Characteristics of pulmonary artery catheter use in multicenter ICUs in Japan and the association with mortality: a multicenter cohort study using the Japanese Intensive care PAtient Database

BACKGROUND: It has been 50 years since the pulmonary artery catheter was introduced, but the actual use of pulmonary artery catheters in recent years is unknown. Some randomized controlled trials have reported no causality with mortality, but some observational studies have been published showing an association with mortality for patients with cardiogenic shock, and the association with a pulmonary artery catheter and mortality is unknown. The aim of this study was to investigate the utilization of pulmonary artery catheters (PACs) in the intensive care unit (ICU) and to examine their association with mortality, taking into account differences between hospitals. METHODS: This is a retrospective analysis using the Japanese Intensive care PAtient Database, a multicenter, prospective, observational registry in Japanese ICUs. We included patients aged 16 years or older who were admitted to the ICU for reasons other than procedures. We excluded patients who were discharged within 24 h or had missing values. We compared the prognosis of patients with and without PAC. The primary outcome was hospital mortality. We performed propensity score analysis to adjust for baseline characteristics and hospital characteristics. RESULTS: Among 184,705 patients in this registry from April 2015 to December 2020, 59,922 patients were included in the analysis. Most patients (94.0%) with a PAC in place had cardiovascular disease. There was a wide variation in the frequency of PAC use between hospitals, from 0 to 60.3% (median 14.4%, interquartile range 2.2–28.6%). Hospital mortality was not significantly different between the PAC use group and the non-PAC use group in patients after adjustment for propensity score analysis (3.9% vs 4.3%; difference, − 0.4%; 95% CI − 1.1 to 0.3; p = 0.32). Among patients with cardiac disease, those with post-open-heart surgery and those in shock, hospital mortality was also not significantly different between the two groups (3.4% vs 3.7%, p = 0.45, 1.7% vs 1.7%, p = 0.93, 4.8% vs 4.9%, p = 0.87). CONCLUSIONS: The frequency of PAC use varied among hospitals. PAC use for ICU patients was not associated with lower hospital mortality after adjusting for differences between hospitals. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13054-023-04702-4.