The number of endocarditis episodes diagnosed worldwide in one year could be as high as 250,000 cases [1], and the mortality rate within one year exceeds 30%. Especially when large vegetations form, often with multiple severe complications, the situation is often life-threatening, requiring early diagnosis and timely intervention.

Vegetations of infective endocarditis can form on the native or prosthetic heart valve. Vegetations $>$10 mm are often treated, making very large vegetations rare in clinical practice and in the literature [2]. The clinical and imaging features of infective endocarditis (IE) with large vegetations are different from those of typical IE [3]. Very large heart vegetations are prone to misdiagnosis as thrombus or tumor, causing clinical difficulties in the diagnosis. It is very important to supplement the clinical characteristics of IE patients with very large vegetations [4]. Alarmingly, the high bacterial density and fragility of massive vegetations usually lead to severe IE complications, valve destruction, heart failure, vessel embolism, sepsis, and immune phenomena [5]. Furthermore, the size of vegetations is associated with in-hospital mortality [6]. However, the guidelines issued by the American College of Cardiology (ACC) and the European Society of Cardiology (ESC) have uncertainty and variation among surgical indications, risks, and timing for $>$30 mm infective valvular vegetation, and these recommendations are limited by the low level of evidence from primary observational studies [7]. In summary, the experience in diagnosis, treatment and management of $>$30 mm vegetation has been largely lacking [8]. Therefore, this study retrospectively analyzed the detailed clinical data and prognosis of 22 patients with IE complicated with $>$30 mm vegetation in our hospital from January 2016 to February 2022. We also retrieved and summarized all IE case reports with $>$30 mm vegetations published between 2020 and 2022. The aims are to improve clinicians’ understanding of IE with very large vegetation and to provide evidence for clinical practice.

Diagnosis and management of IE remains challenging because of its clinical diversity and changing epidemiology [11, 12]. The disease pattern and prognosis of IE vary greatly due to pathogenic microorganisms, basic heart disease, implantation of prosthetic valves and cardiac devices, etc. Therefore, even risk factor classification in international guidelines is inconsistent with clinical practice [13, 14, 15]. In addition, there are only infrequent occurrence and sparse reports of IE with very large vegetations, so the original clinical evaluation can confound its diagnosis. Patients with atypical symptoms visit multiple departments. Less-experienced physicians may consider their diagnosis as chronic infection, rheumatic disease, neurological disease, autoimmune disease, malignant tumor, etc. [16, 17, 18]. Although TTE is a mainstay in the diagnostic toolkit for IE [19], sometimes it is difficult to differentiate one mass from another. Our findings suggested that, in IE patients with atypical clinical presentation, the mass in the left atrium is readily misdiagnosed as myxoma. In this study, Case 1 showed that the pedunculated vegetation seemed to be attached to the mitral valve annulus (Fig. 4). A similarly confusable state also existed in Case 2 with the vegetation located in the posterior mitral valve and in Case 3 with the pedicle of vegetation attached to the junction of the anterior and posterior mitral valves. These were originally considered as left atrial myxoma and were planned to undergo tumor enucleation. During the operation, the surgical plan was changed to vegetectomy and mitral valve replacement. Finally, the postoperative biopsy confirmed the diagnosis of IE. A mass in the right atrium might be wrongly diagnosed as thrombosis. The TTE of Case 16 showed tricuspid septal fixed irregular hypoechoic mass, and the TTE of Case 15 showed fixed fuzzy echoes of the right atrium, surrounded by electrodes incarcerated in the tricuspid valve orifice (Fig. 4). The TTE diagnosis of Case 15 and Case 16 were high possibility of thrombosis, but postoperative pathology confirmed IE (Fig. 5). The blood test results of 22 patients in this study are consistent with the authoritative review [20], which points out that an IE diagnosis may be suspected due to elevated CRP, ESR, anemia, and microscopic hematuria, etc. However, the above indicators lack specificity and have not been included in modern clinical diagnostic criteria. The negative blood culture rate of 31.8% (7/22) in our IE cases with $>$30 mm vegetation seems to be higher than previous reports of those 30 mm [2, 21, 22, 23]. This may be due to previous antibiotic treatment, microbes embedded deep within the giant vegetation and not released into the blood, and the relatively small study sample size. We prefer initial treatment with AMP and CRO [24], and CRO combined with TZP, and VAN combined with CRO are reasonable empirical treatments [25, 26, 27].

Fig. 4.

TTE of IE with very large vegetations are susceptible to misdiagnosis. Arrow: vegetations. (A1) The pedunculated vegetation in Case 1 protruded into the mitral valve orifice during diastole, (A2) and returns to the left atrium during systole. (B1) In Case 15, TTE showed fixed fuzzy echoes of the right atrium near the tricuspid valve orifice, surrounded by electrodes, (B2) and incarcerated in the tricuspid valve orifice.

Fig. 5.

Tricuspid valve vegetations in IE. (A) The gross appearance of a large vegetation on the tricuspid valve in Case 16, as measured in centimeters. (B) Hematoxylin and eosin staining of a microscopic cross section of a vegetation. White arrow: bacteria embedded within the vegetation. Black arrow: inflammatory cells.

In conclusion, in terms of improving early diagnosis, our study suggests that IE accompanied by giant vegetation has a wide variety of first appearances, diverse imaging features, dangerous clinical condition, and a high negative blood culture rate. Clinicians should be more vigilant about IE when a large space-occupying mass is found by TTE. High suspicion of IE in the early stage of clinical admission is appropriate for patients with the following manifestations. (a) With underlying structural heart disease or with intracardiac implants, (b) with vascular manifestations such as arterial embolism, pulmonary embolism, and intracranial hemorrhage, (c) with consumptive disease manifestations such as anemia and hypoalbuminemia, (d) with unexplained manifestations of acute heart failure, (e) with fever and arterial embolism. Patients with unexplained spleen, brain, or kidney infarction as the first symptom should also be alerted to the possibility of IE.

Compared with the 10–20 mm vegetation group, the diagnosis of IE with very large vegetation is more difficult as it is often mixed with basic heart diseases or implantations. Compared with TTE, transesophageal echocardiography (TEE) can more accurately assess the hemodynamic effects of location, attachment site, size and shape of vegetation and valve damage (Fig. 6A1,A2,B), especially in the case of prosthetic valve endocarditis (PVE) and perivalvular abscess. However, cardiac computed tomography/computed tomography angiography (CT/CTA) is more suggestive in detecting perivalvular complications (abscess/pseudoaneurysm) in native valve endocarditis (NVE) and PVE, and non-cardiac abnormalities can be found in a single examination [28]. Case 6 was admitted to hospital due to abdominal pain for 9 days, with no fever and negative blood culture. TTE showed echo mass in the left atriun, and aortic CTA showed mesenteric artery pseudoaneurysm (Fig. 6D). Thrombosis and abscess were seen during aneurysmectomy, and MVR was performed after oral antibiotics for 2 weeks. Mitral valve biopsy was consistent with IE, so CTA was of great significance for prompting diagnosis of IE with atypical clinical features. Case 16 was admitted with negative blood cultures and a paravalvular aortic abscess detected by cardiac CT (Fig. 6D). CT can also be used to differentiate between benign and malignant tumors (Fig. 6E). Magnetic resonance imaging (MRI), on the other hand, can better differentiate tissue components (solid, liquid, hemorrhage, fat, and thrombus), and is especially useful in differentiating giant vegetations from thrombus (Fig. 6F) or myxomas (Fig. 6G). Myocardial perfusion ultrasonography is also useful in excluding thrombus (Fig. 6H). Because there are no lead artifacts in positron emission tomography (PET), its diagnostic ability for CDRIE is higher than CT. Silbiger and his coworkers [29] reported that PET reclassifies 90% of Duke-possible patients with suspected device infections, and PET was also superior in finding primary and extracardiac sepsis [30]. Doctors should be aware of the possibilities offered by the multimodal imaging approach when appropriate.

Fig. 6.

Multimodality imaging in IE. (A) TEE showed a very large mitral valve vegetation in the left atrium. Yellow arrows: vegetations. (A1) TEE of Case 2 showed a pedicle connection to the mitral valve rather than the atria, which was confirmed to be IE by biopsy. (A2) Left atrial vegetation of Case 6. (B) TEE demonstrates pulmonary valve vegetations in a Case of patent ductus arteriosus. (C) Aortic CTA shows superior mesenteric artery aneurysm. Red arrow: rupture of pseudoaneurysm in Case 6. (D) Cardiac CT revealed a perivalvular aortic abscess (blue arrow) of Case 16. (E) CT showing cardiac metastases (primary gastric inflammatory myofibroblastic tumor). Yellow arrow: uneven enhancement. (F) MRI shows abnormal signal foci in the right atrium, with no obvious enhancement after enhancement. Blue arrow: thrombus in the right atrium. (G) Most of the lesions did not enhance during delayed MRI enhancement. Red arrow: Myxoma. (H) Myocardial perfusion ultrasound showed two moderate echoic masses in left ventricle, and there was no contrast agent filling in the irregular masses. Blue arrow: thrombus.

In addition to the emphasis on improving early diagnosis, our study also supports early surgical intervention in IE with $>$30 mm vegetations. Previous studies have shown that more than 80% of deaths from IE were associated with embolic events [31, 32]. In patients with giant vegetations included in the study, the proportions of previous and new embolism (within one week of antibiotic treatment) were significantly increased, which was consistent with the conclusion of many studies that the size of vegetations was related to the occurrence of embolism [33, 34]. The size of vegetation has been used to create an “embolism risk calculator” [35]. In a meta-analysis of 21 studies, patients with vegetation size greater than 10 mm had increased odds of embolic events (OR [Odds Ratio], 2.28), but odds were greater with a cutoff of 15 mm (OR, 4.25) [31, 36]. However, due to the rarity of giant vegetations, a 10 mm size of vegetations was considered to be the best threshold for estimating the risk of embolism in previous studies. There have been no reports on the relative risk of embolism for vegetations $>$30 mm or providing outcome data for further treatment until now.

There are few reported cases of giant IE vegetations worldwide. Radcliffe et al. [2] summarized all twenty-three IE cases with $>$40 mm vegetations reported before 2020, and we retrieved seven published IE cases with $>$30 mm vegetations published after 2020 (Table 6, Ref. [37, 38, 39, 40, 41, 42, 43]). In these 30 published cases, the vegetation was most commonly located in the tricuspid valve. By contrast, in the 22 patients we included, mitral valve vegetations were predominant, which may be attributable to publication bias and small sample size. Surgical intervention was performed in 69.9% (16/23) cases reported before 2020, with an overall mortality rate of 43%. Of the 22 cases we included and the seven cases of giant vegetations published after 2020 that we retrieved, 96.6% (28/29) underwent surgery, there were no cases of in-hospital mortality, and tricuspid regurgitation worsened postoperatively in only one case. These results seem to suggest the important role of early surgery for giant vegetations. Research has confirmed that antibiotic treatment alone was unsatisfactory for very large IE vegetations. IE vegetations $>$10 mm in TTE were associated with significantly lower response rates to specific medical therapy, and 65% of embolisms occurred within two weeks of antibiotic therapy [44, 45]. So despite antibiotic therapy, an increase in vegetation size was not unusual, predicting subsequent embolism [46]. However, the long-standing controversy over the surgical indications and timing in IE with $>$30 mm vegetations has not been resolved, which is reflected in the difference between the two authoritative guidelines [16]. ESC recommends urgent surgery for vegetations $>$30 mm to prevent embolism, but the timing is stipulated within a few days after diagnosis, and the evidence grade is IIa B. The ACC/AHA Guideline uses 10 mm as the dividing line, and does not specifically mention extremely large vegetation cases, with the suggested operation time prior to stopping antibiotics [47]. In fact, surgical indications are narrowed and surgical timing is delayed in clinical practice, as very large vegetations without comorbidities are rarely used as surgical indications [48]. Surgical treatment of IE is often performed after complications have occurred, rather than prophylactically to reduce the risk of future complications [16, 17, 49]. Meanwhile, serious complications are overanxiously considered as surgery contraindications, such that antibiotic therapy under close monitoring in the presence of cardiac function class III or IV is recommended. On the one hand, early surgery is crucial to reduce the risk of additional complications or clinical deterioration [50], and on the other hand, surgical urgency is strongly associated with higher in-hospital mortality [7, 51, 52]. However, more and more recent research supports early surgery (within one week after diagnosis) for IE patients with complex complications [53, 54, 55], and indicates aggressive surgical treatment is associated with better survival (HR = 0.58) and reduced 6-month mortality (11% for early surgery vs. 33% for delayed surgery, OR = 0.18) [56, 57]. A study assigned 76 IE patients with $>$10 mm vegetations and severe valve dysfunction to either early surgery within 48 hours or antibiotic therapy, concluding that early surgery prevented any additional embolic events and did not increase mortality [58].

In our study, patients with $>$30 mm vegetation almost all underwent surgery before the end of a course of antibiotics. On the day of admission, we completed the medical history, physical examination, TTE, TEE and blood culture, evaluated and confirmed within 12 hours. Once we found large vegetations or suspected IE, we tried to improve cardiac and even whole-body imaging with CT/CTA, MRI, and so on. We discussed and addressed with a multidisciplinary team, and started anti-infective treatment promptly. Patients with $>$15 mm vegetations, heart failure, and paravalvular abscess need surgical intervention immediately or within 48 hours (Grade 1a, Grade B). Cardiac glycoside and diuretics were given before surgery to improve cardiac function. All patients underwent surgery on cardiopulmonary bypass. The surgeon completely removed vegetations and infected tissues, repaired or replaced damaged valves, and corrected intracardiac deformities. The standard operation technique was performed according to 2016 American Association for Thoracic Surgery (AATS) guidelines [59]. Surgical specimens were sent for pathogen culture and biopsy. Cardiac and renal function support and pharmacological anticoagulation were given after the operation. TTE was reviewed to evaluate valve function and complications. Antibiotics were administered for at least 4 weeks postoperatively. If the culture results of pathogens were negative, broad-spectrum antibiotics (VAN and aminoglycosides) were preferred.

In terms of the perioperative period, surgical timing, surgical method, details and prognosis, we summarized some experiences of IE with $>$30 mm vegetation that is different from normal-sized vegetations (Table 7).

(1) In terms of preoperative examination, although routine brain imaging screening is reasonable in patients with left-sided IE (IIa B), the 2016 AATS consensus guidelines emphasizes the necessity of preoperative cranial examination in IE with very large vegetations. Because $>$30 mm vegetations often involve higher misdiagnosis rate and more serious infection, multimodality imaging is more emphasized to confirm the diagnosis and to find other infectious lesions or abscesses [60]. Coronary arteriography (CAG) in $>$30 mm aortic vegetations or aortic abscesses theoretically increases the risk of embolism, so CTA should be used instead of CAG (1 C).

(2) Because patients with $>$30 mm vegetations often have poor cardiac function, preoperative cardiotonic and diuretic treatments are more commonly used. However, caution should be taken with regard to the complex basic cardiac conditions in these patients, such as HCM and atrioventricular block.

(3) As for the timing of surgery, the decision of a multidisciplinary team is particularly significant in patients with $>$30 mm vegetations. A $>$30 mm vegetation with imminent embolism is an indication for surgery within 48 hours (IIa B), while normal-sized vegetations with new embolism despite reasonable antibiotic treatment or with severe valve disease should be performed with sub-emergency operation (IIa B). But in patients with $>$30 mm vegetations, if there is a strong indication for early surgery (hemodynamic instability or high risk of new embolic stroke), early surgery may be considered after careful evaluation of the size of the bleeding lesion by the multidisciplinary team [61, 62]. The risk of postoperative neurological deterioration due to ICH appears to be relatively low, even in IE patients who undergo valve surgery within 2 weeks after ICH [63], but very early surgery (within 7 days) should be avoided [64].

(4) The choice of surgical incision and the surgical difficulty are different between the 10–20 mm group and the $>$30 mm group. Compared with one case of minimally invasive MVR in the 10–20 mm group, median sternotomy was used in all patients with $>$30 mm vegetations. Our research suggests that aggressive S. aureus, CIED and PVE are more frequent in patients with $>$30 mm vegetations. Median sternotomy is beneficial for radical debridement and for more difficult operations, such as Bentall if the aortic root is badly damaged [65].

(5) According to 2016 AATS Guidline, when valvular stenosis and regurgitation occur, MVP is the optimal choice instead of MVR, while AVR is better than aortic valve plasty (AVP). Generally speaking, TVP is recommended. But in the patients with $>$30 mm vegetation, TVR in more common than in the 10–20 mm group (4 cases vs. 1 cases), and a biological valve is mostly used. This may be because the $>$30 mm vegetations group had more faulty TV, ICH, or severe stroke, so that it was reasonable to avoid mechanical prostheses, and the use of bioprosthetic valves could avoid postoperative anticoagulation required to reduce other hemorrhagic complications. The selection of valve type should consider the surgical difficulty that the biological valve may need to be completely replaced in case of IE relapse [66].

(6) According to the 2015 ESC IE Guidelines [67], as predictors of poor prognosis in admission assessment, compared patients with 10–20 mm vegetations, those with $>$30 mm vegetations had more complex clinical situations (such as pulmonary edema, pulmonary embolism, poor pulmonary function, cerebral infarction, acute heart failure, shock, and more diverse pathogens). There are many basic conditions in the heart in the $>$30 mm vegetations group (such as cardiomyopathy, pulmonary hypertension, history of heart surgery, prosthesis valves, intracardiac devices and low ejection fraction). The collaborative management of a multidisciplinary team and personalized treatment plan are critical to the outcome of IE patients with $>$30 mm vegetations. Our study supports that early surgery for IE patients with very large vegetations can reduce in-hospital and 6-month mortality, but we need better risk stratification in view of more postoperative cardiac tamponade, DIC, and ICH in patients than in the normal-sized vegetations group [58, 68, 69].

For IE complicated with $>$30 mm vegetations, clinical characteristics are diverse and vegetations on TTE are prone to misdiagnosis as thrombus or tumors. This article emphasizes the use of $>$30 mm IE vegetations as an independent indication for early surgery to improve prognosis.

AATS, American Association for Thoracic Surgery; APTT, activated partial thromboplastin time; AMP, ampicillin; AVP, aortic valve plasty; CAG, coronary arteriography; CRE, carbapenem-resistant Enterobacter; CDRIE, cardiac device-related infective endocarditis; CRTD, cardiac resynchronization-defibrillator device; cTn I, cardial troponin I; CKD5, chronic kidney disease stage 5; CHD, congenital heart disease; CRP, c-reactive protein; CT/CTA, computed tomography/computed tomography angiography; CK, creatine kinase; CK-MB, creatine kinase MB; DCM, dilated cardiomyopathy; DBIL, direct bilirubin; DIC, disseminated intravascular coagulation; DAA, double aortic arch; ESR, erythrocyte sedimentation rate; EuroSCORE II, european System for Cardiac Operative Risk Evaluation II; FDP, fibrinogen degradation product; GAS, group A streptococci; GBS, group B streptococci; HCM, hypertrophic cardiomyopathy; IE, infective endocarditis; ICU-LOS, intensive care unit length of stay; ICH, intracerebral hemorrhage; LDH, lactate dehydrogenase; LOS, length of stay; LY%, lymphocytes percentage; MRI, magnetic resonance imaging; MRSA, methicillin-resistant staphylococcus aureus; MSSA, methicillin-sensitive staphylococcus aureus; NE%, neutrophils percentage; NT-proBNP, N-terminal pro-brain natriuretic peptide; NVE, native valve endocarditis; TEE, transesophageal echocardiography; TZP, piperacillin-Tazobactam; PCT, procalcitonin; PET, positron emission tomography; PT, prothrombin time; PVE, prosthetic valve endocarditis; PE, pulmonary embolism; SP, severe pneumonia; Q1–Q3, the 25th–75th percentiles; ACC, the American College of Cardiology; ESC, the European Society of Cardiology; TBIL, total bilirubin; TTE, transthoracic echocardiography; VAN, vancomycin; VVI, ventricular demand pacemaker; WBC, white blood cells.

FZ and JM designed the research study. XYC performed the research. YQC provided help and advice on data collection. FZ analyzed the data. XYC, JM and YQC wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

This study involved human data only. The name of the ethics committee: medical ethics committee of Xiangya Hospital of Central South University. The reference number: 202203076.

Thanks to all the peer reviewers for their opinions and suggestions.

This study was supported by the National Natural Science Foundation of China (No. 81873585, No. 82090020, No. 82090024, No. 82073918 and No. 82070070) and Natural Science Foundation of Hunan Province (No.2020SK2088, 2018JJ3835).

The authors declare no conflict of interest.