Effect of screening for syphilis on the management of patients with cerebrovascular disease

JStroke Cerebrovasc Dis 1995;5:197-201 © 1995 National Stroke Association

Effect of Screening for Syphilis on the Management of Patients with Cerebrovascular Disease Joseph V. Campellone, M.D., Thomas M. Bosley, M.D., and Dara G. Jamieson, M.D.

Although meningovascular syphilis may result in cerebral ischemia, the effect of syphilis screening on the management of patients with ischemic stroke is unclear. This is due to the variability in syphilis incidence and prevalence, as well as the diverse manifestations that vary with stage of infection. We prospectively screened patients with cerebrovascular disease (CVD) for syphilis and analyzed seropositivity rates as well as assessed the impact of screening on patient management. There were relatively high rates of seropositivity in both CVD (12.5%, n = 104) and control (9.9%, n = 211) populations without a Significant difference between groups. Overall, syphilis screening had little impact on the management of the patients with CVD. No cerebral event was attributed to meningovascular syphilis, and there was no identifiable group in which syphilis screening was more likely to result in treatment for meningovascular syphilis, although there were significantly (p < 0.001) higher rates of seropositivity in black patients than whites. The relatively high incidence of seropositivity in both groups is representative of local syphilis prevalence and suggests that screening is warranted in patients with neurological disease, but not necessarily as part of the evaluation for cerebral ischemia. Key Words: Syphilis-Neurosyphilis-Cerebral infarction.

A diagnosis of meningovascular syphilis has important implications on treatment and prognosis in patients with cerebral ischemia. Although syphilis remains a recognized etiology of ischemic stroke, the diagnosis may be difficult because this complication most commonly occurs 4-7 years after primary infection (1,2) and ranges from months to 12 years (3). The need for excluding syphilis in all patients with symptoms of cerebral ischemia is an important consideration in the development of a rational and cost-efficient diagnostic evaluation. The role of syphilis testing in patients with cerebrovascular disease (CVD) is obFrom the Department of Neurology, Pennsylvania Hospital, Eighth and Spruce Streets, Philadelphia, PA, U.S.A. Address correspondence and reprint requests to Dr. J. V. Campellone at Department of Neurology, University of Pittsburgh Medical Center, 322 Scaife Hall, Pittsburgh, PA 15261, U.S.A.

scured by variability in syphilis incidence across geographic boundaries (1,4) as well as within a given population over time (4-7). The rates of syphilis infection have risen nationally and are particularly high within the Philadelphia region (5,6,8). In consideration of these factors, a study was undertaken to determine the relevance of syphilis screening toward the care of patients admitted with CVD at Pennsylvania Hospital, a SOD-bed, inner-city community hospital.

Methods During a 3D-week period beginning June 1993, consecutive patients admitted to the neurology service or inpatients seen in neurological consultation at Pennsylvania Hospital were prospectively assessed for evidence of syphilis infection. Both Rapid Plasma Reagin (RPR) Assay (Macro-vue RPR Card Tests;

JSTROKE CEREBROVASC DIS, VOL. 5, NO.4, 1995

197

J. V. CAMPELLONE ET AL. Brewer Diagnostic Kits, Becton-Dickinson) and MicroHemagglutination Assay for Treponema pallidum (MHA-TP) (Sera- Tek, Miles) were performed on the sera from all patients by the same laboratory. Patients were categorized within 1 week of entry based on their primary neurological diagnosis. Patients with recent (less than 1 week) ischemic cerebral infarction (CI), transient ischemic attack (TIA), or resolving ischemic neurologic deficit (RIND), as determined by a staff neurologist, were categorized into the CVD group, whereas all other patients were categorized into the neurological control group. Patients with remote ischemic infarction but no acute cerebrovascular event were placed in the control group based on their more immediate neurological symptoms. Patients with subarachnoid hemorrhage or a primary hemorrhagic infarction as suggested by clinical and

Table 1.

radiographic features were also placed in the control group. Attempts were made to obtain both serologic assays on all patients. Patients who did not have both syphilis assays performed were excluded from analysis. Reasons for both assays not being performed were (a) the patient died or left against advice before phlebotomy; (b) the patient refused blood draw; or (c) laboratory or clerical error. Patients noted to have recently documented syphilis serology were also excluded. Values of p were determined by Chi-square analysis and Fisher's exact test when the expected frequency of two or more cells was less than five.

Results A total of 104 CVD patients and 211 neurological controls met criteria for inclusion in the analysis.

Outcome of cerebrovascular disease (CVD) patients with positive syphilis serology

Patient

Age"/raceb/sex

RPRc

MHA d

Outcome

1

70/B/M

Neg

4+

2

69/B/F

Neg

Bdr

3

53/H/M

1:4

Neg

4

621B/F

1:2

3+

5

75/B/F

1:1

Neg

6

74/B/F

Neg

1+

7

8I/B/M

1:8

4+

8

37/B/M

1:2

Bdr

9

78/B/F

1:1

Bdr

10

77/B/F

1:32

3+

11

69/B/F

1:4

4+

12

65/A/M

1:8

4+

13

68/B/M

1:8

3+

Echocardiography revealed a large left ventricular thrombus, necessitating anticoagulation Patient had an anticardiolipin antibody syndrome; results were felt to represent false-positive Had cerebral infarction associated with acute myocardial infarction; there was no clinical evidence of syphilis Patient had a hypercoagulable syndrome in association with disseminated malignancy and expired without eSF examination; there was no clinical evidence of neurosyphilis Atrial septal aneurysm was discovered by echocardiography; there was no clinical evidence of syphilis Further investigation revealed previous treatment for latent syphilis; she had diabetes mellitus, hypertension, heavy smoking history, and diffuse hypokinesis of the left ventricle Patient experienced devastating perioperative cerebral infarction and expired Patient had rheumatic heart disease and severe mitral regurgitation, leaving against medical advice An atrial aneurysm was found by echocardiography; investigation revealed prior treatment for latent syphilis Patient had high-grade carotid stenosis ipsilateral to cerebral infarction and was discovered to have been previously treated for latent syphilis with a prior RPR titer of 1:164 Was discovered to have been previously treated twice for latent syphilis, following normal eSF examinations; she had a history of smoking, poorly controlled hypertension, and new-onset paroxysmal atrial flutter Patient had normal eSF and a history of hypertension; he was treated for latent syphilis Had high-grade ipsilateral carotid stenosis but did not follow up as outpatient for seropositivity

•Age is reported in years. bB, black; H, Hispanic; A, Asian. "Titer of Reactive Plasma Reagin Assay. dTiter of Micro-Hemagluttination Assay for Treponema Pallidum; Neg, negative assay; Bdr, borderline positive result; eSF, cerebrospinal fluid. 198

J STROKE CEREBROVASC DIS, VOL. 5, NO.4, 1995

EFFECT OF SYPHILISSCREENING ON PATIENTS WITH CVD

Male-to-female ratio (0.89:1 CVD group; 0.90:1 controls; p = 0.963) and black-to-white ratio (0.45:1 CVD, 0.37:1 control; p = 0.472) were similar between groupS. Average age of CVD patients (70.24 ± 1.29 years) was significantly older than controls (60.50 ± 1.34; p < 0.001). Positive results for either RPR or MHA-TP or both were found in 13 (12.5%) CVD patients and in 20 (9.5%) controls, representing no significant difference between populations (p = 0.493). Syphilis was not implicated as the etiology of symptoms in any patient with CVD (see Table 1), and there were no patients in the CVD group who received treatment for tertiary syphilis. Controls with positive serology underwent further evaluation based on the clinical judgment of their primary physician. Eleven controls with positive serology underwent lumbar puncture, resulting in one patient (1/211; 0.5%) receiving treatment for neurosyphilis. This represented no significant difference from the CVD group (0/104; 0%) for patients treated for tertiary syphilis (p = 0.670). The control treated for tertiary syphilis was a 72-year-old black male with encephalopathy following an insulin-induced hypoglycemic event. He had a 1:2 RPR titer and a 4+ MHA- TP. Spinal fluid demonstrated a negative VDRL and contained 171 mg/dl protein, normal cell counts, and normal glucose. There were three controls (1.4%) and one CVD patient (0.9%) who received treatment for latent syphilis, also representing no significant difference between groups (p = 0.798).

Three of the 13 patients in the CVD group with a positive MHA-TP had nonreactive RPR assays. One had a weakly positive MHA- TP that was felt to represent a false-positive result due to an anticardiolipin antibody syndrome (Patient 2). Another (Patient 6) was found to have been previously treated for latent syphilis. The remaining patient required anticoagulation for a cardiac intraventricular thrombus, and lumbar puncture was deferred. Among CVD patients, there were two (Patients 3 and 5) with falsepositive RPR titers. In both circumstances, RPR titers were low, there was no clinical evidence of syphilis, and other risks for cerebral ischemic events were present. The results of subgroup analysis for seropositivity rates are presented in Table 2. There was a large and statistically significant difference in seropositivity rates between white and black populations in both the control and CVD patient populations. Among CVD patients, 11/31 (34.5%) black patients had at least one positive serology test compared to 0/69 (0%) white CVD patients (p = 0.0001). Of 56 black control patients, 14 (25.0%) were seropositive for either RPR, MHA-TP, or both, whereas only 6 of 151 (4.0%) white controls had at least one reactive assay (p = 0.0001). (Five Hispanic and three Asian patients were not included in analysis of race.) Among CVD patients, there was no difference between white and black populations in the percentage treated for tertiary syphilis, since patients of neither race were treated for

Table 2. Subgroup analysis of seropositivity rates betuieen control and cerebrovascular

disease (CVD) patients"

Male Female

P Age <45 years Age >45 years

P White h BlacI
P

CVD

Control

p

17/149 (11.4%) 17/166 (10.2%)

6/49 (12.2%) 7/55 (12.7%)

11/100 (11.0%) 10/111 (9.0%)

0.822 0.457

0.739

0.941

0.630

6/61 (9.8%) 28/254 (11.0%)

1/5 (20.0%) 12/99 (12.0%)

5/56 (8.9%) 16/155 (10.3%)

0.788

0.884

0.765

6/220 (2.7%) 25/87 (28.7%)

0.69 (0%) 11/31 (35.4%)

6/151 (3.9%) 14/56 (25.0%)

<0.001

<0.001

<0.001

0.929 0.655

0.101 0.301

4Values of p listed horizontally indicate significance of differences in seropositivity for sex, age, and race, respectively, for total, CVD, and control groups. The p values in the right column represent significance of seropositivity rates between CVD and control groups in various subpopulations. The only significant difference in seropositivity rates was seen between white and black patients. However, these differences were seen in both control and CVD groups. hFiveHispanic and three Asian patients were not included in subgroup analysis with regards to race.

JSTROKECEREBROVASC DIS, VOL. 5, NO.4, 1995 199

t. V. CAMPELLONE IT AL.

such. This was similar to controls, since only 1 of 55 (1.8%) black patients and none of the 149 white control patients were treated for tertiary syphilis (p = 0.270). Further analysis disclosed no significant difference in seropositivity rates with regard to sex (p = 0.739) or age (p = 0.788) in either control or CVD groups (see Table 2).

Discussion The invasion of Treponema pallidum into the central nervous system (CNS) may result in ischemic events through an endarteritis, affecting vessels of any caliber (9). Since many patients may not recall previous syphilis infection (9), physicians should be cautious about relying on history alone. The presence of relatively cheap and safe screening methods for this treatable systemic infection suggests that exclusion of meningovascular syphilis is justified in all patients with symptoms of cerebral ischemia. A recent increase in the frequency of neurosyphilis (2-7), especially vascular and atypical forms (10,11), suggests such screening is warranted, although this is challenged by recent studies demonstrating very few cases of meningovascular syphilis in populations of stroke patients (12-15). Despite increasing syphilis incidence in the Philadelphia area (4), our data suggest that syphilis screening contributes little toward the management of a population of patients with CVD. Serologic testing in our study population resulted in no patient being treated for neurosyphilis. This treatment rate was similar to our group of neurologic controls, in which unconditional screening with both treponemal and reaginic assays resulted in only one patient (0.5%) receiving treatment for neurosyphilis. Strong risk factors for cerebral ischemic events were present in all CVD patients with positive syphilis serologies (Table 2). In combination with clinical judgment, lumbar puncture (LP) is necessary to exclude or confirm neurosyphilis in patients with positive serology and neurological symptoms (2,16). In some of our CVD patients, this was impractical, since the patient's medical condition or his refusal of appropriate evaluation precluded the procedure. Two of our patients (Patients 4 a~d 7) were so ill that seropositivity was of no relevance to their immediate care. In other instances, patients were unaware of prior syphilis infection, and investigation revealed they had been adequately evaluated and treated previously with documented decline in RPR titer. In other circumstances, other stroke risk factors for cerebral infarction were present, and patient's primary physicians took recommendations for LP under advisement. Similar difficulties in excluding neurosyphilis in seropositive CVD patients 200

/ STROKE CEREBROVASC DIS, VOL. 5, NO.4, 1995

were encountered by Kelley et al. (13) and seem inherent in any study designed to screen for neurosyphilis. Because of these limitations, we cannot make an accurate assessment of the true incidence of meningovascular syphilis in this patient population. Accurate knowledge of prior syphilis evaluation and treatment could certainly eliminate many unnecessary serologic tests. In such patients, an RPR titer alone would be sufficient to exclude incompletely treated syphilis. However, screening may be more cost- and time-efficient than contacting every patient's prior physician. Unnecessary testing can also be reduced by not screening patients deemed to be inappropriate candidates for LP or empiric treatment for neurosyphilis. In the largest series of patients with cerebrovascular syphilis (9), 74% of the cases were under the age of 50 years, insinuating that syphilis testing may be of increased utility in younger populations with CVD. However, our data did not suggest that syphilis infection had any increased role in cerebral ischemic symptoms in patients with CVD younger than 45 years old when compared to patients aged 45 years or older. Also, the seropositivity rates among our young patients did not differ between CVD and control groups. This concurs with other studies in which syphilis was an uncommon etiology of ischemic events in young patients not co-infected with human immunodeficiency virus (HIV) (12,15,17,18), although there are regions with significantly higher rates (19). We found a large difference in seropositivity rates between white and black cohorts of patients in both the CVD and control groups. Others have reported similarly higher rates among black patients (13) which is consistent with the generally higher rates of syphilis infection among black patients seen nationally (5). Despite this significant difference, there is no clear advantage of screening black patients with stroke for neurosyphilis, since no cases were detected in either race. Several cases of HIV-infected patients with cerebrovascular syphilis have been reported (13,20,21). It has been suggested that syphilis infection may progress more aggressively in patients coinfected with HIV (22,23), although this has been disputed (24). In our CVD group, none were known to be infected with HIV. Of 15 controls with HIV infection, six had syphilis seropositivity. Five underwent lumbar puncture, revealing no cases of central nervous system involvement. Several were discovered to have been previously treated, and repeat assays were negative in the others. Given the absence of HIV-infected patients in our study group, we can form no conclusions about the role of syphilis screening in this patient

EFFECT OF SYPHILIS SCREENING ON PATIENTS WITH CVD

population, although the available information suggests it is wise. Commonly utilized methods for syphilis screening consist of reaginic and specific treponemal assays. The choice of assay depends on specific circumstances, since there are many variables that can influence sensitivity and specificity. RPR titers may be insensitive in advanced stages of infection (10) independently of treatment, raising concerns that meningovascular syphilis could be neglected when RPRassays are used as the sole screening method. We did not discover that using both assays increased the sensitivity of diagnosing neurosyphilis in CVD patients. There was only one patient in the CVD group with a negative RPR and positive MHA-TP (Patient 1) in whom neurosyphilis could not be excluded. A left cardiac thrombus was the likely cause of his cerebral infarction, and there was no clinical stigmata of syphilis infection. At our institution, MHA- TP assays are approximately fivefold the expense of RPR assays. The groundless use ofMHA-TP as a screening instrument needs to be considered when applied to large numbers of patients.

Conclusions Screening a population of CVD patients for neurosyphilis had little impact on their overall management. In an area with a high prevalence of syphilis, routine testing resulted in no diagnoses of meningovascular syphilis in patients with CVD. In part, this was due to the inability to exclude completely neurosyphilis in some patients. Unnecessary screening may be avoided by appropriate exclusion of patients. in which seropositivity is of little relevance. Since all stages of syphilis are treatable and serologic testing is relatively cheap, the high rates of syphilis seropositivity in both study and control populations suggest that screening patients with any neurological symptoms may be justified, not particularly those with ischemic stroke. Acknowledgment: The authors gratefully appreciate the efforts of Dr. James Reed in performing statistical analysis of our data.

References 1. Hook EW, Marra CM. Acquired syphilis in adults. N Engl ] Med 1992;326:1060-9. 2. Cintron R, Pachner AR. Spirochetal diseases of the

nervous system. CUrT Opin NeuroI1994;7:217-22. 3. Simon RP. Neurosyphilis. Arch NeuroI1985;42:60613. 4. Webster LA, Rolfs RT, Nakashima AK, Greenspan JR. Regional and temporal trends in the surveillance of syphilis, United States. MMWR Surueill SU11l 1991;40: 29-30. 5. CDC. Syphilis and congenital syphilis-United States, 1985-1988. MMWR 1988;37:486-9. 6. CDC. Relationship of syphilis to drug use and prostitution-Connecticut and Philadelphia, Pennsylvania. MMWR 1988;37:755-64. 7. Lukehart SA, Hook EW, Baker-Zander SA, Collier AC, Critchlow CW, Handsfield HH. Invasion of the central nervous system by Treponemapallidum: implications for diagnosis and treatment. AIm Intern Med 1988;109: 855-62. 8. Rolfs RT, Goldberg M, Sharrar RG. Risk factors for syphilis: cocaine use and prostitution. Am] Pub Health 1990;80:853-7. 9. Merrit HH, Adams RO, Solomon HC. Neurosyphilis. New York: Oxford University Press, 1946. 10. Landi G, Villani F, Anzalone N. Variable angiographic findings in patients with stroke and neurosyphilis. Stroke 1990;21:333-8. 11. Burke JM, Schaberg DR. Neurosyphilis in the antibiotic era. Neurology 1985;35:1368-71. 12. Bevan H, Sharma K,Bradley W. Stroke in young adults. Stroke 1990;21:382-6. 13. Kelley RE,Bell L,Kelley SE Lee Soc. Syphilis detection in cerebrovascular disease. Stroke 1989;20:230-4. 14. Rolak LA, Gilmer W, Strittmater WJ. Low yield in the diagnostic evaluation of transient ischemic attacks. Neurology 1990;40:747-8. 15. Chancellor AM, Glasgow GL, Ockelford PA, Johns A, Smith J. Etiology, prognosis, and hemostatic function in young adults. Stroke 1989;20:477-82. 16. Hart G. Syphilis tests in diagnostic and therapeutic decision making. Ann Intern Med 1986;104:368-76. 17. Adams HP, Kappelle LJ,Biller J, et al. Ischemic stroke in young adults. Arch NeuroI1995;52:491-5. 18. Bogousslavsky J, Pierre P. Ischemic stroke in patients under 45. Neurol Gin 1992;10:113-24. 19. Srinivasan K Ischemic cerebrovascular disease in the young-two common causes in India. Stroke 1984;15: 733-5. 20. Berry CD, Hooten TM, Collier AC, Lukehart SA. Neurologic relapse after benzathine penicillin therapy for secondary syphilis in a patient with HIV infection. N Engl ] Med 1987;316:1587-9. 21. Tyler KL,Sandberg E,Baum KF. Medial medullary syndrome and meningovascular syphilis. Neurology 1994; 44:2231-5. 22. Lanska MJ, Lanska OJ, Schmidley JW. Syphilitic polyradiculopathy in an HIV-positive man. Neurology 1988; 38:1297-1301. 23. Johns DR, Tierney M, Felsenstein O. Alteration in the natural history of neurosyphilis by concurrent infection with the human immunodeficiency virus. N Engl] Med 1987;316:1569-72. 24. Simon RP. Neurosyphilis. Neurology 1994;44:222830.

] STROKE CEREBROVASC DIS, VOL. 5, NO.4, 1995

201