Results
Cohort Characteristics: Thirty patients were diagnosed with MIS-C during the study period. Median age was 9.5 ± 5.1 years and 14 (47%) were male. Twenty-four (80%) of patients had significant comorbidities, with the most common being obesity/overweight (n=17, 56%), followed by asthma (n=7, 23%). All patients had either positive SARS-CoV-2 RT-PCR (n=16, 53%) and/or positive antibodies (n=22, 73%), 8 (27%) had both. Eight patients (27%) had an additional documented COVID-19 exposure. Eleven (37%) patients required admission to an ICU, with 6 (20%) requiring vasopressor support, 5 (17%) requiring positive pressure ventilation, and 1 (3%) requiring mechanical ventilation. Median ICU length of stay was 6 days (interquartile range [IQR] 3-9). On echocardiogram, 7 (23%) had coronary artery dilation/aneurysm and 17 (57%) had ventricular dysfunction (mean ejection fraction 54%, standard deviation [SD] 9.2 %).
Laboratory Characteristics: Initial and peak laboratory data are described in Table 1. Markers of inflammation, including ferritin, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP), were elevated on initial presentation. Peak white blood cell count occurred on median day 0 (IQR 0-12) of illness in those not started on steroids, and on median day 8 (IQR 4-12) of illness in those on steroids. Patients did not have significant anemia and were initially mildly thrombocytopenic with a subsequent increase in platelet count later in illness. Liver and renal function was overall normal. Cardiac biomarkers were notable for a mildly elevated median peak b-type natriuretic peptide and a normal median troponin, although 5 (17%) patients had an elevated troponin over the course of illness.
Coagulation profiles are summarized in Table 1. Coagulation profiles were obtained on a median day 0 (IQR 0-1) of illness, prior to initiation of antiplatelet and anticoagulation medication in 21/24 (88%) patients started on those therapies. Three patients had coagulation profiles sent 1-2 days after initiating therapies; aspirin in 3/3 and treatment dose low-molecular weight heparin (LMWH) in 2/3, and had no significant prolongation in their coagulation profiles. Baseline values for activated thromboplastin time (aPTT) and prothrombin time (PT) were slightly prolonged, though international normalized ratio (INR) was within normal limits. Peak values were notable for an elevated mean aPTT and PT, with mean INR slightly increased. D-dimer and fibrinogen were elevated on initial presentation and increased later in illness.
TEG Profiles: Table 2 contains serial TEG data over the course of MIS-C illness with patients’ corresponding thromboprophylaxis regimens. The initial TEG profile was obtained on day 4 (IQR 3 – 8) of illness in 19 (63%) patients. At time of presentation, patients with MIS-C had abnormal TEG profiles compared with age- and gender- matched controls (Figure 1), including decreased K time (1.1 vs. 1.7 min,P <0.01), increased alpha angle (75.0 vs. 65.7 degrees,P <0.01), increased maximum amplitude (70.8 vs. 58.3 mm,P <0.01), and decreased Ly-30 (1.1 vs. 3.7%,P =0.03). These findings are consistent with increased rate of clot formation and clot strength, and slower fibrinolysis (depicted in Figure 2). In correlation analysis (Figure 3), TEG maximum amplitude was directly correlated with ESR (r =0.60, P =0.02), initial platelet count (r =0.67, P <0.01), and peak platelet count (r =0.51, P =0.03). TEG alpha angle was directly correlated with peak platelet count (r =0.54,P =0.02). No other inflammatory markers were associated with TEG maximum amplitude or alpha angle, including D-dimer, CRP, procalcitonin, ferritin, and fibrinogen. There was no association between TEG alpha angle nor amplitude with any markers of disease severity, including ejection fraction, coronary dilation/aneurysm, inotropic support, intubation or ICU stay.
Over the course of weeks 2-6 of illness, a second TEG profile was obtained in 13 (43%) patients, and a third TEG profile was obtained in 6 (20%) patients. By the second TEG, 9/11 (82%) of the patients on ASA had AA inhibition > 50% and by the third TEG, 6/6 (100%) of the patients on ASA had AA inhibition > 50%. There was a significant difference in AA inhibition in those on ASA versus not on ASA; 59.0 (SD 34.6) vs. 4.1 (SD 5.7), P < 0.01; but no significant difference in clot strength. There was a progressive increase in ADP inhibition over time, despite patients not being on any medications known to cause ADP inhibition.
Treatment Regimens and Adverse Events: Patients were treated with antiplatelet and anticoagulation regimens as summarized in Table 3, with guidance set forth by our institution’s Cardiac Antithrombosis Management Program (algorithm used by this team is described in Supplemental Figure S1). Treatment is described for all 30 patients with MIS-C, not solely those who had TEG performed. Twenty-two (73%) patients received aspirin (ASA), 7 (23%) received prophylactic LMWH, 11 (37%) received treatment-dose LMWH, 2 (8%) received unfractionated heparin, and 12 (40%) patients received apixaban. Seventeen (57%) patients were on dual therapy at one point during their treatment: 14/17 (82%) were on ASA and LMWH (9 on treatment-dose LMWH, 5 on prophylactic LMWH) and 3/17 (18%) were on ASA and apixaban (2 were briefly bridged on a heparin drip to apixaban). Only 5 (17%) patients received no thromboprophylaxis. In 4/5 of these cases, the patients had short hospitalizations and normal echocardiograms, and thromboprophylaxis was not provided based on individual provider preference. One patient had significant thrombocytopenia precluding thromboprophylaxis.
No thrombotic events were documented in this population during their hospitalization or clinical follow-up at our institution. Six (20%) patients had a total of 7 bleeding events, of which 2 (7%) were clinically relevant non-major (including bloody stools, epistaxis, or gum bleeding requiring further evaluation or intervention) and 5 (17%) were minor (including epistaxis, gum bleeding, or bleeding from LMWH sites or sites of trauma). All of the bleeding events occurred when the patient was on some form of anticoagulation or antiplatelet agent: 6/7 events were on ASA, 3/7 were on apixaban, and 2/7 were on LMWH (1 treatment-dose, 1 prophylactic). 4/7 events happened while on dual therapy, and 2/4 of these events resulted in discontinuation of the anticoagulant (apixaban) with continuation of ASA. 2/7 events happened while just on ASA, which was stopped as a result. Only one bleeding event occurred on prophylactic LMWH in a patient with bloody stools and a diagnosis of inflammatory bowel disease, and was therefore unlikely to be exclusively attributed to thromboprophylaxis.