C25 PREDICTION OF ALL–CAUSE MORTALITY FOLLOWING PERCUTANEOUS CORONARY INTERVENTION IN BIFURCATION LESIONS USING MACHINE LEARNING ALGORITHMS – THE RAIN–ML PREDICTION MODEL

Aims

Stratifying prognosis following coronary bifurcation percutaneous coronary intervention (PCI) is an unmet need. Machine learning (ML) may identify patterns from multidimensional, non–linear relationships to make outcome predictions. We sought to develop a ML–based risk stratification model built on clinical, anatomical and procedural features to predict all–cause mortality following contemporary bifurcation PCI.

Methods and Results

Multiple ML models to predict all–cause mortality were tested on a cohort of 2,393 patients (training, n = 1,795; internal validation, n = 598) undergoing bifurcation PCI with contemporary stents from the real–world RAIN (veRy thin stents for patients with left mAIn or bifurcatioN in real life) registry. Among 38 commonly available features, 25 (13 patient–related, 12 lesion–related) were selected to train ML models. The best performing model (the RAIN–ML prediction model) was validated in an external validation cohort of 1,701 patients undergoing bifurcation PCI from the DUTCH PEERS (DUrable polymer–based sTent CHallenge of Promus ElemEnt versus ReSolute integrity: TWENTE II) trial and the BIO–RESORT trial cohorts. The area under the receiver operating characteristic curves for the prediction of 2–year mortality was 0.786 (0.74–0.83) in the overall population, 0.736 (0.72–0.847) at internal validation and 0.706 (0.6919–0.794) at external validation. Performance at risk ranking analysis, k–center cross validation, and with continual learning confirmed the generalizability of the models, available also as an online interface.

Conclusions

The RAIN–ML prediction model represents the first tool combining clinical, anatomical and procedural features to predict all–cause mortality among patients undergoing contemporary bifurcation PCI with a good discriminative performance.

Machine learning applied to ambulatory blood pressure monitoring: a new tool to diagnose autonomic failure?

BACKGROUND

Autonomic failure (AF) complicates Parkinson's disease (PD) in one-third of cases, resulting in complex blood pressure (BP) abnormalities. While autonomic testing represents the diagnostic gold standard for AF, accessibility to this examination remains limited to a few tertiary referral centers.

OBJECTIVE

The present study sought to investigate the accuracy of a machine learning algorithm applied to 24-h ambulatory BP monitoring (ABPM) as a tool to facilitate the diagnosis of AF in patients with PD.

METHODS

Consecutive PD patients naïve to vasoactive medications underwent 24 h-ABPM and autonomic testing. The diagnostic accuracy of a Linear Discriminant Analysis (LDA) model exploiting ABPM parameters was compared to autonomic testing (as per a modified version of the Composite Autonomic Symptom Score not including the sudomotor score) in the diagnosis of AF.

RESULTS

The study population consisted of n = 80 PD patients (33% female) with a mean age of 64 ± 10 years old and disease duration of 6.2 ± 4 years. The prevalence of AF at the autonomic testing was 36%. The LDA model showed 91.3% accuracy (98.0% specificity, 79.3% sensitivity) in predicting AF, significantly higher than any of the ABPM variables considered individually (hypotensive episodes = 82%; reverse dipping = 79%; awakening hypotension = 74%).

CONCLUSION

LDA model based on 24-h ABPM parameters can effectively predict AF, allowing greater accessibility to an accurate and easy to administer test for AF. Potential applications range from systematic AF screening to monitoring and treating blood pressure dysregulation caused by PD and other neurodegenerative disorders.

Sphingolipid composition of circulating extracellular vesicles after myocardial ischemia

Sphingolipids are structural components of cell membrane, displaying several functions in cell signalling. Extracellular vesicles (EV) are lipid bilayer membrane nanoparticle and their lipid composition may be different from parental cells, with a significant enrichment in sphingolipid species, especially in pathological conditions. We aimed at optimizing EV isolation from plasma and describing the differential lipid content of EV, as compared to whole plasma. As pilot study, we evaluated the diagnostic potential of lipidomic signature of circulating EV in patients with a diagnosis of ST-segment-elevation myocardial infarction (STEMI). STEMI patients were evaluated before reperfusion and 24-h after primary percutaneous coronary intervention. Twenty sphingolipid species were quantified by liquid-chromatography tandem-mass-spectrometry. EV-ceramides, -dihydroceramides, and -sphingomyelins increased in STEMI vs. matched controls and decreased after reperfusion. Their levels correlated to hs-troponin, leucocyte count, and ejection fraction. Plasma sphingolipids levels were 500-to-700-fold higher as compared to EV content; nevertheless, only sphingomyelins differed in STEMI vs. control patients. Different sphingolipid species were enriched in EV and their linear combination by machine learning algorithms accurately classified STEMI patients at pre-PCI evaluation. In conclusion, EV lipid signature discriminates STEMI patients. These findings may contribute to the identification of novel biomarkers and signaling mechanisms related to cardiac ischemia.

KCNJ5 Mutations: Sex, Salt and Selection

Somatic mutations have been identified in the KCNJ5 gene (encoding the potassium channel GIRK4) in aldosterone-producing adenomas (APA). Most of these mutations are located in or near the selectivity filter of the GIRK4 channel pore and several have been shown to lead to the constitutive overproduction of aldosterone. KCNJ5 mutations in APA are more frequent in women; however, this gender dimorphism is a reported phenomenon of Western but not East Asian populations. In this review we discuss some of the issues that could potentially underlie this observation.

Subtype Diagnosis of Primary Aldosteronism: Approach to Different Clinical Scenarios

Identification and management of patients with primary aldosteronism are of utmost importance because it is a frequent cause of endocrine hypertension, and affected patients display an increase of cardio- and cerebro-vascular events, compared to essential hypertensives. Distinction of primary aldosteronism subtypes is of particular relevance to allocate the patients to the appropriate treatment, represented by mineralocorticoid receptor antagonists for bilateral forms and unilateral adrenalectomy for patients with unilateral aldosterone secretion. Subtype differentiation of confirmed hyperaldosteronism comprises adrenal CT scanning and adrenal venous sampling. In this review, we will discuss different clinical scenarios where execution, interpretation of adrenal vein sampling and subsequent patient management might be challenging, providing the clinician with useful information to help the interpretation of controversial procedures.

Renin and aldosterone measurements in the management of arterial hypertension

Renin-angiotensin-aldosterone system (RAAS) is recognized as the main regulatory system of hemodynamics in man, and its derangements have a key role in the development and maintenance of arterial hypertension. Classification of the hypertensive states according to different patterns of renin and aldosterone levels ("RAAS profiling") allows the diagnosis of specific forms of secondary hypertension and may identify distinct hemodynamic subsets in essential hypertension. In this review, we summarize the application of RAAS profiling for the diagnostic assessment of hypertensive patients and discuss how the pathophysiological framework provided by RAAS profiling may guide therapeutic decision-making, especially in the context of uncontrolled hypertension not responding to multi-therapy.

Primary aldosteronism: who should be screened?

Primary aldosteronism (PA) has a prevalence in the general hypertensive population from 5 to 10%, and is widely recognized as the most frequent form of secondary hypertension. The 2 main PA subtypes are aldosterone producing adenoma (APA) and bilateral adrenal hyperplasia (BAH) that account for 95% of all PA cases. The diagnosis of PA is a 3-step process that comprises screening, confirmatory testing, and subtype differentiation. The different categories of patients at an increased risk of PA who should thus undergo a screening test were described in the first Endocrine Society (ES) Practice Guidelines for diagnosis and treatment of PA published in 2008. These categories include patients with Joint National Committee Stage 2, Stage 3, or drug-resistant hypertension; hypertension, and spontaneous or diuretic-induced hypokalemia; hypertension with adrenal incidentaloma; hypertension and a family history of early-onset hypertension or cerebrovascular accident at a young age and all hypertensive first degree relatives of patients with PA. Recently, a growing number of studies have linked PA with the metabolic syndrome, diabetes, and obstructive sleep apnea that may be partly responsible for the higher rate of cardio and cerobrovascular accidents in PA patients. The aim of this review is to discuss, which patients should be screened for PA, focusing not only on the well-established categories of the ES Guidelines, but also on additional other group of patients with a potentially high prevalence of PA that has emerged from recent research.