Pulse contour analysis for cardiac output measurement in patients after off-pump coronary artery bypass grafting: a comparison of FloTrac and PiCCO_pluswith intermittent thermodilution

Introduction

Pulse contour analysis calibrated by transpulmonary thermodilution (PiCCO_plus; Pulsion Medical Systems, Munich, Germany) has shown in the past years to be a reliable alternative to the pulmonary artery catheter for cardiac output (CO) assessment in different clinical settings [1,2]. A new pulse contour analysis device, which does not need an external calibration (FloTrac/ Vigileo; Edwards Lifesciences, Irvine CA, USA), recently became available. The aim of this study was to compare the CO determined by the FloTrac sensor (FCO) and by the PiCCO_plus system (PCO) with the CO assessed by intermittent thermodilution (ICO).

Methods

With local ethic committee approval, patients after elective off-pump coronary artery bypass grafting were studied under different conditions in the ICU. For one set of data (A = 'haemodynamic stable') the CO was assessed following haemo-dynamic stabilization and calibration of the PiCCO_plus. Triplicate FCO and PCO values were recorded within 3 min before the ICO was determined by three repeated injections at four time points with intervals of 15 min. For the second set of data (B = 'haemo-dynamic changes') triplicate FCO, PCO and ICO measurements were recorded 15 min after inducing CO changes by different body positions (supine 1, 30° head-up, 30° head-down, supine 2). Mean arterial pressure was maintained ≥70 mmHg by adjustment of norepinephrine infusion. Statistical analysis was performed using ANOVA with post-hoc Bonferroni/Dunn correction, t test and Bland–Altman analysis for absolute CO values and percentage changes (Δ = trend analysis) of CO during A and B. P < 0.05 was considered statistically significant.

Results

Seventy-two matched sets of data were obtained from nine patients (ASA III, female/male ratio = 2/7, age = 63.0 ± 9.5 years, BMI = 24.8 ± 2.2 kg/m²). CO values recorded during A ranged from 3.30 to 6.56 l/min, no significant CO changes between measurement points were recorded (ΔFCO = -0.8 ± 14.8%, ΔPCO = -0.9 ± 15.3%, ΔICO = -1.9 ± 12.8%). Bland-Altman analysis revealed a mean bias ± 2SD (limits of agreement) of -0.13 ± 1.08 l/min for FCO-ICO and of 0.08 ± 0.91 l/min for PCO-ICO. Differences of ΔCO were comparable (mean bias ± 2SD = 1.1 ± 24.8% for ΔFCO – ΔICO and 1.0 ± 24.8% for ΔPCO – ΔICO). A range of CO values from 2.85 to 8.60 l/min were obtained during B with significant changes of FCO, PCO and ICO between the measurement points (Table 1 and Fig. 1). The mean bias ± 2SD was -0.14 ± 1.82 l/min for FCO-ICO and -0.17 ± 1.13 l/min for PCO-ICO. For ΔFCO – ΔICO a mean bias ± 2SD of -2.8 ± 36.4% was observed, whereas for ΔPCO – ΔICO the mean bias ± 2SD was 0.9 ± 13.2%.

Table 1

Figure 1

CO changes induced by body positioning.

Conclusion

These preliminary results indicate that cardiac output in patients after off-pump coronary artery bypass grafting can be reliably monitored by both tested pulse contour analysis devices (FloTrac and PiCCO_plus system) during stable haemodynamic conditions. However, the FloTrac system showed a tendency to overestimate rapid decreases and increases of cardiac output when compared with the PiCCO_plus system.