Bulletin of the World Health Organization

Stimulating the development of national Streptococcus suis guidelines in Viet Nam through a strategic research partnership

Peter Horby a, Heiman Wertheim a, Nguyen Hong Ha b, Nguyen Vu Trung b, Dao Tuyet Trinh b, Walter Taylor a, Nguyen Minh Ha b, Trinh Thi Minh Lien b, Jeremy Farrar a & Nguyen Van Kinh b

a. Oxford University Clinical Research Unit, National Hospital of Tropical Diseases, 78 Giai Phong Street, Hanoi, Viet Nam.
b. National Hospital for Tropical Diseases, Hanoi, Viet Nam.

Correspondence to Peter Horby (e-mail: peter.horby@gmail.com).

(Submitted: 29 May 2009 – Revised version received: 18 September 2009 – Accepted: 22 September 2009 – Published online: 26 January 2010.)

Bulletin of the World Health Organization 2010;88:458-461. doi: 10.2471/BLT.09.067959


Streptococcus suis is a bacterial pathogen that has its natural reservoir in pigs but can infect humans, causing meningitis, septicaemia, endocarditis and arthritis.1 Infection can result in severe disease, with mortality of 3–18%, and hearing loss in up to 66% of survivors.2,3 It is probable that exposure to infected pigs and pork products are the main risk factors.2,4

The first reports of human S. suis infection in south-eastern Asia were in China, Hong Kong Special Administrative Region in 19845 and cases have subsequently been reported in Japan and Thailand.6,7 However, it was not until a large outbreak in Sichuan province, China in 2005 that interest in this pathogen grew8 and S. suis is now increasingly recognized as a major cause of bacterial meningitis in south-east Asia.1,2,4,9 Despite this increased interest, it is probable that the burden of S. suis is considerably under-estimated since clinical awareness is low and many smaller hospitals do not possess a microbiology service. Even in hospitals with a microbiology service, infection may be misdiagnosed as a viridans group streptococcus or Enterococcus.

Considerable scope exists for improving the prevention, recognition and treatment of S. suis. The clinical picture and patient demographics can be distinctive and laboratory identification is not complicated, requiring the use of biochemical tests to differentiate S. suis from other streptococcal species.2,9 S. suis isolates are generally sensitive to penicillin and the risk of the most important long-term sequela, deafness, can be reduced by the use of steroids.3,4 Therefore, S. suis is readily treatable with affordable and accessible drugs. The risk of infection might be reduced through education of people who work with and butcher pigs, through enforcement of changes in butchering practices and through efforts to change dietary habits.

S. suis was detected in southern Viet Nam as early as 19973,4 but by 2007 this important pathogen was still not mentioned in national guidelines on the diagnosis, treatment and prevention of meningitis. This paper presents our experience of rapidly influencing the development of national guidelines through a research partnership between an influential national institute and an external academic group.

Local setting

Viet Nam is a low-income country that has made impressive achievements in improving health. The health-care system has seen considerable changes, with improved facilities and services, increased autonomy and the implementation of a social health insurance system. However, significant challenges remain, particularly in the areas of cost containment and quality of care. Although several laws have given more autonomy to health departments and health facilities, parts of the system remain fairly centralized, with the Ministry of Health retaining responsibility for many functions, including the development of guidance on disease prevention, diagnosis and clinical management.


The burden of S. suis in Viet Nam is difficult to quantify and awareness is low since most hospitals do not have the diagnostic resources to identify the organism. Although the international biomedical literature contains some publications on S. suis, language barriers and access difficulties mean that the information is not readily available to most Vietnamese clinicians. Alerting clinicians to the importance of S. suis is difficult since channels for disseminating knowledge and guidance are limited. The Ministry of Health is active in developing clinical and public health guidelines through central directives but there are few other stakeholders, such as professional associations, producing guidelines.


The National Hospital for Tropical Diseases (NHTD) is a 200-bed tertiary care centre for infectious diseases in northern Viet Nam. Unlike most hospitals in Viet Nam, NHTD is a specialist hospital under the direct control of the Ministry of Health and is close to the Ministry of Health both geographically and managerially. In 2006 a research partnership was established between NHTD and the Oxford University Clinical Research Unit. A key reason for developing this partnership was to maximize the impact of clinical research through the close relationship between NHTD and the Ministry of Health.

Shortly after the research unit and laboratories were operational at NHTD, an investigation was started into cases of purulent meningitis identified as being caused by Aerococcus viridans or Streptococcus species. It was decided to re-test a selection of stored strains from late 2006 with API 20 Strep (Biomerieux, Lyon, France). The strains were identified as S. suis and this was confirmed by a S. suis serotype 2 polymerase chain reaction.4 After this finding, it was decided that all streptococci isolates from blood and cerebrospinal fluid should be tested with API 20 Strep. Real time polymerase chain reaction diagnostics for S. suis were introduced in April 2007.


After introducing enhanced diagnostics, S. suis became the most commonly identified pathogen in adults with suspected meningitis at NHTD. Between January and May 2007, 19 cases of S. suis meningitis were detected and 50 cases were detected for the entire year (Fig. 1). Of 562 cerebrospinal fluid specimens submitted for microbiological analysis during 2007, we identified 11 Cryptococcus neoformans, 3 Streptococcus pneumoniae, 3 Streptococcus species, 1 Enterobacter cloace and 43 S. suis. An additional 7 patients had S. suis identified in blood cultures.9 Of the 50 patients with S. suis, 26 (52%) recovered completely, 21 (42%) recovered with sequelae and 3 patients died. Hearing loss was the most common sequela (38%).

Fig. 1. Number of Streptococcus suis cases diagnosed per month during 2007 at the National Hospital for Tropical Diseases
Fig. 1. Number of <em>Streptococcus suis</em> cases diagnosed per month during 2007 at the National Hospital for Tropical Diseases

The identification of S. suis as the commonest cause of bacterial meningitis at NHTD was reported to the Ministry of Health in May 2007. Within 4 months, the Ministry of Health issued national guidelines on S. suis (Ministry of Health Decision 3065/QD-BYT, dated 16 August 2009). The guidelines contained a description of the clinical syndrome and the organism, recommendations on microbiological diagnosis, a recommendation to treat suspected cases with ampicillin, a third generation cephalosporin and intravenous corticosteroids (methylprednisolone 0.5–1 mg/kg/day). These guidelines were sent to all hospitals in Viet Nam and received a lot of local media coverage. Although the treatment of suspected bacterial meningitis with ampicillin and a cephalosporin was standard practice before promulgation of the national S. suis guidelines, steroids were not recommended routinely for non-tuberculous bacterial meningitis in adults.3

In August 2007, the National Institute of Hygiene and Epidemiology, the local World Health Organization office and our research group held a meeting to discuss the public health challenges of S. suis. The National Institute of Hygiene and Epidemiology and three regional public health institutes then received training on the identification of S. suis from human clinical specimens.

In November 2007, the preventive medicine branch of the Ministry of Health issued updated guidelines on the surveillance and control of communicable diseases. For the first time the guidelines included a section dedicated to S. suis. They included advice not to slaughter or consume sick pigs, to cover wounds and wear protective equipment during slaughtering of pigs, and not to consume undercooked pork.

The Ministry of Health has not established a surveillance system for S. suis so it is not possible to assess the impact of the guidelines on the incidence and clinical outcome of S. suis infections. However, 44 cases of S. suis were diagnosed at NHTD during 2008, so it is clear that S. suis remains an important health problem. The main lessons learnt are summarized in Box 1.

Box 1. Summary of main lessons learnt

  • Engaging with research partners who have influence with government and policy-makers will maximize the chances of research having an impact locally.
  • Reporting of simple research data directly to policy-makers can have more impact than publishing in the biomedical literature.
  • It is important to identify local ”pathways to policy” and use prominent figures or institutions to lobby for change.


It is well recognized that developed and developing countries both have difficulties in ensuring that research findings influence policies and practices.10 However, in resource-poor settings, where the population is more vulnerable, the opportunity costs greater and the consequences of poor policy more serious, the relative importance of the influence of research on policy may be greater. Demonstration of the relevance of health research is also essential for the grass roots credibility of the research team.

There are many obstacles to translating evidence into policies and practices that have been well documented.10 The causes are diverse and apply both to the researchers and policy-makers.11 In fact, it could be argued that the duty of policy-makers to seek out the best evidence is equally weighted by the duty of publicly funded researchers to seek to influence policy. Rigorous evidence is not enough if the evidence can only be found in inaccessible journals written in impenetrable scientific language. It is not enough to publish and pray for percolation.12 A common difficulty is that the process of policy formulation is often not well characterized or mapped. And for those sitting outside the policy-making forum, channels to formally challenge or develop policy may be obscure. But Ministries of Health often out-source expertise and it can be very useful to identify these spheres of influence. Indeed, policy-makers have identified personal contact with researchers as the most important factor influencing decisions about use of research information.13

Our experience illustrates that a scientific partnership between an international research group and an influential national institute closely linked to government can have an immediate impact on national policies. Identifying and using the pathways to policy and giving evidence prominence through prominent figures can profoundly improve the chances of effecting change.


We thank Constance Schultsz and Ngo Thi Hoa for helping to set up S. suis diagnostics in NHTD. Authors Peter Horby, Heiman Wertheim, Walter Taylor and Jeremy Farrar are also affiliated with the Centre for Tropical Medicine, Nuffield Department of Clinical Medicine, Oxford University, Oxford, England.


This work was supported by the Wellcome Trust UK (grants 081613/Z/06/Z and 077078/Z/05/Z) and the South East Asia Infectious Disease Clinical Research Network (N01-A0-50042).

Competing interests:

None declared.