Background
Typhoid fever is a significant public health problem with annual estimates of 22 million cases and 216,000 deaths worldwide [
1], although the global burden is known to be underestimated, especially in developing countries where the majority of cases likely remain undiagnosed [
2]. Typhoid fever is caused by
Salmonella enterica serotype Typhi (
S. Typhi, a Gram-negative bacterium, transmitted by ingestion of faecally contaminated food or water. Culture from blood or stool remains the gold standard for typhoid diagnosis, but these methods may not be affordable or practical in low-resource settings, where serological methods have historically been used to diagnose typhoid infection. Even when culture is available, these methods can result in low recovery of the organism (40% blood, 37% stool) and are complicated by the use of antibiotics prior to specimen collection [
3]. Clinical presentation varies from a mild illness with low grade fever, malaise and dry cough to a severe clinical picture with abdominal discomfort, altered mental status and multiple complications [
4]. If not treated, typhoid fever may progress to severe complications like delirium, intestinal haemorrhage, bowel perforation, and death. Humans are the only natural host and reservoir.
Typhoid fever outbreaks have been recorded in central and southern Africa, affecting both children and adults alike, including in the Democratic Republic of Congo [
5], Zambia [
6] and Zimbabwe [
4,
7]. In Zimbabwe, more than 1000 cases of typhoid fever have been reported annually since 2011, demonstrating the endemicity of the disease. In 2009 [
8], a typhoid outbreak primarily affecting two densely populated suburbs of Harare, Mabvuku and Tafara was recorded. Poor sanitation and drinking water quality in these areas and other parts of Zimbabwe were the key risk factors for
S. Typhi transmission and outbreaks [
4]. If detected early and treated with appropriate antibiotics the impact of typhoid fever on an individual and the population is greatly minimised. Antimicrobial susceptibility testing of
S. Typhi is therefore of great importance in ensuring correct treatment regimens and for monitoring the emergence of any drug resistant strains. In Zimbabwe the treatment guidelines recommend the management of typhoid fever using ciprofloxacin and ceftriaxone [
4]. An additional concern is the changing patterns of drug susceptibility for circulating strains of Typhi reported worldwide. Murgia et al. [
9] reported that haplotype 58 (H58) is associated with multidrug-resistance to first line drugs, and is the most diffused and rapidly expanding among
S. Typhi population. The H58 haplotype has also been associated with extremely drug resistance (XDR) Typhoid outbreaks in Pakistan [
10]. In addition to the H58 haplotype, S. Typhi with extended β -Lactamase has also been reported in Democratic Republic of Congo (DRC) [
11] .However in 2016 Murgia et al. [
9] reported that haplotype 58 (H58) is associated with multidrug-resistance to these first line drugs, and is the most geographically dispersed and actively spreading
S. Typhi haplotype. Surveillance of H58
S. Typhi in areas endemic for typhoid fever is therefore key in monitoring the development of resistance to first line drugs and the associated treatment choice in order to effectively minimise the associated morbidity and mortality and prevent large scale outbreaks of
S. Typhi occurring [
9].
Laboratory confirmation of enteric pathogens surveillance was established in Zimbabwe in 1995 and typhoid confirmation was limited to a few laboratories used as sentinel sites.
We present a comprehensive analysis of S. Typhi in Zimbabwe identified between 2009 and 2017, for antimicrobial resistance, presence of H58 haplotype and molecular epidemiology, including strain relatedness both within Zimbabwe and with strains from neighbouring countries.
Discussion
To provide guidance on appropriate treatment choice in order to minimise the morbidity and mortality associated with typhoid fever and prevent large scale outbreaks a phenotypic and genotypic analysis was conducted on S. Typhi isolates collected from 2012 to 2017. To determine the development of drug resistance to first line antibiotics for typhoid fever and the prevalence of
Salmonella enterica serotype Typhi (
S. Typhi) H58 haplotype standardised methodology were performed.
S. Typhi isolates showed a changing pattern in antimicrobial susceptibility across the years for which isolates were available (2012–2017). Fluoroquinolones such as ciprofloxacin are recommended by WHO [
19], as they are reliably effective, inexpensive and well-tolerated drugs for the treatment of typhoid fever [
19]. Ciprofloxacin is used as a first line treatment drug for typhoid in Zimbabwe [
4]. In this study, an increase in resistance to ciprofloxacin was observed from the 2014 (4.2%) to 2017 (22.0%) isolates (Fig.
2). The ciprofloxacin-resistant isolates were from Harare with Budiriro and Glenview having the highest number in 2016. These ciprofloxacin-resistant isolates have spread to others areas like Mbare, Kambuzuma, Kuwadzana and Hatcliff. Also an increase in intermediate resistance (0.5 mg/L) of ciprofloxacin was recorded from 2014 to 2017 (Fig.
2). Intermediate resistance was observed in 5 isolates from Mutare in 2016. An MIC value 0.5 mg/L was recorded in all isolates showing intermediate resistance meaning ciprofloxacin may be effective at higher doses. Though fluoroquinolone resistance is chromosomally mediated [
6], selective pressures exerted by the overuse of these drugs may result in such isolates becoming more common in the future. This may explain the increase in ciprofloxacin resistance in
S. Typhi isolates in Zimbabwe (especially in Harare) where the antibiotic is used as a broad-spectrum drug to treat many diseases. Resistance and intermediate resistance to ciprofloxacin has been reported from many regions worldwide, including Kenya [
20], Cambodia [
21], Bangladesh [
22] and South Africa [
23]. A sharp increase in tetracycline resistance was observed from 2012 (11.0%) to 2017 (46.3%) (Fig.
2). All the tetracycline-resistant
S. Typhi isolates from 2016 were isolated in Harare. In Zimbabwe, tetracycline is not used as a drug of choice for the treatment of typhoid fever but it is heavily used in the poultry industry and may be indicative of human exposure to residual antibiotics in the food chain. Strains that acquire this type of resistance also become co-resistant to other antibiotics such as Beta-lactams and fluoroquinolones, if resistance is plasmid-borne [
24]. A correlation between tetracycline and ciprofloxacin resistance was observed (Table
2). In this study, all the ciprofloxacin resistant strains were susceptible to ceftriaxone and azithromycin (Fig.
2).
All the isolates from 2012 to 2017 were susceptible to ceftriaxone (Fig.
2). Intravenous ceftriaxone is a drug of choice for typhoid treatment in Zimbabwe [
4]. In addition, it is used to treat typhoid fever due to resistant bacteria [
19]. Resistance to older first-line drugs for
S. Typhi such as ampicillin and chloramphenicol remained constantly high ranging from 83.3 to 100% (Fig.
2). In a similar study done in India, 75.5% of
S. Typhi isolates were resistant to amoxicillin [
25]. Ampicillin resistance can be used to predict resistance of
S. Typhi to amoxicillin [
13]. Globally, extremely high resistance to ampicillin and chloramphenicol, [
5,
25,
26] has motivated for the use of alternate antibiotics for typhoid fever, but our results suggest that increasing ciprofloxacin resistance may soon render this antimicrobial ineffective in typhoid fever control programmes. Our findings warrant an adjustment in typhoid treatment guidelines and a shift towards evidence based management and routine antimicrobial resistance surveillance programs in Zimbabwe.
Multidrug resistant strains are a major therapeutic concern for physicians in developing countries. Contributing factors may include antimicrobial misuse and inappropriate prescribing practices [
27] as well as intrinsic plasmid-mediated factors [
22,
28,
29]. Eleven multidrug resistance patterns were observed and the most common pattern, resistotype A (resistance to ampicillin-chloramphenicol) was exhibited by 172 (62.3%) isolates (Table
2). The high level of resistance to first-line antimicrobials for treatment of typhoid fever is worrisome, as 243
S. Typhi isolates (88.0%) were resistant to two or more antimicrobials and 150 of the 161 tested belonged to the H58 haplotype. Results of the study suggest high prevalence of MDR H58 haplotype in clinical
S. Typhi isolates in Zimbabwe. According to a study done by Wong et al. [
29] 63% of
S. Typhi isolates belonged to H58 lineage in Eastern and Southern Africa. The H58 lineages I and II were detected in Kenya, Tanzania, Malawi and South Africa [
30], neighbouring countries to Zimbabwe.
Outbreaks of MDR
S. Typhi strains have been reported around the world. In 2011, researchers in Malawi isolated MDR H58- lineage
S. Typhi in Blantyre, Malawi [
31]. Multidrug resistant strains of S. Typhi have been reported from many African countries, including Kenya, Uganda, Tanzania and Ghana [
32]. Due to the presence of MDR and quinolone-resistant
S. Typhi isolates [
33], it has been recommended that developing countries should use azithromycin as a first-priority drug.
PFGE analysis was used for molecular subtyping of isolates and to determine relatedness of 91
S. Typhi isolates from 2009 to 2016. PFGE is a powerful molecular biology technique which has provided important insights into the epidemiology and population biology of many pathogens in the world [
34]. In the present study, 12 PFGE subtypes were shown amongst the 91 isolates. PFGE is regarded as one of the most reliable techniques for discriminating different strains of
S. Typhi [
35,
36]
. The same subtype observed for the 2009 Mabvuku isolates was consistently seen in South African samples of 2006, 2008, 2009, 2010, 2011, 2012, 2016 and 2012 (Zimbabwe) (Additional file
1: Figure S1, Fig.
3) suggesting that the strain is circulating in Zimbabwe and South Africa. The Mabvuku 2009 subtype was noted to be circulating in Harare (2013; 2016), Mutawatawa (2014), Chitungwiza (2012), Mutare (2016), Rusape (2014) and Inyanga (2013), demonstrating a relationship between isolates across a wide area and timeline. These findings point toward Mabvuku as the source of 2009 typhoid resurgence in Harare, Zimbabwe. Some PFGE subtypes were unique to particular towns such as Masvingo, Mutare and Chegutu.
Resistance traits (e.g. fluoroquinolone resistance) were highly subtype-specific, suggesting predominantly subclonal distribution. Although the proportion of all cases with an available isolate is small due to the sampling process within a country these findings still remain key in advancing our understanding of the genetic structure, ecology, geographic distribution, and emergence of this widely disseminated drug–resistant pathogen, which represents a growing public health threat. It does however point to the need to improve sample collection processes for individuals suspected of having typhoid fever. Our research findings also revealed that there is a common
S. Typhi strain circulating in Zimbabwe, South Africa, Zambia and Tanzania as evidenced by a common subtype in the isolates (Fig.
3). Imanishi et al. [
8] also observed that there was a common subtype circulating in Zimbabwe, Malawi and Tanzania when they analyzed their 2009 and 2011 isolates. Similarities between PFGE subtypes from multiple countries may be the result of population movements in Zimbabwe, Zambia, South Africa and Tanzania where people move easily from one country to another.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.