Comparative Effectiveness of Empiric Antibiotics for Uncomplicated Urinary Tract Infection in Postmenopausal Women

For uncomplicated urinary tract infection in postmenopausal individuals, nitrofurantoin and fosfomycin had similar outcomes; trimethoprim-sulfamethoxazole had more failures and complications.

Abstract

OBJECTIVE:

To compare the effectiveness of antibiotic regimens for acute uncomplicated urinary tract infection (UTI) in postmenopausal women using real-world clinical data.

METHODS:

We conducted a retrospective cohort study in the TriNetX database of women aged 40 years or older with menopausal or postmenopausal status who developed acute uncomplicated cystitis or UTI between March 2006 and March 2026. Patients were stratified by initial antibiotic therapy of nitrofurantoin, trimethoprim-sulfamethoxazole (TMP-SMX), or fosfomycin. The primary outcome was a composite measure of treatment failure within 7 days of treatment initiation, including antibiotic escalation, pyelonephritis, bacteremia, sepsis, hospitalization, critical care services, and mortality. Propensity score matching was used to balance demographics, laboratory values, comorbidities, and prior gynecological procedures. Outcomes were compared between matched groups with a two-sided z test for proportions.

RESULTS:

A total of 236,927 postmenopausal women initiating antibiotics to treat uncomplicated UTI were identified (nitrofurantoin n=152,793; TMP-SMX n=80,815; and fosfomycin n=3,319). After matching, TMP-SMX was associated with a higher risk of composite treatment failure compared with nitrofurantoin (10.2% vs 8.0%, relative risk ratio [RR] 1.28, 95% CI, 1.23–1.32). In addition, TMP-SMX was associated with higher rates of antibiotic escalation, pyelonephritis, bacteremia, sepsis, hospitalization, critical care services, and mortality. Compared with fosfomycin, TMP-SMX was associated with a higher risk of composite treatment failure (13.3% vs 9.6%, RR 1.39, 95% CI, 1.21–1.59), as well as higher rates of sepsis, hospitalization, and critical care services. Nitrofurantoin and fosfomycin had similar rates of composite treatment failure (9.0% vs 9.6%, RR 0.93, 95% CI, 0.80–1.09), although hospitalization was less frequent with nitrofurantoin (1.8% vs 3.0%, RR 0.60, 95% CI, 0.44–0.82).

CONCLUSION:

Among postmenopausal women with acute uncomplicated UTI, treatment with TMP-SMX was associated with higher rates of adverse outcomes and serious infectious complications compared with nitrofurantoin and fosfomycin. Nitrofurantoin and fosfomycin demonstrated similar short-term outcomes. These findings support the need for postmenopausal-specific evidence to inform empiric UTI treatment selection.

Urinary tract infections (UTIs) are common among women and account for millions of outpatient visits annually in the United States.^1,2 Their management is therefore highly relevant to obstetrician–gynecologists who diagnose and treat UTIs in routine practice.

Postmenopausal women may be at increased risk of adverse UTI outcomes. Compared with premenopausal women, they experience higher rates of recurrence and UTI-associated hospitalization.³ Estrogen deficiency is associated with changes in the urogenital epithelium and microbiome that may promote colonization by opportunistic and resistant organisms such as Klebsiella and extended-spectrum β-lactamase (ESBL)–producing Escherichia coli.^3,4 Structural and functional conditions common in this population, including genitourinary syndrome of menopause, pelvic organ prolapse, and urinary incontinence, may further increase susceptibility to UTIs through impaired bladder emptying and urinary retention.^4,5

Current empiric treatment recommendations for acute uncomplicated UTIs are derived largely from studies of predominantly premenopausal women. The American College of Physicians and Infectious Diseases Society of America guidelines recommend nitrofurantoin, trimethoprim-sulfamethoxazole (TMP-SMX), and fosfomycin as first-line agents.^6,7 However, postmenopausal women have not been specifically represented in past studies, and the Infectious Diseases Society of America explicitly limits their scope to premenopausal women.⁷ The American College of Obstetricians and Gynecologists also does not have specific guidelines for the management of uncomplicated UTIs in postmenopausal women.⁸

Whether these empiric antibiotic options perform similarly in postmenopausal women remains unclear. We therefore evaluated the comparative effectiveness of nitrofurantoin, TMP-SMX, and fosfomycin for acute uncomplicated UTI in postmenopausal women using real-world data.

METHODS

This retrospective study used the TriNetX U.S. research network, a large federated database compliant with the Health Insurance Portability and Accountability Act (HIPAA) containing deidentified data from 67 health care organizations and approximately 115 million patients across the nation. TriNetX allows analysis based on standardized administrative coding systems, including the International Classification of Diseases, National Library of Medicine, RxNorm, Anatomical Therapeutic Chemical Classification, and Current Procedural Terminology codes (Appendix 1, available online at https://links.lww.com/AOG/E784). Because of the deidentified nature of the data, this study was determined to be IRB exempt by our institution. All data within the TriNetX platform, whether presented in aggregate or patient-level datasets, consist exclusively of deidentified information in accordance with the HIPAA Privacy Rule deidentification standard, §164.514(a), which is formally certified by a qualified expert as defined under §164.514(b)(1).

Data were queried on March 3, 2026, and included data over a 20-year period (March 2006–March 2026). The study population included patients with a menopause-related diagnosis code. Rather than being classified by diagnosis alone, patients were additionally required to be 40 years of age or older because this is the earliest age at which menopause may be diagnosed. Because of the lack of distinction between menopausal and postmenopausal status in diagnostic coding, these groups were analyzed together and are collectively referred to as postmenopausal women. Patients were also required to have a diagnosis of cystitis or uncomplicated UTI and were stratified into three cohorts based on initial antibiotic therapy: nitrofurantoin, TMP-SMX, and fosfomycin.

To reflect real-world prescribing patterns in which transitioning between first-line antibiotics is rare and to isolate the influence of initial antibiotic choice,⁹ patients who switched between first-line agents were excluded from this analysis. Patients were also excluded if they had conditions associated with complicated UTI, including structural urinary tract abnormalities or diversions (malignant neoplasms of the urinary tract, cystostomy, nephrostomy tube, urostomy, urinary diversion stoma, cutaneous ureterostomy, vesicoureteral reflux), functional bladder disorders (neurogenic bladder), immunocompromised status (human immunodeficiency virus [HIV] disease, neutropenia, organ transplantation, or chemotherapy within the prior year), or prior urologic procedures affecting urinary drainage (sling procedures). Additional exclusions included any recent hospitalization, pyelonephritis, sepsis, bacteremia, bladder catheterization, or calculus of the kidney or ureter within 1 week of UTI to increase the specificity for uncomplicated UTIs.

To reduce confounding, cohorts were balanced using 1:1 propensity score matching with greedy nearest-neighbor matching without replacement. Selected covariates included demographics (current age, age at index, race, ethnicity), laboratory values (body mass index [BMI], serum albumin, hemoglobin A1C), comorbidities (type 2 diabetes mellitus, chronic kidney disease stage 4 or 5, end-stage renal disease), and prior gynecological procedures (hysterectomy, oophorectomy). Matching was performed using a caliper of 0.1, and covariate balance was assessed with standardized mean differences (SMDs), with values less than 0.1 considered indicative of adequate balance. Matching was performed with available data within the TriNetX platform. Variables with unavailable values were handled according to the native matching algorithm of the platform.

The primary outcome was the composite occurrence of adverse outcomes associated with antibiotic failure, including antibiotic escalation (fluoroquinolones or β-lactam antibiotics),^10–12 acute pyelonephritis, bacteremia, sepsis, hospitalization, use of critical care services, and mortality. Secondary outcomes included the individual outcomes used to create the composite score. All outcomes were assessed within an observational window of 7 days after the diagnosis of cystitis or UTI (index event). All statistical analyses were performed with the TriNetX platform. Outcomes were compared between matched groups using a two-sided z test for proportions. Risks and relative risk ratios (RRs) with 95% CIs were calculated. To account for multiple pairwise comparisons, P values were adjusted with the Holm–Bonferroni method, with statistical significance defined as an adjusted value of P<.05. Any outcomes that yielded 10 patients or fewer were censored according to HIPAA compliance regulations associated with use of the database. Because of the inability to compare multiple cohorts simultaneously on the TriNetX platform accessible to our institution, three pairwise analyses were performed. These included nitrofurantoin versus TMP-SMX, TMP-SMX versus fosfomycin, and nitrofurantoin versus fosfomycin. A flowchart describing the cohort construction process is displayed in Figure 1.

Fig. 1. Cohort derivation for postmenopausal women with uncomplicated urinary tract infection (UTI) receiving empiric antibiotic therapy. Inclusion, exclusion, and propensity score matching criteria used to identify the study population within the TriNetX database are shown. TMP-SMX, trimethoprim-sulfamethoxazole.

Lacsamana. Empiric Antibiotics for Postmenopausal Uncomplicated UTI. O&G Open 2026.

RESULTS

Data were queried on March 3, 2026. A total of 236,927 postmenopausal women with uncomplicated UTI were identified, including 152,793 who received nitrofurantoin, 80,815 who received TMP-SMX, and 3,319 who received fosfomycin. The mean age of each cohort was 66.3±11.5, 66.7±11.7, and 73.0±11.4 years, respectively. Most covariates were adequately balanced after matching (SMD less than 0.1). Serum albumin remained modestly imbalanced across pairwise comparisons (SMD more than 0.1) (Fig. 2).

Fig. 2. Covariate balance before and after propensity score matching (PSM) for nitrofurantoin vs trimethoprim-sulfamethoxazole (TMP-SMX) (A), TMP-SMX vs fosfomycin (B), and nitrofurantoin vs fosfomycin (C). A post-PSM standardized mean difference less than 0.1 indicates adequate balance. AI/AN, American Indian/Alaska Native; NH/PI, Native Hawaiian/Pacific Islander; BMI, body mass index; T2DM, type 2 diabetes mellitus; CKD, chronic kidney disease; ICD-10, International Classification of Diseases, 10th Revision; CPT, Current Procedural Terminology.

Lacsamana. Empiric Antibiotics for Postmenopausal Uncomplicated UTI. O&G Open 2026.

When TMP-SMX and nitrofurantoin were compared, each cohort included 77,832 patients after matching. Compared with nitrofurantoin, TMP-SMX was associated with a higher risk of the composite adverse outcome (treatment failure) within 7 days (10.2% vs 8.0%, RR 1.28, 95% CI, 1.23–1.32). In addition, TMP-SMX was associated with higher rates of antibiotic escalation, pyelonephritis, bacteremia, sepsis, hospitalization, critical care services, and mortality (Table 1 and Fig. 3).

Table 1.

Comparison of 7-Day Adverse Outcomes Between Trimethoprim-Sulfamethoxazole and Nitrofurantoin for Uncomplicated Urinary Tract Infections in Postmenopausal Women

TMP-SMX vs Nitrofurantoin
Outcome	TMP-SMX	Nitrofurantoin	RR (95% CI)	Adjusted P
Composite adverse outcome	7,959 (10.2)	6,224 (8.0)	1.28 (1.24–1.32)	.001
Antibiotic escalation	6,816 (8.8)	5,506 (7.1)	1.24 (1.20–1.28)	.001
Pyelonephritis	97 (0.1)	62 (0.1)	1.56 (1.14–2.15)	.006
Bacteremia	167 (0.2)	65 (0.1)	2.57 (1.93–3.42)	.001
Sepsis	478 (0.6)	221 (0.3)	2.16 (1.85–2.54)	.001
Hospitalization	2,306 (3.0)	1,264 (1.6)	1.82 (1.70–1.95)	.001
Critical care services	373 (0.5)	142 (0.2)	2.62 (2.16–3.18)	.001
Mortality	130 (0.2)	49 (0.1)	2.65 (1.91–3.69)	.001

TMP-SMX, trimethoprim-sulfamethoxazole; RR, risk ratio.

Values are n (%) unless indicated otherwise.

Fig. 3. Forest plot showing risk ratios (RRs) and 95% CIs for pairwise comparisons of nitrofurantoin and trimethoprim-sulfamethoxazole (TMP-SMX) (A), TMP-SMX and fosfomycin (B), and nitrofurantoin and fosfomycin (C) among postmenopausal women with uncomplicated urinary tract infection. Values greater than 1.0 indicate higher risk in the first-listed antibiotic. The dashed vertical reference line indicates no difference (RR 1.0).

Lacsamana. Empiric Antibiotics for Postmenopausal Uncomplicated UTI. O&G Open 2026.

When TMP-SMX and fosfomycin were compared, each cohort included 3,317 patients after matching. Compared with fosfomycin, TMP-SMX was associated with a higher risk of the composite adverse outcome (13.3% vs 9.6%, RR 1.39, 95% CI, 1.21–1.59). In addition, TMP-SMX was associated with higher risks of antibiotic escalation, sepsis, hospitalization, and critical care services. Results of pyelonephritis, bacteremia, and mortality could not be reported (Table 2 and Fig. 3).

Table 2.

Comparison of 7-Day Adverse Outcomes Across Fosfomycin-Based Antibiotic Regimens for Uncomplicated Urinary Tract Infections in Postmenopausal Women

TMP-SMX vs Fosfomycin
Outcome	TMP-SMX	Fosfomycin	RR (95% CI)	Adjusted P
Composite adverse outcome	441 (13.3)	318 (9.6)	1.39 (1.21–1.59)	.001
Antibiotic escalation	373 (11.2)	251 (7.6)	1.49 (1.28–1.73)	.001
Sepsis	41 (1.2)	13 (0.4)	3.15 (1.69–5.88)	.001
Hospitalization	146 (4.4)	100 (3.0)	1.46 (1.14–1.87)	.006
Critical care services	32 (1.0)	14 (0.4)	2.29 (1.22–4.28)	.008

Nitrofurantoin vs Fosfomycin
Outcome	Nitrofurantoin	Fosfomycin	RR (95% CI)	Adjusted P
Composite adverse outcome	297 (9.0)	318 (9.6)	0.93 (0.80–1.09)	1.000
Antibiotic escalation	262 (7.9)	251 (7.6)	1.04 (0.88–1.23)	1.000
Sepsis	15 (0.5)	13 (0.4)	1.15 (0.55–2.42)	1.000
Hospitalization	60 (1.8)	100 (3.0)	0.60 (0.44–0.82)	.007
Critical care services	12 (0.4)	14 (0.4)	0.86 (0.40–1.85)	1.000

TMP-SMX, Trimethoprim-sulfamethoxazole; RR, risk ratio.

Values are n (%) unless indicated otherwise.

Outcomes with cell counts of 10 or fewer (pyelonephritis, bacteremia, mortality) for some analyses were censored in accordance with TriNetX reporting requirements.

Finally, when nitrofurantoin and fosfomycin were compared, each cohort included 3,311 patients after matching. Composite adverse outcomes did not differ significantly between nitrofurantoin and fosfomycin (9.0% vs 9.6%, RR 0.93, 95% CI, 0.80–1.09). Most individual outcomes were also similar between groups, although hospitalization was less frequent among patients receiving nitrofurantoin (1.8% vs 3.0%, RR 0.60, 95% CI, 0.44–0.82). Results of pyelonephritis, bacteremia, and mortality could not be reported (Table 2 and Fig. 3).

DISCUSSION

In this large retrospective cohort of postmenopausal women with acute uncomplicated UTI, treatment with TMP-SMX was associated with worse short-term outcomes compared with treatment with nitrofurantoin and fosfomycin. In contrast, nitrofurantoin and fosfomycin demonstrated similar overall effectiveness across most outcomes, although nitrofurantoin was associated with lower rates of hospitalization. In premenopausal populations, TMP-SMX has also been associated with higher rates of treatment failure and progression to pyelonephritis compared with nitrofurantoin,⁹ with our study extending these findings to the more vulnerable postmenopausal population. Nevertheless, treatment failure remains clinically important because of its associated consequences, including increased health care costs and hospitalizations,¹³ and potentially exposes women to additional risks of health care–associated infections, including catheter-associated UTIs and multidrug-resistant pathogens.^14,15

Although the direction of association was consistent with findings in premenopausal populations, the underlying contributors to treatment failure may differ specifically in postmenopausal women. Estrogen deficiency alters the urogenital microbiome, including loss of protective lactobacilli and increased colonization by uropathogenic organisms.¹⁶ Resistance patterns in postmenopausal women also differ from those observed in younger, premenopausal populations.^17,18 Although E coli remains the most common uropathogen, accounting for approximately two-thirds of positive cystitis cultures, postmenopausal women exhibit higher prevalence of colonization with ESBL-producing E coli.^17,19 ESBL-producing E coli have been identified in over 20% of isolates among hospitalized postmenopausal women and are associated with increased rates of antibiotic escalation.³

Although the 2024 Infectious Diseases Society of America guidelines list both nitrofurantoin and TMP-SMX as preferred agents for culture-confirmed ESBL cystitis, this recommendation assumes documented susceptibility.²⁰ In empiric treatment settings, resistance to TMP-SMX remains substantial, with coresistance rates among ESBL-producing E coli reported to be as high as 65%.^21–23 In contrast, resistance to nitrofurantoin and fosfomycin remains comparatively low. These differences may contribute to the higher rates of treatment failure observed with TMP-SMX in postmenopausal women.^17,21–27 A comparative summary of recent surveillance data for ESBL-producing E coli resistance is provided in Appendix 2 (https://links.lww.com/AOG/E784).

These findings have important implications for clinicians managing UTIs in postmenopausal women. Postmenopausal status may represent an underrecognized risk factor for empiric treatment failure, suggesting that age and menopausal status should be considered in the selection of empiric antibiotic therapy.^24,27

Although current guidelines list TMP-SMX, nitrofurantoin, and fosfomycin as equivalent first-line options for acute uncomplicated cystitis, our findings suggest that TMP-SMX may be associated with a higher risk of treatment failure in this population. In contrast, nitrofurantoin and fosfomycin were associated with comparable and more favorable outcomes. However, fosfomycin is often more expensive and difficult to obtain because of variable insurance coverage, which may limit its use and explains the substantially smaller fosfomycin cohort. These findings suggest that nitrofurantoin or fosfomycin may warrant consideration over TMP-SMX as empiric therapy in postmenopausal women, particularly in settings where TMP-SMX resistance may be prevalent.

Our study is not without limitations. As a retrospective analysis, causality cannot be established, and findings are subject to misclassification. Residual confounding may persist. In particular, serum albumin remained modestly imbalanced and may reflect differences in frailty, nutritional status, or underlying illness severity. However, the consistency of associations across multiple outcomes suggests that this imbalance is unlikely to fully explain the findings. Postmenopausal status was identified with diagnostic codes rather than hormonal measurements. It is likely that many postmenopausal women lack specific menopause codes used in this study and are therefore not captured. In addition, information on clinical factors, including UTI history and culture susceptibility results, was not uniformly available. Adherence to treatment also could not be confirmed.

Despite these limitations, this study consists of a large, nationally representative sample, with consistent differences across multiple clinically meaningful end points. Future studies incorporating microbiological and susceptibility data are needed to clarify the mechanisms underlying treatment failure in postmenopausal populations. Prospective longitudinal research such as that described in the published study protocol by Hekker et al²⁸ on urobiomes of postmenopausal women in recurrent UTIs, may further elucidate factors associated with recurrence and treatment response or failure.

In this large, real-world cohort of postmenopausal women with acute uncomplicated UTI, treatment with TMP-SMX was associated with higher rates of antibiotic treatment failure and serious infectious complications compared with treatment with nitrofurantoin and fosfomycin. Nitrofurantoin and fosfomycin demonstrated similar short-term outcomes and may be considered over TMP-SMX when appropriate. These findings support consideration of menopausal status in empiric antibiotic selection and highlight the need for population-specific evidence to inform future treatment guidelines.