Impact of Hedonics and Magnitude Upon Validated Psychophysical Olfactory Tests and Olfactory-Specific Quality of Life

Abstract

Introduction

Patients often describe the olfactory loss as either the inability to detect unpleasant odors (eg, smoke or spoiled food) or noticeable impairment in enjoying pleasant odors (eg, food and fragrances). This complex relationship between the hedonics of various odors, their perceived magnitude in odor strength, and the impact on patient quality of life is not well understood.

Methods

Sixty-five subjects underwent testing with the Henkin olfaction test which assesses hedonics and magnitude for odors traditionally deemed unpleasant (pyridine/dead fish and thiophene/gasoline) and pleasant (amyl acetate/banana and nitrobenzene/almond). Subjects also completed Smell Identification Test-40 (SIT-40), Sniffin’ Sticks (Sniffin’ Sticks), and Snap & Sniffin’ Sticks (Snap) Olfactory Tests, as well as the 17-item Questionnaire for Olfactory Disorders Negative Statements (QODNS) and olfactory symptom visual analog scale (VAS). Spearman's rank correlations were conducted between various olfactory domains, QODNS and VAS.

Results

Mean age was 52.1 years (range: 28-86), with 63.1% females. Magnitude estimation was greatest for pyridine. Most people rated thiophene (75.4%) and pyridine (92.3%) as “unpleasant.” Although most people rated amyl acetate (47.7%) and nitrobenzene (56.9%) as “pleasant,” many rated amyl acetate (52.3%) and nitrobenzene (43.1%) as “neutral” or “unpleasant” scents. Hedonics for unpleasant odors correlated with each other (r_s = 0.60), but not with hedonics for pleasant odors. Hedonics for unpleasant odors also correlated with magnitude estimation and other psychophysical tests. These findings were not seen for hedonics of pleasant odors.

Conclusions

The use of unpleasant odors in Henkin testing showed unique relationship patterns with patient-reported outcome measures and validated olfactory tests, which may provide utility in characterizing olfactory dysfunction. The incorporation of more unpleasant odors in current psychophysical olfactory testing may provide enhanced information regarding the patient impact of olfactory dysfunction.

Keywords

olfaction olfactory tests correlation patient-reported outcome metrics quality of life questionnaire for olfactory disorders Henkin Test VAS hedonics smell

Introduction

Parosmia, or olfactory dysfunction, gained significant attention due to the COVID-19 pandemic, with an estimated 52.7% of patients experiencing loss of smell acutely.¹ Even without an inciting disease process, olfactory dysfunction is common in otherwise healthy adults.² Furthermore, recent studies highlighted that olfaction dysfunction is not a benign symptom and significantly impacts diet, food enjoyment, social isolation, depression, and even mortality.^3–6 Optimal characterization of olfactory loss is thus imperative for both population health and individual disease management.

Patients presenting with olfactory loss often present with one of 2 major complaints. One group may notice olfactory loss when they are unable to detect unpleasant odors, such as smoke, spoiled food, or diapers. The second group notices impairment of pleasant odors which impact their ability to enjoy food, candles, and other fragrances. Both groups express difficulties with hedonics, which is the perceived pleasantness or unpleasantness of an odor. Hedonics is involved in approach-avoidance behavior modification, which may highlight primitive mechanisms for improving safety and survival.⁷ The complex relationship between the hedonics of various odors, their perceived magnitude, and the impact on patient quality of life is not well understood.

The importance of hedonics is particularly highlighted in phantosmia and cacosmia. Phantosmia is the perception of an odor despite the absence of a physical stimulus. For a perceived unpleasant odor, phantosmia is often further specified as “cacosmia.”⁸ Although these symptoms may coincide with hyposmia, phantosmia and cacosmia may exist without quantitative olfactory deficits.⁹ Furthermore, phantosmia and cacosmia have been commonly reported with COVID-19 infection.⁹ Although phantosmia and cacosmia have been shown to effect the quality of life, the impact in comparison to quantitative olfactory disorders remains unclear.¹⁰

Current methods of testing olfaction do not adequately test hedonics, and therefore do not adequately test for phantosmia and cacosmia. The most commonly used psychophysical tests include the 40-item Smell Identification Test (SIT-40), Sniffin’ Sticks, and Snap & Sniff (Snap). These tests measure domains of olfaction (eg, threshold, discrimination, identification) and the majority of employed odorants are pleasurable or neutral odors.^11–14 Psychophysical tests, therefore, provide information on the ability to detect and differentiate odors, but do not necessarily comment on other forms of parosmia, including those related to altered perception in hedonics (eg, phantosmia). One of the earliest olfactory tests reported in scientific literature used a method of chemosensory testing involving 4 odors and 4 domains of olfaction (herein referred to as “Henkin Test”).¹⁵ Henkin Test used 2 known noxious odorants and 2 relatively pleasant odorants. Furthermore, the Henkin Test uniquely quantified perceived magnitude and hedonics. The purpose of this study was to determine how hedonics and magnitude estimation using both noxious and pleasant/neutral odorants relate to common validated psychophysical tests and patient-reported outcome metrics (PROMs).

Methods

Subject Enrollment

This prospective cohort study was approved by the Medical University of South Carolina's institutional review board (#Pro00111948). Subjects who were at least 18 years of age and able to consent to study procedures were included. Exclusion criteria included those with acute allergy exacerbations, acute upper respiratory tract infections, evidence of nasal crusting, active nasal bleeding, or any other ongoing process that would temporarily impair the ability to perform smell testing. Subject recruitment involved the use of flyers and contacting patients who presented to the rhinology clinic and indicated consent to be contacted for future clinical studies. Only subjects with the known existing otolaryngologic disease were examined by an otolaryngologist prior to enrollment in the study. Each subject completed questionnaires on demographics (ie, age, sex, race, ethnicity, weight, height), medication use, medical history, surgical history, social history (eg, alcohol and tobacco use), and self-perceived olfactory and gustatory function. Patients were screened for self-reported parosmia (ie, hyposmia or anosmia) or lack thereof, and further confirmed with psychophysical testing. Patients with self-reported phantosmia or cacosmia were not included in this study, as these variables may be confounders for a pilot study examining hedonics.

Psychophysical Testing

Each subject completed all study procedures within 2 consecutive days. Each participant underwent 4 psychophysical olfaction tests: SIT-40, Sniffin’ Sticks, Snap, and Henkin Test. To account for testing fatigue, the order in which olfaction tests were completed was randomized for each subject.

The SIT-40 involves 40 multiple-choice questions (score range: 0-40) presented as microencapsulated odorant strips in a standardized “scratch and sniff” format. SIT-40 only measures identification and does not have discrimination or threshold domains.¹¹

Sniffin’ Sticks is a series of olfaction tests that use pen-like odor dispensing devices to measure threshold (T) (score range: 1-16), discrimination (D) (score range: 0-16), and identification (I) (score range: 0-16) domains.¹² The composite score (TDI) is the sum of all 3 domain scores and ranges from 1 to 48. Our study defined hyposmia as TDI < 31, and anosmia as TDI < 16.¹⁶ Sniffin’ Sticks was selected to define these categorizations because it has the greatest amount of published normative data.

Snap is a series of olfaction tests similar to Sniffin’ Sticks, but rather than pen-like devices, it delivers odors via a slide mechanism. Like Sniffin’ Sticks, Snap measures threshold (score range: 1.5-9.5), discrimination (score range: 0-22), and identification (score range: 0-16) domains.^13,14 Snap TDI score ranges from 1.5 to 47.5.

Henkin Test is a series of olfaction tests that measures detection threshold, magnitude estimation, and hedonics. Each domain of olfaction involves testing with each of 4 odorants: pyridine (dead fish), nitrobenzene (almond), thiophene (gasoline), and amyl acetate (banana). Each odor is contained in vials of 10⁻⁹ to 10⁰ concentration, as well as a vial of “pure” odorant. Vials are held approximately 2 cm from the nostril, and subjects are instructed to inhale for 2 s.

Detection (score range: 2-9) involves a forced-choice, 3-stimulant, stepwise-staircase technique. Subjects were asked to determine which of 3 vials contains the odorant. Detection began with the most dilute concentration (10⁻⁹) and if the subject could detect it, the detection score was a 9. If subjects were unable to detect the odor, the next strongest concentration was tested.

For magnitude estimation, subjects assign an odor intensity rating on a scale from 0 (least intense) to 10 (most intense). Hedonics measures the perceived pleasantness of each odor from −10 (most unpleasant) to 0 (neutral) to 10 (most pleasant). For this study, “pleasant” was defined as a smell that the participant would not mind smelling again. “Unpleasant” was defined as a smell that the participant would not want to smell again. “Neutral” was defined as having no preference either way. Both magnitude estimation and hedonics scores are the average values when measured using 10⁻², 10⁻¹, 10⁰, and pure concentrations.¹⁷

PROMs

PROMs included the Questionnaire for Olfactory Disorders Negative Statements (QODNS) and visual analog scales (VAS). The QODNS is a 17-item test, with each item scoring 0 to 3 for a total score of 0 to 51.¹⁸ Worse quality of life is indicated by higher scores. QODNS items were combined into 4 groupings according to Mattos et al.¹⁹ Each grouping was based upon the impact of olfactory dysfunction on social (QODNS items 11, 15, 16), eating (1, 3, 13), anxiety (5, 8, 9), and annoyance (4, 7) factors.

A 10 cm VAS (0 = no symptoms, 10 = worst possible symptoms) assessed the impact of olfactory loss (VAS-Olfaction) upon mood, enjoyment of food, social interactions, safety, personal hygiene, sex life, difficulty cooking, change in appetite, and change in weight.⁴ A second VAS on overall nasal and chemosensory symptom severity (VAS-SS) assessed nasal obstruction, overall sense of smell, sense of taste (bitter, sweet, salty, sour), and ability to detect flavors in food and drink.

Statistical Analyses

All data analysis was performed using SPSS v27.0.1 (IBM Corporation, Armonk, NY). Descriptive statistics were calculated for patient demographics, alcohol use, tobacco use, marijuana use, medication use, medical history, surgical history, and olfaction testing results. Continuous data were represented as either mean (standard deviation) for data that fit normality, or median (25%-75% IQR) for data that did not follow normality. Normality was assessed using Shapiro-Wilk tests. Categorical data were presented as counts (percentage of the whole sample). Henkin Testing parameters were compared between odors using related samples Friedman's Two-Way Analysis of Variance (ANOVA) by ranks. Hedonics were compared between anosmics, hyposmics, and normosmics using Kruskall-Wallis Tests. Post-hoc analyses were conducting using Dunn's pairwise tests with the Bonferroni correction. Olfaction testing scores and PROMs were correlated using Spearman's rank correlation. Statistical significance was defined by P < .05.

Multiple imputations by regression method (5 imputed datasets, 200 iterations per imputation) were used to address the incompleteness of the data. The amount of missing data was 0.12% of SIT-40 items. We assumed that the missing cases were not completely randomly distributed among cases. Because data was only missing for dichotomous variables, the most frequently assigned value (ie, mode) across all permutations was used. Each univariate imputation model contained all other variables as predictors, with the binary and categorical variables imputed using logistic and multinomial logistic regression models, respectively.

Results

Study Sample Characteristics

Of the 65 subjects enrolled, 63 (96.9%) completed all study procedures. Two (3.1%) subjects did not complete a total of 3 SIT-40 items. Demographics are summarized in Table 1. The mean age was 52.1 (16.3) years, and most subjects were female (n = 41, 63.1%) and non-Hispanic Caucasian (n = 41, 63.1%). No patients indicated phantosmia or cacosmia. Medical histories are summarized in Table 2.

Table 1.

Demographics.

Age (years)
Mean (SD)	52.1 (16.3)
Sex
Female	41 (63.1%)
Race/ethnicity
Asian	5 (7.7%)
Black or African American	17 (26.2%)
White (non-Hispanic)	41 (63.1%)
White (Hispanic)	1 (1.5%)
Other (Hispanic)	1 (1.5%)
BMI (kg/m²)
Mean (SD)	29.39 (7.97)
Alcohol use
None	23 (35.4%)
Former	4 (6.2%)
Current	38 (58.5%)
Tobacco use
None	41 (63.1%)
Former	21 (32.3%)
Current	4 (4.6%)
Marijuana use
None	41 (63.1%)
Former	16 (24.6%)
Current	7 (10.8%)

Table 2.

Medical History.

Medical history	Present/activeN (%)	Past/inactiveN (%)
Hypertension	23 (35.4)	3 (4.6)
Hypercholesterolemia	18 (27.7)	4 (6.2)
Anxiety	16 (24.6)	5 (7.7)
Depression	13 (20.0)	5 (7.7)
Asthma	10 (15.4)	2 (3.1)
Chronic rhinosinusitis	10 (15.4)	7 (10.8)
Allergic rhinitis	9 (13.8)	10 (15.4)
Positive allergy testing (blood test or skin prick)	8 (12.3)	12 (18.5)
Any cardiac problem (eg, heart failure, heart attack)	5 (7.7)	2 (3.1)
Aspirin sensitivity or allergy	5 (7.7)	2 (3.1)
Migraine	5 (7.7)	17 (26.2)
Noninsulin dependent diabetes	4 (6.2)	0 (0.0)
Head injury	2 (3.1)	13 (20.0)

Present/active indicated the patient has ongoing workup or treatment modifications for the condition listed. Past/inactive indicated that the patient no longer has issues with the disease state or has adequate and unchanging treatment. Disease states were not exclusion criteria unless they actively hindered a patient's ability to perform psychophysical olfaction testing.

Descriptive Results

Although differences in detection scores between odors were not statistically significant (P = .2), most participants were able to detect nitrobenzene at the lowest concentration (Figure 1). However, the magnitude estimation was greatest for pyridine (P < .001). Hedonic scores were worse for thiophene and pyridine (Figure 1). As shown in Figure 2 most people rated thiophene (n = 49, 75.4%) and pyridine (n = 60, 92.3%) as “unpleasant.” Although most people rated amyl acetate (n = 31, 47.7%) and nitrobenzene (n = 37, 56.9%) as “pleasant,” many rated amyl acetate (n = 34, 52.3%) and nitrobenzene (n = 28, 43.1%) as “neutral” or “unpleasant” scents.

Correlation Between Henkin Odor Domains

When examining hedonics, differences were apparent between odors. As shown in Table 3, hedonics for unpleasant odors correlated with each other (r_s = 0.60), but not with hedonics for pleasant odors. Hedonics for unpleasant odors also correlated with magnitude estimation and detection of all tested odors. In contrast, hedonics for amyl acetate and nitrobenzene did not correlate with each other. Furthermore, hedonic scores for pleasant odors generally did not correlate with magnitude estimation or detection.

Table 3.

Correlations Between Henkin Test Domains.

		Detection				Magnitude estimation				Hedonics
		Amyl Acetate	Nitrobenzene	Thiophene	Pyridine	Amyl Acetate	Nitrobenzene	Thiophene	Pyridine	Amyl Acetate	Nitrobenzene	Thiophene	Pyridine
Detection	Amyl acetate		0.73	0.69	0.55	0.39	0.45	0.45	0.47	−0.16	0.22	−0.38	−0.51
	Nitrobenzene	0.73		0.65	0.59	0.37	0.44	0.40	0.44	−0.15	0.10	−0.26	−0.47
	Thiophene	0.69	0.65		0.63	0.42	0.53	0.49	0.58	−0.02	0.09	−0.43	−0.58
	Pyridine	0.55	0.59	0.63		0.33	0.35	0.29	0.34	0.17	0.24	−0.27	−0.42
Magnitude Estimation	Amyl acetate	0.39	0.37	0.42	0.33		0.83	0.88	0.69	0.10	0.35	−0.47	−0.55
	Nitrobenzene	0.45	0.44	0.53	0.35	0.83		0.81	0.72	0.01	0.24	−0.44	−0.55
	Thiophene	0.45	0.40	0.49	0.29	0.88	0.81		0.77	−0.01	0.24	−0.58	−0.60
	Pyridine	0.47	0.44	0.58	0.34	0.69	0.72	0.77		−0.04	0.14	−0.54	−0.80
Hedonics	Amyl acetate	−0.16	−0.15	−0.02	0.17	0.10	0.01	−0.01	−0.04		0.13	−0.01	0.11
	Nitrobenzene	0.22	0.10	0.09	0.24	0.35	0.24	0.24	0.14	0.13		0.00	−0.01
	Thiophene	−0.38	−0.26	−0.43	−0.27	−0.47	−0.44	−0.58	−0.54	−0.01	0.00		0.60
	Pyridine	−0.51	−0.47	−0.58	−0.42	−0.55	−0.55	−0.60	−0.80	0.11	−0.01	0.60

*Data shown are Spearman's r_s. Bolded values were statistically significant (P < .05).

Correlation Between Henkin Odors and Validated Olfactory Tests

When comparing hedonics to validated psychophysical tests, again there were differences among odors. Hedonics for unpleasant odors (thiophene and pyridine) correlated with all domains of validated tests (Table 4). For example, patients rating pyridine with negative hedonic scores typically had normal olfaction capacity on validated testing (r_s −0.52 to −0.71). This contrasts with hedonics for pleasant odors, which mostly did not correlate with validated tests. Subjects with normosmia compared to those with hyposmia and anosmia rated pyridine to be more unpleasant (Figure 3). However, this difference in hedonics was not seen for pleasant odors. When comparing magnitude estimation and detection among odors, all generally had moderate correlations to validated tests.

Table 4.

Correlations Between Henkin Test and Other Psychophysical Olfaction Tests.

		Detection				Magnitude estimation				Hedonics
		Amyl acetate	Nitrobenzene	Thiophene	Pyridine	Amyl acetate	Nitrobenzene	Thiophene	Pyridine	Amyl acetate	Nitrobenzene	Thiophene	Pyridine
TDI	Sniffin’ Sticks	0.63	0.55	0.59	0.67	0.56	0.58	0.58	0.57	0.07	0.22	−0.50	−0.69
TDI	Snap	0.68	0.62	0.67	0.61	0.55	0.62	0.59	0.59	−0.03	0.17	−0.58	−0.70
Threshold	Sniffin’ Sticks	0.55	0.46	0.52	0.59	0.51	0.48	0.52	0.56	0.11	0.17	−0.52	−0.71
Threshold	Snap	0.51	0.54	0.64	0.58	0.54	0.58	0.52	0.59	0.03	0.21	−0.39	−0.65
Discrimination	Sniffin’ Sticks	0.58	0.43	0.59	0.55	0.41	0.51	0.48	0.45	0.03	0.16	−0.38	−0.52
Discrimination	Snap	0.62	0.59	0.69	0.57	0.48	0.57	0.56	0.54	0.01	0.11	−0.57	−0.63
Identification	Sniffin’ Sticks	0.55	0.55	0.48	0.62	0.44	0.48	0.45	0.45	0.04	0.14	−0.44	−0.59
	Snap	0.66	0.60	0.49	0.50	0.43	0.49	0.46	0.48	0.02	0.15	−0.55	−0.61
	SIT-40	0.67	0.58	0.58	0.64	0.48	0.47	0.48	0.50	0.03	0.36	−0.52	−0.63

Abbreviations: TDI: threshold, discrimination; SIT-40, Smell Identification Test-40.

*Data shown are Spearman's r_s. Bolded values were statistically significant (P < .05).

Correlations to PROMs

Hedonics provided insight into olfactory-specific PROMs (Table 5). Patients who rated pyridine as highly unpleasant mostly had normal PROM scores. Interestingly, these correlations were significant for VAS questions asking about safety (r_s = 0.27) and hygiene(r_s 0.31-0.43), which are areas that require awareness of unpleasant odors. Pyridine and thiophene hedonics were not correlated with most other aspects of olfactory VAS such as appetite or cooking. Hedonics of pleasant odors showed correlations to QODNS for amyl acetate (r_s −0.27 to −0.37), but not for nitrobenzene. Other aspects of Henkin odors such as magnitude estimation and detection showed variable associations.

Table 5.

PROM Correlations With Henkin Test Domains.

		Detection				Magnitude estimation				Hedonics
		Amyl Acetate	Nitrobenzene	Thiophene	Pyridine	Amyl Acetate	Nitrobenzene	Thiophene	Pyridine	Amyl Acetate	Nitrobenzene	Thiophene	Pyridine
QODNS	Social	−0.23	−0.08	−0.25	−0.29	−0.25	−0.23	−0.21	−0.29	−0.37	−0.06	0.15	0.26
	Eating	−0.32	−0.26	−0.34	−0.42	−0.38	−0.38	−0.40	−0.41	−0.27	−0.18	0.32	0.38
	Anxiety	−0.26	−0.21	−0.24	−0.37	−0.31	−0.32	−0.30	−0.29	−0.31	−0.17	0.21	0.27
	Annoyance	−0.18	−0.11	−0.20	−0.40	−0.25	−0.25	−0.22	−0.25	−0.27	−0.09	0.24	0.28
	Total	−0.23	−0.22	−0.25	−0.37	−0.30	−0.32	−0.29	−0.33	−0.34	−0.07	0.23	0.32
VAS-Olfaction	Mood	−0.33	−0.31	−0.26	−0.40	−0.27	−0.23	−0.25	−0.20	−0.28	−0.14	0.11	0.18
	Food enjoyment	−0.36	−0.33	−0.32	−0.42	−0.26	−0.25	−0.29	−0.33	−0.15	−0.14	0.27	0.32
	Social interaction	−0.25	−0.28	−0.30	−0.20	−0.31	−0.30	−0.22	−0.25	−0.06	−0.07	−0.02	0.14
	Safety	−0.39	−0.36	−0.39	−0.38	−0.34	−0.43	−0.37	−0.46	−0.14	−0.15	0.27	0.43
	Hygiene	−0.46	−0.41	−0.48	−0.32	−0.34	−0.35	−0.29	−0.32	0.10	−0.14	0.16	0.31
	Sex	−0.09	−0.12	−0.27	−0.09	−0.29	−0.22	−0.18	−0.17	−0.07	−0.21	0.06	0.08
	Cooking	−0.21	−0.23	−0.33	−0.38	−0.16	−0.15	−0.13	−0.14	−0.18	−0.06	0.13	0.17
	Appetite	−0.15	−0.12	−0.28	−0.33	−0.36	−0.28	−0.22	−0.21	−0.20	−0.20	0.08	0.18
	Weight	−0.14	−0.09	−0.32	−0.30	−0.29	−0.21	−0.19	−0.17	−0.31	−0.20	0.06	0.08
VAS-SS	Nasal obstruction	−0.05	−0.11	−0.16	−0.11	−0.19	−0.31	−0.11	−0.21	−0.16	0.06	−0.15	0.09
	Overall smell	−0.47	−0.39	−0.47	−0.49	−0.50	−0.55	−0.52	−0.61	−0.16	−0.22	0.47	0.57
	Taste	−0.25	−0.18	−0.25	−0.19	−0.33	−0.30	−0.43	−0.42	−0.20	0.01	0.35	0.33
	Flavors	−0.30	−0.20	−0.22	−0.25	−0.40	−0.37	−0.46	−0.48	−0.11	−0.13	0.31	0.33

Abbreviations: QODNS, Questionnaire for Olfactory Disorders Negative Statements; VAS, visual analog scale; VAS-SS, VAS on overall nasal and chemosensory symptom severity; PROMs: patient-reported outcome metrics.

*Data shown are Spearman's r_s. Bolded values were statistically significant (P < .05).

Discussion

This study examined the relationship between the hedonics of different odors and 3 commonly used psychophysical tests and olfactory-specific PROMs. These findings help contextualize the complex factors that play a role in chemosensory perception and the impact that olfaction plays in our daily lives. While threshold, discrimination, and identification domains are important to assess, we also found that hedonics provides unique relationship patterns with PROMs and validated olfactory tests, which may provide utility in characterizing olfactory dysfunction.

Our sample confirmed that most people rate amyl acetate and nitrobenzene as “pleasant,” and thiophene and pyridine as “unpleasant.” The hedonics of unpleasant odors were more uniformly interpreted as unpleasant, in contrast to the variability seen for pleasant odors. Specifically, 92% of subjects rated pyridine as unpleasant, whereas amyl acetate, a banana-like fragrance, was only rated as pleasant by 47% of subjects. Thus, there appears to be greater variability in how pleasant a given odor is rated and may be due to cultural or dietary variations.

Unpleasant odors were rated as being stronger or of greater magnitude. Specifically, pyridine was noted to have the greatest magnitude estimation (perceived odor strength) as well as the most unpleasant ratings per hedonics. When examining associations with scores of common olfactory tests, hedonics of unpleasant odors correlated to all tests. However, pleasant hedonics did not. Again, this is likely because nearly half of all subjects rated amyl acetate and nitrobenzene as neutral or even unpleasant.

Unpleasant odor hedonics had the strongest correlations with PROMs. While amyl acetate hedonics did correlate with QOD, there were not many other significant findings. Contrastingly, hedonics for pyridine correlated with VAS of areas asking about safety and hygiene, 2 functions of daily life that require the detection of unpleasant odors. Unpleasant hedonics were also the only hedonics that correlated with overall VAS addressing the overall sense of smell and flavor perception. Hedonics for unpleasant odors had few associations with PROMs addressing positive aspects of olfaction to include cooking and appetite. Altogether, these findings highlight the complexity of hedonics, which involves an interplay of not only olfaction capacity, but also odor strength, perceived pleasantness, and past experiences of an individual.

The psychophysical tests of Sniffin’ Sticks, Snap, and SIT-40 utilize mostly smells that are likely pleasant. For both Sniffin’ Sticks and Snap identification testing, 12 out of 16 (75%) items are related to foods or spices; only one item (fish) is clearly designed to be considered unpleasant.^12–14 For SIT-40, 26 of 40 (65%) items are related to foods or spices, with others considered neutral or noxious.¹¹ Considering that these tests generally focused efforts on nonnoxious stimuli, it is unsurprising that they all correlate strongly.²⁰ However, our study indicates that unpleasant odors may contribute to characterizing olfactory dysfunction impact on quality of life. Altogether, our analyses on hedonics show the need for psychophysical olfaction testing to incorporate smells of varying hedonics.

Most odorants activate both trigeminal and olfactory nerves to some degree.²¹ However, pyridine has strong trigeminal activation, while amyl acetate exhibits predominantly olfactory nerve activation at lower concentrations.^21–24 Less is known for nitrobenzene and thiophene. Noticeably, hedonics differed for pyridine and amyl acetate. However, differences in these odors may provide information regarding which aspect of the olfaction-related nerve is dysfunctional. The nerve selectivity of each odor and its relationships to psychophysical testing and PROMs is an additional dimension for future research.

From a clinical standpoint, methods to quantify the severity of phantosmia and cacosmia have remained elusive. Investigations into these symptoms currently rely on self-reporting of abnormalities because psychophysical testing best quantifies hyposmia. When psychophysical testing finds no abnormalities, it can be easy to disregard patient concerns. Our study provides an impetus to examine how hedonics testing could aid future studies investigating phantosmia and cacosmia. Furthermore, future studies can consider whether hedonics may alter retraining therapy efficacy for phantosmia versus parosmia versus hyposmia. Current smell retraining therapy typically uses pleasant odors (rose, eucalyptus, lemon, and clove).²⁵ The addition of odors with varying hedonics (ie, adding unpleasant or neutral odors) could induce unique results in the rehabilitation of olfaction capacity, especially when the primary symptom relates to the perceived pleasantness/unpleasantness of odors.

Detection testing methodology is similar to the threshold methodology of other psychophysical tests, and thus it was anticipated that these domains would correlate best. However, our study found that threshold, discrimination, and identification all correlated similarly with Henkin Test detection for all odors. Although unexpected, this finding aligns with our previous study showing that threshold, discrimination, and identification domains all correlated well with each other.²⁰ However, PROMs correlated much better to all domains of Sniffin’ Sticks, Snap, and SIT-40 as compared to Henkin Test detection scores.²⁰ Therefore, further usage of the Henkin Test to characterize olfactory function will require additional development and research.

Strengths and Limitations

Our study is unique in the cross-sectional comparison of the Henkin Test to multiple psychophysical tests and PROMs. Each patient completed study procedures in a short timeframe, and thus all data are less likely affected by temporal variations. Additionally, a wide range of olfaction capacities (anosmia to normosmia) and patient characteristics were represented in our sample. However, our study is limited by a small sample size of 65. Furthermore, our findings are limited to US participants. Cultural differences (eg, recognition of certain odors) may limit generalizability to other populations. For example, some identification tests require distinguishing different fruit odors, which may not be recognizable by populations that eat different fruits. Additional limitations stem from the recruitment and screening methods. Our study was limited to patients with known hyposmia/anosmia from the ENT clinic, as well as community volunteers of our institution. Thus, education level and socioeconomic status were not representative of the general population. A future study examining a truly random and generalizable sample would further provide insight into the value of hedonics in olfaction testing.

Additionally, our study does not capture the variability that may be seen in populations with different etiologies, severity, and duration of olfactory dysfunction. For example, our study assessed and summarized common medical comorbidities that could impact olfactory function for informational purposes. However, controlling for each medical condition as a confounder would require thousands of patients. Thus, our study cannot comment on how each medical comorbidity may affect hedonics testing. This is an important area for future research, considering a large amount of published literature involving the Henkin Test. Lastly, Henkin Test only examines 4 odors of varying hedonic levels. Because hedonics constitutes a spectrum, future studies should examine a wider variety of odors to determine what unique utility may arise for characterizing olfactory dysfunction. Specifically, many of our participants rated the “pleasant” odors as “unpleasant” or “neutral,” and thus examining odorants that are more widely considered as “pleasant” may reveal new and interesting information.

Conclusions

The psychophysical olfaction tests Sniffin’ Sticks, Snap & Sniffin’ Sticks, and SIT-40 correlate well with Henkin Test. Therefore, our study aids in contextualizing the findings of the large amount of published literature that used the Henkin Test. However, our study highlights that the detection threshold might be less useful for olfaction testing compared to other threshold testing from other psychophysical tests. The use of unpleasant odors showed unique relationship patterns with PROMs and validated olfactory tests, which may provide utility in characterizing olfactory dysfunction. The incorporation of more unpleasant odors in current psychophysical olfactory testing may provide enhanced information regarding the patient impact of olfactory dysfunction. Future studies should then examine how hedonics may be used to investigate and treat phantosmia and cacosmia.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following conflicts of interest with respect to the research, authorship, and/or publication of this article: Rodney J. Schlosser, MD and Shaun A. Nguyen, MD are consultants for Cyrano Therapeutics, Inc.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Cyrano Therapeutics, Inc. (Delray Beach, Florida, USA).

ORCID iDs

Michael C. Shih

Rodney J. Schlosser

References

Tong

Wong

Zhu

Fastenberg

Tham

. The prevalence of olfactory and gustatory dysfunction in COVID-19 patients: a systematic review and meta-analysis. Otolaryngol Head Neck Surg. 2020;163(1):3–11.

Desiato

Levy

Byun

Nguyen

Soler

Schlosser

. The prevalence of olfactory dysfunction in the general population: a systematic review and meta-analysis. Am J Rhinol Allergy. 2021;35(2):195–205.

Allensworth

Schlosser

Smith

Mace

Soler

. Use of the diet history questionnaire III to determine the impact of dysosmia on dietary quality. Int Forum Allergy Rhinol. 2022;12(6):849–858.

Desiato

Soler

Nguyen

, et al. Evaluating the relationship between olfactory function and loneliness in community-dwelling individuals: a cross-sectional study. Am J Rhinol Allergy. 2021;35(3):334–340.

Kohli

Soler

Nguyen

Muus

Schlosser

. The association between olfaction and depression: a systematic review. Chem Senses. 2016;41(6):479–486.

Choi

Jang

Kim

Hur

Ference

Wrobel

. Association between olfactory dysfunction and mortality in US adults. JAMA Otolaryngol Head Neck Surg. 2021;147(1):49–55.

Perry

Al Ain

Raineki

Sullivan

Wilson

. Development of odor hedonics: experience-dependent ontogeny of circuits supporting maternal and predator odor responses in rats. J Neurosci. 2016;36(25):6634–6650.

Leopold

. Distortion of olfactory perception: diagnosis and treatment. Chem Senses. 2002;27(7):611–615.

Gary

Gallagher

Joseph

Reed

Gudis

Overdevest

. Qualitative olfactory dysfunction and COVID-19: an evidence-based review with recommendations for the clinician. Am J Rhinol Allergy. 2022:19458924221120117. doi: 10.1177/19458924221120117.

10.

Ferdenzi

Bousquet

Aguera

, et al. Recovery from COVID-19-related olfactory disorders and quality of life: insights from an observational online study. Chem Senses. 2021;46:bjab028. doi: 10.1093/chemse/bjab028.

11.

Doty

Shaman

Kimmelman

Dann

. University of Pennsylvania smell identification test: a rapid quantitative olfactory function test for the clinic. Laryngoscope. 1984;94(2):176–178.

12.

Hummel

Sekinger

Wolf

Kobal

. ‘Sniffin' sticks’: olfactory performance assessed by the combined testing of odor identification, odor discrimination, and olfactory threshold. Chem Senses. 1997;22(1):39–52.

13.

Jiang

Liang

. A pilot study of the snap & sniff threshold test. Ann Otol Rhinol Laryngol. 2018;127(5):312–316.

14.

Doty

Wylie

Potter

Beston

Cope

Majam

. Clinical validation of the olfactory detection threshold module of the Snap & Sniff(R) olfactory test system. Int Forum Allergy Rhinol. 2019;9(9):986–992.

15.

Henkin

Powell

. Increased sensitivity of taste and smell in cystic fibrosis. Science. 1962;138(3545):1107–1108.

16.

Oleszkiewicz

Schriever

Croy

Hahner

Hummel

. Updated Sniffin' Sticks normative data based on an extended sample of 9139 subjects. Eur Arch Otorhinolaryngol. 2019;276(3):719–728.

17.

Henkin

. An open-label controlled trial of theophylline for treatment of patients with hyposmia. Am J Med Sci. 2009;337(6):396–406.

18.

Simopoulos

Katotomichelakis

Gouveris

Tripsianis

Livaditis

Danielides

. Olfaction-associated quality of life in chronic rhinosinusitis: adaptation and validation of an olfaction-specific questionnaire. Laryngoscope. 2012;122(7):1450–1454.

19.

Mattos

Edwards

Schlosser

, et al. A brief version of the questionnaire of olfactory disorders in patients with chronic rhinosinusitis. Int Forum Allergy Rhinol. 2019;9(10):1144–1150.

20.

Shih

Soler

Germroth

Snyder

Nguyen

Schlosser

. Comparison of validated psychophysical olfactory tests and olfactory-specific quality of life. Int Forum Allergy Rhinol. 2022;Forthcoming. doi: 10.1002/alr.23012. Epub ahead of print.

21.

Doty

Brugger

Jurs

Orndorff

Snyder

Lowry

. Intranasal trigeminal stimulation from odorous volatiles: psychometric responses from anosmic and normal humans. Physiol Behav. 1978;20(2):175–185.

22.

Foguem

Lemdani

Huart

. Parkinson's disease in elderly patients: lessons from odour detection thresholds on olfacto-trigeminal interaction. Rhinology. 2018;56(2):127–132.

23.

Cui

Evans

. Olfactory event-related potentials to amyl acetate in congenital anosmia. Electroencephalogr Clin Neurophysiol. 1997;102(4):303–306.

24.

Geisler

Murphy

. Event-related brain potentials to attended and ignored olfactory and trigeminal stimuli. Int J Psychophysiol. 2000;37(3):309–315.

25.

Hummel

Rissom

Reden

Hahner

Weidenbecher

Huttenbrink

. Effects of olfactory training in patients with olfactory loss. Laryngoscope. 2009;119(3):496–499.