Association of 24-h Blood Pressure Pattern With Mortality in ICU Patients: A Multicenter Retrospective Study

Abstract

Blood pressure (BP) is a crucial component of the APACHE II scoring system for assessing the severity of illness in ICU patients, and it plays a pivotal role in predicting patient mortality. Based on fluctuations, the 24-h BP patterns of ICU patients can be categorized into dippers (10% ≤ the fall < 20%), extreme-dippers (fall ≥ 20%), non-dippers (0% ≤ the fall < 10%), and reverse-dippers (fall < 0%). This study aims to investigate whether there are statistically significant differences in ICU mortality, in-hospital mortality, 28-day mortality, and 1-year mortality among the dipper, non-dipper, extreme-dipper, and reverse-dipper groups. We enrolled all adult patients with continuous BP monitoring within 24 h of ICU admission. Using Navicat Premium 16 software, we extracted the first 24-h BP values of 10462 patients from the MIMIC IV v2.2 database. Patients were then classified into the dipper group (n = 1244), non-dipper group (n = 6162), reverse-dipper group (n = 2940), and extreme-dipper group (n = 116). Among ICU patients, the non-dipper pattern group constituted the largest proportion (58.90%), followed by the reverse-dipper pattern group (28.10%). After adjusting for relevant confounding factors, we found that the reverse-dipper group had the strongest correlation with in-hospital mortality (OR: 1.592, p < 0.05), 28-day mortality (OR: 1.607, p < 0.01), 90-day mortality (OR: 1.402, p < 0.01), 180-day mortality (OR: 1.403, p < 0.01), and 1-year mortality (OR: 1.525, p < 0.001), with statistical significance observed for all these associations. In the ICU setting, the non-dipper BP pattern is the most prevalent. However, the reverse-dipper pattern is the most significantly associated with mortality.