(P82) CORRELATION OF RAPID PLASMA REAGIN (RPR) VALUES WITH ACTIVE SYPHILIS INFECTION IN A UNIVERSAL SCREENING SETTING

Background:
As syphilis screening programs expand across the country, efficient means to determine active syphilis status remains a challenge. Understanding the clinical utility of positive rapid plasma reagin (RPR) cutoff value may assist with targeting outreach efforts.

Methods:
We examined all cases with reactive quantitative RPR results over two years from an urban hospital emergency department with a universal syphilis screening program. Clinical history in the medical chart and past RPR values from the local health department were used as the gold standard to define cases of syphilis as active or previously treated. We assessed performance of this gold standard to the RPR value alone by calculating the accuracy (and 95% confidence interval), the sensitivity, the specificity and number of false positives and false negatives at various RPR cut points (>=1:1, >=1:2, >=1:4, >=1:8 and >=1:16).

Results:
Over the study period, there were 1,315 individuals with RPR >=1:1 of which 44.9% had active syphilis. RPR >=1:16 had the highest specificity and lowest sensitivity (88.1% and 54.7% respectively), and RPR>=1:1 had the highest sensitivity and lowest specificity (100% and 0.0% respectively). The best performing cut off was an RPR >=1:8, with an accuracy of 73.3%, 95% CI (70.8%, 75.7%), a sensitivity of 67.7%, and a specificity of 78.7%. At RPR >=1:8 we falsely identified 197 persons as active infections and fail to identify 154 active cases. Using an RPR>=1:1 versus an RPR >=1:8 requires reviewing the charts of 527 additional patients who are not active syphilis cases.

Conclusion:
With decreasing RPR cutoff values we saw increased detection of both true and false positives, highlighting the operational burden required for confirmation of low titer RPR results. Using an automatic determination of active cases at a RPR >=1:8 would reduce this burden but would also result in missed active cases of syphilis.