- Email: [email protected]
Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case control study
IJCA-25339; No of Pages 5 International Journal of Cardiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case control study Li-Chuan Sun a, Chih-Cheng Lai b, Cheng-Yi Wang c, Ya-Hui Wang d, Jen-Yu Wang c, Yo-Ling Hsu a, Yin-Lan Hu e, En-Ting Wu f, Ming-Tai Lin f, Leticia B. Sy a,⁎, Likwang Chen g,⁎⁎ a
Department of Pediatrics, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan Department of Intensive Care Medicine, Chi Mei Medical Center, Liouying, Tainan, Taiwan Department of Internal Medicine, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan d Medical Research Center, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan e Department of Dentistry, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan f Department of Pediatrics, National Taiwan University Hospital, National Taiwan University, Taipei, Taiwan g Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Miaoli County, Taiwan b c
a r t i c l e
i n f o
Article history: Received 2 May 2017 Received in revised form 18 July 2017 Accepted 4 August 2017 Available online xxxx Keywords: Infective endocarditis Congenital heart disease Children
a b s t r a c t Background: Infective endocarditis (IE) is uncommon in childhood. Its associated epidemiological characteristics in patients with congenital heart disease (CHD) remain unclear. Methods: The study population included children born in Taiwan during the years 1997 to 2005 who were diagnosed as having CHD before 3 years of age. All children were followed up until the end year of 2010, the diagnosis of infective endocarditis, or death. The demographic characteristics of patients with and without IE, the invasive procedures performed during 6 months before the index date, the prophylactic antibiotics related to dental procedures, and in-hospital mortality were collected. Results: Information of 24,729 children with CHD were retrieved for our analysis and 237 patients with newly diagnosed IE were identified. The incidence rate of IE in all CHD lesions was 11.13 per 10,000 person-years. Taking ASD for reference, the following CHD lesions were at risk for IE: cyanotic CHD (adjusted OR, 9.58; 95% confidence interval, 5.38–17.05), endocardial cushion defect (ECD) (8.01; 2.73–23.50), Left-sided lesions (4.36; 1.90–10.01) and VSD (2.93; 1.64–5.23). Patients who underwent procedures have a higher risk of acquiring IE which include central venous catheter (CVC) insertion (3.17; 2.36–4.27), cardiac catheterization (3.74; 2.67– 5.22), open-heart surgery (2.47; 1.61–3.77), valve surgery (3.20; 1.70–6.02), and shunt surgery (7.43; 2.36– 23.41). However, dental procedures did not increase the risk of IE, irrespective of antibiotic usage. Conclusions: The risk of IE varies markedly among CHD lesions in our study. Invasive heart procedures but not dental procedures, are more significantly associated with IE among children with CHD. © 2017 Published by Elsevier Ireland Ltd.
1. Introduction Infective endocarditis (IE) is uncommon in childhood but is a potentially life-threatening disease. The annual incidence of IE ranged from 0.05 to 0.12 cases per 1000 pediatric admissions [1], and the overall mortality rate is approximately 5% [2,3]. Approximately 35% to 60% of children with IE have congenital heart disease [2–5], which is known to be a predisposing factor for IE in children. The American Heart Association (AHA) guidelines for prevention of IE were revised in 2007 [6]. The prophylaxis for IE has been simplified and restricted to selected patients with predisposing cardiac conditions for whom antibiotic prophylaxis ⁎ Correspondence to: L.B. Sy, Cardinal Tien Hospital, Department of Pediatrics, Cardinal Tien Hospital, Fu Jen Catholic University, College of Medicine, New Taipei City 23148, Taiwan. ⁎⁎ Correspondence to: L. Chen, Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Miaoli County 35053, Taiwan. E-mail addresses: [email protected] (L.B. Sy), [email protected] (L. Chen).
was indicated before the invasive procedure [6,7]. However, concerns still exist about the safety and reliability of these new guidelines from cardiologists, dentists and patients [8–12], especially in developing countries whose populations have unsatisfactory oral hygiene [8,13]. Furthermore, many studies about factors related to IE focused mainly on adult populations, and little information exists in children with congenital heart disease [14]. The purpose of this study was to determine the risk factors for IE, such as dental procedure, open heart surgery, and types of congenital heart disease, in a large population-based cohort of children with IE. 2. Material and methods 2.1. Data source This study used the database constructed by the National Health Research Institutes (NHRI) of Taiwan. Taiwan's National Health Insurance
http://dx.doi.org/10.1016/j.ijcard.2017.08.009 0167-5273/© 2017 Published by Elsevier Ireland Ltd.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
2
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
(NHI) program was implemented in Taiwan since 1995 and 99% of the 23.75 million Taiwanese is under this medical coverage [15]. The NIH offers comprehensive medical coverage, including outpatient, inpatient, emergency, dental services, medical examinations, laboratory tests, drug prescriptions, and interventional procedures. The NHRI used original data from the NHI database to construct a longitudinal database of children with congenital heart disease. 2.2. Study population and control group The patients included in this study were selected from the healthcare records in the NHI database from 1997 to 2010. The study population included children born in Taiwan between year 1997 and 2005 who were diagnosed as having congenital heart disease (CHD) before 3 years of age. In the analysis, 24,729 cases were included and followed up until the end of year 2010, the diagnosis of IE, or death. The CHD was defined in accordance with the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) code. To avoid coding errors caused by incorrect tentative diagnosis, miscoding, and overestimation from insignificant defects (e.g., patent foramen ovale, spontaneous closure of small patent ductus arteriosus [PDA], spontaneous closure of small ventricular septal defect [VSD], and mild peripheral pulmonary artery stenosis of newborn), children with CHD were enrolled only when they had a CHD-specific admission or N3 outpatient clinic visits [16]. For patients with more than one CHD diagnoses, the more frequent and more complex CHD diagnoses were used for further analysis. Children with congenital heart disease who did not have IE during our study period were enrolled in the control cohort. We randomly selected 4725 patients from the control cohort, and matched with the IE patients for same age (1-year strata), sex, and date of hospitalization or outpatient clinic visits to serve as our control group. Twenty control patients were matched with one patient with IE. The IE study group was defined as those having a discharge diagnosis of IE in their first hospitalization, using ICD-9-CM codes (421.0, 421.1, 421.9, and 424.9). The date of hospital admission for IE served as the index date. The control subjects were assigned a date of medical record equal to the index date of their matched case subjects. We recorded all the invasive procedures performed during 6 months before the index date. We also analyzed prophylactic antibiotics related to dental procedures, defined as a prescription of antibiotics on the same day of the dental treatment. 2.3. Definition CHD lesions were classified into eight groups as follows: (1) cyanotic heart disease: Tetralogy of Fallot, single-ventricle heart defects, transposition of great arteries, double outlet of the right ventricle, truncus arteriosus, hypoplastic left heart syndrome, total anomalous pulmonary venous connection, right atrium isomerism (RAI), tricuspid atresia, agenesis of pulmonary artery, and aortic atresia; (2) endocardial cushion defect (ECD); (3) left-sided lesions: coarctation of the aorta, aortic insufficiency, mitral stenosis, and mitral insufficiency; (4) VSD; (5) right-sided lesions: Ebstein anomaly, anomalies of pulmonary artery or pulmonary valve; (6) other CHDs (other or unspecified congenital anomalies); (7) PDA; and (8) arterial septal defect (ASD). Invasive procedures included all open-heart surgeries, shunt operation, valve operation, cardiac catheterization, use of a central venous catheter (CVC), and dental procedures. Valve surgery was defined as surgical repair of cardiac valve that is including annuloplasty, single valve replacement, double valve replacement, three valve replacement, pulmonary valvuloplasty, repair fistula sinus of valsalva, closed mitral or open mitral commissurotomy and Rostalli operation. Percutaneous valve implantation was classified as cardiac catheterization. Dental procedures included tooth extraction, dental scaling, endodontics, tooth pulpotomy, canal enlargement and debridement, root canal therapy, and other operations on the teeth, gums, and alveoli.
2.4. Statistical analysis The baseline characteristics of the children with or without IE in our study cohort were described as absolute numbers and proportions. Overall and CHD lesion-specific incidence rates of IE were calculated as the number of patients with new-onset IE divided by the sum of follow-up person-years at risk for IE in the CHD cohort. The distribution of the risk factors in children with IE and their matched controls were described as absolute numbers and proportions. Odds ratios (ORs) of IE associated with independent variables with 95% confidence intervals (CIs) were computed by using conditional logistic regression that accounted for the matching in the study design. The following variables were assessed: invasive procedures such as CVC, cardiac catheterization, open heart surgery, valve surgery, and shunt surgery, 6 months before index date (yes/no); eight groups of CHD lesions (reference group: ASD); and dental procedures (no treatment, and dental procedure with or without antibiotics) 6 months before index date. All statistical analyses were conducted with the SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA). Statistical significance was defined as a two-sided P value of b 0.05.
3. Results 3.1. Distribution of CHD and incidence of IE Information of 24,729 children with congenital heart disease were retrieved for our analysis and contributed 231,027 person-years. Table 1 shows the demographic and clinical characteristics of the patients, including sex, age at diagnosis, and the distribution of CHD lesions. VSD (8048 cases; 32.54%) was the most common CHD lesions in our cohort, followed by ASD (5620 cases; 22.73%), PDA (3560 cases; 14.4%), and cyanotic CHD lesions (3702 cases; 14.97%). During our observation period, 237 patients with new onset IE were identified, of whom 54.1% were male, and 5.49% were prematures. The median age at IE diagnosis was 1.21 years (interquartile range: 0.59–3.05 years), and 42.19% of IE patients were younger than 1 year. The overall incidence rate of IE in all CHD lesions was 11.13 per 10,000 person-years. The incidence rates of IE per 10,000 children person-years in the major CHD groups were as follows: cyanotic CHD, 35.73; ECD, 27.24; Table 1 The baseline characteristics in CHD cohort (n = 24,729).
Gender Male Female Premature birth No Yes Age of IE diagnosis (year) Median (IQR) 0–1 1–2 2–3 N3 CHD lesions Cyanotic CHD ECD Left-sided lesions VSD Right-sided lesions Other CHD PDA ASD
Total (n = 24,729)
IE (n = 237)
Non-IE (n = 24,492)
No.
%
No.
%
%
%
12,218 12,511
49.41 50.59
128 109
54.01 45.99
12,090 12,402
49.36 50.64
22,883 1846
92.54 7.46
224 13
94.51 5.49
22,659 1833
92.52 7.48
1.21 (0.59–3.05) 100 42.19 56 23.63 18 7.56 63 26.58
– – – –
– – – –
88 5 10 74 5 36 5 14
3614 293 850 7972 1824 778 3555 5606
14.76 1.20 3.47 32.55 7.45 3.18 14.51 22.89
3702 298 860 8048 1829 814 3560 5620
14.97 1.21 3.48 32.54 7.40 3.29 14.40 22.73
37.13 2.11 4.22 31.22 2.11 15.19 2.11 5.91
CHD = congenital heart disease; ECD = endocardial cushion defect; VSD = ventricular septal defect; PDA = patent ductus arteriosus; ASD = atrial septal defect.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx Table 2 The CHD lesion group-specific incidences of infective endocarditis in the study cohort. CHD lesions
No. of IE (%)
Person-year
Incidence ratea
Cyanotic CHD ECD Left-sided lesions VSD Right-sided lesions Other CHD PDA ASD Overall
88 (37.13) 5 (2.11) 10 (4.22) 74 (31.22) 5 (2.11) 36 (15.19) 5 (2.11) 14 (5.9) 237
24,627.97 1835.21 7012.84 73,221.58 16,627.17 7067.1 32,953.83 49,681.47 213,027.17
35.73 27.24 14.26 10.11 3.01 50.94 1.52 2.82 11.13
3
with VSD, 1 patient with PDA, and 1 patient with other congenital heart diseases. In the logistic regression analysis, the in-hospital mortality of the IE patients with cyanotic heart disease was 3.63 times (95% CI, 1.06–12.41) higher than that of the IE patients with non-cyanotic heart disease (data not shown). 4. Discussion
CHD = congenital heart disease; ECD = endocardial cushion defect; VSD = ventricular septal defect; PDA = patent ductus arteriosus; ASD = atrial septal defect. a Per 10,000 person-years.
left-sided lesion, 14.26; VSD, 10.11; right-sided lesion, 3.01; ASD, 2.82; PDA, 1.52; and other CHDs, 50.94 (Table 2). 3.2. Risk of IE According to the previous AHA guideline for IE [19], isolated secundum ASD is consider as a negligible risk of IE, which means the risk of development of IE is not higher than general population. Thus, we take the ASD group as our reference to analyze the relationship between cardiac conditions and risk of infectious endocarditis. Taking ASD as reference, the odds ratio of lesion groups were estimated as follows: cyanotic CHD (adjusted OR, 9.58; 95% CI, 5.38–17.05), endocardial cushion defect (ECD; adjusted OR, 8.01; 95% CI, 2.73–23.50), leftsided lesions (adjusted OR, 4.36; 95% CI, 1.90–10.01), VSD (adjusted OR, 2.93; 95% CI, 1.64–5.23), right-sided lesions (adjusted OR, 1.01; 95% CI, 0.36–2.83), other CHDs (adjusted OR, 21.09; 95% CI, 10.93–40.69), and PDA (adjusted OR, 0.46; 95% CI, 0.17–1.30) (Table 3). 3.3. Invasive procedures, dental procedures, prematurity and IE The risk of IE from invasive and dental procedures are listed in Table 4. The patients who received procedures had a higher risk of IE, including all CVCs (adjusted OR, 3.17; 95% CI, 2.36–4.27), CVCs without surgery (adjusted OR, 2.25; 95% CI, 1.65–3.06), cardiac catheterization (adjusted OR, 3.74; 95% CI, 2.67–5.22), open heart surgery (adjusted OR, 2.47; 95% CI, 1.61–3.77), valve surgery (adjusted OR, 3.20; 95% CI, 1.70–6.02), and shunt surgery (adjusted OR, 7.43; 95% CI, 2.36–23.41). However, dental procedure did not increase the risk of IE, regardless of whether antibiotics were used or not (Table 4). In addition, we found that prematurity did not increase the risk of IE in children with CHD (adjusted OR, 0.69; 95% CI, 0.39–1.21) (Table 4). 3.4. Mortality In-hospital mortality of IE occurred in 12 (5%) of the 237 patients, including 8 patients with cyanotic congenital heart disease, 2 patients
From this large retrospective population-based cohort study of 24,729 children (213,027 person-years) with congenital heart disease, we reported several significant findings. First, the incidence rate of IE in our study was 11.13 cases per 10,000 person-years, higher than that in the previous study [17]. In Quebec, the incidence of IE in children with congenital heart disease was 4.1 cases per 10,000 person-years. This difference in incidence rate may be attributed to the difference in the study population included in the study and the fact that the risk of IE is strongly dependent on the wide variation in the types and severity of the underlying congenital cardiac condition. In our study population, we enrolled only those who had a CHD-specific admission or had N 3 outpatient clinic visits. Insignificant defects such as patent foramen ovale, spontaneous closure of small PDA and small VSD, and mild peripheral pulmonary artery stenosis of newborn were excluded in our study cohort. This can explain why the incidence of IE in our study was slightly higher than that in the previous study. Furthermore, VSD is the most common CHD in our cohort (8048 cases; 32.54%), followed by secundum ASD (5620 cases; 22.73%) and PDA (3560 cases; 14.40%). This CHD spectrum distribution is similar to those in previous studies in Taiwan [16,18] that reported the five most frequently diagnosed CHDs to be VSD, secundum ASD, PDA, pulmonary stenosis, and Tetralogy of Fallot. Second, the risk for IE in our study varies markedly among the types of CHD lesions. Aside from cyanotic heart disease, ECD, left-sided lesions and VSD also had an increased risk for developing IE. Our results are similar with previous pediatric studies [4,20–25] that concluded that congenital heart defects such as aortic valve abnormalities, VSD, and Tetralogy of Fallot were common underlying conditions in children with IE. Rushani et al. [17] reported that relative to ASD, the following lesions were most strongly associated with an elevated risk of IE: cyanotic CHD, ECD, and left-sided lesions. In their study [17], the incidence of IE in VSD patients were lower than our reported rates, this may be attributed in part to some insignificant defects such as spontaneous closure of small VSD for which it was not included in our study cohort. Another result in our study that is worth clarifying is that the incidence rate of IE in the other CHD groups was high (ICD-9:746.8; other unspecified anomalies of the heart). A possible explanation is that some of the specific complex heart diseases might be classified into the other CHD groups. Failure of parents to provide the correct complex heart disease diagnosis and the inability of the general pediatric practitioners to identify the right ICD-9 code for the specific heart disease might contribute to this misclassification. The in-hospital mortality rate in our study for CHD patients with IE
Table 3 Crude and adjusted odds ratio of infectious endocarditis among congenital heart diseases in children.
ASD Cyanotic CHD Endocardial cushion defect Left-sided lesions Ventricular septal defect Right-sided lesions Other CHD Patent ductus arteriosus
Case (n = 237)
Control (n = 4725)
Crude OR
No.
(%)
No.
(%)
(95% CI)
14 88 5 10 74 5 36 5
(5.91) (37.13) (2.11) (4.22) (31.22) (2.11) (15.19) (2.11)
951 635 38 162 1745 335 118 741
(20.13) (13.44) (0.80) (3.43) (36.93) (7.09) (2.50) (15.68)
Reference 9.63 (5.41, 17.14) 8.20 (2.80, 24.04) 4.34 (1.89, 9.96) 2.94 (1.64, 5.25) 1.01 (0.36, 2.82) 21.35 (11.07, 41.18) 0.47 (0.17, 1.30)
P value
Adjusteda OR
P value
(95% CI) b0.001 b0.001 b0.001 b0.001 0.991 b0.001 0.144
Reference 9.58 (5.38, 17.05) 8.01 (2.73, 23.50) 4.36 (1.90, 10.01) 2.93 (1.64, 5.23) 1.01 (0.36, 2.83) 21.09 (10.93, 40.69) 0.46 (0.17, 1.30)
b0.001 b0.001 b0.001 b0.001 0.986 b0.001 0.143
ASD = atrial septal defect; CHD = congenital heart disease. a Adjusted for age and gender.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
4
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Table 4 Invasive procedure dental procedure and prematurity in the study cohort. Control (n = 4725)
Crude OR
No.
(%)
No.
(%)
(95% CI)
Invasive procedure 6 months before index date CVC insertion 104 CVC insertion without OHS, VS or SS 69 Cardiac catheterization 54 Open heart surgery (OHS) 28 Valve surgery (VS) 15 Shunt surgery (SS) 4
(43.88) (29.11) (22.78) (11.81) (5.06) (1.69)
1050 762 369 245 75 11
(22.22) (16.13) (7.81) (5.19) (1.59) (0.23)
3.23 (2.41, 4.34) 2.29 (1.68, 3.12) 3.73 (2.67, 5.20) 2.52 (1.65, 3.85) 3.35 (1.79, 6.28) 7.27 (2.32, 22.84)
b0.001 b0.001 b0.001 b0.001 b0.001 b0.001
3.17 (2.36, 4.27) 2.25 (1.65, 3.06) 3.74 (2.67, 5.22) 2.47 (1.61, 3.77) 3.20 (1.70, 6.02) 7.43 (2.36, 23.41)
b0.001 b0.001 b0.001 b0.001 b0.001 b0.001
Dental Procedure 6 months before index date No dental procedure 224 Dental procedure with antibiotics 10 Dental procedure without antibiotics 3
(94.51) (4.22) (1.27)
4424 144 157
(93.63) (3.05) (3.32)
Reference 1.30 (0.64, 2.65) 0.35 (0.11, 1.17)
0.469 0.088
Reference 1.31 (0.64, 2.66) 0.35 (0.11, 1.17)
0.459 0.088
Premature birth No Yes
94.51 5.49
4355 370
92.17 7.83
Reference 0.69 (0.39, 1.21)
0.685
Reference 0.69 (0.39, 1.21)
0.686
224 13
P value
Adjusteda OR
Case (n = 237)
P value
(95% CI)
CVC = central venous catheter. a Adjusted for age and gender.
was 5% (12/237). Among the 88 IE patients with cyanotic heart disease, 8 patients (9%) died, while the mortality rate of IE patients with non-cyanotic heart disease stands at 2.6% (4/149). Base on the abovementioned data, IE patients with cyanotic heart disease appeared to have a threefold increased risk for inpatients mortality (3.63 times; 95% CI, 1.06–12.41). In summary, we found that children with cyanotic heart disease, ECD, VSD and left-sided heart lesions were at higher risk for developing IE and those IE patients with cyanotic heart disease appeared to be at higher risk for in-hospital mortality than those IE patients with non-cyanotic heart disease. Third, our study showed that invasive procedures, including CVC, cardiac catherization, open heart surgery, valve surgery, and shunt surgery within 6 months after the procedures were significantly associated with increased risk of infectious endocarditis in children with CHD. Previous studies showed that the increased IE risk after cardiac surgery and catheter-related cardiac interventions is mainly linked to implantation of prosthetic material or cardiovascular devices [21,23]. Surgery itself, including such elements as central vascular catheters, intravenous alimentation, and length of stay in the intensive care unit, may be important risk factors for the development of IE [21]. Of all the invasive procedures, shunt surgery was associated with the highest risk of developing IE within the first 6 months post procedure, having a sevenfold increase. This result was similar to the previous study done by Baltimore et al. (21) showing that approximately 50% of children with IE complicating CHD had a previous cardiac surgery, particularly palliative shunt procedures or complex intracardiac repairs. Overall, invasive procedures, including open heart surgery, valve surgery, shunt surgery, cardiac catheterization, and CVC were strongly associated with development of IE during the 6-month post-invasive procedure period. Elevated IE risk in early childhood has been reported in studies of children with CHD [17] which is probably multifactorial in etiology. Our data showed that most of the IE episode occurred in the 1st year of life (100/237; 42.19%). Almost all complex congenital heart patients underwent surgical repair of their CHD lesions early in life which may put these children at risk for IE. Fourth, this is the first large retrospective population-base cohort case-control study conducted to analyze the relationship between dental procedures and IE in children with congenital heart disease. The main finding of our study was that dental procedures are not significantly associated with infectious endocarditis in children with congenital heart disease regardless of antibiotic prophylaxis. In our study, 13 patients (5.49%) had received dental procedures within 6 months preceding their onset of IE. This is consistent with the observation in adult studies [26,27]. In a case-crossover design study
of 739 patients with IE by Chen et al. [26], dental procedures did not significantly contribute to the risk for IE. In another retrospective study from a large national registry of 677 patients by Chirllo et al., the prevalence of dental procedures as a presumed portal of entry of the germ responsible for IE was only 4.7% [28]. This may be explained by the observation of the changing bacterial spectrum causing IE in children is shifting from streptococci to Staphylococcus aureus [2,8,21, 29]. This trend is likely to reflect in changing risk factors for those who acquired IE, with more children having a hospitalization with medical issues predisposing to IE [27]. In addition, children have a much lower prevalence and severity of gingivitis and periodontitis than adults [4] and in general, dental health information is more aggressively provided to parents of children with CHD by pediatric cardiologist and dentist. Although antibiotic prophylaxis for IE has been recommended for persons with congenital heart disease since 1955 [30], so far no published data accurately determine the absolute risk of IE that resulted from a dental procedure nor there was any definite prove in human studies that prophylactic antibiotics treatment can prevent endocarditis [6]. Dental procedures did not increase the risk of IE, irrespective of antibiotic usage, and further study is still needed to confirm our findings. Based on our findings about the association between dental procedure and risk of IE among children with CHD, it should suggest that dental procedures are not important risk factors for IE in children with CHD, regardless of whether antibiotic prophylaxis was given. Because of the low incidence of IE in children, large number of patients per cohort is required to obtain statistical clinical significance when analyzing the risk factors for IE. Our study is a population-base nested case-control design with two major strengths, a large sample size obtained from the database and the ability to longitudinal followup patients to identify the risk factor for IE. Our study has some limitations: First, the NHIRD database contains little clinical information and therefore, we are unable to recognize the causative organism and the mortality of cases after hospital discharge. Furthermore, the diagnosis of IE cases was not based on Duke Criteria and incorrect diagnosis or miscoding may lead to misclassification in CHD. The enrollment of patients with IE relied on ICD-9-CM codes from hospital discharge records. This is in similarity to several large-scale studies designed to obtain absolute measures of IE frequency [17,26]. To avoid errors caused by incorrect tentative diagnosis, miscoding and overestimation from insignificant defects, children with CHD were enrolled only in our study when they have a CHDspecific admission or N 3 times of outpatient clinic visits [16]. Second, our study only included children born in Taiwan during year 1997 to 2005 who were diagnosed to have CHD before 3 years of age. According to Wu et al. report [18] that an upward trend was observed for age
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
specific prevalence of CHD that was maintained until the 3 years old, followed by a plateau at 3–10 years of age in Taiwan. Thus, our study enrolls almost all of children with CHD in Taiwan during the study period. Third, because our study aims to investigate the association between invasive cardiovascular procedures and dental procedures with antibiotic prophylaxis among IE patients in children with CHD, we did not collect other possible risk factors for further evaluation. Furthermore, we can only obtain between 1997 and 2010 from NHIRD due to National Health Research Institute's regulation about the size of data availability. However, our study population included CHD patients with age ranging from birth to age 5–14 and a follow up period of 14 years which is longer than previous studies [3,17,26,27]. Therefore, the findings in our study are still representative of IE risk among children with CHD. 5. Conclusion Our study revealed that the risk for developing IE varies markedly among the types of CHD lesions. Invasive heart procedures were more significantly associated with IE in children with CHDs while dental procedures were not an important risk factors for IE in children with CHD, regardless of whether antibiotic prophylaxis was given. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Sources of funding This study was supported by grants CTH 106A-2A12 (Cardinal Tien Hospital, Taiwan) and National Health Research Institutes (PH-105-SP-08). References [1] S.K. Pasquali, X. He, Z. Mohamad, B.W. Mccrindle, J.W. Newburger, J.S. Li, et al., Trends in endocarditis hospitalizations at US children's hospitals: impact of the 2007 American Heart Association Antibiotic Prophylaxis Guidelines, Am. Heart J. 163 (2012) 894–899. [2] M.D. Day, K. Gauvreau, S. Shulman, J.W. Newburger, Characteristics of children hospitalized with infective endocarditis, Circulation 119 (2009) 865–870. [3] A.L. Ware, L.Y. Tani, H.Y. Weng, J. Wilkes, S.C. Menon, Resource utilization and outcomes of infective endocarditis in children, J. Pediatr. 165 (2014) (807–812.e801). [4] L. Saiman, A. Prince, W.M. Gersony, Pediatric infective endocarditis in the modern era, J. Pediatr. 122 (1993) 847–853. [5] D. Marom, I. Levy, O. Gutwein, E. Birk, S. Ashkenazi, Healthcare-associated versus community-associated infective endocarditis in children, Pediatr. Infect. Dis. J. 30 (2011) 585–588. [6] W. Wilson, K.A. Taubert, M. Gewitz, P.B. Lockhart, L.M. Baddour, M. Levison, et al., Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group, Circulation 116 (2007) 1736–1754.
5
[7] G. Habib, B. Hoen, P. Tornos, F. Thuny, B. Prendergast, I. Vilacosta, et al., Guidelines on the prevention, diagnosis, and treatment of infective endocarditis (new version 2009): the Task Force on the Prevention, Diagnosis, and Treatment of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and the International Society of Chemotherapy (ISC) for Infection and Cancer, Eur. Heart J. 30 (2009) 2369–2413. [8] W.C. Tseng, S.N. Chiu, P.L. Shao, J.K. Wang, C.A. Chen, M.T. Lin, et al., Changing spectrum of infective endocarditis in children: a 30 years experiences from a tertiary care center in Taiwan, Pediatr. Infect. Dis. J. 33 (2014) 467–471. [9] J.M. Embil, K.L. Chan, The American Heart Association 2007 endocarditis prophylaxis guidelines: a compromise between science and common sense, Can. J. Cardiol. 24 (2008) 673–675. [10] M.H. Thornhill, P.B. Lockhart, B. Prendergast, J.B. Chambers, D. Shanson, NICE and antibiotic prophylaxis to prevent endocarditis, Br. Dent. J. 218 (2015) 619–621. [11] T.B. Sivertsen, A.N. Astrom, G. Greve, J. Assmus, M.S. Skeie, Endocarditis prophylaxis and congenital heart defects in the Norwegian Public Dental Service, Acta Paediatr. 102 (2013) 29–34. [12] R.G. Beaulieu, Wait, not so fast. Are the new American Heart Association endocarditis prophylaxis guidelines safe, and where is the proof? Paediatr. Child Health 14 (2009) 173–175. [13] C.P. Owen, W.H. Huang, Antibiotic prophylaxis for dental procedures: is it necessary? SADJ 67 (2012) 413–419. [14] D. Shanson, New British and American guidelines for the antibiotic prophylaxis of infective endocarditis: do the changes make sense? A critical review, Curr. Opin. Infect. Dis. 21 (2008) 191–199. [15] D.N. Taiwan, http://nhird.nhri.org.tw/en/Background.html (Accessed on June 28th 2015). [16] S.J. Yeh, H.C. Chen, C.W. Lu, J.K. Wang, L.M. Huang, S.C. Huang, et al., Prevalence, mortality, and the disease burden of pediatric congenital heart disease in Taiwan, Pediatr. Neonatol. 54 (2013) 113–118. [17] D. Rushani, J.S. Kaufman, R. Ionescu-Ittu, A.S. Mackie, L. Pilote, J. Therrien, et al., Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors, Circulation 128 (2013) 1412–1419. [18] M.H. Wu, H.C. Chen, C.W. Lu, J.K. Wang, S.C. Huang, S.K. Huang, Prevalence of congenital heart disease at live birth in Taiwan, J. Pediatr. 156 (2010) 782–785. [19] A.S. Dajani, K.A. Taubert, W. Wilson, A.F. Bolger, A. Bayer, P. Ferrieri, et al., Prevention of bacterial endocarditis. Recommendations by the American Heart Association, JAMA 277 (1997) 1794–1801. [20] B.J. Mulder, Endocarditis in congenital heart disease: who is at highest risk? Circulation 128 (2013) 1396–1397. [21] R.S. Baltimore, M. Gewitz, L.M. Baddour, L.B. Beerman, M.A. Jackson, P.B. Lockhart, et al., Infective endocarditis in childhood: 2015 update: a scientific statement from the American Heart Association, Circulation 132 (2015) 1487–1515. [22] J.M. Martin, W.H. Neches, E.R. Wald, Infective endocarditis: 35 years of experience at a children's hospital, Clin. Infect. Dis. 24 (1997) 669–675. [23] C.D. Morris, M.D. Reller, V.D. Menashe, Thirty-year incidence of infective endocarditis after surgery for congenital heart defect, JAMA 279 (1998) 599–603. [24] J.A. Stockheim, E.G. Chadwick, S. Kessler, M. Amer, N. Abdel-Haq, A.S. Dajani, et al., Are the Duke criteria superior to the Beth Israel criteria for the diagnosis of infective endocarditis in children? Clin. Infect. Dis. 27 (1998) 1451–1456. [25] A. Hoyer, M. Silberbach, Infective endocarditis, Pediatr. Rev. 26 (2005) 394–400. [26] P.C. Chen, Y.C. Tung, P.W. Wu, L.S. Wu, Y.S. Lin, C.J. Chang, et al., Dental procedures and the risk of infective endocarditis, Medicine (Baltimore) 94 (2015) e1826. [27] R.W. Elder, R.S. Baltimore, The changing epidemiology of pediatric endocarditis, Infect. Dis. Clin. N. Am. 29 (2015) 513–524. [28] F. Chirillo, P. Faggiano, M. Cecconi, A. Moreo, A. Squeri, O. Gaddi, et al., Predisposing cardiac conditions, interventional procedures, and antibiotic prophylaxis among patients with infective endocarditis, Am. Heart J. 179 (2016) 42–50. [29] A.M. Valente, R. Jain, M. Scheurer, V.G. Fowler Jr., G.R. Corey, A.R. Bengur, et al., Frequency of infective endocarditis among infants and children with Staphylococcus aureus bacteremia, Pediatrics 115 (2005) e15–19. [30] L. Baumgartner, T.D. Jones, M.T. Bellows, et al., American Heart Association. Prevention of rheumatic fever and bacterial endocarditis through control of streptococcal infections, Circulation 11 (1955) 317–320.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case control study Li-Chuan Sun a, Chih-Cheng Lai b, Cheng-Yi Wang c, Ya-Hui Wang d, Jen-Yu Wang c, Yo-Ling Hsu a, Yin-Lan Hu e, En-Ting Wu f, Ming-Tai Lin f, Leticia B. Sy a,⁎, Likwang Chen g,⁎⁎ a
Department of Pediatrics, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan Department of Intensive Care Medicine, Chi Mei Medical Center, Liouying, Tainan, Taiwan Department of Internal Medicine, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan d Medical Research Center, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan e Department of Dentistry, Cardinal Tien Hospital and School of Medicine, Colloge of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan f Department of Pediatrics, National Taiwan University Hospital, National Taiwan University, Taipei, Taiwan g Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Miaoli County, Taiwan b c
a r t i c l e
i n f o
Article history: Received 2 May 2017 Received in revised form 18 July 2017 Accepted 4 August 2017 Available online xxxx Keywords: Infective endocarditis Congenital heart disease Children
a b s t r a c t Background: Infective endocarditis (IE) is uncommon in childhood. Its associated epidemiological characteristics in patients with congenital heart disease (CHD) remain unclear. Methods: The study population included children born in Taiwan during the years 1997 to 2005 who were diagnosed as having CHD before 3 years of age. All children were followed up until the end year of 2010, the diagnosis of infective endocarditis, or death. The demographic characteristics of patients with and without IE, the invasive procedures performed during 6 months before the index date, the prophylactic antibiotics related to dental procedures, and in-hospital mortality were collected. Results: Information of 24,729 children with CHD were retrieved for our analysis and 237 patients with newly diagnosed IE were identified. The incidence rate of IE in all CHD lesions was 11.13 per 10,000 person-years. Taking ASD for reference, the following CHD lesions were at risk for IE: cyanotic CHD (adjusted OR, 9.58; 95% confidence interval, 5.38–17.05), endocardial cushion defect (ECD) (8.01; 2.73–23.50), Left-sided lesions (4.36; 1.90–10.01) and VSD (2.93; 1.64–5.23). Patients who underwent procedures have a higher risk of acquiring IE which include central venous catheter (CVC) insertion (3.17; 2.36–4.27), cardiac catheterization (3.74; 2.67– 5.22), open-heart surgery (2.47; 1.61–3.77), valve surgery (3.20; 1.70–6.02), and shunt surgery (7.43; 2.36– 23.41). However, dental procedures did not increase the risk of IE, irrespective of antibiotic usage. Conclusions: The risk of IE varies markedly among CHD lesions in our study. Invasive heart procedures but not dental procedures, are more significantly associated with IE among children with CHD. © 2017 Published by Elsevier Ireland Ltd.
1. Introduction Infective endocarditis (IE) is uncommon in childhood but is a potentially life-threatening disease. The annual incidence of IE ranged from 0.05 to 0.12 cases per 1000 pediatric admissions [1], and the overall mortality rate is approximately 5% [2,3]. Approximately 35% to 60% of children with IE have congenital heart disease [2–5], which is known to be a predisposing factor for IE in children. The American Heart Association (AHA) guidelines for prevention of IE were revised in 2007 [6]. The prophylaxis for IE has been simplified and restricted to selected patients with predisposing cardiac conditions for whom antibiotic prophylaxis ⁎ Correspondence to: L.B. Sy, Cardinal Tien Hospital, Department of Pediatrics, Cardinal Tien Hospital, Fu Jen Catholic University, College of Medicine, New Taipei City 23148, Taiwan. ⁎⁎ Correspondence to: L. Chen, Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Miaoli County 35053, Taiwan. E-mail addresses: [email protected] (L.B. Sy), [email protected] (L. Chen).
was indicated before the invasive procedure [6,7]. However, concerns still exist about the safety and reliability of these new guidelines from cardiologists, dentists and patients [8–12], especially in developing countries whose populations have unsatisfactory oral hygiene [8,13]. Furthermore, many studies about factors related to IE focused mainly on adult populations, and little information exists in children with congenital heart disease [14]. The purpose of this study was to determine the risk factors for IE, such as dental procedure, open heart surgery, and types of congenital heart disease, in a large population-based cohort of children with IE. 2. Material and methods 2.1. Data source This study used the database constructed by the National Health Research Institutes (NHRI) of Taiwan. Taiwan's National Health Insurance
http://dx.doi.org/10.1016/j.ijcard.2017.08.009 0167-5273/© 2017 Published by Elsevier Ireland Ltd.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
2
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
(NHI) program was implemented in Taiwan since 1995 and 99% of the 23.75 million Taiwanese is under this medical coverage [15]. The NIH offers comprehensive medical coverage, including outpatient, inpatient, emergency, dental services, medical examinations, laboratory tests, drug prescriptions, and interventional procedures. The NHRI used original data from the NHI database to construct a longitudinal database of children with congenital heart disease. 2.2. Study population and control group The patients included in this study were selected from the healthcare records in the NHI database from 1997 to 2010. The study population included children born in Taiwan between year 1997 and 2005 who were diagnosed as having congenital heart disease (CHD) before 3 years of age. In the analysis, 24,729 cases were included and followed up until the end of year 2010, the diagnosis of IE, or death. The CHD was defined in accordance with the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) code. To avoid coding errors caused by incorrect tentative diagnosis, miscoding, and overestimation from insignificant defects (e.g., patent foramen ovale, spontaneous closure of small patent ductus arteriosus [PDA], spontaneous closure of small ventricular septal defect [VSD], and mild peripheral pulmonary artery stenosis of newborn), children with CHD were enrolled only when they had a CHD-specific admission or N3 outpatient clinic visits [16]. For patients with more than one CHD diagnoses, the more frequent and more complex CHD diagnoses were used for further analysis. Children with congenital heart disease who did not have IE during our study period were enrolled in the control cohort. We randomly selected 4725 patients from the control cohort, and matched with the IE patients for same age (1-year strata), sex, and date of hospitalization or outpatient clinic visits to serve as our control group. Twenty control patients were matched with one patient with IE. The IE study group was defined as those having a discharge diagnosis of IE in their first hospitalization, using ICD-9-CM codes (421.0, 421.1, 421.9, and 424.9). The date of hospital admission for IE served as the index date. The control subjects were assigned a date of medical record equal to the index date of their matched case subjects. We recorded all the invasive procedures performed during 6 months before the index date. We also analyzed prophylactic antibiotics related to dental procedures, defined as a prescription of antibiotics on the same day of the dental treatment. 2.3. Definition CHD lesions were classified into eight groups as follows: (1) cyanotic heart disease: Tetralogy of Fallot, single-ventricle heart defects, transposition of great arteries, double outlet of the right ventricle, truncus arteriosus, hypoplastic left heart syndrome, total anomalous pulmonary venous connection, right atrium isomerism (RAI), tricuspid atresia, agenesis of pulmonary artery, and aortic atresia; (2) endocardial cushion defect (ECD); (3) left-sided lesions: coarctation of the aorta, aortic insufficiency, mitral stenosis, and mitral insufficiency; (4) VSD; (5) right-sided lesions: Ebstein anomaly, anomalies of pulmonary artery or pulmonary valve; (6) other CHDs (other or unspecified congenital anomalies); (7) PDA; and (8) arterial septal defect (ASD). Invasive procedures included all open-heart surgeries, shunt operation, valve operation, cardiac catheterization, use of a central venous catheter (CVC), and dental procedures. Valve surgery was defined as surgical repair of cardiac valve that is including annuloplasty, single valve replacement, double valve replacement, three valve replacement, pulmonary valvuloplasty, repair fistula sinus of valsalva, closed mitral or open mitral commissurotomy and Rostalli operation. Percutaneous valve implantation was classified as cardiac catheterization. Dental procedures included tooth extraction, dental scaling, endodontics, tooth pulpotomy, canal enlargement and debridement, root canal therapy, and other operations on the teeth, gums, and alveoli.
2.4. Statistical analysis The baseline characteristics of the children with or without IE in our study cohort were described as absolute numbers and proportions. Overall and CHD lesion-specific incidence rates of IE were calculated as the number of patients with new-onset IE divided by the sum of follow-up person-years at risk for IE in the CHD cohort. The distribution of the risk factors in children with IE and their matched controls were described as absolute numbers and proportions. Odds ratios (ORs) of IE associated with independent variables with 95% confidence intervals (CIs) were computed by using conditional logistic regression that accounted for the matching in the study design. The following variables were assessed: invasive procedures such as CVC, cardiac catheterization, open heart surgery, valve surgery, and shunt surgery, 6 months before index date (yes/no); eight groups of CHD lesions (reference group: ASD); and dental procedures (no treatment, and dental procedure with or without antibiotics) 6 months before index date. All statistical analyses were conducted with the SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA). Statistical significance was defined as a two-sided P value of b 0.05.
3. Results 3.1. Distribution of CHD and incidence of IE Information of 24,729 children with congenital heart disease were retrieved for our analysis and contributed 231,027 person-years. Table 1 shows the demographic and clinical characteristics of the patients, including sex, age at diagnosis, and the distribution of CHD lesions. VSD (8048 cases; 32.54%) was the most common CHD lesions in our cohort, followed by ASD (5620 cases; 22.73%), PDA (3560 cases; 14.4%), and cyanotic CHD lesions (3702 cases; 14.97%). During our observation period, 237 patients with new onset IE were identified, of whom 54.1% were male, and 5.49% were prematures. The median age at IE diagnosis was 1.21 years (interquartile range: 0.59–3.05 years), and 42.19% of IE patients were younger than 1 year. The overall incidence rate of IE in all CHD lesions was 11.13 per 10,000 person-years. The incidence rates of IE per 10,000 children person-years in the major CHD groups were as follows: cyanotic CHD, 35.73; ECD, 27.24; Table 1 The baseline characteristics in CHD cohort (n = 24,729).
Gender Male Female Premature birth No Yes Age of IE diagnosis (year) Median (IQR) 0–1 1–2 2–3 N3 CHD lesions Cyanotic CHD ECD Left-sided lesions VSD Right-sided lesions Other CHD PDA ASD
Total (n = 24,729)
IE (n = 237)
Non-IE (n = 24,492)
No.
%
No.
%
%
%
12,218 12,511
49.41 50.59
128 109
54.01 45.99
12,090 12,402
49.36 50.64
22,883 1846
92.54 7.46
224 13
94.51 5.49
22,659 1833
92.52 7.48
1.21 (0.59–3.05) 100 42.19 56 23.63 18 7.56 63 26.58
– – – –
– – – –
88 5 10 74 5 36 5 14
3614 293 850 7972 1824 778 3555 5606
14.76 1.20 3.47 32.55 7.45 3.18 14.51 22.89
3702 298 860 8048 1829 814 3560 5620
14.97 1.21 3.48 32.54 7.40 3.29 14.40 22.73
37.13 2.11 4.22 31.22 2.11 15.19 2.11 5.91
CHD = congenital heart disease; ECD = endocardial cushion defect; VSD = ventricular septal defect; PDA = patent ductus arteriosus; ASD = atrial septal defect.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx Table 2 The CHD lesion group-specific incidences of infective endocarditis in the study cohort. CHD lesions
No. of IE (%)
Person-year
Incidence ratea
Cyanotic CHD ECD Left-sided lesions VSD Right-sided lesions Other CHD PDA ASD Overall
88 (37.13) 5 (2.11) 10 (4.22) 74 (31.22) 5 (2.11) 36 (15.19) 5 (2.11) 14 (5.9) 237
24,627.97 1835.21 7012.84 73,221.58 16,627.17 7067.1 32,953.83 49,681.47 213,027.17
35.73 27.24 14.26 10.11 3.01 50.94 1.52 2.82 11.13
3
with VSD, 1 patient with PDA, and 1 patient with other congenital heart diseases. In the logistic regression analysis, the in-hospital mortality of the IE patients with cyanotic heart disease was 3.63 times (95% CI, 1.06–12.41) higher than that of the IE patients with non-cyanotic heart disease (data not shown). 4. Discussion
CHD = congenital heart disease; ECD = endocardial cushion defect; VSD = ventricular septal defect; PDA = patent ductus arteriosus; ASD = atrial septal defect. a Per 10,000 person-years.
left-sided lesion, 14.26; VSD, 10.11; right-sided lesion, 3.01; ASD, 2.82; PDA, 1.52; and other CHDs, 50.94 (Table 2). 3.2. Risk of IE According to the previous AHA guideline for IE [19], isolated secundum ASD is consider as a negligible risk of IE, which means the risk of development of IE is not higher than general population. Thus, we take the ASD group as our reference to analyze the relationship between cardiac conditions and risk of infectious endocarditis. Taking ASD as reference, the odds ratio of lesion groups were estimated as follows: cyanotic CHD (adjusted OR, 9.58; 95% CI, 5.38–17.05), endocardial cushion defect (ECD; adjusted OR, 8.01; 95% CI, 2.73–23.50), leftsided lesions (adjusted OR, 4.36; 95% CI, 1.90–10.01), VSD (adjusted OR, 2.93; 95% CI, 1.64–5.23), right-sided lesions (adjusted OR, 1.01; 95% CI, 0.36–2.83), other CHDs (adjusted OR, 21.09; 95% CI, 10.93–40.69), and PDA (adjusted OR, 0.46; 95% CI, 0.17–1.30) (Table 3). 3.3. Invasive procedures, dental procedures, prematurity and IE The risk of IE from invasive and dental procedures are listed in Table 4. The patients who received procedures had a higher risk of IE, including all CVCs (adjusted OR, 3.17; 95% CI, 2.36–4.27), CVCs without surgery (adjusted OR, 2.25; 95% CI, 1.65–3.06), cardiac catheterization (adjusted OR, 3.74; 95% CI, 2.67–5.22), open heart surgery (adjusted OR, 2.47; 95% CI, 1.61–3.77), valve surgery (adjusted OR, 3.20; 95% CI, 1.70–6.02), and shunt surgery (adjusted OR, 7.43; 95% CI, 2.36–23.41). However, dental procedure did not increase the risk of IE, regardless of whether antibiotics were used or not (Table 4). In addition, we found that prematurity did not increase the risk of IE in children with CHD (adjusted OR, 0.69; 95% CI, 0.39–1.21) (Table 4). 3.4. Mortality In-hospital mortality of IE occurred in 12 (5%) of the 237 patients, including 8 patients with cyanotic congenital heart disease, 2 patients
From this large retrospective population-based cohort study of 24,729 children (213,027 person-years) with congenital heart disease, we reported several significant findings. First, the incidence rate of IE in our study was 11.13 cases per 10,000 person-years, higher than that in the previous study [17]. In Quebec, the incidence of IE in children with congenital heart disease was 4.1 cases per 10,000 person-years. This difference in incidence rate may be attributed to the difference in the study population included in the study and the fact that the risk of IE is strongly dependent on the wide variation in the types and severity of the underlying congenital cardiac condition. In our study population, we enrolled only those who had a CHD-specific admission or had N 3 outpatient clinic visits. Insignificant defects such as patent foramen ovale, spontaneous closure of small PDA and small VSD, and mild peripheral pulmonary artery stenosis of newborn were excluded in our study cohort. This can explain why the incidence of IE in our study was slightly higher than that in the previous study. Furthermore, VSD is the most common CHD in our cohort (8048 cases; 32.54%), followed by secundum ASD (5620 cases; 22.73%) and PDA (3560 cases; 14.40%). This CHD spectrum distribution is similar to those in previous studies in Taiwan [16,18] that reported the five most frequently diagnosed CHDs to be VSD, secundum ASD, PDA, pulmonary stenosis, and Tetralogy of Fallot. Second, the risk for IE in our study varies markedly among the types of CHD lesions. Aside from cyanotic heart disease, ECD, left-sided lesions and VSD also had an increased risk for developing IE. Our results are similar with previous pediatric studies [4,20–25] that concluded that congenital heart defects such as aortic valve abnormalities, VSD, and Tetralogy of Fallot were common underlying conditions in children with IE. Rushani et al. [17] reported that relative to ASD, the following lesions were most strongly associated with an elevated risk of IE: cyanotic CHD, ECD, and left-sided lesions. In their study [17], the incidence of IE in VSD patients were lower than our reported rates, this may be attributed in part to some insignificant defects such as spontaneous closure of small VSD for which it was not included in our study cohort. Another result in our study that is worth clarifying is that the incidence rate of IE in the other CHD groups was high (ICD-9:746.8; other unspecified anomalies of the heart). A possible explanation is that some of the specific complex heart diseases might be classified into the other CHD groups. Failure of parents to provide the correct complex heart disease diagnosis and the inability of the general pediatric practitioners to identify the right ICD-9 code for the specific heart disease might contribute to this misclassification. The in-hospital mortality rate in our study for CHD patients with IE
Table 3 Crude and adjusted odds ratio of infectious endocarditis among congenital heart diseases in children.
ASD Cyanotic CHD Endocardial cushion defect Left-sided lesions Ventricular septal defect Right-sided lesions Other CHD Patent ductus arteriosus
Case (n = 237)
Control (n = 4725)
Crude OR
No.
(%)
No.
(%)
(95% CI)
14 88 5 10 74 5 36 5
(5.91) (37.13) (2.11) (4.22) (31.22) (2.11) (15.19) (2.11)
951 635 38 162 1745 335 118 741
(20.13) (13.44) (0.80) (3.43) (36.93) (7.09) (2.50) (15.68)
Reference 9.63 (5.41, 17.14) 8.20 (2.80, 24.04) 4.34 (1.89, 9.96) 2.94 (1.64, 5.25) 1.01 (0.36, 2.82) 21.35 (11.07, 41.18) 0.47 (0.17, 1.30)
P value
Adjusteda OR
P value
(95% CI) b0.001 b0.001 b0.001 b0.001 0.991 b0.001 0.144
Reference 9.58 (5.38, 17.05) 8.01 (2.73, 23.50) 4.36 (1.90, 10.01) 2.93 (1.64, 5.23) 1.01 (0.36, 2.83) 21.09 (10.93, 40.69) 0.46 (0.17, 1.30)
b0.001 b0.001 b0.001 b0.001 0.986 b0.001 0.143
ASD = atrial septal defect; CHD = congenital heart disease. a Adjusted for age and gender.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
4
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Table 4 Invasive procedure dental procedure and prematurity in the study cohort. Control (n = 4725)
Crude OR
No.
(%)
No.
(%)
(95% CI)
Invasive procedure 6 months before index date CVC insertion 104 CVC insertion without OHS, VS or SS 69 Cardiac catheterization 54 Open heart surgery (OHS) 28 Valve surgery (VS) 15 Shunt surgery (SS) 4
(43.88) (29.11) (22.78) (11.81) (5.06) (1.69)
1050 762 369 245 75 11
(22.22) (16.13) (7.81) (5.19) (1.59) (0.23)
3.23 (2.41, 4.34) 2.29 (1.68, 3.12) 3.73 (2.67, 5.20) 2.52 (1.65, 3.85) 3.35 (1.79, 6.28) 7.27 (2.32, 22.84)
b0.001 b0.001 b0.001 b0.001 b0.001 b0.001
3.17 (2.36, 4.27) 2.25 (1.65, 3.06) 3.74 (2.67, 5.22) 2.47 (1.61, 3.77) 3.20 (1.70, 6.02) 7.43 (2.36, 23.41)
b0.001 b0.001 b0.001 b0.001 b0.001 b0.001
Dental Procedure 6 months before index date No dental procedure 224 Dental procedure with antibiotics 10 Dental procedure without antibiotics 3
(94.51) (4.22) (1.27)
4424 144 157
(93.63) (3.05) (3.32)
Reference 1.30 (0.64, 2.65) 0.35 (0.11, 1.17)
0.469 0.088
Reference 1.31 (0.64, 2.66) 0.35 (0.11, 1.17)
0.459 0.088
Premature birth No Yes
94.51 5.49
4355 370
92.17 7.83
Reference 0.69 (0.39, 1.21)
0.685
Reference 0.69 (0.39, 1.21)
0.686
224 13
P value
Adjusteda OR
Case (n = 237)
P value
(95% CI)
CVC = central venous catheter. a Adjusted for age and gender.
was 5% (12/237). Among the 88 IE patients with cyanotic heart disease, 8 patients (9%) died, while the mortality rate of IE patients with non-cyanotic heart disease stands at 2.6% (4/149). Base on the abovementioned data, IE patients with cyanotic heart disease appeared to have a threefold increased risk for inpatients mortality (3.63 times; 95% CI, 1.06–12.41). In summary, we found that children with cyanotic heart disease, ECD, VSD and left-sided heart lesions were at higher risk for developing IE and those IE patients with cyanotic heart disease appeared to be at higher risk for in-hospital mortality than those IE patients with non-cyanotic heart disease. Third, our study showed that invasive procedures, including CVC, cardiac catherization, open heart surgery, valve surgery, and shunt surgery within 6 months after the procedures were significantly associated with increased risk of infectious endocarditis in children with CHD. Previous studies showed that the increased IE risk after cardiac surgery and catheter-related cardiac interventions is mainly linked to implantation of prosthetic material or cardiovascular devices [21,23]. Surgery itself, including such elements as central vascular catheters, intravenous alimentation, and length of stay in the intensive care unit, may be important risk factors for the development of IE [21]. Of all the invasive procedures, shunt surgery was associated with the highest risk of developing IE within the first 6 months post procedure, having a sevenfold increase. This result was similar to the previous study done by Baltimore et al. (21) showing that approximately 50% of children with IE complicating CHD had a previous cardiac surgery, particularly palliative shunt procedures or complex intracardiac repairs. Overall, invasive procedures, including open heart surgery, valve surgery, shunt surgery, cardiac catheterization, and CVC were strongly associated with development of IE during the 6-month post-invasive procedure period. Elevated IE risk in early childhood has been reported in studies of children with CHD [17] which is probably multifactorial in etiology. Our data showed that most of the IE episode occurred in the 1st year of life (100/237; 42.19%). Almost all complex congenital heart patients underwent surgical repair of their CHD lesions early in life which may put these children at risk for IE. Fourth, this is the first large retrospective population-base cohort case-control study conducted to analyze the relationship between dental procedures and IE in children with congenital heart disease. The main finding of our study was that dental procedures are not significantly associated with infectious endocarditis in children with congenital heart disease regardless of antibiotic prophylaxis. In our study, 13 patients (5.49%) had received dental procedures within 6 months preceding their onset of IE. This is consistent with the observation in adult studies [26,27]. In a case-crossover design study
of 739 patients with IE by Chen et al. [26], dental procedures did not significantly contribute to the risk for IE. In another retrospective study from a large national registry of 677 patients by Chirllo et al., the prevalence of dental procedures as a presumed portal of entry of the germ responsible for IE was only 4.7% [28]. This may be explained by the observation of the changing bacterial spectrum causing IE in children is shifting from streptococci to Staphylococcus aureus [2,8,21, 29]. This trend is likely to reflect in changing risk factors for those who acquired IE, with more children having a hospitalization with medical issues predisposing to IE [27]. In addition, children have a much lower prevalence and severity of gingivitis and periodontitis than adults [4] and in general, dental health information is more aggressively provided to parents of children with CHD by pediatric cardiologist and dentist. Although antibiotic prophylaxis for IE has been recommended for persons with congenital heart disease since 1955 [30], so far no published data accurately determine the absolute risk of IE that resulted from a dental procedure nor there was any definite prove in human studies that prophylactic antibiotics treatment can prevent endocarditis [6]. Dental procedures did not increase the risk of IE, irrespective of antibiotic usage, and further study is still needed to confirm our findings. Based on our findings about the association between dental procedure and risk of IE among children with CHD, it should suggest that dental procedures are not important risk factors for IE in children with CHD, regardless of whether antibiotic prophylaxis was given. Because of the low incidence of IE in children, large number of patients per cohort is required to obtain statistical clinical significance when analyzing the risk factors for IE. Our study is a population-base nested case-control design with two major strengths, a large sample size obtained from the database and the ability to longitudinal followup patients to identify the risk factor for IE. Our study has some limitations: First, the NHIRD database contains little clinical information and therefore, we are unable to recognize the causative organism and the mortality of cases after hospital discharge. Furthermore, the diagnosis of IE cases was not based on Duke Criteria and incorrect diagnosis or miscoding may lead to misclassification in CHD. The enrollment of patients with IE relied on ICD-9-CM codes from hospital discharge records. This is in similarity to several large-scale studies designed to obtain absolute measures of IE frequency [17,26]. To avoid errors caused by incorrect tentative diagnosis, miscoding and overestimation from insignificant defects, children with CHD were enrolled only in our study when they have a CHDspecific admission or N 3 times of outpatient clinic visits [16]. Second, our study only included children born in Taiwan during year 1997 to 2005 who were diagnosed to have CHD before 3 years of age. According to Wu et al. report [18] that an upward trend was observed for age
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009
L.-C. Sun et al. / International Journal of Cardiology xxx (2017) xxx–xxx
specific prevalence of CHD that was maintained until the 3 years old, followed by a plateau at 3–10 years of age in Taiwan. Thus, our study enrolls almost all of children with CHD in Taiwan during the study period. Third, because our study aims to investigate the association between invasive cardiovascular procedures and dental procedures with antibiotic prophylaxis among IE patients in children with CHD, we did not collect other possible risk factors for further evaluation. Furthermore, we can only obtain between 1997 and 2010 from NHIRD due to National Health Research Institute's regulation about the size of data availability. However, our study population included CHD patients with age ranging from birth to age 5–14 and a follow up period of 14 years which is longer than previous studies [3,17,26,27]. Therefore, the findings in our study are still representative of IE risk among children with CHD. 5. Conclusion Our study revealed that the risk for developing IE varies markedly among the types of CHD lesions. Invasive heart procedures were more significantly associated with IE in children with CHDs while dental procedures were not an important risk factors for IE in children with CHD, regardless of whether antibiotic prophylaxis was given. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Sources of funding This study was supported by grants CTH 106A-2A12 (Cardinal Tien Hospital, Taiwan) and National Health Research Institutes (PH-105-SP-08). References [1] S.K. Pasquali, X. He, Z. Mohamad, B.W. Mccrindle, J.W. Newburger, J.S. Li, et al., Trends in endocarditis hospitalizations at US children's hospitals: impact of the 2007 American Heart Association Antibiotic Prophylaxis Guidelines, Am. Heart J. 163 (2012) 894–899. [2] M.D. Day, K. Gauvreau, S. Shulman, J.W. Newburger, Characteristics of children hospitalized with infective endocarditis, Circulation 119 (2009) 865–870. [3] A.L. Ware, L.Y. Tani, H.Y. Weng, J. Wilkes, S.C. Menon, Resource utilization and outcomes of infective endocarditis in children, J. Pediatr. 165 (2014) (807–812.e801). [4] L. Saiman, A. Prince, W.M. Gersony, Pediatric infective endocarditis in the modern era, J. Pediatr. 122 (1993) 847–853. [5] D. Marom, I. Levy, O. Gutwein, E. Birk, S. Ashkenazi, Healthcare-associated versus community-associated infective endocarditis in children, Pediatr. Infect. Dis. J. 30 (2011) 585–588. [6] W. Wilson, K.A. Taubert, M. Gewitz, P.B. Lockhart, L.M. Baddour, M. Levison, et al., Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group, Circulation 116 (2007) 1736–1754.
5
[7] G. Habib, B. Hoen, P. Tornos, F. Thuny, B. Prendergast, I. Vilacosta, et al., Guidelines on the prevention, diagnosis, and treatment of infective endocarditis (new version 2009): the Task Force on the Prevention, Diagnosis, and Treatment of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and the International Society of Chemotherapy (ISC) for Infection and Cancer, Eur. Heart J. 30 (2009) 2369–2413. [8] W.C. Tseng, S.N. Chiu, P.L. Shao, J.K. Wang, C.A. Chen, M.T. Lin, et al., Changing spectrum of infective endocarditis in children: a 30 years experiences from a tertiary care center in Taiwan, Pediatr. Infect. Dis. J. 33 (2014) 467–471. [9] J.M. Embil, K.L. Chan, The American Heart Association 2007 endocarditis prophylaxis guidelines: a compromise between science and common sense, Can. J. Cardiol. 24 (2008) 673–675. [10] M.H. Thornhill, P.B. Lockhart, B. Prendergast, J.B. Chambers, D. Shanson, NICE and antibiotic prophylaxis to prevent endocarditis, Br. Dent. J. 218 (2015) 619–621. [11] T.B. Sivertsen, A.N. Astrom, G. Greve, J. Assmus, M.S. Skeie, Endocarditis prophylaxis and congenital heart defects in the Norwegian Public Dental Service, Acta Paediatr. 102 (2013) 29–34. [12] R.G. Beaulieu, Wait, not so fast. Are the new American Heart Association endocarditis prophylaxis guidelines safe, and where is the proof? Paediatr. Child Health 14 (2009) 173–175. [13] C.P. Owen, W.H. Huang, Antibiotic prophylaxis for dental procedures: is it necessary? SADJ 67 (2012) 413–419. [14] D. Shanson, New British and American guidelines for the antibiotic prophylaxis of infective endocarditis: do the changes make sense? A critical review, Curr. Opin. Infect. Dis. 21 (2008) 191–199. [15] D.N. Taiwan, http://nhird.nhri.org.tw/en/Background.html (Accessed on June 28th 2015). [16] S.J. Yeh, H.C. Chen, C.W. Lu, J.K. Wang, L.M. Huang, S.C. Huang, et al., Prevalence, mortality, and the disease burden of pediatric congenital heart disease in Taiwan, Pediatr. Neonatol. 54 (2013) 113–118. [17] D. Rushani, J.S. Kaufman, R. Ionescu-Ittu, A.S. Mackie, L. Pilote, J. Therrien, et al., Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors, Circulation 128 (2013) 1412–1419. [18] M.H. Wu, H.C. Chen, C.W. Lu, J.K. Wang, S.C. Huang, S.K. Huang, Prevalence of congenital heart disease at live birth in Taiwan, J. Pediatr. 156 (2010) 782–785. [19] A.S. Dajani, K.A. Taubert, W. Wilson, A.F. Bolger, A. Bayer, P. Ferrieri, et al., Prevention of bacterial endocarditis. Recommendations by the American Heart Association, JAMA 277 (1997) 1794–1801. [20] B.J. Mulder, Endocarditis in congenital heart disease: who is at highest risk? Circulation 128 (2013) 1396–1397. [21] R.S. Baltimore, M. Gewitz, L.M. Baddour, L.B. Beerman, M.A. Jackson, P.B. Lockhart, et al., Infective endocarditis in childhood: 2015 update: a scientific statement from the American Heart Association, Circulation 132 (2015) 1487–1515. [22] J.M. Martin, W.H. Neches, E.R. Wald, Infective endocarditis: 35 years of experience at a children's hospital, Clin. Infect. Dis. 24 (1997) 669–675. [23] C.D. Morris, M.D. Reller, V.D. Menashe, Thirty-year incidence of infective endocarditis after surgery for congenital heart defect, JAMA 279 (1998) 599–603. [24] J.A. Stockheim, E.G. Chadwick, S. Kessler, M. Amer, N. Abdel-Haq, A.S. Dajani, et al., Are the Duke criteria superior to the Beth Israel criteria for the diagnosis of infective endocarditis in children? Clin. Infect. Dis. 27 (1998) 1451–1456. [25] A. Hoyer, M. Silberbach, Infective endocarditis, Pediatr. Rev. 26 (2005) 394–400. [26] P.C. Chen, Y.C. Tung, P.W. Wu, L.S. Wu, Y.S. Lin, C.J. Chang, et al., Dental procedures and the risk of infective endocarditis, Medicine (Baltimore) 94 (2015) e1826. [27] R.W. Elder, R.S. Baltimore, The changing epidemiology of pediatric endocarditis, Infect. Dis. Clin. N. Am. 29 (2015) 513–524. [28] F. Chirillo, P. Faggiano, M. Cecconi, A. Moreo, A. Squeri, O. Gaddi, et al., Predisposing cardiac conditions, interventional procedures, and antibiotic prophylaxis among patients with infective endocarditis, Am. Heart J. 179 (2016) 42–50. [29] A.M. Valente, R. Jain, M. Scheurer, V.G. Fowler Jr., G.R. Corey, A.R. Bengur, et al., Frequency of infective endocarditis among infants and children with Staphylococcus aureus bacteremia, Pediatrics 115 (2005) e15–19. [30] L. Baumgartner, T.D. Jones, M.T. Bellows, et al., American Heart Association. Prevention of rheumatic fever and bacterial endocarditis through control of streptococcal infections, Circulation 11 (1955) 317–320.
Please cite this article as: L.-C. Sun, et al., Risk factors for infective endocarditis in children with congenital heart diseases - A nationwide population-based case contro..., Int J Cardiol (2017), http://dx.doi.org/10.1016/j.ijcard.2017.08.009