Missed Opportunities for Diagnosing Acute Rheumatic Fever

Abstract

We have read with interest the new publication by Rouhiainen and colleagues on missed opportunities for preventing or diagnosing acute rheumatic fever (ARF) [1]. We commend the authors for the rigour, scope, and originality of their approach.

It is clear that episodes of ARF in Australia frequently pass undetected, given that only about half of young people with a diagnosis of rheumatic heart disease (RHD) have recorded evidence that at least one prior episode of ARF has been recognised [2]. Our recently published study complements the findings of Rouhiainen et al., by using routinely collected data from across four Australian jurisdictions to investigate missed opportunities for detection of ARF and RHD during hospital emergency department presentations and inpatient admissions [3]. Our study comprised 1855 individuals who were identified with ARF/RHD when aged 8–24 years, of whom about half (53.0%) resided in the NT and 58.0% in remote or very remote areas. Within the 3 years prior to diagnosis, 7.9% (n = 146) had at least one recorded hospital presentation with a condition mimicking ARF/RHD, among which joint disorders were the most common category. The probability of being diagnosed with highly likely mimics was greater among persons from demographic groups considered to be at low risk, that is, those subsequently diagnosed as young adults (18–24 years) rather than as children (odds ratio [OR] 2.45, 95% confidence interval [CI] 1.34–4.47), and those from Australian-born non-Indigenous persons compared with Aboriginal and Torres Strait Islander peoples (OR 2.44, 95% CI 1.02–5.85).

While our findings support those of Rouhiainen et al. [1], the findings of the latter study greatly extend our own, specifically in relation to populations with a very high risk of developing ARF/RHD during early life, such as those of First Nations communities in remote areas of Australia's Top End.

Our findings and those of Rouhiainen et al. are not directly comparable, considering the differences in populations under investigation, clinical status at diagnosis, data sources, lookback strategy, and range of care settings. Nonetheless, the finding by Rouhiainen et al. that a majority of persons (13/20) with RHD had experienced one or more prior presentations with conditions mimicking ARF underscore that missed opportunities for diagnosis were likely to have been greatly underestimated from the routinely collected data available to us. Although our findings doubtlessly represent only the ‘tip of the iceberg’ of these missed opportunities, they suggest that such occurrences are even more common among populations in whom risk of ARF/RHD is perceived to be low than in the very high-risk community studied by Rouhiainen et al.

Of further note in this regard is the study recently published by Ali et al. [4] on unmasking silent ARF through handheld echocardiography (HHE) screening of febrile children presenting in an RHD-endemic population of Sudan. In this cohort (N = 400, mean age 9 years), subclinical carditis (i.e., lacking auscultatory findings and detectable only by ultrasound) was detected by HHE in 11% of children who had fever but manifested no other major Jones criteria [5] of ARF. These data highlight the importance of subclinical carditis in the diagnosis of ARF, as recognised in the most recent published revision of the Jones Criteria [5]. The findings of Ali et al. suggest that routine echocardiographic examination of children with fever but no other clinical signs may be indicated in populations at very high risk of ARF/RHD in order to detect cases of ARF that would otherwise be missed in the absence of any other major diagnostic criterion.

The feasibility of HHE screening has recently been advanced in the context of non-acute detection of RHD through a task-sharing approach involving simplified case-finding protocols for trained non-expert practitioners supported by offsite expert back-up [6]. To this end, echocardiographic criteria designed specifically to include non-expert screening have been incorporated in the 2023 World Heart Federation guidelines for the echocardiographic diagnosis of RHD [7]. This model of RHD detection, which is being investigated in Australia and neighbouring countries [6, 8], could potentially be expanded to incorporate HHE screening of febrile children for ARF in high-risk populations.

Together, the findings from these studies underscore the shortcomings of current approaches to the diagnosis of ARF, missed opportunities for which may be partially remediable by the education of clinicians to increase their index of suspicion for this condition, along with the promotion of HHE as a non-invasive screening modality in high-risk settings.

J.M.K. conceived the letter. J.A.W. drafted the manuscript. All authors reviewed and contributed to the refinement of the manuscript and have read and approved the final version.

The authors declare no conflicts of interest.