Speculoscopy As Pap Smear Adjunct

Cancer Research and Treatment 2003;35(6):541-548 Speculoscopy as Pap Smear Adjunct

Choong-Hak Park, M.D., Ph.D.

Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, College of Medicine, Dankook University, Cheonan, Korea

Purpose: Speculoscopy, or magnified chemilumines- culoscopy diagnoses of negative, 9 (8.0%) of suspicious, cent examination (MCE), is a new visual method for the and 17 (15.0%) of positive. Of the 38 histologic diagnoses, detection of cervical neoplasia. It utilizes low magnifica- there were 19 (50.0%) diagnoses of negative, 7 (18.4%) of tion and a special “blue-white” chemiluminescent light. LSIL, 1 (2.6%) of HSIL, 10 (26.3%) of SCC, and 1 (2.6%) The purpose of this study was to evaluate the screening of ACC. Pap smear showed sensitivity of 77.8% and effectiveness of speculoscopy combined with the con- specificity of 81.3%, whereas speculoscopy showed 94.1% ventional Papanicolaou smear, as compared with the and 100.0%, respectively, and Pap smear combined with latter alone, through a clinical trial in Korea. speculoscopy showed 100.0% and 81.3%, respectively. Materials and Methods: This prospective, randomized, Conclusion: Speculoscopy showed a higher sensitivity university hospital-based clinical study was performed in rate than Pap smear as a screening test, although the sam - the outpatient clinic of the Department of Obstetrics and ple size was small. Speculoscopy combined with Pap Gynecology at Dankook University Medical Center from smear is thought to be a very effective method for detec- December 1, 2001 to June 30, 2002. Of the 113 patients ting cervical neoplasia. (Cancer Research and Treat- aged 19∼74 years who had undergone both conventional ment 2003;35:541-548) Papanicolaou cervical cytologic test and speculoscopy, ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ 38 cases underwent histologic diagnoses by colposcopy- Key Words: Speculoscopy, Papanicolaou smear, Cervi- directed biopsy. cal neoplasm Results: Of the 113 patients, there were 87 (77.0%) spe-

2,327 cases of carcinoma in situ (CIS) of the cervix, resulting INTRODUCTION in about 900 deaths (3,4). Since Dr. George Papanicolaou introduced this test for cervical cancer in 1939 (5), the mortality rate for cervical Cervical cancer is the seventh most common cancer overall, squamous cell carcinoma in the United States decreased by 70∼ and the third most common in women, in whom it comprises 75% between 1955 and 1992. Today over 50 million tests are 9.8% of all cancers-371,200 new cases per year. In general conducted annually in the United States (6). Unfortunately terms, it is much more common in developing countries, where however, this method contains numerous inherent opportunities 78% of cases occur and where cervical cancer accounts for 15% for error leading to an acceptably high false negative rate. of female cancers, with a lifetime risk of about 3%. In Many researchers have reported a false negative rate from 6% developed countries, on the other hand, it accounts for only to 55% (7). The National Institute of Health reported that 4.4% of new cancers, with a lifetime risk of 1.1% (1). In the roughly half of all women who develop cervical cancer each United States, there are about 12,800 cases of carcinoma, year have had a recent Pap test (8). And in 1999, the American resulting in about 4,600 deaths, annually (2). In the Republic College of Obstetricians and Gynecologists reported that a of Korea in 2,000, there were 3,803 cases of cervix cancer, and single Pap test may miss abnormal cells in nearly half of the women who have abnormal cell growth (9). Hence, many Correspondence: Choong-Hak Park, Division of Gynecologic On- efforts have been made in order to increase the sensitivity and cology, Department of Obstetrics and Gynecology, College of decease the false negative rate of Papanicolaou smear cytology. Medicine, Dankook University, San-29 Anseo-dong, Cheonan Recently, several adjuncts to Papanicolaou smear cytology have 330-714, Korea. (Tel) 041-550-3941, (Fax) 041- 555-6335, been developed and used in the world. (E-mail) [email protected] Speculoscopy is a simple, fast and painless visual examina- Received August 5, 2003, Accepted October 23, 2003 tion of the cervix and vaginal canal that is performed in con- This study was presented at “The 88th Annual Congress of the junction with a routine Papanicolaou cervical cytologic test and Korean Society of Obstetrics and Gynecology” and at “The Inter- produces immediate results. In 1984, the Lonky brothers- M.D.s national Symposium on the New Screening Method for Cervical Stewart and Neal, and Martin who holds a Ph.D. in physics- Cancer Detection” in 2002. first conceived of the idea to use chemical light for anatomic

541 542 Cancer Research and Treatment 2003;35(6) illumination. Following several years of research, the brothers and therefore the patient group had a high prevalence of developed the Speculite, a chemical-based, disposable light source cervical abnormality. All of these patients had undergone both (10). The Speculite received FDA 510 (k) clearance as a conventional Papanicolaou cervical cytologic test and speculo- non-toxic source of illumination in November, 1986. Speculo- scopy. Colposcopy and histology, consisting of colposcopy- scopy is an in-vivo test that is performed by medical practi- directed exocervical biopsy, were conducted. Specific exclusion tioners using the Speculite (SpeculiteⓇ, The Trylon Corpora- criteria consisted of patients under 18 years of age, pregnant tion, Torrance, CA), a special “blue-white” chemiluminescent women, women less than 6 weeks post-partum, patients who disposable light, and low level magnification to visually ex- had active menstruation or active vaginal bleeding, and patients amine the cervix and vaginal wall. The Speculite produces a without cervix. diffuse, glare-free light that illuminates the cervix from all 2) Specimen collection for conventional Papanicolaou sides, allowing the clinician to view tissue at and below the smear cytology surface of the cervix. Normal cervical tissue absorbs the very precise spectrum of light produced by the Speculite and appears The cervical sample was taken in the usual way by the dark blue or purple. Diseased and abnormal cells contain physician using a combination of plastic spatula and endo- enlarged nuclei which reflect the light produced by the Spe- cervical brush (cytobrush). Following collection according to culite and appear bright white, immediately alerting medical the guidelines, the collection device was spread on the surface professionals that further diagnostic testing is necessary, even of the slide as evenly and thinly as possible (Fig. 1). if the Pap test results are normal. The Speculite has no special Immediately the slide was fixed in 95% ethyl alcohol solution. disposal requirements and will naturally lose its luminescence The slide and a case report form containing the patient's in about 15 minutes. The “blue-white” chemiluminescent dispos- identification number, initials and medical history were sent to able light does not emit any heat or electrical energy, is non- the pathology laboratory. toxic, and can be safely placed adjacent to the cervix (11). 3) Papanicolaou cytology slide preparation and evalu- The PapSure exam, a combination of Papanicolaou smear ation cytology and Speculoscopy, is the only direct visual cervical cancer screening method to receive clearance by the FDA, The slides were stained with the laboratory's routine Papa- USA. The PapSure procedure received FDA 510 (k) clearance nicolaou staining method, evaluated and a cytological diagnosis for all women recommended for a pelvic exam and Pap smear was determined according to The Bethesda System (TBS) in November, 1997 (11,12). criteria for each patient enrolled in the trial in the Department Many clinical studies have demonstrated that speculoscopy of Pathology at Dankook University Medical Center. In brief, or the PapSure exam is a more complete cervical cancer scree- specimens that are not within normal limits can contain changes ning method than Pap smear cytology alone (13∼17). The deemed either as benign cellular changes that are thought to purpose of this study was to evaluate the screening effecti- be caused by inflammation or reparative processes, or as cells veness of speculoscopy combined with the conventional Papa- that are consistent with an epithelial cell abnormality. The latter nicolaou smear, as compared with the conventional Papani- category includes precancerous cells that are consistent with colaou smear alone, through a clinical trial in Korea. low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL) or squamous cell car- MATERIALS AND METHODS cinoma (SCC), or cells that are atypical but cannot be categori- zed as intraepithelial lesion. These atypical cells are usually referred to by their acronyms, ASCUS (atypical squamous cells 1) Study design of undetermined significance) and AGUS (atypical glandular The study design was a prospective, randomized, university cells of undetermined significance). Specimens were classified hospital-based clinical design in which the intended-use of as unsatisfactory if they contained little or no interpretable speculoscopy was to be evaluated. This study was performed material owing to scant cellular component, poor fixation, or in the outpatient clinic of the Department of Obstetrics and marked obscuration of 75% or more of the epithelial com- Gynecology at Dankook University Medical Center from ponent by blood, inflammation, or other interfering factors. Pap December 1, 2001 to June 30, 2002. Pap smear test and spe- slides were initially screened by cytotechnologists, and all culoscopy were compared to a reference standard, the histologic slides interpreted as negative were “signed out” by the cytotec- diagnosis which consists of colposcopy-directed exocervical hnologist. All slides considered to potentially contain reactive biopsy. The main outcome measures for the trial are the rate or neoplastic materials were referred to the senior cytotechnolo- of correspondence, sensitivity, and specificity of Pap smear gist for a second opinion and finally confirmed by a path- cytology, speculoscopy, and PapSure Test relative to the ologist. histologic diagnosis. 4) Speculoscopy Two types of patient were enrolled into the trial. One group of 83 cases comprised patients representative of a routine Pap The speculoscopy examination involves a series of con- test screening population. These patients will have a low venient steps. First, the medical practitioner activates the illu- prevalence of cervical abnormality. The other group of 30 minating Speculite (SpeculiteⓇ, The Trylon Corporation), a spe- patients was a referral population enrolled at the time of colpo- cial “blue-white” chemiluminescent disposable light for vaginal scopic examination. These patients had previously had either an illumination. The SpeculiteⓇ, about two inches long and the abnormal Pap test or had been symptomatic of cervical disease, diameter of a pencil, contains non-toxic chemicals within a Choong-Hak Park：Speculoscopy as Pap Smear Adjunct 543

1 2

Fig. 1. Preferred procedure for ob- taining a Papanicolaou smear. ① Expose the entire cervix, including the trans- formation zone through the 3 speculum, as much as possible. ② Rotate Ayre type spatula 360o twice. ③ Ro- 4 tate Endocervical cytobrush 180o only. ④ Spread cells in the spatula on the upper half area of the slide, and roll the brush over the lower half area of the same slide.

Fig. 2. Speculite (a special “blue- white” chemiluminescent disposable light source) and an activated chemiluminescent speculite attached to the upper blade of the speculum.

Fig. 3. Optic (Speculoscopy, 6× 16 Magnifier) and positive diagnosis of speculoscopy. flexible outer plastic capsule and an inner vial. It is bent and of a metal or plastic speculum using a two-sided adhesive strip snapped sharply in order to break the inner capsule. Then it (Fig. 2). Next, the practitioner performs a traditional Pap smear is shaken thoroughly. These procedures result in the two by inserting the speculum and taking an appropriate Pap chemical fluids mixing to produce a chemiluminescent light sample. The cervix then is swabbed with a 5 percent acetic acid with a purplish color lasting for about 15∼20 minutes. The (vinegar) solution. After waiting about 40∼60 seconds for the activated Speculite is attached to the inside, upper dilator blade solution to take effect, the overhead room lights are dimmed 544 Cancer Research and Treatment 2003;35(6) to allow for enhanced visualization. Finally, the medical prac- On PapSure Test (Pap cytology+Speculoscopy), there were 81 titioner visually examines the cervix with a 6× magnification, (78.6%) diagnoses of negative, and 22 (21.4%) of positive monocular optic to inspect any potentially abnormal areas. The (Tables 1 and 2). Table 2 shows the correlation between presence of distinct white areas with at least one sharply cervical Papanicolaou cytologic diagnoses and speculoscopic defined border may indicate potential abnormalities, while diagnoses according to the presence of colposcopy-directed normal tissue may appear a homogeneous blue or purple color biopsy. There were 90 (79.6%) cytologic diagnoses of negative, (Fig. 3). In this study, I had added a diagnosis of ‘suspicious’ 4 (3.5%) of ASCUS, 6 (5.3%) of LSIL, 7 (6.2%) of HSIL, 4 initially. Speculoscopic diagnoses were considered positive if (3.5%) of SCC, 1 (0.9%) of AGUS, and 1 (0.9%) of adeno- there was a distinct acetowhite area, negative if there was no carcinoma (ACC) in the conventional Papanicolaou series. distinct acetowhite area, and suspicious if either the acetowhite Of the 113 women, colposcopy-directed exocervical biopsies lesion disappeared within 30∼60 seconds of being sighted, if were performed in 38 cases. There were 19 (50.0%) histologic the acetowhite lesion was equivocal to define, or if some glare diagnoses of negative, 7 (18.4%) of LSIL, 1 (2.6%) of HSIL, was seen in the endocervical canal and an endocervical neo- 10 (26.3%) of SCC, and 1 (2.6%) of ACC, in the histologic plasm was suspected. series. Ten histologic diagnoses of SCC showed 1 of negative, 1 of AGUS, 4 of HSIL, and 4 of SCC, in the cytologic diagno- 5) Colposcopic examination and exocervical biopsy sis. Seven histologic diagnoses of LSIL showed 3 of negative, At colposcopy, the appearance of the cervix and any ab- and 4 of LSIL, in the cytologic diagnosis. All 3 cases of normal morphology were systematically described and record- ASCUS cytologic diagnosis showed negative histologic diag- ed. Colposcopy-directed exocervical biopsies using punch or noses. When ASCUS and AGUS were excluded, Pap smear LEEP biopsy were performed in cases of a referral population cytology showed 13 true-negative, 4 false-negative, 3 false- and abnormal findings of the cervix on speculoscopy or visual positive, and 14 true-positive, in the histology comparison inspection. The biopsy specimen was fixed in 10% buffered study. The sensitivity of Pap smear cytology was 77.8%, and formalin, paraffin embedded, and prepared as conventional specificity was 81.3% (Table 3). histopathologic slides stained with hematoxylin and eosin. The Of the 38 patients with colposcopy-directed biopsy, there biopsies were interpreted and diagnosed using a combined were 17 (44.7%) speculoscopic diagnoses of negative, 5 (13.2%) reporting system, including TBS and CIN (cervical intra- of suspicious, and 16 (42.1%) of positive, in the speculoscopy epithelial neoplasia) designations by a pathologist. series. Five speculoscopic diagnoses of suspicious showed 3 of negative, and 2 of LSIL, in the histologic diagnosis (Table 4). 6) Data analysis When the 5 cases of suspicious speculoscopic diagnosis were The cytologic diagnoses of conventional Pap smear test and excluded, speculoscopy showed 16 true-negative, 1 false- the diagnoses of speculoscopy were compared with the histo- negative, 0 false-positive, and 16 true- positive, in the histology logic diagnoses by colposcopy-directed exocervical biopsy. Com- comparison study. The sensitivity of speculoscopy was 94.1%, parisons were performed using standard contingency table and specificity was 100.0% (Table 4). analyses. Table 5 shows the results of the PapSure test (Pap smear cytology+Speculoscopy). In the case of ASCUS or AGUS by RESULTS Pap smear cytology, PapSure diagnosis was considered as positive if the speculoscopic diagnosis was positive, as suspicious if the speculoscopic diagnosis was suspicious, and A total of 113 women were studied, with a mean age of 43.2 excluded if the speculoscopic diagnosis was negative. In the years (ranging from 19 to 74 years). The mean gravidity was case of suspicious diagnosis by speculoscopy, PapSure diag- 3.7 (from 0 to 12) and the mean parity was 2.2 (from 0 to nosis was considered as positive if Pap smear cytology was 7). Of these 113 women, there were 90 (82.6%) cases of positive. The PapSure test showed 13 (36.1%) diagnoses of negative and 19 (17.4%) of positive Papanicolaou cytology, negative, 3 (8.3%) of suspicious, and 20 (55.6%) of positive, after the 4 cases of ASCUS had been excluded. in the PapSure test series. When the 5 cases of suspicious On speculoscopy, there were 87 (77.0%) diagnoses of speculoscopic diagnosis were excluded, the PapSure test negative, 9 (8.0%) of suspicious, and 17 (15.0%) of positive. showed 13 true-negative, 0 false-negative, 3 false-positive, and

Table 1. Types of cervical cancer screening ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Cases Types ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total Negative Positive ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Pap cytology 90 (82.6%) 19 (17.4%) 109* Speculoscopy 87 (83.7%) 17 (16.3%) 104† Pap cytology+Speculoscopy (PapSure) 81 (78.6%) 22 (21.4%) 103 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ *4 cases of ASCUS were excluded, †9 cases of suspicious were excluded. Choong-Hak Park：Speculoscopy as Pap Smear Adjunct 545

Table 2. Correlation between Papanicolaou cervical cytologic diagnoses and speculoscopic diagnoses according to presence of colposcopy-directed biopsy ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Speculoscopic diagnosis Papanicolaou ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ cytologic Negative Suspicious Positive Total diagnosis ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ No Bx* Bx No Bx Bx No Bx Bx ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Negative 68 13 4 2 1 2 90 ASCUS† 12 01 00 4 LSIL‡ 12 00 03 6 HSIL§ 00 02 05 7 SCC∥ 00 00 04 4 AGUS¶ 00 00 01 1 ACC** 0 0 0 0 0 1 1 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total 70 17 4 5 1 16 113 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ *biopsy, †atypical squamous cells of undetermined significance, ‡low-grade squamous intraepithelial lesion, §high-grade squamous intraepithelial lesion, ∥squamous cell carcinoma, ¶atypical glandular cells of undetermined significance, **adenocarcinoma

Table 3. Correlation between Papanicolaou cytologic diagnoses and histologic diagnoses ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Papanicolaou Histologic diagnosis cytologic ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total diagnosis Negative LSIL* HSIL† SCC ACC ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Negative 13 3 0 1 0 17 ASCUS 3 0 0 0 0 3 AGUS 0 0 0 1 0 1 LSIL 1 4 0 0 0 5 HSIL 2 0 1 4 0 7 SCC 0 0 0 4 0 4 ACC 0 0 0 0 1 1 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total 19 7 1 10 1 38 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ *HPV (human papillomavirus) infection, flat condyloma, CIN 1, †CIN 2 and CIN 3

Table 4. Correlation between speculoscopic diagnoses and histologic diagnoses ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Histologic diagnosis Speculoscopic diagnosis ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total Negative LSIL HSIL SCC ACC ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Negative 16 1 0 0 0 17 Suspicious 3 2 0 0 0 5 Positive 0 4 1 10 1 16 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total 1971 101 38 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ 546 Cancer Research and Treatment 2003;35(6)

Table 5. Correlation between Papanicolaou cytologic diagnoses combined with speculoscopic diagnoses and histologic diagnoses ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Histologic diagnosis Pap cytology+Speculoscopy ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total (PapSure) diagnosis Negative LSIL HSIL SCC ACC ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Negative 13 0 0 0 0 13 Suspicious 1 2 0 0 0 3 Positive 3 5 1 10 1 20 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total 17 7 1 10 1 36 ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ In the case of ASCUS or AGUS by Pap smear cytology, PapSure diagnosis was considered as positive if speculoscopic diagnosis was positive, as suspicious if speculoscopic diagnosis was suspicious, and excluded if speculoscopic diagnosis was negative. In the case of suspicious diagnosis by speculoscopy, PapSure diagnosis was considered as positive if Pap smear cytology was positive.

Table 6. Correlation between different combinations and histologic diagnoses ꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚꠚ Histologic diagnosis Pap cytology+Speculoscopy ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total (%) (PapSure) diagnosis Negative LSIL HSIL SCC ACC ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Pap (-), Speculoscopy (-) 13 0 0 0 0 13 (43.3) Pap (-), Speculoscopy (+) 0 1 0 1 0 2 (6.7) Pap (+), Speculoscopy (-) 1 1 0 0 0 2 (6.7) Pap (+), Speculoscopy (+) 0 3 1 8 1 13 (43.3) ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ Total 14 5 1 9 1 30 (100.0) ꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏꠏ The cases of ASCUS, AGUS, and suspicious diagnosis were excluded.

17 true-positive, in the histology comparison study. The lesion disappeared within 30∼60 seconds of being seen, when sensitivity of the PapSure test was 100.0%, and specificity was the acetowhite lesion was equivocal to define, or when some 81.3%. glare was seen in the endocervical canal and an endocervical There were 4 different combinations in the PapSure test neoplasm was suspected. When the cases of suspicious diag- results, when the cases of ASCUS, AGUS, and suspicious nosis were excluded, the sensitivity of the PapSure test was diagnosis were excluded. The group of negative Pap cytology 100.0% and the specificity was 81.3%. When the positive and positive speculoscopy revealed 1 histologic diagnosis of PapSure diagnoses included the cases of suspicious diagnosis, LSIL, and 1 of SCC. The group of positive Pap cytology and the sensitivity of the PapSure test was 100.0% and the speci- negative speculoscopy revealed 1 histologic diagnosis of ficity was 76.5%. When the negative PapSure diagnoses negative, and 1 of LSIL (Table 6). included the cases of suspicious diagnosis, the sensitivity of the PapSure test was 89.5% and the specificity was 82.4%. The DISCUSSION sensitivity of the PapSure test ranged from 89.5% to 100.0%, and the specificity from 76.5% to 82.4%. The sensitivity of the PapSure test in any situation was higher than that of Pap smear The results of this study indicate that the addition of cytology alone. This study was a prospective, randomized, speculoscopy to routine, conventional, Papanicolaou smear university hospital-based clinical trial. The high sensitivity of cervical cytology can increase significantly the detection of 100.0% for the PapSure test seems to be due to the many cases cervical lesions, as compared with screening using Pap smear of high-grade cervical pathology in the relatively small sample. cytology alone. The data of this study are consistent with Traditional cervical screening methods are conventional Pap results of other studies of speculoscopy (13∼17). smear cytology and the pelvic examination. However, Pap The present data from this study revealed that Pap smear smear cytology alone has the limitations of relatively low sensi- cytology alone showed sensitivity of 77.8% and specificity of tivity and a high false negative rate. In February 1999, a report 81.3%, while the PapSure test showed sensitivity of 100.0% from AHCPR (the Agency for Health Care Policy and Re- and specificity of 81.3%. In this study I defined the diagnosis search) in the US admitted that estimates of the sensitivity of of speculoscopy as initially ‘suspicious' when the acetowhite conventional Pap screening are not as high as previously re- Choong-Hak Park：Speculoscopy as Pap Smear Adjunct 547 ported (18). The report mentioned that, based on the few (nuclear : cytoplasmic) ratio. These cells will not allow light studies that avoided severe biases, conventional Pap smear to pass through as well as normal cells do, and the underlying cytology showed sensitivity of 51% and specificity of 98%. vascular layer is not seen. Thus the absence of color or Furthermore, the report stated that the conventional Pap test is whitening is seen, while the normal tissue appears pink. Spe- more accurate when an HSIL threshold is used, with the goal culoscopy is a visual examination using low-energy chemilu- of detecting a high-grade lesion, than when lower thresholds, minescent light which highlights acetowhitening areas with a such as a LSIL or ASCUS, are used, with the goal of detecting sharp border between abnormal and normal epithelium, re- low or high-grade dysplasia. Inherent drawbacks in Pap smear gardless of the exfoliation from the surface epithelium of the cytology can lead to inaccurate diagnoses of cervical lesions. cervix. Pap smear cytology and speculoscopy are sensitive to Pap smear cytology is completely reliant upon exfoliated cell different properties of cervical lesions, and so these two tests samples. Errors may occur in the process of sampling, prepara- are theoretically additive from the viewpoint of cellular tion, and interpretation. Researchers have shown that sampling pathophysiology (15,16,22,24). and preparation errors form the major contribution to the high The PapSure test is the only direct, visual, cervical cancer false negative rate (19). While an average sample obtained screening method to have received clearance by the FDA. during a Pap smear test will contain between 300,000 to Clinical studies demonstrate that the PapSure test more than 500,000 cells, abnormal cells may not be present on the slide doubles the sensitivity for detecting abnormalities during in any case of the following: 1) the lesions don't exfoliate cervical screening to 92%, compared to just 41% for a Pap abnormal cells, 2) the exfoliated cells are trapped beneath a smear alone. This increase in sensitivity was accompanied by barrier, 3) the lesions exist in tissue below the surface where an expected decrease in specificity from 95% to 80% (25). If they are unavailable for exfoliation, 4) the abnormal cells may the goal of cervical screening is to identify patients with cer- simply be missed by the collecting device, or 5) the collecting vical lesions, sensitivity is more important than specificity. device retrieves an accurate sampling but it may transfer only Loiudice et al reported that the PapSure test detected 100% of normal cells to the slide (11,12,20). Some researchers reported HSIL and 91% of LSIL, compared with just 76% and 37%, that more than 80% of the cell sample is thrown away with respectively, for the Pap smear alone, in a prospective study the sampling device into the trash at the doctor's office, and of 3,300 women (16). They noted that the PapSure test that the remaining sample on the slide may not necessarily improves cervical screening sensitivity to 90% and is the most represent the condition of the cervix (21). accurate screening method available. Compared to other The data in this study revealed that speculoscopy showed cervical adjuncts, the PapSure test is simple, easy to learn, easy sensitivity of 94.1% and specificity of 100.0%, when the cases to perform, sensitive, and inexpensive. In addition, it takes less of suspicious diagnosis were excluded. When the positive than 2 minutes to perform and is conveniently done at the same speculoscopic diagnoses included the cases of suspicious time as a pelvic examination. Early detection of LSIL and HSIL diagnosis, the sensitivity of speculoscopy was 94.7% and the is very important because some lesions have a high-risk specificity was 84.2%. When the negative speculoscopic potential of developing into more severe lesions, including diagnoses included the cases of suspicious diagnosis, the invasive cancer. Furthermore, many women are surprised and sensitivity of speculoscopy was 84.2% and the specificity was concerned when they learn that Pap smear cytology is less 100.0%. The sensitivity of speculoscopy ranged from 84.2% to accurate than previously thought. 94.7%, and the specificity from 84.2% to 100.0%. When the cases of ASCUS, AGUS, and suspicious diagnosis were CONCLUSIONS excluded, Pap smear cytology detected 57% of LSIL and 91% of HSIL or worse lesions in the histologic diagnosis, and speculoscopy detected 80% of LSIL and 100% of HSIL or The data from this study demonstrate that the combination worse lesions in the histologic diagnosis. Of 10 histologic of Papanicolaou smear cytology and speculoscopy is a very diagnoses of SCC, Pap smear cytology showed 1 of negative, effective screening method for the detection of cervical neo- 1 of AGUS, 4 of HSIL, and 4 of SCC, whereas speculoscopy plasia, as compared with conventional Papanicolaou smear showed all of positive diagnosis. Speculoscopy detected 2∼3 cytology alone. As well as Papanicolaou smear cytology, spe- times as many acetowhite lesions as incandescent light in culoscopy is also a useful in-vivo test for cervical cancer clinical trials (22,23). Edwards et al reported that speculoscopy screening. detected 77% of HSIL and 84% of LSIL, compared with just 67% and 26% for the Pap smear alone (14). Wertlake et al REFERENCES reported that the addition of speculoscopy to the routine Pap smear resulted in detecting 11 of the 32 (34%) women with HSIL and 154 of the 191 (81%) women with LSIL, in a 1. Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA prospective large study of 5,692 women (15). Cancer J Clin 1999;49:33-64. 2. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statis- In-vivo adjunctive tests assess the cervix directly. Visual tests tics, 2000. CA Cancer J Clin 2000;50:7-33. such as colposcopy, speculoscopy, and cervicography evaluate 3. Central Cancer Registry and Ministry of Health and Welfare the cervix for the presence of identifiable abnormalities, and are in Korea. Annual report of the central cancer registry in Korea, not dependent on exfoliation. In these tests diseased areas of 2000, 2002. the cervix, including areas of dysplasia, have larger nuclei than 4. Bae JM, Won YJ, Jung KW, Park JG. Annual report of Korea normal cervical cells. As a result, they have a higher N/C Central Cancer Registry Program 2000: based on registered data from 131 hospitals. Cancer Res Treat 2002;34:77-83. 548 Cancer Research and Treatment 2003;35(6)

5. Papanicolaou GN, Traut HF. The diagnostic value of vaginal using a visual adjunctive procedure: a co-operative Italian smears in carcinoma of the uterus. Am J Obstet Gynecol study on speculoscopy (GISPE). Eur J Cancer Prev 1998;7:295- 1941;42:193-206. 304. 6. Grohs DH. Challenges in cervical cancer screening: what clini- 17. Chang JI, Ou CH, Wu KM, Chen CC, Cheng KC. The evalu- cians, patients and the general public need to know. ActaCytol ation of cervical cancer screening by combining Speculoscopy 1996;40:133-137. with Papanicolaou smear examination in Taiwan. Chinese Med 7. Koss LG. The Papanicolaou test for cervical cancer detection: J (Taipei) 2002;65:430-434. a triumph and a tragedy. JAMA 1989;261:737-743. 18. AHCPR (Agency for Health Care Policy and Research). Evalu- 8. Cervical Cancer. NIH Consensus Statement Online. 1996;April ation of cervical cytology. Evidence report/Technology assess- 1-3[accessed December 2001];43(I)1-38. ment. No. 5. 1999. 9. American College of Obstetricians and Gynecologists. ACOG 19. Joseph MG, Cragg F, Wright VC, Kontozoglou TE, Downing Today 1999;43:1-10. P, Marks FR. Cyto-histological correlates in a colposcopic 10. Trylon Corporation. Trylon Corporation: Background infor- clinic: a 1-year prospective study. Diagn Cytopathol 1991;7: mation. Accessed November 2000. 477-481. 11. Trylon Corporation. PapSureⓇ: Background information and 20. Felix JC, Lonky NM, Tamura K, Yu KJ, Naidu Y, Lai CR, technology description. Accessed November 2000. Available Lonky SA. Aberrant expression of E-cadherin in cervical intra- at http://www.papsure.com/what.asp. epithelial neoplasia correlates with a false negative Papanic- 12. Lonky NM. Current concepts in cervical cancer prevention. olaou smear. Am J Obstet Gynecol 2002;186:1308-1314. Symposium on the new screening method for cervical cancer 21. Tezuka F, Shuki H, Oikawa H, Higashiiwai H. Numerical detection. COEX. Inter-Continental Hotel, Seoul, Korea. Oc- counts of epithelial cells collected, smeared and lost in the con- tober 22, 2002. ventional Papanicolaou smear preparation. Acta Cytol 1995;39: 13. Mann W, Lonky N, Massad S, Scotti R, Blanco J, Vasilev S. 837-838. Papanicolaou smear screening augmented by a magnified 22. Lonky NM, Edwards G. Comparison of chemiluminescent light chemiluminescent exam. Int J Gynecol Obstet 1993;43:289-296. versus incandescent light in the visualization of acetowhite 14. Edwards G, Rutkowski C, Palmer C. Cervical cancer screening epithelium. Am J Gynecol Health 1992;6:11-15. with Papanicolaou smear plus speculoscopy by nurse prac- 23. Suneja A, Mahishee, Agarwal N, Misra K. Comparison of titioners in a health maintenance organization. J Low Gen Tract magnified chemiluminescent examination with incandescent Dis 1997;1:141-147. light examination and colposcopy for detection of cervical 15. Wertlake PT, Francus K, Newkirk GR, Parham GP. Effec- neoplasia. Indian J Cancer 1998;35:81-87. tiveness of the Papanicolaou smear and speculoscopy as com- 24. Lonky NM, Mann WJ, Massad LS, Mutch DG, Blanco JS, pared with the Papanicolaou smear alone: a community-based Vasilev SA, Finan MA, Scotti RJ. Ability of visual tests to clinical trial. Obstet Gynecol 1997;90:421-427. predict underlying cervical neoplasia: Colposcopy and Spe- 16. Loiudice L, Abbiati R, Boselli F, Cecchini G, Costa S, Grossi culoscopy. J Reprod Med 1995;40:530-536. E, Piccoli R, Villani C. Improvement of Pap smear sensitivity 25. PapSure? Full Product Information (PDF). Available at http:// www.papsure.com/product_info.asp.