Material and Method: Out of 37884 cases which were evaluated between 2005-2009 in our laboratory, cases interpreted as LSIL, LSIL/ ASC-H, HSIL and ASC-H were reevaluated and 153 cases for which biopsy materials were available were selected.
Results: The rate of histological CIN2 or worse associated with LSIL/ ASC-H (45%) was between the rates of LSIL (10%) and HSIL (65%), but not significantly different from ASC-H (50%). However, LSIL/ ASC-H was more frequently associated with a definitive histological diagnosis of any CIN2 than ASC-H (30% vs. 8%).
Conclusion: Based on our results, we recommend LSIL/ASC-H to be added to Bethesda System, and Pap test cases of LSIL/ASC-H may need to be clinically followed-up in a manner similar to ASC-H, i.e., with colposcopy for all patients.
A LSIL/ASC-H diagnosis was made in smears with ASCH changes among an LSIL appearance (Figure 1). The following cytomorphological results were used in defining ASC-H cells, based on previously reported studies[9-11]; (1) the presence of usually 5 or less dysplastic cells with HSIL-like findings, (2) the presence of atypical metaplastic squamous cells, (3) the presence of cells with an N/C ratio between LSIL and HSIL (Figure 2).
Figure 1: LSIL/ASC-H (Pap; x200).
The biopsy result distribution between the groups was analyzed by using only SIL cases (a total of 153) to calculate the positive predictive value (PPV).
Table I: Distribution of cervical atypical cytologies [n (%)]
The total rate of atypical cytology as diagnosed by a conventional Pap smear was 1.69% (657 cases).
The percentages of preinvasive cervical neoplasms and invasive neoplasms were 1.63% and 0.06%, respectively.
There were a total of 371 (56%) cases that had received a diagnosis of atypical cytology by Pap smear with a histopathological lesion on biopsy. These cases were distributed as CIN 1 in 47% (175 cases), CIN 2 in 28% (103 cases), CIN 3 in 19% (80 cases) and invasive cancer in 6% (23 cases) (Table II).
Table II: Histopathological correlation following atypical cytology
We found that the percentage of cases which underwent a biopsy within a year after Pap smear was 17.5% (10 cases) for LSIL, 48% (12) for ASC-H, 50% (20 cases) for LSIL/ ASC-H and 74% (23 cases) for HSIL.
1. A CIN2 or higher lesion had been found in 45% of LSIL/
ASC-H cases (PPV=45%). This rate was higher than for
LSIL cases (10%, PPV=10%), similar to ASC-H cases
(50%, PPV=50%) and lower than for HSIL cases (65%,
PPV=65%) (Table III).
2. Classifying high-grade intraepithelial neoplasms as
moderate (CIN2) or severe (CIN3) and higher, we found
a CIN3 or higher lesion in 15% of biopsies from LSIL/
ASC-H cases. This rate was similar to LSIL cases (10%)
and lower than in ASC-H (42%) and HSIL (52%) cases.
3. The rate of CIN2 in the biopsies was 30% in the LSIL/
ASC-H group. This rate was higher than in LSIL (0%),
ASC-H (8%) and HSIL (13%).
4. The rate of CIN1 in the biopsies was 35% for LSIL/ASCH
cases. This rate was higher than in the LSIL group (70%),
similar to the ASC-H group (33%) and higher than in
the HSIL group (10%).
5. We found no significant difference between the benign
diagnosis rates in cases with an LSIL/ASC-H diagnosis
(20%) and the other groups (LSIL 20%, ASC-H 17%,
HSIL 26%).
Table III: Histopathological distribution of LSIL, ASC-H, LSIL/ASC-H, and HSIL diagnoses
Our results showed that CIN2 rate was higher in the biopsies of the LSIL/ASC-H group compared to the ASC-H and HSIL groups; CIN3 rate was lower than the ASC-H and HSIL groups, and CIN2 or higher lesion rate was similar to the ASC-H group but lower than the HSIL group. The rate of CIN1 was lower than in the LSIL group.
Some smears show findings similar to LSIL and HSIL but do not meet all criteria for the diagnosis. The Bethesda 2001 system recommends the two diagnostic categories of ASCUS and ASC-H for such cases. However, there are cases that are difficult to classify as LSIL or HSIL as well. For example, some cases have marked LSIL findings while a few cells have HSIL-like features[3,5-7]. There is no diagnostic category in the abbreviated version of Bethesda 2001 for such cases but the Bethesda 2001 manual mentions that such cases can be evaluated as ‘indeterminate grade'[1]. Laboratories, including our own, define these lesions as LSIL/ASC-H or LSIL-H as they contain both LSIL and ASC-H features.
We compared the ASCUS, LSIL, ASC-H, LSIL/ASC-H, and HSIL diagnosis rates according to conventional Pap smear results and the percentage of finding a lesion in the biopsy results of patients with a diagnosis of LSIL, ASC-H, LSIL/ ASC-H, and HSIL in their smear results, as compared to the literature.
Our total cytological abnormality rate was 1.69% and this was similar to another study from Turkey on the prevalence of cervical cytological abnormalities (1.8%)[12].
Our LSIL/ASC-H ratio (0.10%) was close to that of other authors (0.15-0.53%)[3,5,6].
The rate of diagnosing a CIN2 or higher grade lesion on biopsy following an LSIL/ASC-H diagnosis (45%) was similar to ASC-H (50%), higher than for LSIL (10%) and lower than for HSIL (65%). These results were consistent with recent publications. The potential of LSIL/ASC-H to catch a CIN2 or higher lesion (45%) was especially consistent with some previous studies[3,6,16,17]. The rate of finding a CIN2 or higher lesion in LSIL/ASC-H biopsies was slightly higher than ASC-H in some studies[4,6,7-18] and slightly lower in others[2,3,5,19].
We found a 15% rate for LSIL/ASC-H to define a CIN3 or higher lesion in biopsy results, similar to Difurio et al.[2] and Alsharif et al.[8] (Difurio et al. 14%, Alsharif et al. 11%). This rate was markedly lower than with ASC-H (42%) and HSIL (52%).
The CIN2 detection rate of patients with an LSIL/ ASC-H diagnosis was 30% on follow-up. This rate was higher than with LSIL (0%), ASC-H (8%) and HSIL (13%). Hang et al.[18] and Alsharif et al[8]. have also reported that biopsies of patients diagnosed with LSIL/ASC-H have a higher rate of CIN2 diagnosis when found to have a CIN2 or higher lesion and LSIL/ASC-H probably indicates CIN2 as also reported by these authors.
The rate of receiving a diagnosis of CIN1 was 30% in the LSIL/ASC-H group, 70% for LSIL, 33% for ASC-H and 10% for HSIL. We found the rate of determining the CIN1 diagnosis for the LSIL/ASC-H group to be markedly lower than LSIL. Our results were consistent with previous studies (33.73%).
We obtained similar values between ASC-H, and LSIL and HSIL as regards determining CIN2 or higher lesions of LSIL-ASC-H. Past studies have reported similar results[2,8,16].
Bethesda 2001 contains two definite squamous intraepithelial lesions (LSIL and HSIL) and two indeterminate squamous intraepithelial lesions (ASCUS and ASC-H). The definite cytological findings of LSIL and HSIL have been defined but there is no separate category for cases that are between LSIL and HSIL and especially those that contain HSIL-like cells that are not diagnostic for HSIL in addition to marked LSIL findings.
Some cases meet the criteria for both LSIL and ASC-H. Such intermediate cases are reported as LSIL, ASC-H or LSIL/ASC-H (LSIL-H). However, these definitions lead to confusion in practice for the clinicians. The rate of patients with a LSIL/ASC-H diagnosis varies greatly in the literature (30%-72%). The reason is the inability of the clinician to fully understand these definitions. This rate was 50% in our study. Colposcopy is advised for all cases that receive a diagnosis of ASC-H according to Bethesda 2001 as they are associated with HSIL. There is no defined clinical approach for LSIL/ASC-H and clinicians act according to their own interpretation. They adopt an approach similar to LSIL especially for postmenopausal or adolescent cases. However, the rate of detecting CIN2 is significantly higher in LSIL/ASC-H than with LSIL, ASC-H, HSIL and it therefore needs to be evaluated separately from LSIL and these patients need to be approached as in ASC-H considering it can possibly indicate a high-grade squamous intraepithelial neoplasm.
In conclusion, we believe that LSIL/ASC-H should be classified as a separate category as there is a high chance of defining the underlying CIN2 and that it should therefore be evaluated separately from LSIL. We found that LSIL/ ASC-H had a high risk of a CIN2 or higher neoplasia like ASC-H but that it had low risk for CIN3 in contrast. The clinical management of CIN2 and CIN3 is similar and the clinical approach to LSIL/ASC-H should be undertaken keeping in mind that it can act like ASC-H.
ACKNOWLEDGEMENTS
We would like to thank Sadiye Toraman and Selma Porsuk
for helping us during this study.
1) Solomon D, Davey D, Kurman R, Moriarty A, O'Connor D,
Prey M, Raab S, Sherman M, Wilbur D, Wright T Jr, Young N:
The 2001 Bethesda System: terminology for reporting results of
cervical cytology. JAMA 2002, 287:2114-2119 [ Özet ]
2) Difurio MJ, Mailhiot T, Sundborg MJ, Nauschuetz KK:
Comparison of the clinical significance of the Papanicolaou test
interpretations LSIL cannot rule out HSIL and ASC-H. Diagn
Cytopathol 2010, 38:313-317 [ Özet ]
3) Power P, Gregoire J, Duggan M, Nation J: Low-grade pap smears
containing occasional high-grade cells as a predictor of highgrade
dysplasia. J Obstet Gynaecol Canada 2006, 28:884-887 [ Özet ]
4) Howard MT, Cohen DW, Underwood DL, Deeds DA, Booth
CN:Utilization of ASC-H, LSIL, cannot exclude HSIL (LSIL-H)
and HSIL terminology following implementation of the Thin-
Prep imaging system. Mod Pathol 2007, 20(Suppl 2):71A
5) Elsheikh TM, Kirkpatrick JL, Wu HH: The significance of “lowgrade
squamous intraepithelial lesion, cannot exclude highgrade
squamous intraepithelial lesion” as a distinct squamous
abnormality category in Papanicolaou tests. Cancer 2006,
108:277-281 [ Özet ]
6) Owens CL, Moats DR, Burroughs FH, Gustafson KS: “Lowgrade
squamous intraepithelial lesion, cannot exclude high-grade
squamous intraepithelial lesion” is a distinct cytologic category:
histologic outcomes and HPV prevalence. Am J Clin Pathol
2007,128:398-403 [ Özet ]
7) Shidham VB, Kumar N, Narayan R, Brotzman GL: Should LSIL
with ASC-H (LSIL-H) in cervical smears be an independent
category? A study on SurePath specimens with review of literature.
CytoJournal 2007,4:7 [ Özet ]
8) Alsharif M, Kjeldahl K, Curran C, Miller S, Gulbahce HE,
Pambuccian SE: Clinical significance of the diagnosis of lowgrade
squamous intraepithelial lesion, cannot exclude high-grade
squamous intraepithelial lesion. Cancer Cytopathol 2009,117:
92-100 [ Özet ]
9) Kir G, Cetiner H, Gurbuz A, Karateke A: Reporting of “LSIL
with ASC-H” on cervicovaginal smears: is it a valid category to
predict cases with HSIL follow-up? Eur J Gynecol Oncol 2004, 25:
462-464 [ Özet ]
10) Chivukula M, Shidham VB: ASC-H in Pap test--definitive
categorization of cytomorphological spectrum. Cytojournal 2006,
3:14 [ Özet ]
11) Hunter C, Duggan MA, Duan Q, Power P, Gregoire J, Nation J:
Cytology and outcome of LSIL: cannot exclude HSIL compared to
ASC-H. Cytopathology 2009, 20:17-26 [ Özet ]
12) Turkish Cervical Cancer and Cervical Cytology Research Group:
Prevalence of cervical cytological abnormalities in Turkey. Int J
Gynaecol Obstet 2009, 106:206-209 [ Özet ]
13) Solomon D, Nayar R: The Bethesda System for Reporting
Cervical Cytology: Definitions, Criteria, and Explanatory Notes.
2nd ed. New York, Springer, 2004, 38-52
14) McGrath CM, Kurtis JD, Yu GH: Evaluation of mild-to-moderate
dysplasia on cervical-endocervical (Pap) smear: a subgroup of
patients who bridge LSIL and HSIL. Diagn Cytopathol 2000,
23:245-248 [ Özet ]
15) Deeds D, Underwood DL, Booth CN: Comparison of Pap test fiveyear
retrospective reviews for high grade squamous intraepithelial
lesion (HSIL) and low grade squamous intraepithelial lesion,
cannot exclude HSIL (LSIL-H). Cancer (Cancer Cytopathol)
2007, 111 (suppl 5): 447-448
16) Booth C, Massarani-Wafai R, Salhadar A, Weirick J, Wojcik
EM: Is LSIL, cannot exclude HSIL (LGHSIL) a valid Pap test
interpretation. Mod Pathol 2005,18:61A
17) Jain R, Paedigo MA,Powers CN, Frable WJ: Significance of
reporting ‘LSIL-H' on cervical cytologic specimens. Cancer
Cytopathol 2005, 105:355-356