Content uploaded by David T. Liu
Author content
All content in this area was uploaded by David T. Liu on Nov 19, 2020
Content may be subject to copyright.
Parosmia is Associated with Relevant Olfactory Recovery After
Olfactory Training
David T. Liu, MD ; Maha Sabha; Michael Damm, MD; Carl Philpott, MD ; Anna Oleszkiewicz, PhD ;
Antje Hähner, MD; Thomas Hummel, MD
Objective/Hypothesis: This study aims to determine the association between parosmia and clinically relevant recovery
of olfactory function in patients with post-infectious olfactory dysfunction (PIOD) receiving olfactory training.
Study Design: Retrospective cohort study.
Methods: This was a retrospective cohort study of patients with PIOD that received olfactory training. Adult patients with
the major complaint of quantitative smell loss were recruited and treated at several ENT clinics in German between 2008 and
2018. The outcome was based on the association between smell-loss related factors (including parosmia and phantosmia) and
clinically relevant changes in overall and subdimension olfactory function of threshold, discrimination, and identification using
binary logistic regression analysis.
Results: A total of 153 participants with PIOD were included. Clinically relevant improvements in overall olfactory func-
tion were more likely in those that had lower baseline olfactory function. Relevant improvements in discrimination function
were more likely in those that had lower baseline olfactory function and those that had parosmia at the initial visit. Similarly,
relevant improvements in odor identification were more likely in those that had a lower baseline olfactory function and in
those who had parosmia at the first visit. Clinically significant improvements in odor threshold were more likely in those who
were older in age.
Conclusions: This study demonstrated that the presence of parosmia is associated with clinically relevant recovery in
olfactory discrimination and identification function in patients with PIOD receiving olfactory training.
Key Words: Smell, olfaction, olfactory training, smell loss, anosmia, hyposmia, parosmia.
Level of Evidence: 4
Laryngoscope, 00:1–6, 2020
INTRODUCTION
The olfactory system is important for our response to
the environment and olfactory dysfunction (OD) represents
a critical loss of information. The causes are diverse,
including upper airway respiratory tract infections post-
infectious olfactory dysfunction (PIOD), head traumas, idio-
pathic reasons, and impairments secondary to sinonasal or
neurodegenerative diseases.
1
OD can be categorized into
qualitative and quantitative impairments. Qualitative OD
can be further subdivided into parosmia, defined as distorted
odor perception in the presence of an odor and phantosmia,
defined as odor perception in the absence of an apparent
odor source.
2
Both parosmia and phantosmia can occur alone
but are most commonly present along with quantitative
OD.
1
Parosmia has been associated with better clinical out-
come in terms of spontaneous olfactory recovery.
3–5
How-
ever, literature on the significance of parosmia as a predictor
of olfactory rehabilitation in patients with OD receiving ther-
apy remains sparse. Therefore, further elucidating its role as
prognostic factor in olfactory recovery is needed for clinical
counseling, especially when considering its prevalence of up
to 60% among patients with certain etiologies of OD.
6
Although quantitative impairments of the sense of
smell are common and may affect up to one quarter of the
general population, the prevalence of qualitative impair-
ments appears significantly lower.
7,8
Notably, presence of
parosmia varies among patients with quantitative OD,
depending on the underlying cause of smell loss.
Although parosmia is most commonly found in patients
with PIOD, distorted odor perceptions are also reported
in posttraumatic, idiopathic, and sinonasal causes.
9
Pre-
vious studies on parosmia as prognostic factor in olfactory
This is an open access article under the terms of the Creative
Commons Attribution-NonCommercial License, which permits use, distri-
bution and reproduction in any medium, provided the original work is
properly cited and is not used for commercial purposes.
From the Smell and Taste Clinic, Department of
Otorhinolaryngology, Medical Faculty Carl-Gustav Carus (D.T.L., M.S., A.O.,
A.H., T.H.), Technical University of Dresden, Dresden, Germany;
Department of Otorhinolaryngology, Head and Neck Surgery (D.T.L.),
Medical University of Vienna, Vienna, Austria; Department of
Otorhinolaryngology (M.D.), ENT-Medicine Cologne (HNO-Heilkunde
Köln) and University Hospitals of Cologne, Cologne, Germany; Norwich
Medical School (C.P.), Chancellor’s Drive, University of East Anglia,
Norwich, UK; The Norfolk Smell and Taste Clinic (C.P.), Norfolk and
Waveney ENT Service, Waveney, UK; and the Institute of Psychology (A.
O.), University of Wroclaw, Wroclaw, Poland.
Editor’s Note: This Manuscript was accepted for publication on
November 10, 2020
During the project, DTL was supported by the travel grant of the
Working Group on Olfaction and Gustation of the German Society of Oto-
laryngology (“Reisestipendium der AG Olfaktologie / Gustologie”). AO
received scholarship from the Ministry of Science and Higher Education
(#626/STYP/12/2017).
The authors have no other funding, financial relationships, or con-
flicts of interest to disclose.
Send correspondence to Thomas Hummel, MD, Smell and Taste
Clinic, Department of Otorhinolaryngology, Technical University of Dres-
den, Fetscherstrasse 74, 01307 Dresden, Germany. E-Mail:
thummel@msx.tu-dresden.de
DOI: 10.1002/lary.29277
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
1
The Laryngoscope
© 2020 The Authors. The Laryngoscope
published by Wiley Periodicals LLC on
behalf of American Laryngological,
Rhinological and Otological Society Inc,
"The Triological Society" and American
Laryngological Association (ALA).
recovery provided first evidence, that the presence of par-
osmia at the initial visit might be associated with a
higher number of clinically relevant improvements com-
pared to the parosmia-free group.
3–5
Treatment for smell loss relates to its underlying
cause and pathophysiology. Although treatment strate-
gies for OD secondary to (chronic) sinonasal diseases aim
to resolve the underlying conditions, olfactory training
(OT) aims to enhance olfactory recovery based on the neu-
ronal plasticity of the olfactory system.
10
OT is rec-
ommended as conscious sniffing of at least four different
odors at least twice daily for several months and has
emerged as a simple and side-effect free treatment option
for various causes of smell loss. Previous studies and
meta-analysis provided evidence that OT is effective in
patients with OD, but also healthy subjects of different
age groups to improve olfactory function.
10
It has been
suggested, that etiology of smell loss (i.e., PIOD) and lon-
ger duration of OT might serve as prognostic factor for bet-
ter outcomes in terms of olfactory recovery.
10
However, the
literature on symptoms of qualitative OD as predictor of
olfactory recovery after OT remains sparse. Understanding
its impact would be of great clinical significance in counsel-
ing patients who may otherwise be confused by distorted
odor perceptions in quantitative smell loss. Hence, the aim
of this study was to elucidate the prognostic value of par-
osmia and phantosmia in terms of olfactory rehabilitation
in a cohort of patients with PIOD receiving OT.
MATERIAL AND METHODS
Study Population
This retrospective study followed with the Declara-
tion of Helsinki. Its design was approved by the Ethics
Committee of the Faculty of Medicine at the TU Dresden
(EK251112006). This pooled data analysis included adult
participants from three previously published studies on
OT.
11–13
Adult patients were either self-referrals or
referred from outside institutions to tertiary-care otorhi-
nolaryngology departments between 2008 and 2018. Inclu-
sion criteria were posttraumatic OD, PIOD, and idiopathic
smell loss. Exclusion criteria were TDI above 30.5 (indicat-
ing normal olfactory function), pregnancy, and acute or
chronic sinonasal diseases. At the initial visit, patients
were asked for presence of parosmia (“Do you smell odors
differently compared to previous experiences?”)or
phantosmia (“Do you smell odors in absence of an appar-
ent source?”) based on a binary outcome of yes and no,
time since onset of OD (in month), and possible causes for
their smell loss. Diagnosis was made based on the recent
“Position paper of olfactory dysfunction
1
”(Table I).
Olfactory Testing
Olfactory testing was performed twice birhinally
(before and after training) by means of the validated
Sniffin’Sticks test (Burghart Medical Technology, Wedel,
Germany).
14
The Sniffin’Sticks test is divided into sub-
tests, covering three olfactory dimensions: 1) threshold
(T), 2) discrimination (D), and 3) identification (I).
Summed scores of subdimensions threshold, discrimina-
tion, and identification (TDI) allow the categorization of
olfactory performance into normosmia, hyposmia, and
functional anosmia based on normative data of over 9000
healthy subjects.
15
The test procedure is described in
detail elsewhere.
14
Furthermore, Sniffin’Sticks can also
be used for follow-up testing with minimally clinically
important differences defined for summed scores and
each of the subtests separately.
16
Olfactory Training
All patients included during this study received OT
as a therapy for their smell loss.
11–13
Olfactory training is
defined as conscious sniffing of (usually four) different
odors twice a day for at least 15 seconds each.
13
Partici-
pants either received: 1) four single molecule substances
for the entire study period (anise odor, anethol; eucalyp-
tus odor, eucalyptol; lemon odor, citronella; cloves odor,
eugenol), 2) four multi-molecule substances (mixtures of
single molecule substances) with a dominant scent of the
odors stated hereafter for the entire study period (rose
odor, phenyl ethyl alcohol; eucalyptus odor, eucalyptol;
lemon odor, citronellal; cloves odor, eugenol), or 3) twelve
multi-molecule substances, which were alternated twice
every eight weeks as a group of four (first phase: phenyl
ethyl alcohol, eucalyptol, citronellal, eugenol; second phase:
cinnamon, thyme, chocolate, peach; third phase: coffee, lav-
ender, honey, strawberry). Previous studies have shown
that the effect of OT in olfactory rehabilitation is consistent
within studies that applied different training protocols.
11–13
Statistical Analyses
Binary logistic regression models were computed
in patients with PIOD to assess the associations
TABLE I.
Demographics, Olfactory Test Results, and Smell-loss Related
Variables.
Demographics
Mean age (years) 58.7 (7.3)
Gender 140 Female, 106 Male
Smell-loss related factors
Aetiology
Post-infectious 153 (62.2%)
Posttraumatic 31 (12.6%)
Idiopathic 62 (25.2%)
Duration of OT (weeks) 25.8 (8.0)
Presence of (at first visit)
Parosmia 81 (32.9%)
Phantosmia 43 (17.5%)
Duration of OD (months) 22.0 (41.8)
Olfactory function
Baseline olfactory function (TDI) 17.6 (7.0)
Follow up olfactory function (TDI) 20.8 (7.9)
Continuous data are presented as mean (standard deviation). Categor-
ical data are presented as number (%).
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
2
between demographics, olfactory-related factors, and
clinically relevant changes in overall olfactory function
(TDI) and the sub-dimensions threshold (T), discrimina-
tion (D), and identification (I). Clinically relevant
changes were defined based on the following cut-off
scores: 1) for overall olfactory function: TDI improve-
ment greater or equal 5.5 points at follow up visit, 2)
for threshold function: T improvement greater or equal
2.5 points at follow up visit, and 3) for discrimination
and identification function: improvements greater or
equal 3 points at follow up visit.
16
Olfactory-related
variables included: age (years), gender (male and
female), olfactory function at first visit (baseline olfac-
tory function, TDI), duration of olfactory training
(weeks), duration of smell loss (month), and presence of
parosmia or phantosmia at first visit. All demographics
and olfactory-related variables were entered in the
models, and statistical estimates were generated to cal-
culate adjusted odds ratios (aOR) with 95% confidence
interval to control for the impact of potential con-
founders mentioned above. Data were analyzed using
SPSS (SPSS version 25.0 for Windows; IBM Corp.,
Armonk, NY, USA). This study used a level of signifi-
cance of 0.05.
RESULTS
Participants
The presence of distorted olfactory perception and
improvement of olfactory performance after OT was analyzed
in 246 subjects (106 men, 140 women, mean (±SD) age
58.7 ± 7.3 years). Diagnosis included 153 post-infectious-, 31
posttraumatic-, and 62 idiopathic- related OD (Table I).
Olfactory training was performed for a mean (±SD) period of
25.8 ± 8 weeks. Although 292 participants were initially
included in the study, further analysis was performed on the
basis of ‘listwise’exclusion in case of missing values,
resulting in a total of 46 subjects being excluded from the
final analysis sample (n = 246).
Frequency of Qualitative OD by Etiology
We first sought to determine the presence of par-
osmia and phantosmia for each etiology group separately.
Parosmia was most frequently present in PIOD (40.5%),
followed by posttraumatic OD (25.8%), and idiopathic OD
(17.7%). In contrast, phantosmia was most commonly pre-
sent in idiopathic OD (25.8%), followed by posttraumatic
OD (19.3%) and PIOD (13.7%).
Since the sample sizes of patients with post-
traumatic smell loss (n = 31) and idiopathic OD (n = 62)
were insufficient for further regression analysis, we only
included patients with PIOD (n = 153) in our binary logis-
tic regression models.
Association Between OD with Relevant
Improvement in Overall Olfactory Function
The next step included an analysis of associations
between smell-loss related factors: 1) age, 2) gender, 3)
duration of smell loss (months), 4) duration of training
(weeks), 5) baseline olfactory function, and 6) presence of
parosmia or phantosmia at initial visit with clinically rel-
evant recovery of overall olfactory function (defined as
TDI improvement greater or equal 5.5 points) in patients
with PIOD at follow-up visit. Therefore, a binary logistic
regression model was computed.
Analysis revealed that relevant recovery of overall
olfactory performance was more likely in those that had
lower baseline olfactory function (adjusted odds ratio;
aOR, 0.92; 95%CI, 0.87–0.97; Table II).
Association Between Parosmia and OD with
Relevant Improvement in Discrimination
The next step sought to determine associations
between smell-loss related variables (see above) and rele-
vant changes in discrimination function (defined as D
improvement greater or equal 3.0 points) at follow-up visit.
Logistic regression analysis revealed that relevant
improvements in discrimination function were more
TABLE II.
Factor Associated with Clinically Relevant Changes of Overall and Olfactory Subdimension Function in Patients with PIOD.
Variables TDI Threshold Discrimination Identification
Age (years) 1.01 (0.96–1.07) 0.81 1.07 (1.00–1.13) 0.03 1.00 (0.95–1.05) 0.87 0.99 (0.95–1.05) 0.75
Gender
Female Reference
Male 0.87 (0.42–1.80) 0.70 0.59 (0.28–1.26) 0.17 1.88 (0.90–3.92) 0.09 0.63 (0.28–1.45) 0.28
Baseline olfactory function (TDI) 0.92 (0.87–0.97) 0.002 1.02 (0.96–1.08) 0.52 0.91 (0.86–0.96) 0.001 0.91 (0.86–0.97) 0.004
Duration of smell loss (month) 0.97 (0.94–1.00) 0.11 0.99 (1.00–1.01) 0.66 1.00 (0.99–1.01) 0.96 0.97 (0.93–1.01) 0.17
Presence of
Parosmia 1.12 (0.59–2.46) 0.62 1.11 (0.53–2.33) 0.78 2.88 (1.25–6.11) 0.006 3.38 (1.50–7.60) 0.003
Phantosmia 1.11 (0.41–3.00) 0.94 0.42 (0.13–1.33) 0.14 0.80 (0.27–2.32) 0.68 1.76 (0.61–5.10) 0.30
Duration of OT (weeks) 1.02 (0.97–1.07) 0.11 1.00 (0.91–1.00) 0.05 1.03 (0.98–1.08) 0.22 1.00 (0.95–1.05) 0.96
Adjusted odds ratios, aOR (95%CI) and Pvalues. Statistical significance is set at P< .05. Multivariate analysis was performed using binary logistic regres-
sion models, adjusted for age, baseline olfactory function, gender, duration of training, duration of smell loss, and presence of parosmia or phantosmia.
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
3
likely in those that had lower baseline olfactory function
(aOR, 0.91; 95%CI, 0.86–0.96) and those that had par-
osmia at first visit (aOR, 2.88; 95%CI, 1.25–6.11).
Association Between Parosmia and Gender with
Relevant Improvement in Identification
We were then interested in identifying smell-loss
related factors that are associated with clinically relevant
improvements in odor identification function (defined as
improvement greater or equal 3.0 points) at follow-up
visit.
Binary logistic regression analysis revealed that rel-
evant improvements in identification were more likely in
those that had lower baseline olfactory function (aOR,
0.91; 95%CI, 0.86–0.97) and those that had parosmia at
first visit (aOR, 3.38; 95%CI, 1.50–7.60).
Association Between Age with Relevant
Improvement in Threshold
We were next interested in determining which of
above-mentioned smell loss-related variables were associ-
ated with clinically relevant improvements in olfactory
threshold performance (defined as T improvement greater
or equal 2.5 points) at follow up visit (Fig. 1).
Binary logistic regression analysis revealed that clin-
ically relevant improvements in threshold function were
more likely in those who were older in age (aOR, 1.07,
95%CI 1.00–1.13).
DISCUSSION
Although studies dedicated to assessing the prognos-
tic value of qualitative OD in smell loss provided first evi-
dence that parosmia might serve as a prognostic factor
for spontaneous recovery of olfactory function,
3–5
there
remains a gap of knowledge relating to its predictive
value in patients receiving OT, which is currently the
first-line treatment option for different etiologies of smell
loss.
1
In this study, we showed that the presence of par-
osmia at initial visit was associated with clinically signifi-
cant recovery in suprathreshold olfactory function
discrimination and identification in patients with PIOD
receiving OT. We also found that changes in sup-
rathreshold olfactory functions after OT were distinct
from threshold improvements, possibly indicating that
the improvement of function of olfactory subdimensions
may be based on changes at different stages of olfactory
processing. Specifically, it has been hypothesized that
odor thresholds reflect peripheral function to a higher
degree than odor discrimination and odor identifica-
tion.
17,18
According to this avenue of thought it may be
that the presence of parosmia at the first visit appears to
represent a positive sign in terms of the improvement of
the central nervous extraction of olfactory information.
The most important results emerged from our sub-
group analysis of factors associated with significant recov-
ery of suprathreshold olfactory function discrimination
and identification. Our analyses revealed that both lower
baseline olfactory function and presence of parosmia at
initial visit were prognostic predictors for clinically rele-
vant recoveries in patients with PIOD. The reason for
parosmia as positive predictor in suprathreshold recovery
after OT can only be speculated upon. However, it has
been suggested that OT mainly improves cognitive
processing of olfaction-related sensory information.
19
Recent work based on magnetic resonance imaging (MRI)
further provided evidence, that OT is not only associated
with increase of olfactory bulb and grey matter volume on
a structural level, but also re-established the intensity of
functional connectivity within the olfactory system.
20
Moreover, MRI scanning in posttraumatic olfactory loss
has suggested that recovery of olfactory function after OT
may be largely due to top-down rather than bottom-up
mechanisms.
21
In line with the previously proposed
mechanism of incomplete afferent sensory information in
distorted odor perceptions, it might be speculated that
symptoms of parosmia can be interpreted as early signs
of recovery. Following on from this, OT might effectively
improve cognitive processing of (incomplete) sensory
information, hence resulting in improved outcome of
patients that report parosmia.
Results from threshold, discrimination, and identifica-
tion analysis provide further evidence for the “central-
peripheral”hypothesis of olfactory subdimension processing.
As mentioned above, it has been postulated that threshold
represents peripheral olfactory function to a higher degree
than discrimination and identification.
17,18
Although specu-
lative, these findings stress the importance for future efforts
in experimental and clinical research regarding olfactory
neuron regeneration in different types of olfactory loss.
More importantly, results provide further evidence that the
assessment of both suprathreshold and threshold olfactory
function represent the most meaningful approach to the
human sense of smell.
Prior investigations on the prevalence rates of par-
osmia and phantosmia in patients with various causes of
smell loss showed difference between study centers.
6–9
Since symptoms of isolated qualitative dysfunctions are
hardly ever spontaneously reported by patients,
22
the het-
erogeneity of methods and questionnaires used has been
suggested to be one major reason for this discrepancy.
23
In
addition, qualitative olfactory dysfunction is –like in the
present investigation –typically assessed in terms of the
presence or absence. It has been previously reported that
parosmia is most prevalent in PIOD and one possible
explanation might relate to its pathophysiology. Although
the exact mechanism is only partly delineated, there is at
least preliminary evidence that the number of olfactory
sensory neurons (OSN) is reduced in these patients.
24
Considering the clinical relevance of the current
investigation, results can be implemented effortless into
clinical routine. The awareness for symptoms of qualita-
tive OD must be raised among the medical profession.
Parosmia and phantosmia can be easily assessed based
on straightforward questions with binary outcomes
(yes/no), psychophysical test methods, the use of vali-
dated questionnaires, or the simple grading of parosmia,
with questions on 1) frequency (daily, not daily), 2) inten-
sity (not intense, intense), 3) social impact (present,
absent).
9,25
Since OT has become the recommended first-
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
4
line treatment protocol for certain causes of smell loss,
1
consideration of predictors for relevant recoveries after
OT, such as parosmia might not only calibrate patients’
expectations more appropriately but also comfort patients
with PIOD that may otherwise be distraught by distorted
odor perceptions. Previous studies also provided first evi-
dence that longer duration of treatment (more than
8 months) might also increase the effectiveness of
OT.
11,26,27
Patients that were included in this analysis
received OT for up to 9 months, which is longer than the
usually recommended duration of treatment. Since OT is
usually recommended for at least six months (twice on a
daily base), informing patients (and thus increasing the
likelihood of treatment adherence
28,29
) remains a corner-
stone during counseling.
The present study uses a comprehensive dataset
including relevant olfactory demographics and smell-loss
related variables to assess different factors associated
with clinically relevant improvements after OT in
patients with PIOD. However, this study also has limita-
tions. Firstly, although we were able to depict the train-
ing regimen in all studies, small differences in odors used
might have biased our results. Since previous studies
have shown that the training effect was consistent among
different training protocols, these differences might not
have affected the outcome after OT to a large extent.
11,12
Secondly, although we tried to explore the presence of
parosmia and phantosmia in different causes of smell
loss, binary logistic regression analyses were performed
only for patients with PIOD. Therefore, additional studies
with larger sample sizes of various etiologies (such as
posttraumatic or idiopathic smell loss) are needed to
explore the associations between qualitative OD, reason
for OD, and clinically relevant recovery of olfactory func-
tion after OT. However, our results do provide guidance
regarding the magnitude of potential effects and the
resultant sample sizes needed, as it is one of the few to
examine the impact of parosmia on olfactory rehabilita-
tion. Thirdly, we could not exclude the possibility that the
observed association between lower baseline olfactory
function and higher odds of relevant recovery after OT
might be attributable to the regression to the mean
(RTM) phenomenon, which is frequently described in lon-
gitudinal studies.
30
The RTM is a phenomenon wherein
more unusual/extreme test scores are more likely to be
followed by an average/mean score, regardless of any
“real”change in olfactory function.
Finally, information on the presence of parosmia at
follow-up visit was not available in the current dataset.
Since parosmia is characterized by distorted odor percep-
tions, it might be hypothesized that lower identification
and discrimination function at baseline visit reflected the
interaction and negative effect of parosmia on odor identi-
fication and discrimination tasks, rather than a
Fig. 1. Forest plots showing the associations between olfactory-related variables with relevant changes in overall olfactory function (TDI) and
olfactory subdimensions threshold (T), discrimination (D), and identification (I) in patients with PIOD. Odds ratios (OR) were calculated using
binary logistic regression models adjusted for age, gender (reference group female), presence of parosmia and phantosmia at first visit, base-
line olfactory function, duration of smell loss (month), and duration of olfactory training (weeks). Points represent group-specific OR point esti-
mates, and lines indicate the respective 95% confidence interval (CI).
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
5
quantitative dysfunction. Therefore, recovery from par-
osmia might also explain improved odor identification
and discrimination function at follow-up visit.
CONCLUSION
This study adds to the current literature in three
important ways. First, parosmia was associated with clin-
ically relevant recovery of discrimination and identifica-
tion (suprathreshold) function after OT in patients with
PIOD, which highlights the need to further raise aware-
ness for symptoms of qualitative OD in patients with
smell loss. Secondly, it provides valuable insights into fac-
tors that modulate clinically relevant recovery of olfactory
function after OT, which should be recommended for at
least six months. These variables can further be used in
counseling of patients to calibrate expectations and out-
comes more appropriately. Thirdly, it adds evidence to
the idea that the comprehensive analysis of different
olfactory components, such as threshold and sup-
rathreshold functions during psychophysical testing are
indispensable when evaluating the human sense of smell.
ACKNOWLEDGMENTS
We thank the patients who participated in these studies
and the principal study investigators. Open access
funding enabled and organized by Projekt DEAL.
REFERENCES
1. Hummel T, Whitcroft KL, Andrews P, et al. Position paper on olfactory dys-
function. Rhinol Suppl 2017;54:1–30.
2. Leopold D. Distortion of olfactory perception: diagnosis and treatment.
Chem Senses 2002;27:611–615.
3. Reden J, Maroldt H, Fritz A, et al. A study on the prognostic significance of
qualitative olfactory dysfunction. Eur Arch Otorhinolaryngol 2007;264:
139–144.
4. Cavazzana A, Larsson M, Münch M, Hähner A, Hummel T. Postinfectious
olfactory loss: A retrospective study on 791 patients. Laryngoscope 2018;
128:10–15.
5. Hummel T, Lötsch J. Prognostic factors of olfactory dysfunction. Arch
Otolaryngol Head Neck Surg 2010;136:347–351.
6. Deems DA, Doty RL, Settle RG, et al. Smell and taste disorders, a study of
750 patients from the University of Pennsylvania Smell and Taste Center.
Arch Otolaryngol Head Neck Surg 1991;117:519–528.
7. Landis BN, Konnerth CG, Hummel T. A study on the frequency of olfactory
dysfunction. Laryngoscope 2004;114:1764–1769.
8. Nordin S, Brämerson A, Millqvist E, Bende M. Prevalence of parosmia: The
Skövde population-based studies. Rhinology 2007;45:50–53.
9. Landis BN, Frasnelli J, Croy I, Hummel T. Evaluating the clinical useful-
ness of structured questions in parosmia assessment. Laryngoscope 2010;
120:1707–1783.
10. Sorokowska A, Drechsler E, Karwowski M, Hummel T. Effects of olfactory
training: a meta-analysis. Rhinology 2017;55:17–26.
11. Damm M, Pikart LK, Reimann H, et al. Olfactory training is helpful in
postinfectious olfactory loss: a randomized, controlled, multicenter study.
Laryngoscope 2014;124:826–831.
12. Oleszkiewicz A, Hanf S, Whitcroft KL, Haehner A, Hummel T. Examination
of olfactory training effectiveness in relation to its complexity and the
cause of olfactory loss. The Laryngoscope. 2018;128 (7):1518–1522. http://
dx.doi.org/10.1002/lary.26985.
13. Hummel T, Karo R, Reden J, et al. Effects of olfactory Training in patients
with olfactory loss. Laryngoscope 2009;119:496–499.
14. Rumeau C, Nguyen DT, Jankowski R. How to assess olfactory performance
with the Sniffin’Sticks test®. Eur Ann Otorhinolaryngol Head Neck Dis
2016;133:203–206.
15. Oleszkiewicz A, Schriever VA, Croy I, Hähner A, Hummel T. Updated
Sniffin’Sticks normative data based on an extended sample of 9139 sub-
jects. Eur Arch Otorhinolaryngol 2019;276:719–728.
16. Gudziol V, Lötsch J, Hähner A, et al. Clinical significance of results from
olfactory testing. Laryngoscope 2006;116:1858–1863.
17. Koss E, Weiffenbach JM, Haxby JV, Friedland RP. Olfactory detection and
identification performance are dissociated in early Alzheimer’s disease.
Neurology 1988;38:1228–1232.
18. Frasnelli JA, Temmel AF, Quint C, Oberbauer R, Hummel T. Olfactory
function in chronic renal failure. Am J Rhinol 2002;16:275–279.
19. Haehner A, Tosch C, Wolz M, et al. Olfactory Training in Patients with
Parkinson’s Disease. PLoS One 2013;8:e61680.
20. Han P, Zang Y, Akshita J, Hummel T. Magnetic resonance imaging of
human olfactory dysfunction. Brain Topogr 2019;32:987–997.
21. Pellegrino R, Han P, Reither N, Hummel T. Effectiveness of olfactory train-
ing on different severities of posttraumatic loss of smell. Laryngoscope
2019;129:1737–1743.
22. Reden J, Lill K, Zahnert T, Haehner A, Hummel T. Olfactory function in
patients with postinfectious and posttraumatic smell disorders before and
after treatment with vitamin A: a double-blind, placebo-controlled, ran-
domized clinical trial. Laryngoscope 2012;122:1906–1909.
23. Croy I, Nordin S, Hummel T. Olfactory disorders and quality of life-an
updated review. Chem Senses 2014;39:185–194.
24. Yamagishi M, Fujiwara M, Nakamura H. Olfactory mucosal findings and
clinical course in patients with olfactory disorders following upper respira-
tory viral infection. Rhinology 1994;32:113–118.
25. Liu DT, Lüssen AW, Besser G, et al. Assessment of odor hedonic perception:
the Sniffin’sticks parosmia test ( SSParoT ). Sci Rep 2020;10:18019.
26. Geißler K, Reimann H, Gudziol H, Bitter T, Guntinas-Lichius O. Olfactory
training for patients with olfactory loss after upper respiratory tract infec-
tions. Eur Arch Otorhinolaryngol 2014;271:1557–1562.
27. Altundag A, Cayonu M, Kayabasoglu G, et al. Modified olfactory training in
patients with postinfectious olfactory loss. Laryngoscope 2015;125:
1763–1766.
28. Squier RW. A model of empathic understanding and adherence to treatment
regimens in practitioner-patient relationships. Soc Sci Med 1990;30:
325–339.
29. Liu DT, Besser G, Prem B, Speth MM, Sedaghat AR, Mueller C. Individual
importance of olfaction decreases with duration of smell loss in patients
with olfactory dysfunction. Rhinology 2020. https://doi.org/10.4193/
Rhin20.196. Epub ahead of print. PMID: 32926009.
30. Barnett AG, van der Pols JC, Dobson AJ. Regression to the mean: What it
is and how to deal with it. Int J Epidemiol 2005;34:215–220.
Laryngoscope 00: 2020 Liu et al.: Parosmia and Olfactory Recovery
6
