Trends in antimicrobial resistance amongst Salmonella Typhi in Bangladesh: A 24-year retrospective observational study (1999-2022)

Abstract

Background: Rising antimicrobial resistance (AMR) in Salmonella Typhi restricts typhoid treatment options, heightening concerns for pan-oral drug-resistant outbreaks. However, lack of long-term temporal surveillance data on AMR in countries with high burden like Bangladesh is scarce. Our study explores the AMR trends of Salmonella Typhi isolates from Bangladesh, drawing comparisons with antibiotic consumption to optimize antibiotic stewardship strategies for the country.

Methodology/principal findings: The typhoid fever surveillance from 1999 to 2022 included two pediatric hospitals and three private clinics in Dhaka, Bangladesh. Blood cultures were performed at treating physicians' discretion; cases were confirmed by microbiological, serological, and biochemical tests. Antibiotic susceptibility was determined following CLSI guidelines. National antibiotic consumption data for cotrimoxazole, ciprofloxacin, and azithromycin was obtained from IQVIA-MIDAS database for comparison. Over the 24 years of surveillance, we recorded 12,435 culture-confirmed typhoid cases and observed declining resistance to first-line drugs (amoxicillin, chloramphenicol, and cotrimoxazole); multidrug resistance (MDR) decreased from 38% in 1999 to 17% in 2022. Cotrimoxazole consumption dropped from 0.8 to 0.1 Daily defined doses (DDD)/1000/day (1999-2020). Ciprofloxacin non-susceptibility persisted at >90% with unchanged consumption (1.1-1.3 DDD/1000/day, 2002-2020). Low ceftriaxone resistance (<1%) was observed, with slightly rising MIC (0.03 to 0.12 mg/L, 1999-2019). Azithromycin consumption increased (0.1 to 3.8 DDD/1000/day, 1999-2020), but resistance remained ≤4%.

Conclusion: Our study highlights declining MDR amongst Salmonella Typhi in Bangladesh; first-line antimicrobials could be reintroduced as empirical treatment options for typhoid fever if MDR rates further drops below 5%. The analysis also provides baseline data for monitoring the impact of future interventions like typhoid conjugate vaccines on typhoid burden and associated AMR.

Fig 1. Yearly trends in resistance (-R) and multidrug resistance (MDR) among Salmonella Typhi Cases (1999–2022, Bangladesh).

(A) Amoxicillin resistance, (B) Chloramphenicol resistance, (C) Cotrimoxazole resistance, and (D) MDR (defined as concurrent resistance to amoxicillin, chloramphenicol, and cotrimoxazole). Information on MDR is presented for all recorded cases.

Fig 2. Yearly trends in Salmonella Typhi susceptibility for Ciprofloxacin, Ceftriaxone, and Azithromycin (1999–2022) in Bangladesh.

Antimicrobial susceptibility data were present for 12,431, 12,414, and 8,187 isolates for (A) Ciprofloxacin non-susceptibility (-NS), (B) Ceftriaxone resistance (-R), and (C) Azithromycin resistance (-R), respectively.

Fig 3. Minimum inhibitory concentration (MIC) of ciprofloxacin and ceftriaxone among Salmonella Typhi cases in Bangladesh from 1999–2022.

The MICs were determined for a total of 3,214 and 3,518 isolates for (A) Ciprofloxacin (1999–2016), and (B) Ceftriaxone (1999–2019), respectively. The color of the dots represents resistant (red), intermediate (blue), and susceptible (green) isolates. The generalized additive model (GAM) lines depict the changes in the MICs (black line), with a 95% confidence interval (grey shades) over time. The y-axis is drawn based on a logarithm of 2 scale; MIC values are shown directly (for example, 0.5 instead of 2⁻¹).

Fig 4. Patterns of antimicrobial consumption and resistance patterns in Salmonella Typhi in Bangladesh from 1999–2020.

Patterns of antimicrobial consumption and resistance in (A) Cotrimoxazole, (B) Ciprofloxacin, and (C) Azithromycin among Salmonella Typhi cases in Bangladesh from 1999 to 2020. Consumption data for ciprofloxacin was unavailable from 1999 to 2001. Resistance (-R) or Non-susceptibility (NS) to cotrimoxazole, ciprofloxacin, and azithromycin are presented on the primary Y-axis. The secondary Y-axis displays antibiotic consumption, represented as the Defined Daily Dose (DDD) per 1,000 persons per day (DDD/1,000 day/year).