on 22-Sep-2023 (Fri)

Using the preferred definition of pyuria, which is at least 10 leukocytes/ mm3 of midstream urine by counting chamber, the vast majority of patients with symptomatic or asymptomatic bacteriuria have pyuria. In fact, with symptomatic infection, most have hundreds of leukocytes per cubic millimeter.

A less reliable method uses a urine specimen that is centrifuged for 5 minutes at 2000 rpm and then the sediment is examined under high power. With this method, 5 to 10 leukocytes/ high-power field in the sediment is the upper limit of normal. It should be emphasized that the finding of pyuria is nonspecific, and patients with pyuria may or may not have infection.167 The absence of pyuria should strongly call into question a diagnosis of UTI.

The dipstick leukocyte esterase test is a rapid screening test for detecting pyuria and has largely replaced microscopic methods. Although the sensitivity and specificity are high for detecting more than 10 white blood cells/mm3 of urine (75%–96% and 94%–98%, respectively), a positive test by no means indicates UTI; in patients with a negative leukocyte esterase test and UTI symptoms, a urine microscopic examina- tion for pyuria or a urine culture should be considered.168,169

Microscopic or sometimes gross hematuria is occasionally seen in patients with UTI (i.e., hemorrhagic cystitis). However, red blood cells may be indicative of other disorders, such as calculi, tumor, vasculitis, glomerulonephritis, and renal tuberculosis.

White cell casts in the presence of an acute infectious process are strong evidence for pyelo- nephritis, but the absence of white cell casts does not rule out upper tract infection. White cell casts can also be seen in renal disease in the absence of infection.

Proteinuria is a common although not universal finding in UTI. Most patients with UTI excrete less than 2 g of protein/24 h; excretion of 3 g/24 h or more suggests glomerular disease.

Microscopic examination of a urine specimen for bacteria can be useful for the presumptive diagnosis of UTI. The ability to identify bacteria in the urine depends on whether the specimen has been centrifuged and on whether it has been stained with Gram or methylene blue stain (Table 72.4).

The presence of at least one bacterium per oil immersion field in a midstream clean-catch, Gram-stained, uncentrifuged urine specimen correlates with 105 bacteria/mL of urine or more.

The absence of bacteria in several fields in a stained sedimented specimen indicates the probability of fewer than 104 bacteria/mL

Most common are tests (e.g., dipstick) that detect the presence of urine nitrite, which is formed when bacteria reduce the nitrate that is normally present.168 False-negative test results are common, especially in the detection of low-count bacteriuria (102–103/mL) and with certain bacterial species, but false- positive results are unusual.

The sensitivity and specificity of screening tests for UTI, such as dipsticks, depends on the likelihood of infection in the group being studied (e.g., acutely symptomatic patients vs. those who are asymptomatic) and range widely.170–172 A negative leukocyte esterase test plus a negative nitrite test are strongly predictive of the absence of UTI

Women with acute onset of symptoms of cystitis and without factors that would make the infection complicated can be diagnosed and safely managed on the basis of history only, without diagnostic testing.173

By quantitating bacteria in midstream clean-voided urine, it is possible statistically to separate contamination from UTI. Most patients with infection usually have at least 105 bacteria/ mL in urine in the bladder, and therefore voided urine usually contains at least 105 bacteria/mL. In patients without infection a properly collected, voided urine sample usually contains less than 104 bacteria/mL.

However, it is important to remember that about one-third of young women with cystitis have fewer than 105 bacteria/mL of urine

It is likely that a significant proportion of other patients with symptomatic and asymp- tomatic infection have fewer than 105 bacteria/mL of urine.

The Infectious Diseases Society of America consensus culture definition of cystitis for use in antibiotic treatment studies is 103 colony-forming units (CFU)/ mL or more of a uropathogen (sensitivity 90% and specificity 90%) and, for pyelonephritis, 104 CFU/mL or more (sensitivity 90% and specificity 90%).174 In a subsequent set of practice guidelines, 102 CFU/ mL or more of a uropathogen was used.175

The belief that bladder urine is normally sterile has been increasingly challenged by work that demonstrates the existence of a urinary microbiome.176,177

Other methods, such as the dip inoculum method, in which agar-coated glass slides or paddles are dipped into urine and then incubated, have excellent correlations with calibrated-loop techniques

Acceptable methods for urine collection include (1) midstream clean-catch, (2) “in and out” catheterization, and (3) suprapubic aspira- tion.

The clean-catch method has traditionally been preferred for the routine collection of urine for culture. It avoids the risk of infection inherent in catheterization. The patient must be instructed in the proper technique of obtaining the urine; this is especially important for women. The woman should wash her hands, straddle the commode (facing the back of the commode), wash her vulva from front to back four times with four different sterile gauze pads soaked in green soap or another appropriate cleansing agent, and then rinse with two more sponges soaked in sterile distilled water. She should then spread her labia and void, discarding the first portion of urine and collecting the second. The urine should be processed immediately or, if refrigerated at 4°C, it can be cultured within 24 hours.

Some have challenged the need for cleansing or for using a midstream specimen from women when collecting urine for culture. Similar contamination rates were noted when midstream urine was collected after cleansing versus voiding into a container without cleansing.178 A more recent study demonstrated similar rates of contamination for midstream clean-catch, midstream, and random samples.179

If there are more than 105 bacteria/mL in a clean-catch urine specimen from an asymptomatic woman, there is an 80% probability that this represents true bacteriuria. If two different specimens demonstrate at least 105 of the same bacterium/mL, the probability increases to 95%. Thus two clean-catch specimens should be obtained in an asymptomatic woman to confirm the diagnosis.144

Thus in asymptomatic women, 95% of the time 104 to 105 bacteria/mL represents contamination, with occasional infection manifested by fewer than 105 bacteria/mL of urine.

In asymptomatic men, in whom contamination is less likely, 103 or more organisms/mL in one culture is suggestive of infection, and 105/mL defines bacteriuria.144

False-positive cultures are caused by contamination or incubation of urine before processing. False-negative cultures may be caused by the use of antimicrobial agents, soap from the preparation falling into the urine, total obstruction below the infection, infection with a fastidious organism, renal tuberculosis, and diuresis.

Gram-positive organisms, fungi, and bacteria with fastidious growth requirements may not reach titers of 105/mL in patients with infection and may be in the 104 to 105/mL range. The organism recovered often helps distinguish contamination from true bacteriuria. Samples with counts of less than 104 organisms/mL often contain saprophytic skin organisms, such as diphtheroids, Neisseria, and staphylococci.

Pure growth of Enterobac- teriaceae is uncommonly found in low-titer specimens but is present in over 90% of urine samples containing more than 105 bacteria/mL.

High colony counts containing more than one species of bacteria from the urine of asymptomatic persons often represent contamination but may be more significant in the presence of symptoms. Mixed infection occurs in about 5% of cases.

In patients with symptoms of UTI, one titer of 105 or more bacteria/ mL of urine carries a 95% probability of true bacteriuria. With titers below 105/mL but in the presence of frequency, urgency, and dysuria, women have a 33% chance of having bacterial infection

In women with acute uncom- plicated cystitis, enterococci and group B streptococci isolated from midstream urine were often not found in catheterized urine obtained at the same time.180

Samples obtained by catheterization from noninfected patients are less likely to become contaminated. According to the guidelines, 102 CFU/mL or more is consistent with bacteriuria.144 For diagnosis with an indwelling catheter, see Chapter 302

Most women with an acute onset of frequency, urgency, or dysuria, or all of these, have UTI with 105 or more bacteria/mL of urine (Fig. 72.6). However, up to one-half are found to have fewer than 105 bacteria/mL of urine,78,181 and the term urethral syndrome has been used to refer to this entity.

When suprapubic bladder aspirates are compared with voided midstream urine in acutely dysuric women, 102 or more coliforms/mL in midstream urine have a sensitivity and specificity of 95% and 85%, respectively, for UTI.

Furthermore, up to one-third of young women with symptomatic infection localized to the lower urinary tract have fewer than 105/mL bacteria in their urine.

A study of women coming to a university health clinic for any reason demonstrated that stepwise increases in bacterial counts from 102 to 105 CFU/mL are significantly associated with an increased incidence of symptoms and pyuria182; it was postulated that low-count bacteriuria represents an early phase of urinary infection. This hypothesis appears to have been supported in a clinical study in which 21 women with low-count bacteriuria (102–<82a

The remaining women with the urethral syndrome, after excluding those with bacteria in the bladder and those with genital herpes infection or vaginitis, can be divided into two groups: (1) those with pyuria from urethritis (e.g., caused by Chlamydia trachomatis, Neisseria gonorrhoeae, or Mycoplasma genitalium infection); and (2) those without pyuria in whom all bacterial cultures are negative. The pathogenesis of this latter symptom complex is unknown, but various fastidious microorganisms and noninfectious factors (traumatic, psychological, allergic, and chemi- cal) have been suggested as causes.

Vaginitis is a common cause of dysuria and, accordingly, patients should be questioned regarding vaginal symptoms, particularly if the complaint of burning is external, such as pain felt in the inflamed labia during micturition. If vaginitis is suspected, a pelvic examination should be performed. Dysuria has also been described in 10% of women with initial genital herpes infection.

Although symptoms and the clinical settings cannot reliably distin- guish between the causes of dysuria in women, they can be suggestive.169

Bacterial UTIs tend to have a sudden onset of severe dysuria with frequency and urgency; suprapubic pain and/or hematuria may be present, and there is pyuria.

Clinical clues to urethritis (chlamydial, gonococcal, or mycoplasmal) include the following: a patient with a gradual onset of milder dysuria, with or without frequency and urgency, who is sexually active with a recent new sexual partner; hematuria is absent but vaginal discharge or bleeding may be present from chlamydial or gonococcal cervicitis; and/or pyuria is present.

With herpes, there are usually lesions in the periurethral area.

With vaginitis, the dysuria tends to be mild with gradual onset and is felt externally, frequency and urgency are absent, there is often a vaginal discharge, and pyuria is usually absent in a midstream specimen

The diagnostic testing for UTI described in the previous sections has changed very little in over 60 years, and new diagnostic techniques are currently under investigation to improve the sensitivity and specificity of diagnosis.

More specific markers of infection-associated pyuria—including neutrophil gelatinase-associated lipocalin, a protein from neutrophil granules—are under investigation.185,186

Enhanced quantitative urine culture has been reported to detect bacteria in specimens that are reported as sterile by standard culture techniques, including in almost one-half of women with severe symptoms.183,184

Because clinical microbiology laboratories process a high volume of urine cultures and the majority of these cultures are negative, flow cytometry techniques to screen samples for pyuria and bacteriuria and only culture those likely to be positive have been utilized.187–189

In clinical practice, localization relies mainly on clinical symptoms and signs.

A Cochrane Database review did not find evidence to support the routine use of procalcitonin, C-reactive protein, or erythrocyte sedimentation rate to differentiate pyelonephritis from cystitis in children.191 At present, there is no biomarker that has sufficient evidence to support routine use in clinical practice

All symptomatic UTI is usually treated because the relief of symptoms is a clear-cut benefit. However, symptoms of cystitis often subside spon- taneously or following symptomatic treatment.79,192–194

In one study, treatment of asymptomatic bacteriuria in young women with recurrent UTI increased the frequency of symptomatic infection.195

A rational approach to the treatment of UTI depends on an appreciation of the prognosis of the untreated infection and the long-term results to be expected from therapy.

Asymptomatic bacteri- uria serves as a marker for debilitating diseases, which in turn may contribute to mortality. In addition, bacteriuria is common in older adults, and many of these patients become reinfected or relapse after antimicrobial therapy.

Treatment of pregnant women with asymptomatic bacteriuria is most likely to be beneficial and is strongly recommended; however, it must be acknowledged that the overall quality of evidence to support this recommendation is not high.197

Hospitalized patients with asymptomatic bacteriuria have higher mortality rates than those without bacteriuria.199 This observation may be related to deaths from bacteremia in patients with indwelling urinary catheters.

Some patients have such frequent symptomatic episodes (either relapses or reinfections) that they are almost chronically incapacitated. In these patients, it may be necessary to give prolonged therapy or prophylaxis to prevent recurrent symptoms

In an adult male with a UTI, the possibility of complicated UTI is much higher than in a female, and the existence of a correctable obstructive lesion must be considered. At a minimum, a postvoiding ultrasound of the urinary tract should be obtained to evaluate bladder emptying. For management of patients with an indwelling catheter, see Chapter 302

Although there is a theoretical basis on which to postulate a benefit from hydration and acidification of the urine in the setting of symptomatic UTI, there is no evidence to support the efficacy of these nonphar- macologic interventions, and they should not be recommended.

Two recent randomized controlled trials have studied the potential role of nonsteroidal antiinflammatory agents (NSAIDs) in the management of acute uncomplicated cystitis.200,201 In both studies, NSAIDs were inferior in terms of rapidity of symptom resolution, and women who received NSAIDs had a higher rate of subsequent pyelonephritis. Although the initial use of analgesics clearly results in less antibiotic use, symptom duration is prolonged and the risk of complications (pyelonephritis) may be higher, so this approach should not be routinely recommended at this time.

There may be a role for delayed antibiotic treatment in special circumstances such as multiple antibiotic allergies, or recurrent Clostridioides difficile (formerly Clostridium difficile)–associated diarrhea, since the rate of spontaneous resolution of acute cystitis is substantial

Urinary analgesics such as phenazopyridine hydrochloride (Pyridium) are not recommended because of lack of evidence to support additional benefit in addition to antibiotic therapy.

The use of vaginal estrogens has been shown to be an effective strategy to prevent recurrent UTIs in postmenopausal women and is recommended in a clinical practice guideline.202,203

The benefit of topical estrogens is believed to be related to decreasing the vaginal pH and increasing colonization of the vagina with lactobacilli, thereby decreasing colonization with Enterobacteriaceae.

A Cochrane Database review concluded that there was little or no benefit of cranberry juice.204 However, a meta-analysis published the same year found that there was a protective effect while noting substantial heterogeneity across the included clinical trials.205 A more recently published meta-analysis also suggested a protective effect; however, again the heterogeneity of both the patient populations studied and cranberry products used must be acknowledged.206

At present, there are not sufficient data to make a clear recommendation, and further large trials utilizing a standardized cranberry product are needed

Vaccines to prevent UTI have been under study for several decades, and several promising candidate vaccines are currently in phase III clinical trials.208,209

Given two or more drugs with good activity against the probable infecting microorganism, the agent with the least toxicity, the least likelihood of affecting the normal flora of the vagina and gastrointestinal tract, the narrowest spectrum, and favorable cost should be chosen

There is no evidence to support any superiority of bactericidal drugs over bacteriostatic agents in UTI. However, there may be theoretical reasons for using bactericidal drugs in the treatment of relapsing UTI

A poor correlation exists between the response of bacteriuria and serum levels of antimicrobial agents. In the dosages commonly used for UTI, some oral antimicrobial agents do not achieve serum levels above the minimal inhibitory concentration for most urinary pathogens. The disappearance of bacteriuria is closely correlated with the sen- sitivity of the microorganism to the concentration of the antimicrobial agent achieved in the urine.210

It is probable that the area in which it is most important to reach antibacterial concentrations in pyelonephritis is the intramedullary part of the kidney. Urine concentrations that are inhibitory for sensitive microorganisms are achieved after oral administra- tion of essentially all antimicrobial agents commonly used for UTIs.

Although serum levels do not seem to be important in the treatment of UTI, they may be critical in patients with bacteremia and may be important in the cure of patients with renal parenchymal infection who relapse.

In patients with renal insufficiency, dosage modifications are necessary for agents that are excreted primarily by the kidneys and cannot be cleared by any other mechanism. In renal failure, the kidney may not be able to concentrate an antimicrobial agent in the urine, and there may be difficulty in eradicating bacteriuria.211 This may be an important factor in the failure of therapy for UTI with aminoglycosides

In addition, high concentrations of magnesium and calcium, as well as a low pH level, can raise the minimal inhibitory concentration of aminoglycosides for gram-negative bacilli to levels above those achievable in the urine of patients with renal failure.

In general, the penicillins, cephalosporins, and many fluoroquinolones attain adequate urine concentrations, despite severely impaired renal function. From the point of view of safety, the penicillins and cephalosporins should be considered first-line agents in the treatment of UTIs in patients with renal insufficiency.

If therapy is appropriate, clinical response should occur within 24 hours with treatment of cystitis. With pyelonephritis, response should occur by 48 to 96 hours. Lack of response by 72 hours should be an indication for imaging studies.

Bacteriologic Cure This term is defined as negative urine cultures on chemotherapy and during the follow-up period, usually 1 to 2 weeks. However, it must be understood that many of these patients will develop reinfection at a later time.

Bacteriologic Persistence This term has been used in two ways to describe a response to therapy: (1) persistence of significant bacteriuria after 48 hours of treatment, and (2) persistence of the infecting organism in low numbers in urine after 48 hours. Significant bacteriuria usually persists only if the urinary levels of the antimicrobial agent are below the concentration of the drug needed to inhibit the microorganism.

Bacteria may persist in the urinary tract during therapy without excretion of organisms in the urine. Sites of persistence in the urinary tract are the renal parenchyma, calculi, and prostate. Persistence in bladder cells is a theoretical possibility.212

Bacteriologic Relapse This usually occurs within 1 to 2 weeks after the cessation of chemotherapy and is often associated with renal infection, structural abnormalities of the urinary tract, or chronic bacterial prostatitis.

Relapse indicates that the infecting microorganism has persisted in the urinary tract during therapy. However, an apparent relapse can be related to reinfection (new infection) with the same microorganism. In spite of eradication from the urinary tract, the original infecting organism may still be present in the periurethral area, vagina, or intestine and then may cause a new infection.

Relapses occur- ring within 1 week are usually true relapses (i.e., from within the urinary tract).

Reinfection After initial sterilization of the urine, reinfection may occur during the administration of chemotherapy (also called superinfection) or at any time thereafter. Reinfection is easy to identify when there is a change in bacterial species. However, there may be reinfection with a different strain of the same species (usually E. coli) or even the same strain

With upper tract infection, agents that give antibacterial serum activity, such as fluoroquinolones, TMP-SMX, β-lactam antibiotics, and ami- noglycosides, should be used.

With lower tract infection, additional agents such as nitrofurantoin (with attention to a requirement for creatinine clearance of at least 60 mL/min) and oral fosfomycin can be used.

In female outpatients in the United States, resistance of E. coli urinary isolates to TMP-SMX was 30% among those 18 to 64 years of age and 26% percent among those 65 years of age and older; resistance to ciprofloxacin was 12% and 22% in those two age groups, respectively.213 Nitrofurantoin resistance was less than 5% in both age groups.

The most concerning resistance problem for the management of UTIs is the increasing incidence of community-acquired infection due to ESBL-producing E. coli. While the incidence of community-acquired UTI in the United States is currently low, it is increasing.218,219

There is evidence that cell wall–active agents (e.g., penicillins, cephalosporins) are not as effective in eradicating infection in the kidneys, or for that matter anywhere in the urinary tract, as are TMP-SMX, fluoroquinolones, or aminoglycosides.175,181

UTI in children younger than 3 months of age is usually caused by E. coli or Enterococcus faecalis.

Prophylaxis with an antimicrobial agent following treatment for febrile UTI in infants and young children is controversial. If prophylaxis is given, it should be reserved for children with associated congenital renal anomaly such as high-grade vesicoureteral reflux, particularly with renal scarring.106 Antireflux procedures (endoscopic or open surgery) are considered in those older than 12 months of age with high grades of reflux and scarring.223

Postnatal ultrasounds are mainly restricted to those children at highest risk for kidney damage and underlying abnormalities. These children include those who are seriously ill, those with possible obstruction (e.g., poor urine flow, abdominal or bladder masses, elevated creatinine, lack of response to antibiotics), those infected with non–E. coli organisms, and those with recurrent UTIs.

Patients who are severely ill with pyelonephritis should be hospitalized and treated intravenously; because the response to therapy is often rapid, initial admission to an observation unit is often appropriate. The role of blood cultures in patients who require hospitalization continues to be debated and, provided that the urine culture is obtained prior to the receipt of antibiotic therapy, the utility of blood cultures is questionable. However, for those patients ill enough to require hospitaliza- tion, we continue to recommend blood cultures.

If gram-positive cocci in chains are seen on Gram stain, ampicillin or amoxicillin is probably the agent of choice. When staphy- lococci are implicated on Gram stain, it would be most prudent to use an agent such as vancomycin for inpatients and linezolid or TMP-SMX for outpatients because of the increasing frequency of infection with community-acquired methicillin-resistant staphylococci.

There is good evidence that 7 days of a fluoroquinolone and, with levofloxacin, 5 days of therapy is sufficient for the treatment of uncom- plicated pyelonephritis.224,225 For oral therapy, current international guidelines recommend a 7-day course of ciprofloxacin 500 mg twice daily or 1 g once daily, or a 5-day course of levofloxacin 750 mg once daily if local fluoroquinolone resistance is under 10%.79

When local resistance is greater than 10% fluoroquinolone, therapy should start with an initial, additional, single dose of a parenteral antibiotic such as a 2-g dose of ceftriaxone or a dose of an aminoglycoside or ertapenem to provide coverage while awaiting results of the urine culture.

With use of any agent other than a fluoroquinolone, 14 days of therapy is generally recommended. A recent retrospective study has suggested that shorter (7 day) courses of TMP-SMX may be effective; however, prospective, randomized control trials are needed before this approach can be recommended.226

When a fluoroquinolone cannot be used, oral TMP-SMX (160 mg/800 mg twice daily) is reasonable, but an initial injection of ceftriaxone, ertapenem, or an aminoglycoside is recom- mended pending results of cultures.

Routine imaging studies are not required for women with acute uncomplicated pyelonephritis.

Follow-up urine cultures are not indicated except in pregnancy.

Patients hospitalized for apparent uncomplicated pyelonephritis should receive initial treatment with parenteral antibiotics. Options are a fluo- roquinolone, an extended-spectrum cephalosporin such as ceftriaxone or piperacillin-tazobactam with or without an aminoglycoside initially, or a carbapenem. From a stewardship prospective, ceftriaxone is an appropriate choice in most patients.

In those with severe sepsis or septic shock, initial therapy with a carbapenem should be considered. Again, from a stewardship perspective, since the microbial etiology of a UTI in this setting should always be eventually confirmed by urine and/or blood cultures, initial broad-spectrum empirical antibiotic therapy in severely ill patients should not be considered problematic.

Patients who develop pyelonephritis in a hospital or long-term care facility setting should be assumed to be infected with the resistant flora of that facility, and therapy should be initiated with antibiotics that are likely to be effective.

Those with bacteremia who respond appropriately do not require a more prolonged course of either parenteral or total antibiotic therapy.

The finding of continuing positive blood cultures or persistent high fever and toxicity past the first 3 to 4 days suggests the need for investiga- tion to exclude complications such as urinary obstruction or intrarenal or perinephric abscess formation. Investigation should include renal ultrasonography, CT or MRI scan, and, according to the findings, a urologic consultation.

In the woman with classic symptoms of cystitis, urine dipstick or cultures are not necessary for management. However, if obtained, a negative dipstick or microscopic examination for pyuria would raise great suspicion that the diagnosis of cystitis is incorrect.

The international clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women recommend treatment of uncomplicated cystitis with nitrofurantoin (100 mg every 12 hours for 5 days), fosfomycin (a single dose of 3 g with 3–4 oz of water), and if local resistance is under 20% of isolates, TMP-SMX (1 double-strength tablet every 12 hours for 3 days).79 Clearly one of the challenges in following these recommendations is that local resistance rates are usually not known and data obtained from hospital antibiograms likely overestimate resistance rates. Pivmecillinam (400 mg twice daily for 3–7 days) is also recommended but is not available in the United States. The different durations of administration are based on compara- tive clinical trials.

Nitrofurantoin, TMP-SMX, or fosfomycin represents an appropriate initial choice of therapy for cystitis because of generally low rates of resistance (variable with TMP-SMX) and low impact on microbiologic flora.

It has been reported that the efficacy of single-dose fosfomycin is inferior to 7 days of nitrofurantoin and to other standard short-course regimens, and this agent is the most expensive of the three options.79

A potential advantage of fosfomycin is activity against multidrug-resistant pathogens, including ESBL-producing E. coli, which may be important in certain epidemiologic settings outside the United States.

The guidelines also list β-lactam agents other than pivmecillinam, including amoxicillin-clavulanate, cefdinir, cefaclor, and cefpodoxime- proxetil, in 3- to 7-day regimens as appropriate (but not preferred) choices for therapy when other recommended agents cannot be used.79 Because of inferior efficacy and higher incidence of adverse events, β-lactams should be considered only if nitrofurantoin, fosfomycin, and TMP-SMX cannot be used, with fluoroquinolones reserved as the last resort for therapy

Short-course therapy may not be appropriate for women who have a history of previous urinary infection caused by antibiotic-resistant organisms or more than 7 days of symptoms. In these patients, who have an increased likelihood of upper tract infection, 7 to 10 days of therapy are recommended.

If symptoms do not respond, or if they recur, a urine culture should be obtained and therapy chosen according to antimicrobial susceptibility testing. A symptomatic response followed by relapse after therapy is discontinued indicates the probability of upper tract infection and the need for a culture and therapy for upper tract infection

Complicated Urinary Tract Infection, Including Infection in Men Because of the relative infrequency of UTI in men and the relatively high proportion of complicated UTI in men who get infected, UTI in men should be managed as a complicated infection.1,2,229,230

In a male, presence of fever with only symptoms of cystitis may indicate acute prostatitis.

In addition to complications related to urinary obstruction from diseases of the prostate in men, chronic bacterial prostatitis may be difficult to cure

The management of complicated UTI is problematic for a number of reasons.1,2,229,230 Resistant bacteria (such as ESBL-producing E. coli; urease-producing bacteria, including Proteus, Providencia, and Mor- ganella; Pseudomonas; and Acinetobacter) and polymicrobial infection are found much more commonly in complicated UTI than in uncom- plicated disease

The complication may make persistent or relapsing infection much more likely. For example, in the presence of an infected calculus, relapse from within the calculus is usual after therapy is stopped; the presence of a foreign body, such as a drainage device, makes eradica- tion of infection difficult, probably because of formation of biofilms; and renal insufficiency may result in subinhibitory levels of antibiotic in the urine, as well as more antibiotic side effects.

Long-term suppressive antimicrobial therapy has been advocated for struvite stones that cannot be removed in an attempt to prevent stone enlargement.

Urine cultures and antimicrobial susceptibility tests should be obtained in all patients with complicated UTI who require treatment. Imaging by ultrasound, CT scan, and/or MRI is often important to rule out obstructive uropathy and other abnormalities. A urologic consultation is often indicated.

Accurate initial empirical therapy is paramount in the severely ill patient in whom the urinary tract may be the primary source of illness. Fluoroquinolone resistance rates in many communities are high, and there is little evidence that concentration of the drugs in the urinary tract makes them effective in vivo against resistant organisms.231 Escherichia coli and Klebsiella pneumoniae have a high rate of carriage of ESBLs, which inactivate extended-spectrum cephalosporins.

Some knowledge of local and individual resistance patterns should be used to guide therapy. Consideration should include the increased risk of drug-resistant pathogens in hospital-acquired cases, in men, and in patients ages 65 years or older, and decisions should incorporate the results of urine cultures obtained within the previous 6 months.218,228,233,234

On an empirical basis in the critically ill patient, carbapenems, aminoglycosides, polymyxins (colistin or polymyxin), or IV fosfomycin (not available in the United States) may be necessary

Our preference for empirical initial therapy for lower tract infection is a fluoroquinolone pending antimicrobial susceptibility studies, but nitrofurantoin and fosfomycin are reasonable choices. Fosfomycin in particular has maintained activity against many multidrug-resistant bacteria.87 The duration of therapy should be at least 7 days with a fluo- roquinolone or nitrofurantoin. When fosfomycin is used to treat complicated cystitis, there are data to support a regimen of three doses, administered every other day.228

With creatinine clearances of less than 50 mL/min, nitrofurantoin and sulfamethoxazole urine levels will likely be subtherapeutic, and nitrofurantoin is contraindicated in renal failure related to resultant peripheral neuropathy. Urine concentrations of TMP used alone should be high enough to allow its use in patients with impaired renal function.24

With severe renal insufficiency, the β-lactams are the safest agents to use. In general, in the presence of severe renal insufficiency, doses of virtually all antimicrobials must be adjusted. In addition, with renal insufficiency, levels of antimicrobial agent in the urine may be insufficient to inhibit the infecting organism.24

When upper tract infection is complicated by abscesses, more prolonged therapy and perhaps drainage are indicated (see “Perinephric Abscess” and “Intrarenal Abscess” later)

Emphysematous pyelonephritis is most often seen in older female diabetic patients with chronic urinary infections and renal vascular disease (Fig. 72.7). Because of a high mortality rate in spite of appropriate antibiotics and supportive therapy, immediate nephrectomy is frequently indicated for this condition.242,243

A major exception to not screening for and treating asymptom- atic bacteriuria is in those patients who are to undergo traumatic genitourinary procedures associated with mucosal bleeding, such as transurethral prostatectomy and percutaneous lithotomy. There is a high rate of postprocedure bacteremia and sepsis in these patients. To reduce the incidence of bacteremia, a urine culture is obtained several days before the procedure and therapy with a third-generation cephalosporin or another appropriate agent is started 12 hours to just before the procedure. The therapy is usually stopped after the procedure, but some practitioners continue the therapy until any urethral catheter is removed.2,144,245

Relapsing Urinary Tract Infection If a patient relapses after therapy for UTI, the most likely possibilities are that the patient has (1) renal involvement (pyelonephritis); (2) chronic bacterial prostatitis; or (3) a structural abnormality of the urinary tract (e.g., calculi).

Patients who have a symptomatic relapse should have a urine culture and sensitivity study and be treated with a course of antimicrobial therapy for upper tract infection. In the past, it was demonstrated that a 6-week course of therapy with a β-lactam results in a higher cure rate than a 2-week course in patients who relapse after 2 weeks of therapy.

Obstructive lesions can be corrected surgically and should be sought in the evaluation of patients with relapsing infection. Calculi may be a cause of relapse of UTI. The ultimate success of chemotherapy is dependent on the removal of stones. The same is true with drainage devices or other foreign bodies

In others, surgical correction may not be indicated or feasible or no abnormality may be found. Symptomatic relapses must be treated. In men, chronic bacterial prostatitis should be ruled out.

In carefully selected patients, such as those with frequent symptomatic relapses, prolonged periods of therapy such as 4 weeks or longer should be considered. Some agents that can be used for longer-term therapy are amoxicillin-clavulanic acid, cephalexin, TMP-SMX, TMP, and ciprofloxacin in usual doses, and nitrofurantoin in full dosage for 1 week and then half the usual dosage.

Nitrofurantoin and TMP-SMX have been used for long-term chronic suppression in many older studies. Fosfomycin would seem to be useful for this purpose, but there have been few long-term studies. For example, one study used 3 g of fosfomycin every 10 days.246

Patients with symptomatic reinfections can generally be divided into two groups: (1) those who have relatively infrequent reinfections, perhaps only once every 2 or 3 years to several times a year (the more common situation); and (2) those who develop frequent reinfections. The latter are usually young and middle-aged sexually active women who have recurrent cystitis. With infrequent reinfections, each episode is approached as a new episode of infection. Women with reinfec- tions associated with lower tract symptoms can be managed with self-administration of standard short-course therapy at the onset of symptoms.247,24

In some women, symptomatic reinfections are associated with sexual activity. Voiding immediately after intercourse may help prevent reinfec- tion. However, single-dose prophylactic chemotherapy taken after sexual intercourse (e.g., a single-strength TMP-SMX tablet; 100 mg of nitro- furantoin) is a more effective method of decreasing these episodes.248–25

In older studies, long-term chemoprophylaxis was advocated for asymptomatic patients who reinfect frequently and who were thought to be at risk of developing renal parenchymal damage with each reinfec- tion (e.g., young children with severe vesicoureteral reflux and children and adults with obstructive uropathy). This approach is controversial and not generally recommended.251

Screening for bacteriuria and chemoprophylaxis has been considered for all renal transplantation patients in the postoperative period to prevent symptomatic infection.2,153–155 Transplantation popula- tions demonstrate an increased frequency of multidrug-resistant pathogens, complicating decisions for therapy.253 Definitive studies are lacking, but some evidence supports the contention that routine surveil- lance beyond the second month post–kidney transplantation may not change a primary outcome of acute pyelonephritis during 24 months of follow-up.254

Most Candida UTIs occur in patients with indwelling catheters. Removal of the catheter may result in cure of 30% to 40% of patients with candiduria.255

Continuous amphotericin B bladder irrigation or oral fluconazole, 200 to 400 mg/day for 7 days, in association with removing (if possible) or replacing the catheter may be effective short term in eliminating candiduria.91 However, with longer follow-up, oral fluconazole is no more effective than no therapy.255 There is no demonstrated benefit in the treatment of asymptomatic infection, and therefore therapy is not recommended.

Exceptions include only patients who are to undergo procedures involving the urinary tract that result in mucosal bleeding. In these cases, attempts should be made to eliminate or at least suppress the candiduria with oral fluconazole.

In the past, asymptomatic candiduria in renal transplant recipients was treated with fluconazole on the basis of a perceived risk of a destructive process in the grafted kidney. However, a large study of posttransplantation patients failed to show excessive morbidity associated with candiduria or any benefit from eradication of candiduria.256

Infrequently, ascending infection resulting in pyelo- nephritis occurs, especially in patients with diabetes or obstruction. This obviously requires systemic antifungal therapy and attention to obstruction. Fluconazole-susceptible organisms can be treated with this agent at a dose of 200 to 400 mg daily for 14 days.257

Pyelonephritis due to azole-resistant Candida glabrata and Candida krusei should be managed with systemic amphotericin B desoxycholate, with or without oral flucytosine.257

Azoles other than fluconazole and echinocandins are not recommended for the management of UTIs due to low concentrations of active drug in the urine; for the same reason, lipid formulations of amphotericin B are also not recommended

During pregnancy, there is dilation of the ureters and renal pelves, with markedly decreased ureteral peristalsis.145 These changes begin as early as the seventh week of gestation and progress to term. The bladder also decreases in tone so that, late in gestation, it can contain twice its normal contents without causing discomfort.

The urinary tract tends to revert to normal by the second month after delivery. The urinary tract alterations may be at least partly related to hyperestrogenism. Other possible explanations for the altera- tions are obstruction of the ureters by the gravid uterus and hypertrophy of muscle bundles at the lower end of the ureter.

The prevalence of asymptomatic bacteriuria in pregnancy ranges from 4% to 7%

The prevalence of bacteriuria rises with age, sexual activity, in diabetes mellitus, in women with sickle cell trait, and in women with a past history of UTI; the association of parity and that of socioeconomic status are inconsistent.259,260

Acute cystitis and acute pyelonephritis have been shown to complicate 1% to 2% of pregnancies with a microbial spectrum similar to that of asymptomatic bacteriuria in pregnancy and uncomplicated UTI in nonpregnant women

The development of symptomatic pyelo- nephritis late in pregnancy is usually an expression of asymptomatic bacteriuria that was present earlier in the pregnancy.

It has been reported that as many as 40% of the patients with untreated bacteriuria early in pregnancy develop acute symptomatic pyelonephritis later in pregnancy, although, as discussed previously, more recent studies have reported lower rates of pyelonephritis. In contrast, less than 1% of patients whose urine is uninfected early in pregnancy develop acute infection.

Untreated asymptomatic bacteriuria has been associated with preterm birth and low birth weight, although the association is inconsistent across studies; the association is most likely due to the increased risk of pyelonephritis.260

All women should be screened at 12 to 16 weeks’ gestation or at the first prenatal visit, if later. The goal of therapy is to maintain sterile urine throughout pregnancy and thereby avoid the complications associated with UTI

In the treatment of asymptomatic bacteriuria and cystitis, treatment modalities include single-dose fosfomycin trometamol 3 g or cephalexin 500 mg four times a day for 3 to 5 days. A 7-day course of nitrofurantoin or TMP-SMX for 3 days (when susceptibility tests reveal absence of resistance) can be used; however, ACOG has recently released an opinion bulletin discouraging the use of nitrofu- rantoin and TPM-SMX during the first trimester due to concern regarding the possibility of previously underappreciated risk of birth defects.261

Acute pyelonephritis in pregnancy should be managed initially with parenteral antibiotics, such as third-generation cephalosporins for 14 days.

Urine cultures should be obtained 1 to 2 weeks after discontinuing therapy and then at regular intervals (monthly) for the remainder of the pregnancy. Some experts recommend prophylactic antibiotic therapy until delivery; however, a Cochrane Review did not find evidence of additional benefit as compared with close follow-up alone.262

Perinephric Abscess Perinephric abscess is an uncommon complication of UTI.263 The most common predisposing factors are urinary tract calculi and diabetes mellitus. It usually occurs secondary to obstruction of an infected kidney or calyx or, occasionally, secondary to bacteremia. It may occur insidi- ously, and up to one-third of cases may not be diagnosed until autopsy.

The infecting bacteria are usually gram-negative enteric bacilli and occasionally gram-positive cocci when the infection is of hematogenous origin. Multiple bacterial species are present in about 25% of cases, and occasionally fungi, especially Candida spp., can be cultured from the abscess.

Patients have a syndrome suggestive of acute pyelonephritis, with fever, abdominal and flank pain (usually unilateral), and often symptoms of lower tract infection. However, presenting symptoms are frequently nonspecific. The patient has often been ill for 2 or more weeks. The diagnosis should be strongly considered in any patient with a febrile illness and unilateral flank pain who does not respond to therapy for acute pyelonephritis. A palpable mass may or may not be present. Pyuria and proteinuria are frequently found, but about 30% of patients have a normal urinalysis and about 40% have sterile urine cultures.264

Intrarenal Abscess Intrarenal abscess may occur as a consequence of bacteremia, often caused by S. aureus. Hematogenous lesions are usually unilateral, single, and located in the cortex. However, these focal suppurative lesions are being recognized with increasing frequency as a complication of classic acute pyelonephritis and are located in the cortex, medulla, or both.

The clinical setting is usually that of acute pyelonephritis with high fever, severe flank pain, and tenderness, but with no or slow response to appropriate antimicrobial therapy. In the early stages of abscess formation, contrast-enhanced CT may detect intense parenchymatous inflammation and edema in a lobe of the kidney, termed acute lobar nephronia or acute focal bacterial nephritis (Fig. 72.8). Although antibiotics may arrest progression at this stage, coalescence of micro- abscesses (multifocal bacterial nephritis) can lead to intrarenal abscess (Fig. 72.9)

Emphysematous pyelonephritis is an uncommon, severe, life- threatening necrotizing form of acute multifocal bacterial nephritis in which retroperitoneal, extraluminal gas is seen in the renal parenchyma, urinary collecting system, or perirenal space on an abdominal radiograph or CT scan (see Fig. 72.7).

The presence of gas suggests a gas-forming, gram-negative facultative anaerobic uropathogen and occasionally Candida species. Escherichia coli is the most common organism associated with this complication, but Klebsiella spp., P. mirabilis, and Citrobacter spp. may be involved.

This condition occurs most commonly in diabetic patients with or without urinary obstruction.265 Mortality rates have improved significantly in the past 2 decades, and “nephron-sparing” management (percutaneous drainage of localized gas collections as opposed to nephrectomy) is increasingly utilized.266,267

Emphysematous pyelitis is a distinct clinical entity (class 1 emphysematous pyelonephritis) that occurs as a sequela of infection when gas accumulates in the pelvicalyceal system.

Diabetes and urinary tract obstruction are common predisposing factors, and E. coli and K. pneumoniae are commonly implicated pathogens. These patients are generally far less ill and respond well to antibiotics alone in the majority of cases.267

Xanthogranulomatous pyelonephritis (XGP) is an uncommon but severe focal or diffuse chronic infection of the renal parenchyma. Destroyed tissue is replaced by granulomatous tissue containing lipid- laden macrophages (foam cells). Predisposing factors include renal calculi, urinary obstruction, lymphatic obstruction, renal ischemia, secondary metabolic alterations in lipid metabolism, an abnormal host immune response, and diabetes mellitus.268 XGP has a predilection to extend into the retroperitoneal space, leading to internal and external fistulae. The latter complications and mass effect of XGP often mimic those of a genitourinary malignancy.269

Urinalysis is abnormal in 70% of patients with a corticomedullary abscess, whereas it is usually normal in the patient with a hematogenous cortical or perinephric abscess. Confirmation of the diagnosis requires imaging techniques.

The most common CT findings include thickening of the Gerota fascia, renal enlargement, focal parenchymal decreased attenuation, and fluid or gas, or both, in and around the kidney.