- Email: [email protected]
Atypical squamous cells of undetermined significance: Management patterns at an academic medical center
Atypical squamous cells of undetermined significance: Management patterns at an academic medical center Elizabeth Suh-Burgmann, MD,a Teresa Darragh, MD,c and Karen Smith-McCune, MD, PhDa, b San Francisco, California OBJECTIVE: Our intent was to compare the management of patients with atypical squamous cells of undetermined significance on cytologic screening at an academic center to published guidelines. STUDY DESIGN: We reviewed the management of 223 atypical squamous cells of undetermined significance cervical smears. Patients with a history of dysplasia were excluded. The time interval to and nature of follow-up testing was determined, and the influence of atypical squamous cells of undetermined significance qualifiers and provider specialty analyzed. RESULTS: Initial follow-up consisted of repeat cytologic examination alone in 94% of cases. Of patients with follow-up, 29% were retested within 2 months and 68% within 4 months. No conclusive differences in management were found by qualifier type or by provider specialty. Subsequent high-grade dysplasia was found in 2.6% of patients. CONCLUSIONS: A discrepancy exists between published guidelines and actual management of patients with atypical squamous cells of undetermined significance smears at this medical center. Patients often undergo follow-up testing at shorter intervals than those suggested despite a low likelihood of finding highgrade disease. (Am J Obstet Gynecol 1998;178:991-5.)
Key words: Atypical squamous cells of undetermined significance, ASCUS, cytologic screening, dysplasia
The management of patients with low-grade cervical cytologic abnormalities has received a great deal of attention in both the professional literature and the lay press. Much of the debate has focused on the management of patients whose cytologic testing cannot be diagnosed as either definitively neoplastic or reactive—those classified in the Bethesda terminology as atypical squamous cells of undetermined significance (ASCUS).1-3 The management of these borderline smears remains controversial. The American College of Obstetricians and Gynecologists has recently published a set of recommendations for the management of patients with abnormal cervical cytologic findings. Regarding the category of ASCUS, they state that “when a patient can be relied upon to return for repeat cytologic examinations at scheduled intervals (usually no more frequent than every 6 months) she can be followed by cytology alone.” They recommend that patients with two or more such smears undergo colposcopic examination.4 However, when confronted with a patient with this cytologic diagnosis, many practitioners are not comfortable waiting 6 months or From the Department of Obstetrics, Gynecology and Reproductive Sciences,a Reproductive Endocrinology Center,b and Department of Pathology,c University of California San Francisco. Received for publication August 15, 1997; revised December 2, 1997; accepted December 10, 1997. Reprint requests: Karen Smith-McCune, Box 0546, HSW 1480, UCSF, San Francisco, CA 94143-0546. Copyright © 1998 by Mosby, Inc. 0002-9378/98 $5.00 + 0 6/1/88112
more before performing repeat testing. The National Cancer Institute published interim guidelines for the management of abnormal cervical cytologic findings in 1994 to respond to confusion in this area.5 In these guidelines the follow-up of these borderline smears could be influenced by qualifiers in the cytologic report (e.g., “favor reactive” or “favor squamous intraepithelial lesion”). For those smears suggestive of a reactive process, a screening interval of every 4 to 6 months for 2 years was advocated. The dilemma for both patients and practitioners when faced with a cervical smear showing atypical squamous cells is that it may represent an undetected high-grade lesion, in which case long intervals between follow-up visits could allow progression of disease. This is balanced against the morbidity and cost of aggressive management, as well as the desire to permit spontaneous regression to occur. Few published studies of practice patterns are available to establish the actual clinical management of these patients. We wished therefore to examine how providers within our medical center are currently managing patients who present with their first atypical smear. We conducted a retrospective review directed at the following questions: What is the interval for follow-up testing for patients after an initial cytologic diagnosis of ASCUS? How often is the patient referred immediately for colposcopy? Does management correlate with qualifiers on the cytologic reports? Are there significant differences in management according to specialty? And finally, what are the subsequent rates of dysplasia for this population?
991
992 Suh-Burgmann, Darragh, and Smith-McCune
May 1998 Am J Obstet Gynecol
Table I. Follow-up times after initial ASCUS Papanicolaou smear Interval to follow-up
Provider All By specialty† Obstetrics-gynecology General medicine Family medicine
No.
Excluded*
No. with follow-up
≤2 mo
>2 mo–≤4 mo
>4 mo–≤6 mo
>6 mo
223
42
152
44 (29%)
60 (39%)
33 (22%)
15 (10%)
107 51 18
19 10 3
75 30 11
20 (27%) 5 (17%) 3 (27%)
29 (39%) 13 (43%) 4 (36%)
22 (29%) 7 (23%) 1 (9%)
4 (5%) 5 (17%) 3 (27%)
*Excluded because of current or history of dysplasia. †Proportions of patients for each specialty returning
at ≤4 months versus >4 months were not found to be significantly different as
compared by Fisher’s exact (two-tailed) test: p = 0.65.
Material and methods Approval to conduct the study was obtained from the Committee on Human Research at the University of California San Francisco. The cytology laboratory at the University of California San Francisco receives cervical smears collected at the following clinics in the medical center: gynecology, obstetrics, gynecologic oncology, general internal medicine, family medicine, pediatrics, and surgical specialties. Providers in these clinics include attending physicians, nurse practitioners, and supervised residents and medical students. No protocols dictating follow-up are routinely used in the clinics analyzed. All cytologic reports with a diagnosis of ASCUS during a 2month period (May and June 1995) were retrieved from the laboratory database. This interval provided a sufficient sample size to yield acceptable 95% confidence intervals for the proportions expected at various follow-up times. During these 2 months, 223 cervical smears of a total of 3137 fell into this category (7.1%), a rate that is within the guidelines of being two to three times the rate of squamous intraepithelial lesions for this laboratory. All these smears underwent initial screening by a cytotechnologist and a subsequent review by a cytopathologist. Standard criteria as defined by the Bethesda system were used in determining the cytologic diagnoses.3 A computerized patient database is in use at the medical center that stores data on all clinic visits, laboratory results, pathologic and cytologic reports, radiologic reports, operative reports, discharge summaries, and demographic information for each patient. This database was surveyed for each patient found to have a cytologic diagnosis of ASCUS during the study period. All prior and subsequent cervical cytologic reports, biopsy results, clinic visit dates, and clinic specialty type were recorded. Patients with a history of dysplasia or known human immunodeficiency virus infection were excluded. Follow-up was defined as either subsequent cytologic testing retrieved from the database, colposcopy (with or without biopsy), or a plan for repeat testing documented in clinic notes available in the database. If the patient was found
to have had a prior cervical smear during the preceding 12 months showing ASCUS, this earlier smear was taken as the initial one from which follow-up was determined. Again, a previous smear or biopsy specimen documenting dysplasia excluded the patient from further analysis. The interval to follow-up was then determined for each patient and categorized as either ≤2 months, >2 months and ≤4 months, >4 months and ≤6 months, >6 months, and no follow-up found. Of the patients receiving follow-up, the proportion seen during each interval was calculated. To examine the effect of specialty type, the proportion of patients returning at <4 months was compared with those returning at ≥4 months for gynecology, general medicine, and family medicine. These proportions were compared with Fisher’s exact (two-tailed) test. The initial cytologic report was then reviewed for the presence of qualifying information: “favor benign” (including reactive changes or atrophy) or “favor squamous intraepithelial lesions.” Inflammation noted by itself, with no other qualifying statements or recommendations, was not included; however, if the cytologist’s comments clearly favored a reactive process and recommended repeat cytologic testing, this was taken as a “favor benign” designation. The follow-up intervals for patients with smears suggestive of a benign process were then compared with those whose smears favored squamous intraepithelial lesions. These results were compared for statistical significance (p < 0.05) by Fisher’s exact test. Results A total of 223 cervical smears were given the cytologic diagnosis of ASCUS during the time studied. Of these, 107 (47%) of the cervical smears were collected in the gynecology clinic, 51 (23%) in general medicine, 18 (8%) in family practice, 15 (7%) in the dysplasia clinic, 13 (6%) in gynecologic oncology, 6 (3%) in pediatric and adolescent medicine, and 1 (0.5%) in the surgery clinic. For 14 patients (9%), the exact location of the screening could not be determined. Forty-two patients (19%) were
Volume 178, Number 5 Am J Obstet Gynecol
excluded from further analysis because of a documented concurrent or prior diagnosis of squamous intraepithelial lesions including two patients who were immunocompromised. As expected, 97% of patients seen in the dysplasia and gynecologic oncology clinics were thus excluded for this reason. Of the remaining 181 patients, follow-up was available for 152 (84%). The time to follow-up after the initial smear was determined and categorized as: ≤2 months, >2 and ≤4 months, >4 and ≤6 months, and >6 months (Table I). Ninety percent of these patients (137 of 152) underwent follow-up testing at ≤6 months. For nine patients (6%), colposcopy was included in their initial follow-up visit with the remainder having repeat cytologic testing alone. Twenty-nine percent (44 of 152) of patients were seen at ≤2 months, and 68% (104 of 152) were seen at ≤4 months. When broken down by specialty, these proportions remained similar: the percentage of patients returning at ≤4 months was 65% for gynecology, 60% for general medicine, and 63% for family medicine (p = 0.65, by Fisher’s exact test). Although the small numbers in the subgroups prevented our study from having sufficient power to determine statistical significance of practice pattern by specialty, the differences observed between specialties were so small that they can be considered clinically insignificant. To determine whether qualifiers of the cytologic diagnosis correlated with time to follow-up, those smears that were reported with qualifiers suggesting a benign process (e.g., “favor reactive” [n = 52]) were compared with those reported as “favor squamous intraepithelial lesions” (n = 42) (Table II). The presence of qualifiers was not correlated with interval to follow-up; similar proportions of patients were rescreened at each of the time intervals regardless of whether the initial report favored a reactive or dysplastic process (p = 0.18, Fisher’s exact test). Moreover, referral for immediate colposcopy in this population did not depend on qualifiers; of the nine patients referred, four had smears favoring squamous intraepithelial lesions, two favored a benign process, and the remainder were unqualified. Of the total number of patients who had follow-up (194), 31 (16%) were found on subsequent testing to have biopsy-proven squamous intraepithelial lesions. However, if patients with a prior history of dysplasia are excluded, only 19 (12%) had biopsy-proven squamous intraepithelial lesions, with 4 (2.6%) having high-grade lesions. Of these, 8 of 19 had initial smears qualified by “favor squamous intraepithelial lesions,” 2 of 19 had “favor reactive,” and the remainder were unqualified. Of those with subsequent high-grade lesions, 2 of 4 had “favor squamous intraepithelial lesions” Papanicolaou smears, 1 of 4 had “favor reactive,” and 1 of 4 were unqualified. The positive predictive value of a “favor squamous intraepithelial lesions” qualifier for subsequent dysplasia was calculated to be 19%, whereas the positive
Suh-Burgmann, Darragh, and Smith-McCune 993
Table II. Follow-up patterns by Papanicolaou smear qualifiers (favor benign vs dysplastic process) Favor benign Favor dysplasia Statistical (n = 52) (n = 42) significance No follow-up (n = 16) Repeat Papanicolaou smear ≤2 mo >2 mo–≤4 mo >4 mo–≤6 mo >6 mo
11 (21%)
5 (12%)
13 (25%) 12 (23%) 14 (27%) 2 (4%)
10 (24%) 17 (40%) 6 (14%) 4 (10%)
p = 0.17*
*p Value obtained by comparison of follow-up times for “favor benign” versus “favor squamous intraepithelial lesion” with Fisher’s exact (two-tailed) test.
predictive value of “favor reactive” for benign follow-up was 96%. Although the calculated odds ratio of “favor squamous intraepithelial lesions” compared with “favor reactive” for subsequent dysplasia is 4, the positive predictive value of a “favor squamous intraepithelial lesions” qualifier alone for subsequent dysplasia was 19% and only 9% for high-grade disease. Thus although a “favor reactive” qualifier had a high predictive value for benign follow-up, a qualifier favoring squamous intraepithelial lesions was not by itself highly predictive of dysplasia. Comment The goal of the study was not to characterize the rationale behind patient management but rather to describe how practitioners are actually managing patients with an initial ASCUS diagnosis. Our results indicate that, within this medical center, providers are performing repeat smears at more frequent intervals than recommended by either The American College of Obstetricians and Gynecologists or the National Cancer Institute guidelines, with nearly one third of patients returning at 2 months or earlier. In contrast, only 10% of patients received follow-up at an interval exceeding 6 months, despite the fact that the patients studied did not have a history of dysplasia, and the incidence of underlying high-grade disease in this population was found to be very low (2%). In this study 16% of patients did not return for follow-up in the 18 months after their initial smear. It is possible that patients scheduled for longer follow-up intervals were more likely to fail their appointments, skewing the results toward shorter follow-up times. However, even if all 16% of patients were meant to return after 6 months, the proportion of patients being seen at ≤4 months is still >50%. Because the aim of the study was to describe the follow-up of patients after their first atypical smear, a recent abnormal smear done at an outside institution and not noted in the database would therefore be a potential source of error in determining
994 Suh-Burgmann, Darragh, and Smith-McCune
follow-up. However, patients seen because of an outside abnormal smear are usually referred to the dysplasia or gynecology clinics where such a history would be routinely documented in the database clinic notes. For patients receiving their screening in general medicine and family practice clinics, referral to specialty clinics occurred only when colposcopy was deemed necessary, which usually required more than one prior ASCUS Papanicolaou smear. Thus, in cases in which the initial ASCUS smear was collected in these locations, the repeat smear was performed in the same clinic. If some patients had been referred to specialty clinics for follow-up, delays caused by scheduling may have occurred. Therefore this would have resulted in an overestimate of the interval time to follow-up screening for these patients and would not negate our conclusions that patients are receiving follow-up more frequently than recommended by current guidelines. In this study the overall rate of dysplasia is 16%, which is within the range of values found in other studies.6-9 The interval to follow-up and referral for immediate colposcopy was not found to correlate with the presence of qualifiers on the cytologic report such as “favor reactive” or “favor squamous intraepithelial lesions.” The presence of a “favor squamous intraepithelial lesions” qualifier, however, was a weak predictor of the presence of squamous intraepithelial lesions. Other reports have similarly found a low rate of high-grade dysplasia among patients with “favor reactive” qualifiers.10, 11 The National Cancer Institute guidelines addressed the use of qualifiers in the management of patients with a cytologic diagnosis of ASCUS and recommended that for patients with a report favoring squamous intraepithelial lesions, colposcopy should be considered. However, several studies have found high interobserver variability and poor reproducibility for this cytologic diagnosis, placing in question the usefulness of these qualifiers.12-16 When confronted with a patient with a history of normal smears who now has a smear showing ASCUS, clinicians therefore find little objective data on which to base their management. This report constitutes a first step toward determining the current standard of care for patients with a cytologic diagnosis of ASCUS. Previous studies have used surveys to examine the rate of colposcopy and referral for patients with ASCUS smears.17-20 However, it is not clear whether these surveys allowed respondents the option of differentiating between initial versus repeat ASCUS smears and immediate versus eventual colposcopy. Moreover, the management of these patients by repeat cytologic screening alone was not specifically characterized. The results of studies such as the one described here are therefore useful in providing information about actual clinical management. There are several reasons to explain the discrepancy between the published guide-
May 1998 Am J Obstet Gynecol
lines and the actual practices we observed. Providers may be uncomfortable with the published guidelines or may simply be unaware of them. Data regarding the role of other patient risk factors, apart from a history of dysplasia, such as multiple sexual partners, smoking, and reliability were not collected in this study but clearly may influence the follow-up plan for individual patients. In addition, clinicians may be influenced by patient anxiety and desire for early follow-up testing. Further studies are warranted to both substantiate the current guidelines as written and to determine the bases for the choices practitioner’s are making in clinical practice. REFERENCES
1. National Cancer Institute Workshop. The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. JAMA 1989;262:931-4. 2. Broder S. The Bethesda System for reporting cervical/vaginal cytologic diagnoses: report of the 1991 Bethesda Workshop [rapid communication]. JAMA 1992;267:1892. 3. Kurman RJ, Solomon D. The Bethesda system for reporting cervical/vaginal cytologic diagnoses: definitions, criteria and explanatory notes for terminology and specimen adequacy. New York: Springer-Verlag; 1994. p. 30-8. 4. American College of Obstetricians and Gynecologists. Cervical cytology: evaluation and management of abnormalities. Washington: The College; 1993. p. 1-8. ACOG Technical Bulletin No.: 183. 5. Kurman RJ, Henson DE, Herbst AL, Noller KL, Schiffman MH. Interim guidelines for management of abnormal cervical cytology: the 1992 National Cancer Institute Workshop. JAMA 1994;271:1866-9. 6. Taylor RR, Guerrieri JD, Nash JD, Henry MR, O’Connor DM. Atypical cervical cytology: colposcopic follow-up using the Bethesda system. J Reprod Med 1993;38:443-7. 7. Andrews S, Hernandez E, Miyazawa K. Paired Papanicolaou smears in the evaluation of atypical squamous cells. Obstet Gynecol 1989;73:747-50. 8. Maier RC, Schultenover SJ. Evaluation of the atypical squamous cell Papanicolaou smear. Int J Gynecol Pathol 1986;5:242-8. 9. Linheim SF, Smith-Ngyuen G. Aggressive evaluation for atypical squamous cells in Papanicolaou smears. J Reprod Med 1990;35:971-3. 10. Gonzales D, Hernandez E, Anderson L, Heller P, Atkinson BF. Clinical significance of a cervical cytologic diagnosis of atypical squamous cells of undetermined significance: favoring a reactive process or low grade squamous intraepithelial lesion. J Reprod Med 1996;41:719-23. 11. Collins LC, Wang HH, Abu-Jawdeh GM. Qualifiers of atypical squamous cells of undetermined significance help in patient management. Mod Pathol 1996;9:677-81. 12. Klinkhamer PJ, Vooijs GD, de Haan AF. Intraobserver and interobserver variability in the diagnosis of epithelial abnormalities in cervical smears. Acta Cytol 1988;32:794-800. 13. Sidaway MK, Tabarra SO. Reactive change and atypical squamous cells of undetermined significance in Papanicolaou smears: a cytohistologic correlation. Diagn Cytopathol 1993;9:423-7;discussion 427-9. 14. Davey DD, Naryshkin S, Nielsen ML, Kline TS. Atypical squamous cells of undetermined significance: interlaboratory comparison and quality assurance monitors. Diagn Cytopathol 1994;11:390-6. 15. Lonky NM, Navarre GL, Saunders S, Sadeghi M, Wolde-Tsadik G. Low-grade Papanicolaou smears and the Bethesda system: a prospective cytohistopathologic analysis. Obstet Gynecol 1995;85:716-20. 16. Sherman ME, Schiffman MH, Lorincz AT, Manos MM, Scott DR,
Volume 178, Number 5 Am J Obstet Gynecol
Kuman RJ, et al. Toward objective quality assurance in cervical cytopathology: correlation of cytopathologic diagnoses with detection of high-risk human papillomavirus types. Am J Clin Pathol 1994;102:182-7. 17. Melnikow J, Sierk A, Flocke S, Peters CA. Does the system of Papanicolaou test nomenclature affect the rate of referral for colposcopy? A survey of family physicians. Arch Fam Med 1993;2:253-8.
Suh-Burgmann, Darragh, and Smith-McCune 995
18. Gordon P, Hatch K. Survey of colposcopy practices by obstetrician/gynecologists. J Reprod Med 1992;37:861-3. 19. Chmielewski MJ, Dunton CJ. Internet survey of colposcopy practices. J Low Gen Tract Dis 1997;1:151-3. 20. Wain GV, Ward J, Towler BP. Gynecologic care of women with abnormal Pap smears: How varied is current practice? Med J Aust 1995;162:348-53.
Bound volumes available to subscribers Bound volumes of the American Journal of Obstetrics and Gynecology are available to subscribers (only) for the 1998 issues from the publisher, at a cost of $116.00 for domestic, $148.73 for Canada, and $139.00 for international for vol. 178 (January-June) and Vol. 179 (July-December). Shipping charges are included. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the Journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, USA; phone (800)453-4351 or (314)453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.
Key words: Atypical squamous cells of undetermined significance, ASCUS, cytologic screening, dysplasia
The management of patients with low-grade cervical cytologic abnormalities has received a great deal of attention in both the professional literature and the lay press. Much of the debate has focused on the management of patients whose cytologic testing cannot be diagnosed as either definitively neoplastic or reactive—those classified in the Bethesda terminology as atypical squamous cells of undetermined significance (ASCUS).1-3 The management of these borderline smears remains controversial. The American College of Obstetricians and Gynecologists has recently published a set of recommendations for the management of patients with abnormal cervical cytologic findings. Regarding the category of ASCUS, they state that “when a patient can be relied upon to return for repeat cytologic examinations at scheduled intervals (usually no more frequent than every 6 months) she can be followed by cytology alone.” They recommend that patients with two or more such smears undergo colposcopic examination.4 However, when confronted with a patient with this cytologic diagnosis, many practitioners are not comfortable waiting 6 months or From the Department of Obstetrics, Gynecology and Reproductive Sciences,a Reproductive Endocrinology Center,b and Department of Pathology,c University of California San Francisco. Received for publication August 15, 1997; revised December 2, 1997; accepted December 10, 1997. Reprint requests: Karen Smith-McCune, Box 0546, HSW 1480, UCSF, San Francisco, CA 94143-0546. Copyright © 1998 by Mosby, Inc. 0002-9378/98 $5.00 + 0 6/1/88112
more before performing repeat testing. The National Cancer Institute published interim guidelines for the management of abnormal cervical cytologic findings in 1994 to respond to confusion in this area.5 In these guidelines the follow-up of these borderline smears could be influenced by qualifiers in the cytologic report (e.g., “favor reactive” or “favor squamous intraepithelial lesion”). For those smears suggestive of a reactive process, a screening interval of every 4 to 6 months for 2 years was advocated. The dilemma for both patients and practitioners when faced with a cervical smear showing atypical squamous cells is that it may represent an undetected high-grade lesion, in which case long intervals between follow-up visits could allow progression of disease. This is balanced against the morbidity and cost of aggressive management, as well as the desire to permit spontaneous regression to occur. Few published studies of practice patterns are available to establish the actual clinical management of these patients. We wished therefore to examine how providers within our medical center are currently managing patients who present with their first atypical smear. We conducted a retrospective review directed at the following questions: What is the interval for follow-up testing for patients after an initial cytologic diagnosis of ASCUS? How often is the patient referred immediately for colposcopy? Does management correlate with qualifiers on the cytologic reports? Are there significant differences in management according to specialty? And finally, what are the subsequent rates of dysplasia for this population?
991
992 Suh-Burgmann, Darragh, and Smith-McCune
May 1998 Am J Obstet Gynecol
Table I. Follow-up times after initial ASCUS Papanicolaou smear Interval to follow-up
Provider All By specialty† Obstetrics-gynecology General medicine Family medicine
No.
Excluded*
No. with follow-up
≤2 mo
>2 mo–≤4 mo
>4 mo–≤6 mo
>6 mo
223
42
152
44 (29%)
60 (39%)
33 (22%)
15 (10%)
107 51 18
19 10 3
75 30 11
20 (27%) 5 (17%) 3 (27%)
29 (39%) 13 (43%) 4 (36%)
22 (29%) 7 (23%) 1 (9%)
4 (5%) 5 (17%) 3 (27%)
*Excluded because of current or history of dysplasia. †Proportions of patients for each specialty returning
at ≤4 months versus >4 months were not found to be significantly different as
compared by Fisher’s exact (two-tailed) test: p = 0.65.
Material and methods Approval to conduct the study was obtained from the Committee on Human Research at the University of California San Francisco. The cytology laboratory at the University of California San Francisco receives cervical smears collected at the following clinics in the medical center: gynecology, obstetrics, gynecologic oncology, general internal medicine, family medicine, pediatrics, and surgical specialties. Providers in these clinics include attending physicians, nurse practitioners, and supervised residents and medical students. No protocols dictating follow-up are routinely used in the clinics analyzed. All cytologic reports with a diagnosis of ASCUS during a 2month period (May and June 1995) were retrieved from the laboratory database. This interval provided a sufficient sample size to yield acceptable 95% confidence intervals for the proportions expected at various follow-up times. During these 2 months, 223 cervical smears of a total of 3137 fell into this category (7.1%), a rate that is within the guidelines of being two to three times the rate of squamous intraepithelial lesions for this laboratory. All these smears underwent initial screening by a cytotechnologist and a subsequent review by a cytopathologist. Standard criteria as defined by the Bethesda system were used in determining the cytologic diagnoses.3 A computerized patient database is in use at the medical center that stores data on all clinic visits, laboratory results, pathologic and cytologic reports, radiologic reports, operative reports, discharge summaries, and demographic information for each patient. This database was surveyed for each patient found to have a cytologic diagnosis of ASCUS during the study period. All prior and subsequent cervical cytologic reports, biopsy results, clinic visit dates, and clinic specialty type were recorded. Patients with a history of dysplasia or known human immunodeficiency virus infection were excluded. Follow-up was defined as either subsequent cytologic testing retrieved from the database, colposcopy (with or without biopsy), or a plan for repeat testing documented in clinic notes available in the database. If the patient was found
to have had a prior cervical smear during the preceding 12 months showing ASCUS, this earlier smear was taken as the initial one from which follow-up was determined. Again, a previous smear or biopsy specimen documenting dysplasia excluded the patient from further analysis. The interval to follow-up was then determined for each patient and categorized as either ≤2 months, >2 months and ≤4 months, >4 months and ≤6 months, >6 months, and no follow-up found. Of the patients receiving follow-up, the proportion seen during each interval was calculated. To examine the effect of specialty type, the proportion of patients returning at <4 months was compared with those returning at ≥4 months for gynecology, general medicine, and family medicine. These proportions were compared with Fisher’s exact (two-tailed) test. The initial cytologic report was then reviewed for the presence of qualifying information: “favor benign” (including reactive changes or atrophy) or “favor squamous intraepithelial lesions.” Inflammation noted by itself, with no other qualifying statements or recommendations, was not included; however, if the cytologist’s comments clearly favored a reactive process and recommended repeat cytologic testing, this was taken as a “favor benign” designation. The follow-up intervals for patients with smears suggestive of a benign process were then compared with those whose smears favored squamous intraepithelial lesions. These results were compared for statistical significance (p < 0.05) by Fisher’s exact test. Results A total of 223 cervical smears were given the cytologic diagnosis of ASCUS during the time studied. Of these, 107 (47%) of the cervical smears were collected in the gynecology clinic, 51 (23%) in general medicine, 18 (8%) in family practice, 15 (7%) in the dysplasia clinic, 13 (6%) in gynecologic oncology, 6 (3%) in pediatric and adolescent medicine, and 1 (0.5%) in the surgery clinic. For 14 patients (9%), the exact location of the screening could not be determined. Forty-two patients (19%) were
Volume 178, Number 5 Am J Obstet Gynecol
excluded from further analysis because of a documented concurrent or prior diagnosis of squamous intraepithelial lesions including two patients who were immunocompromised. As expected, 97% of patients seen in the dysplasia and gynecologic oncology clinics were thus excluded for this reason. Of the remaining 181 patients, follow-up was available for 152 (84%). The time to follow-up after the initial smear was determined and categorized as: ≤2 months, >2 and ≤4 months, >4 and ≤6 months, and >6 months (Table I). Ninety percent of these patients (137 of 152) underwent follow-up testing at ≤6 months. For nine patients (6%), colposcopy was included in their initial follow-up visit with the remainder having repeat cytologic testing alone. Twenty-nine percent (44 of 152) of patients were seen at ≤2 months, and 68% (104 of 152) were seen at ≤4 months. When broken down by specialty, these proportions remained similar: the percentage of patients returning at ≤4 months was 65% for gynecology, 60% for general medicine, and 63% for family medicine (p = 0.65, by Fisher’s exact test). Although the small numbers in the subgroups prevented our study from having sufficient power to determine statistical significance of practice pattern by specialty, the differences observed between specialties were so small that they can be considered clinically insignificant. To determine whether qualifiers of the cytologic diagnosis correlated with time to follow-up, those smears that were reported with qualifiers suggesting a benign process (e.g., “favor reactive” [n = 52]) were compared with those reported as “favor squamous intraepithelial lesions” (n = 42) (Table II). The presence of qualifiers was not correlated with interval to follow-up; similar proportions of patients were rescreened at each of the time intervals regardless of whether the initial report favored a reactive or dysplastic process (p = 0.18, Fisher’s exact test). Moreover, referral for immediate colposcopy in this population did not depend on qualifiers; of the nine patients referred, four had smears favoring squamous intraepithelial lesions, two favored a benign process, and the remainder were unqualified. Of the total number of patients who had follow-up (194), 31 (16%) were found on subsequent testing to have biopsy-proven squamous intraepithelial lesions. However, if patients with a prior history of dysplasia are excluded, only 19 (12%) had biopsy-proven squamous intraepithelial lesions, with 4 (2.6%) having high-grade lesions. Of these, 8 of 19 had initial smears qualified by “favor squamous intraepithelial lesions,” 2 of 19 had “favor reactive,” and the remainder were unqualified. Of those with subsequent high-grade lesions, 2 of 4 had “favor squamous intraepithelial lesions” Papanicolaou smears, 1 of 4 had “favor reactive,” and 1 of 4 were unqualified. The positive predictive value of a “favor squamous intraepithelial lesions” qualifier for subsequent dysplasia was calculated to be 19%, whereas the positive
Suh-Burgmann, Darragh, and Smith-McCune 993
Table II. Follow-up patterns by Papanicolaou smear qualifiers (favor benign vs dysplastic process) Favor benign Favor dysplasia Statistical (n = 52) (n = 42) significance No follow-up (n = 16) Repeat Papanicolaou smear ≤2 mo >2 mo–≤4 mo >4 mo–≤6 mo >6 mo
11 (21%)
5 (12%)
13 (25%) 12 (23%) 14 (27%) 2 (4%)
10 (24%) 17 (40%) 6 (14%) 4 (10%)
p = 0.17*
*p Value obtained by comparison of follow-up times for “favor benign” versus “favor squamous intraepithelial lesion” with Fisher’s exact (two-tailed) test.
predictive value of “favor reactive” for benign follow-up was 96%. Although the calculated odds ratio of “favor squamous intraepithelial lesions” compared with “favor reactive” for subsequent dysplasia is 4, the positive predictive value of a “favor squamous intraepithelial lesions” qualifier alone for subsequent dysplasia was 19% and only 9% for high-grade disease. Thus although a “favor reactive” qualifier had a high predictive value for benign follow-up, a qualifier favoring squamous intraepithelial lesions was not by itself highly predictive of dysplasia. Comment The goal of the study was not to characterize the rationale behind patient management but rather to describe how practitioners are actually managing patients with an initial ASCUS diagnosis. Our results indicate that, within this medical center, providers are performing repeat smears at more frequent intervals than recommended by either The American College of Obstetricians and Gynecologists or the National Cancer Institute guidelines, with nearly one third of patients returning at 2 months or earlier. In contrast, only 10% of patients received follow-up at an interval exceeding 6 months, despite the fact that the patients studied did not have a history of dysplasia, and the incidence of underlying high-grade disease in this population was found to be very low (2%). In this study 16% of patients did not return for follow-up in the 18 months after their initial smear. It is possible that patients scheduled for longer follow-up intervals were more likely to fail their appointments, skewing the results toward shorter follow-up times. However, even if all 16% of patients were meant to return after 6 months, the proportion of patients being seen at ≤4 months is still >50%. Because the aim of the study was to describe the follow-up of patients after their first atypical smear, a recent abnormal smear done at an outside institution and not noted in the database would therefore be a potential source of error in determining
994 Suh-Burgmann, Darragh, and Smith-McCune
follow-up. However, patients seen because of an outside abnormal smear are usually referred to the dysplasia or gynecology clinics where such a history would be routinely documented in the database clinic notes. For patients receiving their screening in general medicine and family practice clinics, referral to specialty clinics occurred only when colposcopy was deemed necessary, which usually required more than one prior ASCUS Papanicolaou smear. Thus, in cases in which the initial ASCUS smear was collected in these locations, the repeat smear was performed in the same clinic. If some patients had been referred to specialty clinics for follow-up, delays caused by scheduling may have occurred. Therefore this would have resulted in an overestimate of the interval time to follow-up screening for these patients and would not negate our conclusions that patients are receiving follow-up more frequently than recommended by current guidelines. In this study the overall rate of dysplasia is 16%, which is within the range of values found in other studies.6-9 The interval to follow-up and referral for immediate colposcopy was not found to correlate with the presence of qualifiers on the cytologic report such as “favor reactive” or “favor squamous intraepithelial lesions.” The presence of a “favor squamous intraepithelial lesions” qualifier, however, was a weak predictor of the presence of squamous intraepithelial lesions. Other reports have similarly found a low rate of high-grade dysplasia among patients with “favor reactive” qualifiers.10, 11 The National Cancer Institute guidelines addressed the use of qualifiers in the management of patients with a cytologic diagnosis of ASCUS and recommended that for patients with a report favoring squamous intraepithelial lesions, colposcopy should be considered. However, several studies have found high interobserver variability and poor reproducibility for this cytologic diagnosis, placing in question the usefulness of these qualifiers.12-16 When confronted with a patient with a history of normal smears who now has a smear showing ASCUS, clinicians therefore find little objective data on which to base their management. This report constitutes a first step toward determining the current standard of care for patients with a cytologic diagnosis of ASCUS. Previous studies have used surveys to examine the rate of colposcopy and referral for patients with ASCUS smears.17-20 However, it is not clear whether these surveys allowed respondents the option of differentiating between initial versus repeat ASCUS smears and immediate versus eventual colposcopy. Moreover, the management of these patients by repeat cytologic screening alone was not specifically characterized. The results of studies such as the one described here are therefore useful in providing information about actual clinical management. There are several reasons to explain the discrepancy between the published guide-
May 1998 Am J Obstet Gynecol
lines and the actual practices we observed. Providers may be uncomfortable with the published guidelines or may simply be unaware of them. Data regarding the role of other patient risk factors, apart from a history of dysplasia, such as multiple sexual partners, smoking, and reliability were not collected in this study but clearly may influence the follow-up plan for individual patients. In addition, clinicians may be influenced by patient anxiety and desire for early follow-up testing. Further studies are warranted to both substantiate the current guidelines as written and to determine the bases for the choices practitioner’s are making in clinical practice. REFERENCES
1. National Cancer Institute Workshop. The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. JAMA 1989;262:931-4. 2. Broder S. The Bethesda System for reporting cervical/vaginal cytologic diagnoses: report of the 1991 Bethesda Workshop [rapid communication]. JAMA 1992;267:1892. 3. Kurman RJ, Solomon D. The Bethesda system for reporting cervical/vaginal cytologic diagnoses: definitions, criteria and explanatory notes for terminology and specimen adequacy. New York: Springer-Verlag; 1994. p. 30-8. 4. American College of Obstetricians and Gynecologists. Cervical cytology: evaluation and management of abnormalities. Washington: The College; 1993. p. 1-8. ACOG Technical Bulletin No.: 183. 5. Kurman RJ, Henson DE, Herbst AL, Noller KL, Schiffman MH. Interim guidelines for management of abnormal cervical cytology: the 1992 National Cancer Institute Workshop. JAMA 1994;271:1866-9. 6. Taylor RR, Guerrieri JD, Nash JD, Henry MR, O’Connor DM. Atypical cervical cytology: colposcopic follow-up using the Bethesda system. J Reprod Med 1993;38:443-7. 7. Andrews S, Hernandez E, Miyazawa K. Paired Papanicolaou smears in the evaluation of atypical squamous cells. Obstet Gynecol 1989;73:747-50. 8. Maier RC, Schultenover SJ. Evaluation of the atypical squamous cell Papanicolaou smear. Int J Gynecol Pathol 1986;5:242-8. 9. Linheim SF, Smith-Ngyuen G. Aggressive evaluation for atypical squamous cells in Papanicolaou smears. J Reprod Med 1990;35:971-3. 10. Gonzales D, Hernandez E, Anderson L, Heller P, Atkinson BF. Clinical significance of a cervical cytologic diagnosis of atypical squamous cells of undetermined significance: favoring a reactive process or low grade squamous intraepithelial lesion. J Reprod Med 1996;41:719-23. 11. Collins LC, Wang HH, Abu-Jawdeh GM. Qualifiers of atypical squamous cells of undetermined significance help in patient management. Mod Pathol 1996;9:677-81. 12. Klinkhamer PJ, Vooijs GD, de Haan AF. Intraobserver and interobserver variability in the diagnosis of epithelial abnormalities in cervical smears. Acta Cytol 1988;32:794-800. 13. Sidaway MK, Tabarra SO. Reactive change and atypical squamous cells of undetermined significance in Papanicolaou smears: a cytohistologic correlation. Diagn Cytopathol 1993;9:423-7;discussion 427-9. 14. Davey DD, Naryshkin S, Nielsen ML, Kline TS. Atypical squamous cells of undetermined significance: interlaboratory comparison and quality assurance monitors. Diagn Cytopathol 1994;11:390-6. 15. Lonky NM, Navarre GL, Saunders S, Sadeghi M, Wolde-Tsadik G. Low-grade Papanicolaou smears and the Bethesda system: a prospective cytohistopathologic analysis. Obstet Gynecol 1995;85:716-20. 16. Sherman ME, Schiffman MH, Lorincz AT, Manos MM, Scott DR,
Volume 178, Number 5 Am J Obstet Gynecol
Kuman RJ, et al. Toward objective quality assurance in cervical cytopathology: correlation of cytopathologic diagnoses with detection of high-risk human papillomavirus types. Am J Clin Pathol 1994;102:182-7. 17. Melnikow J, Sierk A, Flocke S, Peters CA. Does the system of Papanicolaou test nomenclature affect the rate of referral for colposcopy? A survey of family physicians. Arch Fam Med 1993;2:253-8.
Suh-Burgmann, Darragh, and Smith-McCune 995
18. Gordon P, Hatch K. Survey of colposcopy practices by obstetrician/gynecologists. J Reprod Med 1992;37:861-3. 19. Chmielewski MJ, Dunton CJ. Internet survey of colposcopy practices. J Low Gen Tract Dis 1997;1:151-3. 20. Wain GV, Ward J, Towler BP. Gynecologic care of women with abnormal Pap smears: How varied is current practice? Med J Aust 1995;162:348-53.
Bound volumes available to subscribers Bound volumes of the American Journal of Obstetrics and Gynecology are available to subscribers (only) for the 1998 issues from the publisher, at a cost of $116.00 for domestic, $148.73 for Canada, and $139.00 for international for vol. 178 (January-June) and Vol. 179 (July-December). Shipping charges are included. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the Journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, USA; phone (800)453-4351 or (314)453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.