Genital Ulcers: Physical Clues Are No Substitute for Lab Tests

Source: DiCarlo RP, Martin DH. Clin Infect Dis 1997;25:292-298.

Dicarlo and martin from new Orleans investigated the sensitivity of key physical exam findings in the diagnosis of primary syphilis, chancroid, and HSV in men with genital ulceration. A total of 446 men were evaluated. Fully 41% had no microbiological etiology identified (negative darkfield and negative cultures for HSV and chancroid), 8.3% had dual infections, and 1.6% had secondary syphilis. All of these patients were excluded from the analysis. The remaining 220 patients (49%) had a single etiology identified, including 45 (20%) with primary syphilis, 57 (26%) with genital herpes, and 118 (54%) with chancroid.

There was no significant difference in the duration of ulcers between the three groups, and there was considerable overlap in the physical exam findings. For example, while more than 50% of the patients in each group had inguinal lymphadenopathy, patients with syphilis were slightly more likely to have LAN than those with herpes, and patients with chancroid were more likely to have very large lymph nodes. Only one patient with chancroid presented with a bubo.

A set of "classic clinical signs" was established for each infection. These classical presentations proved highly specific but poorly sensitive for all three infections: the classic presentations for syphilis (painless, indurated, clean-based ulcers) was 31% sensitive but 98% specific; for herpes (multiple, shallow, tender ulcers) was 35% sensitive but 94% specific; and for chancroid (a deep, undermined, purulent ulcer) was 34% sensitive but 94% specific.

DiCarlo and Martin caution that certain individual signs may be more helpful to clinicians than a constellation of findings. For example, 3+ induration in cases of primary syphilis and shallow ulceration in cases of genital herpes were the most specific findings (95% and 88%, respectively). And, to some degree, the predictive value of these findings depends on the frequency of these infections in your population. Because chancroid was the most common infection causing genital ulcer in their population (54%), patients with negative darkfield examination and who did not meet the criteria for herpes were presumed to have chancroid. More sensitive and more rapid diagnostic tools for the diagnosis of genital ulceration are badly needed.