Episode 54- Is PID in need of a third antibiotic?

Hello and welcome to Episode 54 of ER-Rx. In this week’s Fresh Fruit series, we discuss a potentially practice-changing study published in April 2021 in the journal Clinical Infectious Diseases. This study is entitled; “A randomized controlled trial of ceftriaxone and doxycycline, with or without metronidazole, for the treatment of acute pelvic inflammatory disease.”

Pelvic inflammatory disease (PID) occurs when microorganisms from the vagina or endocervix work their way up to the endometrium and fallopian tubes. If untreated, it can put women at risk of chronic pelvic pain, ectopic pregnancy, and infertility. Eighty-five percent of cases are caused by sexually-transmitted pathogens such as chlamydia and gonorrhea, or by pathogens that cause bacterial vaginosis (BV). Less than 15% of the time, PID is not sexually-transmitted and is instead caused by enteric pathogens such as E.coli, Bacteroides, and group B strep or by respiratory pathogens such as H. influenza or Strep pneumoniae. Importantly for our discussion, anaerobic organisms are also frequently found in infections of the upper genital tract.

The CDC recommends a one-time IM dose of a cephalosporin with oral doxycycline for 14 days for outpatient therapy of PID. This regimen is effective against most of the organisms that cause PID, but it does have limited activity against anaerobic organisms. Although we do commonly isolate anaerobic organisms from women with PID, the need for covering these organisms is still unknown- one reason is that regimens without anaerobic coverage still have excellent clinical cure rates. The CDC guidelines list metronidazole as an additional option, while European guidelines recommend the addition of metronidazole at all times. In a time where antimicrobial stewardship has been a hot topic, and due to concerns about the tolerability of a 3-drug regimen, the authors of this trial wanted to determine if the addition of metronidazole should be incorporated into first-line treatment of PID.

In this randomized, double-blind, placebo-controlled trial, the authors compared ceftriaxone + doxycycline to ceftriaxone + doxycycline + metronidazole for the treatment of acute PID. The study enrolled patients between November 2010 through January 2015 in two EDs and a county STD clinic in Pittsburgh, PA. They included women aged 15-40 years and excluded women who required inpatient treatment, those who were pregnant, those who used systemic/vaginal antibiotics within the last 7 days, or those who had an allergy to any of the medications. At enrollment, patients rated their pain on a visual analog scale, had a clinical tenderness score measured, and had cultures collected. All women received a single 250 mg IM dose of ceftriaxone and 100 mg twice daily doxycycline x 14 days. Remember that since the release of the updated guidance from the CDC in December 2020, the dose of ceftriaxone is now 500 mg for those weighing < 150 kg, or 1 g for those weighing >/= 150 kg—check out Episode 37 for more info. Patients were then randomized in a 1:1 fashion to also received 500 mg twice daily metronidazole or placebo for 14 days.

The primary outcome was clinical improvement at 3-day follow, determined by the reduction of the clinical tenderness score. Secondary outcomes included the presence of anaerobic organisms 30 days after treatment and rates of clinical cure (defined as >70% improvement in the clinical tenderness score and the absence of fever). The study enrolled 233 patients. Of these, 116 received metronidazole and 117 received placebo. The median age was 23 years and about 60% were black. A history of chlamydia (60%), gonorrhea (27%), or PID (29%) was similar between the groups, as were their clinical presentations.

In terms of results, 208 (89%) of women returned for the primary outcome assessment at 3 days. Overall, 91% had clinical improvement, which was similar in both groups. Thirteen women (8 in the metronidazole group and 5 in the placebo group) failed to respond and were either admitted for IV antibiotics or were put on different antibiotic therapy. 184 (79%) of women continued on study medications and returned for the final assessment at 30 days. Again, improvement in pelvic pain was similar between the groups. Clinical cure rates were also similar between groups at 97% vs 90%, P= 0.38. However, pelvic organ tenderness was still present in 20% of women in the placebo group compared to only 9% of women who received metronidazole (P < 0.05).

At enrollment, chlamydia was identified in 15% of women, gonorrhea in 7%, and 18% tested positive for Mycoplasma genitalium. Also, BV was present in 55% of women and 9% had Trichomonas vaginalis, and the proportion of women affected with each of these was similar between groups. However, at 30 days, BV (20% vs 54%, P < 0.001) and T. vaginalis (5% vs 12%, P = 0.10) were less prevalent in women randomized to metronidazole, as was M. genitalium (4% vs 14%, P <0.05)- even though metronidazole does not have activity against this organism. Similarly, at 30 days, women who received metronidazole were less likely to have recovery of anaerobic gram-negative rods or anaerobic gram-positive cocci from the endometrium (8% vs 21%, P < 0.05).

In terms of safety and tolerability, adherence was similar between the groups, and adverse events were reported in about equal numbers. There was a higher rate of vulvovaginal candidiasis in women receiving metronidazole (15.5% vs 6 %, P= 0.02).

In conclusion, although early clinical response at 3 days was similar between groups, fewer women who received metronidazole had pelvic tenderness at one month. Also, women receiving metronidazole had fewer endometrial anaerobic organisms isolated at 1 month after treatment. These benefits came with similar adherence rates and adverse effects -although vulvovaginal candidiasis did occur more in the metronidazole group. The authors concluded that their study supports the routine use of metronidazole in combination with ceftriaxone and doxycycline for the treatment of PID. I would agree generally with this statement- given that metronidazole is discussed in the CDC guidelines as a consideration, especially since it will also treat BV, which is commonly associated with PID. Longer-term and larger studies would be nice, but given the tolerability and efficacy seen in this trial, I’m ready to implement this into practice.