BACKGROUND: To assess preload dependence, the variation of the plethysmographic waveform of pulse oximetry (ΔPOP) has been proposed as a surrogate of the pulse pressure variation (ΔPP). The aim of the study was to assess the ability of the pulse oximeter-derived plethysmographic analysis to accurately trend ΔPP in patients undergoing major abdominal surgery by using standard monitors. METHODS: A continuous recording of arterial and plethysmographic waveform was performed in 43 patients undergoing abdominal surgery. ΔPP and ΔPOP were calculated on validated respiratory cycles. RESULTS: For analysis, 92,467 respiratory cycles were kept (73.5% of cycles recorded in 40 patients). The mean of intrapatient coefficients of correlation was low (r = 0.22). The Bland and Altman analysis showed a systematic bias of 5.21; the ΔPOP being greater than the ΔPP, this bias increased with the mean value of the two indices and the limits of agreement were wide (upper 21.7% and lower -11.3%). Considering a ΔPP threshold at 12% to classify respiratory cycles as responders and nonresponders, the corresponding best cutoff value of ΔPOP was 13.6 ± 4.3%. Using these threshold values, the observed classification agreement was moderate (κ = 0.50 ± 0.09). CONCLUSIONS: The wide limits of agreement between ΔPP and ΔPOP and the weak correlation between both values cast doubt regarding the ability of ΔPOP to substitute ΔPP to follow trend in preload dependence and classify respiratory cycles as responders or nonresponders using standard monitor during anesthesia for major abdominal surgery.
BACKGROUND: To assess preload dependence, the variation of the plethysmographic waveform of pulse oximetry (ΔPOP) has been proposed as a surrogate of the pulse pressure variation (ΔPP). The aim of the study was to assess the ability of the pulse oximeter-derived plethysmographic analysis to accurately trend ΔPP in patients undergoing major abdominal surgery by using standard monitors. METHODS: A continuous recording of arterial and plethysmographic waveform was performed in 43 patients undergoing abdominal surgery. ΔPP and ΔPOP were calculated on validated respiratory cycles. RESULTS: For analysis, 92,467 respiratory cycles were kept (73.5% of cycles recorded in 40 patients). The mean of intrapatient coefficients of correlation was low (r = 0.22). The Bland and Altman analysis showed a systematic bias of 5.21; the ΔPOP being greater than the ΔPP, this bias increased with the mean value of the two indices and the limits of agreement were wide (upper 21.7% and lower -11.3%). Considering a ΔPP threshold at 12% to classify respiratory cycles as responders and nonresponders, the corresponding best cutoff value of ΔPOP was 13.6 ± 4.3%. Using these threshold values, the observed classification agreement was moderate (κ = 0.50 ± 0.09). CONCLUSIONS: The wide limits of agreement between ΔPP and ΔPOP and the weak correlation between both values cast doubt regarding the ability of ΔPOP to substitute ΔPP to follow trend in preload dependence and classify respiratory cycles as responders or nonresponders using standard monitor during anesthesia for major abdominal surgery.