[Skip to Content]
[Skip to Content Landing]
Invited Commentary
Health Care Reform
Dec 12 2011

Pelvic Examinations in Asymptomatic Women: Tipping a Sacred Cow: Comment on “The Pelvic Examination as a Screening Tool”

Author Affiliations

Author Affiliation: Department of Obstetrics, Gynecology & Productive Science, University of California, San Francisco.

Arch Intern Med. 2011;171(22):2054-2055. doi:10.1001/archinternmed.2011.567

In our effort to practice evidence-based medicine, the benefits and harms of all medical interventions have come under scrutiny. Stormo and colleagues1 begin an exploration of one of the most prevalent yet largely unquestioned practices in women's health: the pelvic examination. Using a Web-based survey targeting a variety of clinicians, they posed a fundamental question: how often do you perform pelvic examinations for a given set of purposes?

The most commonly reported purpose, and the least clear, seemed to be rooted in habit alone: as “part of a well-woman exam.” It is not surprising that nearly all obstetrician/gynecologists who responded cited this purpose, since the American College of Obstetricians and Gynecologists includes pelvic examinations as a component of periodic assessments for women aged 21 to 64 years.2 The American College of Obstetricians and Gynecologists states that it is reasonable to discontinue pelvic examinations “when a woman's age and other health issues are such that she would not choose to intervene on conditions detected during the routine examination.”2(p1012) The conditions the examination is targeting, however, are not defined. While clinicians may be adhering to guidance promulgated by professional organizations, they may also be performing these examinations to accommodate patient expectations. On the one hand, it is likely that many women expect a pelvic examination as part of an annual examination, especially by a gynecologist, and some women may wonder if a clinician who omits it is being inattentive or even negligent. On the other hand, women who might have a negative reaction to these examinations3 may be relieved to not receive them.

Using the pelvic examination as a screening test for ovarian cancer in asymptomatic women was also commonly cited. If by “pelvic examination” respondents were including bimanual examinations—a reasonable assumption—then it is unlikely that such an examination has value; a recent randomized trial of 78 216 women aged 55 to 74 years demonstrated that screening with CA-125 and pelvic sonograms (likely more accurate than bimanual examinations) are ineffective in preventing ovarian cancer mortality.4 Screening did, however, identify abnormalities that set in motion a cascade of events that led to harm due to false-positive testing results and subsequent complications of unnecessary surgery. The harms of screening are likely greater in premenopausal women in whom the ovaries are dynamic, regularly making follicular cysts and corpus lutea; these masses can be detected on bimanual examinations, leading to follow-up examinations and undue concern before ultimate (and natural) resolution. Some women will find themselves subjected to surgery to prove definitively that these masses are benign. When cancer is not found (by far the most common scenario), women will be relieved, but clinicians will rarely trace the entire series of events back to an unwise screening decision. The number of women who follow this unfortunate path each year is unknown, but is likely sizable given the sheer number of pelvic examinations performed each year in the United States (63.4 million as reported by Stormo et al1).

In situations in which the pelvic examination is used to make decisions about the provision of hormonal contraception, the potential harms are magnified. Not only are women exposed to the harms of screening, but they are also at risk of unintended pregnancies if they are unable to comply with examination requirements; as Stormo et al1 note, pelvic examinations are unrelated to hormonal contraception provision.5-8

The report by Stormo et al1 is a good first step but has important limitations. Given the study design, it is not possible to determine the proportion of practitioners that actually performs pelvic examinations and under what circumstances. Moreover, “pelvic examination” is not defined, and respondents may have interpreted its meaning in various ways. For example, some may have believed it to include a speculum examination for collection of cervical cytology (for Papanicolaou tests) or specimens to screen for sexually transmitted infections. Finally, given the sampling strategy and response rate, the study population may not reflect the practices of average clinicians in the United States.

Where should we go from here? Future studies should focus on defining the potential benefits and harms of the pelvic examination in asymptomatic women. In the absence of direct evidence of benefits and harms, a thoughtful decision analysis of various aspects of the examination (ie, inspection of the external genitalia, speculum examination, bimanual examination) including potential benefits (eg, identifying neoplastic vulvar/vaginal lesions, identifying benign ovarian lesions that may cause medical emergencies such as rupture and torsion) and harms (eg, complications due to false-positive testing results) would be useful. Given plausible outcomes, clinicians should not be surprised if the examination is deemed to be more harmful than beneficial and ultimately is discouraged as part of a well-woman exam.

Applying a critical lens to preventive interventions is especially important because clinicians must be assured that their interventions will, on average, enhance the future health of currently healthy people without causing illness in the present. In our pursuit of defining such interventions, there are no sacred cows. The benefits and harms of pelvic examinations in asymptomatic women should continue to be scrutinized. If these examinations are found to provide net benefit, they should be continued and promoted; if not, they should be put out to pasture.

Back to top
Article Information

Correspondence: Dr Sawaya, Department of Obstetrics, Gynecology & Productive Science, University of California, San Francisco, 3333 California St, Ste 335, San Francisco, CA 94143 (sawayag@obgyn.ucsf.edu).

Financial Disclosure: None reported.

Stormo AR, Hawkins NA, Purvis Cooper C, Saraiya M. The pelvic examination as a screening tool: practices of US physicians.  Arch Intern Med. 2011;171(22):2053-2054Google ScholarCrossref
American College of Obstetricians and Gynecologists Committee on Gynecologic Practice.  ACOG Committee Opinion No. 483: Primary and preventive care: periodic assessments.  Obstet Gynecol. 2011;117(4):1008-101521422880PubMedGoogle ScholarCrossref
Weitlauf JC, Finney JW, Ruzek JI,  et al.  Distress and pain during pelvic examinations: effect of sexual violence.  Obstet Gynecol. 2008;112(6):1343-135019037045PubMedGoogle ScholarCrossref
Buys SS, Partridge E, Black A,  et al; PLCO Project Team.  Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial.  JAMA. 2011;305(22):2295-230321642681PubMedGoogle ScholarCrossref
Saint M, Gildengorin G, Sawaya GF. Current cervical neoplasia screening practices of obstetrician/gynecologists in the US.  Am J Obstet Gynecol. 2005;192(2):414-42115695980PubMedGoogle ScholarCrossref
Sawaya GF. A 21-year-old woman with atypical squamous cells of undetermined significance.  JAMA. 2005;294(17):2210-221816264163PubMedGoogle ScholarCrossref
Schwarz EB, Saint M, Gildengorin G, Weitz TA, Stewart FH, Sawaya GF. Cervical cancer screening continues to limit provision of contraception.  Contraception. 2005;72(3):179-18116102551PubMedGoogle ScholarCrossref
Stewart FH, Harper CC, Ellertson CE, Grimes DA, Sawaya GF, Trussell J. Clinical breast and pelvic examination requirements for hormonal contraception: Current practice vs evidence.  JAMA. 2001;285(17):2232-223911325325PubMedGoogle ScholarCrossref