A3060
October 14, 2013
8:00:00 AM - 9:00:00 AM
Room Room 104-Area A
Impact of Heart Rate to Respiratory Rate Ratio on the Predictive Value of Pulse Pressure Variation in the Intraoperative Setting
Brenton S. Alexander, B.S., Yannick Le Manach, M.D.,Ph.D., Christoph K. Hofer, M.D., Benoit Tavernier, M.D., Maxime Cannesson, M.D.,Ph.D.
University of California, Irvine, Irvine, California, United States
Introduction: Pulse pressure Variation (PPV) has been shown to be the best predictor of fluid responsiveness in patients undergoing mechanical ventilation. The next logical step is to accurately define the limitations of this technology in order to ensure its proper application. Among other limitations (spontaneous breathing, tidal volume below 8 mL/kg, open chest surgeries, non standard abdominal pressure, left ventricular failure, ARDS), it has recently been illuminated that the ratio between heart rate and respiratory rate (HR/RR) needs to be adequately high in order to allow for proper variations to be recorded1. Our aim is to expand this analysis into the intraoperative setting and determine if this ratio still impacts the ability of PPV to predict fluid responsiveness.

Material and Methods: 413 patients undergoing general anesthesia and mechanical ventilation in four multinational centers were analyzed2. Stroke volume (SV), heart rate (HR), respiratory rate (RR), central venous pressure (CVP) and pulse pressure variation (PPV) were recorded before and after a 500 mL fluid bolus. Responders were defined as those patients having an increase of at least 15% in cardiac output (CO). Patients were then stratified into quartiles (n=89, 125, 100, 99) based on their HR/RR ratio. Individual groups were then analyzed and compared in order to determine which, if any, specific PPV threshold values needed adjustment from the established gray zone values of 9-13%. Receiver operator curves defined optimal threshold values for optimized sensitivity and specificity. In order to determine a gray zone for each threshold value, a population of 1,000 bootstrapped analyses was created and a 95% CI was recorded for each quartile.

Results: Each group’s area under the curve (AUC) was statistically significant (p<0.05). Quartile 1 (HR/RR ≤ 4) had an AUC=0.875 and threshold of 8.7 (95% CI 7.0-11.4), Quartile 2 (4 < HR/RR ≤ 5) had an AUC=0.861 and threshold of 12.3 (95% CI 9.0-13.0), Quartile 3 (5 < HR/RR ≤ 6) had an AUC=0.924 and threshold of 11.0 (95% CI 7.6-12.0), Quartile 4 (6 < HR/RR) had an AUC=0.883 and threshold of 11.7 (95% CI 8.0-12.7)(Figure 1). The percentages of patients for each quartile that fell into their grey zones (95% CI’s) are 34%, 27%, 33%, and 19%, respectively.

Discussion: As shown in the ICU setting, decreasing HR/RR beyond a certain point necessitates an adjustment in the PPV threshold value and gray zone for predicting fluid responsiveness. The change is small but significant for patients with a HR/RR below 4, with a new gray zone being established from 8.7-11.4. The clinical interpretation that should be emphasized as a take-away point from this analysis is that if an intraoperative patient is being fluid optimized using a dynamic predictor of fluid responsiveness, adjustments must be made if they are found to have an unusually high RR or (more likely) an abnormally low HR.



Figure 1
Figure 2

Copyright © 2013 American Society of Anesthesiologists