Trends in Olfactory Threshold Test Results and Limits of Olfactory Testing in Patients with Parosmia

Parosmia, a typical symptom of qualitative olfactory dysfunction, is often associated with quantitative olfactory dysfunction. Manifesting often during the recuperation phase of olfactory dysfunction, it has been noted as a positive factor in some cases [1, 2]. However, the literature on parosmia is notably scarce compared to that on quantitative olfactory dysfunction. The absence of standardized methods for objective and quantitative assessment has led to healthcare professionals underestimating its impact. Nonetheless, the decline in quality of life and depressive tendency due to parosmia can surpass that associated with quantitative olfactory dysfunction [3‒5].

Prevalence figures for parosmia range from 1 to 5% in the general population [6‒8] and escalate to 13 to 59% in individuals with olfactory dysfunction [9‒12]. This wide range reflects differences in parosmia diagnosis assessment methods. Despite limitations in meta-analyses due to diagnostic and assessment method heterogeneity, given its high prevalence, particularly among those with olfactory dysfunction, understanding the pathogenesis and epidemiology of parosmia can enhance olfactory treatment and patients’ quality of life. This report explores the prevalence of parosmia at our clinic, analyzes trends in olfactory threshold test results among parosmia patients at their initial visit, and considers the diagnostic potential of olfactory testing for parosmia.

A total of 467 patients who visited the olfactory outpatient clinic in our hospital from April 2009 to September 2015 were included in the study. The patients were divided into two groups, those without parosmia and those with parosmia, based on their responses to a questionnaire asking, “Do you notice strange smells?” on their first visit. The Jet Stream T&T Olfactometer [13, 14] was used for the olfactory threshold test, with the left and right nostrils tested and measured separately. The T&T Olfactometer, a standard olfaction test in Japan, employs five odorants at 7 to 8 different concentrations to measure detection and recognition thresholds. Administered via the Jet Stream device, it facilitates bilateral olfactory measurements by delivering odorants directly into the nostrils. The average detection threshold, the average recognition threshold, and the difference between the left and right detection thresholds were compared. Based on the average recognition threshold, patients were diagnosed as normosmia for values between −2.0 and 1.0, mild olfactory dysfunction for values between 1.1 and 2.5, moderate olfactory dysfunction for values between 2.6 and 4.0, severe olfactory dysfunction for values between 4.1 and 5.5, and anosmia for values between 5.6 and 5.8 [13]. Patient gender, age, time from symptom onset to hospital visit, and smoking status were compared as patient characteristics. Olfactory detection thresholds were compared to diagnose parosmia from olfactory test results. Patients with mild, moderate, severe olfactory dysfunction, or anosmia were selected for this comparison. The gap between the T&T detection threshold and the recognition threshold values and the number of correct answers in the Open Essence test [13, 15] were compared between the parosmia and non-parosmia groups. The Open Essence test, a card-type odor identification test, consists of 12 distinct odors encapsulated in cards with multiple-choice options and assesses odor identification ability. Mann-Whitney tests and Fisher’s exact tests were employed for statistical comparisons and graphs were created with Prism 9 (GraphPad Software). p values under 0.05 were considered statistically significant. This study complies with the Declaration of Helsinki and was approved by the Ethics Committee of the Jikei University School of Medicine, approval No. 33-159(10774).

Of the 467 patients, 60 had incomplete questionnaire data, 333 were classified as without parosmia, and 74 were classified as parosmia, resulting in an approximate prevalence of 18% for parosmia. Patient characteristics showed that the group with parosmia had a significantly higher proportion of females (p = 0.018), and the duration from symptom onset to hospital visit was significantly shorter (p = 0.036). There were no significant differences in age or smoking status between the two groups. Patient statistics are shown in Table 1.

The prevalence of parosmia differed according to the cause of parosmia. Post-traumatic and post-infectious had the highest percentage of parosmia (28% and 20%, respectively), while sinusitis had the lowest percentage of parosmia (11%). The results are shown in Figure 1.

In olfactory threshold testing, both mean detection and recognition thresholds were significantly lower in the parosmia group compared to the non-parosmia group (p = 0.014 and p = 0.005, respectively). The left-right difference in detection threshold was also significantly greater in the parosmia group (p = 0.012). The results are shown in Figure 2.

No significant differences were found in the identification ability tests, including the gap between detection and recognition threshold values and the number of correct answers in the Open Essence test. The results are shown in Figure 3.

Our findings indicate a higher prevalence of parosmia among females, which is consistent with previous reports [11]. This gender disparity is also observed in COVID-19-related parosmia, where reports [16‒18] highlight a greater occurrence in females. It is conceivable that women are more aware of olfactory dysfunction than men, making them more likely to recognize the symptoms of parosmia and subsequently seek medical attention. Moreover, qualitative dysfunctions may manifest symptoms more prominently or be more perceptible than quantitative dysfunctions, potentially contributing to earlier hospital visits.

The high prevalence rate of parosmia in post-traumatic and post-infectious olfactory dysfunction has been mentioned in prior reports [9, 10, 12, 19‒23]: 29–55% for traumatic, 23–51% for post-infectious, and 17–28% for sinusitis. In the present study, the same trend was observed, although the overall numbers were lower. The higher prevalence of parosmia after trauma or infection implies that central or sensorineural olfactory dysfunction may lead to an increased likelihood of parosmia development.

The better olfactory test results in the parosmia group may indicate that there needs to be some degree of preserved sense of smell for parosmia symptoms to be perceived. In other words, the more severe a quantitative olfactory dysfunction is, the harder it is to perceive qualitative olfactory symptoms. The greater laterality of olfactory function in the parosmia group might be attributed to the heterogeneity of damaged olfactory neurons, potentially resulting in an uneven olfactory perception and manifestation of parosmia symptoms. Considering that specific regions of the olfactory epithelium could be responsible for detecting certain odor molecules [24], heterogeneity in olfactory epithelial damage could theoretically lead to distorted odor perception.

The observed left-right difference in detection thresholds hints at a potential risk for parosmia, prompting a need to explore its underlying mechanisms. It is plausible that during the integration of signals from the left and right nostrils, disparate odors may be perceived and presented as parosmia symptoms. This phenomenon could be indicative of a central mechanism, possibly involving a delicate balance between signals from the left and right olfactory pathways. The controversy surrounding whether olfactory perception is the same with bilateral and ipsilateral stimulation adds complexity to this discussion [25]. Several studies report that the same odorant can be perceived differently by each nostril [26], and split-brain subjects demonstrate different odor perceptions with unilateral olfactory stimulation [27].

Evaluating parosmia solely based on differences between detection and recognition threshold values may be insufficient. Traditional tests, such as the T&T Olfactometer, do not fully capture the qualitative nature of parosmia, which is why subjective measures, like a parosmia score, are increasingly important. A parosmia score – such as the Self-Administered Odor Questionnaire for Parosmia (SAOQ-P) previously reported by the author [28] – could help quantify the patient’s experience by assessing the severity and frequency of parosmia episodes. By incorporating such tools, clinicians can better track symptom progression and recovery, offering more personalized treatment approaches. Future studies should explore the integration of a parosmia scoring system in diagnostic protocols to improve the accuracy of parosmia diagnosis and management.

Given the significant left-right differences observed in the current study, individually measuring olfactory thresholds for each nostril could unveil patterns or asymmetries that contribute to a more comprehensive understanding of parosmia. A unilateral occlusion test could offer therapeutic potential by identifying which nostril is more likely to induce parosmia. For instance, if a patient reports more frequent parosmia episodes when using one nostril, occluding that nostril could alleviate symptoms. A simple approach could involve asking patients to manually occlude each nostril during parosmia episodes to assess symptom changes, providing a practical, non-invasive diagnostic tool. This method could potentially identify which nostril is more likely to induce parosmia, leading to targeted therapeutic interventions such as unilateral olfactory cleft blocking [29]. Additionally, unilateral olfactory tests, such as the Jet Stream T&T Olfactometer used in this study, could be considered for parosmia assessment as they may help detect olfactory laterality and identify potential treatment targets.

The absence of differentiation in the identification test may stem from the fact that the odorants employed in current olfactory tests consist of a single or several types of molecules. Consequently, the number of olfactory receptors stimulated should be notably lower than that triggered by natural odorants that are typically comprised of tens or hundreds of diverse molecules. This limited stimulation might explain the absence of apparent parosmia symptoms in the test results. Moreover, it is crucial to note that the severity of qualitative olfactory dysfunction does not exhibit a direct correlation with quantitative olfactory dysfunction. In fact, this study reveals a potential negative correlation since the parosmia group demonstrated lower olfactory threshold values. Given these intricacies, evaluating parosmia through current olfactory tests proves challenging. Relying solely on standardized tests might not capture the nuanced nature of parosmia and the results emphasize the need for a more subjective and patient-centered approach to assessment.

A notable limitation of this study is that the questionnaire did not distinguish between parosmia and phantosmia; it simply asked whether the patient had perceived strange smells. This aspect holds significant importance as many studies, including the present one, have either failed to make this important distinction or provided ambiguous definitions. Furthermore, the questionnaire’s singular focus on the initial diagnosis with the question “Do you notice strange smells?” restricted the evaluation of symptom progression and severity. In future research, it is imperative to implement a methodology that allows for the continuous assessment of parosmia over time. Such an approach would serve as a valuable tool in gauging the recovery process and conducting prognostic assessments, providing a more comprehensive understanding of parosmia.

In our study, the prevalence of parosmia was 18% and found to have a female-predominance. Parosmia also more often followed trauma or infection episodes and was associated with a shorter time from symptom onset to consultation. Our findings emphasize the importance of preserving olfactory function for parosmia symptoms to manifest, as well as the significant role of subjective assessments in diagnosis. The observed left-right differences in detection thresholds suggest that olfactory lateralization may contribute to the onset of parosmia, warranting further investigation. Our findings also suggest that a unilateral occlusion test, which identifies the nostril more likely to induce parosmia, could facilitate more targeted treatment approaches.

This study was reviewed and approved by the Ethics Committee of the Jikei University School of Medicine, Approval No. 33-159(10,774). This study has been granted an exemption from requiring written informed consent by the Ethics Committee of the Jikei University School of Medicine, Approval No. 33-159(10,774).

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

M.T., E.M., and N.O. designed the study. M.T. wrote the draft of the manuscript. M.T., E.M., R.S., Y.K., H.T., M.N., and Y.T. acquired the data. All authors reviewed and approved the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.