Treatment change and coronary artery abnormality in incomplete Kawasaki disease
Background
Incomplete Kawasaki disease (iKD) showed a higher incidence of coronary artery abnormalities (CAAs) than complete KD. However, the incidence of CAAs among iKD patients may have changed recently.
Methods
We examined KD patients from recent nationwide surveys conducted between 2013 and 2016 and compared them with the results of a previous survey (2001–2002).
Results
Of 63 270 KD patients, 13 770 patients (22%) had iKD. They showed a higher incidence of convalescent‐phase CAAs (cCAAs, 2.8%) than complete KD (2.1%). The incidence of cCAAs in patients with one or two symptoms (6.7%) was significantly higher than those with three or four symptoms (2.6%) (P < 0.0001). Intravenous immunoglobulin (IVIG) treatment was administered to 80% of iKD patients; 30% of them received IVIG before the fifth illness day (early treatment) and 12% of patients received IVIG after the seventh illness day (late treatment). In the previous survey, the incidence of cCAAs was higher in both iKD (5.9%) and cKD (4.4%). Intravenous immunoglobulin was administered to 62% of iKD patients; 26% of them received early treatment, and 16% received late treatment.
Conclusions
The incidence of cCAAs remained higher among iKD patients than cKD patients but this difference was reduced by the increased proportion of iKD patients treated with IVIG and those at an earlier time point. It is important to recognize the possibility that patients may have iKD and perform echocardiography even if they present with a few principal symptoms.
The diagnosis of KD is established using six principal symptoms:1-4 (i) fever persisting for 5 or more days (inclusive of patients in whom the fever had subsided before the fifth day in response to therapy); (ii) bilateral conjunctival congestion; (iii) changes to the lips and oral cavity; (iv) polymorphous exanthema; (v) changes to peripheral extremities; and (vi) acute nonpurulent cervical lymphadenopathy. Patients with five or six principal symptoms are diagnosed with complete KD (cKD) and those with four or fewer symptoms are diagnosed with incomplete KD (iKD).5 It was reported that the incidence of CAAs was higher in iKD patients than in cKD patients.5, 6 Sonobe et al .5 reported that the convalescent phase CAAs (cCAAs) occurred in 5.9% and 4.4% of iKD patients and cKD patients, respectively, based on data from the 17th nationwide survey of KD in Japan (2001–2002). Sudo et al .6 also reported similar findings (proportion of iKD patients with cCAAs: 3.5%, that of cKD patients with cCAAs: 2.5%) based on data from the 20th survey (2007–2008).
The prevalence of KD increased from 19 952 in the 17th survey5, 7 to 31 595 in the 24th survey.7 In parallel with this change, the proportion of iKD patients (in whom the incidence of CAAs is typically higher) increased from 16% in the 17th survey5, 7 to 23% in the 24th survey.7 However, the prevalence of cCAAs among KD patients decreased from 5.0% in the 17th survey to 2.3% in the 24th survey.5, 7 Considering these divergent trends, the incidence of CAAs among iKD patients may have changed. Using recent data, we therefore examined the incidence of CAAs in iKD patients. As the incidence of CAAs in KD is deeply related to their treatment, we also examined the treatments in iKD patients.
Methods
We analyzed data from the 23rd and 24th nationwide surveys of KD,7 which were collected from patients who visited target hospitals in Japan for the treatment of acute KD between 2013 and 2016. Hospitals specializing in pediatrics and hospitals with a pediatric department and 100 or more beds participated in the surveys (1,456 facilities in the 23rd survey and 1,444 in the 24th survey). Diagnostic criteria for KD in this study were based on guidelines issued by the Japan Kawasaki Disease Research Committee (5th revision).1, 4 We examined the incidence and classification of CAAs in iKD patients. According to the definition of the Japanese Ministry of Health,1, 8 coronary arteries were classified as abnormal if the internal lumen diameter was >3 mm in children younger than 5 years or >4 mm in children 5 years or older, when the internal diameter of any segment is at least 1.5 times greater than an adjacent segment, or when the coronary artery lumen is clearly irregular. Giant coronary aneurysm was defined as segments with an internal lumen diameter >8 mm. Coronary artery abnormalities were evaluated using two‐dimensional echocardiography, both during the first month of illness (acute) and at the first month of illness (convalescent phase).
The data on the timing and administration of intravenous immunoglobulin (IVIG) treatment, and the number of principal symptoms were also examined. Intravenous immunoglobulin treatment initiated before the fifth day of illness was defined as early treatment: IVIG treatment initiated between the fifth and seventh day of illness was defined as regular treatment, and IVIG treatment after the seventh day of illness was defined as late treatment.
The incidence of CAAs in iKD and cKD patients in the 23rd and 24th surveys was compared to that of the 17th or 20th surveys. Administration and timing of IVIG treatment were also compared. To evaluate the differences between two groups, we used a chi‐squared test.
Ethics
The 23rd and 24th nationwide surveys of KD were approved by the ethical board of Jichi Medical University, Tochigi, Japan (August 22, 2012, No. 12–18 and October 10, 2014, No. 14–50).
Results
This study included 63 270 KD patients from the 23rd and 24th nationwide surveys. Of these, 13 770 patients (22%) had iKD and 49 447 patients (78%) had cKD. Fifty‐eight percent and 57% of iKD and cKD patients, respectively, were male. The proportion of iKD patients under 1 year of age was 28%. This proportion was significantly higher than that of cKD patients (19%) (P < 0.0001). Most iKD patients presented with four (72%) or three symptoms (22%). Few iKD patients presented with only one (1%) or two (5%) symptoms.
Treatment with IVIG
Intravenous immunoglobulin treatment was administered to 80% of iKD patients and 97% of cKD patients. The proportion of iKD patients treated with IVIG varied according to the number of principal symptoms with which the patients presented: IVIG was provided to 64% of patients with two symptoms, 71% of patients with three symptoms, and 83% of patients with four symptoms (Fig. 1). Early treatment was performed in 30% and 36% of iKD and cKD patients, respectively. Late treatment was performed in 12% and 4% of iKD and cKD patients, respectively.
The proportion of incomplete Kawasaki disease patients treated with intravenous immunoglobulin treatment according to the number of principal symptoms. In both the 23rd and 24th nationwide surveys, the lowest proportion of incomplete Kawasaki disease patients treated with intravenous immunoglobulin (IVIG) was observed in patients who presented with two principal symptoms; the highest proportion of patients treated with IVIG was observed in those who presented with four symptoms. iKD, incomplete Kawasaki disease; IVIG, intravenous immunoglobulin.
Coronary artery abnormalities
Acute phase CAAs (aCAAs) were observed in 9.8% and 6.3% of iKD and cKD patients, respectively. Of the aCAAs in the iKD patients, 0.2% were giant aneurysms, 1.2% were medium aneurysms, and 8.4% were small aneurysms (Table 1). The incidence of cCAAs was slightly higher in iKD patients (2.8%) than in cKD patients (2.1%). Of the cCAAs in the iKD patients, 0.2% were giant aneurysms, 0.7% were medium aneurysms, and 1.9% were small aneurysms, showing that the difference in cCAAs between patients with iKD and cKD may be driven by the number of small aneurysms rather than giant or medium aneurysms. The size distribution of aCAAs and cCAAs was similar to that of cKD patients.
| Giant aneurysm | Medium aneurysm | Small aneurysm | Total (%) | |
|---|---|---|---|---|
| Acute phase | ||||
| Incomplete KD | 0.2% (1.9%) | 1.2% (12.0%) | 8.4% (86.1%) | 9.8 |
| Complete KD | 0.2% (2.4%) | 0.9% (13.9%) | 5.3% (83.7%) | 6.3 |
| Convalescent phase | ||||
| Incomplete KD | 0.2% (5.6%) | 0.7% (26.4%) | 1.9% (67.9%) | 2.8 |
| Complete KD | 0.2% (7.7%) | 0.7% (33.8%) | 1.2% (58.5%) | 2.1 |
- Numbers in parentheses are proportions of the total number of aneurysms. KD, Kawasaki disease.
We observed an inverse relationship between the number of principal symptoms and the incidence of aCAAs. We observed aCAAs in 6.3% of patients with five or six principal symptoms, 9.2% of patients with four symptoms, 9.9% of patients with three symptoms, and 15.8% of patients with one or two symptoms (Fig. 2). The incidence of aCAAs was significantly higher in patients with one or two symptoms than in patients with three or four symptoms (15.8% and 9.4%, respectively) (P < 0.0001). We also observed an inverse relationship between the number of principal symptoms and the incidence of cCAAs. The iKD patients with one or two symptoms had the highest incidence of cCAAs (6.7%) among the iKD patients (3.3% in patients with three symptoms and 2.4% in those with four symptoms) and cKD patients (2.1%), which was similar to the pattern we observed during the acute phase. The incidence of cCAAs was significantly higher in patients with one or two symptoms than in patients with three or four symptoms (2.6%) (P < 0.0001). Convalescent‐phase CAAs were also more severe in patients with fewer symptoms: among the patients who had cCAAs, the incidence of medium or giant aneurysms was significantly higher in patients with one or two symptoms (53%) than in patients with three or four symptoms (29%) (P = 0.001).
The relationship between the incidence of coronary artery abnormalities and the number of principal symptoms in patients with Kawasaki disease. We observed a reverse relationship between the number of principal symptoms with which Kawasaki disease patients presented and the incidence of both the acute and convalescent phase of coronary artery abnormalities. CAA, coronary artery abnormality. (□), Small aneurysm; (
); medium aneurysm; (■), Giant aneurysm.
); medium aneurysm; (■), Giant aneurysm.Relationship between CAAs and IVIG treatment
Most of the iKD patients did not present with aCAAs or cCAAs, regardless of whether they received IVIG treatment (Table 2). The incidence of aCAAs was significantly higher in patients treated with IVIG than in those who were not treated with IVIG (P < 0.0001). The incidence of cCAAs was also higher in patients who were treated with IVIG than in those who were not treated with IVIG, but this difference was not statistically significant (P = 0.134).
| Survey | Acute phase | Convalescent phase | ||
|---|---|---|---|---|
| 23rd | 24th | 23rd | 24th | |
| Total patients | 6,797 (100%) | 7,006† (100%) | 6,797 (100%) | 7,009† (100%) |
| With IVIG | 5,304 (78.0%) | 5,668 (80.9%) | 5,304 (78.0%) | 5,668 (80.9%) |
| CAA (−) | 4,431 (65.2%) | 4.889 (69.8%) | 5,142 (75.7%) | 5.521 (78.8%) |
| CAA (+) | 873 (12.8%) | 779 (11.1%) | 162 (2.4%) | 147 (2.1%) |
| Without IVIG | 1,493 (22.0%) | 1,338† (19.1%) | 1,493 (22.0%) | 1,341† (19.1%) |
| CAA (−) | 1,365 (20.1%) | 1,234 (17.6%) | 1,457 (21.4%) | 1,312 (18.7%) |
| CAA (+) | 128 (1.9%) | 104 (1.5%) | 36 (0.5%) | 29 (0.4%) |
- † Missing data caused the differences; CAA, coronary artery abnormality; IVIG, intravenous immunoglobulin.
Comparison between previous and recent surveys
Sonobe et al .5 reported the incidences of CAAs among patients with KD using data from the 17th nationwide survey in Japan. Sudo et al .6 also reported this information using data from the 20th survey. We compared the incidence of CAAs in the previous two studies and those in the recent surveys (the 23rd and 24th surveys). Between the 17th and 20th surveys, the incidences of CAAs in iKD and cKD patients decreased in both the acute and convalescent phases (Table 3). Further decreased incidence of CAAs was observed in the 23rd and 24th surveys (Table 3). Across all surveys, the incidence of both acute and convalescent phase CAAs was higher in iKD patients than in cKD patients. However, the difference in the incidence of aCAAs and cCAAs between groups was smaller in the 23rd and 24th surveys than that in the 17th or 20th survey.
| 17th (%) | 20th (%) | 23rd (%) | 24th (%) | |
|---|---|---|---|---|
| Acute phase | ||||
| Incomplete KD | 18.4 | 13.1 | 10.3 | 9.2 |
| Complete KD | 14.2 | 8.8 | 6.8 | 5.8 |
| Convalescent phase | ||||
| Incomplete KD | 5.9 | 3.5 | 3.0 | 2.7 |
| Complete KD | 4.4 | 2.5 | 2.4 | 1.9 |
- KD, Kawasaki disease.
Intravenous immunoglobulin treatment was administered to 92% of cKD patients in the study by Sonobe et al .5 and to 93% of cKD patients in the study by Sudo et al .6 (Fig. 3). The proportion of cKD patients treated with IVIG increased to 97% in both the 23rd and 24th surveys. Intravenous immunoglobulin treatment was administered to 62% of iKD patients in the study by Sonobe et al .5 and to 64% of iKD patients in the study by Sudo et al .6 The proportion of iKD patients treated with IVIG increased to 78% and 81% in the 23rd and 24th surveys,7 respectively. The difference in the proportion of cKD and iKD patients treated with IVIG decreased over time, and was 30%, 29%, 19%, and 16% in the 17th 20th, 23rd 24th surveys, respectively.
The proportion of patients with Kawasaki disease treated with intravenous immunoglobulin treatment: 92%, 93%, 97%, and 97% of patients with complete Kawasaki disease (KD) treated with intravenous immunoglobulin (IVIG) in the 17th, 20th, 23rd, and 24th nationwide surveys in Japan, respectively; 62%, 64%, 78%, and 81% of patients with incomplete KD were treated with IVIG in the 17th, 20th, 23rd, and 24th surveys, respectively. IVIG, intravenous immunoglobulin; cKD, patients with complete Kawasaki disease; iKD, patients with incomplete Kawasaki disease. (
), cKD; (
), iKD.
), cKD; (), iKD.
In cKD patients, early treatment was administered to 28% and 27% in the 17th and 20th surveys, respectively; the proportions of patients who received early treatment increased to 36% and 37% in the 23rd and 24th surveys, respectively (Fig. 4). In contrast, the proportions of patients who received late treatment were 7% and 5% in the 17th and 20th surveys, respectively, and the proportion of patients who received late treatment decreased to 4% in both the 23rd and 24th surveys. An increasing trend in the proportion of patients receiving early treatment and a decreasing trend in the proportion of patients with late treatment were also observed among iKD patients; the proportion of patients receiving early treatment was 26% in both previous surveys and this increased to 29% and 30% in the 23rd and 24th surveys, respectively (Fig. 4). The proportions of patients receiving late treatment were 16% and 15% in the 17th and 20th surveys, respectively; these decreased to 12% and 13% in the 23rd and 24th surveys, respectively. Across all the surveys, the proportions of iKD patients receiving early treatment was smaller than that of cKD patients receiving early treatment, and the proportion of patients receiving late treatment was larger than that of cKD patients receiving late treatment.
The proportion of patients with Kawasaki disease who received early, regular, or late immunoglobulin treatment. The proportion of patients receiving early treatment increased from previous surveys (the 17th and 20th nationwide surveys) to recent surveys (23rd and 24th surveys) among both cKD and iKD patients. On the other hand, the proportion of patients receiving late treatment decreased in both cKD and iKD patients. Across all surveys, the proportion of iKD patients receiving early treatment was smaller than that of cKD patients receiving early treatment and the proportion of iKD patients receiving late treatment was larger than that of cKD patients receiving late treatment. Early treatment was intravenous immunoglobulin (IVIG) treatment initiated before the fifth day of illness; regular treatment was IVIG treatment initiated between the fifth and seventh day of illness; late treatment was IVIG treatment initiated after the seventh day of illness. cKD, complete Kawasaki disease; iKD, incomplete Kawasaki disease. (
), with late treatments.
), with late treatments.Discussion
Based on a recent survey, the present study shows that the prevalence of iKD has increased from 16% in 2001–2002 to 22% in 2013–2016, and that the incidence of CAAs remains higher in iKD patients than in cKD patients. The incidence of CAAs in both iKD and cKD patients has decreased; and the difference in the incidence of CAAs between patients with iKD and cKD has also decreased. According to the survey conducted from 2013 to 2016, IVIG treatment was administered to 80% of the iKD patients, which is higher than that reported in previous surveys. Early IVIG treatment has increased and late treatment has decreased, according to the recent surveys. These changes in treatment have reduced the incidence of CAAs among iKD patients. The difference in the incidence of CAAs between patients with iKD and cKD has also decreased. These changes have reduced the incidence of CAAs in Japanese KD patients, despite an increase in the prevalence of iKD.
A diagnosis of iKD is largely related to the existence of CAAs. Some iKD patients are finally diagnosed by the detection of CAA. Without CAA detection, some iKD patients might not be diagnosed, and we do not know the exact number of such undiagnosed patients. It is therefore reasonable for iKD patients with few symptoms to show a higher incidence of cCAAs. The present study shows that the patients with one or two symptoms (6% of the iKD patients) showed a significantly higher incidence of CAAs than the patients with higher number of symptoms. They also had CAAs that were significantly more severe. It has been established that echocardiography is an important component of the diagnostic process for iKD.2 Coronary arterial dilatations are well known findings of echocardiography in KD. Other echocardiographic findings of mild ectasia of coronary arteries, or perivascular brightness of coronary arteries, decreased left ventricular function, mitral regurgitation, or pericardial effusion are also suggestive features for considering KD.2 It is therefore imperative to perform echocardiography in the patients who present with only one or two principal symptoms, taking into consideration that iKD is one of the differential diagnoses.
The present study showed that the highest proportion of patients treated with IVIG were iKD patients with only one symptom. In patients with prolonged fever, even in patients with no other principal symptoms, echocardiography is usually performed in the process of differential diagnosis. When CAAs or other suggestive echocardiographic features for considering KD are observed in such patients, the diagnosis of iKD and treatment with IVIG should be discussed. Among them, patients who are treated with IVIG are reported in the nationwide survey as iKD patients, and patients who are not treated are not reported. These situations resulted in a high proportion of iKD patients with one symptom who were treated with IVIG. The proportion of patients with one symptom treated with IVIG in the 23rd survey was higher than in the 24th survey. However, we do not know whether this difference was significant because the proportion of iKD patients with one symptom was very small (1% of iKD patients); thus, the proportion of patients treated with IVIG might be easily changed.
Sonobe et al .5 and Sudo et al. 6 have reported a higher incidence of CAAs among iKD patients compared with cKD patients. Although the present study has shown that the incidence of CAAs has decreased among both iKD and cKD patients, the occurrence of CAAs remained higher in iKD patients in both the acute and convalescent phases. Compared to the results of previous studies, the incidence of CAAs and the difference in the incidence of CAAs between iKD and cKD patients have decreased. The proportion of iKD patients who received IVIG has increased, and most of iKD patients had been treated with IVIG regardless of whether they had CAAs or not. Early treatment was administered to 29–30% of the iKD patients. This indicates that aggressive treatment was given to iKD patients – even those with few principal symptoms. Such aggressive treatment may be responsible for the decreased incidence of CAAs in iKD patients. Delayed diagnosis and later treatment have been considered as potential causes of CAAs. Sudo et al .6 have reported that hospital attendance after 7 days or more of illness was an independent risk factor for developing CAAs. The proportion of iKD patients who received late treatment in the recent survey was lower than that in the previous studies, and this may explain the reduction in the incidence of CAAs in patients with iKD. Timely and adequate diagnoses are necessary for patients to receive adequate treatments, and to avoid late treatment. It is therefore important to perform echocardiography in patients with a few principal symptoms to confirm the diagnosis of iKD.
Limitations
The present study used data from nationwide surveys, and, therefore, has some limitations. Data on the severity of each CAA were categorical (i.e. small, medium, giant). The doctors evaluated the category of CAA according to their own definition, and they did not report the exact size of the CAA in mm. Further, data on the precise day of illness on which the CAAs had been evaluated by echocardiography were not obtained.
Conclusions
The incidence of CAAs remains higher among iKD patients than in cKD patients but this difference has been reduced by the increased proportion of iKD patients treated with IVIG and those at an earlier time point. Decreasing trends in the incidence of CAAs not only in cKD but also iKD are the reason for decreased incidence of CAAs in Japanese KD patients despite an increase in the prevalence of iKD. To further reduce the incidence of CAAs, timely and adequate diagnoses and adequate treatments in iKD patients are necessary. It is therefore, important to recognize the possibility that patients may have iKD and perform echocardiography even in those who present with a few principal symptoms.
No comments:
Post a Comment