Correspondence
Neurosyphilis in Patients with Human Immunodeficiency Virus Infection
N Engl J Med 1995; 332:1169-1171April 27, 1995
- Article
To the Editor:
The reports by Gordon et al. and Horowitz et al. (Dec. 1 issue)1,2 appropriately question the effectiveness of high-dose penicillin as a treatment for syphilis in persons infected with the human immunodeficiency virus (HIV). However, these reports do not prove that such treatment is ineffective. The polymerase-chain-reaction (PCR) tests used do not necessarily indicate the presence of replicating organisms. In addition, an immunologic marker, especially a nonspecific one such as the Venereal Disease Research Laboratory (VDRL) or rapid plasma reagin titer, cannot be relied on to indicate successful treatment in persons who are immunologically dysfunctional. Patients with severely dysfunctional immune systems are less likely than immunocompetent patients to clear dead organisms. Indeed, in the study by Gordon et al., Treponema pallidum could not be isolated after treatment, and isolation was not attempted by Horowitz et al.
2 ReferencesEdmund C. Tramont, M.D.
University of Maryland Biotechnology Institute, Baltimore, MD 202011
Gordon SM ,Eaton ME ,George R , et al. The response of symptomatic neurosyphilis to high-dose intravenous penicillin G in patients with human immunodeficiency virus infection. N Engl J Med 1994;331:1469-1473
Full Text | Web of Science | Medline2
Horowitz HW ,Valsamis MP ,Wicher V , et al. Cerebral syphilitic gumma confirmed by the polymerase chain reaction in a man with human immunodeficiency virus infection. N Engl J Med 1994;331:1488-1491
Full Text | Web of Science | Medline
To the Editor:
In their report on the use of high-dose intravenous penicillin G for symptomatic neurosyphilis in patients with HIV infection, Gordon et al. conclude, on the basis of clinical, biologic, and serologic data, that this regimen is not consistently effective. However, only one patient had clinical evidence of a therapeutic failure. The rest of the failures were due to lack of either a serologic response or an improvement in the cell count in cerebrospinal fluid six months after treatment.
A serologic response to T. pallidum may not be reliable in patients infected with HIV.1 The same could be said about the cell count and protein concentration in cerebrospinal fluid.2 It has recently been reported that in HIV-negative patients, the Tourtellotte index as a measure of the intrathecal secretion of specific IgG may be a more specific marker of the response to treatment than the total protein concentration.3
In our opinion, the results of the study could be attributed to the lack of sensitive and specific markers of the response to treatment, rather than to the lack of therapeutic efficacy. Until better measures of the response are available, more prolonged clinical observation is needed to conclude that the recommended treatment is not effective.
3 ReferencesJ. Rodriguez-Baño, M.D.
G. Izquierdo, M.D.
M.A. Muniain, M.D.
Hospital Universitario, Virgen Macarena, 41017 Seville, Spain1
Terry PM ,Page ML ,Goldmeier D . Are serological tests of value in diagnosing and monitoring response to treatment of syphilis in patients infected with human immunodeficiency virus? Genitourin Med 1988;64:219-222
Medline2
Augenbraun M, Hyman C, Fishkin E, Benjamin J, Sawchak J. CSF abnormalities in HIV-seropositive patients with latent syphilis. In: Program and abstracts of the 34th Interscience Conference on Antimicrobial Agents and Chemotherapy, Orlando, Fla., October 4–7, 1994. Washington, D.C.: American Society for Microbiology, 1994:181.
3
Martin J ,Aguilar J ,Angulo S ,Rodriguez E ,Izquierdo G . The validity of intrathecal secretion formulas of IgG in the follow-up response to treatment of neurosyphilis. Med Clin (Barc) 1994;103:379-382
Web of Science | Medline
To the Editor:
The data presented by Gordon et al. do not adequately support their conclusions. The authors reviewed 509 cerebrospinal fluid specimens and found evidence of syphilis in specimens from 11 HIV-seropositive and 5 HIV-seronegative patients. Five of the seropositive patients had previously received penicillin G benzathine for early syphilis. The authors interpret these data to suggest that HIV infection increases the risk that early syphilis treated according to standard recommendations may progress to neurosyphilis. Yet this increasingly accepted dictum has never been proved and is based exclusively on case series, such as this one, without HIV-seronegative controls. The authors do little to base their thesis on firmer ground, presenting no comparative data on the treatment histories of their HIV-seronegative patients with neurosyphilis.
The development of neurosyphilis after treatment for early syphilis was documented in the era before HIV.1 The one prospective study comparing the outcomes of early treatment for syphilis in persons with HIV infection and in those without HIV infection found no difference between the two groups.2
The authors also question the efficacy of high-dose intravenous penicillin G as a treatment for neurosyphilis in HIV-infected persons. All their patients had a clinical response except for one with a presumptive diagnosis of meningovascular relapse. In two other patients, whose symptoms resolved after treatment, the treatment was considered to have failed on the basis of the lack of improvement in the serum VDRL titers and the cerebrospinal fluid cell counts, protein levels, and VDRL titers after six months. Yet the cerebrospinal fluid VDRL titer does not reliably indicate the short-term response to treatment,3 and persistent cerebrospinal fluid abnormalities are common in HIV-infected patients, even in the absence of syphilis.
Until careful studies that include HIV-seronegative controls have been reported, the influence of HIV infection on the efficacy of treatment at all stages of syphilis will remain unknown.
3 ReferencesMarc N. Gourevitch, M.D., M.P.H.
Robert S. Klein, M.D.
Ellie E. Schoenbaum, M.D.
Montefiore Medical Center, Bronx, NY 10467-24901
Moskovitz BL ,Klimek JJ ,Goldman RL ,Fiumara NJ ,Quintiliani R . Meningovascular syphilis after `appropriate' treatment of primary syphilis. Arch Intern Med 1982;142:139-140
CrossRef | Web of Science | Medline2
Gourevitch MN ,Selwyn PA ,Davenny K , et al. Effects of HIV infection on the serologic manifestations and response to treatment of syphilis in intravenous drug users. Ann Intern Med 1993;118:350-355
Web of Science | Medline3
Hart G . Syphilis tests in diagnostic and therapeutic decision making. Ann Intern Med 1986;104:368-376
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Dr. Tramont correctly points out that we lack a diagnostic test (or combination of tests) for neurosyphilis that is acceptably sensitive and specific. As a consequence, we lack a method of determining the effectiveness of therapy for treponemal infections of the central nervous system. We tried to address this problem in our discussion, where we pointed out that there is no perfect measure of the response to therapy and that the definition of treatment failure is not clear. However, we cannot put patients on hold until a perfect measure has been devised. Our definition of treatment failure was taken directly from the current recommendations of the Centers for Disease Control and Prevention. If these recommendations are not appropriate, as suggested by Gourevitch et al. as well as by Rodriguez-Baño et al., they should be changed.
Rodriguez-Baño and colleagues also raise the possibility of considering intrathecal antibody as an indicator of active neurosyphilis, noting that the T. pallidum hemagglutination (TPHA) index has been used in this way. Although an elevated TPHA index may improve the specificity of the diagnosis of neurosyphilis, it is insensitive. In two recent studies, the TPHA assay and the microhemagglutination assay for T. pallidum were reactive in only 60 and 71 percent, respectively, of patients with reactive cerebrospinal fluid VDRL tests.1,2 The tests were reactive in 13 and 20 percent, respectively, of patients with possible neurosyphilis.
Gourevitch and colleagues take issue with the suggestion that central nervous system syphilis in patients with concurrent HIV infection may present, progress, and respond to therapy differently from neurosyphilis in those without HIV infection. The relevant data are largely from case reports and small case series like ours.3 We acknowledge that there are no large prospective studies that compare neurosyphilis in patients with HIV infection and in those without HIV infection. The prospective study by Gourevitch et al. did not include examination of the cerebrospinal fluid.4 However, it is likely that the incidence of neurosyphilis is higher and the time to the development of clinical neurosyphilis is shorter in HIV-infected persons than in HIV-negative persons.
In a recent study, patients with HIV infection presented more often in the secondary stage of syphilis, and those with secondary syphilis were more likely to have persistent chancres.5 In addition, we have studied 21 patients who had neurosyphilis with or without HIV infection. The patients with HIV infection were younger and had higher median serum rapid plasma reagin titers at the time of the diagnosis and a shorter time to the development of clinical neurosyphilis. In addition, a significantly higher proportion of the HIV-infected patients presented with ophthalmologic signs of neurosyphilis.6
A number of investigators recommend lumbar puncture as part of the routine evaluation of HIV-infected patients with syphilis. Several larger studies addressing some of these issues are in progress, and we hope that the results will shed more light on the problem of syphilis in the context of HIV infection.
6 ReferencesSteven M. Gordon, M.D.
Cleveland Clinic Foundation, Cleveland, OH 44195Molly E. Eaton, M.D.
Grady Memorial Hospital, Atlanta, GA 30335-3802Sheila A. Lukehart, Ph.D.
Harborview Medical Center, Seattle, WA 98104-24991
Tomberlin MG ,Holtom PD ,Owens JL ,Larsen RA . Evaluation of neurosyphilis in human immunodeficiency virus-infected individuals. Clin Infect Dis 1994;18:288-294
CrossRef | Web of Science | Medline2
Marra CM ,Critchlow CW ,Hook EW III ,Collier AC ,Lukehart SA . Cerebrospinal fluid treponemal antibodies in untreated early syphilis. Arch Neurol 1995;52:68-72
Web of Science | Medline3
Musher DM . Syphilis, neurosyphilis, penicillin, and AIDS. J Infect Dis 1991;163:1201-1206
CrossRef | Web of Science | Medline4
Gourevitch MN ,Selwyn PA ,Davenny K , et al. Effects of HIV infection on the serologic manifestations and response to treatment of syphilis in intravenous drug users. Ann Intern Med 1993;118:350-355
Web of Science | Medline5
Hutchinson CM ,Hook EW III ,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
Web of Science | Medline6
Gordon SM, Eaton M, Kuypers J, et al. Evaluation of neurosyphilis in patients with and without concurrent human immunodeficiency virus. Presented at the 2nd National Conference on Human Retrovirus and Related Infections, Washington, D.C., January 29–February 2, 1995. abstract.
Author/Editor Response
The purpose of our report was to demonstrate that T. pallidum can be found in gumma, heretofore a matter of conjecture. We agree with Dr. Tramont's comment that it is impossible to tell whether treatment with high-dose penicillin failed in our patient, as we note in our article. The reported problems in the use of antibiotic therapy for neurosyphilis in patients with HIV infection, with or without immune suppression, remain to be substantiated. We are aware that DNA-based PCR cannot distinguish live from dead organisms. It is possible that the use of RNA-based PCR will be useful for determining the viability of organisms. However, it is probably fair to conclude that intact organisms (dead or alive) were present in the gumma of our patient, on the basis of the morphologic features of the organisms on stained histologic specimens (Figure 1D in our report). We did not perform rabbit-infectivity testing in our study, since this method is unsuitable for autopsy material.
Little is known about the kinetics of the disappearance of microorganisms and DNA after antibiotic treatment and whether DNA clearance is different for different organisms or dependent on the site of infection. Furthermore, it is unknown whether the presence of DNA implies the presence of an intact cellular structure protecting DNA from digestion and whether an intact cellular structure implies metabolically active and replicating organisms. Knowledge of the rapidity with which immune clearance of organisms occurs in spirochetal infection is important in understanding the pathogenesis of disease-related problems and determining the usefulness of PCR as a diagnostic technique.
It appears that dead spirochetes are rather rapidly removed from local sites of inoculation and from the systemic circulation, on the basis of the studies by Wicher et al. of heat-killed T. pallidum injected into rabbit testicles1 and the report by Malawista et al. of the failure to detect Borrelia burgdorferi DNA by PCR or culture 20 days after antibiotic treatment in infected mice.2 In contrast to these findings, Noordhoek et al.3 detected T. pallidum DNA in cerebrospinal fluid from patients years after they had undergone presumably successful treatment for neurosyphilis; the group included several patients with negative PCR tests before treatment. These data raise questions about the sensitivity of PCR in cerebrospinal fluid, the relation of DNA detection by PCR to the presence of active disease, the kinetics of DNA degradation in the central nervous system, and the efficacy of antibiotic therapy for neurosyphilis. T. pallidum that is sequestered in the brain, without causing apparent clinical disease, may replicate intermittently, leading to intermittent detection in cerebrospinal fluid.
Dr. Tramont hypothesizes that the clearance of dead organisms may be retarded in the immunocompromised host. If the presence of DNA in tissue requires an intact organism, then it is possible that DNA amplification will be able to shed light on this issue in the case of syphilis as well as other infectious diseases.
3 ReferencesHarold Horowitz, M.D.
Gary P. Wormser, M.D.
New York Medical College, Valhalla, NY 10595Konrad Wicher, Ph.D., M.D.
New York State Department of Health, Albany, NY 122011
Wicher K ,Wicher V ,Nakeeb SM ,Dubiski S . Studies of rabbit testes infected with Treponema pallidum. I. Immunopathology. Br J Vener Dis 1983;59:349-358
Medline2
Malawista SE ,Barthold SW ,Persing DH . Fate of Borrelia burgdorferi DNA in tissues of infected mice after antibiotic treatment. J Infect Dis 1994;170:1312-1316
CrossRef | Web of Science | Medline3
Noordhoek GT ,Wolters EC ,De Jonge ME ,Van Embden JD . Detection by polymerase chain reaction of Treponema pallidum DNA in cerebrospinal fluid from neurosyphilis patients before and after antibiotic treatment. J Clin Microbiol 1991;29:1976-1984
Web of Science | Medline
