Misoprostol and estradiol to enhance visualization of the transformation zone during cervical cancer screening: An integrative review

Abstract

The purpose of this integrative literature review was to appraise studies conducted worldwide using misoprostol and estradiol in converting Type 3 transformation zone (TZ) of the cervix into Types 1 or 2 and to assess which regimen could be more feasible in low-and-middle-income countries (LMICs). We reviewed the English language literature for peer-reviewed studies that evaluated strategies to convert Type 3 TZs to Types 1 or 2 for cervical cancer screening. Web of Science and PubMed searches were performed up to July 2020. Search terms included: “cervical colposcopy,” “inadequate colposcopy”, “cervical cancer screening”, “transformation zone,” “estrogen”, “estradiol”, and “misoprostol.” Inclusion criteria were articles published in the English language, original research, and peer reviewed articles. A total of 127 articles were abstracted, 24 articles were reviewed, and 9 articles met all inclusion criteria. We found that intravaginal misoprostol, intravaginal estradiol, and oral estradiol can successfully convert Type 3 TZ to Types 1 or 2. A single dose of vaginal misoprostol had a similar maximum response rate (20–80%) to a multi-dose regimen over several days or weeks of both intravaginal estradiol (64–83%) and oral estradiol (50–70%). Misoprostol administration was associated with more side effects such as abdominal cramping and vaginal bleeding compared to estradiol, although these were generally mild. In conclusion, Oral estradiol, intravaginal estradiol, and intravaginal misoprostol can be used to convert Type 3 TZ to Types 1 or 2. Intravaginal misoprostol is well tolerated and more feasible in LMICs due to availability and shorter treatment schedule compared to oral or intravaginal estradiol.

Introduction

Complete visualization of the transformation zone (TZ) of the cervix is of paramount importance during cervical cancer screening using visual techniques or colposcopy because the majority of cervical neoplasms originate from this area [1]. The TZ is the portion of the cervix between the new squamo-columnar junction (SCJ) and the original SCJ. The International Federation of Cervical Pathology and Colposcopy has proposed a classification system to categorize three types of TZs [2]. These include Type 1 TZ (when the entire new SCJ is located at the ectocervix and can be visualized at 360°), Type 2 TZ (when part or all of the SCJ is in the endocervical canal but can still be visualized at 360°), and Type 3 TZ (when part or all of the SCJ is in the endocervical canal and cannot be visualized at 360°) (Figs. 1a, 1b and 1c. It is Type 3 TZ that poses a problem during cervical cancer screening with visual techniques or colposcopy as the examiner may miss a precancerous or cancerous lesion [3]. Therefore, a strategy is needed to convert these Type 3 TZ into Types 1 or 2 TZ. It is estimated that Type 3 TZs account for approximately 10–15% of women presenting for cervical cancer screening [4,5].

Fig. 1a.

Type 1 transformation zone (TZ) where the entire squamo-columnar junction can be visualized at 360° at the ectocervix.

Fig. 1b.

Type 2 transformation zone (TZ) where part of the squamo-columnar junction is at the endocervix but can still be visualized at 360°.

Fig. 1c.

Type 3 transformation zone (TZ) where the squamo-columnar junction is at the endocervix and cannot be visualized at 360°.

Type 3 TZ has been associated with a four-to-five-fold increase in risk of missed cervical cancers [6]. Studies have previously evaluated methods to improve visualization of the TZ, including oral and intravaginal estrogen, intravaginal misoprostol, and hygroscopic cervical dilators such as laminaria [7,8,9]. Estrogen influences the eversion of the columnar epithelium onto the ectocervix, misoprostol is a prostaglandin E1 analogue that causes cervical ripening with softening and dilation of the cervix, and hygroscopic cervical dilators use osmosis to absorb fluid and expand the endocervix. Hygroscopic cervical dilators are not widely available in LMICs and thus were omitted from our review.

The aim of this integrative literature review was to appraise studies conducted worldwide using misoprostol or estradiol to convert Type 3 TZ of the cervix into Types 1 or 2 and to determine which regimen could be more feasible in LMICs, including the sub-Saharan Africa (SSA).

Materials and methods

This integrative literature review was guided by the integrative review methodology developed by Harris Cooper [10]. An integrative literature review facilitates inclusion of both qualitative and quantitative studies so as to gain full comprehension of the subject under study. The methodology of Cooper comprises five stages: 1) problem formulation, 2) literature search or data collection, 3) data evaluation, 4) data analysis, and 5) and presentation of results and interpretation.

Problem formulation

This integrative literature started with the identification of the problem which was guided by the following clinical questions: 1) What is the success rate of the different pharmacologic regimens? 2) What is the treatment schedule and dosage for each regimen? 3) What are the side effects of each regimen? 4) Are there patient-related factors that predict response to certain regimens?

Literature search and data collection

To answer these clinical questions, the authors reviewed the English language literature for peer-reviewed studies that evaluated strategies to convert Type 3 TZs to Type 1 or Type 2 TZs for cervical cancer screening from 1970 to 2020. Web of Science and PubMed searches of the English literature was performed. Search terms included: “cervical colposcopy,” “inadequate colposcopy”, “cervical cancer screening”, “transformation zone,” “estrogen”, “estradiol”, and “misoprostol”. Inclusion criteria were; articles published in English language, original research, and peer reviewed articles. We used Boolean’s operations; AND, OR, and, “ ”, to combine search terms such as “cervical cancer screening” OR “cervical colposcopy” OR “inadequate colposcopy” AND “transformation zone” AND estrogen OR estradiol OR misoprostol. We also reviewed the reference lists of the articles that were obtained from our initial searches.

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), an evidence-based minimum set of items designed at helping researchers to report an extensive array of systematic reviews and meta-analyses [11] was used to report data extraction and types of studies. This study was conducted with no external funding.

Data evaluation

The authors examined the retrieved articles and selected those that met inclusion criteria. The articles that met inclusion criteria were also assessed for duplicates.

Data analysis

The authors read each article and summarized the findings with regard to response rate, treatment dosage and schedule, side effects, patient factors associated with type of TZ, and response rate.

Results

Presentation of results

A total of 127 articles were abstracted using the search terms, including 99 from PubMed and 28 from Web of Science (Fig. 2). One additional publication was found from the reference list of one article. Twenty-four articles met the inclusion criteria. After duplicates were removed, 17 articles were left that were reviewed. Eight articles were further excluded because they were either reviews (n = 4), not in the English language (n = 1), or not directly related to the topic (n = 3). Therefore, nine articles were included in the study (Fig. 2). Of these nine articles, six were randomized controlled trials (RCTs) and three were prospective cohort studies.

Fig. 2.

PRISMA Flow Diagram.

One study compared oral and intravaginal estradiol [12], two studies examined oral estradiol alone [9,13], two studies examined intravaginal estradiol alone [14,15], one study compared intravaginal estradiol and intravaginal misoprostol [16], and three studies examined intravaginal misoprostol alone 7,8,17]. These studies were conducted in India (n = 3) [12,16,7], England (n = 2) [9,13], Thailand (n = 2) [8,17], Australia (n = 1) [15], and Italy (n = 1) [14]. A summary of the main findings from these studies is shown in Table 1. Of note none of the studies used oral misoprostol.

Table 1.

Summary of Studies (N = 9).

Drug	First Author, Year, Site	Study Design	Study Population	Treatment regimen	Time to Reexamination	Conversion from Type 3 TZ to Types 1 or 2	Side Effects
Oral/Vaginal Estradiol	Beniwal et al. 2016 India	Non-blinded RCT	40	Group 1: 25 ug vaginal estradiol daily × 7 days Group 2: 25 ug oral estradiol daily × 7 days	1 day after treatment	70% vaginal estradiol 50% oral estradiol p = 0.19	Nausea Vaginal discharge (similar in both groups)
Oral Estradiol	Saunders et al 1990 England	Double-blinded RCT	34	Group 1: 30 ug oral estradiol daily × 10 days Group 2: oral placebo daily × 10 days	Not reported	70% oral estradiol 23% oral placebo p < 0.01	None
	Prendiville et al. 1986 England	Prospective cohort	25	50 ug oral estradiol twice daily × 5 days during follicular phase	Last day of treatment (day 5)	64% oral estradiol	Mild nausea (75%)
Vaginal Estrogen	Piccolio et al. 2008 Italy	Prospective cohort	39	1 capsule vaginal promestriene daily × 20 days (followed by 1 week off) for 3 months	7 days after treatment	73% oral promestriene	Not reported
	Richards et al. 2014 Australia	Prospective cohort	54	0.5 mg estradiol gel vaginal twice weekly × 6 weeks	6 weeks after treatment	64% vaginal estradiol gel	Not reported
Vaginal Estrogen/Misoprostol	Makkar et al. 2014 India	Non-blinded RCT	48	Group 1: 200 ug vaginal misoprostol Group 2: 50 ug vaginal estradiol daily × 7 days	6 h after treatment 7 days after treatment	70.8% vaginal misoprostol 82.6% vaginal estradiol group p = 0.4	Nausea (12.5% misoprostol vs. 13% estradiol group, p = 1.0) Abdominal cramping (20.8% misoprostol vs 0% estradiol, p = 0.5) Vaginal bleeding (12.5% misoprostol vs. 0% estradiol, p = 0.07)
Vaginal Misoprostol	Agarwal et al. 2006 India	Double-blinded RCT	40	Group 1: 400 mcg vaginal misoprostol Group 2: vaginal placebo	6 h after treatment	78.9% vaginal misoprostol 30% placebo p = 0.004	Diarrhea (10.5% misoprostol vs 0% placebo)
	Thanapprapasr et al. 2009 Thailand	Double-blinded RCT	60	Group 1: 400 mcg vaginal misoprostol Group 2: vaginal placebo	6 h after treatment	20% vaginal misoprostol 6% vaginal placebo p = 0.172	Pain (13.3% misoprostol vs 0% placebo) Nausea (3.3% misoprostol vs. 0% placebo)
	Tungmunsakulchai et al. 2010 Thailand	Double- blinded RCT	40	Group 1: 200 mcg vaginal misoprostol Group 2: vaginal placebo	4 h after treatment	68.4% vaginal misoprostol 23.6% placebo p = 0.009	Fever (15.8% misoprostol vs 5.3% placebo, p = 0.3) Abdominal pain (42.1% misoprostol vs 0% placebo, p = 0.002) Nausea (15.8% misoprostol vs 0% placebo, p = 0.12) Diarrhea (10.5% in both groups, p = 0.24)

Efficacy

We found three available regimens for converting Type 3 TZs to Types 1 or 2 that had similar maximum response rates reported in the literature. Response rates were as follows: single dose of intravaginal misoprostol 200 mcg to 400 mcg (20–80%) [16,8,7,14], multi-dose regimen of vaginal estradiol (25ug – 0.5 mg) (64–83%) [12,16,18,14], and multi-dose regimen of oral estradiol (30–50 ug) (50–70%) [12,9,13].

Treatment dosage and schedule

Several regimens were proposed for oral and intravaginal estradiol as well as intravaginal misoprostol. The main difference between the two drugs lies in the need for multiple doses of estradiol regardless of route of administration versus a single dose of vaginal misoprostol [12,9,13,16,18,14]. Intravaginal estradiol (25ug – 0.5 mg) was typically used daily for seven days before repeat colposcopy [12,16]; however, some studies suggested six to twelve weeks of use with varying administration schedules prior to re-evaluation [18,14]. Similarly, oral estradiol regimens (30–50ug) typically ranged from 5 to 10 days before repeat examination with daily or twice daily administration [12,9,13]. All intravaginal misoprostol (200–400 mcg) was given in a single dose with colposcopy performed 4–6 h later [16,7,8,17] (Table 1.

Side effects

Misoprostol administration was associated with more side effects including abdominal pain or cramping, vaginal bleeding, and perceived increase in temperature. Abdominal pain was the most frequently reported (13–42%) [16,7,8,17]. Estradiol, both oral and intravaginal, had fewer reported side effects. Nausea and vaginal discharge were the most commonly noted side effects with no difference between route of administration (25% vs. 20%; p = 0.80) [12,13,16]. In the single study that directly compared intravaginal misoprostol and intravaginal estradiol, misoprostol caused significantly more side effects than estradiol (42% vs. 17%; p = 0.04) [16].

Patient factors associated with type of TZ and response rate

Menopausal status and hormonal contraception have been found to be associated with increased frequency of Type 3 TZ. The study of Beniwal et al. revealed that postmenopausal women were more likely to have unsatisfactory colposcopy (54%), which is consistent with Type 3 TZ, compared to premenopausal women (7%) [12].

There is limited information on factors that may predict successful response to pharmacologic methods used to convert Type 3 TZ to Types 1 or 2. Both the studies of Piccolio et al. and Richards et al. described how postmenopausal or hypoestrogenic patients with abnormal cytology have significant improvement in complete visualization of the TZ with the use of intravaginal estrogen therapy (73% and 59%, respectively) [15,14]. Furthermore, 40 of the 54 women in the Richards et al. study had a return to normal cytology after treatment with estrogen. However, the overall response rates to estrogen treatment did not differ by menopausal versus premenopausal status [12]. Response to intravaginal misoprostol administration was not found to be associated with age, body mass index, parity, menopausal status, or cervical cytology [6,8].

Studies have also evaluated whether certain anatomic features of the cervix may predict response to therapy. In regards to location of response, patients may show improvement in the anterior, posterior, or both lips of the cervix. Patients who received intravaginal estradiol were more likely to have a better response at the posterior lip of the cervix than patients who received oral estradiol (100% vs. 33%, respectively), but in patients who had improvement in both lips of the cervix, the overall response was similar (58% vs. 50%) [12]. Another study showed that there was no difference in response to the anterior versus posterior lips when vaginal estrogen was used [16], but that misoprostol was less effective in patients where the posterior lip of the cervix was the non-visualized portion of the unsatisfactory colposcopy compared with the anterior lip of the cervix (65% vs. 94%, p = 0.05). In contrast, the study of Aggarwal et al. demonstrated that 100% of patients who had unsatisfactory colposcopy involving the posterior lip of the cervix responded compared to 85.7% in those involving the anterior lip of the cervix in the misoprostol group [7]. The overall response rate in both lips was 57.1%. Additionally, patients with a slit-like os were more likely to respond than a pinhole external os (81.2% vs. 66.6%) [6]. These findings may be limited by their subjective nature and uncertain reproducibility.

Discussion

Interpretation of results

Our integrative literature review found that intravaginal misoprostol (200–400 mcg) and oral (30–50 ug) or intravaginal (25ug – 0.5 mg) estradiol are effective in converting Type 3 TZ to Types 1 or 2. The maximum success rate of single dose intravaginal misoprostol was similar to multi-dose intravaginal estradiol and oral estradiol. However, there were only two small, non-blinded RCTs directly comparing the response rate of intravaginal misoprostol and intravaginal estradiol (71% vs. 83%; p = 0.40) [16] and oral estradiol versus intravaginal estradiol (50% vs. 70%; p = 0.19) [12], with neither finding reaching statistical significance. Since efficacy may be similar between intravaginal misoprostol, oral estradiol, and intravaginal estradiol, it is necessary to evaluate the feasibility of each of these regimens in LMICs by considering drug dosing and duration, cost, access, and side effects.

The range of successful conversion of Type 3 TZ with intravaginal misoprostol is wide from 20% to 80%. Of the three studies that used intravaginal misoprostol, two studies conducted in India and Thailand had conversion rates of 68.4 – 78.9% [6,16] while the third conducted in Thailand had a conversion rate of 20% [8] (Table 1). It is not clear why this last study had such a low conversion rate given that the dosage and time to reevaluation were similar compared to the other two studies. Another Thai study that did not meet eligibility criteria for this review because it was available in abstract form only had a conversion rate of 55.6% [17]. Thus, the low 20% conversion rate is not consistent with the other studies of intravaginal misoprostol.

A surprising finding is the conversion of Type 3 TZ to Types 1 or 2 in the placebo groups. The placebo used in all the misoprostol trials was pyridoxine, a synthetic vitamin B6 which has a similar appearance to misoprostol but has no topical effect on the female reproductive tract [20]. In the two studies conducted in Thailand, the conversion rate of Type 3 TZ was 68.4% in the misoprostol group versus 23.6% in the placebo group (p = 0.009) [17], and 20% in the misoprostol group versus 6% in the control group (p = 0.172) [7]. In an Indian study, the conversion rate was 78.9% in the misoprostol group versus 30% in the control group (p = 0.004) [7]. In another Thailand study the conversion rate in the control group was 0% [21]. It is unlikely that pyridoxine has some effect on the cervix and differences in the technique used to perform the exam may have contributed. These findings demonstrate that a control arm is critical when assessing the efficacy of these drugs [22].

Based on the United States Food and Drug Administration’s (FDA) available data, the reported side effect profiles are consistent with the adverse events reported for each drug. The side effects associated with all regimens were well tolerated as most patients completed their prescribed treatment and underwent repeat examination. While misoprostol was associated with more reported side effects, it is also widely used in obstetrics for cervical ripening in labor. In the obstetric literature, misoprostol is most commonly associated with nausea, vomiting, diarrhea, and fever and is worse when given sublingually versus intravaginally [28]. In a Norwegian study where women were given up to 1000 mcg of misoprostol intravaginally, the side effects remained well tolerated [24].

Currently, there is limited data on the optimal patient population to target pharmacologic methods to convert Type 3 TZ. The SCJ changes with time in women and it is most commonly visible on the ectocervix during reproductive years as a response to estrogen and then regresses into the endocervical canal once a woman becomes postmenopausal [8]. Therefore, in postmenopausal women, one might postulate a lesser response of the TZ to pharmacological regimens, but this has not been examined in the current literature. While patients who have received treatment for precancerous lesions of the cervix (with cryotherapy or loop electrosurgical excision procedure) have higher rates of Type 3 TZ compared to those who have never received such treatment [29,30], it is unknown how this affects the response to these interventions. Furthermore, although misoprostol induces cervical changes, a search of the entire PubMed database did not yield any study examining whether this effect on the cervix affects the clearance of HPV infection.

Our study had both strengths and limitations. The major strength of this study is that our integrative review includes multiple studies that examined conversion of Type 3 TZ to Types 1 or 2 with different pharmacologic regimens (oral estradiol, intravaginal estradiol, and intravaginal misoprostol). The lone systematic review previously conducted on this subject was limited to intravaginal misoprostol only [20]. The limitation of the study is the small number of articles retrieved from electronic database search. We were interested to find out how pharmacologic regimens aimed at converting Type 3 TZ to Types 1 or 2 could work in LMICs, particularly in SSA, but no study was retrieved from this geographic region, which highlights a knowledge gap. In addition, the specific studies had several weaknesses that follow. None of the studies in this review stratified their participants by age, menopausal status, or history of prior treatment for cervical cancer precancer. All the studies were not powered adequately to evaluate important subgroup efficacy. When reporting the conversion of Type 3 TZ, none of the studies categorized the conversion to either Type 1 or Type 2 TZs.

Whereas in high-income countries, additional sampling with endocervical curettage is used to further investigate women with Type 3 TZ [18], obtaining pathological results from the endocervical canal in LMICs can be challenging because pathologists are scarce and often do not have the experience or equipment to adequately analyze samples from endocervical curettage. Therefore, women with Type 3 TZ in LMICs may have missed dysplastic lesions or historically may have undergone more invasive excisional procedures to completely evaluate the cervix [16,19]. Practical solutions to convert Type 3 TZ to Type 1 or 2 TZ are needed in LMICs.

Since efficacy may be similar between intravaginal misoprostol, oral estradiol, and intravaginal estradiol, it is necessary to evaluate the efficacy of these regimens in LMICs. We propose that such a trial should prioritize the evaluation of vaginal misoprostol over estradiol considering the low cost, ease of availability, and common use of the medication for obstetric reasons in many LMICs. Misoprostol is included on the 2019 WHO Essential Medications List [23], whereas, estradiol is not. The intravaginal (as opposed to oral) route of administration minimizes systemic side effects of misoprostol and avoids other potential side effects of estradiol use. Such studies should also have sufficient sample size and power to allow for women to be stratified according to menopausal status and prior treatment of precancerous lesions in order to determine if certain subgroups respond differently. The cost of misoprostol is lower than that for estradiol [24], and this is further compounded by the multi-dose regimen required for the recommended administration of estradiol, which increases the duration of the visit or requires a second clinic appointment, increasing the risk that women will be lost to follow up [25,26]. While women have to wait 4–8 h after misoprostol administration, the evaluation can be performed in a single visit, which supports the “see-and-treat” approach of cervical cancer screening recommended for LMICs by WHO [27]. The optimal dose of misoprostol to use is not known and we would recommend using a dose of 600 mcg that has demonstrated efficacy in preliminary studies. Furthermore, serial assessment of after placement of the study medication e.g., 4, 6 and 8 h will help determine the optimal duration of misoprostol to allow for conversion. By testing for HPV, follow-up substudies can be done to evaluate whether misoprostol modulation of the cervix influences the natural history of HPV clearance. Additional studies may compare misoprostol and estradiol. We hypothesize that, if successful, the trial protocol outlined above is likely to lead to fewer missed precancer lesions and more effective prevention of cervical cancer in LMICs.

Conclusion

Oral estradiol, intravaginal estradiol, and intravaginal misoprostol are promising pharmacologic methods to convert Type 3 TZ to Types 1 or 2. However, there is no consensus on the optimal dosage or duration of use. Considering the duration, cost, availability of treatment, and the feasibility of completing a “see-and-treat” approach recommended for LMICs by WHO, the intravaginal misoprostol regimen appears preferable for first line evaluation in LMICs. An RCT of the intravaginal misoprostol regimen with adequate sample size is recommended in SSA to investigate the conversion rate of Type 3 TZ to Types 1 or 2.

Abbreviations:

HPV

Human papilloma virus

LMIC

Low- and middle-income countries

TZ

Transformation zone

RCT

Randomized control trial

SCJ

Squamo-columnal junction

SSA

Sub Saharan Africa

VIA

Visual inspection with acetic acid

WHO

World Health Organization