Incomplete clinical manifestation as a risk factor for coronary artery abnormalities in Kawasaki disease: a meta-analysis.

Incomplete Kawasaki disease (KD) comprises a large proportion of the total number of cases. Although it has the potential of delaying diagnosis, it is not conclusive whether an incomplete presentation is a risk factor for coronary artery abnormalities (CAAs). We performed a meta-analysis to establish the risk of CAA in 20 studies including 4,504 cases and 32,519 controls, and the risk of giant aneurysm in two studies including 5,390 cases and 37,648 controls. The pooled results indicated that incomplete KD was associated with an increased risk of CAA [odds ratio (OR) = 1.447, 95 % confidence interval (CI) = 1.158-1.808, p = 0.001]. Subgroup analyses demonstrated higher associations in patients younger than 12 months (OR = 2.023, 95 % CI = 1.252-3.271, p = 0.004), Asians and Indians (OR = 1.57, 95 % CI = 1.234-1.999, p < 0.001 and OR = 7.088, 95 % CI = 1.640-30.631, p = 0.009, respectively). Subgroup analysis according to the period of patient enrollment before and after 2004 showed increased association of incomplete KD with CAA only among studies with patients enrolled after 2004 (OR = 1.969, 95 % CI = 1.240-3.127, p = 0.004). In conclusion, incomplete KD seems to be associated with an increased risk of CAA, and this is more prominent in patients younger than 12 months, Asians and Indians.

Risk factors for Kawasaki disease-associated coronary abnormalities differ depending on age.

INTRODUCTION: The clinical manifestations and risk factors for developing coronary artery abnormalities (CAA) in Kawasaki disease (KD) might differ depending on age. MATERIALS AND METHODS: From January 2001 to July 2007, 161 patients with an age younger than 1 year (younger group) and 60 patients with an age older than 5 years (older group) were diagnosed with KD at the Korea University Medical Center. Their medical records were reviewed retrospectively and the two groups were compared in terms of a number of variables commonly associated with the development of CAA, including clinical manifestations and laboratory findings. RESULTS: While the overall incidence of KD-associated CAA in our hospital was 6.7%, CAA developed in 20 (12.4%) of the younger group and ten (16.7%) of the older group, respectively. The CAA (+) cases of the younger group had a longer duration of total fever (9.1 +/- 3.3 vs 6.3 +/- 1.9 days, p = 0.002) and showed fewer diagnostic symptoms (3.0 +/- 1.2 vs 4.3 +/- 1.1, p < 0.001) than the CAA (-) cases. The CAA (+) cases of the older group had a longer duration of total fever (14.1 +/- 10.4 vs 6.5 +/- 1.9 days, p = 0.045), especially with respect to post-intravenous gamma globulin (IVGG) fever (7.9 +/- 9.6 vs 1.1 +/- 0.8 days, p = 0.052), and had higher total white blood cell counts, erythrocyte sedimentation rates, C-reactive protein levels, total bilirubin levels, and Harada scores and lower serum albumin and sodium levels than the CAA (-) cases. Multivariable logistic regression analysis revealed that the factors that were associated significantly with the development of CAA were the number of total symptoms (OR = 0.494, 95% confidence interval (CI) = 0.281-0.871, p = 0.015) in the younger group and the duration of post-IVGG fever (OR = 1.958, 95% CI = 1.098-3.492, p = 0.023) and the Harada score (OR = 3.455, 95% CI = 1.012-11.796, p = 0.048) in the older group. CONCLUSION: Incomplete clinical manifestations in the younger group and IVGG nonresponsiveness in the older group are associated with the development of KD-associated CAA. These age-specific characteristics could aid the customization of the diagnostic and therapeutic strategies of KD, thereby helping to improve the outcome of this disease.

Incomplete Kawasaki disease in patients younger than 1 year of age: a possible inherent risk factor.

Kawasaki disease (KD) patients younger than 1 year of age are at especially high risk of developing coronary artery abnormalities (CAA). To define the clinical characteristics of this group, as well as the risk factors predisposing them to CAA, we reviewed the medical records of 136 KD patients younger than 1 year of age who were treated at the Korea University Medical Center from January 2001 to July 2006. Of these patients, 16 developed CAA (11.8%). The CAA(+) group had a longer duration of total fever than the CAA(-) group (9.1+/-3.7 days vs. 6.3+/-2.0 days, p=0.011), but did not differ in the duration of pre- and post-intravenous gamma-globulin (IVGG) fever. The CAA(+) group had fewer diagnostic symptoms than the CAA(-) group (2.7+/-1.1 vs. 4.3+/-1.2, p<0.001). Of the hematological findings, the CAA(+) group only differed from the CAA(-) group in having significantly higher total white blood cell (19.2+/-6.0 vs. 14.7+/-4.7 K/mm(3), p=0.007) and platelet (462.9+/-101.0 vs. 383.6+/-121.1 K/mm(3), p=0.014) levels. Multivariable logistic regression analysis showed that the only factors which were significantly associated with the development of CAA were the total number of symptoms (OR=0.493, 95% CI=0.293-0.829, p=0.007) and the duration of total fever (OR=1.405, 95% CI=1.092-1.808, p=0.008). Conclusively, incomplete clinical manifestations and a longer duration of total fever are significantly associated with the development of CAA in KD patients younger than 1 year of age. Therefore, these patients should be monitored for incomplete KD, especially if unexplained fever continues, and treatment to shorten the duration of total fever should be initiated.

Predictive risk factors for coronary artery abnormalities in Kawasaki disease.

Clinical characteristics to predict the development of coronary artery abnormalities (CAA) in Kawasaki disease (KD) were assessed by reviewing medical records of patients diagnosed with KD at Korea University Medical Center from March 2001 to February 2005. Of the 285 patients diagnosed with KD, 19 developed CAA (6.7%). Compared with the CAA(-) group, the CAA(+) group had a longer duration of fever after intravenous gamma-globulin (IVGG) injection (2.4+/-2.9 vs. 1.5+/-1.2 days, p=0.008) and higher C-reactive protein (CRP)(12.3+/-7.8 vs. 8.7+/-7.1 mg/dL, p=0.038). In particular, the CAA(+) group tended to have more than 7 days of fever before IVGG and more than 3 days of fever after IVGG (26.3 vs. 5.3%, p<0.001; 26.3 vs. 6.4%, p=0.002). When the IVGG responsiveness was defined by the presence of defervescence within 3 days after IVGG, IVGG-non-responders showed a higher incidence of CAA (22.7 vs. 5.3%, p=0.002). Non-responders had a longer duration of fever after IVGG (5.5+/-1.9 vs. 1.2+/-0.6 days, p<0.001) and a significantly increased CRP, AST, ALT and total bilirubin. Multivariate regression analysis for CAA showed that the only factor significantly associated with the development of CAA was total fever that lasted for longer than 8 days (OR=4.052, 95% CI=1.151-14.263, p=0.0293). Conclusively, the most important predictor of CAA in KD is total duration of fever longer than 8 days. Early identification of IVGG non-responders and active therapeutic intervention for fever in KD cases might decrease the incidence of CAA.