Nature & Wildlife

A Study on the Frequency of Olfactory Dysfunction

A Study on the Frequency of Olfactory Dysfunction
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  The Laryngoscope Lippincott Williams & Wilkins, Inc. © 2004 The American Laryngological,Rhinological and Otological Society, Inc.  A Study on the Frequency of OlfactoryDysfunction B. N. Landis, MD; C. G. Konnerth, MD; T. Hummel, MD, PhD Objectives/Hypothesis:  Goals of the study were toevaluate the frequency of olfactory dysfunction in alarge representative population without sinonasalcomplaints and to investigate the extent to whichgeneral pathological conditions, medications, and ag-ing influence olfaction.  Study Design:  Prospective.  Methods:  Results based on an odor identification test(“Sniffin’ Sticks”) were reported from 1240 subjects.The subjects presented themselves to an otorhinolar-yngology outpatient clinic with relatively mild andtransitory complaints unrelated to the upper air-ways. A detailed otorhinolaryngological examinationincombinationwithastandardizedinterviewfurtherascertained that these patients had no rhinologicalproblems or symptoms relating to sinonasal disease.  Results:  Apart from the confirmation of the effects of age, gender, and certain otorhinolaryngological dis-eases on the sense of smell, the study results revealedthat certain general diseases (liver diseases, nonoto-laryngological cancers) appear to influence olfactoryfunction, whereas other diseases or disorders havelittle or no impact on olfaction (hypertension, cardio-vascularproblems). Conclusion: Thedatainthestudyrevealed that olfactory dysfunction among subjectsunder 65 years of age is more frequent than previ-ously reported.  Key Words:  Olfactory dysfunction,incidence, medication, drugs, general pathologicalconditions, aging, parosmia, phantosmia, rating, an-esthesia, nasal polyposis.  Laryngoscope,  114:1764–1769, 2004 INTRODUCTION The incidence of olfactory dysfunction in the generalpopulation is a matter of debate. 1–3 However, most au-thors reported frequencies of 1% to 3% of chemosensorydisorders within the groups studied. Most investigationsidentified aging  3,4 and smoking habits 5 as important fac-tors in terms of the occurrence of smell dysfunction. Withfew exceptions, 3,4 most studies used either questionnairesor relatively insensitive tests to assess olfactory function.Thus, the major aim of the present study was to screenolfactory function in a large number of subjects in combi-nation with a structured questionnaire and nasal endos-copy performed by otolaryngological specialists.Most studies on the frequency of different causes of smell dysfunctions 6–9 have been conducted in olfactionoutpatient clinics. Thus, these studies were compromisedby the fact that all investigated subjects had already beenselected because of the presence of chemosensory dysfunc-tion. In addition, because smell and taste clinics are typ-ically associated with otorhinolaryngology departments,patients displaying sinonasal symptoms may have beenoverrepresented 8 or under-represented 10 in the literature.On the other hand, the large epidemiological studiesshowing high frequencies for smell disorders 1,3 did notexclude subjects reporting sinonasal complaints. There-fore, the second goal of the present study was to investi-gate the frequency of chemosensory disorders in patientswithout sinonasal complaints, thereby including subjectswho would probably never have consulted for any chemo-sensory complaint. An otorhinolaryngology outpatient clinic in Germanyhas a wide range of patients and complaints, and most visits are motivated by isolated ear, nose, or throat prob-lems. Usually, otorhinolaryngology-related diseases aretransitory, and children, adults, and elderly patients areseen. Thus, an average otorhinolaryngology outpatientclinic, at least to some degree, represents the normalpopulation. Assuming that general diseases and drug in-take are equally distributed as in the normal population,we screened olfaction with regard to any olfactory disorderdepending on either the amount of drugs taken or thepresence of general diseases. Every general disease occur-ring in more than 2% of our group, thereby offering ameaningful size, was investigated selectively with respectto a chemosensory disorder. Consequently, the most com-mon general diseases in the population could be investi-gated for possible olfactory alterations. Whereas theamount of literature suggesting drugs suspected to causesmell and/or taste dysfunctions is large, the overwhelming majority of reports present single cases. Systematic stud-ies investigating the effect of drug intake and its influenceon chemoreception are scant. Therefore, a special focus From the Smell and Taste Clinic, Department of Otorhinolaryngol-ogy, University of Dresden Medical School, Dresden, Germany.Supported by grant DFG HU441/2-1 from the Deutsche Forschungs-gemeinschaft ( T . H .).Editor’s Note: This Manuscript was accepted for publication March8, 2004.Send Correspondence to Thomas Hummel, MD, Smell and TasteClinic, Department of Otorhinolaryngology, University of Dresden MedicalSchool, Fetscherstrasse 74, 01307 Dresden, Germany. E-mail: thummel@ Laryngoscope 114: October 2004 Landis et al.: Frequency of Olfactory Dysfunction 1764  was on the total number of drugs taken and the possibleeffect of drugs taken on olfaction. Further, the influencesof age and sex on olfactory function were assessed. PATIENTSANDMETHODS  Patients  All participants were patients of the otorhinolaryngologyoutpatient clinic of the Department of Otorhinolaryngology of theUniversity of Dresden Medical School. In all, 1240 patients wereincluded. The mean patient age was 41.7  0.5 years (range, 5 – 86y); sex distribution was almost equal (643 [52%] men vs. 597[48%] women). We excluded patients who required hospitaliza-tion, and patients coming for oncological otorhinolaryngologyfollow-up. Patients consulting for a rhinological problem or com-plaining of chronic sinonasal problems were also excluded. Theremaining 1240 otorhinolaryngological patients were representa-tive for the German population in terms of age and sex according to the data published by the German Federal Statistical Office 11 (Fig. 1). In addition, the investigated subjects were exclusivelyoutpatients with relatively mild and transitory conditions. In themajority of cases the otorhinolaryngology outpatient clinic ful-filled primary care functions. No inpatients were tested. Further,patients reporting a history of sinonasal complaints were ex-cluded. Therefore, considering the remaining patients, it seemsunlikely that they were prone to exhibit higher rates of dysosmia. All patients filled in a questionnaire inquiring about socio-logical and demographic data, smoking habits, general diseases,medication habits, presence of parosmia or phantosmia, ratingsofolfactoryfunction,andratingsofconsequencesonlifequalityincases of smell loss. People rated their olfactory function to beeither  “ absent, ” “ attenuated, ” “ normal, ”  or  “ above average, ”  cor-responding to scores of 0, 1, 2, or 3, respectively. If participantsrated their olfactory function to be lower than normal, they wereasked about their thoughts about the possible origin of theirdysosmia. Similarly, the impact on life quality attributable tosudden olfactory loss was rated as either  “ none, ” “ slight, ”“ strong, ”  or  “  very strong, ”  corresponding to scores of 0, 1, 2, or 3,respectively. After the olfactory testing all participants under-went a detailed, extensive nasal endoscopy. Patients with patho-logical findings on endoscopy were excluded. Olfactory Testing Psychophysical testing of olfactory function was performedby means of the  “ Sniffin ’  Sticks ”  test battery (Burghart, Wedel,Germany). 12 The  “ Sniffin ’  Sticks ”  test is based on commerciallyavailable felt-tip pens. For odor presentation the cap was re-moved by the experimenter for approximately 3 seconds. Thepen ’ s tip was placed approximately 2 cm in front of both nostrilsfor approximately 2 seconds before the pen was capped again.Using 16 common odors, the test is based on the assessment of odor identification abilities. By means of a multiple-choice task,identification of odors was performed from a list of four descrip-tors each. The subject ’ s identification score (IdS) ranged from 0 to16. Normative data for normosmia, hyposmia, and anosmia arebased on multicentric investigations in more than 1000 sub- jects. 12  According to the mentioned study, 12 an IdS below 8 isconsidered  “ functionally anosmic, ”  an IdS of 8 to 12 is consideredhyposmic, and an IdS higher than 12 is thought to representnormosmia. The term  “ functional anosmia ”  was chosen because itrefers both to patients who have little function left and to patientswho have no olfactory function at all. Statistical Analysis Results were analyzed using SPSS version 11.0 for Windows(SPSS Inc., Chicago, IL). Descriptive statistics are presentedwithin the body of text as mean values    SEM. for comparisonsbetween groups,  t  tests for unpaired samples were employed.Control groups were selected to avoid age biases. For analysis of the several factors, control groups were randomly selected accord-ing to age. Specifically, for certain analyses, subjects older oryounger than the subjects of the group of interest were excludedso that there was no group statistical difference between the agesof subjects of the group of interest and the corresponding controlsubjects. For correlations, Spearman analyses were performed.Partial correlations controlled for age as confounding factors. Analyses of frequencies were calculated using     2 tests. The alphalevel was set at 0.05. RESULTS Sinonasal Problems  Although otorhinological patients complaining of si-nonasal symptoms or diseases were excluded, in 4.7% (58patients) nasal endoscopy revealed the presence of nasalpolyps of which subjects were unaware. Within this  “ pol-yposis group, ”  age was higher compared with the otherotorhinological patients (mean age: 54.5   2 y vs. 41.1   0.5 y;  t    6.3 [  P    .001]); in addition, the group wascomposed of more men than women (38 men and 20 wom-en;  t  2.2 [  P  .03]). Comparison between otorhinolaryn-gological patients without and with nasal polyposis indi-cated that the latter ones were hyposmic (IdS inotorhinolaryngological patients without polyps: 12.6   0.1,  t    4.4; IdS in otorhinolaryngological patients withpolyps: 10.2    0.5 [  P    .001]). Otorhinolaryngologicalpatients with nasal polyposis also rated their olfactoryfunction to be lower ( t    5.7 [  P    0001]) and self-ratingscorrelated significantly with measured IdS (n  58, r 58  0.57 [  P    .001]). Those patients were excluded from fur-ther analyses of the effect of general diseases on the senseof smell.  Frequency of Functional Anosmia and Hyposmia  Functional anosmia.  Among the investigated sam-ple (without otorhinolaryngological patients with polyps) Fig. 1. Distribution of the different age groups (in y) shown within theGerman population (black bars) and within the currently studiedsample (gray bars). In addition, the percentage of occurrence ofanosmia (white bars) is shown. Laryngoscope 114: October 2004 Landis et al.: Frequency of Olfactory Dysfunction 1765  55 patients correctly identified less than 8 of 16 odors.These patients were considered functionally anosmic andrepresented 4.7% of all tested subjects (Fig. 1). No signif-icant sex-related difference in the rate of functional anos-mia was detected (   2 test, F-value    2.63; degrees of freedom [ df  ]  1 [  P  .1]). However, anosmic subjects (age,52.9    2.6 y) were older than normosmic subjects (age,40.5    0.5 y) ( t    4.6 [  P    .001]). Anosmics also tooksignificantly more drugs (1.7    0.29 vs. 1.06    0.05;  t   2.1 [  P    .03]) and had more comorbid conditions (0.78   0.1 vs. 0.45  0.02;  t  2.7 [  P  .008]). The percentage of smokers in both groups was not significantly different (   2 test, F    0.06;  df     1 [  P    .78]). However, using anage-controlled analysis, the difference for comorbid condi-tions and medication disappeared. Correlation analysesshowed that anosmic subjects were accurate in rating their olfactory function (r 55    0.34,  P    .01). However,only 20% of the anosmics indicated a possible reason fortheir dysosmia, rendering this information notinterpretable.  Hyposmia.  Sixteen percent (189 participants) iden-tified less than 12 of the 16 odors (Fig. 1). Age-controlledanalysis revealed that significantly more men (   2 test, F  14.9;  df     1 [  P    .001]) were affected by hyposmia,whereas the number of comorbid conditions, drugs, andsmoking habits were the same in the normosmic andhyposmic group. Similar to anosmic subjects, hyposmicsubjects were accurate in rating their olfactory function asindicated by the weak but significant correlation betweenratings of olfactory function and the IdS (r 189  0.18,  P  .02). As in the group of functionally anosmic subjects, only22% of the hyposmic subjects indicated a possible reasonfor their smell impairment, rendering this information notinterpretable.  Parosmia and Phantosmia The frequencies of parosmia and phantosmia were2.1% and 0.8%, respectively. Both parosmia (IdS, 10.2   0.6;  t  3.5 [  P  .002]) and phantosmia (IdS, 10.4  1.4;  t   2.7 [  P    .006]) were correlated with a lower IdS com-pared with normosmic control subjects (IdS, 12.7    0.1).Within the groups of functionally anosmic and hyposmicsubjects the frequencies of parosmia were 6.1% and 1.6%,respectively. Olfactory Ratings and Impact on Life Quality In contrast to anosmic and hyposmic subjects, nor-mosmic subjects were unable to correctly rate their olfac-tory function as indicated by the nonsignificant correla-tion between ratings of olfactory function and the IdS (r 935  0.03,  P  .3).Estimated loss of life quality due to hypotheticalsmell loss was highest among normosmic participants andgradually decreased with hyposmia and anosmia. Differ-ences in loss of quality of life dependent on olfactory func-tion (Fig. 2) were seen between the following two groups:Participants indicating no or low (scores of 0 or 1 on visualanalogue scale) estimated impact of smell dysfunction onquality of life had lower olfactory function compared withpatients with high or very high estimated impact of smelldysfunction on quality of life (   2 test for trend, F  7.7;  df    2 [  P    .009]). In other words, subjects who estimatedthe hypothetical loss of quality of life to be higher also hadhigher olfactory test scores.The indicated loss of quality of life correlated withIdS (r 1140    0.166,  P    .001). No sex-related differenceswere found for these ratings (   2 test, F  6.7;  df   3 [  P  .08]).  Effect of General Pathological Conditions Partial correlation controlling for age revealed asmall but significant correlation between IdS and thenumber of comorbid conditions (r 1176    0.06,  P    .04),whereas no such effect was observed for the number of drugs taken (r 1176   0.04,  P  .1).  Diabetes.  Fifty-one patients had diabetes (4.4%). Age-controlledanalysesrevealednoalteredolfactoryfunc-tion in diabetic patients (age, 63.1  1.6 y; IdS, 11.5  0.3; t  1.16 [  P  .24]). In these patients, ratings of olfactoryfunction were not correlated with IdS (r 51  0.13,  P  .4).  Hypertension.  One hundred ninety-nine patientshad arterial hypertension (16.8%). Age-controlled analy-ses revealed similar olfactory function in patients withand without hypertension patients (age, 59.1  0.8 y; IdS,12.3  0.1;  t  0.69 [  P  .4]).  Depression.  Twenty-nine patients indicated moodchanges in terms of depression (2.9%). Age-controlledanalyses showed no difference between self-identified de-pressive and nondepressive patients in terms of olfactoryfunction (age, 53.4    2.6 y; IdS, 12.8    0.4;  t    0.77[  P   .4]). Cancer (not related to otorhinolaryngology). Seventy-eight patients had undergone therapy or werestill undergoing treatment for malignancies outside theotorhinolaryngological sphere (6.6%). Age-controlled anal-yses showed significant effects of malignant diseases to-ward altering olfactory function (malignancy, IdS    11.7  0.3; no malignancy, IdS  12.4  0.1;  t  2.3,  P  .02).Olfactory self-ratings of the malignancy group correlatedsignificantly with the IdS (r 78   0.3,  P  .01). Cardiovascular problems.  Seventy-two patientsindicated cardiovascular problems (6.1%). Age-controlledanalyses showed no altered olfactory function in patients Fig. 2. Ratings of estimated loss of quality of life (LQ) attributable tofunctional anosmia for three groups of subjects in relation to olfac-tory function. Normosmic subjects estimate the hypothetical LQ tobe higher than do hyposmic and anosmic subjects. Laryngoscope 114: October 2004 Landis et al.: Frequency of Olfactory Dysfunction 1766  with cardiovascular problems compared with those with-out cardiovascular problems (mean age, 65.3  1.5 y; IdS,11.3  0.3;  t  1.5 [  P  .1]).  Liverproblems. Twenty-five patients indicated hav-ing chronic liver diseases (2%). Age-corrected analysesshowed chronic liver problems to significantly alter olfac-tory function (liver disease, IdS    11.2    0.4; no liverdisease, IdS  12.5  0.1;  t  2.19 [  P  .03]). For patientswith liver problems, IdS was not correlated with ratings of olfactory function (r 25   0.25,  P  .3).  Effect of Surgery on Olfactory Function Partial correlation controlling for age, total numberof comorbid conditions, and total amount of drugs takenrevealed small but significant negative correlations be-tween IdS and the total number of interventions (r 1176   0.13,  P  .001), nasal interventions (r 1176  0.06,  P  .02),andnonnasalinterventionssolely(r 1176  0.08,  P  .003).  Effect of Medication on Olfactory Function Partial correlation controlling for age revealed nosignificant correlation between IdS and the total numberof drugs taken (r 1176   0.04,  P  .1). Oral Contraceptives  Age-controlled comparison revealed a small but sig-nificant difference of IdS for women taking oral contracep-tives (the  “ pill group ” ) compared with the  “ non-pill group ” (mean age, 28.9    1.2 y; pill group, IdS    13.7    0.1;non-pill group, IdS  13.1  0.1;  t  2.3 [  P  .02]).  Effects of Age and Sex Olfactory identification ability decreased signifi-cantly with age in the studied sample (r 1176  0.16,  P  .001). Women had significantly better IdS (women, 12.9  0.1; men, 12.4    0.1;  t    3.5 [  P    .001]), although meanage for men and women was similar (men, 40.7    0.8 y;women, 41.6  0.8 y;  t  0.82 [  P  .5]).  Effects of Smoking Twenty percent of the participants declared them-selves as active smokers; the exact consumption had notbeen assessed. Age-controlled comparison revealed no dif-ference in IdS values between nonsmokers and smokers(nonsmokers, 12.9  0.1; smokers, 12.8  0.1;  t  0.5 [  P  .6]). DISCUSSION Theprincipalfindingsofthestudycanbedividedinto1) epidemiological data reflecting frequencies of diseasesand smell disorders and 2) effects of medical and personalfactors affecting olfaction.  Epidemiological Findings One of the exclusion criteria of the present study wassubjectively reported sinonasal complaints. In addition,patients received nasal endoscopy that revealed the pres-ence of nasal polyposis (NP) in almost 5% of the investi-gated otorhinolaryngological patients. These findings sup-port results by Larsen and Tos, 13 who claimed similarhigh incidence of NP based on long-term follow-up studiesand autopsy findings. Further, the present data confirmedthat incidence of NP increases with age, has a male pre-dominance, 14 and is associated with decreased olfactoryfunction. 15  Although the subjects did not appear to beaware of their nasal disease, they rated their state of hyposmia relatively accurately in that ratings of olfactoryfunction correlated with measured olfactory function. Thisconfirms recent findings 16 and corroborates argumentsclaiming olfactory dysfunction to be one of the earliestsigns of NP. 17 Taken together, the present findings em-phasize that human studies on olfaction should be con-trolled by nasal endoscopy to avoid biases because of pa-tients with nondiscovered NP. After having excluded patients with NP, the inci-dence of anosmia and hyposmia in the remaining 1176otorhinolaryngological patients was found to be slightlygreater than 20%, with 16% being hyposmic and 5% being functionally anosmic. These findings are in agreementwith previously reported data on the occurrence of olfac-tory dysfunction. 1,3 Similar to these authors, we found ageto be a main factor leading to chemosensory dysfunction.However, the present data also revealed functional anos-mia and hyposmia to occur in up to 5% and 15%, respec-tively, of subjects 65 years of age or younger (Fig. 1). Theincidence of functional anosmia in the present study issimilar to that of diseases such as diabetes. 18 In compar-ison, blindness is estimated to occur at a frequency of 0.7%to 1.5%. 19 Considering that our study excluded all people withsinonasal complaints, the present findings suggest theoccurrence of smell disorders to be underestimated. Be-cause chronic sinonasal problems are frequent within anormal population 20 and accompanied by hyposmia oranosmia in more than 60% of the cases, 21 the presence of dysosmias could be even higher in a normal populationthan reported in the present study.Ratings of olfactory function have been shown topoorly reflect measured olfactory function, but rather re-flect the perception of nasal patency. 22  Accordingly, nor-mosmic subjects were unable to estimate their olfactoryfunction, whereas hyposmic or anosmic subjects tended tobe quite accurate when the olfactory disorder was accom-panied by nasal discomfort (nasal obstruction, rhinor-rhea). In line with this, accuracy of olfactory rating washighest within the NP group.Parosmia and phantosmia, generally subsumed un-der the term of qualitative olfactory disorders, are oftenreported to occur after postviral and post-traumatic disor-ders. The causes of parosmia and phantosmia remainunknown, and up to the present, the incidence of qualita-tive olfactory disorders has been a matter of debate. 7,23,24 Our data confirmed previous findings of   “ qualitative ”  ol-factory dysfunction occurring, almost always, in combina-tion with  “ quantitative ”  smell disorders (hyposmia andanosmia). 24 Having said this, olfactory distortions werealso reported by normosmic subjects, suggesting thatthese symptoms are experienced more frequently thanestimated. Within the group of anosmic and hyposmicparticipants, 6% reported parosmia and 1.5% reportedphantosmia. Although this is in line with the previously Laryngoscope 114: October 2004 Landis et al.: Frequency of Olfactory Dysfunction 1767  estimated occurrence of smell distortions, 7 some authorshave claimed that as many as 30% of the quantitativelyimpaired patients experience parosmia. 23,25 Because nostandardized tool exists to reliably assess the presence of qualitative olfactory disorders, this controversy about theoccurrence of parosmia seems to reflect, at least in part,methodological issues.One question related to the estimated loss of qualityof life attributable to a hypothetical olfactory loss. Partic-ipants with higher IdS rated potential loss of quality of lifehigher than did those with lower IdS. This may be inter-preted to reflect that subjects with a higher IdS are likelyto attach more importance to the sense of smell 26 and thusmight enjoy daily olfactory impressions more consciouslythan other individuals. Hence, they might envision olfac-tory loss to have a stronger impact on their daily lifequality compared with subjects with lower olfactorysensitivity.  Effects of Medical and Personal Factors Affecting Olfaction In line with previous research, the present study didnot reveal effects of hypertension or cardiovascular prob-lems on olfaction. 27,28 Diabetic patients investigated inthe present study did not differ in terms of their olfactoryfunction compared with healthy controls, which is in con-trast to work by Jorgensen and Buch 29 and Weinstock etal., 30 who found diabetic patients to have a poorer sense of smell compared with healthy control subjects. According to normative data, 12 the IdS of the diabetic patients in thepresent study lies between the 10th and 20th percentilesfor their age group. Without reaching significance, thisindicates a tendency for decreased olfactory function of diabetic subjects. One reason for failing significance couldbe the smaller sample size consisting of only 51 diabeticpatients compared with 111 patients in the study of Wein-stock et al. 30 Nevertheless, in line with the previous stud-ies, the present data seem to confirm that olfactory dys-function in diabetes is not a major element of the disease.Further studies correlating smell dysfunction with thedegree of diabetic polyneuropathy are needed to clarifythis issue.Olfactory impairment was not seen in patients withself-reported depression, although depression has beenshown to interfere with olfactory function. 31 One probablereason for this finding is that the presently investigatedotorhinolaryngological patients either had only mildsymptoms of depression or received medical treatment.Chronic 32 andacute 33 liverdiseaseshavebeenshownto be associated with olfactory dysfunction. Although onlya small group of 25 participants reported liver problems,our findings suggest an association between olfactory dys-function and liver diseases, which may relate to possiblecentral nervous effects of the disease on olfactory function. 34  All patients who came to medical attention for oto-rhinolaryngological cancer follow-up or had an otorhino-laryngological cancer history were excluded from thestudy. Only the remaining participants with a history of other malignant diseases (eg, breast cancer or prostatecancer) were included. Among the tested patients, approx-imately 6% had undergone treatment or were still receiv-ing treatment for malignant disease, or both. Those pa-tients exhibited a significantly lower olfactory functionthan the noncancer control group. However, completetreatment details were not obtained, rendering any inter-pretation difficult. Nonetheless, a possible explanation forthe relatively lower degree of olfactory function could beexposure to cytostatic drugs or irradiation, or both. Al-thoughirradiationhasbeenreportedtohavesome,althoughslight and transient, effects on olfaction, 35 chemotherapieshave been shown to alter chemosensory functions. 36 Partial correlation controlling for age revealed asmall but significant negative correlation between olfac-tory function and the number of comorbid conditions,whereas no such effect was observed for the number of drugs taken. Although the literature of altered smell func-tion in relation to drug intake is abundant, 27,28,36 mostreports have been observations on a few, or even single,cases. Because not all drugs affect olfaction, it is notsurprising that there was no general effect of medicationon chemosensory function. This emphasizes the necessityof further studies focusing on individual drugs and theireffect on olfactory function. Because the range of medica-tions taken was extremely heterogeneous, further analy-ses of drug effects on olfaction appeared to be difficult,with exception of oral contraceptives. Interestingly,women taking oral contraceptives scored higher in theodor identification test. In animal studies, estrogensshowed protective effects towards the olfactory epitheli-um. 37 However, a recent study comparing postmeno-pausal women with and without hormone replacementtherapy could not find differences with respect to olfactoryfunction. 38  Although hormonal interference with olfactoryfunction has been reported, 39,40 the exact mechanism andsite of action are far from clear. One could speculate thatoral contraceptives balance fluctuations of olfactory func-tion during the menstrual cycle, 39,41,42 thereby rendering higher mean IdS in the group taking hormones.The observed negative correlation between the num-ber of surgical interventions and olfactory function re-mained significant after controlling for age, number of drugs taken, number of comorbid conditions, and nasalsurgical interventions. However, considering the weaksignificance, any interpretation of this effect has to becautious. However, the observed effect on chemosensoryfunction seemed to be independent of whether surgerywas performed in the head and neck region or elsewhere.This leads to speculations about an effect of general andlocal anesthesia on the olfactory system. To date, theliterature on olfactory dysfunction following general anes-thesia 36,43 is sparse. The present data may emphasize theneed for further studies on the possible influence of gen-eral and local anesthesia on olfaction.Smoking was not found to significantly affect olfac-tory function. 5 However, this is compatible with the viewthat the smoking-related decrease in olfactory function isnot a major effect but relates mainly to heavy smokers.The present study found sex and age to be the most im-portant determinants of olfactory function, with womenoutperforming men and olfaction decreasing dramaticallywith increasing age. Laryngoscope 114: October 2004 Landis et al.: Frequency of Olfactory Dysfunction 1768
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks