Exercise tolerance test in Kawasaki patients without coronary lesions

Abstract

Introduction

Kawasaki disease (KD) is a systemic vasculitis of unknown origin and the leading cause of acquired heart disease in children in developed countries. If untreated, it can cause coronary artery damage in up to 25% of cases. Lifelong follow-up is essential, even for patients without obvious coronary lesions (risk level-I KD). Since some cardiac abnormalities may not be apparent at rest, stress testing offers valuable insight. This study aimed to evaluate exercise tolerance test (ETT) outcomes in children with a history of risk level-I KD compared to a control group.

Materials and methods

This case-control study included children aged 5–17 years with documented KD without coronary lesions and at least six months since disease onset. After ethics approval, participants underwent a standardized treadmill test (modified Bruce protocol). A control group of healthy children was also included; however, exact age-matched analysis was not possible due to recruitment constraints.

Results

Sixty children were divided equally into KD and control groups. The KD group had significantly lower mean age, height, weight, and BSA compared to the control group. After adjusting for these variables, most exercise indices showed no significant difference between groups, except for chronotropic incompetence which was more prevalent in the KD group. However, most differences became non-significant after adjusting for age and BMI.

Conclusion

Children with KD generally showed normal exercise tolerance, but a reduced chronotropic response may indicate subtle residual cardiac effects. These findings suggest a potential for subtle cardiac effects and may support the consideration of stress-based evaluation in select patients; however, further larger prospective studies are needed to confirm these observations.

Introduction

Kawasaki disease (KD) is an acute systemic vasculitis of unknown etiology that primarily affects young children and predominantly involves small and medium-sized vessels. It is currently the leading cause of acquired heart disease in children in developed countries. Without timely treatment, coronary artery lesions (CAL) occurs in up to 25% of patients and may lead to aneurysm, stenosis, thrombosis, myocardial ischemia, or infarction later in life in 1% of patients [1, 2]. Diagnosis is primarily clinical and supported by laboratory evidence of systemic inflammation, while echocardiography remains the cornerstone of cardiac evaluation and follow-up according to American Heart Association (AHA) guidelines [1, 2].

Although intravenous immunoglobulin (IVIG) therapy significantly reduces coronary complications, a subset of patients still develops long-term cardiovascular sequelae despite optimal treatment [3, 4].

Studies evaluating exercise capacity and cardiorespiratory function in patients with a history of KD are relatively limited and often involve small sample sizes.

However, it remains unclear whether subclinical cardiac dysfunction, detectable only during stress, exists in KD patients without CAL. Therefore, this study aimed to evaluate ETT findings in this population.

Most KD patients without CAL are traditionally considered to have a favorable prognosis. Clinical guidelines provide differing recommendations. The AHA 2017 statement and its 2024 update do not recommend routine long-term cardiology follow-up for patients classified as risk level I (AHA risk level I: without CAL) [1, 5]. In contrast, the Japanese Circulation Society (JCS/JSCS) 2020 guidelines advocate follow-up for up to five years and recommend ETT even in low-risk KD patients [6]. More recent studies employing advanced modalities such as PET imaging have demonstrated evidence of microvascular dysfunction and reduced exercise capacity despite the absence of anatomical coronary abnormalities. Some studies demonstrated that even KD patients without CAL during the acute phase may develop late cardiac abnormalities, including ventricular wall motion abnormalities, ST-segment changes during exercise, and arrhythmias during 5- and 10-year follow-up [7, 8].

Maximal oxygen uptake (Max-VO₂), representing the highest oxygen consumption during peak exercise, is an important index for assessing functional capacity in ETT. In some studies, functional assessments also showed significant reductions in Max-VO₂ and maximal blood pressure (BP) in children with KD compared with the control group, suggesting impaired cardiorespiratory function and an increased risk of long-term cardiovascular complications even among patients without CAL [9, 10].

Pathological studies also demonstrated persistent thickening of the intima of coronary arteries even after apparent regression of aneurysms, indicating ongoing vascular remodeling and potential long-term risk [9].

Baseline cardiac assessments may fail to detect subtle functional cardiovascular changes; instead, during exercise or drug stress, subtle coronary changes may manifest as reduced myocardial perfusion. ETT offers a valuable and noninvasive method for assessing underlying abnormalities in exercise capacity, chronotropic response, hemodynamic regulation, and myocardial perfusion [1, 2].

In summary, existing evidence suggests that while many KD patients without CAL exhibit normal exercise tolerance, subtle and subclinical cardiovascular dysfunction may emerge over time. These findings underscore the importance of functional cardiovascular assessment, especially ETT, and support a more individualized long-term follow-up strategy for patients with a history of KD. This study tested the hypothesis that children with a history of KD without CAL may have detectable differences in cardiorespiratory responses to exercise compared to healthy controls. We specifically evaluated exercise capacity, blood pressure response, electrocardiographic changes, and chronotropic index to identify potential subclinical abnormalities.

Materials and methods

This was a cross-sectional case-control study designed to evaluate cardiovascular functional parameters derived from ETT in children with a history of KD compared with a control group.

Patients were recruited consecutively from the Pediatric Cardiology Clinic of Children’s Medical Center Hospital, Tehran, during 2025. All KD patients at AHA Risk Level-I who met the inclusion criteria were invited to participate. The control group consisted of healthy children randomly referred for evaluation of atypical chest pain or sports clearance, with normal echocardiography and ETT findings.

The study population included children aged 5–17 years with a documented history of KD who were followed at the Pediatric Cardiology Clinic of Children’s Medical Center Hospital, Tehran, during 2025. Inclusion criteria were: (1) history of KD according to AHA diagnostic guidelines; (2) age 5–17 years; (3) time from acute phase 6 months to 5 years; and (4) classification as AHA Risk Level-I, defined as the absence of coronary artery lesions (CAL) with a maximum coronary Z-score < + 2.5 (using the Boston Z-score reference system). All echocardiographic coronary measurements were performed by a single experienced pediatric cardiologist. The control group consisted of healthy children randomly referred for evaluation of atypical chest pain or sports clearance, with normal echocardiography and ETT findings. Exclusion criteria included structural heart disease, heart failure, significant arrhythmias, chronic pulmonary disease and a history of myocarditis or pericarditis related to KD.

All participants underwent treadmill ETT using the Modified Bruce protocol (with 2-minute stages), with continuous electrocardiographic monitoring under the supervision of trained personnel. The treadmill used was a @ AvicinaTreadmill. The test was terminated upon participant discomfort or clinical indication, including: 1- participant’s request due to fatigue or discomfort, 2- achieving ≥ 85% of age-predicted maximal heart rate with no desire to continue, 3- significant ST-segment changes (≥ 2 mm depression or elevation), 4- a drop in systolic blood pressure > 10 mmHg from baseline despite increasing workload, 5- severe arrhythmia, or 6- any sign of respiratory distress. Termination was based on clinical judgment and participant discomfort.

Parameters recorded included baseline and peak heart rate (HR), HR reserve, systolic and diastolic blood pressure (SBP, DBP), double product (DP), metabolic equivalents of task (METs) and estimated Max-VO₂. These parameters reflect hemodynamic responses to exercise, and abnormal rises may indicate impaired cardiac function. Baseline HR, SBP, and DBP were measured prior to exercise testing as indicators of baseline cardiovascular status in children.

Predicted (or expected) maximal heart rate (Max-HR) was estimated for each participant using the Tanaka formula [208 − (0.7 × age)], which is considered appropriate for pediatrics [11]. Reaching at least 85% of predicted Max-HR was considered attainment of target HR. Also, predicted Max-SBP and Max-DBP values (50th, 90th, 95th percentile) were calculated using the Sasaki formula [12, 13]. Achieved Max-HR, Max-SBP, and Max-DBP were defined as the highest HR, SBP, and DBP recorded during ETT. To assess exercise intensity, the percentage of predicted Max-HR achieved (Achieved Max-HR%) was calculated as the ratio of Achieved Max-HR to age-predicted Max-HR.

Heart Rate Reserve (HRReserve) represents the difference between achieved Max-HR and baseline HR, reflecting sympathetic function and the ability to increase cardiac output during physical activity [14, 15].

Predicted HRReserve was defined as predicted maximal heart rate minus baseline heart rate. The Chronotropic Index (CI) evaluates the heart’s ability to increase HR proportionally to metabolic demand, reflecting sinoatrial node function and autonomic regulation [16]. CI is calculated as HRReserve ÷ predicted HRReserve. CI < 0.8 indicates chronotropic incompetence. To ensure that a reduced heart rate response was not merely due to submaximal effort, we verified that all participants achieved ≥ 85% of their age-predicted maximal heart rate, which is considered an adequate exercise effort in pediatric populations [11].

Hypertensive Response to Exercise (HRE), also referred to as Exaggerated BP Response (EBPR), is defined as an abnormal rise in SBP or DBP during or after exercise exceeding the 95th percentile of predicted values. In children, HRE may signal an increased risk for future sustained hypertension [17]. Blunted Response to Exercise (BRE) is characterized by an inadequate increase in SBP (< 10 mmHg from baseline) during exercise, potentially reflecting impaired cardiovascular or neurohormonal regulation [12, 13] Double Product (DP) or Peak Rate-Pressure Product (PRPP) is an indirect measure of myocardial oxygen demand and cardiac workload during exercise, calculated as HR×SBP. Higher DP indicates greater myocardial oxygen consumption, while reduced DP may reflect impaired myocardial workload or coronary perfusion. In children, DP values are lower than those in adults but increase progressively with exercise. Interpretation requires consideration of age, sex, growth, and exercise protocol, highlighting the need for age- and sex-specific reference values [8].

Max-VO₂ is a key indicator of aerobic capacity and cardiopulmonary fitness. Maximal Metabolic Equivalents of Task (Max-METs) quantify exercise energy expenditure relative to rest and reflect relative workload; 1 MET equals baseline oxygen consumption (3.5 mL·kg⁻¹·min⁻¹). Achieved Max-METs reflect the highest METs level reached during ETT, representing peak exercise load capacity. Predicted Max-METs, calculated using Armstrong & Welsman [18], provide age- and sex-adjusted predicted values. Peak percentage of predicted VO₂(Peak-PD), represents the percentage of achieved predicted Max-METs (or Max-VO₂) and was calculated as the ratio of measured Max-METs (or Max-VO₂) during exercise to predicted maximal values. Higher Peak-PD values indicate better cardiorespiratory fitness. Metabolic reserve (MET-Reserve) indicates the functional capacity from rest to maximal exercise (Max-METs minus baseline METs). As explained above, baseline METs are defined as 1 MET. Peak percentage of predicted Metabolic Reserve (MET-Reserve%) quantifies the proportion of an individual’s usable exercise capacity relative to their predicted maximum. It is calculated as:

METs Reserve%=[(Achieved Max-METs minus 1)/(Predicted Max-METs minus 1)]×100.

This index standardizes exercise performance, allowing more accurate comparisons across individuals of different ages and sexes [19].

HR recovery (HRR) and HR recovery index (HRRI%), were assessed to evaluate autonomic function. Recovery HR (HRR) refers to the reduction in HR following exercise termination, commonly measured at specific time points such as 2- and 10-min post-exercise (2-min-HRR, 10-min-HRR). Normal or rapid HRR indicates effective parasympathetic reactivation, while delayed HRR suggests impaired autonomic or cardiovascular function [20, 21]. The HRR Index (HRRI%) quantifies the percentage reduction of peak HR (Achieved Max-HR) after 2-min post-exercise. Using Singh criteria [20], 2-min-HRR < 25 bpm for boys and < 20 bpm for girls indicates delayed 2-min-HRR. Delayed SBP and DBP recovery, defined as values exceeding the 95th percentile at 2 min [12]. The SBP Recovery Index (SBPR), calculated as the ratio of 2-min post-exercise SBP to peak exercise SBP (Achieved Max-SBP), is a key marker of cardiovascular recovery. Values above 0.90 indicate delayed or impaired SBP recovery. SBP and DBP delayed recovery is potentially linked to sympathetic overactivity and, in adults, is associated with an increased risk of hypertension and arterial stiffness [22].

Another important parameter recorded during the whole exercise test is the electrocardiogram. The highest ST-segment elevation (Max-STE) and greatest ST-segment depression (Max-STD) across all ECG leads during exercise testing reflect physiologic or pathologic cardiac responses. Mild ST-segment elevation (STE) in precordial leads (V2–V4) in children, especially asymptomatic adolescents, is commonly physiologic, reflecting rapid depolarization and increased sympathetic activity [23, 24]. Mild, upsloping Max-STD, particularly in inferior or lateral leads, is usually nonspecific and physiologic in pediatric exercise testing [24]. Another parameter, Total Beats, represents the total number of heartbeats recorded during the ETT, reflecting cardiac response to exercise load. High total beats are physiologic in children due to greater cardiovascular capacity and rapid HR response [24]. Abnormal Beats include ventricular and supraventricular beats. Most abnormal beats were isolated, non-sustained, and occurred without clinical symptoms, consistent with physiologic responses to exercise [24].

Statistical analysis was performed using SPSS version 26. Quantitative variables were expressed as mean±standard deviation (SD) and compared using independent samples t-tests or covariate analysis (ANCOVA), while categorical variables were analyzed with chi-square (χ²) tests. To assess the independent effect of Kawasaki disease on chronotropic incompetence while controlling for potential confounders, binary logistic regression analysis was performed. Independent variables included group (KD vs. control), age, sex, and body surface area (BSA). Results were reported as odds ratios (OR) with 95% confidence intervals (CI), and a p-value < 0.05 was considered statistically significant.

Ethical approval was obtained from the Pediatric Ethics Department of Tehran University of Medical Sciences, informed consent was obtained from parents, and all data were handled confidentially.

Results

Out of 911 children diagnosed with KD during 2019–2025, 316 met the age criteria, 151 were classified as AHA Risk Level-I, and of these, 30 children (16 boys, 14 girls) consented to participate, along with 30 healthy children in a control group (18 boys, 12 girls). In total, making 60 participants; there was no significance between groups (p = 0.605).

KD patients had a significantly lower mean age (9.06 ± 1.89) than the control group (11.66 ± 2.50, p < 0.001) (Table 1). Consistent with this finding, differences were observed in height, weight, and body surface area (BSA) between the two groups. KD patients were shorter (132.8 ± 18.73 cm) compared with controls (149.16 ± 16.05 cm), with the difference statistically significant (p = 0.001). KD patients weighed significantly less (34.58 ± 16.40 kg) than the control group (43.93 ± 15.25 kg, p = 0.031). Regarding the mean body mass index (BMI), KD and control groups did not differ significantly (p = 0.574). BMI categories showed 53.3% underweight, 38.3% normal, and 8.3% overweight in all participants. KD patients had significantly lower BSA (1.12 ± 0.33 m²) than controls (1.33 ± 0.29 m², p = 0.014). Other demographic findings, mean daily screen time and daily physical activity, had no significant difference between KD and control groups (p = 0.345 and p = 0.259, respectively).

Table 1.

Comparison of demographic, anthropometric and coronary characteristics between KD patients and control group

Variable	KD group (Mean ± SD)(range)	control group (Mean ± SD)(range)	p-value
Age (years)	9.06 ± 1.89 (7–14)	11.66 ± 2.50 (7–17)	0.00003
Height (cm)	132.8 ± 18.73 (90–171)	149.16 ± 16.05 (113–185)	0.001
Weight (kg)	34.58 ± 16.40 (19–68)	43.93 ± 15.25 (22–79)	0.031
BMI (kg/m²)	18.64 ± 4.23 (12.80–28.10)	19.26 ± 4.27 (12.80–29.70)	0.574
Body Surface Area (BSA, m²)	1.12 ± 0.33 (0.70–1.80)	1.33 ± 0.29 (0.80–1.90)	0.014
Screen Time (hours/day)	3.63 ± 2.12 (1.00–10.00)	4.27 ± 3.15 (0.50–12.00)	0.345
Daily Physical Activity (hours/day)	2.45 ± 1.46 (0.20-6.00)	2.93 ± 1.74 (0.50-6.00)	0.259
LMCA	2.19 ± 0.43 (1.40–3.50)	2.70 ± 0.36 (2.00-3.40)	0.353*
LAD	2.14 ± 0.38 (1.50–3.30)	2.21 ± 0.15 (2.00-3.30)	0.461
RCA	2.85 ± 0.52 (2.00–4.00)	2.60 ± 0.45 (2.10–4.40)	0.051

P-values were calculated using the independent t-test, except variables marked with * (ANCOVA)

Among patients with KD, the mean age at disease onset was 6.15 ± 2.49 years (range: 2–11 years). The mean time since hospital discharge was 3.14 ± 1.59 years. Initial treatment approaches at the acute phase varied, with 26.7% receiving no therapy, 33.3% treated with IVIG alone, 20% receiving methylprednisolone pulse therapy alone, and 20% receiving combination therapy.

Regarding coronary artery diameters, the left main coronary artery (LMCA) diameter was significantly smaller in the KD group compared with controls (2.19 ± 0.43 mm vs. 2.70 ± 0.36 mm). However, after adjustment for age, sex, and BSA using ANCOVA, the group effect was no longer significant (p = 0.353).

For the left anterior descending (LAD) and right coronary arteries (RCA), the mean diameters were 2.14 ± 0.38 mm and 2.85 ± 0.52 mm in patients with KD, and 2.21 ± 0.15 mm and 2.60 ± 0.45 mm in the control group, respectively; no statistically significant differences were observed between groups (Table 1).

Hemodynamic response in children

Baseline HR, SBP, DBP, Achieved Max-HR, Max-SBP, and Max-DBP have no statistically significant difference between groups (Table 2). All values were within normal pediatric ranges. Percent achievement of predicted Max-SBP and Max-DBP were evaluated to assess BP responses to physical stress in children and showed no significant differences between groups (p = 0.216 and p = 0.591, respectively).

Table 2.

Comparison of baseline and exercise hemodynamic parameters between two groups

Variable	KD group(Mean ± SD) (range)	control group (Mean ± SD) (range)	p-value
Baseline HR	105.86 ± 15.06 (76–139)	106.40 ± 15.02 (79–138)	0.8943
Achieved Max-HR	185.60 ± 20.89 (150–220)	191.67 ± 17.31 (155–220)	0.231
Achieved Max-HR%	92.00 ± 10.70 (80–110)	95.93 ± 8.64 (80–110)	0.132
HRReserve	79.73 ± 22.88 (35–157)	85.27 ± 19.73 (44–131)	0.316
2-min-HRR	64.86 ± 28.67 (3-133)	67.06 ± 22.75 (6-116)	0.751
Baseline SBP	114.66 ± 16.87 (80–153)	116.06 ± 10.69 (93–143)	0.7181
Baseline DBP	71.70 ± 12.09 (45–97)	73.46 ± 7.72 (62–92)	0.5049
Achieved Max-SBP	138.96 ± 11.07 (120–155)	143.20 ± 13.78 (125–160)	0.198
Achieved Max-DBP	80.93 ± 8.75 (70–95)	80.40 ± 10.52 (70–100)	0.833
SBP at 2-min post-exercise	121.57 ± 15.94 (98–164)	130.30 ± 22.39 (99–195)	0.084
DBP at 2-min post-exercise	73.86 ± 8.19 (62–97)	73.13 ± 9.62 (56–93)	0.757
SBPR index	0.87 ± 0.08 (0.73-1.00)	0.91 ± 0.14 (0.73–1.54)	0.176
Max-METs	14.29 ± 2.12 (11–17)	16.62 ± 2.52 (12–18)	0.085*
Max-METs% (Peak-PD)	99.27 ± 16.19 (75–125)	117.93 ± 17.68 (80–130)	0.099*
DP	18222.13 ± 4101.14 (9642–27875)	18755.61 ± 2673.11 (12688-25732.5)	0.550
CI	0.82 ± 0.24 (0.20–1.48)	0.92 ± 0.20 (0.52–1.37)	0.089
MET-Reserve%	100.30 ± 20.64 (65–150)	119.33 ± 18.93 (72–151)	0.10*
Total test Time (min)	19 min 44 s ± 1 min 18 s	21 min 5 s ± 1 min 39 s	0.073*

Min minutes

P-values were calculated using the independent t-test, except variables marked with * (ANCOVA)

In regard of HRE, chi-square analysis showed no significant difference between groups (p = 0.603, Table 3).About BRE all participants exhibited normal SBP responses, and no cases of BRE were observed. The mean DP had no significant difference between KD patients (18,222.13 ± 4101.14) and the control group (18,755.61 ± 2673.11, p = 0.550). Most participants fell within an acceptable range (10,000–20,000 mmHg×bpm).

Table 3.

Frequency of achievement of exercise targets and different exercise responses in both groups

Variable	Condition	KD Group (n = 30)	control group (n = 30)	p-value (χ²)
Target HR Achievement	Normal Achievement	23 (76.7%)	27 (90.0%)	0.144
	Failure to Achieve	7 (23.3%)	3 (10.0%)
Acceptable METs Performance	Satisfactory	27 (90.0%)	29 (96.7%)	0.301
	Below 80% of predicted	3 (10.0%)	1 (3.3%)
HRE	Normal Response	18 (60.0%)	16 (53.3%)	0.603
	Exaggerated Response	12 (40.0%)	14 (46.7%)
BRE	Normal Response	30 (100%)	30 (100%)	-
	Blunted Response	0 (0%)	0 (0%)
CI	Normal Response	16 (53.3%)	24 (80.0%)	0.0018
	Chronotropic Incompetence	14 (46.7%)	6 (20.0%)
MET-Reserve%	Maximal Reserve(> 85%)	27 (90.0%)	29 (96.7%)	0.301
	Submaximal Reserve(< 85%)	3 (10.0%)	1 (3.3%)
2-min-HRR	Adequate Recovery	28 (93.3%)	28 (93.3%)	1.000
	Delayed Recovery	2 (7.6%)	2 (7.6%)
Delayed 2-min SBP Recovery	Normal Recovery	30 (100%)	29 (96.7%)	0.313
	Elevated Response	0 (0%)	1 (3.3%)
Delayed 2-min DBP Recovery	Normal Recovery	24 (80.0%)	26 (86.7%)	0.490
	Elevated Response	6 (20.0%)	4 (13.3%)
SBPR Index	Normal Recovery	20 (66.7%)	17 (56.7%)	0.392
	Delayed Recovery	10 (33.3%)	13 (43.3%)

The percentage of predicted Max-HR achieved (Achieved Max-HR% in table-2) was calculated.The KD group reached a mean of 92.0%, while the control group achieved 95.93%, with no significant between-group difference (p = 0.132). Reaching target HR was not statistically different between groups (p = 0.144) (Table 3).

Heart Rate Reserve (HRReserve) Analysis showed no statistically significant difference between groups (p = 0.316). KD patients had a mean CI of 0.81 ± 0.24, with 46.7% showing incompetence, while the control group had a mean CI of 0.91 ± 0.19, with 20% affected. Although an independent t-test showed no significant difference in mean CI (p = 0.089), Chi-square analysis revealed a significantly higher prevalence of chronotropic incompetence in KD patients (p = 0.024). Logistic regression controlling for age, sex, and BSA confirmed that KD independently increased the odds of chronotropic incompetence (CI < 0.8) with an odds ratio of 9.68 (95% CI: 2.34–40.05, p = 0.0018).

Cardiac workload and metabolic reserve assessment

About Max-VO₂, no significant differences were seen between both groups (p = 0.086). KD patients had a mean of 14.29 ± 2.12 METs, and the control group 16.62 ± 2.52 METs (Table 2), which was not statistically significant (p = 0.085). The mean Peak-PD was 99.27% in the KD group and 117.93% in control group; however, the difference was not statistically significant (p = 0.099). Most children (93.3%) achieved ≥ 80% of predicted Max-METs, indicating acceptable performance.

KD patients had lower mean MET-Reserve (100.30%) compared to the control group (119.33%). However, ANCOVA adjusting for age and sex revealed no significant group effect (p = 0.100). Chi-square analysis confirmed no significant difference in submaximal vs. maximal performance (METs Reserve > 85%) between groups (p = 0.301).

Hemodynamic recovery response

No significant differences were found between KD and control groups in 2-min-HRR, 10-min-HRR, and HRRI% (p = 0.751, p = 0.993, and p = 0.819, respectively). In this study, 93.3% of participants had normal HRR, while 6.7% showed delayed or slowed HRR, evenly distributed across groups. SBP at 2- and 10-min post-exercise had no significant differences between both groups (p = 0.084 and p = 0.259, respectively). DBP at 2- and 10-min post-exercise had no significant differences between both groups (p = 0.757 and p = 0.812, respectively). Delayed SBP and DBP recovery have no statistically significant difference between groups (p = 0.313 and p = 0.490, respectively). In KD patients, mean SBPR was 0.874 ± 0.076 with 33.3% exhibiting delayed recovery, compared to 0.911 ± 0.142 in the control group, with 43.3% showing delayed recovery, which were not statistically significant (t-test p = 0.176; Chi-square p = 0.392) .

ETT protocol parameters

The mean duration of the ETT (total test time), maximum treadmill speed, and grade did not show a significant difference in this study (p = 0.073, p = 0.115, and p = 0.151, respectively). Most participants (83.3%) stopped the test after achieving the target HR. Other termination causes were infrequent and heterogeneous, limiting valid statistical analysis.

Electrophysiologic response

In this study, the mean of Max-STE and Max-STD in total participants was 0.42 ± 0.24 mV and − 0.287 ± 0.10 mV, respectively. Leads V2 and V3 most frequently showed Max-STE (20% each). In KD patients, the mean of Max-STE and Max-STD was 0.385 ± 0.254 mV and − 0.285 ± 0.098 mV, and in the control group 0.458 ± 0.243 mV and − 0.291 ± 0.117 mV, with no significant difference (p = 0.267 and p = 0.827, respectively).

About total beats, KD patients had a mean of 2,427 ± 526, while the control group had 2,758 ± 321 beats; the difference was not significant (p = 0.090). KD patients had slightly fewer abnormal beats than the control group, but the differences were not significant (p > 0.05).

Discussion

KD is an acute childhood vasculitis damaging coronary arteries, risking aneurysms and severe heart issues [1, 2]. Treatment is crucial, but about 5% remain at risk for major cardiac events [3, 4]. Even KD patients classified as AHA risk level I may face long-term cardiovascular risks such as arterial stiffness and early atherosclerosis in adulthood, suggesting a need for ongoing monitoring [25].

In the present study, KD patients were significantly younger and had lower height, weight, and BSA than the control group. Overall, while raw differences in exercise capacity were observed between the KD and control groups, many were attenuated after adjusting for demographic and anthropometric variables, emphasizing the importance of age in interpreting pediatric ETT results. These findings underscore the importance of careful interpretation of ETT in KD children, considering anthropometric and clinical factors. Finally, HR, SBP, DBP, VO₂ and METs showed no significant differences between the groups across baseline, predicted, achieved, reserve, and recovery parameters. Electrophysiological responses, including ST-segment changes and abnormal beats, were physiologic and not significant between groups.

Previous studies have demonstrated variable effects of KD on exercise capacity. Kuan et al. found sex differences in exercise response among KD patients, with girls showing higher Max-METs and peak-PD than boys [26]. In the present study, VO₂ and METs showed no significant differences between the groups across Baseline, predicted, achieved, Reserve, and recovery parameters. This aligns with findings by Marchesi et al. and Manlhiot et al., suggesting that KD patients without CAL didn’t exhibit major limitations in physical activity [27].

Study by Tuan et al. shows that KD patients with higher coronary Z-scores have reduced exercise capacity, though those without CAL maintain normal Max-VO₂, peak METs, and DP [8], and our finding was the same. DP (HR×SBP) reflects myocardial oxygen demand and increases physiologically with exercise in children. In our study, there were no significant differences in DP between groups.

Some studies report reduced Max-VO₂ and exercise tolerance in KD patients, even without CAL, possibly due to microvascular dysfunction or chronic endothelial impairment [28]. For instance, Yu-Ling et al. observed lower exercise performance in KD patients [29], whereas our study found no differences, likely due to case selection (KD without CAL) and the timing of assessment during convalescence. Lin et al. and Lu et al. also noted that aerobic capacity and Max-VO₂ in KD patients (adolescents) were lower than controls in follow-up assessments despite normal coronary Z-scores [3, 30]. Long-term studies using PET or angiography reveal persistent microvascular or intimal changes in KD patients even after apparent recovery of coronary anatomy, which may not affect exercise capacity in the absence of CAL [9, 31].

In our study, the mean CI was also similar, but chronotropic incompetence was more prevalent in KD patients (independent of anthropometric factors). This apparent discrepancy is explained by a threshold effect: many KD patients had CI values just below the 0.8 cut-off (e.g., 0.75–0.79), while most controls were above it. Logistic regression, being more sensitive to categorical outcomes, confirmed the association. This finding requires further investigation to confirm or refute. The most plausible mechanism is autonomic dysfunction (reduced sympathetic response), as previously reported in KD. Sinus node dysfunction is unlikely given normal baseline heart rates, and deconditioning is unlikely as physical activity did not differ between groups. Our finding of increased chronotropic incompetence in KD patients without CAL is supported by previous literature. Gravel et al., in a multicenter study of 264 KD patients (including 117 without coronary complications), demonstrated that chronotropic competence is impaired late after Kawasaki disease, regardless of the presence of coronary aneurysms during the acute phase [32]. These findings suggest that autonomic dysfunction may persist in KD patients even in the absence of structural coronary lesions.

Based on this single small study, we cannot change current follow-up guidelines. However, the higher prevalence of chronotropic incompetence in KD patients without CAL raises the question of whether current AHA recommendations might benefit from re-evaluation in future larger prospective studies. While we do not advocate for routine ETT in all low-risk patients, our findings suggest that selective stress testing—particularly in patients with suboptimal performance or concerning symptoms—might help identify those with subtle autonomic dysfunction, a concept that aligns with JCS recommendations. Nevertheless, these observations are preliminary and require confirmation before any clinical recommendations can be made.

This study has several limitations. First, there was a significant age difference between the KD and control groups due to the referral pattern of controls (older children referred for chest pain or sports clearance); although adjustments were made, residual confounding remains possible. Second, only 19.9% of eligible KD patients consented, raising selection bias (healthier families may have participated more, potentially underestimating abnormalities). Third, the control group was not completely asymptomatic(nonspecific atypical chest pain), though all had normal ETT and echocardiography. Fourth, the lack of blinding and the small sample size (30 per group) limit statistical power. Fifth, despite all participants achieving ≥ 85% of predicted maximal heart rate, we cannot exclude the possibility that the higher prevalence of chronotropic incompetence in the KD group reflects reduced effort or deconditioning rather than true autonomic dysfunction. Finally, age and anthropometric differences persisted despite adjustment. Therefore, findings should be interpreted cautiously. Future studies should employ prospective age- and sex-matched designs with larger samples, ideally using propensity score methods.

Conclusion

In conclusion, our findings show no statistically significant differences in exercise performance and cardiovascular responses between children with KD without CAL and controls. The higher prevalence of chronotropic incompetence in the KD group (46.7% vs. 20.0%) suggests possible residual autonomic or cardiac effects; however, given the study’s limitations, this finding is hypothesis-generating and requires confirmation in larger prospective studies before any change in clinical practice can be recommended.

Abbreviations

KD

Kawasaki Disease

ETT

Exercise Tolerance Test

AHA

American Heart Association

JCS

Japanese Circulation Society

CAL

Coronary Artery Lesions

IVIG

Intravenous Immunoglobulin

HR

Heart Rate

SBP

Systolic Blood Pressure

DBP

Diastolic Blood Pressure

BMI

Body Mass Index

BSA

Body Surface Area

MET

Metabolic Equivalent of Task

CI

Chronotropic Index

HRE

Hypertensive Response to Exercise

DP

Double Product

Authors' contributions

E.A.M. contributed to the study conception and design, supervised the clinical management and patient recruitment for cardiac evaluation, provided critical intellectual guidance as the thesis advisor, and reviewed the manuscript. K.S.Z. contributed to the study design, provided supervision as a thesis advisor, and critically reviewed the manuscript. V.Z., F.T., and P.S. contributed as consulting advisors, facilitated patient referral for cardiac evaluation and exercise testing, and critically reviewed the manuscript. F.H.K. contributed to data collection and sample acquisition. S.M.S.* was responsible for project coordination and cardiac fellowship duties, conducted the primary thesis work, performed data collection and analysis, supervised the exercise testing procedures, obtained informed consent, drafted the main manuscript text, and prepared the tables. All authors (E.A.M., V.Z., K.S.Z., F.T., P.S., F.H.K., S.M.S.) reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Funding

This research received no specific grant from any funding agency.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to limitations of participant consent but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted according to the Helsinki Declaration. Ethical approval was obtained from Tehran University of Medical Sciences. Written informed consent was obtained from parents or legal guardians.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.