Background Before January 2010 at the Amsterdam STI clinic, Gram stain microscopy as syndromic test for urogenital gonorrhoea to enable direct treatment was performed in all high risk visitors irrespective of signs and/or symptoms. Since January 2010 Gram stain microscopy is performed only in those with signs and/or symptoms. We examined the diagnostic value of Gram stain microscopy as a syndromic test for urogenital gonorrhoea in high risk visitors with respect to signs and/or symptoms, by comparing the algorithm followed in the 2 periods.

Methods From January 2008–January 2010; 10,952 patients were included and from January 2010–January 2012 26,983. Urethral and cervical Gram stain results were compared with the standard gonorrhoea culture and sensitivity and specificity were calculated by sex. We estimated the rate of lost to follow up cases (visitors with a false negative Gram result who did not come back for treatment) in the 2 periods.

Results Gram strain microscopy had a high specificity, in both time periods regardless of sex (> = 99.8). Sensitivity decreased overall comparing years before and after 2010, from 87.2 to 84.8. This was mainly due to a sharp reduction of the already low sensitivity in women after 2010 from 32.0 to 23.1. Sensitivity in men remained high at 95.9 before 2010 and 95.4 thereafter. Loss to follow-up was estimated at 4.3% (2/47) before 2010 and 5.9% (8/135) after 2010.Overall, over 99% of all culture positive patients were treated.

Discussion The low sensitivity of Gram stain microscopy in females is known, but our results show that sensitivity to diagnose gonorrhoea is even lower when offered solely to symptomatic women. Although it did not have an impact on the number of treated infections, this counter-intuitive result could potentially lead to under-treatment and to be considered syndromic treatment policies.