Kawasaki disease in infants below 6 months: a clinical conundrum?
Abstract
Aim
Kawasaki disease (KD) is a medium vessel vasculitis of childhood. In infancy KD is often characterized by incomplete and atypical forms. There is paucity of literature on KD in children below 6 months and there are no data from any developing country. This study defines the profile of children with KD below 6 months at our centre.
Methods
During January 1994 to March 2015, 460 children were diagnosed with KD and 17 (3.6%) were below 6 months. Diagnosis was based on American Heart Association (AHA) criteria. All children were treated with intravenous immunoglobulin and aspirin; three also received infliximab.
Results
Mucosal changes were present in 11 patients (64%); extremity changes in 11 (64%); rash in nine (53%); conjunctival injection in eight (47%); and cervical lymphadenopathy in three (17%). Irritability was noted in 15 patients (88%); four (23%) had respiratory symptoms; and two (11%) had bacille Calmette–Guérin scar reactivation. Fifteen (88%) had incomplete KD. Twelve patients were diagnosed beyond day 10 of illness. Thrombocytopenia was seen in four. Coronary artery abnormalities were present in six (35%) patients. Two children died from disease-related complications – one of these had giant coronary artery aneurysms.
Conclusion
Our data show that incomplete forms of KD are commonly seen in children below 6 months of age, thereby resulting in delayed diagnoses. Pediatricians need to have a high index of suspicion of KD when dealing with a young infant with unexplained fever beyond 5 days. The AHA criteria appear to be inadequate for diagnosing KD in infants below 6 months.
Introduction
Kawasaki disease (KD) is an acute medium vessel vasculitis of childhood that has a special predilection for the coronary arteries.1, 2 It is one of the most common forms of pediatric vasculitis and the commonest cause of myocardial infarction in children. Current annual incidence rates of KD in Japan, Korea and Taiwan are 264.8, 134.4 and 69 cases per 100 000 children below 5 years, respectively.3-5 In Chandigarh, the estimated incidence is 4.54/100 000 children below 15 years.6
Data from Japan, North America and Europe show that 80% of affected children are below 5 years.1-3 However, we have reported that almost one-third of children with KD at Chandigarh are above 5 years.6, 7
Diagnosis of KD is usually based on the American Heart Association (AHA) guidelines which are widely used.1, 2 KD can result in coronary artery abnormalities (CAA) in up to 25% of untreated patients and 3–5% of treated cases. The long-term consequences of KD in childhood are still obscure.1, 2
It is known that infants with KD tend to have a more severe disease and have a greater predilection for CAA.1, 2, 8-10 Approximately 35% of infants can have incomplete KD, thereby resulting in delayed clinical recognition and treatment.1, 2, 8-10
While several authors have described clinical presentations of KD in infancy,8-10 there is paucity of literature on KD below the age of 6 months and no information is available on this aspect of KD from any developing country. We report the clinical profile and management of 17 children with KD aged below 6 months which we had the occasion to manage in the Pediatric Allergy-Immunology Unit, Advanced Pediatrics Centre, Post Graduate Institute of Medical Education and Research, Chandigarh, India. To the best of our knowledge there are no data on this aspect of KD from any developing country.
Patients and Methods
We diagnosed 460 children with KD during the period January 1994 to March 2015. Diagnosis of KD was based on the accepted standard criteria.1, 6 Our institute is a federally funded not-for-profit referral centre that provides tertiary care to north-west India. Seventeen children in the cohort were below the age of 6 months.
Treatment of KD was based on standard guidelines and included intravenous immunoglobulin (IVIg) and aspirin (30–50 mg/kg/day in the febrile phase followed by 3–5 mg/kg/day during convalescence). IVIg was administered as 0.4 g/kg/day for 4 days during the period 1994–1998; thereafter we have used a single bolus of 2 g/kg. All children were subjected to 2-D echocardiography examination – once during the acute phase and subsequently on follow-up 6–8 weeks later. This study has been approved by the Departmental Publication Review Board.
Details of the demographic profile, clinical presentation, laboratory parameters and echocardiography findings were obtained from the records.
Results
Demographic profile
Mean age of presentation was 4.1 months (range: 1.5–6 months). There were 11 male and 6 female patients (male : female ratio 1.8 : 1.0). The youngest baby was a 40 days old male infant. In this cohort, six patients came from the state of Punjab, four from Himachal Pradesh, two each from Chandigarh and Haryana and one each from Madhya Pradesh, Delhi and West Bengal.
Clinical features
Mucosal changes were present in 11 patients (64%); extremity changes in 11 (64%); rash in nine (53%); conjunctival injection in eight (47%) and lymphadenopathy in three (17%). Fifteen of the 17 children (88%) had incomplete KD. Additional clinical features recorded included: irritability in 15 (88%); respiratory complaints in four (24%); reactivation of the bacille Calmette–Guérin (BCG) site in two (11%); diarrhea in two (11%); seizures in one (5.9%); and scrotal gangrene in one (5.9%). In 12/17 patients the diagnosis of KD was made after day 10 of illness. In two of thes the diagnosis was delayed till day 25 (Table 1).
| Clinical features | |
|---|---|
| Age | Mean – 4.1 months (range 1.5–6.0) |
| Male : female ratio | 1.8 : 1 (11 : 6) |
| Fever > 5 days | 15/17 |
| Mucosal changes | 11/17 (64%) |
| Conjunctival injection | 8/17 (47%) |
| Rash | 9/17 (53%) |
| Desquamation | 11/17 (64%) |
| Lymphadenopathy | 3/17 (17%) |
| Irritability | 15/17 (88%) |
| Cough | 5/17 (29%) |
| Bacille Calmette–Guérin scar reactivation | 2/17 (12%) |
| Shock | 2/17 (12%) |
| Diarrhea | 2/17 (12%) |
| Seizures | 1/17 (6%) |
| Delay in diagnosis (> 10 days) | 12/17 (70%) |
| Laboratory parameters | |
|---|---|
| Parameters | Mean (range) |
| C-reactive protein, mg/L | 116.2 (6.6–273) |
| Erythrocyte sedimentation rate, mm, Westergren | 37.9 (16–64) |
| Platelet count, 109/L | 710 (14 to 1200) |
| Total leukocyte count, 109/L | 18.8 (5.5–33.4) |
| Alanine aminotransferase, U/L | 98.7 (9–299) |
| Aspartate aminotransferase, U/L | 94.2 (9–245) |
| Albumin, g % | 2.53 (1.3–4.2) |
Laboratory features
Of the 13 patients in whom C-reactive protein (CRP) was estimated, 12 had raised levels with a mean of 116.2 mg/L (normal < 6). Mean platelet count was 710 × 109/L (normal 150–400). Thrombocytopenia was seen in four patients with the lowest count being 9.4 × 109/L. Mean total leukocyte count was 18.8 × 109/L, mean alanine transaminase level was 98.7 U/L and mean aspartate transaminase was 94.2 U/L. Coronary artery abnormalities (CAA) on 2D-echocardiography were detected in six (35%) patients. Ectasia and dilatation of the left main coronary artery was the most common abnormality detected (Table 1).
Treatment
All 17 patients received IVIg at a dose of 2 g/kg. Three children had resistant KD and required additional treatment with intravenous infliximab (5 mg/kg). Two children died from disease-related complications during the acute phase – one of these had giant coronary artery aneurysms involving left main and right coronary arteries.
Discussion
The diagnosis of KD in infancy can be very challenging for the pediatrician because of unusual clinical presentations.8-10 Children with KD in infancy often have incomplete forms of the disease with paucity of clinical signs. Atypical presentations, as for instance with seizures and nephritis, are also not unusual in this age group. It is known that infants with KD are more likely to have CAA than older children,1, 2, 8-10 Whether this increased rate of coronary complications is solely because of delayed diagnoses, and consequent delayed treatment, is not clear.
There is paucity of literature on the clinical presentation of KD in early infancy and no data have emanated from developing countries so far.11-14 Among the classical criteria, mucosal changes, rash and extremity changes were more frequently seen, whereas lymphadenopathy was seen in only 17% (3/17) of cases. Although lymphadenopathy is said to be present in 50% of children with KD, it is the least common of all manifestations included in classical criteria. Reactivation of the BCG scar in KD patients is presumedly due to the cross-reactivity of T cells between specific epitopes of mycobacterial and human heat shock protein.15 Although considered to be a specific and early manifestation, reactivation of the BCG scar is rarely seen in older children. In our cohort it was seen in 2/17 (11%) of patients.
Table 2 lists the previous studies on the subject.8, 11-14 Of 120 children with KD reported by Chang et al.,11 17% were below 6 months. Incomplete KD was noted in 35% of children below 6 months, while it was noted only in 12% of the rest. In our cohort, 3.6% of children with KD were below 6 months. An overwhelming majority (88%) had incomplete KD. As the clinical features evolve over a period of time, it is entirely possible some of these findings may have disappeared by the time the parents sought medical attention at our institute. As a result, what appears to be incomplete KD at a given point of time may not actually be so.
| Study | Country | No. patients | Important conclusions |
|---|---|---|---|
| Burns et al. (1986)12 | USA | Unselected series of eight children |
CAA: 6/8 (75%) Mortality: 2/8 (25%) |
| Rosenfield et al. (1995)8 | USA |
Total: 443 Age ≤ 6 months: 14 (3%) |
CAA: 79% < 6 months; 44% 6–12 months |
| Yanagawa et al. (1998)14 | Japan |
Total: 12 531 Age ≤ 6 months: 1402 (11.2%) |
CAA: 18.5% vs. 12.1% Giant aneurysms: 1.6% vs. 0.8% |
| Chang et al. (2006)11 | Taiwan |
Total: 120 Age ≤ 6 months: 20 (17%) |
Incomplete KD: 35% vs. 12% CAA: 65% vs. 19% |
| Park et al. (2008)13 | Korea |
Total: 22 674 Age ≤ 6 months: 1739 (7.7%) |
CAA: 21% vs. 18.7% |
| Present study (2015) | India |
Total: 460 Age ≤ 6 months: 17 (3.6%) |
CAA: 35% Incomplete KD: 88% Mortality: 2/17 (11.8%) |
- CAA, coronary artery abnormalities.
In 12/17 patients (70.6%) the diagnosis of KD was made after 10 days – the cut-off period generally considered most appropriate for administration of IVIg. In two among these, the delay was up to 25 days. Both these children developed CAA. One had giant aneurysms involving the left main and right coronary arteries and he later succumbed to the disease. Children below 6 months are more often diagnosed late.16 Thus it cannot be overemphasized that pediatricians need to have a high index of suspicion of KD whenever they see young infants with unexplained fever for more than 5 days, and especially when there has been no response to antimicrobials. The sterile pyuria so characteristic of KD can easily be, and in our experience is commonly, confused with a urinary tract infection.
CRP levels in the serum are useful not only in the initial assessment of a child with KD but also for follow-up.1, 2 CRP levels were found to be elevated in 12/13 patients in our cohort in which these could be assayed. Thrombocytopenia in the first week of illness is often considered a poor prognostic marker in KD and may presage the development of macrophage activation syndrome.17 Low platelet counts were seen in four of our patients and one of them developed scrotal gangrene. However, none of these patients developed CAA.
CAA were noted in as many as 6/17 children (35%) in our cohort. This is an alarming figure considering the long-term consequences of such abnormalities. It has been documented that infants have much higher incidence of CAA compared to older children.18 Delays in diagnosis, with consequent delays in treatment, could have been responsible for development of these complications.1 The youngest child in our cohort was a 40 days old male infant who developed KD at 40 days of life and had developed an aneurysm in the left main coronary artery. He is currently 6 months old and remains well on follow-up. Two among the 17 children (11.8%) in this cohort died during the acute stage.19
It is obvious that the clinical diagnosis of KD in infants below 6 months can be very challenging as the AHA guidelines may not be fulfilled in the majority (88% in our cohort) of cases.2 We have further shown that this subset of infants with KD is associated with an unusually high incidence of complications. In our opinion, it is high time we had a separate set of diagnostic guidelines for this age group so that diagnostic delays can be minimized and prompt treatment initiated.
To conclude, KD in children below 6 months seems to have significant differences in clinical presentation when compared with older children. Further, it appears to run a more aggressive clinical course with a significant risk of development of CAA even with IVIg treatment. In young infants with unexplained fever lasting more than 5 days, a clinical possibility of KD must be considered and appropriate investigations ordered. Delays in diagnosis and treatment can have catastrophic consequences. The AHA guidelines, in the present form, do not appear to be adequate for diagnosing KD in infants below the age of 6 months and need to be revisited.
Funding
None.
Disclosures
None.
