Phenotypic clustering of patients hospitalised for acute cardiac events in intensive cardiac care unit (ICCU): The ADDICT-ICCU trial

Intensive cardiac care units (ICCU) have been developed initially to monitor and manage ventricular arrhythmias occurring after acute coronary syndrome. However, ICCUs have diversified their activity with a more heterogeneous population, and its characteristics are not well depicted by conventional methods. To identify patient subgroups by phenotypic unsupervised clustering integrating clinical, biological and echocardiographic data to unveil pathophysiological differences between subgroups of consecutive patients admitted to ICCU. From 7 to 22 April 2021, screening for illicit drug was performed prospectively by systematic urinary testing in all consecutive patients admitted to ICCU in 39 centres throughout France. The primary outcome was in-hospital major adverse cardiac events (MAEs) defined by death, resuscitated cardiac arrest or cardiogenic shock. A cluster analysis was performed on clinical, biological, echocardiographic variables using KAMILA algorithm. Of 1499 patients admitted to ICCU (70% male, mean age 63 ± 15 years), 61 experienced a MAE (4.1%). Four distinct phenogroups (PG) were identified by KAMILA algorithm: [PG1, n = 535] non-ST-elevation acute coronary syndrome (NSTEMI) patients, [PG2, n = 444] younger smokers with ST-elevation acute coronary syndrome (STEMI), [PG3, n = 273], elderly patients with conduction disorders and few traditional risk factors, and [PG4, n = 247] patients with acute heart failure. Using regression analysis, the occurrence of MAE differed among the four phenogroups (P < 0.001). Compared to the PG1 as reference, multivariate logistic regression analysis revealed that PG2 and PG3 had a worse prognosis and PG4 had the worst prognosis regarding the occurrence of MAE (PG2: OR = 3.13, 95%CI, 1.16–10.0; PG3: OR = 3.16, 95% CI, 1.02–10.8; PG4: odds-ratio [OR] = 20.5, 95% CI, 8.7–60.8; all P < 0.05) (Fig. 1). Cluster analysis of clinical, biological and echocardiographic variables identified 4 phenogroups of patients admitted to ICCU that were associated with distinct prognostic profiles. Trial Registration: ClinicalTrials.gov Identifier: NCT05063097