Relationship between Minimally Invasive and Non-invasive Cardiac Ouput Measurements In Patients Admitted to Intensive Care Unit – An Observational Study

INTRODUCTION: Shock is most frequently encountered situations in ICU. Measuring Cardiac Output (CO) will help differentiate type of shock and guide interventions.(1) CO CAN BE MEASURED BY: Pulmonary Artery Catheterisation (PAC): gold standard but invasive. Minimally invasive: pulse contour analysis. Non-invasive: Cardiac Ultrasound (CUS). PRIMARY OBJECTIVE: To find relationship between cardiac output obtained by minimally invasive and non-invasive methods in patients admitted to intensive care. MATERIALS AND METHODS: Patients fulfilling the inclusion criteria were recruited and baseline cardiac ultrasound was performed to rule out poor cardiac window. Patient's demographic data was noted. Arterial pressure based cardiac output (APCO) was noted from Edward monitor and transducer, connected to Flotrac sensor which was connected to radial artery catheter of patient and CUS was performed simultaneously. End systolic and end diastolic volumes of the left ventricle were measured in apical four chamber and apical two chamber views. Stroke volume was calculated by the average of the above mentioned two methods. CO was calculated [CO = SV × HR]. Relationship between COAPCO and COCUS analyzed by Spearman coefficient Bias and precision calculated by Bland and Altman analysis. RESULTS: Total of 88 readings (n) were obtained in 30 patients (N). Spearman's correlation coefficient (ρ) between CO obtained by COCUS and APCO was 0.371 which showed weak correlation and good association (r2 = 0.691). One sample t-test was performed for difference between cardiac output obtained by COCUS and APCO to rule out potential proportional bias, which was statistically significant (p=0.001). Considering difference between the two cardiac outputs as the dependant variable and average of the two as independent variable It was further analysed by linear regression which showed p = 0.587. As the p was >0.05, it indicated a fixed bias (Table 1). To validate our study Bland and Altman analysis was done and plots were constructed, with difference between CO obtained by COCUS and APCO on y axis and average of cardiac outputs obtained by COCUS and APCO on x axis. CONCLUSION: CO obtained by APCO correlates weakly and overestimates the CO obtained by COCUS by 2.69L/min. Minimally invasive cardiac output method can be used with non-invasive methods interchangeably after the correction of fixed bias.