Chancroid

INTRODUCTION — Chancroid is a rare infection in the United States and most other developed countries. However, the true incidence of chancroid remains uncertain because a definitive diagnosis requires detection of the causative organism, Haemophilus ducreyi, and few laboratories have the capability of proper microbiologic diagnosis (eg, culture or nucleic acid amplification testing [NAAT]) [1,2]. In addition, many clinicians do not attempt to diagnose genital ulcer disease caused by pathogens other than Treponema pallidum or herpes simplex virus (HSV).

This topic will review the clinical manifestations, diagnosis, and treatment of chancroid. Topic reviews that discuss the approach to patients with genital ulcer disease, syphilis, and genital herpes are found elsewhere. (See "Approach to the patient with genital ulcers" and "Syphilis: Epidemiology, pathophysiology, and clinical manifestations in patients without HIV" and "Epidemiology, clinical manifestations, and diagnosis of genital herpes simplex virus infection".)

MICROBIOLOGY — H. ducreyi is a small, fastidious, gram-negative rod that requires an enriched growth medium containing hemin and usually serum for successful cultivation [3]. Cultures must be delivered expeditiously to the laboratory and incubated at 33° to 35°C in high humidity with CO₂ enrichment. Small, heterogeneous colonies appear on culture medium after 48 to 72 hours. The gray to tan translucent colonies slide intact across the agar plate when pushed.

When examined by Gram stain, organisms from culture often clump in long parallel strands, producing a so-called "school of fish" or "railroad track" appearance. This morphology can occasionally be seen in gram-stained smears from clinical specimens, but it is not a consistent or reliable clinical finding.

PATHOGENESIS — The pathogenesis of chancroid is incompletely understood. In the vast majority of cases, organisms are thought to gain access to tissues via microabrasions in the skin that occur during sexual intercourse since H. ducreyi does not typically infect intact skin. However, the organism has also been identified as a cause of cutaneous ulcers in children and young adults on islands in the South Pacific Islands and in parts of equatorial Africa [4-7].

H. ducreyi is a highly infectious bacterium. The likelihood that inoculation will lead to papule formation appears to be mainly influenced by inoculum dose and host factors [8]. Data from a human model of experimental infection demonstrates that inoculation of a single colony forming unit (cfu) results in papule formation in 50 percent of cases; inoculation of 100 cfu leads to papule formation in 90 percent of cases [9].

The initiation of infection requires attachment of bacteria to susceptible cells, a process that appears to involve the interaction of a protein mediator (probably pili) and lipooligosaccharide with fibronectin contained in the extracellular matrix [10]. Bacterial adherence to cells then occurs via the elaboration of a heat shock protein (GroEL), which may also contribute to the bacterial chaining described above [11]. (See 'Microbiology' above.)

A cytotoxin secreted by H. ducreyi may then play an important role in epithelial cell injury and subsequent development of an ulcer [12,13]. This cytotoxin, present in the majority of strains, has been found to be similar to cytolethal distending toxins (CDTs) from other bacterial species, including Escherichia coli, Shigella, Campylobacter, and Aggregatibacter actinomycetemcomitans [14]. Similar to other CDTs, this toxin causes distension of cultured epithelial cells and fibroblasts, leading to irreversible cell-cycle arrest in the G2 phase of cell growth.

The histologic and immunophenotypic characteristics of chancroid lesions were described in a 1996 report of 11 men, 5 of whom were coinfected with HIV [15]. A superficial purulent exudate was typically present in the epidermis, and a perivascular and interstitial mononuclear cell infiltrate was seen in the dermis. Neutrophils were prominent in most patients without HIV but were less common in those with HIV. The mononuclear cell infiltrate contains many CD4+ lymphocytes, which may explain the increased risk of HIV transmission among persons with chancroid [16-18].

EPIDEMIOLOGY — Chancroid is a rare infection in the United States and most other developed countries, although it can occur in localized outbreaks while remaining endemically absent in most areas. In such settings, there can also be occasional imported cases, which were acquired in endemic areas. However, the true incidence of chancroid remains unclear because of the difficulty isolating H. ducreyi in the laboratory. In addition, most facilities do not test for this organism.

United States and other resource-rich settings — As noted, chancroid is infrequently identified as a cause of genital ulcer disease in developed countries, and, when it has appeared in recent decades, it has generally been in the setting of a localized epidemic, with modest to large numbers of cases. Thus, clinicians and public health agencies must still remain on alert for its reintroduction.

In the United States, chancroid has been diagnosed primarily in certain racial and ethnic groups (African Americans and Hispanic individuals), in heterosexuals, and among female sex workers and their clients [1,3,19]. Cases are more commonly reported in men, at least in part because chancroid is more easily diagnosed in males [20], and uncircumcised men have the highest incidence of infection. Crack cocaine and the exchange of sex for drugs have been implicated in several outbreaks.

Fewer than 20 cases of chancroid per year have been reported in the United States since 2011. A total of 36 cases were reported from 2015 to 2019 [21]; in 2021, only three cases were reported, but the CDC qualified this data by noting that chancroid is not a reportable condition in all jurisdictions. In a report from a sexually transmitted disease (STD) clinic in Paris, France, only 8 of 278 cases of genital ulcer disease evaluated between 1995 and 2005 were diagnosed as being caused by H. ducreyi, and no cases of chancroid were diagnosed after 2002 [22]. However, during recognized outbreaks, chancroid has accounted for up to 20 percent of patients presenting to public STD clinics with genital ulcer disease [19,20,23,24].

While the absence of reported cases of chancroid most likely represents the rarity of this infection, it is important to note that lack of testing may contribute to the apparent disappearance of chancroid [20,23-27]. Limited data suggest that very few patients evaluated in STD facilities are tested for chancroid. As an example, in 1996, a survey of 405 STD treatment facilities in the United States found that only 32 (8 percent) tested patients for chancroid [25]. It is unlikely that this proportion has increased in recent years.

The potential significance of this lack of testing was demonstrated in a study conducted from 1988 to 1990 in New York, where H. ducreyi caused 27 of 65 cases (42 percent) of genital ulcer disease when a microbiologic diagnosis was established [26]. Similar findings were noted in a report of 299 men with nonsyphilitic genital ulcer disease from New Orleans [19]; cultures revealed H. ducreyi in 39 percent and herpes simplex virus (HSV) in 19 percent; culture were negative in 41 percent. These data exemplify the episodic nature of chancroid outbreaks.

Use of nucleic acid amplification testing (NAAT) also suggests that chancroid may be more common than previously believed [20,23,24]. In 1996, data obtained using a multiplex polymerase chain reaction (PCR) test demonstrated that H. ducreyi tended to occur in epidemiologic clusters and accounted for a substantial proportion of genital ulcer cases when present [20]. As an example, cases of chancroid were identified in Memphis and Chicago and accounted for 20 and 12 percent of all cases of genital ulcers in these cities, respectively. In Memphis, the diagnosis of chancroid was not made in any of the patients when clinical criteria were used, and only two of these patients were prescribed an antibiotic regimen that would treat chancroid. However, there are no molecular diagnostic assays for chancroid approved for use by the United States Food and Drug Administration [21]. (See 'Nucleic acid amplification tests' below.)

Resource-limited settings — Although definitive epidemiologic data are not generally available, chancroid has been considered a major cause of genital ulcer disease in sub-Saharan Africa and in many parts of Southeast Asia and Latin America [1,28-30]. However, the prevalence of chancroid has dramatically declined in some countries [31-38], including Thailand, Kenya, Uganda, Namibia, Botswana, and Tanzania. One potential explanation for this observation is the widespread use of the World Health Organization's guidelines for the treatment of sexually transmitted infections, where treatment algorithms are based upon the identification of consistent groups of symptoms and easily recognized signs [39,40].

Nonsexually transmitted cutaneous ulcer disease caused by H. ducreyi has been identified in children and young adults (and in some visitors) from islands in the South Pacific, including Papua New Guinea, Samoa, and Vanuatu, where the cutaneous ulcer disease yaws is endemic [6,40-42]. Identification of the pathogen occurred as part of mass drug administration programs for yaws and trachoma, where azithromycin was given and nucleic acid amplification diagnostic tools were used to assess the etiology of cutaneous ulcers in the treated population [4].

CLINICAL MANIFESTATIONS

Genital ulcers — The incubation period of chancroid is typically 4 to 10 days (range 1 to 35 days). Infection with H. ducreyi leads to an erythematous papule that rapidly evolves into a pustule, which erodes into an ulcer [1,43]. Infected persons commonly have more than one ulcer, and the lesions are almost always confined to the genital area and its draining lymph nodes.

At presentation, a typical chancroid ulcer is about 1 to 2 cm in diameter (picture 1), but the size is variable and there may be more than one ulcer present, especially in patients with HIV. The ulcer is painful and has an erythematous base; the borders are clearly demarcated and sometimes undermined. The base of the ulcer is usually covered with a gray or yellow purulent exudate and bleeds when scraped.

As predicted by the pathogenesis of H. ducreyi infections, the most common sites for chancroid are those that are typically subject to friction during sex. In men, most lesions involve the prepuce, corona, or glans penis. In women, the labia, vaginal introitus, and perianal areas are most commonly affected. Some cases of chancroid may go undiagnosed, especially in asymptomatic women with vaginal or cervical lesions.

Lymphadenopathy — Inguinal lymphadenitis is present in about one-half of infected men (picture 1), but is somewhat less common in women. The involved nodes may undergo liquefaction and present as fluctuant buboes. Most buboes arise one to two weeks after the appearance of the primary ulcer and are often quite painful. Untreated buboes may spontaneously rupture and discharge frank pus.

Nonsexually transmitted cutaneous ulcers — Nonsexually transmitted cutaneous ulcer disease caused by H. ducreyi has been identified in yaws-endemic areas [41], although there is no apparent pathogenetic connection to yaws. (See 'Resource-limited settings' above.)

Compared with the lesions of yaws, cutaneous ulcers caused by H. ducreyi are less circular in shape and are less likely to have central granulating tissue indurated edges. In addition, most lesions are on the limbs rather than the torso. The cutaneous manifestations of yaws are described elsewhere.

DIAGNOSIS

Approach to diagnosis — Most providers rely on clinical criteria to make a diagnosis of chancroid in patients with clinical manifestations consistent with infection [25]. The diagnosis of chancroid is challenging since most sexually transmitted disease (STD) treatment centers do not have the capability of testing for H. ducreyi. (See 'Tests to identify the organism' below.)

However, the sensitivity and specificity of using clinical criteria to make a probable diagnosis of chancroid is variable, and the accuracy of a clinical diagnosis ranges from 33 to 80 percent [19,28,29,44]. Thus, if a diagnosis of chancroid is being considered, patients should also be evaluated for other causes of genital ulcer disease, such as T. pallidum and herpes simplex virus (HSV), which are more likely to occur. In addition, coinfection with H. ducreyi and other sexually transmitted infections have been reported [15,26]. (See "Approach to the patient with genital ulcers".)

Diagnostic criteria — The United States Centers for Disease Control and Prevention (CDC) has developed a case definition for definite and probable chancroid to be used for reporting purposes, and often as a guide for initiation of therapy [45,46]:

●Confirmed – A clinically compatible case that is laboratory confirmed. However, the special culture media required for isolation is not widely available. (See 'Culture' below.)

●Probable – A "probable" diagnosis is made if all four of the following criteria are met:

•The patient has one or more painful genital ulcers

•The patient has no evidence of T. pallidum infection by direct detection (darkfield examination or nucleic acid amplification test [NAAT] of the ulcer exudate or serous fluid) or by serologic testing performed at least 7 to 14 days after the onset of ulcers

•The clinical presentation, appearance of genital ulcers, and if present, regional lymphadenopathy are typical for chancroid (see 'Genital ulcers' above)

•HSV-1 or HSV-2 NAAT or HSV culture performed on the ulcer exudate or fluid is negative

The CDC surveillance case definition does not include NAAT testing for H. ducreyi, even if using a Clinical Laboratory Improvement Amendments (CLIA)-verified test. However, if such a test is clinically available, we would report cases diagnosed in this way as confirmed. (See 'Nucleic acid amplification tests' below.)

Tests to identify the organism — Both culture and NAAT testing are available to identify H. ducreyi. However, these tests are not able to provide rapid results and are not always available. Thus, in most instances, clinicians are unable to base treatment decisions on the results of microbiologic testing and empiric therapy is administered. (See 'Approach to diagnosis' above and 'Antimicrobial therapy' below.)

Gram stain — Gram stain of the exudate from an ulcer can show typical small gram-negative rods in a chain, the so-called "school of fish." However, the sensitivity of the Gram stain is poor [47,48]. (See 'Microbiology' above.)

Culture — A definitive diagnosis of chancroid has traditionally required identification of H. ducreyi on special culture media that is not widely available from commercial sources. The sensitivity of culture for isolating H. ducreyi has been reported to be 60 to 80 percent in patients with clinical symptoms when compared with polymerase chain reaction (PCR) [29,35,49]. Despite modifications in culture media that have improved the yield [1,48], the availability of NAAT testing for many pathogens, including H. ducreyi, has further diminished the value of culture for the diagnosis of chancroid.

Nucleic acid amplification tests — Nonculture methods for the diagnosis of H. ducreyi (eg, PCR) may lead to improvements in the diagnosis of chancroid and H. ducreyi-associated cutaneous ulcer disease [29,49]. Although no United States Food and Drug Administration-cleared nucleic acid amplification tests (NAATs) for H. ducreyi are available in the United States, testing can be performed by commercial laboratories that have developed their own PCR or other NAAT, if the performance of the test has been verified with a CLIA validation study [46]. However, NAAT testing outside of clinical research settings is generally not available in many areas of the world where H. ducreyi cutaneous ulcer disease is found. (See 'Resource-limited settings' above.)

An investigational multiplex PCR assay, which tests for H. ducreyi, T. pallidum, and HSV, was evaluated in 105 patients attending a genitourinary medicine clinic in Lesotho, and was determined to have a sensitivity for H. ducreyi of 95 percent [29]. A similar study, which evaluated 298 genital ulcer swab specimens collected in New Orleans, reported a sensitivity of PCR for H. ducreyi of 98.4 percent and a specificity of 99.6 percent [49].

Despite the potential benefits, PCR or other NAATs may not be a practical diagnostic tool for most publically-funded STD clinics because of the cost and complexity of the test. In addition, NAAT technology for H. ducreyi has not to date been adapted to provide rapid results.

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of genital ulcer disease in sexually active persons is broad and is influenced by the geographic location where the infection was acquired (table 1). In the United States (and increasingly worldwide), most cases are due to herpes simplex virus (HSV) or syphilis. Chancroid remains exceedingly rare, particularly in developed countries. Noninfectious causes include drug eruptions and Behçet syndrome. (See "Approach to the patient with genital ulcers", section on 'Etiologies'.)

A diagnosis based on history and physical examination alone is often inaccurate since findings such as pain, inguinal lymphadenitis, and multiple ulcers are not specific for any one diagnosis. Nonetheless, some findings are more common in certain infections [43]. A detailed discussion of how to evaluate patients with genital ulcers is found elsewhere and summarized below (see "Approach to the patient with genital ulcers"):

●The classic genital presentation of chancroid is with a deep, undermined, purulent ulcer that may be associated with painful inguinal lymphadenitis.

●The classic presentation of genital herpes is with multiple, shallow, tender ulcers that may be vesicular. In addition, only HSV is associated with recurrent disease. (See "Epidemiology, clinical manifestations, and diagnosis of genital herpes simplex virus infection".)

●The classic genital presentation of primary syphilis is with a painless, indurated, clean-based ulcer called a chancre. (See "Syphilis: Epidemiology, pathophysiology, and clinical manifestations in patients without HIV", section on 'Clinical manifestations'.)

●Granuloma inguinale (donovanosis) usually presents as a painless, progressive ulcerative lesion without regional lymphadenopathy [46]. The lesions are highly vascular and beefy red in appearance. As with chancroid, granuloma inguinale is exceedingly rare in the United States but endemic in other areas, such as India, Papua, New Guinea, and southern Africa.

More than one pathogen may be present, further complicating the diagnosis. Nucleic acid amplification testing (NAAT) has shown that this phenomenon may be more common than is evident with culture-based methods used in the past (17 versus 4 percent with culture in one series) [29].

TREATMENT

Antimicrobial therapy — Antimicrobial therapy should be administered to patients with either confirmed or probable chancroid. (See 'Diagnosis' above.)

Several treatment options exist, including azithromycin, ceftriaxone, ciprofloxacin, or erythromycin (table 2) [35,46]. H. ducreyi has developed increasing resistance to some of the antimicrobials that were previously effective, such as trimethoprim-sulfamethoxazole and tetracycline [50-53].

Empiric treatment is reasonable if the clinical manifestations and epidemiology are strongly suggestive of the diagnosis, especially for patients who are unlikely to follow up after their initial evaluation. A detailed discussion of the epidemiology and clinical manifestations of chancroid are found above. (See 'Epidemiology' above and 'Clinical manifestations' above.)

If empiric therapy for chancroid is administered, we typically administer empiric treatment for herpes simplex virus (HSV) and syphilis as well. For most patients, HSV and syphilis are more likely causes of genital ulcer disease. In addition, there are data that describe coinfection with H. ducreyi and T. pallidum [26]. The treatment for syphilis is discussed elsewhere. (See "Syphilis: Treatment and monitoring", section on 'Treatment of early syphilis'.)

The duration of infectivity after initiation of treatment is uncertain. Thus, patients with chancroid should abstain from sex until the ulcer has healed, and condom use should be stressed. (See 'Sex partners' below.)

Preferred regimens — We suggest that patients with proven or suspected chancroid be treated with directly observed single-dose therapy with either azithromycin (1 gram orally) or ceftriaxone (250 mg intramuscularly) (table 2).

Most of the evidence for the efficacy of these regimens comes from uncontrolled observational studies that have shown high cure rates. As examples:

●Azithromycin given as a single 1 gram dose was evaluated as treatment for chancroid in two study sites in Africa. At the first site in Kenya, 127 culture-positive patients were treated with either azithromycin or erythromycin. Cure rates were approximately 90 percent with both drugs. Treatment failure in both groups was significantly associated with HIV infection or being uncircumcised [54]. At the second study site in South Africa, patients were treated in a noncomparative trial [55]. Azithromycin cured 89 percent of patients with culture-positive H. ducreyi infection.

●Ceftriaxone given as a single 250 mg intramuscular injection has been associated with cure rates as high as 98 percent [56]. However, limited data suggest that this regimen may not be as effective in Kenya where failure rates as high as 35 percent have been described [57]. Failure was more common among persons coinfected with HIV. (See 'Patients with HIV' below.)

●Single-dose azithromycin was compared to treatment with single-dose ceftriaxone in a randomized, prospective, unblinded trial of 65 culture-positive patients in New Orleans [58]. Clinical cure or improvement was seen in all patients.

Alternative regimens — A multiple dose regimen of ciprofloxacin (500 mg orally twice daily for three days) is an alternative for patients who cannot take azithromycin or ceftriaxone [46]. Erythromycin base 500 mg orally three times daily for seven days is also effective, but patient adherence to the regimen may be challenging. Both ciprofloxacin and erythromycin have been associated with cure rates above 90 percent [50,59].

As noted, the requirement for multiple doses makes these alternative regimens less desirable. In addition, with erythromycin, there is an appreciable incidence of gastrointestinal side effects, as well as a risk of sudden cardiac death due to QT interval prolongation (especially when other drugs metabolized by CYP3A4 are taken concurrently). (See "Acquired long QT syndrome: Definitions, pathophysiology, and causes".)

Special populations

Patients with HIV — As with other sexually transmitted diseases (STDs), the presence of HIV infection may result in atypical manifestations of chancroid. Such patients may present with numerous lesions, extragenital involvement, and delayed resolution after treatment [47,60]. However, there are no large case series of chancroid in persons with HIV to determine the true incidence or predominant clinical features.

We administer the same antimicrobial regimens to patients with HIV as those described above (table 2). Some experts recommend multiple dose regimens with ciprofloxacin or erythromycin for initial treatment of patients with HIV, based upon a small number of reported treatment failures with single-dose regimens [35]. However, this approach has not been studied prospectively, and we prefer single-dose regimens for patients with HIV, especially those with well-controlled HIV and reasonable immune function [1]. In either case, close follow-up is essential.

Pregnant women — Both azithromycin and ceftriaxone can be used for the treatment of pregnant women with chancroid. The safety of these drugs has been well established based upon studies and clinical experience with the use of azithromycin to treat C. trachomatis and ceftriaxone to treat Neisseria gonorrhoeae in pregnant women [46]. (See "Treatment of uncomplicated gonorrhea (Neisseria gonorrhoeae infection) in adults and adolescents", section on 'Pregnant women' and "Treatment of Chlamydia trachomatis infection", section on 'Pregnant individuals'.)

If neither ceftriaxone nor azithromycin can be used, erythromycin is an alternative, but is poorly tolerated. Ciprofloxacin is generally avoided during pregnancy and lactation due to the small risk of toxicity to the fetus. (See "Fluoroquinolones", section on 'Pregnancy and breastfeeding'.)

Management of buboes — Fluctuant lymphadenitis should be drained, usually by needle aspiration. Data from the era before effective antimicrobial therapy was available suggest that failure to aspirate fluctuant buboes may lead to the development of draining fistulas or secondary ulcers at the site of spontaneous rupture [61]. If the area is not clearly fluctuant, it is reasonable to initiate medical therapy, follow closely, and drain the node if fluctuance develops.

Drainage can be performed by needle aspiration through normal skin (picture 2), or through incision and drainage [35,46]. In a small study performed during a chancroid epidemic in New Orleans, 27 patients with inguinal buboes were randomly assigned to either needle aspiration or incision and drainage [62]. Both groups ultimately did well without any adverse events during the period of follow-up. The patients managed with needle aspiration often required repeated aspirations, but these were well tolerated.

Patient monitoring — Clinical improvement usually occurs promptly after treatment is initiated. Relief of pain is noted by most patients within two to three days, and objective improvement in the ulcers is usually apparent within a week [63]. If no clinical improvement is evident after seven days, then the clinician should consider the following [46]:

●The diagnosis may be incorrect

●The patient is coinfected with another STD (especially syphilis)

●The patient is coinfected with HIV

●The patient was not adherent with medications (if a multiple-dose regimen was prescribed)

●The H. ducreyi strain is drug resistant

For patients who are considered treatment failures, more prolonged treatment with ceftriaxone or a fluoroquinolone may be effective, but there are no definitive data to support this strategy.

The response of chancroid-associated lymphadenitis may occur more slowly. In one study, for example, 8 of 35 patients with inguinal lymphadenitis developed fluctuance that required needle aspiration despite successful treatment of the genital ulcer with erythromycin [64]. In advanced cases, scarring may result despite eradication of infection [46].

SEX PARTNERS — The duration of infectivity after initiation of treatment is uncertain. Thus, patients with chancroid should abstain from sex until the ulcer has dried/resolved, and condom use should be stressed.

Sex partners of patients with chancroid should be treated if they have had sexual contact with the patient within 10 days of symptom presentation [46]. Treatment does not depend upon the presence of symptoms or signs of the disease in the exposed partner. The treatment for exposed partners is the same as for persons with diagnosed infections (table 2), and a single-dose regimen should be administered whenever possible. (See 'Antimicrobial therapy' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sexually transmitted infections".)

SUMMARY AND RECOMMENDATIONS

●The causative agent of chancroid is Haemophilus ducreyi, a small, fastidious, gram-negative rod that requires an enriched growth medium for successful culture. (See 'Microbiology' above.)

●Chancroid is very rare in the United States and most other developed countries. However, the true incidence of chancroid in most areas remains unclear because of the difficulty identifying H. ducreyi in the laboratory. (See 'Epidemiology' above.)

●The incubation period of chancroid is typically 4 to 10 days. Patients typically present with a painful genital ulcer, and multiple ulcers can be present. These ulcers may be associated with regional lymphadenopathy. (See 'Clinical manifestations' above.)

●The diagnosis of chancroid is challenging because testing for H. ducreyi is not routinely available. As such, most providers rely on clinical criteria to make a diagnosis of chancroid in patients with signs and symptoms consistent with infection. In addition, testing for herpes simplex virus (HSV) and syphilis should also be performed. (See 'Diagnosis' above.)

●Empiric treatment is reasonable if the clinical manifestations and epidemiology are strongly suggestive of the diagnosis, especially for patients who are unlikely to follow up after their initial evaluation. If empiric therapy for chancroid is administered, we typically administer empiric treatment for syphilis and HSV as well. (See 'Antimicrobial therapy' above.)

●For patients with confirmed or probable chancroid, we suggest a single-dose regimen using azithromycin (1 gram orally) or ceftriaxone (250 mg intramuscularly) rather than a regimen that requires multiple doses (Grade 2C). Alternative regimens include ciprofloxacin (500 mg orally twice daily for three days) or erythromycin base (500 mg orally three times daily for seven days). (See 'Preferred regimens' above and 'Alternative regimens' above.)

●Fluctuant inguinal lymph nodes should be drained, usually by needle aspiration. (See 'Management of buboes' above.)

●Clinical improvement usually occurs promptly after treatment is initiated. Relief of pain is noted by most patients within two to three days, and objective improvement in the ulcers is usually apparent within a week. (See 'Patient monitoring' above.)

●The presence of HIV infection may result in atypical manifestations of chancroid. In addition, such patients may be at higher risk for treatment failure, and must be followed carefully after the initiation of therapy. (See 'Patients with HIV' above.)

●The duration of infectivity after initiation of treatment is uncertain. Thus, patients with chancroid should abstain from sex until the ulcer has dried/resolved, and condom use should be stressed. In addition, sex partners of patients with chancroid should be treated if they have had sexual contact with the patient within 10 days of symptom presentation. (See 'Sex partners' above.)