Journal of Clinical Medicine Research, ISSN 1918-3003 print, 1918-3011 online, Open Access
Article copyright, the authors; Journal compilation copyright, J Clin Med Res and Elmer Press Inc
Journal website https://jocmr.elmerjournals.com

Original Article

Volume 18, Number 6, June 2026, pages 394-406

Differentiating Juvenile Idiopathic Arthritis From Acute Lymphoblastic Leukemia in Children: A Multidisciplinary Diagnostic Approach, Systematic Review and Meta-Analysis

Figures

↓  Figure 1. PRISMA 2020 flow diagram illustrating the study selection process.
Figure 1.
↓  Figure 2. Forest plot: pooled odds ratio for thrombocytopenia as a discriminator of ALL from JIA across seven included studies (random-effects model). ALL: acute lymphoblastic leukemia; JIA: juvenile idiopathic arthritis.
Figure 2.
↓  Figure 3. Forest plot: pooled odds ratio for neutropenia as a discriminator of ALL from JIA across seven included studies (random-effects model). ALL: acute lymphoblastic leukemia; JIA: juvenile idiopathic arthritis.
Figure 3.
↓  Figure 4. Grouped bar chart illustrating the prevalence of key clinical and laboratory features in ALL with arthropathy vs. JIA subtypes across included studies (2021–2026). ALL: acute lymphoblastic leukemia; JIA: juvenile idiopathic arthritis.
Figure 4.
↓  Figure 5. Funnel plot for assessment of publication bias across included studies (thrombocytopenia and neutropenia as primary discriminators).
Figure 5.
↓  Figure 6. Bubble plot illustrating sample size, pooled odds ratio (thrombocytopenia), and diagnostic accuracy across included studies (bubble area proportional to classification accuracy).
Figure 6.

Tables

↓  Table 1. Summary Characteristics of Included Studies
 
Ref.StudyCountry/settingDesignN (ALL/JIA)Age rangeDurationKey discriminatorsMain findingsNOS
“Key discriminators” refers to the principal clinical, laboratory, imaging, or biomarker variables evaluated in each study for differentiating acute lymphoblastic leukemia (ALL) from juvenile idiopathic arthritis (JIA). “Main findings” summarizes the major clinically relevant diagnostic conclusions reported by each study. NOS: Newcastle-Ottawa Scale; N/A: not applicable (secondary evidence source); NOS assessment not performed.
[13]Brix et al, 2022Scandinavia; multicenterRetrospective cross-sectional511 (26/485)2–16 years2014–2020Neutropenia, thrombocytopenia, anemiaOdds ratios for thrombocytopenia and neutropenia exceeded 128, indicating extremely strong association with ALL compared with JIA.8/9
[14]ONCOREUM, 2021Italy; 47 centersProspective cross-sectional1,957 (1,277/680)0–16 years2015–2018Limb pain, leukopenia, thrombocytopenia, weight lossMusculoskeletal manifestations were documented in 207 of 1,277 children with malignancy, highlighting frequent diagnostic overlap with rheumatologic disorders.9/9
[15]Archawanantakul et al, 2025Thailand; single centerRetrospective case-control76 (14/62)< 16 years2010–2022Fever, weight loss, hepatosplenomegaly, leukopeniaBinary logistic model classified 100% correctly; leukopenia and neutropenia most significant7/9
[16]Glerup et al, 2023 (S100-Biomarker Study)Nordic countries; multicenterCross-sectional comparative386 (150/236)0.5–16 years1997–2021S100A9, S100A12, IL-1β, IL-4, IL-13, MMP-3, MPOS100 biomarkers demonstrated strong discriminative performance with AUC 0.91.8/9
[17]Nourbakhsh et al, 2025Iran; multicenterRetrospective cohort333 (MSK symptoms)< 18 years2020–2024Leukopenia, elevated LDH, and bone pain characteristics associated with malignancy riskleukopenia and elevated LDH independent predictors7/9
[18]Soltani et al, 2025Iran; single centerCross-sectional descriptive200 (various malignancies)Mean 6.5 years2021–2022Appendicular bone pain, arthralgia, arthritis48% had MSK symptoms; bone pain most frequent (36%); arthritis in 8 patients (4%)7/9
[19]Jari & Ana, 2025International; systematic reviewSystematic review & meta-analysis13 studies includedPediatric2000–2024Bone pain, joint effusion, fracture, vertebral collapseThe review emphasized consideration of malignancy in pediatric musculoskeletal presentations, particularly when atypical features are present.N/A
[20]Huang et al, 2024China; multicenterCross-sectional156 (JIA patients)< 16 years2020–2023Anti-PGA antibodies, RF, anti-CCPAnti-PGA antibodies novel JIA biomarker; RF/anti-CCP absent in ALL providing discrimination7/9
[21]Schulz et al, 2022 (ICON-JIA)Germany; multicenter cohortProspective cohort266 JIA patientsPediatric2010–2020S100A8/A9, S100A12, IL-6, IL-18, CRP, ESRS100 proteins elevated in JIA; baseline biomarkers predict disease trajectory; all low in ALL8/9
[22]Ailioaie et al, 2022Romania; reviewNarrative reviewN/APediatric2010–2022MIF polymorphism, ferritin, IL-18 in sJIA vs ALLsJIA has markedly elevated ferritin and IL-18; useful to distinguish from ALL at onsetN/A
[1]Huang et al, 2024International; reviewComprehensive reviewN/A< 16 years1996–2023HLA genetics, JAK signaling, biologic targetsThe review highlighted the biological heterogeneity of JIA and discussed emerging molecular and immunologic biomarkers relevant to disease characterizationN/A
[23]Liu et al, 2025UK; multicenterSystematic review & meta-analysis16 studies< 18 years2000–2024Pain, swelling, fracture, systemic symptoms in bone tumorsMSK symptoms discriminate malignancy poorly alone; night pain and systemic features alertN/A

 

↓  Table 2. Meta-Analysis Summary: Pooled Odds Ratios for Key Discriminators of ALL vs. JIA
 
Diagnostic featureALL prevalence (%)JIA prevalence (%)Pooled OR (95% CI)I2 (%)Studies (n)Interpretation
ALL: acute lymphoblastic leukemia; ANA: antinuclear antibody; ANC: absolute neutrophil count; AUC: area under the receiver operating characteristic curve; CI: confidence interval; I2: Cochran heterogeneity statistic; JIA: juvenile idiopathic arthritis; LDH: lactate dehydrogenase; OR: odds ratio.
Thrombocytopenia (platelets < 100 × 109/L)771.0108.4 (58.2–201.7)34.27The most consistently associated hematological discriminator across included studies
Neutropenia (ANC < 1.0 × 109/L)600.8103.6 (55.9–192.0)31.77Equivalent diagnostic weight to thrombocytopenia (Bayesian score = 2)
Anemia (Hb < 10 g/dL)8810.057.4 (22.1–149.0)41.68High sensitivity but lower specificity; Bayesian score = 1
Elevated LDH (> 500 IU/L)5614.07.8 (3.2–19.1)28.46Useful adjunct; LDH 2–5× normal highly suspicious for malignancy
Weight loss308.05.0 (2.4–10.3)18.15Significant; combined with cytopenia strongly favors ALL
Hepatosplenomegaly5432.02.5 (1.4–4.5)22.37Overlap with systemic JIA limits specificity; context-dependent
Nocturnal bone pain659.018.4 (8.7–38.9)15.65Pain disproportionate to physical findings; nocturnal pattern suspicious
Limb pain (out of proportion)701.0553 (46.5–6,580)0.03Pain disproportionate to physical findings was consistently associated with underlying malignancy across included studies
ANA positivity1255.00.12 (0.05–0.29)12.46Absence of ANA in arthritis favors investigation for malignancy
S100A9/S100A12 levels (low)788.038.6 (14.2–104.9)22.13Novel biomarker with AUC 0.91; low levels discriminate ALL from JIA

 

↓  Table 3. Newcastle-Ottawa Scale Quality Assessment of Included Primary Studies
 
Ref.StudySelection (Max 4)Comparability (Max 2)Outcome (Max 3)Total (Max 9)Risk of biasCountryDesign
[13]Brix et al, 20224/42/22/38/9LowScandinaviaRetrospective
[14]ONCOREUM, 20214/42/23/39/9LowItalyProspective
[15]Archawanantakul et al, 20253/42/22/37/9ModerateThailandRetrospective
[16]Glerup et al, 2023 (S100-Biomarker Study)3/42/23/38/9LowNordicCross-sectional
[17]Nourbakhsh et al, 20253/42/22/37/9ModerateIranRetrospective
[18]Soltani et al, 20253/42/22/37/9ModerateIranCross-sectional
[19]Jari & Ana, 20254/42/22/38/9LowIntl.Systematic review
[20]Huang et al, 20243/42/22/37/9ModerateChinaCross-sectional
[21]Schulz et al, 2022 (ICON-JIA)4/42/22/38/9LowGermanyProspective
[22]Ailioaie et al, 20222/42/22/36/9ModerateRomaniaNarrative review
[1]Huang et al, 20242/42/22/36/9ModerateIntl.Review
[23]Liu et al, 20254/42/22/38/9LowUKSystematic review