Israel Medical Association Journal

Background: Large language models (LLMs) are rapidly advancing, with the potential to improve healthcare. While LLM performance on medical licensing exams were studied extensively, their performance in rheumatology exams requires specific evaluation.

Objective: To assess Chat Generative Pre-Trained Transformer (ChatGPT) performance on 200 validated Israeli rheumatology board exam questions.

Methods: ChatGPT performance was evaluated using 200 multiple-choice questions from the 2023 and 2024 Israeli official rheumatology board examinations. Three gpt-4-turbo based variants were assessed: base model (Model 1), few-shot chain of thought (CoT) model (Model 2), and knowledge-augmented prompting model incorporating rheumatology guidelines (Model 3). Model 1 was assessed using both the original Hebrew and a validated English translated version, while Models 2 and 3 were assessed using the English version only.

Results: Overall, Model 3 achieved the highest numerical accuracy (81%), followed by Model 1 (English, 77%), Model 2 (75%), and Model 1 (Hebrew, 74.5%); however, these differences were not statistically significant. Performance varied markedly by question type. For text-only questions (n=177), accuracies ranged from 78.5% to 83.1%, with Model 3 showing the highest point estimate (83.1%). In contrast, all models demonstrated substantially lower performance on questions that included images (n=23), with accuracies ranging from 34.8% to 65.2%. Model 3 yielded the highest numerical accuracy (65.2%).

Conclusions: The study highlights the potential role of LLMs in rheumatology board examinations but also emphasizes their critical limitations. Future research should focus on addressing limitations, especially image interpretation and management of complex cases to enable efficient application of LLMs in rheumatology.

Background: Immunoglobulin 4 (IgG4) is the least abundant immunoglobulin in the sera of healthy individuals; however, its levels can vary in different diseases such as IgG4-related disease (high) or Sjögren's syndrome (low). While previous studies have suggested the importance of IgG4 in autoimmune diseases, the clinical and biological significance of high or low levels remains unclear.

Objectives: To investigate the association between IgG4 antibody levels and systemic sclerosis (SSc), as well as the clinical features of the disease.

Methods: We measured IgG4 levels in the sera of 74 SSc patients from the years 2000 to 2019 and compared them to IgG4 levels in 80 healthy donors from the Israeli national blood bank. We performed correlation analyses between IgG4 levels and various factors, including age, sex, disease subtype, disease duration, organs involved, and medications taken by the patients.

Results: Our findings revealed significantly lower IgG4 levels in SSc patients compared to healthy participants. SSc patients receiving steroid treatment exhibited prominently lower IgG4 levels. In addition, SSc patients with Raynaud's phenomenon tended to have lower IgG4 levels compared to those without Raynaud's phenomenon.

Conclusions: Our study demonstrates that IgG4 levels are lower in SSc patients. Further research is needed to elucidate whether this observation contributes to the etiology of the disease or if it represents a common manifestation among other autoimmune diseases.

IgA vasculitis, formerly known as Henoch–Schönlein purpura (HSP), is the most common systemic vasculitis in children. It is defined as palpable purpura in the absence of coagulopathy or thrombocytopenia and one or more of the following criteria: abdominal pain, arthritis or arthralgia, biopsy of affected tissue demonstrating predominant IgA deposition, and renal involvement with proteinuria and hematuria or red cell casts [1].

We describe the features of nocebo, and its impact in studies of transition from the originator to the respective biosimilar in inflammatory rheumatic diseases. Investigations in healthy volunteers as well as in the neurology and anesthesiology fields demonstrated the involved cerebral areas and the neurotransmitter pathways responsible for the nocebo response. Whether these findings are applicable to patients with inflammatory rheumatic diseases remains to be demonstrated. Nocebo may account for part of the after-switching biosimilar failures. However, in the absence of validated classification or diagnostic criteria, specific neurochemical and neuroimaging studies, the lack of data on serum tumor necrosis factor and drug levels, and the disease improvement after the switching back to the originator biologic observed in some patients, the nocebo diagnosis remains the role of the individual clinician. Investigations on nocebo pathophysiology and diagnosis are required to address its impact in after-transition biosimilar studies in rheumatology.

Background: Non-infectious uveitis (NIU) leads to severe visual impairment, potentially impacting on health-related quality of life (QoL). 

Objectives: To investigate the impact of NIU on QoL.

Methods: Eighty NIU patients and 23 healthy controls completed the 36-item Short-Form Health Survey (SF)-36. The SF-36 values were statistically analyzed to evaluate differences between patients and healthy controls and to identify correlations between SF-36 subscores and clinical/demographic data. 

Results: NIU patients showed a decrease in the physical component summary score (P < 0.0001) compared to healthy controls, while no difference was highlighted in the mental component summary score (P = 0.97). NIU patients showed a decrease in physical functioning (P = 0.008), role-physical (P = 0.003), bodily pain (P = 0.0001), general health (P < 0.0001), and social functioning (P = 0.01). Physical functioning was lower in patients with acute anterior uveitis (AAU) than in those with panuveitis (P = 0.003). No differences were found between patients with bilateral or unilateral NIU, isolated NIU, or NIU associated with systemic diseases and with or without ocular activity. No correlations were identified between best-corrected visual acuity and SF-36 subscores. Physical functioning (P = 0.02), bodily pain (P = 0.004), and social functioning (P = 0.02) were reduced in males versus females. 

Conclusions: QoL is impaired in individuals with NIU, particularly in the physical domains, general health, and social functioning. AAU affects physical functioning more than panuveitis. NIU seems to affect per se QoL disregarding inflammatory activity, visual impairment, and presence of associated systemic diseases.