ORI GI NALARTI CLERisk factors for development of chronic
rhinosinusitis in patients withallergic rhinitisAhmad R. Sedaghat,
MD, PhD1,2,3, Stacey T. Gray, MD1,3, Claus O. Wilke, PhD4andDavid
S. Caradonna, MD, DMD2,3Background:
Chronicrhinosinusitis(CRS)isaheteroge-neous inammatory condition of
the sinonasal cavity. CRSmay be preceded by other sinonasal
inammatory diseasesincluding allergic rhinitis (AR). It is unclear
what factors maypredispose patients with AR to develop CRS.Methods:
We performed a retrospective review of all pa-tients diagnosed with
AR (and not CRS) that presented toan otolaryngology clinic at a
tertiary care center as part ofa multidisciplinary allergy
evaluation between March 2004and November 2011. Medical records
were evaluated forclinicodemographic factors including age, gender,
smokinghistory, asthma, gastroesophageal reux disease
(GERD),aspirin sensitivity, nasal polyposis, seasonal AR,
perennialAR, categories of positive antigens on formal allergy
test-ing, and the following sinonasal anatomic variants on
com-putedtomography(CT): infraorbital (Haller) cells, con-cha
bullosa, frontal intersinus cells, and anterior ethmoidfrontal
recess cells. Patients who did not develop CRS af-ter at least 4
years of follow-up were grouped into the ARcohort. Patients
whodevelopedCRSaer at least 6 monthsof follow-up were grouped into
the AR-CRS cohort.Results: We found a statistically signicant
association be-tween the presence of infraorbital (Haller) cells
(odds ratio[OR] = 6.27) and frontal intersinus cells (OR = 18.37)
withdevelopment of CRS on both univariate and multivariate
lo-gistical regressions.Conclusion: Sinonasal anatomical variants,
specically in-fraorbital and frontal intersinus cells, are
associated withdevelopment of CRS in patients with AR. The presence
ofthese variants identies patients who should be counseledon
compliance with medical therapy for AR to potentiallyprevent
progression to CRS.C 2012 ARS-AAOA, LLC.Key Words:allergic
rhinitis; chronic rhinosinusitis; computed tomogra-phy;
infraorbital cell; Haller cell; frontal intersinus cellHow to Cite
this Article:Sedaghat AR, Gray ST, Wilke CO, Caradonna DS. Risk
fac-tors for development of chronic rhinosinusitis in
patientswithallergic rhinitis. Int ForumAllergyRhinol, 2012;
2:370375.Chronic rhinosinusitis (CRS) is characterized
byheterogeneous, chronic inammatory processes ofthe sinonasal
mucosa1leading to symptoms of facialpain/pressure, nasal
congestion, and/or purulent
drainage.2Withover10milliondiagnosedindividualsandanesti-1Department
of OtolaryngologyHead and Neck Surgery,Massachusetts Eye and Ear
Inrmary, Boston, MA;2Divison ofOtolaryngology, Beth Israel
Deaconess Medical Center, Boston, MA;3Department of Otology and
Laryngology, Harvard Medical School,Boston, MA;4Section of
Integrative Biology, Center for ComputationalBiology and
Bioinformatics, and Institute for Cell and MolecularBiology, The
University of Texas at Austin, Austin, TXCorrespondence to: Ahmad
R. Sedaghat, MD, PhD, Department ofOtolaryngologyHead and Neck
Surgery, Massachusetts Eye and EarInrmary, 243 Charles St., Boston,
MA 02114; e-mail:ahmad [email protected] conict of
interest: None provided.Received: 4 March 2012; Revised: 3 April
2012; Accepted: 17 April 2012DOI: 10.1002/alr.21055View this
article online at wileyonlinelibrary.com.mated $8.6 billion spent
annually in total health care ex-penditures in the United States
alone, CRS is a major burdenon the healthcare system.3Currently,
the initial treatmentfor uncomplicated CRS is conservative medical
therapy, in-cluding antibiotics and corticosteroids.4Surgical
interven-tion with endoscopic sinus surgery is considered5if
appro-priate medical therapies fail. Treatment of CRS is
targetedtoward long-term symptom control rather than cure, as
theunderlying inammatory causes of CRS in most patients arenever
completely eradicated, and patients often remain onlong-term
maintenance therapy.4Despite such poor
prog-nosisforlong-termdisease-freesurvival,
nopreventativestrategies exist in clinical practice because given
the currentdata on CRS risk factors there is limited scope for
identify-ing individuals at risk for developing the
disease.Allergic rhinitis (AR) has been reported to be
associatedwithCRS.6Althoughevidencefor acausal relationshipbetween
AR and CRS is lacking, there are many lines of ev-idence through
epidemiologic association79and commoncellular and molecular
mediators1which suggest that ARInternational Forum of Allergy &
Rhinology, Vol. 2, No. 5, September/October 2012 370CRS risk
factors in allergic rhinitisis an inammatory instigator or an
exacerbating factor forCRS.We hypothesize that AR, while not
necessarily a causativeagent of CRS, produces a chronic, inammatory
milieu inthe sinonasal mucosa, which in the right setting may
pre-disposepatientstodevelopCRS. Weassessed2cohortsof
patientsdiagnosedwithARwhounderwent
amulti-disciplinaryevaluationandlong-termfollow-up.Oneco-hort
developed CRS during the course of follow-up and theother cohortdid
not developCRS. We hypothesizedthatdirect comparison of medical
comorbidities and sinonasalanatomy in these 2 cohorts may identify
clinical risk fac-tors in patients with AR that may predispose to
subsequentdevelopment of CRS.Patients and methodsPatient
selectionApproval for this study was obtained from the Beth
IsraelDeaconessMedicalCenterInstitutionalReviewBoard.Aconsecutiveseriesofadultpatientshavingonlytheclin-icallydetermineddiagnosisof
ARat initial presentationbetween March 2004 and November 2011 was
screened.ThediagnosisofARwasdependentonsymptomatologyand positive
skin or radioallergosorbent test (RAST) testingby an allergist.
Endoscopic examination of the nasal cavitywas routinely performed
on patients at the time of the ini-tial presentation and diagnosis
of AR. A diagnosis of CRSwas made based on guideline-established
criteria2,10set bythe American Academy of OtolaryngologyHead and
NeckSurgery that are based on patient-reported history and
ob-jective ndings on computed tomography (CT) scan or en-doscopic
examination. All patients followed for 4 or moreyears without
meeting guideline-established criteria for di-agnosis of CRS
comprise the AR cohort. All patients whomet criteria for diagnosis
of CRS after the rst 6 months offollow-up comprise the ARto CRS
(AR-CRS) cohort. Clini-codemographic data collected consisted of
information thatis commonly collected as a part of a routine AR
workup.Demographic data collected included age, gender, and
cur-rent tobacco use. Clinical data recorded included history
ofmedical comorbiditiesassociatedwithCRS: asthma,11,12aspirin
sensitivity, nasal polyposis at the time of presenta-tion (reected
by the Lund nasal polyp score),13history ofrecurrent acute
rhinosinusitis, and gastroesophageal reuxdisease
(GERD).14,15Diagnoses of GERD, asthma, and as-pirin sensitivity
were recorded as positive based on patienthistory, and conrmed with
supporting documentation bythepatientsinternal
medicinephysicianorallergist, in-cluding prescription of
appropriate medication. A patientwas considered to have presented
with recurrent acute rhi-nosinusitis if he or she had greater than
4 distinct episodesof acute sinusitis per year for at least 1 year
preceding theinitial
presentation.2,10Anallergicprolewasdeterminedasfollows. Historyof
seasonal and/or perennial allergies was recorded. More-over,
positive allergy testing for antigens categorized as: (1)a pollen;
(2) dust mite; (3) cat; (4) dog; (5) fungus or
mold;or(6)cockroachwerenotedaswell.Patientswithprevi-oushistoryofCRS,sinussurgery,orcysticbrosiswereexcluded.Image
analysisHeador sinus CT scans, whenavailable, were
evaluated,blinded to which cohort the patient belonged, for the
pres-enceofinfraorbital ethmoidcells(alsoknownasHallercells) and
concha bullosa (pneumatized middle turbinate),both of which may
serve as obstructions of the osteomeatalunit (OMU)
wherethemaxillary, anteriorethmoid, andfrontal sinuses
drain.16,17We evaluated obstruction of
thefrontalrecessbyfrontalintersinuscellsandanterioreth-moidcells.
Afrontal intersinus cell is adistinct air cellwithin the intersinus
septum of the frontal sinuses and hasan outow tract into either the
left or right frontal recess.Frontal intersinus cells may obstruct
frontal sinus outowby impinging on the frontal recess. Anterior
ethmoid cellsinthefrontal
recessareclassiedaccordingtotheKuhnclassication,18in which higher
classes indicate greater en-croachment into the frontal recess or
frontal sinus. For ouranalyses, we considered the highest Kuhn
class from eitherthe patients left or right side. Head or sinus CT
scans wereavailable in the medical record for 26 of 35 patients in
theAR cohort, obtained in order to rule out CRS at the time
ofpresentation in patients. All patients in the AR-CRS
cohorthadsinusCTscansperformedatthetimethatCRSwasdiagnosed.Ofthe24patientsintheAR-CRScohort,23had
available sinus CT scans in the medical record.Statistical
analysisAll analysis was performed with the R statistical
software(www.r-project.org). Associations between development ofCRS
and predictor variables were determined with the lo-gistic
regression function lrm() in the rms package.19Uni-variate logistic
regression was performed for each predic-tor variable. Multivariate
logistical analysis was performedusing all clinicodemographic and
anatomic variables. Mul-tivariatelogistical
analysiswasperformedseparatelyforpredictor variables of the allergy
prole due to sparsity ofdata availabilityin the medicalrecord
comparedto clini-codemographicandanatomiccharacteristics.
Inthemul-tivariatemodel, signicant
predictorswereidentiedviabackward elimination, using a p value
cutoff of 0.100. Foreach variable retained in the nal model, a p
value and alogodds ratio (OR) were
calculated.ResultsCharacteristics of study subjectsWe found 35
patients who met criteria for the AR cohort(Table1). Thiscohort
was40%maleand60%femalewith mean standard deviation (SD) age of 46.4
12.0years at the initial presentation. We found 24 patients who371
International Forum of Allergy & Rhinology, Vol. 2, No. 5,
September/October 2012Sedaghat et al.met criteriafortheAR-CRScohort
(Table1). TheAR-CRS cohort was 41.7% male and 58.3% female with
meanage of 45.7 12.0 years at initial presentation. The me-dian
follow-up time for patients in the AR cohort was 5.3(range,
4.19.3)years.
ThemediantimetodevelopmentofCRSwas3.4yearswith75%diagnosedby4.1years.We
found 5.7% of patients in the AR cohort reported cur-rent smoking
at the time of presentation in comparison to8.3% of patients in the
AR-CRS. There was no signicantdifference in the average age at
presentation, gender, or fre-quency of currently smoking patients
between the 2 cohorts(Table 1).Contribution of medical
comorbiditiesWe evaluated the association of comorbidities
GERD,asthma,aspirinsensitivity,andnasalpolyposis(Table1)with
development of CRS. In the AR cohort, 40.0% of
pa-tientshadGERDand17.1%presentedwithahistoryofrecurrent sinusitis.
In comparison, 29.1% of patients hadGERD and 12.5% had recurrent
sinusitis in the AR-CRScohort. There was no signicant association
between a his-tory of GERD or recurrent sinusitis with the
developmentof CRS (Table 1).We found that 34.3% of patients in the
AR cohort hadasthma and 41.7% of patients in the AR-CRS cohort
hadasthma. The presence of nasal polyps was not common ineither
cohort, occurring in 8.7% of patients in the AR co-hort (average
Lund nasal polyp score of 1.7) and 8.3% inthe AR-CRS cohort
(average polyp score of 1.5). Althoughuncommon, nasal polyposis may
be present in the absenceofCRS(ie, nasal
polyposisdoesnotobligatelyleadtoadiagnosis of CRS according to
published guidelines2,10). Inthe AR cohort, 1 patient was described
as having a singlesmall polyp arising from the middle meatus; the
other pa-tient had more extensive polyposis in the setting of
Samterstriad, although no air-uid level was seen on CT scan
whichwas consistent with the lack of purulence seen on nasal
en-doscopy performed in the clinic. Neither asthma nor
nasalpolyposis signicantly associated with progression to
CRS(Table1). Wealsoconsideredtheassociationof aspirinsensitivity,
because Samters triad may be associated withCRS. We found that 2.8%
of patients in the AR cohort
hadaspirinsensitivityand16.7%ofpatientsintheAR-CRScohort had
aspirin sensitivity. There was a trend toward
asignicantassociationbetweenaspirinsensitivityandde-velopment of
CRS on univariate regression (p = 0.096) butnot on multivariate (p
= 0.858) regression.We furthermore considered atopic
characteristics as po-tential risk factors for development of CRS
(Table 2). Wefound that 74.2% of AR patients had seasonal AR in
com-parisonto78.9%intheAR-CRScohort. Perennial ARwas present in
96.4% of the patients in the AR cohort and89.5% of patients in the
AR-CRS cohort. Neither seasonalAR nor perennial AR was associated
with development ofCRS (Table 2). Similarly, we did not nd any
statisticallysignicant association between positive allergy testing
forany antigens in the specied categories and development ofCRS
(Table 2).Contribution of anatomical factorsWe considered the role
of obstructive sinonasal anatomicvariants in the development of CRS
(Table 3). In the OMU,we found that 19.2% of AR patients had a
concha bullosain comparison to 41.7% of AR-CRS patients.
Associationbetween the presence of concha bullosa and developmentof
CRS trended toward statistical signicance on univariateregression
(p = 0.090) but not on multivariate (p = 0.988)regression. We did,
however, nd that 100% of patients inTABLE 1. Clinicodemographic
dataAR patients, AR-CRS patients, Univariate Univariate
Multivariate Multivariatemean (SD) mean (SD) p value OR (95% CI) p
value OR (95% CI)Agea(years) 46.4 (12.0) 45.7 (12.0) 0.837 1.00
(0.951.04) 0.748 1.01 (0.941.09)Gender (%)M 40.0 (8.3) 41.7 (10.0)
0.898 1.07 (0.373.08) 0.640 1.49 (0.288.05)F 60.0 (8.3) 58.3
(10.0)Tobacco usage (%) 5.7 (3.9) 8.3 (5.6) 0.696 1.50 (0.2011.45)
0.517 0.36 (0.028.12)GERD (%) 40.0 (8.3) 29.1 (9.3) 0.395 0.62
(0.201.87) 0.466 0.56 (0.122.63)Recurrent sinusitis (%) 17.1 (6.4)
12.5 (6.7) 0.627 0.69 (0.153.08) 0.376 0.40 (0.053.00)Asthma (%)
34.3 (8.0) 41.7 (10.1) 0.565 1.37 (0.473.99) 0.348 2.13
(0.4410.33)ASA sensitivity (%) 2.8 (2.8) 16.7 (7.6) 0.096 6.80
(0.7165.16) 0.858 1.03 (0.0331.29)Nasal polyps 0.894 0.93
(0.332.66) 0.473 0.58 (0.162.07)Prevalence (%) 8.7 (4.7) 8.3
(5.6)Score 1.7 1.5aAge at initial presentation and diagnosis of
AR.AR = allergic rhinitis; ASA = aspirin; CI = condence interval;
CRS = chronic rhinosinusitis; F = female; GERD = gastrointestinal
reux disease; M = male; OR = oddsratio; SD = standard
deviation.International Forum of Allergy & Rhinology, Vol. 2,
No. 5, September/October 2012 372CRS risk factors in allergic
rhinitisTABLE 2. AR proleAR Patients, AR-CRS patients, Univariate
Univariate Multivariate Multivariatemean (SD) mean (SD) p value OR
(95% CI) p value OR (95% CI)Seasonal AR (%) 74.2 (7.9) 78.9 (9.4)
0.702 1.30 (0.335.11) 0.965 0.92 (0.0331.01)Perennial AR (%) 96.4
(3.5) 89.5 (7.0) 0.360 0.31 (0.033.74) 0.479 0.33
(0.027.09)Allergen prole (%)Pollens 71.4 (8.5) 80.0 (10.3) 0.541
1.60 (0.357.23) 0.758 0.74 (0.105.16)Dust mite 74.1 (8.4) 68.8
(11.6) 0.707 0.77 (0.203.01) 0.833 1.20 (0.216.83)Cat 55.6 (9.6)
40.0 (12.6) 0.337 0.75 (0.212.70) 0.119 0.17 (0.013.28)Dog 25.9
(8.4) 28.6 (12.1) 0.856 1.14 (0.274.84) 0.271 4.57
(0.4051.14)Fungus/mold 25.9 (8.4) 35.7 (12.8) 0.515 1.59 (0.396.38)
0.418 1.84 (0.428.11)Cockroach 11.1 (6.0) 21.4 (11.0) 0.383 2.18
(0.3812.58) 0.2306 2.75 (0.4019.02)AR = allergic rhinitis; CI =
condence interval; CRS = chronic rhinosinusitis; OR = odds ratio;
SD = standard deviation.TABLE 3. Obstructive sinonasal anatomic
variantsAR patients, AR-CRS patients, Univariate Univariate
Multivariate Multivariatemean (SD) mean (SD) p value OR (95% CI) p
value OR (95% CI)Infraorbital (Haller) cells (%) 19.2 (7.7) 65.2
(9.9) 0.002 7.87 (2.1428.87) 0.013 6.27 (1.4626.87)Concha bullosa
(%) 19.2 (7.7) 41.7 (10.1) 0.090 3.00 (0.8410.67) 0.988 0.98
(0.0910.38)Frontal intersinus cell (%) 4.0 (3.9) 47.8 (10.4) 0.005
22.00 (2.53191.00) 0.011 18.37 (1.93175.04)Kuhn cell class (%)
0.010 2.08 (1.193.64) 0.424 1.37 (0.632.95)0 36.0 30.41 44.0 4.32
16.0 21.73 4.0 43.54 0.0 0.0AR = allergic rhinitis; CI = condence
interval; CRS = chronic rhinosinusitis; OR = odds ratio; SD =
standard deviation.the AR-CRS cohort with concha bullosa had
concomitantmucosal thickening in the ipsilateral maxillary or
anteriorethmoid sinuses on CT scan at the time of CRS diagnosis.We
found that 19.2% of AR patients had at least 1 infraor-bital cell
incomparisonto65.2%of AR-CRSpatients,reecting a signicant
association of infraorbital cells withprogression to CRS on
univariate (p = 0.002) and multi-variate (p =0.013) regressions,
reecting an ORof 6.27 fordevelopment of CRS in AR patients who have
an
infraor-bitalcell.OfpatientsintheAR-CRScohortwhohadaninfraorbital
cell, 81% had concomitant mucosal thickeningin the ipsilateral
maxillary sinus on CT scan at the time ofCRS
diagnosis.Inthefrontalrecess,theaveragehighestKuhnclassofanterior
ethmoid cells from either the right or left side
was0.88intheARcohortand1.78intheAR-CRScohort.TherewasasignicantassociationbetweenhigherKuhnclass
anddevelopment of CRSonunivariate regression(p =0.010), but not on
multivariate regression (p =0.424).We found frontal intersinus
cells in 4.0% of the AR cohortand 47.8% of the AR-CRS cohort,
reecting a statisticallysignicant association with development of
CRS on univari-ate (p = 0.005) and multivariate (p = 0.011)
regressions.Although presence of a frontal intersinus cell
correspondedto an OR = 18.37 for development of CRS, only 50%
ofpatients in the AR-CRS cohort with a frontal intersinus
celldemonstrated mucosal thickening in the frontal sinuses onCT
scan at the time of CRS
diagnosis.Wethereforeconsideredwhetherinfraorbital cellsandfrontal
intersinus cells were correlated, leading to an arti-factual
association of CRS to frontal intersinus cells. Thepresence of
infraorbital cells seemed to correlate with thepresence of frontal
intersinus cells (OR = 3.9, p = 0.088,Fishers exact test). Thus it
is possible that some correlationwith the presence of infraorbital
cells may explain the ap-parent signicance for frontal intersinus
cells in predictingprogression to
CRS.DiscussionTreatmentforCRSiscenteredaroundsymptomcontrolratherthancompletecure.Preventionhasnotbeenafo-cus
of management because factors to predict those at
riskpriortodiagnosisofCRSarenotwell established. CRS373
International Forum of Allergy & Rhinology, Vol. 2, No. 5,
September/October 2012Sedaghat et
al.ishypothesizedtoarisefromanynumberofinstigating,inammatoryprocessesthatprogresstothesameclinicalendpoint.
Previous work has established an association
be-tweenARandCRS.4,6,7WhilethelinkbetweenARandCRS may not be
causal, we hypothesize that AR may pro-duce an inammatory milieu in
the sinonasal mucosa that,in the presence of certain risk factors,
may lead to develop-ment of CRS.In this study, we identify
objective risk factors that maybeusedinclinical
practicetoidentifypatientswithARwhoareat riskfordevelopment of CRS.
Althoughourstudyisretrospectiveinnature,thevastmajorityofclin-icodemographicfactorswesoughtwerereadilyavailableasacommonpartoftheallergyworkup.Weconsideredmedical
comorbidities often associated with CRS6,11,12,15and found no
strong associations with progression of ARto CRS. This may be due
to the fact that conditionssuch as asthma or aspirin sensitivity,
while associated withCRS, are related to an underlying atopic state
that is alsocommonin those with AR. Alternatively,the
presenceofsomecomorbidconditionssuchasnasalpolyposisoras-pirinsensitivitymayreect
distinct inammatorycondi-tions that contributetothedevelopment of
CRSinthesettingofAR,whichourstudymaybeunderpoweredtodetect.Incontrast,ourassessmentofsinonasalanatomicvari-ants
revealed signicant associations between progressiontoCRSandthe
presence of either infraorbital cells orfrontal intersinus cells,
which are associated with obstruc-tionof
outowfromthemaxillaryandfrontal sinuses,respectively.16,17In the
past, the presence of these anatomicvariants has not been found to
be associated with patientswho have CRS.16,20,21However, these
studies have not fo-cused on specic subsets of patients with CRS.
In contrast,we have examined a specic subset of CRS patients,
namelythose with AR that preceded CRS. In this particular subsetof
patients, the presence of obstructive anatomic variants inthe
setting of allergic sinonasal inammation may be con-tributory to
the development of CRS. It is similarly possiblethat the study of
other specic CRS patient subsets may re-veal a signicant
association between the presence of theseanatomic variants and the
development of CRS.Moreover, inadditiontoinfraorbital
cellsandfrontalintersinus cells, weadditionallyobservedtrends
towardsignicantassociationforotherpotentiallysinonasal ob-structive
variants. The presence of these anatomic variantslargely correlated
to mucosal thickening in the correspond-ing sinuses on CTscans
obtained at the time of CRS diagno-sis. In the case of frontal
intersinus cells, which correlatedwith frontal sinus mucosal
thickening only 50%of the time,we hypothesize that because CT scans
capture a picture ofCRS at only 1 time point, any role in
development of CRSmay have occurred prior to imaging.
Alternatively, we
ndthatanunknowncorrelationbetweenthepresenceofin-fraorbitalcellsandfrontalintersinuscellsmayhaveleadto
a noncausal, statistically signicant association of CRSdevelopment
and the frontal intersinus cells.Patients with AR are frequently
treated solely by a
pri-marycarephysician,anditisimportanttoconsiderwhythesepatientsmayhavesought
amultidisciplinaryeval-uationbyanallergistandotolaryngologist,
becauseitispossible that these patients had particularly severe or
con-stant sinonasal symptoms. This is supported by the fact
thatmany of these patients had previous sinus CT scans, whichare
not routinely performed in the workup of simple AR.Moreover,
because not all patients in our AR cohort had aCT scanned
performed, it is likely that those patients in theAR cohort who did
have CT scans performed were patientswith more severe sinonasal
symptoms and the diagnosis ofCRS was being ruled out with imaging
studies at the timeof presentation. A potential skew in our study
populationtoward patients with more severe sinonasal symptoms
mayarticially inate the prevalence of comorbid conditions
orcharacteristics(eithermedicaloranatomical),whichmaybe illustrated
by the high prevalence of both seasonal andperennial
AR.Overall,thiseffectwouldmakeidentica-tion of CRS risk factors
more difcult. Detection of CRSrisk factors is also made difcult
because patients in our ARcohort, which we consider to represent AR
patients who donot progress to CRS, may nonetheless progress to CRS
at alater point in time. Although we would expect the effect ofsuch
confounding issues to be minimized with larger cohortsizes, given
these challenges, the risk factors that we havebeen able to
identify appear all the more signicant. Withlarger cohorts of
patients with AR and even longer follow-up times it would not be
surprising to nd other anatomicor clinical risk factors for
development of CRS.This is the rst study to evaluate and identify
objectiverisk factors for development of CRS in patients with
ARandour ndings have important implications. We have identi-ed
putatively obstructive sinonasal anatomical variants aspotential
CRS risk factors that are easily assessed in
clinicalpracticeonCTscan,whichmaybeobtainedforARpa-tients who have
particularly severe sinonasal symptoms orpoor response to medical
therapy and radiologic studies areobtained to investigate for the
possibility of chronic rhinos-inusitis. Our ndings provide
motivation for and warrantperformance of further prospective study.
For now, how-ever, we suggest that identication of any of these
potentialCRS risk factors in a patient with AR may, in the
contextof the clinical presentation: (1) inform the decision to
pur-sue more aggressive medical treatment of AR as a
rationalattempt to prevent progression to CRS; and (2) serve as
adiscussion point with patients in the course of
counselingstringent compliance with their
therapy.ConclusionAR,intherightsetting,mayleadtothedevelopmentofCRS.
We have identied 2 potentially obstructive sinonasalanatomic
variants as characteristics
associatedwithpa-tientswhoinitiallypresentedwithARandsubsequentlydeveloped
CRS. Easily assessed with imaging, these variantsInternational
Forum of Allergy & Rhinology, Vol. 2, No. 5, September/October
2012 374CRS risk factors in allergic rhinitismay identify AR
patients at increased risk for developmentof CRS and thus may need
more aggressive management oftheir AR. The ability to inform
patients of this risk mightincrease their adherence to treatment
regimens or inuencethe decision to pursue other treatment options,
such as al-lergen immunotherapy.References1. Van Crombruggen K,
Zhang N, Gevaert P, TomassenP, Bachert C. Pathogenesis of chronic
rhinos-inusitis: inammation. J Allergy Clin
Immunol.2011;128:728732.2. Rosenfeld RM, Andes D, Bhattacharyya N,
et al. Clin-ical practice guideline: adult sinusitis.
OtolaryngolHead Neck Surg. 2007;137(3 Suppl):S1S31.3. Bhattacharyya
N. Incremental health care utiliza-tion and expenditures for
chronic rhinosinusitisin the United States. Ann Otol Rhinol
Laryngol.2011;120:423427.4. Hamilos DL. Chronic rhinosinusitis:
epidemiologyandmedical management.
JAllergyClinImmunol.2011;128:693707; quiz 708709.5.
LuongA,MarpleBF.Sinussurgery:indicationsandtechniques. Clin Rev
Allergy Immunol. 2006;30:217222.6.
FokkensW,LundV,BachertC,etal.EAACIposi-tion paper on rhinosinusitis
and nasal polyps execu-tive summary. Allergy. 2005;60:583601.7.
Houser SM, Keen KJ. The role of allergy and smokingin chronic
rhinosinusitis and polyposis. Laryngoscope.2008;118:15211527.8.
Emanuel IA, Shah SB. Chronic rhinosinusitis: al-lergy and sinus
computed tomography relation-ships. Otolaryngol Head Neck Surg.
2000;123:687691.9. Schoenwetter WF, Dupclay L Jr, Appajosyula S,
Bot-teman MF, Pashos CL. Economic impact and quality-of-life burden
of allergic rhinitis. Curr Med Res Opin.2004;20:305317.10.
Benninger MS, Ferguson BJ, Hadley JA, et al. Adultchronic
rhinosinusitis: Denitions, diagnosis, epidemi-ology, and
pathophysiology. Otolaryngol Head NeckSurg. 2003;129(3
Suppl):S1S32.11. Ponikau JU, Sherris DA, Kephart GM, et al.
Featuresof airway remodeling and eosinophilic inammationin chronic
rhinosinusitis: is the histopathology
similartoasthma?JAllergyClinImmunol.2003;112:877882.12. Cruz AA,
Popov T, Pawankar R, et al.
Commoncharacteristicsofupperandlowerairwaysinrhini-tisandasthma:ARIAupdate,incollaborationwithGA(2)LEN.
Allergy. 2007;62(Suppl 84):141.13. LundVJ, FloodJ, Sykes AP,
Richards DH. Effectofuticasoneinseverepolyposis.ArchOtolaryngolHead
Neck Surg. 1998;124:513518.14.
FlookEP,KumarBN.Isthereevidencetolinkacidreux with chronic
sinusitis or any nasal symptoms?A review of the evidence.
Rhinology. 2011;49:1116.15. Wong IW, Rees G, Greiff L, Myers JC,
JamiesonGG, Wormald PJ. Gastroesophageal reux disease andchronic
sinusitis: In search of an esophageal-nasal re-ex. Am J Rhinol
Allergy. 2010;24:255259.16.
BolgerWE,ButzinCA,ParsonsDS.Paranasalsinusbony anatomic variations
and mucosal abnormalities:CTanalysisforendoscopicsinussurgery.
Laryngo-scope. 1991;101(1 Pt 1):5664.17. Stackpole SA, Edelstein
DR. The anatomic rele-vance of the Haller cell insinusitis. AmJ
Rhinol.1997;11:219223.18. Bent JP, Cuilty-SillerC, KuhnFA.
Thefrontal cellas a cause of frontal sinus obstruction. Am J
Rhinol.1994;8:185191.19. Harrell FE. Regression modeling
strategies: With ap-plications tolinear models, logisticregression,
andsurvival analysis. New York: Springer; 2001:568.20. Tonai A,
Baba S. Anatomic variations of the bone insinonasal CT. Acta
Otolaryngol Suppl. 1996;525:913.21. Nouraei SA, Elisay AR,
DimarcoA, et al. Varia-tions in paranasal sinus anatomy:
implications for
thepathophysiologyofchronicrhinosinusitisandsafetyof endoscopic
sinus surgery. J Otolaryngol Head NeckSurg. 2009;38:3237.375
International Forum of Allergy & Rhinology, Vol. 2, No. 5,
September/October 2012
