Treatment of gonorrhoea in males in the Central African Republic with spectinomycin and procaine penicillin

Bulletin ofthe World Health Organization, 62 (1): 89-94 (1984) © World Health Organization 1984 Treatment of gonorrhoea in males in the Central African Republic with spectinomycin and procaine penicillin ANDRE MEHEUS,1 ROSLAW WIDY-WIRSKI,2 JOYE D'COSTA,2 EDDY VAN DYCK,3 RENt DELGADILLO,4 & PETER PIOT3 Gonorrhoea has become a problem in most parts of the world, and valid recommen- dations for treatment are important for control of the disease. In this study in Bangui, Central African Republic, 460 male patients with gonorrhoea were randomly assigned to treatment with either 4.oX 106 units ofprocaine penicillin plus I g ofprobenecid, or 2 g of spectinomycin. Of these patients, 91% returned for follow-up; the failure rate was 4.8%o with the penicillin schedule and 6.2% with spectinomycin (difference not statistically significant). Concomitant Chlamydia trachomatis infection wasfound in 5%0o ofpatients, and almost all of this group developed postgonococcal urethritis. Of the 460 patients, 7 (1.5%) were infected with penicillinase-producing Neisseria gonorrhoeae (PPNG) strains. Penicillin treatmentfailed in these cases, while spectinomycin was highly efficacious. Thefailure ratefor penicillin was considerably higher in infections with strains that were less sensitive to penicillin in vitro. Thefailure ratefor spectinomycin treatment was higher in patients who were infected with a strain that was highly sensitive to penicillin. It is concluded that, once PPNG strains have been found in a country, treatment of gonorrhoea should be based on an antibiotic that cures PPNG infections. Tetracycline can be used as second-line treatment, since it will also cure C. trachomatis infection, which is much less frequently associated with gonorrhoea in Africa than in industrial countries. Gonorrhoea is one of the most common communi- cable diseases in the Central African Republic. National treatment recommendations were first made in 1980 on the basis of a gonococcal susceptibility study (1). Taking into account estimated treatment efficacy and cost, three regimens were recommended: procaine penicillin, 4.8 x 106 units, plus 1 g of pro- benecid; ampicillin, 3.5 g, plus 1 g of probenecid; and tetracycline, 500 mg, 4 times daily for 5 days. The first two regimens have been widely used and have proved highly efficacious (2). Tetracycline is rarely administered, mainly because of the problem of patient compliance with a 5-day regimen. As was feared, the prevalence of penicillinase-producing Neisseria gonorrhoeae (PPNG) is increasing rapidly in Africa and PPNG strains have already been diag- nosed in several countries (2, 3); in the Central African Republic a PPNG strain was first diagnosed early in 1981. l Associate Professor, Epidemiology and Social Medicine, University of Antwerp (UIA), Universiteitsplein 1, 2610 Wilrijk, Belgium. 2 Bacterial and Venereal Infections Intercountry Team, World Health Organization, Bangui, Central African Republic. 3 Department of Microbiology, Institute of Tropical Medicine, Antwerp, Belgium. 4 Department of Ophthalmology, UIA, Wilrijk, Belgium. To examine the possible need for revision of the treatment recommendations, a large clinical trial of the treatment of uncomplicated gonorrhoea was undertaken, using two regimens: procaine penicillin, 4.0 x 106 units, intramuscularly, plus 1 g of pro- benecid by mouth; and spectinomycin, 2 g, intra- muscularly. The susceptibilities of N. gonorrhoeae to these (and three other) antibiotics were determined in vitro and the relationship between cure rate and susceptibility was studied. MATERIALS AND METHODS Male patients who attended the Centre for Sexually Transmitted Diseases in Bangui were considered for the study. Men with signs and symptoms of urethritis were eligible for the trial if they were not allergic to the drugs under study, had not received antimicrobial treatment within the preceding two weeks, and had no complication of the gonococcal infection. The ob- jectives and procedures of the trial were explained and only subjects who gave free and informed consent were included. 4381 -89- A. MEHEUS ET AL. A short history was taken from each patient. With a calcium alginate swab, a specimen of the urethral dis- charge was plated directly on to modified Thayer- Martin medium and another was spread on a slide for Gram staining. Plates were placed in a candle extinction jar and incubated at 36 'C. A blood sample was taken for VDRL (venereal disease re- search laboratory) and TPHA (Treponema pallidum haemagglutination assay) tests. Patients who showed Gram-negative intracellular diplococci were ran- domly assigned to one of two regimens: procaine penicillin, 4.0 x 106 units, intramuscularly, plus 1 g of probenecid by mouth, or spectinomycin hydro- chloride, 2 g, intramuscularly. (It was more con- venient to administer 4 x 106 units of procaine penicil- lin rather than the recommended 4.8 x 106 units, as this could be achieved by diluting one vial of the commercial brand available locally in 10 ml of physiological solution. This resulted in easier admin- istration of the drug with negligible effect on effi- cacy.) Patients were asked to bring their sexual con- tacts to the Centre for treatment, to abstain from any sexual activity, and to return for follow-up after 4-7 days. At follow-up a short history was again taken, a clinical examination was done, a culture and smear were taken from the urethra with calcium alginate swabs, and details were obtained about sexual exposure and side-effects of treatment. Postgono- coccal urethritis (PGU) was diagnosed if, at follow- up, gonorrhoea was excluded (smear and culture negative), but the Gram-stained smear showed 5 or more leukocytes in any microscopic field. Organisms were presumptively identified as N. gonorrhoeae if typical, oxidase-positive colonies containing Gram-negative diplococci grew after 24-48 hours' incubation in 4%0 CO2. Sugar fermen- tation tests were performed only in doubtful cases. All isolates were screened for penicillinase production by the cephalosporin test. Isolates were frozen at - 20 'C and sent in dry ice to the Microbiology Department of the Institute of Tropical Medicine in Antwerp, Belgium, where they were confirmed as N. gonorrhoeae by sugar utiliz- ation tests, rescreened for penicillinase production, and tested for antimicrobial susceptibility using the agar plate dilution technique. Minimal inhibitory concentrations (MICs) were determined for peni- cillin, thiamphenicol, tetracycline, kanamycin, and spectinomycin. Urethral swabs for the isolation of Chlamydia trachomatis were taken before treatment from 141 randomly selected patients; of these, 125 returned for follow-up and a further swab was taken. The speci- mens were transported to Antwerp in sucrose-phos- phate buffer transport medium in liquid nitrogen. Chlamydia were isolated using cycloheximide-treated McCoy cells (4). Only patients with a positive pretreatment culture for N. gonorrhoeae were included in the analysis. Patients with a persistent infection at follow-up were considered treatment failures and were given the alternative regimen (penicillin or spectinomycin). RESULTS In a 4-month period in 1981, a total of 460 patients entered the study, all with a pretreatment culture positive for N. gonorrhoeae (Table 1). Of these, 419 (91%o) returned for follow-up, with similar follow-up rates in both treatment groups. The mean age-of the 460 patients was 23.9 years; 49%o were students, 26% soldiers or policemen, and 25%o belonged to other professional groups. The failure rate was 4.8% with the procaine penicillin regimen and 6.2%o with spectinomycin. This dif- ference is not statistically significant (X2 = 0.4, P>0.5). Table 1. Results of treatment of uncomplicated gonorrhoea in males in Bangui, Central African Republic, with procaine penicillin and spectinomycin Returned for Treatment Post-gonococcal Treatment No. of patients follow-up failures urethritis Side-effects regimen in study No. % No. % No. % No. % Procaine penicillin, 4.0x 106 units, 232a 209a 90 job 4.8 51 24.4 16 7.7 plus 1 g of probenecid Spectinomycin, 2 g 228C 210C 92 13 6.2 43 20.5 6 2.9 ' 5 strains were PPNG. b 4 strains were PPNG. c 2 strains were PPNG. 90 TREATMENT OF GONORRHOEA IN AFRICA Two patients in each treatment group who were not cured admitted sexual re-exposure and may have been reinfected. The frequency of postgonococcal ureth- ritis (PGU) was not significantly different in the two groups (X2 = 0.93, P > 0.3). Only minor side-effects were seen, such as dizzi- ness and pain at the site of injection; they were more frequent with procaine penicillin (X2 = 4.85, P < 0.05). C. trachomatis was isolated from 7 patients (5%), 5 of whom had a positive culture before and after treatment. Of these 7 Chlamydia-positive cases, 6 (85.7%) developed postgonococcal urethritis (4 treated with penicillin and 2 treated with spec- tinomycin), compared with 15 of 118 (12.7/o) Chlamydia-negative cases (P < 0.001). Of460 patients, 7 were infected with a PPNG strain (1.50o). Five of these patients received procaine peni- cillin; 4 were not cured and 1 was lost to follow-up. The remaining 2 patients received spectinomycin treatment and were cured. The 10 patients who were not cured by penicillin (including 4 infected with PPNG strains) were retreated with spectinomycin; 7 were cured (including all 4 PPNG cases) and 3 were lost to follow-up. The 13 patients who were not cured by spectinomycin received procaine penicillin; 7 were cured and 6 were lost to follow-up. Minimal inhibitory concentrations (MICs) of 5 antimicrobials were obtained for 246 pretreatment isolates (Table 2). Twenty-eight strains (11%) were highly sensitive to penicillin (MIC < 0.06 Ag/ml), 75 strains (31%) had a slightly diminished sensitivity (MIC, 0.06-0.25 ,g/ml), and 143 strains (58%) had a highly diminished sensitivity to penicillin (MIC > 0.5 ytg/ml). In this last group were 4 PPNG strains with MIC > 8 sg/ml. All strains were sensitive to 16 or 32 ,g/ml of spectinomycin; 54 strains (22%) had a diminished sensitivity to tetracycline (MIC > 2 ytg/ml). Of 23 treatment failures, pretreatment MICs were available for 17 (8 failures in the penicillin group and 9 in the spectinomycin group). Fig. 1 gives the failure rates according to the pretreatment MIC for Percent Failure 14.3 14.3 7 6 5 4- 3 2 - 1 <0.06 (28) Penicillin fl Spectinomycin 40 1.3 0.06-0.25 (75) 49 ) 0.5 Penicillih (143) Fig. 1. Failure rate with procaine penicillin and spectino- mycin in the treatment of gonorrhoea, according to the MIC of penicillin for the pretreatment isolates of N. gonorrhoeae. The number of isolates tested is given in parentheses. penicillin. The failure rate for penicillin treatment was considerably higher for strains that were less sensitive to penicillin. For spectinomycin a reverse relationship was found, with a high failure rate for the strains that were highly sensitive to penicillin. Post-treatment MICs were available for 11 strains from the 23 treatment failures. Both pre- and post-treatment MICs were available for only 8 of these strains; there was no significant difference between the two values. DISCUSSION Both regimens studied were highly efficacious for the treatment of uncomplicated gonorrhoea. The failure rates were similar to those seen in the United States of America, where the rates were 3.2% with procaine penicillin and probenecid, and 5.2% with Table 2. Antimicrobial MICs for 246 strains of N. gonorrhoeae isolated in Bangui, Central African Republic Minimal inhibitory concentration (&g/ml) Antimicrobial 0.0078 0.015 0.03 0.06 0.12 0.25 0.5 1 2 4 8 16 32 Penicillin 5 14 9 13 36 26 35 67 36 1 4 Thiamphenicol 2 5 59 1 2 81 84 3 Tetracycline 3 19 86 84 36 18 Kanamycin 3 27 113 103 Spectinomycin 26 220 0 MIC > 8 ,g/ml; all 4 strains were PPNG. 91 A. MEHEUS ET AL. the spectinomycin regimen (5). This could be expected from the susceptibility pattern of N. gonor- rhoeae in the Central African Republic which, for both antibiotics, is very similar to that in the USA (6). Few carefully conducted trials of gonorrhoea treat- ment have been done in Africa. The available data show failure rates of 1.0-6.8%/o for high-dose benzyl or procaine penicillin-probenecid schedules (7-10), and 3.9-5.4%o for spectinomycin (10, 11). In agreement with a study carried out in the USA (6), we found a clear relationship between pretreat- ment susceptibility and failure rate for penicillin; such a relationship was not found for spectinomycin (all strains were susceptible to 16 or 32 Itg/ml). There are no other African data available for comparison. It is interesting that failure rates with spectinomycin were higher with strains that were more sensitive to penicillin. These data need to be interpreted with caution because they are based on a relatively small number of subjects. Although not analysed in the publication, the raw data from the USA showed that, for strains with a penicillin MIC < 0.06 yg/ml, the spectinomycin failure rate was 7.507o, while for penicillin MICs of 1.0 and 0.5 tsg/ml, the failure rate was 0-3/0o (6). Later results from the USA did not show any relationship between failure rate with spectinomycin and the pretreatment MIC for peni- cillin (12). A gonococcal susceptibility study was done on 70 strains isolated in the Central African Republic in 1979-80 (1) and our results 15 months later show a slightly higher percentage of strains with diminished sensitivity to penicillin and tetracycline. Predicted failure rates were 4.3%o for procaine penicillin- probenecid and 4.6%7o for spectinomycin (1), which are very close to the rates determined in this study. An important new finding is the 1.507o prevalence of PPNG strains, isolated for the first time in the Central African Republic in 1981. In the first 8 months of 1982, the prevalence increased to 3.1%o (R. Widy-Wirski, unpublished data, 1982). At this low prevalence, the efficacy of penicillin treatment is not yet significantly affected. Procaine penicillin can still be recommended as first-line treat- ment and spectinomycin for patients allergic to penicillin and for all treatment failures and their con- tacts (10). Once the prevalence of PPNG strains reaches 10%o or more, it is generally accepted that penicillin should be abandoned as the first-line treat- ment (13). After the emergence of PPNG strains in south-east Asia and West Africa, most countries continued to use penicillin as the first-line treatment, with an anti- biotic curative for PPNG as back up, mainly for economic reasons. They hoped to contain the PPNG problem at a low level for as long as possible. However, this policy failed and there has been a steady and rapid increase in the prevalence of PPNG strains in both areas. The current prevalences of PPNG strains are 55-60% in the Western Pacific region and 50% in Nigeria (A. Osoba, unpublished data, 1982). An alternative policy would be to use an antibiotic active against PPNG strains as soon as these strains occur in a given area, as first-line treatment for cases and contacts. For most developing countries, the cost of treatment is very important, but spectinomycin has been shown to be not more expensive than unit-dose procaine penicillin (14). Treatment failures with spec- tinomycin could be retreated with tetracycline, pro- caine penicillin-probenecid, thiamphenicol, kana- mycin, or cefoxitin. A concomitant infection of C. trachomatis was found in only 5%o of males with gonorrhoea, which is considerably less than the 20-30%o isolation rate for Chiamydia in gonorrhoea patients in western Europe and North America (15). However, similar studies in Gambia, Kenya, South Africa, and Swaziland also found fairly low isolation rates (3-13Oo) for C. trachomatis in males with gonorrhoea (16-19). These low isolation rates contrast sharply with the high prevalence of anti-chlamydial antibodies in the same populations, including the patients in this trial (results not shown) (17, 18). Further studies are required to elucidate the epidemiology of genital chlamydial infections in Africa. ACKNOWLEDGEMENTS This study was supported in part by grants from the University of Antwerp and from the Upjohn Co., Kalamazoo, MI, USA. 92 TREATMENT OF GONORRHOEA IN AFRICA 93 R-SUMt TRAITEMENT DE LA GONOCOCCIE PAR LA SPECTINOMYCINE ET LA PROCAINE PENICILLINE CHEZ DES PATIENTS CENTRAFRICAINS DE SEXE MASCULIN A des patients de Bangui de sexe masculin porteurs d'une gonococcie aigue confirmee par culture, on a administre, selon une distribution aleatoire, l'un des deux traitements suivants: a) 4,0 x 106 unites de procaine penicilline par voie intramusculaire plus I g de probenicide per os; b) 2 g de spectinomycine v.i. Quatre cent soixante patients au total ont e recrutes pour cette etude et 419 (91 07) se sont pre- sentes pour la visite de controle. Le taux d'echec a e de 4,8% avec le schema base sur la penicilline et de 6,2% avec la spectinomycine; la difference n'est pas statistiquement significative. De meme, la fre- quence de l'urethrite post-gonococcique s'est montree assez semblable dans les deux groupes mais le traitement par la penicilline a provoque sensiblement plus d'effets secon- daires tous mineurs, tels que douleurs a l'injection et ver- tiges. Cinq pour cent des patients presentaient simultane- ment une infection a Chlamydia trachomatis et presque tous ont contacte une ur6thrite postgonococcique. Parmi les 460 malades de l'tude, 7 (1,5%) etaient por- teurs d'une souche de NGPP. Cinq se trouvaient dans le groupe traite par la penicilline et le traitement a echoue; ils ont toutefois W gueris par la suite a l'aide de la spectino- mycine; les deux autres, traites par la spectinomycine, ont e gueris. Les 6preuves de sensibilite aux antibiotiques effectuees par dilution en milieu solide ont montre que 1 I% des. souches etaient tres sensibles a la penicilline (CMI <0,06 tg/ml), 31% avaient une sensibilite legere- ment attenuee (0,06 itg/ml < CMI < 0,25 Lg/ml) et 58% une sensibilite tres attenuee (CMI > 0,5 iLg/ml). Toutes les souches etaient sensibles a 16 ou 32 Ag/ml de spectinomycine et 22% des souches presentaient une sensibilite abaissee a la tetracycline (CMI > 2 yg/ml). Le taux d'echec avec le traitement par la penicilline s'est aver6 considerablement plus grand chez les patients infectes par des souches de sensibilite r6duite vis-a-vis de la penillicine. L'echec de la chimiotherapie par la spectinomycine a e plus frequent chez les malades porteurs de souches tres sensibles a la penicilline. On admet souvent que, tant que les NGPP n'atteignent pas un niveau critique (une fr6quence de 10% par exemple), le traitement de premiere intention de la gonococcie peut s'appuyer l'administration de forte doses de penicilline. 11 est propose, dans le present article, qu'une fois les souches de NGPP introduites dans un pays, le traitement de pre- miere intention des cas de gonococcie et des contacts soit fonde sur l'emploi d'antibiotiques efficaces contre les infec- tions a NGPP, par exemple la spectinomycine. La tetra- cycline peut etre recommandee en seconde intention, car elle permet de traiter simultanement l'infection a C. tracho- matis, qui est beaucoup moins souvent associ&e aux gono- coccies en Afrique que dans les pays industrialises. REFERENCES 1. WIDY-WIRSKI, R. ET AL. Antimicrobial susceptibility of gonococci isolated in the Central African Republic. Bulletin ofthe World Health Organization, 60: 959-963 (1982). 2. WIDY-WIRSKI, R. & D'COSTA, J. Surveillance epid& miologique et therapeutique de Neisseria gonorrhoeae productrices de pEnicillinase dans la R6gion Africaine. Afrique sante, 10(Suppl.): 9-14 (1981). 3. PERINE, P. L. ET AL. Epidemiology and treatment of penicillinase-producing Neisseria gonorrhoeae. Sexually transmitted diseases, 6: 152-158 (1979). 4. RIPA, K. T. & MARDH, P. A. Cultivation of Chlamydia trachomatis in cycloheximide-treated McCoy cells. Journal of clinical microbiology, 6: 328-331 (1977). 5. KAUFMAN, R. E. ET AL. National gonorrhoea therapy monitoring study: treatment results. New England journal of medicine, 294: 1-4 (1976). 6. JAFFE, H. W. ET AL. National gonorrhoea therapy monitoring study: in vitro antibiotic susceptibility and its correlation with treatment results. New England journal of medicine, 294: 5-9 (1976). 7. MASAWE, A. E. J. ETAL. Treatment of gonorrhoea with a combination of probenecid and sodium penicillin G among African Ugandans. African journal of medical sciences, 3: 163-167 (1972). 8. ARYA, 0. P. & BOSA, C. B. In search of an ideal single session penicillin schedule for the treatment of gonor- rhoea in Uganda. British journal of venereal diseases, 49: 460-463 (1973). 9. MEHEUS, A. ET AL. Treatment of gonorrhoea with a combination of probenecid and procaine penicillin in Rwanda. African journal of medical sciences, 5: 209-212 (1976). 10. RATNAM, A. V. ET AL. Penicillin and spectinomycin in treatment of gonococcal urethritis. Sexually trans- mitted diseases, 9: 135-137 (1982). 11. MEHEUS, A. ET AL. Treatment of gonorrhoea with spectinomycin hydrochloride (Trobicin) in African males. Current therapeutic research, 16: 1091-1095 (1974). 12. GUINAN, M. E. ET AL. The national gonorrhoea therapy monitoring study. Review of treatment results and of in vitro antibiotic susceptibility 1972-1978. Sexually transmitted diseases, 6 (Suppl): 93-102 (1979). 13. PIOT, P. & HOLMES, K. K. Sexually transmitted diseases. In: Warren, K. S. & Mahmoud, A. A. F., ed., Tropical and geographical medicine, New York, McGraw Hill, 1983. 94 A. MEHEUS ET AL. 14. BERG, S. W. & HARRISON, W. 0. Spectinomycin as primary treatment of gonorrhoea in areas of high prevalence of penicillinase-producing N. gonorrhoeae. Sexually transmitted diseases, 8: 38-39 (1981). 15. ORIEL, D. Infections of the male genital tract. In: Mardh, P. A. et al., ed., Chiamydial infections. Amsterdam, Elsevier Biomedical Press, 1982, pp. 93-106. 16. MEHEUS, A. Z. ET AL. Epidemiology and aetiology of urethritis in Swaziland. International journal of epidemiology, 9: 239-244 (1980). 17. BALLARD, R. C. ET AL. Urethritis and associated in- fections in Johannesburg. The role of Chiamydia trachomatis. South African journal of sexually trans- mitted diseases, 1: 24-26 (1981). 18. NSANZE, H. ET AL. Chlamydial infections in selected populations in Kenya. In: Mardh, P. A. et al., ed., Chiamydial infections, Amsterdam, Elsevier, Bio- medical Press, 1982, pp. 421-424. 19. MABEY, D. C. W. & WHITTLE, H. C. Genital and neonatal chlamydial infection in a trachoma endemic area. Lancet, 2: 300-301 (1982).