Impact of heart failure and preoperative platelet count on the postoperative short‐term outcome in infective endocarditis patients

Abstract Background Heart failure (HF) and platelet count are often considered risk factors for mortality in patients with infective endocarditis (IE); however, their effects on various complications have not been elucidated. Hypothesis We speculated that HF and platelet count have significant impact on the short‐term outcomes of IE. Methods This single‐center retrospective study analyzed data from 320 IE patients who underwent surgery. A multivariate Cox proportional hazards model was used to identify the risk factors for adverse outcomes. The effect of the platelet count on the prognosis of patients with HF was determined by subgroup analysis and Kaplan–Meier analysis. Results The study population was divided into the HF group (n = 102) and the non‐HF group (n = 218). The median age of the total population was 44.5 years (31–56 years), of which 227 (70.94%) patients were male. The incidence rates of 1‐year all‐cause mortality, cardiac outcomes, and composite outcomes were respectively almost sixfold, fourfold, and threefold higher in the HF group than in the non‐HF group (all p < 0.001). In multivariate Cox regression analysis, HF was an independent risk factor for 1‐year all‐cause mortality, cardiac outcomes, cerebral outcomes, and composite outcomes. The Kaplan–Meier survival curves revealed that the patients with both HF and thrombocytopenia demonstrated the worst composite outcomes than the patients of the other groups (log‐rank p < 0.001). In the HF group, the platelet count was significantly associated with mortality and composite outcomes. Conclusions HF and preoperative platelet count are significantly associated with 1‐year all‐cause mortality and adverse outcomes postoperatively in IE patients. Patients with HF and thrombocytopenia have the worst short‐term prognosis.


| INTRODUCTION
Infective endocarditis (IE) is a systemic infectious disease characterized by valvular destruction and vegetation formation and is accompanied by serious complications and significant mortality. 1,2spite recent improvements in the diagnosis and antimicrobial and surgical management, the incidence of IE continues to increase every year. 3e condition of IE patients is complex and often accompanied by complications, such as acute renal failure, cardiovascular complications, neurological complications, and gastrointestinal complications, which are important causes of death and adverse outcomes. 4though approximately half of IE patients receive surgical treatment and have a satisfactory overall therapeutic effect, the mortality rate remains as high as 8%. 5,6[9] However, in addition to death, other systemic complications also deserve our attention.These complications not only increase the mortality risk but also reduce the quality of life and increase the patient's medical expenses and the length of hospital stay.For instance, cerebral infarction does not necessarily cause death; however, the associated hemiplegia and impaired consciousness greatly reduce the quality of life.Therefore, many details of the preoperative evaluation and management of IE remain to be optimized.In addition to assessing the mortality risk, timely intervention is crucial to reduce the incidence of complications.
There is an interrelationship between HF and platelets.On the one hand, HF affects platelet production, activation, and function. 10 the other hand, platelets interact with leukocytes and endothelial cells and release inflammatory factors that lead to the adhesion and migration of monocytes. 11Activated platelets may lead to disease progression in HF and the development of coronary atherosclerosis and microthrombosis, thereby increasing the risk of sudden death. 12reduction in platelet count is independently associated with mortality and reduced ejection fraction in HF. 13 IE has multiple pathophysiological processes, such as severe inflammatory response, HF, and thrombosis.Moreover, the pathophysiological mechanism of the vicious cycle between HF and thrombocytopenia is likely to affect the prognosis of IE; however, studies in this area are lacking.
This study aimed to investigate the risk factors for postoperative short-term mortality and complications in patients with IE and to evaluate the effect of HF and platelet count on the prognosis of IE. ).The main analysis included 320 patients who underwent cardiac surgery for IE (Figure 1).According to the European Society of Cardiology Guidelines for the management of IE, all patients' surgical indications and medical management were evaluated by the endocarditis team. 4In total, 50 potential variables were collected and analyzed separately, including demographics, medical history, clinical characteristics, laboratory results, echocardiography findings, and surgical details.This study was reviewed and approved by the Institutional Review Board of The First Affiliated Hospital of Sun Yat-sen University (approval number: 2018 [100]) and was consistent with the Declaration of Helsinki.The authors confirm that patient consent is not applicable to this article because it is a retrospective case report using deidentified data.

| Definition and outcomes
All investigators were blinded to the patient information.The preoperative platelet count was obtained from the last routine blood test before surgery, usually within 3 days.Thrombocytopenia, a categorical variable, was defined as platelet count < 150 × 10 9 /L. 14e minimum inpatient value of serum creatinine during the first 7 days of admission was used as the baseline creatinine level.
Preoperative renal insufficiency was defined as acute kidney injury or chronic kidney disease stages 3-5 according to the Kidney Disease Improving Global Outcomes criteria. 15,16The postoperative adverse outcomes were classified as follows: (1) all-cause mortality occurring    The incidence of HF was 31.88% (n = 102).Compared with the non-HF group patients, those in the HF group were older and had more comorbidities such as hypertension, preoperative renal failure, congenital heart disease, and pulmonary hypertension, higher incidence of doublevalve infection, higher frequency of positive blood cultures for

| Baseline characteristics
Staphylococcus aureus, and higher levels of white blood cells, creatinine, urea nitrogen, aspartate aminotransferase, and total bilirubin.However, the hemoglobin, platelets, and serum albumin were lower in the HF group than in the non-HF group (all p < .05).

| Postoperative outcome and Cox regression analysis
During the 1-year follow-up period, composite adverse outcomes were reported in 19.06% of the patients (n = 61).Additionally, 26 patients (8.13%) died, 48 (15%) had cardiac outcomes, 11 (3.44%) had cerebral outcomes, and 26 (8.13%) had two or more adverse events (Table 1).Notably, the incidence of adverse events was significantly higher in the HF group than in the non-HF group.The incidence rates of 1-year all-cause mortality, cardiac outcomes, and composite outcomes were respectively almost sixfold, fourfold, and threefold higher in the HF group than in the non-HF group (all p < .001).

| Subgroup analysis
Subgroup multivariate Cox regression analysis was performed to explore the interaction between HF and preoperative platelet count further (Table 3).In the HF group, the preoperative platelet count continued to be associated with 1-year all-cause mortality (adjusted HR = 0.984; 95% CI, 0.976-0.991;p < .001)and composite outcomes (adjusted HR = 0.994; 95% CI, 0.990-0.999;p = .011).However, in the non-HF group, the platelet count was not an independent risk factor for any outcome (all p > .05).
Based on the stratification by HF, the relationship between the preoperative platelet count and the risk of various outcomes was analyzed using RCS.Using the cardiac outcomes (Figure 2B) and composite outcomes (Figure 2D) as the end events, a U-shaped association was observed between the preoperative platelet count and risk in the HF group, with a higher risk than that in the non-HF group.However, the preoperative platelet count was linearly associated with cardiac outcomes and composite outcomes in the non-HF group (Figures 2B,2D).Using 1-year all-cause mortality (Figure 2A) and cerebral outcomes (Figure 2C) as the end events, an L-shaped association was observed between the preoperative platelet count and risk in the HF group, with a higher risk than that in the non-HF group only when the platelet count was low.
The Kaplan-Meier curves revealed the relationship between adverse outcomes and HF and thrombocytopenia.Notably, the patients with both HF and thrombocytopenia demonstrated higher incidence rates of 1-year all-cause mortality (Figure 3A), cardiac outcomes (Figure 3B), and composite outcomes (Figure 3D) than those of patients with either HF or thrombocytopenia or those with neither HF nor thrombocytopenia (all log-rank p < .001).

| DISCUSSION
This single-center retrospective study evaluated the influence and interaction of HF and preoperative platelet count on short-term postoperative outcomes in IE patients.We found that HF was significantly associated with the incidence of all postoperative adverse events, and preoperative platelet count was associated with 1-year all-cause mortality, cardiac outcomes, and composite outcomes.In the HF subgroup analysis, the preoperative platelet count retained its effect on 1-year all-cause mortality and composite outcomes.More importantly, the adverse effects of thrombocytopenia were amplified by HF.In the Kaplan-Meier survival analysis, patients with HF and thrombocytopenia had the worst short-term prognosis.
Bacteremia and valve vegetation are important parts of the pathophysiology of IE.This vegetation is an excellent shelter for bacteria, which not only makes anti-infective processes difficult but is also one of the main reasons for embolism and hemorrhage. 17atelets, as the main component of vegetation, have been considered a promising target for IE management. 18,19The application of antiplatelet therapy to reduce vegetation and emboli remains controversial in the clinical treatment of IE.In the management of platelets, there may be a transitional zone between promoting vegetation formation and increasing bleeding risk and controlling this balance is difficult.Some investigators have reported that antiplatelet treatment has a limited effect in reducing the risk of embolic events and is probably associated with an increased risk of bleeding. 20,21eeding complications often lead to a worse prognosis in IE patients.
Additionally, the platelet count was a significant predictor of early mortality after surgery in IE patients in other studies. 7Therefore, we should be alert regarding the risk of bleeding complications and death caused by a low platelet count, rather than vegetation formation and embolic events, especially under appropriate antibiotic therapy.
Antimicrobial treatment is considered the best method to reduce the risk of embolic events and improve the prognosis of IE. 22 IE patients encounter the double threat of pathogenic micro- During sepsis, activated platelets have been shown to limit both bacterial growth and dissemination, and reduce bacterial load. 23,24rdiac outcomes were the primary postoperative adverse outcomes, accounting for 78.7% of the composite outcomes.Whether platelets can enhance myocardial protection is an important issue for discussion.It has been reported that platelet-derived sphingosine-1 phosphate can induce myocardial protection, and attenuate ischemia-reperfusion injury. 25In a rat model of myocardial infarction, platelet gel was shown to preserve cardiac function and attenuate adverse ventricular remodeling without exacerbating cardiac inflammation. 26ditionally, activated platelets play a beneficial role in ischemiareperfusion injury. 27Although the aforementioned studies were not based on IE, therapies targeting platelets may be a unique and promising therapeutic approach to improve its prognosis.
HF is the most common complication of IE as well as the most common surgical indication for IE. 1,4In our study, HF was present in almost one-third of the patients, which concurred with previous studies. 28Currently, studies on IE complicated by HF focus on mortality.In addition to mortality, adverse events such as low cardiac output syndrome, malignant arrhythmias, bleeding events, or cerebral infarction deserve our attention.These complications can prolong the patient's hospital stay, increase medical costs, and reduce the quality of life.In our study, the patients with HF were associated with a significantly higher risk of 1-year all-cause mortality and adverse outcomes when compared with those without HF (all p < .05).The heart is the power source of systemic which is associated with high mortality and complication rates. 29cond, pulmonary circulation is obstructed in patients with HF, leading to an accumulation of alveolar fluid that not only impairs bacterial clearance but also undermines the local defenses against infection, thereby increasing the risk of pneumonia. 30,311][32] Finally, the incidence of bivalvular involvement was higher in the HF group than in the non-HF group (22.55% vs. 12.84%; p = .027)in our study, which meant longer cardiopulmonary bypass time and more cardiac structural damage, resulting in an increased incidence of embolic events and gastrointestinal complications. 8,33,34 addition to exploring the impact of HF and platelet count on the short-term prognosis of IE patients, we found a mutual promoting effect between HF and thrombocytopenia.According to our findings, platelets had a greater impact on the prognosis of HF patients than of non-HF patients.Event-free survival in patients with both HF and thrombocytopenia was significantly shorter than that in other patients.Although we have reported a novel association between thrombocytopenia and HF, we acknowledge that the underlying mechanism remains uncertain, especially in the context of coagulopathy, inflammatory storm, and surgical trauma.In these patients, it is necessary to improve cardiac function and correct thrombocytopenia, which may lead to a better prognosis.

| LIMITATIONS
Our study has several limitations.First, since this is a single-center retrospective study, the generalizability of its conclusions needs further validation.However, the frequency of HF, death, and adverse outcomes in our study was similar to that reported in other studies.
Second, all study data were obtained from electronic medical record

| CONCLUSIONS
The results of this study suggest that HF and preoperative platelet count are significantly associated with 1-year all-cause mortality and adverse outcomes postoperatively in IE patients.Patients with HF and thrombocytopenia have the worst short-term prognosis.

Furthermore, we analyzedMETHODS 2 . 1 |
and discussed the influence of HF and platelets on various complications in IE patients.2| Study population and data collectionThis retrospective observational study enrolled 652 consecutive patients diagnosed with IE at The First Affiliated Hospital of Sun Yatsen University between October 2013 and November 2019.All patients definitively fulfilled the modified Duke criteria for IE.The exclusion criteria were as follows: received conservative medical treatment (n = 278), missing data regarding the main variable (n = 13), death before surgery (n = 4), age < 18 years (n = 7), pregnancy (n = 2), advanced malignant tumors (n = 23), and participation in other clinical trials (n = 5 during hospitalization or within 1 year postoperatively; (2) cardiac outcomes, including acute HF, acute myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, malignant arrhythmia, shock, acute pericardial tamponade, or use of ventricular assist devices such as extracorporeal membrane oxygenation (ECMO) or intra-aortic balloon pump; (3) cerebral outcomes, including cerebral hemorrhage, cerebral infarction and cerebral hernia confirmed on computed tomography or magnetic resonance imaging, or transient ischemic attack; and (4) composite outcomes, including gastrointestinal bleeding or ischemia, large retroperitoneal hematoma, pulmonary embolism, septic shock, and other conditions that caused significant symptoms or were life-threatening, in addition to the aforementioned adverse outcomes.When more than two adverse events occurred, the earliest time was recorded as the composite adverse event time.All postoperative complications were confirmed by experienced specialists.Comorbidities were defined according to the International Classification of Diseases, Ninth Revision, Clinical Modification.
Continuous variables are presented as means ± standard deviations or as medians with interquartile ranges based on normality.Categorical variables are presented as group frequencies and percentages.The student's t-test and Mann-Whitney U-test were used to compare continuous variables.The χ 2 test or Fisher's exact test was used to compare categorical variables.Subgroup and multivariate analyses of the association between covariates and adverse outcomes were performed using a Cox proportional hazards model.The assumption of proportional hazards was tested using scaled Schoenfeld residuals.Based on previous research and medical experience, the following covariates were considered to have a potential influence on the adverse outcomes: age, HF, double-valve IE, right-sided IE, atrial fibrillation, vegetation size > 10 mm, and preoperative hemoglobin levels.Kaplan-Meier analysis was conducted to estimate the event-free survival of adverse outcomes, with log-rank tests used to discriminate between the Kaplan-Meier curves.The visualized relationship between the preoperative platelet count and adverse outcomes in different HF subgroups was assessed using restricted cubic spline (RCS) curves.All tests were two-tailed, and p < .05were considered statistically significant.All statistical analyses were performed using the SPSS (version 26.0) software (SPSS Inc.) and the R statistical package (The R Foundation; http:// www.r-project.org;version 4.1.3).
This study included 320 hospitalized IE patients who had undergone cardiac surgery from October 2013 to November 2019.The study population was divided into the HF group (31.88%, n = 102) and the F I G U R E 1 Flow chart for inclusion, exclusion, and grouping of patients with infective endocarditis.IE, infective endocarditis.T A B L E 1 Clinical characteristics and outcomes of IE patients undergoing surgery with and without heart failure.
organisms and surgical trauma.Their baseline condition is poor, making them more vulnerable to HF and death.The role of platelets in the prognosis of IE has not yet been characterized, especially with respect to surgical trauma and severe infection.In this study of postoperative complications of IE, platelet count correlated significantly with cardiac outcomes and composite outcomes in addition to the 1-year all-cause mortality.This suggests that platelets may have a potential role in infection control, myocardial ischemia-reperfusion injury, and organ protection other than thrombosis and bleeding.
blood circulation.Hemodynamic disorder and organ hypoperfusion caused by HF are important causes of death and multiple organ F I G U R E 2 The restricted cubic spline analysis on the association between the preoperative platelet counts and the risk of adverse outcomes in different heart failure subgroups (A) 1-year all-cause mortality (B) Cardiac outcomes (C) Cerebral outcomes (D) Composite outcomes.CI, confidence interval; HR, hazard ratio.dysfunction.First, although surgery is an important treatment to improve the prognosis of IE patients, cardiopulmonary bypass and cardiac arrest during surgery can cause serious trauma to the heart, especially in patients with preoperative HF symptoms.Most patients with HF have improved cardiac function and vegetation clearance postoperatively.However, a small part of patients develop low cardiac output syndrome and cardiogenic shock postoperatively and require ECMO to maintain circulation stability,

F I G U R E 3
The Kaplan-Meier survival curves estimates of event-free survival for adverse outcomes associated with thrombocytopenia and heart failure (A) 1-year all-cause mortality (B) Cardiac outcomes (C) Cerebral outcomes (D) Composite outcomes.systemsand patient follow-up, rather than prospective collection, which have inherent limitations.Several patients were excluded due to missing data, which may have introduced a potential bias.Finally, the platelet count was obtained from the last blood routine results before surgery.Some patients with thrombocytopenia on admission may have recovered after undergoing blood transfusion or medical treatment, and these interventions could potentially affect the prognosis.

Table 3
shows the multivariate Cox regression analysis of the four major postoperative adverse outcomes.After adjustment for age, doublevalve IE, right-sided IE, atrial fibrillation, vegetation size > 10 mm, and preoperative hemoglobin levels, HF was considered an independent risk factor for 1-year all-cause mortality (adjusted hazard ratio [HR] = 5.050; 95% CI, 2.067-12.340;p<.001),cardiacoutcomes(adjustedHR = 3.556; 95% CI, 1.937-6.527;p<.001),cerebraloutcomes(adjustedHR = 3.671;TA B L E 1 (Continued)Note: Bold values indicate statistical differences in variables between the heart failure group and the non-heart failure group (p < .05).Continuous variables are presented as means ± standard deviations or as medians with interquartile ranges based on normality.Categorical variables are presented as group frequencies and percentages.Abbreviations: AST, aspartate transaminase; COPD, chronic obstructive pulmonary disease; HF, heart failure; LVEF, left ventricular ejection fraction.a Statistically significant.p = .007).Thus, both HF and preoperative platelet count were considered independent risk factors for poor prognosis.
T A B L E 2 Surgical indications and details of IE patients with and without heart failure.
Note: Categorical variables are presented as group frequencies and percentages.a Statistically significant.
Multivariable cox regression analysis of adverse outcomes.