International Journal of STD & AIDS 1992: 3: 273-277
ORIGINAL ARTICLE
Strain characteristics and antibiotic susceptibility of isolates of Neisseria gonorrhoeae causing disseminated gonococcal infection in Australia J W Tapsall
FRCPA1, E A PhillifS MAppSci 1, T R Shultz BAppSci\ B Way FRCPA2, K Withnall BSc and Members of the Australian
Gonococcal Surveillance Programme* lDepartment of Microbiology, The Prince of Wales Hospital, Randwick, NSW and Department of Pathology, Royal Darwin Hospital, Casuarina, NT, Australia Summary: The auxotype (A) and serovar (S) distribution and antibiotic and serum sensitivity of 22 strains of Neisseria gonorrhoeae isolated from blood and joints were determined. With one exception, these strains from disseminated gonococcal infections (DGI) belonged to one of 4 serovars of the IA serogroup and were resistant to killing by normal human serum. The auxotype distribution of these Australian strains differed significantly from that reported elsewhere in that 17 of the 22 isolates were proline requirers, but none were of the AHU auxotype. This lack of the AHU auxotype in the DGI strains in Australia was explained by the virtual absence of AHU requirers in a sample of 1560 mucosal strains isolated in Sydney and Darwin from 1987 to 1990. The A/S distribution of these mucosal isolates also helped to account for the low (0.12) percentage of DGI strains in isolates examined by the Australian Gonococcal Surveillance Programme (AGSP) from 1981 to 1991, and the differences in the rates of DGI in Sydney (0.08%) and Darwin (0.87%). There was a relative lack of the IA serogroup strains which are mostly responsible for DGI in the mucosal isolates from Sydney (15%of all strains) but a higher proportion of these serovars (40%) in the Darwin sample. There were 46 cases of DGI in data from the AGSP, 29 of these being women. Seven of the cases diagnosed in Australia were infected with penicillinase-producing gonococci suggesting that antibiotics other than the penicillins should now be used for this condition in this region. Keywords: Neisseria gonorrhoeae; disseminated gonococcal infection; PPNG; antibiotic susceptibility; auxotype; serotype
INTRODUCTION It has been estimated that gonococci may dis-
seminate from between 0.5 and 2% of mucosal infectionsl-I to cause systemic disease manifested most often by skin lesions, tenosynovitis and septic Correspondence to: Dr J W Tapsall, Department of Microbiology, The Prince of Wales Hospital, Randwick, NSW, Australia 2031 *Other current and past members of the AGSP: L M Morris, The Prince of Wales Hospital, Sydney; J R Bates and Y M Cossins, Laboratory of Pathology & Microbiology, Brisbane; R Mallon, C Bieri, J Macrae and C Macleod, Department of Microbiology, ICPMR, Westmead; J Bell and J Raby, Woden Valley Hospital, Canberra; J R L Forsyth, P W Traynor and M M Peel, Microbiological Diagnostic Unit, University of Melbourne; R Lumb and G Handke, Venereal Diseases Laboratory of the IMVS Adelaide; M Blums, G Nicholson and A Ismail, Venereal Diseases Branch Laboratory, State Health Services, Perth; R Tucker and M Gardam, Royal Hobart Hospital.
arthritis and rarely, endocarditis and merungitis", While some patients with disseminated gonococcal infection (DGI) may have a deficiency in late-acting complement components3, the strains of Neisseria gonorrhoeae involved in DGI h~v~ been ~oted to have a number of special characteristics WhICh appear to influence the incidence of this disease. Studies in North America have linked strains of the argininehypoxanthine-uracil (AHU) auxotype to DGI4, although this association is not so pronounced or else absent in other geographic areas5,6. Additionally, strains causing DGI, especially those of the AHU auxotype, have for many years been regarded as highly susceptible to penicillin and resistant to the bactericidal action of serum". Although increased antibiotic resistance in gonococci causing DGI has been reported? and substantial changes in penicillin susceptibility of gonococci isolated in Australia have been documented'U'', there has
Downloaded from std.sagepub.com at CORNELL UNIV on July 16, 2016
274
International Journal of STD & AIDS Volume 3 July/August 1992
been no systematic study of the antibiotic susceptibilities or other characteristics of gonococci causing DGI in Australia. Recent comments'! suggest that the traditional view of the susceptibility of gonococci causing DGI is, understandably, still held by some in Australia and is used as a basis for recommending antibiotic therapy in this disease. For some time the results of determinations of two unrelated characteristics of gonococci, the auxotype (A) and serovar (S), have been used in epidemiological and other studies of gonococcal diseases. An auxotype comprises a subgroup of gonococci with a stable requirement for one or more of a number of specified nutrients and a serovar is determined by the pattern of reactivity of a strain with monoclonal antibodies to epitopes on the protein I of the organism. These two characteristics, when combined, form an auxotype/serovar (A/S) class. This report examines the auxotype and serovar distribution and antibiotic and serum susceptibility of 22 strains of N. gonorrhoeae isolated from cases of DGI. Comparisons are made with the results of overseas studies on the distribution of auxotypes amongst gonococci causing DGI and the patterns of auxotypes and serovars in isolates from DGI are also compared with that seen in strains from a sample of mucosal infections. Data on the antibiotic sensitivity of 18 of the 22 isolates together with information from 28 further cases from the Australian Gonococcal Surveillance Pregramme8,9 are used to review the basis of treatment regimens appropriate to this infection in Australia. MATERIALS AND METHODS Twenty-two strains of N. gonorrhoeae isolated from either blood cultures or joint aspirates and identified by standard methods between 1985 and 1991 were available for examination. Nine of the strains were isolated in Darwin, two in Queensland and four in Papua New Guinea. The remaining strains were isolated in Sydney. One of these patients was originally infected in Papua New Guinea, another in Bangkok and a third in the Philippines while a fourth was the immediate contact of a patient infected in Bangkok (Table 1). All isolates were auxotYEed by the methods of La Scolea and Young 2 using control strains which included those of the AHU auxotype. Serotyping was performed using a panel of 12 monoclonal antibodies to epitopes on protein I (Syva, Palo Alto, USA, kindly supplied by Dr J R L Forsyth, University of Melbourne) using the nomenclature of Knapp'P. Each strain was tested at least twice. Lactamase production was determined by acidometric techniques and antibiotic sensitivities to penicillin, spectinomycin, ceftriaxone and ciprofloxacin by agar plate dilution techniques using an inoculum of 1()4 cfu and lsosensitest agar (Oxoid) containing 8% saponin-lysed horse blood". The serum bactericidal test was performed using a standardized mlcrotechntque>. The gonococcal
isolates examined in this test were all 22 DGIcausing strains and mucosal isolates of the same (n =16) and different (n =19) serovars as the DGI causing strains. Resistance to killing by fresh human serum was defined as growth of more than 50 cfu from a standard inoculum from a well containing a serum dilution of 1: 4 or less. Strains killed by fresh human serum diluted 1: 8 or more were considered serum sensitive". A further 947 strains isolated in Sydney between 1987 and 1990 and 613 isolates cultured in Darwin between 1988 and 1990 were serotyped and auxotyped by the above methods. The records of the Australian Gonococcal Surveillance Programme were reviewed over the period 1981-1991 for additional cases of OGI. For these cases, information on the sex of the patient and penicillin sensitivity were available as well as the site of isolation. RESULTS Auxotype and serovar distribution and antibiotic and serum susceptibility of 22 DGI isolates All but one of the isolates belonged to one of four IA serovars and there was a single isolate of a IB serovar (Table 1). Seventeen of the 22 strains were proline requirers, but no strains of the AHU auxotype were isolated. Six of the proline auxo. trophs showed an additional requirement for isoleucine and three of these for serine as well. Although a requirement for isoleucine or serine can be variably expressed, reproducible results were obtained for these strains on repeat testing. Four of these isolates were fully sensitive to penicillin (minimal inhibitory concentration-MIC
ORIGINAL ARTICLE
Strain characteristics and antibiotic susceptibility of isolates of Neisseria gonorrhoeae causing disseminated gonococcal infection in Australia J W Tapsall
FRCPA1, E A PhillifS MAppSci 1, T R Shultz BAppSci\ B Way FRCPA2, K Withnall BSc and Members of the Australian
Gonococcal Surveillance Programme* lDepartment of Microbiology, The Prince of Wales Hospital, Randwick, NSW and Department of Pathology, Royal Darwin Hospital, Casuarina, NT, Australia Summary: The auxotype (A) and serovar (S) distribution and antibiotic and serum sensitivity of 22 strains of Neisseria gonorrhoeae isolated from blood and joints were determined. With one exception, these strains from disseminated gonococcal infections (DGI) belonged to one of 4 serovars of the IA serogroup and were resistant to killing by normal human serum. The auxotype distribution of these Australian strains differed significantly from that reported elsewhere in that 17 of the 22 isolates were proline requirers, but none were of the AHU auxotype. This lack of the AHU auxotype in the DGI strains in Australia was explained by the virtual absence of AHU requirers in a sample of 1560 mucosal strains isolated in Sydney and Darwin from 1987 to 1990. The A/S distribution of these mucosal isolates also helped to account for the low (0.12) percentage of DGI strains in isolates examined by the Australian Gonococcal Surveillance Programme (AGSP) from 1981 to 1991, and the differences in the rates of DGI in Sydney (0.08%) and Darwin (0.87%). There was a relative lack of the IA serogroup strains which are mostly responsible for DGI in the mucosal isolates from Sydney (15%of all strains) but a higher proportion of these serovars (40%) in the Darwin sample. There were 46 cases of DGI in data from the AGSP, 29 of these being women. Seven of the cases diagnosed in Australia were infected with penicillinase-producing gonococci suggesting that antibiotics other than the penicillins should now be used for this condition in this region. Keywords: Neisseria gonorrhoeae; disseminated gonococcal infection; PPNG; antibiotic susceptibility; auxotype; serotype
INTRODUCTION It has been estimated that gonococci may dis-
seminate from between 0.5 and 2% of mucosal infectionsl-I to cause systemic disease manifested most often by skin lesions, tenosynovitis and septic Correspondence to: Dr J W Tapsall, Department of Microbiology, The Prince of Wales Hospital, Randwick, NSW, Australia 2031 *Other current and past members of the AGSP: L M Morris, The Prince of Wales Hospital, Sydney; J R Bates and Y M Cossins, Laboratory of Pathology & Microbiology, Brisbane; R Mallon, C Bieri, J Macrae and C Macleod, Department of Microbiology, ICPMR, Westmead; J Bell and J Raby, Woden Valley Hospital, Canberra; J R L Forsyth, P W Traynor and M M Peel, Microbiological Diagnostic Unit, University of Melbourne; R Lumb and G Handke, Venereal Diseases Laboratory of the IMVS Adelaide; M Blums, G Nicholson and A Ismail, Venereal Diseases Branch Laboratory, State Health Services, Perth; R Tucker and M Gardam, Royal Hobart Hospital.
arthritis and rarely, endocarditis and merungitis", While some patients with disseminated gonococcal infection (DGI) may have a deficiency in late-acting complement components3, the strains of Neisseria gonorrhoeae involved in DGI h~v~ been ~oted to have a number of special characteristics WhICh appear to influence the incidence of this disease. Studies in North America have linked strains of the argininehypoxanthine-uracil (AHU) auxotype to DGI4, although this association is not so pronounced or else absent in other geographic areas5,6. Additionally, strains causing DGI, especially those of the AHU auxotype, have for many years been regarded as highly susceptible to penicillin and resistant to the bactericidal action of serum". Although increased antibiotic resistance in gonococci causing DGI has been reported? and substantial changes in penicillin susceptibility of gonococci isolated in Australia have been documented'U'', there has
Downloaded from std.sagepub.com at CORNELL UNIV on July 16, 2016
274
International Journal of STD & AIDS Volume 3 July/August 1992
been no systematic study of the antibiotic susceptibilities or other characteristics of gonococci causing DGI in Australia. Recent comments'! suggest that the traditional view of the susceptibility of gonococci causing DGI is, understandably, still held by some in Australia and is used as a basis for recommending antibiotic therapy in this disease. For some time the results of determinations of two unrelated characteristics of gonococci, the auxotype (A) and serovar (S), have been used in epidemiological and other studies of gonococcal diseases. An auxotype comprises a subgroup of gonococci with a stable requirement for one or more of a number of specified nutrients and a serovar is determined by the pattern of reactivity of a strain with monoclonal antibodies to epitopes on the protein I of the organism. These two characteristics, when combined, form an auxotype/serovar (A/S) class. This report examines the auxotype and serovar distribution and antibiotic and serum susceptibility of 22 strains of N. gonorrhoeae isolated from cases of DGI. Comparisons are made with the results of overseas studies on the distribution of auxotypes amongst gonococci causing DGI and the patterns of auxotypes and serovars in isolates from DGI are also compared with that seen in strains from a sample of mucosal infections. Data on the antibiotic sensitivity of 18 of the 22 isolates together with information from 28 further cases from the Australian Gonococcal Surveillance Pregramme8,9 are used to review the basis of treatment regimens appropriate to this infection in Australia. MATERIALS AND METHODS Twenty-two strains of N. gonorrhoeae isolated from either blood cultures or joint aspirates and identified by standard methods between 1985 and 1991 were available for examination. Nine of the strains were isolated in Darwin, two in Queensland and four in Papua New Guinea. The remaining strains were isolated in Sydney. One of these patients was originally infected in Papua New Guinea, another in Bangkok and a third in the Philippines while a fourth was the immediate contact of a patient infected in Bangkok (Table 1). All isolates were auxotYEed by the methods of La Scolea and Young 2 using control strains which included those of the AHU auxotype. Serotyping was performed using a panel of 12 monoclonal antibodies to epitopes on protein I (Syva, Palo Alto, USA, kindly supplied by Dr J R L Forsyth, University of Melbourne) using the nomenclature of Knapp'P. Each strain was tested at least twice. Lactamase production was determined by acidometric techniques and antibiotic sensitivities to penicillin, spectinomycin, ceftriaxone and ciprofloxacin by agar plate dilution techniques using an inoculum of 1()4 cfu and lsosensitest agar (Oxoid) containing 8% saponin-lysed horse blood". The serum bactericidal test was performed using a standardized mlcrotechntque>. The gonococcal
isolates examined in this test were all 22 DGIcausing strains and mucosal isolates of the same (n =16) and different (n =19) serovars as the DGI causing strains. Resistance to killing by fresh human serum was defined as growth of more than 50 cfu from a standard inoculum from a well containing a serum dilution of 1: 4 or less. Strains killed by fresh human serum diluted 1: 8 or more were considered serum sensitive". A further 947 strains isolated in Sydney between 1987 and 1990 and 613 isolates cultured in Darwin between 1988 and 1990 were serotyped and auxotyped by the above methods. The records of the Australian Gonococcal Surveillance Programme were reviewed over the period 1981-1991 for additional cases of OGI. For these cases, information on the sex of the patient and penicillin sensitivity were available as well as the site of isolation. RESULTS Auxotype and serovar distribution and antibiotic and serum susceptibility of 22 DGI isolates All but one of the isolates belonged to one of four IA serovars and there was a single isolate of a IB serovar (Table 1). Seventeen of the 22 strains were proline requirers, but no strains of the AHU auxotype were isolated. Six of the proline auxo. trophs showed an additional requirement for isoleucine and three of these for serine as well. Although a requirement for isoleucine or serine can be variably expressed, reproducible results were obtained for these strains on repeat testing. Four of these isolates were fully sensitive to penicillin (minimal inhibitory concentration-MIC
