International Journal of Cardiology 179 (2015) 465–469
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Epidemiologic features and long-term outcome of dialysis patients with infective endocarditis in Taiwan Ming-Ting Chou a,b,1, Jhi-Joung Wang c,1, Wen-Shiann Wu a,b,1, Shih-Feng Weng c,d, Chung-Han Ho c,d, Zhe-Zhong Lin a, Tain-Junn Cheng e,f, Chih-Chiang Chien g,h,⁎ a
Department of Cardiology, Chi-Mei Medical Center, Tainan, Taiwan The Center of General Education, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan d Department of Hospital and Health Care Administration, Chia Nan University of Pharmacy and Science, Tainan, Taiwan e Department of Neurology, Occupational Medicine, Medical Information Management, Chi Mei Medical Center, Tainan, Taiwan f Department of Occupational Safety and Health, Institute of Industrial Safety and Disaster Prevention, College of Sustainable Environment, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan g Department of Nephrology, Chi-Mei Medical Center, Tainan, Taiwan h Department of Food Nutrition, Chung Hwa University of Medical Technology, Tainan, Taiwan b c
a r t i c l e
i n f o
Article history: Received 4 April 2014 Received in revised form 2 November 2014 Accepted 6 November 2014 Available online 8 November 2014 Keywords: Infective endocarditis Rheumatic heart disease Dialysis
a b s t r a c t Background: The incidence of infective endocarditis (IE) is high in dialysis patients. Limited data are available on the risk factors for IE and long-term outcome after IE in dialysis patients, especially in Asian populations. Methods: We used Taiwan National Health Insurance Research Database to design a longitudinal cohort study. 68,426 ESRD patients who began dialysis between 1999 and 2007 were included. The follow-up period was from the start of dialysis to death, end of dialysis, or end of 2008. Cox proportional hazards models were used to identify the risk factors for IE. Results: IE was diagnosed in 502 patients during follow-up (201.4 per 100,000 person-years). Diabetes mellitus (DM), congestive heart failure (CHF), cerebro-vascular accident (CVA), and rheumatic heart disease (RHD) (HR: 3.07, 95% CI: 1.99–4.75) were associated with an increasing risk of development of IE. The cumulative incidence rate of IE in patients with RHD was 1.4, 2.2, and 3.9% at 1, 3, and 5 years. In-hospital mortality was 23.5%. Cumulative survival rates post-IE were 54.3% at 1 year and only 35.3% at 5 years. Conclusion: Dialysis patients had a higher risk of IE. Those who were older and had DM, CHF, CVA, or especially RHD were at a greater risk. Dialysis patients with IE also had high mortality. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The incidence of mortality and morbidity in dialysis patients is a major global concern because their effect on public health is greater than previously thought [1]. The trend of hospitalization due to infection increases in recent years [1]. Infection is second only to cardiovascular disease as the leading cause of death in patients with end-stage renal disease (ESRD); it occurs in approximately 12–22% of patients [2]. Infective endocarditis (IE) is one of the infectious diseases that always has a high mortality rate unless it is appropriately treated [3].
⁎ Corresponding author at: Department of Nephrology, Chi-Mei Medical Center, and Department of Food Nutrition, Chung Hwa University of Medical Technology, 901 JungHua Road, Yung Kang District, Tainan City 710, Taiwan. E-mail address: [email protected] (C.-C. Chien). 1 Ming-Ting Chou, Jhi-Joung Wang and Wen-Shiann Wu contributed equally and are also first authors for this article.
http://dx.doi.org/10.1016/j.ijcard.2014.11.083 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
Chronic rheumatic heart disease (RHD), which was a prime risk factor for IE in the pre-antibiotic era, is now rare in industrialized countries [4]. However, it continues to be a disease with high prevalence in some developing countries [5]. Even in more industrialized places, namely, Hong Kong and Thailand, RHD was found in 18% and 12%, respectively, of cases of IE [6]. Therefore, RHD probably has a different effect on the occurrence of IE in ESRD patients in Asia and western countries. The cumulative incidence of IE in dialysis patients has been estimated [7] to be approximately 267 cases/100,000 person-years which is much higher than that in the general population. In a study using the United States Renal Data System (USRDS) database, the in-hospital mortality rate was 23.5% and 1-year mortality was 61.6% [8]. The incidence and prevalence rates of patients with ESRD on dialysis are high in Taiwan [9]. In addition, RHD is still an ongoing problem in Asia [6]. However, limited data are available on the risk factors for IE and long-term outcome after IE in ESRD dialysis patients, especially in the Asian population. The purpose of this study was to determine the long-term outcome of dialysis patients hospitalized for IE in Taiwan.
466
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
Using data from 1999 to 2007 in the Taiwan National Health Insurance Research Database (NHIRD) database, we investigated the epidemiologic features and outcome of IE in dialysis patients.
methods of identifying these comorbidities have been extensively employed in various studies using the Taiwan NHIRD as a data source, and many articles based on these studies have been published [10–14].
2. Research design and methods
2.4. Statistical analysis
2.1. Database
The incidence of newly diagnosed IE is expressed as the number of cases of IE per 100,000 person-years. A parametric Pearson's χ2 test was used to compare each variable in the groups of patients with and without IE. Age was entered as a categorical variable (18–44, 45–64, and ≥65 years). Significance was set at p b 0.05. The cumulative proportions of patients with IE and of survivors after IE were calculated using the Kaplan–Meier method. The log rank test was used to analyze significance. Cox proportional hazards models were used to identify the risk factors of IE. Hazard ratios (HRs) and 95% confidence intervals (CIs) were derived from Cox proportional hazards models, which met the assumption of the proportionality of risks. SPSS 17.0 for Windows (SPSS Inc; Chicago, IL, USA) was used for all statistical analyses.
Taiwan's National Health Insurance (NHI) program has provided compulsory universal health insurance since 1995. With the exception of prison inmates, all citizens are enrolled in the program. All contracted medical institutions must submit standard computerized claim documents for medical expenses. Patients with ESRD are eligible for any type of renal replacement therapy free of any charge. All patients on chronic dialysis are covered by NHI. Data were obtained from the NHIRD (Bureau of National Health Insurance; www.doh.gov.tw/statistic/index.htm [in Chinese]; http:// www.doh.gov.tw/EN2006/ index_EN.aspx [in English]) and released for research by the Taiwan National Health Research Institute. The NHIRD covers nearly all (99%) inpatient and outpatient medical benefit claims for Taiwan's 23.3 million residents, is one of the largest and most comprehensive databases in the world, and has been used extensively in various studies [10–14]. Patient identification numbers, gender, birthdate, dates of admission and discharge, medical institutions providing the services, the ICD-9-CM diagnostic and procedure codes (up to five each), and outcomes are encrypted. This study tapped the NHIRD for ambulatory care claims, all inpatient claims, and the updated registry for beneficiaries from 1998 to 2008 for this study. All data sets can be interlinked. 2.2. Patient selection and definition This longitudinal cohort study selected all adult (≥18 years old) patients with ESRD on maintenance dialysis who began renal replacement therapy between January 1, 1999 and December 31, 2007. Maintenance dialysis was defined as having undergone dialysis for more than 90 days. Patients were followed-up from the first reported date of dialysis to the date of death, end of dialysis, or December 31, 2008. The inpatient claims include the records of all hospitalizations and provide a substantial amount of information. We linked the study subjects to the inpatient claim data to identify the first episode of IE (ICD9 codes 421.0, 421.1, and 421.9) after the initiation of dialysis. During the follow-up period, 502 patients were diagnosed with IE. 2.3. Determining the demographic and comorbid variables We linked to the diagnostic codes through the inpatient and outpatient claim databases of the NHI. We included not only survival status, but also date of death, patient demographics, and baseline comorbidities. Baseline comorbidities – listed by ICD-9 code — diabetes mellitus (DM) (250, 357.2, 362.0X, 366.41), congestive heart failure (CHF) (428, 398.91, 422, 425, 402.X1, 404.X1, 404.X3), coronary artery disease (CAD) (410–414), cerebrovascular accident (CVA) (430–438), other cardiac diseases (420–421; 423–424; 429; 785.0–785.3), valvular heart disease (424.0–424.3), rheumatic heart disease (391, 392.0, 393–398) – are important factors affecting episodes of IE and were assessed at the start of dialysis. Which comorbidities the patients had was determined by whether they fit one of the following two definitions: (1) diagnostic codes from outpatient visits if the patient had an initial diagnosis at any time in the 1 year before the start of dialysis and then had one or more additional diagnoses within the subsequent 12 months. The first and the last outpatient visit within 1 year had to have been N30 days apart to avoid accidental inclusion of miscoded patients; (2) diagnostic codes in hospitalization databases at least once in 1 year before the start of dialysis. The accuracy of the diagnoses of major disease in the NHIRD has been validated [15,16]. In addition, the
3. Results 3.1. Demographics and clinical characteristics We enrolled 68,426 adult patients with incident dialysis. During the 9-year follow-up, 502 (0.73%) patients developed IE (men: 236 [0.7%]; women: 266 [0.8%]; p = 0.486) (Table 1). The incidence rate of IE was 201.4/100,000 person-years. There incidence rate of IE increased over the study period (range: 158.81/100,000 person-years [between 1999
Table 1 Patient characteristics and association with (n = 502) and without (n = 67,924) infective endocarditis in end-stage renal disease dialysis patients.
Gender Female Male Age (years) mean ± SD Age (years) 18–44 45–64 ≥65 Dialysis modality PD HD Diabetic mellitus No Yes Congestive heart failure No Yes Coronary artery disease No Yes Cerebrovascular disease No Yes Other cardiac diseasesa No Yes Valvular heart disease No Yes Rheumatic heart disease No Yes
Without IE (n = 67,924)
With IE (n = 502)
n
n
(%)
p
(%) 0.486
34,933 (51.4) 32,991 (48.6) 60.11 ± 14.06
266 (53.0) 236 (47.0) 62.12 ± 13.09
9689 29,604 28,631
(14.3) (43.6) (42.2)
42 216 244
(8.4) (43.0) (48.6)
5078 62,846
(7.5) (92.5)
20 482
(4.0) (96.0)
33,955 33,969
(50.0) (50.0)
204 298
(40.6) (59.4)
50,907 17,017
(74.9) (25.1)
330 172
(65.7) (34.3)
52,325 15,599
(77.0) (23.0)
352 150
(90.1) (29.9)
59,519 8405
(87.6) (12.4)
423 79
(84.3) (15.7)
61,975 5949
(91.2) (8.8)
434 68
(86.5) (13.5)
66,669 1255
(98.2) (1.8)
483 19
(96.2) (3.8)
67,130 794
(98.8) (1.2)
480 22
(95.6) (4.4)
0.007 b0.001
0.003
b0.001
b0.001
b0.001
0.023
b0.001
0.001
b0.001
IE, infective endocarditis; HD, hemodialysis; PD, peritoneal dialysis; SD, standard deviation. a Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices.
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
and 2001], 218.28/100,000 person-years [between 2002 and 2004], 254.51/100,000 person-years [between 2005 and 2007]). Patients with IE were significantly (p b 0.007) older (62.12 ± 13.09 years) than those without IE (60.11 ± 14.06 years). Only 0.4% of those 18–44 years old had IE; however, 0.8% of those ≥ 65 years old had IE (p b 0.001). Patients on hemodialysis (HD) had a higher incidence of IE than did those on peritoneal dialysis (PD) (p = 0.003). Patients with IE tended to have more comorbidities than did those without IE. Many more patients with IE had DM and baseline cardiovascular diseases, including CHF, CAD, CVA, valvular heart disease, and RHD, than did those without IE. The incidence of IE was as high as 791.7/ 100,000 person-years in patients with RHD. 3.2. Cumulative incidence and risk factors for infective endocarditis The cumulative incidence rates of IE for patients with and without RHD were 1.4% vs. 0.3% at one year, 2.2% vs. 0.6% at three years, and 3.9 vs. 0.9% at five years, respectively (Fig. 1). Being older, and having baseline DM, CHF, CVA, and RHD were independent risk factors for IE in dialysis patients (Table 2). Patients on HD had a 42.9% higher risk for IE than did those on PD; however, the difference was not significant after multivariate adjustments (HR: 1.429; 95% CI: 0.910–2.245). The risk of IE increased with age (1.7% per each increment of 1 year). Additionally, factors increasing the risk of developing IE included DM (HR: 1.49; 95% CI: 1.23–1.79), CHF (HR: 1.36; 95% CI: 1.11– 1.66), CVA (HR: 1.31: 95% CI: 1.03–1.68), and RHD (HR: 3.09; 95% CI: 2.00–4.78). 3.3. Outcome data Of the 502 patients with IE, 39 (7.8%) underwent surgery for valve incompetence as a result of the infection (Table 3). The remaining patients were medically managed. Patients given surgery were younger (mean: 52.6 ± 11.7 years vs. 62.6 ± 13.1, p = 0.004). Overall in-hospital mortality was 23.5% (Fig. 2a). The cumulative survival rates after IE were 54.3% at one year and only 35.3% at five years. We also analyzed the outcomes between patients treated with surgery and those treated with antibiotics alone. The in-hospital mortality rate was 25.9% for patients treated with surgery and 21.0% for those
467
Table 2 Risk factor for infective endocarditis after initiation of dialysis in end-stage renal disease dialysis patients (n = 68,426). Covariate
Gender (male vs. female) Age (year; each increment of 1 year) Dialysis modality (HD vs. PD) Diabetic mellitus (yes vs. no) Congestive heart failure (yes vs. no) Coronary artery disease (yes vs. no) Cerebrovascular disease (yes vs. no) Other cardiac diseases (yes vs. no)b Valvular heart disease (yes vs. no) Rheumatic heart disease (yes vs. no)
Univariate analysis
Multivariate analysis
HR (95% CI)
HR (95% CI)a
0.992 (0.833–1.183) 1.016 (0.852–1.212) 1.023 (1.016–1.030)⁎ 1.017 (1.009–1.024)⁎ 1.876 (1.199–2.934)⁎ 1.429 (0.910–2.245) 1.769 (1.479–2.116)⁎ 1.485 (1.232–1.791)⁎ 1.867 (1.551–2.247)⁎ 1.356 (1.107–1.662)⁎ 1.699 (1.402–2.059)⁎ 1.164 (0.945–1.433) 1.637 (1.286–2.083)⁎ 1.313 (1.028–1.678)⁎ 1.746 (1.352–2.255)⁎ 1.309 (0.993–1.725) 2.320 (1.467–3.670)⁎ 1.493 (0.916–2.435) 4.021 (2.622–6.165)⁎ 3.093 (2.000–4.784)⁎
HR, hazard ration; CI: confidence interval; HD, hemodialysis; PD, peritoneal dialysis. a HR adjusted for gender, age, dialysis modality, diabetic mellitus, congestive heart failure, coronary artery disease, cerebrovascular accident, other cardiac diseases, valvular heart disease, and rheumatic heart disease. b Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices. ⁎ p b 0.05.
treated with antibiotics alone. The cumulative survival rates after IE for patients treated with surgery were 66.5% at 3 months and 58.4% at one year, and for patients treated with antibiotics alone these were 69.7% at 3 months and 54.3% at one year, but these differences were not significant (Fig. 2b). An analysis of the outcomes between patients with and without RHD showed nonsignificant differences in the in-hospital mortality rate: 22.7% (with RHD) vs. 21% (without RHD) and the cumulative survival rate after IE: 63.6% (with RHD) vs. 69.7% (without RHD) at 3 months, and 54.5% (with RHD) vs. 54.3% (without RHD) at one year (Fig. 2c).
4. Discussion This is the first national longitudinal cohort study to evaluate the association between RHD and IE and mortality in ESRD dialysis patients. Using the Taiwan NHIRD, which contains data on a nationwide and representative ESRD dialysis population, to investigate the epidemiology of IE, we found that the incidence of IE in dialysis patients is high. RHD was a strong risk factor for IE. We also found that ESRD dialysis patients had a high in-hospital mortality (23.5%) due to IE. Table 3 Patients characteristics in the non-surgical treated (n = 463) and surgical treated (n = 39) groups.
Fig. 1. Cumulative incidence of infective endocarditis stratified by end-stage renal disease dialysis patients with (n = 816) and without (n = 67,580) rheumatic heart disease (RHD).
Male Average age (years) In-hospital death Diabetic mellitus Congestive heart failure Coronary artery disease Cerebrovascular disease Other cardiac diseasesa Valvular heart disease Rheumatic heart disease
Non-surgical group (n = 463)
Surgical group (n = 39)
n
n
(%)
222 (47.9) 62.6 ± 13.1 107 (23.1) 280 (60.5) 159 (34.3) 143 (30.9) 76 (16.4) 60 (13.0) 18 (3.9) 20 (4.3)
p
(%)
14 (35.9) 52.6 ± 11.7 11 (28.2) 18 (46.2) 13 (33.3) 7 (17.9) 3 (7.7) 8 (20.5) 1 (2.6) 2 (5.1)
0.148 0.004 0.471 0.08 0.899 0.09 0.151 0.186 0.677 0.813
a Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices.
468
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
Fig. 2. (a) Overall survival curves after infective endocarditis (n = 502) in end-stage renal disease dialysis patients. (b) Overall survival curves after infective endocarditis stratified by patients treated with surgery and treated with antibiotics alone. (c) Overall survival curves after infective endocarditis stratified by patients treated with and without rheumatic heart disease (RHD).
The incidence of IE in dialysis patients is significantly higher than that in the general population. The mean in-hospital mortality rate reported in a Taiwan nationwide IE study using NHI data from 1997 to 2002 was 18% (mean annual incidence: 7.6/100,000 persons) [17], and renal insufficiency was a strong independent predictor of mortality. In the present study, the mean annual incidence of IE in dialysis patients was 201.4/100,000 persons, about 26.5 times than the overall incidence of IE in Taiwan. A 1-year French IE survey of dialysis patients [18] reported that the incidence of IE was 170–200/100,000 persons, 50–60 times higher than the overall incidence of IE in France. Using the USRDS, Abbott and Agodoa [7] found an age-adjusted incidence ratio of 17.9 for IE in the dialysis population compared with the general population, and Strom et al. [19], reported an adjusted odds ratio 16.9 times higher. Patients with ESRD are inherently prone to bacteremia because of immune system compromise caused by the biochemical abnormalities, malnutrition, and increased risk of malignancy [20]. Malnourishment and older age might also interact with uremia to further impair the immune system, as may comorbid conditions such as DM or underlying systemic disease [3]. These factors all increase the risk of IE in dialysis patients, as the present study shows. Other studies [7,21] have reported that the frequency of IE is not higher in PD patients than in the general population, and that their prognoses of IE are significantly better than are those of HD patients. Our study found that patients on HD had a 42% higher risk for IE than did those on PD after multivariate adjustments. One of the highest risk factor for IE is probably the risk of recurrent bacteremia because of vascular access infection due to increased manipulation. This is probably the reason why PD has lower rates of bacteremia than dose HD and why Staphylococcus aureus is the most common infecting organism. Fernandez-Cean et al. [21] reported that S. aureus was the most common infecting organism (58%) in ESRD dialysis patients and that vascular access is the most frequent identified port of entry. They evaluated the relationship between dialysis modality and the outcome of IE. In-hospital mortality in patients switched to PD (8.3%) was lower than for those who stayed on HD (55.5%). They concluded that the high mortality of IE in HD may be associated with its vascular access [21]. Valvular heart disease and RHD are associated with IE in ESRD dialysis patients in Taiwan. Native valve IE in the general population ≥65 is caused by degenerative cardiac lesions. In patients with ESRD, degenerative heart valve disease is premature because it appears to begin 10–20 years earlier than in the general population [22]. Straumannet et al. [23] reported a 64% prevalence of structural changes in either the mitral or aortic valve after 50 months on HD. In our study, patients ≥ 65 years old had twice the risk of developing IE than did those b45 years old (HR: 2.13; 95% CI: 1.52–3.00). Valvular calcifications were the primary lesions identified. This is thought to be related to the abnormalities of calcium–phosphorus homeostasis in the setting of secondary hyperparathyroidism and to the chronic inflammation of uremia [24]. In patients on HD, any age group is prone to valvular heart disease [25]. In our study, patient with RHD had a higher risk of developing IE (HR: 3.07; 95% CI: 1.99–4.75) than did patients with underlying valvular heart disease. This association between RHD and a higher risk for IE is well documented in the developing world [5,26], but is no longer seen in many developed countries [27]. The studies on IE confirm that in most Asian countries, at least half of all cases of this severe and often fatal disease are due to underlying RHD, although in some countries with emerging middle classes, this proportion is probably lower [6]. A recent study in Hong Kong [28] found RHD in 18% of IE patients, secondary to the underlying cause of intravenous drug addiction (30%). RHD also remains an important factor for the occurrence of IE in Taiwan [29]. Our study showed that patients with RHD and on dialysis had an especially greater risk for IE. The in-hospital mortality rate of IE was high among ESRD dialysis patients. In our study, it was 23.5%, which was higher than the 18% rate in the general population [14]. The mortality of IE was even higher
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
than that of acute coronary syndrome (9.7%) [10]. Shroff et al. [8] analyzed the USRDS database from 1997 to 2000 and also reported a 23.5% in-hospital mortality rate, but lower long-term survival rates at 1 year = 38%, 2 years = 25%, and 3 years = 18%. Leither et al. [30] reported similar survival rates in 2013: 1 year = 40%, 2 years = 30%, and 3 years = 20%. Although the in-hospital mortality is identical to US data [8], the 1-year mortality rate in our database is considerably lower than in the USRDS database. In-hospital death from IE is most often acute and related to the severity of IE, but long-term mortality after IE is often related to patient age and to their baseline comorbidities. In our study, IE occurred in younger patients, particularly those with RHD (mean: 59.9 ± 13.3 years). Thus, it is reasonable to speculate that our lower 1-year mortality after IE was probably because our patients were younger. The survival rate of HD patients with IE has changed little in the past two decades despite improvement in medical and surgical therapy [31]. A recent Taiwan national IE study [10] showed that male gender, older age, DM, heart failure, neurological complications, renal insufficiency, respiratory failure, shock, and Staphylococcus species as the causative microorganism were independent predictors of in-hospital mortality. Shroff et al. [8] reported that the most powerful independent predictors of all-cause mortality were age and DM and CVA as a comorbid condition. In our study, 39 (7.8%) patients underwent surgical treatment with valve replacement (Table 3). The patients in the surgery group were younger (mean: 52.6 ± 11.7 years vs. 62.6 ± 13.1). There was no significant difference in mortality rates between patients treated with surgery and those treated with antibiotics alone, including the in-hospital mortality rate: 25.9% (surgical group) vs. 21.0% (non-surgical group) and the cumulative survival rate after IE: 66.5% (surgical group) vs. 69.7% (non-surgical group) at 3 months, and 58.4% (surgical group) vs. 54.3% (non-surgical group) at one year (Fig. 2b, log rank test, p = 0.819). This is different from the findings of two recent studies [31,32] that showed better survival for patients treated with valve surgery. A limitation in our study was that surgical group is heavily subject to selection bias (only 7.8% of the patients underwent surgery). This study has several limitations. First, the NHIRD includes little clinical data. Information was unavailable on the heart valve involved, as well as clinical criteria on which the diagnosis of bacterial endocarditis was based. Furthermore, the NHIRD contains no information about vascular access routes that are risk factors for IE in patients undergoing HD. Secondly, the diagnosis of the comorbidity relied on claim data and ICD-9-CM codes, which may have resulted in a disease misclassification. Finally, the study provides little insight into the cause of worse outcome of dialysis patients after IE. In conclusion, dialysis patients had a higher risk for IE. Patients who are older and have comorbid DM, CHF, CVA, or, especially, RHD were at greater risk. Dialysis patients with IE also had high mortality rate. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Acknowledgments This study was supported by grants CMFHR10352 and CMNDMC10009 from Chi-Mei Hospital, grant NHRI-NHIRD-99182 from the National Health Research Institutes, and NSC 102-2314-B384-001 from the National Science Council Taiwan. The authors are grateful to the staff in the Department of Disease Classification team, Chi-Mei Medical Center, especially Ya-Ping Chen, who helped with the infromation of ICD-9-CM diagnostic and procedure codes.
469
References [1] A.J. Collins, R.N. Foley, C. Herzog, et al., Excerpts from the US Renal Data System 2009 Annual Data Report, Am. J. Kidney Dis. 55 (1 Suppl. 1) (2010) S1–S420 (A6-A7). [2] N.R. Powe, B. Jaar, S.L. Furth, J. Hermann, W. Briggs, Septicemia in dialysis patients: incidence, risk factors, and prognosis, Kidney Int. 55 (1999) 1081–1090. [3] S. Maraj, L.E. Jacobs, R. Maraj, M.N. Kolter, Bacteremia and infective endocarditis in patients on hemodialysis, Am. J. Med. Sci. 327 (2004) 242–249. [4] J. Normand, A. Bozio, J. Etienne, F. Sassolas, H. Le Bris, Changing patterns and prognosis of infective endocarditis in childhood, Eur. Heart J. 16 (Suppl. B) (1995) 28–31. [5] N. Garg, B. Kandpal, S. Tewari, N. Garg, A. Kapoor, P. Goel, N. Sinha, Characteristics of infective endocarditis in a developing country—clinical profile and outcome in 192 Indian patients, 1992–2001, Int. J. Cardiol. 98 (2005) 253–260. [6] J.R. Carapetis, Rheumatic heart disease in Asia, Circulation 118 (2008) 2748–2753. [7] K.C. Abbott, L.Y. Agodoa, Hospitalizations for bacterial endocarditis after initiation of chronic dialysis in the United States, Nephron 91 (2002) 203–209. [8] G.R. Shroff, C.A. Herzog, J.Z. Ma, A.J. Collins, Long-term survival of dialysis patients with bacterial endocarditis in the United States, Am. J. Kidney Dis. 44 (2004) 1077–1082. [9] W.C. Yang, S.J. Hwang, Taiwan Society of Nephrology, Incidence, prevalence and mortality trends of dialysis end-stage renal disease in Taiwan from 1990 to 2001: the impact of national health insurance, Nephrol. Dial. Transplant. 23 (2008) 3977–3982. [10] M.T. Chou, J.J. Wang, Y.M. Sun, et al., Epidemiology and mortality among dialysis patients with acute coronary syndrome: Taiwan national cohort study, Int. J. Cardiol. 167 (2013) 2719–2723. [11] K.J. Tien, Z.Z. Lin, C.C. Chio, et al., Epidemiology and mortality of new-onset diabetes after dialysis: Taiwan national cohort study, Diabetes Care 36 (2013) 3027–3032. [12] W.C. Kan, J.J. Wang, S.Y. Wang, et al., The new comorbidity index for predicting survival in elderly dialysis patients: a long-term population-based study, PLoS One 8 (2013) e68748. [13] C.C. Chien, C.S. Yen, J.J. Wang, et al., Reverse epidemiology of hypertension–mortality associations in hemodialysis patients: a long-term population-based study, Am. J. Hypertens. 25 (2012) 900–906. [14] C.C. Chien, J.J. Wang, Y.M. Sun, et al., Long-term survival and predictors for mortality among dialysis patients in an endemic area for chronic liver disease: a national cohort study in Taiwan, BMC Nephrol. 13 (2012) 43. [15] C.Y. Wu, F.K. Chan, M.S. Wu, et al., Histamine2-receptor antagonists are an alternative to proton pump inhibitor in patients receiving clopidogrel, Gastroenterology 139 (2010) 1165–1171. [16] C.L. Cheng, Y.H. Kao, S.J. Lin, C.H. Lee, M.L. Lai, Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan, Pharmacoepidemiol. Drug Saf. 20 (2011) 236–242. [17] C.H. Lee, W.C. Tsai, P.Y. Liu, et al., Epidemiologic features of infective endocarditis in Taiwanese adults involving native valves, Am. J. Cardiol. 100 (2007) 1282–1285. [18] B. Hoen, F. Alla, C. Selton-Suty, et al., Changing profile of infective endocarditis; results of a 1-year-survey in France, JAMA 288 (2002) 75–81. [19] B.L. Strom, E. Abrutyn, J.A. Berlin, et al., Risk factors for infective endocarditis: oral hygiene and nondental exposure, Circulation 102 (2000) 2842–2848. [20] M. Haag-Weber, W.H. Hörl, Uremia and infection: mechanisms of impaired cellular host defense, Nephron 63 (1993) 125–131. [21] J. Fernández-Cean, A. Alvarez, S. Burguez, G. Baldovinos, P. Larre-Borges, M. Cha, Infective endocarditis in chronic hemodialysis: two treatment strategies, Nephrol. Dial. Transplant. 17 (2002) 2226–2230. [22] E.C. Madu, I.A. D'Cruz, B. Wall, N. Mansour, S. Shearin, Transesophageal echocardiographic spectrum of calcific mitral abnormalities in patients with end-stage renal disease, Echocardiography 17 (2000) 29–35. [23] E. Straumann, B. Meyer, M. Misteli, A. Blumberg, H.R. Jenzer, Aortic and mitral valve disease in patients with end stage renal failure in long-term hemodialysis, Br. Heart J. 67 (1992) 236–239. [24] E. Umana, W. Ahmed, M.A. Akpert, Valvular and perivalvular abnormalities in endstage renal disease, Am. J. Med. Sci. 325 (2003) 237–242. [25] R. Vijayvargiya, J.H. Veis, Antibiotic-resistant endocarditis in a hemodialysis patient, J. Am. Soc. Nephrol. 7 (1996) 536–542. [26] C.A. Baskerville, B.B. Hanrahan, A.J. Burke, A.J. Holwell, M.G. Rémond, G.P. Maguire, Infective endocarditis and rheumatic heart disease in the north of Australia, Heart Lung Circ. 21 (2012) 36–41. [27] I.M. Tleyjeh, J.M. Steckelberg, H.S. Murad, et al., Temporal trends in infective endocarditis: a population-based study in Olmsted County, Minnesota, JAMA 293 (2005) 3022–3028. [28] K.H. Yiu, C.W. Siu, K.L. Lee, et al., Emerging trends of community acquired infective endocarditis, Int. J. Cardiol. 121 (2007) 119–122. [29] H.C. Lue, M.H. Wu, B.T. Hwang, Establishment of a rheumatic heart disease registry in Taipei: an early appraisal, N. Z. Med. J. 101 (847 Pt2) (1988) 410–411. [30] M.D. Leither, G.R. Shroff, S. Ding, D.T. Gilbertson, G.A. Herzog, Long-term survival of dialysis patients with bacterial endocarditis undergoing valvular replacement surgery in the United States, Circulation 128 (2013) 344–351. [31] G. Nucifora, L. Badano, P. Viale, et al., Infective endocarditis in chronic haemodialysis patients: an increasing clinical challenge, Eur. Heart J. 28 (2007) 2307–2312. [32] D.A. Jones, L.A. McGill, K.S. Rathod, et al., Characteristics and outcomes of dialysis patients with infective endocarditis, Nephron. Clin. Pract. 123 (2013) 151–156.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Epidemiologic features and long-term outcome of dialysis patients with infective endocarditis in Taiwan Ming-Ting Chou a,b,1, Jhi-Joung Wang c,1, Wen-Shiann Wu a,b,1, Shih-Feng Weng c,d, Chung-Han Ho c,d, Zhe-Zhong Lin a, Tain-Junn Cheng e,f, Chih-Chiang Chien g,h,⁎ a
Department of Cardiology, Chi-Mei Medical Center, Tainan, Taiwan The Center of General Education, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan d Department of Hospital and Health Care Administration, Chia Nan University of Pharmacy and Science, Tainan, Taiwan e Department of Neurology, Occupational Medicine, Medical Information Management, Chi Mei Medical Center, Tainan, Taiwan f Department of Occupational Safety and Health, Institute of Industrial Safety and Disaster Prevention, College of Sustainable Environment, Chia-Nan University of Pharmacy and Science, Tainan, Taiwan g Department of Nephrology, Chi-Mei Medical Center, Tainan, Taiwan h Department of Food Nutrition, Chung Hwa University of Medical Technology, Tainan, Taiwan b c
a r t i c l e
i n f o
Article history: Received 4 April 2014 Received in revised form 2 November 2014 Accepted 6 November 2014 Available online 8 November 2014 Keywords: Infective endocarditis Rheumatic heart disease Dialysis
a b s t r a c t Background: The incidence of infective endocarditis (IE) is high in dialysis patients. Limited data are available on the risk factors for IE and long-term outcome after IE in dialysis patients, especially in Asian populations. Methods: We used Taiwan National Health Insurance Research Database to design a longitudinal cohort study. 68,426 ESRD patients who began dialysis between 1999 and 2007 were included. The follow-up period was from the start of dialysis to death, end of dialysis, or end of 2008. Cox proportional hazards models were used to identify the risk factors for IE. Results: IE was diagnosed in 502 patients during follow-up (201.4 per 100,000 person-years). Diabetes mellitus (DM), congestive heart failure (CHF), cerebro-vascular accident (CVA), and rheumatic heart disease (RHD) (HR: 3.07, 95% CI: 1.99–4.75) were associated with an increasing risk of development of IE. The cumulative incidence rate of IE in patients with RHD was 1.4, 2.2, and 3.9% at 1, 3, and 5 years. In-hospital mortality was 23.5%. Cumulative survival rates post-IE were 54.3% at 1 year and only 35.3% at 5 years. Conclusion: Dialysis patients had a higher risk of IE. Those who were older and had DM, CHF, CVA, or especially RHD were at a greater risk. Dialysis patients with IE also had high mortality. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The incidence of mortality and morbidity in dialysis patients is a major global concern because their effect on public health is greater than previously thought [1]. The trend of hospitalization due to infection increases in recent years [1]. Infection is second only to cardiovascular disease as the leading cause of death in patients with end-stage renal disease (ESRD); it occurs in approximately 12–22% of patients [2]. Infective endocarditis (IE) is one of the infectious diseases that always has a high mortality rate unless it is appropriately treated [3].
⁎ Corresponding author at: Department of Nephrology, Chi-Mei Medical Center, and Department of Food Nutrition, Chung Hwa University of Medical Technology, 901 JungHua Road, Yung Kang District, Tainan City 710, Taiwan. E-mail address: [email protected] (C.-C. Chien). 1 Ming-Ting Chou, Jhi-Joung Wang and Wen-Shiann Wu contributed equally and are also first authors for this article.
http://dx.doi.org/10.1016/j.ijcard.2014.11.083 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
Chronic rheumatic heart disease (RHD), which was a prime risk factor for IE in the pre-antibiotic era, is now rare in industrialized countries [4]. However, it continues to be a disease with high prevalence in some developing countries [5]. Even in more industrialized places, namely, Hong Kong and Thailand, RHD was found in 18% and 12%, respectively, of cases of IE [6]. Therefore, RHD probably has a different effect on the occurrence of IE in ESRD patients in Asia and western countries. The cumulative incidence of IE in dialysis patients has been estimated [7] to be approximately 267 cases/100,000 person-years which is much higher than that in the general population. In a study using the United States Renal Data System (USRDS) database, the in-hospital mortality rate was 23.5% and 1-year mortality was 61.6% [8]. The incidence and prevalence rates of patients with ESRD on dialysis are high in Taiwan [9]. In addition, RHD is still an ongoing problem in Asia [6]. However, limited data are available on the risk factors for IE and long-term outcome after IE in ESRD dialysis patients, especially in the Asian population. The purpose of this study was to determine the long-term outcome of dialysis patients hospitalized for IE in Taiwan.
466
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
Using data from 1999 to 2007 in the Taiwan National Health Insurance Research Database (NHIRD) database, we investigated the epidemiologic features and outcome of IE in dialysis patients.
methods of identifying these comorbidities have been extensively employed in various studies using the Taiwan NHIRD as a data source, and many articles based on these studies have been published [10–14].
2. Research design and methods
2.4. Statistical analysis
2.1. Database
The incidence of newly diagnosed IE is expressed as the number of cases of IE per 100,000 person-years. A parametric Pearson's χ2 test was used to compare each variable in the groups of patients with and without IE. Age was entered as a categorical variable (18–44, 45–64, and ≥65 years). Significance was set at p b 0.05. The cumulative proportions of patients with IE and of survivors after IE were calculated using the Kaplan–Meier method. The log rank test was used to analyze significance. Cox proportional hazards models were used to identify the risk factors of IE. Hazard ratios (HRs) and 95% confidence intervals (CIs) were derived from Cox proportional hazards models, which met the assumption of the proportionality of risks. SPSS 17.0 for Windows (SPSS Inc; Chicago, IL, USA) was used for all statistical analyses.
Taiwan's National Health Insurance (NHI) program has provided compulsory universal health insurance since 1995. With the exception of prison inmates, all citizens are enrolled in the program. All contracted medical institutions must submit standard computerized claim documents for medical expenses. Patients with ESRD are eligible for any type of renal replacement therapy free of any charge. All patients on chronic dialysis are covered by NHI. Data were obtained from the NHIRD (Bureau of National Health Insurance; www.doh.gov.tw/statistic/index.htm [in Chinese]; http:// www.doh.gov.tw/EN2006/ index_EN.aspx [in English]) and released for research by the Taiwan National Health Research Institute. The NHIRD covers nearly all (99%) inpatient and outpatient medical benefit claims for Taiwan's 23.3 million residents, is one of the largest and most comprehensive databases in the world, and has been used extensively in various studies [10–14]. Patient identification numbers, gender, birthdate, dates of admission and discharge, medical institutions providing the services, the ICD-9-CM diagnostic and procedure codes (up to five each), and outcomes are encrypted. This study tapped the NHIRD for ambulatory care claims, all inpatient claims, and the updated registry for beneficiaries from 1998 to 2008 for this study. All data sets can be interlinked. 2.2. Patient selection and definition This longitudinal cohort study selected all adult (≥18 years old) patients with ESRD on maintenance dialysis who began renal replacement therapy between January 1, 1999 and December 31, 2007. Maintenance dialysis was defined as having undergone dialysis for more than 90 days. Patients were followed-up from the first reported date of dialysis to the date of death, end of dialysis, or December 31, 2008. The inpatient claims include the records of all hospitalizations and provide a substantial amount of information. We linked the study subjects to the inpatient claim data to identify the first episode of IE (ICD9 codes 421.0, 421.1, and 421.9) after the initiation of dialysis. During the follow-up period, 502 patients were diagnosed with IE. 2.3. Determining the demographic and comorbid variables We linked to the diagnostic codes through the inpatient and outpatient claim databases of the NHI. We included not only survival status, but also date of death, patient demographics, and baseline comorbidities. Baseline comorbidities – listed by ICD-9 code — diabetes mellitus (DM) (250, 357.2, 362.0X, 366.41), congestive heart failure (CHF) (428, 398.91, 422, 425, 402.X1, 404.X1, 404.X3), coronary artery disease (CAD) (410–414), cerebrovascular accident (CVA) (430–438), other cardiac diseases (420–421; 423–424; 429; 785.0–785.3), valvular heart disease (424.0–424.3), rheumatic heart disease (391, 392.0, 393–398) – are important factors affecting episodes of IE and were assessed at the start of dialysis. Which comorbidities the patients had was determined by whether they fit one of the following two definitions: (1) diagnostic codes from outpatient visits if the patient had an initial diagnosis at any time in the 1 year before the start of dialysis and then had one or more additional diagnoses within the subsequent 12 months. The first and the last outpatient visit within 1 year had to have been N30 days apart to avoid accidental inclusion of miscoded patients; (2) diagnostic codes in hospitalization databases at least once in 1 year before the start of dialysis. The accuracy of the diagnoses of major disease in the NHIRD has been validated [15,16]. In addition, the
3. Results 3.1. Demographics and clinical characteristics We enrolled 68,426 adult patients with incident dialysis. During the 9-year follow-up, 502 (0.73%) patients developed IE (men: 236 [0.7%]; women: 266 [0.8%]; p = 0.486) (Table 1). The incidence rate of IE was 201.4/100,000 person-years. There incidence rate of IE increased over the study period (range: 158.81/100,000 person-years [between 1999
Table 1 Patient characteristics and association with (n = 502) and without (n = 67,924) infective endocarditis in end-stage renal disease dialysis patients.
Gender Female Male Age (years) mean ± SD Age (years) 18–44 45–64 ≥65 Dialysis modality PD HD Diabetic mellitus No Yes Congestive heart failure No Yes Coronary artery disease No Yes Cerebrovascular disease No Yes Other cardiac diseasesa No Yes Valvular heart disease No Yes Rheumatic heart disease No Yes
Without IE (n = 67,924)
With IE (n = 502)
n
n
(%)
p
(%) 0.486
34,933 (51.4) 32,991 (48.6) 60.11 ± 14.06
266 (53.0) 236 (47.0) 62.12 ± 13.09
9689 29,604 28,631
(14.3) (43.6) (42.2)
42 216 244
(8.4) (43.0) (48.6)
5078 62,846
(7.5) (92.5)
20 482
(4.0) (96.0)
33,955 33,969
(50.0) (50.0)
204 298
(40.6) (59.4)
50,907 17,017
(74.9) (25.1)
330 172
(65.7) (34.3)
52,325 15,599
(77.0) (23.0)
352 150
(90.1) (29.9)
59,519 8405
(87.6) (12.4)
423 79
(84.3) (15.7)
61,975 5949
(91.2) (8.8)
434 68
(86.5) (13.5)
66,669 1255
(98.2) (1.8)
483 19
(96.2) (3.8)
67,130 794
(98.8) (1.2)
480 22
(95.6) (4.4)
0.007 b0.001
0.003
b0.001
b0.001
b0.001
0.023
b0.001
0.001
b0.001
IE, infective endocarditis; HD, hemodialysis; PD, peritoneal dialysis; SD, standard deviation. a Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices.
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
and 2001], 218.28/100,000 person-years [between 2002 and 2004], 254.51/100,000 person-years [between 2005 and 2007]). Patients with IE were significantly (p b 0.007) older (62.12 ± 13.09 years) than those without IE (60.11 ± 14.06 years). Only 0.4% of those 18–44 years old had IE; however, 0.8% of those ≥ 65 years old had IE (p b 0.001). Patients on hemodialysis (HD) had a higher incidence of IE than did those on peritoneal dialysis (PD) (p = 0.003). Patients with IE tended to have more comorbidities than did those without IE. Many more patients with IE had DM and baseline cardiovascular diseases, including CHF, CAD, CVA, valvular heart disease, and RHD, than did those without IE. The incidence of IE was as high as 791.7/ 100,000 person-years in patients with RHD. 3.2. Cumulative incidence and risk factors for infective endocarditis The cumulative incidence rates of IE for patients with and without RHD were 1.4% vs. 0.3% at one year, 2.2% vs. 0.6% at three years, and 3.9 vs. 0.9% at five years, respectively (Fig. 1). Being older, and having baseline DM, CHF, CVA, and RHD were independent risk factors for IE in dialysis patients (Table 2). Patients on HD had a 42.9% higher risk for IE than did those on PD; however, the difference was not significant after multivariate adjustments (HR: 1.429; 95% CI: 0.910–2.245). The risk of IE increased with age (1.7% per each increment of 1 year). Additionally, factors increasing the risk of developing IE included DM (HR: 1.49; 95% CI: 1.23–1.79), CHF (HR: 1.36; 95% CI: 1.11– 1.66), CVA (HR: 1.31: 95% CI: 1.03–1.68), and RHD (HR: 3.09; 95% CI: 2.00–4.78). 3.3. Outcome data Of the 502 patients with IE, 39 (7.8%) underwent surgery for valve incompetence as a result of the infection (Table 3). The remaining patients were medically managed. Patients given surgery were younger (mean: 52.6 ± 11.7 years vs. 62.6 ± 13.1, p = 0.004). Overall in-hospital mortality was 23.5% (Fig. 2a). The cumulative survival rates after IE were 54.3% at one year and only 35.3% at five years. We also analyzed the outcomes between patients treated with surgery and those treated with antibiotics alone. The in-hospital mortality rate was 25.9% for patients treated with surgery and 21.0% for those
467
Table 2 Risk factor for infective endocarditis after initiation of dialysis in end-stage renal disease dialysis patients (n = 68,426). Covariate
Gender (male vs. female) Age (year; each increment of 1 year) Dialysis modality (HD vs. PD) Diabetic mellitus (yes vs. no) Congestive heart failure (yes vs. no) Coronary artery disease (yes vs. no) Cerebrovascular disease (yes vs. no) Other cardiac diseases (yes vs. no)b Valvular heart disease (yes vs. no) Rheumatic heart disease (yes vs. no)
Univariate analysis
Multivariate analysis
HR (95% CI)
HR (95% CI)a
0.992 (0.833–1.183) 1.016 (0.852–1.212) 1.023 (1.016–1.030)⁎ 1.017 (1.009–1.024)⁎ 1.876 (1.199–2.934)⁎ 1.429 (0.910–2.245) 1.769 (1.479–2.116)⁎ 1.485 (1.232–1.791)⁎ 1.867 (1.551–2.247)⁎ 1.356 (1.107–1.662)⁎ 1.699 (1.402–2.059)⁎ 1.164 (0.945–1.433) 1.637 (1.286–2.083)⁎ 1.313 (1.028–1.678)⁎ 1.746 (1.352–2.255)⁎ 1.309 (0.993–1.725) 2.320 (1.467–3.670)⁎ 1.493 (0.916–2.435) 4.021 (2.622–6.165)⁎ 3.093 (2.000–4.784)⁎
HR, hazard ration; CI: confidence interval; HD, hemodialysis; PD, peritoneal dialysis. a HR adjusted for gender, age, dialysis modality, diabetic mellitus, congestive heart failure, coronary artery disease, cerebrovascular accident, other cardiac diseases, valvular heart disease, and rheumatic heart disease. b Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices. ⁎ p b 0.05.
treated with antibiotics alone. The cumulative survival rates after IE for patients treated with surgery were 66.5% at 3 months and 58.4% at one year, and for patients treated with antibiotics alone these were 69.7% at 3 months and 54.3% at one year, but these differences were not significant (Fig. 2b). An analysis of the outcomes between patients with and without RHD showed nonsignificant differences in the in-hospital mortality rate: 22.7% (with RHD) vs. 21% (without RHD) and the cumulative survival rate after IE: 63.6% (with RHD) vs. 69.7% (without RHD) at 3 months, and 54.5% (with RHD) vs. 54.3% (without RHD) at one year (Fig. 2c).
4. Discussion This is the first national longitudinal cohort study to evaluate the association between RHD and IE and mortality in ESRD dialysis patients. Using the Taiwan NHIRD, which contains data on a nationwide and representative ESRD dialysis population, to investigate the epidemiology of IE, we found that the incidence of IE in dialysis patients is high. RHD was a strong risk factor for IE. We also found that ESRD dialysis patients had a high in-hospital mortality (23.5%) due to IE. Table 3 Patients characteristics in the non-surgical treated (n = 463) and surgical treated (n = 39) groups.
Fig. 1. Cumulative incidence of infective endocarditis stratified by end-stage renal disease dialysis patients with (n = 816) and without (n = 67,580) rheumatic heart disease (RHD).
Male Average age (years) In-hospital death Diabetic mellitus Congestive heart failure Coronary artery disease Cerebrovascular disease Other cardiac diseasesa Valvular heart disease Rheumatic heart disease
Non-surgical group (n = 463)
Surgical group (n = 39)
n
n
(%)
222 (47.9) 62.6 ± 13.1 107 (23.1) 280 (60.5) 159 (34.3) 143 (30.9) 76 (16.4) 60 (13.0) 18 (3.9) 20 (4.3)
p
(%)
14 (35.9) 52.6 ± 11.7 11 (28.2) 18 (46.2) 13 (33.3) 7 (17.9) 3 (7.7) 8 (20.5) 1 (2.6) 2 (5.1)
0.148 0.004 0.471 0.08 0.899 0.09 0.151 0.186 0.677 0.813
a Includes pericarditis, endocarditis, myocarditis, other complications of heart disease, heart transplant, heart valve replacement, and cardiac devices.
468
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
Fig. 2. (a) Overall survival curves after infective endocarditis (n = 502) in end-stage renal disease dialysis patients. (b) Overall survival curves after infective endocarditis stratified by patients treated with surgery and treated with antibiotics alone. (c) Overall survival curves after infective endocarditis stratified by patients treated with and without rheumatic heart disease (RHD).
The incidence of IE in dialysis patients is significantly higher than that in the general population. The mean in-hospital mortality rate reported in a Taiwan nationwide IE study using NHI data from 1997 to 2002 was 18% (mean annual incidence: 7.6/100,000 persons) [17], and renal insufficiency was a strong independent predictor of mortality. In the present study, the mean annual incidence of IE in dialysis patients was 201.4/100,000 persons, about 26.5 times than the overall incidence of IE in Taiwan. A 1-year French IE survey of dialysis patients [18] reported that the incidence of IE was 170–200/100,000 persons, 50–60 times higher than the overall incidence of IE in France. Using the USRDS, Abbott and Agodoa [7] found an age-adjusted incidence ratio of 17.9 for IE in the dialysis population compared with the general population, and Strom et al. [19], reported an adjusted odds ratio 16.9 times higher. Patients with ESRD are inherently prone to bacteremia because of immune system compromise caused by the biochemical abnormalities, malnutrition, and increased risk of malignancy [20]. Malnourishment and older age might also interact with uremia to further impair the immune system, as may comorbid conditions such as DM or underlying systemic disease [3]. These factors all increase the risk of IE in dialysis patients, as the present study shows. Other studies [7,21] have reported that the frequency of IE is not higher in PD patients than in the general population, and that their prognoses of IE are significantly better than are those of HD patients. Our study found that patients on HD had a 42% higher risk for IE than did those on PD after multivariate adjustments. One of the highest risk factor for IE is probably the risk of recurrent bacteremia because of vascular access infection due to increased manipulation. This is probably the reason why PD has lower rates of bacteremia than dose HD and why Staphylococcus aureus is the most common infecting organism. Fernandez-Cean et al. [21] reported that S. aureus was the most common infecting organism (58%) in ESRD dialysis patients and that vascular access is the most frequent identified port of entry. They evaluated the relationship between dialysis modality and the outcome of IE. In-hospital mortality in patients switched to PD (8.3%) was lower than for those who stayed on HD (55.5%). They concluded that the high mortality of IE in HD may be associated with its vascular access [21]. Valvular heart disease and RHD are associated with IE in ESRD dialysis patients in Taiwan. Native valve IE in the general population ≥65 is caused by degenerative cardiac lesions. In patients with ESRD, degenerative heart valve disease is premature because it appears to begin 10–20 years earlier than in the general population [22]. Straumannet et al. [23] reported a 64% prevalence of structural changes in either the mitral or aortic valve after 50 months on HD. In our study, patients ≥ 65 years old had twice the risk of developing IE than did those b45 years old (HR: 2.13; 95% CI: 1.52–3.00). Valvular calcifications were the primary lesions identified. This is thought to be related to the abnormalities of calcium–phosphorus homeostasis in the setting of secondary hyperparathyroidism and to the chronic inflammation of uremia [24]. In patients on HD, any age group is prone to valvular heart disease [25]. In our study, patient with RHD had a higher risk of developing IE (HR: 3.07; 95% CI: 1.99–4.75) than did patients with underlying valvular heart disease. This association between RHD and a higher risk for IE is well documented in the developing world [5,26], but is no longer seen in many developed countries [27]. The studies on IE confirm that in most Asian countries, at least half of all cases of this severe and often fatal disease are due to underlying RHD, although in some countries with emerging middle classes, this proportion is probably lower [6]. A recent study in Hong Kong [28] found RHD in 18% of IE patients, secondary to the underlying cause of intravenous drug addiction (30%). RHD also remains an important factor for the occurrence of IE in Taiwan [29]. Our study showed that patients with RHD and on dialysis had an especially greater risk for IE. The in-hospital mortality rate of IE was high among ESRD dialysis patients. In our study, it was 23.5%, which was higher than the 18% rate in the general population [14]. The mortality of IE was even higher
M.-T. Chou et al. / International Journal of Cardiology 179 (2015) 465–469
than that of acute coronary syndrome (9.7%) [10]. Shroff et al. [8] analyzed the USRDS database from 1997 to 2000 and also reported a 23.5% in-hospital mortality rate, but lower long-term survival rates at 1 year = 38%, 2 years = 25%, and 3 years = 18%. Leither et al. [30] reported similar survival rates in 2013: 1 year = 40%, 2 years = 30%, and 3 years = 20%. Although the in-hospital mortality is identical to US data [8], the 1-year mortality rate in our database is considerably lower than in the USRDS database. In-hospital death from IE is most often acute and related to the severity of IE, but long-term mortality after IE is often related to patient age and to their baseline comorbidities. In our study, IE occurred in younger patients, particularly those with RHD (mean: 59.9 ± 13.3 years). Thus, it is reasonable to speculate that our lower 1-year mortality after IE was probably because our patients were younger. The survival rate of HD patients with IE has changed little in the past two decades despite improvement in medical and surgical therapy [31]. A recent Taiwan national IE study [10] showed that male gender, older age, DM, heart failure, neurological complications, renal insufficiency, respiratory failure, shock, and Staphylococcus species as the causative microorganism were independent predictors of in-hospital mortality. Shroff et al. [8] reported that the most powerful independent predictors of all-cause mortality were age and DM and CVA as a comorbid condition. In our study, 39 (7.8%) patients underwent surgical treatment with valve replacement (Table 3). The patients in the surgery group were younger (mean: 52.6 ± 11.7 years vs. 62.6 ± 13.1). There was no significant difference in mortality rates between patients treated with surgery and those treated with antibiotics alone, including the in-hospital mortality rate: 25.9% (surgical group) vs. 21.0% (non-surgical group) and the cumulative survival rate after IE: 66.5% (surgical group) vs. 69.7% (non-surgical group) at 3 months, and 58.4% (surgical group) vs. 54.3% (non-surgical group) at one year (Fig. 2b, log rank test, p = 0.819). This is different from the findings of two recent studies [31,32] that showed better survival for patients treated with valve surgery. A limitation in our study was that surgical group is heavily subject to selection bias (only 7.8% of the patients underwent surgery). This study has several limitations. First, the NHIRD includes little clinical data. Information was unavailable on the heart valve involved, as well as clinical criteria on which the diagnosis of bacterial endocarditis was based. Furthermore, the NHIRD contains no information about vascular access routes that are risk factors for IE in patients undergoing HD. Secondly, the diagnosis of the comorbidity relied on claim data and ICD-9-CM codes, which may have resulted in a disease misclassification. Finally, the study provides little insight into the cause of worse outcome of dialysis patients after IE. In conclusion, dialysis patients had a higher risk for IE. Patients who are older and have comorbid DM, CHF, CVA, or, especially, RHD were at greater risk. Dialysis patients with IE also had high mortality rate. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Acknowledgments This study was supported by grants CMFHR10352 and CMNDMC10009 from Chi-Mei Hospital, grant NHRI-NHIRD-99182 from the National Health Research Institutes, and NSC 102-2314-B384-001 from the National Science Council Taiwan. The authors are grateful to the staff in the Department of Disease Classification team, Chi-Mei Medical Center, especially Ya-Ping Chen, who helped with the infromation of ICD-9-CM diagnostic and procedure codes.
469
References [1] A.J. Collins, R.N. Foley, C. Herzog, et al., Excerpts from the US Renal Data System 2009 Annual Data Report, Am. J. Kidney Dis. 55 (1 Suppl. 1) (2010) S1–S420 (A6-A7). [2] N.R. Powe, B. Jaar, S.L. Furth, J. Hermann, W. Briggs, Septicemia in dialysis patients: incidence, risk factors, and prognosis, Kidney Int. 55 (1999) 1081–1090. [3] S. Maraj, L.E. Jacobs, R. Maraj, M.N. Kolter, Bacteremia and infective endocarditis in patients on hemodialysis, Am. J. Med. Sci. 327 (2004) 242–249. [4] J. Normand, A. Bozio, J. Etienne, F. Sassolas, H. Le Bris, Changing patterns and prognosis of infective endocarditis in childhood, Eur. Heart J. 16 (Suppl. B) (1995) 28–31. [5] N. Garg, B. Kandpal, S. Tewari, N. Garg, A. Kapoor, P. Goel, N. Sinha, Characteristics of infective endocarditis in a developing country—clinical profile and outcome in 192 Indian patients, 1992–2001, Int. J. Cardiol. 98 (2005) 253–260. [6] J.R. Carapetis, Rheumatic heart disease in Asia, Circulation 118 (2008) 2748–2753. [7] K.C. Abbott, L.Y. Agodoa, Hospitalizations for bacterial endocarditis after initiation of chronic dialysis in the United States, Nephron 91 (2002) 203–209. [8] G.R. Shroff, C.A. Herzog, J.Z. Ma, A.J. Collins, Long-term survival of dialysis patients with bacterial endocarditis in the United States, Am. J. Kidney Dis. 44 (2004) 1077–1082. [9] W.C. Yang, S.J. Hwang, Taiwan Society of Nephrology, Incidence, prevalence and mortality trends of dialysis end-stage renal disease in Taiwan from 1990 to 2001: the impact of national health insurance, Nephrol. Dial. Transplant. 23 (2008) 3977–3982. [10] M.T. Chou, J.J. Wang, Y.M. Sun, et al., Epidemiology and mortality among dialysis patients with acute coronary syndrome: Taiwan national cohort study, Int. J. Cardiol. 167 (2013) 2719–2723. [11] K.J. Tien, Z.Z. Lin, C.C. Chio, et al., Epidemiology and mortality of new-onset diabetes after dialysis: Taiwan national cohort study, Diabetes Care 36 (2013) 3027–3032. [12] W.C. Kan, J.J. Wang, S.Y. Wang, et al., The new comorbidity index for predicting survival in elderly dialysis patients: a long-term population-based study, PLoS One 8 (2013) e68748. [13] C.C. Chien, C.S. Yen, J.J. Wang, et al., Reverse epidemiology of hypertension–mortality associations in hemodialysis patients: a long-term population-based study, Am. J. Hypertens. 25 (2012) 900–906. [14] C.C. Chien, J.J. Wang, Y.M. Sun, et al., Long-term survival and predictors for mortality among dialysis patients in an endemic area for chronic liver disease: a national cohort study in Taiwan, BMC Nephrol. 13 (2012) 43. [15] C.Y. Wu, F.K. Chan, M.S. Wu, et al., Histamine2-receptor antagonists are an alternative to proton pump inhibitor in patients receiving clopidogrel, Gastroenterology 139 (2010) 1165–1171. [16] C.L. Cheng, Y.H. Kao, S.J. Lin, C.H. Lee, M.L. Lai, Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan, Pharmacoepidemiol. Drug Saf. 20 (2011) 236–242. [17] C.H. Lee, W.C. Tsai, P.Y. Liu, et al., Epidemiologic features of infective endocarditis in Taiwanese adults involving native valves, Am. J. Cardiol. 100 (2007) 1282–1285. [18] B. Hoen, F. Alla, C. Selton-Suty, et al., Changing profile of infective endocarditis; results of a 1-year-survey in France, JAMA 288 (2002) 75–81. [19] B.L. Strom, E. Abrutyn, J.A. Berlin, et al., Risk factors for infective endocarditis: oral hygiene and nondental exposure, Circulation 102 (2000) 2842–2848. [20] M. Haag-Weber, W.H. Hörl, Uremia and infection: mechanisms of impaired cellular host defense, Nephron 63 (1993) 125–131. [21] J. Fernández-Cean, A. Alvarez, S. Burguez, G. Baldovinos, P. Larre-Borges, M. Cha, Infective endocarditis in chronic hemodialysis: two treatment strategies, Nephrol. Dial. Transplant. 17 (2002) 2226–2230. [22] E.C. Madu, I.A. D'Cruz, B. Wall, N. Mansour, S. Shearin, Transesophageal echocardiographic spectrum of calcific mitral abnormalities in patients with end-stage renal disease, Echocardiography 17 (2000) 29–35. [23] E. Straumann, B. Meyer, M. Misteli, A. Blumberg, H.R. Jenzer, Aortic and mitral valve disease in patients with end stage renal failure in long-term hemodialysis, Br. Heart J. 67 (1992) 236–239. [24] E. Umana, W. Ahmed, M.A. Akpert, Valvular and perivalvular abnormalities in endstage renal disease, Am. J. Med. Sci. 325 (2003) 237–242. [25] R. Vijayvargiya, J.H. Veis, Antibiotic-resistant endocarditis in a hemodialysis patient, J. Am. Soc. Nephrol. 7 (1996) 536–542. [26] C.A. Baskerville, B.B. Hanrahan, A.J. Burke, A.J. Holwell, M.G. Rémond, G.P. Maguire, Infective endocarditis and rheumatic heart disease in the north of Australia, Heart Lung Circ. 21 (2012) 36–41. [27] I.M. Tleyjeh, J.M. Steckelberg, H.S. Murad, et al., Temporal trends in infective endocarditis: a population-based study in Olmsted County, Minnesota, JAMA 293 (2005) 3022–3028. [28] K.H. Yiu, C.W. Siu, K.L. Lee, et al., Emerging trends of community acquired infective endocarditis, Int. J. Cardiol. 121 (2007) 119–122. [29] H.C. Lue, M.H. Wu, B.T. Hwang, Establishment of a rheumatic heart disease registry in Taipei: an early appraisal, N. Z. Med. J. 101 (847 Pt2) (1988) 410–411. [30] M.D. Leither, G.R. Shroff, S. Ding, D.T. Gilbertson, G.A. Herzog, Long-term survival of dialysis patients with bacterial endocarditis undergoing valvular replacement surgery in the United States, Circulation 128 (2013) 344–351. [31] G. Nucifora, L. Badano, P. Viale, et al., Infective endocarditis in chronic haemodialysis patients: an increasing clinical challenge, Eur. Heart J. 28 (2007) 2307–2312. [32] D.A. Jones, L.A. McGill, K.S. Rathod, et al., Characteristics and outcomes of dialysis patients with infective endocarditis, Nephron. Clin. Pract. 123 (2013) 151–156.
