Antimicrobial Chemotherapy, 4th Edition
The therapeutic use of antimicrobial agents
Sexually transmitted infections
- C. B. Slack
Venus has a lot to answer for if she was responsible for the conditions that traditionally bear the name of venereal diseases. These afflictions of love are a motley collection of microbial diseases (Table 30.1); some, such as syphilis, are potentially very serious; others, such as trichomonal vaginal discharge, are merely a nuisance. When two human beings are in close contact, their mucous membranes in apposition, transfer of microbes is facilitated. Not all sexually transmitted infections, however, are genital infections; varieties in sexual technique allow for many different parts, not necessarily private, to be in sufficient contact to lead to a clinical lesion. The preferred name for venereal diseases is now sexually transmitted infections (STI) or sexually transmitted diseases (STD), and the specialty dealing with them is genito-urinary medicine (GUM).
Table 30.1 Common sexually transmitted infections and their treatment
The explosive increase in STI world-wide makes it important for all doctors to have a knowledge of their treatment. In some parts of the world STIs are so common that their treatment constitutes a disproportionate part of the total health care budget. Moreover, the recognition that the virus responsible for the acquired immune deficiency syndrome (AIDS) is usually spread sexually and transmission is enhanced by STIs has greatly increased public awareness of these conditions.
Microscopy will give accurate rapid confirmation of a clinical diagnosis in many cases. The examination of exudate from a syphilitic chancre must be done by dark ground microscopy within a few minutes of collecting the specimen; the presence of motile spirochaetes confirms the diagnosis as it is not possible to cultivate these organisms in artificial media. Typical Gram-negative intracellular diplococci in a Gram-stained film of a urethral discharge is strongly supportive of the diagnosis of acute gonorrhoea in a man. Culture of cervical swabs from women and from extragenital sites in both sexes is necessary because examination
of Gram-stained smears is unreliable. The unstained ‘wet’ film of vaginal secretions is the most widely used method for the diagnosis of trichomoniasis and may also reveal Candida and cells suggestive of non-specific bacterial vaginosis.
Direct microscopic examination is of utmost importance in GUM as it confirms many clinical diagnoses, and for this reason many GUM clinics have laboratory facilities on site. In most cases a sufficiently accurate microbiological cause can be ascribed to the patient's complaints to enable specific chemotherapy to be given.
Few genital pathogens can be cultivated easily. The most commonly sought, Neisseria gonorrhoeae, is a fastidious organism requiring special media and growth conditions. Selective media containing antibiotics to inhibit commensal bacteria are used. Isolation of Chlamydia trachomatis requires cell culture. These obligate intracellular bacteria may also be seen by direct fluorescence microscopy; alternatively, the antigenic particles may be detected by enzyme immunoassay techniques and DNA amplification methods are widely available.
Treatment of sexually transmitted infections
Many patients with STIs default treatment. It is therefore important to render as many patients as possible non-infectious after a single visit to the clinic. Concomitant treatment of the sexual partner(s) is essential to prevent reinfection, and contact tracing can help to keep the spread of the disease within bounds, particularly in the control of spread of antibiotic-resistant strains of N. gonorrhoeae.
Curing ‘the pox’ holds an important place in the history of chemotherapy. Heavy metals, in particular mercury, were used for many centuries and the development in the first decade of the twentieth century of arsenicals such as Salvarsan heralded the start of modern chemotherapy. It is difficult to imagine nowadays the horror which was felt about the disease and the importance of finding a cure. It held something of the mystery and fear in the minds of laymen later occupied by cancer and now by AIDS. In its medieval heyday syphilis was apparently a more virulent infection than it is today and earned the adjective ‘great’, diminishing the importance of ‘smallpox’, a disease feared later in history. Syphilis is a chronic disease capable of involving nearly every organ of the body. The old student adage ‘know syphilis and you will know medicine’ indicates how widespread the disease may be and how it can mimic other conditions. The progression of the infection varies greatly; even in untreated cases, latent periods of many years frequently occur.
Penicillin has been the mainstay of therapy since 1943 when the drug was first used to treat syphilis. The primary sore (chancre) will respond to relatively low doses of many antimicrobials, but such treatment may only suppress the disease, which reappears in its later manifestations. This danger exists in treating a patient with non-syphilitic STI who may also be incubating syphilis. For this reason serological tests for syphilis should be done on all high-risk patients attending clinics and repeated after 6 weeks.
Animal work suggests that Treponema pallidum is exquisitely sensitive to penicillin: as little as 0.002 mg/l is bactericidal. There is no evidence of variation in sensitivity and resistance is not known to occur. Other antibiotics with trepone-micidal activity include erythromycin, tetracyclines, chloramphenicol, and cephalosporins, but none is thought to be as active as penicillin, which is the drug of choice in all cases except those where there is definite evidence of allergy.
Even in the early stages of the disease, spirochaetes divide relatively slowly and in latent and tertiary syphilis division presumably occurs infrequently. Since β-lactam antibiotics require active growth to achieve a killing effect, a low concentration of penicillin is required over a prolonged period. This is achieved by using forms of the drug which are released slowly from an intramuscular depot. Aqueous procaine penicillin (600 000 units) will give an adequate concentration for 24 h; benzathine penicillin gives greatly prolonged release, but is no longer used in the UK. In tertiary syphilis treatment for several weeks is necessary. Slow-release penicillins do not achieve adequate CSF concentrations, and frequent high doses of benzylpenicillin are recommended in the treatment of neurosyphilis.
A common hazard of syphilis therapy is the Jarisch–Herxheimer reaction observed within a few hours of treatment with penicillin (or arsenicals). This is a hypersensitivity reaction due to spirochaetal endotoxin and is not related to penicillin allergy. The Herxheimer response is of little significance in primary cases, but may occasionally be fatal in some tertiary or late cases.
In 1829 the Lancet published a series of lectures to students at St Bartholomew's Hospital, London, by Mr Lawrence. In his paragraph on how to cure a clap he proclaims ‘If anyone could find a speedy and effectual mode of accomplishing this he would undoubtedly immortalise himself. The ladies of Fleet Street and the Strand would be inclined to erect a statue to his memory!
Modern antibiotics have revolutionized the practice of GUM and no more so than in ‘curing the clap’. Acute gonococcal urethritis occurs 2–10 days after contact and in men is nearly always obvious, presenting as a visible thick yellow discharge accompanied by dysuria and itching. Asymptomatic cases represent less than 5 per cent of male infections, but about 50 per cent of female infections. Prompt treatment with appropriate antibiotics will cure patients with no residual effects: it is hard to imagine that only 65 years ago gonorrhoea was treated by weeks of local irrigation and many sufferers were left with urethral strictures. Nowadays, the major problems of the disease are seen in women, especially with disseminated gonococcaemia. It is one of the main causes of infertility in the world.
- gonorrhoeaeis sensitive to many antimicrobial agents, but penicillin has remained the drug of choice since it replaced sulphonamides in the later stages of the Second World War. Its value has been maintained by increasing the dose to keep ahead of bacterial resistance. This strategy worked successfully until the emergence of strains completely resistant to penicillin in 1976. In the late 1950s in-vitro testing showed that some strains of N. gonorrhoeaewere becoming less sensitive to penicillin. In one of the clearest demonstrations of the relevance of laboratory tests to clinical practice, Curtis and Wilkinson demonstrated that failure of penicillin was directly related to the MIC of penicillin for the infecting strain. In this study 300 000 units of penicillin were unable to cure any patient infected with gonococci for which the penicillin MIC was 0.5 units/ml (0.3 mg/l) whereas all those infected with strains displaying an MIC <0.015 units/ml were cured. Proportionately reduced cure rates were observed with organisms of intermediate susceptibility.
In the 1960s the widespread and increasing resistance of gonococci to penicillin became well recognized and standard therapy for acute gonorrhoea increased to 4.8 or 5 megaunits in some parts of the world. In countries where antibiotics can be purchased in the market place the prevalence of resistant gonococci is very high. For many of these strains the MIC of penicillin is greater than 1 mg/l, which makes treatment with that drug almost impossible, and resistance to sulphonamides, tetracyclines, and newer drugs is common. Since the discovery of N. gonorrhoeae capable of producing β-lactamase, the position has worsened. These gonococci contain a transmissible plasmid (R-factor) from Enterobacteriaceae which renders them completely resistant to penicillin, ampi-cillin, and amoxycillin. In parts of the Far East and Africa over one-third of all strains are β-lactamase producers. Recommendations for the therapy of gonor-rhoea must be tempered with a knowledge of the antimicrobial susceptibilities of local strains.
In acute disease, in areas in which resistance is uncommon, a single dose of a penicillin giving high tissue concentrations for 12 h is sufficient. Intramuscular injections of procaine penicillin were often used for this purpose, but these have been largely replaced by single oral doses of amoxycillin together with probenecid to delay renal excretion. Oral drugs are mistrusted by some patients who think they are not as good as the injection, and by doctors who worry that patients will not comply, especially if multiple doses have to be given without supervision. Patients with acute gonorrhoea are among the highest defaulters in GUM clinics. There is therefore great importance in a single curative dose and more studies have been performed comparing different regimens for treatment of gonor-rhoea than almost any other condition. Most give a greater than 90 per cent cure rate and many ‘failures’ may in fact be re-infections as patients do not always give accurate information.
For patients infected with β-lactamase-producing strains of N. gonorrhoeae, the newer cephalosporins (especially ceftriaxone), co-amoxiclav, or a fluoroquino-lone such as ciprofloxacin are commonly used. If there is known hypersensitivity to penicillin, cephalosporins may be used, but if the reaction was previously severe the danger of cross-allergy is too great and a non-β-lactam alternative (e.g. co-trimoxazole or ciprofloxacin) should be employed.
Antibiotics give a speedy and complete cure in most cases of acute gonorrhoea. Occasional complications such as epididymitis, arthritis, and pelvic infection in women require admission to hospital and prolonged antibiotics. Non-genital gono-coccal infection also requires more than a single dose of penicillin to effect a cure.
A more common complication following penicillin monotherapy is post-gonococcal urethritis, which is really a double infection with gonococci and the microbes causing non-gonococcal urethritis (NGU; see below). Many physicians prefer to treat all cases of gonorrhoea with amoxycillin plus a tetracycline to cover both infections. This is a good pragmatic public health method for disease control, but offends the principle of making a definitive diagnosis before prescribing antibiotics. If rapid, accurate tests for NGU become available, it may be possible to tailor treatment more scientifically.
This occurs within a few days of birth in babies born to infected mothers. Gonococci in the female genital tract are implanted in the conjunctivae during delivery and the neonate develops a purulent discharge from one or both eyes. There may be considerable cellulitis and if untreated the infection may lead to destruction of the cornea. Treatment should be prompt, with parenteral penicillin if the strain is sensitive and frequent local instillations of saline. The condition may be prevented by silver nitrate drops placed in the eyes immediately after birth. This therapy (Credé's method) is still used in areas where the danger of infection is high but does have a risk of inducing a chemical conjunctivitis especially if the concentration of silver nitrate is too high.
Neonatal conjunctivitis due to chlamydiae is a less severe form of ophthalmia neonatorum than that caused by gonococci. It may be so mild as to be unsuspected clinically and, like all the conditions due to chlamydiae, it is under-diagnosed. In spite of its mild, self-limiting course it can cause permanent eye damage and, whenever suspected, chlamydial conjunctivitis should be treated. Without appropriate laboratory facilities an accurate diagnosis cannot be made, but an index of suspicion is an indication for using tetracycline eye ointment. Local treatment is often difficult to apply adequately and many clinicians also advise giving erythromycin orally to prevent the development of chlamydial pneumonia. Erythromycin is preferred in the infant because systemic tetracy-cline stains teeth and bones. Therapy needs to be for at least 3 weeks as with all complicated chlamydial infections. It is self-evident that the parents should be examined and treated as for non-specific urethritis (see below).
In many men presenting with urethral discharge, gonococci cannot be demonstrated and a diagnosis of non-specific (NSU) or non-gonococcal urethritis (NGU) is made. Such cases are seen more frequently than gonorrhoea in the UK and are common everywhere. Although the primary disease is less severe and the complications fewer than with gonorrhoea, satisfactory treatment is more difficult. Thorough microbiological investigation has determined the aetiological role of Chlamydia trachomatis in nearly half these cases. Some are probably due to ureaplasma, a form of mycoplasma, and a few to Trichomonas vaginalis, herpes-virus, urinary tract infection, and local causes such as trauma or tumours.
Even with elaborate microbiological study, the aetiology of many cases is unknown, but these infections appear to respond to treatment appropriate to chlamydial infection.
The most useful therapeutic agents are tetracyclines, especially doxycycline, but short courses are ineffective. The duration of therapy should be at least 7–10 days. In spite of this, failures are common; some are due to failure of compliance or reinfection, but many genuine relapses occur, confirming the view that chlamydial infections may have a latent phase. Since laboratory facilities for the diagnosis of chlamydial infection are not universally available (and even where they are, results take some time to obtain), ‘blind’ therapy is necessary. The new macrolide azithromycin (p. 41) appears, because of favourable pharmacological properties, to be effective in single dose treatment, although cost limits its use world-wide.
Pelvic inflammatory disease
In women, upper genital tract infection with chlamydiae and other venereal pathogens commonly results in pelvic inflammatory disease (PID), which gives rise to serious complications of tubal blockage or dysfunction. Infertility, ectopic pregnancy, or chronic PID, a condition with considerable morbidity, commonly occur.
Acute salpingitis may respond to treatment with tetracyclines for 7–10 days, but many authorities would advocate longer treatment. Because PID may be caused by a variety of microbes, especially strict anaerobes, combination therapy with metronidazole or co-amoxiclav is necessary. Chronic PID, although initiated by chlamydiae, is more often caused by commensal bacteria, and antimicrobial agents alone seldom have a curative effect.
Genital infection with chlamydiae is commonly undiagnosed and untreated. In the UK it has been estimated that there are about 20 000 untreated infections in women each year. Throughout the world, these organisms cause substantial morbidity in terms of pelvic infection and infertility as well as the blinding eye disease, trachoma, for which 1 per cent tetracycline eye ointment (applied twice daily for 6 weeks) is effective. In severe cases, oral erythromycin may be used. Single-dose azithromycin is also often curative in trachoma.
The normal bacteria flora of the adult vagina before the menopause consists of numerous lactobacilli, diphtheroids, and anaerobes. This maintains locally a pH between 4 and 5 which is inhibitory to coliforms. However, yeasts flourish in acid conditions, as wine-makers know.
Candida albicans, the commonest pathogenic yeast, may be found in up to a quarter of healthy women of child-bearing age and frequently the delicate balance between the resident flora and intruding Candida is disturbed to produce clinical ‘thrush’. Oral antibiotics, in particular tetracyclines, are prone to produce this side-effect which is also more common in pregnancy. Men, especially if uncircumcised, may occasionally have clinical balanitis due to Candida and the organism is not infrequently carried by healthy individuals. Venereal transfer is probable in these circumstances, but thrush can occur without intimate contact. Local applications of nystatin (p. 61) or one of the imidazoles (p. 62) are suffi-cient, but prolonged and repeated courses are required in a few intransigent cases and oral fluconazole (p. 63) is used. The partner must be seen and treated, as with all venereally spread conditions.
Trichomonas vaginalis is a flagellate protozoon commonly found throughout the world. It favours a more alkaline pH than Candida and causes a foul-smelling yellow vaginal discharge often noticed because of staining of clothes and itching. It has been found in a high proportion of asymptomatic women in antenatal clinics, but may cause symptoms subsequently, especially after menstruation. In
some patients the organism invades the anterior urethra and symptoms of dysuria and frequency may lead the clinician to make a tentative diagnosis of urinary tract infection. Some patients labelled as having ‘urethral syndrome’ (p. 239) may be suffering from trichomoniasis. The organism is sometimes carried transiently and asymptomatically by men, but a low-grade non-specific urethritis may occur.
The treatment of trichomonal infection has been revolutionized by the advent of metronidazole. Treatment is given orally. Metronidazole should not be used during pregnancy because of possible teratogenic effects; patients on metron-idazole should also avoid taking alcohol because of a reaction similar to that caused by disulfiram (Antabuse). However, most patients under treatment for venereal diseases are asked to abstain from alcohol and sex: the former to reduce the willingness for the latter and so reduce spread.
This is a term employed for a symptomatic discharge for which no obvious cause can be found. As with NSU there are likely to be many possible aetiological agents—not all microbial. There is evidence that a proportion of these cases are associated with a pleomorphic, Gram-variable rod previously named Haemophilus vaginalis or Corynebacterium vaginale, but now called Gardnerella vaginalis. Like many a potential pathogen it can be found in normal healthy individuals. Metronidazole appears to improve symptoms associated with this organism, but its role may be to inhibit associated anaerobic, curved bacteria, called Mobiluncus. Topical clindamycin may be useful, especially if metronida-zole is contraindicated.
Other genital lesions
Every genital sore thought to be venereal in origin must be considered potentially syphilitic. The long-term complications of treponemal infection are so serious that it is essential not to miss an early infection. Although with clinical experience it is possible to distinguish between the causes of such lesions, mixed infections are not rare in STI.
Herpes simplex type 2
This virus causes vesicles, usually on the penis or labia, similar to ‘cold sores’ found around the mouth. Proctitis is common in passive homosexuals. The painful vesicles burst to form superficial erosions which can be secondarily infected. Women may carry the virus in the cervix where there is a possible complication of infection of the newborn which may occasionally be fatal. Aciclovir is sometimes helpful (seeChapter 28).
Genital warts (condylomata acuminata) are similar to the common skin complaint and local therapy is palliative in many cases. A long course of chemical applications such as podophyllin, or burning the lesions with diathermy or liquid nitrogen, is often required; in some patients the warts disappear spontaneously, giving rise to the myths of ‘charmers’. There is a strong association between these warts, which are caused by some types of human papillomaviruses, and cancers of the cervix and penis. The treatment of genital warts occupies a good part of the work of GUM clinics and is often unrewarding. Imiquimod cream (p. 333) has recently been introduced as topical therapy.
Chancroid (soft sore)
This is caused by Haemophilus ducreyi. The genital lesions are painful and often multiple with large associated inguinal glands which may suppurate to form a ‘bubo’. Erythromycin for 7 days is usually effective; alternatively single-dose azithromycin or ceftriaxone can be used. Tetracyclines and co-trimoxazole work in most cases unless bacterial resistance is common. Short courses of co-amox-iclav or fluoroquinolones have also been successfully used. Chancroid is rarely seen in the UK. but in warmer countries there have been several epidemics and the infection enhances the spread of HIV.
Lymphogranuloma venereum (LGV)
This is also a predominantly tropical condition, but caused by a chlamydial agent. It starts as a small ulcer which may be unnoticed until inguinal glands enlarge and become matted together. Associated inflammation may give the appearance of elephantiasis as a late complication and breakdown of abscesses may give recto-vaginal fistulae. Tetracyclines, sulphonamides or erythromycin may be used but, as with other chlamydial infection, 2–3 weeks of therapy is required.
The treatment of STIs has been complicated by the AIDS pandemic. HIV infection, itself, may require (or demand) drug treatment (seeChapter 28); the opportunist infections associated with the syndrome also require specific therapy (see Chapter 29). STIs are an important link in the spread of the infection. Every occasion on which a patient presents with an STI should be used as an opportunity to reach the high-risk community with a health education message. In those already at risk of contracting HIV infection, counselling and advice should be available at the clinic. Those with full-blown AIDS may also have intractable STIs such as herpes. This may, indeed, be the signal for the diagnosis, so that clinicians expert in the field may be alerted to deal with the patient.