The Scottish Bacterial Sexually Transmitted Infections Reference Laboratory (SBSTIRL) provides universal surveillance data on antimicrobial resistance for all gonococci isolated in Scotland. All cultured organisms are tested against seven antibiotics using the agar dilution method1, and E-tests (AB Biodisk) where the minimum inhibitory concentration (MIC) exceeds the dilution series. Since April 2004 isolates have also been sequence typed using Neisseria gonorrhoeae multi-antigen sequence typing2.
Episodes of gonorrhoea in Scotland
A total of 926 episodes of gonococcal infection were reported by SBSTIRL in 2006. Antibiotic susceptibility and sequence typing was performed on isolates from 904 episodes (97.6%). Isolates were not available for 22 episodes which were confirmed as N. gonorrhoeae using a nucleic acid amplification test or NAAT (Aptima, Genprobe). Table 1 shows gonorrhoea culture-positive episodes and trends by gender. Between 2000 and 2006 there was a 6.6% increase in the overall level of gonorrhoea but there were differences in trends by gender. Episodes of infection in men increased by 4.1% between 2005 and 2006 and in women decreased by 12%. The increase in episodes of infection in men was mainly in the men who have sex with men (MSM) patient group3.
General antibiotic susceptibility trends
Table 2 provides the resistance trends for the antibiotics tested over the period 2001-2006 and Table 3 gives the pattern of resistance for isolates in 2006. As there is no standardised definition of resistance for cefixime, ceftriaxone and azithromycin, a decreased susceptibility category has been used. WHO recommends that once resistance to an antibiotic is greater than 5%, continued use of that antibiotic for empiric treatment should be reconsidered. Resistance to penicillin, tetracycline and ciprofloxacin exceeded 5%, but these antibiotics are not recommended for first line use. Use of ciprofloxacin was no longer recommended after 2003 when resistance was over 5% in most NHS board areas and all patient groups4.
In 2006 the number of isolates resistant to one or more antibiotics decreased to 46.0% from 49.2% in 2005. This was largely due to a decrease in chromosomal resistance to tetracycline but masked a large increase in ciprofloxacin resistance. Isolates with antibiotic resistance to two or more antibiotics increased significantly from 25.1% to 33.6% (p<0 .001). There was no significant change in the level of decreased susceptibility to azithromycin. There were no isolates with reduced susceptibility to cefixime or ceftriaxone or resistance to spectinomycin.
Penicillin and tetracycline resistance
Overall penicillin resistance (plasmid and chromosomal) increased significantly from 11.0% (99/903) in 2005 to 17.4% (157/904) in 2006 (p<0 .001). There is no comparable data from England and Wales for 2006. However, the increase in Scotland between 2005-6 mirrors that seen in England and Wales between 2004-5 where penicillin resistance increased from 11.4% to 17.9% (data from GRASP; gonococcal resistance to antimicrobials sentinel surveillance programme). Between 2005 and 2006 there was no significant change in the total plasmid mediated penicillin resistant isolates (8.1% vs 8.8%). Most (69/80, 86.3%) of the PPNG and PPNG/TRNG were resistant to ciprofloxacin (MIC > 1mg/l). Chromosomal resistance to penicillin increased from 2.9 % to 8.5% from 2005 to 2006.
Overall tetracycline resistance (plasmid and chromosomal) has decreased to 39.5% (357/904) in 2006 from 48.1% (434/903) in 2005 (p<0 .001). There was no significant change in the level of strains with plasmid mediated resistance. However, strains with chromosomal resistance to both penicillin and tetracycline increased significantly (p<0.001) from 2.9% to 8.2% from 2005 to 2006 and strains with chromosomal tetracycline resistance but no penicillin resistance have decreased significantly from 35.4% in 2005 to 21.9% in 2006 (p<0.001). Levels of tetracycline resistance in 2005 in England and Wales were very similar to those seen in Scotland (48% overall, 7.1% plasmid mediated).
Ciprofloxacin resistance
There was a significant increase in ciprofloxacin resistance from 23.6% in 2005 to 34.8% in 2006 (p<0 .001). Sequence typing data indicates that this was mainly attributable to a small number of commonly occurring strains (see below). Although the ciprofloxacin resistance for England and Wales is not available for 2006, the level was 21.7% in 2005.
Sequence type and antibiotic resistance
Sequence types represented by five or more isolates that were associated with antibiotic resistance categories are given in Table 4. The increase in ciprofloxacin resistance and chromosomally mediated resistance to penicillin was largely due to increased transmission of isolates of the sequence type ST225 which was the most commonly occurring sequence type in 2006. An increase in the number of isolates of ST147 also contributed to the increase in ciprofloxacin resistance in 2006. ST147 was the most common sequence type seen in 2004, declined in 2005 but has increased once more in 2006. Several sequence types associated with a resistance phenotype were present in 2006 that were not previously detected in 2005; of these ST1440 was represented by the most isolates (11) and is ciprofloxacin-resistant.
Overall, the antibiotic resistance category for isolates within a given ST is very consistent. Thus, sequence typing data could be used as a predictor of antibiotic resistance for common sequence types where identification by NAAT only would preclude antibiotic sensitivity determination5. However, this prediction should be made with caution since selective pressures may influence the resistance phenotype over time. The proportion of ciprofloxacin-resistant isolates in any given ST (Table 4) is extremely high (> 99%), which suggests that ciprofloxacin resistance determinants are maintained in strains even in the absence of widespread use of this antibiotic for treatment of gonorrhoea. In contrast, the proportion of isolates with reduced susceptibility to azithromycin within a given ST is more variable (MIC range 0.03 - >256mg/L amongst isolates of ST470). This is an exception to the general trend and might be attributable to selective pressure resulting from the widespread use of azithromycin to treat genital chlamydia infections. Although there is some variability in the levels of chromosomal tetracycline and penicillin resistance within ST225, the majority of isolates have an MIC one dilution either side of the resistance category threshold.
Table 5 shows the proportion of all resistant isolates that belong to unique sequence types (i.e. a sequence type that occurred only once). Chromosomal resistance to penicillin and tetracycline and reduced susceptibility to azithromycin have a very low proportion of unique sequence types suggesting that these forms of resistance are endemic. Chromosomal resistance to ciprofloxacin is also associated with a relatively low level of unique sequence types suggesting a high level of endemicity but with a greater importation of strains than the other forms of chromosomal resistance. Plasmid mediated resistance was associated with a high proportion of isolates that belonged to unique sequence types which is consistent with importation of these strains with little onwards transmission within the UK. Only one strain with plasmid-mediated resistance was spread within Scotland (ST368, see Table 4), predominantly within Tayside and Grampian.
Summary
Continued surveillance for antibiotic resistance is essential for guiding the choice of effective therapeutic regimens for gonorrhoea. In 2006 there was a further increase in ciprofloxacin resistance and resistance to one or more antibiotics remained high. No resistance was observed to the first line antibiotics cefixime and ceftriaxone. Sequence typing data demonstrated how the spread of the common sequence types influences the antibiotic resistance pattern. In the future, the widespread introduction of molecular testing will provide a challenge in maintaining comprehensive antibiotic susceptibility surveillance in Scotland, since both culture and non-culture diagnoses will be possible.
References
- Young H, Moyes A, Robertson DHH, et al. Gonococcal infection within Scotland: antigenic heterogeneity and antibiotic susceptibility of infecting strains. European Journal of Epidemiology 1990; 6:1-8.
- Martin IMC, Ison CA, Aanensen DM, Fenton KA, Spratt BG. Rapid sequence-based identification of gonococcal transmission clusters in a large metropolitan area. Journal of Infectious Diseases 2004; 189: 1479-1505.
- Wallace L, Young H, Codere G, Goldberg DJ. Genital herpes simplex, genital Chlamydia and gonorrhoea infection in Scotland: laboratory diagnoses 1997-2006. HPS Weekly Report 2007; 41:79-85.
- Young H, Palmer H, Winter A. Ciprofloxacin resistant gonorrhoea: the situation in Scotland and implications for therapy. SCIEH Weekly Report 2003; 37: 178-180.
- Young H, Palmer H. Can antibiotic resistance in Neisseria gonorrhoeae be predicted by molecular typing using NG-MAST? International Journal of STD & AIDs 2006;17 (supplement 1):27.