The prozone phenomenon with syphilis and HIV-1 co-infection.

Abstract: The prozone phenomenon in syphilis testing refers to a false negative response resulting from overwhelming antibody titers which interfere with the proper formation of the antigen-antibody lattice network necessary to visualize a positive flocculation test. This prozone effect in syphilis testing can be expected in cases of disproportionately high antibody titers, such as secondary syphilis, or with human immunodeficiency virus (HIV) coinfection. Clinicians need to remain familiar with the protean manifestations of syphilis to be able to exclude the prozone phenomenon.

Key Words: HIV, prozone phenomenon, syphilis

**********

The diagnosis of syphilis, caused by the spirochete Treponema pallidum, is usually straightforward. The prozone phenomenon, a false negative reaction due to high levels of antibody titers, can be seen in secondary syphilis, human immunodeficiency virus (HIV) infection, and pregnancy. (1) We report a case of secondary syphilis presenting with HIV coinfection which demonstrated a false negative reaction to the serum rapid plasma reagin (RPR) test, and discuss the expected increased incidence of the prozone effect of syphilis testing in HIV-infected patients.

Case Report

A 21-year-old homosexual man presented with fevers to 103[degrees]F, chills, nausea, vomiting, diarrhea, a reported 20-pound weight loss, red eyes, and a rash on his arms, hands, and feet, thought to have begun three months earlier. He was noted to have conjunctivitis (Fig. 1), thrush, generalized lymphadenopathy, patchy alopecia (Fig. 2), a maculopapular eruption on his arms and hands (Fig. 3), and an extensive papulosquamous rash on his feet (Fig. 4). His laboratory testing demonstrated a serum aspartate aminotransferase level of 87 (normal 17-59) and a serum alkaline phosphatase level of 173 (normal 38-126). A serum RPR was negative, but when diluted at 1:4 became positive and remained so to a dilution of greater than 1:2,056. The result of the serum fluorescent treponemal antibody (FTA) absorption test was reactive.

Additional findings included a positive HIV-1 serology with an HIV RNA quantitative viral load by polymerase chain reaction of 267,457 copies per mL and a CD4 absolute count of 6/mL. His cerebrospinal fluid (CSF) findings demonstrated a total protein of 20, glucose of 68, a white blood cell count of 0, with a positive CSF-FTA. A CSF-VDRL test was requested but not actually performed.

After pretreatment with 20 mg of oral prednisone, the patient was treated with intravernous ampicillin 2 g every 4 hours for 14 days. Outpatient follow-up has demonstrated resolution of his clinical features with sequential RPR titers at 4, 6, and 12 months of 1:256, 1:128, and 1:64, respectively.

Discussion

Syphilis has been referred to as "the great imitator" due to its wide variety of clinical presentations. Therefore, prompt diagnosis and treatment are essential not only to lower transmission rates, but also to avoid the complications seen in the later stages of the disease.

[FIGURE 1 OMITTED]

Our patient's CSF-VDRL was not tested due to an error of omission, and his CSF-FTA was positive. Uncertainty regarding the significance of his CSF-FTA at that time led to treatment for presumptive neurosyphilis. The diagnosis of neurosyphilis in the individual infected with HIV can be difficult. The sensitivity of the CSF-VDRL can vary widely, and the treponemal tests (microhemagglutination assay for T pallidum [MHA-TP] and the FTA) lack specificity for neurosyphilis. In a 1994 study (2) of neurosyphilis in HIV-infection, Tomberlin et al identified 5 subjects out of 58 who had evidence of a local CSF antibody response to neurosyphilis but had negative CSF-VDRL results. These authors argue that measuring the intrathecal antireponemal antibody production using a quantitative CSF MHA-TP measurement and the T pallidum hemagglutinin (THPA) index can assist in suggesting or excluding the diagnosis of neurosyphilis in HIV-infected patients. It is worth noting that three of their five patients who had a negative CSF-VDRL but a positive THPA index had no CSF abnormalities, while the other two had either CSF pleocytosis or an elevated protein. Thus, neurosyphilis could not have been definitively excluded in our patient by virtue of his normal CSF protein and cell count, and further testing was therefore warranted.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

There are three basic methods used in screening for syphilis. These include direct observation of the spirochete by dark field microscopy, and both nontreponemal and treponemal serologic antibody studies. More sensitive nontreponemal tests such as the rapid plasma reagin (RPR) and the Venereal Disease Research Laboratory (VDRL) are used for initial screening, while specific treponemal tests such as the fluorescent treponemal antibody absorption (FTA-ABS) are used to confirm the diagnosis. The importance of these screening tools is shown by past clinical studies that have demonstrated 78% accuracy for the clinical diagnosis of primary syphilis by experienced clinicians. (3) The RPR test is of particular interest in our case because although our patient exhibited many of the characteristic findings of secondary syphilis, his initial screening test for syphilis yielded a false negative as a result of the prozone phenomenon.

[FIGURE 4 OMITTED]

In the RPR test, charcoal particles are added to a cardiolipin antigen suspension. Although the charcoal is not attached to the antigen, it becomes trapped in the lattice created when antibody in positive sera forms an antibody-antigen complex with the test media. This allows the test to be read macroscopically, as opposed to the VDRL test, which is microscopic. The study can be performed as a qualitative test, which uses undiluted patient serum, and also as a quantitative test, which uses serial dilutions of the patient's serum with the endpoint being the highest dilution at which the test is positive. An agglutination or precipitation reaction will be "positive" (visible through the lattice formation) when the optimal ratio of antigen to antibody yields an insoluble precipitate that is visible. The "zone of equivalence" defines this optimal ratio at which the immunologic test is positive. The zones of antibody excess ("prozone") or antigen excess ("postzone") will yield false negative test results because they result in faulty formation of the lattice needed to trap the charcoal particles. (1) All positive RPR tests should have a quantitative test done to document the titer before treatment. Subsequent titers can then be obtained to document response to therapy.

Our patient exhibited a negative result on initial screening by the RPR card test, although the clinical suspicion for secondary syphilis remanded high. It is often stated that fewer than 2% of patients with secondary syphilis will have a result that is nonreactive, a finding known as the prozone phenomenon. (5,1) This phenomenon may occur in those patients with disproportionately high antibody titers. In our case, a high clinical suspicion for secondary syphilis led to a quantitative RPR being requested, which returned positive when diluted and remained positive past a dilution of 1: 2056.

In a 1994 study by El-Zaatari et al, (6) the incidence of the prozone phenomenon was found to be much lower at 0.3%. In this study, 3,504 females and 824 males were screened, with 312 returning positive results. Only one male patient had a false negative result, which became positive upon quantitative examination to a dilution of 1:1,024. In an earlier study from 1971 by Walker, (7) 6,225 serum samples were screened, with 445 being positive and one showing a false negative result. This gives an incidence of 0.2%. The incidence of the prozone phenomenon could be expected to be higher in the HIV-infected population because of the higher serologic tests for syphilis in HIV-infected individuals, even when stratified for syphilis history. (8)

The prozone phenomenon has been noted in five other cases with HIV coinfection in four reports. (4,9-11) In these other reports the diagnosis of syphilis has been made by skin biopsy or diluting the serum nontreponemal test (VDRL or RPR) to as high as 1:64 before becoming positive (Table 1). The nontreponemal titers in these cases are quite high.

Several factors could lead to an increased incidence of the prozone phenomenon in the HIV-infected population. HIV and syphilis each increase the chance of a subject contracting the other disease, and concomitant infection is not uncommon. During 2000 and 2001 the number of cases of primary and secondary syphilis decreased among women but increased among men, accompanied by case reports from several cities from 1997 to 2002 describing syphilis outbreaks among men who have sex with men. (12) These outbreaks were associated with high rates of HIV coinfection and high-risk sexual behavior among subpopulations of men who have sex with men. In addition, early in HIV disease, as in our patient, anomalous B-cell function can lead to an overpopulation of antibody in response to an antigen. (13) The excess antibody produced in response to a new syphilis infection may then lead to inhibition of flocculation on the RPR test card, resulting in the prozone phenomenon.

Conclusion

Although no large studies have been undertaken to determine the incidence of this prozone phenomenon in HIV-infected patients, particular care in this regard should be exercised in those patients in whom the clinical suspicion for syphilis is high when the RPR test returns negative. Our case exhibited many of the prominent clinical findings associated with secondary syphilis. The astute physician who maintains a high index of suspicion and a continued familiarity with the protean manifestations of secondary syphilis will look for the prozone phenomenon in suspected cases. The RPR cards in these situations should then be performed in a quantitative manner with serial dilutions to exclude the prozone phenomenon.

Table 1. Prozone phenomenon with syphilis and HIV-1 co-infection (a)

Report,
Year             Clinical Features         Diagnostic Studies

Hicks (9), 1987  31-year-old male, Kaposi  Three different serum tests:
                   sarcoma, diffuse        1) Negative VDRL
                   erythematous            2) Negative VDRL to 1:32
                   maculopapular eruption     dilutions, serum FTA
                                              negative
                                           3) 3rd test 20 days after 1st
                                              test, VDRL positive at
                                              1:8, FTA positive, skin
                                              biopsy demonstrated
                                              spirochetes
Jurado (4) 1993  Case 1: 31-year-old male  Serum RPR positive upon
  (2 cases)        seen in 1983, no HIV      dilution
                   test performed, oral    Remained so to 1:512
                   thrush, no rash,          dilutions
                   generalized
                   lymphadenopathy
                 Case 2: 30-year-old       Serum RPR became reactive at
                   male, oropharyngeal       1:64
                   ulcers, circular        Serum FTA positive
                   lesions on palms and
                   soles
Jurado (10)      Female, HIV positive      Serum RPR undiluted was
  1994                                       negative
                                           Serum RPR diluted became
                                             positive to 1:512.
Haslett (11)     41-year-old male,         Serum RPR became positive at
  1994             hyperkeratotic rash,      1:4
                   bilateral anterior      Final titer 1:256
                   uveitis, difficulty
                   word-finding, neuro-
                   ophthalmic syphilis
Current Case
Smith 2004       21-year-old male, HIV-    Serum RPR negative
                   positive, rash,         Serum RPR positive when
                   conjunctivitis, fever,    diluted 1:4, remains
                   weight loss               positive to > 1:2056
                                           Serum FTA positive

Report,
Year             CSF

Hicks (9), 1987  Two mononuclear cells, normal protein,
                   CSF-VDRL not reported.
Jurado (4) 1993  Not performed
  (2 cases)
                 No white blood cells, normal protein,
                   negative CSF-VDRL
Jurado (10)      Not reported
  1994
Haslett (11)     White blood cells 66, protein 202 mg/dL,
  1994             CSF VDRL = 1:64
Current Case
Smith 2004       No white blood cells, normal protein, CSF
                   VDRL not tested, CSF FTA positive

(a) CSF, cerebrospinal fluid.

Acknowledgments

All photographs were taken by Robert P. Holman, MD. The authors wish

to express appreciation to Nicole M. Monserrate, MD, for her assistance in reviewing this manuscript.

Accepted June 25, 2003.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9704-0379

Please see William M. Geisler's editorial on page 327 of this issue.

References

1. Berkowitz K, Baxi L, Fox HE. False-negative syphilis screening: The prozone phenomenon, non-immune hydrops, and the diagnosis of syphilis during pregnancy. Am J Obstet Gynecol 1990;163:975-977.

2. Tomberlin MG, Holtom PD, Owens JL, et al. Evaluation of Neurosyphilis in Human Immunodeficiency Virus-Infected Individuals. Clin Inf Dis 1994;18:288-294.

3. Chapel TA, Brown WJ, Jefferies C, et al. How reliable is the morphological diagnosis of penile ulcerations? Sex Transm Dis 1997;4:150-152.

4. Jurado RL, Campbell J, Martin PD. Prozone Phenomenon in Secondary Syphilis-has its time arrived? Arch Int Med 1993;153:2496-2498.

5. Tramont, EC. Treponema pallidum (Syphilis), in Mandell GL, Bennett JE, Dolin R (eds): Principles and Practice of Infectious Diseases. Philadelphia, Churchill Livingstone, 2000, ed 5, pp 2484-2485.

6. El-Zaatari MM, Martens MG, Anderson GD. Incidence of the Prozone Phenomenon in Syphilis Serology. Obstet Gynecol 1994;84:609-612.

7. Walker AN. Rapid plasma reagin (RPR) card test, a screening method for treponemal disease. Br J Ven Dis 1971;47:259-262.

8. Hutchinson CM, Rompalo AM, Reichart CA, et al. Characteristics of Patients with Syphilis Attending Baltimore STD Clinics. Arch Intern Med 1991;151:511-516.

9. Hicks CB, Benson PM, Lupton GP, et al. Seronegative Secondary Syphilis in a Patient Infected with the Human Immunodeficiency Virus (HIV) with Kaposi Sarcoma. Ann Int Med 1987;107:492-495.

10. Jurado RL, In reply to Haslett P. The Prozone Phenomenon in Syphilis Associated with HIV Infection. Arch Int Med 1994;154:1644.

11. Haslett P. The Prozone Phenomenon in Syphilis Associated with HIV Infection. Arch Int Med 1994;154:1643-1644.

12. Centers for Disease Control. Primary and Secondary Syphilis--United States, 2000-2001. MMWR Morb Mortal Wkly Rep 2002;51:971-973.

13. Lane HC, Masur H, Edgar LC, et al. Abnormalities of B cell activation and immunoregulation in-patients with the acquired immunodeficiency syndrome. N Engl J Med 1983;309:453-458.

RELATED ARTICLE: Key Points

* The prozone effect in syphilis testing refers to a false negative result seen in cases of high titers of antibodies interfering with the proper formation of the antigen-antibody lattice network needed to visualize a positive flocculation reaction.

* HIV patients could be expected to have disproportionately higher antibody levels to syphilis and, thus, a higher incidence of the prozone phenomenon in syphilis.

* Clinicians need to remain familiar with the protean manifestations of syphilis and exclude the prozone effect when necessary.

Gregory Smith, MD, and Robert P. Holman, MD

From the Department of Medicine, Georgetown University Medical Center, Washington, DC, and the Virginia Hospital Center, Arlington, VA.

Reprint requests to Robert P. Holman, MD, 1715 N. George Mason Drive, Suite #108, Arlington, VA 22205.