Abstract
Background
Infective endocarditis (IE) is a common and potentially serious disease. Although it is an illness that affects populations around the world, narrower descriptions of this disease as it impacts specific regions are uncommon. We analyzed the clinical characteristics of IE patients from two eastern counties in Taiwan and studied the relationship between the isolated pathogens and clinical outcomes in these patients.
Methods
This is a retrospective chart review study which enrolled patients who received services between January 2007 and December 2010. Subsequent to chart review, IE was confirmed in a total of 55 patients by the modified Duke criteria.
Results
Of these patients, 17 (31%) had previous traumatic open skin wounds. Pre-existing cardiac abnormalities were found in 47 (85%) patients, 28 of whom had valvular abnormalities. Staphylococcus aureus was isolated from the blood as the leading pathogen in 25 (45%) patients (including 23 methicillin-sensitive and 2 methicillin-resistant). Septic emboli and shock occurred in 27 (49%) of 55 patients; surgery was performed on 11 (20%) of those patients, and 4 (36%) of them died post-operatively. The total in-hospital mortality rate was 40% (n = 22). Staphylococcus aureus infection was associated with significantly higher complication and mortality rate than non-Staphylococcus aureus infection (59% vs. 41% and 64% vs. 36%, respectively; p < 0.05). In addition, patients with complications had a very high mortality rate (81.5%).
Conclusions
We found that Staphylococcus aureus was the most common pathogen of IE in Eastern Taiwan, and was associated with higher rates of morbidities and mortality.
Keywords: Infective endocarditis, Septic shock, Staphylococcus aureus, Systemic embolization
INTRODUCTION
Infective endocarditis (IE) is a serious, potentially life-threatening disease. Successful treatment is dependent upon early and accurate diagnosis. Streptococcus viridans was previously reported to be the main causative pathogen of IE.1-3 However, there has been a shift in the microbiological etiology of IE due to a variety of reasons, such as regional variation of living environment, an increasingly elderly population, and existing patient cardiac abnormalities and medical comorbidities.4 Although IE is an illness affecting patients worldwide, descriptions of this disease and its ramifications for specific regions were not commonly found. To know the characteristics and pathogens of IE in a specific region (such as Eastern Taiwan) could more familiarize primary physicians with its epidemiology, and better assist both physician and patient with their choice of the most appropriate antibiotic treatment. We retrospectively analyzed the clinical features of IE in patients who were referred to our tertiary-care medical center from the two eastern Taiwan counties of Hualien and Taitung over a 4-year period. Specifically, we studied the relationship between the isolated pathogens and the clinical outcomes in these patients.
METHODS
Patients
We retrospectively reviewed the medical records of patients over 18 years of age with IE who were admitted to the Buddhist Tzu Chi General Hospital (a 950-bed teaching hospital) from January 2007 to December 2010. The hospital provides both primary and tertiary medical services in Eastern Taiwan, an area with approximately 0.6 million residents. Medical charts of patients who met the International Classification of Diseases, Ninth Revision (ICD-9) diagnostic codes for endocarditis (ICD-9 codes 421.0, 421.1, and 421.9) were reviewed. A diagnosis of IE was confirmed based on the modified Duke criteria,5 and only those patients with definite IE participated in the study. The Protection of the Human Subjects Institutional Review Board of Buddhist Tzu Chi General Hospital approved this study.
IE diagnosis and evaluation
Standard blood cultures (each set containing one for aerobic and another for anaerobic incubation) were obtained in each patient on admission, prior to the administration of antibiotics. The site and timing of venous puncture for blood culture varied for each patient. Additional blood cultures were performed whenever a patient’s temperature was > 38.5 °C, or if the patient developed symptoms of chill or rigor during the hospital stay. Transthoracic echocardiography (TTE) was performed on all 55 patients (100%) and transesophageal echocardiography (TEE) was performed on 46 patients (84%). A 5.0-MHz transducer was used for multi-plane transesophageal studies. Review of TTE revealed a good quality of imaging showing clear vegetation; echocardiographic criteria for the definition of vegetation were detailed elsewhere.6 In general, vegetation was defined as a fixed or oscillating mass with a distinct echogenic structure and independent motion adherent to a leaflet or other cardiac structure. The flexibility, thickness, and echogenicity of the leaflet was used to distinguish pre-existing valvular abnormalities from the vegetation. Computed tomographic scan of the head was used to evaluate possible brain emboli in patients with sudden onset of neurological deficit or change of consciousness.
Statistical analysis
Continuous variables are expressed as mean ± standard deviation (SD). The chi-square test and Fisher’s exact test were used to compare the categorical variables between the two groups. A p value < 0.05 was considered statistically significant.
RESULTS
Clinical characteristics and risk factors
Fifty-five patients (40 men and 15 women, mean age of 51 ± 21 years) who fulfilled the diagnosis of IE defined by the modified Duke criteria5 were enrolled in the study. All patients were referred (n = 6, 11%) or directly (n = 49, 89%) admitted to our institution from the two counties (Hualien and Taitung) in the eastern region of Taiwan.
The clinical characteristics and demographics of all 55 IE patients are shown in Table 1. Their comorbidities included: alcoholism (n = 7, 13%), liver cirrhosis (n = 7, 13%), diabetes (n = 5, 9.1%), chronic kidney disease (n = 3, 5.5%) and hemodialysis (n = 1, 1.8%). Of the 55 patients studied, the most common symptoms were fever (n = 52, 95%), weight loss (n = 33, 60%), anorexia (n = 30, 55%) and shortness of breath (n = 22, 40%). The most common signs observed were cardiac murmurs (n = 55, 100%), splenomegaly (n = 36, 65%), pallor (n = 32, 58%), Osler’s nodes (n = 10, 18%), Janeway lesions (n = 4, 7.3%) and Roth’s spots (n = 1, 1.8%). High ESR (mean = 74 mm/hr) and leucocytosis (mean = 13,500/μL) were the most frequent abnormalities noted by laboratory investigation.
Table 1. Clinical characteristics and demographics in the 55 patients with infective endocarditis .
| Sex (male/female) | 40/15 |
| Age (year)* | 51 ± 21 (30-72) |
| Pre-existing risk factors# | |
| Traumatic open skin wound | 17 (31%) |
| Intravenous drug abuse | 8 (15%) |
| Dental therapy | 1 (1.8%) |
| Catheter insertion for dialysis | 1 (1.8%) |
| Pre-existing heart | |
| disease# | |
| Pre-existing valve disease | 28 (51%) |
| Mitral valve prolapse | 9 (16%) |
| Prosthetic valve | 9 (16%) |
| Ventricular septal defect | 1 (1.8%) |
| Comorbidity# | |
| Alcoholism | 7 (13%) |
| Liver cirrhosis | 7 (13%) |
| Diabetes | 5 (9.1%) |
| Chronic renal failure | 3 (5.5%) |
| Hemodialysis | 1 (1.8%) |
* Mean ± SD. # Values are expressed as number of cases (%).
Pre-existing cardiac abnormalities were found in 47 patients (85%), in whom 28 (51%) had pre-existing valve disease (defined as thickening of the valves with or without calcification by echocardiography or by pathology of the removed valves); also, 9 (16%) patients had mitral valve prolapse (MVP) identified by echocardiography, 9 (16%) had prosthetic mechanical valve replacement, and 1 (1.8%) had congenital ventricular septal defect (VSD). Eight (15%) patients had no significant cardiac abnormalities (Table 1). The potential risk factors for IE were found in 27 patients (49%) (Table 1). These risk factors include a history of recent traumatic open skin wounds (defined as an open skin wound related to a recent history of trauma in the past one month prior to the diagnosis of IE, the mean period of 10.7 days) in 17 patients (31%), intravenous drug abuse in 8 (15%), recent dental treatment in 1 (1.8%), and catheter insertion for hemodialysis in 1 (1.8%). The remaining 28 patients (51%) had no history of pre-existing risk factors. The valves affected by IE are shown in Table 2. The mitral valve, especially the anterior leaflet, was the most commonly affected (n = 32, 58%), followed by the aortic valve (18%), tricuspid valve (15%) and pulmonary valve (3.6%). Moreover, tricuspid valves were affected in those patients with a history of intravenous drug abuse (n = 8, 15%). Both the tricuspid valve and pulmonic valve were involved in one patient (1.8%). The pulmonic valve was also affected in the patient with a VSD.
Table 2. Valves affected by infective endocarditis in the 55 patients .
| Valves | |
| Mitral | 32 (58%) |
| Anterior mitral leaflet | 21 (38%) |
| Posterior mitral leaflet | 11 (20%) |
| Aortic | 10 (18%) |
| Tricuspid | 8 (15%) |
| Pulmonic | 2 (3.6%) |
| Prosthetic | 4 (7.3%) |
Values are expressed as no. of patients (%).
The pathogens
The pathogens responsible for IE in the 55 patients are shown in Table 3. Staphylococcus aureus was the pathogen most often isolated (n = 25, 45%), followed by Streptococcus species (n = 16, 29%). Gram negative bacteria were isolated in 7 patients (13%), and blood culture was negative in the remaining 7 patients (13%). Moreover, there were 23 methicillin-sensitive and 2 methicillin-resistant patients in the Staphylococcus aureus group. For further analysis of pre-existing risk factors in 25 patients with Staphylococcus aureus, there were 15 (60%) patients with traumatic open skin wounds, 8 (32%) patients with intravenous drug abuse, and 1 (4%) patient with catheter insertion for dialysis. Most interestingly, two patients with methicillin-resistant Staphylococcus aureus both had traumatic open skin wounds.
Table 3. Isolated pathogens of infective endocarditis in the 55 patients .
| Pathogen | |
| Staphylococcus aureus | 25 (45%) |
| Methicillin-sensitive | 23 (42%) |
| Methicillin-resistant | 2 (3.6%) |
| Streptococcus spp. | 16 (29%) |
| Gram-negative bacteria | 7 (13%) |
| Negative blood culture | 7 (13%) |
Values are expressed as no. of patients (%).
Prognosis - major complications
Brain emboli and septic shock were the major complications found in this study (Table 4). A total of 27 patients (49%) had major complications, in which 15 (27%) had brain emboli and 12 (22%) had septic shock. None of the patients had significant brady- or tachyarrhythmias other than sinus tachycardia, and one had atrioventricular block. Of the 27 patients with major complications, 16 (59%) were infected with Staphylococcus aureus, 7 (26%) with Streptococcus, and 4 (15%) with Gram-negative microorganisms. No major complications were found in patients with a negative blood culture. For comparison among isolated pathogens, patients with Staphylococcus aureus infection had a higher rate of brain emboli or septic shock (n = 16, 59%) than the patients infected with other pathogens (n = 11, 41%, p < 0.05) (Table 4).
Table 4. Comparisons among isolated pathogens and clinical outcomes .
| Total (n = 55) | S. aureus (n = 25) | Strept. (n = 16) | G(-) (n = 7) | NBC (n = 7) | |
| Complications | 27 | 16 (59%)* | 7 (26%) | 4 (16%) | 0 (0%) |
| Brain emboli | 15 | 8 (53%) | 4 (27%) | 3 (20%) | 0 (0%) |
| Septic shock | 12 | 8 (67%) | 3 (25%) | 1 (8%) | 0 (0%) |
| In-hospital mortality | 22 | 14 (64%)* | 6 (27%) | 2 (9%) | 0 (0%) |
| Surgery | 11 | 6 (55%) | 3 (27%) | 0 (0%) | 2 (18%) |
Values are expressed as no. of cases (%).
G(-), Gram-negative bacteria; NBC, negative blood culture; S. aureus, Staphylococcus aureus; Strept., Streptococcus spp.
* p < 0.05: The S. aureus group compared with the non-S. aureus group.
Prognosis - mortality
Twenty-two patients died in the hospital with a mortality rate of 40% (Table 4). Of the 22 patients who expired, more than a half of them had complications: 5 (23%) were related to brain emboli and 7 (32%) to septic shock, respectively. For comparisons among isolated pathogens, patients with Staphylococcus aureus infections also had a higher mortality (n = 14, 64%) than the other non-Staphylococcus aureus infected patients (n = 8, 36%, p < 0.05). Eleven patients (20%) underwent cardiac surgery (valvuloplasty or prosthetic valve replacement) due to either refractory heart failure (n = 8) or due to non-responsiveness to antibiotic treatment (n = 3). Of the 11 patients who underwent surgery, 4 had aortic and 7 had mitral valve replacement. None had tricuspid or combined valve surgery. Four patients died postoperatively in the hospital.
DISCUSSION
In this retrospective study, we report the clinical characteristics of the 55 patients who were diagnosed with IE and treated in our hospital from January 2007 to December 2010. All of these patients lived in two counties (Hualien and Taitung) in the region of Eastern Taiwan. Staphylococcus aureus was the leading pathogen (45%) in this population. The distribution of pathogens responsible for IE in our patients was different from that reported previously.3,4,7,8 Recently, the incidence of IE caused by Streptococcus viridans has decreased while IE caused by Staphylococcus aureus has increased. This shift in the microbiological spectrum is thought in part to be due to an aging population, increased intravenous drug abuse, and increased incidence of intensive and invasive medical procedures.4,9 The higher incidence of Staphylococcus aureus endocarditis in our series cannot be well-explained by the above-mentioned factors. Approximately 75% of patients with IE were reported to have structural heart diseases.9-11 In the present study, we also found that 85% of the patients had pre-existing cardiac abnormalities. However, approximately 50% of our patients had pre-existing risk factors for IE development. In particular, 17 (31%) of our 55 patients had a history of traumatic open skin wounds. A further analysis showed that 60% of patients with Staphylococcus aureus infections had traumatic open wounds and 32% of patients had intravenous drug abuse. Similar to intravenous drug abuse,8 skin wounds may be an important risk factor for Staphylococcus aureus IE. Consequently, the high percentage of traumatic open skin wounds may (at least in part) account for the higher incidence of Staphylococcus aureus endocarditis in our series. Interestingly, a history of recent dental procedure was found in only 1.8% of patients with IE in our study population.
In all 25 patients with Staphylococcus aureus infections, there were 23 (92%) methicillin-sensitive and 2 (8%) methicillin-resistant. The low incidence rate (8%) of methicillin-resistant Staphylococcus aureus is also consistent with other prior reports of community-acquired Staphylococcus aureus infection.12,13 For empiric antibiotic treatment of IE patients in Eastern Taiwan, oxacillin/nafcillin plus gentamicin are therefore superior to penicillin G plus gentamicin for synergistic coverage of methicillin-sensitive Staphylococcus aureus, especially in patients with traumatic open skin wounds and intravenous drug abuse.
Brain emboli and septic shock are the most common complications that are associated with a high mortality in IE patients.14-16 The rates of complications (49%) and mortality (40%) in our IE patients were even higher than those previously reported.1,3,9,17 Because our patients were mostly referred from rural areas in Eastern Taiwan, delayed medical treatment may be considered as an important factor associated with a higher complication and mortality rate. A skin wound is commonly infected with Staphylococcus aureus, and it often occurs in rural areas with lower hygiene standards and delayed treatment for the skin wound. Nearly 45% of pathogens isolated in our patients were Staphylococcus aureus, which is a virulent pathogen that often damages tissues rapidly and causes septic emboli in the early phase of infection. Moreover, we found that the complication and mortality rates of Staphylococcus aureus IE are much higher than those of non-Staphylococcus aureus IE (59% vs. 41% and 64% vs. 36%, respectively, p < 0.05). These results were also reported in one study of Taiwan’s National Health Insurance database.13 Thus, a high percentage of Staphylococcus aureus IE could explain the higher rates of complications and mortality in our study.
The incidence of negative blood culture in IE patients has decreased with the development of better microbiologic techniques and procedures. Slow-growing or fastidious microorganisms that are difficult to culture may result in an increased risk of septic emboli and an elevated mortality rate.18-22 Although we did not use special culture techniques to identify the fastidious microorganisms, none of the patients with negative blood culture experienced complications or died in the present study. However, the true complication and mortality rates of the culture-negative IE group may be underestimated in the study due to the sparse number of cases in the culture-negative IE group (n = 7).
Reviewing the literature of IE epidemiology in Taiwan, Lee et al.13 reported that Streptococci species (61%) followed by Staphylococcal species (32%) were the most common causative pathogens based on Taiwan’s National Health Insurance database from 1997 through 2002. The mean in-hospital mortality rate was 18%. For the individual districts, Chen et al.23 reported that the most frequent pathogen found was Streptococcus viridians (55%) followed by Staphylococcus aureus (21%) in Changhua County (Mid-Taiwan). The total mortality rate was 29.3%. Moreover, one report relating to Southern Taiwan showed that Staphylococcus aureus (57.5%) was the most frequent pathogen with a high mortality rate of 25%.24 Overall, the most common IE pathogens throughout the entire island of Taiwan were still were Streptococci species. However, Staphylococcus aureus became the leading pathogen of IE patients in some individual districts, such as Eastern Taiwan (our study) and Southern Taiwan.
There are several limitations in our study. This was a retrospective study and our study population (n = 55) is considered relatively small, albeit the study period covered the most recent 4 years. Our institution is the only tertiary-care medical center in Eastern Taiwan and the majority of our patients were referred from rural areas with delayed treatment. Thus, our study may not represent true clinical presentations and prognosis of IE in this region. Nonetheless, our data shows some interesting findings that are different than those previously reported. Our study is also the first to report the unique clinical characteristics of IE in the eastern region of Taiwan in the two counties (Hualien and Taitung). Further study on a larger-scale and prospective manner would be needed.
CONCLUSION
We identified Staphylococcus aureus as the leading pathogen responsible for a high rate of complications and mortality in our IE patients in Eastern Taiwan. We thought that traumatic open skin wounds could be a significant contributing factor for Staphylococcus aureus IE, especially for those patients with known cardiac abnormalities. In patients with an open skin wound and existing valvular abnormality, it is appropriate that Staphylococcus aureus sensitive antibiotics (e.g., oxacillin or nafcillin) be administered early once the diagnosis of IE is suspected and blood cultures are obtained. However, it remains unclear whether or not to routinely use prophylactic antibiotic against Staphylococcus aureus in patients with an open skin wound and with an existing valvular abnormality, in the absence of a positive blood culture or clinical symptoms and signs suggestive of IE.
Acknowledgments
This work was supported in part by Grants TCRD 102-16 from the Tzu Chi General Hospital, Hualien, Taiwan. The authors thank Prof. Shoei K. Stephen Huang for his critical review of this manuscript and study.
REFERENCES
- 1.Chao CH, Liu CY, Cheng DL, et al. Overview of viridans streptococcal endocarditis:clinical analysis of 99 cases. Zhonghua Yi Xue Za Zhi (Taipei) 1991;48:351–358. [PubMed] [Google Scholar]
- 2.Harris LF. Viridans streptococcal endocarditis. Ala Med. 1989; 59:13–17. [PubMed] [Google Scholar]
- 3.Di Salvo GD, Habib G, Pergola V, et al. Echocardiography predicts embolic events in infective endocarditis. J Am Coll Cardiol. 2001;37:1069–1076. doi: 10.1016/s0735-1097(00)01206-7. [DOI] [PubMed] [Google Scholar]
- 4.Cabell C, Abrutyn E. Progress toward a global understanding of infective endocarditis. Lesson from the international collaboration on endocarditis. Cardiol Clin. 2003;21:147–158. doi: 10.1016/s0733-8651(03)00033-x. [DOI] [PubMed] [Google Scholar]
- 5.Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30:633–638. doi: 10.1086/313753. [DOI] [PubMed] [Google Scholar]
- 6.Sachdev M, Peterson GE, Jollis JG. Imaging techniques for diagnosis of infective endocarditis. Cardiol Clin. 2003;21:185–195. doi: 10.1016/s0733-8651(03)00026-2. [DOI] [PubMed] [Google Scholar]
- 7.Steckelberg JM, Wilson WR. Risk factors for infective endocarditis. Infect Dis Clin North Am. 1993;7:9–19. [PubMed] [Google Scholar]
- 8.Nakatani S, Mitsutake K, Hozumi T, et al. Current characteristics of infective endocarditis in Japan; an analysis of 848 cases in 2000 and 2001. Circ J. 2003;67:901–905. doi: 10.1253/circj.67.901. [DOI] [PubMed] [Google Scholar]
- 9.Ako J, Ikari Y, Katori M, et al. Changing spectrum of infective endocarditis:review of 194 episodes over 20 years. Circ J. 2003; 67:3–7. doi: 10.1253/circj.67.3. [DOI] [PubMed] [Google Scholar]
- 10.Miro JM, Rio Ad, Mestres CA. Infective endocarditis and cardiac surgery in intravenous drug abusers and HIV-1 infected patients. Cardiol Clin. 2003;21:167–184. doi: 10.1016/s0733-8651(03)00025-0. [DOI] [PubMed] [Google Scholar]
- 11.Sia SK, Wu YL, Wu DJ, et al. Subaortic-right atrial fistula after endocarditis in hypertrophic cardiomyopathy. Acta Cardiol Sin. 2013;29:366–369. [PMC free article] [PubMed] [Google Scholar]
- 12.Chen CH, Lo MC, Hwang KL, et al. Infective endocarditis with neurological complication:10 year experience. J Mcrobiol Immunol Infect. 2001;34:119–124. [PubMed] [Google Scholar]
- 13.Lee CH, Tsai WC, Liu PY, et al. Epidemiologic features of infective endocarditis in Taiwanese adults involving native valves. Am J Cardiol. 2007;100:1282–1285. doi: 10.1016/j.amjcard.2007.05.053. [DOI] [PubMed] [Google Scholar]
- 14.Daiani AS, Tauberi KA, Wilson W. Prevention of bacterial endocarditis. Recommendation by the American Heart Association. JAMA. 1997;277:1794–1801. [PubMed] [Google Scholar]
- 15.Steckelberg JM, Murphy JG, Ballard D, et al. Emboli in infective endocarditis:the prognostic value of echocardiography. Ann Intern Med. 1991;114:635–640. doi: 10.7326/0003-4819-114-8-635. [DOI] [PubMed] [Google Scholar]
- 16.Stewart JA, Silimpery D, Harris P, et al. Echocardiographic documentation of vegetative lesions in infective endocarditis:clinical implication. Circulation. 1980;61:374–380. doi: 10.1161/01.cir.61.2.374. [DOI] [PubMed] [Google Scholar]
- 17.Cabell CH, Pond KK, Peterson GE, et al. The risk of stroke and death in patients with aortic and mitral valve endocarditis. Am Heart J. 2001;142:75–80. doi: 10.1067/mhj.2001.115790. [DOI] [PubMed] [Google Scholar]
- 18.Di Salvo G, Thuny F, Rosenberg V, et al. Endocarditis in the elderly:clinical,echocardiographic,and prognostic features. Eur Heart J. 2003;24:1576–1583. doi: 10.1016/s0195-668x(03)00309-9. [DOI] [PubMed] [Google Scholar]
- 19.Houpikian P, Raoult D. Diagnostic methods current best practices and guidelines for identification of difficult-to-culture pathogens in infective endocarditis. Cardiol Clin. 2003;21:207–217. doi: 10.1016/s0733-8651(03)00028-6. [DOI] [PubMed] [Google Scholar]
- 20.Berbari EF, Cockerill FR, 3 rd, Steckelberg JM. Infective endocarditis due to unusual or fastidious micro organisms. Mayo Clin Proc. 1997;72:532–542. doi: 10.4065/72.6.532. [DOI] [PubMed] [Google Scholar]
- 21.Barnes PD, Crook DW. Culture negative endocarditis. J Infect. 1997;35:209–213. doi: 10.1016/s0163-4453(97)92662-1. [DOI] [PubMed] [Google Scholar]
- 22.Zamonano J, Sanz J, Almeria C, et al. Differences between endocarditis with true negative blood cultures and those with previous antibiotic treatment. J Heart Valve Dis. 2003;12:256–260. [PubMed] [Google Scholar]
- 23.Chen CH, Tsai CD, Wu SL, et al. A retrospective study of infective endocarditis at Changhua Christian Hospital during the years 1989-1999. J Intern Med Tawian. 2001;12:177–183. [Google Scholar]
- 24.Wu KS, Lee SS, Tsai HC, et al. Non-nosocomial healthcare-associated infective endocarditis in Taiwan:an underrecognized disease with poor outcome. BMC Infect Dis. 2011;11:221–228. doi: 10.1186/1471-2334-11-221. [DOI] [PMC free article] [PubMed] [Google Scholar]