SPV, dDown increased from 4.8 ± 1.4 mm Hg and 2.0 ± 1.1 mm Hg, respectively, at Step 1 to 11.2 ± 1.8 mm Hg and 9.2 ± 1.3 mm Hg at Step 3 and then returned to 4.6 ± 1.6 mm Hg and 2.6 ± 1.4 mm Hg after complete retransfusion at Step 5 (Fig. 1, A and C). During hypervolemia, SPV and dDown were unchanged at Steps 6 and 7 but increased again at Steps 8 and 9 when hypervolemia was corrected by blood withdrawal to the baseline level. Although dUp changed significantly, these changes were small and could not be used to differentiate hypovolemia and hypervolemia (Fig. 1C). The changes in SPV were attributed mostly to changes in dDown.
SVV and dPP showed similar changes with dDown or SPV. They increased during hypovolemia and then returned to the baseline level with blood transfusion at Step 5 (Fig. 2, A and B). They decreased at Steps 6 and 7 during hypervolemia but increased again to 13.9% ± 4.6% and 11.8% ± 5.1%, respectively, at Step 9, when hypervolemia was corrected with blood withdrawal to the baseline level. Although FTc did not change during hypovolemia from 255 ± 29 ms at Step 1, it increased to 356 ± 117 ms at Step 5 when hypovolemia was corrected and to 434 ± 124 ms, 406 ± 109 ms, and 363 ± 75 ms at Steps 6, 7, and 8, respectively, during hypervolemia (Fig. 2C.
The mean difference between SVV and dPP was +1.9%, and the limits of agreements (mean ± 2sd) were +9.9 and −6.2% (Fig. 3). There was a significant correlation between SVV and dPP (y = 0.64x + 5.4; r2 = 0.67; P < 0.05).
The comparison of different variables of cardiac preload in this study demonstrated that SPV, dDown, dPP, and SVV are equally sensitive for detecting hypovolemia and for predicting fluid responsiveness during mechanical ventilation, but they do not correctly reflect blood volume status during hypervolemia. However, FTc did not change significantly during hypovolemia but increased significantly during hypervolemia, suggesting that it is more sensitive to blood volume changes during hypervolemia than during hypovolemia.
Attention should be paid to the fact that SPV, dDown, dPP, and SVV may still indicate hypovolemia when intravascular blood volume is corrected to the normovolemic level from severe hypervolemia. Increased venous capacitance due to preceding hypervolemia and, therefore, relative hypovolemia may explain this discrepancy. The increases in these variables should not be interpreted as indications of hypovolemia, but merely as a prediction of fluid responsiveness, when the blood volume state is acutely normalized from the hypervolemic state.
The drawbacks of CVP and PAOP monitoring are 1) relatively large individual variances and 2) invasiveness associated with catheter placement. CVP and PAOP at single points are hence considered poor indicators for the intravascular blood volume state or for predicting the responsiveness to intravascular fluid administration 3). However, CVP and PAOP changed in parallel with the intravascular blood volume status in this study, indicating their value for following an acutely changing intravascular blood volume state across hypovolemia and hypervolemia, at least under conditions with normal cardiac function.
An automated system for the measurement of SPV was first invented by Schwid and Rooke in 2000 9). With our system, both SPV and dPP are measured automatically. Such a system is appropriate for use in all mechanically-ventilated patients whose arterial blood pressure has been monitored via an arterial catheter, because it does not require additional invasion or cost. Whereas Perel et al. 1) and others 10,11) used systolic arterial blood pressure as a reference value after a 5- to 12-second apneic pause, the reference value with this system is calculated from the systolic blood pressures just before and after end-expiration 12). This method has the advantage of not requiring interruption of mechanical ventilation. Whatever the reference methods used, the respiratory setting should not be changed, and the absence of spontaneous respiration should be confirmed, because SPV is affected by the magnitude of tidal volume and respiratory effort 13).
SPV, dDown, dPP, and SVV are referred to as dynamic variables because they reflect respiration-induced cyclic changes in preload, whereas CVP and PAOP are called static variables3). The exact mechanism of SPV is complex and is thought to be a combined reflection of pleural pressure changes and left ventricular stroke volume changes, whereas dPP is thought to reflect changes in left ventricular stroke volume more directly 14). This study confirmed a close relation between SVV and dPP. A study by Reuter et al. 6) showed that SVV correlates well with the retrospective offline quantification of SPV. In line with their study, in this study SVV and SPV changed in parallel. On the basis of these results, it is thus considered that these dynamic variables are equally useful as indicators of hypovolemia in mechanically-ventilated patients.
The EDM has been recognized as a less invasive technique for CO monitoring and preload assessment by means of FTc. FTc is the time required for the left ventricle to eject the stroke volume with correction for the HR. FTc is considered to be closely related to the left ventricular end-diastolic volume and thus can be used as an indicator of the preload. Actually, FTc has been shown to be useful in guiding optimal cardiac filling in patients undergoing surgery 5,7). The results of this study, however, indicated that FTc is not a reliable indicator of hypovolemia because it did not decrease during hypovolemia, as indicated by comparison with the value of normovolemia (Step 1). However, restoration of the blood volume after hypovolemia and the subsequent hypervolemic state resulted in a longer FTc, thus confirming the utility of this variable for assessing cardiac preload during a hypervolemic state. We defined the starting blood volume status as normovolemia, whereas optimal cardiac filling is usually achieved at the point of maximal stroke volume after intravascular fluid administration 7). The FTc (255 ± 29 ms) at baseline in our animals suggests suboptimal cardiac filling, because the normal range of FTc is 330–360 ms in humans 15). In other words, the baseline in this study was normovolemic in terms of the dynamic variables (i.e., SPV, dPP, and dDown) and, at the same time, suboptimal in terms of FTc. This discrepancy in the blood volume state at the baseline may explain the lower sensitivity of FTc during hypovolemia than during hypervolemia. In addition, the increases in vascular resistance during hypovolemia may have blunted a decrease in FTc, because FTc is dependent not only on cardiac preload, but also on vascular resistance 15).
There were some limitations in this experimental study. First, we evaluated the response of the newer variables during induced hypovolemia and hypervolemia. This study does not provide a comparison of these variables for predicting fluid responsiveness. Because the validity of arterial pulse contour-derived SVV as a predictor of fluid responsiveness was contradicted in a recently published article 16), further studies are required to confirm the validity of automatically measured SPV and dPP for predicting fluid responsiveness in different clinical settings. Second, the hypovolemia and hypervolemia induced in our experiment were regarded as moderate to severe, because 200 and 350 mL of blood or hydroxyethyl starch are estimated to be 25% and 43% of circulating blood volume, respectively, in dogs of 10 kg body weight. This is associated with increased sympathetic activity and a compensatory volume shift into and from the intravascular compartment.
In summary, findings in this study, which measured cardiac preload variables simultaneously in an animal model of graded hypovolemia and hypervolemia, indicated that SPV, dDown, dPP, and SVV are useful indicators of hypovolemia, but not of hypervolemia. Further, their values were found to be unreliable when blood volume was restored to normovolemia after severe hypervolemia. FTc is not appropriate for detecting hypovolemia, but it does reflect increases in blood volume during a hypervolemic state, and its usefulness to guide preload optimization is well accepted. Although online SPV and dPP measurements do not require any additional costs or invasion, beyond arterial cannulation, their limits must be considered for the monitoring of blood volume status in mechanically-ventilated patients.
The authors thank T. Ohta, laboratory technician, and M. Takahashi and M. Taguchi, students of Kawasaki Medical College of Allied Health Professions, for their help in the experiment.
1. Perel A, Pizov R, Cotev S. Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded hemorrhage. Anesthesiology 1987;67:498–502.
2. Tavernier B, Makhotine O, Lebuffe G, et al. Systolic pressure variation as a guide to fluid therapy in patients with sepsisinduced hypotension. Anesthesiology 1998;89:1313–21.
3. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest 2002;121:2000–8.
4. Fujita Y, Sari A, Yamamoto T. On-line monitoring of systolic pressure variation. Anesth Analg 2003;96:1529–30.
5. Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: randomised controlled trial. BMJ 1997;315:909–12.
6. Reuter DA, Felbinger TW, Kilger E, et al. Optimizing fluid therapy in mechanically ventilated patients after cardiac surgery by on-line monitoring of left ventricular stroke volume variations: comparison with aortic systolic pressure variations. Br J Anaesth 2002;88:124–6.
7. Gan TJ, Soppitt A, Maroof M, et al. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology 2002;97:820–6.
8. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10.
9. Schwid HA, Rooke GA. Systolic blood pressure at endexpiration measured by the automated systolic pressure variation monitor is equivalent to systolic blood pressure during apnea. J Clin Monit Comput 2000;16:115–20.
10. Pizov R, Cohen M, Weiss Y, et al. Positive end-expiratory pressure-induced hemodynamic changes are reflected in the arterial pressure waveform. Crit Care Med 1996;24:1381–7.
11. Preisman S, DiSegni E, Vered Z, Perel A. Left ventricular preload and function during graded haemorrhage and retranfusion in pigs: analysis of arterial pressure waveform and correlation with echocardiography. Br J Anaesth 2002;88:716–8.
12. Rooke G, Schwid H, Shapira Y. The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation. Anesth Analg 1995;80:925–32.
13. Szold A, Pizov R, Segal E. The effect of tidal volume and intravascular volume state on systolic pressure variation in ventilated dogs. Intensive Care Med 1989;5:368–71.
14. Gunn SR, Pinsky MR. Implications of arterial pressure variation in patients in the intensive care unit. Curr Opin Crit Care 2001;7:212–7.
15. Singer M. Esophageal Doppler monitoring of aortic blood flow: beat-by-beat cardiac output monitoring. Int Anesthesiol Clin 1993;31:99–125.
© 2004 International Anesthesia Research Society
16. Wiesenack C, Prasser C, Rodig G, Keyl C. Stroke volume variation as an indicator of fluid responsiveness using pulse contour analysis in mechanically ventilated patients. Anesth Analg 2003;96:1254–7.