HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by CHURCH, L. Articles by ELLSWORTH, A. Search for Related Content PubMed PubMed Citation Articles by CHURCH, L. Articles by ELLSWORTH, A.

Analgesia for Colposcopy: Double-Masked, Randomized Comparison of Ibuprofen and Benzocaine Gel

From the Department of Family Medicine, School of Medicine, University of Washington, Seattle, Washington.

Address reprint requests to: Lili Church, MD Department of Family Medicine UWMC Roosevelt, Box 354775 University of Washington School of Medicine 4245 Roosevelt Way NE Seattle, WA 98105 E-mail: llchurch{at}u.washington.edu

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Objective: To evaluate pain relief effectiveness of oral ibuprofen and topical benzocaine gel during colposcopy.

Methods: In a double-masked, randomized controlled trial, women who attended a family medicine colposcopy clinic received one of four treatments, 800 mg of oral ibuprofen, 20% topical benzocaine, both, or placebos. Using visual analog scales, women recorded their pain after speculum placement, endocervical curettage (ECC), and cervical biopsy. Participants were 18–55 years old, spoke English, and were not taking other pain or psychotropic medications. Demographic and historical information was collected from each participant.

Results: Ninety-nine subjects participated. Twenty-five received oral ibuprofen and topical benzocaine (median pain scores on a 10-point scale for speculum placement, ECC, and biopsy were 0.75, 3.00, and 3.38, respectively), 24 received oral placebo and topical benzocaine (1.00, 3.75, and 2.63), 24 received oral ibuprofen and topical placebo (0.63, 3.75, and 2.25), and 26 received oral and topical placebos (0.75, 3.50, and 3.00). There were no statistically significant differences in patient visual analogue pain scale scores across the four groups (statistical power, ECC = 0.74, cervical biopsy = 0.62). Younger women and women who had pain with speculum placement were more likely to have increased pain during ECC. Increased pain during biopsy was associated with history of severe dysmenorrhea but no other demographic or historical factors. Women overall reported ECC and biopsy to be mildly painful, with median scores of 3.5 for ECC and 2.75 for biopsy on a 10-point scale. The range in pain scores was large, with some women reporting severe pain (for ECC minimum = 0.25, maximum = 10.0; biopsy: minimum = 0.0, maximum = 9.0).

Conclusion: Colposcopy is perceived as somewhat painful, but oral ibuprofen and topical benzocaine gel, alone or together, provided no advantage over placebo in decreasing colposcopy pain.

Clinical use of anesthetic and analgesic agents for colposcopy is sporadic and without clear guidelines. Despite lack of supportive evidence in the literature, many colposcopists attempt to decrease pain by giving an oral nonsteroidal anti-inflammatory drug before the procedure or applying a topical anesthetic to the cervix before biopsy. Those techniques are inexpensive, easy to administer, and have few risks. They provide analgesia without the discomfort or potential risk of anesthetic injections, and they do not have the additional risk of conscious sedation.

Nonsteroidal anti-inflammatory drugs have been used successfully to treat primary and intrauterine device (IUD)–induced dysmenorrhea by inhibiting prostaglandin production and decreasing uterine contractions.¹ Topical anesthetics such as benzocaine, highly effective in dermatologic and dental procedures, are absorbed through mucosal surfaces and block neural transmission of painful stimuli. Whether they are efficacious in gynecologic procedures is not known.

Prior investigations have conflicting results. Uncontrolled studies suggested that oral nonsteroidal anti-inflammatory drugs are beneficial for colposcopy² but not for cervical laser procedures.³ In a placebocontrolled study, Rabin et al found a topical anesthetic effective during colposcopy,⁴ but two other placebocontrolled trials did not.^5,6 Studies of topical anesthetics in other gynecologic procedures also yielded conflicting results.^7–10

There is no consensus about the amount of pain associated with colposcopy. Some studies concluded that pain is insignificant.⁵ Others suggested pain is mild to moderate.^2,4,6 Few scored colposcopic pain as severe.^2,6 Patient factors might also influence pain. Nulliparity, anxiety, and history of dysmenorrhea have correlated with increased pain accompanying certain gynecologic procedures.^11,12

We conducted a randomized, double-masked, placebo-controlled trial with oral ibuprofen and topical benzocaine gel, alone and together, to study their effects on pain during colposcopy. Prior studies did not compare them to each other and placebo. To better predict which women were at higher risk for significant discomfort from the procedure, we analyzed factors possibly associated with different responses to pain. We hypothesized that increasing age and parity would be associated with less pain.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
The study was conducted at the Family Medical Center, the ambulatory care teaching site for the University of Washington Family Medicine Residency Program. Two hundred thirty-eight consecutive women, referred for colposcopy from within the practice, were screened between July 1995 and June 1998. Subjects were excluded if they were younger than 18 or older than 55 years, did not speak English, or were pregnant. They were also excluded if they had taken analgesics in the previous 24 hours, were taking psychotropic or anticoagulant medications, had histories of bleeding disorders or bleeding ulcers, or adverse reactions to nonsteroidal anti-inflammatory or local anesthetic drugs. One hundred seven women were excluded, including 18 non–English speakers and 49 who took disqualifying medications. Two unanticipated reasons for exclusion were inability to swallow pills and one organ transplant recipient. The study protocol was reviewed and approved by the University of Washington Human Subjects Review Committee, and subjects gave written informed consent.

Participants completed questionnaires on demographic and historical information, including age, education, race or ethnicity, obstetric and gynecologic history, current medications, history of previous sexual abuse, current or past substance abuse, history of depression or anxiety, and numerous questions related to experiences with various types of pain (dysmenorrhea, irritable bowel syndrome, back pain, headaches, dyspareunia, pelvic pain, other chronic pain).

Participants were randomly assigned to one of four treatment groups, 800-mg oral ibuprofen capsule and 20% benzocaine topical gel, placebo oral capsule and 20% benzocaine topical gel, 800-mg oral ibuprofen and placebo topical gel, or placebo oral capsule and placebo topical gel. This design allowed testing for individual and potential combined treatment effects. It was possible that age and previous childbirth might affect pain perception, so a blocked randomization scheme was used. To ensure that randomization was balanced with respect to parity and age, four categories were created, younger than 30 years and parity zero, younger than 30 years and parity more than zero, older than 30 years and parity zero, and older than 30 years and parity more than zero. Treatments were distributed randomly within each category. To do that, a computer-generated randomization scheme in blocks of eight was used within each category to ensure balance (for every block of eight women, there were two assignments to each of the four treatments).¹³ Individuals were assigned within the appropriate category in chronological order as they presented.

Study drugs were prepared and the randomization scheme code maintained by Pharmaceutical Services, University of Washington Medical Center. Ibuprofen and placebo oral capsules were indistinguishable, as were the benzocaine and placebo white petrolatum topical gels. Patients, physicians, pharmacists, nurses who administered oral agents, the research assistant, and data analysts were all unaware of the identity of the agents used. The code was not broken until the study was completed.

Each woman was prepared for colposcopy in the usual manner and received the oral study drug at least 30 minutes before endocervical curettage (ECC) or cervical biopsy. The research assistant and one of three physician authors were present at all procedures. A medium-sized Graves speculum was placed. At least 2 minutes before ECC, approximately 2 mL of topical study ointment was applied with a cotton swab into the endocervical canal and onto the adjacent exocervix.

The primary outcome was patient-described pain level, measured with a 5-inch-long, unmarked, continuous horizontal line as a visual analog scale. Subjects were told that the far left point represented "no pain" and the far right point represented "the worst pain you can imagine." They were asked to mark a vertical line on the visual analog scale three times during the procedure to note levels of discomfort. Separate scales were presented immediately after speculum placement, after ECC, and after the first ectocervical biopsy.

For analysis, the visual analog scale was converted into ten points, each subdivided into fourths, resulting in a 40-unit scale. With those 40 units, it was statistically reasonable to treat the scale as continuous. Medians and ranges were used because the data were skewed and the standard deviations large. Median responses for each arm were compared for each procedure (speculum placement, ECC, biopsy). Nonparametric Wilcoxon and Kruskal-Wallis tests were used for all comparisons. Although we hypothesized a decrease in pain with medication, medication might also have negative effects; therefore, the more conservative two-tailed tests were done for statistical analysis. No statistically significant differences were found across the treatment arms, so we combined data from all arms to describe overall pain perception among subjects and analyze other factors associated with mean pain perceived for each procedure. Normal linear models were used for tests exploring demographic and historical predictors of pain.

Using an analysis of variance model, the study was powered to have an 80% chance of detecting a true difference of two of ten points in the mean perceived pain between the largest and smallest means of the four treatment groups ( = .05). We reasoned that a difference of two points on a ten-point pain scale would be clinically meaningful. Based on findings of Ward et al, we assumed a large standard deviation (2.25 of ten) to calculate sample size.¹⁴ The sample calculation was done with JMP (SAS Institute Inc., Cary, NC).

	Results

Top Abstract Materials and Methods Results Discussion References

 
Among 131 eligible women, 99 agreed to participate. All had speculum placement, 95 (96%) had ECC, and 93 (94%) had at least one cervical biopsy. Demographics and historical information are presented in Table 1. There were no statistically significant differences in characteristics across arms. Although we did not ask for reasons for declining participation, we compared age and gravidity between women who participated and women who declined participation. Women who participated were significantly younger than eligible women who did not participate (31.7 versus 36.3 years, P = .02). There were no statistically significant differences in gravidity. The mean age of women who participated was not significantly different than women who were ineligible (31.7 versus 34.9 years, P = .17), although a difference was expected because age was an exclusion criterion. Several questionnaires were incomplete, as noted in Table 1.

There were few predictors of pain. Pain with speculum placement was a positive predictor of pain with ECC (correlation coefficient = 0.29, P < .01) but not biopsy (correlation coefficient = 0.19, P = .07). Older age also correlated with less pain from ECC (correlation coefficient = -0.31, P < .01) and biopsy (correlation coefficient = -0.21, P < .05) but not speculum placement (correlation coefficient = - 0.08, P = .41). Decreasing pain with advancing age was independent of parity, race or ethnicity, and all other demographic and historical factors. However, the effect was small and equivalent to a decrease of 0.75 on a ten-point scale for every 10-year advance in age. Women with histories of menstrual cramps reported more pain from biopsy than women with no such histories (P = .02) but did not report significantly more pain from speculum placement or ECC (Table 3). No other demographic or historical factors correlated significantly. Pain might vary with technique, so patient pain scores were compared among the three physicians. There were no significant differences in any of the groups for speculum placement (P = .80), ECC (P = .39), or biopsy (P = .29; not shown).

	Discussion

Top Abstract Materials and Methods Results Discussion References

For many women, colposcopy is painful, but its measurement and interpretation in the literature have been variable, making comparisons difficult. Two previous studies that used ten-point visual analog scales showed mean pain scores with nonmedicated colposcopy of 1.92⁵ and 4.6,² with ranges of 0–7.3⁵ and 0–10.² Our ECC median of 3.5, with range of 0.75–9.5, and biopsy median of 2.75, with range of 0–7.75, are comparable. Other studies that used qualitative descriptions reported pain with nonmedicated colposcopy to average "mild" to "moderate," and range from "none" to "severe."^4,6 Clinically relevant change in pain is also difficult to define, and with one exception, studies of gynecologic procedures have not discussed it.⁵ Work in other areas suggested that a decrease of at least 1.3 on a ten-point visual analog scale¹⁵ or a 16% change in pain¹⁶ is a clinically relevant outcome for an intervention. Nonetheless, there is no reason to argue against attempting to decrease pain if it is simple, safe, and cost effective.

To control for the possibility that ECC and biopsy have different pain mechanisms, we measured each separately. Most other studies included all women "completing colposcopy," whether they had ECC or biopsy. Those studies measured pain once and delayed measurement until after the procedure was completed.^2,5,6 In our study, by measuring pain immediately after each part of the procedure, we showed that both are painful and that study interventions did not affect pain perception after ECC or biopsy.

We included patient perception of pain at speculum placement as a measurement of baseline pain. We hypothesized that pain with speculum placement would be positively correlated with pain at other times of the procedure, which it did for ECC but not for biopsy. We also hypothesized that demographic or historical factors, eg, history of sexual abuse, would be correlated with predicting pain with speculum placement, but for most factors that was not confirmed, and numbers were small (eg, five of 97 women reported histories of sexual abuse). The exceptions were age (inversely), a history of severe menstrual cramps, and discomfort with speculum placement.

This study had several limitations. Women who took any pain medications or psychotropic medications were excluded, which resulted in fewer eligible, representative subjects. A significant proportion (32 of 131) of eligible women also declined participation. We limited our study to pain assessment and did not address symptoms such as anxiety and cramping.^17–19 However, pain perception is complex and also affected by psychologic factors. We did not study presence of a support person, which also might influence pain perception. Individual physicians’ bedside manner is also variable. In our study three different faculty physicians did colposcopies and were assisted by rotating senior family medicine residents. A research assistant introduced women to the study, was present at the bedside, and administered the visual analog scale. The wide variability in pain for all treatment and demographic groups in our study further supports that pain is complex and difficult to predict for any individual.

The study was designed to have 80% power to detect two of a ten-point difference in perceived pain at a 5% significance level, but the actual power was slightly less. For pain at ECC, the approximate power to detect a difference of two points between the largest and smallest of the four medians was 74%. For pain at biopsy, the approximate power was 62%. The loss in power for pain at biopsy was caused by the sample of 93 instead of the calculated 100, the slightly larger than expected standard deviation (2.5 instead of 2.25), and the use of nonparametric tests. A larger study would detect smaller differences, but we contend that our study was adequate to detect moderate to large treatment effects.

Age was an exclusion criterion, but we did not observe a significant difference between participants and women who were excluded. However, the mean age of 36.3 years for subjects who were eligible but declined participation was significantly greater than the mean age of 31.7 years for participants (P = .02). If increasing age does correlate with decreased pain, then our median pain scores might have been higher than if those women had chosen to participate. Likewise, a younger population than ours might report higher median pain scores. Participants in all treatment arms were well matched for other demographic and historical characteristics.

Collection of data took 3 years at our small, single-study site. Exclusion criteria eliminated a much larger proportion of women than was anticipated. The long enrollment period potentially introduced biases, including practitioner technique and other subtle changes in style over time. Colposcopy is a procedure with a standardized format, so in this study there were no changes in colposcopic procedure protocols, instruments, faculty physician participation, or procedure rooms. Our study provided placebos for oral ibuprofen and topical benzocaine gel but did not control for placebo effect. We did not measure pain levels when no oral tablet and no topical gel was given.

Although this study did not show a reduction of pain with the study medications, there might be other significant and potentially treatable factors correlated with pain we did not address. The intervention of ibuprofen and benzocaine might still be helpful in a subpopulation of women. Other studies might answer that question.

For colposcopy, it might be appropriate to use different anesthetics and analgesics. Future work also needs to assess submucosal cervical injection, which decreased pain in studies with cryotherapy and loop electrosurgical excision procedure.^20–23 Although paracervical blocks and transcutaneous electrical nerve stimulation units have not been shown to decrease pain of cryotherapy and laser,^19,24,25 it would be reasonable to test their use in colposcopy. Light or conscious sedation with oral or parenteral medications, methods with different risks and added costs, have not been adequately studied during colposcopy.

	Footnotes

 
Support for this research came from the American College of Clinical Pharmacy-Bristol Meyers Squibb Family Medicine Research Award.

PII S0029-7844(00)01084-X

Received April 21, 2000. Received in revised form July 21, 2000. Accepted September 13, 2000.

	References

Top Abstract Materials and Methods Results Discussion References

 
1. Dawood MY. Nonsteroidal anti-inflammatory drugs and changing attitudes toward dysmenorrhea. Am J Med 1988;84:23–9.

2. Rodney WM, Huff M, Euans D, Hutchins C, Clement K, McCall JW III. Colposcopy in family practice: Pilot studies of pain prophylaxis and patient volume. Fam Pract Res J 1992;12:91–8.[Medline]

3. Al-Kurdi M, Hare MJ, Lowles I, Douglas CP, English JR. The effect of a prostaglandin-synthetase inhibitor, naproxen sodium, and a placebo on the pain response to carbon dioxide laser treatment of the uterine cervix. J Obstet Gynaecol 1985;5:260–2.

4. Rabin JM, Spitzer M, Dwyer AT, Kaiser IH. Topical anesthesia for gynecologic procedures. Obstet Gynecol 1989;73:1040–4.[Medline]

5. Clifton PA, Shaughnessy AF, Andrews S. Ineffectiveness of topical benzocaine spray during colposcopy. J Fam Pract 1998;46:242–6.[Medline]

6. Prefontaine M, Fung-Kee-Fung M, Moher D. Comparison of topical Xylocaine with placebo as a local anesthetic in colposcopic biopsies. Can J Surg 1991;34:163–5.[Medline]

7. Sarker PK. Topical anaesthesia with lignocaine-prilocaine cream (EMLA) for carbon dioxide treatment to the cervix—A pilot study. Br J Clin Pract 1990;44:352–3.[Medline]

8. Mikhail MG, Bevan JR. A randomized trial of the use of cocaine spray to provide pain relief during laser vaporization of the cervix. Br J Obstet Gynaecol 1988;95:469–72.[Medline]

9. Munsick RA. Topical anesthesia of the uterine cervix or corpus. Obstet Gynecol 1975;46:613–5.[Abstract/Free Full Text]

10. Lipscomb GH, McCord ML, Bain KW, Ling FW. The effect of topical 20% benzocaine on pain during loop electrosurgical excision of the cervix. Am J Obstet Gynecol 1995;173:772–4.[Medline]

11. Johnson N, Crompton AC, Parker J. The correlation between dysmenorrhea and the pain experienced during laser ablation of a cervical lesion. Gynecol Oncol 1990;36:215–6.[Medline]

12. Johnson N, Crompton AC. Who finds cervical laser therapy painful? Gynecol Oncol 1994;52:44–9.[Medline]

13. Snedecor GW, Cochran WG. Statistical methods. Ames: Iowa State University Press, 1980.

14. Ward SE, Goldberg N, Miller-McCauley V, Mueller C, Nolan A, Pawlik-Plank D, et al. Patient-related barriers to management of cancer pain. Pain 1993;52:319–24.[Medline]

15. Todd KH, Funk KG, Funk JP, Bonacci R. Clinical significance of reported changes in pain severity. Ann Emerg Med 1996;27:485–9.[Medline]

16. Wells GA, Tugwell P, Kraag GR, Baker PR, Groh J, Redelmeier DA. Minimum important difference between patients with rheumatoid arthritis: The patient’s perspective. J Rheumatol 1993;20:557–60.[Medline]

17. Marteau TM, Walker P, Giles J, Smail M. Anxieties in women undergoing colposcopy. Br J Obstet Gynaecol 1990;97:859–61.[Medline]

18. Rickert VI, Kozlowski KJ, Warren AM, Hendon A, Davis P. Adolescents and colposcopy: The use of different procedures to reduce anxiety. Am J Obstet Gynecol 1994;170:504–8.[Medline]

19. Harper DM. Paracervical block diminishes cramping associated with cryosurgery. J Fam Pract 1997;44:71–5.[Medline]

20. Chanen W. The efficacy of electrocoagulation diathermy performed under local anaesthesia for the eradication of precancerous lesions of the cervix. Aust N Z J Obstet Gynaecol 1989;29:189–92.[Medline]

21. Whiteley PF, Olàh KS. Treatment of cervical intraepithelial neoplasia: Experience with the low-voltage diathermy loop. Am J Obstet Gynecol 1990;162:1272–7.[Medline]

22. Sammarco MJ, Hartenbach EM, Hunter VJ. Local anesthesia for cryosurgery on the cervix. J Reprod Med 1993;38:170–2.[Medline]

23. Harper DM, Dobb JL. Cervical mucosal block effectively reduces the pain and cramping from cryosurgery. J Fam Pract 1998;47: 285–9.[Medline]

24. Johnson N, Crompton AC, Ramsden SV. The efficacy of paracervical injections of lignocaine before laser ablation of the cervical transformation zone. A randomized placebo-controlled double-blind clinical trial. Br J Obstet Gynaecol 1989;96:1410–2.[Medline]

25. Crompton AC, Johnson N, Dudek U, Batra N, Tucker A. Is transcutaneous electrical nerve stimulation of any value during cervical laser treatment? Br J Obstet Gynaecol 1992;99:492–4.[Medline]

This article has been cited by other articles:

				Does Analgesia or Anesthesia Make Colposcopy Less Painful? Journal Watch Women's Health, March 6, 2001; 2001(306): 2 - 2. [Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by CHURCH, L. Articles by ELLSWORTH, A. Search for Related Content PubMed PubMed Citation Articles by CHURCH, L. Articles by ELLSWORTH, A.