Abstract PR617

Abstract

Background & Objectives: The main objective of this work was to develop a parameter able to predict the fluid responsiveness of a patients. The photoplethismographic (PPG) waveform was studied. This parameter would allow the early identification of alterations in the state of the patient that could result in severe injuries. Similarly to the Pleth Variability Index (1) (PVI), our gold standard, two new indexes were computed from the PPG waveform. These indexes were expected to be able to specify the degree of fluid responsiveness in mechanically ventilated patients. Materials & Methods: The new indexes were computed by two different methods: method 1 and method 2, based on the analysis of changes in the amplitude of the PPG waveform in constant time windows. While method 1 takes into account only the maximum and the minimum values of the envelope of the signal, method 2 computes the mean of the envelope. In order to record the physiological parameters, an anesthetic monitor that synchronizes and integrates both hemodynamic and respiratory parameters was used. This monitor allowed exporting the data to a computer. Once exported, the data was converted into ASCII format, and processed using a numerical computing environment for mathematical calculations. The data of 20 patients (30 to 80 years old), with 5658 measurements for each method used were obtained during a month. Results: The histogram showed that the threshold which determines the fluid responsiveness of a patient, stated in 14% for the PVI, decreased for both methods. Both methods showed high correlation and low p-values, although results were better for method 1. Besides, the Bland-Altman test performed better for method 1, showing that the probability of obtaining the same results with method 1 and the gold standard was higher.Conclusion: Method 2 could not be used interchangeably with the gold standard. However, results for method 1 suggested that, eventually, it could be used as a fluid responsiveness predictor.