Background/Purpose: It is common view that macrophage activation syndrome (MAS) bears close similarities with primary hemophagocytic lymphohistiocytosis (pHLH). The resemblance of their clinical and laboratory manifestations may make it difficult to differentiate the two conditions, particularly when pHLH occurs at a later age or MAS develops at onset of systemic juvenile idiopathic arthritis (sJIA), when arthritis is not yet present. However, early recognition is important because pHLH is often more severe than MAS and the therapeutic approaches are different. The aim of our study is to develop and validate a diagnostic score that discriminates pHLH from sJIA-associated MAS

Methods:  The clinical, laboratory and histopathologic features of 362 patients with sJIA-associated MAS and of 258 patients with pHLH were collected in a multinational collaborative project involving pediatric rheumatologists and pediatric hematologists. 80% of the study population was used to develop the score and the remaining 20% constituted the validation sample. The features with the strongest association with pHLH in univariate analyses (odds ratio > 5) were further scrutinized in multivariate logistic regression procedures. Each variable that entered the best fitting model was then assigned a score, based on its statistical weight. The MH score was made up with the individual scores of the selected variables. The cut-off in the MH score that discriminated best pHLH from MAS was calculated by means of ROC curve analysis. The sensitivity (SE), specificity (SP), area under the curve (AUC) and kappa value of the MH score were calculated for both the developmental and validation samples

Results: The following 6 variables entered the best fitted model of logistic regression analysis, that is, were most closely associated with a diagnosis of pHLH: age at disease onset ≤ 1.6 years, neutrophil count ≤ 1400/µl, fibrinogen ≤ 131 mg/dl, splenomegaly, platelet count ≤ 78000/µl, hemoglobin ≤ 8.3 g/dl. The MH score ranged from 0 to 123. Its median value was 97 (IQR 75-123) in pHLH patients and 12 (IQR 11-34) in MAS patients. The probability of a diagnosis of pHLH ranged from < 1% for a score < 11 to > 99% for a score ≥ 123. A cut-off value > 59 revealed the best performance in discriminating pHLH from MAS (SE=91%, SP=93%, AUC=0.92, kappa=0.85). The strong diagnostic power of the MH score was confirmed in the validation sample

Conclusion:  The MH score is a powerful tool that facilitates timely discrimination of pHLH from MAS. Its application in routine clinical care may aid practitioners to identify those patients who are more likely to have pHLH and may, thus, deserve diagnostic confirmation with appropriate genetic and functional testing

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/development-and-initial-validation-of-the-mh-score-a-new-diagnostic-tool-that-differentiates-primary-hemophagocytic-lymphohistiocytosis-from-macrophage-activation-syndrome/