David McCreary
Emergency Fellow

Alfred Health

A 26-year-old female represents to the ED 24 hours after an initial assessment for lower abdominal pain which was attributed to likely pelvic inflammatory disease (PID), for which antibiotics had been commenced by GP and bolstered to an eTG-happy regimen in ED before being discharged.

On representation, the patient was reporting increasing lower back and abdominal pain. PV and speculum exam on the second presentation showed purulent discharge from the cervix and right adnexal tenderness.

WCC was normal on both presentations. CRP was 155 on first presentation and had increased to 205 the following day.

The concern at this point was for tubo-ovarian abscess (TOA) and so a pelvic ultrasound was performed quicker than you could say “fill your bladder” and was reported as:


Moderate volume free fluid in the pelvis which is mildly complex, with possible dilated right fallopian tube. This area was mobile, and non-tender.

No collection, abscess or features of appendicitis.

At which point the patient was discussed with our friendly neighbourhood gyanecologists, who were happy to continue her investigation and management under their care as a likely tubo-ovarian abscess.

So, what can we learn from this case? Well, first let’s talk a little about TOA…

What is it?
It’s a complex infectious mass of the adnexa that forms as a sequela of PID.
How does it present?
Classically, lower abdominal-pelvic pain ± PV discharge, fever, ↑ WCC and an adnexal mass on PV exam.

While this is the textbook presentation, a study in 1983(1) reported 35% patients were afebrile, 23% had normal WCC, only 50% complained of fevers/chills, 28% PV discharge, 265 with nausea and 21% with abnormal vaginal bleeding.

Exam findings?
Mucopurulent discharge and cervical motion tenderness → PID

Add uterine or adnexal tenderness and you’re thinking TOA.

Risk factors?
  • Sexually active
  • Multiple partners
  • IUD insertion
  • Previous history of PID
What are your differentials?
All the usual suspects:

  • Appendicitis
  • Diverticulitis
  • IBD
  • PID
  • Torsion
  • Ectopic
  • Ruptured ovarian cyst
  • Pyelonephritis
  • Cystitis
What investigations are you going to do?
Aside from the usual (bloods, HCG etc), options are US (sensitivity 75-82%) or CT (sensitivity up to 100%). Though for obvious reasons we’re usually choosing US as first line in these patients, knowing that CT is there if you have ongoing diagnostic uncertainty.
Does it matter?
Hell yes. If it ruptures the resulting intra-abdominal sepsis can be life threatening and then we’re in strife.
What bugs cause it?
As you would expect – Chlamydia and Gonorrhoea are on there. But so are E.Coli, Bacteroides, Peptococcus and Peptostreptococcus. Actinomyces can be associated with presence of IUD.
From us: Ceftriaxone 2g daily as per eTG (empirical treatment for severe PID)

From our specialty friends:

  • Percutaneous draining by radiology will suffice for most
  • Alternatively, laparoscopic drainage or laparotomy may be needed if things are really bad
So, let’s talk utility of pelvic exams…

We had a bit of spirited discussion during our grand round on this case as to when we should be performing pelvic examinations on female patients presenting with lower abdominal pain. The text book answer is always. Why? You can assess for cervical motion tenderness, adnexal tenderness, purulent discharge, and you can take a high vaginal swab. But in the uncomplicated patient where you are already past the pre-test probability where you have decided to treat them – does it need to be performed on every patient?

Well, I’ll be the first to put my hand up and admit that I don’t perform pelvic examination on every female patient whom I treat for pelvic inflammatory disease, though it’s thoroughly on a case-by-case basis and I have to be pretty convinced that it is uncomplicated PID.  So, I cannot stress enough that the following argument is not for patients with abdominal tenderness, abnormal vital signs or a stonking high CRP (as was the case in the patient). But as I love a bit of EBM, let’s look at what I was on about.

There is a little bit of evidence for this approach with a study in the Annals of Emergency Medicine 2018(2). You can see my summary of the paper, and listen to my RCEMLearning podcast segment on it here(3).

This was a study of 288 female patients aged 14-20 years with lower abdominal pain and/or PV discharge and normal vital signs. All patients had urine testing for chlamydia, gonorrhoea and trichomonas. Following standardised history to assess for cervicitis/PID, the clinicians recorded their likelihood of disease on a VAS. The same clinician then performed pelvic exam and again recorded their likelihood of PID with the additional information gained.

They looked at the VAS before and after examination for both STI +ve and STI -ve patients. They also looked at how many cases where the examination would have changed management (by moving the VAS across the half way mark that they had decided was diagnostic).  They then worked out the test characteristics of pelvic exam using the urine results as the standard.

What did they find?

History alone had a sensitivity of 54.5% and specificity of 59.8% for cervicitis/PID.

History plus pelvic exam had a sensitivity of 48.1% and specificity of 60.7%.

So, no difference.

The information the clinicians gained from the pelvic examination changed their management in 71 cases. Of those, 35 actually had a STI and 36 didn’t…so it was a coin flip.

They concluded “For young female patients with suspected cervicitis or PID, the pelvic examination does not increase the sensitivity of diagnosis of chlamydia, gonorrhoea, trichomonas compared with taking a history alone.  Test characteristics for pelvic examination are not adequate, its routine performance should be reconsidered.”

Why does it matter?

Well, that’s a valid question. It is the textbook work up for a reason. Despite this study, I don’t think diagnostically you’re ever going to be worse off for doing an examination. However, as the authors point out: pelvic exam is uncomfortable and emotionally distressing for most women, particularly teenage girls – so are the not less likely to engage in health care if the think they are routinely going to have to undergo an intimate exam? It’s the same argument for not doing ABGs on asthmatics or diabetics if you don’t have to.

Caveat, disclaimer, “don’t blame me” closing statement…

Remember this study, and my argument is entirely for uncomplicated PID presentations. If there’s a sniff of unwell or not quite right then this is not the paper for you.


  1. Landers DV, Sweet RL. Tubo-ovarian abscess: contemporary approach to management. Rev Infect Dis. 1983 Oct;5(5):876–84.
  2. Farrukh S, Sivitz AB, Onogul B, Patel K, Tejani C. The Additive Value of Pelvic Examinations to History in Predicting Sexually Transmitted Infections for Young Female Patients With Suspected Cervicitis or Pelvic Inflammatory Disease. Ann Emerg Med. 2018;7
  3. McCreary D, Neill A. Pelvic Exam for PID. (RCEMLearning Podcast). December 2018. Available from: https://www.rcemlearning.co.uk/foamed/december-2018/
  4. Kairys N, Roepke C. Tubo-Ovarian Abscess. [Updated 2020 Jun 27]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2020 Jan-.



David McCreary

David McCreary

Emergency Fellow, Alfred Health

Dave is an Emergency Physician who completed training between the UK and Australia and completed an MSc in Trauma Science with QMUL. His clinical interests include trauma, critical care, evidence based medicine and human factors. Dave is a regular contributor the RCEMLearning podcast and is a FOAMed editor for RCEMLearning. He dislikes coriander, decaff coffee and dermatology.