Reduced aerobic capacity in children with multisystem inflammatory syndrome in children (MIS-C) after PICU admission: a retrospective cohort study, 2020–2022

Abstract

Objective

This study aimed to evaluate aerobic capacity in children and adolescents following a diagnosis of Multisystem Inflammatory Syndrome in Children (MIS-C) compared to healthy matched controls.

Design

Prospective cohort study.

Setting

Quaternary Pediatric intensive care unit and Pediatric cardiology unit.

Participants

14 children diagnosed with MIS-C.

Interventions

None.

Measurements

Cardiopulmonary fitness parameters at the time of post-Pediatric intensive care unit follow-up (mean 3.6 months) such as maximum oxygen uptake (VO_2max) and the first ventilatory anaerobic threshold (VAT), as a marker of muscular deconditioning.

Main Results

A total of 14 patients (12 boys) with a confirmed diagnosis of MIS-C and 70 healthy children were included. The median age was 13.2 years (range 10.7-13.6 years). All participants had a normal echocardiogram and normal lung function at the time of cardiopulmonary exercise test. As measured by VO_2max Z-score, exercise capacity was significantly lower in the MIS-C group compared to healthy controls (median-0.91 vs 0.13, p < 0.01), and a significantly higher proportion of children in the MIS-C group had impaired aerobic capacity (VO_2max Z-score<-1.64) compared to controls (29% vs. 3%, p < 0.01). The VAT was also significantly lower in the MIS-C group with a higher proportion of children with an impaired VAT (VAT Z-score<-1.64) compared to controls (43% vs. 3%, p = 0.03). Impaired aerobic capacity in the MIS-C group was associated with higher BMI, higher PELOD 2 score and lower platelet count at the PICU admission, and lower hemoglobin level at the cardiopulmonary exercise test time.

Conclusions

This study suggests that children with MIS-C experience significant reductions in aerobic capacity compared to healthy controls, primarily due to muscular deconditioning. These findings highlight the importance of considering post-ICU consultations and implementing strategies to address physical deconditioning in this population.

Strengths and limitations of this study

Strengths

• This study aimed to evaluate aerobic capacity in children and adolescents following a diagnosis of MIS-C compared to healthy matched controls.

• Assessing the impact of a disease by patient-related outcomes was encouraged.

• Higher BMI, higher PELOD 2 score, higher PICU length of stay at the PICU admission, and lower haemoglobin level during CPET were associated with impaired aerobic capacity in the MIS-C group.

• Findings highlight the importance of considering post-ICU consultations and implementing strategies to address physical deconditioning in this population.

Limitations

• The sample size was modest.

• The control group consisted of children who were not hospitalized in the PICU, making it uncertain whether the observed findings are attributable to MIS-C, PICU admission, or both.

Introduction

The emergence of the SARS-CoV-2 virus and the subsequent COVID-19 pandemic significantly impacted children’s global health and well-being. Multisystem Inflammatory Syndrome in Children (MIS-C) is a rare but severe post-infectious complication associated with COVID-19 infection [1].

The pathophysiology of MIS-C remains under investigation but is thought to involve a cytokine storm and a hyperactive immune response that can affect multiple organs. Several definitions of MIS-C exist, with criteria established by the World Health Organization (WHO), and the Centres for Disease Control and Prevention (CDC) [2, 3]. Common elements include persistent fever, laboratory evidence of inflammation, evidence of multisystem organ involvement, and a plausible or proven prior SARS-CoV-2 infection [4–6]. Cardiac injury is one of the most frequent features of MIS-C with several manifestations such as myocarditis, pericarditis, myocardial dysfunction, and/or coronary artery abnormalities [7–10]. Clinically, MIS-C can overlap with Kawasaki disease (KD) and Toxic Shock Syndrome (TSS), leading to classification into three categories: shock-like, KD-like, and undefined inflammatory presentation [6, 7]. The syndrome predominantly affects males with a mean age around 9 years and carries a low mortality rate of approximately 1–2% [6, 11]. The severity of the initial presentation often necessitates hospitalization in the paediatric intensive care unit (PICU) for treatment with medications that have demonstrated efficacy, including intravenous immunoglobulin (IVIG), corticosteroids, or biological agents such as tocilizumab, infliximab or anakinra [6, 7].

Monitoring these patients involves multidisciplinary assessments, frequently focusing on cardiac function, for a period ranging from several weeks to a year following their discharge from the hospital [7]. Despite the initial severity, MIS-C typically shows good recovery, including cardiac outcomes. Several studies using electrocardiography (ECG), magnetic cardiac resonance imaging (MRI), and echocardiography have reported complete resolution of cardiac injury caused by MIS-C, even within two months of the acute illness [12–18]. The cardiopulmonary exercise test (CPET) assesses aerobic capacity and provides the most comprehensive physical capacity evaluation in children and adolescents with chronic conditions. It offers insight into cardiac, respiratory, and peripheral exercise limitations [19]. Notably, maximum oxygen uptake (VO_2peak) measured by CPET predicts prognosis and health-related quality of life (HRQoL) in various paediatric diseases [20–24]. However, research on aerobic capacity in children and adolescents post-MIS-C remains limited [25], partly due to recommendations advising against physical activity for at least three months following MIS-C diagnosis in children with cardiac involvement [7].

This study aimed to evaluate aerobic capacity in children and adolescents following a diagnosis of MIS-C compared to healthy matched controls. We further explored potential factors associated with impaired aerobic capacity within the MIS-C group. We hypothesized that children with MIS-C would demonstrate decreased aerobic capacity primarily due to muscular deconditioning resulting from the initial illness and subsequent restrictions on physical activity.

Materials and methods

Study design and population

This is a retrospective study of prospectively collected data. The MIS-C cohort was enrolled between September 2020 and March 2022 at the Geneva Children’s Hospital, a quaternary care academic institution located in Switzerland. Participants included children and adolescents aged 8 to 17 diagnosed with MIS-C, as defined by the WHO. MIS-C diagnosis criteria involved fever, at least two organ-specific injuries (dermatologic, gastrointestinal, cardiovascular, respiratory, neurologic, renal), and elevated inflammatory markers. Additionally, participants had confirmed SARS-CoV-2 infection or had contact with a COVID-19 case and lacked an alternative plausible diagnosis. CPET was proposed to all children with previous MIS-C diagnosis during routine cardiological follow-up visits 3 months after discharge from hospital. These 3 months corresponded to the period during which physical activity was restricted. Only subjects with CPET data were included in the study. Healthy controls matched for age, sex, and Body Mass Index (BMI) were sourced from Montpellier’s paediatric CPET laboratory database and enrolled between 2010 and 2020 [26, 27]. Control participants were children referred for non-severe functional symptoms related to exercise or for a medical sports certificate, following a comprehensive normal check-up including physical examination, electrocardiogram, echocardiography, and spirometry.

Study variables

For MIS-C participants, data collected from electronic health records included the need for PICU hospitalization and PICU length of stay (LOS), the Paediatric Logistic Organ Dysfunction 2 (PELOD 2) score (validated score system used to assess the severity of organ dysfunction in paediatric patients) at admission [28], a requirement for vasoactive infusion, cardiac involvement indicators (minimal Left Ventricular Ejection Fraction (LVEF), coronary damage, maximal ultra-sensitive (US) troponin T and pro-BNP levels), MIS-C classification (Kawasaki-like, shock-like, or undefined inflammatory presentation), and treatment details (IVIG infusion, corticosteroid administration). Biological data at admission (CRP, uraemia and creatinine levels, platelets and lymphocyte counts) were also recorded. Participants were categorized into cohorts based on the timing or suspected timing of COVID-19 infection, aligned with CDC data on variant prevalence (Alpha, Delta, Omicron) [9]. Clinical data including sex, age, weight, height, BMI, echocardiographic parameters, and haemoglobin level were collected at the time of CPET. Normal echocardiography was defined as the presence of normal left and right ventricles structure, size, and function; as well as the absence of coronary anomalies. For control participants, only anthropometric and CPET data were collected.

Cardiopulmonary exercise test procedure

CPET was performed on a paediatric cycle ergometer, and oxygen consumption was measured by direct gas analysis (MetaLyzer^® 3B, Cortex, Leipzig, Germany). MetaLyzer^® 3B is a breath-by-breath device. Data were analysed using ABRM on the manufacturer software (MetaSoft^®, Cortex, Leipzig, Germany). The paediatric cycle ergometer used was the ERGOSELECT 100P (Ergoline, Germany). The spirometry was performed with the MetaLyzer 3B and analysed by the MetaSoft software. The spirometry was considered as interpretable if fulfilling the (ATS/ERS) criteria [29].

The following lung function parameters at rest were collected: forced expiratory volume in one second (FEV1), forced vital capacity (FVC), and FEV1/FVC. Measurements were normalized using paediatric Z-score reference values [30]. We used a CPET paediatric cycle ergometer standardized protocol adapted to the patient’s age, with a homogeneous incremental overall duration between 10 and 12 min: a 1-min rest; a 3-min warm-up (10–20 watts) in increments of 10, 15, or 20 watts each minute; a pedalling rate of 60–80 revolutions per minute; a 3-min active recovery (20 watts); and a 2-min rest [31]. The CPET was considered maximal if at least 3 of the following criteria were met: maximum heart rate ≥ 80% of predicted, 8/10 on the Borg scale; reaching a VO₂ plateau; respiratory exchange ratio (RER) >1.05; and inability to maintain a pedalling rate above 60 despite verbal encouragement [19].

The main cardiopulmonary fitness parameter used in this study was the VO_2peak. When the VO_2max did not reach a plateau, the peak VO₂ (VO_2peak) was collected, as usual in paediatrics [32]. VO_2peak was expressed in raw values and Z-score paediatric reference values. These reference values are derived from a cohort of 1,241 patients, including those who are overweight or obese, making them the most reliable available. Impaired aerobic capacity was defined as a VO_2peak Z-score<−1.64, based on the 5th percentile values to distinguish abnormal from normal performance [26]. We also collected other maximal parameters (heart rate, workload, respiratory rate, breathing reserve, respiratory exchange ratio, VE/VCO₂ slope, OUES) which were expressed in raw values and Z-score paediatric reference values [27]. Particular attention was paid to the first ventilatory anaerobic threshold (VAT), e.g. the point at which minute ventilation increases disproportionally relative to VO₂, which was calculated using the V-slope method (which corresponds to an increase in the respiratory equivalent VE/VO₂ compared with VE/VCO₂) [33]. This parameter is used to explore muscular deconditioning and serves as a benchmark for rehabilitation programs [34].

Formal aspects

The study adhered to Good Clinical Practices and Declaration of Helsinki principles. Approval was obtained from the Cantonal Ethics Committee (CCER 2020 − 00414) for inclusion of the MIS-C group, with written informed consent obtained from participants or their guardians. The Montpellier’s CPET database was approved by the Institutional Review Board of Montpellier University Hospital (2019_IRB-MTP_10–20), with informed consent obtained from all parents or legal guardians.

Statistical analysis

Continuous variables were presented as medians with the interquartile ranges (IQR) and binary variables were described as frequencies and percentages. Continuous variables were compared using Mann-Whitney test and binary variables were compared using Fisher’s exact test. The absolute differences and their 95% confidence interval were calculated. The Hodge-Lehman method was used for estimated median difference. For MIS-C group, age, gender and BMI matched control comparisons were performed. For each child with MIS-C, five control subjects were identified with the same sex, age (within one year of the MIS-C case), and BMI (within 1 kg/m2 of the MIS-C case’s BMI). This matching strategy aimed to control for potential confounding factors that could influence the results, such as sex-based differences in physical capacity or the impact of body composition on exercise performance. The case with their matched controls was considered as a cluster. Therefore, to compare CPET parameters between the two groups, mixed models were performed in which the clusters were introduced as a random effect. Statistical significance was set at 0.05 and analyses were performed using software SAS Enterprise Guide, version 7.13 (SAS Institute, Cary, NC, USA).

Results

A total of 14 patients (12 boys) with a confirmed diagnosis of MIS-C were included in the study. All 14 patients underwent CPET testing during their follow-up visits. The median age was 13.2 years (range 10.7–13.6 years). Indices of maturation were assessed as being in line with chronological ages during clinical evaluation. The median BMI percentile was 50.8% (range 38.9–82.4%) and 3 patients were classified as obese (BMI ≥ 95th percentile). A control group of 70 healthy children was matched by sex, age, and BMI to the MIS-C group. Anthropometric data were similar between the two groups (Table 1).

Table 1.

Baseline characteristics

Variables	MIS-C group (N = 14)	Age-sex-BMI matched healthy controls (N = 70)	P-value
Sex ratio (male/female)	6	6	1
Age (years)	13.2 [10.7; 13.6]	13.2 [10.2; 14.0]	0.85
Weight (kg)	51.1 [37.2; 65.8]	45.0 [33.0; 62.0]	0.43
Height (cm)	159.0 [147.3; 176.0]	153.0 [139.0; 170.0]	0.28
BMI (kg/m2)	18.9 [17.0; 21.2]	19.0 [16.9; 21.3]	0.95
BMI (percentiles)	50.8 [38.9; 82.4]	50.1 [38.6; 78.4]	0.91

Legend: BMIBody mass index

Values are numbers with (%) and medians with [IQR]

Clinical characteristics of the MIS-C group

MIS-C group characteristics are described in Table 2. Twelve (86%) children required PICU hospitalization with a median PICU LOS of 3.0 days (range 2.0–5.3.0.3 days). Half of the patients (7 children) needed vasopressor administration. All patients receiving vasopressors presented with a shock-like form of MIS-C, while the remaining patients had a Kawasaki-like clinical presentation. Cardiac involvement was observed in 7 children of the MIS-C group, with an LVEF below 55%. One child had coronary artery dilatation. Median levels of US-troponin T and NT-proBNP were elevated (150 ng/L and 3392 ng/L respectively), indicating potential cardiac injury. The median PELOD 2 score at admission was 2.0 (range 1.0–5.5.0.5). All cases presented with elevated inflammatory markers, as evidenced by C-reactive protein (CRP) levels with a median at 206 mg/L (ranges 170; 257 mg/L). Thrombopenia (platelet count < 150 G/L) was observed in 10 children in the MIS-C group. The median hospital LOS was 6.0 days (ranges 7.0–8.8.0.8 days).

Table 2.

MISC group characteristics

Variables	MIS-C group (N = 14)
Hospitalization
PELOD 2 score		2.0 [1.0; 5.5]
PICU hospitalization (Yes)		12 (86)
	PICU length of stay (days)	3.0 [2.0; 5.3]
	Vasoactive infusion (yes)	7 (50)
Hospital length of stay (days)		6.0 [7.0; 8.8]
MISC characteristics
Vaccinated for COVID		0 (0)
Organ dysfunction	Pulmonary	3 (21)
	Digestive	8 (57)
	Kidney	2 (14)
	Neurologic	2 (14)
MIS-C classification	Kawasaki-like	7 (50)
	Shock-like	7 (50)
COVID-19 variants	Alpha	10 (72)
	Delta	2 (14)
	Omicron	2 (14)
COVID testing	Serology	12 (86)
	PCR	2 (14)
Initial cardiac evaluation
Minimal LVEF (%)		55 [49; 62]
Maximal NT-proBNP (ng/L)		3392 [1814; 5933]
Maximal US-Troponine T (ng/L)		150 [113; 199]
Biological characteristics
Maximal CRP maximal (mg/L)		206 [170; 257]
Maximal Uraemia (mmol/L)		6.2 [4.0; 8.1]
Maximal Creatinine (µmol/L)		62.5 [48.8; 85.8]
Minimal Lymphocytes (G/L)		0.48 [0.26; 0.59]
Minimal Platelets (G/L)		113 [77; 148]
Treatments
Intravenous immunoglobulins (Yes)		11 (79)
Corticosteroid (Yes)		11 (79)
CPET conditions
Delay MISC-CPET (months)		3.6 [2.8; 5.6]
Haemoglobin (g/dL)		13.0 [12.1; 14.9]
LVEF at CPET (%)		59 [57; 63]

Legend: CPETCardiopulmonary exercise test, LVEF Left Ventricular Ejection Fraction, PELOD 2 Paediatric Logistic Organ Dysfunction 2 score, PICU Paediatric intensive care unit, US Ultra-sensitive

Values are numbers with (%) or medians with [IQR]

Treatment for MIS-C included intravenous immunoglobulin (IVIG) for 11 patients, with one patient receiving three courses because of the presence of coronary artery involvement. Additionally, 11 patients received corticosteroid therapy. No child had been vaccinated for COVID.

CPET results

All participants in both groups were able to complete the CPET protocol without complications. All CPET was considered as maximal. All participants had a normal echocardiography and electrocardiogram at the time of CPET. Lung function testing at rest (FEV₁, FVC) showed no significant differences between the MIS-C and control groups. No electrical abnormalities were detected on the stress electrocardiogram during CPET. As measured by VO_2peak Z-score, exercise capacity was significantly lower in the MIS-C group compared to healthy controls (median − 0.91 vs. 0.13, p < 0.01). A significantly higher proportion of children in the MIS-C group had impaired aerobic capacity (VO_2peak Z-score < −1.64) compared to controls (29% vs. 3%, p < 0.01) (Table 3).

Table 3.

Main spirometry and CPET data: comparison between MIS-C and matched healthy control groups

Variables	MIS-C group (N = 14)	Healthy controls (N = 70)	Absolute difference [CI 95%]	P-value *
Pulmonary function: at rest
FEV1 (L)	2.73 [1.92; 3.68]	2.47 [1.87; 3.55]	0.1 [−0.5;0.8]	0.48
FEV1 (Z-score)	−0.66 [−1.37; 0.63]	−0.31 [−0.84; 0.47]	−0.3 [−1.1;0.5]	0.50
FVC (L)	3.13 [2.10; 4.03]	2.96 [2.26; 4.06]	0.1 [−0.7;0.8]	0.58
FVC (Z-score)	−0.93 [−1.99; 0.18]	−0.29 [−1.04; 0.81]	−0.7 [−1.5;0.3]	0.11
FEV1/FVC (%)	93.3 [90.5; 96.5]	86.3 [81.9; 91.8]	0.1 [0.0;0.1]	< 0.01
FEV1/FVC (Z-score)	0.93 [0.33; 1.69]	−0.04 [−0.83; 0.96]	0.9 [0.1;1.6]	0.02
CPET: performance parameters
VO2peak (mL/kg/min)	37.5 [30.0; 44.0]	43.5 [36.6; 48.9]	−5.9 [−11.9;−0.5]	< 0.01
VO2peak (Z-score)	−0.91 [−1.89; −0.34]	0.13 [−0.34; 0.74]	[−1.9;−0.6]	< 0.01
VO2peak Z-score < −1.64	4 (29)	2 (3)	25.7 [1.7; 49.7]	< 0.01
Peak heart rate (/min)	184 [171; 195]	190 [186; 193]	−5.0 [−13.0;2.0]	0.01
Peak heart rate/Predicted Peak heart rate (%)	94.4 [89.3; 98.9]	96.7 [94.4; 98.9]	[−7.2;1.1]	0.02
RERpeak	1.09 [1.06; 1.14]	1.17 [1.11; 1.23]	−0.1 [−0.1;−0.0]	< 0.01
Workloadpeak (Watts)	115 [87; 148]	155 [125; 220]	−42.5 [−75.0;−13.0]	< 0.01
Workloadpeak (Z-score)	−1.69 [−2.22; −0.85]	0.32 [−0.17; 0.73]	−1.9 [−2.4;−1.3]	< 0.01
VAT (mL/kg/min)	21.5 [18.0; 27.0]	30.0 [25.1; 34.3]	−8.0 [−11.8; −4.1]	< 0.01
VAT (Z-score)	−1.47 [−2.90; −0.73]	0.08 [−0.40; 0.48]	−1.81 [−2.66; −1.10]	< 0.01
VAT Z-score < −1.64	6 (43)	2 (3)	40 [14; 66]	< 0.01
CPET: ventilatory parameters
Respiratory ratepeak (/min)	40 [35; 47]	52 [46; 59]	−10.0 [−16.0;−3.0]	0.02
Breathing reserve (%)	39 [22; 45]	20 [10; 30]	15 [7; 25]	< 0.01
VE/VCO2 slope	30.5 [28.5; 34.6]	30.0 [28.1; 32.0]	0.70 [−1;72; 3.70]	0.26
VE/VCO2 slope (Z-score)	0.39 [−0.41; 1.03]	0.13 [−0.42; 0.75]	0.18 [−0.40; 0.90]	0.38
OUES (Z-score)	−0.49 [−1.10; 0.00]	−0.20 [−0.65; 0.67]	−0.61 [−1.20; −0;00]	0.03

Legend: FEV1 Forced Expiratory Volume in 1 s, FVC Forced Vital Capacity, FEV1/FVC Tiffeneau index, OUES Oxygen Uptake Efficiency Slope, RER Respiratory exchange ratio, VAT Ventilatory anaerobic threshold, V_E/VCO2 Minute ventilation/carbon dioxide production, VO₂ Oxygen uptake

Values are numbers with (%) and medians with [IQR]. The absolute difference was the median difference except for VO2peak Z-score and VAT Z-score<−1.64 where it was a percentage difference

*Comparison between MIS-C group and age, gender and BMI matched healthy controls (random cluster)

Most CPET parameters, including workload_peak, peak heart rate, RER_peak, peak respiratory rate, and OUES, were significantly lower in the MIS-C group compared to controls. Notably, the breathing reserve was higher in the MIS-C group. VE/VCO₂ slope did not differ significantly between the groups.

The ventilatory anaerobic threshold (VAT) was also significantly lower in the MIS-C group compared to controls (p < 0.01). A higher proportion of children in the MIS-C group had an impaired VAT (VAT Z-score < −1.64) compared to controls (43% vs. 3%, p < 0.01).

Impaired aerobic capacity (defined by VO_2peak Z-score < −1.64) in the MIS-C group was associated with higher BMI, higher PELOD 2 score, and higher PICU length of stay at the PICU admission, and lower haemoglobin level at the CPET time (Table 4).

Table 4.

Univariate analysis in MIS-C group according to VO_2max impairment

Variables	Impaired VO2peak (N = 4)	Normal VO2peak (N = 10)	Absolute difference [CI 95%]	P-value
BMI (percentiles)	89.1 [76.9; 95.8]	42.4 [30.1; 52.2]	44.1 [2.2; 67.2]	0.02
PELOD 2 score	6.0 [5.0; 6.5]	1.0 [1.0; 2.0]	4.5 [−6.0; −2.0]	< 0.01
PICU length of stay (days)	6.5 [4.0; 9.5]	2.0 [1.0; 3.0]	4.5 [−9.0; −0.1]	0.01
Vasoactive infusion (yes)	4 (100)	3 (30)	70.0 [9.7; 95.1]	0.07
Initial cardiac damage
Minimal LVEF (%)	52 [44; 58]	55 [50; 62]	−0.1 [−0.2; 0.1]	0.38
Maximal NT-proBNP (ng/l)	1658 [771; 13969]	4795 [2722; 5933]	−2024 [−5893; 21329]	0.33
Maximal US-Troponine T (ng/L)	98 [22; 318]	159 [118; 199]	−85 [−259; 101]	0.37
Biological characteristics
Maximal CRP (mg/L)	213 [190; 293]	206 [153; 264]	37 [−74; 135]	0.31
Minimal Platelets (G/L)	71 [55; 100]	123 [104; 158]	−52x [−129; 15]	0.10
Treatments
Intravenous immunoglobulins (yes)	3 (75)	8 (80)	−5 [−61; 53]	1.00
Corticosteroid (yes)	3 (75)	8 (80)	−5 [−61; 53]	1.00
CPET parameters
Haemoglobin (g/dL)	11.3 [10.8; 12.3]	14.3 [12.9; 15.4]	−2.6 [−4.7; −0.1]	0.03
Workloadmax (Z-score)	−2.15 [−3.61; −1.88]	−1.40 [−1.73; 0.76]	−1.1 [−3.5; 0.0]	0.05
VAT (Z-score)	−3.03 [−3.73; −2.04]	−1.27 [−2.60; −0.35]	−1.4 [−3.3; 0.1]	0.05

Legend: BMI Body mass index, CI Confidence interval, CPET Cardiopulmonary exercise test, LVEF Left Ventricular Ejection Fraction, PICU Paediatric intensive care unit, US Ultra-sensitive, VAT Ventilatory anaerobic threshold

Values are numbers with (%) or medians with [IQR]. The absolute difference was the median difference except for vasoactive infusion, which was a percentage difference

Discussion

This cohort study demonstrated that children with a history of MIS-C exhibited significantly lower aerobic capacity compared to healthy controls, despite the absence of detectable cardiac or respiratory dysfunction at the time of evaluation. As previous studies have reported, cardiac abnormalities are transient and often resolve without lasting sequelae [9, 15]. Our findings indicate that muscular deconditioning may be a contributor to the observed decline in exercise capacity. Interestingly, higher BMI, a higher PELOD 2 score, and higher PICU length of stay at the PICU admission, and lower haemoglobin level during CPET were associated with impaired aerobic capacity in the MIS-C group. These factors may represent markers of heightened severity and inflammation, which could potentially contribute to muscular dysfunction and reduced exercise tolerance [35]. However, the observed muscular deconditioning in the MIS-C population might be also due, at least in part, to the recommended restriction of physical activity following the acute illness.

The observed decrease in VO_2peak in our MIS-C population compared to healthy controls (magnitude of the difference in VO_2peak at 6 mL/Kg/min) is particularly concerning in light of existing research. Studies in healthy adults have shown that a VO_2peak reduction exceeding 3.5 mL/Kg/min is associated with an increased risk of all-cause mortality and cardiovascular events [36]. Furthermore, the prevalence of impaired exercise capacity in our MIS-C group is comparable to rates reported in children with chronic health conditions such as paediatric asthma [37], congenital heart disease [21], and childhood cancer survivors [22]. These parallels suggest the severity of the decline in aerobic capacity observed in MIS-C survivors and the potential long-term health implications in this patients’ population [38]. Further studies will be needed to confirm if this decline persists over the long term.

The restricted exercise capacity primarily stems from muscle deconditioning. Specifically, a notable difference in ventilatory anaerobic threshold (VAT) of 8.5 ml/kg/min compared to that of healthy children, resulting in impaired VAT (<−1.64 Z-score) in 43% of children post-MIS-C. This finding is reinforced by a significant Workload_peak decrease in the MIS-C group. Notably, despite the absence of cardiac and respiratory damage during follow-up, this discrepancy stands out as the key factor in the decline of aerobic capacity. This muscular deconditioning likely stems from not engaging in physical activity as advised according to international guidelines. While these guidelines are crucial to prevent complications of cardiac involvement [7, 39–42], they can lead to deconditioning and potentially contribute to a sedentary lifestyle. Muscular dysfunction and reduced exercise tolerance could be secondary to inflammation [35], or corticosteroid treatment [43]. However, this hypothesis remains unanswered due to limitations in the study design and the small sample size. Furthermore, our analysis revealed an association between lower haemoglobin levels and impaired VO_2peak. The Fick principle, which underlies the assessment of VO_2peak during CPET, highlights the critical role of haemoglobin in oxygen transport. A decrease in haemoglobin concentration can lead to impaired oxygen content in the blood, ultimately resulting in a lower VO_2peak. Previous research in the general population has consistently demonstrated the independent association between haemoglobin levels and VO_2max during CPET [44, 45].

These results underscore the importance of early intervention and proactive management strategies to address physical deconditioning and improve exercise tolerance in this population. While muscle deconditioning may not be the sole contributor to the observed impairment in VO_2peak, factors such as elevated inflammation, lower haemoglobin levels, and prolonged PICU stay also likely play significant roles. However, muscle deconditioning, particularly in patients with a higher BMI, remains the most important modifiable factor. Our study on a small number of patients suggests that early identification of patients at risk—those with higher PELOD 2 scores, higher PICU length of stay at PICU admission, lower haemoglobin levels at the time of CPET, and higher BMI—can help guide targeted interventions aimed at improving outcomes. The concept of post-ICU consultations emerges in adult ICU survivorship with outcomes such as quality of life, anxiety, depression, posttraumatic stress disorder, physical ability, and cognitive function [46]. However, the findings of a recent randomized controlled trial cast doubt on the efficacy of such consultations, highlighting the infancy of this concept, primarily reliant on expert opinions, as well as the absence of comprehensive observational data regarding patients’ requirements and progressions in the post-ICU period [47]. Existing paediatric rehabilitation programs [34] focused on exercise training, education (including dietetic education), and physical therapy could potentially be adapted to address the specific needs of MIS-C survivors. These initiatives, such as supervised exercise training by a specialized physical activity instructor, a comprehensive educational program led by psychologists, dieticians, and specialized nurses, and personalized physical therapy sessions provided by a physiotherapist, could be proposed to MIS-C patients and possibly to all paediatric PICU survivors to potentially mitigate the decline in physical capacity resulting from muscular deconditioning. Currently, there is no standardized or protocolized approach to rehabilitation strategies at the time of PICU discharge.

Study limitations

Our study has limitations. The control group consisted of children who were not hospitalized in the PICU, making it uncertain whether the observed findings are attributable to MIS-C, PICU admission, or both. However, at the time of the study, PICU controls were not available, as CPET was not routinely performed in the post-PICU period. Moreover, the sample size was modest, which diminished the strength of our findings and hindered our ability to pinpoint specific predictors of impaired aerobic capacity through multivariate analysis. Additionally, the gender distribution disparity observed in our sample may be influenced by recruitment or referral patterns that were not fully captured by our study design. Furthermore, the absence of severely ill patients, such as those necessitating mechanical ventilation, limits the broader applicability of our findings. To address these limitations, future studies should encompass larger, multicentre cohorts to delve deeper into these aspects. Additionally, our study relied solely on a single CPET assessment, and muscle deconditioning was primarily evaluated based on VAT and workload, which are the usual monitoring tools for muscle conditioning in rehabilitation programs [34]. However, the addition of instrumented assessments such as handheld or isokinetic dynamometry would have provided a more comprehensive evaluation. Employing longitudinal follow-up with repeated CPET testing and muscular instrumented assessment would offer valuable insights into the trajectory of aerobic capacity recovery and the efficacy of potential rehabilitation interventions.

Conclusion

This study suggests that children with MIS-C experience reductions in aerobic capacity compared to healthy controls (matched for age, sex, and BMI), with muscular deconditioning appearing to play a contributing role. While the MIS-C group exhibited higher BMI overall, controlling for this factor in our analysis allowed us to isolate a higher PELOD 2 score and higher PICU length of stay at the PICU admission, and lower haemoglobin level during CPET with impaired aerobic capacity. These findings highlight the importance of considering post-ICU consultations and implementing strategies to address physical deconditioning in this population. The potential benefits of physical rehabilitation programs warrant further exploration, especially considering the potential for these programs to improve physical function and overall well-being in this vulnerable population.

Authors’ contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by A. Maggio, H Huguet, A Polito, A Gavotto. The first draft of the manuscript was written by A. Maggio, A Polito, A Gavotto and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

Open access funding provided by University of Geneva. This work was supported by the Geneva University Hospital, Switzerland.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study adhered to Good Clinical Practices and Declaration of Helsinki principles. Approval was obtained from the Cantonal Ethics Committee (CCER 2020 − 00414) for inclusion of the MIS-C group, with written informed consent obtained from participants or their guardians. The Montpellier’s CPET database was approved by the Institutional Review Board of Montpellier University Hospital (2019_IRB-MTP_10–20), with informed consent obtained from all parents or legal guardians.

Competing interests

The authors declare no competing interests.

Declarations of interest

the authors have no competing interests to declare.

Consent to participate

Written informed consent was obtained from the parents.