Using machine learning to improve the diagnostic accuracy of the modified Duke/ESC 2015 criteria in patients with suspected prosthetic valve endocarditis - a proof of concept study.

INTRODUCTION: Prosthetic valve endocarditis (PVE) is a serious complication of prosthetic valve implantation, with an estimated yearly incidence of at least 0.4-1.0%. The Duke criteria and subsequent modifications have been developed as a diagnostic framework for infective endocarditis (IE) in clinical studies. However, their sensitivity and specificity are limited, especially for PVE. Furthermore, their most recent versions (ESC2015 and ESC2023) include advanced imaging modalities, e.g., cardiac CTA and [18F]FDG PET/CT as major criteria. However, despite these significant changes, the weighing system using major and minor criteria has remained unchanged. This may have introduced bias to the diagnostic set of criteria. Here, we aimed to evaluate and improve the predictive value of the modified Duke/ESC 2015 (MDE2015) criteria by using machine learning algorithms. METHODS: In this proof-of-concept study, we used data of a well-defined retrospective multicentre cohort of 160 patients evaluated for suspected PVE. Four machine learning algorithms were compared to the prediction of the diagnosis according to the MDE2015 criteria: Lasso logistic regression, decision tree with gradient boosting (XGBoost), decision tree without gradient boosting, and a model combining predictions of these (ensemble learning). All models used the same features that also constitute the MDE2015 criteria. The final diagnosis of PVE, based on endocarditis team consensus using all available clinical information, including surgical findings whenever performed, and with at least 1 year follow up, was used as the composite gold standard. RESULTS: The diagnostic performance of the MDE2015 criteria varied depending on how the category of 'possible' PVE cases were handled. Considering these cases as positive for PVE, sensitivity and specificity were 0.96 and 0.60, respectively. Whereas treating these cases as negative, sensitivity and specificity were 0.74 and 0.98, respectively. Combining the approaches of considering possible endocarditis as positive and as negative for ROC-analysis resulted in an excellent AUC of 0.917. For the machine learning models, the sensitivity and specificity were as follows: logistic regression, 0.92 and 0.85; XGBoost, 0.90 and 0.85; decision trees, 0.88 and 0.86; and ensemble learning, 0.91 and 0.85, respectively. The resulting AUCs were, in the same order: 0.938, 0.937, 0.930, and 0.941, respectively. DISCUSSION: In this proof-of-concept study, machine learning algorithms achieved improved diagnostic performance compared to the major/minor weighing system as used in the MDE2015 criteria. Moreover, these models provide quantifiable certainty levels of the diagnosis, potentially enhancing interpretability for clinicians. Additionally, they allow for easy incorporation of new and/or refined criteria, such as the individual weight of advanced imaging modalities such as CTA or [18F]FDG PET/CT. These promising preliminary findings warrant further studies for validation, ideally in a prospective cohort encompassing the full spectrum of patients with suspected IE.

The impact of the multidisciplinary Endocarditis Team on the management of infective endocarditis.

BACKGROUND: In their latest guidelines for infective endocarditis (IE) (2015), the European Society of Cardiology (ESC) introduced the implementation of the Endocarditis Team (ET) to facilitate the management of IE. This study presents our experiences and the diagnostic and therapeutic impact of the ET on the management of IE. METHODS: From 2016-2020, data of all patients with suspected IE referred to the ET were prospectively collected. The final diagnosis was defined by the ET as either rejected, possible or definite IE. Diagnostic impact was scored as any change in initial diagnosis, the frequency of additional diagnostic tests advised by the ET and any change in diagnosis after these tests. Therapeutic impact was scored as any change in antibiotic therapy or change from conservative to invasive therapy or vice versa. RESULTS: A total of 321 patients (median age 67 [55-77] years, 71% male) were enrolled. The final diagnosis was rejected IE in 47 (15%), possible IE in 34 (11%) and definite IE in 240 (75%) patients. A change of initial diagnosis was seen in 53/321(17%) patients. Additional microbiological tests were advised in 69/321 (21%) patients, and additional imaging tests in 136/321 (42%) patients, which resulted in subsequent change in diagnosis in 23/321 (7%) patients. Any change in antibiotic treatment was advised in 135/321 (42%) patients, and change from initial conservative to additional surgical treatment in 15/321 (5%) patients. CONCLUSION: The ET had a clear impact on the therapeutic policy for patients with suspected IE and is useful in the management of this life-threatening disease. Broad implementation is warranted.

Implementation of the 2015 European Society of Cardiology guidelines for the management of infective endocarditis in the Netherlands.

Because the occurrence of infective endocarditis (IE) continues to be associated with high mortality, a working group was created by the Dutch Society of Cardiology to examine how the most recent European Society of Cardiology (ESC) guidelines for IE management could be implemented most effectively in the Netherlands. In order to investigate current Dutch IE practices, the working group conducted a country-wide survey. Based on the results obtained, it was concluded that most ESC recommendations could be endorsed, albeit with some adjustments. For instance, the suggested pre-operative screening and treatment of nasal carriers of Staphylococcus aureus as formulated in the ESC guideline was found to be dissimilar to current Dutch practice, and was therefore made less restrictive. The recently adapted ESC diagnostic criteria for IE were endorsed, while the practical employment of the relevant diagnostic techniques was simplified in an adapted flowchart. In addition, the presence of a multidisciplinary, so-called 'endocarditis team' in tertiary centres was proposed as a quality indicator. An adapted flowchart specifically tailored to Dutch practice for microbiological diagnostic purposes was constructed. Lastly, the working group recommended the Stichting Werkgroep Antibioticabeleid (SWAB; Dutch Working Party on Antibiotic Policy) guidelines for IE treatment instead of the antibiotic regimens proposed by the ESC.

Infective endocarditis in the Netherlands: current epidemiological profile and mortality : An analysis based on partial ESC EORP collected data.

INTRODUCTION: Infective endocarditis (IE) is associated with a high in-hospital and long term mortality. Although progress has been made in diagnostic approach and management of IE, morbidity and mortality of IE remain high. In the latest European guidelines, the importance of the multi-modality imaging in diagnosis and follow up of IE is emphasized. AIM: The aim was to provide information regarding mortality and adverse events of IE, to determine IE characteristics and to assess current use of imaging in the diagnostic workup of IE. METHODS: This is a prospective observational cohort study. We used data from the EURO-ENDO registry. Seven hospitals in the Netherlands have participated and included patients with IE between April 2016 and April 2018. RESULTS: A total of 139 IE patients were included. Prosthetic valve endocarditis constituted 32.4% of the cases, cardiac device related IE 7.2% and aortic root prosthesis IE 3.6%. In-hospital mortality was 14.4% (20 patients) and one-year mortality was 21.6% (30 patients). The incidence of embolic events under treatment was 16.5%, while congestive heart failure or cardiogenic shock occurred in 15.1% of the patients. Transthoracic and transoesophageal echocardiography were performed most frequently (97.8%; 81.3%) and within 3 days after IE suspicion, followed by 18F­fluorodeoxyglucose positron emission tomography/computed tomography (45.3%) within 6 days and multi-slice computed tomography (42.4%) within 7 days. CONCLUSION: We observed a high percentage of prosthetic valve endocarditis, rapid and extensive use of imaging and a relatively low in-hospital and one-year mortality of IE in the Netherlands. Limitations include possible selection bias.

Incidence and outcome of prosthetic valve endocarditis after transcatheter aortic valve replacement in the Netherlands.

BACKGROUND: Transcatheter aortic valve replacement (TAVR) is increasingly being used as an alternative to conventional surgical valve replacement. Prosthetic valve endocarditis (PVE) is a rare but feared complication after TAVR, with reported first-year incidences varying from 0.57 to 3.1%. This study was performed to gain insight into the incidence and outcome of PVE after TAVR in the Netherlands. METHODS: A multicentre retrospective registry study was performed. All patients who underwent TAVR in the period 2010-2017 were screened for the diagnosis of infective endocarditis in the insurance database and checked for the presence of PVE before analysis of general characteristics, PVE parameters and outcome. RESULTS: A total of 3968 patients who underwent TAVR were screened for PVE. During a median follow-up of 33.5 months (interquartile range (IQR) 22.8-45.8), 16 patients suffered from PVE (0.4%), with a median time to onset of 177 days (IQR 67.8-721.3). First-year incidence was 0.24%, and the overall incidence rate was 0.14 events per 1000 person-years. Overall mortality during follow-up in our study was 31%, of which 25% occurred in hospital. All patients were treated conservatively with intravenous antibiotics alone, and none underwent a re-intervention. Other complications of PVE occurred in 5 patients (31%) and included aortic abscess (2), decompensated heart failure (2) and cerebral embolisation (1). CONCLUSION: PVE in patients receiving TAVR is a relatively rare complication and has a high mortality rate.

Anticoagulant bridging in left-sided mechanical heart valve patients.

BACKGROUND: In preparation for an invasive procedure with a high bleeding risk, patients with a mechanical heart valve temporarily have to discontinue their anticoagulant therapy and are usually bridged with either intravenous unfractionated heparin (UFH) or subcutaneous low-molecular-weight heparin (LMWH). In this study we retrospectively analyzed the safety of UFH versus LMWH as bridging strategy in left-sided mechanical heart valve patients. METHODS: We performed a retrospective multicenter study in four surgical centers in The Netherlands. Patients with a mechanical heart valve implantation bridged from January 2010 until January 2015 were included. The cumulative incidence of adverse events in the 30days following the procedure was recorded. Main outcomes were major bleeding according to International Society on Thrombosis and Haemostasis (ISTH) criteria, symptomatic thromboembolism, and mortality. RESULTS: In total, 238 (174 aortic, 42 mitral, 22 aortic+mitral) bridging episodes were included. The incidence of major bleeding was 16 (19%) events in the UFH group versus 29 (19%) events in the LMWH group (p=0.97). Incidences of thromboembolism were 2 (2.4%) versus 1 (0.6%). The incidence of death was 1 (1.2%) patient in the UFH group versus 3 (1.9%) patients in the LMWH group. More than 50% of all bleeding complications were categorized as a major bleeding. CONCLUSIONS: Bridging anticoagulation in patients with aortic and mitral mechanical valves is associated with considerable risk, but no difference was apparent between UFH and LMWH strategy. The rate of thromboembolism and death was low with either strategy and the vast majority of adverse events were bleedings.

Novel imaging strategies for the detection of prosthetic heart valve obstruction and endocarditis.

Prosthetic heart valve (PHV) dysfunction remains difficult to recognise correctly by two-dimensional (2D) transthoracic and transoesophageal echocardiography (TTE/TEE). ECG-triggered multidetector-row computed tomography (MDCT), 18-fluorine-fluorodesoxyglucose positron emission tomography including low-dose CT (FDG-PET) and three-dimensional transoesophageal echocardiography (3D-TEE) may have additional value. This paper reviews the role of these novel imaging tools in the field of PHV obstruction and endocarditis.For acquired PHV obstruction, MDCT is of additional value in mechanical PHVs to differentiate pannus from thrombus as well as to dynamically study leaflet motion and opening/closing angles. For biological PHV obstruction, additional imaging is not beneficial as it does not change patient management. When performed on top of 2D-TTE/TEE, MDCT has additional value for the detection of both vegetations and pseudoaneurysms/abscesses in PHV endocarditis. FDG-PET has no complementary value for the detection of vegetations; however, it appears more sensitive in the early detection of pseudoaneurysms/abscesses. Furthermore, FDG-PET enables the detection of metastatic and primary extra-cardiac infections. Evidence for the additional value of 3D-TEE is scarce.As clinical implications are major, clinicians should have a low threshold to perform additional MDCT in acquired mechanical PHV obstruction. For suspected PHV endocarditis, both FDG-PET and MDCT have complementary value.

Multimodality Imaging Assessment of Prosthetic Heart Valves.

Echocardiography and fluoroscopy are the main techniques for prosthetic heart valve (PHV) evaluation, but because of specific limitations they may not identify the morphological substrate or the extent of PHV pathology. Cardiac computed tomography (CT) and magnetic resonance imaging (MRI) have emerged as new potential imaging modalities for valve prostheses. We present an overview of the possibilities and pitfalls of CT and MRI for PHV assessment based on a systematic literature review of all experimental and patient studies. For this, a comprehensive systematic search was performed in PubMed and Embase on March 24, 2015, containing CT/MRI and PHV synonyms. Our final selection yielded 82 articles on surgical valves. CT allowed adequate assessment of most modern PHVs and complemented echocardiography in detecting the obstruction cause (pannus or thrombus), bioprosthesis calcifications, and endocarditis extent (valve dehiscence and pseudoaneurysms). No clear advantage over echocardiography was found for the detection of vegetations or periprosthetic regurgitation. Whereas MRI metal artifacts may preclude direct prosthesis analysis, MRI provided information on PHV-related flow patterns and velocities. MRI demonstrated abnormal asymmetrical flow patterns in PHV obstruction and allowed prosthetic regurgitation assessment. Hence, CT shows great clinical relevance as a complementary imaging tool for the diagnostic work-up of patients with suspected PHV obstruction and endocarditis. MRI shows potential for functional PHV assessment although more studies are required to provide diagnostic reference values to allow discrimination of normal from pathological conditions.

Spontaneous coronary artery dissection: current insights and therapy.

Spontaneous coronary artery dissection (SCAD) is a very rare cause of acute coronary syndromes in young otherwise healthy patients with a striking predilection for the female gender. The pathological mechanism has not been fully clarified yet. However, several diseases and conditions have been associated with SCAD, such as atherosclerosis, connective tissue disorders and the peripartum episode. In this paper we present a review of the literature, discussing the possible mechanisms for SCAD, therapeutic options and prognosis. The review is illustrated with two SCAD patients who had a recurrence of a spontaneous dissection in another artery within a few days after the initial event. Because of the susceptibility to recurrent spontaneous dissections we propose at least one week of observation in hospital. Further, we will elaborate on the possible conservative and invasive treatment strategies in the acute phase of SCAD. Primary percutaneous coronary intervention remains the reperfusion strategy of choice; however, in small and medium-sized arteries with normalised flow conservative treatment is defendable. In addition, after the acute phase evaluation of possible underlying diseases is necessary, because it affects further treatment. (Neth Heart J 2008;16:344-9.).

Human hepatic lymph nodes contain normal numbers of mature myeloid dendritic cells but few plasmacytoid dendritic cells.

We investigated whether the hyporesponsiveness of the adaptive arm of the liver immune system is related to the composition of the dendritic cell (DC) population in hepatic lymph nodes. Noninflamed human hepatic lymph nodes (LN) were obtained from multiorgan donors, inflamed hepatic LN from liver transplant recipients with autoimmune cholestatic liver diseases, and inguinal LN from kidney transplant recipients. Quantitative immunohistochemistry showed that all three types of LN contained comparable numbers of mature and immature myeloid DC, but that noninflamed and inflamed hepatic LN contained significantly fewer plasmacytoid DC as compared to inguinal LN. Likewise, DC-enriched cell preparations from hepatic LN contained relatively few plasmacytoid DC. The difference in numbers of plasmacytoid DC was confirmed in comparisons of hepatic LN and ileacal LN from the same organ-donors. Myeloid DC from hepatic LN showed similar expressions of HLA-DR, CD83, and CD86, and higher expression of CD80 compared to myeloid DC from inguinal LN. In conclusion, hepatic LN contain similar numbers of myeloid DC as muscle/skin lymph-draining LN, with no signs of immaturity, but relatively few plasmacytoid DC.