Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors

INTRODUCTION — Pelvic inflammatory disease (PID) refers to acute infection of the upper genital tract structures in females, involving any or all of the uterus, oviducts, and ovaries; this is often accompanied by involvement of the neighboring pelvic organs. By convention, PID is initiated by a sexually transmitted agent, which ascends into the upper genital tract, distinguishing it from pelvic infections caused by trans-cervical medical procedures, pregnancy, and other primary abdominal processes that can extend to pelvic organs.

The prevalence of PID in the United States and many other resource-rich countries had been decreasing in the early 2000s [1,2], but subsequent evidence suggests a plateau and potentially an increase in prevalence trends [3,4]. In the United States, around 4 percent of females aged 18 to 44 years report having had PID; it accounts for approximately 90,000 outpatient visits every year and is a similarly frequent cause of emergency department visits [4].

The pathogenesis and microbiology of PID as well as risk factors for PID will be reviewed here. The clinical features, diagnosis, treatment, and sequelae of this disorder are discussed separately. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis" and "Pelvic inflammatory disease: Treatment in adults and adolescents" and "Pelvic inflammatory disease: Long-term complications".)

PATHOGENESIS — The vaginal flora of most normal, healthy females includes a variety of potentially pathogenic bacteria [5]. Among these are species of Prevotella, Leptotrichia, Atopobium, and other anaerobes [6]. Compared with the dominant, non-pathogenic, hydrogen peroxide-producing Lactobacillus species, these other organisms are present in low numbers, and ebb and flow under the influence of hormonal changes (eg, pregnancy, menstrual cycle), contraceptive method, sexual activity, vaginal hygiene practices, and other as yet unknown factors.

The endocervical canal functions as a barrier protecting the normally sterile upper genital tract from the organisms of the dynamic vaginal ecosystem. Endocervical infection with sexually transmitted pathogens can disrupt this barrier. Disturbance of this barrier provides vaginal bacteria access to the upper genital organs, infecting the endometrium, then endosalpinx, ovaries, pelvic peritoneum, and their underlying stroma. The resulting infection may be subclinical or manifest as the clinical entity of pelvic inflammatory disease (PID). The reasons why lower genital tract bacteria cause PID in some females but not others is not fully understood but may relate to genetic variations in immune response, estrogen levels affecting the viscosity of cervical mucus, and the bacterial load of potential pathogens [7,8].

Patients with PID can present with clinical disease at any point along a continuum from endometritis (with normal tubes, ovaries, and peritoneum) to salpingitis (with inflammation of the fallopian tubes and adjacent pelvic structures). This limits the sensitivity of direct visualization of the fallopian tubes through laparoscopy when making a diagnosis of PID because endometritis and mild intratubal inflammation cannot be seen [9,10]. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis".)

MICROBIOLOGY — Neisseria gonorrhoeae and Chlamydia trachomatis are commonly identified pathogens in pelvic inflammatory disease (PID) among sexually active pre-menopausal females. Mycoplasma genitalium is also likely to be a cause in the pre-menopausal group [11]. E. coli and colonic anaerobes may be responsible for the rare cases of PID seen in post-menopausal females. Very rare pathogens identified include Mycobacterium tuberculosis, Haemophilus influenzae, Streptococcus pneumoniae, and the agents of actinomycosis. However, in most cases, the precise microbial etiology of PID is unknown. Regardless of the initiating pathogen, PID is clinically considered a mixed polymicrobial infection.

Neisseria gonorrhoeae — N. gonorrhoeae was the first identified cause of PID. (See "Epidemiology and pathogenesis of Neisseria gonorrhoeae infection".)

Approximately 15 percent of females with an endocervical N. gonorrhoeae infection go on to develop PID [12,13]. The proportion of PID caused by N. gonorrhoeae varies widely and reflects the underlying prevalence of gonorrhea in the local population. Its importance tends to be higher in the southeastern United States, less in the northwestern United States, and much less in western Europe [2,14]. Gonococcal PID has also been reported more commonly among certain indigenous populations [15]. There has been a steady evolution of multidrug-resistant strains since the early 1980s, and treatment options are increasingly limited. (See "Treatment of uncomplicated gonorrhea (Neisseria gonorrhoeae infection) in adults and adolescents" and "Pelvic inflammatory disease: Treatment in adults and adolescents".)

Gonococcal PID tends to be clinically more severe than chlamydia PID [16], which may lead to earlier diagnosis and treatment. (See "Pelvic inflammatory disease: Clinical manifestations and diagnosis".)

Chlamydia trachomatis — Genital chlamydia is the most common bacterial sexually transmissible disease. In the United States, it is also the most common reportable infectious disease, with over 1.5 million cases reported annually [17]. C. trachomatis accounts for about one-third of cases of PID, with less geographic variation in the prevalence than that seen for gonorrhea-associated PID.

As seen with genital gonococcal infections in females, about 10 to 15 percent of endocervical C. trachomatis infections produce PID, but asymptomatic subclinical infections are also common, and these may present years later as chronic pelvic pain or infertility. (See "Clinical manifestations and diagnosis of Chlamydia trachomatis infections", section on 'Pelvic inflammatory disease'.)

Females aged 16 to 24 years account for most cases of chlamydia. Many, but not all studies have shown a reduction in PID rates following the introduction of chlamydia screening in young females. This is discussed in greater detail elsewhere. (See "Screening for sexually transmitted infections", section on 'Females'.)

Mycoplasma genitalium — The role of M. genitalium in sexually transmitted infections among females is emerging. There is growing evidence supporting an association with cervicitis and PID in females. The proportion of PID cases that are associated with M. genitalium is uncertain; in one study of females with mild to moderate PID in the United Kingdom, 10 percent tested positive for M. genitalium [18-20]. (See "Mycoplasma genitalium infection", section on 'Pelvic inflammatory disease'.)

Other initiating pathogens — Other agents that initiate PID, while almost certainly sexually transmitted and probably infectious, remain obscure. Using molecular amplification with generic primers, a number of novel bacteria have been identified in the fallopian tubes of females with PID, including Atopobium, Sneathia, and Leptotrichia [21]. The precise role of these and other anaerobic bacteria in the pathogenesis of PID remains unclear, but is likely consistent with their role in the anaerobic microbiology of bacterial vaginosis. Evidence suggesting that anaerobic coverage in PID treatment improves outcomes supports the potential pathogenic role of anaerobes [22].

Mixed infection — Regardless of the initiating pathogen, the microbiology of PID, especially for clinical purposes, should be viewed and treated as a mixed (facultative and anaerobic) polymicrobial infection. Older studies isolated groups A and B streptococci (rarely enterococci), E. coli, Proteus mirabilis, Haemophilus spp, Bacteroides/Prevotella spp, Peptococcus, and Peptostreptococcus spp from females with PID [23-25]. These studies found that, among cases of PID initiated by N. gonorrhoeae, a mixed polymicrobial infection was seen in approximately 35 percent. Another study, which employed particularly stringent microbiologic techniques, identified other organisms in more than 50 percent of patients with gonococcal PID [26].

In a given patient with PID, different organisms can be isolated from the various levels of the genital tract (ie, the organisms isolated from the upper genital tract may be distinct from those isolated from the lower genital tract) [23,25-28].

RISK FACTORS — Sex is the primary risk factor for pelvic inflammatory disease (PID). Sexually abstinent females are not at risk for PID [29], and females with longstanding monogamous relationships rarely develop PID. On the other hand, females with multiple sexual partners are at the highest risk. Younger age, past infection with chlamydia, a partner with a STI, and previous PID are other important risk factors. The frequency of PID is also affected by the method of contraceptive used. Barrier contraceptives (when used correctly) are protective.

African-American or Black-Caribbean ethnicity has been associated with a higher risk of STIs, including PID, and the reasons for this are likely multifactorial [30,31]. These racial differences in PID risk may be decreasing over time [32].

Multiple partners — The importance of multiple partners was illustrated by a study that compared 712 females hospitalized for PID with 2719 controls [33]. Having four or more sexual partners within the prior six months increased the risk of PID 3.4 times, and having sex with a single partner six or more times per week increased it 3.2 times. Other studies have confirmed multiple partners as a risk factor, associated with an increased frequency of PID ranging from 3- to 20-fold [34-36]. One report, however, did not confirm a role for coital frequency [34].

STI in the partner — Approximately a third of males with gonococcal or chlamydial urethritis are asymptomatic. Having a symptomatic (dysuria, urethral discharge) male partner may increase a female’s risk of PID, most likely because of the associated increase in bacterial load [23].

Age — PID occurs in highest frequency among those 15 to 25 years of age [13,37]; the incidence in females older than the age of 35 years is only one-seventh that in younger females [38]. The initiating pathogens of PID in both the United States and Europe, particularly C. trachomatis, are densely concentrated among adolescent and young adult females, with prevalence around 3 to 5 percent [2,39]. C. trachomatis and N. gonorrhoeae are less likely to be identified in post-menopausal females, in whom the risk of PID is very low. In such females, it is important to consider alternate diagnoses including ovarian cancer, fibroids, diverticulitis, and colorectal cancer [40].

Reinfection with genital C. trachomatis is also a function of age. In one series, for example, the risk of reinfection among females whose first infection occurred at <15 years and 15 to 19 years was eightfold (at 54 percent) and fivefold (at 30 percent) greater, respectively, compared to those whose first infection occurred older than age 30 [41].

Previous PID — Approximately one in four females with PID will suffer recurrence [42]. In one study, a previous episode of PID increased the risk for subsequent episodes by a factor of 2.3 [34]. However, these data must be used cautiously in practice, since PID is associated with an increased risk of subsequent chronic pelvic pain in general, even in the absence of an identifiable new infection.

Contraceptive method — The choice of contraceptive method does not clearly affect the risk of pelvic infection, although consistent and correct use of condoms offers a significant reduction of risk.

Barrier contraception — Barrier contraception protects against PID [43]. Condoms are the most effective form, preventing over 50 percent of endocervical gonococcal and chlamydial infections if used correctly. One large study suggested that consistent condom use following a diagnosis of PID can lower the rate of PID sequelae [44].

However, most females and their partners do not consistently use condoms. According to survey data reported by the United States Centers for Disease Control and Prevention from 2002 to 2007, consistent condom use was reported in only 31 percent of never-married females aged 20 to 24 years [45].

Oral contraceptives — Oral contraceptives (OCs) have a complex interaction with PID. Several studies have shown that OC use nearly doubles the prevalence of both chlamydia and gonococcal infection of the cervix [46]. However, OC use has traditionally been associated with a 50 percent reduction in PID risk [47].

Among OC users with cervical infection, asymptomatic endometritis is fourfold more common than among their counterparts not using OCs [48], although gross salpingitis is reduced fivefold [49]. Thus, females using OCs appear to develop PID about as frequently as other females, but the severity of the infection is substantially diminished.

Intrauterine device and tubal ligation — Modern intrauterine devices (IUDs) cause little, if any, increased risk for PID [50-52]. The risk of PID is primarily limited to the first three weeks after IUD insertion and is associated with the physical introduction of the device; PID is uncommon thereafter [51]. In females who develop PID with an IUD already in place, one methodologically limited trial suggested that IUD removal results in better clinical outcomes [53], but most guidelines recommend leaving the IUD in place while treating acute PID with antibiotics with close follow-up [52]. The IUD should, however, be removed if clinical improvement is delayed beyond a few days.

Long-term indwelling IUDs have been associated with pelvic actinomycosis, a rare disease that can present as a pelvic mass with weight loss and constitutional symptoms [54]. In a study of 475 isolates of actinomyces species, 30 percent of the clinical specimens originated in association with IUDs [55]. The decision to treat a patient for possible pelvic actinomycosis is influenced by the presence or absence of clinical symptoms, since actinomyces are part of normal vaginal flora. This is discussed elsewhere. (See "Intrauterine contraception: Management of side effects and complications", section on 'Actinomyces and related organisms'.)

Tubal ligation may protect the distal oviducts from involvement, but the clinical syndrome of PID is otherwise unaffected.

Other conditions — Complete disruption of the vaginal ecosystem can occur, in which anaerobic bacteria assume predominance over the desirable strains of lactobacilli in the lower genital tract. This condition is known as bacterial vaginosis and affects 15 to 30 percent of American females, one-half of whom are asymptomatic [56]. The presence of bacterial vaginosis likely increases the risk of PID [57], particularly in females with certain subtypes of bacterial vaginosis based on the pattern of bacteria present [58].

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Pelvic inflammatory disease (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Definition – Pelvic inflammatory disease (PID) refers to acute infection of the upper genital tract structures in females, involving any or all of the uterus, oviducts, and ovaries, and sometimes other surrounding pelvic organs. It is caused by a sexually transmitted agent, distinguishing it from pelvic infections caused by medical procedures, pregnancy, and primary abdominal processes. (See 'Introduction' above and 'Pathogenesis' above.)

●Causative pathogens – Initiating pathogens in PID include Neisseria gonorrhoeae, Chlamydia trachomatis, and likely Mycoplasma genitalium, which account for an estimated 30 to 40 percent of all cases, although in the majority of cases the specific microbial etiology of PID is unknown. Regardless of the initiating event, the microbiology of PID should be viewed and treated as a mixed (facultative and anaerobic) polymicrobial infection. (See 'Microbiology' above.)

●Risk factors – The main risk factor for pelvic inflammatory disease is having unprotected intercourse with multiple sex partners. The choice of contraceptive method does not clearly affect the risk of pelvic infection, though consistent and correct use of condoms offers a significant reduction of risk. (See 'Risk factors' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Charles Livengood, MD, who contributed to an earlier version of this topic review.