The Clinical Profile of Kawasaki Disease in Algerian Children: A Single Institution Experience

Abstract

Kawasaki disease (KD) in an acute vasculitis of unknown etiology. The epidemiological data available for Algerian patients remains insufficient.

Objective: To describe the demographic, clinical features of children with KD and to identify the risk factors for developing coronary artery lesions (CAL).

Methods: This retrospective study included children admitted with KD at the pediatric hospital in Algiers from January 2005 to December 2014.

Results: One hundred thirty-three patients (82 boys and 51 girls) with a mean age of 31 months were identified. The most common sign was fever, rash, oral changes and conjunctivitis. The cardiac complications were CAL (22.5%), pericarditis (2%) and myocarditis (1.5%). The independent variable for prediction of CAL was duration of fever >10 days, male gender and platelet count >450 000/mm³.

Conclusion: The incidence of cardiovascular complications is high. Knowledge of KD among Algerian pediatricians should be enhanced to guarantee appropriate treatment of this disease.

INTRODUCTION

Kawasaki disease (KD), also known as the infantile acute febrile mucocutaneous lymph node syndrome, was originally described as a distinct clinical entity in Japanese children by Dr Tomisaku Kawasaki in 1967 and was first mentioned in the English literature in 1974 [1, 2]. KD is now recognized as the leading cause of acquired heart disease among children in developed countries. Up until now, KD has only been reported in Algeria as anecdotal cases or brief reports [3], leaving its incidence unknown and causing cases to go undiagnosed and untreated. The aim of this study was to identify the number of cases of KD among Algerian children in an urban setting, and describing the epidemiological features and consequences of this disease.

MATERIALS AND METHODS

Following institutional ethics committee approval, medical records of 133 patients with KD hospitalized at the pediatric unit of Birtraria hospital, Algiers, between January 2005 and December 2014 were retrospectively reviewed.

The Birtraria hospital is a public tertiary care hospital with a pediatric department. Our institute serves as a tertiary care referral hospital for the city of Algiers (800 000 children <15 years). Patients who fulfilled the classical clinical diagnostic criteria for KD established by the American Heart Association were enrolled in our study [4].

Complete KD is defined by fever lasting 5 days and more than or equal to four of the five principal clinical features (bilateral conjunctival injection, cervical lymphadenopathy, polymorphous skin rash, changes in the lips or oral mucosa and changes in the distal extremities). Cases with incomplete diagnostic criteria, sometimes referred to as 'atypical' cases, were defined as such when the fever lasted ≥5 days at diagnosis and fewer than four criteria in the absence of other etiology for the symptoms. Laboratory tests, including complete blood counts, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels were assessed as potential risk factors for coronary complications along with age, sex and number of days of fever at the time of the diagnosis.

The cardiac complications of KD were assessed with echocardiography. This included the presence and degree of valvular regurgitation, myocarditis and pericardial effusion determined, as well as the presence and type of coronary artery lesions (CAL).

Definitions of CAL were based on the following criteria: (i) diffuse ectasia or dilation (diameter greater than that expected for the body surface area), (ii) coronary aneurysm (segmental dilation >1.5 times larger than the adjacent segment), considered small (<5 mm), medium-sized (5–8 mm) or giant (≥8 mm), and (iii) coronary stenosis. A repeat echocardiography was performed following the initial study, at 4–6 weeks from the onset of symptoms, every 6 months in those who had coronary aneurysms, and 1 year after onset in patients with no early cardiac findings.

Statistical analyses

Qualitative data are presented as frequencies with percentages and quantitative data as means with standard deviations (SD). Logistic regression analysis was performed to select factors significantly associated with cardiovascular sequelae. The odds ratio (OR) and the 95th confidence intervals were calculated to evaluate the variables analyzed as probable risk factors to develop coronary disease. A p  <  0.05 was considered statistically significant. Data were analyzed using SPSS 19.0.

RESULTS

During the 10 year study period, 133 children referred with KD fulfilled the diagnostic criteria, with 1.6 male to female ratio. Among those, 131 (98.4%) were diagnosed with complete and 2 (0.015%) with incomplete KD. The age at diagnosis ranged between 5 and 132 months (median 31 months). The vast majority, 90% (120 of 133), of the patients were <5 years old, and 18 of them were <1 year old at the time of diagnosis. Half of the cases, 54.1% (72 of 133), presented in winter and spring, i.e. between December and May inclusively.

Diagnosis was established within the first 10 days of onset of fever in 71 of 133 (53%) cases, and mean duration of fever was 13 ± 6 days. Alternative diagnostic hypotheses entertained in most of the cases before referral to our center included bacterial infections, viral infections and other rheumatologic diseases (Table1). The presenting clinical criteria for KD are summarized in Table 2. Other non-cardinal symptoms reported (Table 3) involved the gastrointestinal system (vomiting, diarrhea, abdominal pain, pancreatitis, gallbladder hydrops), the genitourinary tract (sterile pyuria), the musculoskeletal system (arthritis, arthralgia), the central nervous system (aseptic meningitis, facial paralysis) and the skin [perineal desquamation and Bacille Calmette-Guérin (BCG) reactivation]. Laboratory investigations (Table 4) included leukocytosis in 82 cases (61%), thrombocytosis with platelet count above 500 × 10³/mm³ in 53% cases (69 of 130) and increased sedimentation rate (≥40 mm/h) and CRP (≥30 mg/l) in 78% and 67%, respectively.

Table 1.

Patient diagnoses at first medical visit

Diagnosis	n (%)
Bacterial infections	107 (80)
Viral syndrome	9 (6.7)
Allergic reaction	5 (3.7)
Rheumatic fever	2 (1.5)
Idiopathic juvenile arthritis	1 (0.7)
KD	9 (6.7)

Table 2.

Clinical characteristics of patients at diagnosis

Diagnosis criteria	n	%
Fever lasting at least 5 days	133	100
Changes in extremities	128	97
Oral mucosal changes	130	98
Rash	130	98
Conjunctivitis	121	91
Cervical lymph node	38	28.50

Table 3.

Non-cardinal symptoms

Symptoms	n	(%)
Gastrointestinal
Abdominal pain	30	22.5
Vomiting	28	21
Diarrhea	20	15
Gallbladder Hydrops	5	3.7
Pancreatitis	1	0.7
Genitourinary
Sterile pyuria	9	6.7
Musculoskeletal
Arthritis	10	7.5
Arthralgia	32	24
Central nervous
Aseptic meningitis	35/87 lumbar Punctures	26
Facial paralysis	1	0.7
Skin	111
Perineal eruption	2	83
BCG reactivation		1.5

Table 4.

Laboratory values of patients at diagnosis

Variables	Mean value  ±  SD	Range
Leukocyte × 103 (/mm3)	17.81 ± 6.45	6.5–40.7
Platelet × 103 (/mm3)	543.79 ± 213.44	104–1431
ESR (mm/1st h)	85 ± 30.6	20–150
CRP (mg/l)	74.50 ± 43.24	6–192

SD = standard deviation.

Of the 133 patients, 32 (24%) had evidence of cardiac complications, such as coronary artery abnormalities (CAA) in 30 (22.5%), mitral regurgitation in 1, myocarditis in 2 and pericardial effusion in 3. Of those with CAA, 11 (8%) had coronary dilatation and 19 (14%) had coronary aneurysms (13 with small to medium, 6 with giant aneurysms). These abnormalities had regressed in 16 of 30 cases during the follow-up period. Coronary thrombosis occurred in one case with giant aneurysm 6 months into follow-up and stenosis in 2 cases with giant aneurysms 2 years into follow-up requiring surgical revascularization in one patient. No fatal outcomes were recorded however.

Only 88 of 133 (66.1%) patients received intravenous immunoglobulin (IVIG) therapy and 80 mg/kg/day of aspirin at admission. The IVIG posology was 1 g/kg once in 24 cases (27%), 1 g/kg twice in 63 cases (71%) and 400 mg/kg × 5 days in a single case. Refractoriness to IVIG was the case in 8 of 88 (9.1%), 7 were treated with methylprednisolone and 1 with methotrexate. Of the 45 patients who did not receive IVIG, 38 patients (28.6% of the total cohort) were started on 80 mg/kg/day aspirin alone. The risk factors for coronary aneurysms based on single variable comparison was the duration of fever >10 days (OR = 5.132, 95% CI = 1.67–15.77; p = 0.002), sex male (OR = 3.103, 95% CI = 1.17–8.23; p = 0.019), platelet count  > 450 000/mm³ (OR = 3.28, 95% CI = 1.15–9.30; p = 0.02). A multivariate analysis incorporating these elements identified a significant association between acute CAA with duration of fever > 10 days and platelet count > 450 000/mm³.

In contrast, there was no statistically significant association between the frequency of coronary sequelae and age, leukocyte count, ESR or CRP (Table 5).

Table 5.

Risks of CAL among subjects with KD

Variables	Univariate			Multivariate
	OR	95% CI	p	OR	95% CI	p
<1 year of age at the time of presentation	0.393	0.137–1.125	0.075	0.379	0.078–1.840	0.229
Males	3.103	1.170–8.233	0.019	4.075	0.928–17.896	0.063
>10 days fever duration at the time of presentation	5.132	1.670–15.773	0.002	9.063	1.214–67.671	0.032
White blood cell count >15 000/mm3	0.88	0.373–2.077	0.77	0.205	0.045–0.927	0.04
Platelet count >450 000/mm3	3.28	1.157–9.30	0.02	9.56	1.084–84.329	0.042
ESR ≥40 mm 1st h	2.21	0.260–18.885	0.456	0.287	0.007–11.176	0.504
CRP ≥30 mg/l	1.69	0.453–6.336	0.429	2.23	0.296–16.804	0.436

Note. CAL = coronary artery lesions; OR = odds ratio; CI = confidence interval.

^ap value is estimated by logistic regression.

DISCUSSION

Nearly 50 years ago, KD was described as an acute, febrile multisystem vasculitis that almost exclusively affected young children. It is generally accepted that a ubiquitous infectious agent causes the disease in a genetically predisposed individual with no known direct cause to it [5–7]. Since then, KD has been described in almost every country replacing rheumatic fever as the most common cause of acquired heart disease in North America, Japan and Europe [8]. As of now, detailed information about this disease was still unavailable in Algeria and scarce in neighboring countries. In our series, KD occurred mainly in male patients and in those <5 years of age in our series similar to previous reports' results from other countries [8]. The rather long mean duration of fever (13 days) in our series demonstrates a marked delay in diagnosis of this disease in referral pediatric centers in Algeria. Most delays were owing to difficulties in distinguishing KD from bacterial and viral infections, which together constituted the initial diagnostic hypotheses in most cases. Besides the persistent fever, the most frequent clinical manifestation of KD was changes in the lips and the oral cavity, followed by changes in peripheral extremities and bilateral conjunctival congestion. Cervical lymphadenopathy was the least common symptom among our patients, which is in agreement with other studies [9]. The disturbing reality is, however, that cases with incomplete criteria for KD were diagnosed in only 0.01% of this series compared with a reported prevalence of such presentation between 15 and 36.2% [10, 11]. Missing the diagnosis of incomplete KD is one of the major risk factors for development of coronary artery disease [12]. On the other hand, CAA were detected in 22.5% of 133 cases during the acute stage in our series. In comparison, the prevalence of coronary artery dilatation, aneurysm and giant aneurysm within 30 days after KD onset were 0.72%, 1.8% and 0.18%, respectively, in a recent study from Japan between 2011 and 2012 [13].

Risk factors for coronary involvement detected in our study were male gender and high platelet count. Various risk factors for cardiac sequelae have been reported elsewhere including male gender [14], extreme ages [15], recurrent cases [16], late IVIG treatment [17], refractoriness to initial IVIG treatment and abnormal laboratory data such as CRP, ESR, anemia and platelet count [18]. From this perspective, the index of suspicion of an orphan disease such as KD relies on knowledge of the medical community in a specific population. In fact, despite current recommendations that all children with suspected KD receive IVIG [4], we are concerned that only 66% patients received this treatment. This is mainly owing to late presentation or missed diagnosis, which indicates that knowledge of KD among Algerian pediatricians should be enhanced to increase diagnostic sensitivity and guarantee appropriate treatment of this disease.

Finally, and despite the limitation of this study pertaining to being a single center retrospective cohort, atypical or incomplete KD diagnosis is likely underreported owing to failure to recognize and hospitalize such category of patients. Nevertheless, there is an important message for the benefit of the general public health in our country and other nations with similar issues.

CONCLUSIONS

The results of this study confirm that KD is present in Algeria. Larger epidemiologic studies at a national lever are essential to draw a more precise picture of its distribution and characteristics in Algeria. Nevertheless, the significantly delayed diagnosis in a teaching hospital raises a deficient awareness among our medical community thus far.