Efficacy of vaginal estrogen for recurrent urinary tract infection prevention in hypoestrogenic women

Abstract

Vaginal estrogen is considered to be the standard of care for recurrent urinary tract infection prevention in women with hypoestrogenism. However, literature supporting its use is limited to small clinical trials with narrow generalizability. This study aimed to assess the association between vaginal estrogen prescription and the frequency of urinary tract infections over the following year in a diverse population of women with hypoestrogenism. Secondary objectives included evaluation of medication adherence and predictors of postprescription urinary tract infection. This multicenter retrospective review included women who were prescribed vaginal estrogen for the indication of recurrent urinary tract infection from January 2009 through December 2019. Recurrent urinary tract infection was defined as having ≥3 positive urine cultures (separated by at least 14 days) in the 12 months preceding the index vaginal estrogen prescription. Patients were asked to fill their prescriptions and continue care within Kaiser Permanente Southern California system for at least 1 year. Exclusion criteria included anatomic abnormalities, malignancy, or mesh erosion of the genitourinary tract. Data on demographics, medical comorbidities, and surgical history were collected. Adherence was captured through refill data after the index prescription. Low adherence was defined as no refills; moderate adherence was defined as 1 refill; high adherence was defined as ≥2 refills. Data were abstracted from the electronic medical record system using the pharmacy database and diagnosis codes. A paired t test was used to compare pre- and postprescription urinary tract infections over the year preceding and following the vaginal estrogen prescription. A multivariate negative binomial regression was used to evaluate predictors of postprescription urinary tract infection. The cohort included 5638 women with a mean (±standard deviation) age of 70.4 (±11.9) years, body mass index of 28.5 (±6.3) kg/m2, and baseline urinary tract infection frequency of 3.9 (±1.3). Most of the participants were White (59.9%) or Hispanic (29.7%) and postmenopausal (93.4%). The mean urinary tract infection frequency in the year following the index prescription decreased to 1.8 (P<.001) from 3.9 in the year preceding the prescription, which is a 51.9% reduction. During the 12 months after the index prescription, 55.3% of patients experienced ≤1 urinary tract infections, and 31.4% experienced no urinary tract infections. Significant predictors of postprescription urinary tract infection included age of 75 to 84 years (incident rate ratio, 1.24; 95% confidence interval, 1.05-1.46) and >85 years (1.41; 1.17-1.68), increased baseline urinary tract infection frequency (1.22; 1.19-1.24), urinary incontinence (1.14; 1.07-1.21), urinary retention (1.21; 1.10-1.33), diabetes mellitus (1.14; 1.07-1.21), and moderate (1.32; 1.23-1.42) or high medication adherence (1.33; 1.24-1.42). Patients with high medication adherence demonstrated more frequent postprescription urinary tract infections than patients with low adherence (2.2 vs 1.6; P<.0001). In this retrospective review of 5600 women with hypoestrogenism who were prescribed vaginal estrogen for the prevention of recurrent urinary tract infections, the frequency of urinary tract infection decreased by more than 50% in the following year. Baseline urinary tract infection frequency, increasing age, urinary incontinence or retention, and diabetes were associated with an increased risk of postprescription urinary tract infection. The paradoxical finding that women with moderate and high medication adherence experienced the lowest-magnitude reduction in urinary tract infection frequency may represent unobserved selection or unmeasured confounding.