In Response: We thank Dr. Perel for his comments on our study. He raises important issues concerning the effect of volume status on Delta up and the difficulty in applying analysis of systolic pressure variation during spontaneous ventilation. We, too, were surprised to find a statistically significant relationship between Delta up and volume status in mechanically ventilated humans. Our study provides no obvious clues as why this was so, but we do think that species differences cannot be dismissed. All the articles concerning Delta up cited by Dr. Perel were studies performed on dogs [1-4], including the study by Brower et al. [4]. It seems plausible that the lungs of our subjects remained in Zone III throughout the study, given that they remained supine and were never profoundly hypovolemic. Assuming our patients had a typical compliance under anesthesia of 85 mL/cm H2 O [5], static pressure should not have exceeded 8 mm Hg for a typical tidal volume of 900 mL. Extremely low pulmonary artery pressures would have been required to place much of these human lungs into Zone II. The other possible explanations mentioned for the correlation between Delta up and volume status appear unlikely. None of our subjects had any suggestion of cardiac dysfunction, and so we doubt that Delta up was exaggerated on that basis. Failure to achieve a true equilibration of systolic pressure by the end of the expiratory period would indeed shift some of the systolic pressure variation from Delta down to Delta up. Although we did not use a full 5 s of apnea, our records show an unchanging systolic pressure, usually for several beats, prior to the onset of the next inspiration. Regardless of the mechanism behind the correlation between Delta up and volume status, the magnitude of the change was much smaller than for Delta down, and therefore Delta down is clearly the better of the two indices for the assessment of volume status in humans. The remaining issue brought up by Dr. Perel concerned the definitions of Delta up and Delta down during spontaneous ventilation. The physiological mechanisms responsible for Delta up and Delta down are clearly different during mechanical versus spontaneous ventilation; however, we felt constrained by our interpretation of the definitions of Delta up and Delta down. We agree that the distinction between Delta up and Delta down is probably meaningless during spontaneous ventilation and that such compartmentalization of systolic pressure ventilation should be limited to mechanical ventilation. Perhaps it was fortuitous that we could not adequately divide systolic pressure variation into a Delta up and a Delta down during spontaneous ventilation. In summary, we suspect, but cannot prove, that the relationship we observed between Delta up and intravascular volume status was due to differences in the anatomy between dogs and humans, with Zone III dominating the lungs of humans but not dogs. G. Alec Rooke, MD, PhD Howard A. Schwid, MD Department of Anesthesiology University of Washington School of Medicine Seattle, WA 98195