Predicting target lesion failure following percutaneous coronary intervention through machine learning risk assessment models.

Aims: Central to the practice of precision medicine in percutaneous coronary intervention (PCI) is a risk-stratification tool to predict outcomes following the procedure. This study is intended to assess machine learning (ML)-based risk models to predict clinically relevant outcomes in PCI and to support individualized clinical decision-making in this setting. Methods and results: Five different ML models [gradient boosting classifier (GBC), linear discrimination analysis, Naïve Bayes, logistic regression, and K-nearest neighbours algorithm) for the prediction of 1-year target lesion failure (TLF) were trained on an extensive data set of 35 389 patients undergoing PCI and enrolled in the global, all-comers e-ULTIMASTER registry. The data set was split into a training (80%) and a test set (20%). Twenty-three patient and procedural characteristics were used as predictive variables. The models were compared for discrimination according to the area under the receiver operating characteristic curve (AUC) and for calibration. The GBC model showed the best discriminative ability with an AUC of 0.72 (95% confidence interval 0.69-0.75) for 1-year TLF on the test set. The discriminative ability of the GBC model for the components of TLF was highest for cardiac death with an AUC of 0.82, followed by target vessel myocardial infarction with an AUC of 0.75 and clinically driven target lesion revascularization with an AUC of 0.68. The calibration was fair until the highest risk deciles showed an underestimation of the risk. Conclusion: Machine learning-derived predictive models provide a reasonably accurate prediction of 1-year TLF in patients undergoing PCI. A prospective evaluation of the predictive score is warranted. Registration: Clinicaltrial.gov identifier is NCT02188355.

Prevalence, Treatment, and Control Rates of Conventional and Ambulatory Hypertension Across 10 Populations in 3 Continents.

Hypertension is a major global health problem, but prevalence rates vary widely among regions. To determine prevalence, treatment, and control rates of hypertension, we measured conventional blood pressure (BP) and 24-hour ambulatory BP in 6546 subjects, aged 40 to 79 years, recruited from 10 community-dwelling cohorts on 3 continents. We determined how between-cohort differences in risk factors and socioeconomic factors influence hypertension rates. The overall prevalence was 49.3% (range between cohorts, 40.0%-86.8%) for conventional hypertension (conventional BP ≥140/90 mm Hg) and 48.7% (35.2%-66.5%) for ambulatory hypertension (ambulatory BP ≥130/80 mm Hg). Treatment and control rates for conventional hypertension were 48.0% (33.5%-74.1%) and 38.6% (10.1%-55.3%) respectively. The corresponding rates for ambulatory hypertension were 48.6% (30.5%-71.9%) and 45.6% (18.6%-64.2%). Among 1677 untreated subjects with conventional hypertension, 35.7% had white coat hypertension (23.5%-56.2%). Masked hypertension (conventional BP <140/90 mm Hg and ambulatory BP ≥130/80 mm Hg) occurred in 16.9% (8.8%-30.5%) of 3320 untreated subjects who were normotensive on conventional measurement. Exclusion of participants with diabetes mellitus, obesity, hypercholesterolemia, or history of cardiovascular complications resulted in a <9% reduction in the conventional and 24-hour ambulatory hypertension rates. Higher social and economic development, measured by the Human Development Index, was associated with lower rates of conventional and ambulatory hypertension. In conclusion, high rates of hypertension in all cohorts examined demonstrate the need for improvements in prevention, treatment, and control. Strategies for the management of hypertension should continue to not only focus on preventable and modifiable risk factors but also consider societal issues.

Renal artery and parenchymal changes after renal denervation: assessment by magnetic resonance angiography.

OBJECTIVES: Relatively little is known about the incidence of long-term renal damage after renal denervation (RDN), a potential new treatment for hypertension. In this study the incidence of renal artery and parenchymal changes, assessed with contrast-enhanced magnetic resonance angiography (MRA) after RDN, is investigated. METHODS: This study is an initiative of ENCOReD, a collaboration of hypertension expert centres. Patients in whom an MRA was performed before and after RDN were included. Scans were evaluated by two independent, blinded radiologists. Primary outcome was the change in renal artery morphology and parenchyma. RESULTS: MRAs from 96 patients were analysed. Before RDN, 41 renal anomalies were observed, of which 29 mostly mild renal artery stenoses. After a median time of 366 days post RDN, MRA showed a new stenosis (25-49% lumen reduction) in two patients and progression of pre-existing lumen reduction in a single patient. No other renal changes were observed and renal function remained stable. CONCLUSIONS: We observed new or progressed renal artery stenosis in three out of 96 patients, after a median time of 12 months post RDN (3.1%). Procedural angiographies showed that ablations were applied near the observed stenosis in only one of the three patients. KEY POINTS: • The incidence of vascular changes 12 months post RDN was 3.1%. • No renal vascular or parenchymal changes other than stenoses were observed. • Ablations were applied near the stenosis in only one of three patients.

Prevalence and Determinants of Masked Hypertension Among Black Nigerians Compared With a Reference Population.

Hitherto, diagnosis of hypertension in sub-Saharan Africa was largely based on conventional office blood pressure (BP). Data on the prevalence of masked hypertension (MH) in this region is scarce. Among individuals with normal office BP (<140/90 mm Hg), we compared the prevalence and determinants of MH diagnosed with self-monitored home blood pressure (≥135/85 mm Hg) among 293 Nigerians with a reference population consisting of 3615 subjects enrolled in the International Database on Home Blood Pressure in Relation to Cardiovascular Outcomes. In the reference population, the prevalence of MH was 14.6% overall and 11.1% and 39.6% in untreated and treated participants, respectively. Among Nigerians, the prevalence standardized to the sex and age distribution of the reference population was similar with rates of 14.4%, 8.6%, and 34.6%, respectively. The mutually adjusted odds ratios of having MH in Nigerians were 2.34 (95% confidence interval, 1.39-3.94) for a 10-year higher age, 1.92 (1.11-3.31) and 1.70 (1.14-2.53) for 10- or 5-mm Hg increments in systolic or diastolic office BP, and 3.05 (1.08-8.55) for being on antihypertensive therapy. The corresponding estimates in the reference population were similar with odds ratios of 1.80 (1.62-2.01), 1.64 (1.45-1.87), 1.13 (1.05-1.22), and 2.84 (2.21-3.64), respectively. In conclusion, MH is as common in Nigerians as in other populations with older age and higher levels of office BP being major risk factors. A significant proportion of true hypertensive subjects therefore remains undetected based on office BP, which is particularly relevant in sub-Saharan Africa, where hypertension is now a major cause of death.