Introduction
Macrophage activation syndrome (MAS) is a severe, potentially fatal form of secondary hemophagocytic lymphohistiocytosis that complicates systemic juvenile idiopathic arthritis (sJIA), formally termed Still's disease, and is characterized by uncontrolled activation of macrophages and T lymphocytes leading to fever, cytopenias, hyperferritinemia, transaminitis, hypertriglyceridemia, and coagulopathy [1]. Up to 10% of MAS episodes are fatal, underscoring the importance of early recognition and prompt treatment [2].
The 2016 EULAR/ACR/PRINTO classification criteria require known or suspected Still's disease with fever, ferritin > 684 ng/mL, and at least two of: platelet count ≤ 181 × 109/L, AST > 48 U/L, triglycerides > 156 mg/dL, or fibrinogen ≤ 360 mg/dL [1, 2, 3]. These criteria were designed for classification rather than diagnosis. Fibrinogen is an acute‐phase reactant that is expected to be raised during active systemic inflammation and typically falls only as MAS evolves.
The criteria can therefore be satisfied through the other components before fibrinogen falls, and a normal or high fibrinogen should not delay recognition.
MAS may complicate sJIA that has not yet been recognized, and a variety of infectious triggers, most notably viral pathogens, have long been implicated in its pathogenesis [4]. Since the emergence of the COVID‐19 pandemic, SARS‐CoV‐2 has been recognized both as a trigger of MAS in patients with established rheumatic disease and as a driver of multisystem inflammatory syndrome in children (MIS‐C), a distinct hyperinflammatory entity that shares overlapping clinical and laboratory features with MAS, including fever, elevated inflammatory markers, cytopenias, and coronary artery abnormalities [4, 5, 6, 7]. This overlap complicates the evaluation of children presenting with hyperinflammation and SARS‐CoV‐2 positivity.
We describe a child with MAS complicating an sJIA‐like illness of several weeks' duration, with coronary artery dilation and SARS‐CoV‐2 positivity, in whom sJIA‐associated MAS and MIS‐C could not be fully separated, and we discuss the limits of the evidence for a SARS‐CoV‐2 trigger.
Case History/Examination
A previously healthy boy, aged approximately 5 years and 5 months, with no significant past medical or family history, presented 1 month prior to his first hospital admission with an acute febrile polyarticular illness. He was unvaccinated against SARS‐CoV‐2, and no known COVID‐19 contacts were reported at the time. SARS‐CoV‐2 IgG was reported positive at PICU admission, but this cannot be taken as evidence of prior exposure because its timing relative to IVIG administration is unknown (see Investigations).
He initially presented with swelling and erythema of the left ankle, followed within days by similar involvement of the right ankle and an inability to bear weight, accompanied by high‐grade daily fevers, lethargy, poor appetite, and a generalized rash. The fever pattern was spiking, with a documented peak temperature of 40.5°C, and the rash was evanescent, salmon‐colored, distributed over the trunk and extremities, appearing with fever spikes and fading between episodes. Cervical lymphadenopathy was present, with the largest node measuring 1.5 cm.
He also had lip changes consistent with cheilitis, peripheral extremity edema, abdominal pain, and diarrhea. He was started on oral prednisone (1.5 mg/kg/day), with dramatic improvement followed by prompt recurrence of symptoms on tapering.
Approximately 4 weeks later, he was readmitted with polyarticular joint involvement, inability to walk, and high‐grade fever. Two days after this readmission he was transferred to a tertiary referral center for further evaluation, and approximately 2 weeks after readmission he required transfer to a pediatric intensive care unit (PICU) in the context of rapid clinical deterioration, prompting extensive diagnostic evaluation (see Investigations). The sJIA‐like features (fever, arthritis, rash) therefore preceded the PICU deterioration by several weeks, and the first positive SARS‐CoV‐2 test was obtained on PICU Day 8, approximately 6 weeks after the onset of arthritis, fever, and rash.
Because this positive test followed prolonged hospitalization, hospital‐acquired infection cannot be excluded.
Differential Diagnosis
Systemic Juvenile Idiopathic Arthritis
From the second admission onward, sJIA was the leading clinical diagnosis. The illness included daily spiking fever (peak 40.5°C) over several weeks, polyarthritis involving both ankles and subsequently additional joints, and an evanescent, salmon‐colored, fever‐associated generalized rash distributed over the trunk and extremities. The documented duration of arthritis exceeded 6 weeks.
Cervical lymphadenopathy (largest node 1.5 cm) and hepatosplenomegaly were reported; imaging also showed minimal pericardial and pleural effusions. Rheumatoid factor and antinuclear antibodies were negative, and no personal or family history of psoriasis or inflammatory bowel disease was reported. These findings supported a sJIA phenotype, although they were not specific and could overlap with MIS‐C.
Classification criteria should not be interpreted as independently establishing the diagnosis. Applied retrospectively, the documented features (daily spiking fever for several weeks, arthritis persisting beyond 6 weeks, evanescent rash, lymphadenopathy, hepatosplenomegaly and serositis) satisfy the ILAR [8] and PRINTO 2019 [9] definitions of sJIA. Both sets of criteria, however, presuppose that infection, malignancy, and other mimics have been excluded, which was incomplete here (see below).
Infectious and Reactive Arthritides
Reactive arthritis, post‐streptococcal arthritis, and other infectious arthritides were considered. ASO/anti‐DNase B titers, throat culture, and serology for parvovirus B19, EBV, CMV, and Mycoplasma were not performed, and joint aspiration was not undertaken. These unperformed tests limit the extent to which these alternatives can be excluded.
The small serosal effusions and hepatosplenomegaly are not specific to sJIA and may also occur in other systemic inflammatory conditions, including MIS‐C.
Malignancy, principally leukemia, was suspected at readmission and again during the period of deterioration but was excluded by two bone marrow biopsies and a peripheral blood film, all without blasts; hemophagocytosis was not reported in either marrow specimen.
Primary HLH was considered because of the prolonged fever and subsequent deterioration. Whole‐exome sequencing did not identify pathogenic variants in the primary HLH‐associated genes assessed; the scope and coverage of testing were not reported. A negative exome does not exclude primary HLH because some deep intronic, structural, and copy‐number variants may not be detected.
MIS‐C and Kawasaki Disease
MIS‐C was considered during the PICU admission because of coronary artery dilation and SARS‐CoV‐2 IgG positivity, followed by a positive rapid antigen test on PICU Day 8. The available findings also prompted consideration of Kawasaki disease. Fever duration exceeded the five‐day threshold used in the American Heart Association framework [10].
Reported features included oral mucosal changes (cheilitis), a polymorphous rash, and peripheral extremity edema; cervical lymphadenopathy (1.5 cm) was at the diagnostic threshold. Conjunctival injection was not documented and therefore cannot be classified as present or absent. Hypoalbuminemia, elevated transaminases, anemia, and leukocytosis were reported, and echocardiography showed dilation of the right coronary artery and left main coronary artery (LMCA; reported diameter 3.2 mm); coronary Z‐scores were not reported.
The criteria assessment should therefore be considered provisional. At least three principal clinical features were present together with coronary dilation, so the AHA criteria for incomplete Kawasaki disease could be met if coronary Z‐scores confirmed dilation (Z‐score ≥ 2); this cannot be verified.
Application of MIS‐C Criteria
By the CDC case definition [11] (age < 21 years, fever, laboratory evidence of inflammation, clinically severe illness with multisystem involvement, and evidence of SARS‐CoV‐2 infection or exposure), the child had fever, markedly raised inflammatory markers, and cardiac, gastrointestinal, mucocutaneous, hematologic and hepatic involvement, with a positive antigen test and IgG; however, the requirement for no plausible alternative diagnosis was not met, because sJIA‐associated MAS is such an alternative. By the WHO case definition [12] (age 0–19 years; fever for at least 3 days; two of rash or mucocutaneous inflammation, hypotension or shock, cardiac involvement including coronary abnormality, coagulopathy, or acute gastrointestinal symptoms; raised inflammatory markers; no other obvious microbial cause; and evidence of SARS‐CoV‐2), the child had rash and mucocutaneous changes, coronary dilation, raised D‐dimer and gastrointestinal symptoms (hypotension or shock was absent), but an alternative non‐microbial cause (sJIA) was not excluded and the evidence of SARS‐CoV‐2 is limited (see Investigations). The criteria are therefore met on their face but are not specific in this setting.
MIS‐C could not be excluded. A monophasic course with sustained drug‐free remission is compatible with MIS‐C but is not sufficient to distinguish it from sJIA, which can also have a monophasic course.
Bacterial infection and sepsis were considered during recurrent fever episodes. Repeated blood, urine, and cerebrospinal fluid cultures were reported as negative, as were several infectious serologies (Brucella, hepatitis A/B/C, Leishmania); empiric antibiotics were nevertheless administered during the clinical course.
Macrophage Activation Syndrome
The 2016 EULAR/ACR/PRINTO classification criteria were reviewed across available time points (Table 2) [3]. At readmission, ferritin (1675 ng/mL) and AST (108 U/L) met criteria, but platelets (688 × 109/L) and fibrinogen (610–712 mg/dL) remained elevated rather than low, and triglycerides were not documented—criteria were not yet fulfilled at this stage. At PICU admission, ferritin was 83,430 ng/mL, AST was 2068 U/L, and triglycerides were 937 mg/dL; in the presence of fever, AST and triglycerides alone satisfied the two‐criterion requirement, so the criteria were fulfilled without any contribution from fibrinogen or platelets, which had not yet fallen.
Fibrinogen later declined to 166 mg/dL on PICU Day 17, alongside a platelet nadir of 73 × 109/L on Day 16, confirming ongoing/worsening MAS activity rather than marking the first fulfillment of criteria. Overall, the findings were consistent with MAS complicating an sJIA‐like illness, while the contribution of SARS‐CoV‐2 and the possibility of MIS‐C remained uncertain.
Investigations
Echocardiography during the PICU admission demonstrated dilation of the right coronary artery and LMCA, with the LMCA diameter reported as 3.2 mm; the RCA diameter was not recorded. Coronary Z‐scores, the reference equation used, and ventricular function (ejection fraction/shortening fraction) were not available. Troponin was reported as 11.3 ng/mL.
NT‐proBNP was not performed. Follow‐up echocardiography approximately 2 months after discharge showed complete normalization of coronary artery dimensions (LMCA 2.8 mm, LAD 2.2 mm), with no residual ectasia, allowing discontinuation of aspirin. The patient did not develop shock and did not require inotropic support.
Contrast‐enhanced CT of the chest, abdomen, and pelvis revealed ground‐glass opacities, minimal pericardial and pleural effusion, hepatosplenomegaly, and paratracheal lymphadenopathy.
SARS‐CoV‐2 Testing
A rapid antigen test was positive on PICU Day 8, suggesting active infection. SARS‐CoV‐2 IgG was reported positive at PICU admission (Day 1), based on its appearance within the same admission laboratory panel as D‐dimer, troponin, and CK‐MB; the exact sampling time, assay, and antigen target were not available. As IgG was apparently sampled at admission, IVIG dosing dates were not available to confirm timing relative to this result; passive antibody transfer from IVIG therefore cannot be excluded as a contributor to IgG positivity.
Because IVIG preparations may contain anti‐SARS‐CoV‐2 antibodies, the IgG result should be regarded as uninterpretable as evidence of infection. No PCR result was documented; only rapid antigen and IgG serology were performed. Consequently, no Ct values are available, the assay target (spike or nucleocapsid) is unknown, and the antigen result cannot distinguish community‐acquired from hospital‐acquired infection.
Serial laboratory monitoring showed progressive severe neutropenia and anemia requiring packed red blood cell transfusion, followed by thrombocytopenia and a marked decline in fibrinogen from an initially elevated level. Ferritin followed a biphasic course: 83,430 ng/mL at PICU admission (Day 1), declining to a nadir of 7090 ng/mL by Day 11 in parallel with initial clinical improvement on anakinra and corticosteroids, then rebounding to 61,760 ng/mL by Days 16–17, coinciding with the onset of thrombocytopenia and hypofibrinogenemia which indicated worsening MAS activity (the criteria having already been met at PICU admission), before declining to 236 ng/mL at discharge (Tables 1, 2).
The first bone marrow biopsy, performed 5 days after readmission, showed normocellular trilineage hematopoiesis without blasts. As neutropenia worsened, concern for malignancy was reconsidered. A second biopsy showed a slightly hypocellular marrow with reactive CD3‐positive T‐lymphocytosis and no blasts; the peripheral blood film also showed no blasts.
Hemophagocytosis was not reported in either specimen.
Whole‐exome sequencing did not identify pathogenic variants in the primary HLH‐associated genes assessed. The specific genes tested, sequencing coverage, and whether copy‐number analysis was included were not available in the records reviewed. Functional or immunologic testing (NK‐cell function, CD107a degranulation, perforin, SAP, or XIAP) was not performed.
Treatment
Initial treatment was oral prednisone (1.5 mg/kg/day), with marked clinical improvement followed by recurrence during tapering. After readmission and deterioration, treatment included pulse methylprednisolone, dexamethasone, two doses of IVIG, prophylactic enoxaparin for elevated D‐dimer, and empiric broad‐spectrum antibiotics despite repeatedly negative cultures. Anakinra was started during the PICU admission, with significant initial improvement allowing dose weaning; however, the patient deteriorated coinciding with confirmed SARS‐CoV‐2 infection, prompting re‐escalation of anakinra dosing without adequate response.
Cyclosporine was subsequently added on (PICU Day 17) for ongoing disease activity (see Table 3 for doses, start days relative to PICU admission, and durations).
Pulse methylprednisolone was given for severe systemic inflammation and suspected MAS, whereas dexamethasone was started as part of an HLH‐directed protocol; the records do not document a single rationale for using the two together or the length of their overlap, and combined high‐dose corticosteroid exposure should not be taken as a recommended approach. Aspirin was given for coronary dilation and was continued despite a platelet nadir of 73 × 109/L and concurrent viral infection, accepting a risk of bleeding (a platelet transfusion was needed after bone marrow biopsy‐site oozing) and a theoretical risk of Reye syndrome; this was a clinical risk–benefit judgment rather than a standard recommendation for thrombocytopenic patients.
Outcome and Follow‐Up
Following multimodal immunomodulatory treatment, white‐cell count, platelet count, fibrinogen, and ferritin gradually normalized. Cyclosporine was tapered first, followed by gradual withdrawal of corticosteroids and anakinra over approximately 1 year of outpatient follow‐up. At discharge, the child remained on aspirin, reduced‐dose corticosteroids, and anakinra.
Follow‐up echocardiography approximately 2 months after discharge showed complete normalization of coronary artery dimensions (left main coronary artery 2.8 mm, left anterior descending artery 2.2 mm), with no residual coronary ectasia, and aspirin was discontinued at that visit. He has reportedly remained in drug‐free clinical remission for more than 4 years, without recurrence of sJIA features or MAS. The observed recovery cannot be attributed to any single agent or regimen in this single case.
Discussion
MAS should be considered in children with persistent fever, cytopenias, hepatic dysfunction, and hyperferritinemia, including those with an evolving sJIA‐like illness. An initially elevated fibrinogen does not exclude evolving MAS when other findings are compelling, and serial monitoring—as illustrated by this patient's biphasic ferritin course—may better capture disease activity than isolated values. In children with SARS‐CoV‐2 positivity and coronary artery dilation, sJIA‐associated MAS and MIS‐C may remain difficult to distinguish [5, 6, 13]; causality should not be attributed to SARS‐CoV‐2 without adequate temporal and virologic evidence.
Here, the positive antigen test on PICU day 8 came about 6 weeks after the onset of arthritis, fever and rash, so SARS‐CoV‐2 could at most have contributed to the later deterioration, and even that cannot be established given the possibility of hospital acquisition, the uninterpretable IgG result and the absence of PCR data. The child's recovery and sustained four‐year drug‐free remission are clinically encouraging but do not establish the efficacy of the multi‐agent regimen.
The profound neutropenia (ANC 0.005 × 109/L on PICU Day 14) with a slightly hypocellular marrow is unusual for MAS, in which cytopenias more often reflect peripheral consumption and hemophagocytosis than marrow failure, and hemophagocytosis was not reported on either marrow specimen. Drug‐induced neutropenia is a plausible contributor: anakinra had been started before the nadir, and he had also received empiric broad‐spectrum antibiotics and other agents, whereas cyclosporine was started after the nadir and cannot explain it. Virus‐related marrow suppression, whether from SARS‐CoV‐2 or from an untested virus such as parvovirus B19, EBV, or CMV, is another possibility.
The neutropenia therefore remains unexplained and probably multifactorial; malignancy was excluded by two marrow examinations.
Treatment was heterogeneous and cannot be reproduced exactly from the available records (Table 3). Anakinra is used in both sJIA‐associated MAS and MIS‐C [14], so a response to it does not discriminate between them, and concurrent corticosteroid, IVIG, and cyclosporine exposure further limits any inference about individual agents.
Limitations
The main limitation is that the available retrospective data do not allow definitive separation of sJIA‐associated MAS from MIS‐C with secondary hyperinflammation. The timing and assay details of SARS‐CoV‐2 IgG testing, including its relation to IVIG, are incomplete; hospital‐acquired or incidental positivity cannot be excluded. Coronary Z‐scores and some cardiac investigations (ventricular function, NT‐proBNP) are missing, though follow‐up echocardiography confirmed complete resolution of coronary ectasia approximately 2 months after discharge.
The negative exome does not exclude all forms of primary HLH; the scope of genetic testing and functional immunologic testing was not available in the records reviewed, and although hemophagocytosis was not reported on either marrow biopsy, its specific assessment in the original pathology reports requires confirmation. The profound neutropenia was not fully explained and may have been multifactorial (drug‐related, virus‐related or MAS‐related). No cytokine profiling was performed.
The clinical rationale for specific treatment decisions—including combining pulse methylprednisolone with dexamethasone and continuing aspirin despite a platelet nadir of 73 × 109/L, during which platelet transfusion was administered—could not be fully reconstructed from the available records. In addition, classification criteria were applied retrospectively and are not diagnostic rules. As a single case, this report cannot establish causality or generalize treatment efficacy.
Conclusion
MAS may complicate sJIA that has been evolving for weeks, and it can be recognized from AST, triglycerides, and ferritin even when fibrinogen is still normal or high. When SARS‐CoV‐2 positivity and coronary artery dilation coexist, sJIA‐associated MAS and MIS‐C may be indistinguishable, and a SARS‐CoV‐2 trigger should not be assumed without adequate temporal and virologic evidence. The favorable outcome is encouraging, but a single case cannot show that the regimen produced the four‐year drug‐free remission.
Author Contributions
Ayham Slimi: conceptualization, data curation, writing – review and editing, writing – original draft. Haya Jaradat: data curation, conceptualization, writing – original draft, writing – review and editing. Baraa Nassar: conceptualization, data curation, writing – original draft, writing – review and editing.
Fawzy M. Abunejma: supervision.
Funding
The authors have nothing to report.
Ethics Statement
This study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was reviewed and approved by the Institutional Ethics Committee of Palestine Polytechnic University. A formal approval number was not issued for this case report.
Written informed consent was obtained from the patient's parent or legal guardian for publication of all clinical details; no images are included.
Conflicts of Interest
The authors declare no conflicts of interest.