Lateralisation of nasal cycle is not reflected in the olfactory bulb volumes and cerebral activations

Nasal cycle (NC) is a rhythmic change of lateralised nasal airflow mediated by the autonomous nervous system. Previous studies reported the dependence of NC dominance or more patent side on handedness and hemispheric cerebral activity. We aimed to investigate firstly the possible lateralised effect of NC on olfactory bulb volume and secondly the association of NC with the lateralised cerebral dominance in terms of olfactory processing. Thirty‐five subjects (22 women and 13 men, mean age 26 ± 3 years) participated in the study. NC was ascertained using a portable rhino‐flowmeter. Structural and functional brain measurements were assessed using a 3T MR scanner. Vanillin odorant was presented during functional scans using a computer‐controlled olfactometer. NC was found to be independent of the olfactory bulb volumes. Also, cerebral activations were found independent of the NC during odorant perception. NC potency is not associated with lateralised structural or functional differences in the cerebral olfactory system.


| INTRODUCTION
The nasal cycle is a shift of nasal airflow mediated by alternating dilation and constriction of veins in the nasal mucosa, by action of the sympathetic nervous system (Hanif et al., 2000).The nose exhibits an asymmetrical airflow with the more patent nostril alternating from one nasal passage to the other over a period of hours (Eccles, 2000).The physical reason for such a change is the asymmetric congestion of the nasal swelling bodies, which leads to partial blockage of one side of the nose (Widdicombe, 1986).The functional purpose of the nasal cycle remains uncertain, but it may involve an increased influx of odour molecules into the nasal cavity, resulting in heightened perception.This heightened perception is then likely to manifest as observable ipsilateral brain activations in fMRI scans.Among many things, air conditioning and filtering of the air have been discussed as one of its possible functions (Abolmaali et al., 2013;Hanif et al., 2000;Letzel et al., 2022).Sobel and colleagues showed that the nasal cycle may help in more differentiated olfactory perception by mediating lateralised nasal airflow (Sobel et al., 1999).Results from their study showed that the majority of the study population (17 of 20) exhibited a nasal cycle though no preferential side of nostril was established.Another study by the same group, using a portable rhino flowmeter or 'Nasal Holter' device, examined 33 subjects and found that the left side was patent for a longer time than the right side (Kahana-Zweig et al., 2016).The nasal cycle changes with body posture (Hasegawa, 1982), age (Williams & Eccles, 2015), handedness (Searleman et al., 2005), ultradian rhythms of cerebral activity (Werntz et al., 1983) and is also evident in various non-olfactory related brain measures as seen using EEG (Shannahoff-Khalsa, 1993) and cognitive task performance.Another EEG study revealed nasal cycle as an important parameter for left and right cerebral differences when awake and during sleep.This study showed a correlation between hemispheric dominance and the nasal cycle (Shannahoff-Khalsa, 1993).Cerebral hemispheres exhibit functional and structural asymmetry in terms of performance based on spatial processing and logical processing, which might be due to handedness differences (Goel, 2019;Sun & Walsh, 2006).Previous studies showed a correlation between nasal airflow and handedness (Searleman et al., 2005).For the majority of the population, the more patent side of nostril was found to be positively correlated with the dominant hand.Evidence from a study of 19 subjects discovered greater EEG activity in the hemisphere contralateral to the patent nostril (Werntz et al., 1983).This was confirmed in another, lager study (n = 126, right-handed), which also observed a link of more patent side with the contralateral hemisphere.Enhanced verbal task performance was found at times of right nostril being more patent whereas greater spatial task performance was found at times of left nostril (Telles & Samantaray, 2008).Given evidence from previous studies, it is clear that NC has important role in upper airway humidification and keeping out foreign bodies.However, only a few studies focus on functional implications on ipsilateral brain activations, which mainly utilising EEG too.Hence, we assume that if the nasal cycle has a more patent side, then one side should receive more odorant stimulus than the other side, leading to more olfactory activations in the ipsilateral hemisphere.We hypothesise that the nasal cycle could affect the anatomy of central brain structures such as OB volume (Buonviso et al., 2006) or the functionality of higher order olfactory processing areas of the brain in humans.The OB is a specialised structure and is the first odorant decoding centre (Scott et al., 1993).It is the most important relay station in the odorant pathway, integrating peripheral and central olfactory information.OB volume is associated with olfactory function (Hummel et al., 2011;Mazal et al., 2016).Reduced OB volumes have been related to a decrease in olfactory sensitivity (Negoias et al., 2010).OB volume increases significantly after recovery from olfactory loss or with olfactory training in healthy subjects (Negoias et al., 2017).
The aim of the present study was firstly to examine effects of a lateralised nasal cycle on OB volumes.In other words, we expected subjects with more patent nasal cycle to have larger OB volume.Secondly, we aimed to investigate the lateralised effect of nasal cycle on brain activations when perceiving a selective olfactory stimulant.

| METHODS
Thirty-five subjects (22 women and 13 men, mean age 26 ± 3 years) participated in the fMRI study.The study was approved by the ethics committee at the local institute in Dresden, Germany (approval number EK558122019) and was conducted according to the Declaration of Helsinki.All participants provided written informed consent.Subject recruitment was done with the help of flyers displayed at the university campus.All participants were right-handed as ascertained with the Edinburgh Handedness Inventory (Oldfield, 1971).We recruited healthy participants with exclusion criteria of olfactory dysfunctions, clinically diagnosed depression, physical or mental illness, regular intake of medication (apart from contraceptives), alcohol consumption on a daily basis and history of any neurological diseases affecting the sense of smell.We additionally also asked about the menstrual cycle in female participants to eliminate any bias towards sensitivity to odorants.The health status of participants was ascertained using a detailed medical history (Welge-Luessen et al., 2013).Normosmia (normal olfactory sensitivity) was ascertained using a 16-item odorant identification test (score ranging from 0 to 16) from the 'Sniffin' Sticks' olfactory test kit.This test includes a forced choice paradigm where subjects have to identify 16 odorants at supra-threshold concentrations using flash cards with four descriptors each (Oleszkiewicz et al., 2019).Mean identification score of subjects was (mean ± standard deviation) 13.3 ± 1.5, score rank ranging from (1-16).Nasal dominance or more patent side was ascertained using a portable rhinoflowmeter, the 'Nasal Holter', developed by Sobel and colleagues (Kahana-Zweig et al., 2016).This small, portable and easy to use device enables long-term recordings.Laterality index (LI), its value reflecting more patent side of nasal airflow, was calculated using MATLAB code developed by Sobel and colleagues (Kahana-Zweig et al., 2016).The laterality index measures the flow ratio between the left and right nostrils.LI is calculated using the following equation: LI = (Flow R À Flow L )/ (Flow R + Flow L ) for every minute.'LI' was used to ascertain which was the more patent side during awake phases (Figure 1).Positive values of the LI refer to more patent right sided, and negative values refer to more patent left-sided.To ascertain the NC patterns, 45 participants initially put on the Nasal Holter for 24 h, and these participants were then scanned according to their 24 h NC patterns and all, except 10 participants who were not included in the fMRI study, even on the day of the scan showed similar NC patterns (from 24 h), which was ascertained by putting on the device briefly before scanning.We looked only at the patterns of NC during the awake phase (Figure 1).
Data were acquired on a 3T MRI scanner (model 'Trio', Siemens Medical Systems, Erlangen, Germany) using a 32-channel head coil.For the assessment of the OB, we used a 2D coronal T2-weighted sequence with parameters: 30 volumes, TR = 6770 ms, TE = 84 ms, slice thickness = 1 mm, voxel size = 0.7*0.5*1.0 mm.Additionally, a T2-weighted sequence, focusing particularly at the nasal cavity to rule out signs of nasal significant septum deviation or nasal polyposis, among others and to confirm the presence of NC (Figure 2).It has been shown that nasal cycle is affected by septal deviation and returns to normal post septoplasty (Letzel et al., 2022).Comparing functional differences between the groups, Vanillin odorant (undiluted concentration) (JVA178677K/DGI, provided by Takasago Inc.) was delivered using a computer controlled olfactometer at a constant airflow of 2 L/min (Sommer et al., 2012).Odorant was delivered in a block pattern with alternating on and off sessions.During ON blocks, vanillin odorant was delivered for 8 s; during OFF blocks, clean air was presented for 12 s in a total of 12 blocks; 248 functional images were collected using a T2 echo planar sequence: TR = 1000 ms, TE = 38 ms, 58 flip angle, no interslice gap, 210 Â 210 mm field of view.A high-resolution structural T1 image was acquired using a 3D magnetisation prepared gradient rapid acquisition gradient echo (MPRAGE) sequence (TR = 2000 ms, TE = 1.95 ms, 256 Â 256 mm 2 field of view, voxel size 1 Â 1 Â 1 mm 3 ).
OB volume was measured by box-frame method (Joshi et al., 2020) using ITK-SNAP software (version 3.8.0,University of Pennsylvania & University of Utah, www.itksnap.org).Two experts measured the bulb volumes independently.A third expert measured volume for bulbs when the difference between the first two observers was more than 10%.The closest two volumes with less than 10% difference were selected at the end.
Task driven general linear model (GLM) approach using Statistical Parametric Mapping (SPM) (version 12; http://www.fil.ion.ucl.ac.uk/spm) was used to analyse functional data.SPM12 is a MATLAB R2018b (The Math-Works Inc., Natick, MA, USA) based software.Preprocessing steps were set as default.Individual level and group level analyses were done for more patent side (left side dominant subjects with right side dominant subjects).We looked at the brain activations for contrast left potent > right potent and vice versa.MRIcroGL (https:// www.nitrc.org/projects/mricrogl)was used for visualisation and display of fMRI results.We also performed region of interest (ROI)-based within-group analysis to look at the lateralised olfactory activations within more patent nostril (right-and left-sided individuals).ROIs chosen were part of primary and secondary olfactory network: left and right piriform cortex and left and right orbitofrontal cortex (OFC), respectively (Seubert et al., 2013).Unless otherwise stated, all neuroimaging results are reported at (family-wise error corrected) FWEcorrected p < 0.05 fewat cluster size (K) > 10 voxels.Statistical analysis done using SPSS v. 27 (Armonk, NY, USA: IBM Corp).The level of significance was set at p value < 0.05.

| RESULTS
In total, 35 subjects participated in the study (22 women and 13 men, mean age 26 ± 3 years); 22 participants had a more patent right-sided nasal cycle and 13 had a leftsided nasal cycle.All women, during the time of the study, were not menstruating.The mean LI for the leftsided nasal cycle participants was À0.12 ± 0.1, and for right-sided participants, it was 0.17 ± 0.2.We used EHI scores (0-100) for determining handedness, and the mean score was 92.1 ± 5.3 corresponding to right-hand preference among the groups.

| Between-group analysis (comparing more patent side)
Functional response to vanilla odorant for 22 subjects with right dominance and 13 subjects with left dominance, using age and gender as covariates, showed overlapping activations (Table 1).We did not find any voxel differences for left > right-sided nasal cycle as well as for the reverse contrast both at FWE-corrected p < 0.05 and p-uncorrected <0.001 and cluster size >20 voxels.For this second level analysis, comparison was made for first level contrasts odorant versus baseline.

| DISCUSSION
Since all the participants (n = 35) in our study were right-handed, our study does not support the dependence of nasal cycle on handedness (Searleman et al., 2005).Twenty-two subjects (62.8%) of the sample had a rightsided nasal cycle, and 13 being left-sided.Contrary to our hypothesis, nasal cycle potency was found to be independent of the contralateral and ipsilateral OB volumes.Also, cerebral activations during odorant perception were found to be independent of the nasal cycle potency.Results from the within-group analysis showed activations in the bilateral piriform cortex and OFC, suggesting the absence of lateralised differences during odorant perception.However, it is important to note that the cluster size was on average higher for right side lateralisation.Comparison of functional activations between left and right dominant participants revealed overlapping activation patterns suggesting that brain processing is similar and the hemispheric differences is not strongly dependent on the nasal cycle.
Sobel and colleagues reported that the nostril with higher airflow is more sensitive to high-sorption odorants and the nostril with lower airflow is more sensitive to low-sorption odorants.They concluded that different airflow between the nostrils results in disparity of olfactory perception (Sobel et al., 1999).Odorant propagation relies on nasal airflow, directly affecting the number of odorant molecules reaching the olfactory epithelium, which relays the olfactory information to the OB, the first central processing region of the brain.However, results from our study suggest OB volumes are independent of NC either ipsilaterally or contralaterally.The OB volumes did not differ interindividually between sides showing symmetrical OBs in young normosmic individuals.However, previous research reports higher sensitivity of left nostril  (decreased odorant threshold) with higher left side OB volumes (Hummel et al., 2013).The authors hypothesised that a third variable, inflammation or developmental errors, could explain such a change.Yao et al. (2020) provided evidence that the subjective intensity of unilaterally presented odorant decreases with increase of perceived contralateral airflow, independent of sniff vigour and odorant transportation.These findings indicated that the central processing of odorants involves the processing of nasal airflow information, which is relayed from branches of the trigeminal nerve (Yao et al., 2020).With regard to the cerebral processing of olfactory information, hemispheres are specialised for various processes, referred to as hemispheric specialisation or brain lateralisation (Rogers, 2021).For example, Gotts et al. stated that the right hemisphere, involved in visuospatial and attentional processing, has a relatively stronger interaction with the left hemisphere, whereas the left hemisphere exhibits a lower degree of interactions with the contralateral hemisphere (majorly in linguistics and motor coordination) (Gotts et al., 2013).In our study, we did not find any lateralised differences when subjects processed odorant-induced information.The nasal cycle dominance was not found to be correlated with the ipsilateral or contralateral cerebral activations.As evident from the activity patterns, the brain activations to olfactory stimuli are roughly similar.It may be speculated that the presentation of a bimodal odorant with combined properties of olfactory and trigeminal stimuli would enhance differential activations and better shed light on the interplay of the nasal cycle and hemispherical activations, because both olfactory and trigeminal nerves are involved in nasal chemosensitivity and serve different functions of the nose (Doty et al., 1978).
The results of the present study are in line with research where functional changes can be due to either sinonasal surgery resulting in brain alterations (Whitcroft et al., 2021) or to nasal septum deviation (Altundag et al., 2014).Absence of lateralised differences could possibly be explained by the fact that nasal cycle is highly dynamic and is unstable.
The current study has certain limitations, notably a relatively small sample size that might introduce bias and potentially underestimate the observed effect.However, a power analysis demonstrated that the sample size was adequate for detecting an effect.Another limitation pertains to the use of a single odorant to evoke brain responses.The perception of an odour is more associated with its deposition in the nasal cavity than its physiochemical characteristics, making it independent of nasal airflow.Employing various odours would likely yield comparable results.
The negative results in terms of an influence of nasal cycle dominance on brain morphology and function are of significance.Although the nasal cycle is of importance as suggested by previous research, its functional role in olfaction is still debatable.Concluding from our study it appears that nasal cycle dominance does not result in significant structural or functional lateralisation in the olfactory system.

F
I G U R E 1 Outputs from the device 'nasal halter' for mentioned period, awake (during day) and sleep.The participants were scanned according to their patterns based on awake phase.F I G U R E 2 Presence of nasal cycle seen using MRI.Red elliptical depicts the area where the fluctuations in nasal mucosa occur, leading to changes in airflow.

F
I G U R E 3 Enhanced cerebral brain activations in response to 'vanillin' odorant in right-side potent versus left-side potent.Areas depicted by red and yellow elliptical include entorhinal cortex, piriform cortex, and amygdala.
T A B L E 1 The coordinates of brain activations and their labels each for left side NC and right-side NC.It is important to note here that similar activations patterns are seen.Abbreviations: AAL3, labelling atlas; MNI, Montréal neurological institute; NC, nasal cycle; OFC, orbitofrontal cortex. Note: