Background Pelvic inflammatory disease (PID) occurs when pathogens (often sexually transmitted) ascend from the cervix to the upper genital tract. No pathogen is detected in up to two thirds of PID cases yet few studies have assessed the characteristics of pathogen-negative PID. We assessed the characteristics of pathogen-negative compared to pathogen-positive PID using data from females attending a large urban sexual health clinic in Melbourne.

Methods Data were extracted from the clinic’s electronic patient database for women aged 16–49 and tested for chlamydia (CT), gonorrhoea (NG), mycoplasma genitalium (MG) and bacterial vaginosis (BV) and diagnosed with PID at first clinic visit between Jan 2006–June 2013. PID diagnosis was clinical: based on uterine, cervical motion, or adnexal tenderness in sexually active women with pelvic pain where no cause beside PID was identified. Multivariable logistic regression was conducted to identify demographic, sexual behavioural factors and whether vaginal inflammation (defined as ≥5 polymorphonuclear leukocytes (PMNL) per 1000 high powered field from high vaginal swabs) was associated with pathogen- negative PID.

Results Between 2006–2013, a total of 326 new females patients were diagnosed with PID and tested for CT, NG, MG and BV; 203 (62%; 95% CI: 57%–68%) tested negative for the four infections (pathogen-negative PID). Among pathogen-positive PID cases, 49.6% (95% CI: 40.6–58.6) had CT, 6.5% (95% CI: 2.1–10.9) had NG, 12.2% (95% CI: 6.3–18.1) had MG, 56.1% (95% CI: 47.2–65.0) had BV. Multivariate analysis showed that pathogen-negative PID cases were less likely to have vaginal inflammation (OR = 0.5, 95% CI: 0.3–0.8), to report unprotected sex with non-regular sexual partners in the past three months (OR = 0.6, 95% CI: 0.4–1.0) or to present as an STI contact (OR = 0.4, 95% CI: 0.2–1.0).

Conclusion These findings suggest that pathogen-negative PID is associated with a lower sexual risk and inflammation then pathogen-positive PID cases. In the clinical setting, consideration of PMNS in diagnosing PID may improve diagnostic precision.

Disclosure of interest statement The authors declare that they have no commercial or other association that might pose a conflict of interest.