Effects of injection site on the accuracy of thermal washout right ventricular ejection fraction measurements in clinical and model investigations

Modified Thermodilution for Simultaneous Cardiac Output and Recirculation Assessment in Veno-venous Extracorporeal Membrane Oxygenation: A Prospective Diagnostic Accuracy Study

BACKGROUND

Thermodilution is unreliable in veno-venous extracorporeal membrane oxygenation (VV-ECMO). Systemic oxygenation depends on recirculation fractions and ratios of extracorporeal membrane oxygenation (ECMO) flow to cardiac output. In a prospective in vitro simulation, this study assessed the diagnostic accuracy of a modified thermodilution technique for recirculation and cardiac output. The hypothesis was that this method provided clinically acceptable precision and accuracy for cardiac output and recirculation.

METHODS

Two ECMO circuits ran in parallel: one representing a VV-ECMO and the second representing native heart, lung, and circulation. Both circuits shared the right atrium. Extra limbs for recirculation and pulmonary shunt were added. This study simulated ECMO flows from 1 to 2.5 l/min and cardiac outputs from 2.5 to 3.5 l/min with recirculation fractions (0 to 80%) and pulmonary shunts. Thermistors in both ECMO limbs and the pulmonary artery measured the temperature changes induced by cold bolus injections into the arterial ECMO limb. Recirculation fractions were calculated from the ratio of the areas under the temperature curve (AUCs) in the ECMO limbs and from partitioning of the bolus volume (flow based). With known partitioning of bolus volumes between ECMO and pulmonary artery, cardiac output was calculated. High-precision ultrasonic flow probes served as reference for Bland-Altman plots and linear mixed-effect models.

RESULTS

Accuracy and precision for both the recirculation fraction based on AUC (bias, -5.4%; limits of agreement, -18.6 to 7.9%) and flow based (bias, -5.9%; limits of agreement, -18.8 to 7.0%) are clinically acceptable. Calculated cardiac output for all recirculation fractions was accurate but imprecise (RecirculationAUC: bias 0.56 l/min; limits of agreement, -2.27 to 3.4 l/min; and RecirculationFLOW: bias 0.48 l/min; limits of agreement, -2.22 to 3.19 l/min). Recirculation fraction increased bias and decreased precision.

CONCLUSIONS

Adapted thermodilution for VV-ECMO allows simultaneous measurement of recirculation fraction and cardiac output and may help optimize patient management with severe respiratory failure.

EDITOR’S PERSPECTIVE

Right atrial effects on right ventricular ejection fraction derived from thermodilution measurements

OBJECTIVE

The thermodilution technique provides a convenient means to monitor cardiac output, right ventricular (RV) ejection fraction (EF), and volumes at the bedside. To calculate RVEF from the pulmonary artery temperature curve, the bolus thermodilution technique assumes that right atrial (RA) temperature returns to baseline value within 1 beat following the cold saline injection. The authors hypothesized that this assumption is the reason why the thermodilution technique consistently underestimates RVEF.

DESIGN

A theoretical analysis and animal study.

SETTING

Laboratory, university, multi-institutional.

PARTICIPANTS

Animals.

INTERVENTIONS

Cold saline injections.

MEASUREMENTS AND MAIN RESULTS

In 2 porcine experiments, after a rapid injection of cold saline into right atrium, RA temperature took several heartbeats to return to baseline. In a theoretical analysis, if after the cold saline injection RA temperature returned to baseline in 1 beat (RAEF = 1), then thermodilution-derived RVEF(T) = actual RVEF(A). In contrast, if RA temperature took several beats to return to baseline (RAEF = RVEF), then RVEF(T) consistently underestimated RVEF(A). A least square fit of RVEF(A) versus RVEF(T) resulted in RVEF(A) = 1.0 x RVEF(T) + 0.11. Applying this correction (adding 0.11 to RVEF(T)) to the data gave relatively small errors in estimating RVEF over a wide EF range.

CONCLUSIONS

After injecting cold saline into the right atrium, RA temperature takes several heart beats to return to baseline temperature, leading to underestimating RVEF and overestimating RV volumes. The pulsed thermal energy approach by injecting heat into the RV avoids these problems, but the impact of its small temperature signal on RVEF measurements needs to be determined.

Right atrial pressure predicts hemodynamic response to apneic positive airway pressure

OBJECTIVE

To evaluate if the preexistant filling state, assessed by right atrial pressure (RAP), pulmonary artery occlusion pressure (PAOP), and right ventricular end-diastolic volume index (EDVI), would define the subsequent hemodynamic effects of increases in airway pressure (Paw).

DESIGN

Prospective open clinical study.

SETTING

Postoperative intensive care unit, university hospital.

PATIENTS

Twenty-two consecutive ventilator-dependent patients with mild to severe acute lung injury with Murray scores (scoring infiltrates on chest radiograph, oxygenation index, lung compliance, and the level of positive end-expiratory pressure) ranging from 0.5 to 3.0 without history of preexisting cardiopulmonary disease.

INTERVENTIONS

Paw varied during apnea from 0 to 10, 20, and 30 cm H2O using inspiratory hold maneuvers of 15 secs.

MEASUREMENTS AND MAIN RESULTS

Cardiac index and right ventricular ejection fraction were measured by the thermodilution technique. We made measurements in triplicate using manual injection of iced saline. Right ventricular volumes were calculated. Increasing Paw induced variable changes in cardiac index among subjects (+6% to -43% change from baseline 0 cm H2O Paw values), which correlated with percentage changes in both stroke index (r2 = .89) and right ventricular EDVI (r2 = .75), whereas heart rate and right ventricular ejection fraction did not change. The change in cardiac index from 0 to 30 cm H2O Paw correlated with baseline values for RAP, PAOP, and right ventricular EDVI (r2 = .68, .43, and .34, respectively, p < 0.01). Increases in RAP correlated with lung compliance if baseline RAP was >10 mm Hg but did not if it was < or =10 mm Hg. Similarly, patients with baseline RAP < or =10 mm Hg had a greater decrease in cardiac index than patients with a RAP >10 mm Hg (for 30 cm H2O Paw: -30% +/- 9% vs. -8% +/- 7%, p < .01).

CONCLUSIONS

Apneic positive Paw decreased cardiac output mainly by reducing venous return. From the investigated filling variables, RAP was most sensitive in predicting the hemodynamic response, followed by PAOP and right ventricular EDVI. Patients with RAP < or =10 mm Hg, if subjected to aggressive positive pressure ventilation, are at risk of hemodynamic deterioration and organ hypoperfusion.

Measurement of right ventricular performance during apnea in patients with acute lung injury

BACKGROUND

Mechanical ventilation resulted in increased, decreased or unchanged end-diastolic volumes together with either profoundly decreased or unchanged right ventricular ejection fraction (RVEF). The goal of our study was, therefore, to evaluate the effects of positive end-expiratory pressure on measurements of RVEF performed during apneic periods with different levels of positive end-expiratory pressure.

METHODS

Fifteen consecutive patients suffering from acute lung injury after major surgery or trauma were included. Measurements were performed during 15 seconds of apnea at airway pressure levels of 0 (baseline), 10, 20, and 30 cm H2O. Cardiac output and RVEF were determined using the thermodilution technique.

RESULTS

Lung inflation to an airway pressure of 30 cm H2O caused a 22 +/- 14% decrease of cardiac output resulting from a 20 +/- 14% decrease of stroke volume index. The decrease of stroke volume index was induced by a 17 +/- 11% decrease of right ventricular end-diastolic volume index, while RVEF remained virtually unchanged (0.49 +/- 0.10 vs. 0.47 +/- 0.12 at 0 and 30 cm H2O, respectively). Relative changes of cardiac output were closely correlated with changes of right ventricular end-diastolic volume index (p < 0.05, r2 = 0.78).

CONCLUSIONS

Right ventricular systolic function was well maintained despite substantially decreased end-diastolic volumes. In our study, during apneic conditions, higher levels of positive end-expiratory pressure did not worsen RVEF in patients with acute lung injury. The proposed technique of apneic lung inflation may serve as an alternative approach to obtain comparable measurements of RV function in patients with acute lung injury.

Thermodilution cardiac output measurement during simultaneous volume infusion through the venous infusion port of the pulmonary artery catheter

OBJECTIVE

To determine the effect on thermodilution cardiac output (TDCO) measurements of continuous volume infusion through the right atrial venous infusion port of the pulmonary artery catheter.

DESIGN

Prospective, blinded, randomly allocated crossover.

SETTING

Surgical intensive care unit in a university hospital.

PARTICIPANTS

Forty-nine hemodynamically stable patients between the ages of 43 and 84 in the intensive care unit and with a pulmonary artery catheter in place.

INTERVENTIONS

Each patient received two room-temperature infusions (250 or 500 mL/hr) of 5% dextrose. Consecutive TDCO measurements were taken during each infusion and during a control period (no infusion). The sequence of infusion rates and control period was randomly determined. At both infusion rates, most patients exhibited a decrease in TDCO measurement of between 0.1 and 3.0 L/min. Although some patients exhibited an increase in measured TDCO, this increase was primarily between 0.1 and 0.5 L/min.

MEASUREMENTS AND MAIN RESULTS

The mean TDCO measurements at infusion rates of 0, 250, and 500 mL/hr were 6.77, 6.49, and 6.47 L/min, respectively. Measured TDCO decreased on volume infusion by 0.3 +/- 0.13 L/min (mean +/- SD, p < 0.02).

CONCLUSION

Rapid continuous infusion of fluid through the venous infusion port of the pulmonary artery catheter significantly limits the accuracy of simultaneous TDCO measurements. Optimally, TDCO measurements should be avoided during rapid volume infusion.

Monitoring of right ventricular function using a conventional slow response thermistor catheter

OBJECTIVE

To investigate whether determination of right ventricular end-diastolic volume (RVEDV) and right ventricular ejection fraction (RVEF) can be performed with reasonable accuracy and reproducibility using a conventional slow response thermistor pulmonary artery catheter (CPAC) applying an adaptive algorithm.

DESIGN

To study RVEDV and RVEF simultaneously with pulmonary artery catheters equipped with slow and fast response thermistors (FRPAC) under a broad range of cardiac output.

SETTING

Laboratory of Institute of Experimental Surgery, Technical University.

ANIMALS

11 anaesthetised piglets.

INTERVENTIONS

Hypovolemia (V-) was induced by withdrawal of blood up to 50 ml/kg, hypervolemia (V+) was produced by retransfusing blood and adding up to 30 mg/kg hydroxyethyl starch. In 5 animals in phases V- and V+ beta-adrenergic stimulation was achieved with dobutamine. Finally pulmonary artery hypertension was induced by infusion of small air bubbles.

MEASUREMENTS AND RESULTS

Cardiac output (CO), RVEDV and RVEF were determined simultaneously with FRPAC and CPAC placed in the same pulmonary artery branch. Measurements were repeated 8 times sequentially in steady state normovolemia. A total of 130 measurements could be analysed. The coefficient of variation was 6.7 +/- 4.2% for CO(FRPAC) and 4.6 +/- 1.7% for CO(CPAC); for RVEF it was 9.7 +/- 6.2% (FRPAC) and 9.9 +/- 3.9% (CPAC); for RVEDV it was 11.6 +/- 4.8% (FRPAC) and 8.54 +/- 3.2 (CPAC). Mean difference (bias) was 0.06 +/- 0.39 l/min for CO measured with both methods, 19 +/- 35 ml for RVEDV and -3.3 +/- 6.5% for RVEF. CO(CPAC) displayed a strong correlation to CO(FRPAC) (R = 0.97, p = 0.001) as well as RVEF (R for RVEF(CPAC) versus RVEF(FRPAC) = 0.90, p = 0.001). R for RVEDV(CPAC) versus RVEDV(FRPAC) was 0.67, p = 0.001. We conclude that this animal study demonstrates good agreement between RVEF and RVEDV obtained with catheters equipped with a fast response thermistor or with a conventional slow response thermistor allowing accurate monitoring of right ventricular function with a conventional pulmonary artery catheter.

Influence of constant sustained positive airway pressure on right ventricular performance

OBJECTIVE

The detrimental effect of positive airway pressure on right ventricular (RV) performance is controversial and the aim of this study was to determine the effects of constant positive airway pressure without ventilatory fluctuation on RV performance with the aid of a pulmonary arterial catheter equipped with a rapid response thermistor for measuring RV ejection fraction (RVEF) and RV end-diastolic volume index (RVEDVI).

DESIGN

A prospective, clinical study.

SETTING

The central operating theatre of a university hospital.

PATIENTS

Nine patients who had major surgery and required right heart catheterization for normal clinical management.

MEASUREMENTS AND RESULTS

Cold indicator was injected into the RV 4 or 5 times for each airway pressure (0, 10 or 20 cmH2O) which was maintained manually stable for 15 s, and 9 paired data were analyzed by repeated-measures analysis of variance. They are separated into two groups; RVEF at zero airway pressure greater (A group) or less (B group) than 0.4. In A group (7 patients), increasing airway pressures (0 vs 10 vs 20 cmH2O) did not affect RVEF (0.55 +/- 0.05 vs 0.54 +/- 0.06 vs 0.56 +/- 0.04), RVEDVI (69 +/- 36 vs 73 +/- 29 vs 58 +/- 20 ml.m-2), or stroke volume index (SVI: 38 +/- 18 vs 40 +/- 17 vs 33 +/- 13 ml.beat-1.m-2); however, in B (2 patients), RVEF (0.35 and 0.38 vs 0.31 and 0.28 vs 0.19 and 0.17) and SVI (35 and 28 vs 32 and 27 vs 27 and 23) decreased, while RVEDVI increased (99 and 73 vs 103 and 97 vs 146 and 132).

CONCLUSIONS

In most patients, the changes in RVEF, SVI, and RVEDVI did not occur under constant positive airway pressure, therefore the changes reported in mechanically ventilated patients may not attributable to the extent of positive airway pressure but rather to abrupt increases in airway pressure. These appears, however, to be patients whose RV function is so disturbed that they cannot cope with increased afterloads.

Combined dose ratios of dopamine and dobutamine and right ventricular performance after cardiac surgery

The effect of combined administration of different dose ratios of dobutamine (DB) and dopamine (DA) (DB/DA ratio of 1:1; 1.5:0.5; 2:0; 0.5:1.5; and 0:2), with the added dose kept constant (10 micrograms/kg/min-20 micrograms/kg/min), on right ventricular function (measured by the thermal washout method with the aid of a rapid-response thermistor) was determined in ten patients after cardiac surgery (between 12 and 24 h after surgery). The following values represent the mean +/- SD of DB only and of the DB/DA-equal combination vs DA only. The DB/DA-equal or DB-dominant combination increased the right ventricular ejection fraction vs DA only (0.39 +/- 0.12 [p less than 0.01] and 0.37 +/- 0.11 [p less than 0.05], respectively, vs 0.32 +/- 0.12) and the stroke volume index (43 +/- 12 ml/m2 [p less than 0.01] and 41 +/- 15 ml/m2, respectively, vs 38 +/- 14 ml/m2) and decreased right ventricular end-diastolic pressure (RVEDP) (10 +/- 4 mm Hg [p less than 0.01] and 11 +/- 4 mm Hg [p less than 0.05], respectively, vs 13 +/- 5 mm Hg) and pulmonary capillary wedge pressure (10 +/- 4 mm Hg [p less than 0.01] and 12 +/- 5 mm Hg [p less than 0.05], respectively, vs 14 +/- 6 mm Hg) to the same degree as DB alone. The DB/DA-equal or DB-dominant combination did not induce tachycardia (heart rate, 105 +/- 11 [p less than 0.05] and 95 +/- 14 beats per minute, respectively, vs 90 +/- 17 beats per minute) or have any effect on the right ventricular end-diastolic volume index (RVEDVI) (115 +/- 30 ml/m2 and 117 +/- 33 ml/m2, respectively, vs 127 +/- 42 ml/m2). Moreover, the diastolic parameters of the right ventricle (the ratio of RVEDVI/RVEDP: 15 +/- 8 [p less than 0.05] and 13 +/- 7, ml/mm Hg/m2, respectively, 11 +/- 5 ml/mm Hg/m2) decreased as the ratio of DA increased. This change in the diastolic properties of the right ventricle might have been caused by release of norepinephrine in the myocardium by DA or by improved coronary perfusion with DB. The DB/DA-equal and DB-dominant combinations were superior to DB or DA alone and to the DA-dominant combination in obtaining enhanced right ventricular performance.