RS

SKa

Db

Uc

S

AfmTdstobaisocstprc©

Ko

ItvisesIs

povr

*4EAsv

Neuroscience 169 (2010) 415–421

0d

EDUCED OLFACTORY BULB VOLUME AND OLFACTORY

ENSITIVITY IN PATIENTS WITH ACUTE MAJOR DEPRESSION

tnbcnotss(eSicoosawsr(tttbstrt(te

dqfsecpoTlaitsa

. NEGOIAS,a I. CROY,a,b J. GERBER,c S. PUSCHMANN,a

. PETROWSKI,b P. JORASCHKYb AND T. HUMMELa*

Smell and Taste Clinic, Department of Otorhinolaryngology, University ofresden Medical School, Dresden, Germany

Clinic and Policlinic for Psychotherapy and Psychosomatic Therapy,niversity of Dresden Medical School, Dresden, Germany

Department of Neuroradiology, University of Dresden Medicalchool, Dresden, Germany

bstract—The purpose of this study was to assess olfactoryunction and olfactory bulb volume in patients with acute

ajor depression in comparison to a normal population.wenty-one patients diagnosed with acute major depressiveisorder and 21 healthy controls matched by age, sex andmoking behavior participated in this study. Olfactory func-ion was assessed in a lateralized fashion using measures ofdor threshold, discrimination and identification. Olfactoryulb volumes were calculated by manual segmentation ofcquired T2-weighted coronal slices according to a standard-

zed protocol. Patients with acute major depressive disorderhowed significantly lower olfactory sensitivity and smallerlfactory bulb volumes. Additionally, a significant negativeorrelation between olfactory bulb volume and depressioncores was detected. Their results provide the first evidence,o our knowledge, of decreased olfactory bulb volume inatients with acute major depression. These results might beelated to reduced neurogenesis in major depression thatould be reflected also at the level of the olfactory bulb.2010 IBRO. Published by Elsevier Ltd. All rights reserved.

ey words: olfaction, smell, major depression, olfactory bulb,lfactory sensitivity.

n recent years a series of studies has investigated olfac-ory function in various neuropsychiatric disorders (for re-iew see Atanasova et al., 2008). Motivating this researchs the partial overlap between emotional and olfactory braintructures, especially the limbic system and prefrontal ar-as. Dysfunctions in emotional processing areas are con-equently expected to also affect the olfactory perception.n turn, olfactory probes could provide a surrogate mea-ure of some neuropsychiatric symptoms.

Olfactory deficiencies have been established in schizo-hrenia (Turetsky et al., 2009a) whereas in affective dis-rders a review of the literature research provides contro-ersial results. Major depression is one such case, withesults covering almost the entire spectrum of olfactory

Corresponding author. Tel: �49-351-458-4189; fax: �49-351-458-326.-mail address: thummel@mail.zih.tu-dresden.de (T. Hummel).bbreviations: BDI, Beck’s depression inventory; MDD, major depres-

aive disorder; OB, olfactory bulb; SD, standard deviation; SVZ, sub-entricular zone.

306-4522/10 $ - see front matter © 2010 IBRO. Published by Elsevier Ltd. All rightoi:10.1016/j.neuroscience.2010.05.012

415

esting outcomes. Odor identification ability ranges fromormal (Amsterdam et al., 1987; Warner et al., 1990; Lom-ion-Pouthier et al., 2006; Swiecicki et al., 2009) to de-reased performance (Serby et al., 1990), compared to aormosmic control group. For threshold measurements,ne study reported no difference and no correlation be-ween olfactory sensitivity and depression quantificationcales for patients with acute depression, but better sen-itivity for patients after 42 days after initiation of treatmentGross-Isseroff et al., 1994) supporting the results from anarlier study (Serby et al., 1990). In line with these results,wiecicki et al. showed no olfactory sensitivity impairments

n unipolar depression (Swiecicki et al., 2009). On theontrary, Pause and colleagues reported strongly reducedlfactory sensitivity in acute-phase major depressive dis-rder (MDD) that correlated negatively with depressioncores. When investigating the same group of patientsfter successful therapy, no olfactory sensitivity differenceas found compared to controls (Pause et al., 2001). Theame dynamic was in cerebral processing as assessed byeduced amplitudes of olfactory event related potentialsERPs) peaks in patients with MDD at the beginning of theherapy, without reproduction of the effect after successfulreatment (Pause et al., 2003). Decreased olfactory sensi-ivity in MDD has been replicated in a recent study (Lom-ion-Pouthier et al., 2006). Seasonal affective disordertudies have also studied the relationship between olfac-ory function and depression, with the same contrastingesults. One study reported no difference in olfactoryhresholds and identification between patients and controlsPostolache et al., 1999), while another showed betterhresholds for patients, regardless of season (Postolachet al., 2002).

It seems that the relationship between olfaction andepression is reciprocal. Olfactory impairment alters theuality of life (Hummel and Nordin, 2005); loss of olfactoryunction is typically associated with increased depressiveymptoms (Deems et al., 1991; Gudziol et al., 2009a; Seot al., 2009). A recent report showed depressive symptomsorrelate negatively with olfactory sensitivity in healthyarticipants (Pollatos et al., 2007) analogous to the resultsf Pause et al. on patients with MDD (Pause et al., 2001).he question whether olfactory impairment alone could

ead to or trigger a major depression episode remains to benswered. Two considerations are crucial when trying to

nvestigate olfactory function in MDD. First, it is importanto discriminate between testing patients with acute depres-ion and patients in a remission, or between medicatednd unmedicated patients. Differences in methodological

pproaches, number of participants and heterogeneity in

s reserved.

mi

im(1rMPfotmdwtat(atdrctclMctt

vi(iaAe1mec

tcat

P

THMDp

d

sbaa(ispmmtnt(itaot

poT2yatsMice((lppo

pr6(tBrggi

s(B

b

T(

BAA

DD

S. Negoias et al. / Neuroscience 169 (2010) 415–421416

edication and moment of testing could account for thenconsistencies in results seen in previous studies.

Second, some authors divide olfactory function testsnto those focusing on peripheral processing (threshold

easurement) and those focusing on cognitive processinge.g., odor discrimination or odor identification; Cain,979), concluding that they should be examined sepa-ately. Most of the studies investigating olfactory function inDD used identification tests, mainly the University ofennsylvania Smell identification Test (UPSIT). While ol-

actory identification scores correlate with olfactory thresh-lds (Doty et al., 1984; Hummel et al., 1997), independentesting of various olfactory functions could provide infor-ation on the level of the interaction between olfaction andepression. On one hand, cognitive impairment associatedith major depression could interfere with odor identifica-

ion or discrimination abilities (Austin et al., 2001; Marvelnd Paradiso, 2004). On the other hand, it has been shownhat bulbectomized rodents provide a model for depressionSong and Leonard, 2005), with correction of symptomsfter chronic administration of antidepressants. The au-hors of this study presumed that bulbectomy inducedysfunction in the cortical-hippocampal-amygdalar circuit,esponsible for modulating behavioral responses. Specifi-ally, the olfactory bulb (OB) sends inhibitory projections tohe amygdala, which is involved preferentially in the pro-essing of fear and sadness (Costafreda et al., 2008). Inight of these ideas some authors have speculated that in

DD, dysfunction at the level of the olfactory bulb couldause not only the reduced olfactory sensitivity, but alsohe increased sadness and fear, through disinhibition ofhe amygdala (Pause et al., 2001).

The human OB is a highly plastic structure whoseolume reflects changes in olfactory sensitivity, as shownn patients with post-traumatic chemosensory deficitsYousem et al., 1996a, 1999; Rombaux et al., 2006), post-nfectious olfactory deficits (Mueller et al., 2005b; Rombaux etl., 2006), congenital anosmia (Yousem et al., 1996b;bolmaali et al., 2002), neurodegenerative disorders (Muellert al., 2005a) and in normosmic participants (Yousem et al.,998). A recent study provided normative data in a normos-ic population (Buschhuter et al., 2008). To our knowl-dge, no study has yet been published addressing theonnection between the OB and depression in humans.

The purpose of the present study was to assess olfac-ory function and OB volume in patients with acute MDDompared to a normal population matched for age, sex,nd smoking behavior, using a standardized test for odorhreshold, discrimination and identification.

EXPERIMENTAL PROCEDURES

articipants and experimental protocol

he study was performed in accordance to the Declaration ofelsinki on Biomedical Studies Involving Human Subjects (Worldedical Association, 1997) and was approved by the University ofresden Medical Faculty Ethics Review Board. All participantsrovided written informed consent before inclusion in the study.

Twenty-five inpatients of the Clinic for Psychosomatic Disor-

ers and 22 healthy controls were invited to participate in this

tudy. All patients had been admitted to the hospital and treatedecause of acute MDD. They had been previously diagnosed withcute MDD by the treating physician /clinical psychologist (KP)fter completing a Composite International Diagnostic InterviewCIDI, DIA-X German version; Wittchen and Pfister, 1997) accord-ng to DSM IV criteria. Healthy controls were recruited via posterset in the University Clinic area and reimbursed for their partici-ation. Before proceeding with olfactory and volumetric measure-ents, the testing protocol for all participants included a detailededical history review as well as otorhinolaryngological examina-

ion, comprising nasal endoscopy, which ensured exclusion ofasal or internal pathology potentially causing olfactory dysfunc-ion. All participants underwent a mini mental state examinationMMSE; Folstein et al., 1975) to screen for possible cognitivempairment. Additionally, all participants were asked to completehe German version of Beck’s Depression Inventory (BDI; Beck etl., 1961 German version; Hautzinger et al., 1995) and to rate theirlfactory function on a 7 points scale ranging from “extremely bad”o “very good.”

Four patients were excluded because of concomitant nasalathology or existence of comorbidities known to interfere withlfactory function (severe septum deviation, sinonasal disease).he final group included four men and 17 women, aged between1 and 55 years (mean�standard deviation (SD)�36.86�10.13ears); 14 were non-smokers, and seven smokers. Demographicnd illness-related parameters are shown in Table 1. Comorbidi-ies included: somatoform disorders (12 cases) posttraumatictress disorder (nine cases) and anxiety disorders (17 cases).edication of the patients included selective serotonin reuptake

nhibitors (SSRI: Citalopram, Escitalopram, Paroxetine) in fiveases, tricyclic antidepressants in three cases (Mirtazapin, Dox-pin, Primipramin), serotonin-norepinephrine reuptake inhibitorsSNRI: Venlafaxin) in three cases, anticonvulsants in seven casesCarbamazepin, Pregabalin, Valproat), one with zinc and one withithium. Additionally four patients received neuroleptic drugs (Ris-eridon and Quetiapin), as well as analgesics (five cases) androton pump inhibitors (three cases), while five patients were freef medication.

One subject from the control group was excluded because ofresumption of incidental MRI findings and failure to completeequired tests. The final control group was composed out of

males and 15 females, aged between 20 and 52 yearsmean�SD� 39.62�11.39 years), including 19 non-smokers andwo smokers. None of the controls scored higher than nine on theDI questionnaire (mean�SD�3.12�2.91) or reported psychiat-

ic diagnoses in their personal history. Data for part of the controlroup (15 participants) were randomly selected according to ageroups from a database of a previous study that followed the same

nclusion criteria as the present study (Buschhuter et al., 2008).The groups did not differ in terms of age (T40�0.83, P�0.41),

ex distribution (Chi-square�.52, P�0.47) and smoking habitsChi-square�4.30, P�0.12) but significantly differed regardingDI scores (T40��9.8, P�.001).

Olfactory function was assessed using the “Sniffin’ Sticks” testattery (Burghart GmbH, Wedel, Germany) following a standard-

able 1. Demographic and illness related parameters for patientsn�21)

Minimum Maximum Mean SD

DI 11 51 29.67 10.84ge (y) 21 55 36.86 10.13ge at debut ofdisease (y)

9 42 20.25 9.65

uration of disease (y) 2 40 15.70 11.88uration of current 16 75 43.55 19.95

episode (d)

iiPaicnswmsTT

(cbsoqisas(pwvntstti(na

S

DC

a““DGdrora

Tmc

ftv0nwtsaP2

f2tco3

Fra

Ttaa

T

T

T

D

I

O

O

O

S. Negoias et al. / Neuroscience 169 (2010) 415–421 417

zed procedure (Hummel et al., 1997; Kobal et al., 2000). The testncluded measurements for odor thresholds (T �16 dilutions ofEA, single staircase procedure, with No. 16 marking the lowestvailable odor concentration, therefore the best olfactory sensitiv-

ty), odor discrimination (D �16 triplets, 3-alternative forcedhoice) and odor identification (I �16 common odors, four alter-ative forced choice). Threshold measurements were performedeparately for the left and right nostril (Buschhuter et al., 2008),hile discrimination and identification were assessed in a birhinalode. The sum of the scores from the three individual tests (TDI

core) defined the olfactory function (Wolfensberger et al., 2000).he results of the best nostril were used to calculate the overallDI score for group comparisons (compare; Frasnelli et al., 2002).

MRI measurements were performed with a 1.5-Tesla scannerSonata Vision; Siemens, Erlangen, Germany) using an eighthannel-head coil. The investigation protocol included one wholerain anatomical sequence without interslice gap (5-mm-thicktandard T1-weighted 3D sequence) for every participant to ruleut any organic brain disorders. The OB sequence included ac-uisition of 2-mm-thick T2-weighted fast spin-echo images without

nterslice gap in the coronal plane covering the anterior and middleegments of the base of the skull. Images were processed offlinend left and right OBs limits were drawn manually on each coronallice using the AMIRA 3D visualization and modeling systemVisage Imaging, Carlsbad, USA). OB volumes were calculated bylanimetric manual contouring (surface in mm2) and all surfacesere added and multiplied by 2 (2-mm slice thickness) to obtain aolume in cubic millimeters. The change of diameter at the begin-ing of the olfactory tract was used as the distal demarcation ofhe OB. OB measurements of all data were performed by theame experimenter (SN), blinded to the group category or olfac-ory test results. It included re-analyses of the data selected fromhe laboratory database (15 participants). For reliability purposes,ntra-class correlation coefficients for these data were calculatedICC�0.70, P�0.003). The OB volume corresponding to the bestostril was defined as “best OB” and was used for correlationnalyses with different functional parameters.

tatistical analyses

ata was analyzed using SPSS 15.0 for Windows (SPSS Inc.,

able 2. Results (mean, SD) and group comparison (t-Test) for odorhresholds (T-left, right and best), discrimination (D), identification (I)nd olfactory bulb volume (OB-left, right and best) between controlsnd patients (df�degrees of freedom)

Group Number ofparticipants

Mean SD t-test sign(2-tailed)

_right Control 21 8.54 2.19 .038Patients 18 6.97 2.37

_left Control 21 8.40 1.68 .041Patients 18 6.84 2.84

_best Control 21 9.14 1.89 .032Patients 21 7.56 2.67Control 21 12.76 2.14 .890Patients 21 12.66 2.28Control 21 13.76 1.09 .557Patients 21 14.00 1.48

B_right Control 21 65.30 14.97 .028Patients 21 55.84 11.77

B_left Control 21 63.85 11.51 .026Patients 21 55.64 11.47

B_best Control 21 64.83 13.58 .061Patients 18 57.30 10.69

hicago, IL, USA). Results were submitted to analyses of vari-2i

nce for repeated measures, adopting “side of testing” (left/right),test” (threshold, OB volume) as within-subjects-factors; the factorgroup” (patients/controls) was used as a between-subject-factor.egrees of freedom were adjusted according to Greenhouse-eisser. Additional comparisons for the functional and volumetricata were performed using t-tests for independent samples. Theesults may be prone to type 1 statistical error because the levelf significance was not corrected for multiple tests. Pearson cor-elations between volumetric and functional measurements werelso calculated. The level of significance was set at 0.05.

RESULTS

he results for psychophysical and volumetric measure-ents are listed in Table 2 separately for patients and

ontrols.No difference was found in terms of self-rated olfactory

unction (Chi-square�6.72, P�0.35). A significant effect ofhe factor “group” was present only for thresholds and OBolume (F[1,37]�5.32, P�0.027, and F[1,40]�5.73, P�.021, respectively), unlike for measures of odor discrimi-ation and identification. No “group X side” interactionsere found significant. Regarding olfactory thresholds, t-

ests for paired samples revealed significantly lower sen-itivity for patients compared to controls for both later-lized (right: T[37]�2.15, P�0.038, left: T[37]�2.11,�0.041) (Fig. 1) and best nostril measurements (T[37]�.21, P�0.032).

The patients group also exhibited smaller OB volumesor side-to-side comparisons with controls (right T[40]�.27, P�0.028; left T[40]�2.21, P�0.026) (Fig. 2), while a

rend towards significance was revealed for “best bulb”omparison (T[37]�1.93, P�0.061). OB volumes variedver a wide range: 37.4–98.1 mm3 for controls and6.4–86 mm3 for patients. Intraindividual variation for

ig. 1. Box plot of olfactory thresholds measured separate for left andight nostril in patients and controls (where 16 represents the lowestvailable odor concentration). The box-whisker plots show the 5th,

5th, 50th, 75th and 95th percentiles. The asterisk (*) indicates signif-

cant differences between patients and controls at a level of P�0.05.

Otsr

uwsrwioolrOtsmr

Tptga

d2bta

sttv

p

FatsP

Fvctsr(

F

S. Negoias et al. / Neuroscience 169 (2010) 415–421418

B volume was much smaller, with a significant correla-ion emerging (P�.001) between left and right-sided mea-urements for both patients and controls (r21�0.87 and

21�0.79, respectively).A significant correlation between left-sided OB vol-

mes and left-sided odor thresholds (r39�0.37, P�0.02)as observed: the lower the olfactory sensitivity, themaller the OB volume. However the correlation betweenight-sided OB volumes and right-sided odor thresholdsas not significant (r39�0.19, P�0.26) (Fig. 3). No signif-

cant correlations were found between OB volume anddor discrimination or odor identification scores. Furthern, depression scores (BDI) showed high negative corre-

ations with OB volumes (left: r37��0.40, P�0.014; right:

37��0.37, P�0.025, best: r34��0.34, P�0.044) (Fig. 4).nly the olfactory sensitivity as reflected in the best nostril

hreshold was found to significantly correlate with the BDIcore (r37��0.34, P�0.04), while lateralized thresholdeasures missed statistical significance (left or right:

34��0.28, P�0.10).

DISCUSSION

he present study points to the following major results: (1)atients with acute MDD have a reduced olfactory sensi-

ivity and smaller OB volumes than a normosmic controlroup; (2) BDI scores correlate negatively with OB volumend olfactory sensitivity.

These results are in line with previous findings of re-uced olfactory sensitivity in acute MDD (Pause et al.,001, 2003; Lombion-Pouthier et al., 2006). A correlationetween olfactory function and depression scores/symp-oms has also been shown in previous studies (Pause etl., 2001; Pollatos et al., 2007). In addition, we demon-

ig. 2. Box plot of olfactory bulb volumes measured separately for leftnd right nostril in patients and controls. The box-whisker plots showhe 5th, 25th, 50th, 75th and 95th percentiles. The asterisk (*) indicates

ignificant differences between patients and controls at a level of�0.05 for the same measures between patients and controls.

Or

trate for the first time that the decrease of olfactory func-ion is accompanied by smaller OB volumes in MDD pa-ients, and that depression scores correlate with OBolume.

A range of theories has been proposed to explain aotential olfactory deficit in MDD. Some authors point to

ig. 3. Left and right odor thresholds plotted against left and right OBolume, respectively (where 16 represents the lowest available odoroncentration). Left-sided OB volumes exhibited a significant correla-ion with left-sided odor thresholds (r39�0.37, P�0.02) (dotted regres-ion line), while correlations between right-sided OB volumes andight-sided odor thresholds did not reach the level of significancer39�0.19, P�0.26, continuous regression line).

ig. 4. BDI scores plotted separately against left and right OB volume.

B volume shows significant correlation with BDI scores (left:

37��.40, P�0.01; right: r37��.37, P�0.02).

tdcntpOtcaspcp

baseat1hsw(fMaetpc2ntdnnrttt(dsahsmrdm(av(so

a2vielerfbpsbptdIastr1fwaws2ptitiitAat

pkaoctapmodtitiTnds

S. Negoias et al. / Neuroscience 169 (2010) 415–421 419

he primary olfactory cortex and the OB as the site ofysfunction, since tests involving cognitive levels of pro-essing, like olfactory discrimination and identification, doot reveal any difference in patients with MDD compared

o controls. These results are also confirmed by theresent study. An additional model postulates that primaryB dysfunction in MDD is responsible for reduced olfac-

ory perception leading to amygdala disinhibition whichonsequently would affect emotional responses (Pause etl., 2001), in analogy to the bulbectomy model of depres-ion in rats (see Introduction). Abnormal functionality in theara-limbic brain areas (mainly amygdala and orbitofrontalortex) has also been hypothesized to affect early olfactoryrocessing in MDD (Pause et al., 2003).

The present results raise the possibility of a relationetween neurogenesis and MDD. Neurogenesis is presentt the level of dentate nucleus of the hippocampus andubventricular zone (SVZ) in mammalian brains (Erikssont al., 1998; Gould and Gross, 2002). Interestingly, someuthors consider the dentate nucleus of the hippocampuso be closely linked to the olfactory system (Vanderwolf,992, 2001). Decreased neurogenesis at the level of theippocampus has been shown in animal models of depres-ion (Jaako-Movits et al., 2006; Kronenberg et al., 2009),hile human studies report atrophic hippocampus in MDD

McKinnon et al., 2009). Neurogenesis in the hippocampusormation is suppressed by factors that predispose toDD, like prolonged stress and this process is reversed byntidepressant therapy (see; Paizanis et al., 2007a; Pererat al., 2008; Boldrini et al., 2009). Consequently, the idea

hat impaired neurogenesis could be involved in theathoetiology of MDD has drawn substantial attention andontroversy (see; Paizanis et al., 2007b; Perera et al.,008). It has been suggested that newborn cells might beeeded in the adult hippocampus for linking external con-ext to emotions; disruption of this process by stress-in-uced suppression of neurogenesis is thought to lead toegative mood that could be resolved by the stimulation ofeurogenesis by antidepressant drugs. In the same theo-etical framework, we might discuss the connection be-ween OB, neurogenesis and MDD. Animal studies showhe rostral migratory stream provides progenitor cells fromhe SVZ to the olfactory bulb in both rodents and primatesAltman, 1969; Gheusi and Lledo, 2007). Although underebate, some authors include the OB within the regionshowing cell proliferation in the human brain by analogy tonimal studies (Curtis et al., 2007; Kam et al., 2009). Aigh incidence of OB ventricles found in a recent studyupports this theory (Smitka et al., 2009). The OB in hu-ans has been shown to be a highly plastic structure

esponding to sensory input. Mueller et al. showed re-uced OB volumes in patients with post viral and posttrau-atic olfactory loss compared to a normosmic population

Mueller et al., 2005b). The same population measuredfter 15 months showed significant increase of the OBolume that paralleled the recovery of olfactory sensitivityHaehner et al., 2008). Similarly, a recent longitudinaltudy provided evidence of increased OB volume and

lfactory function in patients with chronic rhinosinusitis C

fter functional endoscopic sinus surgery (Gudziol et al.,009b). It has been speculated that the increase of OBolume is partly regulated by a bottom-up process involv-

ng sensory input from the olfactory epithelium. Our currentvidence suggests potential simultaneous top-down regu-

ation of the OB volume. In MDD, suppressed neurogen-sis could also be reflected at the OB level and cause theeduced olfactory sensitivity. Support for this idea comesrom animal studies, where reduced neurogenesis haseen shown also at the level of the SVZ which providesrogenitor cells to the OB, in mice exposed to chronictress (Mineur et al., 2007). This speculation is sustainedy a correlation of BDI scores with OB volume, found in theresent study. Similar to decreased hippocampus volumeshat resolve after antidepressant treatment, one could pre-ict an increase of the OB volume with successful therapy.mprovement of olfactory sensitivity with treatment of ancute depressive episode, regarding of the level of thetarting point (reduced, the same or increased) comparedo a normosmic control population has been previouslyeported (Suffin and Gitlin, 1986; Gross-Isseroff et al.,994; Pause et al., 2001; Postolache et al., 2002). Aollow-up study is currently ongoing to further investigatehether this effect is reflected also in the OB volume. Anlternative hypothesis was proposed by Turetsky et al.ho observed decreased OB and decreased olfactory sen-itivity in patients with schizophrenia (Turetsky et al.,000). Recent results coming from the same laboratoryoint to a potential olfactory receptor neuron dysfunction inhese patients (Turetsky et al., 2009) as showed by thencreased ORN depolarization responses following olfac-ory stimulation. Corroborating the clinical and morpholog-cal data, the authors interpreted these findings as proof ofncreased neuronal proliferation associated with dysfunc-ional olfactory receptor development in schizophrenia.nalogously, the reduced bulb volume in patients withcute depression could reflect a pre-existing condition athe level of the olfactory epithelium.

A limitation of the present study may relate to theatients’ use of drugs. Although antidepressants arenown to affect gustatory function in humans (Schiffmannd Graham, 2000), there is no evidence of quantitativelfactory impairment coming from human studies. Onease report on Citalopram associates this drug with induc-ion of parosmia (“unfavourable and intolerable smell”) thatppeared 7 weeks after the treatment initiation and disap-eared upon termination (Ghanizadeh, 2007). Some ani-al studies suggest, nevertheless, that SSRIs might affectlfactory function: Rolipram was shown in mice to impairetection accuracy of 1-propanol at relatively low concen-rations (Pho et al., 2005). Citalopram and Clomipraminenduced a decrease in olfactory sensitivity after 3 weeks ofreatment in mice (Lombion et al., 2008). The patientsncluded in this study had a long history of the disease.herefore possible chronic effects of the medication can-ot be excluded. However, as in the present study, re-uced olfactory sensitivity was also reported in antidepres-ant drug free patients with MDD (Pause et al., 2005).

onsequently a potential confounding effect of the antide-

prdivtapwc

Trcvdcutmr

ASt

A

A

A

A

A

B

B

B

C

C

C

D

D

E

F

F

G

G

G

G

G

G

H

H

H

H

J

K

K

K

L

S. Negoias et al. / Neuroscience 169 (2010) 415–421420

ressant medication does not seem to play a significantole in the present study. Lack of motivation in patients withepression could have possibly led to poorer performance

n the olfactory tests; the correlation with depression se-erity seems to support this possibility. Nevertheless, pa-ients participated on a voluntary basis and were free tobandon the study at any time. In fact, the investigatedatients showed a clear interest in the olfactory testshich—apparently—was not different from that of healthyontrols.

CONCLUSION

he present results provide further evidence supporting aeduced olfactory sensitivity with maintained olfactory dis-rimination and identification abilities in patients with se-ere MDD compared to healthy controls. Furthermore,ecreased OB volume was shown in MDD patients, ac-ompanied by a significant correlation between OB vol-me, olfactory thresholds and depression scores, respec-

ively. A hypothetical mechanism explaining these resultsight involve reduced neurogenesis in MDD that could be

eflected also at the OB level.

cknowledgments—We would like to thank Artin Arshamian, Anjaymank and Dorothee Buschhüter for their contribution to collec-

ion of the data.

REFERENCES

bolmaali ND, Hietschold V, Vogl TJ, Huttenbrink KB, Hummel T(2002) MR evaluation in patients with isolated anosmia since birthor early childhood. AJNR Am J Neuroradiol 23:157–164.

ltman J (1969) Autoradiographic and histological studies of postnatalneurogenesis. IV. Cell proliferation and migration in the anteriorforebrain, with special reference to persisting neurogenesis in theolfactory bulb. J Comp Neurol 137(4):433–457.

msterdam JD, Settle RG, Doty RL, Abelman E, Winokur A (1987)Taste and smell perception in depression. Biol Psychiatry 22(12):1481–1485.

tanasova B, Graux J, El Hage W, Hommet C, Camus V, Belzung C(2008) Olfaction: a potential cognitive marker of psychiatric disor-ders. Neurosci Biobehav Rev 32(7):1315–1325.

ustin MP, Mitchell P, Goodwin GM (2001) Cognitive deficits in de-pression: possible implications for functional neuropathology. Br JPsychiatry 178:200–206.

eck AT, Ward CM, Mendelson M, Mock JE, Erbaugh JK (1961) Aninventory for measuring depression. Arch Gen Psychiatry 4:561–571.

oldrini M, Underwood MD, Hen R, Rosoklija GB, Dwork AJ, JohnMann J et al. (2009) Antidepressants increase neural progenitorcells in the human hippocampus. Neuropsychopharmacology34(11):2376–2389.

uschhuter D, Smitka M, Puschmann S, Gerber JC, Witt M, AbolmaaliND et al. (2008) Correlation between olfactory bulb volume andolfactory function. Neuroimage 42(2):498–502.

ain WS (1979) To know with the nose: keys to odor identification.Science 203:467–470.

ostafreda SG, Brammer MJ, David AS, Fu CH (2008) Predictors ofamygdala activation during the processing of emotional stimuli: ameta-analysis of 385 PET and fMRI studies. Brain Res Rev58(1):57–70.

urtis MA, Kam M, Nannmark U, Anderson MF, Axell MZ, Wikkelso Cet al. (2007) Human neuroblasts migrate to the olfactory bulb via a

lateral ventricular extension. Science 315(5816):1243–1249.

eems DA, Doty RL, Settle RG, Moore-Gillon V, Shaman P, MesterAF et al. (1991) Smell and taste disorders: a study of 750 patientsfrom the University of Pennsylvania Smell and Taste Center. ArchOtorhinolaryngol Head Neck Surg 117:519–528.

oty RL, Shaman P, Dann M (1984) Development of the University ofPennsylvania smell identification test: a standardized microencap-sulated test of olfactory function. Physiol Behav 32(3):489–502.

riksson PS, Perfilieva E, Bjork-Eriksson T, Alborn AM, Nordborg C,Peterson DA et al. (1998) Neurogenesis in the adult human hip-pocampus. Nat Med 4(11):1313–1317.

olstein MF, Folstein SE, McHugh PR (1975) “Mini-mental state.” Apractical method for grading the cognitive state of patients for theclinician. J Psychiatr Res 12:189–198.

rasnelli J, Livermore A, Soiffer A, Hummel T (2002) Comparison oflateralized and binasal olfactory thresholds. Rhinology 40:129–134.

hanizadeh A (2007) Unfavorable smell with citalopram?. J Clin Psy-chopharmacol 27(5):528–529.

heusi G, Lledo PM (2007) Control of early events in olfactory pro-cessing by adult neurogenesis. Chem Senses 32(4):397–409.

ould E, Gross CG (2002) Neurogenesis in adult mammals: someprogress and problems. J Neurosci 22(3):619–623.

ross-Isseroff R, Luca-Haimovici K, Sasson Y, Kindler S, Kotler M,Zohar J (1994) Olfactory sensitivity in major depressive disorderand obsessive compulsive disorder. Biol Psychiatry 35(10):798–802.

udziol V, Wolff-Stephan S, Aschenbrenner K, Joraschky P, HummelT (2009a) Depression resulting from olfactory dysfunction is asso-ciated with reduced sexual appetite—a cross-sectional cohortstudy. J Sex Med 6(7):1924–1929.

udziol V, Buschhuter D, Abolmaali N, Gerber J, Rombaux P, Hum-mel T (2009b) Increasing olfactory bulb volume due to treatment ofchronic rhinosinusitis—a longitudinal study. Brain 132(Pt 11):3096–3101.

aehner A, Rodewald A, Gerber JC, Hummel T (2008) Correlation ofolfactory function with changes in the volume of the human olfac-tory bulb. Arch Otolaryngol Head Neck Surg 134(6):621–624.

autzinger M, Bailer M, Worall H, Keller F (1995) Beck-depressions-inventar (BDI). Göttingen: Hogrefe.

ummel T, Nordin S (2005) Olfactory disorders and their conse-quences for quality of life—a review. Acta Otolaryngol 125:116–121.

ummel T, Sekinger B, Wolf S, Pauli E, Kobal G (1997) “Sniffin’sticks”: olfactory performance assessed by the combined testing ofodor identification, odor discrimination and olfactory threshold.Chem Senses 22:39–52.

aako-Movits K, Zharkovsky T, Pedersen M, Zharkovsky A (2006)Decreased hippocampal neurogenesis following olfactory bulbec-tomy is reversed by repeated citalopram administration. Cell MolNeurobiol 26(7–8):1559–1570.

am M, Curtis MA, McGlashan SR, Connor B, Nannmark U, Faull RL(2009) The cellular composition and morphological organization ofthe rostral migratory stream in the adult human brain. J ChemNeuroanat 37(3):196–205.

obal G, Klimek L, Wolfensberger M, Gudziol H, Temmel A, Owen CMet al. (2000) Multicenter investigation of 1,036 subjects using astandardized method for the assessment of olfactory function com-bining tests of odor identification, odor discrimination, and olfactorythresholds. Eur Arch Otorhinolaryngol 257:205–211.

ronenberg G, Kirste I, Inta D, Chourbaji S, Heuser I, Endres M et al.(2009) Reduced hippocampal neurogenesis in the GR(�/�) ge-netic mouse model of depression. Eur Arch Psychiatry Clin Neu-rosci 259(8):499–504.

ombion-Pouthier S, Vandel P, Nezelof S, Haffen E, Millot JL (2006)Odor perception in patients with mood disorders. J Affect Disord

90(2–3):187–191.

L

M

M

M

M

M

P

P

P

P

P

P

P

P

P

P

R

S

S

S

S

S

S

S

T

T

T

V

V

W

W

W

Y

Y

Y

Y

S. Negoias et al. / Neuroscience 169 (2010) 415–421 421

ombion S, Morand-Villeneuve N, Millot JL (2008) Effects of anti-depressants on olfactory sensitivity in mice. Prog Neuropsychop-harmacol Biol Psychiatry 32(3):629–632.

arvel CL, Paradiso S (2004) Cognitive and neurological impairmentin mood disorders. Psychiatr Clin North Am 27(1):19–36, vii–viii.

cKinnon MC, Yucel K, Nazarov A, MacQueen GM (2009) A meta-analysis examining clinical predictors of hippocampal volume inpatients with major depressive disorder. J Psychiatry Neurosci34(1):41–54.

ineur YS, Belzung C, Crusio WE (2007) Functional implications ofdecreases in neurogenesis following chronic mild stress in mice.Neuroscience 150(2):251–259.

ueller A, Abolmaali ND, Hakimi AR, Gloeckler T, Herting B, Reich-mann H et al. (2005a) Olfactory bulb volumes in patients withidiopathic Parkinson’s disease a pilot study. J Neural Transm112(10):1363–1370.

ueller A, Rodewald A, Reden J, Gerber J, von Kummer R, HummelT (2005b) Reduced olfactory bulb volume in post-traumatic andpost-infectious olfactory dysfunction. Neuroreport 16(5):475–478.

aizanis E, Kelai S, Renoir T, Hamon M, Lanfumey L (2007a) Life-longhippocampal neurogenesis: environmental, pharmacological andneurochemical modulations. Neurochem Res 32(10):1762–1771.

aizanis E, Hamon M, Lanfumey L (2007b) Hippocampal neurogen-esis, depressive disorders, and antidepressant therapy. NeuralPlast 2007:73754.

ause B, Lembcke J, Reese I, Hinze-Selch D, Aldenhoff J, Ferstl R(2005) Reduced olfactory sensitivity in antidepressant drug freepatients with major depression. Z Klin Psychol Psychother 34:79–85.

ause BM, Miranda A, Goder R, Aldenhoff JB, Ferstl R (2001) Re-duced olfactory performance in patients with major depression.J Psychiatr Res 35(5):271–277.

ause BM, Raack N, Sojka B, Goder R, Aldenhoff JB, Ferstl R (2003)Convergent and divergent effects of odors and emotions in depres-sion. Psychophysiology 40(2):209–225.

erera TD, Park S, Nemirovskaya Y (2008) Cognitive role of neuro-genesis in depression and antidepressant treatment. Neuroscien-tist 14(4):326–338.

ho V, Butman ML, Cherry JA (2005) Type 4 phosphodiesteraseinhibition impairs detection of low odor concentrations in mice.Behav Brain Res 161(2):245–253.

ollatos O, Kopietz R, Linn J, Albrecht J, Sakar V, Anzinger A et al.(2007) Emotional stimulation alters olfactory sensitivity and odorjudgment. Chem Senses 32(6):583–589.

ostolache TT, Doty RL, Wehr TA, Jimma LA, Han L, Turner EH et al.(1999) Monorhinal odor identification and depression scores inpatients with seasonal affective disorder. J Affect Disord 56(1):27–35.

ostolache TT, Wehr TA, Doty RL, Sher L, Turner EH, Bartko JJ et al.(2002) Patients with seasonal affective disorder have lower odordetection thresholds than control subjects. Arch Gen Psychiatry59(12):1119–1122.

ombaux P, Mouraux A, Bertrand B, Nicolas G, Duprez T, Hummel T(2006) Retronasal and orthonasal olfactory function in relation toolfactory bulb volume in patients with posttraumatic loss of smell.

Laryngoscope 116(6):901–905.

chiffman SS, Graham BG (2000) Taste and smell perception affectappetite and immunity in the elderly. Eur J Clin Nutr 54 (Suppl3):S54–S63.

eo HS, Jeon KJ, Hummel T, Min BC (2009) Influences of olfactoryimpairment on depression, cognitive performance, and qualityof life in Korean elderly. Eur Arch Otorhinolaryngol 266(11):1739–1745.

erby M, Larson P, Kalkstein D (1990) Olfactory sense in psychoses.Biol Psychiatry 28(9):830.

mitka M, Abolmaali N, Witt M, Gerber JC, Neuhuber W, BuschhueterD et al. (2009) Olfactory bulb ventricles as a frequent finding inmagnetic resonance imaging studies of the olfactory system. Neu-roscience 162(2):482–485.

ong C, Leonard BE (2005) The olfactory bulbectomised rat as amodel of depression. Neurosci Biobehav Rev 29(4–5):627–647.

uffin SC, Gitlin M (1986) Olfaction in depression and recovery: a newmarker. Abstracts of the American Psychiatry Association Meeting,pp 87 Washington, DC American Psychiatry Press.

wiecicki L, Zatorski P, Bzinkowska D, Sienkiewicz-Jarosz H, Szyn-dler J, Scinska A (2009) Gustatory and olfactory function in pa-tients with unipolar and bipolar depression. Prog Neuropsychop-harmacol Biol Psychiatry 33(5):827–834.

uretsky BI, Hahn CG, Arnold SE, Moberg PJ (2009) Olfactory recep-tor neuron dysfunction in schizophrenia. Neuropsychopharmacol-ogy 34(3):767–774.

uretsky BI, Hahn CG, Borgmann-Winter K, Moberg PJ (2009a)Scents and non-sense: olfactory dysfunction in schizophrenia.Schizophr Bull 35: 1117–1131.

uretsky BI, Moberg PJ, Yousem DM, Doty RL, Arnold SE, Gur RE(2000) Reduced olfactory bulb volume in patients with schizophre-nia. Am J Psychiatry 157:828–830.

anderwolf CH (1992) Hippocampal activity, olfaction, and sniffing: anolfactory input to the dentate gyrus. Brain Res 593(2):197–208.

anderwolf CH (2001) The hippocampus as an olfacto-motor mecha-nism: were the classical anatomists right after all?. Behav BrainRes 127(1–2):25–47.

arner MD, Peabody CA, Csernansky JG (1990) Olfactory functioningin schizophrenia and depression. Biol Psychiatry 27(4):457–458.

ittchen H, Pfister H (1997) DIA-X-interviews: manual für screening-verfahren und interview; interviewheft. Frankfurt: Swets & Zeitlinger.

olfensberger M, Schnieper I, Welge-Lussen A (2000) Sniffin’Sticks:a new olfactory test battery. Acta Otolaryngol 120:303–306.

ousem DM, Geckle RJ, Bilker WB (1996a) Post-traumatic olfactorydysfunction: MR and clinical evaluation. Am J Neuroradiol17:1171–1179.

ousem DM, Geckle RJ, Bilker W, McKeown DA, Doty RL (1996b) MRevaluation of patients with congenital hyposmia or anosmia. Am JRadiol 166:439–443.

ousem DM, Geckle RJ, Bilker WB, Doty RL (1998) Olfactory bulb andtract and temporal lobe volumes. Normative data across decades.Ann N Y Acad Sci 855:546–555.

ousem DM, Geckle RJ, Bilker WB, Kroger H, Doty RL (1999) Post-traumatic smell loss: relationship of psychophysical tests and vol-umes of the olfactory bulbs and tracts and the temporal lobes.

Acad Radiol 6:264–272.

(Accepted 5 May 2010)(Available online 13 May 2010)