Thermodilution vs pressure recording analytical method in hemodynamic stabilized patients

• Published in: Journal of critical care 2014-04, Vol.29 (2), p.260-264
• Main Authors: Donati, Abele, MD, Carsetti, Andrea, MD, Tondi, Stefania, MD, Scorcella, Claudia, MD, Domizi, Roberta, MD, Damiani, Elisa, MD, Gabbanelli, Vincenzo, MD, Münch, Christopher, MD, Adrario, Erica, MD, Pelaia, Paolo, MD, Cecconi, Maurizio, MD
• Format: Article
• Language: English
• Subjects: Adult; Aged; Blood Pressure; Cardiac output; Cardiac Output - physiology; Catheterization, Swan-Ganz; Catheters; Confidence intervals; Critical Care; Critical Illness; Female; Germany; Heart rate; Hemodynamic monitoring; Humans; Intensive Care Units; Male; Methods; Middle Aged; Monitoring, Physiologic - instrumentation; Monitoring, Physiologic - methods; Pressure recording analytical method; Prospective Studies; Reproducibility of Results; Thermodilution; Thermodilution - methods; Veins & arteries
• ISSN: 0883-9441; 1557-8615
• DOI: 10.1016/j.jcrc.2013.11.003

Abstract Purpose Many mini-invasive devices to monitor cardiac output (CO) have been introduced and, among them, the pressure recording analytical method (PRAM). The aim of this study was to assess the agreement of PRAM with the intermittent transpulmonary thermodilution and continuous pulmonary thermodilution in measuring CO in hemodynamically stabilized patients. Materials and Methods This is a prospective clinical study in a mixed medical-surgical intensive care unit (ICU) and in a postcardiac surgical ICU. Forty-eight patients were enrolled: 32 patients to the medical-surgical ICU monitored with PiCCO (Pulsion Medical System AG, Munich, Germany) and 16 were cardiac patients monitored with Vigilance (Edwards Lifesciences, Irvine, CA). Results A total of 112 measurements were made. Ninety-six comparisons of paired CO measurements were made in patients hospitalized in medical-surgical ICU; 16, in cardiac surgical patients. The mean Vigilance-CO was 4.49 ± 0.99 L/min (range, 2.80-5.90 L/min), and the mean PRAM-CO was 4.27 ± 0.88 L/min (range, 2.85-6.19 L/min). The correlation coefficient between Vigilance-CO and PRAM-CO was 0.83 (95% confidence interval, 0.57-0.94; P < .001). The bias was 0.22 ± 0.55 L/min with limits of agreement between 0.87 and 1.30 L/min. The percentage error was 25%. Mean TP-CO was 6.78 ± 2.04 L/min (range, 4.12-11.27 L/min), and the mean PRAM-CO was 6.11 ± 2.18 L/min (range, 2.82-10.90 L/min). The correlation coefficient between PiCCO-CO and PRAM-CO was 0.91 (95% confidence interval, 0.83-0.96; P < .0001). The bias was 0.67 ± 0.89 L/min with limits of agreement − 1.07 and 2.41 L/min. The coefficient of variation for PiCCO was 4% ± 2%, and the coefficient of variation for PRAM was 10% ± 8%. The percentage error was 28%. Conclusions The PRAM system showed good agreement with pulmonary artery catheter and PiCCO in hemodynamically stabilized patients.