Arterial versus plethysmographic dynamic indices to test responsiveness for testing fluid administration in hypotensive patients: a clinical trial.

In the present study, we compared indices of respiratory-induced variation obtained from direct arterial blood pressure measurement with analogous indices obtained from the plethysmogram measured by the pulse oximeter to assess the value of these indices for predicting the cardiac output increase in response to a fluid challenge. Thirty-two fluid challenges were performed in 22 hypotensive patients who were also monitored with a pulmonary artery catheter. Hemodynamic and plethysmographic data were collected before and after intravascular volume expansion. Patients were classified as nonresponders if their cardiac index did not increase by 15% from baseline. Nonresponding patients had both lower arterial pulse variation ([10 +/- 4]% vs [19 +/- 13]%, P = 0.020) and lower plethysmographic pulse variation ([12 +/- 7]% vs [21 +/- 14]%, P = 0.034) when compared with responders. Fluid responsiveness was similarly predicted by arterial and plethysmographic pulse variations (area under ROC curve 0.74 vs 0.72, respectively, P = 0.90) and by arterial and plethysmographic systolic variation (area under ROC curve 0.64 vs 0.72, respectively, P = 0.50). Nonresponders were identified by changes in pulse variation both on arterial and plethysmographic waveform (area under ROC curve 0.80 vs 0.87, respectively, P = 0.40) and by changes in arterial and plethysmographic systolic variations (area under ROC curve 0.84 vs 0.80, respectively, P = 0.76). In the population studied, plethysmographic dynamic indices of respiratory-induced variation were just as useful for predicting fluid responsiveness as the analogous indices derived from direct arterial blood pressure measurement. These plethysmographic indices could provide a noninvasive tool for predicting the cardiac output increase by administering fluid.