Carol E. Chenoweth
Sanjay Saint
Urinary tract infection (UTI) is the most frequently reported healthcare-associated infection (HAI), accounting for up to 40% of all HAIs [1,2,3,4]. The vast majority of nosocomial UTIs is associated with urinary catheters. Urinary catheters are widely used in health care today, especially in the intensive care units (ICU), in postsurgical patients, in long-term care facilities, and increasingly in home care patients [5,6,7,8]. Up to 25% of patients have a urinary catheter placed at some time during their hospital stay [9,10]. Worldwide, approximately 96 million urethral catheters are sold every year; nearly a quarter of these in the United States. Like other indwelling catheters, urinary catheters disrupt the normal host immune mechanisms and allow for the formation of biofilm [3,11,12]. Urinary catheter-related infection results in substantial morbidity and significantly increases hospital costs [13,14]. This chapter reviews the pathogenesis, epidemiology, and preventive measures for catheter-associated urinary tract infections (CA-UTIs).
Pathogenesis
Urinary catheters readily develop a biofilm composed of clusters of microbial organisms on the internal and external catheter surface surrounded by an extracellular matrix made up of primarily polysaccharide materials [12,15,16,17]. The biofilm allows for microbial attachment and adherence to catheter surfaces. Microorganisms gain access to the catheter and attach to the biofilm via one of two routes, extraluminally or intraluminally (Figure 30-1). Extraluminal organisms are primarily endogenous, originating from the patient's gastrointestinal tract and colonizing the patient's perineum. Organisms ascend the catheter by direct inoculation at the time of catheter insertion or by migration in the mucous sheath surrounding the external aspect of the catheter [16,18,19]. Approximately 70% of episodes of bacteriuria in catheterized women are felt to occur through extraluminal entry of organisms [18]. In a recent prospective study of 173 CA-UTIs, 115 (66%) were likely acquired through the extraluminal route [19].
|
Figure 30-1 Entry points for bacteria causing catheter-associated urinary tract infection. |
Microorganisms also enter the catheter intraluminally when organisms gain access to the internal lumen of the catheter through failure of a closed drainage system or contamination of the collection bag [16,19,20]. These organisms, usually introduced from exogenous sources, are often the result of cross-transmission of organisms on the hands of healthcare workers (HCWs) [16,20,21,22]. Intraluminal contamination of the collecting system was recently found
P.508
to account for 34% of CA-UTIs [19]. Once microorganisms attach and multiply, the resultant sheet of organisms secretes an extracellular matrix of bacterial glycocalyces, imbedding the microorganisms [11,12,15,19,23,24].
Bacteria within the biofilm grow much more slowly than planktonic bacteria and secrete chemical signals that mediate population density–dependent gene expression [11,12,15,19,23,24]. The migration of the biofilm over the inner surface of the catheter to the bladder occurs within 1–3 days or more quickly by swarming organisms, such as Proteus mirabilis [11,23,24,25]. Most biofilms are composed of single organisms; however, biofilms may contain a mixture of up to five organisms [16,26]. Some organisms, such asProvidencia stuartii, Pseudomonas sp., enterococci, or Proteus sp., persist in the urine for up to 10 weeks while other organisms appear to spontaneously cycle in and out [16,26]. Several studies suggest that planktonic bacteria found in cultures obtained from the catheter may not reflect bacterial population growing within the biofilm [3,26]. Proteus sp.,Pseudomonas aeruginosa, Klebsiella pneumoniae, and Providencia sp. have the ability to hydrolyze urea in the urine to free ammonia. The resulting increase in local pH allows precipitation of minerals such as hydroxyapetite or struvite. Mineral deposition within the catheter biofilm causes encrustations that are unique to biofilms formed on urinary catheters [24,27,28]. Encrustations on the inner surface of the catheter can build to completely block catheter flow or act as a nidus for formation of renal calculi [29,30].
The urinary biofilm provides a protective environment from the activity of antimicrobial agents [31,32]. First, the extracellular matrix may prevent the penetration of antimicrobials into the biofilm. For example, both ciprofloxacin and tobramycin have poor diffusion into biofilms. Secondly, organisms growing at a slower rate within the biofilm are more resistant to the effects of antimicrobial agents that require active growth [31,32,33]. Finally, chemical signaling from organisms growing within the biofilm appears to regulate genes that alter the molecular targets of antimicrobials [33]. The features of biofilm as described have important implications for both prevention and treatment of CA-UTIs.
Epidemiology
Microbial Etiology
Enterobacteriaceae, including Escherichia coli, Klebsiella sp., and Enterobacter sp., are the most common pathogens associated with HAI CA-UTIs (Table 30-1). Other pathogens, more common in the ICU setting, include Pseudomonas aeruginosa, enterococci and Candida sp. [6,7,34,35]. European hospitals report a similar spectrum of bacteria associated with nosocomial UTIs, except for Pseudomonas sp., which were isolated in only 7% of urine cultures [36]. While 80% of infections associated with short-term indwelling urinary catheters are due to single organisms, infections in long-term catheters are frequently polymicrobial. UTI in long-term catheters are associated with ≥2 organisms in 77–95% of episodes, and 10% have >5 species of organisms present [16,26].
|
TABLE 30-1 |
|||||||||||||||||||||||||||||||||||
|
Enterococci emerged as a significant cause of nosocomial UTIs between 1975 and 1984 [37]. Enterococcal UTIs may derive from an endogenous source such as the patient's fecal flora or may be acquired exogenously [38,39,40]. The emergence and spread of antimicrobial-resistant strains have compounded the problem in many centers. Candida spp. are prevalent in the ICU setting where 25% of UTI are associated with Candida spp. [6,7]. Risk factors for candiduria include prolonged catheterization and use of broadspectrum antimicrobials. Most candida UTIs are asymptomatic, but candidemia may result in the setting of urinary tract abnormalities or procedures. Other potential complications include fungus balls of the bladder or kidney, renal abscesses, or disseminated candidiasis [41]. Coagulase-positive staphylococci (CPS) are an infrequent
P.509
cause of CA-UTI [6,7], but when CPS-UTIs occur, secondary bacteremia may result [42]. Conversely, CPS bacteremia or endocarditis may result in secondary infection of the urinary tract. In one study, 27% of CPS bacteremias were associated with secondary bacteriuria [43]. The presence of CPS in the urine should prompt consideration of coinciding bacteremia and endocarditis [42,43].
Incidence of Catheter-Associated Urinary Tract Infection
In hospitalwide data, UTIs have accounted for approximately 40% of all HAIs [1,44], but UTIs make up a smaller proportion of HAIs occurring in the ICU setting. UTIs account for 15–21% of HAIs in pediatric ICU patients, 23% of HAIs in adult U.S. ICU patients, and 18% of ICU infections in the European EPIC study [6,7,45,46].
The prevalence of UTI varies by ICU type; rates of CA-UTIs reported through the Centers for Disease Control and Prevention's (CDC) National Nosocomial Infections Surveillance (NNIS) system between January 2002 and June 2004 ranged from 3.0 infections/1,000 catheter-days in cardiothoracic ICUs to 6.7 infections/1,000 catheter-days in burn or neurosurgical ICUs [47]. The rate of UTI in pediatric ICUs was 4.0 infections/1,000 catheter-days, lower than the rate seen in an equivalent adult medical ICU population of 5.1 infections/1,000 catheter-days [47]. Nosocomial UTI is infrequently identified in neonatal ICUs [47,48,49]. In data collected in a non-ICU setting in 42 German hospitals, the rate of infection was similar, 6.8 infections per 1,000 urinary-catheter days [50].
Risk Factors
The most important and consistent risk factor for bacteriuria is the duration of urinary catheterization (odds ratio [OR] = 2.3–22.4, depending on duration) [51,52,53,54,55,56]. Urinary catheters are associated with the vast majority of nosocomial UTI; up to 97% of UTIs in ICUs are associated with a urinary catheter [6,13]. Bacteriuria occurs quickly and frequently in catheterized patients with an average daily risk of 3–10% per day [57,58,59]. In patients with a catheter indwelling for 2–10 days, 26% will develop bacteriuria. Nearly all patients catheterized for a month will have bacteriuria, making this the dividing line between short-term and long-term catheterization [3,26].
Females have a higher risk of bacteriuria than males (relative risk [RR] = 1.7–3.7) [51,52,53,54,56,60]. Systemic antibiotics have a protective effect on bacteriuria; therefore, the absence of systemic antimicrobials increases the risk of bacteriuria (RR = 2.0–3.9) [51,52,53,54,56,60]. Non adherance to catheter care recommendations has been associated with increased risk of bacteriuria [20,51,53]. Other risk factors identified in ≥1 studies include rapidly fatal underlying illness (RR = 2.5) [51]; age >50 years (RR = 2) [51,54]; nonurgical disease (RR = 2.2) [51]; hospitalization on an orthopedic (RR = 51) or urological service (RR = 4) [55]; catheter insertion after the sixth day of hospitalization (RR = 8.6) [55]; catheter inserted outside the operating room (RR = 5.3) [53]; diabetes mellitus (OR = 2.3) [53]; or serum creatinine >2 mg/dL at the time of catheterization (OR = 2.1) [53]. Heavy periurethral colonization with bacteria also has been associated with increased risk of bacteriuria [61]. Significant risk factors for CA-UTI are summarized in Table 30-2.
|
TABLE 30-2 |
|||||||||||||
|
Risk factors for UTI-related bacteremia are less clearly defined than for catheter-related bacteriuria because catheter-related bacteremia occurs infrequently (<4% of patients with catheter-related bacteriuria develop bacteremia) [62,63,64]. In a 23-month prospective study by Krieger et al., 1,233 patients with nosocomial UTI were identified [63]. BSIs from a urinary tract origin were found in 32 patients (2.6%). Univariate analysis identified risk factors for secondary nosocomial blood stream infection (BSI) due to Serratia marcescens,compared with other organisms (RR = 3.5) and male sex (RR = 2.0) [63]. No other factors (e.g., age, race, underlying disease, hospital service) were found to significantly predispose to bacteremia [63].
Clinical Manifestations
CA-UTI presents clinically with a spectrum from asymptomatic bacteriuria to urosepsis and death [3,14,57,64]. Only 10–32% of patients with catheter-associated bacteriuria experience symptoms attributable to infection; thus, most patients can be classified as asymptomatic bacteriuria [3,14,57,64]. In a study of 235 patients with nosocomial catheter-related bacteriuria, approximately 90% of infections were asymptomatic [64]. Patients, with and without infection, had no significant differences in fever, dysuria, urgency, flank pain, or leukocytosis.
P.510
When present, local symptoms of UTI include lower abdominal discomfort, dysuria, urgency, frequency, or hematuria [14]. Fever, flank pain, or other clinical manifestations of pyelonephritis develop in <1% of patients with catheter-associated bacteriuria [62,65,66]. Clinically recognized infections, including prostatitis, epididymitis, seminal vesiculitis, or renal infection, may arise from bacteriuria originating during catheterization, but the frequency of such infections remains ill defined [62,65,66]. In general, these complications arise primarily in patients with long-term indwelling catheters and are rare in patients whose catheterization lasts <10 days. Signs and symptoms of sepsis, including fever, hypotension, mental status changes, or organ system dysfunction may be associated with secondary bacteremia, especially those due to gram-negative bacilli.
Sequelae
From the short-term clinical perspective, most CA-UTIs appear benign. Few patients with catheter-associated bacteriuria have undergone localization studies, and thus the proportion of patients with bladder, prostate, or kidney infections has not been determined. Patients with bacteriuria who die have had autopsy findings of acute pyelonephritis, renal calculi, or perinephric abscesses [62,65,66].
The major systemic complication of catheter-associated bacteriuria is secondary bacteremia. Secondary bacteremia occurs only infrequently (0.4–3.9%) in patients with nosocomial UTI [57,62,63,64,67]. Bacteremia is less likely to occur with asymptomatic bacteriuria and is more likely to be associated with major underlying disease and co-morbidities [64]. Reflecting how often urinary catheters are used in hospitalized patients, the urinary tract is the source of 11–40% of nosocomial BSI [68,69,70]. A series of hospitalized patients with gram-negative BSI demonstrate that 30–40% of all gram-negative BSI acquired in the hospital originate in the urinary tract [63,68], making this the most commonly recognized source of gram-negative sepsis. Interestingly, the onset of bacteremia usually occurs within 24 hours of the onset of bacteriuria except for Serratia marcescens infections, in which bacteremias most commonly originate days after the onset of bacteriuria [63].
Nosocomial UTI is associated with a mortality rate of 14-19% in prospective studies [14,67,71]. Patients with UTI are nearly 3 times more likely to die during hospitalization than patients without infection [67,71]. UTI-related nosocomial BSIs have an attributable case-fatality rate of approximately 12.7% [62], with severely ill patients at highest risk of death. The use of a urinary catheter alone has been independently associated with an increased risk of death in an elderly population residing in long-term care facilities [72,73].
Recent evidence indicates that hospital stay associated with nosocomial UTI is increased by approximately 1 day, with an average cost of infection between $558 and $676 [14,74]. In a retrospective study of adults in acute care hospitals in the United States, Haley et al. estimated that nosocomial UTI occurred in 2.39 patients per 100.00 admissions, prolonged hospitalization 1 day, and cost $593 [44,75]. In another study conducted almost three decades ago, nosocomial UTI resulted in an average increase in length of stay of 2.4 days and an associated cost of $558 [76]. More recent studies suggest that an episode of bacteriuria increases costs by $676 and that urinary catheter-related BSI increases costs $2,836 per episode [14].
Diagnosis
The terms bacteriuria and UTI are often used interchangeably in published reports of CA-UTI. The term bacteriuria or candiduria implies the presence of a significant number of microorganisms in quantitative urine cultures [59,64,77]. Low levels of Candida sp. or bacteria in urine grow within 72 hours to concentrations of >105 cfu/mL unless antibiotic therapy is administered [59]. Therefore, most studies of CA-UTI use bacteriuria as the primary outcome, and growth of ≥102 cfu/mL of a predominant pathogen from a catheterized urine specimen collected aseptically from a sampling port is a standard definition for CA-bacteriuria [20,78].
The CDC has developed surveillance definitions for identifying nosocomial UTI that allow for interhospital comparison of infection rates [6,7,79]. The definitions differentiate symptomatic (presence of fever, urgency, frequency, dysuria, or suprapubic tenderness) from asymptomatic infection but do not allow for classification of asymptomatic bacteriuria with <105 CFU/mL.
Pyuria is an important indicator of UTI in the noncatheterized patient; in the catheterized patient, however, pyuria is not strongly correlated with UTI [77,80]. In one study, pyuria was uniformly present with bacteriuria in catheterized men, but pyuria also was present in 30% of catheterized patients without bacteriuria [80]. A recent prospective study of 761 catheterized patients found that pyuria was most strongly associated with infection caused by gram-negative bacilli; infections caused by coagulase-negative staphylococci, enterococci, or yeast were less frequently associated with pyuria [77]. Urinary white blood cell (WBC) counts >5/high-power field had a specificity of 90% for predicting infections but had a sensitivity of <37% [77].
Neither urinalysis nor urine cultures are reliable tests for diagnosing symptomatic UTI in patients with long-term indwelling urinary catheters [81]. Cultures from these catheters are universally positive and may not reflect bladder cultures [26,82]. Fever and chills are the most consistent symptom of CA-UTI [65,66,83]. UTI in patients with spinal cord lesions may be particularly difficult to diagnose because of the inability of the patient to sense localizing symptoms [83].
P.511
Treatment
Most patients with CA-bacteriuria are asymptomatic and do not require treatment unless the patient is at high risk for complications (e.g., BSI or renal infection [64,84]. Treatment of asymptomatic bacteriuria may be useful in patients with neutropenia, renal transplants, pregnancy or if the patient is undergoing transurethral resection of the prostate or other urological procedures likely to induce bleeding [84]. In most patients without such complicating clinical features, bacteriuria often resolves spontaneously with removal of the catheter. After catheter removal, the patient can be observed and subsequently treated if the bacteriuria does not resolve spontaneously after 48 hours [84]. The latter may be particularly useful in elderly women in whom the risk of development of symptomatic infection is particularly high [85]. Because the antimicrobial susceptibility patterns of strains causing CA-bacteriuria vary widely, choice of a specific antimicrobial agent should be guided by the in vitro antimicrobial susceptibilities of the infecting organism(s).
Treatment with the catheter in place often results in emergence of resistant strains, and eradication of bacteriuria in the presence of an indwelling catheter has been largely unsuccessful [31,86,87]. A recent prospective, randomized, controlled trial of patients with symptomatic UTI found that patients who underwent indwelling catheter replacement before initiation of antibiotic therapy had a significant decrease in bacteriuria and improved clinical outcome when compared with patients who had no catheter replacement [87]. Their findings support the recommendation that catheters present for at least one week before onset of CA-infection should be replaced (or removed if no longer required) before antimicrobial therapy [17,88].
Prevention
Despite existence and knowledge of guidelines emphasizing the use of aseptic technique and closed urinary drainage for the prevention of nosocomial UTI [89,90], adherence with guidelines varies among institutions. Errors in compliance with guidelines were found in 11% of catheter-days and overall in 29% of catheterized patients at one institution [91]. Surveillance and feedback of nosocomial UTI rates to staff may be useful as an adjunct for improving compliance with recommendations [44,92,93]. Several measures for the prevention of CA-UTI are discussed next (Table 30-3).
|
TABLE 30-3 |
|||||||
|
Avoidance of Use of Indwelling Catheters
Because as many as 80% of nosocomial UTIs and 97% of UTIs in ICUs are associated with a urinary catheter, the most important prevention strategy is decreasing the use of urinary catheters [4,16,63]. Data from the NNIS system between January 1992 and June 2004 reveal urinary catheter utilization in participating ICUs ranging from 0.30 to 0.91 urinary catheter-days/patient-days. Utilization was highest in trauma, neurosurgical, and cardiothoracic ICUs (0.91, 0.85, and 0.84 catheter-days/patient-days, respectively) and lowest in pediatric ICUs, 0.30 catheter-days/patient-days [47]. The duration of catheterization varies by hospital ward and patient population, but the mean and median durations in acute care hospitals are 2 and 4 days, respectively. Catheters are removed within 7 days in 70% of patients [94].
Overall, urinary catheters are overutilized, and documentation surrounding catheterization is inconsistent [2,95,96,97,98,99,100]. In recent prospective studies of catheterized patients, the indication for catheterization was judged to be inappropriate 21% to 50% of the time [2,95,97,99,100]. Furthermore, a written order or procedure note was frequently not documented in the medical record [96,97]. A study conducted at four different institutions found that 28% of HCWs were unaware that their patient had an indwelling urinary catheter. Lack of awareness increased with the level of training and correlated with inappropriate catheter use. The authors found that 22% of interns, 28% of residents, and 38% of attending physicians were unaware of catheters in their patients [2].
In addition, urinary catheters may be unpleasant and restrictive to patients. Elderly men at a Veterans Affairs medical center more frequently responded that a condom catheter was comfortable (86%) compared to patients with an indwelling urethral catheter (58%, P = 0.04) [101]. Patients also felt that condom catheters were less painful or restrictive of activities of daily living (24% vs. 61%, P = 0008) [101]. Another survey of patients and family of residents in long-term care facilities revealed that 85% preferred diapers and 77% preferred prompted voiding to indwelling urinary catheterization [102].
Nevertheless, urinary catheters are important for patients requiring drainage of anatomic or physiologic outlet obstruction, patients undergoing surgery of the genitourinary tract, patients requiring accurate urinary output measurements, and patients with sacral or perineal
P.512
wounds [2,99,103] (Table 30-4). However, limiting urinary catheter use to appropriate indications has important implications. Initial efforts at decreasing unnecessary catheterization have been promising [104,105,106,107]. Cornia et al. evaluated a computerized reminder, which shortened the duration of catheterization by 3 days (approximately 30% reduction in catheter days) while not affecting recatheterization rates [104]. Huang et al. evaluated a nurse-based reminder system in the adult ICU of a large hospital in Taiwan and found that nurse reminders to remove unnecessary catheters significantly reduced the duration of catheterization (7 days vs 4.6 days; P < 0.001) and UTI (11.5 vs 8.3 per 1000 catheter-days; P = .009) [105]. Finally, Saint et al. completed a pilot study in which a nurse-based reminder after 48 hours of catheterization signficantly reduced the proportion of time patients were catheterized [106].
|
TABLE 30-4 |
|||||||||||
|
Use of Aseptic Insertion and Catheter Care Techniques
Proper aseptic technique, including aseptic insertion and maintenance of the catheter and drainage bag, is another essential strategy for preventing CA-UTI [20,51,89,90]. Cleansing the meatus at urinary catheter insertion has been widely recommended but has not been well studied. A recent randomized study comparing water with 0.1% chlorhexidine cleansing of the periurethral area before catheter insertion revealed no difference in the development of bacteriuria [108]. In addition, routine meatal cleaning of catheterized patients has shown no benefit [109,110]. The collection bag should always remain below the level of the bladder to prevent reflux of urine (and bacteria introduced into the bag) into the bladder. Proper hand hygiene and use of gloves for insertion and manipulation of catheters is critical to prevent introduction of exogenous pathogens [21,22,111,112,113].
Use of Closed Drainage Systems
An important advance in the prevention of CA-UTI was the introduction of the closed catheter drainage system that includes the use of sealed urinary catheter junctions [58,71,114,115,116,117]. A recent evaluation of two closed drainage systems compared a complex system (including a pre attached catheter, antireflux valve, drip chamber, and povidone iodine–releasing cartridge) with a two-chamber system. The authors found no difference in the rate of bacteriuria between the two systems [118]. Improper catheter care and breaches of the closed system remain an important risk factor for the development of bacteriuria [51,91].
Other Catheter Care Practices
Other interventions, such as irrigating the bladder or instilling antibacterial solutions into the urinary collecting bag, have not been shown to have benefit when used on closed urinary collecting systems [119,120,121,122,123]. These practices allow for flow of organisms colonizing the catheter into the bladder and require opening the closed system and are therefore not routinely recommended [124,125,126]. In addition, use of meatal lubricants and creams (both antibacterial and nonantibacterial) [127,128,129,130,131] or urinary catheters that have been coated with heparin [132] or polymer have not shown benefit for prevention of UTIs [133].
Use of Anti-Infective Urinary Catheters
Several studies support the use of anti-infective (latex-based silver alloy or nitrofurazone impregnated) urinary catheters as adjunct to the preceding proven methods of prevention in patients at high risk for CA-UTI [134,135,136,137,138,139,140]. A recent prospective trial of a silicone-based, silver-coated urinary catheter, however, showed no effect in preventing UTIs when compared to a silicone-based hydrogel catheter [141]. Recent analysis of the clinical and economic consequences of urinary catheters indicates that latex-based silver alloy catheters, which cost substantially more than standard catheters, may provide both clinical and economic benefits in patient populations receiving indwelling catheterization for 2–10 days, including the critically ill [10,142]. Al-Habdan et al. evaluated a nitrofurazone-impregnated urinary catheter among 100 postoperative and trauma patients in a hospital in Saudi Arabia. In this randomized trial, patients given the nitrofurazone-impregnated catheter were significantly less likely to develop bacteriuria compared to those given regular latex urinary catheters (12% vs 0%; P = 0.028) [143]. Anti-infective urinary catheters thus appear to be a promising method of reducing bacteriuria in the catheterized patient, especially in those patients at highest risk of either bacteriuria or complications associated with bacteriuria. The effect anti-infective urinary catheters will have on the more important clinical outcomes of urinary catheter-related BSI and mortality is not clear [10,135,142].
P.513
Use of Systemic Antimicrobials
Systemic antimicrobial therapy may lower the risk of developing a CA-UTI [51,52,53,55]. However, most experts do not recommend routine use of prophylactic antibiotics for catheterized patients because of their cost, potential adverse effects, and potential for selection of antibiotic-resistant organisms [16,17,144]. Several studies have demonstrated that antibiotic prophylaxis increases the rate of isolation of antibiotic-resistant organisms in catheterized patients [26,145,146,147]. Prophylaxis with trimethoprim-sulfamethoxazole (TMP/SMX) has been shown to be beneficial for the prevention of UTI after renal transplantation [148,149]. Antibiotic prophylaxis is justifiable for men undergoing transurethral resection of the prostate, especially in those with an indwelling catheter or bacteriuria preoperatively [150].
Methenamine, available as a salt of mandelate or hippurate, has been used for preventing CA-UTI for >30 years [4]. Its breakdown products, hippuric acid and formaldehyde, acidify the urine and have antibacterial properties. In small studies, oral methenamine hippurate therapy (2–6 grams daily) has been found to reduce the incidence of bacteriuria [151,152], symptomatic UTI [153], or pyuria [152]. In recent studies in patients undergoing gynecological surgery, prophylactic treatment with methenamine hippurate significantly reduced the incidence of post operative bacteriuria and symptomatic UTI [154,155]. While methenamine hippurate is not currently recommended, randomized controlled trials evaluating this intervention should be considered [4].
References
P.514
P.515
P.516