Approach to Dysuria and Urinary Tract Infections in Women

Leslie S.-T. Fang

Among adult women, urinary tract infection (UTI) is the most common bacterial infection, with an annual incidence of 12%. More than 50% of women have a UTI by the time they are 32 years of age, and 50% of women with one infection have a recurrence, 25% within 6 months. UTI accounts for 8 to 9 million office visits to physicians annually, with an estimated annual cost approaching $2 billion. As such, UTIs represent a significant source of discomfort for women, much work for primary care practices, and considerable health system expense. In addition, UTIs account for much community antibiotic exposure, necessitating careful consideration of drug selection and duration of therapy to minimize adverse ecologic consequences spurring antibiotic resistance. The primary care clinician needs to take all these factors into consideration when encountering the woman presenting with dysuria and the possibility of UTI.

PATHOPHYSIOLOGY AND CLINICAL PRESENTATION (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 and 16)

Pathophysiology

Uncomplicated UTI in adult women results from ascending infection. Colonization of periurethral tissue occurs with bacteria from the bowel and vagina, which ascend to the bladder via the urethra, occasionally extending to the kidneys through the ureters. Hematogenous seeding of the kidney occurs only in the context of high-grade bacteremia, usually due to invasive organisms such as Staphylococcus aureus.

This colonization of the vaginal introitus has been shown to be the essential first step in the production of bacteriuria and plays an important role in recurrent UTIs. Bacteria that cause UTI (predominantly uropathogenic strains of Escherichia coli) are found in the periurethral area in up to 20% of adult women. In premenopausal women, sexual intercourse plays an important role in colonization and infection; in postmenopausal women, changes in the vaginal mucosa are believed to contribute to susceptibility leading to colonization with Enterobacteriaceae.

Entry of bacteria into the bladder through the relatively short female urethra can occur spontaneously, but certain activities such as sexual intercourse are strongly associated with increased risk of infection. Significant increases in bacteriuria follow 30% of intercourse episodes. In one study, the average 24-year-old woman who had sexual intercourse 3 days of the week had a risk of UTI that was 2.6 times greater than that of a similar female who had not had intercourse during that week.

Other risk factors for infection and recurrence include history of UTIs (especially before age 15 years), having a mother or other first-degree female relative with a history of UTIs, use of a spermicide, having a new sex partner, and exhibiting Lewis blood group antigen nonsecretor phenotypes, Le(a + b-) and Le(a-b-). The use of tampons, wiping from back to front, douching, sitting in a hot tub, and various voiding patterns have been suspected risk factors, but not confirmed when studied.

The establishment of a bladder infection also depends on the virulence of the bacteria introduced, the number of organisms introduced, and, most important, a lapse in normal host defense mechanisms. In patients with normal urogenital tracts, upper tract infections are caused almost exclusively by uropathogenic bacteria with virulence determinants. In contrast, only about 50% of cystitis strains have virulence determinants. While E. coli account for the vast majority of uncomplicated UTIs, infection with Staphylococcus saprophyticus is also noted with some frequency, with risk increased by spermicide application, often in conjunction with use of a diaphragm or condom. Uropathogenic bacteria may remain even after clinical cure.

The most common urinary pathogen in community-acquired UTI is E. coli, accounting for 75% to 95% of organisms grown. S. saprophyticus accounts for 5% to 15% of urinary pathogens. Occasionally, other Enterobacteriaceae (e.g., Klebsiella) are responsible, isolated most often in postmenopausal women. Other isolates such as enterococci, S. aureus, and group B streptococci are believed to be contaminants as are diphtheroids, lactobacilli, and α-hemolytic streptococci.

A number of host defense mechanisms normally act together to decrease the likelihood of infection. Normal voiding eliminates some organisms. Certain chemical properties of the urine are antibacterial; urine with a high concentration of urea, low pH, and high osmolarity supports bacterial growth poorly. The most important host defense mechanism is phagocytosis of bacteria that come into contact with the bladder mucosal surface. Vaginal epithelial cell characteristics also contribute. Susceptibility to UTI correlates with increases in cellular bacterial adhesiveness. Abnormalities in these host defense mechanisms result in recurrent and complicated UTIs.

In about 3% of otherwise healthy women, the lower urinary infection extends proximally through the ureters into the kidneys causing clinical pyelonephritis. The presence of reflux increases the chance that infection will ascend. Once infected urine gains access to the renal pelvis, it can enter the renal parenchyma through the ducts of Bellini at the papillary tips and then spread outward along the collecting ducts to cause parenchymal infection. Anatomic abnormalities and presence of stones may also predispose to upper tract infection and adversely affect ability to clear the organism.

Two basic patterns of recurrence are recognized: (a) relapse, in which the original organism is suppressed but not eliminated by antimicrobial therapy and then clinical infection reappears shortly after the antibiotic is stopped, and (b) reinfection, in which the original organism is eradicated from the urinary tract by antimicrobial therapy and clinical recurrence is caused either by a new bacterial strain or by the same strain persisting in the fecal flora. Approximately 80% of recurrences represent reinfection. Ureteral catheterization studies have demonstrated that most reinfections occur in patients in whom infection is restricted to the bladder, whereas most relapses occur in patients with renal parenchymal infection.

Noninfectious etiologies can also cause urethral and bladder discomfort syndromes simulating cystitis, the most common being the poorly understood but potentially disabling interstitial cystitis/bladder pain syndrome. It is defined by the Society for Urodynamic and Female Urology as “an unpleasant sensation (pain, pressure, discomfort) perceived to be related to the urinary bladder, associated with lower urinary tract symptoms of more than 6 weeks’ duration, in the absence of infection or other identifiable causes.” The underlying pathophysiology remains largely unknown but appears to involve both abnormal nociception and abnormalities of the bladder urothelium (e.g., cystoscopic dilation reveals submucosal hemorrhages and other abnormalities of unclear significance).

Clinical Syndromes

UTIs are associated with a number of clinical syndromes, ranging from acute urethral syndrome to pyelonephritis. Most are accompanied by dysuria, frequency, urgency, and suprapubic or flank discomfort. Other features are unique to each syndrome.

Complicated versus Uncomplicated UTI

The adjectives “uncomplicated” and “complicated” are sometimes used to guide selection and duration of antibiotic therapy (see later discussion). “Uncomplicated” UTI refers to cystitis or pyelonephritis occurring in an otherwise healthy, nonpregnant, premenopausal women with a normal urinary tract, but some persons with so-called complicated disease may respond to the same treatments as those designated “uncomplicated” (see later discussion)

Acute Urethral Syndrome (Symptomatic Abacteriuria/Early Infection and Chronic Interstitial Cystitis/Bladder Pain Syndrome)

This syndrome occurs in about 10% to 15% of women who present with symptoms suggestive of UTI but fewer than 10⁵ organisms per milliliter on urine culture and on urinalysis, few white blood cells (WBCs) and no bacteria. These patients can be subdivided into two groups: those with infection and those without.

Early Infection.

Approximately 70% have some degree of pyuria (more than two to five WBCs per high-power field) and true infection, either with bacterial counts of fewer than 10⁵ organisms or with Chlamydia trachomatis. Those with bacterial counts in the range of 10² to 10⁴ may have early UTI with infection not yet established in the bladder.

Interstitial Cystitis/Bladder Pain Syndrome.

The remaining 30% of women presenting with bladder and urethral discomfort in the absence of pyuria, bacteriuria, and other known urinary tract pathology fall into the category of interstitial cystitis/bladder pain syndrome. Their characteristic presentation in addition to dysuria and urinary frequency is one of chronic suprapubic or pelvic pain relieved by voiding, often in association with dyspareunia suprapubic tenderness to palpation. Some note that consumption of alcoholic, caffeinated, diet, or citrus beverages or eating tomatoes, spicy or acidic foods, or chocolate can trigger or worsen symptoms. Patients have normal findings on urinalysis. The course is chronic and can become disabling. Other nociceptive syndromes may be present such as chronic fatigue syndrome, fibromyalgia, or irritable bowel syndrome.

Asymptomatic Bacteriuria (see also Chapter 127)

Among healthy nonpregnant women, asymptomatic bacteriuria is common, but not a risk factor for adverse outcomes. Even though prevalence of asymptomatic bacteriuria is increased in diabetic women, there are no differences between those who are initially bacteriuric and those who are not with regard to symptomatic UTI, mortality, or progression to diabetic complications. Similarly, among older women residing in the community or in geriatric apartment complexes, there is no excess risk associated with asymptomatic bacteriuria. Women in long-term care facilities with asymptomatic bacteriuria show no increase in mortality compared to those who are abacteriuric; even those who have indwelling catheters have no observed increase in risk of bacteremia, though mortality is increased (believed related to factors unassociated with bacteriuria—no reduction in mortality associated with treatment of infection).

Symptomatic Bacteriuria (Cystitis and Pyelonephritis)

Symptomatic bacteriuria is the most common of the clinical syndromes. In otherwise healthy women, most cases represent cystitis (infection confined to the bladder), traditionally believed to present primarily as frequency, urgency, dysuria, hematuria, and bacteriuria. Pyelonephritis (infection that has spread to the kidney) accounts for about 3% of uncomplicated cases and classically manifests additional symptoms of flank pain, fever, and sometimes nausea and vomiting. However, numerous investigations have found considerable overlap in symptoms between the two forms of UTI. Many patients with documented upper tract infection present with symptoms supposedly characteristic of lower tract infection. Moreover, patients whose infection is limited to the bladder may occasionally have fever, flank pain, and systemic symptoms usually associated with pyelonephritis. Thus, the traditional clinical clues are at best imprecise for identifying the site of infection, but their presence does help establish the presence of UTI (see Workup). Hematuria may be noted in up to half of cases.

Clinical Course

In most cases of uncomplicated cystitis, the clinical course is benign, with as many as 40% of patients experiencing spontaneous resolution of symptoms without curative treatment and only about 3% progressing to pyelonephritis. However, it is not uncommon for such patients to experience recurrent infection, especially if sexually active.

Groups frequently bothered by recurrent infections include (a) sexually active women, who report a temporal relationship of urinary symptoms to intercourse; (b) patients with host defenses compromised by underlying systemic illness or residual urine in the bladder; (c) patients with upper tract infections; and (d) pregnant women.

The consequences of recurrent uncomplicated infections are, for the most part, minimal and rarely result in progressive renal impairment. Patients with risk factors for complicated disease (e.g., diabetes, stone disease, pregnancy, immunocompromise) are at greater risk. However, patients with infections in the setting of vesicoureteral reflux, pregnancy, or diabetes are at greatest risk. Vesicoureteral reflux is associated with residual urine in the bladder, ascending infection, chronic pyelonephritis, and a high risk for renal scarring, which leads to focal glomerulosclerosis, proteinuria, and progressive renal failure. Patients most likely to have vesicoureteral reflux are those who report a long history of UTIs beginning in childhood. UTI during pregnancy has been linked to increased rates of fetal complications and prematurity, especially when the infection occurs within 2 weeks of delivery. The mother has an enhanced risk for pyelonephritis. Patients with diabetes show an increased susceptibility to upper tract infection.

DIFFERENTIAL DIAGNOSIS (9,11,15,17,18)

Dysuria

The differential diagnosis of dysuria includes UTI, vaginitis, and urethritis. Patients with vaginitis may occasionally be mistakenly believed to have UTI. Vaginal discharge, “external” discomfort (from urinary irritation of inflamed labial tissue), absence of frequency or urgency, and urine cultures negative for bacteria distinguish vaginitis from UTI. Trichomonas vaginalis and Candida albicans are the most commonly responsible organisms. Women with dysuria and an absence of bacterial growth on routine urine culture may have urethritis caused by Neisseria gonorrhoeae or herpes simplex virus, although, as previously noted, most cases are caused by C. trachomatis (see Chapter 125). The onset is
usually gradual, dysuria is mild, and vaginal discharge may be present. Pelvic pain and vaginal or cervical discharge suggest spread of infection into the cervix and fallopian tubes, a serious development (see Chapters 116 and 117).

Pyuria typically accompanies gonococcal and trichomonal infection as well as chlamydial infection. Patients with acute urethral syndrome and no pyuria may have dysuria on the basis of local trauma or irritation rather than infection, problems that occur in postmenopausal women secondary to desiccation of vaginal and urethral tissue.

Flank Pain

Patients with renal calculi or embolic infarction may present with flank pain and hematuria, mimicking pyelonephritis. However, urine cultures are sterile, and no bacteria are seen on Gram stain, as they are in UTI.

WORKUP (9,11,15,17, 18, 19, 20, 21, 22, 23 and 24)

The pace, extensiveness, and order of the evaluation are largely dictated by the patient’s clinical presentation. Patients presenting with fever, flank pain, and systemic symptoms require prompt evaluation for the possibility of urinary tract obstruction and superimposed infection. They should be questioned about a history of diabetes, sickle cell anemia, and excessive analgesic use—risk factors for renal papillary necrosis and subsequent obstruction by sloughed papillae. Likewise, a history of renal calculi is cause for concern in this setting. Any person with such risk factors who appears to be toxic (high temperature, prostration) and restless on examination and who has marked tenderness in the costovertebral angle requires immediate hospitalization for urinary tract imaging and initiation of parenteral antibiotic therapy (see Management and Indications for Admission). Infection behind an obstruction constitutes a medical and urologic emergency necessitating urgent urologic consultation and consideration of interventional therapy. In the absence of such a presentation, the workup can proceed on an outpatient basis and in some instances might even be conducted over the telephone or by e-mail.

History

The acutely dysuric but otherwise nontoxic woman should first be questioned about vaginal discharge, external irritation on urination, and pain on intercourse as well as absence of urinary frequency and urgency to differentiate vaginal causes of dysuria from those referable to the urinary tract. Also helpful is a sexual history to identify risk factors for chlamydial urethritis, including new sexual partners, a partner with a penile discharge or recent urethritis, mucoid vaginal discharge, and gradual onset of symptoms. A recent history of gonorrhea or exposure to gonorrhea should be elicited.

Patients giving a history of dysuria, frequency, urgency, or back pain without vaginal discharge or vaginal irritation have a very high likelihood of UTI (96%). On the other hand, patients without dysuria and only vaginal symptoms are much less likely to have a UTI. Checking the history fever, flank pain, nausea, and vomiting can help increase the pretest probability of UTI, but as noted, such symptoms do not have strong discriminant value for a diagnosis of pyelonephritis.

History findings suggestive of interstitial cystitis/bladder pain syndrome include bladder pain, relief with voiding, dietary precipitation of symptoms, and concurrent nociceptive conditions such as fibromyalgia, irritable bowel syndrome, and chronic fatigue syndrome.

Laboratory Studies

For persons with mild symptoms of uncomplicated cystitis, laboratory testing is unnecessary; characteristic clinical features are sufficient for diagnosis and choice of treatment (see Management). For persons with dysuria accompanied by severe symptoms or risk factors for complicated disease (see earlier discussion), urinalysis and culture are of prime importance.

Urinalysis

Specimen Collection.

Proper collection of the urine specimen is essential. The clean-voided technique has withstood the test of time and minimizes contamination from vaginal and labial sources. The patient is told to straddle or squat over the toilet and to spread the labia with the nondominant hand. This position is maintained throughout collection. With the other hand, the vulva is swabbed front to back with three sterile gauze pads soaked in sterile water or with a sponge soaked in a mild nonhexachlorophene soap. A small amount of urine is then passed. This is a urethral specimen and can be saved if bacterial or protozoan urethritis is suspected. More urine is voided and collected in a sterile cup. Alternatively, the patient can be told to slide the cup into a freely flowing stream to collect a true midstream specimen. The adequacy of collection can be confirmed by examining for epithelial cells; their presence indicates vulvar or urethral contamination.

With an elderly patient, the assistance of a family member or a nurse may be needed. When repeated contamination is suspected, straight catheterization of the bladder can be performed with relatively little risk.

Dipstick Testing.

Dipstick testing of the urine can be helpful in diagnosis if there is doubt clinically. A dipstick urinalysis positive for leukocyte esterase and/or nitrites in a midstream-void specimen reinforces the clinical diagnosis of UTI; a completely negative dipstick result has a high negative predictive value. The leukocyte esterase component of dipstick testing is moderately to highly sensitive (75% to 96%) and specific (94% to 98%) for UTI (when defined as >10⁶ CFU/mL). Sensitivity of the nitrite component test is lower (35% to 85%), reduced by the presence of urinary pathogens that do not reduce nitrate (e.g., enterococci, S. saprophyticus, Acinetobacter) or by a very dilute urine. However, specificity is 95%. Although microscopic examination of urine sediment or examination of Gram stain of the unspun urine might add a modicum of specificity and sensitivity (e.g., using criteria such as a single organism on unspun urine), these tests are usually impractical in the busy primary care office setting and rarely necessary.

Only gold members can continue reading. Log In or Register to continue