Infective endocarditis: a tertiary referral centre experience from Turkey

Abstract

Introduction: We aimed to define the current characteristics of infective endocarditis (IE) in a part of Turkey. Methods: All patients who were hospitalized in our hospital with a diagnosis of IE between 2009 and 2014 were included in the study. Data were collected from archives records of all patients. Modified Duke criteria were used for diagnosis. Results: There were 85 IE cases during the study period. The mean age of patients was 52 years. Fourty eight of patients were males. Native valves involved in 47%, prostetic valves involved in 40% and pacemaker or ICD lead IE in 13% of patients. Mitral valve was the most common site of vegetationb (38%). The most common valvular pathology was mitral regurgitation. The most common predisposing factor was prosthetic valve disease (40%). Positive culture rate was 68%. Staphylococci were the most frequent causative microorganisms isolated (27%) followed by Streptococcus spp. (11%). In-hospital mortality rate was 36%. Conclusion: In Turkey, IE occurs in relatively young patients. In high developed part of Turkey, prosthetic and dejenerative valve disease is taking the place of rheumatic valve disease as a predisposing factor. Surgery is an important factor for preventing mortality.

Introduction

Infective endocarditis (IE) is the infection of the endothelial surface of the heart. It is still a life threatening disease which has high morbidity and mortality [1-3]. Although the disease was first described in sixteenth century, which was Osler’s Gulstonian Lectures to the Royal College of Physicians in 1885 that created the impetus for systematic study of IE [4-6]. The profile of the disease differs among developed and developing countries. In developed countries, decrease in rheumatic heart diseases and increase in degenerative heart disease has led to an increase in patient age, accompanying comorbid diseases and increasing incidence of Staphylococcus aureus [3,7]. Rheumatic heart disease is still a major risk factor in developing countries, and degenerative valvular heart lesions are the most frequent anatomic abnormalities [1,8,9]. IE is frequently associated with rheumatic valvular disease in Turkey [10]. Classification of the disease is based on the activation pattern, presence of recurrence, pathogenesis, anatomic localization and the causative microbiological agent which are among the Duke criteria [11].

In this study we aimed to investigate the clinical manifestations, microbiological profile, echocardiographic findings, management strategies and complications of IE retrospectively, in our centre, which is a tertiary care hospital located in the South Marmara Region, with a population over 3 million people.

Materials and methods

Design

The study was designed as a retrospective observational single-center study.

Study population

Study population consisted of 85 consecutive adult patients treated for definite IE in our hospital. We evaluated the patients retrospectively, using archive records between the years 2009-2014. Inclusion criteria were; definite IE, according to the modified Duke criteria [11]. All of the patients were examined by transthoracic (TTE) and transesophageal echocardiography (TEE) on admission. TTE was performed at regular intervals and also in case of a change in clinical status. Echocardiographic data included routine parameters and presence of vegetation and abscess. Patients with possible or probable IE were not enrolled in the study.

Data collection

Data on demographic characters, age, sex, underlying heart disease, results of biochemical and microbiological investigations, echocardiographic findings, surgical requirements and in-hospital mortality data were collected.

Microbiologic culture and identification

Blood cultures were collected from all the patients, at least three samples from each patient. Blood was cultured in BACTEC (Beckton Dickinson, USA) and incubated in BACTEC FX 200 (Beckton Dickinson, USA). Identification of Staphylococcus spp. was performed using standard methods. Identification of streptococci and enterococci, Gram-negative enteric rods, Gram-negative non-fermentative rods, and Candida spp. were done with API Strep, API 20E, API NE, and API c AUX (bioMe´rieux, France), respectively, along with standard methods. Brucella spp. were identified using standard methods (biotyping, sensitivity to dyes, H₂S production, CO₂ requirement,) and species-specific antisera.

Statistical analyses

We used SPSS v 20 (IBM, Chicago, IL, USA) for statistical analyses.

Results

Baseline charactersitics of patients are shown in Table 1. There were 48 male (56.5%) and 37 female (43.5%) patients. The mean age was 52, 2 years, which ranged between 21 and 84. In 57 patients (67%), NYHA score was class III-IV on admission. Mean haemoglobin levels of patients were 10.1±2.3 g/dl and white blood cell (WBC) counts were found 13.5±7.2 × 10³/ml. Mean erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were found; 69.2±47.3 mm/h and 107.2±88.0 mg/l, respectively. There were 40 native (47.1%), 34 prostetic valve (40%) endocarditis in patients. The remaining 11 patients had pacemaker endocarditis. Rheumatic heart disease (22.4%) was the primary reason of native valve endocarditis, followed by degenerative heart disease (17.6%) (Table 2). There were 72 left sided endocarditis and 13 right sided endocarditis. Most common sites of vegatations are mitral valve which was detected in 32 patients (37.6%) followed by aortic valve with 24 patients (28.2%) and both mitral and aortic valve in 16 patients (18.8%) (Table 3). The most common valvular pathology was mitral regurgitation, which was detected in 38 patients.

Table 1.

Demographic characters, clinical features and laboratory tests of patients

Variables	Case; n (%)
Gender (M/F)	48/37 (56.5/43.5)
NYHA III/IV	57 (67.1)
Diabetes mellitus	15 (17.6)
Previous IE	1 (1.2)
Atrial fibrillation	15 (17.6)
Age (years)	52.2±15.0
Hemoglobin, g/dl	10.1±2.3
White blood cell x103/ml	13.5±7.2
RDW, %	17.3±3.7
MCV, fl	84.1±8.7
Platelet n/ml	238.741±127.983
Neutrophile/lymphocyte ratio	10.9±11.3
ESR, mm/h	69.2±47.3
CRP, mg/dl	107.2±88.0
Creatinine, mg/dl	1.4±0.2
Uric acid, mg/dl	6.7±2.7
Albumin, g/dl	3.1±0.8
ALT, U/l	55.8±121.4

M/F: Male/female, NYHA: New York Heart Association, RDW: Red cell distribution width, MCV: Mean corpuscular volume, ESR: Erythrocyte sedimentation rate, CRP: C reactive protein, ALT: Alanin amino transpherase.

Table 2.

Predisposing conditions of patients

Variables	Case n (%)
Native valve endocarditis	40 (47.1)
Rheumatic heart disease	19 (22.4)
Degenerative heart disease	15 (17.6)
Congenital heart disease	2 (2.4)
Other	4 (4.7)
Prosthetic valve endocarditis	34 (40.0)
Pacemaker/ICD lead endocarditis	11 (12.9)

ICD: implantable cardiac defibrillator.

Table 3.

Sites of vegetations in IE patients

Site	n	%
Mitral valve	32	37.6
Aortic valve	24	28.2
Aortic + Mitral valve	16	18.8
Tricuspid valve	2	2.4
Pulmonic valve	1	1.2
Tetralogy of Fallot	1	1.2

All patients had blood culture studies, but only 58 (68%) had positive blood cultures results. Most common pathogen was Staphlyococci which were detected in 23 patients (27.1%). Staphlyococci consisted of 16 coagulase negative Staphylococcus spp. and, 7 Staphylococcus aureus. Streptococcus spp. (10.5%), Brucella spp. (5.8%) and Enterococcus spp. (4.7%) and Candida spp. (5.8%) are other common pathogens (Table 4).

Table 4.

Distribution of causative microorganisms isolated from blood cultures of IE patients

Organism	Case n (%)
Defined causative agent	58 (68.2)
Stahylococci	23 (27.1)
MRSA	2 (2.4)
MSSA	5 (5.8)
MRCNS	10 (11.8)
MSCNS	6 (7.1)
Enterococcus spp.	4 (4.7)
Streptococcus spp	9 (10.5)
Brucella spp.	5 (5.8)
Gram negative	4 (4.7)
Candida spp.	5 (5.8)
Corynebacterium spp.	1 (1.1)
Mixed	3 (3.5)

Mixed infections; Candida spp. + Corynebacterium spp. 1 patient, Streptococcus spp + HACEK 1 patient, MRSA + Candida spp. 1 patient) (MRSA: Methicillin resistant Staphylococcus aureus, MSSA: Methicillin susceptible Staphylococcus aureus, MRCNS: Methicillin resistant coagulase negative Staphylococcus spp., MSCNS: Methicillin susceptible coagulase negative Staphylococcus spp.

Congestive heart failure was the most common complication which was detected in 23 patients (27%) during the disease course (Table 5). This was also the most common endication for surgery. Serebral embolism detected in 12 patients (14%) and 2 of them had reccurent embolism. Three patients had splenic abscess, two patients had perivalvular prostetic valve dehisence and one patiens had perivalvular abscess. Septic shock occured in 9 patients (10%) and multiple organ failure occured in 15 patients (17%).

Table 5.

Major complications led to surgery of IE

Indication for surgery	n	%
Congestive heart failure/severe valvular regurgitation	23 (52.2)	52.2
Resistance to medical treatment	16 (36.5)	36.5
Reccurent serebral embolism	2 (4.5)	4.5
Perivalvular Abscess	1 (2.3)	2.3
Dehisence	2 (4.5)	4.5
Total surgery	44 (100)	100

The presence of prostetic valve did not differ statistically in mortality. Mortality rates in left and right sided endocarditis were 35 and 38% respectively and the difference was not statistically significant. Mortality was higher in mitral valve endocarditis than aortic valve with 21 and 52% rate, which was statistically significant (P<0.05).

Surgical intervention was performed in 44 patients (52%) patients. Thirty one patients died during hospital follow-up. In-hospital mortality rate was 36%. The mortality rate was 46% with medical treatment and 27% with surgical treatment (P<0.05) (Table 6). In univariate analysis, risk factors that increased mortality in cases of IE were female gender, diabetes mellitus, mitral valve endocarditis, higher levels of RDW, creatinine and AST, lower levels of albumin (P<0.05) (Table 7). Decreased mortality in cases of IE was associated with surgical intervention for the treatement of IE (P<0.05) (Table 7). In multivariate analysis the independent risk factors for mortality were hipoalbuminemia, female gender, presence of DM, and high levels of creatinine. Surgical treatment for IE is found to be the protective factor against mortality.

Table 6.

Treatment strategies of IE patients

Treatment	Total Case	Mortality of cases n (%)
Overall	85	31 (36)
Medical treatment	41	19 (46)
Surgery	44	12 (27)
AVR	8	2 (25)
MVR	19	6 (32)
AVR + MVR	12	2 (17)
Lead extraction	4	2 (50)
PVR	1	0 (0)

AVR: Aortic valve replacement, MVR: Mitral valve replacement, PVR: Pulmonic valve replacement.

Table 7.

Baseline characteristics, predisposing conditions, clinical and laboratory findings of infective endocarditis patients who died and survived

Features	Survived	Death	Univariate P value	Multivariate
				beta	OR (95% CI)	P value
Age	49.4	57.1	0.357	-	-	-
Sex	M: 35, F: 20	M: 12, F: 18	0.041	-	-	0.028
Diabetes mellitus	5	10	0.005	0.20	(0.19-0.45)	0.005
Atrial fibrillation	7	8	0.138	-	-	-
Aortic regurgitation	16	7	0.487	-	-	-
Mitral regurgitation	21	17	0.158	-	-	-
Left sided endocarditis	47	25	0.437	-	-	-
Right sided endocarditis	8	5	0.513	-	-	-
Presence of prosthetic valve	21	13	0.275	-	-	-
Prothesis duration (years)	3.1	2.9	0.157	-	-	-
Cardiac surgery for IE	34	10	0.016	0.20	(0.34-0.65)	0.023
Haemoglobin, g/dl	10.4	9.6	0.065	-	-	-
White blood cell, nx103/l	12.2	15.6	0.610	-	-	-
Thrombocyte, n/l	250.278	218.161	0.535	-	-	-
Neutrophile/ lymphocyte ratio	9.0	14.5	0.103	-	-	-
RDW, %	17.1	17.4	0.017	0.15	(16.12-18.42)	0.593
MPV, fl	8.7	8.8	0.711	-	-	-
ESR, mm/h	65.2	76.0	0.261	-	-	-
CRP, mg/l	100.9	118.3	0.431	-	-	-
Albumin, g/dl	3.4	2.6	0.005	0.49	(2.82-3.15)	<0.001
Creatinine, mg/dl	1.2	1.9	<0.001	0.06	(1.13-1.81)	0.004
ALT, U/l	40.9	81.6	0.038	0.15	(38.5-99.2)	0.124

RDW: Red cell distribution width, MPV: Mean platelet volume, ESR: Erythrocyte sedimentation rate, CRP: C reactive protein, ALT: Alanin amino transpherase.

The mortality rate in culture positive and culture negative patients were similar; 39 and 32% respectively (P=ns) (Table 6). Mortality rate according to microorganisms were calculated and it was highest in S. aureus grown blood culture positive patients as 75%, but as the number of patients infected were low for each microorganism, we could not obtain a statistically significant result. There were seven culture positivity in 13 right sided endocarditis and the causative microorganisms were S. aureus (n=2, both MSSA), Brucella spp. (n=2), Streptococcus spp. (n=2), and one Enterococcus faecalis (n=1).

Discussion

This is the first study investigating infective endocarditis in the South Marmara Region of Turkey. Infective endocarditis is still related with high mortality, despite advances in diagnostic methods, antibiotic treatment, blood culture techniques and surgical treatment options. Prosthetic valve endocarditis was found to be the most common underlying disease for IE in our study, which is similar to the findings of Simsek-Yavuz et al. [12], but different from Elbey et al. which are the last two studies about IE from Turkey. Both of Simsek-Yavuz et al. and our study, investigated the patient’s, located in the high developed part of Turkey, that may be the reason for similarity in the results. Elbey et al. [1], has investigated 13 different centers from all around Turkey and have found that, rheumatic valvular diseases as the major underlying diseases which may be due to the fact that as they investigated low developed parts of Turkey. In developing countries, patient age, place of acquisition of infection and causative microorganisms are usually different from developed countries according to the higher rate of rheumatic heart disease. In developed counties there is a decrease in rheumatic heart disease and increase in degenerative heart disease which led to increase in patient age and incidence of Staphylococcus aureus [12-14]. Recent studies from developed countries found the mean age higher than 60 [14,15]. Though mean age of IE was found between 36 and 51 in Turkey, our study group consisted the highest mean age as 52, which may be due to the fact that South Marmara Region is a high developed part of Turkey [1,10,12,16-18]. The major factor contributing to the younger age of patients with IE may be because of the higher rate of chronic rheumatic heart disease in Turkey. In a recent study from Turkey, chronic rheumatic heart disease was found to be the leading cause of chronic valvular disease, which constituted 46% of all cases [19]. The rate of chronic rheumatic heart disease declined from 64% in 1990s to 18-36% to 2000s, of which our result 22.4% is one of the lowest [1,10,16]. Chronic rheumatic heart diseases are less than 10% of IE cases in developed countries [15]. As a underlying disease, our result is between developing and developed countries results. The number of patients with prostetic valve disease is high in our country as a result of higher prevelence of cronic rheumatic heart disease in previous years.

In our study, the most frequent causative microorganisms were staphylococci, streptococci, Brucella spp., Candida spp. and enterococci. Recent studies from the developed countries reported staphylococci and streptococci to be the most frequently isolated microorganisms [3,20]. Causative microorganisms also differed relative to the age ranges in our study. Streptococci were more frequent among patients aged <40 years and enterococci were more frequent among patients aged >50 years; this is in accordance with the current literature [12,21]. The determination rate of causative microorganisms in cases of IE differs between countries, probably due to the availability of advanced technologies. Causative microorganisms are determined in up to 90% of cases of IE in developed countries, but only 41-67% of cases of IE in developing countries [3,22-27]. We found 68% culture positive patients which is in accordance with previous studies from Turkey [1,12,16-18]. Our hospital is a referral hospital, to which patients are referred from peripheral hospitals, after beginning of antibiotic treatment, which diminishes our culture positivity rate.

Antimicrobial resistance was not a major problem among the microorganisms isolated from community-acquired endocarditis: all of the streptococci were sensitive to penicillin and all of the enterococci were sensitive to penicillin, high-level gentamicin, and vancomycin. Despite increasing reports of strains of viridans group streptococci that are resistant to penicillin and other antibiotics, penicillin-resistant viridans group streptococci are reported very rarely as a cause of IE, but we did not isolate penicillin-resistant viridans group streptococci [28]. Methicillin resistance was found in 3 of 8 (38%) of S. aureus and 10 of 16 (63%) of coagulase negative staphylococci.

The mortality rate of our patients was very high (36%). Our hospital is a referral centre for the medical and surgical treatment of IE, thus this higher mortality rate could be due to the referral of more complicated cases. The use of a surgical intervention was found to be a protective factor against mortality in our study. Our study was not designed to establish the impact of surgery on the mortality of patients with IE. Early surgical intervention has been reported as a protective factor for mortality in recent studies which is concordant with our study [29-31].

Our study, which reflects the developed part of Turkey, revealed that, mean patient age is higher than developing countries which is approaching to developed countries, prosthetic valve is the most common predisposing condition, valvular regurgitation and related congestive heart failure is the most common complication and leading cause of surgery. Staphylococci are the most common causative microorganism. Mortality rate is often high. Hipoalbuminemia, female gender, presence of DM, and high levels of creatinine were found to be independent risk factors for mortality in multivariate analysis. Surgical treatment for IE is found to be the protective factor against mortality.

Study limitations

The main limitations of our study were retrospective design, low patient number and being a referral centre. Being a referral centre might also have resulted in more culture negative results due to possible antibiotic usage prior to our hospital admission. Despite the higher prevalence of prosthetic valve IE, dehisence and abscess rates were lower compared with previous data. Due to local restrictions and retrospective limitations we could only gather data from the study institutions. We evaluated only the index hospitalization period and did not further follow outcome (such as surgery after discharge, reinfection or relapse) apart from mortality. Furthermore, the cardiac pacemaker leads were extracted with percutaneous procedures and were not grouped under the title of surgery.

Conclusion

In conclusion, present study demonstrated that IE remains a severe disease with a high mortality rate. Younger age, high prevalence of prosthetic valve endocarditis followed by rheumatic heart disease, more frequent staphylococci infection and higher rates of culture negativity were other important aspects of IE epidemiology in Turkey.

Disclosure of conflict of interest

None.