Uncertainty in the diagnosis and prognosis of adult-onset Still’s disease

1. Introduction

Adult-onset Still’s disease (AOSD), now unified with its pediatric counterpart, systemic juvenile idiopathic arthritis (sJIA), under the umbrella term ‘Still’s disease,’ continues to pose significant diagnostic challenge in clinical practice [1–3]. Its hallmark clinical triad – high spiking fever, transient salmon-colored rash, and polyarthralgia – are strikingly nonspecific and overlap significantly with a spectrum of infectious, auto-immune/-inflammatory, and malignant disorders. Fever, one of the most prominent symptoms, typically spikes once or twice daily, resembling patterns seen in infections such as malaria or enteric fever [4,5]. The associated rash, often transient and nonpruritic, appears with fever spikes and can be mistaken for drug reactions or viral exanthems [1,4,5]. Joint involvement, commonly manifesting as polyarthralgia, may mimic systemic lupus erythematosus or acute rheumatic fever. Organ involvement in AOSD, such as hepatosplenomegaly, lymphadenopathy, and serositis, can mimic malignancies like lymphoma or chronic infections such as tuberculosis [4,5]. Consecutively, differentiating AOSD from these mimics, especially in resource-limited settings, frequently results in diagnostic delays and uncertainty.

2. Problems in diagnosis

The lack of a single definitive diagnostic test makes AOSD a diagnosis of exclusion [2]. The Yamaguchi criteria (1992) are the most widely used, offering high sensitivity and specificity but requiring the exclusion of infections, malignancies, and autoimmune diseases, which often leads to delays [6]. Fautrel’s criteria (2002) incorporate novel markers like glycosylated ferritin to improve specificity; however, these tests are not routinely available, limiting their practicality [7]. Cush’s point-based system (1987) captures a broad range of clinical features but adds complexity to an already intricate diagnostic process [8]. The recent criteria proposed by Karima Daghor-Abbaci et al. (2023) aim to improve diagnostic accuracy with parameters like the neutrophil-to-lymphocyte ratio (NLR) and a refined weighted scoring system, though these too require further validation [9]. Notably, none of these criteria have been universally accepted or validated against a consistent gold standard, often relying on retrospective data or expert consensus based on a heterogeneous set of alternative diagnoses, which further limits their reliability.

The presence of distinct phenotypic clusters in Still’s disease, such as systemic, hyperferritinemic, uncomplicated, and catastrophic subtypes, further complicates its diagnosis and management [10]. These clusters exhibit diverse clinical manifestations and immune activation profiles, contributing to heterogeneity in disease presentation [10]. One classical example of this diagnostic challenge is its presentation as fever of unknown origin (FUO), characterized by persistent fever without a clear diagnosis after extensive investigations [4]. AOSD is a significant contributor to FUO, particularly in younger populations, yet its nonspecific symptomatology often misleads clinicians [4,11,12]. Crispín et al. (2005) developed a clinical scale to differentiate AOSD from other causes of FUO, emphasizing features like arthritis, pharyngitis, splenomegaly, and neutrophilic leukocytosis [13]. However, subsequent studies have questioned its utility, necessitating further refinement and validation [4,13]. Shuni Ying et al. (2023) introduced the AOSD/FUO (AF) score using Bayesian modeling to assign weighted probabilities to unique features, such as persistent pruritic eruptions [14]. Although the AF score has shown improved diagnostic accuracy over traditional criteria, it remains a probabilistic model rather than a definitive diagnostic framework.

Recognition of macrophage activation syndrome (MAS) or secondary hemophagocytic lymphohistiocytosis (HLH), a severe and potentially life-threatening complication of AOSD, compounds the diagnostic challenge [2,4]. Unlike sJIA, the pediatric counterpart of AOSD, which benefits from the structured EULAR/ACR/PRINTO criteria (2016) for diagnosing MAS, AOSD lacks an equivalent, adult-specific framework [2,15]. Given the shared pathophysiological mechanisms between AOSD and sJIA, adapting pediatric MAS criteria for adults appears reasonable [2,4]. However, diagnostic markers such as ferritin and liver dysfunction often present more severely in adults, with ferritin levels in AOSD-MAS frequently exceeding 10,000 ng/mL, far above the 684 ng/mL cutoff used in sJIA-MAS [15,16]. Although the HLH-2004 criteria, designed for primary HLH in children, are often used for diagnosing secondary HLH or MAS in adults, they remain unvalidated in the context of AOSD [17]. Similarly, the HLH-probability calculator (HScore) is an emerging tool for adult HLH diagnosis, but its utility in AOSD-related MAS is yet to be proven [17,18]. This lack of validated, age-appropriate criteria highlights the urgent need for more reliable diagnostic tools. Until such criteria are developed, clinicians may need to rely on adapted sJIA criteria and informed clinical judgment as a practical interim strategy for early recognition and management of MAS in AOSD.

2.1. Emerging biomarkers

Emerging biomarkers hold promise in addressing the diagnostic, prognostic, and therapeutic challenges associated with AOSD [2,10,19]. Serum ferritin, particularly when markedly elevated, remains a cornerstone diagnostic clue but lacks specificity, as it is also elevated in other inflammatory conditions, malignancies, and HLH or MAS [4,11]. Glycosylated ferritin levels below 20% add specificity but are underutilized due to limited accessibility and the lack of standardized testing [2,19]. Novel biomarkers, particularly interleukin-18 (IL-18) and S100 proteins (S100A8/A9 and S100A12) have shown significant potential in aiding diagnosis and predicting systemic involvement [2,19,20]. IL-18, a cytokine central to inflammasome activation, not only correlates strongly with disease activity but also plays a pivotal role in the pathogenesis of MAS by driving interferon-gamma (IFN-γ) overproduction [2,19,20]. Caspase-1, a key regulator of IL-18 activation, has recently been linked to disease activity, further highlights the central role of inflammatory cytokines in the pathogenesis of AOSD [21]. Alarmins such as S100 proteins, produced by innate immune cells (monocytes and neutrophils), act as danger-associated molecular patterns that amplify inflammation and are strongly associated with disease severity [2,19]. Similarly, elevated CXCL9 levels and ADA2 activity, both associated with IFN-γ, have been implicated in MAS [2,19]. Genetic insights, such as the association of HLA-DRB1 × 15 with lung involvement in Still’s disease, further underscore the role of genetic predisposition in organ-specific manifestations [2,10,19]. These biomarkers, when integrated with established diagnostic criteria such as Yamaguchi or Fautrel, have the potential to enhance diagnostic specificity and facilitate early recognition of severe complications like MAS. However, further validation, standardization of thresholds, and large-scale real-world studies are essential for their widespread implementation in clinical practice.

Imaging modalities, particularly whole-body 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET/CT), have proven valuable in assessing systemic inflammation, ruling out malignancies, and guiding tissue biopsies [4]. However, the high cost and limited availability of PET/CT restrict its use in routine clinical settings. When uncertainty persists, tissue sampling, such as bone marrow or lymph node aspiration or biopsy, often becomes necessary to definitively exclude malignancies or other mimicking diseases [4].

Kaya et al.’s recent study evaluates the utility of inflammatory and nutritional indices, such as the CRP-to-albumin ratio (CAR), lactate dehydrogenase-to-albumin ratio (LAR), ferritin, ferritin-to-ESR ratio (FER), and the prognostic nutritional index (PNI), in predicting organ involvement in AOSD [22]. Their findings suggest associations between CAR and hepatomegaly, LAR and lymphadenopathy, ferritin and splenomegaly, and both FER and PNI with serositis. While these insights add diagnostic precision, it is crucial to recognize that basic imaging, such as ultrasound and chest X-ray, alongside clinical examinations, are often sufficient for detecting organ involvement [4,22]. Thus, future research should prioritize identifying markers that can predict clinically relevant organ dysfunction or outcomes, such as MAS, or guide risk stratification for hospitalization and intensive care. The newly developed Still’s Disease Activity Score (2024), which utilizes readily available clinical and laboratory parameters, is emerging as a promising tool for categorizing disease severity and guiding treatment strategies [23]. However, it requires external validation and real-world studies to confirm its effectiveness and utility.

3. Conclusion

In conclusion, AOSD remains a clinical challenge, marked by diagnostic uncertainty that extends from initial symptom recognition to the identification of severe complications like MAS. The absence of universally validated diagnostic criteria and the reliance on exclusion-based methods underscore the need for continued research into specific biomarkers and improved imaging modalities [2,10,19]. Kaya et al.’s study contributes to our understanding by proposing indices that may aid in assessing organ involvement; however, there remains a pressing need for research that focuses on outcomes impacting patient management, such as the prediction of MAS [22].

Despite advances in understanding AOSD pathophysiology and the development of novel biomarkers such as IL-18 and S100 proteins, critical gaps persist in diagnostic precision and timely management. This editorial synthesizes recent advancements and highlights the critical need for multidisciplinary collaboration and innovative research to refine diagnostic tools, optimize management strategies, and ultimately improve outcomes for patients with this complex disease.

Funding Statement

This paper was not funded.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Author contributions

¹AKP: conceptualization, drafted and revised the manuscript