Jean-Philippe Empana

Aims: We examined the predictive value of soluble levels of triggering receptor expressed on myeloid cells 1 (sTREM-1) for incident cardiovascular disease (CVD). Methods and results: The main analysis was conducted in an observational prospective cohort. Adjudicated CVD events included coronary heart disease (CHD), stroke, heart failure, and peripheral artery disease. sTREM-1 measures were centralized using a protocol optimized for Meso Scale Discovery technology. Among 9097 CVD-free participants (39% females, mean age 59.6 ± 6.3 years), median sTREM-1 concentration was 193.41 pg/mL (IQR = 158.13–241.54) at baseline (2008–12). Over 10.1 years of median follow-up (IQR = 9.2–12.0), 444 participants experienced 479 CVD events. Higher sTREM-1 concentrations were associated with higher risk of CVD in a multivariable Cox proportional hazard model [hazard ratio (HR) 200–290 vs. 290 vs. <200 pg/mL = 10.40; 95% CI: 7.72–14.02]. Results were consistent across CVD subtypes. Adding sTREM-1 to SCORE2–SCORE2-OP improved the Harrell C index (C index 0.79 vs. 0.71; bootstrapped C-index difference: 0.084; 95% CI: 0.083–0.085) overall including among individuals at low to moderate 10-year risk of CHD or stroke, and reclassification: categorical NRI: 0.34; 95% CI: 0.28–0.40. Findings were similar to other recommended risk scores. Conclusion: In this prospective study, higher circulating sTREM-1 levels were independently associated with incident CVD. Addition of sTREM-1 to established risk scores was associated with improvements in discrimination and reclassification metrics, supporting its potential relevance as a candidate biomarker for further evaluation in primary prevention.

BACKGROUND: The aim of the DESAC study (Devenir des Survivants d’Arrets Cardiaques) was to describe the survival rate and neurological status of out-of-hospital (out-of-hospital cardiac arrest) survivors, and to assess the influence of prehospital and intrahospital factors. METHODS: Between July 2015 and October 2018, consecutive nontraumatic out-of-hospital cardiac arrest patients aged over 18 years discharged alive from 33 intensive care units in Paris and the suburbs (France) were included. Their neurological status was assessed every 6 months using the Cerebral Performance Category (CPC) scale. Prognostic factors associated with a favorable neurological outcome at 24 months (CPC category of 1 or 2) were assessed by covariate-adjusted logistic regression. Trajectories of CPC level over time were derived using a linear mixed model in survivors with at least 3 repeated assessments. RESULTS: Out of 593 survivors, 525 were followed up for 24 months (mean age, 58.1±15.3 years, 79% males), and 449 of 525 (86%) had a favorable neurological outcome. Initial shockable rhythm (adjusted odds ratio, 3.33 [95% CI, 1.57–7.07]) and previous practice of sport (adjusted odds ratio, 3.13 [95% CI, 1.21–8.12]) were associated with favorable outcome, whereas factors with reduced odds ratio were older age (adjusted odds ratio per 10 years, 0.53 [95% CI, 0.41–0.69]) and prior percutaneous coronary intervention (adjusted odds ratio, 0.38 [95% CI, 0.17–0.85]). Therapeutic hypothermia or acute coronary interventions were not related to prognosis ( P >0.05). Overall, among the 514 patients with at least 3 repeated assessments, 5 distinct CPC trajectories (2 reflecting early death were combined) were derived: 64% remained in CPC1, 13.4% in CPC2, 11% improved from CPC2 to CPC1, and 11.6% died either early or lately (>3 months) during follow-up. CONCLUSIONS: This multicentric prospective study indicates that 86% of out-of-hospital cardiac arrest survivors had a favorable neurological outcome at 24 months. Pre-event and intraevent factors (including previous practice of sport and shockable rhythm), but not in-hospital management (therapeutic hypothermia or acute coronary interventions), were associated with a favorable outcome. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02494414.

Aims: The vast majority of sudden cardiac death (SCD) cases occur in the general population with few known risk factors instead of just patients already identified to be at high risk, making the prediction of SCD very difficult. Therefore, a better screening tool should be developed to facilitate early identification. Methods and results: To estimate the risk of SCD, we trained and validated a machine learning model on electronic health record (EHR) data covering 17 172 359 drug prescriptions and 1 639 057 hospital diagnoses up to 5 years for cases and controls. Training was done on data obtained from a cohort of 12 338 SCD cases in Greater Paris and 12 338 controls from 2011 to 2015. We then validated the results on two external cohorts: a temporal cohort in the same area from 2016 and 2020 with 11 620 SCD cases and 11 620 controls and a geographical cohort from the University of Washington (Seattle, USA) with 892 SCD cases and 892 controls from 2013 to 2021. In the 5 years preceding the SCD, cardiovascular diagnoses were prevalent in only a few patients, scarce in many patients, and totally nonexistent for 25.7% of subjects. Our model achieved an area under the curve of 0.81 [95% confidence interval (CI), 0.80–0.82] and 0.66 (95% CI, 0.58–0.73) in the validation and geographical cohort, respectively. The prediction model discriminated SCD from the general population, especially in the highest decile, where the model detected 26% and 33% of all SCD in the Paris and Seattle datasets, respectively. Our prediction model was specific to SCD and was not predictive of myocardial infarction. In addition to classical cardiovascular risk factors, various non-cardiovascular drugs and diagnoses contributed to the prediction model. Conclusion: We propose a prediction model designed to identify individuals at high risk of SCD among the general population. When combined with EHR, this artificial intelligence model has the potential to assist in risk stratification and may help inform the prioritization of preventive strategies, contributing to more targeted use of cardiovascular health resources.

Aims: The rising popularity of endurance races underscores the need to explore the risks of sports-related sudden cardiac arrest (Sr-SCA). Although rare, Sr-SCA is significantly more prevalent in men than in women. The mechanisms underlying these sex differences remain unclear. We aimed to investigate the incidence rates, clinical characteristics, aetiologies, sex differences, and exercise performances among SCA cases during major endurance races in Paris over a 10-year period. Methods and results: We Analysed the Paris Sudden Death Expertise Centre Registry Data (Covering 2011–2024, excluding 2020). This included SCA cases from the half marathon, full marathon and 20 km Parisian race events. We calculated the incidence rates for men and women, with performance analyses focusing on acceleration patterns and the relative risk of SCA in the final kilometre. Among the 1.2 million participants, 17 SCA cases (88% male) were identified, yielding crude incidences of 16.9 and 5.7 per million for men and women, respectively. Sr-SCA was overrepresented in the final kilometres of short races. Men exhibited twice the acceleration rate that women did. Despite extensive medical investigations, no cause was identified in 47.1% of the cases, underscoring the idiopathic nature of Sr-SCA. After hospitalization, 88% (15/17) of the cases survived, all with excellent neurological outcomes [cerebral performance category (CPC) 1], except for one CPC 2. Conclusion: SCA incidences during endurance races are low, with male predominance, high survival rates, and a high proportion of unexplained cases. The male-specific acceleration in the final kilometre may suggest that physiological and behavioural factors influence SCA risk.

Background: We investigated the prevalence, temporal trends and associated factors of overweight and obesity among adults in the WHO African region. Methods: We analysed individual-level data from 54 nationally/sub-nationally representative STEPS surveys conducted between 2003 and 2022 among adults aged 18–69 years. Prevalence estimates were weighted and age-standardised. Time trends were estimated using a Bayesian spatiotemporal modelling approach. Factors associated with body mass index (BMI) categories were identified in hierarchical multinomial mixed-effects logistic regression with random effects for country and survey year, using the normo-weighted as the reference group. Results: The study population included 198 901 adults (50.3% women) with a mean age of 36.3 years. The mean BMI was 23.3±2.0 kg/m ² (24.23±1.60 in women and 22.11±1.53 in men, p for sex difference <0.001). The prevalence of overweight and obesity was 17.8% and 9.0%, respectively, higher in women (20.8% and 13.3%) than in men (14.9% and 4.6%). There was no significant time trend in mean BMI (23.25 kg/m² (95% CI 20.1 to 26.6) in 2003 and 23.43 kg/m² (95% CI 19.3 to 27.8) in 2022, p for trend=0.75). However, obesity prevalence increased from 15.39% to 16.71% (p for trend <0.001), and underweight from 12.07% to 12.76% (p for trend <0.001), whereas overweight plateaued. In multivariate analysis, sex, older age, higher education, physical inactivity and low fruit and vegetable consumption increased the odds of overweight and obesity, whereas past and current smoking showed inverse associations. Specifically, adjusted odds ratios for overweight and obesity for females versus males were 2·07 [(95% CI: 1·83– to 2·34]) and 4.92 [(95% CI: 4·13– to 5·89]); for tertiary education versus no education, they were 2·07 [(95% CI: 1·63– to 2·63]) and 3·77 [(95% CI: 2·77– to 5·11]), respectively. Conclusion: These findings support the urgent need to intensify preventive programmes to fight obesity in the WHO African region.

Aims: Identifying individuals at high risk of sudden cardiac death (SCD) is key, and knowledge on health care consumption prior to the event may be relevant to address this challenge. This study aimed to evaluate healthcare consumption patterns in the year preceding SCD. Methods and results: SCD cases were collected from the Paris Sudden Death Expertise Centre (SDEC) from 2011 to 2020. Using electronic health records from the French National Health Insurance System, all medical interactions were analysed in the year prior to SCD occurrence. To contextualize our findings, the SDEC population was compared with national-level data from the French population, categorized as either above or below the population average, across 2 dimensions: number of hospital diagnoses (primary diagnosis or emergency room visit) and outpatient visits (general practitioner or cardiologist visit). 21 912 SCD were included in the study. Compared with the general population, three distinct patterns of healthcare consumption were identified. The low-interaction group (14%) had minimal healthcare contact, with 3.8 times fewer outpatient and 13.5 times fewer inpatient visits. The intermediate group (40%) showed modest engagement, recording 1.5 times fewer outpatient visits and 5.8 times more inpatient visits. The high-interaction group (46%) had 2.7 times more outpatient and 7.9 times more inpatient visits compared with the general population. Conclusion: This study highlights significant variability in healthcare use during the year preceding SCD. Nearly half of patients had frequent healthcare contacts, suggesting opportunities for earlier identification and prevention.

BACKGROUND: Type 2 diabetes is associated with elevated large artery stiffening, but arterial stiffening may also precede the development of type 2 diabetes. Whether the change in arterial stiffness over time is associated with dysglycemia across the entire spectrum of blood glucose concentration is unknown. Therefore, the primary objective of this study was to quantify the association between the change in arterial stiffness (exposure) and glycemic concentration (outcome). METHODS: Within the prospective WHS II (Whitehall II Study), arterial stiffness was first measured in the 2007/09 examination round (baseline for the current study) and again in 2012/13, using the noninvasive, gold-standard method of carotid to femoral pulse wave velocity (cfPWV). Change in cfPWV was determined as cfPWV in 2012/13-cfPWV−cfPWV in 2007/09. Fasting blood glucose concentrations were measured at baseline (2007/09) and follow-up in 2012/13 and 2015/16. RESULTS: There were 2632 participants without prior type 2 diabetes and cfPWV measured at 2 timepoints with follow-up in 2012/13 and 2282 with follow-up in 2015/16. Greater change in cfPWV was associated with significantly higher fasting glucose concentration at follow-up in 2012/13 and 2015/16 independently of confounders including cfPWV and fasting glucose levels at baseline ( P <0.001 for both). Greater change in cfPWV was significantly and independently associated with a greater change in fasting glucose concentration over the same period (2007/09–2012/13) and beyond (2007/09–2015/16). CONCLUSIONS: Minimizing premature arterial stiffening over time could be a valuable strategy for the prevention of glucose dysregulation and overt type 2 diabetes.

Background: We examined the association between clusters of vascular aging manifestations and ultrasensitivity cardiac troponin I in individuals without cardiovascular disease. Methods and Results: A cross‐sectional analysis was conducted using baseline data from PPS‐3 (Paris Prospective Study III), a French cohort of 10 157 participants. Cardiac troponin I was measured with an ultrasensitive immunoassay with a limit of detection of 0.013 pg/mL. Vascular aging manifestations were assessed via echotracking of the right common carotid artery to measure structural and functional parameters. Hierarchical clustering was used to identify clusters of vascular aging. Multinomial regression assessed the association between vascular aging clusters and cardiac troponin I quintiles. The study included 8722 cardiovascular disease–free participants (mean±SD age, 59.5±6.3 years; 39% women). Three vascular aging clusters were identified. Cluster 1 (n=4158; 47.4%) was characterized as healthy vascular aging with the lowest arteriosclerosis and atherosclerosis indices; cluster 2 (n=2237; 25.5%) was characterized by the highest arteriosclerosis indices, including increased pulse wave velocity, β index, and Young elastic modulus and lowest distensibility coefficient; and cluster 3 (n=2377; 27.0%) was characterized by the highest atherosclerosis indices, including more frequent plaque prevalence and greater intima‐media thickness. Compared with healthy vascular aging, arteriosclerosis and atherosclerosis clusters showed a graded positive association with cardiac troponin I quintiles, independent of traditional risk factors. The adjusted odds ratio for belonging to the highest quintile (quintile 5 versus quintile 1) was 1.55 (95% CI, 1.31–1.92) for arteriosclerosis and 2.66 (95% CI, 2.18–3.23) for atherosclerosis clusters. Conclusions: Vascular aging manifestations of arteriosclerosis and atherosclerosis may partly explain the release of troponin I into the bloodstream in adults without clinical cardiovascular disease. Registration: URL: https://clinicaltrials.gov ; Unique identifier: NCT00741728.

Aims: There is conflicting evidence as to whether patients with pre-existing heart failure (HF) are at increased risk of developing cancer, especially because of common risk factors. We aimed to assess the incidence of cancer in patients with pre-existing HF compared with patients without known HF. Methods and results: The French National Administrative Health Data System, containing all healthcare information for 99% of the French population, was used to identify adult patients with a first diagnosis of HF between 2010 and 2019 and without a history of cancer before HF diagnosis. HF patients were matched for sex and age to HF-free and cancer-free individuals (3:1 ratio). We found 330 867 HF patients, and 992 601 matched controls (54.7% women, mean age 77.7 ± 13.5 years). A first cancer was diagnosed in 28 151 (8.5%) HF patients over a mean follow-up of 4.3 ± 2.8 years compared with 77 325 (7.8%) in the controls over 4.9 ± 2.8 years of follow-up (unadjusted sub-distribution hazard ratio, sHR: 1.12 [1.11–1.13], P < 0.001). The higher risk of new cancer in HF patients remained after full adjustment for major comorbidities, age, sex, year of diagnosis, region of residence, tobacco use, and alcohol consumption (adjusted sHR = 1.06, [1.04–1.07]; P < 0.0001). Overall, the calculated attributable risk of new cancer after HF was 16.5% (13.3.9–20.4%). This increased risk was observed for most solid malignancies (especially colorectal and lung cancer), and for multiple myeloma. Conclusion: Patients with a history of HF have a higher risk of developing cancer than the general population. Cancer screening strategies should be advocated in patients with HF.