- Email: [email protected]
Urinalysis in the Diagnosis of Urinary Tract Infections
Urinalysis
Urinalysis in the Diagnosis of Urinary Tract Infections Leslie S. T. Fang, MD"
Among women, acute uncomplicated urinary tract infection is the most common of all bacterial infections, affecting an estimated 6 per cent of women each year.30An additional 3 per cent develop so-called acute urethral syndrome: The patients would develop symptoms of lower urinary tract infection (dysuria, frequency, and suprapubic discomfort) without significant bacteriuria on culture (less than 10 organisms per mL of urine).35 Uncomplicated urinary tract infections are uncommon in males. Very few men acquire urinary tract infections owing to uncomplicated superficial mucosal infection of the bladder. Most who do have had the urinary tract instrumented, or have had a prostatic infection or some other abnormality of the urinary tract. In all instances, the first laboratory test usually used to confirm the clinical suspicion of urinary tract infection is the urinalysis. Urinalysis can be used to diagnose the presence of an infection, to localize the site of infection in some instances, and even to guide therapy in some patients. URINALYSIS IN THE DIAGNOSIS OF URINARY TRACT INFECTIONS Urinalysis is extremely helpful in the diagnosis ofbacterial infections of the urinary tract, but it is important that the specimen be collected correctly. If urinary tract infection is suspected either on the basis of clinical symptoms or on a random urine specimen, the patient should be instructed to collect a mid-stream clean voided specimen for urinalysis and for culture. In the female, the specimen should be discarded and repeated if many squamous cells are seen on the examination of the urinary sediment, because the specimen is in all likelihood contaminated by vaginal flora.
" Assistant Professor of Medicine, Harvard Medical School, and Associate Director, Dialysis Unit, Massachusetts General Hospital, Boston, Massachusetts Clinics in Laboratory Medicine-Vol.
8, No. 3, September 1988
567
568
LESLIE S.T. FANG
In a correctly collected specimen, urinary tract infections can easily be diagnosed in most instances. On dipstick, positive tests for blood and leukocytes would raise the concern for active infection. Examination of the sediment is important in the confirmation of the diagnosis: Red cells, white cells, and bacteria are usually seen under the high-power field. In some instances, white cell casts can be seen in patients with acute pyelonephritis. Other helpful tests include Gram stain of a dried unspun sample of urine. Presence of bacteria in such an examination usually denotes clinically significant urinary tract infection. Usually, more than 1 0 organisms per mL of urine would be found on culture of the urine specimen. URINALYSIS IN THE DETECTION O F BACTERIAL INFECTION IN ASYMPTOMATIC PATIENTS The prevalence of bacteriuria depends on the age and sex of the patient. Among neonates, positive cultures are found in about 1per cent of both males and females. During the school-age years, the prevalence of bacteriuria among boys is less than 0.03 per cent, compared to 1to 2 per cent among girls. Prevalence among females increases by 1per cent per decade of life. Through the child-bearing age, the prevalence is about 3 to 4 per cent, and by age 5 0 the prevalence is as high as 1 0 per cent. Geriatric men have a high prevalence of bacteriuria because of the high incidence of prostatic infection and urologic procedures that can lead to colonization and infection. Geriatric women also have a high prevalence of bacteriuria b e c a ~ s eof large post-void residuals secondary to pelvic floor relaxation. In some series, the prevalence of bacteriuria is as high as 20 per cent in patients above the age of 65. Urinalysis can be used for the detection of asymptomatic bacteriuria. However, it is not clear that routine screening is important for everybody. Although some concerns were originally raised about the relationship between recurrent urinary tract infection with renal failure and h v ~ e r t e n s i o n recent .~~ studies on bacteriuria in both men and womehlfail to demonstrate progressive renal disease in the absence of anatomic abnormalities, hypertension, analgesic abuse, or other systemic conditions predisposing patients to renal insufficiency.', l4 At one time routine screening had been advocated for children because of concern over the effect of bacteriuria on progressive renal damage. However, the currently available information suggests that the renal damage " is rarelv due to infection alone, but rather to infection su~erimposed upon an anatomic abnormality, such as vesicoureteral reflux. Equally important is the finding that the majority of the renal scarring occurs before aee 5. Two recent studies have suggested that if renal scarring has not Gccurred by age 5 , even in the f a c E f continuing bacteriuria and vesicoureteral reflux, renal growth will continue unim~ a i r e d .22~It . is therefore difficult to advocate routine screening. " of children for urinary tract infection. In elderly patients, asymptomatic bacteriuriais quite common. Male patients may have prostatic outlet obstruction and female patients may 139
&
have large post-void residuals in the bladder because of pelvic floor relaxation, predisposing the patient to infection. Prevalence of urinary tract infection may b e as high as 1 0 to 20 per cent in these patients. However, it is not clear that treatment of asymptomatic bacteriuria in these patients would substantially change the natural history of the disease. The original concern of chronic bacteriuria resulting in chronic renal failure, however, has not been borne out in both prospective and retrospective studies. Prospective studies of the natural history of recurrent bacteriuria in over 1000 patients for periods of up to 1 2 years have not demonstrated progressive renal damage in the absence of significant anatomic abnormalities. A retrospective study of 101 consecutive cases of renal failure due to chronic interstitial nephritis seen at a major medical center has likewise failed to reveal a single case in which infection was the primary cause of the renal disease.25These studies suggest that aggressive interventions in asymptomatic patients are probably not warranted. In practice, asymptomatic bacteriuria often becomes symptomatic, and at least one attempt at eradicating the infection is probably indicated if the responsible organism is susceptible to oral antimicrobial agents. If toxic or parenteral agents are required, treatment is probably not indicated. However, in certain groups of patients, urinary tract infection carries considerable morbidity and routine screening would be important. These groups include pregnant females, patients with known structural anatomic abnormalities in the genitourinary tract, and patients with recent instrumentation of the urinary tract. In pregnant women, the prevalence of bacteriuria is at least twice that among similarly aged nonpregnant females, and some 4 to 1 0 per ~, baccent of pregnancies are marked by b a ~ t e r i u r i a .2s~ Asymptomatic teriuria will result in symptomatic urinary tract infections in at least half of the patients who were untreated, and up to 40 per cent will develop full-blown acute pyelonephritis. Bacteriuria has also been associated with an increased rate of fetal loss and low birth weight^.^, 26* 40 Randomized trials of treatment of asymptomatic bacteriuria of pregnancy have demonstrated efficacy in reducing the incidence of pyelonephritis and low birth weight at delivery. Routine screening is therefore important for both maternal and fetal health. Routine screening is also important in patients with known anatomic abnormalities in the urinary tract. Whereas long-term prospective studies have failed to demonstrate progressive renal disease in the absence of anatomic abnormalities, infection in patients with known genitourinary tract abnormalities will cause rapid renal scarring. In patients with spinal cord injury and vesicoureteral reflux, infection resulted in renal scarring and the development of hypertension in 30 per cent ofthe p a t i e n t ~ . ~is~ I t therefore reasonable to screen for infection in patients with known bladder outlet obstruction, vesicoureteral reflux, and nephrolithiasis. Finally, routine screening should also be done in patients who have recently been instrumented. The risk of introducing infection during catheterization of the bladder may be as high as 5 per cent. Prompt therapy in these patients can eliminate the morbidity associated with symptomatic urinary tract infections. 317
570
LESLIE S.T. FANG
URINALYSIS IN THE DIAGNOSIS OF URINARY TRACT INFECTION IN SYMPTOMATIC PATIENTS The most important role of the urinalysis is in the diagnosis of urinary tract infection in patients presenting with symptoms. Seventy per cent of patients presenting with dysuria, frequency, and suprapubic pain will have bacterial infection. The remaining 30 per cent would have insignificant growth on urine cultures and are diagnosed to have acute urethral syndrome. In patients presenting with bacterial infections, the presence of hematuria, pyuria, and bacteriuria would be adequate for the initiation of therapy. Clinically significant bacteriuria would be expected if bacteriuria is detected on an unspun specimen. This is particularly relevant if a Gram stain is performed on an unspun, dried urine specimen. The presence of organism under high-power field examination is highly suggestive of significant bacteriuria. An added advantage of the urinary Gram stain is a preliminary delineation of the organism involved (gram-positive cocci versus gram-negative rods) so as to guide initial therapy.
URINALYSIS IN ACUTE URETHRAL SYNDROME IN ADULT WOMEN Dysuria and urinary frequency are the most common complaints in females with urinary tract infections. However, fully one third of patients with these complaints do not have significant bacteriuria on culture. These patients are diagnosed to have acute urethral syndrome.30, 349 35 The urinalysis is quite helpful in the management of these patients. Patients with acute urethral svndrome can be divided into two maior groups on the basis of the uri&alysis: those with pyuria and those wGhAbout 70 per cent of patients with acute urethral syndrome have pyuria on urinalysis and the remainder do not. The patients with pyuria have true microbial infection due to Chlamydia trachomatis, l-form of gonococci, Staphylococcus saprophyticus, or some other classical uropathogen, but in smaller number than conventionally recognized (less than 10 organisms per mL of urine). Patients without pyuria have no known microbial cause of their symptoms, and the dysuria and frequency may be related to mechanical trauma or i r r i t a t i ~ n . ~ ~
URINALYSIS IN THE LOCALIZATION OF THE SITE OF INFECTION Urinalysis is at times helpful in the localization of the site of urinary tract infection. It is now clear that infection at different sites within the urinary tract has very different pathologic characteristics and different treatment requirement^.'^-'^ Bladder infection is a superficial mucosal infection at a body site where high concentrations of antibiotics can be delivered. Considerable data have accumulated indicating that these
infections can be treated with a single dose of an appropriate a n t i b i o t i ~ . ~ , 19,38 This mode of therapy is less expensive, is accompanied by a high rate of patient compliance, and is associated with fewer In contrast, renal and prostatic infections are deep, side effects2, parenchymal infections requiring a more intensive and prolonged therapy. In patients with documented upper tract infections, conventional therapy with a 14-day course of antibiotics failed in a significant prsportion of ~ a t i e n t s . ' ~ - ' ~ ~ e c a u s the e therapeutic responses are in large part dependent upon the site of the urinary tract infection, numerous invasive and noninvasive ~ r o c e d u r e shave been devised to h e l ~to localize the site of infection (Table 1). Unfortunately, none of the noninvasive tests are extremely sensitive or reliable in the localization of infection. However, two features easily examined by the urinalysis can provide some clues of the site of infection: concentrating ability and the finding of white cell casts on urinalysis. One of the earliest noninvasive tests devised to localize the site of infection utilizes the observation that patients with upper tract infections differ from those with lower tract infections in that they often have an abnormality in renal concentrating ability. The inability to concentrate is most obvious when maximal urinary concentrating ability is examined. In one representative study, Ronald et al. (Am J Med 46:126, 1968) studied the maximal concentrating ability in 38 patients whose site of infection was directlv localized bv ureteral catheterization. Thev demonstrated that renal but not bladdLr bacteriuria was associated with a decreased concentrating ability. Bilateral infection was associated with a greater defect than unilateral infection. With eradication of infection, concentrating ability improves. Even in this study, however, the overlap in maximal urinary osmolarity achieved in patients with bladder, unilateral, and bilateral renal infections is considerable. The test is therefore neither a sensitive nor a consistentlv reliable method of localization. Urinary specific gravity provides an easy means of examining the concentrating ability. Persistingly dilute urine may suggest upper tract infection. However, because many patients would be forcing fluids with the onset of symptoms of urinary tract infection, it would be hard to interpret the significance of the urinary specific gravity in this setting. Urinary concentrating ability is therefore a poor index, at best, in the delineation of the site of infection. In contrast, presence of white cell casts is highly suggestive of an upper tract infection in the appropriate patients. The differential diagnosis ofwhite cell casts is quite limited, incorporating (1) pyelonephritis; Table 1. Invasive and Noninvasive Studies in the Localization of the Site of Urinay Tract Infection INVASIVE STUDIES
NONINVASIVE STUDIES
Bilateral ureteral catheterization Bladder washout studies
Maximal urinary concentrating ability C-reactive protein Urinary enzymes Antibody-coated bacterial assay
572
LESLIE S.T. FANG
( 2 ) acute allergic interstitial nephritis, usually drug induced; and (3) rarely in patients with cholesterol emboli to kidneys. White cell casts are most commonly seen in patients with acute upper tract infections. In the correct clinical scenario, the presence of white cell casts is pathognomonic of upper tract infections. However, the absence of white cell casts does not rule out upper tract infections. In invasive studies that have been done to localize the site of urinary tract infections in females, fully one third of the infections with only lower tract symptoms and urinary findings turned out to have renal bacteriuria.'. 9, 30- 33 Conversely, the majority of patients with clear-cut upper tract symptoms including fever, chills, flank pain, and costovertebral angle tenderness do not have white cell casts in the urine. The presence of white cell casts is therefore a relatively specific but a very insensitive means of localization of site of infection. In summary, urinalysis is a very imprecise way of localizing the site of infection in the urinary tract. However, a careful search for the presence of white cell casts should be made in all instances of urinary tract infection so as not to miss an occult upper tract infection. ANTIBODY-COATING OF BACTERIA ASSAY IN THE LOCALIZATION O F THE SITE O F INFECTION Renal infection is associated with the net synthesis of specific antibody directed against antigens of the infecting organism.16 Cotran,7 using fluorescent antibody techniques, demonstrated gamma-globulincontaining cells in the kidney during the course of experimental pyelonephritis. Lehrman et a1.21 injected E. coli into rabbits with one ureter transiently occluded and studied the local immune response. Using in vitro incubation techniques, they were able to demonstrate protein synthesis in the pyelonephritic kidney, with most of this protein being IgG class globulin, significant portions of which were specific antibodies directed against the infecting strain of organism. Given this information, it is not surprising that immunologic techniques have been applied to the problem of infection localization. Percival et al.,27 using a bacterial agglutination test, found elevated serum antibody levels in patients with symptoms of acute pyelonephritis, with these titers falling in response to antibiotic therapy. Patients with clinically inapparent pyelonephritis also had high antibody levels, whereas those with infection limited to the lower tract had normal titers. Clark et localized the site of infection by ureteral catheterization and examined hemagglutinating antibody response and confirmed that some patients with renal infection had higher hemagglutinating antibody titers than patients with bladder bacteriuria. However, a wide range of titers and a considerable overlap between the two groups were seen, so that the determination of antibody titer was of limited usefulness in the individual patient. Thomas,36 Jones,17 and their colleagues, using an immunofluorescence assay, showed that bacteria originating from the kidney were coated with antibody, whereas bacteria associated with lower urinary
tract infection were antibody-negative. In the Jones study, in particular, there was at least 95 per cent correlation with the results of bladder washout tests. The high degree of sensitivity and reliability of this assay were confirmed by other workers. However, both false-positive and false-negative results may occur. Vaginal23 or rectal contaminating a urine specimen may give false-positive results. Patients with heavy proteinuria may also have false-positive results. In males, prostatitis may also be associated with the antibody coating of the infecting bacteria." False-negative results in adults may occur early in the course of renal infection, during the time necessary to generate the antibody response. In experimental hematogenous pyelonephritis in the rabbit, Smith et al.32showed a time lag of 11 to 15 days between introduction of the organism and the development of a positive assay. Finally, for reasons as yet unclear, the test appears to be unreliable in children. With the realization of these limitations, the test can be areasonable means of localization of the site of infection. URINALYSIS AS A GUIDE TO THERAPY O F URINARY TRACT INFECTIONS In some instances, in addition to the diagnosis of the presence of infection, the urinalysis can actually guide therapy of infection: (1) In adult females with acute urethral syndrome, urinalysis is of great therapeutic importance. Patients with acute urethral syndrome and pyuria usually respond to a 14-day course of doxycycline. Those patients with acute urethral syndrome without pyuria are best treated with analgesics and attention to personal hygiene and sexual practices. (2) In patients with symptoms of urinary tract infection and white cell casts on urinalysis, considerable concern should be raised for the possibility of acute pyelonephritis, and the patients should be considered for a more intensive and prolonged course of antibiotics. In the appropriate patient, urologic and radiologic studies may be necessary. (3) In patients with pyuria and bacteriuria, a Gram stain of the urine can yield some preliminary information about therapy. Gram-positive cocci are likely to be enterococci or staphylococci and should be treated differently than gram-negative bacilli. (4) In patients with pyuria and bacteriuria, an alkaline pH on urinalysis should raise the index of suspicion for an infection with a urea-splitting organism such as Proteus and, less commonly, Pseudomonas and Klebsiella. Concerns should also be raised about the possibility of an infected stone, because the urea-splitting organisms would provide an ideal environment for the formation of struvites. (5) Urinalysis is particularly useful in the follow-up evaluation of patients who have been treated with a course of antibiotics for their urinary tract infection. A benign urinalysis can often obviate the need for a follow-up urine culture. Urinalysis is therefore an easy and important tool in the management of patients with urinary tract infections.
LESLIE S.T. FANG
574 SUMMARY
Urinary tract infection is the commonest human bacterial infection. Bacteriuria alone does not appear to produce progressive renal damage or hypertension. However, it can produce considerable morbidity. Urinalysis is a simple, relatively sensitive, and reliable way of diagnosing urinary tract infection. It is not clear that routine screening should be performed in all patients, but pregnant females, patients with known anatomic abnormalities, and patients with recent genitourinary instrumentation should be screened. The major determinant of therapeutic success in patients with urinary tract infections is the anatomic site of infection. Superficial mucosal infection of the bladder is well treated with a single dose of an appropriate antibiotic, whereas deep tissue infection of the kidney or prostate should be treated with a prolonged and intensive course of therapy. Urinalysis is an insensitive tool in the localization of infection. However, the presence of white cell casts on the examination of the urinary sediment is pathognomonic of upper tract infection and would lead one to pursue an aggressive course of therapy. Examination of the concentrating ability is of limited help in this regard because of the wide range of overlap of concentrating ability in patients with upper and lower tract infections. In selected instances, urinalysis is of help in guiding therapy of urinary tract infections. This is particularly true of the patients with acute urethral syndrome where therapy is guided by the presence or absence of pyuria. Urinalysis, a simple front-line test, is of paramount importance in the evaluation and management of the patient with urinary tract infection. REFERENCES 1. Asscher AW, Chick S, Radfors N, et al: Natural history of asymptomatic bacteriuria in non-pregnant women. In Brumfritt W, Asscher AW (eds): Urinary Tract Infection. London, Oxford University Press, 1973, p 5 1 2. Bailey RR: Single dose therapy of urinary tract infection. Sydney, Australia, ADIS Health Science Press, 1983 3. Braude R, Block C: Proteinuria and antibody-coated bacteria in the urine. N Engl J Med 297:617,1977 4. Brumfritt W: The effects of bacteriuria in pregnancy on maternal and fetal death. Kidney Int (Suppl) 8:S113, 1975 5. Cardiff-Oxford Bacteriuria Study Group: Sequelae of covert bacteriuria in school girls: A four-year follow-up study. Lancet 1:889, 1978 6. Clark H, Ronald AR, Turck M: Serum antibody response in renal versus bladder bacteriuria. J Infect Dis 123:539, 1971 7. Cotran RS: Retrograde Proteus pyelonephritis in rats. Localization of antigen and antibody in treated sterile pyelonephritic kidneys. J Exp Med 117:813, 1963 8. Fairley KF, Bond AG, BrownRB, et al: Simple test to determine the site ofurinary tract infection. Lancet 2:427, 1967 9. Fairley KF, Carson NE, Gutch RC, et al: Site of infection in acute urinary tract infection in general practice. Lancet 2:615, 1971 10. Fang LST, Tolkoff-Rubin NE, Rubin RH: Efficacy of single dose and conventional amoxicillin therapy in urinary tract infection localized by the antibody-coated bacteria technique. N Engl J Med 298:413, 1979
11. Fang LST, Tolkoff-Rubin NE, Rubin RH: Localization and antibiotic management of urinary tract infection. Annu Rev Med 30:225, 1979 12. Fang LST, Tolkoff-Rubin NE, Rubin RH: Urinary tract infections in women. Compr Ther 5:20, 1979 13. Freedman LR, Andriole V: The long-term follow-up of women with urinary tract infection. In Villarreal H (ed): Proc Fifth Int Cong Nephrology. Basel, Karger, 1972, p 230 14. Freedman RB, Smith WM, Richardson JA, et al:.Long-term therapy for chronic bacteriuria in men: U.S. Public Health Service Cooperative Study. Ann Intern Med 83:133, 1975 15. Hellerstein S, Kennedy E, Nussbaum L, Rice K: Localization of the site of urinary tract infections by means of antibody-coated bacteria in the urinary sediments. J Pediatr 92:188. 1978 16. Holmgren J, Smith JW: Immunological aspects of urinary tract infection. Prog Allergy 18:289, 1975 17. Jones SR, Smith JW, Sanford JP: Localization of urinary tract infections by detection of antibody-coated bacteria in urine sediment. N Engl J Med 290:591, 1974 18. Jones SR: Prostatitis as a cause of antibody-coated bacteria in the urine. N Engl J Med 297:365, 1977 19. Kallenius G, Winburg J: Urinary tract infection treated with a single dose of short-acting sulfonamide. Br Med J 1:1175, 1979 20. Kunin CM: Detection, Prevention and Management of Urinary Tract Infection. Philadelphia, Lea & Febiger, 1979 21. Lehrman JD, Smith JW, Miller TE,et al: Local immune response in experimental pyelonephritis. J Clin Invest 47:2541, 1968 22. Lindberg U, Claesson I, Hanson LA, Hodal U: Asymptomatic bacteriuriain schoolgirls. VIII. Clinical course during a %year follow-up. J Pediatr 92:194, 1978 23. Montplaisir S, Cote PA, Martineau B, et al: Localisation du site de l'ingection urinaire chez l'enfant par la recherche de bacteries recouvertes d'anticorps. Can Med Assoc T 115:1096,1676 24. Montplaisir S, Courteau C, Roche AJ: Antibody-coated bacteria in contaminated urine svecimens. N Engl T Med 296:758, 1977 a ~oldb berg M: Chronic interstitial nephritis: Etiologic factors. Ann Intern 25. ~ u i r T, Med 82:453,1975 26. Naeye RL: Cause of the excessive rates of perinatal mortality and prematurity in pregnancies complicated by maternal urinary-tract infections. N Engl J Med 300:819, 1979 27. Percival A, Brumfitt W, DeLouvois J: Serum antibody levels as an indication of clinically inapparent pyelonephritis. Lancet 2:1027, 1964 28. Polk BF: Urinary tract infection in pregnancy. Clin Obstet Gyrlecol 22:285, 1979 29. Ryan RW, Kowalski I, Tilton RC: Mechanisms of the antibody-coated bacteria test (abstract C2). Presented at 77th Annual Meeting Am Soc Microbiol, New Orleans, 1977 30. Sanford JP: Urinary tract symptoms and infections. Annu Rev Med 26:485, 1975 3 1. Sever JL, Ellenberg JH, Edmonds D: Urinary tract infection during pregnancy: Maternal and pediatric findings. In Kass EH, Brumfritt W (eds): Infections of the Urinary Tract. Chicago, University of Chicago Press, 1979, p 1 2 32. Smith JW, Jones SR, Kaijser B: Significance of antibody-coated bacteria in urinary sediment in experimental pyelonephritis. J Infect Dis 135:577, 1977 33. Stamey TA, Govan DE, Palmer JM: The localization and treatment of urinary tract infections: The role ofbactericidal urine levels as opposed to serum levels. Medicine 44:1, 1965 34. Stamm WE, Running K, McKevitt M, et al: Treatment of acute urethral syndrome in women. N Engl J Med 304:956,1981 35. Stamm WE, Wagner KF, Amsel R, et al: Causes of acute urethral syndrome in women. N Engl J Med 303:409,1980 36. Thomas V, Shelokov A, Forland M: Antibody-coated bacteriain the urine and the site of urinary tract infection. N Engl J Med 290:591, 1974 37. Thomas VL, Harris RE, Gilstrap LC 111: Antibody-coated bacteria in the urine of obstetrical patients with acute pyelonephritis. J Infect Dis 131 (Suppl):S57, 1975
576
LESLIE S.T. FANG
38. Tolkoff-Rubin NE, Rubin RH: Urinary tract infection. In Cotran RS (ed):Tubulointerstitial Nephropathies. New York, Churchill Livingstone, 1983, p 49 39. Weiss S, Parker F Jr: Pyelonephritis: Its relation to vascular lesions and to arterial hypertension. Medicine 18:221, 1939 40. Zinner SH: Bacteriuria and babies revisited. N Engl J Med 300:835, 1979 8 Hawthorne Place Suite 104 Boston, Massachusetts 021 14
Urinalysis in the Diagnosis of Urinary Tract Infections Leslie S. T. Fang, MD"
Among women, acute uncomplicated urinary tract infection is the most common of all bacterial infections, affecting an estimated 6 per cent of women each year.30An additional 3 per cent develop so-called acute urethral syndrome: The patients would develop symptoms of lower urinary tract infection (dysuria, frequency, and suprapubic discomfort) without significant bacteriuria on culture (less than 10 organisms per mL of urine).35 Uncomplicated urinary tract infections are uncommon in males. Very few men acquire urinary tract infections owing to uncomplicated superficial mucosal infection of the bladder. Most who do have had the urinary tract instrumented, or have had a prostatic infection or some other abnormality of the urinary tract. In all instances, the first laboratory test usually used to confirm the clinical suspicion of urinary tract infection is the urinalysis. Urinalysis can be used to diagnose the presence of an infection, to localize the site of infection in some instances, and even to guide therapy in some patients. URINALYSIS IN THE DIAGNOSIS OF URINARY TRACT INFECTIONS Urinalysis is extremely helpful in the diagnosis ofbacterial infections of the urinary tract, but it is important that the specimen be collected correctly. If urinary tract infection is suspected either on the basis of clinical symptoms or on a random urine specimen, the patient should be instructed to collect a mid-stream clean voided specimen for urinalysis and for culture. In the female, the specimen should be discarded and repeated if many squamous cells are seen on the examination of the urinary sediment, because the specimen is in all likelihood contaminated by vaginal flora.
" Assistant Professor of Medicine, Harvard Medical School, and Associate Director, Dialysis Unit, Massachusetts General Hospital, Boston, Massachusetts Clinics in Laboratory Medicine-Vol.
8, No. 3, September 1988
567
568
LESLIE S.T. FANG
In a correctly collected specimen, urinary tract infections can easily be diagnosed in most instances. On dipstick, positive tests for blood and leukocytes would raise the concern for active infection. Examination of the sediment is important in the confirmation of the diagnosis: Red cells, white cells, and bacteria are usually seen under the high-power field. In some instances, white cell casts can be seen in patients with acute pyelonephritis. Other helpful tests include Gram stain of a dried unspun sample of urine. Presence of bacteria in such an examination usually denotes clinically significant urinary tract infection. Usually, more than 1 0 organisms per mL of urine would be found on culture of the urine specimen. URINALYSIS IN THE DETECTION O F BACTERIAL INFECTION IN ASYMPTOMATIC PATIENTS The prevalence of bacteriuria depends on the age and sex of the patient. Among neonates, positive cultures are found in about 1per cent of both males and females. During the school-age years, the prevalence of bacteriuria among boys is less than 0.03 per cent, compared to 1to 2 per cent among girls. Prevalence among females increases by 1per cent per decade of life. Through the child-bearing age, the prevalence is about 3 to 4 per cent, and by age 5 0 the prevalence is as high as 1 0 per cent. Geriatric men have a high prevalence of bacteriuria because of the high incidence of prostatic infection and urologic procedures that can lead to colonization and infection. Geriatric women also have a high prevalence of bacteriuria b e c a ~ s eof large post-void residuals secondary to pelvic floor relaxation. In some series, the prevalence of bacteriuria is as high as 20 per cent in patients above the age of 65. Urinalysis can be used for the detection of asymptomatic bacteriuria. However, it is not clear that routine screening is important for everybody. Although some concerns were originally raised about the relationship between recurrent urinary tract infection with renal failure and h v ~ e r t e n s i o n recent .~~ studies on bacteriuria in both men and womehlfail to demonstrate progressive renal disease in the absence of anatomic abnormalities, hypertension, analgesic abuse, or other systemic conditions predisposing patients to renal insufficiency.', l4 At one time routine screening had been advocated for children because of concern over the effect of bacteriuria on progressive renal damage. However, the currently available information suggests that the renal damage " is rarelv due to infection alone, but rather to infection su~erimposed upon an anatomic abnormality, such as vesicoureteral reflux. Equally important is the finding that the majority of the renal scarring occurs before aee 5. Two recent studies have suggested that if renal scarring has not Gccurred by age 5 , even in the f a c E f continuing bacteriuria and vesicoureteral reflux, renal growth will continue unim~ a i r e d .22~It . is therefore difficult to advocate routine screening. " of children for urinary tract infection. In elderly patients, asymptomatic bacteriuriais quite common. Male patients may have prostatic outlet obstruction and female patients may 139
&
have large post-void residuals in the bladder because of pelvic floor relaxation, predisposing the patient to infection. Prevalence of urinary tract infection may b e as high as 1 0 to 20 per cent in these patients. However, it is not clear that treatment of asymptomatic bacteriuria in these patients would substantially change the natural history of the disease. The original concern of chronic bacteriuria resulting in chronic renal failure, however, has not been borne out in both prospective and retrospective studies. Prospective studies of the natural history of recurrent bacteriuria in over 1000 patients for periods of up to 1 2 years have not demonstrated progressive renal damage in the absence of significant anatomic abnormalities. A retrospective study of 101 consecutive cases of renal failure due to chronic interstitial nephritis seen at a major medical center has likewise failed to reveal a single case in which infection was the primary cause of the renal disease.25These studies suggest that aggressive interventions in asymptomatic patients are probably not warranted. In practice, asymptomatic bacteriuria often becomes symptomatic, and at least one attempt at eradicating the infection is probably indicated if the responsible organism is susceptible to oral antimicrobial agents. If toxic or parenteral agents are required, treatment is probably not indicated. However, in certain groups of patients, urinary tract infection carries considerable morbidity and routine screening would be important. These groups include pregnant females, patients with known structural anatomic abnormalities in the genitourinary tract, and patients with recent instrumentation of the urinary tract. In pregnant women, the prevalence of bacteriuria is at least twice that among similarly aged nonpregnant females, and some 4 to 1 0 per ~, baccent of pregnancies are marked by b a ~ t e r i u r i a .2s~ Asymptomatic teriuria will result in symptomatic urinary tract infections in at least half of the patients who were untreated, and up to 40 per cent will develop full-blown acute pyelonephritis. Bacteriuria has also been associated with an increased rate of fetal loss and low birth weight^.^, 26* 40 Randomized trials of treatment of asymptomatic bacteriuria of pregnancy have demonstrated efficacy in reducing the incidence of pyelonephritis and low birth weight at delivery. Routine screening is therefore important for both maternal and fetal health. Routine screening is also important in patients with known anatomic abnormalities in the urinary tract. Whereas long-term prospective studies have failed to demonstrate progressive renal disease in the absence of anatomic abnormalities, infection in patients with known genitourinary tract abnormalities will cause rapid renal scarring. In patients with spinal cord injury and vesicoureteral reflux, infection resulted in renal scarring and the development of hypertension in 30 per cent ofthe p a t i e n t ~ . ~is~ I t therefore reasonable to screen for infection in patients with known bladder outlet obstruction, vesicoureteral reflux, and nephrolithiasis. Finally, routine screening should also be done in patients who have recently been instrumented. The risk of introducing infection during catheterization of the bladder may be as high as 5 per cent. Prompt therapy in these patients can eliminate the morbidity associated with symptomatic urinary tract infections. 317
570
LESLIE S.T. FANG
URINALYSIS IN THE DIAGNOSIS OF URINARY TRACT INFECTION IN SYMPTOMATIC PATIENTS The most important role of the urinalysis is in the diagnosis of urinary tract infection in patients presenting with symptoms. Seventy per cent of patients presenting with dysuria, frequency, and suprapubic pain will have bacterial infection. The remaining 30 per cent would have insignificant growth on urine cultures and are diagnosed to have acute urethral syndrome. In patients presenting with bacterial infections, the presence of hematuria, pyuria, and bacteriuria would be adequate for the initiation of therapy. Clinically significant bacteriuria would be expected if bacteriuria is detected on an unspun specimen. This is particularly relevant if a Gram stain is performed on an unspun, dried urine specimen. The presence of organism under high-power field examination is highly suggestive of significant bacteriuria. An added advantage of the urinary Gram stain is a preliminary delineation of the organism involved (gram-positive cocci versus gram-negative rods) so as to guide initial therapy.
URINALYSIS IN ACUTE URETHRAL SYNDROME IN ADULT WOMEN Dysuria and urinary frequency are the most common complaints in females with urinary tract infections. However, fully one third of patients with these complaints do not have significant bacteriuria on culture. These patients are diagnosed to have acute urethral syndrome.30, 349 35 The urinalysis is quite helpful in the management of these patients. Patients with acute urethral svndrome can be divided into two maior groups on the basis of the uri&alysis: those with pyuria and those wGhAbout 70 per cent of patients with acute urethral syndrome have pyuria on urinalysis and the remainder do not. The patients with pyuria have true microbial infection due to Chlamydia trachomatis, l-form of gonococci, Staphylococcus saprophyticus, or some other classical uropathogen, but in smaller number than conventionally recognized (less than 10 organisms per mL of urine). Patients without pyuria have no known microbial cause of their symptoms, and the dysuria and frequency may be related to mechanical trauma or i r r i t a t i ~ n . ~ ~
URINALYSIS IN THE LOCALIZATION OF THE SITE OF INFECTION Urinalysis is at times helpful in the localization of the site of urinary tract infection. It is now clear that infection at different sites within the urinary tract has very different pathologic characteristics and different treatment requirement^.'^-'^ Bladder infection is a superficial mucosal infection at a body site where high concentrations of antibiotics can be delivered. Considerable data have accumulated indicating that these
infections can be treated with a single dose of an appropriate a n t i b i o t i ~ . ~ , 19,38 This mode of therapy is less expensive, is accompanied by a high rate of patient compliance, and is associated with fewer In contrast, renal and prostatic infections are deep, side effects2, parenchymal infections requiring a more intensive and prolonged therapy. In patients with documented upper tract infections, conventional therapy with a 14-day course of antibiotics failed in a significant prsportion of ~ a t i e n t s . ' ~ - ' ~ ~ e c a u s the e therapeutic responses are in large part dependent upon the site of the urinary tract infection, numerous invasive and noninvasive ~ r o c e d u r e shave been devised to h e l ~to localize the site of infection (Table 1). Unfortunately, none of the noninvasive tests are extremely sensitive or reliable in the localization of infection. However, two features easily examined by the urinalysis can provide some clues of the site of infection: concentrating ability and the finding of white cell casts on urinalysis. One of the earliest noninvasive tests devised to localize the site of infection utilizes the observation that patients with upper tract infections differ from those with lower tract infections in that they often have an abnormality in renal concentrating ability. The inability to concentrate is most obvious when maximal urinary concentrating ability is examined. In one representative study, Ronald et al. (Am J Med 46:126, 1968) studied the maximal concentrating ability in 38 patients whose site of infection was directlv localized bv ureteral catheterization. Thev demonstrated that renal but not bladdLr bacteriuria was associated with a decreased concentrating ability. Bilateral infection was associated with a greater defect than unilateral infection. With eradication of infection, concentrating ability improves. Even in this study, however, the overlap in maximal urinary osmolarity achieved in patients with bladder, unilateral, and bilateral renal infections is considerable. The test is therefore neither a sensitive nor a consistentlv reliable method of localization. Urinary specific gravity provides an easy means of examining the concentrating ability. Persistingly dilute urine may suggest upper tract infection. However, because many patients would be forcing fluids with the onset of symptoms of urinary tract infection, it would be hard to interpret the significance of the urinary specific gravity in this setting. Urinary concentrating ability is therefore a poor index, at best, in the delineation of the site of infection. In contrast, presence of white cell casts is highly suggestive of an upper tract infection in the appropriate patients. The differential diagnosis ofwhite cell casts is quite limited, incorporating (1) pyelonephritis; Table 1. Invasive and Noninvasive Studies in the Localization of the Site of Urinay Tract Infection INVASIVE STUDIES
NONINVASIVE STUDIES
Bilateral ureteral catheterization Bladder washout studies
Maximal urinary concentrating ability C-reactive protein Urinary enzymes Antibody-coated bacterial assay
572
LESLIE S.T. FANG
( 2 ) acute allergic interstitial nephritis, usually drug induced; and (3) rarely in patients with cholesterol emboli to kidneys. White cell casts are most commonly seen in patients with acute upper tract infections. In the correct clinical scenario, the presence of white cell casts is pathognomonic of upper tract infections. However, the absence of white cell casts does not rule out upper tract infections. In invasive studies that have been done to localize the site of urinary tract infections in females, fully one third of the infections with only lower tract symptoms and urinary findings turned out to have renal bacteriuria.'. 9, 30- 33 Conversely, the majority of patients with clear-cut upper tract symptoms including fever, chills, flank pain, and costovertebral angle tenderness do not have white cell casts in the urine. The presence of white cell casts is therefore a relatively specific but a very insensitive means of localization of site of infection. In summary, urinalysis is a very imprecise way of localizing the site of infection in the urinary tract. However, a careful search for the presence of white cell casts should be made in all instances of urinary tract infection so as not to miss an occult upper tract infection. ANTIBODY-COATING OF BACTERIA ASSAY IN THE LOCALIZATION O F THE SITE O F INFECTION Renal infection is associated with the net synthesis of specific antibody directed against antigens of the infecting organism.16 Cotran,7 using fluorescent antibody techniques, demonstrated gamma-globulincontaining cells in the kidney during the course of experimental pyelonephritis. Lehrman et a1.21 injected E. coli into rabbits with one ureter transiently occluded and studied the local immune response. Using in vitro incubation techniques, they were able to demonstrate protein synthesis in the pyelonephritic kidney, with most of this protein being IgG class globulin, significant portions of which were specific antibodies directed against the infecting strain of organism. Given this information, it is not surprising that immunologic techniques have been applied to the problem of infection localization. Percival et al.,27 using a bacterial agglutination test, found elevated serum antibody levels in patients with symptoms of acute pyelonephritis, with these titers falling in response to antibiotic therapy. Patients with clinically inapparent pyelonephritis also had high antibody levels, whereas those with infection limited to the lower tract had normal titers. Clark et localized the site of infection by ureteral catheterization and examined hemagglutinating antibody response and confirmed that some patients with renal infection had higher hemagglutinating antibody titers than patients with bladder bacteriuria. However, a wide range of titers and a considerable overlap between the two groups were seen, so that the determination of antibody titer was of limited usefulness in the individual patient. Thomas,36 Jones,17 and their colleagues, using an immunofluorescence assay, showed that bacteria originating from the kidney were coated with antibody, whereas bacteria associated with lower urinary
tract infection were antibody-negative. In the Jones study, in particular, there was at least 95 per cent correlation with the results of bladder washout tests. The high degree of sensitivity and reliability of this assay were confirmed by other workers. However, both false-positive and false-negative results may occur. Vaginal23 or rectal contaminating a urine specimen may give false-positive results. Patients with heavy proteinuria may also have false-positive results. In males, prostatitis may also be associated with the antibody coating of the infecting bacteria." False-negative results in adults may occur early in the course of renal infection, during the time necessary to generate the antibody response. In experimental hematogenous pyelonephritis in the rabbit, Smith et al.32showed a time lag of 11 to 15 days between introduction of the organism and the development of a positive assay. Finally, for reasons as yet unclear, the test appears to be unreliable in children. With the realization of these limitations, the test can be areasonable means of localization of the site of infection. URINALYSIS AS A GUIDE TO THERAPY O F URINARY TRACT INFECTIONS In some instances, in addition to the diagnosis of the presence of infection, the urinalysis can actually guide therapy of infection: (1) In adult females with acute urethral syndrome, urinalysis is of great therapeutic importance. Patients with acute urethral syndrome and pyuria usually respond to a 14-day course of doxycycline. Those patients with acute urethral syndrome without pyuria are best treated with analgesics and attention to personal hygiene and sexual practices. (2) In patients with symptoms of urinary tract infection and white cell casts on urinalysis, considerable concern should be raised for the possibility of acute pyelonephritis, and the patients should be considered for a more intensive and prolonged course of antibiotics. In the appropriate patient, urologic and radiologic studies may be necessary. (3) In patients with pyuria and bacteriuria, a Gram stain of the urine can yield some preliminary information about therapy. Gram-positive cocci are likely to be enterococci or staphylococci and should be treated differently than gram-negative bacilli. (4) In patients with pyuria and bacteriuria, an alkaline pH on urinalysis should raise the index of suspicion for an infection with a urea-splitting organism such as Proteus and, less commonly, Pseudomonas and Klebsiella. Concerns should also be raised about the possibility of an infected stone, because the urea-splitting organisms would provide an ideal environment for the formation of struvites. (5) Urinalysis is particularly useful in the follow-up evaluation of patients who have been treated with a course of antibiotics for their urinary tract infection. A benign urinalysis can often obviate the need for a follow-up urine culture. Urinalysis is therefore an easy and important tool in the management of patients with urinary tract infections.
LESLIE S.T. FANG
574 SUMMARY
Urinary tract infection is the commonest human bacterial infection. Bacteriuria alone does not appear to produce progressive renal damage or hypertension. However, it can produce considerable morbidity. Urinalysis is a simple, relatively sensitive, and reliable way of diagnosing urinary tract infection. It is not clear that routine screening should be performed in all patients, but pregnant females, patients with known anatomic abnormalities, and patients with recent genitourinary instrumentation should be screened. The major determinant of therapeutic success in patients with urinary tract infections is the anatomic site of infection. Superficial mucosal infection of the bladder is well treated with a single dose of an appropriate antibiotic, whereas deep tissue infection of the kidney or prostate should be treated with a prolonged and intensive course of therapy. Urinalysis is an insensitive tool in the localization of infection. However, the presence of white cell casts on the examination of the urinary sediment is pathognomonic of upper tract infection and would lead one to pursue an aggressive course of therapy. Examination of the concentrating ability is of limited help in this regard because of the wide range of overlap of concentrating ability in patients with upper and lower tract infections. In selected instances, urinalysis is of help in guiding therapy of urinary tract infections. This is particularly true of the patients with acute urethral syndrome where therapy is guided by the presence or absence of pyuria. Urinalysis, a simple front-line test, is of paramount importance in the evaluation and management of the patient with urinary tract infection. REFERENCES 1. Asscher AW, Chick S, Radfors N, et al: Natural history of asymptomatic bacteriuria in non-pregnant women. In Brumfritt W, Asscher AW (eds): Urinary Tract Infection. London, Oxford University Press, 1973, p 5 1 2. Bailey RR: Single dose therapy of urinary tract infection. Sydney, Australia, ADIS Health Science Press, 1983 3. Braude R, Block C: Proteinuria and antibody-coated bacteria in the urine. N Engl J Med 297:617,1977 4. Brumfritt W: The effects of bacteriuria in pregnancy on maternal and fetal death. Kidney Int (Suppl) 8:S113, 1975 5. Cardiff-Oxford Bacteriuria Study Group: Sequelae of covert bacteriuria in school girls: A four-year follow-up study. Lancet 1:889, 1978 6. Clark H, Ronald AR, Turck M: Serum antibody response in renal versus bladder bacteriuria. J Infect Dis 123:539, 1971 7. Cotran RS: Retrograde Proteus pyelonephritis in rats. Localization of antigen and antibody in treated sterile pyelonephritic kidneys. J Exp Med 117:813, 1963 8. Fairley KF, Bond AG, BrownRB, et al: Simple test to determine the site ofurinary tract infection. Lancet 2:427, 1967 9. Fairley KF, Carson NE, Gutch RC, et al: Site of infection in acute urinary tract infection in general practice. Lancet 2:615, 1971 10. Fang LST, Tolkoff-Rubin NE, Rubin RH: Efficacy of single dose and conventional amoxicillin therapy in urinary tract infection localized by the antibody-coated bacteria technique. N Engl J Med 298:413, 1979
11. Fang LST, Tolkoff-Rubin NE, Rubin RH: Localization and antibiotic management of urinary tract infection. Annu Rev Med 30:225, 1979 12. Fang LST, Tolkoff-Rubin NE, Rubin RH: Urinary tract infections in women. Compr Ther 5:20, 1979 13. Freedman LR, Andriole V: The long-term follow-up of women with urinary tract infection. In Villarreal H (ed): Proc Fifth Int Cong Nephrology. Basel, Karger, 1972, p 230 14. Freedman RB, Smith WM, Richardson JA, et al:.Long-term therapy for chronic bacteriuria in men: U.S. Public Health Service Cooperative Study. Ann Intern Med 83:133, 1975 15. Hellerstein S, Kennedy E, Nussbaum L, Rice K: Localization of the site of urinary tract infections by means of antibody-coated bacteria in the urinary sediments. J Pediatr 92:188. 1978 16. Holmgren J, Smith JW: Immunological aspects of urinary tract infection. Prog Allergy 18:289, 1975 17. Jones SR, Smith JW, Sanford JP: Localization of urinary tract infections by detection of antibody-coated bacteria in urine sediment. N Engl J Med 290:591, 1974 18. Jones SR: Prostatitis as a cause of antibody-coated bacteria in the urine. N Engl J Med 297:365, 1977 19. Kallenius G, Winburg J: Urinary tract infection treated with a single dose of short-acting sulfonamide. Br Med J 1:1175, 1979 20. Kunin CM: Detection, Prevention and Management of Urinary Tract Infection. Philadelphia, Lea & Febiger, 1979 21. Lehrman JD, Smith JW, Miller TE,et al: Local immune response in experimental pyelonephritis. J Clin Invest 47:2541, 1968 22. Lindberg U, Claesson I, Hanson LA, Hodal U: Asymptomatic bacteriuriain schoolgirls. VIII. Clinical course during a %year follow-up. J Pediatr 92:194, 1978 23. Montplaisir S, Cote PA, Martineau B, et al: Localisation du site de l'ingection urinaire chez l'enfant par la recherche de bacteries recouvertes d'anticorps. Can Med Assoc T 115:1096,1676 24. Montplaisir S, Courteau C, Roche AJ: Antibody-coated bacteria in contaminated urine svecimens. N Engl T Med 296:758, 1977 a ~oldb berg M: Chronic interstitial nephritis: Etiologic factors. Ann Intern 25. ~ u i r T, Med 82:453,1975 26. Naeye RL: Cause of the excessive rates of perinatal mortality and prematurity in pregnancies complicated by maternal urinary-tract infections. N Engl J Med 300:819, 1979 27. Percival A, Brumfitt W, DeLouvois J: Serum antibody levels as an indication of clinically inapparent pyelonephritis. Lancet 2:1027, 1964 28. Polk BF: Urinary tract infection in pregnancy. Clin Obstet Gyrlecol 22:285, 1979 29. Ryan RW, Kowalski I, Tilton RC: Mechanisms of the antibody-coated bacteria test (abstract C2). Presented at 77th Annual Meeting Am Soc Microbiol, New Orleans, 1977 30. Sanford JP: Urinary tract symptoms and infections. Annu Rev Med 26:485, 1975 3 1. Sever JL, Ellenberg JH, Edmonds D: Urinary tract infection during pregnancy: Maternal and pediatric findings. In Kass EH, Brumfritt W (eds): Infections of the Urinary Tract. Chicago, University of Chicago Press, 1979, p 1 2 32. Smith JW, Jones SR, Kaijser B: Significance of antibody-coated bacteria in urinary sediment in experimental pyelonephritis. J Infect Dis 135:577, 1977 33. Stamey TA, Govan DE, Palmer JM: The localization and treatment of urinary tract infections: The role ofbactericidal urine levels as opposed to serum levels. Medicine 44:1, 1965 34. Stamm WE, Running K, McKevitt M, et al: Treatment of acute urethral syndrome in women. N Engl J Med 304:956,1981 35. Stamm WE, Wagner KF, Amsel R, et al: Causes of acute urethral syndrome in women. N Engl J Med 303:409,1980 36. Thomas V, Shelokov A, Forland M: Antibody-coated bacteriain the urine and the site of urinary tract infection. N Engl J Med 290:591, 1974 37. Thomas VL, Harris RE, Gilstrap LC 111: Antibody-coated bacteria in the urine of obstetrical patients with acute pyelonephritis. J Infect Dis 131 (Suppl):S57, 1975
576
LESLIE S.T. FANG
38. Tolkoff-Rubin NE, Rubin RH: Urinary tract infection. In Cotran RS (ed):Tubulointerstitial Nephropathies. New York, Churchill Livingstone, 1983, p 49 39. Weiss S, Parker F Jr: Pyelonephritis: Its relation to vascular lesions and to arterial hypertension. Medicine 18:221, 1939 40. Zinner SH: Bacteriuria and babies revisited. N Engl J Med 300:835, 1979 8 Hawthorne Place Suite 104 Boston, Massachusetts 021 14