Introduction Syphilis reinfections are playing an increasing role in syphilis transmission in a number of populations. The assessment of reinfection and response to treatment depends on accurately measuring intraindividual changes in non-treponemal tests (delta-NTTs). In a 0 to 6 month delta-RPR determined by routine RPR testing (RT), samples would be tested 6 months apart with differences in reagent batches, environmental temperatures and observers all leading to measurement errors. We hypothesised that conducting paired RPR (PT) would enable a more accurate determination of delta-RPR than RT.

Methods 120 patients with a new diagnosis of syphilis were followed up at 0,3,6,9,12,18 and 24 months with RPRs performed via RT at each study visit and at any suspected reinfection. RPR PT was performed at 0 and 6 months and at any suspected reinfection.

Results The quantitative agreement +/-1 dilution among PT and RT was 97.4%. There was no difference in the proportion with an incomplete serological response at 6 months: 21 (19.4%) and 19 (17.6%) according to PT and RT, respectively (p=0.726). There was no statistically significant difference between 0 to 6 month delta-RPR as determined by PT and RT in predicting seroresponse at 12 months (86.1% and 91.6% agreement with 12 month classification, respectively, p=0.262. PT did not reduce the numbers of those classified with asymptomatic reinfections.

Conclusion In our setting routine PT is unlikely to be worth the considerable effort and cost it entails. Further research is required to assess its utility in specific circumstances.