482
Point-of-Care Tests for Diagnosis of Neurosyphilis
Christina Marra*, K Hall, L Tantalo, S Sahi, and T Jones
Univ of Washington, Seattle, US
Background: Syphilis is endemic in regions with the
highest HIV prevalence. As many as 30% of patients with syphilis have
symptomatic neurosyphilis. The cerebrospinal fluid (CSF) -VDRL is specific for
neurosyphilis diagnosis, and the CSF-FTA-ABS may be sensitive. Both tests
require equipment that may not be available where neurosyphilis is most
prevalent. Point-of-care tests could facilitate neurosyphilis diagnosis.
Methods: CSF was collected from 173 HIV-infected
patients with syphilis. CSF-VDRL (n = 173) and CSF-FTA-ABS (n = 110) were
performed using standard methods. CSF-RPR (n = 100) was performed as for
CSF-VDRL. The Syphicheck immunochromatographic strip test (ICS) (n = 145) was
performed according to the manufacturers’ instructions except that 90 µL CSF
and 52 µL buffer were used. CSF-VDRL was used as the laboratory gold standard
and symptomatic neurosyphilis (eye disease or meningitis) was used as the
clinical gold standard for neurosyphilis diagnosis.
Results: Of 173 subjects, 106 had early
syphilis and 67 had late latent syphilis; 47 (27%) had a reactive CSF-VDRL and
60 (35%) had symptomatic neurosyphilis.
|
Comparison Test
|
Gold Standard for Diagnosis
|
|
CSF-RPR
|
CSF-VDRL
|
Symptomatic Neurosyphilis
|
|
Sensitivity, %
|
75
|
40
|
|
Specificity, %
|
97
|
85
|
|
CSF-VDRL
|
|
|
|
Sensitivity, %
|
|
48
|
|
Specificity, %
|
|
82
|
Agreement between CSF-RPR and CSF-VDRL was substantial (k = 0.76), but in 7 instances, the CSF-RPR
was negative when the CSF-VDRL was positive. In 2 cases, the CSF-RPR was
positive when the CSF-VDRL was negative. CSF-RPR was always positive when
CSF-VDRL titer was >1:2.
|
Comparison Test
|
Gold Standard for Diagnosis
|
|
CSF-ICS
|
CSF-VDRL
|
Symptomatic Neurosyphilis
|
|
Sensitivity, %
|
72
|
41
|
|
Specificity, %
|
78
|
72
|
|
CSF-FTA-ABS
|
|
|
|
Sensitivity, %
|
91
|
52
|
|
Specificity, %
|
83
|
75
|
The CSF-ICS was less sensitive than the CSF-FTA-ABS using
CSF-VDRL as the gold standard. Sensitivities were low and comparable for the 2
tests using the clinical gold standard. When the ICS was used with a 4-fold CSF
dilution, specificity increased to 92% and 79% using the same gold standards,
at the expense of sensitivity.
Conclusions: A reactive CSF-RPR establishes the
diagnosis of neurosyphilis with a high degree of certainty, but a negative
result does not rule out the diagnosis. Using a clinical gold standard, the
specificity of ICS using a 1:4 CSF dilution is comparable to the CSF-VDRL, but
offers no advantage over the CSF-RPR as a point-of-care test in this
HIV-infected population.
|
