- Email: [email protected]
Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years
IJCA-25594; No of Pages 6 International Journal of Cardiology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years K. Thom a,d, A. Hanslik d, J.L. Russell b, S. Williams a, P. Sivaprakasam a, U. Allen c, C. Male d, L.R. Brandão a,⁎ a
Pediatric Haematology/Oncology, The Hospital for Sick Children, Toronto, Canada Pediatric Cardiology, Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, Canada Infectious Disease, The Hospital for Sick Children, Toronto, Canada d Division of Pediatric Cardiology, Department of Children and Adolescent Medicine, Medical University Vienna, Austria b c
a r t i c l e
i n f o
Article history: Received 27 April 2017 Received in revised form 13 October 2017 Accepted 23 October 2017 Available online xxxx Keywords: Pediatric endocarditis Thrombosis Embolism Heart failure Mortality
a b s t r a c t Background: Pediatric infective endocarditis (IE) has been associated with high morbidity and mortality, mostly related to thromboembolic complications (TEC). The objective of our study was to describe the experience in children with IE and to review the changes over a thirty-year period, regarding origin of IE, incidence of vegetations, TEC and their respective morbidity and mortality rates. Methods: A retrospective chart review of children aged 0–18 years with IE defined by the Duke Criteria and admitted to The Hospital for Sick Children, was conducted. Data were divided into three periods (P); P1 (1979–1988); P2 (1989–1998); and P3 (1999–2008). Results: The study included 113 patients, median age 7 yrs.; females: 46 (41%), congenital heart defects 95 (84%), comparable in all periods. Overall, cardiac vegetations were found in 68/113 patients (60%); large vegetations (≥1 cm) in 32 patients (28%). Fourty-five (45/133 [40%]) TEC were documented, 22 patients (20%) developed cerebrovascular events (CVE) and 23 patients (20%) had non-CVE. Patients diagnosed during P3 were older, had more vegetations (p b 0.05), and a higher incidence of community acquired-IE (p b 0.05). Overall, mortality was 15%, comparable in all periods. Significant risk factors for mortality were vegetations (HR 6.44; 95% CI: 2.07– 20.01, p = 0.002) and heart failure (HR 28.39; 95% CI: 10.49–76.85, p b 0.001). Conclusions: Over the study period, we report a growing incidence of community acquired pediatric IE in older children accompanied by an increasing rate of TEC. Heart failure and vegetations were associated with an increased mortality. These preliminary data need to be confirmed by prospective data. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Infective endocarditis (IE) is a rare but serious condition in children with an incidence estimated between 0.34 and 0.64 cases per 100,000 per year [1]. To date, the diagnosis of pediatric IE is still challenging and frequently delayed, with a reported mortality in children ranging from 4 to 25% [2,3]. According to adult data, systemic embolism occurs in 13 to 50% of patients with IE. Thromboembolic complications (TEC) into major arteries and organs significantly contribute to disease morbidity [4–6]. Cerebro-vascular events (CVE), including acute ischemic
Abbreviations: IE, infective endocarditis; TEC, thromboembolic complication; P1, 2, 3, periods 1, 2 and 3; (Non)-CVE, (non)-cerebrovascular events; CA-IE, community acquired-infective endocarditis; AIS, arterial-ischemic stroke; CHD, congenital heart disease; PE, pulmonary embolism; TTE, transthoracic echocardiography; TEE, transesophageal echocardiography; BC, blood culture. ⁎ Corresponding author at: The Hospital for Sick Children, 555 University Avenue, Black Wing, Room 10412, Toronto, ON M5G 1X8, Canada. E-mail address: [email protected] (L.R. Brandão).
strokes (AIS), are reported in 20 to 40% of patients with IE [7–9]. For children, a small retrospective series reported 6% IE related AIS [10]. Location and size of intracardiac vegetations have been identified as risk factors in various studies for cardioembolic AIS in adults [11]. Initiation of prompt antimicrobial treatment is considered the mainstay in both the treatment of IE and in the prevention of thrombotic complications, including AIS [12]. Despite the importance of thrombotic events in patients with IE, the use of anticoagulation and antiplatelet therapy is highly controversial. On the one hand, an inverse dose-dependent response to acetylsalicylic acid on vegetation growth, microbial proliferation and renal embolization has been described for experimental Staphylococcus aureus (S. aureus)-induced IE [13]. On the other hand, uncertainty about the appropriateness of such treatment modalities continues, as reflected by the most recent pediatric and adult antithrombotic and IE guidelines [12,14,15]. Currently there is no evidence to support anticoagulation or antiplatelet therapy for patients with IE [16]. The objective of the current study was to review a single-center experience with IE in children and to describe changes over 30 years,
https://doi.org/10.1016/j.ijcard.2017.10.085 0167-5273/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
2
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
regarding origin of IE, incidence of vegetations and TEC. In addition, we aimed to describe risk factors for IE-related TEC and mortality. 2. Methods A retrospective chart review including only pediatric patients with definite IE, as per modified Duke Criteria, was conducted [17] In patients without vegetations, IE diagnosis was made according to positive blood culture (BC) or to minor criteria including prolonged fever, CHD, new regurgitation murmur, and vascular or embolic phenomena. Thromboembolic complications were included if confirmed radiologically and occurred after confirmation of IE as per the modified Duke Criteria for vascular complications [17]. Consecutive patients diagnosed with IE, admitted to The Hospital for Sick Children, Toronto (Canada) between January 1978 and January 2008 were eligible. Patients whose admitting and/or discharge diagnoses included IE were retrieved from either the Cardiology or the Thrombosis service database. Permission to use and analyze the data was obtained from the local Research Ethics Board of The Hospital for Sick Children, Toronto (Canada). Demographic data obtained included: age, gender, underlying disease, congenital heart (CHD) type (cyanotic or acyanotic), previous cardiac surgery, cardiac interventions or previous IE as well as non-cardiac interventions prior to IE diagnosis (e.g. other surgery, vaccinations, dental procedures). A standardized case report form was completed for each patient to maximize data collection consistency and quality. 2.1. Definitions Nosocomial infection consisted of any infection, occurring ≥72 h after hospital admission [18]. Postoperative IE was defined as IE diagnosed within 6 months after surgery [19]. Thromboembolic complications (TEC) included cerebrovascular events (CVE) or nonCVE, as follows: a) CVE, including AIS, mycotic aneurysms or hemorrhages; and b) nonCVE, defined as radiologically-confirmed thromboembolism outside the central nervous system. The indication to perform radiological investigation was the clinical suspicion of a TEC. Major bleeding events secondary to either antiplatelet or anticoagulant use were defined as per internationally accepted criteria [20]. Surgical intervention encompassed any surgery related to IE: early surgery comprised surgery up to three days after diagnosis of IE. Heart failure was defined and confirmed by cardiology staff physician; this information was confirmed in the consultation notes. Due to the retrospective study design, mortality included only perioperative deaths or endocarditis-related mortality up to 30 days after the diagnosis of IE was confirmed. 2.2. Statistical methods To determine trends in childhood IE over the surveyed 30 years, the total observation time was divided into equal time periods of 10 years, each: period 1 (P1; 1979–1988), period 2 (P2; 1989–1998) and period 3 (P3; 1999–2008). Descriptive statistics including percentages and medians were calculated. Mann– Whitney-U-test was used for comparison between groups with continuous variables and Chi-square-test for comparison of categorical variables. Univariable and stepwise multivariable logistic regression were conducted for exploratory analysis of risk factors associated with TEC and mortality. Variables showing a trend for an association with the dependent variable (p value ≤ 0.1) in the univariable analysis were included in the multivariable analysis. Significance level was set up at alpha b0.5.
3. Results 3.1. Demographic data The initial cohort using the diagnostic code for IE comprised 152 patients. Thirty nine patients were excluded for the following reasons: i) no definite IE according to Duke criteria (n = 30), ii) unavailable or incomplete data (n = 7), iii) postmortem diagnosis of IE (n = 2). The final study cohort consisted of 113 patients, aged 1 day to 18 yrs., with 41 patients (36%) aged ≤3 yrs. The median age was 7.0 yrs., 46 patients (41%) were females. Underlying CHD was present in 95/113 patients (84%), acyanotic CHD in 61/113 patients (54%) and cyanotic CHD in 34/113 children (30%). 3.2. Characteristics of the surveyed periods Table 1 shows characteristics of patients in the surveyed time periods. Period 1 (P1) comprised 31 (27%) patients, P2: 47 (42%) and P3: 35 (31%) children. Median age at diagnosis was significantly (p b 0.001) higher in P3 (12.3 years) compared to P1 (3.0 years) and P2 (4.0 years, p b 0.001). In P3, a significantly higher incidence of community-acquired IE (77% [P3] vs. 48% [P1] vs. 57% [P2], p = 0.01)
existed compared to the prior two study decades. Additionally, patients in P3 had higher incidences of vegetations (71% [P3] vs. 45% [P1] vs. 62% [P2], p = 0.01) with comparable CVE, but increasing non-CVE (26% [P3] vs. 13% [P1], p = 0.43). 3.3. Thromboembolic complications Fourty-five thrombotic events (45/113 [40%]) were documented. Cerebrovascular events (CVE) were documented in 22/113 children (19%). An acute ischemic stroke was diagnosed 18/113 patients (16%) and cerebral mycotic aneurysms in 4/113 children (3%). Non-CVE was present in 23/113 cases (20%). Among those with non-CVE, 9/113 children (8%) developed pulmonary embolism, 6 (5%) had kidney infarcts, 5 (4%) had splenic infarcts, and 3 (3%) had osteomyelitis complicated by bone abscess formation. Overall, children above 3 years had longer (median 17 days) fever periods with prolonged clinical courses compared to patients younger than 3 years (median 12 days, n.s.). An exploratory analysis was conducted to identify risk factors associated with TEC. The logistic regression found age above 3 years to be significantly associated with TEC, including CVE and nonCVE (p = 0.01). Congenital heart diseases (p = 0.33), causative microorganisms (p = 0.95) and heart failure (p = 0.51) were not associated with TEC. 3.4. Vegetations Vegetations were identified by either transthoracic (TTE) or transesophageal echocardiography (TEE) in 68/113 cases (60%). Location of vegetations is given in Table 1. During P1, TEE was not done whereas in P2 15/29 patients (52%) with vegetations had TEE with 13/15 (87%) positive exams for vegetations (vegetation only seen in TEE [n = 1], lesion only visible in TTE [n = 1]). During P3 17/25 children (68%) who had developed vegetations underwent TEE (vegetation only seen by TEE [n = 2]). Overall, from 68 patients with vegetations, 45/68 (66%) developed TEC. Whereas of the 42 patients without vegetations 12/42 (29%) developed TEC (8 CVE 6 non-CVE, 2 both; p b 0.001). In another three patients there was no information regarding vegetation. In 32/113 children (27%) large (≥1 cm) vegetations were documented and of these, 18 children (18/32 [56%]) developed TEC. 3.5. Causative microorganisms The causative microorganism was identified by BC in 105/113 children (93%). Staphylococcus aureus was found to be the causative microorganism in 43 cases (38%). The remaining patients had positive BC for Streptococcus viridans (29 [26%]), Staphylococcus epidermidis (14 [12%]), Streptococcus pneumoniae (3 [3%]), Candida albicans (2 [2%]), combined organisms (6 [5%]) and miscellaneous (8 [7%]) microorganisms (Haemophilus influenza [1], Enterococcus [2], Actinobacillus sp. [1], Pseudomonas aeruginosa [2], Pasteurella sp. [1], and Streptococcus abiotrophia [1]). Eight patients (7%) had BC-negative endocarditis. Of these, 5 children were diagnosed by a biopsy after surgery and 3 children by clinical criteria according to the Duke Criteria. During the entire study period, an increasing rate of S. viridans and decreasing rate of S. aureus as the causing organisms were found. 3.6. Origin of IE and antibiotic prophylaxis Overall, community acquired-IE was seen in 69/113 (61%) and nosocomial IE in 44/113 cases (39%) with comparable proportions of CHD over time. Children with community acquired-IE were significantly older than in the nosocomial group (median age 10.8 yrs. vs. 0.7 yrs., p b 0.001), had more vegetations (p = 0.016) and a higher incidence of IE-related surgical interventions (p = 0.009). S. viridans was the dominant causative microorganism in CA-IE (p b 0.001) compared with IE in the nosocomial group (Table 2).
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Characteristics of all patients in the surveyed periods. Characteristics
Overall n (%)
Period 1 (1979–1988)
Period 2 (1989–1998)
Period 3 (1999–2008)
n Age at diagnosis (y) 0–3 3–11 11–18 Median age (Q1;Q3) Endocarditis type Community acquired Nosocomial Underlying disease Cyanotic CHD Acyanotic CHD No CHD Prior cardiac surgery/interv. Prior non-cardiac surgery Causing organisms Staphyococcus aureus Streptococcus viridans Vegetations Multiple locations Aortic/Mitralvalve Right sided, conduits, prosthetic valves, BTS Septal Septic aneurysms CVE Non CVE IE surgical intervention IE related mortality
113
31 (27%)
47 (42%)
35 (31%)
41 (36) 34 (30) 38 (34) 7.0 (1.1;12.8)
15 (48) 13 (42) 3 (10) 3.0 (1.0;8.6)
21 (45) 13 (28) 13 (28) 4.0 (0.4; 11.6)
5 (14) 8 (23) 22 (63) 12.3 (7.8; 14.8)
69 (61) 44 (39)
15 (48) 16 (52)
27 (57) 20 (42)
27 (77) 8 (23)
34 (30) 61 (54) 18 (16) 50 (44) 3 (3)
13 (42) 14 (45) 4 (13) 22 (71)b 0
14 (30) 26 (55) 7 (15) 19 (40) 3 (6)
7 (20) 21 (60) 7 (20) 9 (26) 0
43 (38) 29 (26) 68 (60) 18 38 40 6 3 22 (19) 23 (20) 40 (35) 17 (15)
14 (45) 4 (13) 14 (45) 6 5 9 2 2 5 (16) 4 (13) 8 (26) 5 (16)
22 (47) 12 (25) 29 (62) 7 14 19 3 0 9 (19) 10 (21) 17 (36) 7 (15)
7 (20) 13 (37) 25 (71) 5 19 12 1 1 8 (23) 9 (26) 15 (43) 5 (14)
p value
b0.001a
0.01a
0.01a
0.43 0.09c 0.72
IE = infective endocarditis; CHD = congenital heart defect; TEC = thromboembolic complication; CVE = cerebrovasculare events; non-CVE = TEC below the neck. Italic only significant p-values. a P1 vs. P3 and P2 vs. P3. b Inclusive 5 cardiac catheterizations c P1 vs P3.
Of the 69 children with community acquired-IE, 12/69 (17%) had dental work 3 days to 4 weeks prior to IE diagnosis; 8 of these 12 patients (67%) had underlying CHD and 4/12 (33%) had received antibiotic prophylaxis. In contrast, of the 44 patients with nosocomial endocarditis, 26/44 (59%) had surgery prior to their IE diagnosis (cardiac surgery [n = 23], tonsillectomy [n = 1], scoliosis surgery [n = 1], liver transplant [n = 1]) and in 5/44 cases (11%) cardiac catheterization was performed. Overall, prior to IE diagnosis 50/113 (44%) patients had cardiac
surgery (including 5 patients with cardiac catheterizations) and 3/113 (3%) had non-cardiac surgery procedures as stated earlier. The IE prevalence of patients without prior surgery was comparable (60/113 [53%]. Regarding antibiotic prophylaxis, 43/113 patients (38%) received antibiotic prophylaxis prior to surgery and dental procedures and in 70/113 children (62%) no antibiotics were given. Of the 43 patients with prophylaxis, 12/43 (28%) had IE with TEC vs 27/70 (38%; p = 0.31) in the non-antibiotics group. 3.7. Antiplatelet or anticoagulation therapy
Table 2 Characteristics of patients with community acquired and nosocomial IE. Characteristics
Community acquired IE n (%)
Nosocomial IE n (%)
p value
n Median age (y) Female Underlying disease Cyanotic CHD Acyanotic CHD No CHD Thromboembolic complications CVE Non-CVE Vegetations Large (≥1 cm) Microorganisms Staphylococcus aureus Streptococcus viridans Surgical intervention Heart failure Mortality
69 (61) 10.8 32 (46)
44 (39) 0.7 14 (54)
b0.001 0.124
20 (29) 38 (55) 11 (16) 31 (44) 15 (22) 18 (26) 47 (68) 24 (35)
14 (32) 23 (52) 7 (16) 14 (32) 7 (16) 5 (11) 21 (47) 8 (18)
0.749 0.771 0.996 0.165 0.479 0.058 0.016 0.056
19 (27) 26 (38) 31 (45) 6 (9) 5 (7)
24 (54) 3 (7) 9 (20) 19 (44) 12 (27)
0.003 b0.001 0.009 b0.001 0.006
IE = infective endocarditis, CVE = cerebro-vascular event, CHD = congenital heart defect. Italic only significant p-values.
Eleven (10%) patients were on antiplatelet therapy (IE-related AIS [n = 5], stent implantation/vascular intervention [n = 4], antiinflammatory treatment for post-pericardiotomy syndrome [n = 2]). Eight (7%) children were treated with warfarin for mechanical valve replacement prior to their IE or after IE-related surgery. Two more patients received temporary anticoagulation for PE and septic peripheral TEC. Two additional patients were on unfractionated heparin for TEC and extracorporeal membrane oxygenation, respectively. Our review did not identify any major bleeding events related to antiplatelet or anticoagulant agents in the 20% of patients in this cohort that received either antiplatelet or anticoagulant agents. 3.8. Surgical intervention and time of surgery Surgical intervention was necessary in 40/113 patients (35%). Early surgery was performed in 23/40 cases (58%). Mortality between patients with early and late surgery was comparable (5/23 (22%) vs 4/17 (23%). Median age in the surgery group was significantly higher (10.3 years vs. 4.0 years, p b 0.001) compared to the non-surgery cases. Similarly, in this group more patients had community-acquiredIE (45% vs 20%) and significantly more vegetations (21/40; 52%; p b
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
4
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Table 3 Factors associated with IE-related mortality. Risk factor
Mortality n/n(%)
Univariate analysis
Multivariate analysis
p value
OR
95% CI
p value
b0.001
378.5
32–999
b0.0001
16/68 (24) 1/45 (2)
0.005
50
3.9–692
0.002
7/45 (16) 10/68 (15)
0.53
Age (y) b3 N3
8/41 (19) 9/72 (12)
0.32
CHD Yes No
12/95 (13) 5/18 (28)
0.09
Heart failure Yes 16/25 (64) No 1/88 (1) Vegetations Yes No TEC Yes No
Surgical intervention Yes 7/40 (17) No 10/73 (14) Total 17/113
0.59
CI = confidence interval; CHD = congenital heart defect; TEC = thromboembolic complication. Italic only significant p-values.
0.001). Children undergoing IE-related surgery had a significantly higher frequency of aortic (p = 0.01) and mitral (p = 0.08) valve involvement and severe valve dysfunction compared to the patients who did not need surgery. S. aureus-related IE was seen in 45% of cases (vs. 34% in the non-surgery group, p = 0.02). Patients with IErelated surgical intervention had a slightly higher mortality (20% vs. 17%, p = 0.31). Dominant causes for deaths were multiorgan failure and sepsis-related cardiac arrests. 3.9. Infective endocarditis-related mortality The overall mortality was 15% (17/113) and remained stable over the three study periods (Table 1). Within the deceased patients 8/17 (47%) were younger than 3 years of age, 10/17 (59%) had CHD (4 cyanotic), and 10/17 (59%) patients had a S. aureus infection identified (remaining germs were: S. viridans [1], C. albicans [2], H. influenza [1], Pseudomonas [1], BC negative [2]). Surgery due to the IE was done in 7/17 (41%) children (aortic valve destruction [n = 2], aortic root abscess [n = 1], mitral valve repair [n = 3], 3-valve disease [n = 1]). All 17 patients (100%) had vegetations and sepsis with multiorgan failure and 16/17 (94%) developed heart failure. In Table 3 univariable and multivariable analysis is presented. Heart failure was documented in 25/113 cases (22%) and was significantly more frequent in patients with nosocomial IE (44% vs 9%; p b 0.001) In univariable analysis, factors associated with IE-related mortality were heart failure and the presence of vegetations. Exploratory analysis showed vegetations and heart failure as the two risk factors independently associated with IE-related death. 4. Discussion Pediatric infective endocarditis is a rare disease, however its incidence appears to be increasing in the last decades [2,12,21–22]. This study demonstrates considerable epidemiological changes in pediatric IE at The Hospital for Sick Children (Toronto, Canada) from 1978 to 2008, a tertiary pediatric centre receiving IE cases from both the community and from other pediatric facilities with lower complexity care levels. In regards to the IE origin, in the most recent period (P3), 77% community acquired-IE were documented in patients who were
significantly older compared to patients from the earlier decades. This age profile is likely to have influenced our results, being contrary to the findings of Ashkenazi et al., who described a trend for higher IE rates in younger patients [23]. Consistent with the high median age in P3, we found the highest incidence of community acquired-IE with positive histories for dental procedures, usually done in older children. The higher proportion of older children in P3 was accompanied by an increasing rate of S. viridans associated CA-IE in this period. Conversely, in our population S. aureus caused nosocomial IE decreased over the periods. This results are comparable with data from Gupta et al. [24]. In our cohort 38% (43/113) received antibiotic prophylaxis. Prophylaxis did not reduce the onset of IE with TEC. Antibiotic management during the observed periods was according to the prior guidelines [18]. Due to the retrospective design, we cannot comment on the effectiveness of these prior guidelines. The distribution and change of IE-causing organisms over the periods may have been influenced by the changing antibiotic guidelines. Since our data collection ended in 2008, we were not able to study the influence of the AHA guidelines published 2007 [25]. Most recent guidelines recommend antibiotics only for prosthetic valves or material, cyanotic CHD and prior IE [15]. The rate of CHD was comparable in all decades which is consistent with data from Day et al. [2], but contrary to published data, suggesting an increasing incidence of CHD-related IE [3,26] With the improvement in survival rates of patients with complex CHD, there will appear more patients after surgical treatment with shunts or implanted devices, developing an inreasing risk population inclusive pediatric and adult patients with complex heart disease. However, the improvement of outcome in patients with CHD could also partially explain the increasing age of affected children over the periods. In 60% of our reviewed cases, vegetations were found by TTE or TEE. Over time TEE was performed more frequently. However, with a sensitivity up to 97%, TTE has been shown a sufficient method to detect endocarditis in young children, particularly useful for serial studies. For patients N 10 yrs. and N60 kg TEE has better sensitivity [27]. However, for some children with complex CHD, computerized tomography or magnetic resonance imaging with 3-D-reconstruction might optimize preoperative diagnostic evaluation. Over the three time periods (P1 to P3), an increasing rate of vegetations were found. This finding might be explained by the advances in imaging techniques but may also be influenced by the higher rate of CA-IE with delayed diagnosis. The overall rate of TEC (40%) as well as the high TEC incidence in patients with large vegetations was consistent with published data, reporting embolization rates between 13% and 50% [4,5]. Our study showed a trend for an increasing incidence of non-CVE over the 3 decades. This finding may be related to the use of more sensitive diagnostic tools and to the delayed diagnosis of some CA-IE cases. Older children presenting in the community with fever are usually initially treated with antibiotics for suspected other causes of infection, such as tonsillitis or pneumonia. Cerebrovascular events were documented in 19% and non-CVE in 20% of our cases, prevalences which are comparable to previously published adult data [7,9]. However, contrary to existing data [9], TEC was not associated with mortality in our investigation. In univariable analysis for TEC we only found a significant association of age above three years, reflecting the increasing age and rate of community acquired-IE over the periods. Our data revealed a surgical intervention rate of 35% with 58% within 3 days after diagnosis of IE. The high rate of vegetations and aneurysms is reflected by the rate of S. aureus infection in the surgery group. Usually S. aureus causes more serious and destructive endocarditis. [28] We found comparable mortality rates in patients with early vs late surgery. However, early surgery is currently recommended to prevent heart failure, uncontrolled infection and TEC. These are also the most common reasons for surgical intervention [12,29]. Heart failure has been reported to be independently associated with increased risk for mortality in adult patients [5,30]. Our data revealed heart failure as significant risk factor for mortality, which has also
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
been recently reported by Jomaa et al. [31]. Consistent with the study of Yoshinaga et al. we found a high rate of heart failure in the patients with nosocomial and postoperative IE, indicating underlying acute illness and acute courses, e.g. perforation of valve leaflets, increased regurgitation with ventricular dysfunction and heart failure in this significantly younger patient group [21]. The overall mortality rate in our population was 15%, comparable to recently published data [31]. Consistent with published literature [7,9, 32], the present data suggest an increased mortality for patients with vegetations. The presence of vegetations alone, regardless of their size, may have caused enough valve destruction and regurgitation to lead to significant impaired survival. Despite differences in age, the rate of community acquired-IE, surgical intervention and the incidence of TEC, mortality remained stable over the 3 study time periods. One of several factors influencing mortality may have been the improved intensive care management of sepsis and heart failure which were the most common causes for death. Our study has some limitations. The study was performed at a single tertiary care institution including a patient cohort from a large referral center. The fact that many patients with IE were referred from other hospitals may have led to patient selection bias with limitations to the generalizability of these results. In conclusion, our single tertiary centre study reports considerable epidemiological changes in pediatric IE over 30 years. Higher age of affected children and an increasing incidence of CA-IE and TEC may reflect delayed diagnosis in older children. Mortality rate remained stable over time; heart failure and the presence of vegetations were associated with an increased mortality. These preliminary data need to be confirmed by prospective data.
Funding source K. Thom is a recipient of the Baxter Bioscience Fellowship funding in Pediatric Hemostasis and Thrombosis at the Hospital for Sick Children, Toronto/Canada. There was no external funding for this project and manuscript. No conflicts of interests.
Declaration of interest Financial disclosure statement: The authors have no financial relationship relevant to this article.
Conflict of interest statement The authors have no conflicts of interest relevant to this article to disclose.
Contributors' statement Dr. Thom conceptualized and designed the study, carried out the initial analyses, drafted and finalized the manuscript, and approved the final manuscript as submitted. Dr. Hanslik contributed substantially in finalizing the manuscript and approved the final manuscript as submitted. Drs. Russel, Williams, Sivaprakasam and Allen contributed to design the study, and approved the final manuscript as submitted. Dr. Male contributed substantially to the final and new calculations requested by the reviewers and approved the final manuscript as submitted. Dr. Brandão carried out additional analyses, reviewed and revised the manuscript and approved the final manuscript as submitted.
5
Acknowledgement We thank Derek Stephens from the Hospital for Sick Children, Toronto, Canada for assistance with the statistical analysis. References [1] W. Knirsch, D. Nadal, Infective endocarditis in congenital heart disease, Eur. J. Pediatr. 170 (9) (2011) 1111–1127. [2] M.D. Day, K. Gauvreau, S. Shulman, et al., Characteristics of children hospitalized with infective endocarditis, Circulation 119 (6) (2009) 865–870. [3] L.B. Rosenthal, K.N. Feja, S.M. Levasseur, et al., The changing epidemiology of pediatric endocarditis at a children's hospital over seven decades, Pediatr. Cardiol. 31 (6) (2010) 813–820. [4] G. Habib, Embolic risk in subacute bacterial endocarditis: determinants and role of transesophageal echocardiography, Curr. Cardiol. Rep. 5 (2) (2003) 129–136. [5] F. Thuny, G. Di Salvo, O. Belliard, et al., Risk of embolism and death in infective endocarditis: prognostic value of echocardiography: a prospective multicenter study, Circulation 112 (1) (2005) 69–75. [6] A. Saxena, N. Aggarwal, P. Gupta, et al., Predictors of embolic events in pediatric infective endocarditis, Indian Heart J. 61 (3) (2009) 242–245. [7] D.R. Murdoch, G.R. Corey, B. Hoen, et al., Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study, Arch. Intern. Med. 169 (5) (2009) 463–473. [8] S.J. Lee, Oh. SS, D.S. Lim, et al., Clinical significance of cerebrovascular complications in patients with acute infective endocarditis: a retrospective analysis of a 12-year single-center experience, BMC Neurol. 14 (2014) 30. [9] E. Garcia-Cabrera, N. Fernandez-Hidalgo, B. Almirante, et al., Neurological complications of infective endocarditis: risk factors, outcome, and impact of cardiac surgery: a multicenter observational study, Circulation 127 (23) (2013) 2272–2284. [10] C. Venkatesan, M.S. Wainwright, Pediatric endocarditis and stroke: a single-center retrospective review of seven cases, Pediatr. Neurol. 38 (4) (2008) 243–247. [11] E.E. Hill, P. Herijgers, P. Claus, et al., Infective endocarditis: changing epidemiology and predictors of 6-month mortality: a prospective cohort study, Eur. Heart J. 28 (2) (2007) 196–203. [12] R.S. Baltimore, M. Gewitz, L.M. Baddour, et al., Infective endocarditis in childhood: 2015 update: a scientific statement from the American Heart Association, Circulation 132 (15) (2015) 1487–1515. [13] L.I. Kupferwasser, M.R. Yeaman, S.M. Shapiro, et al., Acetylsalicylic acid reduces vegetation bacterial density, hematogenous bacterial dissemination, and frequency of embolic events in experimental Staphylococcus aureus endocarditis through antiplatelet and antibacterial effects, Circulation 99 (21) (1999) 2791–2797. [14] P. Monagle, A.K. Chan, N.A. Goldenberg, et al., Antithrombotic therapy in neonates and children: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines, Chest 141 (2 Suppl) (2012) e737S–801S. [15] G. Habib, P. Lancellotti, M.J. Antunes, et al., ESC Guidelines for the management of infective endocarditis: the Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM), Eur. Heart J. 36 (44) (2015) 3075–3128 2015. [16] K.L. Chan, J.G. Dumesnil, B. Cujec, et al., A randomized trial of aspirin on the risk of embolic events in patients with infective endocarditis, J. Am. Coll. Cardiol. 42 (5) (2003) 775–780. [17] J.S. Li, D.J. Sexton, N. Mick, et al., Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis, Clin. Infect. Dis. 30 (4) (2000) 633–638. [18] A.S. Dajani, K.A. Taubert, W. Wilson, et al., Prevention of bacterial endocarditis. Recommendations by the American Heart Association, Circulation 96 (1) (1997) 358–366. [19] P. Ferrieri, M.H. Gewitz, M.A. Gerber, et al., Unique features of infective endocarditis in childhood, Circulation 105 (17) (2002) 2115–2126. [20] L.G. Mitchell, N.A. Goldenberg, C. Male, et al., Definition of clinical efficacy and safety outcomes for clinical trials in deep venous thrombosis and pulmonary embolism in children, J. Thromb. Haemost. 9 (9) (2011) 1856–1858. [21] M. Yoshinaga, K. Niwa, A. Niwa, et al., Risk factors for in-hospital mortality during infective endocarditis in patients with congenital heart disease, Am. J. Cardiol. 101 (1) (2008) 114–118. [22] K. Keller, R.S. von Bardeleben, M.A. Ostad, et al., Temporal trends in the prevalence of infective endocarditis in Germany between 2005 and 2014, Am. J. Cardiol. 119 (2017) 317–322. [23] S. Ashkenazi, O. Levy, L. Blieden, Trends of childhood infective endocarditis in Israel with emphasis on children under 2 years of age, Pediatr. Cardiol. 18 (6) (1997) 419–424. [24] S. Gupta, A. Sakhuja, E. McGrath, et al., Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States, Congenit. Heart Dis. 12 (2) (2017) 196–201. [25] W. Wilson, K.A. Taubert, M. Gewitz, 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 (15) (2007) 1736–1754. [26] D. Rushani, J.S. Kaufman, R. Ionescu-Ittu, et al., Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors, Circulation 128 (13) (2013) 1412–1419.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
6
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
[27] J.S. Penk, C.L. Webb, S.T. Shulman, et al., Echocardiography in pediatric infective endocarditis, Pediatr. Infect. Dis. J. 30 (12) (2011) 1109–1111. [28] C. Chirouze, C.H. Cabell, V.G. Fowler Jr., et al., Prognostic factors in 61 cases of Staphylococcus aureus prosthetic valve infective endocarditis from the International Collaboration on Endocarditis merged database, Clin. Infect. Dis. 38 (9) (2004) 1323–1327. [29] D.H. Kang, Y.J. Kim, S.H. Kim, et al., Early surgery versus conventional treatment for infective endocarditis, N. Engl. J. Med. 366 (26) (2012) 2466–2473.
[30] R. Hasbun, H.R. Vikram, L.A. Barakat, et al., Complicated left-sided native valve endocarditis in adults: risk classification for mortality, JAMA 289 (15) (2003) 1933–1940. [31] W. Jomaa, I. Ben Ali, D. Abid, et al., Clinical features and prognosis of infective endocarditis in children: insights from a Tunisian multicentre registry, Arch. Cardiovasc. Dis. (2017) (in press). [32] F. Thuny, J.F. Avierinos, C. Tribouilloy, et al., Impact of cerebrovascular complications on mortality and neurologic outcome during infective endocarditis: a prospective multicentre study, Eur. Heart J. 28 (9) (2007) 1155–1161.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30 years K. Thom a,d, A. Hanslik d, J.L. Russell b, S. Williams a, P. Sivaprakasam a, U. Allen c, C. Male d, L.R. Brandão a,⁎ a
Pediatric Haematology/Oncology, The Hospital for Sick Children, Toronto, Canada Pediatric Cardiology, Labatt Family Heart Centre, The Hospital for Sick Children, Toronto, Canada Infectious Disease, The Hospital for Sick Children, Toronto, Canada d Division of Pediatric Cardiology, Department of Children and Adolescent Medicine, Medical University Vienna, Austria b c
a r t i c l e
i n f o
Article history: Received 27 April 2017 Received in revised form 13 October 2017 Accepted 23 October 2017 Available online xxxx Keywords: Pediatric endocarditis Thrombosis Embolism Heart failure Mortality
a b s t r a c t Background: Pediatric infective endocarditis (IE) has been associated with high morbidity and mortality, mostly related to thromboembolic complications (TEC). The objective of our study was to describe the experience in children with IE and to review the changes over a thirty-year period, regarding origin of IE, incidence of vegetations, TEC and their respective morbidity and mortality rates. Methods: A retrospective chart review of children aged 0–18 years with IE defined by the Duke Criteria and admitted to The Hospital for Sick Children, was conducted. Data were divided into three periods (P); P1 (1979–1988); P2 (1989–1998); and P3 (1999–2008). Results: The study included 113 patients, median age 7 yrs.; females: 46 (41%), congenital heart defects 95 (84%), comparable in all periods. Overall, cardiac vegetations were found in 68/113 patients (60%); large vegetations (≥1 cm) in 32 patients (28%). Fourty-five (45/133 [40%]) TEC were documented, 22 patients (20%) developed cerebrovascular events (CVE) and 23 patients (20%) had non-CVE. Patients diagnosed during P3 were older, had more vegetations (p b 0.05), and a higher incidence of community acquired-IE (p b 0.05). Overall, mortality was 15%, comparable in all periods. Significant risk factors for mortality were vegetations (HR 6.44; 95% CI: 2.07– 20.01, p = 0.002) and heart failure (HR 28.39; 95% CI: 10.49–76.85, p b 0.001). Conclusions: Over the study period, we report a growing incidence of community acquired pediatric IE in older children accompanied by an increasing rate of TEC. Heart failure and vegetations were associated with an increased mortality. These preliminary data need to be confirmed by prospective data. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Infective endocarditis (IE) is a rare but serious condition in children with an incidence estimated between 0.34 and 0.64 cases per 100,000 per year [1]. To date, the diagnosis of pediatric IE is still challenging and frequently delayed, with a reported mortality in children ranging from 4 to 25% [2,3]. According to adult data, systemic embolism occurs in 13 to 50% of patients with IE. Thromboembolic complications (TEC) into major arteries and organs significantly contribute to disease morbidity [4–6]. Cerebro-vascular events (CVE), including acute ischemic
Abbreviations: IE, infective endocarditis; TEC, thromboembolic complication; P1, 2, 3, periods 1, 2 and 3; (Non)-CVE, (non)-cerebrovascular events; CA-IE, community acquired-infective endocarditis; AIS, arterial-ischemic stroke; CHD, congenital heart disease; PE, pulmonary embolism; TTE, transthoracic echocardiography; TEE, transesophageal echocardiography; BC, blood culture. ⁎ Corresponding author at: The Hospital for Sick Children, 555 University Avenue, Black Wing, Room 10412, Toronto, ON M5G 1X8, Canada. E-mail address: [email protected] (L.R. Brandão).
strokes (AIS), are reported in 20 to 40% of patients with IE [7–9]. For children, a small retrospective series reported 6% IE related AIS [10]. Location and size of intracardiac vegetations have been identified as risk factors in various studies for cardioembolic AIS in adults [11]. Initiation of prompt antimicrobial treatment is considered the mainstay in both the treatment of IE and in the prevention of thrombotic complications, including AIS [12]. Despite the importance of thrombotic events in patients with IE, the use of anticoagulation and antiplatelet therapy is highly controversial. On the one hand, an inverse dose-dependent response to acetylsalicylic acid on vegetation growth, microbial proliferation and renal embolization has been described for experimental Staphylococcus aureus (S. aureus)-induced IE [13]. On the other hand, uncertainty about the appropriateness of such treatment modalities continues, as reflected by the most recent pediatric and adult antithrombotic and IE guidelines [12,14,15]. Currently there is no evidence to support anticoagulation or antiplatelet therapy for patients with IE [16]. The objective of the current study was to review a single-center experience with IE in children and to describe changes over 30 years,
https://doi.org/10.1016/j.ijcard.2017.10.085 0167-5273/© 2017 Elsevier B.V. All rights reserved.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
2
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
regarding origin of IE, incidence of vegetations and TEC. In addition, we aimed to describe risk factors for IE-related TEC and mortality. 2. Methods A retrospective chart review including only pediatric patients with definite IE, as per modified Duke Criteria, was conducted [17] In patients without vegetations, IE diagnosis was made according to positive blood culture (BC) or to minor criteria including prolonged fever, CHD, new regurgitation murmur, and vascular or embolic phenomena. Thromboembolic complications were included if confirmed radiologically and occurred after confirmation of IE as per the modified Duke Criteria for vascular complications [17]. Consecutive patients diagnosed with IE, admitted to The Hospital for Sick Children, Toronto (Canada) between January 1978 and January 2008 were eligible. Patients whose admitting and/or discharge diagnoses included IE were retrieved from either the Cardiology or the Thrombosis service database. Permission to use and analyze the data was obtained from the local Research Ethics Board of The Hospital for Sick Children, Toronto (Canada). Demographic data obtained included: age, gender, underlying disease, congenital heart (CHD) type (cyanotic or acyanotic), previous cardiac surgery, cardiac interventions or previous IE as well as non-cardiac interventions prior to IE diagnosis (e.g. other surgery, vaccinations, dental procedures). A standardized case report form was completed for each patient to maximize data collection consistency and quality. 2.1. Definitions Nosocomial infection consisted of any infection, occurring ≥72 h after hospital admission [18]. Postoperative IE was defined as IE diagnosed within 6 months after surgery [19]. Thromboembolic complications (TEC) included cerebrovascular events (CVE) or nonCVE, as follows: a) CVE, including AIS, mycotic aneurysms or hemorrhages; and b) nonCVE, defined as radiologically-confirmed thromboembolism outside the central nervous system. The indication to perform radiological investigation was the clinical suspicion of a TEC. Major bleeding events secondary to either antiplatelet or anticoagulant use were defined as per internationally accepted criteria [20]. Surgical intervention encompassed any surgery related to IE: early surgery comprised surgery up to three days after diagnosis of IE. Heart failure was defined and confirmed by cardiology staff physician; this information was confirmed in the consultation notes. Due to the retrospective study design, mortality included only perioperative deaths or endocarditis-related mortality up to 30 days after the diagnosis of IE was confirmed. 2.2. Statistical methods To determine trends in childhood IE over the surveyed 30 years, the total observation time was divided into equal time periods of 10 years, each: period 1 (P1; 1979–1988), period 2 (P2; 1989–1998) and period 3 (P3; 1999–2008). Descriptive statistics including percentages and medians were calculated. Mann– Whitney-U-test was used for comparison between groups with continuous variables and Chi-square-test for comparison of categorical variables. Univariable and stepwise multivariable logistic regression were conducted for exploratory analysis of risk factors associated with TEC and mortality. Variables showing a trend for an association with the dependent variable (p value ≤ 0.1) in the univariable analysis were included in the multivariable analysis. Significance level was set up at alpha b0.5.
3. Results 3.1. Demographic data The initial cohort using the diagnostic code for IE comprised 152 patients. Thirty nine patients were excluded for the following reasons: i) no definite IE according to Duke criteria (n = 30), ii) unavailable or incomplete data (n = 7), iii) postmortem diagnosis of IE (n = 2). The final study cohort consisted of 113 patients, aged 1 day to 18 yrs., with 41 patients (36%) aged ≤3 yrs. The median age was 7.0 yrs., 46 patients (41%) were females. Underlying CHD was present in 95/113 patients (84%), acyanotic CHD in 61/113 patients (54%) and cyanotic CHD in 34/113 children (30%). 3.2. Characteristics of the surveyed periods Table 1 shows characteristics of patients in the surveyed time periods. Period 1 (P1) comprised 31 (27%) patients, P2: 47 (42%) and P3: 35 (31%) children. Median age at diagnosis was significantly (p b 0.001) higher in P3 (12.3 years) compared to P1 (3.0 years) and P2 (4.0 years, p b 0.001). In P3, a significantly higher incidence of community-acquired IE (77% [P3] vs. 48% [P1] vs. 57% [P2], p = 0.01)
existed compared to the prior two study decades. Additionally, patients in P3 had higher incidences of vegetations (71% [P3] vs. 45% [P1] vs. 62% [P2], p = 0.01) with comparable CVE, but increasing non-CVE (26% [P3] vs. 13% [P1], p = 0.43). 3.3. Thromboembolic complications Fourty-five thrombotic events (45/113 [40%]) were documented. Cerebrovascular events (CVE) were documented in 22/113 children (19%). An acute ischemic stroke was diagnosed 18/113 patients (16%) and cerebral mycotic aneurysms in 4/113 children (3%). Non-CVE was present in 23/113 cases (20%). Among those with non-CVE, 9/113 children (8%) developed pulmonary embolism, 6 (5%) had kidney infarcts, 5 (4%) had splenic infarcts, and 3 (3%) had osteomyelitis complicated by bone abscess formation. Overall, children above 3 years had longer (median 17 days) fever periods with prolonged clinical courses compared to patients younger than 3 years (median 12 days, n.s.). An exploratory analysis was conducted to identify risk factors associated with TEC. The logistic regression found age above 3 years to be significantly associated with TEC, including CVE and nonCVE (p = 0.01). Congenital heart diseases (p = 0.33), causative microorganisms (p = 0.95) and heart failure (p = 0.51) were not associated with TEC. 3.4. Vegetations Vegetations were identified by either transthoracic (TTE) or transesophageal echocardiography (TEE) in 68/113 cases (60%). Location of vegetations is given in Table 1. During P1, TEE was not done whereas in P2 15/29 patients (52%) with vegetations had TEE with 13/15 (87%) positive exams for vegetations (vegetation only seen in TEE [n = 1], lesion only visible in TTE [n = 1]). During P3 17/25 children (68%) who had developed vegetations underwent TEE (vegetation only seen by TEE [n = 2]). Overall, from 68 patients with vegetations, 45/68 (66%) developed TEC. Whereas of the 42 patients without vegetations 12/42 (29%) developed TEC (8 CVE 6 non-CVE, 2 both; p b 0.001). In another three patients there was no information regarding vegetation. In 32/113 children (27%) large (≥1 cm) vegetations were documented and of these, 18 children (18/32 [56%]) developed TEC. 3.5. Causative microorganisms The causative microorganism was identified by BC in 105/113 children (93%). Staphylococcus aureus was found to be the causative microorganism in 43 cases (38%). The remaining patients had positive BC for Streptococcus viridans (29 [26%]), Staphylococcus epidermidis (14 [12%]), Streptococcus pneumoniae (3 [3%]), Candida albicans (2 [2%]), combined organisms (6 [5%]) and miscellaneous (8 [7%]) microorganisms (Haemophilus influenza [1], Enterococcus [2], Actinobacillus sp. [1], Pseudomonas aeruginosa [2], Pasteurella sp. [1], and Streptococcus abiotrophia [1]). Eight patients (7%) had BC-negative endocarditis. Of these, 5 children were diagnosed by a biopsy after surgery and 3 children by clinical criteria according to the Duke Criteria. During the entire study period, an increasing rate of S. viridans and decreasing rate of S. aureus as the causing organisms were found. 3.6. Origin of IE and antibiotic prophylaxis Overall, community acquired-IE was seen in 69/113 (61%) and nosocomial IE in 44/113 cases (39%) with comparable proportions of CHD over time. Children with community acquired-IE were significantly older than in the nosocomial group (median age 10.8 yrs. vs. 0.7 yrs., p b 0.001), had more vegetations (p = 0.016) and a higher incidence of IE-related surgical interventions (p = 0.009). S. viridans was the dominant causative microorganism in CA-IE (p b 0.001) compared with IE in the nosocomial group (Table 2).
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
3
Table 1 Characteristics of all patients in the surveyed periods. Characteristics
Overall n (%)
Period 1 (1979–1988)
Period 2 (1989–1998)
Period 3 (1999–2008)
n Age at diagnosis (y) 0–3 3–11 11–18 Median age (Q1;Q3) Endocarditis type Community acquired Nosocomial Underlying disease Cyanotic CHD Acyanotic CHD No CHD Prior cardiac surgery/interv. Prior non-cardiac surgery Causing organisms Staphyococcus aureus Streptococcus viridans Vegetations Multiple locations Aortic/Mitralvalve Right sided, conduits, prosthetic valves, BTS Septal Septic aneurysms CVE Non CVE IE surgical intervention IE related mortality
113
31 (27%)
47 (42%)
35 (31%)
41 (36) 34 (30) 38 (34) 7.0 (1.1;12.8)
15 (48) 13 (42) 3 (10) 3.0 (1.0;8.6)
21 (45) 13 (28) 13 (28) 4.0 (0.4; 11.6)
5 (14) 8 (23) 22 (63) 12.3 (7.8; 14.8)
69 (61) 44 (39)
15 (48) 16 (52)
27 (57) 20 (42)
27 (77) 8 (23)
34 (30) 61 (54) 18 (16) 50 (44) 3 (3)
13 (42) 14 (45) 4 (13) 22 (71)b 0
14 (30) 26 (55) 7 (15) 19 (40) 3 (6)
7 (20) 21 (60) 7 (20) 9 (26) 0
43 (38) 29 (26) 68 (60) 18 38 40 6 3 22 (19) 23 (20) 40 (35) 17 (15)
14 (45) 4 (13) 14 (45) 6 5 9 2 2 5 (16) 4 (13) 8 (26) 5 (16)
22 (47) 12 (25) 29 (62) 7 14 19 3 0 9 (19) 10 (21) 17 (36) 7 (15)
7 (20) 13 (37) 25 (71) 5 19 12 1 1 8 (23) 9 (26) 15 (43) 5 (14)
p value
b0.001a
0.01a
0.01a
0.43 0.09c 0.72
IE = infective endocarditis; CHD = congenital heart defect; TEC = thromboembolic complication; CVE = cerebrovasculare events; non-CVE = TEC below the neck. Italic only significant p-values. a P1 vs. P3 and P2 vs. P3. b Inclusive 5 cardiac catheterizations c P1 vs P3.
Of the 69 children with community acquired-IE, 12/69 (17%) had dental work 3 days to 4 weeks prior to IE diagnosis; 8 of these 12 patients (67%) had underlying CHD and 4/12 (33%) had received antibiotic prophylaxis. In contrast, of the 44 patients with nosocomial endocarditis, 26/44 (59%) had surgery prior to their IE diagnosis (cardiac surgery [n = 23], tonsillectomy [n = 1], scoliosis surgery [n = 1], liver transplant [n = 1]) and in 5/44 cases (11%) cardiac catheterization was performed. Overall, prior to IE diagnosis 50/113 (44%) patients had cardiac
surgery (including 5 patients with cardiac catheterizations) and 3/113 (3%) had non-cardiac surgery procedures as stated earlier. The IE prevalence of patients without prior surgery was comparable (60/113 [53%]. Regarding antibiotic prophylaxis, 43/113 patients (38%) received antibiotic prophylaxis prior to surgery and dental procedures and in 70/113 children (62%) no antibiotics were given. Of the 43 patients with prophylaxis, 12/43 (28%) had IE with TEC vs 27/70 (38%; p = 0.31) in the non-antibiotics group. 3.7. Antiplatelet or anticoagulation therapy
Table 2 Characteristics of patients with community acquired and nosocomial IE. Characteristics
Community acquired IE n (%)
Nosocomial IE n (%)
p value
n Median age (y) Female Underlying disease Cyanotic CHD Acyanotic CHD No CHD Thromboembolic complications CVE Non-CVE Vegetations Large (≥1 cm) Microorganisms Staphylococcus aureus Streptococcus viridans Surgical intervention Heart failure Mortality
69 (61) 10.8 32 (46)
44 (39) 0.7 14 (54)
b0.001 0.124
20 (29) 38 (55) 11 (16) 31 (44) 15 (22) 18 (26) 47 (68) 24 (35)
14 (32) 23 (52) 7 (16) 14 (32) 7 (16) 5 (11) 21 (47) 8 (18)
0.749 0.771 0.996 0.165 0.479 0.058 0.016 0.056
19 (27) 26 (38) 31 (45) 6 (9) 5 (7)
24 (54) 3 (7) 9 (20) 19 (44) 12 (27)
0.003 b0.001 0.009 b0.001 0.006
IE = infective endocarditis, CVE = cerebro-vascular event, CHD = congenital heart defect. Italic only significant p-values.
Eleven (10%) patients were on antiplatelet therapy (IE-related AIS [n = 5], stent implantation/vascular intervention [n = 4], antiinflammatory treatment for post-pericardiotomy syndrome [n = 2]). Eight (7%) children were treated with warfarin for mechanical valve replacement prior to their IE or after IE-related surgery. Two more patients received temporary anticoagulation for PE and septic peripheral TEC. Two additional patients were on unfractionated heparin for TEC and extracorporeal membrane oxygenation, respectively. Our review did not identify any major bleeding events related to antiplatelet or anticoagulant agents in the 20% of patients in this cohort that received either antiplatelet or anticoagulant agents. 3.8. Surgical intervention and time of surgery Surgical intervention was necessary in 40/113 patients (35%). Early surgery was performed in 23/40 cases (58%). Mortality between patients with early and late surgery was comparable (5/23 (22%) vs 4/17 (23%). Median age in the surgery group was significantly higher (10.3 years vs. 4.0 years, p b 0.001) compared to the non-surgery cases. Similarly, in this group more patients had community-acquiredIE (45% vs 20%) and significantly more vegetations (21/40; 52%; p b
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
4
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
Table 3 Factors associated with IE-related mortality. Risk factor
Mortality n/n(%)
Univariate analysis
Multivariate analysis
p value
OR
95% CI
p value
b0.001
378.5
32–999
b0.0001
16/68 (24) 1/45 (2)
0.005
50
3.9–692
0.002
7/45 (16) 10/68 (15)
0.53
Age (y) b3 N3
8/41 (19) 9/72 (12)
0.32
CHD Yes No
12/95 (13) 5/18 (28)
0.09
Heart failure Yes 16/25 (64) No 1/88 (1) Vegetations Yes No TEC Yes No
Surgical intervention Yes 7/40 (17) No 10/73 (14) Total 17/113
0.59
CI = confidence interval; CHD = congenital heart defect; TEC = thromboembolic complication. Italic only significant p-values.
0.001). Children undergoing IE-related surgery had a significantly higher frequency of aortic (p = 0.01) and mitral (p = 0.08) valve involvement and severe valve dysfunction compared to the patients who did not need surgery. S. aureus-related IE was seen in 45% of cases (vs. 34% in the non-surgery group, p = 0.02). Patients with IErelated surgical intervention had a slightly higher mortality (20% vs. 17%, p = 0.31). Dominant causes for deaths were multiorgan failure and sepsis-related cardiac arrests. 3.9. Infective endocarditis-related mortality The overall mortality was 15% (17/113) and remained stable over the three study periods (Table 1). Within the deceased patients 8/17 (47%) were younger than 3 years of age, 10/17 (59%) had CHD (4 cyanotic), and 10/17 (59%) patients had a S. aureus infection identified (remaining germs were: S. viridans [1], C. albicans [2], H. influenza [1], Pseudomonas [1], BC negative [2]). Surgery due to the IE was done in 7/17 (41%) children (aortic valve destruction [n = 2], aortic root abscess [n = 1], mitral valve repair [n = 3], 3-valve disease [n = 1]). All 17 patients (100%) had vegetations and sepsis with multiorgan failure and 16/17 (94%) developed heart failure. In Table 3 univariable and multivariable analysis is presented. Heart failure was documented in 25/113 cases (22%) and was significantly more frequent in patients with nosocomial IE (44% vs 9%; p b 0.001) In univariable analysis, factors associated with IE-related mortality were heart failure and the presence of vegetations. Exploratory analysis showed vegetations and heart failure as the two risk factors independently associated with IE-related death. 4. Discussion Pediatric infective endocarditis is a rare disease, however its incidence appears to be increasing in the last decades [2,12,21–22]. This study demonstrates considerable epidemiological changes in pediatric IE at The Hospital for Sick Children (Toronto, Canada) from 1978 to 2008, a tertiary pediatric centre receiving IE cases from both the community and from other pediatric facilities with lower complexity care levels. In regards to the IE origin, in the most recent period (P3), 77% community acquired-IE were documented in patients who were
significantly older compared to patients from the earlier decades. This age profile is likely to have influenced our results, being contrary to the findings of Ashkenazi et al., who described a trend for higher IE rates in younger patients [23]. Consistent with the high median age in P3, we found the highest incidence of community acquired-IE with positive histories for dental procedures, usually done in older children. The higher proportion of older children in P3 was accompanied by an increasing rate of S. viridans associated CA-IE in this period. Conversely, in our population S. aureus caused nosocomial IE decreased over the periods. This results are comparable with data from Gupta et al. [24]. In our cohort 38% (43/113) received antibiotic prophylaxis. Prophylaxis did not reduce the onset of IE with TEC. Antibiotic management during the observed periods was according to the prior guidelines [18]. Due to the retrospective design, we cannot comment on the effectiveness of these prior guidelines. The distribution and change of IE-causing organisms over the periods may have been influenced by the changing antibiotic guidelines. Since our data collection ended in 2008, we were not able to study the influence of the AHA guidelines published 2007 [25]. Most recent guidelines recommend antibiotics only for prosthetic valves or material, cyanotic CHD and prior IE [15]. The rate of CHD was comparable in all decades which is consistent with data from Day et al. [2], but contrary to published data, suggesting an increasing incidence of CHD-related IE [3,26] With the improvement in survival rates of patients with complex CHD, there will appear more patients after surgical treatment with shunts or implanted devices, developing an inreasing risk population inclusive pediatric and adult patients with complex heart disease. However, the improvement of outcome in patients with CHD could also partially explain the increasing age of affected children over the periods. In 60% of our reviewed cases, vegetations were found by TTE or TEE. Over time TEE was performed more frequently. However, with a sensitivity up to 97%, TTE has been shown a sufficient method to detect endocarditis in young children, particularly useful for serial studies. For patients N 10 yrs. and N60 kg TEE has better sensitivity [27]. However, for some children with complex CHD, computerized tomography or magnetic resonance imaging with 3-D-reconstruction might optimize preoperative diagnostic evaluation. Over the three time periods (P1 to P3), an increasing rate of vegetations were found. This finding might be explained by the advances in imaging techniques but may also be influenced by the higher rate of CA-IE with delayed diagnosis. The overall rate of TEC (40%) as well as the high TEC incidence in patients with large vegetations was consistent with published data, reporting embolization rates between 13% and 50% [4,5]. Our study showed a trend for an increasing incidence of non-CVE over the 3 decades. This finding may be related to the use of more sensitive diagnostic tools and to the delayed diagnosis of some CA-IE cases. Older children presenting in the community with fever are usually initially treated with antibiotics for suspected other causes of infection, such as tonsillitis or pneumonia. Cerebrovascular events were documented in 19% and non-CVE in 20% of our cases, prevalences which are comparable to previously published adult data [7,9]. However, contrary to existing data [9], TEC was not associated with mortality in our investigation. In univariable analysis for TEC we only found a significant association of age above three years, reflecting the increasing age and rate of community acquired-IE over the periods. Our data revealed a surgical intervention rate of 35% with 58% within 3 days after diagnosis of IE. The high rate of vegetations and aneurysms is reflected by the rate of S. aureus infection in the surgery group. Usually S. aureus causes more serious and destructive endocarditis. [28] We found comparable mortality rates in patients with early vs late surgery. However, early surgery is currently recommended to prevent heart failure, uncontrolled infection and TEC. These are also the most common reasons for surgical intervention [12,29]. Heart failure has been reported to be independently associated with increased risk for mortality in adult patients [5,30]. Our data revealed heart failure as significant risk factor for mortality, which has also
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
been recently reported by Jomaa et al. [31]. Consistent with the study of Yoshinaga et al. we found a high rate of heart failure in the patients with nosocomial and postoperative IE, indicating underlying acute illness and acute courses, e.g. perforation of valve leaflets, increased regurgitation with ventricular dysfunction and heart failure in this significantly younger patient group [21]. The overall mortality rate in our population was 15%, comparable to recently published data [31]. Consistent with published literature [7,9, 32], the present data suggest an increased mortality for patients with vegetations. The presence of vegetations alone, regardless of their size, may have caused enough valve destruction and regurgitation to lead to significant impaired survival. Despite differences in age, the rate of community acquired-IE, surgical intervention and the incidence of TEC, mortality remained stable over the 3 study time periods. One of several factors influencing mortality may have been the improved intensive care management of sepsis and heart failure which were the most common causes for death. Our study has some limitations. The study was performed at a single tertiary care institution including a patient cohort from a large referral center. The fact that many patients with IE were referred from other hospitals may have led to patient selection bias with limitations to the generalizability of these results. In conclusion, our single tertiary centre study reports considerable epidemiological changes in pediatric IE over 30 years. Higher age of affected children and an increasing incidence of CA-IE and TEC may reflect delayed diagnosis in older children. Mortality rate remained stable over time; heart failure and the presence of vegetations were associated with an increased mortality. These preliminary data need to be confirmed by prospective data.
Funding source K. Thom is a recipient of the Baxter Bioscience Fellowship funding in Pediatric Hemostasis and Thrombosis at the Hospital for Sick Children, Toronto/Canada. There was no external funding for this project and manuscript. No conflicts of interests.
Declaration of interest Financial disclosure statement: The authors have no financial relationship relevant to this article.
Conflict of interest statement The authors have no conflicts of interest relevant to this article to disclose.
Contributors' statement Dr. Thom conceptualized and designed the study, carried out the initial analyses, drafted and finalized the manuscript, and approved the final manuscript as submitted. Dr. Hanslik contributed substantially in finalizing the manuscript and approved the final manuscript as submitted. Drs. Russel, Williams, Sivaprakasam and Allen contributed to design the study, and approved the final manuscript as submitted. Dr. Male contributed substantially to the final and new calculations requested by the reviewers and approved the final manuscript as submitted. Dr. Brandão carried out additional analyses, reviewed and revised the manuscript and approved the final manuscript as submitted.
5
Acknowledgement We thank Derek Stephens from the Hospital for Sick Children, Toronto, Canada for assistance with the statistical analysis. References [1] W. Knirsch, D. Nadal, Infective endocarditis in congenital heart disease, Eur. J. Pediatr. 170 (9) (2011) 1111–1127. [2] M.D. Day, K. Gauvreau, S. Shulman, et al., Characteristics of children hospitalized with infective endocarditis, Circulation 119 (6) (2009) 865–870. [3] L.B. Rosenthal, K.N. Feja, S.M. Levasseur, et al., The changing epidemiology of pediatric endocarditis at a children's hospital over seven decades, Pediatr. Cardiol. 31 (6) (2010) 813–820. [4] G. Habib, Embolic risk in subacute bacterial endocarditis: determinants and role of transesophageal echocardiography, Curr. Cardiol. Rep. 5 (2) (2003) 129–136. [5] F. Thuny, G. Di Salvo, O. Belliard, et al., Risk of embolism and death in infective endocarditis: prognostic value of echocardiography: a prospective multicenter study, Circulation 112 (1) (2005) 69–75. [6] A. Saxena, N. Aggarwal, P. Gupta, et al., Predictors of embolic events in pediatric infective endocarditis, Indian Heart J. 61 (3) (2009) 242–245. [7] D.R. Murdoch, G.R. Corey, B. Hoen, et al., Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study, Arch. Intern. Med. 169 (5) (2009) 463–473. [8] S.J. Lee, Oh. SS, D.S. Lim, et al., Clinical significance of cerebrovascular complications in patients with acute infective endocarditis: a retrospective analysis of a 12-year single-center experience, BMC Neurol. 14 (2014) 30. [9] E. Garcia-Cabrera, N. Fernandez-Hidalgo, B. Almirante, et al., Neurological complications of infective endocarditis: risk factors, outcome, and impact of cardiac surgery: a multicenter observational study, Circulation 127 (23) (2013) 2272–2284. [10] C. Venkatesan, M.S. Wainwright, Pediatric endocarditis and stroke: a single-center retrospective review of seven cases, Pediatr. Neurol. 38 (4) (2008) 243–247. [11] E.E. Hill, P. Herijgers, P. Claus, et al., Infective endocarditis: changing epidemiology and predictors of 6-month mortality: a prospective cohort study, Eur. Heart J. 28 (2) (2007) 196–203. [12] R.S. Baltimore, M. Gewitz, L.M. Baddour, et al., Infective endocarditis in childhood: 2015 update: a scientific statement from the American Heart Association, Circulation 132 (15) (2015) 1487–1515. [13] L.I. Kupferwasser, M.R. Yeaman, S.M. Shapiro, et al., Acetylsalicylic acid reduces vegetation bacterial density, hematogenous bacterial dissemination, and frequency of embolic events in experimental Staphylococcus aureus endocarditis through antiplatelet and antibacterial effects, Circulation 99 (21) (1999) 2791–2797. [14] P. Monagle, A.K. Chan, N.A. Goldenberg, et al., Antithrombotic therapy in neonates and children: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines, Chest 141 (2 Suppl) (2012) e737S–801S. [15] G. Habib, P. Lancellotti, M.J. Antunes, et al., ESC Guidelines for the management of infective endocarditis: the Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM), Eur. Heart J. 36 (44) (2015) 3075–3128 2015. [16] K.L. Chan, J.G. Dumesnil, B. Cujec, et al., A randomized trial of aspirin on the risk of embolic events in patients with infective endocarditis, J. Am. Coll. Cardiol. 42 (5) (2003) 775–780. [17] J.S. Li, D.J. Sexton, N. Mick, et al., Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis, Clin. Infect. Dis. 30 (4) (2000) 633–638. [18] A.S. Dajani, K.A. Taubert, W. Wilson, et al., Prevention of bacterial endocarditis. Recommendations by the American Heart Association, Circulation 96 (1) (1997) 358–366. [19] P. Ferrieri, M.H. Gewitz, M.A. Gerber, et al., Unique features of infective endocarditis in childhood, Circulation 105 (17) (2002) 2115–2126. [20] L.G. Mitchell, N.A. Goldenberg, C. Male, et al., Definition of clinical efficacy and safety outcomes for clinical trials in deep venous thrombosis and pulmonary embolism in children, J. Thromb. Haemost. 9 (9) (2011) 1856–1858. [21] M. Yoshinaga, K. Niwa, A. Niwa, et al., Risk factors for in-hospital mortality during infective endocarditis in patients with congenital heart disease, Am. J. Cardiol. 101 (1) (2008) 114–118. [22] K. Keller, R.S. von Bardeleben, M.A. Ostad, et al., Temporal trends in the prevalence of infective endocarditis in Germany between 2005 and 2014, Am. J. Cardiol. 119 (2017) 317–322. [23] S. Ashkenazi, O. Levy, L. Blieden, Trends of childhood infective endocarditis in Israel with emphasis on children under 2 years of age, Pediatr. Cardiol. 18 (6) (1997) 419–424. [24] S. Gupta, A. Sakhuja, E. McGrath, et al., Trends, microbiology, and outcomes of infective endocarditis in children during 2000–2010 in the United States, Congenit. Heart Dis. 12 (2) (2017) 196–201. [25] W. Wilson, K.A. Taubert, M. Gewitz, 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 (15) (2007) 1736–1754. [26] D. Rushani, J.S. Kaufman, R. Ionescu-Ittu, et al., Infective endocarditis in children with congenital heart disease: cumulative incidence and predictors, Circulation 128 (13) (2013) 1412–1419.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085
6
K. Thom et al. / International Journal of Cardiology xxx (2017) xxx–xxx
[27] J.S. Penk, C.L. Webb, S.T. Shulman, et al., Echocardiography in pediatric infective endocarditis, Pediatr. Infect. Dis. J. 30 (12) (2011) 1109–1111. [28] C. Chirouze, C.H. Cabell, V.G. Fowler Jr., et al., Prognostic factors in 61 cases of Staphylococcus aureus prosthetic valve infective endocarditis from the International Collaboration on Endocarditis merged database, Clin. Infect. Dis. 38 (9) (2004) 1323–1327. [29] D.H. Kang, Y.J. Kim, S.H. Kim, et al., Early surgery versus conventional treatment for infective endocarditis, N. Engl. J. Med. 366 (26) (2012) 2466–2473.
[30] R. Hasbun, H.R. Vikram, L.A. Barakat, et al., Complicated left-sided native valve endocarditis in adults: risk classification for mortality, JAMA 289 (15) (2003) 1933–1940. [31] W. Jomaa, I. Ben Ali, D. Abid, et al., Clinical features and prognosis of infective endocarditis in children: insights from a Tunisian multicentre registry, Arch. Cardiovasc. Dis. (2017) (in press). [32] F. Thuny, J.F. Avierinos, C. Tribouilloy, et al., Impact of cerebrovascular complications on mortality and neurologic outcome during infective endocarditis: a prospective multicentre study, Eur. Heart J. 28 (9) (2007) 1155–1161.
Please cite this article as: K. Thom, et al., Incidence of infective endocarditis and its thromboembolic complications in a pediatric population over 30years, Int J Cardiol (2017), https://doi.org/10.1016/j.ijcard.2017.10.085