Limits...
Evaluation of cardiac output by 5 arterial pulse contour techniques using trend interchangeability method

View Article: PubMed Central - PubMed

ABSTRACT

Cardiac output measurement with pulse contour analysis is a continuous, mini-invasive, operator-independent, widely used, and cost-effective technique, which could be helpful to assess changes in cardiac output. The 4-quadrant plot and the polar plot have been described to compare the changes between 2 measurements performed under different conditions, and the direction of change by using different methods of measurements. However, the 4-quadrant plot and the polar plot present a number of limitations, with a risk of misinterpretation in routine clinical practice. We describe a new trend interchangeability method designed to objectively define the interchangeability of each change of a variable. Using the repeatability of the reference method, we classified each change as either uninterpretable or interpretable and then as either noninterchangeable, in the gray zone or interchangeable. An interchangeability rate can then be calculated by the number of interchangeable changes divided by the total number of interpretable changes. In this observational study, we used this objective method to assess cardiac output changes with 5 arterial pulse contour techniques (Wesseling's method, LiDCO, PiCCO, Hemac method, and Modelflow) in comparison with bolus thermodilution technique as reference method in 24 cardiac surgery patients. A total of 172 cardiac output variations were available from the 199 data points: 88 (51%) were uninterpretable, according to the first step of the method. The second step of the method, based on the 84 (49%) interpretable variations, showed that only 18 (21%) to 30 (36%) variations were interchangeable regardless of the technique used. None of pulse contour cardiac output technique could be interchangeable with bolus thermodilution to assess changes in cardiac output using the trend interchangeability method in cardiac surgery patients. Future studies may consider using this method to assess interchangeability of changes between different methods of measurements.

No MeSH data available.


Simulated data in a 4-quadrant graphical representation using reference method repeatabilities of 5% (A) and 20% (B). A specific color is applied to each change: uninterpretable (blue), noninterchangeable (red), in the gray zone of interpretation (orange), and interchangeable (green). ΔRM = changes in reference method, ΔTM = changes in test method.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4998299&req=5

Figure 5: Simulated data in a 4-quadrant graphical representation using reference method repeatabilities of 5% (A) and 20% (B). A specific color is applied to each change: uninterpretable (blue), noninterchangeable (red), in the gray zone of interpretation (orange), and interchangeable (green). ΔRM = changes in reference method, ΔTM = changes in test method.

Mentions: Three hundred simulated data points were analyzed. A wide distribution of measurements was observed, ranging from 0.01 to 8.12 units for the RM and from −0.22 to 7.80 units for the TM; the mean value for both methods was 3.67 units. The changes of measurements for the RM and TM were −3.17 to 5.50 units and −3.24 to 5.58 units, respectively. When the repeatability coefficient of the RM was set at 5%, 50 (16%) changes were uninterpretable, 149 (50%) were noninterchangeable, 45 (15%) were situated in the gray zone, and 56 (19%) were interchangeable. In contrast, when the repeatability coefficient of the RM was set at 20%, 117 (39%) changes were uninterpretable, 11 (4%) were noninterchangeable, 49 (16%) were situated in the gray zone, and 123 (41%) were interchangeable. Graphical representation using the previous color code is presented for R = 5% (Fig. 5A) and R = 20% (Fig. 5B), respectively. According to the previous definition, the trend interchangeability rate was then calculated as 56/250 (22%) for R = 5%, and 123/183 (67%) for R = 20%. All data and calculations are presented in Appendix 2.


Evaluation of cardiac output by 5 arterial pulse contour techniques using trend interchangeability method
Simulated data in a 4-quadrant graphical representation using reference method repeatabilities of 5% (A) and 20% (B). A specific color is applied to each change: uninterpretable (blue), noninterchangeable (red), in the gray zone of interpretation (orange), and interchangeable (green). ΔRM = changes in reference method, ΔTM = changes in test method.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4998299&req=5

Figure 5: Simulated data in a 4-quadrant graphical representation using reference method repeatabilities of 5% (A) and 20% (B). A specific color is applied to each change: uninterpretable (blue), noninterchangeable (red), in the gray zone of interpretation (orange), and interchangeable (green). ΔRM = changes in reference method, ΔTM = changes in test method.
Mentions: Three hundred simulated data points were analyzed. A wide distribution of measurements was observed, ranging from 0.01 to 8.12 units for the RM and from −0.22 to 7.80 units for the TM; the mean value for both methods was 3.67 units. The changes of measurements for the RM and TM were −3.17 to 5.50 units and −3.24 to 5.58 units, respectively. When the repeatability coefficient of the RM was set at 5%, 50 (16%) changes were uninterpretable, 149 (50%) were noninterchangeable, 45 (15%) were situated in the gray zone, and 56 (19%) were interchangeable. In contrast, when the repeatability coefficient of the RM was set at 20%, 117 (39%) changes were uninterpretable, 11 (4%) were noninterchangeable, 49 (16%) were situated in the gray zone, and 123 (41%) were interchangeable. Graphical representation using the previous color code is presented for R = 5% (Fig. 5A) and R = 20% (Fig. 5B), respectively. According to the previous definition, the trend interchangeability rate was then calculated as 56/250 (22%) for R = 5%, and 123/183 (67%) for R = 20%. All data and calculations are presented in Appendix 2.

View Article: PubMed Central - PubMed

ABSTRACT

Cardiac output measurement with pulse contour analysis is a continuous, mini-invasive, operator-independent, widely used, and cost-effective technique, which could be helpful to assess changes in cardiac output. The 4-quadrant plot and the polar plot have been described to compare the changes between 2 measurements performed under different conditions, and the direction of change by using different methods of measurements. However, the 4-quadrant plot and the polar plot present a number of limitations, with a risk of misinterpretation in routine clinical practice. We describe a new trend interchangeability method designed to objectively define the interchangeability of each change of a variable. Using the repeatability of the reference method, we classified each change as either uninterpretable or interpretable and then as either noninterchangeable, in the gray zone or interchangeable. An interchangeability rate can then be calculated by the number of interchangeable changes divided by the total number of interpretable changes. In this observational study, we used this objective method to assess cardiac output changes with 5 arterial pulse contour techniques (Wesseling's method, LiDCO, PiCCO, Hemac method, and Modelflow) in comparison with bolus thermodilution technique as reference method in 24 cardiac surgery patients. A total of 172 cardiac output variations were available from the 199 data points: 88 (51%) were uninterpretable, according to the first step of the method. The second step of the method, based on the 84 (49%) interpretable variations, showed that only 18 (21%) to 30 (36%) variations were interchangeable regardless of the technique used. None of pulse contour cardiac output technique could be interchangeable with bolus thermodilution to assess changes in cardiac output using the trend interchangeability method in cardiac surgery patients. Future studies may consider using this method to assess interchangeability of changes between different methods of measurements.

No MeSH data available.