Skip to content

Is Pelvic Exam in the Emergency Department Useful?


graves-vag-speculum-lg-30-20-miltexWomen with undifferentiated abdominal pain and/or vaginal bleeding commonly present to the emergency department. Many textbooks advocate for the pelvic exam as an essential part of the history and physical exam. Performance of the pelvic exam is time consuming to the physician and uncomfortable for the patient. It is with great regularity that emergency physicians make clinical decisions, based on information derived from the pelvic examination, but is this information reliable and does it effect the clinical plan of patients?

How reliable is the pelvic exam?

Study Year of Study Number of Patients What Was Evaluated Results
Close RJH et al 2003 186 Percentage agreement for cervical motion tenderness, uterine tenderness, adnexal tenderness, adnexal mass 71 – 84% agreement of pelvic exam, but only
17 – 33% agreement of positive findings
Houry D et al 2001 87 Accuracy of pelvic exam for ovarian torsion 29% of pts had no pain on exam, and 53% had no palpable mass
Padilla LA et al 2000 140 Patients undergoing laparoscopy or laparotomy had pelvic exams performed under general anesthesia Adnexal Mass Detection:
Sensitivity 15 – 36%,
Specificity 79 – 92%,
PPV 26 – 69%
Dart RG et al 1999 441 History and physical exam findings predictive of ectopic pregnancy (EP) No constellation of findings could confirm or exclude EP

Is pelvic exam in the emergency department useful? 1

What they did:

  • Prospective cohort study
  • 183 female patients with abdominal pain and/or vaginal bleeding prospectively evaluated
  • Providers were asked to predict the findings of the pelvic exam
  • Providers predictions were compared with actual findings of exam

Exclusion criteria:

  • Patients with suspected vaginal foreign body
  • Patients requiring pelvic exam for cultures
  • Patients whom the exam required visualizing products of conception


  • Pelvic exam findings were as predicted in 72% of cases
  • Pelvic exam findings were not as predicted in 22% of cases, but no change in clinical plan
  • Pelvic exam findings were not as predicted in 6% of cases but did change the clinical plan
    • 7 with normal exam
    • 1 not specified
    • 1 adenexal tenderness
    • 1 IUD requiring removal,
    • 1 less uterine bleeding than expected
    • 1 cervical motion tenderness


  • Management plans were not specified prior to completion of pelvic exam
  • The way in which unexpected findings changed plans was not documented
  • No objective criteria were used to determine need for cervical cultures and this group excluded from the analysis

Conclusion: In 94% of female patients with acute abdominal pain and/or vaginal bleeding, the results of the pelvic exam were predictable or had no effect on clinical plan.

Take Home Point

Although this is controversial and contrary to traditional teaching, this study suggests that basing decisions on female pelvic exams in the emergency department is not reliable and may not have any effect on the clinical plan.

Brown J, Fleming R, Aristzabel J, Gishta R. Does pelvic exam in the emergency department add useful information? West J Emerg Med. 2011;12(2):208-212. [PubMed]
Salim Rezaie, MD

Salim Rezaie, MD

ALiEM Associate Editor
Clinical Assistant Professor of EM and IM
University of Texas Health Science Center at San Antonio
Founder, Editor, Author of R.E.B.E.L. EM and REBEL Reviews
  • Michelle Lin

    Great discussion on Twitter today. Thought I’d share:

    Finally, and answer to the "Have you done a PV in the ED?" question #FOAMed— Tim Leeuwenburg (@KangarooBeach) January 30, 2014

  • Michelle Lin

    Interesting discussion today on Twitter. Thought I’d embed here to share:

  • Michelle Lin

    This is a discussion that happened on Twitter today. Fascinating discussion!

  • Graham Walker

    Agree with you, Salim, mostly. The pelvic exam is almost never helpful — except when it is (<2%?) Has anyone else had the patient with rip-roaring, fulminant, pus-from-the-cervix PID that you would have probably sent home otherwise? I have several times in my career; they're usually teens or young adults too embarrassed to give you an honest sexual history or tell you about their vaginal discharge (even if you ask), and imaging, bloods, and urine don't always come up with anything. So you do the pelvic, and you're just flabbergasted.

    Those times are the rare times it's quite helpful. It's also helpful in knowing if it's a threatened vs inevitable abortion, and in knowing a few hours earlier which patients need to go to the OR for a D&C because they're hemorrhaging to death vaginally (but they're young and so can compensate with normal vitals for a long time).

    • Salim R. Rezaie

      Hello Graham,

      I like to think of the pelvic exam like any other test we would order. What is in my Ddx? I order imaging and lab tests based on this, not just a “shotgun” technique. If PID or threatened/inevitable abortion is there, or I don’t have a better answer from what I have ordered, then yes it should be done. By no means am I advocating that this part of the exam should be completely eliminated, but do you have to do it on every female with lower abdominal pain or vaginal bleeding? Again it depends on your Ddx. It is an invasive portion of the exam that causes distress and discomfort in patients., The stats from the above studies are horrible, even in the OR under complete anesthesia by OB/Gyn physicians.

      All I am saying is that some clinical judgement should be used and not every single female with lower abdominal pain or vaginal bleeding needs the pelvic exam as stated in many textbooks. Thank you for your comments and reading the post.


      • Shannon McNamara

        This post does a great job of framing the pelvic exam like all the other tests preformed in the ED – we should preform these tests with a specific question in mind, and be thoughtful about the impact any test has on our patients.

        I do think there are many situations where the pelvic exam answers helpful clinical questions. For miscarriages, is the cervix open/closed? Are there POC? For pelvic pain: Is there pus? Is there CMT/adnexal tenderness? For discharge: is there trich? BV?

        An issue that I think we often overlook in the ED is how subtle PID can actually be. The CDC strongly encourages treating even mild PID (pelvic pain in young woman with risk factors, and CMT or adnexal or uterine tenderness on pelvic exam). Many cases of PID are not even associated with STDs.

        I have a low threshold to diagnose PID, but find the pelvic exam essential in making that diagnosis, and important when committing the patient to 14 days of doxy +/- flagyl. It’s hard for me to differentiate which young women have mild PID based on history alone, as they often present with undifferentiated lower abdominal pain, so the pelvic is important.

        For example: a patient presents with vaginal discharge and mild pelvic pain, urine GC is sent, but she is very well appearing, so she doesn’t get a bimanual to pick up CMT. She’s treated for cervicitis, but actually had PID. She may have a worsening clinical course of PID, or develop some infertility and increased risk of ectopic pregnancy form her incompletely treated infection. Not ideal.

        I do agree, however, that it is very reasonable to modify the exam based on what information is important to change the clinical management of the patient. For many patients, I won’t do a speculum exam if I don’t think it will add any useful information. For patients with painless discharge, they can self-swab for a wet mount and avoid the pelvic. The urine GC testing is very helpful in collecting specimens who don’t need a pelvic for any other reasons.

        Thanks for raising these important questions to challenge our conventional practice! Great discussion.

        • Salim R. Rezaie

          Hello Shannon,

          Very much appreciate your thoughts and the important point of PID. And I think you hit the nail on the head:

          “The pelvic exam should be treated like all the other tests performed in the ED – we should perform these tests with a specific question in mind.”

          TY for reading the post and your comments.


  • Matthew DeLaney


    Great post! There are plenty of shifts when I get slammed and it would be tremendously helpful to not take the time to do a pelvic exam. Also it has to be one of the most unpleasant things we do to patients, likely right behind NG tubes. I think bleeding in early pregnancy is another area where we could make a decent argument to omit the pelvic exam. The patient is already upset because she might be having a miscarriage, has already had a fairly invasive ultrasound, is there really a need to do a pelvic exam and make a bad day that much worse. Here is a decent study that is at least food for thought:

    Graham, I agree that sometimes the pelvic exam can give us a huge piece of the puzzle. I think we should probably do more pelvics on females with pain and discharge, but a whole lot less pelvics on females who have pain and vaginal bleeding.

    • c-daks

      I tend to disagree. The state of the cervical os is an important piece of information in early preg + bleeding, esp if it has a POC hanging out of it. And a pelvic exam, while time-consuming, is not one of the most unpleasant things we do to patients if done right (speaking as a woman as well as a doc). I definitely agree that it, like all other tests, should have a specific question in mind. Every now and again we get fooled though. I’ve definitely found a very unexpected bleeding vaginal wall lesion that required gyne f/u (to r/o carcinoma) when expecting to find a very unremarkable trickle of blood from a cervical os!

  • njoshi8

    I think that it cannot be stated enough that the stats on these papers are terrible! And therefore to make conclusions off of this doesn’t make much sense. As a new attending, and as a woman I get very frustated when I feel that newer docs try to find excuses as to why they don’t need to do a pelvic exam. I agree – if it was on the nose, we would examine it and not even care. We should be diligent on our exams. And if we don’t practice them, we will lose the skills. We are bad enough describing what the os looks like! Yes pre and post test probability is important, but we must not try to “justify” a pretest probability because pelvic exams are time consuming, or difficult to coordinate..

  • CC

    I am a PA who just started in the ED and can think of 2 examples already where I definitely am glad I went ahead and did the pelvic exam. One was a young pregnant female who had a retained tampon which she did not inform me of. The other was a vaginal bleeder who’s H&H did not yet correlate with the copious amount of clotting blood I found pouring out on exam. Just to give you some thoughts from the newby…

  • @befune

    I actually did this topic as my Journal club this past November. The general consensus from the group was that this study was not nearly rigorous enough to make any strong statements. The number sound great, but there was a lot of criticism of the methods. I think it shows that this is likely an area where with good research we have room to make the test more beneficial for each patient that requires it. I do believe the results of future studies would probably follow this data though.

    I also remember trying to find a solid paper on the diagnosis of PID and there really isn’t anything concrete. The gold standard is laparoscopy, biopsy or culture (GC/chlamydia are poor), CMT, purulence, etc all don’t have the best PPV. Pelvic pain seems to be the most sensitive which is unfortunately awfully nonspecific, which does put the pelvic exam in a terrible position to be successful.

    Also, there are some good papers that show self swab and trich PCR are as good or better to speculum swab. Urine gonorrhea testing is terrible (~55%) for females, but an acceptable replacement in men

    1.Eval of single intravaginal swab in women –
    2.Cook, Robert Et al – Systematic review of noninvasive gc/ch testing
    3.Practical approach to dx of PID –

    It’s also important to remember this paper doesn’t apply to pregnancy as some of the comments below mention.

    • Michelle Lin

      Sent from Mailbox