Analysis of a New Algorithm for the Diagnosis of Syphilis among Patients with HIV Infection

Background: Syphilis is a substantial health problem especially among men who have sex with men (MSM). CDC guidelines recommend testing with a non treponemal test (VDRL/RPR), followed by a treponemal test (FTA, TPPA, EIA). Studies have shown that reverse algorithms for testing syphilis (reactive treponemal test followed by a non-treponemal test), have a higher efficacy in detecting syphilis. The study objective is to analyze the impact of the reverse algorithm for screening and diagnosis in HIV patients.

Methods: Retrospective chart review of HIV outpatients was conducted for the period 2012-16. 1265 charts were reviewed. Patients were screened by CMIA for syphilis. A negative CMIA was recorded as negative. Positive results were followed by RPR. A negative RPR was followed by FTA. Outcome variables recorded were: Sex, Race, Ethnicity, Age, HIV Viral Load, CD4 count, History of STIs, and HIV Risk Factor.

Patients were further stratified into 4 groups for statistical analysis:

Group 1:CMIA+, RPR (-), and FTA+ with no evidence of prior treatment for syphilis and received treatment syphilis; N=28.Group 2:CMIA+, RPR (-) and FTA+ and prior treatment for syphilis; N=111Group 3:CMIA+ and RPR+; N=157Group 4:(-); N=952 (17 patients false + and included in Group 4)Pearson Chi square tests (2-sided, p<.05) were conducted to determine group differences

Results:

In Group 1, 39.3% had detectable HIV viral load compared with 17% in the other groups (p=.002). Patients in Group 3 were more likely to be younger; with 36.9% being aged 50 or older vs 53.1% in the other groups (p<.0001), more likely to be male (93.6% vs 59.5%; p<.0001), and more likely to be MSM (77.1% vs 29.1%). Patients in Group 4 were less likely to be male (56.8% vs 86.4% other groups, p<.0001).

Conclusion: Our study shows that a reverse algorithm to serologically screen for syphilis produced higher sensitivity and specificity in diagnosis, similar to studies in non HIV patients. 91 patients received treatment for syphilis, of whom 28 (31%) would have been missed if screened by the traditional algorithm. Patients positive for syphilis are more likely to be male and MSM. It is unknown why patients in Group 1 were more likely to have a detectable viral load. Further research is needed on this subset of patients.