Cardiac surgery timing on the prognosis of patients with infective endocarditis Authors: Wu Liu, Yongqin Li, Heng Yang, Jingyu Wang, Wanqi Lan, Congcong Li, Yanhua Tang Published: 2025 ABSTRACT AbstractBackgroundInfective endocarditis (IE) is a disease caused by the involvement of pathogenic microorganisms in the endocardium, primarily characterized by the formation of growths and progressive valve damage. Despite advances in diagnostic and therapeutic methods, the mortality rate remains high. Conventional treatment typically favors surgery after 4-6 weeks of infection control. Although a series of articles on early surgical treatment have been published in recent years, outcomes for early surgical treatment continue to vary.MethodsBy collecting the clinical data of 166 patients who underwent cardiac surgery for IE between February 1, 2017, and January 31, 2023, we classified the patients into three groups: Group A (radical surgery within 1-7 days after admission), Group B (surgical treatment 8-13 days after admission), and Group C (surgical treatment 14-28 days after admission) according to different treatment strategies. To compare the effects of different times of surgical intervention on the early prognosis of patients, logistic multivariate regression analysis was used to determine the risk factors associated with surgical mortality, and the survival of different groups of patients was compared using Kaplan-Meier survival analysis.ResultsThe results showed no significant differences in perioperative reinfarction, endocarditis recurrence, heart failure, or in-hospital mortality among the three treatment groups. However, ICU length of stay (P=0.015, 1.30 [1.05-1.61]), cardiac malfunction (P=0.015, 22.28 [1.82-273.04]), and other organ malfunctions (P=0.007, 29.21 [2.46-346.49]) were identified as risk factors affecting patient mortality.ConclusionsEarly surgical intervention did not increase the risk of death in patients with IE, nor did it increase the risk of recurrence and reoperation within one year. Therefore, we recommend early surgical intervention before completing a cycle of antibiotics in patients with a clear diagnosis, unstable cardiac function, or risk of recurrent embolism.Keywords: Endocarditis, Timing of surgery, Early prognosis Introduction Infective endocarditis (IE) is a disease caused by pathogenic microorganisms that infect the endocardium and is characterized by the formation of growths and progressive valve damage. The incidence of IE is reported to be about 3–10 cases per 100,000 people, with a mortality rate of about 20% during hospitalization, increasing to 25–30% at 6 months and up to 40% at 5 years [1, 2]. Despite improvements in diagnosis, antibiotics, and surgical treatment, early surgical intervention is recommended when antibiotics alone are not curative, when the redundant organism is large and at risk of dislodging, when peripheral arterial embolism has already occurred, or when severe cardiac insufficiency follows valvular destruction. Both American and European guidelines indicate that all decisions regarding the timing of surgical intervention for IE should be made by a multidisciplinary endocarditis team consisting of experts in infectious diseases, cardiology, and cardiac surgery, who ultimately develop an individualized treatment plan for patients with IE [3]. The traditional concept of treatment favors conventional surgery after 4–6 weeks of infection control. In recent years, a series of articles have been published comparing the therapeutic effects of early surgical treatment with those of conventional treatment, but the findings on early surgical treatment vary. Early surgical intervention, as recommended by the American College of Cardiology/American Heart Association (ACC/AHA) guidelines, is defined as ‘during the initial hospitalization and before completion of a full course of antibiotic therapy’, especially when IE complications such as severe cardiac destruction, inadequate response to antibiotic therapy, or prevention of an embolic event occur [3]. The European Society of Cardiology (ESC) guidelines, on the other hand, are more specific concerning the timing of surgery and classify the timing into the following categories: super-emergency (within 24 h), emergency (within a few days), and elective situations (after at least 1–2 weeks of antibiotic therapy) [4]. To date, there is insufficient evidence that early surgery within days of an IE diagnosis is associated with lower mortality, and uncertainty remains about the appropriate timing of surgery. Despite the clear definition of indications for surgery in international guidelines [5], there is no consensus on the timing of surgical intervention during the active phase of infection due to the lack of evidence-based data. This is because of the risks of uncontrolled sepsis, shock, and organ failure during the acute or active phase of IE, as well as the possibility of high operative mortality and recurrence of IE caused by early surgical intervention. On the other hand, delaying surgery to complete a course of antimicrobial therapy may increase the risk of embolism, leading to extensive cardiac tissue damage, increased repair difficulties, progressive cardiogenic shock, organ failure, and ultimately increased mortality. Studies have shown that multidisciplinary endocarditis teams recommend that the time between indication for surgery and intervention should be shortened [6], but it is not clear whether early intervention is independently associated with lower mortality. Overall, the decision to operate on patients with IE is complex. Therefore, we conducted a single-center systematic review based on the tradition of multidisciplinary endocarditis team treatment of patients with IE in our center to compare the impact of surgical timing interventions on early prognosis in patients with IE, and to summarize the effectiveness of early surgical treatment options. Methods Study population and clinical data This study collected clinical data, blood cultures, and serological results from February 1, 2017, to January 31, 2023. The study was approved by the Ethics Committee of the Second Affiliated Hospital of Nanchang University (No.2023074). Inclusion criteria for this study cohort were patients diagnosed with definite IE according to the latest Duke criteria [7]. All surgical patients met the indications for surgery: (i) involvement and valves with moderate-to-severe regurgitation (acute, severe left-sided regurgitation), aortic abscess or destructive penetrating lesions, congestive heart failure (New York Heart Association classification, NYHA III-IV); (ii) persistent bacteremia (not responding to antibiotic therapy), fungal infections; (iii) active valvular excrescence greater than 10 mm in diameter with clinical evidence of recurrent embolization; and (iv) prosthetic valve involvement. To maintain the independence of observation, only the first medical visit of patients with IE was used. Exclusion criteria : (i) surgery should be delayed for at least 4 weeks in the following cases: short life expectancy, multiple organ failure, combined with hemorrhagic stroke, brain abscess, or infectious cerebral aneurysm. (ii)The patient refused surgery. All patients underwent preoperative transthoracic echocardiography (TTE) or transesophageal echocardiography (TEE). The maximum length of the Vegetation length was measured by cardiac ultrasound [8] and checked for fistulae, abscesses, and perforations. The endpoints were death and a composite endpoint at 1 year. Composite endpoint events were defined as death from any cause, postoperative embolic event, recurrence of IE, postoperative valve dysfunction (PVD), and rehospitalization for heart failure. Definition Pathogenic organisms that were detected positively from heart valves, blood, and sputum included Staphylococcus spp., Streptococcus spp., Enterococcus spp., and fungi (Candida glabrata, Proximo Smooth Pseudoalbicans, Bacillus thiaminolyticus, and fungal spores). Diagnosis to admission: time from the first diagnosis of definite IE according to the latest Duke criteria to admission to our center. History of other systemic infarcts: including pulmonary, hepatic, splenic, and renal mycotic infarcts. History of the peripheral arterial system: except for vascular lesions other than large vessels, e.g., atherosclerosis, stenosis, arteriovenous fistulae, and aneurysms. Other procedures: repair of patent foramen ovale failure, pericardial stripping, atrial/ventricular septal repair, atrial thrombus removal, single internal mammary artery-bypass grafting, left auricle ligation, valvular annuloplasty, and other procedures necessary for the patient. Perioperative reinfarction: including sub/acute cerebral infarction, arterial occlusion, pulmonary embolism, etc. The degree of aortic, mitral, tricuspid, and pulmonary regurgitation was assessed using semi-quantitative or quantitative methods and categorized as mild, moderate, and severe [9]. Patient Patients were divided into three groups: Group A (radical surgery within 1–7 days of admission), Group B (surgical treatment 8–13 days after admission), and Group C (surgery 14–28 days after admission). The timing of the patient’s surgery was determined through a multidisciplinary team (MDT) discussion by an endocarditis team, comprising an infectious disease specialist, a neurologist, a general surgeon assessing embolic damage to abdominal organs, and a vascular surgeon for peripheral embolism with limb ischemia. All patients with infective endocarditis (IE) who met the surgical indications for cardiac surgery were retrospectively analyzed; 166 patients had complete data. Of these, 124 (74.7%) were males and 42 (25.3%) were females. Aortic valve endocarditis occurred in 84 (50.6%) patients, and mitral valve endocarditis in 89 (53.6%) patients, with some patients having both aortic and mitral valve involvement. Six patients (3.6%) had previously undergone valve replacement surgery (mitral or aortic valve) and developed endocarditis on the implanted prosthesis. All patients were contacted for follow-up one year after surgery, and regular clinical and echocardiographic examinations were performed according to guidelines. Operation Surgery was performed under general anesthesia with extracorporeal circulation, utilizing mild hypothermia (28–32 °C). Twenty-two cases were conducted with small thoracoscopic incisions, while the remainder were performed with a median sternotomy. Myocardial protection was achieved using cold blood cardioplegia administered either anterogradely or retrogradely, followed by periodic administration every 30 min and local hypothermia to protect the myocardium. The total operative time for the cohort was 300 ± 37.5 min, with a cardiopulmonary bypass (CPB) time of 109 ± 25.75 min and an aortic cross-clamp time of 76 ± 21 min. The cardiac chambers and valves were meticulously examined to excise redundant tissue, damaged valves, and necrotic tissue, remove perivalvular abscesses, and based on the extent of the lesions, repair the annulus and valves, replace prosthetic chordae tendineae and valves, and repair the interventricular septum. Perivalvular septal defects and residual fistulas were also repaired. The native or prosthetic valves were completely removed and sent for microbiological culture. Fifty-seven (34.34%) patients received bioprostheses, while 109 (65.66%) received mechanical valve prostheses. Intraoperative irrigation with antibiotic solution and saline was repeatedly performed to prevent postoperative infection recurrence. Statistical method Preoperative, intraoperative, and postoperative indicators were collected from the healthcare system. First, the assumption of normality of the variable distributions was tested using the Shapiro-Wilk test. Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Next, risk factors associated with postoperative mortality were analyzed. All available variables were entered into a univariate logistic regression analysis to identify significant differences between survivors and deceased patients. In the second step, all variables with p-values less than 0.1 were included in a multivariate analysis using stepwise selection. Additionally, variables expressing or containing similar parameters were intentionally omitted in constructing the multivariate model to avoid covariance. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests. The overall probability of survival and the corresponding survival curves were analyzed using the Kaplan-Meier method. Analyses were performed using IBM SPSS Statistics 26 and GraphPad Prism 9 software. Results Patient characteristics Of the 308 patients diagnosed with infective endocarditis (IE), 191 (62.01%) had at least one indication for surgery, and a total of 96 patients (50.26%) were referred by the hospital. Twenty-five patients (13.09%) were excluded either because they did not undergo surgery until after the completion of antimicrobial therapy or because they refused surgery due to the high surgical risk. The remaining 166 patients (86.91%) who underwent surgery during antimicrobial treatment comprised our study cohort, with 86 patients in Group A, 51 in Group B, and 29 in Group C. Tables 1, 2, and 3 display the characteristics of these groups. Table 1Preoperative patient basic information on infective endocarditisA(D1-7)No. (%)C(D14-28)P ValueB(D8-13)N = 86N = 51N = 29Age,mean, years49 ± 15.0350 ± 13.9348 ± 15.960.735Female sex24 (27.9)11 (21.6)7 (24.1)0.703Serum Creatinine (mg/dl)78.77 ± 60.2384.4 ± 51.6778.24 ± 24.90.514eGFR(ml/min/1.73 m2)90 ± 39.790.46 ± 61.6295.85 ± 29.240.545White blood cell count(10^9/L)7.27 ± 4.468.08 ± 3.597.93 ± 3.480.259Platelet Count(10^9/L)193 ± 85.54236 ± 90.76195 ± 100.240.026Hemoglobin(g/dl)105 ± 21.18105 ± 21.6299 ± 19.850.167Lactate(mmol/L)2.5 ± 2.742.8 ± 1.832.0 ± 1.50.096Causal organism (+)46 (53.5)35 (68.6)19 (65.5)0.061Vegetation planting location0.136Aortic valve46(53.5)23(45.1)15(51.7)0.631Mitral valve43(50)30(58.8)16(55.2)0.596Vegetation length(mm)7.2 ± 8.848.1 ± 7.78.0 ± 70.700Valve perforation37(43)15(29.4)13(44.8)0.731Echocardiographic findingsLeft ventricular diastolic dysfunction46(53.5)27(52.9)16(55.2)0.962Pulmonary artery pressure28(32.6)11(21.6)7(24.1)0.398Tricuspid regurgitation72(83.7)46(90.2)25(86.2)0.773Mitral regurgitation38(44.2)19(37.3)20(69)0.039Aortic regurgitation27(31.4)16(31.4)7(24.1)0.740Pulmonic regurgitation10(11.6)1(2)3(10.3)0.134Left ventricular ejection fraction(%)64 ± 8.3963 ± 6.7464 ± 8.220.839NYHA III-IV Class71(82.6)42(82.4)24(82.8)0.759Duration between IE diagnosis and hospitalization (days)10 ± 23.54 ± 20.877 ± 21.060.195Clinical manifestations: fever46(53.5)25(49)19(65.5)0.619Duration of symptoms < 1 month before diagnosis37(43)26(51)17(58.6)0.304History of diseaseHeart disease29(33.7)14(27.5)10(34.5)0.621Valvular disease32(37.2)13(25.5)11(37.9)0.161Heart failure6(7)8(15.7)1(3.4)0.161Atrial fibrillation12(14)9(17.6)4(13.8)0.825Neurological disease18(20.9)12(23.5)8(27.6)0.511Abscess, Emboli4(4.7)6(11.8)8(27.6)0.003Chronic lung disease28(32.6)22(43.1)13(44.8)0.619Hepatic disease9(10.5)10(19.6)5(17.2)0.304Renal insufficiency43(50)26(51)12(41.4)0.675Peripheral arterial system diseases6(7)4(7.8)1(3.4)0.841Type 2 diabetes mellitus4(4.7)7(13.7)0(0)0.054Hypertension17(19.8)6(11.8)9(31)0.109Smoker8(9.3)5(9.8)3(10.3)1.000Drink8(9.3)7(13.7)4(13.8)0.667Preoperative treatment time(days)5 ± 1.6810 ± 2.119 ± 7.360.000Antibiotic therapy in hospital66(76.8)47(92.2)29(100)0.001Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Preoperative patient basic information on infective endocarditis Age, mean, years eGFR (ml/min/1.73 m2) Platelet Count (10^9/L) Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Table 2Surgical status of infective endocarditisA(D1-7)No. (%)C(D14-28)P ValueB(D8-13)N = 86N = 51N = 29Operation procedure Mitral valve replacement52(60.5)37(72.5)14(48.3)0.226 Aortic valve replacement49(57)23(45.1)19(65.5)0.204 Tricuspid annuloplasty33(38.4)22(43.1)8(27.6)0.384 Other operations35(40.7)25(49)9(31)0.284 mechanical valve replacement56(65.1)33(64.7)20(69)0.463Operation time(min)300 ± 64.47300 ± 72.3315 ± 54.390.849 Aortic clamping time(min)76 ± 29.1479 ± 35.9674 ± 28.170.832 Extracorporeal circulation time(min)108 ± 39.15110 ± 43.9106 ± 31.070.852Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Surgical status of infective endocarditis Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Table 3In-hospital condition of patients after surgery for infective endocarditisA(D1-7)No. (%)C(D14-28)P ValueB(D8-13)N = 86N = 51N = 29Postoperative intubation periods < 24 h70(81.4)44(86.3)24(82.8)0.696ICU hospitalization time(days)3(2.5–3.5)3(2.75–3.25)3(2.75–3.25)0.997Length of hospitalization(days)10(3–13)9(6.5–11.5)10(8.25–11.75)0.258Embolic event2(2.3)6(11.8)2(6.9)0.071Symptomatic heart failure11(12.8)6(11.8)0(0)0.132Other organ dysfunction8(9.3)4(7.8)1(3.4)0.684Arrhythmia10(11.6)5(9.8)2(6.9)0.762In-hospital mortality5(5.8)3(5.9)0(0)0.562Echocardiographic findingsLeft ventricular ejection fraction (%)58 ± 7.9859 ± 7.6760 ± 9.120.733Tricuspid regurgitation78(91.8)38(77.6)25(86.2)0.629Mitral regurgitation45(52.9)25(51)18(62.1)0.256Aortic regurgitation32(37.6)16(32.7)12(41.4)0.651Pulmonic regurgitation10(11.8)9(18.4)3(10.3)0.506Left ventricular diastolic dysfunction44(51.8)30(61.2)20(69)0.172Aortic transvalvular pressure difference(mmHg)15 ± 10.0811 ± 9.1917 ± 12.420.035NYHA III-IV Class73(85.9)37(75.5)24(82.8)0.674Postoperative death/reoperation within 1 year5(5.8)4(7.8)1(3.4)0.665Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests In-hospital condition of patients after surgery for infective endocarditis Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests In the preoperative data, we identified differences in platelet count (P = 0.026), mitral regurgitation status (P = 0.039), and history of other systemic infarcts (P = 0.003). However, differences in the duration of preoperative treatment < 1 month and in-hospital antibiotic treatment were attributable to subgroup variations. Intraoperative data did not reveal any significant differences. Regarding perioperative data, only the in-hospital aortic transvalvular pressure difference showed a significant difference (P = 0.035). Additionally, no significant differences were observed in in-hospital mortality (P = 0.562) and death/reoperation within 1 year (P = 0.665) (Supplement Fig. 1). Effect of operation time on 1-year mortality All patients were followed up at our hospital several times within 1 year after surgery according to standard guidelines. Follow-up examinations included physical examinations, 12-lead electrocardiograms, and transthoracic echocardiograms. The in-hospital mortality rate was 4.82% (8 patients) and the reoperation rate was 1.2% (2 patients) (Table 4). A univariate analysis was conducted for the death and survival groups, revealing that preoperative blood creatinine (P = 0.015), leukocyte count (P = 0.008), lactate levels (P = 0.007), NYHA classification III-IV (P = 0.04), and heart failure (P = 0.012), as well as intraoperative operative time (P = 0.017), aortic cross-clamp time (P = 0.015), and extracorporeal circulation time (P = 0.001), were significantly associated with mortality. Postoperative factors such as endotracheal intubation for < 24 h (P = 0.001), ICU length of stay (P = 0.000), cardiac dysfunction (P = 0.000), and other organ dysfunction (P = 0.000) were also significantly associated with mortality (Table 5). Subsequent multifactorial analyses, adjusted for confounding variables, indicated that ICU length of stay (P = 0.015), cardiac dysfunction (P = 0.015), and other organ dysfunction (P = 0.007) remained significant predictors of mortality (Table 6). Table 4Cause of death and reoperation within 1 year in the three groups of patientsA(D1-7)B(D8-13)C(D14-28)In-hospital mortalityN = 5N = 3N = 0Septic shock100Low cardiac output symptom320Renal insufficiency110Cerebral hemorrhage/herniation100Pyemia020multiple organ failure110Cumulative results within 1 yearDeathN = 5N = 3N = 0ReoperationN = 1N = 1N = 0Statistics on causes of death and reoperations within 1 year for all three groups, counting only Cause of death and reoperation within 1 year in the three groups of patients Statistics on causes of death and reoperations within 1 year for all three groups, counting only Table 5Univariate analysis of patient mortality factorsDeath N = 8Alive N = 158Logistic P ValueOR for early surgery (95% CI)Serum Creatinine(mg/dl)146.19(70.65–155.66.65.66)80.41(61.48–99.34)0.0151.01(1.00–1.02.00.02)White blood cell count(10^9/L)11.14(5.64–16.65)7.78(5.73–9.84)0.0081.16(1.04–1.30)Lactate(mmol/L)5.84(1.10–15.00)3.03(0.40–12.70)0.0071.29(1.07–1.55)NYHA III-IV Class8(100)129(81.65)0.046.25(1.09–36.01)Heart disease3(37.5)12(7.6)0.0127.3(1.55–34.31)Operation time(min)315(187.5–442.5.5.5)300(262.5–337.5.5.5)0.0171.01(1.01–1.02)Aortic clamping time(min)97.5(56–139)76(55.5–96.5)0.0151.02(1.00–1.04.00.04)Extracorporeal circulation time(min)165.5(96.13–234.88.13.88)107.5(84.38)0.0011.02(1.01–1.04)Postoperative intubation periods < 24 h4(50)134(84.8)0.0012.60(1.44–4.69)ICU hospitalization time(days)9.38(1–25)3.44(2–13)0.0001.56(1.23–1.98)Symptomatic heart failure6(75)11(7)0.00040.09(7.23–222.45.23.45)Other organ dysfunction4(50)9(5.7)0.00016.56(3.55–77.27)NYHA III-IV Class5(100)129(81.6)0.00714.38(2.06–100.38.06.38)Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Univariate analysis of patient mortality factors Serum Creatinine (mg/dl) 146.19 (70.65–155.66.65.66) 80.41 (61.48–99.34) White blood cell count (10^9/L) 11.14 (5.64–16.65) 7.78 (5.73–9.84) Lactate (mmol/L) 5.84 (1.10–15.00) 3.03 (0.40–12.70) 315 (187.5–442.5.5.5) 300 (262.5–337.5.5.5) 97.5 (56–139) 76 (55.5–96.5) 165.5 (96.13–234.88.13.88) 107.5 (84.38) 9.38 (1–25) 3.44 (2–13) Continuous normally distributed variables were reported as mean ± standard deviation. Continuous non-normally distributed variables were reported as median with the interquartile range in parentheses, or mean with the range of values in parentheses. Categorical variables were reported as numbers and percentages. Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Table 6Multifactorial analysis of patient mortality factorsLogistic P ValueOR for early surgery (95% CI)ICU hospitalization time0.0151.30(1.05–1.61)Symptomatic heart failure0.01522.28(1.82–273.04.82.04)Other organ dysfunction0.00729.21(2.46–346.49.46.49)Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Multifactorial analysis of patient mortality factors Differences between groups were compared using the chi-square (χ²) test or Fisher’s exact test. Continuous variables were compared using the unpaired t-test, and the Mann-Whitney U test or Kruskal-Wallis test was used in other cases. A two-sided p-value of 0.05 was considered the level of statistical significance for all tests Discussion The ACC/AHA recently updated its recommendations for the surgical treatment of IE in the 2020 ACC/AHA Guidelines for the Management of Valvular Heart Disease, following earlier revisions in 2017 [3]. However, the surgical recommendations for IE have remained unchanged when compared with the 2014 ACC/AHA guidelines. The role of surgery in managing IE continues to evolve, with current guidelines advocating surgical intervention for complicated left-sided infective endocarditis. Nevertheless, the optimal timing for surgical intervention remains uncertain, and the decision between completing an adequate course of antibiotic therapy and opting for early surgical intervention varies significantly across surgical centers. Hill et al. observed a four-fold increase in mortality among patients who underwent surgery within 7 days of IE diagnosis; however, this increased mortality was attributed to the severity of the IE rather than the timing of the surgery [10]. Embolism and heart failure are major contributors to mortality in IE, and the potential mortality benefit of early surgical intervention may stem from mitigating these complications [11]. Recent studies have demonstrated that early surgical intervention is associated with lower mortality compared to surgery performed 8–20 days after diagnosis. Specifically, surgical intervention within 7 days of initiating antibiotic treatment is linked to lower mortality compared to surgery performed 20 days after the start of antibiotic therapy [12]. Consequently, there is a trend towards favoring early surgical intervention in clinical practice. To assess the relationship between the timing of surgery and outcomes in IE, it is crucial to accurately determine the date of both diagnosis and surgery. While recording the date of surgery in observational studies is straightforward, accurately capturing the dates of hospitalization, definitive IE diagnosis, and the onset of complications that indicate the need for surgery can be challenging. Similarly, while the modified Duke criteria can confirm a diagnosis of definitive IE, the timing of surgical indication may be delayed by several days until complications such as heart failure symptoms, embolic events, persistent bacteremia, or intracardiac abscesses arise. This time interval between IE diagnosis and the indication for surgery can be further complicated if the patient was initially diagnosed and treated at another hospital before being transferred to a surgical center. Therefore, we conducted a detailed review of the data on symptom onset, initial medical consultation, first hospital treatment, and referral to our center for each patient with IE to comprehensively assess the disease course. Avoidance of all-cause mortality, disabling stroke, or recurrence of IE at 1 year is a more critical outcome measure for IE surgery compared to in-hospital mortality alone. Therefore, we collected and analyzed detailed data from each group of IE patients. The results indicated no significant differences in composite primary endpoints, such as mortality, incidence of postoperative complications, postoperative cardiac function, and organ dysfunction among the three groups. Additionally, there were no significant differences in postoperative endotracheal intubation duration, ICU length of stay, and total postoperative hospital days. These improvements in in-hospital clinical outcomes did not correspond to an increased risk of surgical mortality or recurrence of IE at 1 year. Studies have shown that [4] pathogenic microorganisms can be resistant to microbicidal peptides or antimicrobial agents released locally by platelets. The hypercoagulable state associated with nonbacterial thrombotic endocarditis lesions promotes fibrinogen deposition, platelet aggregation, and microbial proliferation, which contributes to infectious redundancy. Thus, while platelets may reflect the patient’s physical status to some extent, they are not likely a significant factor in the primary endpoint of the mortality composite. Most cases of infective endocarditis occur in patients with pre-existing heart disease and typically involve the heart valves, with the aortic and mitral valves being the most commonly affected, while the tricuspid and pulmonary valves are less frequently involved. In patients with a longer disease duration, there is often a comorbidity with rheumatic heart disease, leading to differences in mitral regurgitation between groups. This suggests that the management of IE in patients with underlying cardiac disease may be more complex and prolonged. The results also revealed that approximately one-fifth of patients had central nervous system involvement, and systemic embolism occurred in about 10% of patients, including pulmonary, hepatic, splenic, and renal mycotic infarcts, with splenic infarcts accounting for 61% of these cases. Splenic dysfunction is generally less impactful on overall health compared to dysfunction of other organs and thus is not a significant factor in the mortality composite endpoint. It is noteworthy that the differences observed in preoperative treatment duration and in-hospital antibiotic treatment are largely attributable to subgroup variations, with extended preoperative hospital stays often resulting from prolonged antibiotic therapy and the complexity of the condition. Subsequently, we analyzed the risk factors associated with postoperative mortality by entering all variables into a logistic regression model to identify significant differences between survivors and deceased patients. The analysis revealed that ICU length of stay, postoperative cardiac malfunction, and other organ dysfunction were significant risk factors for composite endpoints such as death. ICU length of stay is known to correlate with disease severity, and postoperative organ dysfunction increases the need for advanced life support. The severity of heart failure has been shown to elevate mortality rates in patients with IE [13]. Hill et al. suggested that early surgical intervention within 7 days might reduce the incidence of heart failure [10]. Overall, the primary mortality risk factor for patients remains the state of organ dysfunction. Our 1-year follow-up results corroborate this, with organ dysfunction accounting for 69.23% of death causes. We found that the in-hospital and 1-year mortality rates were significantly lower in all three groups compared to previously reported rates [11]. Several factors may explain the lower mortality observed in our study. First, the proportion of patients with poor prognostic factors, such as moderate-to-severe congestive heart failure and staphylococcal infection (12.2% in this study), was lower than in previous studies [14]. Second, 85.5% received effective antibiotics tailored to in-hospital bacterial sensitivity. This suggests that the correct decision-making by the multidisciplinary endocarditis team and the appropriate timing of interventions may be associated with the observed low mortality rates. Third, blood cultures and echocardiography were conducted within 24 h of hospitalization for all patients with suspected infective endocarditis. The prompt diagnosis based on established guidelines was likely associated with favorable outcomes. Finally, the patient cohort was relatively young, with a mean age close to 50 years, and the rate of reoperation and other mortality-increasing risk factors was low. However, our study has several limitations. First, this study is a retrospective data collection, which limits the design and may introduce biases. Second, the relatively small sample size prevented more detailed stratified analyses, such as propensity score matching, potentially leading to biases related to certain factors. The timing of surgical interventions after hospitalization could have influenced the study results, with patients in poorer clinical conditions possibly experiencing delays or opting against surgery. Nonetheless, all variables were rigorously reviewed to ensure high data quality. Third, preoperative treatment timing, influenced by patient symptoms, may affect prognosis. Overall, retrospective observational studies, including this one, are susceptible to survivor bias. Early surgical intervention did not increase the risk of death in patients with IE compared to those undergoing late or elective surgery, which may be attributable to the lower incidence of adverse clinical features in the early surgery group. Importantly, 1-year mortality rates were similar across all three groups and were lower than the overall patient mortality rates reported in recent literature. These findings suggest that early cardiac surgery for active infective endocarditis is not contraindicated and can be performed as soon as it is indicated, offering optimal postoperative benefits to patients. In conclusion, our team believes that establishing a strong standardised endocarditis team is essential to reduce mortality and improve patient prognosis. Early surgical intervention does not increase the risk of death in patients with IE, nor does it increase the risk of recurrence and reoperation within one year. Therefore, in patients with a clear diagnosis, unstable cardiac function, or risk of recurrent embolism, we recommend early surgical intervention before completing a cycle of antibiotic therapy. Supplementary Information Supplementary Material 1 Author contributions Wu Liu: Writing – original draft, Resources, Methodology, Investigation, Data curation.Yongqin Li: Writing – review & editing, Writing – original draft, Heng Yang: Resources, Methodology, Data curation, Conceptualization.Congcong Li: Writing – review & editing, Resources, Methodology.Jinyu Wang: Writing – original draft, Methodology, Investigation.Wanqi Lan: Writing – original draft, Software, Resources.Yanhua Tang: Writing – review & editing, Writing – original draft, Resources, Methodology, Formal analysis, Datacuration. Funding Key Research and Development Program of Jiangxi Province, 20223BBG71010. The Second Affiliated Hospital Of NanChang University Funding Program (No. 2021efyA02) Data availability Data will be made available on request. Declarations Ethics approval and consent to participate These experiments were conducted according to established ethical guidelines, and informed consent obtained from the participants. Consent for publication All authors are aware of the content of this article, guarantee originality and authenticity, and agree to publish. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Wu Liu and Yongqin Li co-first author. Supplementary Information The online version contains supplementary material available at 10.1186/s13019-025-03766-3. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement Data will be made available on request.