Vascular and blood pressure alterations in children two years after multisystem inflammatory syndrome

Abstract

Multisystem inflammatory syndrome in children (MIS-C) is a severe complication of SARS-CoV-2 infection. The long-term impact on vascular and cardiac health in post-MIS-C patients remains unclear. We aimed to evaluate subclinical cardiovascular changes in children two years after MIS-C. This cross-sectional study included 42 children diagnosed with MIS-C (29 boys, 13 girls, median age 10.7 years) and 38 age- and sex-matched healthy controls. Participants underwent comprehensive cardiovascular assessment, including biomarkers of endothelial dysfunction (galectin-3, sVCAM-1, sICAM-1), arterial damage (pulse wave velocity, carotid intima-media thickness [cIMT]), echocardiography, and blood pressure measurements. Compared to controls, children post-MIS-C had higher fasting glucose levels (RE = 0.79; 95% CI 0.68–0.89; p < 0.05), higher markers of endothelial dysfunction (sVCAM-1, RE = 0.64; 95% CI 0.52–0.77; p = 0.03; galectin-3, RE = 0.74; 95% CI 0.63–0.85; p < 0.001), and increased peripheral and central blood pressure (RE from 0.64 to 0.77; all p < 0.05). Thickened cIMT (> 95th percentile) was observed in 29.3% of post-MIS-C children versus 3.1% of controls (odds ratio 12.83, 95% CI 1.57-104.95, p = 0.004). Some MIS-C severity markers (troponin, NT-proBNP, inflammatory markers) correlated with long-term cardiovascular changes. Two years after MIS-C, children are exposed to long-term biochemical changes and alterations in the circulatory system. Our findings highlight the importance of long-term follow-up and cardiovascular monitoring in post-MIS-C children.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-45878-3.

Introduction

Multisystem inflammatory syndrome in children (MIS-C) is a complication of infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), characterized by multi-organ involvement and a particular predilection for the cardiovascular system. Many affected children require admission to a pediatric intensive care unit (PICU) due to the severity of their illness¹.

However, because MIS-C is a recently recognized condition, data on its long-term cardiovascular consequences are limited. Kawasaki disease (KD), a pediatric vasculitis with overlapping clinical features, organ involvement, and treatment strategies, is known to result in persistent arterial damage, as assessed by flow-mediated dilation, pulse wave velocity (PWV), and carotid intima-media thickness (cIMT)². Whether similar long-term vascular changes occur after MIS-C remains unclear.

Central (aortic) blood pressure, which reflects the hemodynamic load on the heart, aorta, and vital organs, is more strongly associated with cardiovascular risk than peripheral blood pressure³. Its usefulness has also been revealed in pediatric patients with primary hypertension⁴ and autosomal polycystic kidney disease⁵.

Therefore, in this study, we aimed to evaluate subclinical cardiovascular changes in children two years after MIS-C and to compare these findings with those of age- and sex-matched healthy controls. Our primary outcome was the evaluation of aortic (central) systolic blood pressure (AoSBP), and secondary outcomes included peripheral blood pressure, echocardiographic and vascular parameters, and concentrations of intracellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), and galectin-3 in MIS-C children and healthy peers.

Results

Characteristics of the study group

Of the 60 post-MIS-C children initially considered, 42 patients (29 boys [69.0%], 13 girls [31.0%]) with a median age of 10.7 years (IQR 6.7) were ultimately included in the study (see flow chart in Fig. 1). The median interval between MIS-C onset and cardiovascular evaluation was 2.0 years (IQR 0.08).

Fig. 1.

Participant Flow Diagram.

Clinical and laboratory features of the study group during MIS-C are presented in Table 1. At the time of MIS-C diagnosis, seven children were overweight and one was obese. No other risk factors for severe SARS-CoV-2 infection were identified in the available medical records. Cardiovascular complications were frequent: all children had elevated N-terminal pro-B-type natriuretic peptide (NT-proBNP), and most (30 [71.43%]) had elevated troponin. Hypotension was observed in 23 (54.76%) patients and decreased ejection fraction (EF) in 26 (61.9%). Five patients (11.9%) required PICU admission. Intravenous immunoglobulin was administered to 39 (92.9%) children, and most received glucocorticosteroids: lower doses (≤ 2 mg/kg/day) were used in eight (19.05%) and high doses (≥ 10 mg/kg/day) in 23 (54.76%).

Table 1.

Laboratory and clinical characteristics of the MIS-C group during the acute phase of the illness.

Characteristics	Median (IQR)
Age (years)	8.7 (6.6)
Lymphocytes minimum [×103/µL]	0.85 (0.9)
Hemoglobin minimum [g/dL]	10.6 (1.7)
PLT minimum [×103/µL]	165.5 (115.8)
CRP maximum [mg/dL]	17.55 (19.1)
PCT maximum [ng/mL]	2.27 (6.1)
Ferritin maximum [µg/L]	318.8 (406.7)
D-dimers maximum [µg/L]	2870.4 (2898.2)
Albumin minimum [g/L]	3 (0.8)
Troponin maximum [µg/L]	59.0 (134.5)
NT-pro-BNP maximum [pg/mL]	4191 (4395.5)
SBP minimum [mmHg]	84 (19)
EF minimum [%]	50 (14.8)

Comparison of MIS-C and control group

The control group included 38 patients (26 boys [68.42%], 12 girls [31.58%]), with a median age of 10.6 years (IQR 6.9). Tables 2 and 3 compare the demographic, laboratory, and cardiovascular parameters between the MIS-C and control groups. Although the mean body mass index (BMI) did not differ between the groups, abnormal BMI values were more frequently observed in the study group. In the MIS-C group, six children were overweight and three were obese, whereas in the control group three children were overweight and none was obese. The number of available observations varied slightly between variables due to missing data, and it is indicated in the corresponding tables.

Table 2.

Demographic and laboratory features of the MIS-C and the control groups.

Median (IQR) or N (%)
		MIS-C	HC	Relative effect (95% CI)	p-value*
DEMOGRAPHIC
Age [years]		10.7 (6.7)	10.9 (6.9)	0.51 (0.38–0.64)	0.943
Sex	Female	13 (30.95%)	12 (31.58%)	1.03 (0.40–2.65)	1.000
	Male	29 (69.05%)	26 (68.42%)
BMI [kg/m2]		17.15 (6.7)	17.83 (4.64)	0.58 (0.45–0.71)	0.211
LABORATORY EVALUATION
Hemoglobin [g/dL]		13.5 (1.90)	13.0 (1.38)	0.59 (0.46–0.71)	0.157
Total cholesterol [mg/dL]		166.0 (40.25)	163.0 (35.0)	0.55 (0.42–0.68)	0.414
HDL-C [mg/dL]		55.0 (14.75)	55.0 (15.25)	0.52 (0.39–0.66)	0.732
non-HDL-C [mg/dL]		108.0 (35.75)	108.0 (25.5)	0.57 (0.44–0.71)	0.263
Triglycerides [mg/dL]		64.5 (30.5)	58.5 (25.5)	0.58 (0.45–0.71)	0.224
Glucose [mg/dL]		93.0 (7.0)	85.0 (6.0)	0.79 (0.68–0.89)	< 0.001
VASCULAR INJURY MARKERS
Galectin-3 [ng/mL]		5.60 (3.57)	3.91 (2.17)	0.74 (0.63–0.85)	< 0.001
sICAM-1 [ng/mL]		45.95 (10.00)	50.36 (11.43)	0.40 (0.27–0.53)	0.127
sVCAM-1 [ng/mL]		1012.90 (290.85)	880.93 (222.05)	0.64 (0.52–0.77)	0.030

BMI, body mass index; HC, healthy controls; HDL, high-density lipoprotein cholesterol; IQR, interquartile range; MIS-C, multisystem inflammatory syndrome in children; non-HDL-C, non-high-density lipoprotein cholesterol; sICAM-1, soluble intercellular adhesion molecule 1; sVCAM-1, soluble vascular cell adhesion molecule 1.

* Brunner–Munzel’s Test (random permutation test).

Table 3.

Cardiovascular parameters in the MIS-C and the control group.

Median (IQR)
	MIS-C	HC	Relative effect (95% CI)	p-value*
PERIPHERAL BP
SBP [mmHg]	112 (15)	106 (17)	0.64 (0.51–0.77)	0.032
DBP [mmHg]	68 (6)	62 (5)	0.77 (0.65–0.89)	< 0.001
SBP Z-score	0.40 (0.89)	−0.16 (1.31)	0.65 (0.52–0.78)	0.026
DBP Z-score	0.77 (1.20)	−0.04 (1.08)	0.73 (0.62–0.85)	0.001
VASCULAR EVALUATION
cIMT [mm]	0.42 (0.08)	0.40 (0.04)	0.62 (0.50–0.75)	0.056
cIMT Z-score	0.61 (1.37)	0.28 (0.91)	0.64 (0.51–0.78)	0.037
PWV [m/s]	4.38 (1.13)	4.30 (0.97)	0.50 (0.36–0.63)	0.975
PWV Z-score	−0.56 (1.00)	−0.61(1.77)	0.50 (0.35–0.64)	0.954
AoSBP [mmHg]	96 (11)	92 (13)	0.66 (0.53–0.78)	0.015 0.02*
AoDBP [mmHg]	69 (8)	64 (6)	0.76 (0.64–0.87)	< 0.001 0.002*
AoMAP [mmHg]	83 (6)	79 (8)	0.73 (0.61–0.84)	< 0.001 0.002*
AoPP [mmHg]	27 (11)	28 (9)	0.5 (0.37–0.63)	0.946 0.945*
AIx75HR	3.5 (17.33)	8.0 (24.4)	0.42 (0.28–0.55)	0.217
ECHOCARDIOGRAPHIC EVALUATION
EF [%]	60.0 (5.75)	59.0 (5.75)	0.56 (0.43–0.69)	0.341
LVMI [g/m2]	67.01 (15.35)	69.84 (17.3)	0.43 (0.29–0.56)	0.244
LVM Z-score for height	−0.72 (1.28)	−0.66 (1.21)	0.47 (0.34–0.61)	0.669

AoSBP, aortic systolic blood pressure; AoDBP, aortic diastolic blood pressure; AoPP, aortic pulse pressure; AIx75HR, augmentation index at heart rate 75 bpm; cIMT, common carotid artery intima-media thickness; DBP, diastolic blood pressure; EF, ejection fraction; HC, healthy controls; LVM, left ventricular mass; LVMI, left ventricular mass indexed to body surface area; MAP, mean arterial pressure; MIS-C, multisystem inflammatory syndrome in children; PWV, pulse wave velocity; SBP, systolic blood pressure.

*Nominal p-values were obtained from the Brunner–Munzel test (two-sided). qBH values were calculated using the Benjamini–Hochberg procedure (FDR control at q = 0.05) applied only within central blood pressure family and they are presented with asterisk mark.

Routine laboratory results

Patients from the MIS-C group had higher glucose concentration than healthy controls (93.0 vs. 85.0 mg/dL, p < 0.001). Moreover, eight children (19.05%) in the MIS-C group had impaired fasting glucose, compared to none in the control group.

Abnormal lipid values were observed in both groups. Twelve (28.57%) MIS-C patients and 14 (36.84%) controls had borderline total cholesterol levels (OR 0.69, 95%CI 0.27–1.75, p = 0.480), while high total cholesterol was noted in eight (19.05%) MIS-C patients and two (5.26%) controls (OR 4.24, 95%CI 0.84–21.38, p = 0.092). Borderline non-HDL-C levels were more frequent in the MIS-C group than controls (26.19% vs. 7.89%, OR 4.14, 95%CI 1.06–16.21; p = 0.041). Borderline triglyceride levels were found in seven (16.67%) MIS-C patients and four (10.53%) controls (OR 1.7, 95%CI 0.46–6.34, p = 0.524), while high triglycerides were observed in four (9.52%) MIS-C patients and three (7.89%) controls (OR 0.89, 95%CI 0.17–4.71, p = 1.000).

Vascular injury markers

Galectin-3 and sVCAM-1 levels were higher in MIS-C patients than controls (p < 0.001 and p = 0.03, respectively). No significant difference was observed between the groups in sICAM-1 levels (see Table 2; Fig. 2).

Fig. 2.

Violin plots illustrating the concentrations of galectin-3, sVCAM-1, and sICAM-1 in patients with multisystem inflammatory syndrome in children (MIS-C) and the control group. The plots depict the distribution of values, where the width represents the probability density, and the embedded box plots indicate the median and interquartile range. Statistically significant differences were observed in galectin-3 and sVCAM-1 concentrations between the two groups. *** p < 0,001; * p < 0,05.

Peripheral blood pressure

Children from the MIS-C group had higher peripheral systolic blood pressure (SBP) (p = 0.032) and diastolic BP (DBP) (p < 0.001) values, as well as SBP and DBP Z-scores (p = 0.026, and p = 0.001, respectively), compared to healthy controls (Table 3).

Central blood pressure and arterial damage

Median values of cIMT did not differ between the groups (p = 0.056). However, the median cIMT Z-score was higher in the MIS-C group (p = 0.037). Moreover, 12 patients (29.27%) from the MIS-C group had cIMT > 95th percentile compared to one patient (3.13%) in the control group (OR 12.83, 95%CI 1.57–104.95.57.95, p = 0.004).

Central BP parameters: AoSBP, aortic diastolic blood pressure (AoDBP), and aortic mean arterial pressure (AoMAP) were higher in the MIS-C group (p = 0.015, p < 0.001, and p < 0.001, respectively). We found no differences between the groups regarding PWV, PWV Z-score, and augmentation index adjusted for a heart rate of 75 per minute (Aix75HR).

Echocardiographic evaluation

No differences were observed between the groups in EF, left ventricle mass index (LVMI), or LVM Z-scores. No coronary abnormalities were detected in post-MIS-C children.

Age and sex influence on differences between the MIS-C and the control group

Age and sex influenced the difference in abnormal cIMT frequency between the groups. Boys from the MIS-C group had cIMT > 95th percentile more often than boys from the control group (OR 9.9, 95%CI 1.15–85.2, p = 0.030). Similarly, post-MIS-C children aged ≥ 10 years were more likely to have cIMT > 95th percentile (OR 23.34, 95%CI 1.27–429.8, p = 0.003) than their healthy peers. No other differences in cardiovascular injury parameters between the groups were stratified by age or sex. Moreover, age and sex did not moderate differences between the MIS-C and control groups for any of the analyzed variables.

The associations between severity markers during acute MIS-C and cardiovascular status at follow-up

Inflammatory markers

The maximum procalcitonin (PCT) concentration during MIS-C was correlated with SBP Z-score (rho= −0.443, p = 0.005) and sVCAM-1 concentration (rho = 0.35, p = 0.026) two years later. The maximum ferritin concentration during MIS-C was associated with DBP Z-score (rho = 0.35, p = 0.032), and AoDBP (rho = 0.32, p = 0.042).

Cardiac injury markers

Maximum troponin and NT-proBNP concentrations during MIS-C were correlated with LVMI two years later (rho = 0.442, p = 0.04 and rho = 0.342, p = 0.027, respectively). Moreover, maximum NT-proBNP was correlated with sVCAM-1 concentration (rho = 0.31, p = 0.045).

The minimum SBP value during acute MIS-C was associated with sVCAM-1 concentration (rho=−0.43, p = 0.005). The minimum EF during MIS-C was correlated with SBP (rho=−0.37, p = 0.019), AoSBP (rho=−0.40, p = 0.009), and AoMAP (rho=−0.40, p = 0.009). Moreover, children with decreased EF during acute MIS-C had higher LVMI and LVM Z-score two years later, compared to those with normal EF (72.01 v 61.62 g/m², RE 0.81, 95%CI 0.67–0.94, p = 0.001, and − 0.46 v −1.21, RE 0.76, 95%CI 0.61–0.92, p = 0.004, respectively).

Discussion

Two years post-MIS-C children exhibit subtle but significant cardiovascular changes

Our study revealed significant differences in cardiovascular injury parameters between children who had MIS-C two years prior and healthy controls. While some late cardiovascular complications of MIS-C have been reported, follow-up durations are heterogeneous, and control groups are frequently absent^6–8. In our study, the time span between the acute disease and follow-up examination was nearly identical for all subjects, and their results were compared to those of a healthy population.

However, it must be emphatically stressed that, given the nature of the study, it is not possible to draw definitive conclusions about cause and effect based on it; we can only point to associations.

Laboratory markers

Children in the MIS-C group had higher glucose concentration, and a substantial proportion of them had impaired fasting glucose. Although not a direct marker of vascular injury, glucose concentration may contribute to cardiovascular complications. Impaired glucose during the acute phase of MIS-C has been reported, potentially associated with the widespread use of steroids in treatment⁹. D’auria et al. found persistent insulin resistance in one third of post-MIS-C patients at 12 months follow-up¹⁰. Apart from that, no long-term abnormalities related to glucose metabolism after MIS-C have been documented.

Notably, more children in the MIS-C group were overweight or obese than in the control group. Given the positive association between BMI and fasting blood glucose in both adults and children¹¹, we cannot exclude that body composition, rather than MIS-C history, is the main driver of serum glucose concentration.

Similarly, we observed that post-MIS-C children were more likely to have borderline non-HDL-C levels. Elevated non-HDL-C is a well-known cardiovascular risk factor. Moreover, childhood non-HDL-C levels are associated with atherosclerotic disease in adulthood¹². Abnormal lipid profiles during MIS-C have been reported, likely reflecting complications associated with the systemic inflammation, similar to KD¹³. However, we found no reports on persistent lipid abnormalities following MIS-C. Some studies have described significant increases in BMI and other markers of adipose tissue accumulation after MIS-C^14,15. It is important to note that our MIS-C patients did not differ in BMI from controls.

We found that both galectin-3 and sVCAM-1 concentrations were higher in children after MIS-C. Galectin-3 directly induces endothelial dysfunction and promotes fibrotic remodeling of vascular smooth muscle cells¹⁶. In adults, elevated galectin-3 levels are associated with increased arterial stiffness in various diseases¹⁷. In children, galectin-3 is increased in several cardiovascular conditions, including heart failure, ventricular arrhythmias, and KD¹⁸. Furthermore, young adults with a history of KD have higher galectin-3 levels than healthy controls, particularly those who developed coronary artery aneurysms¹⁹. The increased galectin-3 levels observed in our patients may therefore reflect persistent vascular damage.

Increased serum concentration of sVCAM-1 is a well-established marker of endothelial activation and dysfunction, associated with numerous cardiovascular diseases in both adults and children²⁰. sVCAM-1 is elevated years after KD²¹. In our study group sVCAM-1 concentrations were significantly higher than in controls, also indicating long-term vascular wall injury.

Serum sICAM-1 concentration is also associated with increased cardiovascular risk²². Elevated sICAM-1 levels have been reported during various phases of KD²³, but data on its long-term levels are lacking. In our study, we found no significant difference in sICAM-1 concentrations between the MIS-C and control groups.

Peripheral and central blood pressure

Interestingly, post-MIS-C children had significantly higher central and peripheral BP parameters than their healthy peers. Central BP is derived from the contraction of the LV and the elasticity of the aorta and arterial tree²⁴. Elevated central BP suggests that central hemodynamics are also impaired in children who underwent MIS-C. Healthy endothelium generates various vasoactive substances, lowering the tonus of smooth muscles, thereby reducing peripheral resistance and BP²⁵. Elevated BP in our patients could be explained by chronic endothelial dysfunction, suggested by elevated galectin-3 and sVCAM-1.

Carotid intima-media thickness

Thickening of the cIMT is one of the earliest indicators of organ damage in patients with increased cardiovascular risk. It has been demonstrated in children who underwent KD². The main determinants of cIMT in children are SBP and BMI²⁶. The observed cIMT thickening in patients after MIS-C suggests the presence of structural vascular changes, which may be associated with endothelial damage and elevated BP. This finding was influenced by sex and age, with post-MIS-C cIMT thickening more commonly observed in a subgroup of boys and children aged ≥ 10 years. This could have been influenced by age-dependent increase in cIMT which is more abrupt in pre-pubertal boys²⁷. However, these results should be interpreted with caution due to very wide confidence intervals. In addition, as previously mentioned, BMI is an important determinant of cIMT values in children and adults, and in the group of MIS-C survivors, more children were overweight or obese, which may have further contributed to the difference in this parameter between the groups.

Since absolute differences in cIMT are often very small – on the order of hundredths of a millimeter – all recommendations call for comparing Z-scores, which correspond more closely to actual changes in the arteries than absolute values²⁸. This may only indicate a trend toward the significance of absolute values and a significant difference in cIMT Z-scores.

Arterial stiffness

Interestingly, we did not observe elevated arterial stiffness parameters (PWV, Aix@75HR) in post-MIS-C children. This finding may be influenced by the small sample size or the relatively short follow-up period. Since stiffness parameters primarily assess central (aortic) stiffness, the presence of carotid artery lesions (cIMT) but not aortic stiffness suggests that arterial damage post MIS-C may vary depending on vessel size. Further studies with larger cohorts are needed to clarify these findings.

The associations between severity markers during acute MIS-C and cardiovascular status after two years

We identified multiple correlations between acute MIS-C severity and cardiovascular damage parameters two years later. Inflammatory markers were positively correlated with DBP and sVCAM-1 levels. Additionally, patients with more severe systolic dysfunction during the acute phase had greater LVM, higher BP, and elevated sVCAM-1 at follow-up. These findings suggest that a more severe course of MIS-C may be associated with a greater risk of long-term cardiovascular damage. It is difficult to explain the negative correlation between systolic pressure and maximal procalcitonin concentration. It is noteworthy, however, that systolic pressure depends less on peripheral resistance and vascular damage than diastolic pressure.

Comparison with other data on long-term cardiovascular status after MIS-C

Several studies discussing long-term cardiovascular outcomes following MIS-C are available in the global literature. Truong et al. analyzed data from 1,024 children 6 months after MIS-C. Their results are very encouraging. The authors note that even in severe cases, reduced EF, or coronary artery dilation resolve within six months, and physical fitness returns to pre-illness levels in nearly all children²⁹. In another study addressing psychological and neurological outcomes within 2 years after MIS-C, somatic symptoms were more common in post-MIS-C children than in matched controls. However, these symptoms tended to improve over time³⁰. Finally, a recently published systematic review by Alvarado-Gamarra et al. revealed that cardiovascular outcomes after MIS-C improved over time, but certain subclinical cardiac abnormalities (e.g., myocardial edema and/or fibrosis) persisted up to two years³¹. The authors of all these studies concluded that MIS-C-survivors require a long-term follow-up. To the best of our knowledge, except two studies evaluating flow-mediated dilation^32,33, there are no data on arterial function and structure in children after MIS-C.

Uncertain direction of association

The lack of baseline (pre-MIS-C or early post-acute) vascular measurements limits causal inference, and it cannot be determined whether the abnormalities observed at follow-up developed after MIS-C or were already present beforehand. Certain known risk factors for MIS-C, such as obesity³⁴, are also associated with less favorable cardiovascular profiles, and therefore, a pre-existing predisposition cannot be excluded. Nevertheless, we believe the results remain clinically relevant, as they indicate that children with a history of MIS-C may require long-term cardiovascular monitoring regardless of the underlying mechanism.

Strengths and limitations

This is the first study to comprehensively evaluate cardiovascular damage in children following MIS-C. The study group was highly homogeneous. All patients received standardized treatment, and cardiovascular assessments were performed at a consistent time point. Importantly, the circulatory system was thoroughly evaluated using gold-standard methods. Results were compared with a matched control group and, where applicable, established pediatric reference standards.

However, several limitations of our study should be acknowledged. The homogeneity of the group may limit the generalizability of the results to other ethnic populations. Nevertheless, MIS-C severity in Poland has generally been milder compared with reports from Western countries³⁵, suggesting that our findings may underestimate rather than overestimate the disease’s long-term impact. Children with MIS-C were referred to our center from a wide geographic area, including both urban and rural regions, which resulted in a more heterogeneous clinical population. In contrast, the control group was intended to represent generally healthy children without a history of MIS-C, and recruitment from a primary care setting was a pragmatic and feasible approach within the study design. However, this could have resulted in a selection bias related to socioeconomic factors or health-seeking behavior. Moreover, a higher prevalence of overweight and obesity in the MIS-C group may have influenced glucose metabolism and cardiovascular outcomes. The study group size was relatively small, though it exceeded the calculated sample size for the primary endpoint (AoSBP). We are aware that analyses of other cardiovascular parameters were secondary/exploratory and may be underpowered for smaller effects; therefore, non-significant findings are interpreted cautiously with emphasis on effect sizes and 95% confidence intervals. Also, we have not analyzed pubertal status in the subjects examined. Additionally, patients were evaluated at only two time points, and it remains possible that some cardiovascular changes may resolve, or new abnormalities may emerge over longer follow-up periods.

Finally, our study lacks assessment of baseline cardiovascular, especially arterial parameters, and we cannot exclude, e.g., higher central blood pressure or thicker carotid artery wall before MIS-C episode in the studied children. Thus, the blood pressure and vascular abnormalities should be rather considered associations rather than consequences of MIS-C.

Methods

We conducted a cross-sectional study at the Paediatric Teaching Clinical Hospital in Warsaw, Poland, between September 2022 and March 2023. The study was approved by the Bioethical Committee of the Medical University of Warsaw (AKBE/219/2022). All procedures involving human participants adhered to the highest ethical standards of the aforementioned institutional research committee and were performed in accordance with the Declaration of Helsinki and its subsequent amendments regarding the treatment of human subjects. Informed consent was obtained from the guardians of participating children and, if aged ≥ 16 years, from the children themselves.

Study population – patients with MIS-C

Children diagnosed with MIS-C between October 2020 and February 2021 (during the first MIS-C wave in Poland) were included in the study group. MIS-C was diagnosed according to the World Health Organization definition³⁶:

1. Age 0–19 years,

2. Fever > 3 days,

3. At least two of the following:

• - Rash or bilateral non-purulent conjunctivitis or muco-cutaneous inflammation signs (oral, hands or feet),

• - Hypotension or shock,

• - Features of myocardial dysfunction, pericarditis, valvulitis, or coronary abnormalities (including ECHO findings or elevated Troponin/NT-proBNP),

• - Evidence of coagulopathy (by PT, PTT, elevated d-Dimers),

• - Acute gastrointestinal problems (diarrhea, vomiting, or abdominal pain).

• d.Elevated markers of inflammation such as ESR, C-reactive protein, or procalcitonin,
• e.No other obvious microbial cause of inflammation, including bacterial sepsis, staphylococcal or streptococcal shock syndromes,
• f.Evidence of COVID-19 (RT-PCR, antigen test or serology positive), or likely contact with patients with COVID-19.

The sample size was determined by the availability of eligible patients: all children hospitalized at our hospital with MIS-C were invited to participate, and some were excluded according to the exclusion criteria.

Exclusion criteria.

1. Age ≥ 18 years old at the time of the study,

2. Congenital heart defects or other chronic cardiovascular diseases preceding MIS-C diagnosis,

3. Other chronic significant diseases (defined as conditions that limit the child’s daily activities or development).

Data collection at the time of MIS-C Diagnosis

To evaluate the association between MIS-C severity and cardiovascular status two years later, we analyzed clinical and laboratory data from the acute phase of MIS-C, obtained from hospital electronic records. Laboratory variables included: C-reactive protein (CRP), PCT, D-dimer, ferritin, albumin, lymphocyte count, platelet count, troponin, and NT-proBNP. Most of them were identified as markers of MIS-C severity³⁷. We chose lymphocyte count rather than total white blood cell count because leukocytosis may be substantially influenced by corticosteroid therapy and may not reliably reflect disease severity, whereas lymphopenia has been reported as a severity marker in MIS-C³⁸. Additionally, we recorded the minimum SBP, EF, and the requirement for PICU admission. In our center, PICU admission is based on clinical condition and is generally reserved for patients with cardiovascular and/or respiratory failure, need for ventilatory support, or requirement for vasoactive/inotropic therapy.

Elevated NT-proBNP was defined as above 324 pg/mL for children younger than 4 years, 374 pg/mL for children aged 4–7 years, 163 pg/mL for children aged 7–10 years, 296 pg/mL for children aged 10–13 years, and 145 pg/mL for older children³⁹. The troponin threshold of 19 µg/L was adopted according to the instructions provided with the reagent used in our laboratory. Hypotension was defined as minimum SBP < (70 + 2×[age in years]) mmHg in children up to 10 years old, or < 90 mmHg in older patients⁴⁰. Decreased EF was defined as < 55%⁴¹.

Control group

Healthy control children were recruited during routine visits in a single general practitioner’s office, including vaccination appointments and visits for mild self-limiting infections. All participants were clinically well at the time of study assessment. The exclusion criteria mirrored those of the study group.

Laboratory evaluation

All assessments were obtained during a single study visit performed approximately two years after the MIS-C episode. Blood samples were obtained from all participants after an overnight fast (≥ 8 h). The laboratory assessments encompassed hemoglobin concentration, lipid profile, and glucose concentration. These parameters were selected as part of the long-term cardiovascular assessment, as they are recognized markers associated with early cardiovascular risk and subclinical cardiovascular injury in pediatric populations^42,43. Troponin was measured using the Abbott ARCHITECT analyzer with the STAT Troponin-I reagent (catalog number 2K41, IVD).

Impaired fasting glucose was defined as a level of 100–125 mg/dL. High total cholesterol was defined as ≥ 200 mg/dL, while borderline was 170–199 mg/dL. Non-HDL cholesterol (non-HDL-C) was defined as elevated if ≥ 145 mg/dL, and borderline if 120–144 mg/dL. Triglycerides in children < 10 years old were defined as high if ≥ 100 mg/dL and borderline if 75–99 mg/dL. In children aged 10–18 years, triglycerides were defined as high if ≥ 130 mg/dL and borderline if 90–129 mg/dL. HDL-C was defined as low if < 40 mg/dL, and borderline if 40–45 mg/dL⁴⁴.

Vascular injury biochemical markers

We measured serum concentrations of galectin-3, sVCAM-1, and sICAM-1. Measurements were performed using the Enzyme-Linked Immunosorbent Assay (ELISA) with standardized kits provided by Invitrogen (Thermo Fisher Scientific). All tests were conducted following the manufacturer’s instructions. Absorbance readings were obtained using a UVM340 plate reader (ASYS, Biogenet), and the results were analysed using MikroWin2000 v4 software (Mikrotek Laborsysteme GmbH, Biogenet, Overath, Germany).

Arterial damage assessment

After at least a 3-minute resting period, peripheral blood pressure was measured using the validated oscillometric device OMRON HBP-1320 (OMRON, Kyoto, Japan)⁴⁵. BP was measured three times on the right arm using appropriately sized cuffs, and the mean of the last two measurements was reported. The obtained values were expressed in mmHg and Z-scores^46,47.

Peripheral pressure waveforms were recorded from the radial artery at the right wrist using applanation tonometry. After acquiring 20 sequential waveforms, a validated generalized transfer function was applied to derive the corresponding central aortic pressure waveform. The following parameters were analyzed: AoSBP, AoDBP, and AoMAP [mmHg], and the AIx75HR [%]. Only high-quality recordings (in-device quality index > 80%) were included. Aortic PWV was calculated as the difference in carotid-to-femoral path length divided by the difference in R wave to the foot of the pressure wave from superimposed electrocardiographic and pressure tracings. Path length was measured over the body surface using an external tape measure: distance from the right carotid sampling site to the manubrium subtracted from the distance from the manubrium to the right femoral site. All measurements were performed in the sitting position in a quiet, temperature-controlled room (20 ± 5 °C) after a 5-min rest. Assessments were conducted in all patients by two experienced investigators three times, and the mean value was obtained using a SphygmoCor device (AtCor Medical Pty Ltd., Sydney, Australia).

We measured cIMT [mm] using the Aloka Prosound Alpha 6 (Hitachi Aloka Medical, Mitaka, Japan) equipped with a standard 13 MHz linear transducer. cIMT was defined as the mean distance between the leading edge of the lumen–intima interface and the leading edge of the media–adventitia interface at the far wall of the common carotid artery, measured approximately 1 cm proximal to the carotid bulb. Six measurements were obtained in total (three on each side), and the average value was used for analysis, in accordance with previously published recommendations⁴⁸.

PWV and cIMT values were expressed as absolute values and Z-scores, using pediatric normative data, whenever possible to assess^28,49.

Echocardiographic assessment

Echocardiography was performed using a Philips EPIQ 9 ultrasound system (software version 9.0.1) with X5-1, S5-2, and S8-3 transducers. Cardiac morphology was evaluated according to the sequential segmental analysis approach commonly used in the pediatric population. The assessment included left ventricular (LV) M-mode parameters (LV systolic and diastolic diameters, LV wall thickness, LVM, and LVMI) as well as LVEF calculated using the Simpson method. LVMI (g/m²) was indexed to body surface area. Coronary artery anatomy was also assessed. Measurements were expressed as absolute values and Z-scores. Z-scores were calculated using the “Cardio Z” application in accordance with the recommendations of the Association for European Paediatric and Congenital Cardiology⁵⁰.

Statistical analysis

The results were expressed as medians with interquartile ranges (IQR) for continuous variables and frequencies with percentages for categorical variables.

An a priori sample size estimation for the prespecified primary outcome, central (aortic) systolic blood pressure (AoSBP) was calculated, using published pediatric data (AoSBP 98.6 ± 9.7 vs. 90.5 ± 6.9 mmHg; Δ = 8.2 mmHg; pooled SD = 8.4 mmHg). Assuming a two-sided α = 0.05 and 90% power, the required sample size was 22 participants per group (44 total) for a two-sample comparison of means.

Comparisons between the MIS-C and control groups were performed using the Brunner–Munzel test, a non-parametric alternative to the Wilcoxon-Mann-Whitney test, to assess differences in continuous variables. This approach was chosen due to its robustness in handling unequal variances and sample sizes. 95% confidence intervals (95%CI) were calculated to evaluate the relative effect (RE) sizes. Categorical variables were analyzed using Fisher’s exact test.

To address multiplicity, we applied false discovery rate (FDR) control (Benjamini–Hochberg) within the prespecified central blood pressure family (AoSBP, AoDBP, AoMAP, AoPP), as these parameters are physiologically related and closely linked to the primary endpoint (AoSBP). All remaining analyses (biomarkers, vascular indices, echocardiographic parameters, subgroup analyses and correlations) were considered secondary/exploratory and were therefore not subjected to global FDR correction; corresponding p-values are reported as nominal and interpreted cautiously, with emphasis on effect sizes and 95% confidence intervals.

Correlations between cardiovascular parameters and inflammatory markers from the acute phase of MIS-C were analyzed using Spearman’s rank correlation coefficient (rho). This method was selected because it captures monotonic relationships without assuming normal distribution.

Sensitivity analyses were performed using robust ANOVA, employing the bootstrap method with a median estimator and Mahalanobis distances, to assess the impact of potential confounders such as age, sex, and clinical severity markers on cardiovascular parameters at follow-up⁵¹. Effect modification was evaluated by including product (interaction) terms between study group and each candidate modifier (e.g., group×sex, group×age category) in factorial robust ANOVA/ANCOVA models; statistical inference was obtained using bootstrap-based robust tests⁵².

Measurement bias was minimized by using standardized protocols and validated devices for all cardiovascular assessments.

All statistical analyses were conducted using Jamovi software (version 2.3), with a predetermined significance threshold set at α = 0.05.

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, or reporting of this study. Individual results were communicated to all participants, and they will be invited for a follow-up visit five years after MIS-C.

Conclusions

Our study has shown that patients who have had MIS-C are at risk of long-term biochemical changes (such as high blood glucose and dyslipidemia) and alterations in the circulatory system, including elevated peripheral and central BP, thickening of the common carotid artery wall, and markers of vascular endothelial injury. It remains unclear whether the abnormalities identified in our study are a direct consequence of MIS-C or reflect pre-existing conditions that may have predisposed children to develop MIS-C and the design of our study precludes drawing final conclusions of the causative relationships. Nevertheless, our findings suggest that these patients require regular assessments of BP, blood glucose, and lipid profile. The reversibility of the changes we observed, and their clinical implications remain open questions.

.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, W.W., P.S., M.P., and M.O-N.; data curation, P.S., M.S., C.N., B.K., A.S-E., A.Z., and K.T.; formal analysis, M.P.; funding acquisition, E.K.; investigation, W.W., P.S., M.S., M.P., K.T., and M.O-N.; methodology, W.W., P.S., A.S-E., M.P., and M.O-N.; project administration, W.W., M.S., and A.Z.; supervision, E.K.; writing – original draft, W.W., P.S., C.N., A.S-E., M.P., and M.O-N.; writing – review & editing, W.W., P.S., M.S., E.K., and M.O-N.

Funding

This study was funded by the Medical University of Warsaw. The funding covered only the costs of reagents and equipment necessary for the assessment of vascular injury markers. The funding institution had no influence on the study design, data collection, analysis, interpretation, or the decision to publish the results.

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.