S106

Multidrug-Resistant Salmonella typhi: A Worldwide Epidemic

Bernard Rowe, Linda R. Ward, and E. John Threlfall
From the WHO Collaborating Centre for Phage Typing and Drug Resistance in Enterobacteria, Laboratory of Enteric Pathogens, Central Public Health Laboratory, London, United Kingdom

Since 1989, strains of Salmonella typhi resistant to chloramphenicol, ampicillin, and trimethoprim (i.e., multidrug-resistant [MDR] strains) have been responsible for numerous outbreaks in countries in the Indian subcontinent, Southeast Asia, and Africa. MDR strains have also been isolated with increasing frequency from immigrant workers in countries in the Arabian Gulf, as well as in developed countries from returning travelers. In all MDR strains so far examined, multiple resistance has been encoded by plasmids of the H 1 incompatibility group. As a result of the widespread dissemination of such strains, chloramphenicol can no longer be regarded as the first-line drug for typhoid fever. Because strains are also resistant to ampicillin and trimethoprim, the efficacy of these antibiotics has also been impaired, and ciprofloxacin is now the drug of choice for typhoid fever. Chromosomally encoded resistance to ciprofloxacin has now been observed in a small number of strains isolated in the United Kingdom from patients returning from the Indian subcontinent, and in at least one case the patient did not respond to treatment with this antibiotic. It is regrettable that resistance to ciprofloxacin has now emerged in MDR S. typhi, and it is of paramount importance to limit the unnecessary use of this vital drug so that its efficacy should not be further jeopardized.

Downloaded from http://cid.oxfordjournals.org/ by guest on December 5, 2011

Although precise figures are not available, in 1986 the number of cases of typhoid fever worldwide was estimated at 33 million [1]. The disease is endemic in many developing countries, particularly in the Indian subcontinent, South and Central America, and Africa, with an estimated mean incidence (per 100,000 people per annum) of 150 in South America and 900 in Asia [2]. Travelers to these areas are at risk, and in developed countries of northwest Europe, most cases involve returning travelers. In Britain, for example, 200-300 cases occur each year, an overall incidence of 0.5 per 100,000 [3], and at least 70% of the patients have a history of recent foreign travel.
Emergence of Drug Resistance

chloramphenicol-resistant strain was that which occurred in Mexico in 1972 [6, 7]. At about the same time, there was a second substantial outbreak in Kerala, India [8], and in both outbreaks mortality was high. In the succeeding 5 years outbreaks occurred in several other countries, notably, Vietnam, Indonesia, Korea, Chile, and Bangladesh (for review, see [9]). A notable feature of all chloramphenicol-resistant strains from outbreaks in Mexico, the Indian subcontinent, and Southeast Asia was that although the strains belonged to different Vi phage types, resistance to chloramphenicol was encoded by a plasmid of the H 1 incompatibility group, often in combination with resistance to streptomycin, sulfonamides, and tetracyclines (antibiogram, CSSuT; see footnote to table 1) [7, 9].
Outbreaks of Multidrug-Resistant Typhoid

Treatment with an appropriate antibiotic is essential for typhoid fever, and because of the nature of the disease it should commence as soon as the clinical diagnosis is made rather than after the results of antimicrobial susceptibility tests are available. Until the mid-1970s chloramphenicol was the undisputed drug for treatment of typhoid fever, and in developed countries the use of this antibiotic resulted in a reduction in mortality from 10% to <2% [4]. Although before 1970 a few sporadic isolations of chloramphenicol-resistant Salmonella typhi had been reported from Aden, Chile, and Kuwait [5], the first epidemic caused by a

Reprints or correspondence: Dr. E. John Threlfall, WHO Collaborating Centre for Phage Typing and Drug Resistance in Enterobacteria, Laboratory of Enteric Pathogens, Central Public Health Laboratory, 61 Colindale Avenue, London NW9 5HT, United Kingdom.
Clinical Infectious Diseases 1997; 24(Suppl 1):S106-9
1997 by The University of Chicago. All rights reserved. 1058-4838/97/2401-0046$02.00

Since 1989 outbreaks caused by strains of S. typhi resistant to chloramphenicol, ampicillin, and trimethoprim and with additional resistance to streptomycin, sulfonamides, and tetracyclines (antibiogram, ACSSuTTm; see footnote to table 1) have been reported in many developing countries, especially Pakistan [10, 11] and India [9, 12-14]. Such strains have been termed multidrug-resistant (MDR). MDR strains have also caused outbreaks in Bangladesh [15], several countries in Southeast Asia [16], and both North and South Africa [17, 18], and in 1991 there was a small outbreak associated with a foodhandler in a take-out restaurant in Britain [19] (table 1). MDR strains linked to immigrant workers from the Indian subcontinent have been isolated with increasing frequency in several countries in the Arabian Gulf [20-22]. The most common Vi phage types have been those of El and M 1, but outbreaks have been caused by MDR strains of Vi phage types 51 and 0 in India [13, 14], Vi phage type M1

CID 1997;24 (Suppl 1)

Multidrug-Resistant Typhoid Fever

S107

Table 1. Data regarding outbreaks due to multiresistant S. typhi, 1989 1995.

-

Reference(s) [10, 11] [12-14] [20-22] ... [19] [18] [17] ... [16] [15]

Year(s) of outbreak 1989 1990-95 1990-95 1990-93 1991 1991 1991-92 1992-94 1993 94 1994 1994-95

Country or area Pakistan India Arabian Gulf Kuala Lumpur United Kingdom South Africa Egypt Vietnam Philippines Bangladesh Pakistan

Vi phage type(s) M1 El, 51, 0 El, Ml, 51 El M1 A E2, Cl, Dl-N

Resistance type (antibiogram)* ACSSuTTm ACSSuTTm ACSSuTTm ACSSuTTm CSTTm ACKSSuT ACSSuTTm ACSSuTTm CKSSuTTm ACSSuTTm ACSSuTTm

Plasmid type H1 Hi H1 Hi li, Hi H1

El El

H1 H1

* A = ampicillin; C = chloramphenicol; K = kanamycin; S = streptomycin; Su = sulfonamides; T = tetracyclines; Tm = trimethoprim. Downloaded from http://cid.oxfordjournals.org/ by guest on December 5, 2011

in the United Kingdom [19], Vi phage type A in South Africa [18], and Vi phage types E2, Cl, and Dl-N in Egypt [17]. The phage types of MDR strains isolated in the Arabian Gulf (E 1, M 1, and 51) correspond to those that have caused outbreaks in the Indian subcontinent. Strains of Vi phage types El and M1 have also been isolated in Australia [19] and Canada [23]. One outbreak of particular interest was that which occurred in the Philippines (metropolitan Manila) from July 1993 to April 1994, in which 252 cases of MDR S. typhi in 13 hospitals were reported [16]; the strains were not phage-typed. The strains were resistant to chloramphenicol, co-trimoxazole, kanamycin, streptomycin, and tetracyclines (antibiogram, CKSSuTTm; see footnote to table 1) but susceptible to ampicillin. Although the vehicle of infection was not microbiologically confirmed, epidemiological findings suggested an association between typhoid fever and the consumption of flavored-ice drinks purchased from street vendors. The recent explosive emergence in developing countries of strains of S. typhi with resistance to trimethoprim and ampicillin has caused many problems, as since 1980 these antibiotics had been used extensively for the treatment of patients infected with chloramphenicol-resistant strains [24]. Without exception, in all outbreaks of MDR S. typhi so far studied, the complete spectrum of multiple resistance has been encoded by plasmids of the H 1 incompatibility group. Drug Resistance in British Isolates In the United Kingdom, of 2,356 strains of S. typhi isolated between 1978 and 1985, only six (0.25%) were resistant to chloramphenicol at a clinically significant level (MIC, >32 mg/L). On the basis of these findings, in 1987 it was recommended that in the United Kingdom chloramphenicol should remain the first-line drug for treatment of typhoid fever and, in particular, that it should be used until the results of laboratory susceptibility tests are known [25]. In the succeeding 4-year period (1986-1989), the isolation of chloramphenicol-resistant

S. typhi from patients in the United Kingdom increased slightly,

from 0.25% to 1.5% [26], but the increase was not considered sufficient to justify altering the recommendation made in 1987. The situation has changed dramatically since 1990. In that year 20% of strains were resistant not only to chloramphenicol but also to trimethoprim (MIC, > 125 mg/L) and ampicillin (MIC, > 125 mg/L) [21]. The situation has worsened in the succeeding 5 years, and since 1994 about 35% of strains from patients with typhoid fever have been resistant to chloramphenicol (table 2); the majority of chloramphenicol-resistant strains also are resistant to ampicillin and trimethoprim [27]. In 1990 the majority of chloramphenicol-resistant strains isolated in the United Kingdom belonged to Vi phage type M1 and were from patients recently returned from Pakistan, although some strains of Vi phage type El were also isolated [21]. In subsequent years the proportion of MDR strains belonging to Vi phage type El has increased, and in 1995 almost all MDR strains belonged to this phage type (table 3). From 1990 to 1993 the majority of patients infected with MDR strains of Vi phage type El had recently returned from India, but since 1994 an increasing number of patients infected
Table 2. Isolations of chloramphenicol-resistant S. typhi in the United Kingdom, 1978-1995.
Year(s) 1978-85 1986-89 1990 1991 1992 1993 1994 1995 No. of isolates studied 2,345 790 248 226 204 194 259 291 No. (%) of isolates resistant* to chloramphenicol 6 (0.3) 12 (1.5) 50 (20) 48 (21) 49 (24) 49 (25) 94 (36) 100 (34)

NOTE. Data are from the Laboratory of Enteric Pathogens (London). * MIC, >32 mg/L.

S108

Rowe, Ward, and Threlfall

CID 1997; 24 (Suppl 1)

Table 3. Phage-type distribution among chloramphenicol-resistant S. typhi isolates recovered in the United Kingdom, 1978-1995.
Total no. of chloramphenicolresistant isolates 6 12 50 48 49 49 94 100 No. of isolates of indicated phage type A 2 1 8 20 27 29 81 96 2 36 25 12 18 8 8 2 El M1 0 Others 4 1 4 3 10 1 4 4

Year(s) 1978-85 1986-89 1990 1991 1992 1993 1994 1995

NOTE. Data are from the Laboratory of Enteric Pathogens (London). Downloaded from http://cid.oxfordjournals.org/ by guest on December 5, 2011

with MDR Vi phage type El had recently returned from Pakistan. Other countries visited by patients from whom MDR S. typhi has been isolated include Sri Lanka, Bangladesh, Nepal, and Somalia [27]. In all MDR S. typhi isolated in the United Kingdom since 1990, resistance to chloramphenicol, ampicillin, and trimethoprim, together with resistance to streptomycin and sulfonamides (antibiogram, ACSSuTTm; see footnote to table 1), has been encoded by a plasmid of the H 1 incompatibility group [9, 21].
Recommendations for Therapy MDR S. typhi is now endemic in many developing countries but also has been isolated from returning travelers in developed countries. As a result of the proliferation of such strains, the use of chloramphenicol has been compromised, and that of ampicillin and trimethoprim similarly impaired. In October 1990, because 20% of strains of S. typhi isolated in the United Kingdom were resistant to chloramphenicol, it was suggested that physicians in the United Kingdom should consider ciprofloxacin as an alternative to chloramphenicol for the treatment of enteric fever [26]. In 1991, in light of the increasing isolation of MDR strains in the United Kingdom after reports of outbreaks in the Indian subcontinent and Arabian Gulf, the use of ciprofloxacin for the treatment of typhoid was recommended, particularly for patients returning from areas where MDR strains are endemic [21]. Ciprofloxacin is now used extensively for the treatment of typhoid in both developing and developed countries [12, 24, 28, 29].

0.30 mg/L) was reported [19]. The patient was a 1-year-old child who had been infected in India and did not respond to treatment with ciprofloxacin [29]. Concern was expressed that resistance to this important antibiotic had appeared in an isolate of MDR S. typhi [19]. Subsequently, chromosomally encoded ciprofloxacin resistance has been observed in a total of 14 strains of S. typhi isolated in the United Kingdom since 1991, of which 10 have also been resistant to chloramphenicol, ampicillin, and trimethoprim [27]. All MDR strains with additional resistance to ciprofloxacin have belonged to Vi phage type El . Patients infected with such strains had recently returned from several countries in the Indian subcontinent, where this antibiotic has been used to control outbreaks of MDR typhoid since 1990 [12]. Although it is regrettable that resistance to ciprofloxacin is now emerging, ciprofloxacin still remains the drug of choice for the treatment of MDR typhoid fever. To maintain the efficacy of ciprofloxacin for typhoid fever, however, unnecessary usage should be avoided. In particular, any prophylactic use of this important drug should be strongly discouraged. It is unfortunate that a range of fluoroquinolones, including ciprofloxacin, have been recommended for prophylaxis against traveler's diarrhea [30].
References 1. Edelman R, Levine MM. Summary of an international workshop on typhoid fever. Rev Infect Dis 1986; 8:329-49. 2. Ivanoff B. Typhoid fever: global situation and WHO recommendations. Proceedings of the 2nd Asia-Pacific Symposium on Typhoid Fever and Other Salmonellosis. Bangkok: Infectious Disease Association of Thailand, 1994:39. 3. Anonymous. Typhoid and paratyphoid fevers. Office of Population Census and Surveys Monitor 1985; MB2 85/2:9-11. 4. Woodward TE, Smadel JE, Ley HL, et al. Preliminary report on the beneficial effect of chloromycetin on the treatment of typhoid fever. Ann Intern Med 1948; 29:131-3. 5. Anderson ES, Smith HR. Chloramphenicol resistance in the typhoid bacillus. BMJ 1972;3:329-31.

Resistance to Ciprofloxacin

In 1992, the isolation in the United Kingdom of an MDR strain of S. typhi Vi phage type El with plasmid-encoded resistance to chloramphenicol, ampicillin, and trimethoprim (and chromosomal resistance to ciprofloxacin at a concentration of

CID 1997;24 (Suppl 1)

Multidrug-Resistant Typhoid Fever

S109

6. Centers for Disease Control. Typhoid fever-Mexico. Morb Mortal Wkly Rep 1972;21:177-8. 7. Anderson ES. The problem and implication of chloramphenicol resistance in the typhoid bacillus. J Hyg 1975; 74:289-99. 8. Paniker CKJ, Vilma KN. Transferable chloramphenicol resistance in Salmonella typhi. Nature 1972;239:109 -10. 9. Threlfall EJ, Rowe B, Ward LR. Occurrence and treatment of multiresistant Salmonella typhi. Public Health Laboratory Service Microbiology Digest 1991; 8:56-9. 10. Karamat KA. Multiple drug resistant Salmonella typhi and ciprofloxacin. In: Proceedings of the 2nd Western Pacific Congress on Infectious Diseases and Chemotherapy. Jakarta, Indonesia: Western Pacific Society of Chemotherapy, 1990:480. 11. Mandal BK. Treatment of multiresistant typhoid fever [letter]. Lancet
1990; 336:1383.

17. Mourad AS, Metwally M, Nour El Deen A, et al. Multiple-drug-resistant Salmonella typhi. Clin Infect Dis 1993; 17:135-6. 18. Coovadia YM, Gathiram V, Bhamjee A, et al. An outbreak of multiresistant Salmonella typhi in South Africa. Q J Med 1992; 82:91-100. 19. Rowe B, Threlfall EJ, Ward LR. Ciprofloxacin and typhoid fever. Lancet
1992;339:740.

20. Anonymous. Spread of multiresistant Salmonella typhi [letter]. Lancet

1990; 336:1065-6.

12. Anand AC, Kataria VK, Singh W, Chatterjee SK. Epidemic multiresistant enteric fever in eastern India [letter]. Lancet 1990; 335:352. 13. Threlfall EJ, Ward LR, Rowe B, et al. Widespread occurrence of multiple drug-resistant Salmonella typhi in India. Eur J Clin Microbiol Infect Dis 1992;11:989-93. 14. Prakash K, Pillai PK. Multidrug-resistant Salmonella typhi in India. Alliance for the Prudent Use of Antibiotics Newsletter 1992;10:1-3. 15. Albert MJ, Haider K, Nahar S, Kibriya AK, Hossain MA. Multiresistant Salmonella typhi in Bangladesh. J Antimicrob Chemother 1991;27: 554 -5. 16. Tinya-Superable JF, Castillo MTG, Magboo FP, et al. Multidrug resistant Salmonella typhi outbreak in metro Manila, Philippines. Southeast Asian J Trop Med Public Health 1995; 26(suppl 2):37-8.

21. Rowe B, Threlfall EJ, Ward LR. Treatment of multiresistant typhoid fever. Lancet 1991; 337:1422. 22. Wallace MR, Yousif AA, Mahroos GA, et al. Ciprofloxacin versus ceftriaxone in the treatment of multiresistant typhoid fever. Eur J Clin Microbiol Infect Dis 1993;12:907-10. 23. Hartnett N, McLeod S, AuYong Y, et al. Emergence in Ontario, Canada, of multiresistant Salmonella typhi from South Asia. Lancet 1992; 340: 177 24. Mandal BK. Modern treatment of typhoid fever. J Infect 1991;22:1-4. 25. Rowe B, Threlfall EJ, Ward LR. Does chloramphenicol remain the drug of choice for typhoid? Epidemiol Infect 1987; 98:379-83. 26. Rowe B, Threlfall EJ, Ward LR. Spread of multiresistant Salmonella typhi. Lancet 1990; 36:1065. 27. Rowe B, Threlfall EJ, Ward LR. Ciprofloxacin-resistant Salmonella typhi in the UK. Lancet 1995; 346:1302. 28. Eykyn SJ, Williams H. Treatment of multiresistant Salmonella typhi with oral ciprofloxacin [letter]. Lancet 1987;2:1407-8. 29. Umasankar S, Wall RA, Berger J. A case of ciprofloxacin-resistant typhoid fever. CDR Review 1992;2:R139-40. 30. DuPont HL, Ericsson CD. Prevention and treatment of traveler's diarrhea. N Engl J Med 1993; 328:1821-7.

Downloaded from http://cid.oxfordjournals.org/ by guest on December 5, 2011