Recommendations and Rationale for the Treatment of Pelvic Inflammatory Disease

Oluwatosin Jaiyeoba; Gweneth Lazenby; David E Soper


Expert Rev Anti Infect Ther. 2011;9(1):61-70. 

In This Article

Expert Commentary

Treatment of PID should take into account the short-term goals of clinical and microbiological cure and the long-term goals of prevention of infertility, ectopic pregnancy, recurrent infection and chronic pelvic pain. Optimal antimicrobial regimens should be well tolerated with little to no gastrointestinal side effects and tailored for compliance, for example, single- to short-course regimens administered once a day or even less frequently if possible. Oral therapy is preferred over parenteral therapy for reasons of convenience and cost as well as to avoid the pain of needle injection. Cost is an important consideration and relative expense of the antimicrobials under consideration is noted in Table 1.

Several quinolone antimicrobials (e.g., moxifloxacin, ciprofloxacin and ofloxacin) alone or in combination with other agents have been studied and shown to be effective in the treatment of PID.[19–22] However, ciprofloxacin appeared less effective in clearing BV-associated microorganisms from the endometrium despite the patients' clinical cure.[20] This causes some concern that persistent anaerobic infection could lead to a relapse of endometritis and salpingitis. In addition, owing to increasing N. gonorrhoeae resistance, quinolone use is no longer recommended for the treatment of N. gonorrhoeae in the USA and therefore cannot be considered a primary monotherapy option for the treatment of PID in areas with quinolone-resistant N. gonorrhoeae (QRNG).[23,24] In areas where QRNG is uncommon, moxifloxacin is an antibiotic that should be considered for the treatment of acute PID.

Moxifloxacin has been found in three randomized controlled trials to be an effective and well-tolerated oral treatment in women with acute PID.[25] Ross et al. found that moxifloxacin once-daily monotherapy was as clinically and bacteriologically effective as twice-a-day ofloxacin–metronidazole combination.[22] Heysteck et al. also demonstrated the efficacy of moxifloxacin monotherapy in a comparative study with a combination of doxycycline, metronidazole and ciprofloxacin.[26]

Moxifloxacin rapidly penetrates uterine tissue where it remains in high concentrations and is sufficiently high to eliminate the principal pathogens of PID.[27] In addition, Bradshaw et al. reported that moxifloxacin was able to eradicate M. genitalium in patients who had failed azithromycin treatment.[28,29] Thus, moxifloxacin represents a useful addition to the treatment armamentarium for acute PID in women.

Owing to persistent concern about the current outpatient antimicrobial regimens lacking comprehensive coverage of anaerobic bacteria, metronidazole has been suggested as an addition to doxycycline, particularly for those women with PID and BV. However, the combination of doxycycline and metronidazole has been associated with low clinical and microbiological cure rates of 75 and 71%, respectively, for PID.[24,30] These observations suggest that the combination of doxycycline and metronidazole is a suboptimal choice for an oral antibiotic regimen used to treat PID.[24,31] The low efficacy rates may be due to the poor coverage of these drugs against N. gonorrhoeae, and the addition of an effective anti-gonoccocal antibiotic to the regimen should improve the efficacy rates. Piyadigamage et al. demonstrated that by adding ceftriaxone to doxycycline and metronidazole, there was a significant clinical cure rate improvement from 55 to 72%.[31]

Azithromycin provides excellent coverage of chlamydia and moderate-to-good coverage for a range of aerobic and anaerobic bacteria, including Gram-negative anaerobes.[30,32] It is also at least 100-fold more active in vitro against M. genitalium than the quinolones or tetracyclines.[33,34] Although not included in the updated CDC 2007 guidelines for treatment of PID, several studies suggest that azithromycin can be used for the treatment of acute PID.[30,32] In a randomized controlled trial, cure rates of azithromycin for chlamydial infections, concomitant gonorrhea and gonorrhea only were high (95.8, 84.2 and 98.2%, respectively).[35] Another randomized trial comparing the efficacy and safety of azithromycin as monotherapy or combined with metronidazole showed cure rates of 97.6 and 95.5%, respectively.[36] Malhotra et al. reported a clinical cure rate of 93% among those given a one-time dose of fluconazole, azithromycin and secnidazole.[37] In the macaque animal model, azithromycin was more effective in eradicating chlamydial organisms from the upper and lower reproductive tract tissues than doxycycline, and also exerted a significant anti-inflammatory effect.[38] Azithromycin's advantages over doxycycline include single-dose administration and fewer side effects.[30] However, N. gonorrhoeae resistance to azithromycin has been reported and the higher 2 g dose recommended to treat this pathogen is associated with significant gastrointestinal side effects. Despite these concerns, multidose regimens of azithromycin monotherapy reliably eradicated N. gonorrhoeae when isolated in women with PID in the studies cited. Savaris et al., in a randomized controlled trial conducted in Brazil, showed that when combined with ceftriaxone, 1 g of azithromycin weekly for 2 weeks was equivalent to ceftriaxone plus a 14-day course of doxycycline for treating mild PID.[39] The discussion on azithromycin would be incomplete without discussing azithromycin failure in the treatment of M. genitalium urethritis. A total of 15% of patients experienced treatment failure with azithromycin versus 55% treatment failure with doxycycline.[40] Azithromycin treatment failure of 15–30% has also been reported in patients with M. genitalium urethritis in Australia and the use of moxifloxacin in these cases resulted in rapid symptom resolution and eradicated the infection.[28,29]

Given the above commentary, it appears that optimal regimens for the treatment of mild-to-moderate PID should include an antibiotic with reliable activity against N. gonorrhoeae in addition to an antibiotic with reliable activity against C. trachomatis and M. genitalium. There is not a single agent with these characteristics so only combination therapy can be recommended for the treatment of PID. Given the increasing concern about M. genitalium as an etiologic agent for PID, coverage of this microorganism is desirable.[11] Finally, broad-spectrum coverage of the BV-associated microorganisms appears less important.[15,24] In fact, the combination of doxycycline with metronidazole appears to be inferior and therefore one might consider avoiding it.

The extended-spectrum cephalosporins offer excellent coverage of N. gonorrhoeae. For the reasons of convenience, cost and ease of administration, oral cefixime 400 mg in a single dose is our first extended cephalosporin of choice. Ceftriaxone 250 mg intramuscularly in a single dose is our second choice owing to the fact that, unlike cefoxitin, it does not require the addition of oral probenecid. However, it should be pointed out that both of these cephalosporins fail to cover the Gram-negative anaerobes often associated with BV that cefoxitin covers. Cefoxitin is high on our list but given the need to split the intramuscular dose into two 1 g intramuscular injections plus the need to add oral probenecid, this is a less attractive alternative. These choices reliably treat a gonococcal infection. Optimally, one would use cefixime or ceftriaxone in those women with PID and no BV present, but opt for cefoxitin, owing to its superior anaerobic coverage in the presence of PID with concurrent BV (Box 6).

The second antibiotic should be either doxycycline or azithromycin owing to their effectiveness against C. trachomatis and M. genitalium. Doxycycline has been studied in the gold standard of PID studies, the PEACH study, and along with cefoxitin has been shown to be highly efficacious. However, doxycycline requires a twice-a-day dosing regimen for 14 days, making compliance improbable. Azithromycin, on the other hand, can be dosed in either a single daily dose (500 mg followed by 250 mg daily) for 1 week or in two single doses (1 g each) 1 week apart. In addition, azithromycin appears to work more efficiently in eradicating in vitro infection and has the added benefit of a significant anti-inflammatory effect.[38] Moreover, there is some concern about persistent M. genitalium infection following doxycycline.[11,14] Given these observations, azithromycin appears to be the superior choice.

Finally, we have to ask ourselves, how important is it to cover the anaerobic bacteria associated with BV and isolated from the upper genital tracts of women with PID? The antibiotics discussed above have modest activity against these microorganisms. The antibiotic, cefoxitin, with the best coverage against Gram-negative anaerobes, is administered in a single dose. Antibiotic regimens prospectively studied with little to no activity against anaerobic bacteria have uniformly performed well, while regimens containing metronidazole and doxycycline, which should enhance anaerobic coverage, have performed comparatively poorly. This leads us to suggest that dual therapy with an extended-spectrum cephalosporin and doxycycline or azithromycin is an excellent regimen for the treatment of women with PID who can be treated as outpatients.

Women with clinically severe PID should be considered for imaging to evaluate for TOA. As mentioned earlier, antibiotic regimens recommended for clinically severe disease should cover Gram-negative aerobic and anaerobic bacteria. If a TOA is present, regimens should have the ability to penetrate abscess cavities while remaining stable in an acidic, hypoxic abscess environment.

Although regimens containing an aminoglycoside have been used effectively in women with pelvic abscesses, this class of antibiotic has its activity significantly reduced at low pH, low oxygen tension and in the presence of drug-binding purulent debris.[41] McNeeley et al. showed that the combination of clindamycin and gentamicin was associated with a significantly lower cure rate (47%) than the combination of clindamycin/ampicillin/gentamicin (87.5%) when used to treat patients with TOA.[42] For these reasons, an extended-spectrum cephalosporin, for example, ceftriaxone, may be a better choice to combine with either clindamycin or metronidazole in treating women with severe PID with or without a TOA. Not only do extended-spectrum cephalosporins maintain their activity in an abscess environment, but they have a much higher serum level to MIC ratio than the aminoglycosides (Box 7).

Clindamycin is actively transported into polymorphonuclear leukocytes and macrophages and is present in relatively high concentrations compared with peak serum levels in experimental abscesses.[43] Clindamycin has long been used in combination therapy for PID because of its activity against anaerobes and continues to be recommended in the treatment of PID on the basis of earlier studies and previously successful experience, but resistance to clindamycin has recently been observed among isolates recovered from the lower genital tract.[44] Among nonpregnant women with BV, 17% of isolates demonstrated clindamycin resistance at baseline, a rate that increased to 53% after clindamycin therapy. Although there are no data indicating that these observations of resistance are associated with PID treatment failures with clindamycin-based regimens, there is concern that resistance to clindamycin may be on the rise.[45] Clinicians should be aware of bacterial resistance when selecting antimicrobial therapy. In this light, renewed interest has been focused on ampicillin/sulbactam or metronidazole which, in contrast to clindamycin, does not appear to have the same problems with selective pressure for microbial resistance.[46]

For the reasons noted earlier, we recommend metronidazole or clindamycin in combination with ceftriaxone as our regimen of choice. Other antimicrobials with similar broad-spectrum coverage and activity within the hostile environment of the abscess include β-lactam agents with β-lactamase inhibitors (e.g., ampicillin/sulbactam, ticarcillin/clavulanate and piperacillin/tazobactam) and the carbapenems (e.g., ertapenem, merepenem and imipenem/cilistatin). Of these alternative choices, we recommend ertapenem owing to our ability to dose it once a day. Finally, for those women with an immediate hypersensitivity to penicillin, we would recommend the combination of a quinolone (e.g., ciprofloxacin or levofloxacin) plus metronidazole. It would be important to test for N. gonorrhoeae by culture in these patients as susceptibility to quinolones would need to be ascertained (Box 7).

Patients admitted with severe PID and/or TOA should be discharged on a broad-spectrum oral antimicrobial regimen to complete a 14-day course. Recommended oral regimens for discharge include amoxicillin/clavulanate (875 mg twice daily) or the combination of trimethoprim/sulfamethoxazole (160/800 mg twice daily) and metronidazole (500 mg twice daily), owing to excellent polymicrobial coverage.

The most severe form of clinical PID-ruptured TOA should be considered in patients with PID presenting with an acute abdomen and signs of septic shock. TOAs in patients undergoing medical management of PID may rupture and require emergent surgical therapy. Surgical exploration with extirpation of the involved adnexa and drainage of purulent loculations is lifesaving. Hysterectomy is usually not necessary.[47,48]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.