Introduction

Because olfactory dysfunction is common in the general population (approx. 20 % of the population exhibit hyposmia; [1]) there is need for adequate assessment of olfactory function as an essential basis for its therapy. Patients’ self-estimation about this sense is notoriously unreliable and is problematic for studies requiring follow-up investigations [15]. Up to now, a large number of tests exist which allow estimation of smell function. Many of them depend on identification of specific odors (e.g., UPSIT, CC-SIT, ETOC), others rely additionally on threshold testing or discrimination of different odorants (“Sniffin’ Sticks”). These psychophysical tests require good compliance and get to their limits when dealing with children, patients with cognitive impairment, or patients with different linguistic/cultural background. For these cases, electrophysiological methods are available. Recording of olfactory event-related potentials, for instance, provides objective evidence for the presence of olfactory function [9, 18]. These techniques are complex, time-consuming and costly, and are only used routinely in highly specialized centers.

Another approach to olfactory function is to measure the sniff pressure over time while inhaling either non-odorized air or an odor stimulus. It is well known that odor characteristics affect sniffing behavior, i.e., sniff airflow and duration increases with lower odor concentrations and more pleasant odors and vice versa [17]. Based on these ideas, a test kit was developed where the negative pressure during one sniff was measured, assuming that the area under the curve would be smaller the more intense the subject/patient perceives an odor. Using well-defined odorants/concentrations and comparing the results to non-odorized air, one could estimate the smell ability or rather the degree of olfactory impairment. This Sniff Magnitude Test (SMT, CompuSniff, LLC, Cincinnati, Ohio, USA) has already been shown to have good correlation with established tests such as the UPSIT, an odor memory test (OMT), and an odor detection threshold test [5, 19].

The “Sniffin’ Sticks”, the most extensively used olfactory test kit in Europe, comprises the three categories: odor threshold, odor discrimination and odor identification. It has not been studied for correlation with the Sniff Magnitude Test, yet. Therefore, the aim of this study was to compare these two tests and investigate the usability of the Sniff Magnitude Test in clinical routine.

Materials and methods

Participants of this study were healthy subjects and patients recruited from the Smell and Taste division of the ENT university clinic in Dresden/Germany. After a detailed interview, all patients were examined thoroughly by an ENT specialist, including endoscopy of the nasal cavity. Hereafter, they were informed about intention and risks of the study. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study. The study was approved by the Ethics Committee of the Medical Faculty of the TU Dresden.

Out of the 70 participants, there were 21 participants without subjective olfaction disorder (the control group, consisting of 10 female; mean age 54 years) and 49 patients (29 female; mean age 59 years). After signing the informed consent, patients were asked to fill out two neurophysiologic tests: the “Minimal Mental Status Test” and the “Trailmaking Test”. The former aimed to rule out dementia. A maximum of 30 points can be achieved, inclusion criterion for this study was a score >25 points. For editing the Trailmaking test, the participant had to connect numbers (1–25) on a sheet of paper (part a) and alternating numbers and letters 1–13 and A–L; part b) in a limited time (<60 s). Here, processing speed and concentrated attention were tested. Hereafter, smell ability was tested.

Olfactory testing

“Sniffin’ Sticks” In the first session, the “Sniffin’ Sticks” test kit was applied, a validated psychophysical method to assess the olfactory function consisting of phenylethanol odor threshold testing, odor discrimination and odor identification tests [8]. The results of the individual tests were summated to the so-called TDI score (with a maximum score of 48 points), which reliably indicates the degree of olfactory function. A score higher than 30.5 points represents normosmia, a score higher than 16.5 and less than 30.75 points stands for reduced olfactory function in terms of hyposmia, and a score of less than 16.75 points indicates functional anosmia [10]. In total, performing the “Sniffin’ Sticks” requires approximately 30 min.

Sniff Magnitude Test (see [4]): In a second session, the Sniff Magnitude Test (WR Medical Electronics, Maplewood, MN, USA) was applied to every participant. The test kit consists of four canisters, one containing no odor and three containing an odor diluted in mineral oil. The odor stimuli included methylthiobutyrate (MTB, 98 %) with an unpleasant smell like rotten meat, isoamylacetate (IAA, 99 %), reflecting a more pleasant odor like banana and ethylmercaptopropionate (EMP (99 %) that smells burnt. The participants wore a birhinal nasal cannula which measured the negative air pressure associated with sniffing. Having connected the canisters and the nasal cannula to a laptop computer, the lid of a canister would open milliseconds after detecting a sniff via a pressure transducer and the odor stimulus/odorless air within the canister would be exposed. The participants held the respective canister approx. 2 cm under the nose and were told to inhale a single breath through the nose until a smell was perceived. Every 10 ms pressure was measured and recorded until reaching ambient values, which finally led to a closure of the canister-lid. Each participant passed through three blocks of trials with each of the four canisters (the odorless canister was presented twice, resulting in three trials for each of the odorants and six odorless trials). Each trial generated a measure of sniff magnitude defined as the sum of negative pressure values during one sniff. The “Sniff Magnitude Ratio” (SMR) is the sum of the sniff magnitudes for the odor stimuli divided by the sniffs of odorless air. For evaluation, the mean SMR (for the three different odorants) was taken into account as well. In case of suppression of the airflow due to perception of an odor, the SMR would be smaller than one. The higher the SMR, the less the suppression of a sniff of an odor, reflecting a decreased sense of smell (Fig. 1).

Fig. 1
figure 1

SMR scores versus TDI scores, separately for controls (squares) and patients (circles)

Full size image

Statistical analysis

For statistical analyses, SPSS (Statistical Packages for Social Sciences, version 19.0, SPSS Inc., Chicago, Ill., USA) was used. Comparisons between two groups of data were performed using t tests. Univariate analyses of variance were used for comparisons of olfactory function between the three groups of participants with normosmia, hyposmia, and anosmia (defined by the Sniffin’ Sticks scores); post hoc testing was based on Bonferroni tests. Correlation analyses were performed according to Pearson. The alpha level was set at 0.05.

Results

All together, the results of 70 participants were evaluated. Within the group of patients (n = 49), there were 37 % idiopathic, 37 % postinfectious, 16 % posttraumatic and 10 % sinonasal caused olfactory dysfunctions. One of them were normosmic, 39 were hyposmic, and 9 were functionally anosmic. Within the control group, 17 participants exhibited normosmia and 4 of them had a decreased sense of smell (hyposmia). All subjects fulfilled the inclusion criteria by passing the “Mini Mental Status Test” and the “Trailmaking Test”.

Regarding the “Sniffin’ Sticks” scores, statistical analyses showed the expected significant difference between controls and patients, in the total TDI score as well as for all individual subtests (t 68 > 5.8, p < 0.001). Among the participants from the control group a negative correlation between age and the TDI score was found (r 21 = −0.48; p = 0.028).

With regard to the “Sniff Magnitude Test” analyses revealed a significant difference between healthy controls and patients (t 54 = 3.09, p < 0.003). In addition, it discriminated between anosmic, hyposmic, and normosmic participants [F(2,55) = 6.22, p = 0.004; Bonferroni tests: anosmia vs normosmia: p = 0.003; anosmia vs hyposmia: p = 0.17; hyposmia vs normosmia: p = 0.051] [normosmia: M = 0.48, SEM (standard error of means) = 0.07; hyposmia: M = 0.65, SEM = 0.03; anosmia: M = 0.81, SEM = 0.08], meaning normosmic participants revealed the smallest SMR values, whereas anosmic participants showed significantly higher SMR. Importantly, the sniff magnitude values showed significant correlation with the TDI score (r 56 = −0.53, p < 0.001), and each subtest (threshold: r 56 = −0.47, p < 0.001; discrimination: r 56 = −0.44, p = 0.001; identification: r 114 = −0.38, p = 0.004).

Discussion

The present study aimed to investigate the correlation between a standard olfactory test (“Sniffin’ Sticks”) and a test kit based on a different principle, the Sniff Magnitude Test. Furthermore, the validity of the latter test was investigated.

The “Sniffin’ Sticks” are a well-established tool for assessing olfactory function since more than 15 years [11, 14] and have been investigated for their reliability and validity (e.g., [10]). Comparing it to the “Sniff Magnitude Test” a significant correlation could be seen between the two tests, i.e., the higher the TDI score the lower the SMR. Overall, SMR values discriminated between patients with anosmic and normosmic patients. Nevertheless, it has to be stated that some of the subjects showed a striking difference between the results of the two tests, i.e., some of the normosmic subjects happened to gain high SMR values, whereas other patients with a low TDI score still showed small SMR. A possible explanation for this phenomenon lays in the high intensity of the odor stimuli of the “Sniff Magnitude Test” compared to the odorants used in the “Sniffin’ Sticks”.

It is a well-established fact that the sniffing behavior can be influenced by an odor. Gudziol, for example, showed a significant alteration of the breathing pattern after presenting hydrogen sulfide [6]. Here, the concentration of an odorant seems to play an important role, i.e., sniffing volume is inversely proportional to stimulus concentration [13].

Another aspect that has to be considered for individual differences between the two tests is that tests of odor discrimination or odor identification (as used in the “Sniffin’ Sticks” test as well as the UPSIT) are closely related to cognitive functions (Hedner, Larsson et al.) and therefore are subject to various modulating factors. According to Dulay, the SMR test is not affected by any of the major cognitive variables such as working memory, cognitive speed, and verbal retrieval [2]. This factor gains relevance when dealing with elderly, neurologic or psychiatric patients [3, 12]. It is known that patients with neurologic diseases exhibit olfactory loss to a certain extent, e.g., Parkinson’s disease, Alzheimer’s disease, or Huntington’s disease [7, 16, 20]. To address the olfactory components of those, a stable test kit is required that minimizes the influence of cognitive co-factors.

In this study, to extend the pool of olfactory diagnostic tools, for the first time the “Sniff Magnitude Test“ was compared to the well-established “Sniffin’ Sticks” in a controlled trial. Practically, the “Sniff Magnitude Test” was easy to handle and could be performed in a comparatively short time. Furthermore, neither linguistic problems nor concentration deficits did affect the test results. The usefulness could be confirmed by detecting a correlation between these tests and the capability of grouping patients into normosmia, hyposmia and anosmia, when regarding the whole study population. Nevertheless, on an individual level, results of some of the patients were not in accordance with the results produced by the “Sniffin’ Sticks”. The reason here for probably lies in the fact, that a complex psychophysiological test as the “Sniffin’ Stick” inquires different aspects of olfaction, i.e., threshold, discrimination and identification, asking for processing on an integrative level. The “Sniff Magnitude Test” in contrast is a reflex-driven test which might as well be altered by hedonic aspects of an odor.

In the current experiment, we did not look for effects of sex, mostly because the sample size is too small to be really meaningful. In addition, when looking for sex-related effects for either group (controls or patients) no significant effect came up (all p’s > 0.16). So, on the one hand, sex-related effects were not the focus of the manuscript, and on the other hand, the sample size became probably too small to provide meaningful results, so that we did not include such analyses in the current manuscript.

The major inclusion criterion for controls was that people told us that their sense of smell was normal. So, although this approach made it more difficult for the psychophysical tests and the SMT to discriminate pathology from non-pathology still the tests were able to do just that. The currently used inclusion criterion seems also to be a good reflection of the clinical situation where we see a continuum between a normal sense of smell, hyposmia and functional anosmia. Having said that, even when we used only healthy subjects with normosmic scores for analyses (which resulted in the exclusion of four subjects scoring in the upper hyposmic range) the results did not change.

In conclusion, compared to the “Sniffin’ Sticks”, the “Sniff Magnitude Test” gives less information on clinically relevant olfactory function and impairment and is less sensitive for detecting this purpose. The same aspect was seen by Frank et al. when comparing the UPSIT with the Sniff Magnitude Test. Even hough the diagnostic value of the Sniff Magnitude Test could be proven it showed a smaller sensitivity than the UPSIT [5]. However, addressing certain patients (e.g., with neurological/psychiatric diseases, linguistic problems, elderly) could be a helpful tool for diagnostics. The focused application on these patients and its informative value needs to be investigated in further studies.