Gynecologic Oncology Consult

When NOT to perform a Pap test


Pap tests have the reputation of being a simple, noninvasive, low-cost test to offer patients, and, therefore, it is understandable to believe there is no harm in offering it in all situations. However, if inappropriately applied in isolation, performing the Pap test may do more harm than good.

Doctor Talking To Serious Female Patient In Exam Room monkeybusinessimages/iStock/Getty Images

I recently saw a patient in consultation for cervical cancer. Her story was similar to one I’ve seen many times before. She was a 30-year-old non–English-speaking Hispanic woman who received regular care from the health department clinics.

In April of the prior year, she had noticed abnormal bleeding symptoms including intermenstrual and postcoital bleeding. She visited the health department and reported these symptoms to the provider who performed an examination. According to the provider’s notes, the cervix appeared “abnormal” and a Pap test was done. The result of this Pap test was high-grade dysplasia. The patient was promptly notified of the result and an appointment was arranged with the local ob.gyn. for a consultation, presumably for colposcopy and subsequent appropriate excisional procedure. Unfortunately, the patient did not attend that scheduled appointment. She later recounted to me that it was because she had not understood that it was important. She had a long history of abnormal Pap tests which, in the past, had only required repeat testing or minor interventions such as “freezing.”

Her bleeding symptoms became worse, and she developed abnormal discharge and pain. In November, she presented again for evaluation to the same provider. Now her cervix appeared very abnormal and was described as a “crater.” Again a Pap test was done. This time the Pap test showed “carcinoma,” and the patient was informed that she had cancer and was referred to gynecologic oncology. When I examined this unfortunate young woman, I discovered a 10 cm, stage IIB very locally advanced tumor. She is currently receiving primary chemotherapy/radiation with an approximately 60% probability of cure, and a high likelihood of lifelong sequelae of this toxic therapy.

This case highlights that, even when patients are engaged within our health care system, we can miss the opportunity to diagnose early-stage cancers if we are not utilizing screening versus diagnostic tests appropriately.

The purpose of a Pap test is as a screening test, which are designed to detect disease in asymptomatic individuals. The accuracy of these tests is determined in low-risk (asymptomatic) populations, which influences the pretest probability of disease. In asymptomatic patients with a normal screening test, it is safe to wait out the interval of time for the repeat screening test, because the combination of a low pretest probability and a high sensitivity of the test in finding disease means that there is a very low chance of missing disease.

Dysplasia rarely causes bleeding. However, invasive cervical cancer does. If a patient has a symptom such as abnormal bleeding, they no longer fit into the population with a low pretest probability for having cervical cancer. This same sensitivity of the Pap test in finding disease, combined with the now-higher pretest probability can raise the level of false-negative results to unacceptably high levels.

Patients with symptoms of cervical cancer should not receive screening tests exclusively; they should receive diagnostic tests. For example, Pap tests should not be used in isolation to diagnose pathology in patients with abnormal bleeding or discharge, just as screening mammograms should not be ordered in patients with symptomatic breast lumps, nipple discharge, retraction, etc. (these women should be referred for diagnostic mammography and ultrasound). It is not unusual for gynecologic oncologists to see patients with visible invasive cervical cancer who have only cervical intraepithelial neoplasia grade 3 on the preceding Pap test. There is a 34% positive predictive value that a cervical cancer will be found with a high-grade dysplastic Pap test.1 Cytology is an inferior diagnostic tool, compared with histology, in determining invasive cancer from preinvasive lesions. Cytology is an inferior diagnostic tool, compared with histology, in determining invasive cancer from preinvasive lesions. It analyzes individual cells rather than a piece of tissue with intercellular relationships.

Dr. Emma C. Rossi is an assistant professor in the division of gynecologic oncology at UNC-Chapel Hill.

Dr. Emma C. Rossi

The take-home message for this column is that, if a provider sees an abnormal lesion on a cervix, they should biopsy the visible lesion to obtain a histologic diagnosis. Simply performing a Pap test alone may result in false reassurance and in underestimating the severity of disease.

Some providers will tell me that they have concerns about performing a biopsy on a grossly abnormal cervix for fear that the subsequent bleeding will be difficult to manage in the outpatient setting. This is understandable, although it is unlikely that an office equipped with the ability to perform colposcopy or excisional procedures would not have the necessary equipment to manage this. Prolonged pressure applied to the cervix with topical hemostatic agents or – in extreme cases – vaginal packing with gauze always has been effective for me in these circumstances.

The additional benefit of establishing histologic confirmation prior to referral is expediting care, including additional imaging and referrals to treating providers. If the diagnosis is inadequately established prior to their appointment with a gynecologic oncologist, it can add further delays before definitive surgical or nonsurgical management can be initiated, which is particularly problematic if the patient is experiencing severe bleeding. If the provider feels uncomfortable with proceeding with biopsy, they should inform the patient very clearly that they suspect that there is a cancer of the cervix, and it needs attention from a cancer specialist to confirm the diagnosis. This clear communication will minimize the likelihood that the patient may not show up for the subsequent appointments before her diagnosis is definitively established.

Another common scenario in which Pap tests are inappropriately applied is in the surveillance of endometrial cancer. In 2013, the Society of Gynecologic Oncology released its five “Choosing Wisely” recommendations. This included the recommendation to not perform Pap tests in the surveillance of endometrial cancer. This recommendation was based on a body of evidence that demonstrates screening for endometrial cancer recurrence with Pap smears does not detect vaginal mucosal recurrences any sooner than visualization of lesions on speculum examination.2,3 These Pap-positive recurrences almost always are visible on exam. Additionally, false positives are common in this population, particularly among women who have had radiation or have atrophic tissues.

Using Pap tests for the surveillance of cervical cancer is somewhat more complicated. Similarly, they do not detect cervical cancer recurrence any sooner than comprehensive examination does. However, this population may suffer from chronic human papillomavirus (HPV) infection, and there remains a role of the Pap test in screening for future, new HPV-related preinvasive vaginal disease. Therefore, Pap tests, and/or HPV testing can be offered to cervical cancer survivors in accordance with the American Society for Colposcopy and Cervical Pathology guidelines for noncervical cancer patients, with the caveat that, if radiation has been given, false positives are more likely.2

Pap tests clearly have an important role as a screening test in asymptomatic individuals. However, when the patient has a symptom that might be cervical cancer or a visibly suspicious lesion, she should receive a diagnostic test, and Pap tests are not designed for that purpose.

Dr. Rossi is assistant professor in the division of gynecologic oncology at the University of North Carolina at Chapel Hill. She said she had no conflicts of interest. Email Dr. Rossi at [email protected].


1. Cytopathology. 2016 Jun;27(3):201-9.

2. Gynecol Oncol. 2017 Jul;146(1):3-10.

3. Gynecol Oncol. 2011 Nov;123(2):205-7.

Next Article: