Symptomatic Early Neurosyphilis Among HIV-Positive Men Who Have Sex with Men --- Four Cities, United States, January 2002–June 2004

Symptomatic early neurosyphilis is a rare manifestation of syphilis that usually occurs within the first 12 months of infection (1). Most neurologic symptoms of early neurosyphilis result from acute or subacute meningitis, abnormalities in cranial nerve function, and inflammatory vasculitis leading to a cerebrovascular accident. Symptomatic early neurosyphilis essentially disappeared in the United States after the introduction of penicillin treatment for syphilis in the late 1940s but reappeared in the 1980s among persons with human immunodeficiency virus (HIV) infection (1–3). The disease burden from neurosyphilis is unknown because national reporting of this disease is incomplete. Because the increase in syphilis cases during the past 5 years has occurred primarily among MSM, many of whom were infected with HIV, CDC conducted a review of possible neurosyphilis cases to describe the clinical course of symptomatic early neurosyphilis and to better characterize the risk for this illness among HIV-infected MSM. The review included health department records from four U.S. cities (Los Angeles, California; San Diego, California; Chicago, Illinois; and New York, New York) for the period January 2002–June 2004. This report describes the results of that review, which identified 49 HIV-positive MSM with symptomatic early neurosyphilis during that 30-month period. Among HIV-positive MSM with early syphilis, the estimated risk for having symptomatic early neurosyphilis was 1.7%, and the risk for having early neurosyphilis with persistent symptoms 6 months after treatment was 0.5%. These findings emphasize the importance of preventing syphilis in HIV-infected persons. HIV-infected persons with cranial nerve dysfunction or other unexplained neurologic symptoms should be evaluated for early neurosyphilis.

Possible neurosyphilis cases were identified using health department surveillance reports and interview records of all syphilis cases reported in the four cities during January 2002–June 2004. These records were selected for review if 1) records of early syphilis case interviews administered by disease investigators indicated neurologic signs or symptoms, 2) laboratory reports included cerebrospinal fluid (CSF) tests for neurosyphilis, or 3) the record indicated that the patient was treated with a regimen used for neurosyphilis (e.g., intravenous [IV] penicillin for 10–14 days). Next, additional data were abstracted from medical records and, when indicated, supplemented by interviews with the patients’ physicians. The patients’ HIV status and the sex of their sex partners were determined from medical records, syphilis case interview records, or information provided by the patients’ physicians. Because of limited staff availability, a convenience sample of possible neurosyphilis cases was used for the reviews in New York and Chicago (sampling fractions were 0.68 and 0.55, respectively); cases were more likely to be included if they were reported by health-care providers who reported large numbers of syphilis cases.

A total of 170 possible neurosyphilis cases were reviewed in Los Angeles (74), New York (47), Chicago (32), and San Diego (17); 147 (86%) had serologic evidence of syphilis and met the CDC surveillance case definition for neurosyphilis^* (4) (121 confirmed and 26 probable cases). Ninety-nine (67%) of these 147 cases were in patients with symptoms compatible with early neurosyphilis (e.g., cranial nerve dysfunction, acute meningitis, cerebrovascular accident, headache, or altered mental status of recent onset); of these, 57 (58%) were in patients identified as MSM, 49 (86%) of whom were reported to be HIV positive (39 with confirmed neurosyphilis and 10 with probable neurosyphilis) and included in the case-series analysis.

The 49 HIV-positive MSM with symptomatic early neurosyphilis had a mean age of 38.4 years (range: 21–50 years); 63% were non-Hispanic white, 18% were non-Hispanic black, 14% were Hispanic, and 5% were of other or unknown race/ethnicity. Neurologic complaints were visual disturbances, 25 patients (51%); headache, 16 (32%); gait difficulty, two (4%); hearing loss, two (4%); meningismus, one (2%); altered mental status, one (2%); and unknown symptoms, two (4%). After medical evaluation, the final clinical diagnoses could be grouped into four previously reported syndromes (3): cranial nerve dysfunction (34), meningitis (six), meningovascular syndrome (two), and other syndromes (seven) (Table).

TABLE.

Characteristics of symptomatic early neurosyphilis cases^* among HIV-positive men who have sex with men. by syndrome — four U.S. cities,^† January 2002–June 2004

Syndrome§	Cases		Median no. of hospital days	Median CD4 count (cells/mm3)¶	Median HIV viral load (copies/mL)¶	Median no. of days from symptom onset to start of treatment	Cases in patients with persistent symptoms 6 months after treatment**
	No.	(%)					No. of cases/No. of patients	(%)
Cranial nerve dysfunction
Ocular	25	(51.0)	6	376	28,100	26	6/19	(31.6)
Auditory	6	(12.2	6	322	31,072	24	2/4	(50.0)
Ocular and auditory	1	(2.0)	17	345	866	34	1/1	(100.0)
Other	2	(4.1)	1	14	—††	86	0/1	(0.0)
Meningitis
Acute meningitis	6	(12.2)	7	446	10,020	27	1/6	(16.7)
Meningovascular syndrome
Cerebrovascular accident	2	(4.1)	11	66	1,350	5	1/2	(50.0)
Other syndromes
Headache, altered mental status, or both	7	(14.3)	14	148	24,784	16	0/4	(0.0)
Overall	49	(100.0)	7	312	27,570	25	11/37	(29.7)

^*Information on ODC’s surveillance case definition for neurosyphilis is available at http://www.odc.gov/std/syphsurvrcoo.pdf. Confirmed: syphilis of any stage, a reactive serologic test for syphilis, and a reactive Vanereal Disease Research Laboratory (VDRL) test in cerebrospinal fluid (CSF). Probable: syphilis of any stage, a nonreactive VDRL test in CSF, and both of the following: 1) elevated CSF protein (>40 mg/dL) or leukocyte count (>5 cell/mm³) in the absence of other known causes of these abnormalities and 2) clinical symptoms or signs consistent with neurosyphilis in the absence of other known causes of these abnormalities.

^†Los Angeles, California; San Diego, California; Chicago, Illinois; and New York, New York.

^§Syndrome classification definitions: ocular (inflammation of ocular tissue [e.g., uveitis, retinitis, or optic neuritis]), auditory (e.g., decreased hearing or tinnitus), other cranial nerves (e.g., diplopia or imbalance), acute meningitis (e.g., fever or meningismus), and other syndromes that were thought by the clinician to be manifestations of neurosyphilis.

^¶Measure obtained closest to date of diagnosis within the period 45 days before to 15 days after that date.

^**Denominators based on 37 patients for whom 6-month follow-up was available; three patients with ocular neurosyphilis had a symptomatic neurosyphilis relapse after appropriate treatment and were not included in the persistent symptoms analysis.

^††Data not available.

At the time of neurosyphilis diagnosis, 23 (47%) patients had secondary syphilis, five (10%) had signs of secondary syphilis within 1 week after neurosyphilis diagnosis, 12 (24%) had early latent syphilis, and nine (18%) were reported as having late latent syphilis, including five patients who had nontreponemal syphilis serologic titers of ≥1:32, suggesting probable early syphilis. Twelve (24%) patients reported a previous history of syphilis; of these, nine had adequate previous penicillin treatment documented. Seven of the nine had follow-up nontreponemal titers; of these, six had at least a fourfold decline in titer, suggesting they had an appropriate response to treatment.

Neurosyphilis signs and symptoms often were the only indication that a patient had a syphilis infection; 53% (26 of 49) had no other signs or symptoms of syphilis (21 with latent syphilis and five who had initial signs of secondary syphilis within 1 week after neurosyphilis diagnosis). All patients had one or more lumbar punctures (median: one; range: one to four). Of the 28 patients who had received a radiologic examination of the brain (i.e., computerized tomography or magnetic resonance imaging), 11 (39%) had abnormal results. HIV infection was newly identified (i.e., within 45 days before or after the neurosyphilis diagnosis) in 12 (24%) patients. Among the 28 patients with HIV diagnosed more than 1 year before the onset of neurosyphilis symptoms, 11 (45%) were receiving highly active antiretroviral therapy (HAART) at the time of neurosyphilis diagnosis. Forty-two (86%) patients had been hospitalized (median number of hospital days: 7 days; range: 1–17 days). Forty-seven (96%) patients had received IV penicillin, consistent with CDC’s Sexually Transmitted Disease Treatment Guidelines, 2006 (5), and 41 (81%) had documentation of receiving IV penicillin for at least 10 days. The median time from neurosyphilis symptom onset to start of treatment was 25 days (range: 2–256 days) (Table).

Symptoms often persisted months after treatment. Among the 49 patients, three had experienced a neurosyphilis relapse; all had been retreated. Among the remaining 46 patients, 37 (80%) had 6-month follow-up information available; of these, 11 (30%) had experienced persistent symptoms (Table). Persistence of symptoms at 6 months was not associated with time from neurosyphilis symptom onset to treatment, receipt of HAART, initial CD4 count, or initial HIV viral load (chi-square test for categorical variables [receiving HAART], Kruskal-Wallis test for continuous variables [all others]).

Of the 49 HIV-positive MSM with symptomatic early neurosyphilis, 40 were reported as having early syphilis during January 1, 2002–June 30, 2004. During the same period, in these jurisdictions, 4,776 cases of early syphilis were reported among males (adjusted by applying the sampling fractions in New York [0.68] and Chicago [0.55] to the total number of males reported with early syphilis in those jurisdictions), an estimated 3,916 (82%) cases were in MSM, of whom an estimated 2,380 (61%) were HIV positive. The estimated risk for having symptomatic early neurosyphilis in this population with early syphilis was 1.7% (40 of 2,380), and the risk for having neurosyphilis with persistent symptoms 6 months after treatment was 0.5% (12 [30% of 40] of 2,380).

Editorial Note:

Neurosyphilis is a serious condition that can involve substantial consumption of health-care resources and persistent disabilities for patients. In this case series of symptomatic early neurosyphilis among HIV-positive MSM, most of the patients were admitted to a hospital, received 10–14 days of treatment with IV penicillin, and likely had numerous office visits, expenses, and loss of work time. At 6-month follow-up, 30% of patients had persistence of their principal neurosyphilis symptom. Intensity of persistent symptoms compared with initial complaint was not determined systematically, and most patients were reported as improved, but some will likely have permanent disabilities (6). No relationship was observed between persistent symptoms and promptness of neurosyphilis treatment or HIV-related immunodeficiency; however, only 49 patients were included in this case series.

Symptomatic early neurosyphilis is known to have protean manifestations and variable clinical presentations (7–10). Approximately 75% of the patients in this case series reported visual disturbances or new onset headaches, and 12% had acute meningitis syndrome with fever and meningismus. Approximately half of the patients had no other clinical signs or symptoms of syphilis, and the HIV infections in nearly one quarter were undiagnosed. Because the clinical spectrum of symptomatic early neurosyphilis is varied, health-care providers caring for MSM should consider neurosyphilis in the differential diagnosis of any patient with compatible signs or symptoms or any unexplained neurologic symptom. The recommended treatment for early neurosyphilis (symptomatic and asymptomatic) is aqueous crystalline penicillin G, 18–24 million units per day, administered as 3–4 million units intravenously every 4 hours or by continuous infusion, for 10–14 days, regardless of HIV status.^†

The likelihood that syphilis infection will progress to symptomatic early neurosyphilis in HIV-infected MSM is difficult to determine because neurosyphilis occurs in persons with previously undiagnosed and untreated syphilis. Estimates based on series of reported neurosyphilis cases, as in this analysis, might be inaccurate because certain neurosyphilis cases were undiagnosed or unreported and because the number of persons with undiagnosed syphilis is unknown. However, crude estimates from this analysis indicate a risk of 1.7% for having symptomatic early neurosyphilis and 0.5% for having symptomatic early neurosyphilis with persistent symptoms.

The findings in this report are subject to at least three limitations. First, cases reported from two cities (New York and Chicago) were selected as part of a convenience subsample and might reflect bias toward patients with more severe illness. Second, the number of HIV-positive MSM with symptomatic early neurosyphilis and the estimated population at risk (i.e., HIV-positive MSM with early syphilis) are underestimated. Thirteen possible symptomatic early neurosyphilis cases were not included because information on patients’ HIV status or the sex of their sex partners was missing or because a CSF test was not documented. In addition, certain neurosyphilis and early syphilis cases likely were undiagnosed or unreported. Finally, because medical records were not standardized and had varying levels of completeness, establishing a relationship between clinical and laboratory findings is difficult. Prospectively collected data are needed to more clearly describe the complexities of symptomatic early neurosyphilis.

Health-care providers should be alert to signs and symptoms of neurosyphilis among MSM and should counsel MSM about the various symptoms of neurosyphilis and the risk for illness and permanent disability. Counseling about neurosyphilis and its consequences might promote safer sexual behaviors and decrease transmission of syphilis and other sexually transmitted infections.