Session Information
Session Type: ACR Concurrent Abstract Session
Session Time: 4:30PM-6:00PM
Background/Purpose: Macrophage activation syndrome (MAS) constitutes over 5% of multi-organ dysfunction syndrome in adults, leading to 60-70% mortality without early treatment. H-Score, the only validated diagnostic tool in adults, is often too complex for bedside use. Our objective was to determine clinical predictors of MAS and build a simple yet accurate risk calculator for use by clinicians to facilitate early recognition.
Methods: Electronic Health Records were used to identify patients hospitalized at a single academic center between 2009-12 with serum ferritin levels (SFL) of ≥ 2,000 ng/dL. Patients with chronic hemolytic disorders and those receiving chronic transfusions were excluded. Diagnosis of MAS was determined using a) ≥ 5/8 of HLH-2004 or ≥4/5 of adapted HLH-2004 criteria, or b) H score of ≥169, or c) treatment modified specifically for MAS. Control patients met none of these criteria. Organ Dysfunction and INfection (ODIN) definitions determined presence of organ dysfunction. Coagulopathy was determined based on score of ≥4 on “International Society of Thrombosis and Hemostasis Diagnostic Algorithm”. Hepatobiliary dysfunction was defined as ≥2 of: ALP, ALT, AST elevation more than 3 times normal limits, or bilirubin ≥ 5.8 mg/dL (sepsis definitions). The optimum SFL cut point was identified based on prediction accuracy for MAS at various thresholds. Following standard univariate analysis, logistic regression determined independent predictors of MAS. Parameter estimates were used for a weighted risk score for MAS.
Results: Of the 527 hyperferritinemic patients identified, 67 MAS cases [70% women, 60% non-white, median age 41 (18-83)] were identified; 69 controls with similar demographics were selected from remaining patients [67% women, 60% non-white, median age 42 (19-84)]. Univariate analysis of clinical and outcome parameters associated with presence of MAS is presented in Table-1. A SFL cut-off of 5,000ng/dL had maximal accuracy predicting MAS. Independent predictors of MAS were coagulopathy, hematologic dysfunction, SFL≥ 5,000 ng/dL, and fever≥ 101.3° F. Using a simple count of signs positive, MAS incidence was 0% (0/19) in patients with 0 signs positive versus 93% (27/29) in patients with 4 signs positive (p<.0001).
Conclusion: MAS increases the need for critical care admission and intervention. Presence of 3 of: coagulopathy, SFL≥ 5,000, cytopenia, or fever≥ 101.3° F should alert the clinicians to the possibility of MAS in hospitalized patients.
|
Table 1. MAS is significantly associated with higher likelihood of fever, elevated ferritin level, organomegaly, and organ dysfunction, as well as higher probability of death, prolonged hospitalization, and higher need for critical interventions. |
|||
|
MAS N= 67 |
non-MAS N= 69 |
p value* |
|
| Median Age (Range) |
41 (18-83) |
42 (19-84) |
0.9999 |
| Male Gender N (%) |
20 (30) |
23 (33) |
0.7144 |
| Non-White N (%) |
27 (40) |
28 (40) |
1.000 |
| Mean Ferritin (95% CI) |
28,891 14,319-43,462 |
6,140 (3,973-8,307) |
0,0022 |
| Hematologic N (%) |
55 (82) |
39 (57) |
0.0015 |
| Hepatobiliary N (%) |
48 (72) |
32 (46) |
0.0032 |
| DIC N (%) |
45 (67) |
16 (23) |
<0.0001 |
| CNS N (%) |
39 (58) |
25 (36) |
0.0158 |
| Infection N (%) |
48 (72) |
32 (46) |
0.0032 |
| Renal N (%) |
31 (46) |
29 (42) |
0.73 |
| Respiratory N (%) |
34 (51) |
21 (30) |
0.0228 |
| CVS N (%) |
41 (61) |
29 (42) |
0.0274 |
| Fever ° (SD) |
102.9 (1.9) |
100.6 (1.8) |
0.0001 |
| Triglyceride: mg/dL (SD) |
N= 59; 295(184) |
N= 15; 166 (98) |
0.0108 |
| Hepatomegaly N (%) |
35 (52) |
20 (29) |
0.0085 |
| Splenomegaly N (%) |
20 (30) |
5 (7) |
0.0008 |
| CRP mg/dL (SD) |
N=40; 159 (118) |
N=11; 142 (114) |
0.6719 |
| ESR mmHg/Hr (SD) |
N=44; 33 (49) |
N=10; 37 (21) |
0.8024 |
| Immune-suppressed N (%) |
37 (55) |
20 (29) |
0.0030 |
| Dead |
31 (46) |
16 (23) |
0.0066 |
| Treated for MAS |
36 (54) |
0 (0) |
na |
| Length of Hospital Stay Mean (SD) Median (Range) |
34.8 (36); 25 (3-178) |
12.3 (14.2); 9 (2-105) |
0.0001 |
| Critical Care Admission N (%) |
48 (72) |
29 (42) |
0.0006 |
| Mechanical Ventilation N (%) |
40 (60) |
19 (28) |
0.0002 |
| Vasopressor Use |
36 (54) |
22 (32) |
0.0149 |
| Renal Replacement Therapy (new) |
29 (43) |
10 (14) |
0.0003 |
| * pvalue is set at 0.05 MAS: macrophage activation syndrome; CI: confidence interval; SD: standard deviation; DIC: disseminated intravascular coagulation; CNS: central nervous system; CVS: cardiovascular system; CRP: C-reactive protein; ESR: erythrocyte sedimentation rate. | |||
|
Table 2: High ferritin, fever, coagulopathy, and cytopenia are independent predictors of MAS with equal weight. |
|||
| Predictor |
Parameter Estimate (SE) |
Odds Ratio (95% CI) |
p value* |
| Ferritin > 5000 |
0.78 (.28) |
4.75 (1.60-14.11) |
.0051 |
| Fever > 101.3 |
1.16 (.32) |
10.26 (2.95-35.74) |
.0003 |
| Hematology |
0.81 (.32) |
5.05 (1.45-17.55) |
.011 |
| DIC |
0.88 (.27) |
5.77 (1.99-16.70) |
.0012 |
| Infection |
0.48 (.29) |
2.64 (0.85-8.19) |
.09 |
| Renal |
0.50 (.30) |
2.72 (0.83-8.89) |
.098 |
| * pvalue is set at 0.05
MAS: macrophage activation syndrome; CI: confidence interval; DIC: disseminated intravascular coagulation |
|||
To cite this abstract in AMA style:
Shakoory B, Mohtasham N, Mullen M, Amdur R, Chatham WW. Macrophage Activation Syndrome Is Identified By Coagulopathy, Hyperferritinemia, Fever, and Cytopenia in Hospitalized Patients, Resulting in Poor Outcome [abstract]. Arthritis Rheumatol. 2016; 68 (suppl 10). https://acrabstracts.org/abstract/macrophage-activation-syndrome-is-identified-by-coagulopathy-hyperferritinemia-fever-and-cytopenia-in-hospitalized-patients-resulting-in-poor-outcome/. Accessed .« Back to 2016 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/macrophage-activation-syndrome-is-identified-by-coagulopathy-hyperferritinemia-fever-and-cytopenia-in-hospitalized-patients-resulting-in-poor-outcome/