Primary Prevention of Acute Rheumatic Fever and Rheumatic Heart Disease

Primordial Prevention

Acute rheumatic fever (ARF) and rheumatic heart disease (RHD) are predominantly diseases of social, environmental, and economic poverty. Primordial prevention, defined as improving socioeconomic and living conditions, and having well-organized, effective health systems, is arguably the most important and effective population-based strategy for prevention of both ARF and RHD.

Even before the introduction of penicillin in the 1940s and 1950s, the incidence of ARF was declining and largely disappeared from countries as they developed economically. The best recorded data come from Denmark, where a dramatic decline in ARF incidence over a 100 year period from 200 per 100,000 population in 1862, to 11 per 100,000 in 1962 was documented, with the decline largely occurring before the introduction of penicillin ( Fig. 10.1 ).

Deaths from ARF similarly declined in the United States between 1910 and 1977, the steepest decline seen before penicillin was introduced ( Fig. 10.2 ).

Early in the 20th century, researchers recognized the link between improved housing and hygiene, less household crowding, improved nutrition, and declining rates of many infectious diseases including ARF—although the specific environmental and/or host factors contributing to the decline in ARF remain elusive. Social and economic improvements coupled with better healthcare systems and access to effective antibiotic treatment for group A streptococcus (GAS) infections have led to ARF becoming a rare disease in almost all high-income countries, although remaining endemic and significant in many low- and middle-income countries. In more recent years, case-control and cohort studies have provided some empirical evidence of an association between poor housing conditions and household crowding and ARF, as described in Chapter 1 . However, as noted, although poverty and social deprivation are strong predictors of developing ARF, it is difficult to identify what are the most important modifiable proximal risk factors where preventive actions could be focused.

Effectiveness of Primary Prevention

Primary prevention of ARF/RHD is defined as early diagnosis and prompt treatment of GAS infections to prevent the first episode of ARF. Vaccination against GAS will be an important primary prevention intervention, once developed (see Chapter 14 ).

To be implemented as a primary healthcare policy, primary prevention needs to be effective in reducing the incidence of ARF. A simple, reliable, affordable, and sensitive method for diagnosis, and available treatment of primary GAS infections, the most common of which is GAS pharyngitis, is required. In recent years, evidence of the contribution of GAS skin infections to the development of ARF in some countries such as Australia, New Zealand, and Fiji has become stronger (see Box 10.1 for clinical features). Although the role of skin infections is less clear in other countries , consideration of GAS skin infection management as a preventive measure for ARF may need to be evaluated in certain settings.

Prompt use of antibiotics is effective in reducing the incidence of ARF following an episode of suspected GAS pharyngitis. A systematic review and meta-analysis comprising studies mainly conducted among US Military personnel in the 1950s showed a protective effect of 68% (RR = 0.32; 95% CI = 0.21–0.48) when using antibiotics to reduce the incidence of ARF following a GAS throat infection. The absolute risk reduction was 1.67% with a number-to-treat (NNT) of 53. Penicillin alone, assessed in a subgroup analysis, had a protective effect of 80% (fixed effect RR = 0.20, 95% CI = 0.11–0.36) with an NNT of 60.

Experience from several regions of the world has shown that widespread use of antibiotics to treat GAS pharyngitis led to a decrease in ARF and RHD. In Cuba (1986–2002) and the French Caribbean islands, primary prevention was integrated into comprehensive programs that included public awareness, training of health personnel and establishment of ARF/RHD registries, which was associated with a dramatic decline in the ARF rate. In Costa Rica, prompt clinical diagnosis of GAS pharyngitis without doing throat cultures and using a single penicillin injection was also associated with a decrease in ARF incidence. Similarly, in Tunisia, an RHD control program based on early notification, early treatment of GAS pharyngitis using a single penicillin injection, and a strong secondary prevention program was associated with a decline in ARF from 8.7 (1985) to 1.5 per 100,000 population (2001). Consequently, RHD is shown to have declined significantly over the corresponding period.

Of course, ecological studies such as these do not provide definitive evidence for efficacy of the intervention itself, given that other factors such as economic development, improved housing or health systems strengthening may also have contributed to the decline in ARF and RHD. However, collectively, these reports suggest that integration of multiple approaches including education of the public and health personnel, and active management of GAS pharyngitis successfully contributed to the prevention of ARF in these settings. However, other similar attempts at primary prevention through school-based programs did not demonstrate a reduction in ARF in Hawaii and had only a modest effect among Navajo Indian communities in the 1960s and 1970s, suggesting there are other contributing factors and that the specific context factors including the incidence of ARF and health system delivery mechanisms are important to consider.

A meta-analysis that included six studies investigated the role of treatment of GAS throat infection at the community and school levels to prevent ARF reported a relative risk of 0.41 (95% CI: 0.23–0.70) and concluded that treatment of GAS pharyngitis in school children could reduce the cases of ARF by 60%. However, many of the studies included were noted to be of poor quality. In New Zealand, a randomized controlled trial assessing the efficacy of school-based clinics for the primary preventions of ARF implemented from 1998 to 2001 reported a nonsignificant 21% ( P = .47) to 28% ( P = .27) reduction in ARF cases. New Zealand recently implemented a national Rheumatic Fever Prevention Programme with several facets of primary prevention for ARF among high-risk populations. These included school-based sore throat clinics, where children self-identified as having a sore throat, had a throat swab, and any children whose swabs cultured GAS were given a 10-day course of oral amoxicillin. An evaluation of the school-based sore throat clinic component reported an overall effectiveness of 23% (95% CI: −6%–44%) (rate ratio (RR) 0.77 (95% CI: 0.56–1.06)). In one high-risk area with high coverage of the program, effectiveness was greater (46%, 95% CI: 16%–66%; RR 0.54 (95% CI: 0.34–0.84)) with this finding supported by other evaluations. The authors concluded that population-based primary prevention of ARF through sore throat management in schools where high-risk populations are geographically concentrated may be effective in well-resourced settings. However, where high-risk populations are dispersed, a school-based primary prevention approach appears ineffective and is expensive.

Taken together, these findings suggest that incorporation of a strategy of primary antibiotic prophylaxis into a comprehensive program working in concert for disease control, including awareness and surveillance, may reduce the incidence of ARF and RHD but requires careful consideration of costs and feasibility.

This provides support for the efforts of the Pan African Society of Cardiology (PASCAR), which is convening a series of meetings with the World Health Organization: Africa region (WHO Region) to develop and endorse an action plan to implement a control program for RHD that includes primary and secondary prevention in African countries with a policy of “treat all” sore throats with one injection of benzathine penicillin G (BPG).

Diagnosis of Group A Streptococcal Pharyngitis

Given that the signs and symptoms of GAS pharyngitis are difficult to distinguish from viral causes of sore throat, microbiological culture of a throat swab is the gold standard for diagnosing GAS pharyngitis. Given the necessary delay between the swab being taken and culture results becoming available, rapid antigen diagnostic tests (RADTs) are an alternative method for diagnosing GAS pharyngitis, although are associated with variable sensitivity (70%–90%) that depends on the clinical likelihood of GAS infection is an identified drawback. However, the need for an accurate and specific test that is affordable and validated in low-resource settings is clear. More recent progress in this area include newer tests such as the nucleic acid amplification tests that have much better sensitivity and specificity and are now being increasingly used in low-resource settings for other diagnoses. We anticipate that this need will generate new data in the near future, hopefully with applicability to low-resource but endemic areas of the world.

A third alternative for diagnosing GAS pharyngitis is using clinical predictive rules ( Box 10.2 ). Various clinical decision or predictive rules have been developed over the years. Systematic reviews and meta-analyses have concluded that symptoms alone are not sufficient to rule in or rule out GAS pharyngitis diagnosis and only the rule of Joachim et al. was clinically useful to exclude GAS pharyngitis. However, in low-resource settings where the use of laboratory facilities or RADTs may not be available or feasible, there is recent evidence that clinical diagnosis of GAS pharyngitis using clinical predictive rules can be reliably implemented in RHD endemic countries, obviating the need for more expensive bacteriological diagnostic methods. See Chapter 3 for a more detailed discussion of the diagnosis of GAS infection.

The American Heart Association (AHA) guidelines indicate that clinical and epidemiological findings have to be considered before deciding on using a microbiological test and only if these findings suggest GAS then throat culture or RADTs should be performed. It is stated that if there is evidence of viral infection (coryza, cough) then bacteriological investigations should not be performed. Therefore, clinical rules may be applied even in high-resource settings, emphasizing their usefulness alongside bacteriological investigations.

The WHO included sore throat in the Integrated Management of Childhood Illness (IMCI) Program, which has been implemented in many developing countries. The diagnosis of GAS pharyngitis in IMCI is based on the presence of fever or sore throat and two of the following: red congested throat, white or yellow pharyngeal exudates, or enlarged anterior cervical lymph nodes. Of note, the algorithm suggested by the WHO did not perform well when applied to children in Brazil, Croatia, and Egypt, missing up to 96% of children with positive cultures. This indicates that this clinical predictive rule is too specific to be utilized in areas with a high incidence of ARF. Tailoring decision rules to a specific population allows for improvement in performance characteristics such as the Turkish experience where less strict criteria were used.

The New Zealand ARF Guidelines recommend the protocol depicted in Fig. 10.3 . The most important features of this protocol are the considerations given to the ethnic/geographical and socioeconomic background of the patient (risk stratification), where less strict criteria are used for high-risk populations and if any criteria is positive then either empiric treatment is started immediately or bacteriological tests are performed and patients followed with treatment as indicated.

In South Africa, a “treat-all” strategy, where all children presenting with pharyngitis were treated with penicillin, was found to be cost-effective and affordable. The authors concluded that both the treat-all strategy and a clinical predictive rule called the clinical decision rule 2+ (CDR2+), where two or more criteria in a clinical decision rule are met, are affordable and simple and missed few cases of GAS pharyngitis at the primary level of care. In settings with a similar GAS prevalence and ARF attack rate in the Cape Town study area, the CDR2+ strategy was the most cost-effective. The authors recommended that a strategy for primary prevention of ARF and RHD in urban South Africa should be adopted to complement strategies to improve primordial prevention (better housing and hygiene) and secondary prevention (regular penicillin for those with a history of ARF). This strategy recommends that all patients complaining of a sore throat should receive antibiotics in the form of injectable BPG without requiring bacteriological investigations.

In Sudan, a protocol for the management of pharyngitis was based on a mapping analysis, which showed clustering of RHD in certain geographical areas. Similar to New Zealand, risk stratification is used and in high-risk patients two criteria are taken to diagnose GAS pharyngitis (age 3–18 years and absence of viral symptoms) without bacteriological tests. One dose of BPG is recommended.

Steinhoff et al. tested a clinical predictive rule that utilizes two points (sore throat with either a pharyngeal membrane or cervical lymphadenopathy) and found that it has a sensitivity of 80% and specificity of 40%. This rule was reviewed for the Cape Town setting and found to have a higher than acceptable missed diagnosis (MDx) to wrong diagnosis (WDx) ratio, whereas the locally developed Cape Town prediction rule (used in the primary prevention cost-effectiveness study) had a MDx/WDx ratio closest to 0.044, which met the target parameters of a sensitivity of ≥90% and a specificity of ≥40%.

In settings with research infrastructure, an approach may be to consider local investigation into clinical findings of sore throat parameters and a review of suitable clinical prediction rules and treat-all policies in line with local findings. However, it may be more feasible and cost-effective to consider likely similarities with other similar or nearby geographical settings and adapt a recommendation from one of those settings. The AFRO Strep registry is seeking to define the clinical, microbiological, epidemiological, and molecular characteristics of GAS infection in Africa. An active surveillance component will collect comprehensive data on GAS in African settings and may define locally accurate and adaptable clinical predictive rules in the future.