Article Text


Short report
Characteristics of pelvic inflammatory disease where no sexually transmitted infection is identified: a cross-sectional analysis of routinely collected sexual health clinic data
  1. Jane L Goller1,
  2. Alysha M De Livera1,
  3. Christopher K Fairley2,
  4. Rebecca J Guy3,
  5. Catriona S Bradshaw2,
  6. Marcus Y Chen2,
  7. Jane S Hocking1
  1. 1Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, Parkville, Victoria, Australia
  2. 2Central Clinical School, Monash University and Melbourne Sexual Health Centre, Carlton, Victoria, Australia
  3. 3Kirby Institute, University of New South Wales, Kensington, New South Wales, Australia
  1. Correspondence to Jane Louise Goller, Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Level 3, 207 Bouverie Street, Parkville VIC 3010, Australia; jane.goller{at}


Objectives Pelvic inflammatory disease (PID) occurs when pathogens, often sexually transmitted, ascend to the upper genital tract, yet a causative pathogen is not detected in a substantial proportion of diagnosed PID. We assessed the characteristics associated with PID in women in whom chlamydia, gonorrhoea, Mycoplasma genitalium (MG) and bacterial vaginosis (BV) were not detected (‘pathogen-negative-PID’).

Methods Cross-sectional analysis of routinely collected clinical data from new female patients attending a sexual health clinic between 2006 and 2013. Women were eligible if they had been diagnosed with PID and tested for genital chlamydia, gonorrhoea, MG and BV. Logistic regression was conducted to identify characteristics associated with pathogen-negative-PID.

Results Among 330 women with clinically diagnosed PID, 204 (61.8%, 95% CI 56.3% to 67.1%) had pathogen-negative-PID. Compared with pathogen-positive-PID, pathogen-negative-PID cases were more likely to be aged ≥30 years (adjusted odds ratio (AOR) 1.7, 95% CI 1.0 to 3.0), had less evidence of vaginal inflammation (AOR 0.5, 95% CI 0.3 to 0.9) and reported less unprotected sex (AOR 0.6, 95% CI 0.4 to 1.0).

Conclusions These findings highlight uncertainties around PID diagnosis and aetiology. Pathogen-negative-PID could represent (i) a false positive diagnosis where the woman does not have a sexually transmitted infection (STI) or PID, (ii) PID of another microbiological aetiology or associated with a past STI or (iii) PID where the cervical infection has cleared. However, until diagnostic biomarkers are available, PID treatment should be based on clinical features and sexual risk.


Statistics from


Pelvic inflammatory disease (PID) is a clinical syndrome involving female upper genital tract (UGT) inflammation following pathogen ascent from the lower genital tract. The sexually transmitted infections (STIs), Chlamydia trachomatis (chlamydia) and Neisseria gonorrhoeae (gonorrhoea) are well-known causes of PID1 with an aetiological role established for Mycoplasma genitalium (MG).2 A range of bacteria, many endogenous to the vaginal microbiota or occurring in increased concentrations with bacterial vaginosis (BV) have been isolated from the UGT of women with PID.1

Despite this array of microbial species, recent studies in high-income countries have not detected a causative pathogen in a substantial proportion of PID. In the USA, 17% of emergency department PID had chlamydia and/or gonorrhoea3 and 23% of PID in an inpatient/outpatient treatment trial had chlamydia, gonorrhoea and/or MG.4 In the UK, a case–control study found no aetiological agent in 64% PID.5 Although PID features vary by microbial cause1 ,4 its characteristics without an identified infectious pathogen have been infrequently reported.

In a recent analysis of Australian sexual health clinic (SHC) data examining the association between chlamydia or gonorrhoea and PID we detected neither infection in over half of PID cases,6 prompting questions whether the characteristics of PID without an identified pathogen differed to PID with a pathogen and if this information might enhance patient triage. In this study, we sought to assess the characteristics associated with PID in women where no concurrent STI or BV was detected.


We conducted a cross-sectional study using retrospective routinely collected SHC data. New female, non-sex worker patients between January 2006 and June 2013, aged 16–49 years, diagnosed with PID and tested for chlamydia, gonorrhoea, MG and BV were eligible. The study population here which consisted of 204 (pathogen-negative-PID) cases and 126 (pathogen-positive-PID) controls would allow us to detect an OR of 2 at a significance level of 5% with 85% power, assuming that 50% of (pathogen-positive-PID) controls were exposed to the risk factor of interest.

Melbourne Sexual Health Centre is the major public SHC in the Australian state of Victoria. Attendees assessed at high sexual risk via their sexual history, self-reported risk behaviours, symptoms suggesting an STI, STI contacts or women with pelvic pain are triaged in. Chlamydia testing is offered to new female patients. Specific signs/symptoms guide testing women for other STIs: gonorrhoea for symptomatic women with vaginal discharge, cervicitis or suspected PID; MG for mucopurulent cervicitis, MG contacts or suspected PID; BV for vaginal discharge or suspected PID and trichomonas for symptomatic, high-risk women. High vaginal swabs (HVSs) for microscopy are indicated by vaginal discharge, odour, pain or intermenstrual bleeding.

The primary study outcome was ‘pathogen-negative-PID’ (a PID diagnosis where chlamydia, gonorrhoea, MG and BV tests conducted during the visit were negative). Pathogen-negative-PID cases were compared with pathogen-positive-PID (PID and testing positive for at least one of chlamydia, gonorrhoea, MG, BV). Clinical PID diagnosis was guided by minimum criteria of uterine, cervical motion or adnexal tenderness in sexually active women with pelvic pain.7 STI diagnoses were laboratory confirmed: chlamydia by strand displacement amplification from urine, HVS and cervical swabs; gonorrhoea by microscopy of Gram-stained cervical swab and culture; MG by polymerase chain reaction on urine, HVS or cervical swab and BV by microscopy, predominantly a Nugent score of 7–10 or 4–6 with clue cells. Presence of vaginal inflammation was assessed by HVS polymorphonuclear leucocyte (PMNL) count and defined as ≥5 PMNL per 1000 high-powered field.

Demographic, sexual behaviour (condom use, number of male sexual partners (MSPs)), clinical information (intrauterine device (IUD) in situ, PMNL, chlamydia, gonorrhoea, MG, BV results) were extracted from the electronic record. Records with incomplete sexual behaviour (n=9) or no PMNL (n=1) were excluded. Trichomonas results were extracted where available.

The Alfred Health Human Research Ethics Committee (EC00315) provided study approval.

Statistical analysis

Age, number of MSP, unprotected sex with non-regular MSP, IUD in situ and presenting as an STI contact were identified a priori as potential confounders. Univariable and multivariable logistic regression was conducted to identify characteristics associated with pathogen-negative-PID. As clinical PID diagnosis can vary by clinician8 we accounted for clustering within clinician by obtaining robust variance estimates that adjust for within-cluster correlation.9 Results are reported as OR with 95% CIs. Analyses were conducted using STATA V.13.0.


During the study period, PID was diagnosed in a total of 330 new patients who were tested for chlamydia, gonorrhoea, MG and BV; 312(95%) had trichomonas tests. Of these 330 women, 204 (61.8%; 95% CI 56.3% to 67.1%) had pathogen-negative-PID and 126 had pathogen-positive-PID (49.2% (95% CI 40.4% to 58.1%) had chlamydia, 6.3% (95% CI 2.0% to 10.7%) had gonorrhoea, 11.9% (95% CI 6.2% to 17.6%) had MG and 56.3% (95% CI 47.6% to 65.1%) had BV). Two women tested trichomonas positive (one each from the pathogen-negative and pathogen-positive-PID groups).

Compared with pathogen-positive PID, women with pathogen-negative PID (table 1) were more likely to be aged ≥30 years (AOR 1.7, 95% CI 1.0 to 3.0) and less likely to have vaginal inflammation (AOR 0.5, 95% CI 0.3 to 0.9) or report recent unprotected sex (AOR 0.6, 95% CI 0.4 to 1.0).

Table 1

Characteristics of the study population and associations for pathogen-negative PID


This study found no aetiological agent in 62% of women with clinically diagnosed PID. Similar levels of PID without an identified pathogen have been reported.3–5 Women in this study, with clinical evidence of PID but no pathogen were older, less likely to have vaginal inflammation, or report recent unprotected sex than women with chlamydia, gonorrhoea, MG and/or BV-associated PID. We also found no difference between pathogen-negative and pathogen-positive-PID regarding an IUD in situ or presenting as STI contacts.

The high proportion of pathogen-negative-PID here raises two key questions. First, are these cases PID or are they false positive diagnoses? It is well known that clinical PID diagnosis is imprecise, which is our study's main limitation. There is no specific non-invasive diagnostic test or physical feature both sensitive and specific to PID diagnosis1 which can also vary by clinician and experience.8 Pelvic pain is a key PID characteristic but its duration, intensity and cause can vary. Clinicians must consider whether the sexual history, symptoms and signs suggest PID or another diagnosis. Laparoscopy is often considered the diagnostic gold standard but is invasive and impractical for most outpatient settings and can miss early infection.1 Minimum clinical diagnostic criteria of uterine, cervical motion or adnexal tenderness in sexually active women with recent onset pelvic pain is the mainstay of PID diagnosis and initiating mild–moderate PID treatment internationally.7 Local guidelines in this SHC are based on these minimum criteria7 and aim to promote a low threshold for prescribing PID pharmacotherapy, minimise PID cases that are missed and prevent future reproductive morbidity. This may result in overtreatment, but it has been argued PID overdiagnosis is better than underdiagnosis (providing other diagnoses are excluded) as the risks of antibiotics are potentially less than from possible sequelae associated with ongoing untreated infection.1 ,7 Although we found a lower sexual risk in pathogen-negative-PID, the overall study population was high risk; representing <2% of new female patients during the study and over half (52%) reporting three or more MSP in a year compared with 3.4% women surveyed nationally.10 In this high-risk population, clinicians may have been sensitive to PID risk and consequently overcautious when diagnosing PID, although overdiagnosis may have varied by clinician experience.8

Second, are these PID cases caused by an unidentified pathogen or past infection? Respiratory, enteric1 or novel bacteria11 have occasionally been isolated in non-gonococcal/non-chlamydial PID as have chlamydia or MG from the UGT of salpingitis cases without cervical infection.12 Further, a past STI has been associated with increased PID risk.5 Although guidelines chiefly recommend testing suspected PID cases for chlamydia and gonorrhoea7 there is a need to understand via clinical or genetic microbiome studies11 the potential for other known or novel pathogens to ascend to the UGT and cause symptoms.

The main characteristic differentiating women with pathogen-negative and pathogen-positive-PID was reduced evidence of vaginal inflammation. Other studies have assessed purulent vaginal/cervical discharge as PID diagnostic tests, but, with mixed results their use in improving diagnostic certainty has not been shown.13 Uncertainty around clinical PID diagnosis highlights the need for objective diagnostic methods, with development of sensitive and specific non-invasive biomarkers for female-UGT infection a research priority.1 Promisingly, different ribonucleic acid biosignatures (many involved in inflammation) were recently identified in peripheral blood of a small sample of PID cases.14

This study's key strength is that pathogen-negative and pathogen-positive-PID cases are from the same SHC, limiting variability in practice between patient group for STI/BV and PID diagnoses. The analysis is further strengthened by accounting for correlation in clinical practice within clinicians.

These findings highlight an area of considerable uncertainty around PID diagnosis and aetiology. Pathogen-negative-PID could be because: (i) these women do not have an STI or PID; (ii) an STI had cleared from the cervix but had ascended to the UGT causing PID or (iii) it is PID caused by an unidentified pathogen or past STI. Until we have well-validated rapid biomarkers that improve precision of clinical PID diagnosis, it would be prudent to maintain the basis for PID diagnosis and treatment on clinical features and sexual risk rather than microbiological findings.


The authors would like to thank Mr Afrizal Afrizal and Mr Jun Kit Sze for the data extraction.


View Abstract


  • Handling editor Jackie A Cassell

  • Contributors JLG collected the data, conducted the analysis and wrote the manuscript. AMDL supervised the analysis and contributed to the manuscript. JH, RG, CKF, CB and MYC contributed to the design and/or interpretation, provided feedback and contributed to the manuscript. All authors approved the final submitted version of the manuscript.

  • Funding JLG is supported by an Australian Postgraduate Award from the University of Melbourne.

  • Competing interests None declared.

  • Ethics approval Alfred Health Human Research Ethics Committee (EC00315).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.