Are You Confident of the Diagnosis?
What you should be alert for in the history
Granuloma inguinale (GI), or donovanosis, should be considered when persons in those parts of the world where it is endemic, such as South Africa, Australia, Brazil, and India, present with a genital ulcer. In developed countries, GI rarely is a cause of genital ulcers; however, the classic skin findings noted below should prompt suspicion of the disease. In endemic areas, clinicians should be aware of the atypical presentations described below.
Characteristic findings on physical examination
GI most commonly presents as a painless beefy-red (vascular) granulomatous ulcer that bleeds readily to the touch. After an incubation period that averages 2-10 weeks, GI starts as a firm papule or nodule that then ulcerates (Figure 1). Four types of GI have been described:
Ulcerogranulomatous or classic—most common, as descibed above
Hypertrophic or verrucous ulcer—dry irregular growths
Deep ulcers causing tissue destruction
Sclerotic or cicatricial lesions with fibrous and scar tissue
Most lesions occur in the genital and inguinal area (psuedobubos). A small minority will present with lesions in the extragenital sites noted below. True inguinal adenopathy is rare in GI, and if seen, should prompt another diagnosis.
Expected results of diagnostic studies
The diagnosis of GI is confirmed by the finding of typical intracellular Donovan bodies within large mononuclear cells in smears either from tissue or from biopsy specimens. Donovan bodies are 0.5-0.7 by 1-1.5µm bipolar safety-pin-shaped rods, best seen with Giemsa or Wright staining. Polymerase chain reaction (PCR) testing is not generally available and, along with culture and serology, remain mostly research tools.
The diagnosis is best made by Giemsa or Wright staining of surface smears, or by staining of biopsy specimens (for detailed description of techniques, see O’Farrell reference).
Granuloma inguinale can be clinically differentiated from other causes of genital ulcer disease:
– Secondary syphilis (condyloma lata)—bleeding is unusual and rapid plasma reagin (RPR) is uniformly positive
– Primary syphilis—less granulomatous and associated with inguinal lymphadenopathy, darkfield positive
– Chancroid—more typically painful and often with inguinal lymphadenopathy
– Genital warts (human papillomavirus)—rarely bleed, can be confused with verrucous GI
– Penile or vulvar cancer—see below under complications
Coinfection of GI with one or more of the above diseases has been reported.
Who is at Risk for Developing this Disease?
GI is generally regarded as a sexually transmitted disease; however, coinfection rates of sexual contacts is low. GI has been described in young children and sexually inactive adults, suggesting the possibility of fecal transmission and innoculation by contact. GI tends to be found in populations that are impoverished and in those with poor hygiene. GI lesions increase the individual’s suseptibility to human immunodeficiency virus (HIV) infection.
What is the Cause of the Disease?
GI is caused by infection with the gram-negative rod Klebsiella granulomatis, formerly known as Calymmatobacterium granulomatis.
As noted above, the disease is, in most cases, transmitted by sexual contact, although repeated exposure may be needed for innoculation. In general, there is a paucity of data concerning the pathophysiology of Klebsiella granulomatis infection and transmission.
Systemic Implications and Complications
Extragenital lesions account for less than 10% of reported lesions. Sites of infection include the lips, gums, nose, larynx, chest, palate, neck, and gums. Disseminated donovanosis has been noted, with spread to bones and the liver. Dissemination has been most associated with pregnancy. Coinfection of lesions with either typical bacterial skin organisms or other sexually transmitted diseases such as syphilis has been reported.
A rare but important complication of GI is carcinoma (squamous cell carcinoma being more common than basal cell carcinoma), occuring in upwards of 0.25% of cases, according to one series. The histologic differentiation between carcinoma and GI may be difficult, in which case a therapeutic trial of antibiotics may be indicated.
Both elephantiasis and psuedo-elephantiasis (direct infiltration) of external genitalia in women with GI have been described.
Antibiotic choices are based on small series and clinical experience. Durations of therapy, as noted, should be taken as estimates and lesions should, in general, be treated until resolution.
Azithromycin, 1g orally per week for at least 3 weeks, or until healing is complete (Centers for Disease Control [CDC] guidelines)
Azithromycin, 1g initially, then 500mg daily until healing is complete (World Health Organization [WHO] guidelines)
Docycycline, 100 mg twice daily for at least 3 weeks, or until lesions have healed
Ciprofloxacin 750mg daily for at least 3 weeks, or until healing is complete
Trimethoprim-sulfamethoxazole (160mg/800mg), twice daily for at least 3 weeks, or until healing is complete
Erythromycin base, 500mg four times a day for at least 3 weeks, and until all lesions have healed
In the pregnant patient or those coinfected with HIV, adding a parenterally administered aminoglycoside such as gentamycin should be considered. Parenteral gentamycin should be considered in patients slow to respond to therapy. Surgical debridement may be needed in cases of extensive fibrotic or invasive lesions.
Optimal Therapeutic Approach for this Disease
Although definitive diagnosis is recommended, empiric treatment with azithromycin (based on clinical features) is reasonable, either by CDC guidelines or WHO guidelines.
Since both secondary and primary syphilis are much more common than GI, initial therapy with doxycycline as described above, which treats both diseases, would be the initial treatment for possible GI. In nonendemic areas with known outbreaks of chancroid, the addition of 1g of azithromycin, one time, should be added until the diagnosis of GI is made. In nonendemic areas, all efforts should be made to make a definitive diagnosis.
PREGNANT AND LACTATING WOMEN
The erythromycin regimen described above is the treatment of choice. Azithromycin is a possible choice, but published data is lacking. Doxycyline, ciprofloxacin, and trimethoprim-sulfamethoxasole should not be used.
All patients should be tested for HIV at time of initial diagnosis.
All patients treated for GI should be monitored with at least weekly follow-up and treated until lesions are completely healed. Recurrence has been noted in rare cases, despite a good initial response.
Patients who have had sexual contact with the person with GI within 60 days should be monitored and considered for antibiotic therapy.
Unusual Clinical Scenarios to Consider in Patient Management
Some experts have suggested that because of the long half-life of azithromycin, follow-up to complete healing is not mandatory.
In endemic areas, public health initiatives have helped to control the disease.
Psychological counselng of infected patients may be necessary due to the disfiguring nature of the disease and the often slow response to therapy.
In the event of disfigurement, surgical reconstruction of external genitalia may be needed..
What is the Evidence?
O’Farrell, N. “Donavanosis”. Sex Transm Inf. vol. 78. 2002. pp. 452-7. (An up-to-date and detailed review of GI, with attention to diagnostic techniques in different settings)
“Sexually transmitted diseases treatment guidelines”. MMWR. vol. 55. 2006. pp. 1-94.
Rashid, R, Janjua, SA, Khachamoune, A. “Granuloma inguinale: a case report”. Dermatology Online Journal. vol. 12. pp. 14(Case descriptions from the United States, where GI is rare. Includes a brief discussion of the history of GI.)
Mander, S, Baxter, JD. “Granuloma Inguinale and HIV: A unique presentation and novel treatment regimen”. J Am Acad Dermatol. vol. 37. 1997. pp. 494-6. (Case descriptions from the United States, where GI is rare. Demonstrates the recurrence of disease in a treated patient and the difficulty of eradicating GI in the setting of HIV.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.