Vaginal Estrogen for the Prevention of Recurrent Urinary Tract Infection in Postmenopausal Women

This was an investigator-initiated, multicenter, single-blind, randomized placebo-controlled trial of vaginal estrogen (delivered by estradiol ring or conjugated estrogen cream) compared to placebo cream. This study was conducted by the Divisions of Female Pelvic Medicine and Reconstructive Surgery at three academic institutions. Potential participants were recruited from women receiving care for rUTIs in the clinics of each institution. Women were postmenopausal as defined by amenorrhea for \>12 months, history of bilateral salpingoophrectomy, or in patients with prior hysterectomy defined as menopausal symptoms for \>1 year or age \>55. Participants had documented rUTI (three or more in one year or two in six months) by positive urine culture. Participants were excluded if they had undergone urologic surgery within three months of screening or planned surgery within one year of screening, had the diagnosis of painful bladder syndrome, history of UTI requiring the use of intravenous antibiotics or one oral antibiotic available for treatment based on allergies and resistance profiles, known etiology of infections (e.g. urologic stones, fistulas, fecal incontinence, catheterization or poorly controlled diabetes), urothelial cancer, or actively treated estrogen sensitive tumor. Participant enrollment was deferred in the setting of undiagnosed hematuria or vaginal bleeding, but enrollment could occur after negative workup for malignancy. Participants deferred enrollment if they used vaginal androgens, estrogens or progestins within six months or if they used medications or supplements known to prevent UTI (e.g. cranberry products, D mannose, prophylactic antibiotics, methenamine hippurate) within three months of enrollment, but could enroll after washout. Those with a remote history of estrogen-sensitive tumor required approval by her oncologist or primary care physician. Participants were initially randomized to receive either vaginal estrogen (via estradiol ring or conjugated estrogen cream) or placebo cream in a 1:1:1 fashion and were told that there were placebo and vaginal estrogen arms, but were unaware that there was not a placebo ring. Participants were randomized to either conjugated estrogen cream at a strength of 0.625 mg/g dosed at 0.5g (0.312 mg) twice a week, estradiol ring containing 2mg of estradiol placed vaginally every three months by study personnel, or placebo cream which was an over-the-counter lubricant containing mineral oil, petrolatum, ceresin and paraffin dosed at 0.5g twice a week. If participants in any arm developed three UTI in the six-month treatment period, they were unblinded. If on placebo, they were provided active study drug or if they were in the vaginal estrogen arm, they were treated according to the provider's clinical practice and their outcomes recorded for the remainder of the study. The primary outcome of occurrence of UTI was assessed at six months or the end of study blinding, whichever occurred first. In order to improve recruitment and assess long-term compliance, satisfaction and efficacy, we offered open label use of estradiol ring or conjugated estrogen cream after unblinding until month 12 of the study. Clean-catch urine cultures were collected when participants were symptomatic: fever (\>38 degrees C), urgency, frequency, dysuria or suprapubic tenderness. Women were asked to have all urine labs done at the parent site as per standard clinical practice. The primary outcome of the study was the presence of a UTI in the randomized six months. We also assessed rates of UTI over the course of the 12-month study in all participants as-treated. Quality of life (QOL) questionnaires were administered at baseline, six months or unblinding, and 12 months. Questionnaires included the Pelvic Floor Distress Inventory (PFDI-20), Pelvic Floor Impact Questionnaire (PFIQ-7), Medical, Epidemiologic, and Social Aspects of Aging Questionnaire (MESA), Female Sexual Function Index (FSFI) and Patient Satisfaction Questionnaire (PSQ), Global Perception of Improvement (GPI) and Estimated Percent Improvement (EPI) rating scales. Baseline medical history and concomitant pelvic medicine diagnoses were also collected. Participants using a cream treatment were asked to keep medication diaries and return tubes for weighing at months three, six, nine and 12. Adherence was defined as presence of the estrogen ring or tube weights within 20% of expected at follow up visits. In addition, before unblinding, the participants were asked whether they believed they were on active treatment, placebo, or were unsure.