IVIG Compared With IVIG Plus Infliximab in Multisystem Inflammatory Syndrome in Children

Eligible patients included any patient diagnosed with and treated for MIS-C based on CDC criteria between April 1, 2020, and February 28, 2021, at our 528-bed system of care, which includes an academic quaternary care pediatric hospital and affiliated sites.¹⁶  Study investigators extracted data from the electronic health record via systematic retrospective chart review, including sociodemographic information, clinical characteristics, laboratory and cardiac evaluations, medications, and outcomes as defined below. Study data were collected and managed by using Research Electronic Data Capture tools hosted at the University of Colorado.¹⁷  Research Electronic Data Capture is a secure Web-based application designed to support data capture for research studies. Approval was obtained from the Colorado Multiple Institutional Review Board. Patients were separated into 2 groups for comparison on the basis of choice of initial therapy: treatment group receiving intravenous immunoglobulin alone (IVIG-Alone) and treatment group receiving intravenous immunoglobulin and infliximab (IVIG+Infliximab).

Initial therapy for MIS-C was defined as the earliest administered immunomodulatory medication(s), including IVIG, biologic agents (eg, infliximab), and/or corticosteroids. If ≥2 immunomodulatory medications were administered, they were considered part of the initial therapy if they were given within 24 hours of treatment initiation, defined as the start time of the earliest administered agent. Patients were included in the study if their initial therapy comprised IVIG or IVIG plus infliximab and were excluded if initial therapy included alternative medications. IVIG dosing was 2 g/kg, often but not always to a maximum of 70 g, and was administered over 10 to 12 or 20 to 24 hours depending on cardiac function and patient fluid status. If given, the infliximab dose was 10 mg/kg. Additional therapy was defined as administration of an immunomodulatory agent >24 hours after treatment initiation. Additional therapies were given for any of the following indications: fever (≥38.0°C) >24 hours after completion of initial therapy, continued need for vasoactive medication, worsening echocardiogram findings, and/or laboratory evidence of persistently high inflammation. Most patients also received aspirin for coronary artery protection; exceptions included patients with thrombocytopenia (platelet count of <150 000/m³).

Treatment decisions were made by clinical team discretion with assistance from institutional MIS-C guidelines. Initial guidelines suggested the use of IVIG alone as initial therapy unless patients had coronary artery dilation of the right coronary artery (RCA) and/or left anterior descending artery (LAD) (with z scores ≥2.5). In December 2020, institutional guidance changed to recommend initial therapy with IVIG plus infliximab for patients with RCA and/or LAD dilation (with z scores ≥2.0), left ventricular (LV) dysfunction (ejection fraction [EF] <55%), and/or hypotension.

The primary outcome was the need for additional immunomodulatory therapy. Secondary outcomes included hospital length of stay, ICU length of stay, new development or worsening of LV dysfunction, new vasoactive medication requirement, duration of vasoactive medications, change in inflammation as measured by the C-reactive protein (CRP) level, and need for diuretics. New or worsened LV dysfunction was defined as newly developed EF <55% at ≥24 hours after treatment initiation or, for those with EF <55% on the initial echocardiogram, a decrease in EF by ≥5% at ≥24 hours after treatment initiation. New vasoactive medication was defined as initiation of a new or additional vasoactive medication ≥24 hours after MIS-C treatment initiation, not including escalation of a previously prescribed medication. Effect of treatment on the CRP level was measured by the percentage change from the pretreatment level at 2 time points, ≥24 to <48 hours and ≥48 to <72 hours after treatment initiation.

In addition to the 2-group analysis described above comparing IVIG-Alone and IVIG+Infliximab, a 3-group subanalysis was performed. IVIG-Alone remained the same, but IVIG+Infliximab was divided into 2 groups on the basis of timing of administration of infliximab relative to IVIG administration: the IVIG-followed-by-infliximab group and the infliximab-followed-by-IVIG group. The purpose was to investigate potential effects of infliximab administration timing relative to IVIG administration on outcomes.

Medians with interquartile ranges (IQRs) and counts with percentages were used to summarize continuous and categorical variables, respectively. For 2- and 3-way comparisons of continuous variables, Wilcoxon rank tests and nonparametric Kruskal-Wallis tests were performed, respectively. All statistical tests were 2-sided, and P < .05 was considered significant. If 3-way comparisons were significant, Wilcoxon rank tests were performed to determine which groups significantly differed. For categorical variables, we made comparisons using the Pearson χ² tests or, when assumptions for χ² tests were not met, Fisher’s exact test or a generalization of Fisher’s exact test for m-by-n groups. All analyses were performed in SAS 9.4 (SAS Institute, Inc, Cary, NC).

During the study period, 79 patients met the CDC case definition and received treatment of MIS-C. Seven (9%) patients were excluded from analysis: 6 whose initial therapy included other immunomodulatory medications in addition to IVIG with or without infliximab and 1 who received corticosteroids for pharyngitis 2 and 4 days before initiation of MIS-C-directed initial therapy. The 72 remaining patients were divided into 2 groups on the basis of initial therapy, with 20 (28%) patients in IVIG-Alone and 52 (72%) in IVIG+Infliximab (Fig 1).

There were no statistically significant differences between groups in age, race and ethnicity, presence of underlying conditions, organ system involvement, overall need for respiratory support, or need for vasoactive medications (Table 1). Groups differed in admit location, with 2 of 20 (10%) in IVIG-Alone and 29 (56%) in IVIG+Infliximab initially admitted to the ICU (P < .001) (Table 1). Small but statistically significant differences were found in incomplete KD criteria fulfillment, as well as initial ferritin, sodium, and albumin between groups (Table 2). An abnormal echocardiogram finding during hospitalization, defined as presence of coronary artery dilation and/or LV dysfunction, was more prevalent in IVIG+Infliximab at 37 of 52 (71%) compared with IVIG-Alone at 8 of 20 (40%) (P = .03) (Table 2).

The epidemiological curve of MIS-C, stratified by initial therapy choice and need for additional therapy, is shown in Fig 2. At our institution, a dramatic increase in MIS-C cases followed the surge in SARS-CoV-2 infections in Colorado, which peaked in November 2020 at >6000 per day, continuing into early January 2021.¹⁸  Outcomes are reported in Table 3. Additional therapy was required in a total of 29 patients, including 13 of 20 (65%) treated with IVIG alone (IVIG-Alone) compared with 16 of 52 (31%) treated with IVIG plus infliximab (IVIG+Infliximab) (P = .01). Specific description of additional therapies administered can be found in Fig 1. Of the 49 patients who received infliximab at any point, 1 (2%) developed a rash; otherwise, there were no adverse events.

Median (IQR) hospital lengths of stay in days were 4.5 (3.4–5.9) and 3.7 (3.2–4.5) in IVIG-Alone and IVIG+Infliximab, respectively (P = .10) (Table 3). Median (IQR) ICU lengths of stay in days were 3.3 (2.2–3.8) and 1.8 (1.1–2.1) in IVIG-Alone and IVIG+Infliximab, respectively (P = .001). New or worsened LV dysfunction developed in 4 of 20 (20%) and 2 of 52 (4%), respectively (P = 0.05). New vasoactive medication requirement developed in 3 of 20 (15%) and 2 of 52 (4%), respectively (P = 0.13). Diuretics were required in 10 of 20 (50%) and 23 of 52 (44%), respectively (P = .79). The median (IQR) percentage changes in the CRP level at 24 hours after treatment initiation compared with pretreatment were 0% (−29% to 66%) and −46% (−62% to −15%) (P < .001), and at 48 hours after treatment initiation were −5% (−41% to 57%) and −70% (−79% to −49%), for IVIG-Alone and IVIG+Infliximab, respectively (P < .001). The median (IQR) days to fever resolution were 3 (2 to 4) and 2 (1 to 3), respectively (P = .12). The median (IQR) vasoactive medication duration was 36 hours (15 to 64) and 19 hours (14 to 33), respectively (P = .39).

A subanalysis was performed to compare patients on the basis of timing of infliximab administration relative to IVIG (Supplemental Tables 4 through 6). This was a 3-group subanalysis in which IVIG-Alone remained the same (n = 20), but IVIG+Infliximab was divided into 2 groups on the basis of timing of administration of infliximab relative to IVIG administration: the IVIG-followed-by-infliximab group (n = 16) and the infliximab-followed-by-IVIG group (n = 36). The IVIG-followed-by-infliximab group received infliximab at a median of +14 hours after initiation of IVIG (IQR 8 to 17), and the infliximab-followed-by-IVIG group received infliximab at a median of −2.5 hours before initiation of IVIG (IQR −3 to −2). Results of this 3-group subanalysis were similar to results of the 2-group analysis, although there was also a statistically significant difference in need for diuretics, which were required in 10 of 20 (50%) in IVIG-Alone, 12 of 16 (75%) in the IVIG-followed-by-infliximab group, and 11 of 36 (29%) in the infliximab-followed-by-IVIG group (P = .01) (Supplemental Table 6). This difference in diuretic requirements remained significant when directly comparing the IVIG-followed-by-infliximab group with the infliximab-followed-by-IVIG group (P = .006). Additionally, in comparing the IVIG-followed-by-infliximab group with the infliximab-followed-by-IVIG group, there were statistically significant differences in percentage change in the CRP level at 24 hours after treatment initiation compared with pretreatment, with medians (IQRs) of −16% (−60% to 26%) and −54% (−64% to −23%), respectively (P = .02). New or worsened LV dysfunction and/or new vasoactive medication requirements occurred in 3 of 16 (19%) compared with 1 of 36 (3%), respectively (P = .08).

Initial intensified therapy consisting of IVIG plus infliximab compared with IVIG alone was associated with improvement in several outcomes, including decreased need for additional therapy, shorter median ICU length of stay, decreased development of LV dysfunction, and more rapid resolution of inflammation. Our results expand on a smaller observational study describing the use of infliximab in patients with MIS-C.¹⁹  In that cohort of 29 patients, all received IVIG as first-line therapy, and of the 12 who received infliximab as second-line treatment of ongoing inflammation or cardiac dysfunction, all responded and recovered well without adverse reactions.¹⁹ 

Infliximab is a chimeric monoclonal antibody that binds tumor necrosis factor α (TNF-α) and inhibits interaction with its receptor, preventing downstream proinflammatory effects. Its use in KD has revealed improved outcomes in high-risk patients.^11–13,20  Although there appear to be differences in the cytokine profile in patients with KD compared with MIS-C; patients with MIS-C have been shown to have elevated TNF-α levels.^14,21–25  Given the complexities of the proinflammatory signaling cascade, the question remains whether the elevated TNF-α level represents cause or effect or both, but nevertheless it provides some pathophysiologic rationale for the use of anti–tumor necrosis factor therapy. Notably, our patients who received IVIG plus infliximab showed more rapid improvement in CRP levels compared with those who received IVIG alone. This suggests that a targeted anti–tumor necrosis factor approach is effective in reducing the inflammatory response represented by CRP, which is largely reflective of elevations in interleukin 6 (IL-6) levels. IL-6 levels have also been shown to be elevated in patients with MIS-C, prompting some centers to use anti-IL-6 therapies, such as tocilizumab.^14,21,24  Our results suggest infliximab may also indirectly affect IL-6-mediated inflammatory pathways.

Importantly, our study revealed improved cardiac outcomes in patients who received IVIG plus infliximab. Only 8% in IVIG+Infliximab developed new or worsened LV dysfunction and/or new vasoactive medication requirement beyond 24 hours after treatment initiation, compared with 25% in IVIG-Alone. Similarly, although vasoactive medication duration was not statistically significantly different between the groups, likely limited by small numbers, there was a clinically significant difference favoring use of infliximab. In the 3-group subanalysis comparing outcomes among patients in IVIG-Alone, the IVIG-followed-by-infliximab group, and the infliximab-followed-by-IVIG group, the composite cardiac outcome of new or worsened LV dysfunction and/or new vasoactive medication requirement developed in 5 of 20 (25%) in IVIG-Alone, 3 of 16 (19%) in the IVIG-followed-by-infliximab group, and 1 of 36 (3%) in the infliximab-followed-by-IVIG group (P = .02). When directly comparing the cardiac outcomes in the IVIG-followed-by-infliximab group versus the infliximab-followed-by-IVIG group, there was no statistically significant difference in the outcomes (P = .08). This suggests that for cardiac outcomes, it may not matter whether infliximab is given before or after IVIG; however, it is also possible that giving infliximab first may result in better outcomes that may be demonstrated in a larger study. Benefit of early infliximab administration is clearer in the comparison of the percentage change in CRP at 24 hours posttreatment initiation compared with pretreatment, with medians (IQRs) of −16% (−60% to 26%) in the IVIG-followed-by-infliximab group and −54% (−64% to −23%) in the infliximab-followed-by-IVIG group (P = .02). The more rapid drop in CRP levels in the group that received infliximab earlier suggests that the effect of infliximab may be maximized when given early, before administration of high-dose IVIG, which, in the setting of a hyperinflammatory state with presumed capillary leak and myocarditis with ventricular dysfunction, can lead to fluid overload.²⁶  This in turn may explain the significant difference seen in diuretic requirements between the IVIG-followed-by-infliximab group (12 of 16 patients; 75%) and the infliximab-followed-by-IVIG group (11 of 36 patients; 29%) (P = .006).

Other literature has supported the use of intensified initial therapy in MIS-C, specifically with IVIG plus corticosteroids. In January 2021, a French observational study revealed a lower risk of treatment failure (defined as fever 48 hours after initiating treatment), need for additional therapy, and need for vasoactive medications in patients treated initially with IVIG plus methylprednisolone compared with those who received IVIG alone.²⁷  This combination therapy also shortened ICU length of stay by 2 days.²⁷  Notably, the need for additional therapy in the French study was only 9% (3 of 34) in patients who received IVIG plus methylprednisolone, compared with 31% (16 of 52) of patients in our IVIG+Infliximab. However, the need for additional therapy in the French study’s group of patients receiving IVIG alone was also lower at 46% (33 of 72), compared with 65% (13 of 20) in IVIG-Alone in our study.²⁷  This suggests there may be other institutional practices that account for this difference, including a lower threshold for additional therapy or patient population differences. Encouragingly, the median ICU length of stay in our study was 1.8 days in IVIG+Infliximab, which is substantially shorter than the reported 4 days in the IVIG-plus-methylprednisolone group in the French study.²⁷  More recently, the Overcoming COVID-19 collaborative investigators published data from 518 US patients with MIS-C revealing improved cardiovascular outcomes in patients who received initial IVIG plus corticosteroids compared with IVIG alone, with 17% vs 31% developing LV dysfunction (risk ratio 0.56; 85% confidence interval 0.34–0.94).²⁸  In that study, the 34 patients who received a biologic as initial therapy were excluded from comparison.²⁸  Somewhat contradictory results were published simultaneously in a study of 614 patients with MIS-C from 32 countries revealing no difference in outcomes among patients treated with IVIG plus glucocorticoids versus IVIG or glucocorticoids alone.²⁹  Again, patients receiving biologic medications as initial therapy were not included in the analysis.

Potential benefits of infliximab over corticosteroids include single-dose administration without need for a taper, which can be cumbersome for families and providers and may require several weeks of therapy, allowing greater potential for side effects. With a half-life of 7 to 12 days, antiinflammatory effects of infliximab are long-lasting, and infliximab has an excellent safety profile.^30,31  Potential risks of giving both IVIG and infliximab versus IVIG alone are unknown and should be specifically explored in further studies.

Our study has several limitations. First, this is a single-center, retrospective, nonrandomized observational study, which may limit generalizability. Second, there were changes in institutional guidelines over time, and treatment was at the discretion of the clinical providers. This along with diagnostic familiarity and experience over time might have led to variability in management over the course of the study period. Third, not all patients received high-dose aspirin, and maximum dosage and administration time of IVIG varied. Fourth, patients in IVIG+Infliximab were more likely to be admitted directly to the ICU and to have LV dysfunction on the echocardiogram. However, these markers of increased severity would be expected to bias the findings against the demonstrated improved outcomes in IVIG+Infliximab. We did not perform propensity score matching because of the small sample size and relative similarity between groups. Finally, without a definitive diagnostic test, some patients who did not have MIS-C might have been included. However, only patients who met the CDC MIS-C case definition were included, and the demographic, clinical, and laboratory characteristics of our cohort were similar to those previously reported.^32,33 

Patients with MIS-C treated with intensified initial therapy of IVIG plus infliximab were less likely to require additional therapy and had decreased development of LV dysfunction, more rapid decline in inflammation, and decreased ICU length of stay compared with those treated with IVIG alone. Future randomized controlled trials are necessary to confirm these findings and to compare initial therapy with IVIG alone, IVIG plus infliximab, and IVIG plus corticosteroids. Comparative effectiveness trials, which may be more feasible, may also provide insight into optimal initial MIS-C therapy.

We thank Dr Mary Glode for critical review of the article, the Children’s Hospital Colorado Scientific Advisory Council for continued review of our institutional MIS-C clinical care guidelines, the clinical teams who expertly and compassionately cared for these patients, and Michelle Hite for research and administrative support. Seven patients included in this study were part of descriptive cohorts of patients with MIS-C in previous publications (N Engl J Med 2020;383(4):334–336; JAMA 2021;325(11):1074–1087; JAMA Neurology 2021;e210504).

CDC

Centers for Disease Control and Prevention

CRP

C-reactive protein

EF

ejection fraction

IL-6

interleukin-6

IQR

interquartile range

IVIG

intravenous immunoglobulin

IVIG-Alone

treatment group receiving intravenous immunoglobulin alone

IVIG+Infliximab

treatment group receiving intravenous immunoglobulin and infliximab

KD

Kawasaki disease

LAD

left anterior descending artery

LV

left ventricular

MIS-C

multisystem inflammatory syndrome in children

RCA

right coronary artery

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

TNF-α

tumor necrosis factor α