Noninvasive Hemodynamic Monitoring

The initial promise of functional hemodynamic monitoring1 was to advise the bedside clinician in either the operating room or acute care environment whether his or her patient was volume responsive. The point, of course, was to guide therapy for the hypotensive patient requiring increased cardiac output. If the patient would respond to volume, then resuscitation should start with fluids. To that end we2–4 and others5,6 showed that threshold values of arterial pulse pressure variation (PPV) during positive pressure accurately separated volume responders from nonresponders. Subsequently, stroke volume variation (SVV) estimated from esophageal Doppler7 or derived from arterial waveform analysis8 also proved predictive of volume responsiveness. With greater scrutiny, it became clear that these robust measures were limited to patients on mechanical ventilation with tidal volumes of 8 mL/kg or greater9 (although they still remain predictive of fluid responsiveness when >13% in this setting), and that they could not predict fluid responsiveness accurately in conditions such as increased intraabdominal pressure,10 atrial fibrillation,11 and acute cor pulmonale. In addition, it was recently identified that when PPV is between 8% and 13% (“gray zone”), other tests should be used to aid the bedside clinician in defining volume responsiveness.3 Two recent developments simplify these issues. First, one may use the magnitude of the PPV or SVV as a surrogate for cardiac reserve. If a patient has a good cardiac reserve, then the SVV or PPV will be higher. The inverse is also true. If a patient has poor cardiac reserve, then SVV and PPV will be lower. This is expected, of course, because the Starling mechanism is the underlying physiology responsible for PPV and SVV. Maas et al.12 demonstrated that postoperative cardiac surgery patients given norepinephrine to increase mean arterial pressure by 20 mm Hg also increased their cardiac output if their initial SVV was >12%. Thus, one may use PPV and SVV as functional parameters of cardiac reserve. Second, although the absolute values of PPV and SVV may be altered by tidal volume and other disease states, as listed above, the dynamic relationship between changes in stroke volume and arterial pulse pressure is independent of these limitations. As quantified by the ratio of PPV to SVV and approximate dynamic arterial elastance, PPV/SVV appears to define central arterial tone. Monge Garcia et al. demonstrated that hypotensive but volume-responsive septic patients had a variable blood pressure response to fluid resuscitation that was dependent on PPV/SVV or dynamic arterial elastance. A dynamic arterial elastance <0.9 identified patients whose mean arterial pressure would not increase with fluid loading even though cardiac output did.13 Dynamic arterial elastance is typically about 1.2 to 2, so values ≤0.9 are very low. That study was important because it showed that arterial pulse pressure and aortic flow, both measured separately, could accurately identify central vasomotor state. However, these measures of pressure and flow require the use of separate devices to minimize mathematical coupling potentially present if one used arterial pressure waveforms to calculate stroke volume. Still, measuring aortic flow independently of arterial pressure is cumbersome without the use of an esophageal Doppler probe, and thus not commonly available to practicing clinicians. That is why the present study by Cecconi et al.14 is so interesting as a follow-up on the clinical trial to the initial one by Monge Garcia (co-author on this trial) et al. in septic patients.13 These investigators used a noninvasive finger plethysmographic monitor of arterial pressure, the Nexfin (BMEye, Amsterdam, The Netherlands) to estimate arterial pressure and PPV. From these measurements, they calculated the stroke volume and SVV in surgical patients. They found that in subjects who were volume responsive, an arterial elastance <1.06 identified patients whose blood pressure would not increase after a fluid challenge. This study is important because the investigators used a noninvasive device to estimate pressure and flow, thus making the application of much greater utility in clinical practice. Furthermore, they applied this monitoring as part of their usual care to a standardized volume challenge. However, the study has several limitations. The most important limitation is that the investigators estimated stroke volume from the pulse pressure; thus, the 2 signals are necessarily mathematically coupled. Despite this limitation, PPV/SVV predicted blood pressure response, and the threshold values to predict blood pressure increases were similar across studies. Second, the number of test subjects who were actually hypotensive in the entire cohort was small, decreasing the ability of the study to detect statistically significant increases in mean arterial pressure in the hypotensive subjects. Finally, the Nexfin device requires a good pressure signal in the fingers to allow for calculation of both pressure and flow. If vasopressors are being used, this may not be the case.15 These important results leave us with 2 remaining questions: First, are these results reproducible in a more diverse group of patients? And second, when cardiovascular reserve and vasomotor tone are assessed using the PPV and SVV signals, does the information demonstrably improve patient resuscitation, more rapidly restore hemodynamic stability, and reduce the incidence of untoward events? DISCLOSURES Name: Maxime Cannesson, MD, PhD. Contribution: This author helped write the manuscript. Attestation: Maxime Cannesson approved the final manuscript. Conflicts of Interest: Maxime Cannesson consulted for Edwards Lifesciences, received research funding from Edwards Lifesciences, consulted for Masimo, received research funding from Masimo, has equity interest in Sironis, has equity interest in Gauss, and consulted for Covidien. Name: Michael Pinsky, MD, CM, Dr hc, MCCM, FCCP. Contribution: This author helped write the manuscript. Attestation: Michael Pinsky approved the final manuscript. Conflicts of Interest: Michael Pinsky received honoraria from Edwards Lifesciences and received research funding from Edwards Lifesciences. RECUSE NOTE Dr. Maxime Cannesson is the Section Editor for Technology, Computing, and Simulation for the Journal. This manuscript was handled by Dr. Steven L. Shafer, Editor-in-Chief. Dr. Cannesson was not involved in any way with the editorial process or decision.