Rheumatic fever in Aboriginal children.

Abstract

In this issue, Steer et al. review the role of environmental factors in rheumatic heart disease.1 While low socio-economic status is a marker of regions where rheumatic fever and rheumatic heart disease remain common, the specific contributions of overcrowding, urbanization, nutrition and access to medical services have been difficult to assess. It is notable that rheumatic fever and rheumatic heart disease continue to be important causes of morbidity and mortality in the Indigenous population of central and northern Australia, as well as in Maori and Pacific Islanders in New Zealand, despite access to high-quality health services. In contrast, rheumatic fever is now extremely rare in non-Indigenous children throughout Australia. The rates of acute rheumatic fever (ARF) in Aboriginal children in the Top End of the Northern Territory are among the highest ever recorded, and current enhanced surveillance in central Australia is showing similar rates. In remote Aboriginal communities in the Northern Territory, 1−3% of the population has established rheumatic heart disease, usually as a result of cumulative valve damage from episodes of recurrent rheumatic fever. While rheumatic fever is thought to occur only in a subset of the population, this apparent immunological predisposition, which may be genetically based, is not restricted to any specific ethnic group. Rheumatic fever was commonly seen in the non-Aboriginal paediatric population in all Australian urban centres in the first half of last century. The current continuing high rates of rheumatic fever in Aboriginal children parallel overcrowding and poor living conditions, with very high levels of exposure to group A streptococci. The World Health Organization has recommended the use of rheumatic fever registers, primarily to co-ordinate individual patient management and improve availability of and adherence to secondary prophylaxis for the prevention of recurrent rheumatic fever.2 In a retrospective review from the Top End of the Northern Territory, there were 555 episodes of ARF in 367 people, of which 543 were episodes in 355 Aboriginal patients, nearly all from remote communities.3 None of the 12 non-Aboriginal patients had a relapse, whereas 39.2% of the episodes in Aboriginal patients were recurrent ARF among 107 individuals with previously diagnosed rheumatic fever or rheumatic heart disease. All of these recurrences are a failure of availability of, or compliance with, secondary prophylaxis with penicillin. These data resulted in the establishment of a Rheumatic Fever Register and Programme for the Top End in November 1997. The proportion of ARF notifications that were recurrent ARF decreased to 28% in 1998 and 16% in 1999, suggesting benefits from the emphasis on secondary prophylaxis protocols.4 The Rheumatic Fever Register and Programme continues to result in increased ascertainment of people with established rheumatic heart disease, with the number on the Register increasing from 403 in 1997 to 580 in 2000. New Zealand has had established rheumatic fever registers and programmes for much longer than Australia, with much of the New Zealand experience influencing the Northern Territory initiatives. In the early 1980s, recurrent ARF made up approximately 20% of individuals hospitalized in New Zealand with ARF.5 Factors contributing to this included a lack of recognition that 4-weekly IM benzathine penicillin is superior to oral penicillin, poor compliance with oral penicillin and the inability of clinicians to retain long-term patient contact to ensure regular prophylaxis. In New Zealand, there have been a number of register-based ARF prevention programmes, with considerable variation in register roles and configuration, based on local issues and service delivery structures. In 2001, there were six programmes using register-based initiatives for ARF prophylaxis, and during 1995−2000 the estimated proportion of ARF cases that were recurrent ARF was 8.9%.55 Acute rheumatic fever is also common in Indigenous communities in the Kimberley region of northern Western Australia,6 and in north Queensland.7 As in New Zealand,5 further development of regional register-based recurrent ARF prevention programmes in central and northern Australia will be dependent on local action, often mobilized through the efforts of key individuals. Circumstances will vary and the effectiveness of programmes will require ongoing evaluation to help optimize outcomes and secure ongoing funding. The diagnosis of an initial attack of rheumatic fever can be missed easily, so diagnoses are based on the Jones criteria, which requires two major, or one major and two minor, criteria, as well as evidence of a recent streptococcal infection. The major criteria are carditis, migratory polyarthritis, erythema mar­ginatum, chorea and subcutaneous nodules. Minor criteria are fever, arthralgia (in the absence of arthritis), raised C-reactive protein or erythrocyte sedimentation rate (ESR) and pro­­longed PR interval on electrocardiograph (in the absence of carditis). However, in Top End Aboriginal children, there were definite episodes of rheumatic fever that did not satisfy the strict application of the Jones criteria. In particular, low-grade fever and monoarthritis were important manifestations of rheumatic fever.3 In the Top End study, excluding cases of Sydenham’s chorea, monoarthritis occurred in 17% of confirmed ARF cases and 35% of unconfirmed ARF cases. Fifteen cases in the latter group subsequently had another episode that did fulfil the Jones criteria, suggesting that the initial episode was indeed ARF and an opportunity for prophylaxis against recurrence was missed. Similarly, only 71% of confirmed cases had a fever of 38 ϒC or higher, and only 25% had a high fever ( > 39 ϒC). These findings suggest that in regions with high rates of ARF, where the consequences of missed diagnosis outweigh those of over-diagnosis, the positive predictive value of less stringent criteria for ARF diagnosis may be acceptable. Such practice is being increasingly adopted in the Top End, where the diagnosis of acute rheumatic fever has often been missed in children with low-grade fever and arthritis who were considered to have transient synovitis or septic arthritis. The Jones criteria are based on clinical evidence of valvulitis, generally involving the mitral valve with a regurgitant murmur. The advent of echocardiography has delineated a group of patients with subclinical valvular insufficiency. Studies from Turkey,8 Chile,9 New Zealand,10 and Qatar,11 all indicate that subclinical valvulitis detected by echocardiography should be accepted as evidence of carditis in the diagnosis of acute rheumatic fever. However, technical expertise is required in interpretation of the echocardiogram, as non-specific trivial regurgitation may lead to over-diagnosis. As a minimum, early follow-up echocardiography and clinical assessment of a suspicious echocardiogram, in the absence of a murmur, is essential. In Top End Aboriginal children, approximately 10−15% of new cases of rheumatic fever have no clinical evidence of carditis, yet pathological mitral regurgitation is found on echocardiography (C Kilburn, pers. comm.). Follow-up echocardiography is also important to assess development of valve disease in those without evident carditis initially, and to determine if valve damage is progressing in those with carditis on presentation. Conventional wisdom deems that the absence of carditis on the initial episode means that recurrences will usually spare the heart.12 However, in the Top End study, 56% of patients without clinical carditis at the initial episode had carditis with at least one recurrence, indicating that this group of patients is also at risk of rheumatic heart disease.3 An important consideration from the epidemiological studies in northern Australia is a possible role for streptococcal pyoderma (not just pharyngitis) in rheumatic fever patho­genesis. Although this has not been proven, the suggestive findings are the very low rate of history of recent sore throat and positive throat swabs and high anti-DNAase B titres in acute rheumatic fever in Top End Aboriginal children.13 Scabies with streptococcal pyoderma is extraordinarily common in Top End Aboriginal children, whereas streptococcal pharyngitis is rarely recognized. Molecular typing of group A streptococci from the Top End has shown that molecular types usually identified as pharyngeal isolates are more commonly found in skin sores in remote communities.14 Therefore, even if throat colonization is still required to trigger ARF, the burden of bacteria is on the skin and should be amenable to skin health programmes directed at scabies and skin sores. There has been an important advance in the surgical management of rheumatic heart disease with the introduction of mitral valve repair (rather than replacement). This obviates the need for long-term anticoagulant therapy, which is a major difficulty in Top End Aboriginal patients. This surgical procedure has been carried out for almost 30 years in France, and long-term results are excellent.15 A number of Top End Aboriginal children have had valvular repairs for rheumatic heart disease over the last 3 years, allowing earlier surgery before there is deterioration in left ventricular function. However, in one case, recurrent rheumatic fever following poor compliance with benzathine penicillin resulted in further severe carditis, valve repair breakdown and death. This emphasizes that any valve surgery must be followed by meticulous attention to secondary prophylaxis as these patients are at high risk of recurrent ARF. Rheumatic heart disease morbidity and mortality in Abor­iginal communities can be improved in the short term through register-based programmes aimed at identifying those who require secondary prophylaxis and careful cardiac follow-up. While this is best achieved by providing resources at a regional level, there remains a national responsibility to address the underlying causes of the persisting high ARF rates, which are overcrowding and poor living conditions in remote communities.