Kawasaki disease (KD) is the most common disease causing pediatric acquired heart disease. According to the epidemiological statistics of Japan in 1997-1998, the incidence rate was 112/100,000 in children under 5 years of age, and the incidence rates in Shanghai and Beijing in mainland China were 16.8-36.8/100,000 and 26-31/100,000, respectively, but the total incidence rate in China is unknown. In recent years, incomplete Kawasaki disease (iKD) has received increasing attention as some cases with atypical presentation or not fully meeting the diagnostic criteria of Kawasaki disease have been reported in China and abroad. Because of the lack of gold standard for diagnosing iKD, its early diagnosis is difficult to be accomplished in a timely manner. In this paper, we selected children with KD admitted to our hospital for a prospective controlled clinical trial to investigate the value of CRP, ESR, plasma BNP, HDL and urinary LDH for the diagnosis of incomplete KD, in order to facilitate the early diagnosis of iKD and to start treatment within 10 days of onset of disease as much as possible to reduce the chance of cardiovascular and other complications.
1. Clinical data and methods
1.1 General data: A total of 40 children with KD admitted to our hospital from January 2007 to June 2008 were selected, and the diagnosis of the cases was based on the 5th revision of the diagnosis of Kawasaki disease issued by the Kawasaki Disease Committee of Japan in 2005 and the diagnosis and treatment guidelines of iKD jointly formulated by the AAP and AHA in 2004 [1, 4]. There were 20 cases in the typical KD group, 12 males and 8 females, with a male-to-female sex ratio of 1.5:1 and ages 4 months-9 years (median: 15.5 months); and 20 cases in the iKD group, 14 males and 6 females, with a male-to-female sex ratio of 2.33:1 and ages 3 months-10 years (median: 16.0 months).
1.2 Methods
1.2.1 Samples were collected in the acute phase for selected typical KD and iKD cases, respectively, with emphasis on plasma BNP, HDL and urinary LDH values, in addition to routine measurement of CRP and ESR. Twenty normal children attending our hospital were randomly selected to establish a control group with informed consent from their families, and the blood collection time was synchronized with that of the KD cases. The plasma BNP, HDL and urinary LDH were compared for statistical differences among the three groups of KD, iKD and normal controls, and whether there were correlations between these three indicators and CRP and ESR levels.
1.2.2 Statistical processing methods: SAS 8.02 statistical software was used for data analysis. Data of measurement data were expressed as mean ± standard deviation or M, and t-test with lognormal distribution was used; P0.05 was considered statistically significant difference.
2, Results
Among the selected case samples, 4 cases in the KD group had coronary artery dilation and 7 cases had hypoproteinemia; only 2 cases in the iKD group had coronary artery dilation and no hypoproteinemia was seen.
There was no statistical difference (P0.05) between the KD group and the iKD group in terms of age, with a P value of 0.7449. In the three groups, the values of blood BNP, urinary LDH, blood HDL and the statistical results of the two controls are shown in Table 1, Table 2. linear correlation analysis is shown in Table 3.
Table 1 Plasma BNP, HDL and urinary LDH values in the three groups (mean ± standard deviation) Blood HDL (lg) Urinary LDH (lg) Blood BNP (lg)
KD group-0.43±0.251.80±0.472.87±0.48
iKD group-0.19±0.211.89±0.502.55±0.28
Control group 0.11±0.231.00±0.401.68±0.31
Table 2 Two-by-two comparison of plasma BNP, HDL and urinary LDH values in the three groups (P value) Blood HDL (lg) Urinary LDH (lg) Blood BNP (lg)
KD VSiKD group0.00141)0.52990.00571)
KD VS control group 0.00011)0.00011)0.00011)
iKD VS control group 0.00011)0.00011)0.00011) Note: 1) P<0.01, statistically significant difference.
Table 3 Correlation analysis of plasma BNP, HDL and urinary LDH values with CRP and ESR levels
Blood HDL(lg) urinary LDH(lg) blood BNP(lg) ESRCRP blood HDL(lg)r-0.483)-0.603)-0.06-0.19P0.00011)0.00011)0.70310.2361 urinary LDH(lg)r0.502)0.17-0.1358P0.00011) 0.27620.4097 blood BNP(lg)r0.220.422)P0.17320.00811)ESRr0.332)P0.04071)Note: 1) P<0.05, statistically different; 2) 1> r>0, positive correlation; 3) -1
In several current retrospective studies, iKD accounted for 10%-36% of children with KD, with studies from the country of Japan in recent years showing an incidence of 13.8%. Data from a domestic study showed that 19.4% of children with KD had an atypical or incomplete presentation. iKD has a high incidence in infancy and the chance of coronary artery disease is comparable to that of typical KD, but because the diagnosis of children with iKD is often delayed, the chance of cardiovascular damage tends to be higher. Therefore, the early diagnosis and treatment of iKD is of great importance.
According to the diagnostic criteria of typical KD, iKD is defined as a child with fever ≥ 5 d but only 2 or 3 out of 5 other clinical features, and febrile diseases such as scarlet fever, drug allergy syndrome, Stevens Johnson syndrome, toxic shock syndrome, adenovirus infection, and Epstein Barr (EB) virus infection need to be excluded.
Several studies have now found that children with iKD may have fewer clinical features than children with typical KD, but their laboratory diagnostic indicators are consistent with those of typical KD children. Therefore, although the changes in laboratory parameters in children with KD are not specific, it is helpful to focus on the changes in laboratory parameters such as CRP, ESR, D, D-dimer, and echocardiography showing coronary artery abnormalities. Recently, plasma BNP, HDL and urinary LDH have been found to be significantly elevated or decreased during the acute phase of iKD in children with iKD, and they have certain specificity, which helps in the diagnosis of iKD.
Plasma BNP was first recognized to be secreted by the atria, and in children with congenital heart disease, elevated plasma BNP levels can promote vasodilation and diuresis, thereby reducing anterior and posterior loads. Takeuchi et al. suggested that abnormal myocardial mechanical exercise may lead to elevated plasma BNP levels. In our case, comparative analysis of plasma BNP data suggested that the increase in blood BNP in the acute phase of KD was statistically significant and had diagnostic value. In the linear analysis of plasma BNP and CRP in both KD groups, there was a significant positive correlation, suggesting an association between myocardial inflammation and elevated inflammatory factors in the acute phase in children with Kawasaki disease, which is consistent with the literature.
As a “protective” apolipoprotein, HDL regulates the secretion of endothelial cells and facilitates the increase of endothelial blood flow. It also interferes with the expression of various inflammatory factors, counteracts many oxidative stress reactions inside and outside the body, prevents blood cells from adhering to the endothelium, reduces platelet aggregation and coagulation, and maintains the integrity of the endothelium; and inhibits endothelial progenitor cells. (endothelial progenitor cells, EPCs) and maintain the dynamic repair function of endothelium. In recent years, it has been reported that endothelial function is impaired to varying degrees in patients with reduced HDL-C, and low HDL-C has been an independent risk factor for the development of ischemic diseases. As a systemic vascular injury disease, several studies have shown a significant decrease in blood HDL during the acute phase of KD. In our group, the decrease in plasma HDL was statistically significant, suggesting that the acute phase of KD is associated with endothelial injury, which has diagnostic value.
In addition to aseptic pus urine, there have been specific case reports of increased activity of urinary LDH in the acute phase of KD. Urinary LDH is an enzyme reflecting metabolism, with a molecular weight of 12×104 Dalton, and is not easily excreted from the blood into the urine via the kidneys, so urinary LDH activity is low in normal subjects. Due to the damage of intima of blood vessels in children with KD, the permeability increases, which provides the basis for LDH excretion from urine. All cases in this group excluded primary damage diseases of the urinary system, and there was a statistically significant difference compared with the control group, suggesting that the increase of urinary LDH in the acute phase of KD has diagnostic value.
In this study, a significant linear negative correlation was found between plasma HDL values and BNP and urinary LDH, and a significant linear positive correlation was found between urinary LDH values and plasma BNP, which may be related to the synchronous changes of these three indicators due to the specific pathophysiological damage of KD disease.
In this group, plasma BNP, HDL and urinary LDH data were significantly changed in the typical KD group compared with the iKD group. The reason for this may be related to the fact that there were more cardiovascular and other complications in the typical KD group than in the iKD group, and the severity of the disease was slightly higher. Therefore, further data from multicenter and large sample studies are needed to support this paper, and this is the limitation of this paper.
In general, iKD has the same pathophysiological changes as typical KD, and although the clinical manifestations are incomplete, from the results of this clinical study, the detection rate of early diagnosis of iKD in the acute phase can be improved by improving relevant laboratory tests and increasing the detection of plasma BNP, HDL and urinary LDH, so that early treatment can be given to avoid complications.