DESPITE MORE THAN 15 years of observation, the clinical course and therapeutic response of syphilis in HIV-coinfected persons remains uncertain. Early in the HIV epidemic, several case reports of HIV-infected patients documented aggressive ophthalmologic complications of syphilis, 1 unusual serologic responses, 2 and penicillin treatment failures resulting in the development of neurosyphilis. 3 Such reports prompted the Centers for Disease Control and Prevention (CDC) to recommend pretreatment lumbar puncture in HIV-infected patients with early syphilis (< 1 year duration) or late latent syphilis (> 1 year duration) regardless of neurologic signs or symptoms, and more frequent posttreatment follow-up evaluation with serologic tests for syphilis to document an inadequate titer decline. 4 Despite these reports and admonitions, syphilis has not been regarded as a serious opportunistic infection that predictably progresses among most HIV-coinfected patients as their immunity declines. In fact, several retrospective studies have documented similar clinical presentations and therapeutic responses among HIV-infected patients who presented with and were treated for early syphilis. 5,6 In addition to questions about the effect of HIV on syphilis, limited systematically collected data were available to determine if gender influences syphilis disease manifestations or therapeutic response among HIV-infected patients. To address these issues, the CDC coordinated a prospective, multicenter, randomized, controlled trial of enhanced versus standard therapy to compare the benefit of enhanced therapy, the clinical importance of central nervous system (CNS) involvement, and the clinical manifestations of early syphilis infection among HIV-infected and HIV-uninfected patients. 7 Serologic response among HIV-infected patients with primary and secondary syphilis was slower than among HIV-uninfected patients, clinically defined failure was uncommon, and enhanced therapy did not improve outcome regardless of HIV status. Whereas the initial analyses focused on therapeutic response, in the analyses reported here we focused on clinical manifestations to define and describe differences in the presentation and response to treatment of early syphilis among HIV-infected and HIV-uninfected patients, to describe any differences by gender, and to determine if clinical presentation or CNS involvement predicted serologic failure.
Patient Enrollment and Evaluation
This study has been described in detail previously. 7 It was designed as a prospective, multicenter, randomized, controlled trial of therapy for patients with primary, secondary, and early latent syphilis (< 1 year duration). The study was approved by the boards reviewing research on human subjects at the CDC and at each of eight study sites. At each site, consenting patients with untreated primary, secondary, or early latent syphilis were enrolled.
Primary syphilis was diagnosed if patients had a genital ulcer with typical spirochetes observed on dark-field microscopy of the lesion exudate or with a newly documented, reactive serologic test for syphilis. Secondary syphilis was diagnosed in patients with a rash, mucous patches, condylomata lata, or other signs associated with secondary syphilis manifestations and a newly reactive serologic test for syphilis. Early latent syphilis was diagnosed when the patient presented with no apparent clinical signs of syphilis infection but with a document conversion of a previously negative serologic test result for syphilis. Infections in patients with a history of previously treated syphilis were diagnosed and staged as mentions, but had to show a documented fourfold or greater increase in either rapid protein reagin card test (RPR) or VDRL test titers. Infections in patients with secondary syphilis, but with concomitant genital ulcers, were staged as secondary disease.
Patients were excluded if they were younger than 18 years, pregnant, unable to take penicillin, had taken antibiotics effective against Treponema pallidum within the preceding 2 weeks, or if they required such therapy at enrollment in addition to syphilis therapy.
Patients were interviewed and examined at the initial visit, and scheduled for follow-up visits at 2 weeks, and at 1, 2, 3, 6, 9, and 12 months. Patients were seen by trained clinicians who used standard protocols and data-collection forms. Blood was obtained for syphilis serology at each visit and for lymphocyte analysis at the second visit. Lumbar punctures were recommended for all patients at the initial visit, and also at the 6-month visit for HIV-infected patients and for any other patients who had abnormal cerebrospinal fluid (CSF) findings at the initial visit. At each study visit, patients were asked specific questions regarding symptoms and signs of primary and secondary syphilis, and about any neurologic signs and symptoms, sexual activity, and alcohol and drug use behavior. At each visit, patients also underwent complete physical examinations that included genitourinary, pelvic, and complete neurologic evaluations and fundoscopy.
The serologic tests for syphilis used in these analyses were performed at the CDC with frozen serum samples and included the RPR and the microhemagglutination assay for antibodies to T pallidum (MHA-TP). 8 The HIV-antibody tests were performed using standard methods. 9 Conventional tests for CSF (assays for protein and glucose, leukocyte counts, VDRL tests) were performed at each study site. The criteria for an abnormal result were the presence of a protein concentration greater than 50 mg/dl, a leukocyte count greater than 5/mm3, or a reactive CSF VDRL test.
Statistical methods included chi-square tests for differences in proportions, Wilcoxon rank sum test for medians, Breslow-Day test for homogeneity of odds ratios to evaluate interaction in a stratified analysis, and logistic regression to examine the effect of any neurologic complaint on serologic treatment failure. 10 Serologic treatment success was defined as a RPR that decreased by two or more dilutions or became nonreactive. Logistic regression was performed to examine the effect of any neurologic complaint on serologic treatment failure at the 6-month visit, adjusted for potential confounders (age, sex, stage of syphilis, history of syphilis, HIV status, treatment assignment, initial RPR titer, study site, degree of compliance with medication, incidental use of antibiotics). Because of the small number of patients who underwent lumbar punctures and CSF testing, logistic regression for CSF abnormalities was not performed. Odds ratios (OR) with the corresponding 95% confidence intervals (95% CI) were also computed.
Among 541 patients enrolled between January 1991 and June 1994, 101 (18.7%) were HIV infected (85 men,16 women) and 440 were HIV uninfected (298 men, 142 women) (Table 1). Evaluation of interaction (homogeneity of OR) between gender and HIV status for selected categorical characteristics (Table 1) indicated significant interactions (P < 0.10) for the following characteristics: (1) vaginal sex practices in past 3 months (OR [odds of vaginal sex practices in men versus in women], 0.42 in HIV-infected patients versus 1.00 [odds of vaginal sex practices in men versus in women] in HIV-uninfected patients); (2) crack use in the past 3 months (OR, 0.22 in HIV-infected patients versus 0.77 in HIV-uninfected patients); and (3) giving or receiving drugs or money for sex (OR, 0.32 in HIV-infected patients versus 1.80 in HIV-uninfected patients).
Regardless of HIV status, men were older than women, but did not differ significantly in race, marital status, or years of education. Regardless of HIV status, a larger percentage of men had primary syphilis than did women (P < 0.05). Median serologic syphilis test titers at enrollment were higher among HIV-infected patients than HIV-uninfected patients (P < 0.05). Regardless of gender, more than 80% of HIV-infected patients with early syphilis were aware of their HIV status at presentation, and HIV-infected patients had more syphilis in the past than did HIV-uninfected patients (P < 0.05). Most patients presented to the study sites for evaluation of symptoms. HIV-uninfected women were less likely to have presented for a sexually transmitted disease checkup and more likely to have been referred to the clinic than were HIV-uninfected men. Although not statistically significant, similar patterns were observed among HIV-infected patients.
The HIV-uninfected women reported significantly fewer sex partners in the past 3 months, year, and lifetime than did HIV-uninfected men. In the past 3 months, HIV-uninfected men were more likely to have practiced oral sex and less likely to have practiced rectal sex than were HIV-uninfected women. In the past 3 months, HIV-infected men were less likely to report engaging in vaginal sex than were HIV-infected women, whereas similar rates of vaginal sex were reported between men and women among HIV-uninfected patients. Regardless of gender, rectal sex was more frequently reported among HIV-infected patients than among HIV-uninfected patients (P < 0.05). Regardless of HIV status, rectal sex was more frequently reported by men who had sex with men than by heterosexual men or women (56.5%, 0.4%, and 14.6%, respectively;P < 0.05; data not shown).
In general, HIV-infected patients reported high-risk behavior more frequently than did HIV-uninfected patients. The HIV-infected men reported less use of crack or cocaine in the previous 3 months than did HIV-infected women, whereas a weaker association was reported in HIV-uninfected patients (OR, 0.22 and 0.77, respectively). Women also reported receiving drugs or money for sex more frequently than did the men and men reported providing drugs or money for sex more frequently than did the women. The HIV-infected men reported either providing or receiving drugs or money less frequently than did the HIV-infected women (OR, 0.32); this association was significantly reversed in HIV-uninfected patients (OR, 1.80).
Primary syphilis was diagnosed in 139 patients (25 HIV-infected and 114 HIV-uninfected patients) (Table 2). The median number of ulcers was significantly greater among HIV-infected patients than HIV-uninfected patients, as was the percent of HIV-infected patients with multiple ulcers. The duration of lesions at initial presentation was similar between HIV-infected and HIV-uninfected patients (8.5 days versus 10 days, respectively). Among patients who presented with primary syphilis, the median RPR titer was similar regardless of HIV status (1:8 among HIV-infected versus 1:16 among HIV-uninfected patients).
Among patients with primary syphilis, those with a history of syphilis were more likely to have multiple genital ulcers, [15/32 (47%) versus 34/105 (32%)], but ulcers among patients with past syphilis tended to be smaller (64 mm3 versus 86 mm3). These differences were not statistically significant (data not shown).
Among 253 patients in whom secondary syphilis was diagnosed, a higher percentage of HIV-infected patients presented with genital ulcers [13/53 (25%)] than did HIV-uninfected patients [27/200 (14%)]. The median size of the largest ulcer, the median number of ulcers, the percentage of patients with multiple ulcers, and median duration of ulcer were similar among HIV-infected and HIV-uninfected patients (Table 2). The median duration of rash was slightly longer among HIV-infected patients than HIV-uninfected patients (21 days versus 14 days, P = 0.052), as was the median RPR titer of HIV-infected patients. No differences between HIV-infected and HIV-uninfected patients were detected for other secondary syphilis manifestations. The majority of patients presented with macular or maculopapular rashes (81% of HIV-infected and 76% of HIV-uninfected patients). Rashes involved the trunk and extremities in almost 60% of HIV-infected patients and in more than 40% of HIV-uninfected patients; more than 70% of patients had a palmar-plantar rash, regardless of HIV status. Mucous patches were observed in 8% of patients (12% HIV infected, 7% HIV uninfected), and condyloma lata in 17% (14% and 18%, respectively; data not shown).
When stratified by history of syphilis, the only noted difference in the clinical characteristics of patients who presented with secondary stage syphilis was that patients with a history of syphilis were more likely to present with condylomata lata than were patients with no history of syphilis (32% versus 15%, respectively;P < 0.05; data not shown). Comparing patients with a history of syphilis with those without a history of syphilis, 29% (10/38) versus 26% (56/211) had genital ulcers and secondary manifestations. The distribution of the rash was similar.
Influence of gender and sexual preference on clinical manifestations.
Only 16 of the 158 women in this study (10%) presented with primary syphilis (two HIV-infected and 14 HIV-uninfected women), compared with 32% of men (123/383) (data not shown). A higher proportion of women presented with secondary syphilis [94/158 (59%)] compared with men [159/383 (42%)], regardless of whether men identified their sexual preference as heterosexual or homosexual. Because of these small numbers, only clinical manifestations of secondary syphilis were analyzed by gender. Women who presented with secondary syphilis had a shorter ulcer duration than did all men (Table 3). Among men, heterosexuals had a longer ulcer duration than homosexuals. Men who had sex with men were more likely to present with a rash distribution that included the extremities than were women or heterosexual men. No other differences were detected among patients by gender or sexual preference for other secondary syphilis manifestations.
Neurologic signs and symptoms were infrequent and unrelated to HIV status or history of syphilis. Overall, a history of memory loss was reported by 7.3%, headaches by 24.8%, stiff neck by 7.4%, impaired concentration by 6.7%, visual changes by 6.5%, hearing changes by 3.9%, weakness in any limb by 7.1%, clumsiness in any limb by 3.2%, difficulty with bowel or bladder control by 3.6%, dizziness by 4.8%, or slurred speech by 1.5% in the past 3 months. Among 409 patients who underwent a fundoscopic examination, the results in 396 (97%) were recorded as normal (data not shown). Thirteen patients had reported abnormal fundoscopic examination results, which included chamber clouding in four patients, papilledema in two patients, optic nerve atrophy in three patients, retinal lesions in three patients, and the type of abnormality was not recorded for five patients. Two patients had overlapping abnormalities. Of these 13 patients, 10 were HIV uninfected and 3 were HIV infected; 11 were males and 2 were females, 3 patients had primary syphilis, 8 had secondary syphilis, and 2 had early latent disease.
To evaluate the association between neurologic complaints at initial presentation and other characteristics, we selected a set of complaints that were suggestive of a neurologic disorder due to syphilis (headaches, stiff neck, visual and hearing changes, and weakness or clumsiness in the limb). Neurologic complaints (any of the above complaints) were reported most frequently among patients with secondary syphilis [103/248 (42%)] compared with patients with primary syphilis [32/136 (24%)] and early latent syphilis [48/142 (34%)] (P < 0.05, data not shown). No other differences in neurologic complaints were apparent by HIV status, history of syphilis, or CSF abnormalities. The only symptom that was significantly associated with presence of a CSF abnormality was a stiff neck (Table 4). In a multivariate logistic analysis adjusting for age, sex, HIV status, history of syphilis, stage of syphilis, and study site, stiff neck was associated with CSF abnormalities (OR, 3.86; 95% CI, 1.04–14.26). As presented previously, 7 HIV-infected patients were significantly more likely than HIV-uninfected patients to have > 5 leukocytes/mm3 and somewhat more likely to have reactive VDRL tests of CSF and increased protein levels. In multivariate logistic analysis, there was no association found between any neurologic complaint and CSF abnormalities (OR, 0.99; 95% CI, 0.49–2.02). The frequency of CSF abnormalities did not differ according to the stage or history of syphilis, and was not associated with risk of serologic treatment failure at 6 months (data not shown).
Neurologic predictors of serologic failure.
None of the individual neurologic complains was significantly associated with serologic treatment failures at 6 months (Table 4). Examined by stage of syphilis, presence of any of the examined neurologic complaints was associated with serologic treatment failure at 6 months only for patients with early latent syphilis (48% treatment failure among patients with complaints versus 24% in other patients, P < 0.05, data not shown). However, this finding may be due to confounding because this association was not statistically significant in a multivariate analysis adjusting for age, sex, stage of syphilis, history of syphilis, HIV status, treatment assignment, initial RPR titer, study site, degree of compliance with medication, and incidental use of antibiotics.
In this analysis, HIV infection had a small effect on the clinical manifestations of primary and secondary syphilis. Compared with HIV-uninfected patients, HIV-infected patients with primary syphilis tended to present more frequently with multiple ulcers, and HIV-infected patients with secondary syphilis presented with concomitant genitals ulcers more frequently. Overlapping stages of syphilis have been clinically well documented in early literature, 11,12 but in this study it is difficult to determine if HIV-infected patients progressed to secondary stage of disease more rapidly or had primary chancres that healed more slowly compared with HIV-uninfected patients. In both stages, patients generally presented with similar duration of ulcers and secondary syphilis rashes regardless of HIV status. In contrast to the many case reports and small series reporting unusual manifestation of syphilis among HIV-infected persons, 1–3,5,6 this study showed minimal HIV effect. Furthermore, many of the previous reports were based on observations among patients who required hospitalization for more aggressive management, whereas this study involved outpatients whose results may be more generalizable to the majority of patients who acquire early syphilis.
The more common history of syphilis in HIV-infected patients may have influenced the differences detected in reasons for visits between HIV-infected and HIV-uninfected patients. The HIV-infected patients may have identified symptoms and a need for checkups more frequently because of past experiences. Nonetheless, HIV-infected patients presented with slightly more severe disease clinically and, as we previously reported, 7 failed therapy by titer decline more frequently than did HIV-uninfected patients. Possible explanations may involve failure to mount an adequate protective immune response to new strains, or failure to make the correct immune response. 13 Another explanation may involve an absence or suppression of T-cell–mediated immunity to control infection, as has been shown by the in vitro studies of Marra and Lukehardt 14 in rabbit models. We were unable to analyze the effect of low CD4 cell count on clinical presentation because of the small number of available counts. Only 80 HIV-positive patients have CD4% information. Among HIV-infected patients with primary syphilis, 15 had CD4 cell counts, six of whom (40%) had a CD4 of < 14%. Among 44 HIV-infected patients with secondary syphilis and CD4 cell counts, 14 (32%) had a CD4 < 14%.
Otherwise, HIV infection did not appear to have significant impact on the secondary clinical manifestations of syphilis. Rash distribution and character were similar among all patients regardless of HIV status. A greater proportion of women had palmar plantar rashes than did men, and women and men who had sex with men were more likely to present with secondary syphilis and concomitant genital ulcers of shorter duration than were heterosexual men. Otherwise gender did not influence the clinical manifestations among primary and secondary syphilis cases. In 1934, Stokes and Beerman 12 stated that “clinically, all the manifestations of syphilis tend to be milder and less outspoken in women.” Women were thought to have relatively greater immunity to syphilis, which was hypothesized to be due to the greater degree of lymphocytosis in women, or protective hormonal influences as shown on autopsy studies reporting an apparent resistance of ovaries to syphilis compared with the testes. Also, in more recent reviews, HIV-uninfected women tended to present less frequently in the primary stage and more frequently in the secondary stage of syphilis than did men. 15,16 It has been postulated that this may occur because primary lesions in women may not be as readily noticed given the classically painless nature of the primary chancre and the possibility that lesions can be asymptomatic if located on the cervix. Similarly, rectal chancres may be easily missed or unnoticed in men who engage in receptive anal intercourse with other men. 17 Although 41% of men who had sex with men in this study practiced receptive anal intercourse, women still presented more frequently in the secondary stage than did homosexual men. We did not characterize the appearance or severity of rash beyond the macular, papular, or pustular descriptions; therefore, we are unable to compare our findings with that of Stokes, and more data are needed to investigate gender influence.
Severe neurologic symptoms were not common in this cohort. A few patients with severe neurologic complaints were initially considered to have neurosyphilis and were excluded from enrollment and managed accordingly. Nonetheless, mild neurologic symptoms were commonly reported. Among those reported symptoms, none was associated with serologic treatment failure.
Demographics were similar between HIV-infected and HIV-uninfected patients. As noted for many sexually transmitted diseases, 18 women with syphilis tended to be younger than men, regardless of HIV status. A greater proportion of HIV-infected persons reported a history of high-risk behaviors. Although most HIV-infected patients (80%) knew were aware of their infection, a higher percentage of infected patients reported current high-risk behavior than did those who were not infected with HIV. These data indicate that HIV-infected patients with syphilis may be among the most important transmitters of HIV infection based on the biologic effect of genital ulceration, and compounded by their continued risky behavior. Data indicating an association between genital ulcer diseases and increased blood HIV type 1 RNA levels have been reported. 19–21 Research to quantitate HIV viral load in secretions of patients infected with syphilis that is similar to the work done among HIV-infected patients with incident gonorrhea 22 is needed to evaluate the HIV transmission potential of these coinfected patients. Targeted interventions to implement and sustain behavior change among these core transmitters should be pursued.
1. Radolf JD, Kaplan RP. Unusual manifestations of secondary syphilis and abnormal humoral immune response to Treponema pallidum
antigens in a homosexual man with asymptomatic human immunodeficiency virus infection. J Am Acad Dermatol. 1988; 18: 423–428.
2. Berry CD, Hooton TM, Collier AC, Lukehart SA. Neurologic relapse after benzathine penicillin therapy for secondary syphilis in a patient with HIV infection. N Engl J Med 1987; 316: 1587–1589.
3. Johns DR, Tierney M, Felsenstein D. Alteration in the natural history of neurosyphilis by concurrent infection with the human immunodeficiency virus. N Engl J Med 1987; 316: 1569–1572.
4. Centers for Disease Control, Prevention. Sexually transmitted disease treatment guidelines. MMWR Morb Mortal Wkly Rpt 1993; 42(RR-14): 1–102.
5. Goeman J, Kivuvu M, Nzila N, et al. Similar serologic response to conventional therapy for syphilis among HIV-positive and HIV-negative women. Genitourin Med 1995; 71: 275–279.
6. Telzak EE, Greenberg MS, Harrison J, Stoneburner RL, Schultz S. Syphilis treatment response in HIV-infected individuals. AIDS 1991; 5: 591–595.
7. Rolfs R, Joesoef R, Hendershot E, et al. Early syphilis treatment: a randomized trial of enhanced therapy in HIV infected and uninfected persons. N Engl J Med 1997; 337: 307–314
8. Larsen SA, Hunter EF, Kraus SJ, eds. A Manual of Tests for Syphilis. 8th ed. Washington, DC: American Public Health Association, 1990.
9. Centers for Disease Control, Prevention. Update: serologic testing for antibody to human immunodeficiency virus. MMWR Morb Mortal Wkly Rpt 1988; 36: 833–840.
10. Rosner B. Fundamentals of Biostatitics. Boston: Duxbury Press, 1982.
11. Fuimara NJ. Reinfection primary, secondary and latent syphilis. Sex Transm Dis 1980; 7: 111–114.
12. Stokes HJ, Beermen H, Ingraham NR. Modern Clinical Syphilology. 3rd ed. Philadelphia: WB Saunders, 1944.
13. Blanco DR, Miller JN, Lovett MA. Surface antigens of the syphilis spirochete and their potential as virulence determinants. Emerg Infect Dis 1997; 3: 11–20.
14. Marra CM, Handsfield HH, Kuller L, Morton WR, Lukehart SA. Alterations in the course of experimental syphilis associated with concurrent simian immunodeficiency virus infection. J Infect Dis 1992; 165: 1020–1025.
15. Hutchinson CM, Rompalo AM, Reichart CA, Hook EWIII. Characteristics of patients with syphilis attending Baltimore STD clinics. Arch Intern Med 1991; 151: 511–516.
16. Hutchinson CM, Hook EWIII, Shepherd M, Verley J, Rompalo AM. Altered clinical presentation of early syphilis in patients with human immunodeficiency virus infection. Ann Intern Med 1994; 121: 94–99.
17. Wile UJ, Holman HH. A survey of sixty-eight cases of extragenital chancres. Am J Syph 1941; 25: 58–51.
18. Centers for Disease Control, Prevention. Sexually Transmitted Disease Surveillance 1998. Atlanta, GA: Division of STD Prevention, Department of Health and Human Services, 1999.
19. Ballard RC, Htun YE, Dangor Y, et al. HIV and genital ulcer disease: determinants of HIV shedding from lesions and consequences of therapy (abstract no. 055). In: Abstracts of the 13th Meeting of the International Society for Sexually Transmitted Disease Research. Denver, July 11–14, 1999.
20. Htun YE, Bredell H, Martin D, et al. Influence of reproductive tract infections and their treatment on vaginal shedding of HIV (abstract no. 056). In: Abstracts of the 13th Meeting of the International Society for Sexually Transmitted Disease Research. Denver, July 11–14, 1999.
21. Dyer JR, Eron JJ, Hoffman IF, et al. Association of CD4 cell depletion and elevated blood and seminal plasma human immunodeficiency virus type 1 (HIV-1) RNA concentrations with genital ulcer disease in HIV-1-infected men in Malawi. J Infect Dis 1998; 177: 224–227.
22. Cohen MS, Hoffman IF, Royce RA, et al. Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1. AIDSCAP Malawi Research Group. Lancet 1997; 349: 1868–1873.