CLINICAL REVIEW

Lower Genitourinary in Women

Alfred 0. Berg, MD, MPH, and Michael P. Soman, MD, MPH Seattle, Washington

Vaginitis, cystitis, urethritis, and cervicitis are common diagnoses made in women attending family physicians’ offices. Recent research has fundamen­ tally altered available information on the diagnosis and management of these common genitourinary infections. This clinical review discusses presenting symptoms, physical findings, laboratory diagnostic aids, treatment, and follow-up for each lower genitourinary syndrome in women concluding with a summary flow chart illustrating an overall recommended approach.

aginitis, cystitis, urethritis, and cervicitis collec­ toms predicts diagnosis accurately enough so that one V tively account for between 5 and 15 percent of all or more diagnoses can be eliminated from considera­ visits by women to family physicians.1 Women rarely, tion at the outset. Evidence is mixed. Komaroff2 found however, volunteer such diagnoses as complaints: ap­ that differentiating dysuria into external and internal pointments- are made for vaginal discharges and was helpful in establishing diagnoses of vaginitis and malodor, dysuria, urgency, vulvar pruritus, and other cystitis-urethritis, respectively. He further identified symptoms. Translating symptoms into treatable diag­ three clinical groups that would allow the physician to noses is an important part of the physician’s task dur­ focus the remainder of the examination: ing the office visit. Unfortunately, clinical syndromes 1. Vaginal discharge and irritation absent, internal overlap, physical findings may be nonspecific, and dysuria and frequency present. This group had a very microbiological confirmation of diagnoses may be both low probability of vaginitis and high probability of uri­ expensive and untimely. The picture is further compli­ nary tract , thus allowing the elimination of cated by recent challenge to many long-held beliefs pelvic examination and urinalysis, and necessitating about the diagnosis and treatment of these common only urine culture. problems. This article reviews recent diagnostic and 2. Vaginal discharge or irritation present, internal therapeutic advances in managing lower genitourinary dysuria and frequency absent. This group had a low infections in women and outlines some of the unan­ probability of , allowing the swered questions that hamper physicians’ ability to elimination of urinalysis and culture. deal with these problems and that require further in­ 3. Vaginal discharge or irritation present, internal vestigation. dysuria or frequency present. This group had equal probabilities of both vaginitis and urinary infection, thus requiring pelvic examination, urinalysis, and PRESENTING COMPLAINTS urine culture. Generalizability of Komaroff s recommendations to Symptoms are the ticket for admission to an office visit other primary care settings is hampered by patient and largely determine the physician’s initial approach. selection and by the lack of uniform database on all The list of potential complaints for lower genitourinary patients seen. A more recent study performed in infections is long: dysuria, vaginal discharge or itch­ Seattle3 found that up to one third of women could not ing, urgency, frequency, incontinence, dyspareunia, distinguish between internal and external dysuria or hematuria, and many others. The immediate question claimed to have both. In the same study, although for the physician is whether any combination of symp- Komaroff s clinical strategies were generally helpful, some women were misclassified by these criteria. In summary, presenting complaints are suggestive Submitted, revised, December 7, 1984. but not diagnostic of specific groups of infections. horn the Department of Family Medicine, University of Washington, Further research in the family practice setting will be Seattle, Washington. Requests for reprints should be addressed to Dr. necessary to accurately assess predictive value of Alfred Berg, Department of Family Medicine, University of Washington, 110-30, Seattle, WA 98195. symptoms for specific diagnoses.

c 1986 Appieton-Century-Crofts

THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1: 61-67, 1986 61 LOWER GU INFECTIONS IN WOMEN

VAGINAL INFECTIONS aration. Additional use of cultures for Candida albi­ cans has been recommended by some.8 Demonstration Vaginitis is the most common lower genitourinary clin­ of Candida by culture is the gold standard, a standard ical syndrome seen by the primary care practitioner, compared with which the potassium hydroxide prep­ accounting for two thirds of all visits for genitourinary aration performs poorly.9 Additionally, it is well problems. Recent studies have fundamentally changed known that asymptomatic women may harbor yeast the diagnosis and treatment of nonspecific vaginitis organisms.8 Thus, the physician is left with a dilemma: and have suggested new ways of evaluating women the potassium hydroxide preparation is insensitive in with yeast and trichomonal infections as well. identifying yeast in symptomatic women, yet use of Vaginal discharge and vulvar irritation are the usual the culture technique routinely would undoubtedly presenting complaints of vaginal infection. Certain falsely identify some carriers of Candida as infected. symptoms are associated with particular infections— An answer to this dilemma awaits further research pruritus with yeast, purulent and copious discharge testing the efficacy of treatment in symptomatic with trichomonas, and foul-smelling discharge with women who are potassium hydroxide negative but cul­ nonspecific vaginitis. Such symptoms, though charac­ ture positive. In the interim, some have suggested that teristic of given infections once the diagnosis is estab­ cultures for Candida be used selectively in symptoma­ lished, may have poor predictive value during the ini­ tic women with negative potassium hydroxide prep­ tial evaluation phase. Thus, of all women with vaginal arations.9 discharge and vulvar pruritus, only one half may be The diagnosis of vaginitis caused by Trichomonas found to have yeast on examination. Recent attempts vaginalis has problems similar to those in making the to correlate groups of symptoms with specific diag­ diagnosis of Candida. Traditionally, the diagnosis of noses have been only marginally successful. trichomonal infection has been based upon direct mi­ On physical examination, vaginal infections are croscopic observation of the organism in a bit of the associated with specific findings. Thick, curdlike dis­ discharge mixed with saline. Culture methods are charge and external vulvitis are typical of yeast vag­ available that detect the organism in a larger propor­ initis; copious purulent discharge with cervical tion of women, however.1011 Because women may petechiae and friability are characteristic of tricho­ have asymptomatic Trichomonas infections, the situa­ monal infections; and a thin, adherent, foul-smelling tion is similar to that confronting the physician at­ discharge is the hallmark of nonspecific vaginitis. tempting to make the diagnosis of vaginitis caused by Again, however, the predictive value of these charac­ yeast. Again, trichomonal cultures may be used selec­ teristic physical findings for the three diagnoses is un­ tively in symptomatic women with negative micro­ known, and a recent Seattle study suggests that addi­ scopic examinations. tional factors need to be explored.3'4 Treatment protocols for vaginal infections are fairly Significant changes in laboratory aids to the diag­ straightforward. Yeast infections are effectively nosis of vaginal infections have been recently pro­ treated using or creams containing the posed, especially in making the diagnosis of polyene (nystatin) or imidazole (clotrimazole, mi­ nonspecific vaginitis. Formerly a diagnosis of exclu­ conazole) . Oral ketoconazole for resis­ sion, nonspecific vaginitis may now be diagnosed tant infections is currently under study. Trichomonal using Amsel’s5 recommendation that at least three of vaginitis may be effectively treated using single-dose the following four criteria be present: (1) thin, adher­ metronidazole in a 2-g dose. The treatment of bacterial ent, homogenous discharge, (2) vaginal pH of 4.5 or vaginosis formerly included several regimens, includ­ higher, (3) amine odor upon application of potassium ing topical sulfa creams and oral or hydroxide to discharge, or (4) clue cells on micro­ amoxicillin. It is now clear that metronidazole is more scopic examination of saline . Recent mi- effective, although exact dosage and duration of crobiologic investigation has strongly implied that the treatment are still being studied.12 Currently recom­ Gardnerella vaginalis organism (formerly Hemophilus mended is 500 mg of metronidazole twice daily for vaginalis) may play only a partial role in a mixed infec­ seven days, longer if symptoms persist (metronidazole tion of several bacteria collectively responsible for is not approved for this application by the Federal symptoms.6 Finally, at a recent international sym­ Drug Administration).13 The necessity of treatment of posium, nonspecific vaginitis was renamed to be sexual partners for these infections is still being bacterial vaginosis, reflecting the mixed bacterial studied. Most yeast infections do not require treatment noninvasive nature of this condition.7 of the sexual partner, although men with yeast The diagnosis of yeast vaginitis has been based upon balanitis may occasionally serve as a reservoir for rein­ demonstration of budding yeast and pseudohyphae fection and should be treated accordingly. Most au­ after application of potassium hydroxide to a bit of the thorities agree that trichomonas infections should be discharge and microscopic examination of the prep­ treated with a similar single 2-g dose of metronidazole

62 THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1,1986 lower g u in f e c t io n s in w o m e n

in the partner. Virtually no data are available to sup­ per milliliter identifies only one half of symptomatic port treatment of sexual partners of women who have women with proven coliform infection of the bladder. nonspecific vaginitis, although several studies are cur­ Among symptomatic women, the criterion of at least rently being conducted. 102 coliforms per milliliter of clean voided urine has a high sensitivity (.95), specificity (.85), and predictive value (.88) and is preferable to previous criteria.17 In CYSTITIS AND URETHRITIS addition, in many symptomatic women "mixed” cul­ ture results (in which one of the isolates is a coliform) CLINICAL SPECTRUM have been shown to correlate with true bacterial infec­ Urinary tract infections affect at least 10 to 20 percent tion of the bladder. With “ mixed” or lower level of women in the general population during their life­ (fewer than 105/mL) culture results, the presence of times.14 Though there is some overlap, the clinical pyuria increases the likelihood of true infection. Al­ spectrum of urinary tract infection includes three pri­ though a reasonable clinical definition of pyuria is at mary syndromes. The first is asymptomatic , least 10 leukocytes per high-power field of clean, void­ in which patients have no symptoms but cultures of urine ed, unspun urine, the hemacytometer determination reveal pathogenic organisms. The second syndrome, of pyuria has been shown to be simple and highly re­ lower tract infection, is characterized by various com­ producible and may have a greater role in the future.18 binations of frequency, nocturia, urgency, suprapubic The acute has traditionally been discomfort and tenderness, and sometimes hematuria. defined as acute dysuria and frequency in women This syndrome can be further divided into bacterial whose clean voided urine has fewer than 105 microor­ infection of the bladder with or without “ silent” renal ganisms per milliliter. More recently,19 this syndrome involvement and acute urethral syndrome. Both of has been categorized into three subgroups. The first these subdivisions have undergone recent redefinition. group is characterized by bladder bacteriuria with less The third clinical syndrome is acute , than 103 bacteria per milliliter, usually accompanied by which often presents with costovertebral angle pain pyuria. This differs from acute cystitis only in and tenderness, fever, rigors, and sometimes with ac­ terms of the quantitative bacterial count and in most companying signs and symptoms of cystitis. As this populations constitutes the largest category of the review is limited to lower genitourinary syndromes, acute urethral syndrome. The second group is charac­ acute pyelonephritis will not be discussed further. terized by pyuria and no bacteriuria on culture. The following problems are included in the clinical Though several studies of young sexually active categorization of urinary tract infection: (1) History and women have documented a high rate of chlamydia in­ physical examination do not reliably distinguish be­ fection,20 a prospective study of women with tween the acute urethral syndrome and other forms of genitourinary symptoms presenting to a university- urinary tract infection; (2) it is not possible to distin­ based family practice did not find a high rate of guish between upper tract infection and lower tract chlamydia infection in women with pyuria and no bac­ infection on clinical grounds alone14,15; and (3) up to 50 teriuria.3 Neisseria gonorrhoeae should be considered percent of all patients with bacteriuria have been when evaluating these patients. The third group, with­ shown to have silent renal involvement by localization out bacteriuria or pyuria, usually has no identifiable techniques.16 Because of these problems, the labora­ etiologic agent. tory has taken on a greater role in the evaluation of urinary tract infections. UPPER VS LOWER TRACT INFECTION The differentiation of upper from lower tract infection LABORATORY FINDINGS can be accomplished accurately using bladder washout Although the presence in an unspun sample of clean or ureteral catheterization, but these are expensive voided urine of leukocytes, erythrocytes, or bacteria is and time-consuming procedures. A more recent tech­ consistent with urinary tract infection, the absence of nique, testing for -coated bacteria, has signifi­ any or all does not preclude the diagnosis. Leukocyte cant numbers of false-positive and false-negative re­ casts, if present, confirm the presence of renal infec­ sults and should be reserved for research.18'21 At pres­ tion. ent, the most practical guide to anatomic site of infec­ In uncomplicated urinary tract infections aerobic tion may be the response to single-dose therapy (dis­ gram-negative coliforms are the most common cussed later).22 In addition, the failure to seek medical etiologic agents, with Escherichia coli being responsi­ attention within six days of the onset of symptoms or ble for 80 to 90 percent of cases. The definition of findings consistent with upper tract or “ complicated” significant bacteriuria is in the process of radical re­ infection increases the likelihood of upper tract infec­ vision. The traditional criterion of at least 105 bacteria tion.

THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO 1, 1986 63 LOWER GU INFECTIONS IN WOMEN

MANAGEMENT women treated with single-dose TMP-SMZ had a Many regimens will “ cure” the majority of higher relapse rate than those treated with a multiple- acute, uncomplicated urinary tract infections, and, in dose regimen.31 Clinically stable patients who fail fact, the high urinary concentration of many single-dose therapy, as evidenced by a positive urine antibiotics results in eradication of many organisms culture two to three days after treatment, should re­ with in vitro drug resistance. In discussing cure, it ceive a 10- to 14-day course of a standard antibiotic must be remembered that placebo treatment has re­ based on susceptibility testing as described below. If sulted in sterile urine obtained within five months in 80 this therapy fails, or if a woman has signs or symptoms percent of nonpregnant women with acutely symp­ of upper tract infection after failed single-dose tomatic urinary tract infections.23 Another important therapy, a four- to six-week course of antibiotics is aspect of cure is that because of the high rate of recur­ indicated. In addition, urologic and radiologic investi­ rence and because of occasional persistence of bac­ gations of the urinary tract should be considered. teria in the urinary tract, follow-up urine culture In acute, uncomplicated urinary tract infection with should be obtained several days to a week after the any of the stated contraindications for single-dose completion of treatment. Failure to respond is gener­ therapy, a seven- to ten-day course of antibiotics is ally due to patient noncompliance, reinfection, or indicated. Choice of antibiotic should be governed by more rarely to bacterial persistence. In all aspects of urine culture and sensitivity reports, but pending re­ management, health care providers must be aware that sults, empiric therapy based on symptoms and women at risk for severe morbidity or renal scarring urinalysis may be started with sulfisoxazole, ampicil- from urinary tract infections include those with (1) in­ lin, tetracycline, nitrofurantoin, TMP-SMZ, trimeth­ fection with urea-splitting organisms (usually Proteus) oprim alone, cephalexin, or various other agents. that cause struvite renal stones, (2) congenital In patients with early recurrence of urinary tract in­ anomalies of the genitourinary system that become fection, many previous episodes that have been dif­ secondarily infected, (3) bacteriuria in the presence of ficult to control, underlying anatomic abnormality, or obstruction of the urinary tract (such as by neph­ history of renal transplant, a four- to six-week course rolithiasis), (4) analgesic nephropathy, (5) diabetes, of antibiotics is indicated. Although many authors rec­ (6) neurogenic bladder, and (7) pregnancy. ommend a similar regimen for diabetics, an initial In general, asymptomatic bacteriuria in adult seven- to ten-day course is justified for acute, uncom­ women requires treatment only during pregnancy. plicated infection. The acute urethral syndrome is best treated by Therapy of pyelonephritis and catheter-associated single-dose therapy with amoxicillin or trimethoprim- urinary tract infection is beyond the scope of this re­ sulfamethoxazole (TMP-SMZ), as described below, view. when coliforms are the etiologic agent. If contraindi­ Recent studies have questioned the traditional indi­ cations exist or if a coliform infection does not respond cations for investigation of the urinary tract in women to single-dose therapy, a 7- to 14-day course of with urinary tract infections. Procedures such as intra­ antibiotic therapy is indicated. When chlamydia is the venous pyelography, cystoscopy, and cystography have suspected pathogen, therapy should consist of a ten- a very low yield and are cost ineffective for adult women day course of doxycycline, 100 mg orally twice daily, with uncomplicated lower urinary tract infections.32 or tetracycline, 500 mg orally four times daily. Treat­ These studies appear most useful in women with ment of sexual partners should be considered. In possible nephrolithiasis, history of multiple childhood symptomatic women without pyuria, antibiotic infections (if not already evaluated), relapsing infec­ therapy is of no proven benefit.24 tion when repeated lack of response to an effective A major recent advance in the therapy of acute, un­ agent has been documented; in diabetics with relapses; complicated urinary tract infections in adult women or in women with pyelonephritis that does not respond has been the discovery that most such infections can quickly to treatment, neurogenic bladder, or painless be eradicated with single-dose oral therapy with hematuria. amoxicillin (3 g), TMP-SMZ (two double-strength tab­ lets), or sulfisoxazole (2 g). 18,22,25-30 T h j s therapy is in­ PREVENTION appropriate for diabetics, pregnant women, or women who seem unlikely to return for follow-up culture, Prevention of urinary tract infections is cost effective have had symptoms for over six days, have signs or and efficacious in women who have three or more symptoms suggesting upper tract infection, or have a episodes yearly. Effective regimens include daily ad­ history of multiple urinary tract infections. However, ministration of TMP-SMZ (one-half ) or nitrofu­ an important cautionary note has been struck by the rantoin (100 mg) or trimethoprim (100 mg). Thrice recent publication of the largest prospective series to weekly and postintercourse dosing have also been ef­ date comparing single- and multiple-dose therapy: fective, but as with the daily regimens, the beneficial

64 THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1, 1986 LOWER g u in f e c t io n s in w o m e n

effect of the drugs stops when prophylaxis is discon­ many communities. Other techniques, such as Gram tinued. An initial six-month trial to establish tolerance stains, Tzanck preparations, and Papanicolaou and effectiveness should be followed by individualiza­ smears, are not sufficiently sensitive or specific for tion of treatment determined by baseline rate of infec­ routine clinical use. In the case of herpes infections, tion. Periodic urine cultures during periods of even though cultures may be unavailable, typical ap­ prophylaxis are needed to confirm efficacy.33,3,4 pearance of the lesions with a good exposure history may be sufficient to establish the diagnosis. Develop­ PREGNANT WOMEN ment of monoclonal antibody techniques holds the greatest promise for quick and inexpensive diagnosis Because of hormone-induced dilatation of the renal of most cervical infections.40,41 , calyces, and , there; is a high rate of Treatment protocols for gonorrhea infection vary urinary tract infection in pregnancy. Asymptomatic depending upon region of the country (because of the bacteriuria occurs in 4 to 7 percent of pregnant prevalence of -resistant strains), severity in­ women, and up to 40 percent of these episodes, if un­ cluding presence of systemic symptoms, and individ­ treated, may go on to pyelonephritis.35 Pyelonephritis ual drug sensitivity. Recommended regimens use in pregnancy has been reported to cause many compli­ amoxicillin, ampicillin, tetracycline, doxycycline, cations, including septic shock36 and fetal death.35 erythromycin, spectinomycin, ceftriaxone, and tri­ Thus screening all pregnant women for asymptomatic methoprim-sulfamethoxazole. Tetracycline, doxycy­ bacteriuria and treating all episodes detected with an cline, and erythromycin are effective for the treatment initial 10- to 14-day course of antibiotics are essential. of Chlamydia trachomatis, but treatment may need to Frequent follow-up cultures should be obtained, as the be extended beyond seven days, especially if symp­ relapse rate is high. If relapse occurs, a four- to six- toms and signs are persistent. Symptoms and signs week antibiotic course is indicated. Tetracyclines are accompanying the first clinical episode of genital in­ contraindicated in pregnancy. Sulfa preparations are fection by herpesvirus may be reduced by treatment not contraindicated in the first and second trimesters with oral acyclovir, but the use of acyclovir for recur­ but when used in late pregnancy may contribute to rent infections is less effective. Readers are encouraged neonatal kernicterus.37 Drugs in the penicillin family to keep up to date with the current treatment recom­ and the cephalosporin family have no known con­ mendations for all of these infections by reading the traindications specifically related to pregnancy.37,3S annually published summary from the Centers for Dis­ ease Control.42 CERVICITIS

Cervicitis may present with vaginal symptoms or may RECOMMENDED CLINICAL STRATEGY be incidentally found on pelvic examination performed for other reasons. Purulent cervical discharge, friabil­ Figure 1 summarizes the clinical implications of this ity, and cervical lesions, such as petechiae and vesi­ review by presenting a diagnostic strategy for women cles, are common in cervical infections, but are not visiting primary care physicians for the initial evalua­ sufficiently characteristic to allow specific diagnosis tion of lower genitourinary symptoms. Based upon the without further evaluation. Depending upon the popu­ clinical history, including exposure to sexually trans­ lation seen, gonorrhea may be present in from less mitted disease, nature of dysuria, vaginal discharge, than 1 percent to more than 50 percent of women, with and past episodes, women are initially divided into one lower prevalences most common in primary care set­ of three groups for further evaluation: (1) probability tings. Chlamydia trachomatis may be present in from 4 of vaginitis high, (2) probability of urinary infection to 10 percent of women, again depending upon charac­ high, or (3) inability to separate into groups 1 or 2. teristics of the population seen.3,20,39 Herpes simplex Evaluation for vaginitis includes a normal saline prep­ has high prevalence in many sexually transmitted dis­ aration, a potassium hydroxide preparation, and ease clinics and tertiary care settings but may be rarely assessment of criteria for bacterial vaginosis (vaginal seen in some established family practices.3 Tricho- pH, appearance of discharge, amine odor, clue cells), monal infections typically cause vaginal symptoms but and may include evaluation for sexually transmissible may cause isolated cervicitis. Unfortunately, the diag­ disease. The evaluation, if initially nondiagnostic, may nosis of all cervical infections is dependent upon cul­ be followed by further cultures for Candida, ture techniques for N gonorrhoeae, Chlamydia trichomonas, and others. Evaluation for urinary tract trachomatis, and herpesvirus. Although cultures for infection includes urinalysis and culture and evalua­ gonorrhea are widely available, those for chlamydia tion for sexually transmissible disease, if appropriate. and herpes are just beginning to penetrate hospital The workup for sexually transmissible disease in both microbiology laboratories and may be unavailable in groups should include cultures for gonorrhea and

THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1, 1986 65 LOWER GU INFECTIONS IN WOMEN

PATIENT PRESENTS WITH LOWER GENITOURINARY SYMPTOMS I Take appropriate history, including STD exposure, nature of dysuria, vaginal discharge, and past episodes of genitourinary infection

r------1 Probability of vaginitis felt to Unable to separate Probability of UTI felt to be high be high due to any of the following: into high-probability due to any of the following: 1. History of previous episodes vaginitis 1. History of previous episodes, or 2. Komaroff criteria, or or UTI group 2. Komaroff criteria 3. STD exposure

Perform pelvic examination with: ------Evaluate for both ------Obtain routine urinalysis 1. Normal saline and KOH preparations vaginitis and UTI and urine culture 2. BV criteria *3. STD evaluation ______|______Initial presumptive therapy if ______1 I UA has WBCs and bacteria Diagnosis made Diagnosis uncertain ______

Organisms at least 105/mL Fewer organisms than 105/mL Treat Other cultures (classical UTI) (acute urethral syndrome) (Candida, trichomonas) Treat No growth (sterile pyuria) 1. Reculture Organisms are more than 102/mL but *2. STD evaluation fewer than 105/mL 3. Empirical therapy (after STD evaluation)

Coliforms No coliforms 1. Reculture *2. STD evaluation Treat 3. Empirical therapy

'STD evaluation (gonorrhea and chlamydia cultures, herpes optional) if any of the following: (1)STD exposure with any genitourinary symptoms; (2) purulent cervicitis; (3) acute urethral syndrome if one of the following exist: (a) sterile pyuria, (b) STD exposure, (c) high-prevalence population for STD Figure 1. Diagnostic strategy for patients with lower genitourinary symptoms. STD, sexually transmissible disease; UTI, urinary tract infection; BV, bacterial vaginosis; UA, urinalyses; WBC, white blood cell

chlamydia and possibly herpes. The group of women criteria for the diagnosis of bacterial vaginosis, by who are not initially categorized into high-probability selective use of cultures for Candida and trichomonas, vaginitis or urinary tract infection group should be by employing different criteria in the diagnosis of uri­ evaluated for both. nary infection, and by recommending the use of Several other points should be emphasized. This gonorrhea and chlamydia cultures in women at risk. proposed clinical strategy differs from those previ­ Further recommended is that physicians consider the ously published by recommending more rigorous use of empirical therapy in the management of urinary

66 THE JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1, 1986 lower g u in f e c t io n s in w o m e n

tract infection when diagnostic tests (including evalu­ 19. Stamm WE, Wagner KF, Amsel R, et al: Causes of the acute ation for sexually transmissible disease) have failed to urethral syndrome in women. N Engl J Med 1980; 303:409- 415 reach closure. 20. Pauja S: Urethral syndrome in women attending a clinic for sexually transmitted . Br J Vener Dis 1983; 59:179- 181 21. Turck M: New concepts in genitourinary tract infections. JAMA 1981; 246:2019-2023 References 22. Rubin RH, Fang LST, Jones SR, et al: Single-dose amoxicil­ lin therapy for urinary tract infection. JAMA 1980; 244:561- 1. Rosenblatt RA, Cherkin DC, Schneeweiss R, et al: The struc­ 564 ture and content of family practice: Current status and fu­ 23. Mabeck CE: Treatment of uncomplicated urinary tract in­ ture trends. J Fam Pract 1982; 15:681-722 fection in non-pregnant women. Postgrad Med J 1972; 2. Komaroff AL, Pass TM, McCue JD, et al: Management strat­ 48:69-75 egies for urinary and vaginal infections. Arch Intern Med 24. Stamm WE, Running K, McKevitt M, et al: Treatment of the 1978; 138:1069-1073 acute urethral syndrome. N Engl J Med 1981; 304:956-958 3. Berg AO, Heidrich FE, Fihn SD, et al: Establishing the etiol­ 25. Anderson JD, Aird MY, Johnson AM, et al: The use of a ogy of genitourinary symptoms in women in a family prac­ single large dose of amoxycillin for the treatment of acute tice. JAMA 1984; 251:620-625 urinary tract infections. J Antimicrob Chemother 1979; 4. Bergman JJ, Berg AO: How useful are symptoms in the 5:481-483 diagnosis of Candida vaginitis? J Fam Pract 1983; 16:509- 26. Bailey RR, Abbott GD: Treatment of urinary tract infection 511 with a single dose of amoxycillin. Nephron 1977; 18(6):316- 5. Amsel R, Totten PA, Spiegel CA, et al: Nonspecific vaginitis: 320 Diagnostic criteria and microbial and epidemiological 27. Bailey RR, Abbott GD: Treatment of urinary tract infection associations. Am J Med 1983; 74:14-23 with a single dose of trimethoprim-sulfamethoxazole. Can 6. Spiegel CA, Amsel R, Eschenbach DA, et al: Anaerobic bac­ Med Assoc J 1978; 118:551-552 teria in nonspecific vaginitis. N Engl J Med 1980; 303:601- 28. Buckwold FJ, Ludwig P, Harding GK, et al: Therapy for 607 acute cystitis in adult women. JAMA 1982; 247:1839-1842 7. Mardh PA, Taylor-Robinson D (eds): Bacterial Vaginosis. 29. Counts GW, Stamm WE, McKevitt M, et al: Treatment of Stockholm, Almquist & Wikseli International, 1984 cystitis in women with a single dose of trimethoprim- 8. Odds FC: Candida and Candidosis. Baltimore, University sulfamethoxazole. Rev Infect Dis 1982; 4:484-490 Park Press, 1979 30. Stamm WE: Single-dose treatment of cystitis. JAMA 1980; 9. Bergman JJ, Berg AO, Heidrich FE, Schneeweiss RS: A clin­ 244:591-592 ical comparison of microscopic and culture techniques in 31. Schultz HJ, McCaffrey LA, Keys TF, Nobrega FT: Acute cys­ the diagnosis of Candida vaginitis. J Fam Pract 1984; titis: A prospective study of laboratory tests and duration of 18:549-552 therapy. Mayo Clin Proc 1984; 59:391-397 10. McLennan MT, Smith JM, McLennan CE: Diagnosis of vagi­ 32. Fowler JE, Pulaski ET: Excretory urography, cystography, nal mycosis and trichomoniasis. Obstet Gynecol 1972; and cystoscopy in the evaluation of women with urinary- 40:231-234 tract infection. N Engl J Med 1981; 304:462-465 11. McCann JS: Comparison of direct microscopy and culture 33. Stamm WE, Counts GW, Wagner KF, et al: Antimicrobial in the diagnosis of trichomoniasis. Br J Vener Dis 1974; prophylaxis of recurrent urinary tract infections. Ann Intern 50:450-452 Med 1980; 92:770-775 12. Eschenbach DA, Critchlow CW, Watkins H, et al: A dose- 34. Kunin CM: Duration of treatment of urinary tract infections. duration study of metronidazole for the treatment of Am J Med 1981; 71:849-854 nonspecific vaginosis. Scand J Infect Dis (suppl) 1983; 35. Davison JM, Lindheimer MD: Renal disease in pregnant 40:73-80 women. Clin Obstet Gynecol 1978; 21:411-427 13. Pheifer TA, Forsyth PS, Durfee MA, et al: Nonspecific vag­ 36. Cunningham FG, Morris GB, Mickal A: Acute pyelonephritis initis: Role of Hemophilus vaginalis and treatment with met­ of pregnancy: A clinical review. Obstet Gynecol 1973; ronidazole. N Engl J Med 1978; 298:1429-1434 41:112-117 14. Rubin RH: Infections of the urinary tract. In Scientific Amer­ 37. Schwarz RH: Considerations of antibiotic therapy during ican Medicine. New York, Scientific American, 1982; vol 2, pregnancy. Obstet Gynecol 1981; 58(suppl):95-99 sec 7, chapt 23 38. Howard FM, Hill JM: Drugs in pregnancy. Obstet Gynecol 15. Fairley KF, Carson NE, Gutch RC, et al: Site of infection in Surv 1979; 34:643-653 acute urinary-tract infection in general practice. Lancet 39. Sorbie J, O’Shaughnessy MU: Chlamydia trachomatis in­ 1971; 2:615-618 fections in women with urogenital symptoms. Can Med 16. Stamey T: Urinary tract infections in the female: A perspec­ Assoc J 1982; 127:974-976 tive. In Remington J, Swartz M (eds): Current Clinical Topics 40. Nowinski RC, Tam MR, Goldstein LC, et al: Monoclonal in Infectious Diseases, No. 2. New York, McGraw-Hill, 1981, for diagnosis of infectious diseases in humans. PP 31-53 Science 1983; 219:637-644 17. Stamm WE, Counts GW, Running KR, et al: Diagnosis of 41. Stamm WE, Harrison HR, Alexander ER, Cles LD, et al: Di­ coliform infection in acutely dysuric women. N Engl J Med agnosis of Chlamydia trachomatis infections by direct im­ 1982; 307:463-468 munofluorescence staining of genital secretions. Ann Intern 18. Stamm WE: Recent developments in the diagnosis and Med 1984; 101:638-641 treatment of urinary tract infections. West J Med 1982; 42. 1985 STD Treatment Guidelines. MMWR 1985; 34(4S):75S- 137:213-220 108S

the JOURNAL OF FAMILY PRACTICE, VOL. 23, NO. 1, 1986 67