Secondary Prevention of Acute Rheumatic Fever and Rheumatic Heart Disease

Definition of Secondary Prevention

Individuals within a vulnerable age group who have had an episode of acute rheumatic fever (ARF), or already have established rheumatic heart disease (RHD), are at high risk of recurrence of ARF if further episodes of group A streptococcal (GAS) infection occur. Secondary prevention of ARF and RHD is therefore warranted ( Box 11.1 ). A holistic view of secondary prevention comprises both secondary prophylaxis with an antibiotic (discussed here), plus enhanced primordial and primary preventive activities targeted for individuals and households affected by ARF and/or RHD (discussed in Chapter 10 ).

Enhanced primordial and primary prevention for individuals and households affected by ARF and/or RHD includes activities to tackle the socioeconomic drivers of RHD, such as infrastructure support for households to mitigate crowding and ensure access to washing facilities. Health literacy support and other educational activities described in Chapters 10 and 15 can enable early health seeking behavior for syndromes consistent with potential GAS infection (e.g., sore throat, impetigo), which may help ensure timely receipt of antibiotic treatment ( Box 11.2 ).

Historical Background

Before the introduction of antibiotics, ARF and RHD resulted in significant morbidity and premature mortality, particularly for children. One study in the preantibiotic era followed 588 children for 10 years after their initial episode of ARF: 58% had a recurrence in the first 5 years; 23% progressed to RHD and 24% died from ARF or bacterial endocarditis. One of the largest prepenicillin cohorts included 1000 patients who were followed for 20 years: 19% had recurrences in the first 5 years after admission to hospital with ARF or RHD, and the vast majority (89%) who had RHD at baseline had persistent evidence of valvular disease at the 10 years follow-up review.

Historically, implementation of antibiotic secondary prophylaxis has been the only medical intervention to have altered the course of disease for individuals after an initial ARF diagnosis. The risk of ARF after the first attack of GAS is approximately 0.3%–3%, but with subsequent infection this risk rises to 25%–75%, emphasizing the critical importance of secondary prophylaxis.

Evidence of Effectiveness

The following section and Table 11.1 review the data on the effectiveness of secondary prophylaxis with penicillin, showing that it is effective in reducing ARF recurrences, reducing RHD progression, increasing the likelihood of regression of existing RHD, and reducing overall mortality. Emerging data from Africa also suggests that high adherence (>80% of scheduled injections) is associated with reduced mortality from RHD. Secondary prophylaxis with penicillin has also been shown to be the most cost-effective RHD prevention strategy. In early studies of ARF prophylaxis using sulfonamides, 1.5% of treated cases developed ARF recurrences, compared to 20% of untreated cases. Subsequently, penicillin was found to be more efficacious than sulfonamides in preventing ARF recurrences (Level I evidence). A Cochrane meta-analysis concluded that penicillin, compared to no therapy, significantly reduces the rate of ARF recurrences, and that intramuscular BPG is superior to oral penicillin.

There is strong evidence that secondary prophylaxis also reduces the severity of RHD by preventing RHD progression. Data on the extent to which RHD can regress with secondary prophylaxis are mixed. In people receiving secondary prophylaxis with good adherence, some studies indicate that RHD severity improves in up to 50%–70% of people, with higher regression rates in mild RHD than in moderate or severe disease, and a mortality reduction has also been noted. Tompkins et al. reported in 1972 that acute mitral regurgitation resolved in 70% of patients with ARF who were adherent to 4-weekly BPG for at least 5 years. These early studies used auscultation to determine disease severity rather than echocardiography, so these are less accurate than more recent work using echocardiography to grade RHD severity. A more recent study measured changes in valve lesion severity and all patients prescribed secondary prophylaxis ( n = 6) had regression of mild lesions. A further study used valvular regurgitation as an indication of disease severity; regurgitation improved in 7 of the 13 people who received ≥75% of doses and did not improve in any of the people who were not adherent ( n = 4).

Numerous studies have reported that in nontrial (programmatic) settings, adherence to secondary prophylaxis reduces the likelihood of ARF recurrence and can prevent progression of RHD ( Table 11.1 ). Many of these studies have also reported rare recurrences occurring despite high adherence to secondary prophylaxis. In one study from New Zealand, which closely examined the circumstances of apparent “breakthrough” episodes, it was shown that there had been “days at risk” due to an injection being administered late by several days or more. In another from Australia's Northern Territory, while nearly all ARF recurrences occurred due to nonreceipt of BPG, some genuine “breakthrough” episodes did occur despite absence of “days at risk.”

Limitations and Challenges in Delivery of Secondary Prophylaxis

Secondary prophylaxis can only prevent RHD if implemented before RHD has developed, yet most RHD diagnoses are made in individuals with no prior diagnosed episodes of ARF. In Australia's Northern Territory, for example, approximately 75% of people diagnosed with RHD have not had a documented prior ARF diagnosis. This group of patients represents the missed opportunities for prevention. However, if the person is still in an age group vulnerable to recurrent GAS infection and ARF ( Table 11.3 ), implementation of secondary prophylaxis is still of high value to reduce the chance of further RHD progression.

A second limitation is that high adherence to secondary antibiotic prophylaxis long term is very challenging. Programmatic approaches to support adherence, including register-based recall systems and use of pain minimization techniques, are detailed later.

The potential for “breakthrough” ARF recurrences is another potential limitation—that is, ARF that occurs despite high adherence to the 4-weekly BPG regimen. This is very uncommon but has been noted, in which case a switch to a 3-weekly regimen is recommended. Details on the pharmacokinetics and pharmacodynamics of penicillin underpinning this decision-making are discussed later. A final challenge is that people with penicillin allergy are usually recommended a macrolide antibiotic regimen. In many settings, macrolides are prescribed in the absence of proven penicillin allergy, bringing with it difficulties in ensuring oral adherence, which is potentially even more challenging and difficult to measure. Furthermore, the proportion of GAS isolates resistant to macrolides is as high as 30% in some settings.

Penicillins and Pharmacokinetics

Fortunately, GAS remain fully susceptible to penicillin. This is likely to be because the organism lacks capacity to express β-lactamase or low-affinity penicillin-binding proteins; an alternative hypothesis is that expression of either is toxic to the organism. Resistance does, however, readily develop to non β-lactam antibiotics, as mentioned earlier in relation to macrolides such as erythromycin.

BPG is the first-line antibiotic for secondary prophylaxis. BPG is a fusion of two penicillin G molecules, characterized by very low solubility and low hydrolysis in vivo. These features allow for slow absorption from a depot intramuscular injection, producing low but adequate serum concentrations for streptococcal prophylaxis, for a number of weeks. There are two main formulations of BPG: a low-cost lyophilized powder (produced by many brands) and a high-cost cold-chain-dependent premixed suspension (produced by Pfizer alone). It is important to note that Pfizer has recently changed the naming and expression of product strength regarding BPG ( Box 11.3 ). Powdered BPG may suffer from variable quality and efficacy compared to the suspension form. Moreover, significant pharmacokinetic differences were found between two brands of powdered BPG, with no clear explanation found. Where available, the premixed suspension is preferred over powdered BPG.

Box 11.3

Terminology and Dosing of Premixed Intramuscular Penicillin for Secondary Prophylaxis of Acute Rheumatic Fever and Rheumatic Heart Disease

Reproduced (and modified) with permission from Pfizer and the Health Quality and Safety Commission, New Zealand.

• The name and strength description for the premixed suspension of benzathine penicillin G (BPG), otherwise known as benzathine benzylpenicillin or Bicillin® LA, has recently been changed by Pfizer, following changes in international practice

• The expression of the name of the product will now be benzathine benzylpenicillin tetrahydrate (see images later)

• The expression of product strength will now be 1,200,000 units/2.3 mL

	New Labeling	Old Labeling
Product name	Benzathine benzylpenicillin tetrahydrate Alternative name: benzathine penicillin tetrahydrate	Benzathine benzylpenicillin
Strength expression	1,200,000 units/2.3 mL	900 mg/2.3 mL

Reproduced with permission from the Health Quality and Safety Commission New Zealand

• There are no other changes: the active ingredient, the strength of the active ingredient, the product information, the dosing, and administration all remain the same

• Do not confuse long-acting intramuscular benzathine benzylpenicillin tetrahydrate with:

  • Rapidly acting benzylpenicillin (otherwise known as penicillin G), which is administered via the intravenous route

  • Short-acting intramuscular procaine penicillin G

New syringe artwork introduced in January 2019:

New carton artwork introduced in January 2019:

Anecdotes are commonly heard about doctors mistakenly prescribing other forms of penicillin, such as procaine penicillin G. This form is also administered intramuscularly, but achieves high blood concentrations for only 24 h and is therefore ineffective as a form of secondary prophylaxis against ARF.

In cases where BPG is refused, oral penicillin V (phenoxymethylpenicillin) may be prescribed, although this is associated with a higher risk of ARF recurrence. Although the lower effectiveness of oral administration may be partly attributable to greater difficulty in adhering to such a regimen compared with a 4-weekly supervised injection regimen, the half-life of oral penicillin V is only around 1 h, so even full adherence to a twice-daily oral penicillin V regimen has been considered to be inferior to intramuscular BPG in preventing recurrent ARF.

Although the standard secondary prophylaxis BPG regimen comprises administration every 4 weeks (28 days), serum penicillin levels may be low or undetectable much earlier than 28 days, often within 14 days, following a standard BPG dose of 1.2 million units (900 mg). Indeed, fewer streptococcal infections and ARF recurrences have been reported among individuals receiving 3-weekly BPG compared with a 4-weekly regimen. Moreover, the 3-weekly regimen resulted in greater resolution of mitral regurgitation in a long-term randomized study in Taiwan (66% vs. 46%).

As noted earlier, prospective data from New Zealand, and other analyses from Australia, show that recurrences are extremely rare among people who are fully adherent to a 4-weekly BPG regimen. 3-weekly as opposed to four-weekly penicillin has important workforce implications. It is likely that with a four-weekly regimen, any streptococcal infection acquired after serum penicillin concentrations have dropped below the protective level would be treated early by the next administered BPG dose. Historical work suggested that treating GAS infection within 9 days was likely to reduce most ARF episodes, but these data derive from a small cohort, and are somewhat immunologically implausible, especially in highly immunologically primed hosts in whom the abnormal immune response causing ARF would likely be triggered early after onset of symptomatic disease. Nevertheless, despite the pharmacokinetic data and uncertainty about the “grace period” in which to treat GAS to avoid ARF, 4-weekly regimens work very effectively, even in environments where GAS transmission is very high.

A pharmacokinetic study in Aboriginal children and adolescents in Australia has recently been undertaken to better understand serum concentrations after intramuscular dosing in children of a variety of ages and weights. Serum concentration of penicillin was tested using liquid chromatography–mass spectroscopy assay on dried blood spots. Counterintuitively, the study found that blood concentrations of penicillin remained above the minimum inhibitory concentration (protective level) for only 9.8 days on average in normal weight children (body mass index <25 kg/m ² ) after administration of a BPG dose intended to provide protection for 4 weeks. The authors concluded that more knowledge is therefore needed to reconcile this finding with the observed protectiveness conferred by 4-weeks dosing regimens of BPG.

Penicillin reformulation, for instance as an implant or long-acting depot that could provide slow release of penicillin over months, has been under investigation in recent years. Despite technical challenges, there is optimism that reformulated penicillin may be a reality in the coming years, which would dramatically improve the ability to deliver effective secondary prophylaxis for ARF.

Penicillin Supply

BPG is an off-patent, generally inexpensive antibiotic. Stock-outs occur repeatedly in both low- and high-income settings, posing major challenges to ARF prevention. As recently noted, attention to the supply, manufacture, and accessibility of BPG has declined as its use in high-income settings has become supplanted by newer, broader-spectrum antibiotics. Attention must be given to BPG availability and quality in low-resource settings. Although BPG is included in the World Health Organization's Essential Medicines List, practitioners in many countries lack access. In 2011, a survey of healthcare providers who manage patients with ARF or RHD was undertaken by the World Heart Federation in African countries, the Asia–Pacific region, and Central and South America. Responses indicated that many settings had minimal or no access to BPG. Supply of the preferred formulation of BPG (premixed liquid formulation) has been precarious in many settings, with stock-outs documented in Australia in 2001, 2004, and 2006. The problem is far more pronounced in low-income settings.

Penicillin Adverse Reactions

BPG should not be given to patients with a documented allergy to penicillins or cephalosporins. Extreme caution should be exercised in giving BPG to patients with decompensated heart failure, as this can lead to collapse and death, mimicking an “anaphylactic reaction” (discussed later and also chapter 10 ).

Global experience with long-acting penicillin administration is extensive. As well as being administered for ARF and RHD prophylaxis, BPG is used for treatment of streptococcal infections, syphilis, and yaws. Anaphylaxis to BPG is rare. This section provides reported rates of adverse reactions, and concerns about increased risk for adverse reactions in low-resource settings; the following section details the diagnosis and clinical management of anaphylaxis.

Hypersensitivity reactions to BPG have been reported after multiple monthly injections, and anaphylaxis has been reported to occur in patients who have previously tolerated the injection for months and years without incident. In a prospective international study after 32,430 injections (given specifically for secondary prevention of ARF and RHD) during 2736 patient years of observation, 57 (3.2%) of the 1790 patients had an allergic reaction and four (0.2% or 1.2 per 10 000 injections) developed anaphylaxis. A meta-analysis of 13 studies that reported on adverse reactions after BPG administration (for any indication; the study referred to above was included) found that the risk of death across studies was 0 (there were 4 deaths in a sample of 2,108,117 patients) and the absolute risk of serious adverse events including anaphylaxis was 0.17%. The authors indicated that the risk of adverse reactions was higher for people who received multiple BPG injections; however, the results for this group were not reported separately.

A recent review documented that alongside rising concerns about the quality of available BPG internationally, anecdotal reports of adverse reactions appear to have increased in recent years. The review cites three deaths occurring in Zimbabwe in 2000 associated with BPG from three different manufacturers. In the World Heart Federation's BPG survey, 26% of 39 clinicians surveyed internationally reported being involved in at least 1 anaphylactic reaction, and 21% of providers reported that they had had a patient die due to anaphylaxis after BPG injection.

This clinical experience has not been replicated in other high-use settings, including Australia and New Zealand. Indeed, the more commonly recognized problem in these settings is erroneous labeling of individuals as being penicillin allergic when in fact they are not–when investigated, only around 10% of individuals with a penicillin allergy label in fact are allergic. Reports of anaphylaxis occurring after BPG in Australian and New Zealand ARF/RHD registers are vanishingly rare.

It is impossible to determine whether the adverse drug reactions reported in Zimbabwe and in the World Heart Federation survey can be attributed to penicillin, reactions to other components that might be present in the vial (such as contaminants), or incorrect classification of reactions. Indeed, patients with underlying cardiac conditions may experience fatal or near-fatal reactions (with no clinical features of anaphylaxis) seconds to minutes after BPG administration. Fatal cases have included young patients (10–12 years old) with severe mitral regurgitation who were receiving BPG for secondary prophylaxis of ARF. The mechanism for death in these circumstances remains obscure. A suggested hypothesis is that some reactions could represent vasovagal reactions (± bradycardia) to intramuscular injection in general, not related to penicillin specifically, which may precipitate cardiovascular collapse in patients with significant valvular heart disease. For example, mitral stenosis is a preload-dependent state and any reduction in blood pressure from vasovagal syncope may precipitate a cardiac arrest ; or in patients with pulmonary hypertension (secondary to left-sided valve disease), a vasovagal reaction may result in life-threatening arrhythmias or acute right heart failure. The vasovagal hypothesis is supported by the observation that almost all severe adverse reactions appear to be in cases with significant valvular heart lesions, whereas anaphylaxis should be equally common in people with mild or no RHD. It is important that systems are in place for the reporting of adverse drug reactions to better understand safety of BPG and potentially, as a proxy guide to BPG quality.

At the time of writing, a consensus guideline for delivery of secondary prophylaxis in cases of moderate or severe RHD was being prepared, highlighting the extreme care needed in these cases, and that in some circumstances oral penicillin prophylaxis may be preferred to BPG.

Care should also be taken to ensure that BPG is not inadvertently given by the intravascular route (FDA black-box warning). Intravenous administration has been associated with hyperkalemia, arrhythmias, and cardiac arrest, whereas intraarterial administration has also been associated with cardiac arrest (attributed to emboli after injection) and life-threatening pulmonary edema.

Other forms of adverse reaction to penicillin have also been noted. In qualitative work in Australia's Northern Territory, most individuals receiving regular BPG did not experience side effects, but a small number described taking to bed for several days afterward and feeling generally unwell.