To the Editor:
We read the article by Cannesson1,et al. and the accompanying editorial2with great interest, and praise the effort to better define the clinical utility and applicability of pulse pressure variation (PPV), not only as a tool to predict volume responsiveness but also to move away from a single-threshold value for conducting intraoperative volume optimization and perioperative goal-directed fluid therapy. We would like to make three points.
First, the most commonly used index to assess volume responsiveness, by far in the United States, appears to be stroke volume variation (SVV). Although the area under the receiver-operating curve, in systematic reviews,3shows that both PPV and SVV have excellent sensitivity and specificity in patients who are mechanically ventilated with normal tidal volumes and in a regular sinus rhythm, the threshold values discriminating between fluid responders and nonresponders are not the same for these two parameters.3,4Thus, although strategies using the “gray zone approach” applied to PPV identify a range of values where volume responsiveness cannot reliably be predicted, this range may not be applicable when SVV is used for determining volume responsiveness. Although it is clear from the work of Joseph Erlanger, the father of the pulse pressure concept,5that pulse pressure in man is proportional to left ventricular stroke volume (depending both on arterial tone and cardiac contractility) it is plausible that variations in pulse pressure may also be proportional to variations in stroke volume. To the extent that PPV and SVV are affected differently by changes in arterial tone, given the same degree of volume responsiveness,4these values may lose their direct proportionality as vascular tone changes. Pinsky described the potential utility of a SVV-to-PPV ratio to reflect ventricular-arterial coupling that might be helpful when extrapolating PPV threshold data to SVV.6Overall, applying PPV's “gray zone” to SVV seems clinically appealing but may be misleading. The “gray zone” defined for PPV should not simply be applied to patients who are being optimized using stroke volume variation. The “gray zone” for SVV requires its own definition.
Second, the paper does not address the issue of volume responsiveness versus needing to give volume or using a dynamic index to actively restrict fluids or to administer diuretics to patients under certain clinical circumstances. That is, not all patients who are volume responsive require volume therapy. Conversely, there may be untapped utility for SVV- and/or PPV-guided fluid restriction and diuretic use (consider patients with acute lung injury or acute respiratory distress syndrome).7The goal would be to identify patients with a PPV that is “too low” who, consequently, should receive restrictive fluid therapy or diuretics.
Third, what is, perhaps, not considered by the author2is the fact that PPV goals may vary within individual patients across changing clinical settings in relatively short intervals of time, such as during single lung ventilation for thoracic surgery, laparoscopic pneumoperitoneum, and conditions of pathologic intraabdominal hypertension. For example, the goals for intraoperative PPV during esophagectomy vary during the case, such that the goal during the abdominal part of the procedure is fluid liberal and during the thoracic part of the procedure is fluid restrictive.8
In summary, we praise Cannesson et al. 1for their work and we agree that there is a “gray zone” well represented by the picture in the editorial2(i.e. , the Golden Gate Bridge with fog going across the middle represents “a static view of dynamic indices”). We contend that PPV and SVV “gray zones” are probably better described as the view one gets of the Golden Gate Bridge, with fog going across the middle as one is driving along winding hilly roads. This is a “dynamic view of dynamic indices,” such that the gray zone changes at different times depending on one's changing vantage point, an analogy closer to clinical reality.