Aetiology/microbiology

PID has classically been associated with the STIs Neisseria gonorrhoeae and C. trachomatis. Studies in the 1980s and 1990s de­tected chlamydia and/or gonorrhoea from over 50% of cases (15­17) and in up to 77% of tubo-ovarian abscesses at laparoscopy (18).

Identifying which other organisms are implicated in the patho­genesis of PID is problematic. Firstly it must be considered whether the upper female genital tract is normally sterile. There is some evi­dence that some microorganisms may be present in healthy, asymp­tomatic individuals. Even when identified, the role these organisms may play in the disease process is not understood.

Figure 43.2 PID caused by C. trachomatis diagnosed in England 2001-2014.

Source data from https://www.gov.uk/government/statistics/sexually-transmitted-infections-stis-annual-data-tables.

Acute PID is now considered a polymicrobial infection with in­volvement of a wide variety of microorganisms, many of which are not routinely tested for in clinical practice.

The exact role of anaerobic bacteria and BV in the pathogenesis of PID remains unclear. Lactobacilli are important in the host defence against STIs, and BV has been shown to increase the risk of inci­dent gonorrhoea and chlamydial infection (20). BV has also been noted to be frequently present in women with PID (up to 70%). But whether BV-associated bacteria independently cause PID, whether they facilitate the ascension of N. gonorrhoeae or C. trachomatis in­fection, or whether they are simply innocent bystanders following infection with other pathogens are yet to be determined. There are data, however, which suggest anaerobes are capable of causing local tissue damage at the level of the fallopian tube when present (21). The proportion of women with PID who have anaerobes isolated at laparoscopy ranges from 13% to 78% (20). The wide differences are thought to be related to diverse patient populations, different definitions of PID, varying degrees of disease severity, and different microbiological techniques.

Several studies have investigated the role of BV in endometritis or clinically suspected PID. The only prospective study showed no increase in the risk of developing incident PID in women with BV over a 3-year period (22). This study used standard Gram-stained slides and microscopy for the BV diagnosis. However, 1 year later the same authors concluded that women with a heavy growth of BV- associated microorganisms and a new sexual partner appeared to be at particularly high risk of PID (adjusted rate ratio 8.77; 95% confi­dence interval (CI) 1.11-69.2) (23). Several other studies have also found an association with the presence of anaerobic Gram-negative rods, independent of chlamydia or gonorrhoea infection (24, 25). The differences in these findings have been explained by variable definitions of BV. Although BV as determined by Gram stain is rela­tively reliable, it does not capture all the microbiological components of the BV ecosystem, in particular anaerobic Gram-negative rods. So when BV was identified by microbial culture, a combination of BV- related microorganisms was shown to significantly increase the risk of acquiring PID (23). There is substantial evidence that anaerobic infection is associated with more severe disease, in particular the formation of tubo-ovarian abscesses (26).

M. genitalium was first identified in the early 1980s in men with non-gonococcal urethritis. The organism was extremely difficult to isolate from clinical specimens, prompting the development of nucleic acid amplification tests (NAATs) which have improved the understanding of pathogenesis and the association with disease pro­cesses. M. genitalium has been shown to be sexually transmittable in partner studies which demonstrate concurrent infection with con­cordant strain types (27). Infection with M. genitalium is as common as C. trachomatis among high-risk women and women with clinical PID, with rates ranging from 2.1% to 16% (28, 29). Studies have in­creasingly shown an association with cervicitis, PID, and infertility.

2%; P = 0.02) (30). Similarly, in the PID Evaluation and Clinical Health (PEACH) study women testing positive for M. genitalium were more than twice as likely to have histologically confirmed endometritis compared to women without infection (29). More re­cently, a meta-analysis looking at M. genitalium infection and female reproductive tract disease showed this organism is associated with a significantly increased risk of PID, with a pooled odds ratio (OR) of 2.14 (95% CI 1.31-3.49) (31). Women also had an increased risk of cervicitis (OR 1.66; 95% CI 1.35-2.04), preterm birth (OR 1.89; 95% CI 1.25-2.85), and miscarriage (OR 1.82; 95% CI 1.10-3.03). A significant association between M. genitalium and PID has impli­cations for current testing algorithms and recommended therapies which specify the use of antibiotics with poor efficacy against this organism.