Vol. 77 No.
`
`I
`
`January
`
`1994
`
`ORAL DIAGNOSIS
`
`Editor:
`
`Pete G. Fotos
`
`The Cytobrush Plus cell collector
`
`in oral cytology
`
`Anne Cale Jones, DDS,a Frank E. Pink, DDS, MS,b Pamela L. Sandow, DMD,C
`Carol M. Stewart, DDS, MS,d Cesar A. Migliorati, DDS, MS,a and
`Ronald A. Baughman, DDS, MSD,e Gainesville, Fla.
`UNIVERSITY
`OF FLORIDA
`COLLEGE
`OF DENTISTRY
`
`compared
`was
`(cytobrush)
`Plus cell collector
`The Cytobrush
`the convenience
`cytology.
`The degree
`of patient
`discomfort,
`exfoliative
`analysis
`of variance
`cells collected
`were
`evaluated.
`Two-factor
`epithelial
`in this study,
`we
`Because
`of
`the
`favorable
`findings
`revealed
`performed.
`when
`obtaining
`diagnostic
`cytologic
`smears
`from
`the oral mucosa.
`(ORAL
`
`oral
`during
`depressor
`tongue
`the wooden
`with
`the quantity
`and distribution
`to
`the clinician,
`and
`and parametric
`and nonparametric
`analyses
`were
`that
`recommend
`the cytobrush
`be considered
`for use
`SURC ORAL MED OUL
`PATHOL
`1994;77z1014)
`
`of
`
`the oral mucous
`Many pathologic processes affect
`membranes. Although
`several of these diseases are
`clinically distinctive,
`the majority of lesions require a
`definitive diagnosis before appropriate
`therapy can be
`instituted. The most accepted clinical
`technique
`for
`establishing
`the diagnosis of an oral mucosal disease
`is an incisional or excisional biopsy. However,
`in spe-
`cific clinical conditions,
`including primary herpetic
`gingivostomatitis,
`recurrent
`herpes
`simplex
`infec-
`tions, herpes zoster, and candidiasis, a presumptive
`diagnosis can be established
`readily with a cytologic
`smear.
`In addition, a cytologic smear may be benefi-
`cial in establishing
`the diagnosis of erythematous
`or
`ulcerative mucosal
`lesions that are suspected of being
`premalignant
`or malignant.
`In each of these situa-
`tions, a cytologic smear provides
`important
`informa-
`tion concerning
`the etiology of
`the lesions being
`examined.
`It facilitates prompt
`therapy without
`re-
`sorting
`to more time-consuming
`and expensive proce-
`dures, and it is useful
`in the selection of a represen-
`tative area to be biopsied
`if indicated.
`Any
`instrument
`that is chosen for oral exfoliative
`
`by a grant
`
`from Medscand
`
`(U.S.A.)
`
`Inc.
`
`was supported
`Study
`Hollywood,
`Fla.
`aAssistant
`Professor,
`bAssistant
`Professor,
`=Clinical
`Assistant
`ences.
`dAssociate
`eProfessor
`ences.
`Book,
`@ 1994 by Mosby-Year
`Copyright
`0030-4220/94/$1.00
`+ .lO
`7/17/49256
`
`Sciences.
`of Oral Diagnostic
`Department
`of Operative
`Dentistry.
`Department
`Professor,
`Department
`of Oral Diagnostic
`
`Sci-
`
`Sciences.
`of Oral Diagnostic
`Department
`Professor,
`and Chairman,
`Department
`of Oral
`Diagnostic
`Sci-
`
`Inc.
`
`cytology should satisfy several requirements. The in-
`strument should be easy to use in any intraoral
`loca-
`tion, it should not elicit significant patient discomfort,
`and it should collect an adequate number of epithelial
`cells that can be distributed
`evenly on a glass slide.
`The latter requirement
`is essential
`to facilitate an ac-
`curate cytopathologic
`diagnosis.
`A wooden
`tongue depressor, a metal spatula, and a
`cotton-tipped
`applicator are the most commonly used
`instruments
`for oral exfoliative
`cytology. However,
`these
`instruments
`are associated with
`some minor
`disadvantages. They may cause pain and tenderness
`when
`they are scraped or rolled over sensitive oral
`mucosa, and they are awkward
`to use in selected in-
`traoral
`locations such as the mandibular
`lingual gin-
`giva because of their
`inflexible and long handles.
`Furthermore,
`the quality and quantity of epithelial
`cells collected with
`these instruments
`can vary mark-
`edly. Epithelial cells collected with
`the wooden
`tongue
`depressor and metal spatula often exhibit significant
`nuclear and cytoplasmic distortion. This finding is re-
`lated to the fact that these instruments
`tend to cluster
`the cells in thick aggregates. On the other hand, few
`epithelial cells are collected with
`the cotton-tipped
`applicator, possibly because of its nonadhesive
`sur-
`face.
`the past several years, physicians have used
`During
`the Cytobrush Plus cell collector
`(cytobrush)
`(Med-
`stand U.S.A.
`Inc., Hollywood,
`Fla.) to obtain speci-
`mens for cervical cancer screening. The cytobrush has
`been shown
`to be a consistently effective and accurate
`instrument
`for collecting endocervical
`cells, either
`
`1
`
`

`

`102
`
`Jones et al.
`
`Table 1. Characteristics of patient population
`
`28 (56%)
`22 (44%)
`
`50.07
`51.14
`
`Gender
`Male
`Female
`Mean
`age
`Male
`Female
`evaluated
`Diseases
`Vesiculobullous
`or ulcerative
`Candidiasis
`Erythroplakia
`Squamous
`Normal
`
`cell carcinoma
`control
`
`of patients
`Number
`(20%)
`10
`(30%)
`15
`5 (10%)
`10
`(20%)
`10
`(20%)
`50
`(100%)
`
`when used alone or in combination with other cyto-
`logic
`instruments.
`‘-’ Because of the reliable and
`encouraging
`results obtained
`in
`these studies, an
`evaluation of the usefulness of the cytobrush
`for
`intraoral cytology was undertaken.
`The purpose of this study was to evaluate the degree
`of patient discomfort associated with the cytobrush
`and wooden tongue depressor, to determine
`the con-
`venience of these instruments
`to the clinician per-
`forming
`the cytologic smears, and to assess the quan-
`tity and distribution
`of epithelial
`cells collected dur-
`ing oral exfoliative cytology.
`
`MATERIAL AND METHODS
`The study population
`consisted of 40 patients
`referred to the oral medicine and oral oncology clin-
`ics at the University of Florida College of Dentistry
`and 10 normal control patients. The sex and mean age
`of all patients enrolled
`in this study are listed in Ta-
`ble
`I. Patients under
`the age of 18 years were
`excluded. Patients were classified into five groups-
`vesiculobullous
`or ulcerative,
`candidiasis, erythro-
`plakia, squamous cell carcinoma, normal control-on
`the basis of the clinical presentation and natural his-
`tory of each oral
`lesion and any response to prior
`therapy. The specific diseases evaluated and number
`of patients
`included
`in each category are listed in Ta-
`ble I. Patients with squamous cell carcinoma were
`enrolled
`in the study at the time of their referral
`to the
`Health Center Head and Neck Tumor Conference.
`According
`to their referring physicians, each of these
`patients had a previous diagnosis of an infiltrating
`epithelial neoplasm. Normal
`control patients did not
`display evidence of oral mucosal disease.
`
`Clinical procedure
`After appropriate consent and with the approval of
`the University of Florida Health Center Institutional
`Review Board, each patient was examined by one of
`four faculty members
`from
`the Department of Oral
`Diagnostic Sciences. Before the cytologic smears were
`performed, each patient’s medical history was re-
`
`ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
`January 1994
`
`examination was
`intraoral
`viewed and a thorough
`conducted. After
`these evaluations,
`two cytologic
`smears were obtained
`from
`the pathologic area in
`question using a cytobrush and a wooden tongue de-
`pressor (Fig.
`I). Both cytologic smears were per-
`formed in the same area. The wooden tongue depres-
`sor was chosen arbitrarily
`because of its historic and
`accepted use in oral cytology and the availability
`of
`this instrument
`in the College of Dentistry.
`The surface of the lesions were rolled or scraped
`uniformly,
`and
`the collected epithelial
`cells were
`smeared on an appropriately
`labeled glass slide and
`fixed with a 95% ethyl alcohol cytology
`fixative. The
`cytobrush was rolled on the glass slide by applying a
`continuous motion
`from one end of the slide to the
`other. The wooden tongue depressor was scraped in a
`continuous motion
`from one end of the slide to the
`other. To balance which cytologic
`instrument was
`used first during
`the cytologic smears, a permuted
`block randomization was followed. This procedure
`controlled
`for sequencing bias.
`Data concerning
`the degree of patient discomfort
`during each cytologic smear were collected by means
`of a 100 mm visual analogue scale. Zero mm was la-
`beled as no discomfort and 100 mm was labeled as
`extreme discomfort. After each cytologic smear, the
`patient was read a prepared statement
`intended
`to
`serve as a guideline
`for the patient
`in discriminating
`degrees of discomfort. The patient was then asked to
`mark
`the visual analog scale on the basis of the
`amount of discomfort
`felt during each cytologic
`smear.
`each cytologic
`after performing
`Immediately
`smear, each clinician was asked to rate the conve-
`nience of the cytobrush versus the wooden tongue de-
`pressor. This information was collected by marking a
`5-point categoric scale (1, easy to use; 5, difficult
`to
`use). The same clinician performed both cytologic
`smears for any given patient.
`In those situations
`in which the clinical diagnosis
`was equivocal, a 5 mm punch biopsy was performed
`after the cytologic smears. The specimens were placed
`in 10% formalin
`for subsequent routine hematoxylin
`and eosin staining. All cytologic smears and biopsy
`specimens were submitted
`to the oral pathology ser-
`vice at the University of Florida College of Dentistry.
`
`laboratory procedure
`All cytologic smears were stained with the Papani-
`colaou’s stain except for those cases of suspected can-
`didiasis
`that were stained with
`the periodic acid-
`Schiff stain. All cytologic smears were reviewed in an
`independent and blind
`fashion by two board certified
`oral pathologists
`(A.C.J. and R.A.B.). Each oral pa-
`thologist was unaware of the cytopathologic
`findings
`of the other, which disease category
`the cytologic
`
`2
`
`

`

`ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
`Volume 77, Number
`1
`
`Jones et al.
`
`103
`
`Table
`
`II. Degree of patient discomfort
`
`Table
`
`III. Nonparametric
`
`analyses
`
`Cytobrush
`
`Wooden tongue
`depressor
`
`Subject
`
`area
`
`Cytobrush
`
`Wooden tongue
`depressor
`
`Mild
`
`Mild
`
`Degree of patient
`discomfort (p = 0.97)*
`Lesion categoryt
`19.2 mm
`Vesiculobullous
`17.6 mm
`Candidiasis
`28.0 mm
`Erythroplakia
`30.7 mm
`Squamous cell carcinoma
`5.0 mm
`Normal control
`Method of cell collection: p = 0.212
`Interaction
`(lesion category X method of cell collection):
`p = 0.314
`
`29.6 mm
`15.1 mm
`17.4 mm
`25.6 mm
`8.3 mm
`
`*Parametric analysis (I test)
`on visual analogue scale
`t~N0v.A
`
`smears represented (except for cases of candidiasis),
`and what cytologic
`instrument was used to collect the
`epithelial
`cells. The following cytopathologic param-
`eters were evaluated:
`(1) quantity of epithelial
`cells
`collected (that
`is, adequate or inadequate number of
`cells for a cytopathologic
`diagnosis); and (2) distri-
`bution of epithelial
`cells collected
`(that
`is, cells
`arranged
`in thick aggregates or evenly dispersed on
`the slide).
`The biopsy specimens were processed routinely and
`stained with hematoxylin and eosin. In those cases in
`which both cytologic smears and a biopsy specimen
`were performed,
`the results of the cytologic smears
`were compared with the results of the biopsy specimen.
`
`RESULTS
`Evaluation of all 50 cases revealed no significant
`difference in the patient’s perception of pain between
`the cytobrush and the wooden tongue depressor dur-
`ing oral exfoliative cytology (t test, p = 0.972). Fur-
`ther analysis was performed with
`the use of a
`two-factor analysis of variance (ANOVA) with pain,
`as assessed on the visual analog scale as the dependent
`variable (Table
`II). Table II demonstrates a trend to-
`ward less pain in the normal control groups with ei-
`ther method of cell collection versus the other lesion
`categories; however, this finding was not statistically
`significant. No significant difference in pain was noted
`
`Convenience to clinician
`(Wilcoxon’s signed rank
`test, p = 0.001)
`Quantity of epithelial cells
`collected (sign test,
`p = 1.00)
`Distribution of epithelial cells
`collected (sign test,
`p = 0.000)
`Ability of evaluator to
`interpret collected epithelial
`cells (sign test, p = 0.97)
`
`Best
`
`Average
`
`Adequate
`
`Adequate
`
`Even
`
`Uneven
`
`Acceptable
`
`Acceptable
`
`(two-factor
`between the methods of cell collection
`interaction
`ANOVA, p = 0.212) and no significant
`was evident (two-factor ANOVA, p = 0.314).
`Table
`III
`lists
`the results of all nonparametric
`analyses performed. The cytobrush was determined
`to
`be significantly more convenient
`to the clinician
`than
`the wooden tongue depressor when performing oral
`cytologic
`smears
`(Wilcoxon’s
`signed
`rank
`test,
`p = 0.001). Evaluation of the quantity of epithelial
`cells collected (that
`is, adequate or inadequate num-
`ber of cells) revealed no significant difference between
`the cytobrush and wooden tongue depressor; both
`methods proved adequate (sign test, p = 1 .OO). How-
`ever, an evaluation of the distribution
`of epithelial
`cells collected (that
`is, cells arranged
`in thick aggre-
`gates or evenly dispersed) revealed a significant dif-
`ference between the cytobrush and wooden tongue
`depressor. The cytobrush provided more evenly dis-
`persed cells in a greater number of cases than
`the
`wooden tongue depressor (sign test, p = 0.000). No
`significant difference was noted in the ability of the
`two participating
`oral pathologists
`to render an accu-
`rate cytopathologic
`diagnosis on the basis of which
`cytologic
`technique was used (that
`is, the cytobrush
`or the wooden tongue depressor) (sign test, p = 0.97).
`An analysis of interexaminer
`reliability
`revealed no
`significant difference in opinions concerning either the
`quantity or distribution
`of epithelial
`cells collected
`
`3
`
`

`

`104
`
`Jones et al.
`
`instrument
`
`(sign
`
`test, p-value
`
`with each cytologic
`range 0.257 to 1.00).
`le-
`In four cases (one vesiculobullous or ulcerative
`sion and three erythroplakia
`lesions), the clinical di-
`agnosis was equivocal and a 5 mm punch biopsy was
`performed after the cytologic smears. In each of these
`cases, the cytologic
`smears demonstrated
`atypical
`squamous epithelial
`cells whereas the biopsy speci-
`mens displayed varying degrees of epithelial dyspla-
`sia. Although a punch biopsy was performed
`in only
`four of the 50 cases examined,
`the results of this di-
`agnostic technique correlated and supported
`the cy-
`topathologic
`findings noted on exfoliative cytology.
`
`DISCUSSION
`that the Cy-
`The results of this study demonstrate
`tobrush Plus cell collector is a practical, effective, and
`reliable
`instrument
`for collecting epithelial cells dur-
`ing intraoral
`cytology. When
`the degree of patient
`discomfort was evaluated, no significant difference
`was demonstrated
`between
`the cytobrush and
`the
`wooden tongue depressor. Both
`instruments elicited
`only mild patient discomfort. As expected, the least
`amount of discomfort was demonstrated
`in the nor-
`mal control group as compared with the four exper-
`imental groups.
`the cyto-
`that
`found
`each clinician
`Statistically,
`brush was more convenient
`to use in the oral cavity
`than the wooden tongue depressor. The cytobrush has
`a flexible and adjustable handle that allows a clinician
`to obtain epithelial
`cells from any intraoral
`location.
`The cytobrush can be shortened by snapping
`the pre-
`formed serrated notch on the handle. This feature al-
`lows a clinician
`to use this instrument
`in difficult
`to
`reach intraoral
`locations, such as the mandibular
`lin-
`gual gingiva.
`An accurate cytologic diagnosis was possible when
`specimens were collected with either the cytobrush or
`the wooden tongue depressor because an adequate
`number of cells was obtained with both instruments.
`However, the epithelial cells collected with the cyto-
`brush were dispersed in a thin, uniform
`layer unlike
`the epithelial
`cells obtained with the wooden tongue
`depressor that were arranged in thick clumps display-
`ing nuclear and cytoplasmic distortion. The cytologic
`uniformity
`seen in the epithelial
`cells collected with
`the cytobrush
`is an important
`advantage over the
`wooden tongue depressor because this uniformity
`fa-
`cilitates an accurate cytopathologic diagnosis.
`A cytologic smear is an advantageous diagnostic
`procedure because it is noninvasive, relatively painless,
`inexpensive, and requires a minimum of technical skill.
`It is useful in those situations when a patient refuses to
`have a biopsy performed or when medically compro-
`mised patients would be exposed to unnecessary surgi-
`cal risks. In addition, anxious patients can be reassured
`
`ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY
`Januarv
`I994
`
`quickly about the nature of oral mucosal changes, es-
`pecially when a fear of cancer or a family history of
`cancer accounts
`for their apprehension, With
`the
`availability
`of new diagnostic
`techniques using cell
`markers (such as, monoclonal antibodies and DNA
`in
`situ hybridization),
`cytologic
`smears are likely
`to
`become an increasingly
`important
`clinical
`technique
`for establishing
`the diagnosis of oral mucosal lesions.
`
`1.
`
`2.
`
`3.
`
`4.
`
`CONCLUSIONS
`We reached the following conclusions:
`The cytobrush was shown to be a more convenient
`cytologicinstrumenttotheclinicianthan
`thewooden
`tongue depressor during oral exfoliative cytology.
`The cytobrush provided a more even distribution of
`epithelial
`cells on a glass slide than the wooden
`tongue depressor.
`Both instruments were found to facilitate an accu-
`rate cytopathologic diagnosis.
`No significant difference existed in the minimal de-
`gree of patient discomfort between the cytobrush
`and the wooden tongue depressor during oral exfo-
`liative cytology.
`No significant difference was noted between the cy-
`tobrush and the wooden tongue depressor in the
`quantity of epithelial cells collected.
`Because of the favorable
`findings revealed in this
`study, we propose that the cytobrush be added to the
`list of diagnostic
`instruments currently available
`for
`use in oral exfoliative cytology.
`We thank Dr. Lewis Fannon and Mr. Wilson Jones for
`reviewing
`the manuscript
`and Ms. Patricia Baughman
`for
`histotechnical
`assistance.
`
`5.
`
`REFERENCES
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`1.
`
`of
`
`of
`J
`
`follow-up
`and
`KE, Losh DP, Folley A. Evaluation
`Miller
`1992;45:143-50.
`abnormal
`Pap smears. Am Fam Physician
`Mitchell
`H, Medley G. Reporting
`of an endocervical
`component
`after a previous
`cervical
`biopsy. Cytopathology
`1991;2:177-80.
`Hutchinson
`M, Fertitta
`L, Goldbaum
`3, Hamza M, Vanerian
`S, Isenstein
`L. Cervex-Brush
`and Cytobrush:
`comparison
`their
`ability
`to sample
`abnormal
`cells
`for cervical
`smears.
`Reprod Med
`1991;36:581-6.
`KE. Controlled
`Murata
`PJ, Johnson RA, McNicoh
`of implementing
`the Cytobrush
`technique
`to improve
`colaou
`smear
`quality.
`Obstet Gynecol
`1990;75:690-5.
`T,
`van Erp EJ, Dersjant-Roorda
`MC,
`Arentz
`NP, Stijnen
`Trimbos
`JB. Should
`the Cytobrush
`be used
`in routine
`screening
`for cervical
`natholoav?
`Int J Gvnaecol
`Obstet
`1989;30:139-44.
`Boon ME,
`IAlons-van
`Kordelaar
`JJ, Rietveld-Scheffers
`PE.
`Consequences
`of
`the
`introduction
`of combined
`spatula
`and
`Cytobrush
`sampling
`for cervical
`cytology:
`improvements
`in
`smear quality
`and detection
`rates. Acta Cytol
`1986;30:264-70.
`Trimbos
`JB, Arentz
`NP. The efficiency
`of the Cytobrush
`ver-
`sus the cotton
`swab
`in the collection
`of endocervical
`cells
`in
`cervical
`smears. Acta Cytol
`1986;30:261-3.
`
`evaluation
`Papani-
`
`requests:
`Reprint
`Anne Cale Jones, DDS
`University
`of Florida
`College
`Department
`of Oral Diagnostic
`Box 100414,
`JHMHC
`Gainesville,
`FL 32610
`
`of Dentistry
`Sciences
`
`4
`
`