Rollback of Salmonella enterica Serotype Typhi Resistance to Chloramphenicol and Other Antimicrobials in Kolkata, India

Multidrug-resistant (MDR) Salmonella enterica serotype Typhi created a significant therapeutic problem in the late 1980s and early 1990s (1, 7, 8). As serotype Typhi strains were resistant to commonly used antimicrobials for the treatment of typhoid fever, such as ampicillin, amoxicillin, chloramphenicol, and cotrimoxazole, ciprofloxacin began to be used widely for MDR typhoid fever (2). But after some years, decreased clinical responsiveness of typhoid fever cases to ciprofloxacin was observed. Ceftriaxone continues to be uniformly effective against typhoid fever, including cases in which ciprofloxacin treatment has failed (3). The purpose of the study was to compare the antimicrobial resistance patterns of Salmonella enterica serotype Typhi strains isolated from typhoid fever cases in Kolkata, India over the past few years.

From 1 May 2003 to 31 July 2004, as part of a prospective surveillance for typhoid fever in two urban slums in Kolkata (formerly Calcutta), a total of 4,400 blood samples were collected in BACTEC culture vials (Becton Dickinson) from patients with fever lasting 3 or more days. Incubation and identification of organisms were carried out according to conventional procedures. Serotype Typhi was isolated from 132 (3%) of the samples collected and processed. The antimicrobial susceptibility of the serotype Typhi isolates was determined by the Kirby-Bauer disk diffusion technique, and the MICs were determined by the E test (AB Biodisk, Solna, Sweden). The following resistance pattern of serotype Typhi strains was observed: chloramphenicol, 13%; ampicillin, 13%; amoxicillin-clavulanic acid, 0%; cotrimoxazole, 15%; ciprofloxacin, 10%; ofloxacin, 2%; and ceftriaxone, 0%. We compared our present observations with those from previous years (Table 1). The chi-square test was used to compare the pro-portion of resistant isolates. A significant decrease over the years in resistance to chloramphenicol, ampicillin, and cotrimoxazole was noticed (P < 0.0001).

TABLE 1.

Antimicrobial resistance pattern of Salmonella enterica serotype Typhi, Kolkata, India^a

Antimicrobial	No. of resistant strains/total no. tested (%)
	1991 to 1992b	1998 to 1999c	2003 to 2004d
Chloramphenicol	221/221 (100)	13/26 (50)	17/132 (13)e
Ampicillin	221/221 (100)	15/26 (58)	17/132 (13)e
Amoxicillin-clavulanic acid	ND	ND	0/132 (0)
Cotrimoxazole	221/221 (100)	17/26 (65)	19/130 (15)e
Ciprofloxacin	0/221 (0)	0/26 (0)	13/131 (10)f
Ofloxacin	ND	ND	2/132 (2)g
Ceftriaxone	ND	0/26 (0)	0/129 (0)

^aND, not done.

^bSee references 7 and 8.

^cSee reference 3. Due to the limited number of typhoid cases reported to the hospital, there were few serotype Typhi isolates.

^dObserved in the present study.

^eSignificant change over the years (P < 0.0001).

^fThe MIC at which 90% of the isolates tested were inhibited (MIC₉₀) was 16 μg/ml for 2 resistant strains, and the MIC₉₀ range was 0.125 to 0.38 μg/ml for 11 intermediately susceptible strains.

^gThe MIC₉₀ was 16 μg/ml for two resistant strains.

Our findings indicate a remarkable reversal in the resistance pattern of serotype Typhi in Kolkata since the early 1990s. This reversal may be due to the emergence of de novo susceptible strains or the loss of a high-molecular-weight self-transferable plasmid encoding chloramphenicol, ampicillin, and cotrimoxazole resistance in recently isolated serotype Typhi strains (5). There are reports which indicate the coexistence of antibiotic-sensitive and MDR serotype Typhi strains as distinct independent clones (4, 10). The MDR serotype Typhi isolates may also belong to different pulsed-field gel electrophoresis genotypes (6) or emerge from a single clone (9). In the present study, molecular characterization of the isolated strains was not carried out. The change in the resistance patterns of chloramphenicol, ampicillin, and cotrimoxazole is noteworthy and calls for future studies to determine molecular relatedness of or differences between the present and past strains and possible loss of the resistance plasmid.

It may be concluded that this new information could help clinicians to decide on the drug of choice for the treatment of typhoid fever. Chloramphenicol and other antimicrobials may be of use again. The emergence of isolates resistant to ciprofloxacin and ofloxacin suggests that if present use remains unchanged, it is probably only a matter of time before the organisms develop widespread resistance to the fluoroquinolones.