Antimicrobial susceptibility of gonococci isolated in the Central African Republic

Bulletin of the World Health Organization, 60 (6): 959-963 (1982) Antimicrobial susceptibility of gonococci isolated in the Central African Republic R. WIDY-WIRSKI,' J. D'COSTA,' J. BIDDLE,2 & S. BROWN3 Using agar dilution techniques, we determined the minimum inhibitory concentrations (MIC) of 11 antimicrobials for 70 isolates of Neisseria gonorrhoeae obtained in Bangui, CentralAfrican Republic. Thesegonococci werefound to befairlysusceptible to commonly used antibiotics: only 3 isolates (4%) had a penicillin MIC > 1.0 pg/ml and 6 (9 1%) had a tetracycline MIC > 2. 0 ug/ml. With regard to other antibiotics, 54 isolates (77%) had an erythromycin MIC > 0.25 /tg/ml, all had a spectinomycin MIC < 16 sAg/ml, and 32 (46%) had a sulfamethoxazole/trimethoprim MIC > 9.5/0.5 Ag/ml. None of these isolates produced penicillinase. This study has helped to provide a rational basisfor establishing national gonorrhoea treatment recommendations in the Central African Republic. Gonorrhoea is one of the most common communi- cable diseases in the world, producing serious compli- cations in 10-20% of infected women, including pelvic inflammatory disease (PID) with pelvic abscess, ectopic pregnancy, and infertility (1). The complications of gonococcal infection in men may be equally common if the disease remains untreated, and include epididymitis, urethral stricture, and infertility. The social and economic consequences of gonor- rhoea may be greater for many countries of Africa than elsewhere. Infertility is a major problem in many areas (2), and in several countries, 25-4407o of hospitalized women have PID (3, 4). Urethral stric- tures are so common in some countries that special clinics have been established for their treatment. The incidence of gonorrhoea in the Central African Republic is unknown, but surveys conducted in various population groups throughout the country have all identified a high gonococcal prevalence (5). In Bangui, 327o of men attending the Sexually Trans- mitted Diseases clinic, 31 /o of male university students, 17% of schoolboys, and 9.5% of women attending the antenatal clinic were found to be infected with gonorrhoea. Furthermore, in studies carried out in rural areas, 37.5% of male villagers, 29% of male agricultural workers, 24% of Pygmy ' Bacterial and Venereal Infections Intercountry Team, World Health Organization, Bangui, Central African Republic. 2 STD Research Laboratory, Center for Infectious Diseases, Centers for Disease Control, Atlanta, GA, USA. 3 Coordinator for International VD Control Activities, Venereal Disease Control Division, Centers for Disease Control, Atlanta, GA 30333, USA. Requests for reprints should be addressed to this author. men, and 19% of male forest workers were found to be infected. There are no national recommendations for treat- ment of gonococcal infection in the Central African Republic; consequently, many patients receive expen- sive but ineffective therapy (R. Widy-Wirski, unpub- lished observations, 1977-81). As a preliminary basis for establishing such recommendations, we deter- mined the in vitro susceptibility to antimicrobials of gonococci isolated in Bangui and used these to estimate the efficacy of various treatment regimens. MATERIALS AND METHODS Between September 1979 and February 1980, selected specimens of urethral discharge or endo- cervical material were obtained and inoculated onto modified Thayer-Martin (MTM) medium within 3 h of collection. The plates were then placed in candle jars and incubated at 36 ± 1 'C. The plates were inspected for Neisseria gonorrhoeae after 24 and 48 h, and oxidase-positive colonies with typical morphology and containing typical Gram-negative diplococci were presumed to be N. gonorrhoeae. All isolates were subcultured and tested for 3-lactamase production, using the rapid iodometric test (1). Some strains were lyophilized, while others were suspended in trypticase soybroth (TSB) containing 200 ml of glycerol per litre and frozen in an ice-alcohol bath in a vacuum flask. The specimens were then transported by mail or by hand to the Centers for Disease Control (CDC), Atlanta, Georgia. At the CDC, organisms were identified as N. gonor- 420 -959- 960 R. WIDY-WIRSKI ET AL. rhoeae by their ability to utilize glucose, but not sucrose, lactose, or maltose. The Nitrocefin testa for penicillinase production was performed on all iso- lates. Susceptibility testing was performed as described previously, using agar dilution techniques (6). Plates containing doubling dilutions of anti- biotics were used to determine minimum inhibitory concentration (MIC) values. The range of antibiotic concentrations for each drug were: penicillin, ampicillin, and amoxicillin, 0.008-64 ,g/ml; tetra- cycline, 0.06-4 jg/ml; chloramphenicol, 0.25-16 ug/ ml; spectinomycin, 2-32 jig/ml; erythromycin, 0.015-2 ug/ml; gentamicin, 0.25-16 ,g/ml; cefox- itin, 0.25-8 tzg/ml; cefuroxime, 0.015-2 ytg/ml. Special media (free of folate and folate derivatives) were used to test the co-trimoxazole (sulfameth- oxazole/trimethoprim) susceptibility (7). Co-trim- oxazole concentrations, tested using a fixed 19:1 ratio of sulfamethoxazole (SMZ) to trimethoprim (TMP), ranged from SMZ:TMP 0.6:0.03 to 38:2 zg/ml. Test plates were spot inoculated with approximately 104 organisms from an 18-h culture, using a Steer's replicator device. Inoculated plates were incubated for 24 h in candle extinction jars at a BBL Microbiology Systems, Cockeysville, MD, USA. 35 'C. The MIC was taken as the lowest concen- tration of antibiotic that permitted the growth of no more than one colony. Control strains were included in each susceptibility test. RESULTS The recovery rate of isolates varied for the different transportation methods used. Only 14 (29%) of 48 lyophilized specimens mailed from the Central Afri- can Republic were recovered at CDC. Of 38 ly- ophilized isolates hand-carried to CDC, 37 (977o) survived, and 19 of 25 (76Go) preserved in TSB- glycerol and maintained in an ice-alcohol bath also survived hand delivery. All 70 of the presumptively identified N. gonor- rhoeae were confirmed by carbohydrate utilization tests. None of the 70 isolates produced 3-lactamase. Penicillin, ampicillin, and amoxicillin susceptibilities showed a high correlation; susceptibility to other anti- microbials showed lower correlation with that to peni- cillin. The distribution of susceptibility to the various antimicrobials is shown in Table 1. Table 1. Antimicrobial MICs for 70 isolates of N. gonorrhoeae from Bangui, 1979-80 MIC (yg/ml) Antimicrobial S 0.008 0.015 0.03 0.06 0.125 0.25 0.5 1 2 4 8 12 16 32 Penicillin 9 7 6 5 4 20 16 2 1 0 0 Ampicillin 3 7 14 13 21 12 0 0 Amoxicillina 0 2 12 12 19 22 2 0 0 Tetracycline 0 0 2 9 9 12 32 4 2 0 0 Erythromycin 0 0 5 11 16 19 17 2 0 0 Gentamicin 0 0 4 19 47 0 0 Spectinomycin 0 0 1 1 19 34 15 0 Cefoxitin 0 0 3 13 13 23 15 3 0 0 Cefuroxime 0 27 7 25 8 3 0 0 Chloramphenicol 0 0 11 12 8 16 20 3 0 0 MIC (ILg/ml) 0.6: 1.2: 2.4: 4.8: 9.5: 0.03 0.06 0.125 0.25 0.5 19:1 38:2 Co-trimoxazole 7 3 5 23 24 8 0 a 69 strains tested. GONOCOCCAL SUSCEPTIBILITY IN CENTRAL AFRICAN REPUBLIC Detailed information is available from studies carried out in the United States of America on the effi- cacy of several treatment regimens at various levels of in vitro susceptibility (8-10) (Table 2). Using this information and the observed MIC of the 70 gono- coccal isolates, we calculated the theoretical efficacy of seven antimicrobial regimens in the treatment of gonorrhoea in the Central African Republic (Table 3). It was found that 3 regimens (4.8 x 106 units of procaine penicillin plus 1 g of probenecid, 10 g of tetracycline, and 2 g of spectinomycin) would cure more than 95% of patients; two regimens (3.5 g of ampicillin plus 1 g of probenecid and 27 tablets of co- trimoxazole) would cure 90-95% of patients; and two regimens (2.4 x 106 units of procaine penicillin for 2 days and 10 g of erythromycin) would cure 82% or less of our patients. DISCUSSION Gonorrhoea is an important health problem in many countries of Africa. A careful description of the disease problem is required for the development of feasible strategies to prevent disease transmission and minimize complications by providing effective treat- ment to patients and their sexual partners. Many patients with gonorrhoea already receive treatment, but the existing treatment system could be made more efficient by (a) defining criteria for treatment that can be adopted by all health care personnel; (b) defining effective treatment regimens that are available and acceptable to all health workers; and (c) encouraging the referral and treatment of sexual partners of treated cases. This study formed part of an effort to implement appropriate treatment strategies in the Central Afri- can Republic. Similar studies on a larger number of isolates, collected under more carefully defined conditions, are needed to ensure that these findings represent the true distribution of gonococcal suscepti- bility in the country. However, the present study has been useful in establishing preliminary national treatment recommendations, since it showed that gonorrhoea in the Central African Republic probably requires larger doses of antimicrobial drugs than in European countries but lower doses than in south- east Asian countries. Ineffective antimicrobial regimens for gonorrhoea Table 2. Therapeutic efficacy of various antibiotic regimens for gonorrhoea, according to the MIC for the infecting organism. These data are taken from studies carried out in the USA, 1972-79 Antibiotic regimen Reference MIC" (yg/ml) No. tested No. failed Failure rate 1%) Procaine penicillin, 6 0.125 1739 43 2.5 4.8 x 106 units, (8) 0.25-0.5 658 32 4.9 plus 1 g of probenecid > 1.0 91 10 11 Procaine penicillin, < 0.06 110 2 1.8 2.4x 106 units (8) 0.125 29 2 6.9 daily for 2 days > 0.25 45 14 31 Ampicillin, 3.5 g, < 0.06 728 19 2.6 plus 1 gofprobenecid (8) 0.125-0.25 650 56 8.6 orally .0.5 59 9 15.25 Tetracycline, 500 mg, .0.5 329 3 0.9 4 times daily for (8) 1.0 72 2 2.8 5 days > 2.0 15 2 13 Erythromycin, 500 mg, .0.125 29 1 3.4 4 times daily for (9) 0.25-0.5 67 13 19 5 days .1.0 24 17 71 Spectinomycin, S16 847 39 4.6 2 g (8) .32 4 1 25 Co-trimoxazole, 9 tablets .4.8:0.25 305 10 3.3 once daily for 3 days (10) > 9.5:0.5 114 13 11.4 a The MIC groups were established arbitrarily on the basis of regimen efficacy. 961 R. WIDY-WIRSKI ET AL. Table 3. Calculation of the failure rate of various antibiotic regimens in the treatment of gonorrhoea in the Central African Republic MIC No. of Proportion Predicted number Failure rate Regimena group' isolates of failures of failures (%) Procaine Low 31 0.025 0.78 penicillin Middle 36 0.049 1.76 (4.8 x 1 6 units) High 3 0.11 0.33 Total 70 2.87 4.1 Procaine Low 27 0.018 0.49 penicillin Middle 4 0.069 0.28 (2.4x 106 units) High 39 0.31 12.09 Total 70 12.86 18.4 Ampicillin Low 37 0.026 0.96 Middle 33 0.086 2.84 High 0 0.1525 0 Total 70 3.8 5.4 Tetracycline Low 32 0.009 0.29 Middle 32 0.028 0.90 High 6 0.13 0.78 Total 70 1.97 2.8 Erythromycin Low 16 0.034 0.54 Middle 35 0.19 6.65 High 19 0.71 13.49 Total 70 20.68 29.5 Spectinomycin Low 70 0.046 3.2 High 0 0.25 0 Total 70 3.2 4.6 Co-trimoxazole Low 38 0.033 1.25 High 32 0.114 3.65 Total 70 4.90 7.0 a See Table 2 for details. select for progressively less susceptible N. gonor- rhoeae, thus compounding an already difficult con- trol problem for many African countries. Conversely, three reports have described the greater susceptibility of gonococci to antibiotics after highly effective treat- ment (> 9407 cures) for gonorrhoea had been imple- mented (11-13). It is hoped to monitor gonococcal susceptibility in the Central African Republic in future years, in order to evaluate changes occurring after implementation of the national treatment recommendations. RESUME SENSIBILITP AUX MEDICAMENTS ANTIMICROBIENS DES GONOCOQUES ISOLES EN REPUBLIQUE CENTRAFRICAINE A I'aide des techniques de dilution en gelose, nous avons determine les concentrations minimales inhibitrices (CMI) de 11 medicaments antimicrobiens pour 70 isolements de Neisseria gonorrhoeae obtenus a Bangui, Republique centrafricaine. Ces gonocoques sont relativement sensibles aux antibiotiques d'usage courant: pour trois isolements seulement (407o), la CMI de la penicilline etait > 1,0 Ag/ml et pour six (9%), la CMI de la tetracycline etait > 2,0 tg/ml. En ce qui concerne d'autres antibiotiques, la CMI de l'erythromycine etait > 0,25 ml pour 54 isolements (77%); la CMI de la spectinomycine etait < 16 yg/ml pour tous les isolements, et la CMI de l'association sulfa- methoxazole/trimethoprime etait > 9,5/0,5 ug/ml pour 32 (46%). Aucun de ces isolements de gonocoques ne produi- 962 GONOCOCCAL SUSCEPTIBILITY IN CENTRAL AFRICAN REPUBLIC 963 sait de penicillinase. Sur la base des donnees relatives a 1'efficacite des schemas therapeutiques a divers niveaux de sensibilite in vitro et des CMI observees pour nos isolements, nous avons calcule l'efficacite theorique de sept posologies antimicrobiennes pour le traitement de la gonococcie en Republique centrafri- caine. Trois schemas therapeutiques doivent guerir plus de 95% des malades: 4,8 x 106 unites de penicilline procaine plus I g de probenecide, 2 g de tetracycline quotidienne- ment en doses fractionnees pendant cinq jours, et 2 g de spectinomycine. Deux schemas doivent permettre de guerir 90 a 95% des malades: 3,5 g d'ampicilline plus 1 g probe- necide et 9 comprimes de co-trimoxazole en une dose quotidienne unique pendant trois jours. Deux schemas therapeutiques devraient guerir 82% ou moins de nos malades: 2,4 x 106 unites de penicilline procaine quoti- diennement pendant deux jours et 2 g d'erythromycine quotidiennement, en doses fractionnees, pendant cinq jours. Cette etude a contribue a fournir une base rationnelle en vue de la formulation de recommandations nationales pour le traitement de la gonococcie en Republique centrafri- caine. REFERENCES 1. WHO Technical Report Series, No. 616, 1978. (Neis- seria gonorrhoeae and gonococcal infections: report of a WHO Scientific Group). 2. MUIR, D. G. & BELSEY, M. A. Pelvic inflammatory disease and its consequences in the developing world. American journal of obstetrics and gynecology. 138: 913-928 (1980). 3. NSANZE, H. Problems and approaches in the surveil- lance and control of sexually transmitted agents associ- ated with pelvic inflammatory disease in Africa. American journal of obstetrics and gynecology, 138: 1088-1090 (1980). 4. OSOBA, A. 0. Microbiologic techniques for the diag- nosis of pelvic inflammatory disease in developing countries. American journal of obstetrics and gynecology, 138: 1091-1095 (1980). 5. WIDY-WIRSKI, R. & D'COSTA, J. Gonococcal surveys in Central African Republic. 13 Conference technique de I'OCEAC, Yaounde, 1980. Yaounde, OCEAC, 1980. 6. JAFFE, H. W. ET AL. National gonorrhea therapy moni- toring study. In vitro antibiotic susceptibility and its correlation with treatment results. NewEnglandjournal of medicine, 294: 5-9 (1976). 7. ELLIOTT, W. C. ET AL. Treatment of gonorrhea with tri- methoprim-sulfamethoxazole. Journal of infectious diseases, 135: 939-943 (1977). 8. GUINAN, M. E. ET AL. The national gonorrhea therapy monitoring study: I. Review of treatment results and in vitro antibiotic susceptibility, 1972-1978. Sexually transmitted diseases, 6: 93-102 (1979) 9. BROWN, S. T. ET AL. Therapy of gonorrhea. Comparison of erythromycin base and estolate. Journal of the American Medical Association, 238: 1371-1373 (1977). 10. BROWN, S. T. ET AL. Treatment of uncomplicated gonococcal infection with trimethoprim/sulfameth- oxazole. Sexually transmitted diseases, 9: 9-14 (1982). 11. JAFFE, H. W. ET AL. Trends and seasonality of anti- biotic resistance of Neisseria gonorrhoeae. Journal of infectious diseases; 136: 684-688 (1977). 12. OLSEN, G. A. Consumption of antibiotics in Green- land, 1964-70. IV. Changes in the sensitivity of N. gonorrhoeae to antibiotics. British journal of venereal diseases, 49: 33-41 (1973). 13. EVANS, A. J. ET AL. Prolonged use of the Greenland method of treatment of gonorrhea. British journal of venereal diseases, 56: 88-91 (1980).