Assessment of Coronary Function in Children With a History of Kawasaki Disease Using 15O-Water Positron Emission Tomography
Background— Coronary abnormalities after Kawasaki disease (KD) may be associated with endothelial dysfunction due to intimal hypertrophy. The purpose of this study was to evaluate myocardial flow reserve (MFR) and endothelial function in regressed aneurysmal regions after KD.
Methods and Results— Subjects were 12 patients aged 16.0±2.6 years who suffered from KD at 1.7±1.5 years and 12 normal subjects aged 26.5±3.4 years. MFR and endothelial function were estimated, respectively, by changes in myocardial blood flow (MBF) during ATP infusion and by that during cold pressor test using 15O-water positron emission tomography. Data from 24 regressed aneurysmal regions were compared with those from the corresponding regions (n=36) in the control group. Although the MBF at rest in the regressed aneurysmal regions was similar to that in controls, the MBF at a hyperemic state induced by ATP infusion in the regressed aneurysmal regions was significantly lower than that in the control regions. Therefore, the MFR in regressed aneurysmal regions was significantly lower than that in controls (3.53±0.95 versus 4.60±1.14; P<0.05). MBF at rest and during the cold pressor test did not change in the control regions, but it was significantly reduced in regressed aneurysmal regions. The ratio of MBF during the cold pressor test to MBF at rest was significantly lower in regressed aneurysmal regions than in control regions (0.67±0.15 versus 1.00±0.15; P<0.05).
Conclusions— MFR and endothelial function are often impaired in regressed aneurysmal regions after KD, and tomography enables the noninvasive evaluation of coronary function.
Received January 31, 2002; revision received April 4, 2002; accepted April 4, 2002.
Kawasaki disease (KD) is an acute, febrile, pediatric vasculitis of unknown cause that was first described in 1967.1,2⇓ Its most common and serious complications involve coronary arterial abnormalities, which develop in ≈25% of KD patients.3 Approximately half of the cases of coronary aneurysm seemed to resolve spontaneously. Although coronary arterial aneurysms and stenosis may lead to myocardial ischemia, infarction, and sudden death, they are not considered a cause of coronary events in regressed aneurysmal regions in general.
Recently, intimal hypertrophy has discovered in the coronary arteries of patients with a history of KD using intravascular ultrasound imaging and histopathology.4–6⇓⇓ Because endothelial dysfunction seems to play a major role in coronary events, endothelial function was assessed by cardiac catheterization using acetylcholine infusion in patients with a history of KD.7–9⇓⇓ These data indicated that regressed coronary arteries show constriction with acetylcholine infusion.
Positron emission tomography (PET) with 15O-labeled water is a noninvasive method of accurately quantifying regional myocardial blood flow (MBF). This method has been validated in previous reports.10,11⇓
ATP is a short-acting drug that mainly dilates coronary arterial smooth muscle. This drug induces hyperemic blood flow and is used to evaluate coronary artery disease.12 The cold pressor test (CPT) is useful for evaluating endothelial function,13–16⇓⇓⇓ and it induces coronary vascular dilation with normal endothelial cells.
This study was designed to estimate both myocardial flow reserve (MFR) and endothelial function during pharmacological stress (ATP infusion) and during cold stress (CPT) in patients with a history of KD by PET using 15O-labeled water.
Twenty patients (14 males and 6 females aged 17.5±3.2 years; range, 10 to 24 years) who suffered from KD at 1.7±1.4 years, and 12 healthy control subjects (aged 26.5±3.4 years; range, 23 to 35 years) who had no evidence of cardiovascular disease or risk factors such as hyperlipidemia, hypertension, diabetes mellitus, and smoking, as well as a normal ECG at rest, were examined. Of the 20 patients with KD, 12 patients (16.0±2.6 years) with regressed aneurysmal regions were enrolled. The mean duration from the onset of KD to the PET study was 14.2±3.1 years. Sedation was not necessary for any subject. Using echocardiography, we determined that all patients had coronary abnormalities in the acute phase.17 Treatment consisted of intravenous administration of high-dose gamma globulin and aspirin in 3 patients and only aspirin in the remaining 9 patients.
All the subjects refrained from consuming caffeine-containing beverages for at least 24 hours before the PET examination. The purpose and potential risks of this study were explained to all the subjects and their parents before informed consent was obtained.
Positron Emission Tomography
MBF at rest, during ATP infusion, and during CPT was determined noninvasively by PET using 15O-labeled water. All PET scans were performed with ECAT EXACT HR+ (Siemens/CTI).
The study protocol is shown in Figure 1. A transmission scan was performed to correct the photon attenuation for 6 minutes with a 68Ge source. Next, the subject inhaled 15O-labeled CO for 1 minute to obtain a blood volume image. The total inhaled dose of 15O-labeled CO was 2000 MBq. After inhalation of the tracer, 3 minutes were allowed for CO to combine with hemoglobin before a static scan for 5 minutes was started. A 10-minute period was allowed for 15O-labeled CO radioactive decay before the flow measurement; then, 1000 MBq of 15O-labeled water was infused into an antecubital vein. Simultaneously, a 24-frame dynamic PET scan was performed for 6 minutes; this consisted of 18×10-s and 6×30-s frames. Twelve minutes after the first infusion of 15O-labeled water, an intravenous drip infusion of ATP (0.16 mg · kg–1 · min–1) was started. It lasted until the end of the second PET scan using 15O-labeled water. Fifteen minutes later, the PET scan in the resting state using 15O-labeled water was performed following the same image acquisition sequence. Twelve minutes later, CPT was performed as follows. The patient’s left foot was immersed in ice water for 4 minutes while PET scanning using 15O-labeled water was performed following the same acquisition sequence. Finally, a transmission scan for 6 minutes was performed with a 68Ge source after CPT to correct the photon attenuation of the last PET scan using 15O-labeled water. The subject’s motion was minimized by fastening a Velcro strap across the chest and abdomen.
Heart rate, blood pressure (BP), and a 12-lead ECG were recorded at rest and at 1-minute intervals during and after the administration of ATP and CPT.
All emission sinograms were reconstructed with filtered back projection using a Hann filter (cutoff frequency, 0.3 cycle/pixel). The in-plane resolution has a 4.5-mm full width of half-maximum in images reconstructed into a 128×128 matrix. All data were corrected for dead time, decay, and measured photon attenuation.11
Quantification of MBF
The left ventricular cavity time-activity curve was used as the input function.10 The myocardial time-activity curves were fitted by a single-compartment kinetic model that estimates MBF.10 The regional MBF was quantified according to previously published methods.10,11⇓ The entire myocardial region of interest was set on the image obtained with 15O-labeled water.
To quantify regional MBF, the entire left ventricle was divided into the following 3 major coronary territories: left anterior descending, left circumflex, and right coronary arteries. Twenty-four regressed aneurysmal regions from 12 patients with a history of KD and 36 regions from 12 control subjects were assessed using MBF at rest, under ATP infusion, and under CPT. All PET data were analyzed by 3 expert doctors who were blinded to the patients’ clinical data.
Calculation of Myocardial Vascular Resistance and Rate-Pressure Product
Heart rate, diastolic BP (DBP), and systolic BP (SBP) were determined for each subject. The rate-pressure product (RPP) was calculated as heart rate×SBP. Myocardial vascular resistance was calculated by dividing the mean BP by MBF. The mean BP was calculated as [2×(DBP+SBP)]/3.
All data are expressed as mean±SD. The changes in hemodynamic parameters between rest, during pharmacological stress, and during CPT in all groups were compared with a paired ttest. Moreover, MBF at rest during ATP-induced hyperemia and during CPT in both groups was compared with an unpaired t test. P<0.05 was considered statistically significant.
All hemodynamic parameters are shown in Table 1. No significant differences in SBP, DBP, mean BP, and RPP at rest, during ATP stress, and during cold stress were observed between the 2 groups. Heart rate at rest and during ATP stress did not differ between the 2 groups, but heart rate during cold stress in the KD group was significantly higher than that in the control group. All parameters during ATP stress or cold stress significantly increased compared with those at rest.
Hyperemic Blood Flow and MFR
All patient and control data are shown in Table 2⇓. Because MBF at rest is influenced by RPP as the index of cardiac work,18 it was corrected against RPP to account for individual differences in cardiac work as follows: MBF at rest was divided by RPP and multiplied by 6512, which is the average RPP at rest of healthy normal volunteers aged 23.0±5.7 years. MBF at a hyperemic state induced by pharmacological stress was not corrected for RPP because ATP causes direct coronary vasodilation and acts independently of blood flow from oxygen demand (ie, cardiac work).18
The corrected MBF at rest did not differ between regressed aneurysmal regions and the corresponding regions in the control group (0.91±0.19 versus 0.84±0.18 mL · g–1 · min–1; P=NS; Figure 2). MBF at a hyperemic state caused by ATP stress significantly increased in both regressed aneurysmal regions and corresponding control regions; however, it was significantly lower in the regressed aneurysmal regions than in control regions (3.16±0.99 versus 3.83±1.02 mL · g–1 · min–1; P<0.05). Therefore, MFR, defined as the ratio of MBF at a hyperemic state to that at rest in regressed aneurysmal regions, was significantly lower in aneurysmal regions than that in control regions (3.53±0.95 versus 4.60±1.14, P<0.05), although some regions between the 2 groups overlapped (Figure 3).
Coronary Vasomotor Response to CPT
All the patient and control data are shown in Table 2⇑. MBF during cold stress significantly increased in regressed aneurysmal regions (from 0.95±0.29 to 1.11±0.42 mL · g–1 · min–1; P<0.05) and in control regions (from 0.82±0.15 to 1.12±0.27 mL · g–1 · min–1; P<0.05).
The magnitude of the flow response to cold stress varied between participants. Because the response of MBF to cold stress depends on RPP, which in turn is related to sympathetic nerve activation, MBF at rest and MBF during CPT were corrected for RPP.19 The corrected MBF at rest was similar in both the regressed aneurysmal regions and control regions (0.93±0.21 versus 0.81±0.12 mL · g–1 · min–1; P=NS). Although the corrected MBF after cold stress in the control regions showed no significant difference from the MBF at rest (from 0.81±0.12 to 0.81±0.13 mL · g–1 · min–1; P=NS), in regressed aneurysmal regions, the corrected MBF after cold stress significantly decreased compared with the MBF at rest (from 0.93±0.21 to 0.61±0.16 mL · g–1 · min–1; P<0.05; Figure 4). Therefore, the ratio of the corrected MBF during CPT to the MBF at rest in the regressed aneurysmal regions was significantly lower than that in the control regions (0.67±0.15 versus 1.00±0.15; P<0.05), although 3 of these values in the regressed aneurysmal regions were within the SD of controls (Figure 5).
Coronary Vascular Resistance
To exclude the effect of changes in coronary perfusion pressure, the index of coronary vascular resistance (mm Hg · mL–1 · g–1 · min–1) was calculated (see Methods). No difference in coronary vascular resistance was observed at rest in either regressed aneurysmal regions or control regions (82.6±17.7 versus 91.5±19.4 mm Hg · mL–1 · g–1 · min–1; P=NS). Furthermore, the coronary resistance during ATP stress did not differ between the regions (22.7±9.1 versus 18.7±5.7 mm Hg · mL–1 · g–1 · min–1; P=NS).
The coronary resistance in the control regions did not show significant changes between cold stress and rest (from 86.6±14.9 to 82.6±16.8 mm Hg · mL–1 · g–1 · min–1; P=NS). However, the coronary resistance during cold stress in the regressed aneurysmal regions was significantly increased from that at rest (from 76.4±19.0 to 92.2±27.3 mm Hg · mL–1 · g–1 · min–1; P<0.05).
The present study indicated that regressed aneurysmal regions after KD had partly impaired MFR and endothelial function, as determined by PET examination.
KD is known to involve systemic vasculitis and to induce coronary arterial abnormalities. Our findings support the major pathophysiology of KD, which is intimal hypertrophy as determined by histological examination or intravascular ultrasound imaging.4–6⇓⇓ To our knowledge, this is the first study designed to evaluate coronary endothelial-dependent and endothelial-independent vasomotion by PET.
Endothelial-Independent Coronary Function
ATP mainly induces endothelial-independent vasodilation by relaxing arterial smooth muscle. This diagnostic value is similar to the vasodilation induced by adenosine.12
In this study, the hyperemic blood flow induced by ATP and MFR in regressed aneurysmal regions was significantly lower than that in the control regions. This observation was similar to that of previous reports.20,21⇓ Ohmochi et al20 reported that the MFR of patients with coronary aneurysms after KD was significantly lower than that of patients with normal coronary angiograms after KD. Muzik et al21 compared the MFR between patients without coronary arterial lesions after KD and healthy volunteers and found that the MFR after KD was significantly lower than that of healthy volunteers. They suggested that reduced hyperemic blood flow and MFR can be attributed to obstruction in small vessels resulting from fibrous thickening of the intima.
Because diffuse intimal hypertrophy was observed by intravascular ultrasound4,5⇓ or histopathology,6 annular stricture or obstruction in small vessels may also occur in the regressed aneurysmal regions. Therefore, the occurrence of impaired coronary microcirculation after KD may contribute to reduced hyperemic blood flow and MFR.
Endothelial-Dependent Coronary Function
The CPT is useful for noninvasive evaluation of endothelial-dependent functions.13,14⇓ Cold stimulation activates the sympathetic nervous system, with an increase in the level of catecholamines, and both α- and β-adrenergic activations induce neural effects on the coronary vascular bed. In the coronary artery with an intact and normally functioning endothelium, β-adrenergic activation induces direct coronary vasodilation and an increase in coronary blood flow. An increased coronary blood flow induces an increment in shear stress on endothelial cells, which then induces α2-adrenergic stimulation and releases an endothelium-derived relaxing factor (EDRF). The released EDRF acts on coronary smooth muscle cells and induces vasodilation. However, in the coronary artery with an impaired and dysfunctioning endothelium, coronary vasodilation does not occur because the EDRF is not released from impaired endothelial cells.
In this study, MBF at rest in both the regressed aneurysmal regions and the corresponding regions in the control group significantly increased after CPT. Because the individual responses to cold were different, both MBF at rest and during CPT in all the subjects was corrected for RPP. The ratio of corrected MBF after CPT to the MBF at rest in regressed aneurysmal regions was significantly lower than that in the control regions (P<0.05). These findings suggest that the endothelial function in regressed aneurysm regions is impaired.
A number of studies have revealed that epicardial coronary arteries with regressed aneurysms are constricted by acetylcholine infusion.7–9⇓⇓ These studies concluded that endothelial- dependent vasomotion is impaired in regressed aneurysmal vessels. Pathologically, diffuse abnormal fibrous thickening of the intima without inflammatory changes had been observed in regressed coronary aneurysms.6 Sugimura et al4 reported that marked thickening of the intima was demonstrated at the site of the regressed coronary artery by intravascular ultrasound. Panvasculitis resulting from KD may affect resistance vessels, and intimal thickening may occur even in the resistance vessels.22
Our patients had contracted KD at 1.7±1.5 years, and 14.2±3.1 years had passed from diagnosis to evaluation in this study. Suzuki et al23 reported that active remodeling of the coronary arterial lesions continues in the form of luxuriant intimal proliferation for several years after the onset of KD. This fact may help explain why endothelial function did not recover late after the onset of KD.
First, control subjects were older than patients because the administration of radioactive agents to normal children is ethically difficult. Unfortunately, no published MBF data are available on normal children. However, MFR (endothelial-independent function) and endothelial-dependent function decrease linearly with increasing age.18,24,26,27⇓⇓⇓ Therefore, impairment of endothelial-dependent or -independent function is clinically significant. In a previous report, the reduced MFR was due to a high MBF at rest with increasing age.18 In the present study, MBF at rest in the KD patients was similar to that in the control subjects. Consequently, impairment of MFR was mainly due to reduced hyperemic blood flow in KD patients.
Second, the coronary arteries of all the patients with a history of KD were examined only by 2D echocardiography in the acute phase. Satomi et al17 reported that accurate coronary diagnosis should be done by multiple echo planes.
PET examination was of value in noninvasive and short-term assessment of coronary endothelial-dependent and -independent function in the late phase of KD. The MFR and endothelial function are often impaired in regressed aneurysmal regions after KD, despite the fact that, angiographically, the regions exhibit smooth and normal arteries.
We would like to thank Ken-ichi Nishijima and Kotaro Suzuki for excellent work in providing isotopes and PET scanner handling. We would also like to thank Yachio Ohta and Isamu Hamada of the Hokkaido Kawasaki Disease Research Society for referring patients with a history of KD to us.
- ↵Kawasaki T. Acute febrile mucocutaneous syndrome with lymphoid involvement with specific desquamation of the fingers and toes in children. Jpn J Allergy. 1967; 16: 178–222.Japanese.
- ↵Kawasaki T, Kosaki F, Okawa S, et al. A new infantile acute febrile mucocutaneous lymph node syndrome (MCLS) prevailing in Japan. Pediatrics. 1974; 54: 271–276.
- ↵Kato H, Sugimura T, Akagi T, et al. Long-term consequences of Kawasaki disease: a 10- to 21-year follow-up study of 594 patients. Circulation. 1996; 94: 1379–1385.
- ↵Sugimura T, Kato H, Inoue O, et al. Intravascular ultrasound of coronary arteries in children: assessment of the wall morphology and the lumen after Kawasaki disease. Circulation. 1994; 89: 258–265.
- ↵Fujiwara T, Fujiwara H, Nakano H. Pathological features of coronary arteries in children with Kawasaki disease in which coronary arterial aneurysm was absent at autopsy. Circulation. 1988; 78: 345–350.
- ↵Mitani Y, Okuda Y, Shimpo H, et al. Impaired endothelial function in epicardial coronary arteries after Kawasaki disease. Circulation. 1997; 96: 454–461.
- ↵Iemura M, Ishii T, Sugimura T, et al. Long term consequences of regressed coronary aneurysms after Kawasaki disease: vascular wall morphology and function. Heart. 2000; 83: 307–311.
- ↵Iida H, Kanno I, Takahashi A, et al. Measurement of absolute myocardial blood flow with H215O and dynamic positron emission tomography: strategy for quantification in relation to the partial-volume effect. Circulation. 1988; 78: 104–115.
- ↵Katoh C, Ruotsalainen U, Laine H, et al. Iterative reconstruction based on median root prior in quantification of myocardial blood flow and oxygen metabolism. J Nucl Med. 1999; 40: 862–867.
- ↵Nabel EG, Ganz P, Gordon JB, et al. Dilation of normal and constriction of atherosclerotic coronary arteries caused by the cold pressor test. Circulation. 1988; 77: 43–52.
- ↵Uren NG, Crake T, Tousoulis D, et al. Impairment of the myocardial vasomotor response to cold pressor stress in collateral dependent myocardium. Heart. 1997; 78: 61–67.
- ↵Czernin J, Muller P, Chan S, et al. Influence of age and hemodynamics on myocardial blood flow and flow reserve. Circulation. 1993; 88: 62–69.
- ↵Campisi R, Czernin J, Schoder H, et al. Effects of long-term smoking on myocardial blood flow, coronary vasomotion, and vasodilator capacity. Circulation. 1998; 98: 119–125.
- ↵Masuda H, Kanda M, Naoe S, et al. Coronary arterial lesions after Kawasaki disease: assessment of the intramural coronary vessels. Clin Immunol. 1982; 14: 441–450.Japanese.
- ↵Suzuki A, Miyagawa-Tomita S, Komatsu K, et al. Active remodeling of the coronary arterial lesions in the late phase of Kawasaki disease: immunohistochemical study. Circulation. 2000; 101: 1935–1941.
- ↵Zeiher AM, Drexler H, Saubier B, et al. Endothelium-mediated coronary blood flow modulation in humans: Effect of age, atherosclerosis, hypercholesterolemia, and hypertension. J Clin Invest. 1993; 92: 652–662.
- ↵Di Carli MF, Bianco-Bartlles D, Landa ME, et al. Effects of autonomic neuropathy on coronary blood flow in patients with diabetes mellitus. Circulation. 1999; 100: 813–819.
- ↵Yasue H, Matsuyama K, Matsuyama K, et al. Responses of angiographically normal human coronary arteries to intracoronary injection of acetylcholine by age and segment: possible role of early coronary atherosclerosis. Circulation. 1990; 81: 482–490.
- ↵Egashira K, Inou T, Hirooka Y, et al. Effects of age on endothelium-dependent vasodilation of resistance coronary artery by acetylcholine in humans. Circulation. 1993; 88: 77–81.