p < 0.0001), as shown in Fig. 2. For a cut off of 10 cm, the AST showed a sensitivity of 88% and specificity of 41%, resulting in an Odds Ratio of 9.7 (95% CI 3.3–28.1) (p < 0.001) and a Positive Predictive Value of 69.4% for PVOD (Fig. 2).

Fig. 2.

Receiver operating characteristic (ROC) curve for the acohol sniff-test (AST). Area under the curve (A) = 0.83 (IC 95% 0.74–0.91 and p < 0.0001)

Discussion

Post-viral olfactory dysfunction (PVOD) has been described as one of the main causes of hyposmia/anosmia. Several studies have reported high rates of olfactory alterations in patients infected with SARS-CoV-2, which further support this hypothesis [17–20].

In a review study, the prevalence of olfactory loss in patients with COVID-19-associated post viral olfactory dysfunction ranged from 19.4% to 74% when questionnaires or self-reports were used as methods [21]. In a Brazilian study, Mendonca et al. (2021) evaluated 261 individuals with COVID-19-associated PVOD and found that approximately 89% of patients with mild COVID-19-related flu-like symptoms reported hyposmia/anosmia [22]. In studies that used the University of Pennsylvania Smell Identification Test (UPSIT), the prevalence of olfactory dysfunction was 82.7% [23] and 98% [24]. In the present study, also conducted in patients with mild symptoms, using SARS-CoV-2 infection as a model for non-conductive olfactory loss, 57.1% and 26.3% of individuals with positive and negative COVID-19-associated PVOD, respectively, reported olfactory loss (p < 0.05). For a cut off of 10 cm, 88.57% of patients with PVOD + and 64.10% of patients with PVOD- had olfactory detection thresholds on the Alcohol Sniff Test (AST) less than 10 cm (p < 0.05).

There was a statistically significant difference in the mean distance between the PVOD + and PVOD- groups (p < 0.05). This relationship also remained significant between the PVOD + group and controls (p < 0.05), as well as between the PVOD- group and controls (p < 0.05). These findings indicate that patients with flu-like syndrome and non-conductive olfactory loss have significantly shorter distances on the Alcohol Sniff Test (AST) compared to patients with flu-like syndrome and conductive olfactory dysfunction. To the best of our knowledge, this is the first study to use this methodology to evaluate olfactory dysfunction in individuals with PVOD.

Calvo-Henriquez et al. (2020) used ethyl alcohol, diluted in different concentrations, to test the olfaction of patients with confirmed and suspected diagnosis of COVID-19. The alcohol was presented on a gauze soaked in this solution at the level of the labiomental fold. For an alcohol concentration greater than or equal to 25%, in the evaluation of patients suspected of COVID-19, a test sensitivity of 75% and specificity of 67.62% were observed, with an odds ratio of 6.56 (95% CI 2.38–19.75) [25]. In the present study, for a cut off of 10 cm, the Alcohol Sniff Test (AST) showed a sensitivity of 88% and specificity of 41%, resulting in an odds ratio of 9.7 (95% CI 3.3–28.1) (p < 0.001) and a positive predictive value of 69.4% for PVOD. To the best of our knowledge, this is the first study to use this methodology to evaluate olfactory dysfunction in individuals with PVOD.

PVOD can be secondary to lesions in the olfactory epithelium and olfactory nerves or central lesions, and it appears to be similar among various respiratory viruses, including SARS-CoV-2 [15]. PVOD caused by this virus has shed light once again on the cellular pathophysiological mechanisms of this alteration. It is postulated that the infectivity of this microorganism depends on binding to angiotensin-converting enzyme 2 (ACE2) receptors, which are present in non-neuronal structures of the olfactory epithelium, such as sustentacular cells, Bowman's glands, and progenitor cells (stem cells).

Sustentacular cells are anatomically and functionally linked to olfactory receptor neurons (ORNs): they protect ORNs from volatile chemical substances by expressing enzymes from the cytochrome P450 family, provide glucose to the cilia of ORNs, and participate in the transduction process of these neurons. When these cells are affected, the cascade of processes involved in olfactory perception can be interrupted. After viral infection affects sustentacular cells, ORNs can be subsequently affected through contiguity. Simultaneously, an inflammatory process is initiated in ORNs and microvillar cells, recruiting cells such as macrophages and granulocytes and releasing inflammatory mediators, which can lead to damage of olfactory neurons. The formation of scar tissue, atrophy, disorganization, and reduction of ORNs can inhibit the regrowth and regeneration of the olfactory epithelium, resulting in delayed recovery of olfactory function. In some individuals, apoptosis of progenitor cells due to the inflammatory process can lead to permanent loss of olfactory function [26, 27]. Finally, some studies have visualized areas of bleeding and increased contrast uptake in the olfactory bulb region in the MRI of patients with COVID-19 and olfactory loss. The virus would cross the cribriform plate through the olfactory neurons to reach the olfactory bulb, where the second-order neurons responsible for this sense are located [7, 28].

A review study indicates a high prevalence of persistent olfactory loss following viral infection with SARS-CoV-2 [29, 30], in untreated patients. In a recent study, the rates of complete and incomplete objective recovery at 6 months after SARS-CoV-2 infection, as assessed by olfactory tests, were 44% and 56%, respectively. Meanwhile, the subjective recovery rate, described as a good, very good, or excellent sense of smell, was 81% during the same period [30]. In previous studies conducted before the COVID-19 pandemic, the rate of persistent olfactory alteration following URTI ranged from 18.6 to 42.5% [4, 8].

Therefore, treating the inflammatory process following viral aggression in patients with potential PVOD could minimize this sequelae or improve olfactory perception in these individuals. This is the main prognostic factor described in this group of individuals [12]. In a recent review study of patients with PVOD due to COVID-19, phosphodiesterase inhibitors (caffeine, theophylline), intranasal insulin, and corticosteroids (systemic and/or nasal) appear to be promising drugs for the treatment of this dysfunction [31].

In sudden hearing loss, audiometry is the initial diagnostic test of choice and should be performed within 140 days of symptom onset [32]. Similarly, in cases of suspected PVOD, AST could be used as a screening test in patients with flu-like symptoms. Thus, treatment with anti-inflammatory medications could be recommended to reduce the possibility of permanent sequelae when the AST is less than 10 cm.

AST is a test that is relatively simple, easily reproducible, painless, provides immediate results, and is cost-effective (only requiring a numbered ruler and an alcohol swab). It can be routinely employed in the evaluation of patients with flu-like symptoms. In the present study, individuals with an olfactory threshold on the AST test equal to or less than 10 cm had an almost 10 times higher chance of having PVOD.

Olfaction is dependent on the first (qualitative perception of odors) and fifth cranial nerves (somatosensory sensations) [33]. The function of the trigeminal nerve is explored using irritants (ammonia, acetic acid, and alcohol), while the function of the olfactory nerve is assessed using aromatic fragrances. Alcohol has a characteristic odor that is easily recognized by individuals. When this substance is very close to the nose (< 3 cm), it usually provokes a somatosensory sensation (itching). When it is at an appropriate distance, it stimulates the olfactory nerve, resulting in the perception of the odor of alcohol [14]. In the present study, these differences were considered during the testing of patients.

There are other well-established tests in the literature for evaluating olfactory function, such as the University of Pennsylvania Smell Identification Test (UPSIT), which assesses an individual's olfactory discrimination by identifying odors in a multiple-choice format using microencapsulated essences in a booklet [34]. However, despite its high reliability [34], these tests are not always accessible in some healthcare settings, especially in areas with limited resources.

One limitation of this study is that the sample consisted only of patients with mild flu-like symptoms, which may introduce selection bias. It should be noted that the value obtained in the AST test depends on the understanding and cooperation of the tested individual to perceive the exact distance at which the odor of alcohol is no longer detected. No distractors for the alcohol, such as saline, were used during the test. The study did not aim to evaluate the rate of olfactory recovery in this sample. No otolaryngological physical examination was performed at the initial evaluation to exclude local obstructive factors, although patients with previous nasal symptoms before the flu-like syndrome were excluded from the study.

Conclusion

For a cut-off of 10 cm, the Alcohol Sniff Test demonstrated a sensitivity of 88% and an odds ratio of 9.7 for post-viral olfactory loss. It should be considered as a screening test for studying olfaction up to 10 days from the onset of symptoms in mild flu-like syndrome, helping guide the treatment of these patients and facilitate timely interventions to reduce the likelihood of persistent hyposmia.

Funding

The research did not receive any external funding, and all the materials used in the study were financed by the researchers themselves.

Declarations

Conflict of Interest

There are no conflicts of interest to declare.

Ethical Statement

This manuscript has not been published or presented elsewhere in part or in entirety and is not under consideration by another journal. All authors have approved the manuscript and agree with submission. The article is in compliance with ethical standards.

Ethical Approval

The study was approved by the ethics committee under the number 37266520.0.0000.5463, and all participants signed an informed consent form. The individuals were selected through non-probabilistic judgment sampling.