Antimicrobial Resistance Surveillance among Nosocomial Pathogens in South Africa: Systematic Review of Published Literature

Article Outline

• Abstract
• 1. Introduction
• 2. Methodology
  • 2.1. Online search strategy
  • 2.2. Search engines, dates of publications, and search words used
  • 2.3. Manual search strategy
• 3. Results
  • 3.1. Antimicrobial resistance surveillance for invasive pathogens in South Africa
  • 3.2. Description of study settings and study designs12,15–22
  • 3.3. Description of microbiological methods12,16–19,22
  • 3.4. Resistance rates for different pathogens
    • 3.4.1. Staphylococcus aureus12,16–19,21,22
    • 3.4.2. Klebsiella pneumoniae12,16–22
    • 3.4.3. Pseudomonas aeruginosa16,17,21
  • 3.5. Presence of extended-spectrum beta-lactamases
• 4. Discussion
• 5. Conclusions
• Authors’ contributions
• Acknowledgments
• References
• Copyright

There has been a significant increase in the prevalence of antimicrobial drug resistance in sub-Saharan Africa. This may increase health-care costs due to patients' needs for more diagnostic tests, longer hospitalization, and poor outcome. Therefore, monitoring systems for resistance patterns are needed to effectively minimize poor outcome. A systematic review was conducted to find out the prevalence of antimicrobial drugs' resistance among Staphylococcus aureus, Klebsiella pneumoniae, and Pseudomonas aeruginosa, and to understand whether or not such data were part of an ongoing surveillance system for nosocomial infections in South Africa. An online search of main databases, including Cochrane Library, PUBMED, and MEDLINE, was done using the following search terms: “antimicrobial resistance” and “surveillance”; “antimicrobial susceptibility” and “surveillance”; Staphylococcus aureus or Klebsiella pneumoniae or Pseudomonas aeruginosa; “nosocomial” or “hospital acquired”; or South Africa or Africa. We also performed manual search of local conferences, theses, and dissertations to identify relevant articles. In total, 41 manuscripts were identified of which eight were analyzed. There is no evidence of any ongoing antimicrobial resistance surveillance for nosocomial pathogens in South Africa. Data reported in this review seem to have been analyzed on an ad hoc basis and do not show a particular resistance pattern; however, data show evidence of resistance to commonly used antimicrobial drugs in this population: for S aureus, resistance to cloxacillin was 29% and to erythromycin 38%; for K pneumoniae, resistance to ciprofloxacillin was 35% and to ampicillin 99%; and for P aeruginosa, the mean resistance to ciprofloxacillin was 43% and to amikacin 35%. Surveillance of antimicrobial resistance is essential to better understand the complexity of antimicrobial resistance development. Such evidence would be used in developing an effective surveillance program to monitor patterns and trends of resistance over time.

Key words: antimicrobial resistance, bacterial pathogens, nosocomial infections, surveillance

Back to Article Outline

1. Introduction 

Antimicrobials are essential for the treatment of infectious diseases. However, a high prevalence of resistance impacts patient outcomes negatively. Antimicrobial resistance increases health-care costs due to a need for more diagnostic tests, additional drugs for treatment, and longer duration of hospitalization.¹, ² Therefore, the emergence and spread of antimicrobial-resistant organisms from hospital to the community is a growing public health challenge in South Africa and worldwide. It is associated with a high level of morbidity and mortality, and for this reason, antimicrobial resistance requires effective monitoring to determine patterns and trends over time.³, ⁴, ⁵, ⁶ For South Africa, such information is particularly important because of the HIV/AIDS epidemic and increased antimicrobial consumption due to frequent episodes of opportunistic infections.

Antimicrobial resistance surveillance is crucial for evaluating the use of empirical antimicrobials for treatment.⁷ Continuous monitoring, and a better understanding of the profile and magnitude of antimicrobial resistance are therefore required. This will help address the problem of increasing rates of antimicrobial resistance in South Africa. The European Antimicrobial Resistance Surveillance System (EARSS) is an electronic laboratory information system that has been used as a tool for identifying emerging antimicrobial resistance.⁸ In South Africa, an equivalent national surveillance system to monitor the status of antimicrobial resistance for nosocomial pathogens has not yet been established. For this reason and as an interim exercise, this review was initiated to gather scientific evidence of the extent and patterns of antimicrobial resistance in selected hospital-acquired pathogens in South Africa.

Back to Article Outline

2. Methodology 

2.1. Online search strategy 

A comprehensive search of biomedical databases was carried out to find all relevant manuscripts published in English. The search aimed at identifying relevant peer-reviewed epidemiological studies that would provide adequate information on antimicrobial surveillance initiatives in South Africa.

2.2. Search engines, dates of publications, and search words used 

The following search terms were using: “antimicrobial resistance” and “surveillance”; “antimicrobial susceptibility” and either “surveillance” or “Staphylococcus aureus” or “Klebsiella pneumoniae” or “Pseudomonas aeruginosa”; “nosocomial” or “hospital acquired”; or “South Africa” or “Africa.” We focused on searching pathogen-specific literature and data for this review using manuscripts identified through such an extensive search of the following databases: Cochrane Library (July 2011); MEDLINE (1966 to July 2011); African Journals Online (AJOL) (1980 to July 2011), EMBASE (1980 to July 2011); and LILACS (1982 to July 2011) on www.bireme.br.

2.3. Manual search strategy 

We also carried out a manual search and review of the reference lists of the identified articles. Additionally, as findings of studies are not always published conventionally, we manually searched the abstracts and proceedings within the past 10 years for the following conferences: “OIE International Conference on Antimicrobial Resistance,” “Conference on Antibiotic Resistance Prevention and Control” (ARPAC), “Public Health Association of Southern Africa” (PHASA), “Federation of Infectious Diseases Society of South Africa” (FIDSSA), “Global Antimicrobial Resistance Program” (GARP), “Congress of the European Society of Clinical Microbiology and Infectious Diseases” (ESCMID), and the “Congress of the International Society for Infectious Diseases.” Such conference proceedings outline major group sessions for microbiology and infectious disease specialists working within the field of antimicrobial resistance. We did not obtain any relevant data from these searches. In addition, informal approaches were made to individuals and organizations within the field of hospital infection control and antimicrobial resistance surveillance for information regarding unpublished data, dissertations, and theses.

This search yielded four of the eight papers that were included for analysis. Data for rates of antimicrobial resistances were presented as means.

Back to Article Outline

3. Results 

3.1. Antimicrobial resistance surveillance for invasive pathogens in South Africa 

A good surveillance system for antimicrobial resistance monitoring should involve ongoing collection and collation of both clinical and microbiological data, with an emphasis on timeliness, accuracy, consistent and standardized methods of collection, and analysis, using a centralized laboratory with appropriate control measures, with a focus on reporting on nosocomial pathogens. Such a system has not been present in South Africa. However, although different methods were used, they were all approved by the National Committee for Clinical Laboratory Standards (NCCLS), predecessor of the Clinical Laboratory Standards Institute (CLSI), and therefore suitable for trend analysis e.g. ciprofloxacin resistance in K. pneumoniae increased in academic hospitals from 18% (24/1324 isolates) in 1999 to 28% (498/1778) in 2007.

From the included studies, lack of clinical data and quality assurance information are deficiencies requiring attention; nonetheless, some steps have been taken to contain resistance development. Prudent use of antimicrobials (antimicrobial stewardship) has been looked at through the South African Society of Clinical Microbiology, formerly the National Antimicrobial Surveillance Forum (NASF), using passively collating antimicrobial data in public through the National Health Laboratory Services (NHLS) and in private health-care sectors through private microbiology laboratories.

The Antibiotic Study Group of South Africa has been active since 1976⁹; this group joined private sector surveillance in 2002 as NASF, meeting and sharing information, and several publications in the area of antimicrobial resistance have been released.⁹, ¹⁰, ¹¹, ¹² More recently, the Group for Enteric, Respiratory and Meningeal Diseases Surveillance (GERMS-SA), an established entity within the National Institutes for Communicable Diseases (NICD), has been established, which operates in all nine provinces, focusing on surveillance of community-acquired pathogens and monitoring resistance profiles. As of 2010, a surveillance to monitor resistance among S aureus and K pneumoniae was established as part of GERMS-SA. Another initiative was introduced in KwaZulu Natal for surveillance of Escherichia coli in 2000/2001,¹³ and the Veterinary Surveillance of Antimicrobial Resistance in South Africa has been involved in monitoring resistance among zoonotic infections.¹⁴ Table 1 illustrates hospitals and laboratories that contributed antimicrobial susceptibility data for the studies that were included in this review.

Table 1. Public and private sector laboratories that participated in antimicrobial susceptibility data over the period 2000–2011

NHLS = National Health Laboratory Service.

3.2. Description of study settings and study designs¹², ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²² 

A total of 41 manuscripts were identified: 26 identified through database searches and 14 through manual searches in libraries and among personal contacts. Twenty-four manuscripts were excluded, leaving 18 that had full-text article reviews to further assess for eligibility, and 10 more were further excluded. Eight manuscripts published between 2000 and 2011 were identified and included in this review (Figure 1). Of the eight manuscripts, five were published prior to 2007. All manuscripts identified for this review included susceptibility data from only four of the nine provinces of South Africa. Five of these studies were from public sector tertiary hospitals and three were from private sector laboratories, predominantly from urban settings across South Africa (Table 2).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1 

  Flow diagram of antimicrobial resistance studies included in the review. Note. From PRISMA: www.prisma-statement.org.

Table 2. Characteristics of antimicrobial resistance studies in South Africa

BAL = bronchial alveolar lavage; CSF = cerebral spinal fluid; EC = E coli; HI = Haemophilus influenzae; KP = K pneumoniae; MEF = middle ear fluid; NHLS = National Health Laboratory Service; PA = P aeruginosa; SA (pathogen) = S aureus; SA = South Africa; SP = Streptococcal pneumonia.

Seven of these studies produced results from surveillance data aggregated from more than seven sites nationwide, while one study produced results from surveillance data from 16 hospitals within KwaZulu Natal province. None of the eight studies detailed the study design used, other than stating that the study was “multi-site and used data of blood culture isolates from microbiology laboratories.” Only one study used isolates from respiratory aspirates²⁰; all except one study from various public sector hospitals within KwaZulu Natal province used retrospective laboratory data²¹ (Table 2).

3.3. Description of microbiological methods¹², ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²² 

Seven of the studies used data from blood and cerebral spinal fluid (CSF) cultures¹², ¹⁶, ¹⁷, ¹⁸, ¹⁹; one study used data from respiratory aspirates.²² The methodologies of antibiotic susceptibility testing were described in seven studies, all of which mentioned the use of the CLSI breakpoints, formerly NCCLS, to determine antimicrobial susceptibilities. Two studies described in detail other methods used for susceptibility testing of various antibiotics such as Kirby–Bauer disk diffusion, Broth microdilution, E-test, and use of automated Vitek 2 system.¹⁷, ²⁰ Only one study mentioned quality control in identification and susceptibility testing as per CLSI recommendations.¹⁷ All studies used only one sample per patient; hence, duplicate samples were excluded to minimize over-representation of the cases that had multiple and frequent cultures. Two studies that reported antimicrobial susceptibility of respiratory tract pathogens mentioned intermediate- and high-level resistance for such organisms.¹⁸, ²⁰

3.4. Resistance rates for different pathogens 

Susceptibility data for S aureus were reported in seven studies (Table 2). Five of these studies were from public sector laboratories and two from private sector laboratories.¹², ¹⁶, ¹⁸, ¹⁹, ²² Geographically all studies identified were performed in urban areas except one study done in Durban, which included isolates from district and regional hospitals. Specimen types included blood and CSF, except one study that included respiratory aspirates (Table 2). The resistance rate of S aureus to cloxacillin was 29%, erythromycin 38%, and gentamicin 20%, and methicillin resistance (MRSA) was 33%. No resistance has been reported to linezolid since its introduction in 2000, while frequency of resistance to glycopeptides is uncertain due to disagreement on optimization of vancomycin susceptibility testing (Figure 2).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 2 

  Prevalence of antimicrobial resistance among S aureus. Note. From seven published studies between 2000 and 2009. ∗Different methods used to determine MRSA status (Cloxacillin resistance of 29% vs 33% MRSA).

Most studies that reported on susceptibility patterns for K pneumoniae were published by the Antibiotic Study Group that used data mostly from large public sector academic hospitals that provide services to a diverse population group. Clinical isolates were predominantly from blood and CSF culture (four studies), blood culture only (one study), blood and urine culture (one study), and respiratory aspirates (one study). The resistance of K pneumoniae to ciprofloxacillin was 35%, cefuroxime 52%, gentamicin 50%, and ampicillin 99%. Resistance was almost nonexistent for imipenem, meropenem, and moxifloxacin (Figure 3).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 3 

  Prevalence of antimicrobial resistance among K pneumoniae. Note. From eight published studies from 2000 to 2009.

Three studies reported resistance rates for P aeruginosa, two of which were from blood culture isolates and one from nonspecific sources.¹⁶, ¹⁷, ²¹ The resistance among P aeruginosa to ciprofloxacillin was 43%, gentamicin 50%, amikacin 35%, and aztreonam 42%. Resistance to polymyxin was <5% and was reported in a single study.¹⁶, ¹⁷, ²¹ Resistance rates to almost all drugs tested were greater than 30% (Figure 4). A study conducted by Perovic et al using data from 1998 to 1999 at Chris Hani – Baragwanath Hospital showed that there was an association between P aeruginosa bacteremia and outbreaks caused by multiple-resistant genotypes. In this study, the proportion of nosocomially acquired infection was 57.1%.²⁴ The resistance profiles and incidence of disease are likely to have changed during the 10-year period, and the current status may be different but is unknown. This review shows high resistance rates of P aeruginosa to most conventional antibiotics.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 4 

  Prevalence of antimicrobial resistance among P aeruginosa. Note. From three published studies from 2005 to 2009.

3.5. Presence of extended-spectrum beta-lactamases 

Seven studies reported on extended-spectrum beta-lactamases (ESBLs) in K pneumoniae. In academic hospitals the rates of ESBLs increased from 33% (436/1324) in 1999 to 49% (869/1778) in 2007. These studies used the double-disk method and reported resistance rates as high as 59% and 62% in private hospitals and public sector hospitals, respectively. A study conducted by Essack Sabiha at a teaching hospital in Durban between 1994 and 1996 investigated ESBL-mediated resistance in South African nosocomial origin of K pneumoniae and demonstrated that each of the isolates expressed 1–6 beta-lactamases.²³

Back to Article Outline

4. Discussion 

This systematic review assessed the prevalence of resistance to commonly used antimicrobials as well as whether or not such data were part of an ongoing surveillance system for nosocomial infections in South Africa. We found that no national surveillance system exists that collates and collects data year on year to assess trends and resistance patterns for nosocomial pathogens. In addition, we found that the overall prevalence of resistance to antimicrobials used for empirical treatment is high. Except for polymyxin, with a resistance rate of <5%, most other antibiotics showed high prevalence of P aeruginosa resistance to commonly available antimicrobials. The study found a low level of resistance among K pneumoniae to moxifloxacin and carbapenems, and a high pattern of resistance to other classes of antimicrobials that are commonly prescribed. S aureus showed no resistance to teicoplanin, vancomycin, and linezolid, but high resistance to other classes of antimicrobials. This is similar to the resistance pattern in Central African countries, as shown in a review by Vlieghe et al,²⁵ even though their study focused mostly on community-acquired pathogens.

Several limitations have been observed in this study: Firstly, studies included in this review reported laboratory data on antimicrobial-resistant isolates, with no clinical data; hence, they could not link resistant isolates to clinical findings. Secondly, most studies aggregated data from different laboratories which employed varied laboratory techniques. This was not ideal for surveillance purposes but all methods were NCCLS/CLSI approved. Thirdly, data used were collected retrospectively, except for a single study by Brink et al that collected data prospectively.¹⁷ Use of retrospective data has several limitations, including incomplete data that are subject to numerous biases. Fourthly, most, if not all, studies lacked demographic data; hence, it was difficult to compare community-acquired versus hospital-acquired infections. Lastly, variation in clinical specimens, taking practices between different institutions, might alter representativeness of data reported from these various studies. Furthermore, this study included invasive pathogens from blood cultures as well as pathogens from respiratory specimens and, in the case of P. aeruginosa, also from other sources, including burns.”

In spite of the limitations mentioned above, there is growing evidence of escalating rates of antimicrobial resistance to several conventional antimicrobials. Even though vancomycin resistance is still negligible, ESBL and MRSA rates are high in these urban academic centers and private institutions. This emphasizes the fact that surveillance is essential to further our understanding of antimicrobial resistance development and how it relates to prescription practice.²³, ²⁵ Such undertaking will pave the way for designing interventions that could overcome resistance development to established antimicrobial agents.

Back to Article Outline

5. Conclusions 

Evidence indicates that antimicrobial resistance rate to nosocomial pathogens are generally high in South Africa. This is an emerging threat to public health and clinical management of patients with such infections in the face of dwindling antimicrobial development. We believe that a good surveillance system would enhance effective monitoring of emerging resistance and changes in resistance profiles, and identify significant differences in trends and distribution of antimicrobial resistance.

Back to Article Outline

Authors’ contributions 

PN searched the relevant papers and drafted the manuscript. JM proposed the topic for this review and helped draft the manuscript. OP and HK participated in critically reviewing the manuscript on intellectual content and scholarly writing. All authors read and approved the final manuscript.

Back to Article Outline

Acknowledgments 

We would like to acknowledge the contribution of Professor Essack Sabiha, Dr Colleen Bamford, Duduzile Mditshwa, Angeline Zwane, and Thando Mabeqa in identifying relevant articles and Professor Stanley Luchters for critically reviewing the draft manuscript.

Back to Article Outline

References 

1. Goering R, Nord CE, Hare R, Sabatelli F. Ziracin susceptibility testing group. In vitro activity of evernimicin and selected antibiotics against methicillin-resistant staphylococci: a 24-country study. Clin Microbiol Infect. 2000;6:549–556
   • View In Article
   • CrossRef
2. Jeena P, Thompson E, Nchabeleng M, Sturm A. Emergence of multi-drug-resistant Acinetobacter anitratus species in neonatal and paediatric intensive care units in a developing country: concern about antimicrobial policies. Ann Trop Paediatr. 2001;21:245–251
   • View In Article
   • MEDLINE
   • CrossRef
3. Roca A, Quinto L, Abacassamo F, Morais L, Valles X, Espasa M, et al Invasive Haemophilus influenzae disease in children less than 5 years of age in Manhica, a rural area of southern Mozambique. Trop Med Int Health. 2008;13:818–826
   • View In Article
   • CrossRef
4. Sigauque B, Roca A, Sanz S, Oliveiras I, Martínez M, Mandomando I, et al Acute bacterial meningitis among children, in Manhica, a rural area in southern Mozambique. Acta Trop. 2008;105:21–27
   • View In Article
   • CrossRef
5. Shapiro RL, Kumar L, Phillips-Howard P, Wells JG, Adcock P, Brooks J, et al Antimicrobial-resistant bacterial diarrhea in rural western Kenya. J Infect Dis. 2001;183:1701–1704
   • View In Article
6. Klugman KP. Emerging infectious diseases—South Africa. Emerg Infect Dis. 1998;4:517–520
   • View In Article
   • MEDLINE
7. Marais E, Aithma N, Perovic O, Oosthuysen WF, Musenge E, Duse AG. Antimicrobial susceptibility of methicillin-resistant Staphylococcus aureus isolates from South Africa. S Afr Med J. 2009;99:170–173
   • View In Article
8. The European Antimicrobial Resistance Surveillance System. EARSS manual. 2005. Available from: http://www.rivm.nl/earss/. [accessed 10.01.10].
   • View In Article
9. Naude duT E, Van den Ende J, Botha P, Forder A, Hyland J, de Klerk HC, et al A multicentre study on the susceptibility of a variety of bacteria to cephalothin, cefamandole, tobramycin and gentamicin. S Afr Med J. 1977;52:798–800
   • View In Article
10. Van den Ende J, Rotter MI. An analysis of blood culture isolates from 7 South African teaching hospital centres. S Afr Med J. 1986;69:89–93
    • View In Article
11. Antibiotic Study Group of South Africa . Number of isolates and antibiotic resistance from seven academic training hospitals in South Africa. S Afr Med J. 1986;80:366
    • View In Article
12. Antibiotic Study Group of South Africa . Susceptibility of invasive pathogens from academic in hospitals in South Africa to selected antimicrobial agents. South Afr J Epidemiol Infect. 2000;15:51–55
    • View In Article
13. Mocktar C, Govinden U, Sturm AW, Essack S. Complexity and diversity of beta-lactamase expression in inhibitor-resistant Escherichia coli from public hospitals in KwaZulu-Natal, South Africa. South Afr J Epidemiol Infect. 2009;24:29–33
    • View In Article
14. Nel H, van Vuuren M, Swan GE. Towards the establishment and standardization of a veterinary antimicrobial resistance surveillance and monitoring programme in South Africa. Onderstepoort J Vet Res. 2004;71:239–246
    • View In Article
    • MEDLINE
15. Holloway K, Mathai E, Gray A, Chandy S, Essack SY, Gous A, et al Surveillance of antimicrobial resistance in resource-constrained settings—experience from five pilot projects. Trop Med Int Health. 2011;16:368–374
    • View In Article
    • CrossRef
16. Bamford C, Badenhorts I, Duse AG, Hoosen AA, Oliver S, Perovic O, et al Antimicrobial susceptibility patterns of selected invasive pathogens from public sector hospitals in South Africa. South Afr J Epidemiol Infect. 2007;2009(24):28–30
    • View In Article
17. Brink A, Moolman J, da Silva MC, Botha M. National Antibiotic Surveillance Forum. Antimicrobial susceptibility profile of selected bacteraemic pathogens from private institutions in South Africa. S Afr Med J. 2007;97:273–279
    • View In Article
18. Crewe-Brown HH, Coovadia Y, Dove MG, Hanslo D, Hoosen AA, Kooornhof HJ, et al Susceptibility of invasive pathogens from academic in hospitals in South Africa to selected antimicrobial agents for the year 2000. South Afr J Epidemiol Infect. 2001;16:91–95
    • View In Article
19. Sein PP, Hoosen AA, Crewe-Brown HH, Coovadia Y, Dove MG, Heidi O, et al Antimicrobial susceptibility profile of selected invasive pathogens from academic in hospitals in South Africa for the year 2001–2004. South Afr J Epidemiol Infect. 2005;20:85–89
    • View In Article
20. Liebowitz LD, Slabbert M, Huisamen A. National surveillance programme on susceptibility patterns of respiratory pathogens in South Africa: moxifloxacin compared with eight other antimicrobial agents. J Clin Pathol. 2003;56:344–347
    • View In Article
21. Essack SY, Connolly C, Sturm WA. Antibiotic use and resistance in public-sector hospitals in KwaZulu-Natal. S Afr Med J. 2005;95:865–870
    • View In Article
22. National Antimicrobial Surveillance Forum . Surveillance data: National antimicrobial surveillance forum—private susceptibility data July–December 2007. South Afr J Epidemiol Infect. 2008;23:44–48
    • View In Article
23. Essack SY. Unprecedented resistance to B-lactam antibiotics evident in Durban, South Africa—surveillance-based antibiotic policies imperative!. South Afr J Epidemiol Infect. 2000;15:48–50
    • View In Article
24. Perovic O, Koornhof HJ, Crewe-Brown HH, Duse AG, van Nierop W, Galpin JS. Pseudomonas aeruginosa bacteraemia in an academic hospital in South Africa. S Afr Med J. 2008;98:626–632
    • View In Article
25. Vlieghe E, Phoba MF, Tamfun JJM, Jacobs J. Antibiotic resistance among bacterial pathogens in Central Africa: a review of the published literature between 1955 and 2008. Int J Antimicrob Agents. 2009;34:295–303
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (490 KB)
    • CrossRef