|Sexually Transmitted Diseases (STD) Syndromic Management (AIDSCAP/FHI, 1997, 54 p.)|
|DESIGN OF STD FLOWCHARTS|
Described below are examples of syndromic case management protocols developed by a variety of countries in Latin America and the Caribbean.16 The original WHO flowcharts were developed for six syndromes: urethral discharge, vaginal discharge, pelvic inflammatory disease, genital ulcer disease, swollen scrotum, and neonatal conjunctivitis. The first four are discussed below.
Gonorrhea is the main cause of urethritis among clinic attendees in most developing countries. In recent years, however, as diagnostic techniques for chlamydia have become more sensitive, the role of chlamydial and mixed infections in causing urethritis in developing countries is also becoming better defined. Some clinicians rely on the characteristics of urethral discharge to differentiate between gonococcal and non-gonococcal urethritis (NGU). Gonococcal urethritis tends to be more purulent and NGU more mucoid. However, these clinical signs are not sufficiently discriminatory to predict the etiology or cause of urethral discharge in a given patient.17 In addition, they can be confounded by prior, ineffective treatments patients may have taken before coming to the clinic.
Two examples of flowcharts for urethral discharge are shown in Figures 2 and 3. The first example (Figure 2) is a simple syndromic management, treating every man with a complaint of urethral discharge for gonorrhea and NGU. A sequential treatment (first, treatment for gonorrhea and if this fails, treatment for NGU) has been the policy in the past in some countries in order to limit unnecessary treatments. However, because of a large proportion of missed chlamydial infections, and because many patients fail to come back, this approach can no longer be recommended.
Figure 2 - Management of Urethral Discharge - HONDURAS
In the second flowchart (Figure 3), Gram stain is added to a syndromic approach. Depending on the result of the Gram stain, a syndromic treatment or a treatment for NGU will be given. This approach offers the advantage of reducing unnecessary treatments (including expensive gonorrhea drugs) for the patient and his partner by increasing the specificity of the flowchart.
Figure 3 - Management of Urethral Discharge - JAMAICA
A flowchart including Gram stain can only be considered when laboratory facilities are available. Results should be given within a reasonable time so patients do not have to return to the health facility for treatment the next day. This approach reduces the risk of serious complications, acute morbidity associated with either gonorrhea or chlamydia, and further transmission of the causative organism.
The symptoms of cervicitis and vaginitis overlap. Abnormal (in amount, color or odor) vaginal discharge is the symptom most commonly presented, but it is more predictive for vaginitis than for cervicitis.18,19 The sensitivity of the symptom vaginal discharge for cervicitis varies from 25 percent (prostitutes in Zaire) to 48 percent (STD patients in USA). Cervical mucopus and induced endocervical bleeding have a high specificity (83 to 99 percent) but a low sensitivity (1 to 43 percent) as clinical signs for cervicitis. Examples of flowcharts for vaginal discharge are shown in Figures 4 and 5.
Figure 4 is a flowchart for situations in which a speculum examination is not possible. The most probable cause of a woman complaining of vaginal discharge is vaginitis. Cervicitis is a less frequent cause of consultation for vaginal discharge, but the complications of untreated cervicitis are much more serious.
Figure 4 - Example of a Flowchart for the Management of Vaginal Discharge - HAITI (without speculum)
The accuracy and cost-effectiveness of syndromic diagnosis of vaginitis can be improved significantly in some settings by adding a risk assessment component to the case management protocols (for instance, determining whether an individual has had a new sexual partner or more than one sexual partner in the past three months). Using this approach, a woman with vaginal discharge and positive risk assessment for STDs would be treated for gonorrhea and chlamydia cervicitis as well as for vaginitis; a woman with no risk factors for STDs would be treated only for vaginitis, which requires a much less expensive treatment regimen. A recent analysis of data from pregnant women and sex workers in Zaire suggested that a simple case management protocol based on reported vaginal discharge and a risk assessment could be a useful tool for symptomatic women at high and low risk for STDs.20
In situations where a speculum examination is possible, the clinician can try to differentiate between various etiologies of vaginal discharge. The clinical sign mucopus, however, is not sensitive enough to be the only indication for cervicitis treatment. Figure 5 is an example of a flowchart utilizing a speculum exam and a risk assessment.
Figure 5 - Example of a Flowchart for the Management of Vaginal Discharge - JAMAICA (without speculum)
An alternative for differentiating etiologies for vaginitis can be offered by simple laboratory tests, if the infrastructure is available. Direct examination of a vaginal wet mount is useful for detecting trichomonads and yeast forms. Determination of the vaginal pH and amine odor with 10 percent potassium hydroxide solution can be helpful in the diagnosis of bacterial vaginosis. However, no simple laboratory test has been developed so far for detecting cervicitis. Adding Gram stain for the detection of intracellular gram-negative diplococci or leukocytes in the endocervix does not offer any advantage, as the sensitivity will drop dramatically. The leukocyte esterase dipstick, which has a good sensitivity for detecting male urethritis, had a sensitivity of only 47 percent for the detection of cervicitis.15
Pelvic inflammatory disease (PID) is a common complication of untreated gonococcal and/or chlamydial cervicitis and results in tubal scarring and occlusion. This can lead to ectopic pregnancy a serious, possibly life-threatening complication. Most infertility problems in the developing world are attributed to prior upper genital tract infections.21
An example of a clinical flowchart for detecting PID is shown in Figure 6. Because of the serious complications of PID, the flowchart should start with a very sensitive symptom. Lower abdominal pain is more sensitive for PID than fever. It is important that surgical and obstetrical emergencies, such as peritonitis and extra-uterine pregnancy, are immediately referred.
Figure - Example of a Flowchart for the Management of Abdominal Pain - DOMINICAN REPUBLIC
Many studies have tried to describe a "typical" clinical picture for the different etiological diagnoses of genital ulcer disease (GUD) but have failed. Descriptions, such as regular shape, smooth base, undermined edge, friability, tenderness and purulence, are not sufficiently discriminatory (even for experienced clinicians) to make an etiological diagnosis in most cases. In a study in South Africa of 210 patients with genital ulcers, clinical diagnosis was compared with a gold standard laboratory test. Clinical diagnosis had a positive predictive value of 89 percent for chancroid, 47 percent for syphilis, and 19 percent for genital herpes.13 Dual infections were common, making an etiological diagnosis even more difficult. Without sophisticated laboratory tests, an etiological diagnosis of GUD is impossible.
The relative frequencies of the different causes of GUD vary between geographical areas but can also vary in time. For example, two studies on the etiologies of GUD, in Rwanda in 1986 and 1992, found there was a shift in the relative frequencies of different etiologies. As the prevalence of HIV infection increased, herpes became more important as an etiology of GUD.22
In many developing countries, the etiologies of GUD most frequently found are syphilis and chancroid. Both are treated with simple antibiotics (erythromycin and benzathine penicillin, respectively).
An antiviral therapy for herpes is not available in most primary health care settings in developing worlds. It is important to treat for chancroid and syphilis, even if some of the genital ulcers treated are actually caused by herpes.
In Rwanda, three different approaches were compared for the management of syphilis and/or chancroid. The syndromic approach adopted by most developing countries, illustrated in Figure 7, resulted in 99 percent of the patients with syphilis and/or chancroid correctly managed. For the approach based on the result of a Rapid Plasma Reagin (RPR) test (if RPR positive, treat for syphilis; if RPR negative, treat for chancroid) and for a clinical etiological approach, the proportions of correctly managed patients were 82 percent and 38 percent, respectively.
Figure - Example of a Flowchart for the Management of Genital Ulcers - BRAZIL
Including an RPR test in a hierarchic model is not an improvement in genital ulcer case management because many chancroid cases are missed. However, based on the Rwanda data, including an RPR test in a syndromic approach (treating all RPR positive patients for both syphilis and chancroid, and all RPR negative patients for chancroid alone), leads to a reduction in unnecessary syphilis treatment of patients and their partners.