2. Materials and Methods
This retrospective case-control study took place at a tertiary care pediatric children’s hospital in a metropolitan area. Medical records of patients diagnosed with COVID-19 or MIS-C per the World Health Organization (WHO) criteria from March 2020 to June 2020 were reviewed.
Patients were included in the study if they met the following criteria: 1) age 0-21 years old, 2) MIS-C per WHO criteria [
6] 3) hospitalized at Children’s Memorial Hermann Hospital, 4) suspected COVID-19 infection. Patients were excluded if: 1) age >21 years, 2) the patient was not hospitalized, 3) patient did not meet criteria for MIS-C per WHO guidelines. Children presented to the ED and MIS-C were diagnosed once hospitalized and found to have laboratory findings consistent with MIS-C.
Demographic data obtained were age, sex, and race. Other metrics studied included obesity (defined as body mass index [BMI] of 30 or greater), leukopenia (defined as <1,000/ml), neutrophilia (defined as >7,700/ml), lymphopenia (defined as <1,000/ml), elevated troponin (defined as >0.4ng/ml), elevated CRP (defined as >3.0mg/L), elevated ferritin (defined as >275ng/mL), and elevated d-dimer (defined as <0.50mcg/mL). Outcomes were given numerical scores based on the Pediatric Overall Performance Category scale as seen in
Table 3 (1 = Good performance, 2 = Mild overall disability, 3 = Moderate overall disability, 4 = Severe overall disability, 5 = Coma/Vegetative state, 6 = Brain death) [
8]. We considered a scale of 2-6 as a good outcome and a scale of 1 as a good outcome.
Our study compared patients who met MIS-C criteria who tested positive for COVID on polymerase chain reaction (PCR) at least twice and those who met MIS-C criteria but tested negative on PCR. Control cases were defined as patients who met the criteria for diagnosis of MIS-C but were found to be negative on COVID-19 PCR testing. These control cases were compared to those patients diagnosed with MIS-C who also had a positive COVID result on PCR.
All patients diagnosed with MIS-C were also divided into subgroups based on their disposition from the ED. Patients admitted to the PICU were compared to patients who were not admitted to the PICU (either discharged home from the ED or admitted to the pediatric floor).
Statistics Analysis
Our first aim was to compare patients with MIS-C who were admitted to the PICU to those who were either discharged home or admitted to the pediatric floor. The two subgroups were compared regarding obesity, neutrophilia, lymphopenia, and the presence of an elevated troponin. To assess the association between the groups and obesity, Pearson’s Chi-square test was conducted. To assess the association between the groups regarding neutrophilia, lymphopenia, and elevated troponin, Fisher’s exact test was conducted.
Our second aim was to compare patients based on outcomes. We assessed the association of outcomes with the diagnoses of obesity, neutrophilia, lymphopenia, and elevated troponin. Outcomes were based on the Pediatric Overall Performance Category scale and an outcome ≥ 1 was considered poor. We conducted Fisher’s exact test on performance outcome with diagnoses of obesity, neutrophilia, lymphopenia, and elevated troponin.
Our final aim was to determine the association between the control group versus the MIS-C group concerning obesity, neutrophilia, lymphopenia, and elevated troponin. We conducted Pearson’s Chi-square test on control versus MIS-C grouping with diagnoses of obesity, neutrophilia, and lymphopenia. We conducted Fisher’s exact test on control versus MIS-C grouping with a diagnosis of elevated troponin.
3. Results
A total of 80 patients were included in the study. Of the total, 41 were diagnosed with COVID-19 with confirmatory PCR that was performed twice and met the MIS-C criteria and 39 met all MIS-C criteria but had a negative PCR for COVID-19 and were considered the control group. The MIS-C patients included five patients seen in the ED and discharged home, 12 admitted to the pediatric floor, nineteen admitted to the PICU initially, and one who was only in the PICU for all of the hospital stays.
Only six patients had an abnormal echocardiogram (19%). All of those patients were admitted to the PICU. Five patients had cardiac dysfunction and one had no cardiac dysfunction and only tricuspid valve regurgitation. For those with abnormal echocardiograms, two had a good outcome, two had a mild overall disability and two had a moderate overall disability.
In reference to treatment, all of those with cardiac dysfunction had treatment with IVIG and steroids. All but one patient with an abnormal echocardiogram received IVIG and steroids. The one patient who did not receive IVIG and steroids had only mild tricuspid valve regurgitation on an echocardiogram. For the whole group with MIS-C, only one patient required high flow nasal cannula (HFNC), five non-HFNC, and none were intubated. The one patient who was on HFNC had no echocardiogram abnormalities but was hypoxic and had pneumonia on chest x-ray.
Only one patient required pressor support and did have cardiac dysfunction. No patients in the MIS-C or control group required ventilation or extracorporeal membrane oxygenation (ECMO).
Patients with MIS-C admitted to the PICU were compared to those discharged from the ED or admitted to the pediatric floor.
Table 1a shows the comparison between the two regarding obesity, neutrophilia, lymphopenia, and elevated troponin. There were significant associations between patients admitted to the PICU and neutrophilia (p = 0.03) and between patients admitted to the PICU and elevated troponin (p = 0.01).
Table 1b demonstrates the association of outcomes of those with MIS-C with obesity, neutrophilia, lymphopenia, and elevated troponin. There were no significant associations between having a non-good outcome and obesity, neutrophilia, lymphopenia, or elevated troponin. No patients died nor had moderate to severe disability, coma/vegetative, or brain death. Of those MIS-C patients in our group, 18 (58%) had a good performance, 6 (19%) had a mild overall disability and 3 (1%) had moderate overall disability.
The vital signs as reported upon presentation to the ED were compared for those admitted to the PICU to those who were discharged from the ED or admitted to the pediatric floor. There were no significant differences among the group for temperature elevation, hypoxia, hypotension, or tachycardia. Patients who were admitted to the PICU had significantly more tachypnea than those who were not admitted to the PICU.
General demographics between patients with MIS-C with diagnosed COVID-19 were compared to the control group and are displayed in
Table 2a. No significant difference or association was found between those with MIS-C and controls about age, gender, or race. The median age for the MIS-C group was 14 years. The MIS-C group was 55% female, 45% male, 48% Hispanic, 25% other race, 16% white, and 10% Black.
MIS-C patients were compared to the control group in terms of elevated troponin, CRP, and neutrophilia as seen in
Table 2b. There were significant associations between patients with MIS-C and obesity, as well as between patients with MIS-C and elevated troponin.
4. Discussion
As more literature becomes available regarding the novel MIS-C syndrome, this study adds critical information regarding the initial presentation in the ED, and progression of the disease during hospitalization, and compares outcomes of pediatric patients based on severity. No study has been published examining the presenting clinical and laboratory findings of those with MIS-C upon presentation to the ED and how they may differ in their presentation if they required admission to the PICU. Our findings found: 1) there was a significant association between MIS-C patients admitted to the PICU and neutrophilia and elevated troponin, 2) there was no association between having a non-good outcome and obesity, neutrophilia, lymphopenia, or elevated troponin in MIS-C patients, 3) demographics between MIS-C patients and controls were not significantly different, and 4) when MIS-C and controls were compared, elevated troponin and obesity were significantly associated with MIS-C patients.
There was a significant association between those admitted to the PICU and neutrophilia and elevated troponin, but not for obesity or lymphopenia. Elevated troponin has been reported as the most abnormal MIS-C marker [
9]. Prior studies have shown an association between MIS-C and obesity, neutrophilia, lymphopenia, and elevated troponin levels [
7,
8,
10,
11]. An association between PICU admission and elevated troponins is to be expected, as the presence of elevated troponins alone would warrant ICU admission due to cardiac involvement [
12]. Neutrophilia, however, which can be seen across all severities of sepsis as well as other diseases with a systemic inflammatory response syndrome (SIRS), is characterized by a marked increase (up to ten-fold) and is significantly associated with PICU admission [
13]. Neutrophilia has also been seen in severe viral infections such as RSV [
14].
These same four variables (obesity, neutrophilia, lymphopenia, and elevated troponin levels) were assessed to see if there was any association with the outcome at the time of hospital discharge. None of these variables were associated with outcomes at discharge. Thus, it could be inferred that while neutrophilia and elevated troponin may initially be associated with the need for a higher level of care, there was no relationship with the outcome at discharge. Elevated troponin levels in individuals with COVID-19 are independent prognostic markers of poor outcomes [
15]. However, it is unknown if troponin levels correlate or can prognosticate specific abnormalities in a patient with MIS-C. Our study showed no association of troponin or other laboratory or clinical abnormalities such as obesity, neutrophilia, and lymphopenia related to MISC outcome.
Our study’s outcomes were similar to other studies finding that death is rare, with reports of death as low as 1.7% in the U.S. and 1.4% in Europe [
7,
8,
13]. We did not have any reported deaths, those requiring ventilation, or those requiring ECMO, which are markers of severity. This could possibly be explained by our patients receiving glucocorticoids and intravenous immunoglobulin (IVIG) quickly upon admission. Of those patients who received IVIG and glucocorticoids, all received treatments within 48 hours of presentation to the ED. Unlike our study, Feldstein and colleagues found their MIS-C patients received IVIG and glucocorticoids on days five to eight [
11]. The severity of MIS-C has been associated with cardiac dysfunction, intubation, pressor support, and ECMO [
10,
11]. We had only 19% of patients with cardiac involvement per echocardiogram, with previous studies reporting 50-80% [
16]. Previous reports of required intubation were up to 18%, whereas our study did not have any patients intubated. Also, pressor support has been reported to be 47% whereas it was only required in one of our patients. A study by Ahmed did report 1% requiring pressor support which is comparable to our study [
17]. In previous studies, ECMO was necessary for 4% of patients [
17].
Our study showed no difference in the incidence of MIS-C among racial groups. This is unlike prior studies which have shown a higher incidence of MIS-C amongst Black, Hispanic, and South Asian people (8, 10, 13). In our study, the highest incidence of patients with MIS-C was Hispanic, followed by other races, White, then Black. Surprisingly, there were not as many Black patients affected as in previous literature. Houston is the most diverse city in the nation; thus, it is unlikely that our sample was affected by racial bias.
There was a significant association between MIS-C patients and obesity as well as elevated troponin. This association of obesity and MIS-C is interesting since we have a high percentage of obesity in our city. Nationally, Houston is number seven in overweight and obese residents in the U.S. Obesity has been noted to be a risk factor for COVID and MIS-C. Proposed mechanisms explaining why obesity may be a risk factor include the accumulation of inflammatory cells in adipose tissue, fat tissue-associated cytokines being proinflammatory, impaired respiratory function, and adipose cells having more SARS-CoV-2 binding receptors [
18,
19,
20,
21,
22].
Most patients with MIS-C recover with intensive care and medications such as IVIG, glucocorticoids, anti-tumor necrosis factor (TNF), interleukin-1 (IL-1) or 6 inhibitors (IL-6). Of the patients in our study who were in the PICU, eleven received IVIG, ten received steroids, one received pressors, one received IL-6 inhibitor, none received IL-8 inhibitor, and none received anti-TNF. None of the patients who were admitted to the floor received IVIG, glucocorticoids, anti-TNF, IL6, or IL-8 inhibitor. All of our patients with cardiac dysfunction received IVIG and glucocorticoids. Of those without cardiac dysfunction, three patients received IVIG only one patient received glucocorticoids only, and one received both IVIG and glucocorticoids. Full recovery in MIS-C patients at discharge has been reported) with the majority of the patients with cardiac involvement having full recovery [
23,
24]. A recent study by Son and colleagues showed a lower risk of cardiovascular dysfunction after the administration of IVIG and glucocorticoids (17% vs 31%) [
17]. This was similar to our study results, with the six patients with abnormal echocardiograms who received IVIG and glucocorticoids having good recovery by the Pediatric Overall Performance Category Scale by the time they were discharged from the hospital even though they had cardiac dysfunction while in the PICU. Unlike other studies, five of our MIS-C patients were discharged home from the ED without any treatment. This can be explained by the period in which we did our study. Our study started when MIS-C was minimally apparent. These patients looked well and did not return to our ED from our chart review. Hospital follow-up communication with families also did not inform us of the patients presenting to an outside hospital with severe illness.
In reference to those who were not well appearing and were instead admitted, any patient initially admitted to the PICU was eventually transferred to the pediatric floor before discharge home. This indicates the progression of illness was from severe to less severe. Markers of inflammation were consistent with this, with CRP being elevated upon presentation to the ED. We can assume these patients were studied at the peak of illness since none of the patients with MIS-C were upgraded from the floor to the PICU. In addition, their CRP was elevated when they presented to the ED. This indicates that we saw them in their acute versus recovery phase, which may have changed their treatment and outcome.
Our findings of comparable vital signs upon presentation of those admitted to the PICU versus those discharged home from the ED or admitted to the pediatric floor are very interesting. Fever, tachycardia, tachypnea, and hypotension have been strongly associated with shock [
25]. All of the vital signs were similar among those admitted to the PICU versus those not requiring PICU, except for tachypnea. One would think the vital signs would be abnormal among those who were admitted to the PICU. We could not explain why tachypnea was the only vital sign abnormal when the groups were compared. It is possible that since it became our hospital’s practice to admit children with MIS-C with elevated troponin, their vital signs may not have been significantly different. In our study, those with an initially elevated troponin had normal vital signs upon presentation but then became abnormal during their hospital stay.
Limitations to the study include its retrospective nature where all the variables may have not been collected. Additionally, MIS-C is a rare disease, limiting our data. The incidence of long-term follow-up was not always available, including echocardiograms. Also, mild cases due to lack of recognition or lack of COVID testing were not included. Therefore, the numbers in the study may have underestimated the true number of cases of MIS-C.
In conclusion, this study was done to help distinguish patients diagnosed with MIS-C who required PICU admission and those who did not and to look at the outcomes of each group. This study adds useful information about how those with MIS-C versus those with MIS-C-like symptoms but with negative COVID testing present to the ED and progress during hospitalization.