Early Mortality Predictors in İnfective Endocarditis Patients: A Single-Center Surgical Experience

Introduction Infective endocarditis is a disease that progresses with morbidity and mortality, afecting 3-10 out of 100,000 people per year. We conducted this study to review the early outcomes of surgical treatment of infective endocarditis. Methods In this retrospective study, 122 patients who underwent cardiac surgery for infective endocarditis in our clinic between November 2009 and December 2020 were evaluated. Patients were divided into two groups according to in-hospital mortality. Demographic, echocardiographic, laboratory, operative, and postoperative data of the groups were compared. Results Between November 3, 2009, and December 7, 2020, 122 patients were operated for infective endocarditis in our hospital. Emergency surgery was performed in nine (7.3%) patients. In-hospital mortality occurred in 23 (18.9%) patients, and 99 (81.1%) patients were discharged. In-hospital mortality was related with older age, presence of periannular abscess, New York Heart Association class 3 or 4 symptoms, low albumin level, high alanine aminotransferase level, and longer cross-clamping time (P<0.05 for all). Conclusion The presence of paravalvular abscess was the most important prognostic factor in patients operated for infective endocarditis.


INTRODUCTION
Infective endocarditis (IE) is a disease that progresses with morbidity and mortality, affecting 3-10 out of 100,000 people per year [1] . Despite early diagnosis and surgical interventions, hospital mortality was 17.1% in the European Infective Endocarditis (or EUROENDO) study published in 2019 [2] . Surgical intervention in patients with IE is required because of heart failure, uncontrollable infection, and prevention of embolism. Almost half of IE patients undergo heart surgery during hospitalization [3] . We conducted this study to review the early outcomes of surgical treatment of IE and to explain the impact of demographic, clinical, echocardiographic, and intraoperative parameters on in-hospital mortality and morbidity of IE patients after surgical treatment.

METHODS
In this retrospective case-control study, patients who underwent cardiac surgery for IE in our clinic between November 2009 and December 2020 were evaluated. During this period, a total of 122 patients were operated for IE in our clinic. Their baseline demographic data, echocardiographic data, performance status, laboratory data, operative data, and postoperative status were comprehensively collected. Blood cultures were taken from all patients at the time of admission. Empirical broad-spectrum antibiotics were administered to patients with no known recent blood culture results. Then, specific treatment was arranged according to hemoculture and antibiogram results. Routine laboratory tests and blood, valves, and vegetations cultures were performed during hospitalization and treatment. Transesophageal echocardiography was performed after transthoracic echocardiography in all patients to determine surgical strategy. Surgical procedures for IE were performed using conventional cardiopulmonary bypass. All infected tissue was resected, and a physiological or anatomical surgical reconstruction was performed. We defined emergency surgery as an operation with a refractory cardiac problem, which will not respond to any treatment other than cardiac surgery, and where there should be no delay in operative intervention. Hospital mortality was defined as mortality occurring within 30 days postoperatively or without discharge. This study was approved by the local ethics committee of our hospital (2018-23) and complies with the standards of the Declaration of Helsinki and current ethical guidelines.

Statistical Analysis
Statistical analyses were performed using the IBM Corp. Released 2015, IBM SPSS Statistics for Windows, version 23.0, Armonk, NY: IBM Corp. Continuous variables are expressed as mean ± standard deviation and categorical data as proportions throughout the manuscript. Categorical variables were compared using the χ2 test or Fisher's exact test, and independent continuous variables were compared by the unpaired Student's t-test or Kruskal-Wallis test as appropriate. Logistic regression analysis was performed to determine the predictors of in-hospital mortality. P-value < 0.05 was considered statistically significant.
In-hospital mortality occurred in 23 (18.9%) patients, and 99 (81.1%) patients were discharged. Demographic, preoperative laboratory, and clinical characteristics of patients with and without in-hospital mortality were compared ( Table 2). In-hospital mortality was related with older age, presence of periannular abscess, NYHA class 3 or 4 symptoms, low albumin level, high alanine aminotransferase (ALT) level, and longer crossclamping time (P<0.05 for all).
Except for wound complication and focal neurological deficit, other complications were higher in the group of patients with in-hospital mortality. But there was no statistically significant difference between the two groups in terms of intensive care unit stay (Table 3). Postoperative mesenteric hemorrhage was seen in one patient. Tracheostomy was performed in one patient due to prolonged hospitalization. Nine patients were rehospitalized due to pleural or wound complications.
Univariate and multivariate analyses were performed to identify independent risk factors related to in-hospital mortality. Univariate variables with P<0.05 were included in the multivariate analysis. Table 4 shows that older age, NYHA class 3-4 symptoms, and presence of periannular abscess were independently associated with in-hospital mortality after IE surgery in the multivariate analysis.

DISCUSSION
The epidemiology of IE has changed significantly over the past three decades [4] . Mean age of the patients in our study was 52.53 years; compared to developed countries, patients with IE were mostly young [5,6] . The male rate was found to be 68.9%, which is consistent with male prevailing in the literature [7] . In our study, 7.3% of the patients were operated under emergency conditions. The duration of antibiotic use before the operation was 17.18 days in all patients.
In our cohort, 67.3% of the patients were operated for native valve endocarditis and 32.7% for prosthetic valve or cardiac  Table 2. Comparison of demographic, preoperative laboratory, and clinical characteristics between patients with in-hospital mortality and patients without in-hospital mortality.   [8] .

Patients' Characteristics n (%)/mean±SD n (%)/mean±SD
The most common pathogen in our study was staphylococci, similar to the literature [9] . The reason for the high culture negativity in our study may be the patients referred to our hospital. It can be explained by the negative blood culture taken from patients diagnosed with IE in another hospital and started on antibiotic therapy [10] .
In-hospital mortality rate was 18.9%, close to the two large international registries [2,8] .
As shown in our study, mortality was higher in elderly patients with IE. The high incidences of IE in the elderly and its clinical and echocardiographic features have been emphasized in many studies [11][12][13] .
No significant relationship was found between vegetation length > 1 cm and in-hospital mortality. However, there are   [14] .
In our study, we found a relationship between periannular abscess and in-hospital mortality. The presence of periannular abscess increased four times in-hospital mortality. Radical debridement of abscess cavities is an essential procedure in cardiac surgery and is important in active IE. In most cases, reconstruction using a pericardial patch is required to close the abscess cavity [15] . Necessary aggressive surgical treatment results in high complication and mortality rates.
In the in-hospital mortality group, the serum albumin level was found to be low in the preoperative period. Hypoalbuminemia increases mortality as it is associated with malnutrition and frailty [16] .

Limitations
The limitations of this study were the small number of patients, the review of short-term results, and the fact that it was a retrospective study. The study offered the opportunity to evaluate surgically treated IE patients from a single referral tertiary care center.

CONCLUSION
Although in-hospital mortality was related with older age, presence of periannular abscess, NYHA class 3 or 4 symptoms, low albumin level, high ALT level, and longer cross-clamping time in univariate analysis, multivariate analysis showed that the presence of paravalvular abscess was the most important prognostic factor in patients operated for IE.