Women with mild or moderate PID can be treated in primary care

• There is evidence from a large randomized trial that women with mild-to-moderate PID can be safely treated at home. Outpatient and inpatient treatments are equally effective for women with clinically mild-to-moderate PID, and there is evidence that long-term reproductive outcomes are also similar.

Importance of excluding pregnancy

• National guidance from the Royal College of Obstetrics and Gynaecology stresses the importance of excluding an ectopic pregnancy in all women suspected of having PID [RCOG, 2009].

Referral for contact tracing and screening

• These recommendations are based on guidance from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009] and the British Association for Sexual Health and HIV [BASHH, 2005].

Analgesia

• This recommendation is based on expert advice in the European guideline for the management of pelvic inflammatory disease [Ross et al, 2008] and the British Association for Sexual Health and HIV [BASHH, 2005].

Importance of prompt treatment

• It is likely that delaying treatment increases the risk of the PID becoming worse, and increases the risk of complications such as ectopic pregnancy, subfertility, or chronic pelvic pain occurring in the future [BASHH, 2005; CDC, 2006; Ross, 2008; RCOG, 2009].
• A low threshold for commencing an empirical antibiotic (that is, before the results of tests are available) is recommended because of the risk of serious complications if treatment is delayed. This is particularly important in women younger than 25 years of age because there is a higher incidence of PID in this age group.
• Negative swabs do not exclude PID and should not necessarily influence the decision to treat.
• There is limited evidence from a case-controlled study that fertility may be impaired and ectopic pregnancy may be more likely if treatment for acute pelvic inflammatory disease is delayed for 3 days or more after the onset of symptoms.

Recommended antibiotic regimens

• The antibiotic regimens included are those which are practical for use in primary care (that is, they do not require intravenous antibiotics) and are recommended in guidelines produced by the Royal College of Obstetricians and Gynaecologists and the British Association for Sexual Health and HIV [BASHH, 2005; RCOG, 2009].
• These guidelines recognize that there is uncertainty about the optimum treatment regimen [RCOG, 2009] due to limited evidence from clinical trials. However, the recommended regimens are likely to be effective against the most likely pathogens involved (Chlamydia trachomatis, Neisseria gonorrhoeae, anaerobes).
• There is a better evidence base for the use of cefoxitin than ceftriaxone but, as it is not easily available in the UK, intramuscular ceftriaxone is recommended in its place [RCOG, 2009].
• Metronidazole is included in the intramuscular ceftriaxone plus oral doxycycline regimen to improve coverage for anaerobic bacteria.
• Ofloxacin and azithromycin should be avoided in women at high risk of gonorrhoea because of increasing resistance in the UK [BASHH, 2005; BASHH, 2005].
• Although PID presenting in primary care may be less severe than in other settings, there is no published evidence to support the use of less intensive regimens.
• Recent guidance from the Royal College of Obstetricians and Gynaecologists recommends the combination of intramuscular ceftriaxone and oral azithromycin but states that clinical trial evidence for this regimen is less strong [RCOG, 2009]. There is limited evidence from two randomized controlled trials that azithromycin used alone or in combination with ceftriaxone is effective in the treatment of PID.
• Oral cefixime as a single dose (off-label use) is recommended as an alternative to the intramuscular ceftriaxone component of the regimens by the Health Protection Agency, for practical issues of administration in primary care [HPA and Association of Medical Microbiologists, 2010].
• Ceftriaxone is included in the regimen primarily to cover gonorrhoea. A single dose of oral cefixime 400 mg is an alternative treatment option recommended for gonorrhoea by British Association for Sexual Health and HIV (BASHH) [BASHH, 2005], a primary care guideline [RCGP and BASHH, 2006], and a European guideline [Bignell, 2009].

Antibiotic regimens not recommended

• A combination of metronidazole and doxycyline (without intramuscular ceftriaxone). Although this regimen has been recommended in the past and is still used in primary care it is not recommended in the guidelines produced by the British Association for Sexual Health and HIV or the Royal College of Obstetricians and Gynaecologists [BASHH, 2005; RCOG, 2009]. Guidelines from the Royal College of Obstetricians and Gynaecologists state that there are no clinical trials adequately assessing the effectiveness of this regimen, and its use in isolation (that is, without intramuscular ceftriaxone) is not recommended [RCOG, 2009]. There is evidence from a retrospective study and a systematic review that metronidazole combined with doxycycline is not suitable for the treatment of acute PID in the UK. The addition of intramuscular ceftriaxone as an initial dose significantly improved the clinical cure rate in the one retrospective study [Piyadigamage and Wilson, 2005].
• Azithromycin used alone (that is, without intramuscular ceftriaxone). Data supporting its use alone are limited and it is not recommended [RCOG, 2009].
• Moxifloxacin alone. There is evidence from one high quality randomized controlled trial that moxifloxacin has similar efficacy to a combination of ofloxacin and metronidazole. CKS does not recommend its use as it is a black triangle drug (under intensive surveillance by the Medicines and Healthcare products Regulatory Agency [MHRA]), and there are concerns that it may have similar issues with gonococcal resistance to ofloxacin. In addition, although moxifloxacin is now licensed for the treatment of PID, the Medicines and Healthcare products Regulatory Agency (MHRA) has restricted its indications to second-line use to when other medicines cannot be prescribed or have failed. This is because of an increased risk of life-threatening liver reactions and other serious risks associated with its use [MHRA, 2008; MHRA, 2011 ].

Drug choice during very early pregnancy (before a positive pregnancy test)

• Experience suggests that paracetamol can be used at any time during pregnancy or breastfeeding [NTIS, 2004; Schaefer et al, 2007].
• The UK Teratology Information Service (UKTIS, formerly the National Teratology Information Service [NTIS]), reviewed safety data of NSAIDs from published research and postmarketing surveillance systems. They concluded that 'the available data do not indicate that exposure to ibuprofen before 30 weeks of pregnancy is associated with an increased risk of congenital defects or spontaneous abortions' [NTIS, 2004].
• The Royal College of Obstetricians and Gynaecologists recommends that the risk of giving any of their recommended antibiotic regimens in very early pregnancy (before a positive pregnancy test) is low, since significant drug toxicity results in failed implantation [RCOG, 2009].
• Ceftriaxone has not specifically been studied during pregnancy. However, the risk associated with use of cephalosporins during pregnancy is thought to be low and, although data are limited for individual agents such as ceftriaxone, the cephalosporins as a class are considered to be an appropriate choice during pregnancy [NTIS, 2008]. Although ceftriaxone crosses the placental barrier, it has not been associated with adverse events on fetal development in laboratory animals [ABPI Medicines Compendium, 2009].
• Doxycycline is contraindicated beyond the 15th week of gestation because, from the 16th week of pregnancy it causes tooth and bone discolouration and inhibits bone growth. However, it can be used in the first trimester. Inadvertent first-trimester use of tetracyclines occurs frequently and has not been associated with an increased risk of congenital malformations [Schaefer et al, 2007].
• Metronidazole has been in clinical use for a long time, and experience suggests that it is not teratogenic in humans [Schaefer et al, 2007]. A recent prospective controlled study in 228 women exposed to metronidazole in pregnancy, 86% of whom had first trimester exposure, confirms these findings [Schaefer et al, 2007].
• Ofloxacin has only limited pregnancy-exposure data. Quinolones have caused arthropathy in animal studies. However, a recent study (most of the data are on ciprofloxacin and norfloxacin, but some are on ofloxacin) did not find that quinolone use in the first trimester of pregnancy was associated with an increased risk of malformations or other adverse effects on pregnancy outcome [Schaefer et al, 2007].
• There are fewer published data on the use of azithromycin rather than erythromycin during pregnancy and breastfeeding. The limited published data and follow-up data collected by the UK Teratology Information Service (UKTIS, formerly the National Teratology Information Service [NTIS]), do not demonstrate an increased risk of congenital malformations following exposure to azithromycin in human pregnancy [NTIS, 2008].

Management of women with HIV

• Women with HIV may experience more severe symptoms of PID, but will usually respond to the standard antibiotic regimens. No change in treatment from that recommended for women without HIV infection is required [BASHH, 2005]. There is no evidence to suggest that HIV-positive women benefit from hospitalization or parenteral treatment [Sweet, 2009].

Urgent admission

• These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009], the British Association for Sexual Health and HIV [BASHH, 2005a], the European guideline for the management of pelvic inflammatory disease [Ross et al, 2008], and guidelines from the Department of Health and Human Services Centres for Disease Control and Prevention [CDC, 2006].

Pregnant women

• If the pregnancy is intrauterine, then pelvic inflammatory disease (PID) is rare except in the case of septic abortion [RCOG, 2009].
• The recommendation that pregnant women with PID should be admitted to hospital under the care of an obstetrician is based on expert opinion, as intravenous antibiotics are required because of the increased risk of maternal and fetal morbidity and pre-term delivery [BASHH, 2005a; CDC, 2006; Ross et al, 2008; RCOG, 2009].
• Neonatal complications can occur as a result of perinatal transmission of infection, such as ophthalmia neonatorum (due to Chlamydia trachomatis or Neisseria gonorrhoeae infection) and chlamydial pneumonitis [Brocklehurst and Rooney, 1998].

Women with HIV

• Women with HIV may experience more severe symptoms of PID. Expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists and the British Association for Sexual Health and HIV [BASHH, 2005a; RCOG, 2009] recommend that only women with severe disease should be admitted.

Suspected peri-hepatitis

• Guidelines from the Royal College of Obstetricians and Gynaecologists and the British Association for Sexual Health and HIV [BASHH, 2005a; RCOG, 2009] state that although laparoscopic division of adhesions is sometimes performed, there is insufficient trial evidence to make any recommendations other than giving the usual treatment for PID. CKS therefore recommends seeking specialist advice if peri-hepatitis is suspected.

These recommendations are based on expert advice in guidelines from the Royal College of Obstetricians and Gynaecologists, the British Association for Sexual Health and HIV [BASHH, 2005a], and the European guideline for the management of pelvic inflammatory disease [Ross et al, 2008].

Complications of pelvic inflammatory disease (PID)

• Tubal infertility:
• There is evidence from prospective cohort studies that the risk of infertility is related to the number of episodes of pelvic inflammatory disease (PID) and their severity: 0.6% of women had tubal-factor infertility after an episode of mild PID, and 21.4% of women had tubal-factor infertility after an episode of severe PID.
• The risk of ectopic pregnancy is increased ten-fold after one episode of PID [Oakeshott, 2003].
• Estimates vary but chronic pelvic or abdominal pain (lasting for more than 6 months) develops in 18% of women who have had PID [Drife and Magowan, 2004]. There is evidence from a large randomized trial of 831 women who have had an episode of mild-to-moderate PID that chronic pelvic pain may be present in 34% of women at a mean follow up of 35 months [Ness et al, 2002].
• About a third of women have repeated infections [Drife and Magowan, 2004].

Discontinuation of metronidazole if it is not tolerated

• Although anaerobes may have a role in the pathogenesis of PID, they are probably of more importance in women who have severe PID. Expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009], the British Association for Sexual Health and HIV [BASHH, 2005a], and the European guideline for the management of pelvic inflammatory disease [Ross et al, 2008], based on studies which have not included metronidazole but have had good outcomes, is that if the PID is not severe and metronidazole is not tolerated then it may be stopped.

Referral for contact tracing and screening

• These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009] and the British Association for Sexual Health and HIV [BASHH, 2005a].

Avoiding unprotected sexual intercourse

• These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009] and the British Association for Sexual Health and HIV [BASHH, 2005a].

• These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009] and the British Association for Sexual Health and HIV [BASHH, 2005a].

These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009], the British Association for Sexual Health and HIV [BASHH, 2005a], and the Department of Health and Human Services — Centres for Disease Control and Prevention [CDC, 2006].

Admission, if the woman is failing to improve

• Failure to improve substantially after 72 hours suggests a need for further investigation, intravenous therapy, or surgical intervention [BASHH, 2005a].

Discontinuation of metronidazole if it is not tolerated

• Although anaerobes may have a role in the pathogenesis of pelvic inflammatory disease (PID), they are probably of more importance in women who have severe PID. Expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009], the British Association for Sexual Health and HIV [BASHH, 2005a], and the European guideline for the management of pelvic inflammatory disease [Ross et al, 2008], based on studies which have not included metronidazole but have had good outcomes, suggests that if the PID is not severe and metronidazole is not tolerated then it may be stopped.

Removal of an intrauterine device (IUD)

• Evidence on whether or not to remove an intrauterine device in a woman who has pelvic inflammatory disease is limited and conflicting. There are no long term data on the effects on fertility.
• Expert opinion differs regarding women with pelvic inflammatory disease (PID) who have an IUD:
• Expert advice in the Faculty of Sexual and Reproductive Healthcare clinical guideline on intrauterine contraception does not routinely recommend the removal of an IUD. The Faculty's Clinical Effectiveness Unit supports the continued use of intrauterine contraception and appropriate antibiotic treatment if PID is suspected; there is no need to remove the IUD unless symptoms fail to resolve within 72 hours or the woman wants it removed [FFPRHC, 2006; FSRH, 2007].
• Expert opinion in guidelines on the management of acute PID from the Royal College of Obstetricians and Gynaecologists [RCOG, 2009] advises that consideration be given to removing the IUD especially if symptoms have not resolved within 72 hours.
• The British Association for Sexual Health and HIV recommend considering removing the IUD if the woman develops PID. They advise balancing the decision to remove the IUD against the risk of pregnancy if the woman has had sexual intercourse in the preceding 7 days [BASHH, 2005a].
• Selected practice recommendations for contraceptive use from the World Health Organization state that [WHO, 2004]:
• There is no need for removal of the IUD if the woman wishes to continue its use.
• If the woman wishes removal, remove it after antibiotic treatment has been started.
• If the infection does not improve then generally the course would be to remove the IUD and continue antibiotics. If the IUD is not removed then the antibiotic should be continued and the woman should be monitored closely.

Presence of actinomyces-like organisms (ALOs)

• Expert opinion in guidelines from the Faculty of Sexual and Reproductive Healthcare is that [FSRH, 2007]:
• The role of ALOs in infection in women using intrauterine contraception is unclear. Actinomyces israelii is a commensal organism in the female genital tract and although these organisms may be found on cervical smears or swabs, their presence is not diagnostic or predictive of disease. Therefore, there is no need to remove an IUD if the woman does not have symptoms.
• If PID is suspected in a woman who has a history of ALOs on a cervical smear, it is important to consider that the infection may be due to other organisms.
• It may be appropriate to remove the IUD.

Choice of contraception

• These recommendations are based on the UK Medical Eligibility Criteria (UKMEC) for contraceptive use [FFPRHC, 2006] and a guideline from the Faculty of Sexual and Reproductive Healthcare [FSRH, 2007].

Intrauterine device (IUD)

• A review of the World Health Organization's experience of pelvic inflammatory disease (PID) associated with IUD use from around the world included 12 randomized studies and one non-randomized study, and found that [FSRH, 2007]:
• Amongst 22,908 IUD insertions and during 51,399 woman-years of follow up, the overall rate of PID was 1.6 cases per 1000 years of use.
• The risk of PID was six-fold higher during the 20 days after IUD insertion than during later times.

Testing for bacterial vaginosis

• CKS recommends taking a high vaginal swab to look for other vaginal infections such as bacterial vaginosis.

Recommendation on antibiotic prophylaxis

• The recommendation to test for sexually transmitted infections (STIs), or prescribe prophylactic antibiotics if testing for STIs is not possible or has not been completed, before IUD insertion is based on expert opinion in guidelines from the National Institute for Health and Clinical Excellence [NICE, 2005] and the Faculty of Sexual and Reproductive Healthcare [FSRH, 2007] as the risk of PID following insertion of an IUD where infection is present is unknown.
• Both guidelines recommend testing for STIs and prescribing prophylactic antibiotics if testing for STIs is not possible or has not been completed before an IUD is inserted in women at increased risk of STIs. Woman who have a history of PID are at increased risk of STIs.
• However, there is good evidence from a Cochrane systematic review that the use of doxycycline or azithromycin orally before IUD confers little benefit even in populations with a high prevalence of STIs.

Sterilization

• These recommendations are based on the UKMEC for contraceptive use [FFPRHC, 2006].

Protection against STIs

• Evidence supports the use of male or female condoms to reduce the risk of several sexually transmitted infections (STIs). For more information see the section on Barrier methods in preventing STIs in the CKS topic on Contraception. However, even with consistent and correct use, infections may still (rarely) be transmitted [FFPRHC, 2007a].

Other investigations in secondary care

• Transvaginal ultrasound is most useful in detecting large tubal swellings or fluid collections which only occur in women with severe pelvic inflammatory disease (PID) in whom irreversible tubal damage may already have occurred [Ross, 2003]. Power Doppler ultrasound is able to detect changes in blood flow associated with the hyperaemia that occurs with tubal inflammation [Ross, 2003].
• The recommendation that transvaginal ultrasound scanning supported by power Doppler may be helpful is based on expert opinion in a guideline from the Royal College of Obstetricians and Gynaecologists who also commented that there is limited evidence that magnetic resonance imaging and computerized tomography can assist in making a diagnosis [RCOG, 2009].
• Laparoscopy may provide information on the severity of the condition. However there is potential difficulty in identifying mild intra-tubal inflammation or endometritis and there is diagnostic variability between clinicians [Ross, 2003; BASHH, 2005a; RCOG, 2009].
• Endometrial biopsy may be helpful, but there is insufficient evidence to support its routine use [Ross et al, 2008].

Surgical treatment

• These recommendations are based on expert opinion in guidelines from the Royal College of Obstetricians and Gynaecologists and the British Association for Sexual Health and HIV [BASHH, 2005a; RCOG, 2009].
• The British Association for Sexual Health and HIV [BASHH, 2005a] comment that although it is possible to perform adhesiolysis in women with peri-hepatitis there is no evidence that this is superior to antibiotic treatment alone.