Recurrent Urinary Tract Infection in Older Outpatient Women

Scope of the problem

Recurrent urinary tract infection (rUTI) presents unique diagnostic and management dilemmas in older women in outpatient settings. Conventionally defined by symptomatic, culture-confirmed UTI occurring at least twice in 6 months or three times in a year, rUTI is twice as common among women over age 65 as in the general population of women.¹ Diagnosis of recurrent bladder infection is more challenging in older women due to higher rates of chronic genitourinary symptoms that can be confused with acute cystitis symptoms as well as asymptomatic bacteriuria that decreases the discriminatory value of urine tests. Treatment and prevention can be complicated by frequent adverse effects of antibiotics, drug-drug interactions in the setting of polypharmacy, and antimicrobial resistance from cumulative antimicrobial exposure. In this article, we discuss special considerations in rUTI diagnosis and management in older women.

Diagnostic considerations

Classic symptoms of bladder infection include acute dysuria, urgency, frequency, and suprapubic pain, sometimes accompanied by hematuria or fever. Over a third of older community-dwelling women have chronic overactive bladder or genitourinary syndrome of menopause, however, and the waxing and waning symptoms of these syndromes should not be confused with acute cystitis symptoms.¹ Empiric treatment based on patient self-diagnosis may be appropriate in reproductive-age women, but clinicians should question older women more carefully about the timing, severity, and quality of symptoms to gauge how likely these are to represent true infection.²

Urine testing can be misleading among older women due to high rates of bacterial colonization and contamination. Asymptomatic bacteriuria occurs in 15%−20% of community-dwelling older women compared to 5% of healthy, premenopausal women,³ and over 90% of older women with asymptomatic bacteriuria also have some degree of pyuria, which limits the discriminant value of urine testing. To avoid overtreatment, clinicians should avoid unnecessary testing in older women without symptoms strongly suggesting UTI.⁴ However, the sensitivity of pyuria or bacteriuria on microscopic urinalysis is higher for symptomatic UTI in adults over 70 years old than in the general population, while specificity is lower.⁵ Thus, even though presence of pyuria or bacteriuria on urine tests cannot rule in an infection, their absence can be useful in ruling out rUTI.

Many older women are treated for presumptive UTI in the setting of non-specific problems such as falls or functional status changes, but these non-localizing symptoms have not been shown to correlate with a positive urine culture in frail older populations.^5,6 Consequently, even when older patients with these non-specific presentations are found to have bacteriuria, acute UTI is unlikely to be the specific cause in the absence of signs of sepsis. In such situations, careful monitoring and assessment for other potential etiologies are more appropriate than presumptive treatment.⁴

Unique diagnostic challenges arise when bacteriuria is detected in older women with dementia or other functional impairments that prevent them from perceiving or reporting cystitis symptoms.⁷ Management should include hydration and close monitoring for other symptoms such as fever, hypotension, or tachycardia, along with thoughtful shared decision-making discussions. Clinicians should acknowledge diagnostic uncertainty, consider past evidence of the chronicity of pyuria or bacteriuria, and weigh the potential harms of treating bacteriuria when UTI-specific symptoms cannot be confirmed.

Treatment considerations

Clinicians should consider the potential adverse effects of antibiotic treatment in older women, especially those with reduced kidney function. These include increased risk of kidney failure and hyperkalemia with trimethoprim-sulfamethoxazole, confusion and tendonitis with ciprofloxacin, and pulmonary/hepatic toxicity with nitrofurantoin. Regardless of the agent, however, women do not require more extended treatment for rUTI by virtue of age alone; no meaningful differences in efficacy have been detected between short (3–6 days) and long (7–14 days) antibiotic courses in older women.⁸

Among community-dwelling older women, empiric antibiotic therapy is associated with increased risk of antimicrobial resistant organisms, bacteremia, and death.⁹ Consequently, clinicians should consider delaying initiation of antimicrobial therapy while awaiting culture sensitivities in older immunocompetent women (while emphasizing hydration, offering bladder analgesics, and discussing return precautions). However, evidence to guide decisions about delaying antibiotic therapy for UTI in older women is limited. A retrospective analysis of older patients reported that delaying antimicrobial therapy was associated with subsequent bacteremia,¹⁰ but conclusions were limited by potential confounding and misdiagnosis.¹¹

Prevention considerations

Sustained low-dose antibiotic therapy is the conventional approach to rUTI prophylaxis.¹² While effective in reducing UTI recurrence, the optimal duration of antibiotic prophylaxis is unclear given that trials have involved only 6 to 12 months of therapy. This approach is also associated with rapid development of antimicrobial resistance, based on research demonstrating resistance in 90% of urinary and fecal isolates after only 1 month of trimethoprim-sulfamethoxazole therapy.¹³ Prolonged antibiotic therapy also increases the potential for drug-drug interactions, although rates vary by antibiotic agent (Table).

Table.

Strategies to Prevent Recurrent Urinary Tract Infection in Older Women in Outpatient Settings

I. Antibiotic strategies
Specific agents	Common formulations/regimens12	Special considerations for older women
Nitrofurantoin	Daily oral 50 to 100 mg at bedtime, or 100 mg as a single dose within 2 hours of sexual intercoursea	Risk of pulmonary toxicity, hepatotoxicity, and peripheral neuropathy if CrCl <30 mL/min; conversely, lower rates of drug-drug interactions and antimicrobial resistanceb
Trimethoprim-sulfamethoxazole	Daily or three times weekly oral 40 mg/200 mg (half a single-strength tablet)a	Increased risk of kidney failure and hyperkalemia if CrCl <30 mL/min, particularly with concomitant angiotensin-converting enzyme inhibitor or receptor antagonist therapy; multiple potential other drug-drug interactions; potential for antimicrobial resistanceb
Trimethoprim	Daily oral 100 mga	50% dose decrease indicated if CrCl 15–30 mL/minute; multiple potential drug-drug interactions; potential for antimicrobial resistanceb
Fosfomycin	Oral 3 grams every 7 to 10 days (or alternately every 3 days to maintain higher blood levels)a	Dose adjustment indicated if CrCl <50 mL/minute; lower rates of drug-drug interactions than some alternatives; potential for antimicrobial resistanceb
Cephalexin	Daily oral 125 mg to 250 mg, or 250 mg as a single dose just before or after sexual intercoursea	Potential for increased nephrotoxic effects with concomitant loop diuretic therapy; potential greater serum concentration of metformin and anticoagulant effect of warfarin; and potential for antimicrobial resistanceb
II. Antibiotic-sparing strategies
Specific agents	Common formulations/regimens	Special considerations for older women
Vaginal estrogen	Insufficient evidence to compare formulations, but evidence from trialsc of vaginal estriol cream (0.5 mg nightly for 2 weeks then twice weekly)14 or estradiol ring (2 mg, placed every 3 months)15	In contrast to vaginal estrogen, no evidence of reduced UTI recurrence with oral or systemic estrogen, which has other long-term health risks in older postmenopausal women
Methenamine	Oral methenamine hippurate 1 gram tablet twice dailyd	Evidence of non-inferiority based on trials in women of all ages (only 0.49 more episodes per year with methenamine versus nitrofurantoin),17 but no age-specific data available
Cranberry supplements	No one definitive regimen—past trials reporting benefit with juice, tablets, or powdere	Relative risk 0.74 in a meta-analysis of trials of women of all ages,18 but no significant benefit detected in the subgroup of older adults in long-term care
Hydration	No one definitive approach—one trial reporting benefit with 1.5 L water over daily intake, but other trials using juice or other oral solutions19	Hydration with water preferable to electrolyte beverages in older women with conditions worsened by salt load (e.g., congestive heart failure); diuretic therapy may be held in older women during acute UTI episodes
Toileting practices	No definitive evidence of benefit (but no obvious associated harm) with front-to-back wiping after using the toilet or with early post-coital voiding	Despite lower self-reported frequency of vaginal intercourse in older women, vaginal sexual intercourse is a risk factor for UTI across all ages20

UTI = urinary tract infection; CrCl = creatinine clearance

^aOptimal duration of antibiotic prophylaxis not established, but prior trials limited to 6 to 12 months of therapy

^bConsider evaluation of antimicrobial sensitivities through urine culture before initiating antibiotics in clinically stable older patients

^cPast randomized trials demonstrating efficacy of vaginal estrogen therapy limited to 8 to 9 months of therapy

^dPast randomized trials demonstrating non-inferiority of methenamine to antibiotics limited to 12 months of therapy

^ePast randomized trials supporting efficacy of cranberry products limited to 12 months of therapy

Vaginal estrogen decreases rUTI risk in postmenopausal women, based on a double-blinded trial of vaginal estriol¹⁴ and an open-label trial of a vaginal estradiol ring.¹⁵ Nevertheless, vaginal estrogen therapy may not be as effective as antibiotic prophylaxis, given evidence of over 2-fold higher recurrent cystitis rate among women using a 0.5 mg estradiol vaginal pessary compared to nitrofurantoin.¹⁶

Methenamine, an agent metabolized to formaldehyde in the distal tubules of the kidney, may be an option for rUTI prophylaxis, although prior methenamine research has not focused on older women with comorbidities. In non-inferiority trials in women of all ages, rates of UTI recurrence were not significantly worse with methenamine than with chronic trimethoprim or nitrofurantoin therapy.¹⁷

Cranberry products are also reported to reduce rUTI risk based on a meta-analysis of trials suggesting an average 26% relative risk reduction among women of all ages.¹⁸ No one cranberry regimen is known to be superior to any other, however, and a subgroup analysis of over 1,400 older adults in long-term care facilities found no benefit of cranberry supplementation.¹⁸ At this time, no other non-prescribed supplements such as D-mannose or Lactobacillus probiotics have been proven effective for rUTI prevention.

Although not supported by robust evidence, behavioral and non-pharmacologic prevention strategies are low risk, low cost, and generally low burden. Fluid intake has been shown to reduce rUTI risk in reproductive-age women¹⁹ and may be equally appropriate for older women without chronic health conditions worsened by fluid overload. Because rUTI is associated with sexual intercourse in both postmenopausal and reproductive-age women,²⁰ early postcoital voiding may offer benefits for all women across the aging spectrum.

Summary

Prevention and management of rUTI pose unique challenges in older community-dwelling women. Rather than testing and treating liberally, clinicians should focus on acute, localizing urinary symptoms that are most likely to indicate recurrent infection, use clinical tests judiciously to rule out infection while considering alternate explanations for positive results, and consider the greater risks of antibiotic therapy for both acute intermittent treatment and chronic suppression.

Shared decision-making is especially important when diagnosis of a rUTI episode in older women is unclear, such as when patients have functional limitations that preclude recognition of localizing symptoms. In these cases, clinicians should acknowledge limitations in the evidence and invite older women or their caregivers to discuss their preferences about presumptive treatment, recognizing that some may place greater value on the possibility of earlier symptom relief and decreased risk of complications, while others may prefer to minimize adverse drug effects and future multidrug resistance.