Gynecologic Cytopathology

Effects of Gel Lubricant on Cervical Cytology

Kittipat Charoenkwan, M.D., Kanchana Nimmanahaeminda, M.S.,

Surapan Khunamornpong, M.D., Jatupol Srisomboon, M.D., and
Paul S. Thorner, M.D., Ph.D.

Objective cytologic evaluation, the discordance in cytologic diagnosis

To determine whether the use of lubricating gel during between the gel-contaminated and uncontaminated smears
vaginal speculum examination affected cytologic interpreta- from the same patient was 0.3%.
tion in the conventional Pap smear.
Conclusion
Study Design Lubricating gel contamination of the cervix can adversely
Two consecutive cervical affect adequacy and cytologic
smears were obtained from diagnosis in the conventional
1,334 patients undergoing Use of more than the usual Pap smear. (Acta Cytol 2008;
Pap smear screening. The first amount of gel lubricant during the 52:654–658)
smear (uncontaminated) was
obtained using the routine col- pelvic examination can interfere Keywords: cancer, cervix;
lection technique. The second with the adequacy of cervical cytology; Papanicolaou smear;
smear (gel-contaminated) was speculum; vaginal gel.
taken after applying a 1- to
cytology smears through a
reduction in cellularity….
1.5-cm ribbon of lubricating
gel onto the external cervical
os. Adequacy of Pap smear and
C
ervical cancer contin-
ues to be one of the
most common cancers in
discordance in diagnosis between the paired smears were ex- women and a leading cause of cancer death worldwide,
amined. despite the availability of an effective cervical cytolog-
ic screening test, the Pap smear. Early diagnosis and
Results management of precancerous cervical lesions in
The proportion of unsatisfactory smears was significantly women who have regular Pap tests has resulted in a
higher in the gel-contaminated smears, 12.1% vs. 1.7% decrease in incidence and mortality of cervical can-
(p < 0.01). This difference was consistent across all repro- cer.1 A historical review of women diagnosed with in-
ductive groups. For patients who had smears satisfactory for vasive cervical cancer demonstrated that 29% had

From the Departments of Obstetrics and Gynecology and Pathology, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand; and
Department of Laboratory Medicine, Hospital for Sick Children and University of Toronto, Toronto, Canada.
Dr. Charoenkwan is Assistant Professor, Department of Obstetrics and Gynecology, Faculty of Medicine, Chiang Mai University.
Ms. Nimmanahaeminda is Assistant Professor, Department of Pathology, Faculty of Medicine, Chiang Mai University.
Dr. Khunamornpong is Associate Professor, Department of Pathology, Faculty of Medicine, Chiang Mai University.
Dr. Srisomboon is Professor, Department of Obstetrics and Gynecology, Faculty of Medicine, Chiang Mai University.
Dr. Thorner is Professor, Department of Laboratory Medicine, Hospital for Sick Children and University of Toronto.
Supported by the Thailand Research Fund (grant no. MRG4880074).
Presented at the 33rd European Congress of Cytology, Madrid, Spain, October 14–17, 2007.
Address correspondence to: Kittipat Charoenkwan, M.D., Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Fac-
132.239.1.231 - 1/8/2017 3:18:36 PM

ulty of Medicine, Chiang Mai University, Chiang Mai 50200, Thailand (kicharoe@mail.med.cmu.ac.th).
Univ. of California San Diego

Financial Disclosure: The authors have no connection to any companies or products mentioned in this article.
Received for publication April 24, 2007.
Downloaded by:

Accepted for publication May 4, 2007.

654 ACTA CYTOLOGICA 0001-5547/08/5206-0654/$21.00/0 © The International Academy of Cytology


never had a screening test and 33% had not had a Pap
smear in the 5 years before diagnosis.2 Pain and dis-
comfort with pelvic examination appeared to be the
most common barrier to compliance with Pap smear
follow-up in a study of adolescents.3 Lubricating the
speculum with a water-based gel can reduce pain and
discomfort during speculum insertion and pelvic ex-
amination, theoretically improving Pap smear compli-
ance. However, this practice has not been encouraged
because of the concern about possible interference
with cytologic interpretation of the Pap smear.4 This
concept has become a clinical dogma in most gynecol-
ogy textbooks.
Comparative studies have shown no difference in
Pap smear adequacy and degree of disagreement on
Pap smear diagnosis between the smears taken with
and without a small amount of gel lubricant on the
vaginal speculum.4-7 However, it has also been shown
that lubrication of the external surfaces of the specu-
lum blades with a small amount of gel does not signif-
icantly reduce patient discomfort during speculum ex-
amination.8 Although increasing the amount of gel
lubricant should reduce the discomfort associated
with speculum insertion, data are lacking concerning
the effects on the Pap smear from contamination of
the cervix by greater than normal amounts of gel lu-
bricant. The aim of this study was to explore those ef-
fects by determining the proportion of unsatisfactory
Pap smears and the discordance in diagnoses between
gel-contaminated smears and uncontaminated smears
from the same patients.
Materials and Methods
Women undergoing conventional Pap smear at the
gynecology and family planning clinic of our universi-
ty hospital from July 2005 to April 2006 were invited
to participate. A total of 1,334 cases were recruited.
Informed consent was obtained from all participants
before examination. Women were grouped according
to reproductive status into 1 of 4 groups: normal re-
productive, lactation/postpartum, postmenopausal
and hormone replacement. Exclusion criteria includ-
ed (1) no identifiable cervix, (2) genital bleeding, (3)
obvious cervical cancer, (4) prior pelvic radiotherapy,
(5) current use of intravaginal medication except for
hormone replacement cream and (6) known allergy to
gel lubricant. The project received prior approval by
the Faculty of Medicine institutional review board.
The pelvic examination commenced with inspec-
tion and palpation of the external genitalia. The water-
lubricated speculum was then placed as usual. Two
cervical smears were collected from each patient using
an Ayre spatula for each. The pointed end of the spat-
ula was inserted and rotated a full turn clockwise. The
first sample (designated uncontaminated) was taken,
Volume 52
Gel Lubricant and Pap Smear
smeared on a glass slide and fixed with 95% alcohol as
usual. Any patients with significant contact bleeding
visible after obtaining the first cervical sample were
excluded from the study and the rest of the usual ex-
amination process completed. In the remaining pa-
tients, a 1- to 1.5-cm ribbon of K-Y Lubricating Jelly
(Johnson & Johnson, Sezanne, France) was applied
onto the external cervical os using a cotton-tipped
swab. The lubricant ingredients include chlorhexidine
gluconate, glucono delta lactone, glycerin, hydroxy-
ethylcellulose, methylparaben, sodium hydroxide and
purified water. The second sample (designated gel-
contaminated) was then obtained with another Ayre
spatula, smeared on another glass slide and fixed using
the same process as with the first sample. The glass
slides were kept in separate containers to prevent con-
tamination of gel lubricant to the uncontaminated
slides. Each slide was labeled with an identification
code without indicating gel status. All samples were
processed and stained in the same manner in the cy-
tology laboratory, except that uncontaminated and
gel-contaminated slides were kept separate to avoid
cross-contamination by the gel. All slides were inter-
preted by a single cytopathologist, who was blinded
to the gel status of the slides. After decoding of the
specimens, the results were compared for specimen
adequacy and cytologic diagnosis. Smears were con-
sidered unsatisfactory if there were < 8,000 well-
preserved and well-visualized squamous epithelial
cells or ≥ 75% of the epithelial cells were obscured by
blood, inflammation, thick areas, air-drying artifact,
poor preservation or foreign material. The satisfacto-
ry smears were further divided into 3 categories based
on cellularity: low, intermediate and high. The esti-
mated number of squamous epithelial cells identified
in a smear, according to the reference images, was
used as a criterion for each cellularity category, as fol-
lows: low, < 18,000 cells; intermediate, 18,000–50,000
cells; and high, > 50,000 cells.9 For statistical analysis,
the McNemar test was used to determine the differ-
ence between study groups. A p value < 0.05 was con-
sidered significant.
Results
The proportion of unsatisfactory smears was signifi-
cantly higher in the gel-contaminated specimens,
12.1% vs. 1.7% (p < 0.01). All unsatisfactory gel-
contaminated smears resulted from scant epithelial
cellularity. Subgroup analysis based on reproductive
status consistently demonstrated a significantly higher
proportion of unsatisfactory specimens in the gel-
contaminated slides (p < 0.01). This effect was most
132.239.1.231 - 1/8/2017 3:18:36 PM
evident in the lactation/postpartum and postmeno-
Univ. of California San Diego
pausal subgroups (Table I).
The relationship of Pap smear adequacy between
Downloaded by:
Number 6 November–December 2008 ACTA CYTOLOGICA 655
Charoenkwan et al

Table I Comparison of Proportion of Unsatisfactory Pap Smears Between Gel-Contaminated and Uncontaminated Specimens
Gel-contaminated Uncontaminated
No. of unsatis- Percentage No. of unsatis- Percentage
Reproductive status No. factory smears (95% CI) factory smears (95% CI) p Value
Normal reproductive 737 50 6.8 (5.1, 8.8) 10 1.4 (0.7, 2.5) < 0.01
Lactation/postpartum 303 61 20.1 (15.8, 25.1) 6 2.0 (0.7, 4.3) < 0.01
Postmenopausal 265 48 18.1 (13.7, 23.3) 7 2.6 (1.1, 5.4) < 0.01
Hormone replacement 29 3 10.3 (2.2, 27.4) 0 0 (0, 11.9) 0.25
Total 1,334 162 12.1 (10.4, 14.0) 23 1.7 (1.1, 2.6) < 0.01

the gel-contaminated and the uncontaminated smears
is demonstrated in Table II. The smear adequacy sta-
tus was the same (satisfactory/satisfactory or unsatis-
factory/unsatisfactory) in 87.3% of the patients. In
154 (11.5%) patients, the gel-contaminated smears
were unsatisfactory and the uncontaminated smears
were adequate, and in 15 (1.1%) patients the reverse
was true.
Of the 1,157 women whose paired cervical smears
were considered adequate for cytologic interpretation,
754 (65.2%) cases had comparable epithelial cellulari-
ty between the gel-contaminated and uncontaminated
smears. However, the epithelial cellularity was higher
in the uncontaminated smears (high/low, intermedi-
ate/low, high/intermediate) in 368 (31.8%) women
(Table III). The reverse situation of higher cellularity
in the gel-contaminated smears compared to the un-
contaminated occurred only in 3% of cases.
The gel was grossly and microscopically visible on
the gel-contaminated slides in 312 (23.4%) and 330
(24.8%) cases, respectively. Nevertheless, the gel con-
sistently appeared translucent and did not interfere
with cytologic evaluation. The degree of discordance
in cytologic diagnosis between the gel-contaminated
and uncontaminated smears from the same patient
was only 0.3% when only the satisfactory smears were
considered. Of note, a cytologic diagnosis of squa-
mous cell carcinoma, low grade squamous intraepithe-
lial lesion (LSIL) and atypical squamous cells of unde-
termined significance (ASCUS) was missed in the
Table II Comparison of Pap Smear Adequacy Between
Gel-Contaminated and Uncontaminated Smears for the
Same Patient
Uncontaminated
No. satis- No. unsatis-
Gel-contaminated factory (%) factory (%)
gel-contaminated smears from 1 patient each (Table
IV).
Discussion
Several recent well-designed studies support the use of a
small amount of water-based gel to lubricate the vagi-
nal introitus or speculum blades during cervical spec-
imen collection by demonstrating comparable smear
adequacy between the gel and no-gel groups.4-8,10
However, no significant alteration in pain/discom-
fort level was demonstrated with the use of gel in the
only study that examined the possible benefit of such
practice.8 Our findings show that if more than the
usual amount of gel lubricant is used in order to re-
duce pain and discomfort during the examination, this
can compromise cervical smear evaluation.
The proportion of unsatisfactory smears was signif-
icantly higher in the gel-contaminated group and was
always on the basis of scant epithelial cellularity. In ad-
dition, for almost one third of the cases with satisfac-
tory paired cervical smears, the epithelial cellularity
was lower in the gel-contaminated smears. This could
be the result of the gel acting as a barrier between the
spatula and the underlying cervical epithelium, or a di-
lutional effect due to the amount of gel used. One
might also predict that when 2 consecutive cervical
smears are obtained, the second smear (in this case,
the gel-contaminated smear) would contain a lower
number of epithelial cells simply because the first
scraping had removed most of the cells. However, the
literature suggests the contrary. In studies that found
increased sensitivity of Pap smears by performing 2
consecutive smears at the same visit, a comparable
proportion of unsatisfactory smears between the first
and the second smear was also noted.11,12 In one
study, the proportion of unsatisfactory smears was
5.3% for the first smear and 3.7% for the second
smear.11 It was suggested that although a significant
Total
number of desquamated cells were removed by the
first scraping, the second scraping obtained more cells
132.239.1.231 - 1/8/2017 3:18:36 PM

No. satisfactory 1,157 (86.7) 15 (1.1) 1,172 (87.9)

from deeper layers. In our study, there were a much
No. unsatisfactory 154 (11.5) 8 (0.6) 162 (12.1)
Univ. of California San Diego

higher proportion of unsatisfactory smears (12%) for

Total 1,311 (98.3) 23 (1.7) 1,334 (100)
the gel-contaminated specimens. Considering these
Downloaded by:

656 ACTA CYTOLOGICA Volume 52 Number 6 November–December 2008

 Gel Lubricant and Pap Smear

Table III Comparison of Epithelial Cellularity in the Satisfactory Smears Between Gel-Contaminated and Uncontaminated Smears
Uncontaminated smears
Gel-contaminated smears No. low (%) No. intermediate (%) No. high (%) Total (%)
No. low 32 (2.8) 337 (29.1) 6 (0.5) 375 (32.4)
No. intermediate 28 (2.4) 721 (62.3) 25 (2.2) 774 (66.9)
No. high 0 (0) 7 (0.6) 1 (0.1) 8 (0.7)
Total 60 (5.2) 1,065 (92.0) 32 (2.8) 1,157 (100)
White area = epithelial cellularity in the gel-contaminated smears was higher, italics = epithelial cellularity was comparable between the groups, bold = epithelial
cellularity in uncontaminated smears was higher.

data along with the paired sampling design of this
study, we believe that the application of more than the
usual amount of gel was the major factor contributing
to the higher number of unsatisfactory specimens in
the gel-contaminated group.
The proportion of unsatisfactory gel-contaminated
smears in the postpartum/lactation and postmeno-
pausal subgroups was approximately one fifth of the
patients in each of those subgroups, higher than that
of the normal reproductive subgroup. This was most
likely the result of postmenopausal and postpartum
hormonal change on cervical epithelium, which is
characterized by a predominance of parabasal cells and
scanty intermediate and superficial cells that are more
readily dislodged during cervical scraping. This raises
a clinical dilemma in that menopausal women would
have a higher chance of experiencing pain and dis-
comfort during speculum examination because of
vaginal dryness, yet the use of gel lubricant during the
examination is more likely to compromise the results
of the cervical smear in this group of women. Differ-
ence in the proportions of unsatisfactory smears be-
tween the gel-contaminated and uncontaminated
smears in the patient group with hormone replace-
ment also showed the same trend, but the sample size
for this subgroup was too small to exhibit statistical
significance.
The degree of discordance in cytologic diagnosis
between the gel-contaminated and uncontaminated
smears in the patients whose paired cervical smears
were considered adequate for cytologic interpretation
was very low. Even though almost a quarter of the
cases had evidence of residual gel on the smeared slide,
the gel did not interfere with the cytologic evaluation.
This indicates that the main negative effect of in-
creased gel use is in smear cellularity rather than
smear interpretation. Of importance, however, a cyto-
logic diagnosis of squamous cell carcinoma and LSIL
was missed in the gel-contaminated smears from 1 pa-
tient each. The prevalence of abnormal cervical
smears in this study was too low to make any conclu-
sion on this issue.
A possible limitation of this study is that we did not
have enough cases with epithelial abnormalities to ad-
dress the question of whether the use of gel increases
the false negative rate. In addition, we examined only
conventional cervical cytology smears and not liquid-
based Pap testing. The conventional smear is still the
major technique used in developing countries where
the incidence of cancer is high. There is evidence to
suggest that the information from this study would be
applicable to the liquid-based techniques. In a study
that examined the effect of gel lubricant on liquid-
based Pap testing, 2 simultaneous cervical specimens
were collected by the same technique and 1 specimen
was randomly assigned to be contaminated with the

Table IV Comparison of Cytologic Diagnosis in the Satisfactory Smears Between Gel-Contaminated and Uncontaminated Smears
Uncontaminated smears
Gel-contaminated
smears No. normal (%) No. ASCUS (%) No. LSIL (%) No. HSIL (%) No. SCC (%) Total (%)
No. normal 1,149 (99.3) 1 (0.1) 1 (0.1) 0 (0) 0 (0) 1,151 (99.5)
No. ASC-H 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.1) 1 (0.1)
No. LSIL 0 (0) 0 (0) 1 (0.1) 0 (0) 0 (0) 1 (0.1)
No. HSIL 0 (0) 0 (0) 0 (0) 2 (0.2) 0 (0) 2 (0.2)
No. SCC 0 (0) 0 (0) 0 (0) 0 (0) 2 (0.2) 2 (0.2)
132.239.1.231 - 1/8/2017 3:18:36 PM

Total 1,149 (99.3) 1 (0.1) 2 (0.2) 2 (0.2) 3 (0.3) 1,157 (100)

Univ. of California San Diego

ASC-H = atypical squamous cells, cannot exclude HSIL; HSIL = high grade squamous intraepithelial lesion; SCC = squamous cell carcinoma; bold = discordance in
cytologic diagnosis between the gel-contaminated and uncontaminated smears from the same patient.
Downloaded by:

Volume 52 Number 6 November–December 2008 ACTA CYTOLOGICA 657

Charoenkwan et al
gel after the collection and before being placed in the
cervical cytology solution.7 No interference in cervi-
cal cytology evaluation from the gel was shown. This
finding is in agreement with our results that the gel
had little effect on smear interpretation. However, the
design of that other study did not address the effect of
the gel on cervical cell collection.
In conclusion, the use of more than the usual
amount of gel lubricant during the pelvic examination
can interfere with the adequacy of cervical cytology
smears through a reduction in cellularity, although in-
terpretation of smears is not noticeably compromised.
Applying a small amount of gel to lubricate the specu-
lum is an option, based on individual patient prefer-
ence, although the benefit of doing this has not been
established. Use of large amounts of gel to lubricate
the speculum or fingers during speculum examination
or before Pap smear bimanual examination should be
avoided.
References
1. Eddy DM: Screening for cervical cancer. Ann Intern Med
1990;113:214–226
2. Janerich DT, Hadjimichael O, Schwartz PE, Lowell DM,
Meigs JW, Merino MJ, Flannery JT, Polednak AP: The screen-
ing histories of women with invasive cervical cancer, Connecti-
cut. Am J Public Health 1995;85:791–794
3. Kahn JA, Chiou V, Allen JD, Goodman E, Perlman SE, Emans
658 ACTA CYTOLOGICA Volume 52 Number 6 November–December 2008
SJ: Beliefs about Papanicolaou smears and compliance with Pa-
panicolaou smear follow-up in adolescents. Arch Pediatr Ado-
lesc Med 1999;153:1046–1054
4. Harer WB, Valenzuela G Jr, Lebo D: Lubrication of the vagi-
nal introitus and speculum does not affect Papanicolaou smears.
Obstet Gynecol 2002;100:887–888
5. Amies AM, Miller L, Lee SK, Koutsky L: The effect of vaginal
speculum lubrication on the rate of unsatisfactory cervical cy-
tology diagnosis. Obstet Gynecol 2002;100:889–892
6. Tavernier LA, Connor PD, Gates D: Water vs gel lubricant for
cervical cytology specimens. J Fam Pract 2003;52:701–704
7. Hathaway JK, Pathak PK, Maney R: Is liquid-based pap testing
affected by water-based lubricant? Obstet Gynecol 2006;107:
66–70
8. Gilson M, Desai A, Cardoza-Favarato G, Vroman P, Thornton
JA: Does gel affect cytology or comfort in the screening Papani-
colaou smear? J Am Board Fam Med 2006;19:340–344
9. Birdsong GG, Davey DD, Darragh TM, Elgert PA, Henry M:
Specimen adequacy. In The Bethesda System for Reporting
Cervical Cytology. Edited by D Solomon, R Nayar. New York,
Springer, 2004, pp 1–20
10. Griffith WF, Stuart GS, Gluck KL, Heartwell SF: Vaginal
speculum lubrication and its effects on cervical cytology and
microbiology. Contraception 2005;72:60–64
11. Shulman JJ, Hontz A, Sedlis A, Walters AT, Balin H, LoScuito
L: The Pap smear: Take two. Am J Obstet Gynecol 1975;121:
1024–1028
12. Luthy DA, Briggs RM, Buyco A, Eschenbach A: Cervical cy-
tology: Increased sensitivity with a second cervical smear. Ob-
stet Gynecol 1978;51:713–717
132.239.1.231 - 1/8/2017 3:18:36 PM
Univ. of California San Diego
Downloaded by: