Office Cervicoscopy versus Stationary Colposcopy in Suspicious Cervix: A Randomized Controlled Trial.

STUDY
OBJECTIVE: The objective of the study was to estimate the diagnostic accuracy and doctor satisfaction of small caliber office cervicoscopy versus stationary colposcopy in diagnosis of ectocervical as well as endocervical lesions in women clinically presented with suspicious cervix. PATIENTS AND METHODS: Eligible 112 cases with clinically suspicious cervix were randomized into Group A (56 cases) and Group B (56 cases) who were subjected to small caliber office cervicoscopy and stationary colposcopy, respectively. The outcome was the diagnostic accuracy and safety of both tools for detection of ectocervical and endocervical cervical lesions.
RESULTS: There was no statistically significant difference between both groups regarding parity, previous abortion, age at marriage, duration of marriage, and age at menarche and menopause. On unaided naked eye examination of the cervix (UNEE), there were no statistically significant differences between both groups. Diagnostic indices were similar in both groups apart from the finding that office cervicoscopy was more sensitive for detection of endocervical abnormalities. Doctors were significantly more satisfied with stationary colposcopy than office cervicoscopy.
CONCLUSIONS: Office cervicoscopy is a good complementary tool added to stationary colposcopy for detection of cervical lesions in cases with suspicious cervix as an example of high-risk group for cervical cancer. Due to its small caliber, cervicoscopy offers a better evaluation of the endocervical canal, especially in cases of Type 2 and 3 transformation zone with a possibility of examination of the endometrial cavity. Copyright:

INTRODUCTION

Formerly, Papanicolaou (Pap) test was the most common and cost-effective screening method for cervical cancer.[12345] Whenever access to Pap smear is limited, unaided naked-eye examination (UNEE) with or without visual inspection after application of acetic acid is an acceptable alternative screening approach.[6] Colposcopy is the gold standard tool for further management of abnormal Pap smears and is the second step of the diagnostic approach.[7] Nevertheless, the real problem of endocervical neoformations and transformations in the deep glandular cells represents a diagnostic difficulty of both Pap smear and colposcopy which often escape their diagnosis. Standard office hysteroscopy allows proper evaluating of the endocervical canal (cervicoscopy)[8] to screen for premalignant cervical lesions.[9] However, randomized comparison of office cervicoscopy (OC) and stationary colposcopy (SC) is missing in the medical literature. Moreover, previously published studies on cervicoscopy [89] evaluated the endocervical canal without comment on the ectocervix and the transformation zone (TZ) in patients with negative or unsatisfactory colposcopy. Nowadays, the 2011 International Federation for Cervical Pathology and Colposcopy (IFCPC) nomenclature classification [10] is the global reference standard for cervical colposcopic examination findings. This study aimed to estimate diagnostic indices and doctor satisfaction on using a small caliber OC versus SC in the diagnosis of cervical lesions in women with clinically unhealthy-looking (suspicious) cervix as guided by the 2011 IFCPC guidelines.[10]

PATIENTS AND METHODS

Study design and setting

This is a prospective cross-sectional interventional randomized controlled trial (RCT) performed at the Early Cervical Cancer Detection Unit of the Woman's Health University Hospital, Assiut, Egypt, between July 2016 and December 2017. It was approved by the Institutional Review Board of the Faculty of Medicine on June 2016.

Trial registration

This study is internationally registered on clinical trial.gov (NCT03150745).

Subject selection

It comprised 250 nonpregnant women with clinical diagnosis of unhealthy-looking (suspicious) cervix attending gynecologic, family planning, and infertility clinics at the Woman's Health Hospital, Assiut University, Egypt [Figure 1]. Patients gave a written consent to participate in this study.

Figure 1

Flowchart of the studied cases

Sample size calculation

The sample size calculation was carried out using Epi Info software (version 7), Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia (US), using confidence interval 95% and power 80%. Hence, total sample size was 112 patients that were divided into two groups.

Method of randomization

Eligible 112 cases were randomized using cards; 112 were sequentially numbered. Opaque sealed envelopes were used which containing 56 cards were labeled office cervicoscopy and 56 cards were labeled stationary colposcopy. All envelops were mixed together randomly in a box. The cards were selected randomly and once selected never changed.

Clinical work-up

A thorough history was taken with special emphasis on age and parity of the patients, age and duration of marriage, menstrual pattern, contraceptive history (with stress on type and duration of use), color and odor of vaginal discharge, pelvic pain, backache, or any associated symptoms. Examination in lithotomy position was done using a nonlubricated bivalve vaginal speculum. The cervix was cleaned from any secretions or blood using sterile 0.9% saline. A thorough naked-eye assessment (UNEE) of the cervix was done as previously described by our team.[6] If a cervical lesion was seen, it was recorded and its site was marked on a special diagram. Cervical Pap smears were obtained using Ayre's spatula as the tip was inserted into the endocervical canal, while the shoulder was placed on the ectocervix. The spatula was gently rotated around the cervix (360°) so that a representative sample of the whole cervix was obtained. The samples were then immediately plated on a slide, fixed by immersing the slide in 95% ethyl alcohol fixative for 15–20 min, and stained by modified Pap stain using a hand staining procedure. Each smear was screened using modified Bethesda System terminology [11] and checked for cytological evidence of inflammation and identification of cellular atypia, dysplasia, or carcinoma.

Patient grouping

A total of 112 cases with clinically suspicious cervix were included in this study. They were classified into Group A (56 cases) who were subjected to office cervicoscopy and Group B (56 cases) who were examined by the stationary colposcope. In Group A, office cervicoscopy was done using the same instruments used for office hysteroscopy (2.6 mm telescope, outer sheath 3.6 mm, a digital endocamera, and high-resolution monitor, KARL STORZ, Tuttlingen, Germany) and includes six steps: (a) 0.9% saline technique to assess the cervical lesion and vasculature of the cervix, (b) 5% acetic acid technique to determine acetowhite-positive areas, (c) Schiller's iodine technique to visualize high glycogen-containing cells, (d) endocervical canal assessment, (e) endometrial cavity evaluation whenever possible, and (f) biopsy using the Punch Biopsy Forceps was obtained from every abnormal office cervicoscopic examination. In Group B, stationary colposcopic examination was done using Karl Kaps Colposcope model (SOM 52), Germany, and included saline technique, acetic acid application, Schiller's iodine test, and endocervical canal assessment using an endocervical speculum or counter pressure with Q stick, and finally, Punch Biopsy Forceps from every abnormal colposcopic examination as previously published by our team.[3]

To perform complete colposcopic examination, the TZ should be fully visible and ectocervical (Type 1 TZ), while a TZ that is partially or completely endocervical but is fully visible is a Type 2 TZ. A TZ that is partially or completely endocervical but is not fully visible is a Type 3 TZ.[10]

In both groups, women with positive findings were advised to continue follow-up care after proper management protocols of our institution [312] and they were scheduled for follow-up examination at 3, 6, and 12 months thereafter until negative tests. Data entry was done using Epi Info software version.

Outcomes of the study

The outcome was the diagnostic indices and doctor satisfaction of small caliber office cervicoscopy versus stationary colposcopy for detection of both ectocervical and endocervical cervical lesions.

Statistical analysis

Data collection and statistical analyses were performed using Excel (Microsoft, Redmond, USA) and the statistical package for the social sciences (Windows version 15.0; SPSS Inc., Chicago, IL, USA). Quality control was done at the stages of coding and data entry. Data were presented using descriptive statistics in the form of frequencies and percentages for qualitative variables and ranges, means and standard deviations, medians, and quartiles for quantitative variables in box plots. The comparability of baseline characteristics according to the outcome was ascertained by Student's t-test (unpaired t-test) for continuous variables and Mann–Whitney U-test when appropriate, and W2 test for categorical variables. To evaluate the univariate relation between variables, Pearson correlation coefficient was calculated. Values were considered statistically significant if P ≤ 0.05. In the aim of evaluating and comparing the diagnostic accuracy of UNEE, cytology, and colposcopy, office hysteroscopy and biopsy, the sensitivity and specificity tests were applied for each. Furthermore, the predictive values of positive and negative results were calculated.

RESULTS

This study comprised 112 women with clinically suspicious cervix subjected to either office cervicoscopy (Group A) or stationary colposcopy (Group B). There was no statistically significant difference between both groups regarding parity, previous abortion, age at marriage, duration of marriage, and age at menarche and menopause [Table 1]. However, women of the Group A reported less regular menstrual pattern than Group B which was statistically significant (P = 0.020). There was no statistically significant difference between both groups regarding risk factors for preinvasive cervical lesions. On UNEE of the studied women, ectopy was the most common finding, and there were no statistically significant differences between both groups as regard cervical ectopy, cervical ectropion, cervical mass, nabothian cyst, cervical ulcer, hypertrophied cervix, cervical polyp, white spots, inflammatory spots, and cervical warts. Most cytology results were negative (67.3% and 67.9%), unsatisfactory (32.7% and 28.6%) in both groups, respectively. Atypical glandular cells and atypical squamous cells were diagnosed in two cases in Group B. Table 2 describes office cervicoscopic and colposcopic findings, while description of cervical biopsy in both groups is shown in Table 3. Details of histopathologic examinations are shown in Figure 2. Figure 3 shows some office cervicoscopic findings, while Figure 4 shows abnormal cervical vasculature on stationary colposcopic examination. Stationary colposcopy was unsatisfactory in 7.1% due to nonvisualization of TZ (TZ3). No pain or tolerable pain was reported during office cervicoscopy in 85.7% of patients. Cervicoscopy succeeded to visualize the whole TZ in all cases (100%). Diagnostic indices were similar in both groups apart from the finding that office cervicoscopy was more sensitive for detection of endocervical abnormalities [Table 4]. Doctors were significantly more satisfied with ectocervical examination using stationary colposcopy than office cervicoscopy (49 cases [75.3%] vs. 28 cases [50%], respectively). Doctors expressed better satisfaction with the use of stationary colposcopy due to one or more of the following: it allowed proper visualization of cervical vasculature using green filter (30 cases), more details of the cervical lesion due to controlled magnification (26 cases), free both hands of the surgeons allowed skillful maneuvers including taking cervical biopsy (16 cases), and steady view devoid of hand tremor effect (11 cases). On the other hand, doctors were more satisfied with cervicoscopy for diagnosis of endocervical lesions (47 cases, 83.9%) if compared to stationary colposcopy with endocervical speculum (31 cases, 55.3%).

Table 1

Sociodemographic and obstetrical data of the studied women

	Group A (Cervicoscope) (n=56)		Group B (Colposcope) (n=56)		P
	No.	%	No.	%
Age: (years)					0.863
Mean±SD	33.25±8.52		32.36±6.26
Range	19.0-58.0		22.0-50.0
Residence					0.815
Urban	12	21.4	11	19.6
Rural	44	78.6	45	80.4
Education					0.088
Educated	30	53.6	21	37.5
Not educated	26	46.4	35	62.5
Parity					0.155
Nullipara	3	5.4	0	0.0
Para 1-2	20	35.7	17	30.4
Para 3-4	22	39.3	31	55.4
Para 5 or more	11	19.6	8	14.3
Previous abortion					0.844
No abortion	30	53.6	33	58.9
Once	14	25.0	12	21.4
Two or more	12	21.4	11	19.6
Age at marriage: (years)					0.911
<18	14	25.0	16	28.6
18-20	27	48.2	26	46.4
>20	15	26.8	14	25.0
Mean±SD	19.12±2.74		19.21±2.79		0.888
Duration of marriage: (years)					0.974
Mean±SD	13.72±9.37		12.77±6.83
Range	1.5-43.0		1.0-31.0
Age at menarche: (years)					0.889
Mean±SD	12.48±1.33		12.36±1.07
Range	11.0-16.0		11.0-16.0
Menstrual pattern:					0.020*
Regular	35	62.5	46	82.1
Irregular	21	37.5	10	17.9
Menopause:					0.364
Yes	4	7.1	1	1.8
No	52	92.9	55	98.2

*Statistically significant

Table 2

Office cervicoscopic and colposcopic findings in both groups

	Group A (Office cervicoscope) (n=56)		Group B (Colposcope) (n=56)		P
	No.	%	No.	%
Examination feasibility
Suspicious abnormalities	8	14.3	6	10.7	0.568
Inflammatory	48	85.7	46	82.2	0.607
Unsatisfactory	0	0.0	4	7.1	0.118
Types of suspicious abnormalities
Abnormal vasculature	1	1.7	12*	21.4	0.001**
Punctuation	1	12.5	1	16.7	1.000
Mosaic appearance	1	12.5	1	16.7	1.000
Acetowhite areas	3	37.5	2	33.3	1.000
Schiller’s iodine					0.067
Positive	9	16.1	3	5.4
Negative	47	83.9	53	94.6
Accessibility of squamocolumnar junction (satisfactory)	56	100%	48	86%	0.011
Accessibility of endometrial cavity					-
Accessible	49	87.5	-	-
Not accessible due to internal os stenosis	7	12.5	-	-
Pain during procedure
Painful#	8	14.3	7	12.5%	0.481
No pain	26	46.4	43	76.7%	0.001*
Tolerable	22	39.3	6	10.7%	0.000*

#Pain at colposcopy is attributed to the use of endocervical speculum. *Highly significant. **With green filter

Table 3

Cervical biopsy in both groups

	Group A (Cervicoscope) (n=56)		Group B (Colposcope) (n=56)		P
	No.	%	No.	%
No. of biopsy samples					0.560
Single	36	64.3	33	58.9
Multiple	20	35.7	23	41.1
Bleeding after biopsy					0.654
No bleeding	14	25.0	12	21.4
Bleeding	42	75.0	44	78.6
Bleeding control
Cautery with diathermy	9	21.4	7	15.9	0.911
Compression	10	23.8	18	40.9	0.091
Compression and diathermy	2	4.8	3	6.8	0.684
Compression and Trichloroacetic acid (11)	1	2.4	5	11.4	0.203
Trichloroacetic acid (11)	20	47.6	11	25.0	0.029*

*Statistically significant

Figure 2

Histopathology of biopsy specimen

Figure 3

Office cervicoscopic appearance of immature metaplasia (a) and condyloma acuminata (b)

Figure 4

Abnormal stationary colposcopic examination of the cervical vasculature using the green filter

Table 4

Diagnostic indices of different diagnostic tools compared to the histopathology

	Office cervicoscope	Stationary colposcope	Pap smear	P
Sensitivity	75	60	50	0.031*
Specificity	90.38	94.12	100.0	0.288
+PV	37.5	50	100.0	0.171
-PV	97.9	96	98.2	0.736
Diagnostic accuracy	89.3	91.1	98.2	0.412
AUC (area under curve)	0.827	0.771	0.750	0.315

*Statistically significant due to relatively high unsatisfactory colposcopy (7.1%)

DISCUSSION

Over the past 50 years, routine use of the Pap test to screen for cervical cancer has reduced deaths from the disease by ≥70%. Limitations include false-negative rate of 15%–20%, inadequate sampling of the TZ, poor collection and fixation of the specimen, inclusion of excessive blood, inflammatory material, or necrotic material that can obscure or preclude the correct cytopathologic diagnosis and deficient cytologists and well-equipped laboratories.[13] In this study, Pap test was unsatisfactory in 18 (32.7%) and 16 (28.6%) in both groups, respectively. This means that about 1/3 of cases of cervical malignancy can be missed by Pap test. In a recent study, Pap test was unsatisfactory in 38% of cases.[14] Moreover, many clinicians encounter cervical lesions that may or may not be associated with cytologic abnormalities.[4615] The guidelines for the management of abnormal cervical cancer screening tests and cancer precursors [16] apply only to women undergoing routine screening with adequate visualization of the cervix and directed sampling with acceptable collection instruments. Unsatisfactory cytologic or colposcopic examinations represent real obstacles against proper screening. The 2011 IFCPC terminology can improve the diagnostic accuracy for all lesion severities. The categorization of major changes and minor changes is appropriate. However, colposcopic diagnosis remains unsatisfactory.[17] Hence, adding cervicoscopy to the protocol of management would erase all cases of unsatisfactory colposcopy (TZ 2 and TZ 3).

This study aims to test if using cervicoscopy would overcome the problem of unsatisfactory colposcopy in women with suspicious cervix.

At our institution, due to financial restrictions as well as technical difficulties accomplished with Pap test screening, we focus on high-risk women since 1998. One of these groups is suspicious cervix which includes one or more of the following lesions; white or red patches, ectopy, polyps, nodular cervix with retention cyst, hypertrophied cervix, ulcer, purulent or persistent discharge, and bleeding on touch or postcoital bleeding.[3] This concept is supported by the 2012 risk-based guidelines [18] based on big data collected from 1.4 million women and classified women into high risk and low risk. Despite we follow these guidelines at our institution, yet it lakes any algorithm or plan of management of cases of unsatisfactory colposcopy in both low-risk and high-risk patients. A relatively uncommon problem of colposcopy is unsatisfactory examination of the endocervical canal, especially in nulliparous and postmenopausal women. To overcome this problem, we usually use an endocervical speculum which adds little value, particularly in nulliparous women with cervical stenosis. Others [1920] perform endocervical curettage which is a blind painful technique associated with excessive bleeding and high percentage of false-negative results. Here comes the valuable role of office cervicoscopy as an excellent tool to meticulously evaluate the endocervical canal. Despite being poorly reported in the literature,[8] it allows excellent view of the mucosa of whole endocervical canal and the internal OS. In this study, cervicoscopy could evaluate the TZ in all cases as it allows excellent view of the original (native) and new squamocolumnar junction (SCJ) almost in every case like others,[21] who reported nonvisualization of the SCJ in only 4 patients (3.4%) because of external uterine orifice stenosis caused by previous cold knife conization. One of the advantages of this study is the use of office hysteroscope (2.6 mm telescope) as a cervicoscope helps overcome cervical stenosis in most of cases because of the advantage of under vision negotiation of the cervix. In a previous study,[22] we found that cervicoscopy was a simple, cheap, and office procedure that can be used as a quick screening tool of human papillomavirus (HPV) infection in women with suspicious cervix. It had the advantage of screening of the rest of the genital tract. In the absence of laboratory HPV testing, performing cervicoscopy in conjunction with cytology screening of high-risk cases were encouraged.

In this study, office cervicoscopy succeeded to diagnose abnormal and inflammatory lesions in 14.3% and 85.7%, respectively, with no case of unsatisfactory examination. Contrarily, stationary colposcopy was unsatisfactory in 7.1%. This high sensitivity of office cervicoscopy can be attributed to the ability of the office telescope to freely examine the whole cervix and go inside the endocervical canal unlike colposcopic lens which is fixed at a far distance from the cervix with the possibility of magnification again away from the cervix. In a previous study,[23] cervicoscopic examination revealed 7.8% of cervical intraepithelial neoplasia 2–3 in low-grade intraepithelial lesion patients with inadequate or negative colposcopy. On the other hand, colposcopy was unsatisfactory in 30% of women at 12 months after large loop excision of the TZ (LLETZ),[23] in 34.3% in another study,[24] and in 30% of nonpregnant women with abnormal cervical cytology.[25]

An additional valuable advantage of this study is the use of a 2.6-mm telescope rather than using conventional 4-mm telescope is to minimize pain during the examination of the endocervical canal. We reported no pain or tolerable pain during office cervicoscopy in 85.7% of patients. Using office endoscope has a high success rate and a low complication rate, even when performed by a group of gynecologists with limited experience in the procedure. Compared to the conventional 4-mm hysteroscopy, it expresses greater diagnostic accuracy and lower cost. In one study,[26] the mean charges, excluding professional fees, for the hospital were $1799 versus $62 for conventional and office hysteroscopy, respectively.

The results of our study indicated that ≥60% of patient complained of vaginal discharge followed by backache in about 60% and dyspareunia in about 55% in both groups which is in agreement with our previous study.[6] On UNEE of the cervix, cervical ectopy was the most common suspicious lesion which should not be ignored during routine speculum examination. In addition to recurrent cervicitis and postcoital bleeding, cervical ectopy is a precursor of cervical metaplasia which means change of columnar epithelium to squamous epithelium (cellular activity) with a possibility of abnormal cellular pathway to cellular dysplasia at the most dangerous part of the cervix (TZ).

A senior examiner (AD) has an experience with colposcopy for ≥31 years. He and other authors expressed more satisfaction with stationary colposcopy due to better visualization of cervical vasculature using green filter, more details of the cervical lesion due to controlled magnification, free both hands of the surgeons allowed delicate maneuvers including taking cervical biopsy, and steady view devoid of hand tremor effect. These marvelous advantages particularly more magnification and better description of the blood vessels and gland openings would favor stationary colposcopy as a gold standard tool for accurate examination of the cervix.

In this study, we performed cervical biopsy in all cases as histopathology is the cornerstone for diagnosing the nature of cervical lesions. Since a long time, it has been recommended that at colposcopy, two or more biopsies should be taken.[25] Endocervicoscopy was successfully performed on 77 women having a Type 3 TZ with orientated punch biopsy and consequent LLETZ.[27]

This study was a prospective RCT design including consecutive patients referred according to well-defined national guidelines that resulted in a low risk of selection bias. Moreover, the patient evaluation including was uniformly structured. As all histology was revised, the risk of misclassification and random error were probably low. Single or multiple biopsies were taken from all patients, not just those that had abnormal Pap test results this prevent missing any precancerous lesions. However, the main limitations of this study could be focusing on only one high-risk group (suspicious cervix) and relatively low sample size.

In modern practice, office hysteroscopy becomes more popular than colposcopy due to expanding indications of its use. Office hysteroscopy is now available in almost all hospitals and gynecologic clinics. Most doctors do not buy colposcope but have office hysteroscope as an important diagnostic tool for abnormal uterine bleeding, recurrent pregnancy loss, and infertility. In this study, we tested whether office cervicoscopy can replace colposcopy for accurate diagnosis of premalignant cervical lesions, but we failed to support this assumption due to examiners' dissatisfaction.

From this study, it is concluded that office cervicoscopy is a good complementary but not alternative tool added to stationary colposcopy for detection of cervical lesions in cases with suspicious cervix as an example of high-risk group for cervical cancer. Due to its small caliber, cervicoscopy offers a better evaluation of the endocervical canal, especially in cases of unsatisfactory colposcopy (TZ 2 and TZ 3), and possible examination of the endometrial cavity.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.