OBJECTIVEKawasaki disease (KD) is a kind of febrile disorder without definite etiology. The pathologic change of KD is characterized by nonspecific vasculitis, which mainly involves the coronary artery. Some patients may have coronary angioma formation, and some of them will result in the coronary narrowing or embolism. Notwithstanding that KD has been one of the most common causes for acquired heart diseases in childhood in addition to the rheumatic fever, the pathogenesis of the vascular damage remains unknown. This study was conducted to explore the pathophysiological role of cell adhesion molecules (P-selectin and E-selectin) on the endothelial lesions in KD, and to look for the evidence of direct relationship between the plasma levels of soluble cell adhesion molecules (P- and E-selectin) and the incidence of the coronary artery lesion (CAL).

METHODSSoluble P-selectin (PS), E-selectin (ES), thromboxane-B(2)(TXB(2)), 6-keto-PGF(1)alpha (6-KPGF(1)alpha) were measured in 36 patients with KD, 20 patients with febrile disease and 30 healthy children by using double antibody sandwich enzyme linked immunosorbent assay (ELISA) and radioimmunoassay. Patients with KD were separated into acute phase group, subacute phase group, recovery phase group, coronary artery lesion group (CAL), non-coronary artery lesion group (NCAL), intravenous immunoglobulin (IVIG) effective group (body temperature back to normal after 48 hours of using IVIG), and IVIG ineffective group.

RESULTSPlasma PS and ES levels in the acute phase group [(211 +/- 28 and 186 +/- 14) ng/ml], subacute phase group [(238 +/- 27 and 151 +/- 13) ng/ml] and recovery phase group [(198 +/- 21 and 1008 +/- 9) ng/ml] were significantly higher than those in the healthy group [(102 +/- 36 and 72 +/- 10) ng/ml, P < 0.01]. The plasma PS levels remained higher after the treatment, but in IVIG effective group, the PS and ES levels declined significantly (P < 0.01) compared with those in acute phase group. Plasma PS and ES levels of CAL group [(281 +/- 78 and 210 +/- 52) ng/ml] were significantly higher than those of NCAL group [(217 +/- 15 and 108 +/- 10) ng/ml, P < 0.01]. In contrast to 1 week after the treatment, the PS and ES in IVIG effective group at the time point of 2 weeks after the treatment decreased more significantly (P < 0.01). While the PS and ES in IVIG ineffective group remained higher at the time point of 2 weeks after the treatment, which showed no significant difference compared with those 1 week after the treatment (P > 0.05). One week after the treatment, the PS levels of IVIG effective and ineffective groups did not descend, and there was no significant difference in PS between these two groups at this time point. Two weeks after the treatment, the PS and ES in IVIG ineffective group remained higher than those in IVIG effective group, and there was a significant difference between them. The peak level of PS appeared in the subacute phase. TXB(2) levels of KD in acute phase group increased markedly, which were significantly higher than those of healthy group [(345 +/- 127 and 190 +/- 69) ng/L, P < 0.01]. There was no significant difference between subacute phase group and healthy group. No significant difference was found between CAL group and NCAL group (P > 0.05). The levels of TXB(2) declined quickly after the treatment. The 6-KPGF(1)alpha level in KD of acute phase group, subacute phase group and recovery phase group [(7.1 +/- 2.8, 10.8 +/- 3.7 and 11.3 +/- 4.0) ng/L, respectively] was significantly lower than that of healthy group [(17.7 +/- 5.8) ng/L, P < 0.01], and the levels did not recover to normal even 2 weeks after the treatment. There was no significant difference 6-KPGF(1)alpha levels between CAL group and NCAL group (P > 0.05). In the febrile group, PS and ES levels showed no significant differences compared with healthy children (P > 0.05). ES level of KD patients was significantly correlated with CRP levels (r = 0.79 P < 0.01). In febrile group, there was no significant correlation between ES and CRP. There was a significant correlation between PS and PLT levels in KD patients (r = 0.75 P < 0.01), and no significant correlation between PS and PLT levels in febrile patients.

CONCLUSIONThe increase of plasma PS and ES levels in KD acute phase and subacute phase might play an important role in the pathophysiology of the endothelial damage. E- and P-selectin may potentially be a predictor of CAL in patients with KD.