ブックタイトル第43回日本集中治療医学会学術集会プログラム・抄録集

概要

第43回日本集中治療医学会学術集会プログラム・抄録集

－165－used arterial pressure waveform analysis from an indwelling arterial catheter or an estimate of the arterial waveform using aplethysmographic signal from a finger cuff. Hadian et al. compared the three most commonly used minimally-invasive devices:PiCCO, LiDCO and FloTrac to each other and a paired thermodilution cardiac output measured from a pulmonary arterycatheter in 17 post-operative surgical patients. They showed that although all devices gave similar mean cardiac output values,their trending and dynamic responses correlated poorly with each other. In a recent review, Stagt examined the accuracy of thelatest generation software Flotrac system and found it acceptable for clinical use in the operating room［20］. This is important,because FloTrac is often used to compare cardiac output and SVV accuracy of other devices. For example, others showed thata similar minimally-invasive device, IntelliVue MP was as accurate as FloTrac in defining mean cardiac output values in 47septic shock patients［21］. Most likely any new devise that used arterial pulse pressure to estimate flow will show similardegrees of general accuracy and poorer degrees of trending accuracy. Similarly, Vos et al.［22］compared the ability of a noninvasiveplethysmographic measure of arterial pressure（Massimo Radical）with invasive FloTrac to estimate cardiac output in30 patients undergoing major hepatic resection. They found that the new non-invasive device gave similar cardiac output andSVV values. In contrast to these authors, Monnet et al.［23］showed that when arterial tone was altered by norepinephrine incritically ill patients, FloTrac poorly tracked changes in cardiac output. Presumably the FloTrac algorithm accuracy degradeswhen large changes in arterial impedance occurs, as may occur with the use of norepinephrine or when acute endotoxin inducesshock［24］.Neither PPV nor SVV are accurate predictors of fluid responsiveness in the setting of forceful spontaneous inspiratory efforts,because the inspiration-induced increase in RV filling causes LV diastolic compliance to decrease, thus for the same wall stress,LV end-diastolic volume is less. To circumvent this very real limitation, two alternative physiologic preload stresses are used.First a PLR test will transiently increase venous return equal to approximately 300 ml in a 70 kg patient. In the volumeresponsive patient this will result in a transient increase in cardiac output. Similarly, one can merely stop positive pressurebreathing and note the transient increase in systolic arterial pressure, which will also reflect volume responsiveness.Another insightful study was done by Marik et al.［25］who used a combined non-invasive estimate of cardiac output bybioreactance（NICOM）with regional measures of cerebral blood flow by carotid Doppler in 34 critically ill patients. Volumeresponsiveness was assessed by the PLR maneuver in non-ventilated patients and SVV in 19 mechanically ventilated patients.Only half their patients were volume responsive. Importantly, only the volume responders displayed an increase in carotid flow（79± 32%）. Furthermore, if carotid flow were used as the non-invasive estimate of volume responders to a PLR maneuver, athreshold value of 20% separated responders from non-responders with a high sensitivity and specificity. Since sustainingadequate cerebral blood flow is an important resuscitation target, these data suggest that in severely ill patients measures ofcarotid blood flow changes in response to a PLR maneuver may greatly augment the functional hemodynamic useful of thesemeasures.Fluid therapy is one of the first steps in the goal-directed therapy［26］and is a central part of the Surviving Sepsis Guidelines［27］. Thus, an adequate assessment of fluid responsiveness should improve the therapy. One study［19］conducted in high risksurgery patients showed that a volume loading guided to the goal of PPV minimization improved postoperative outcome anddecreases length of hospital stay. Patients in the interventional group received more fluid than the control group and had alsoless postoperative complications, lower duration of mechanical ventilation and lower stay in the intensive care unit.Assessing arterial toneAlthough PPV and SVV cannot be accurate interpreted as measures of volume responsiveness in patients with atrial fibrillation,their ratio always defines dynamic central arterial elastance（Ea）［28］. If the arterial circuit becomes stiffer, then for the samestroke volume change, arterial pulse pressure will change more and vice versa. Using this approach, Monge et al.［29］assessedthe effect of volume loading on arterial pressure in hypotensive septic shock patients whose PPV predicted that they werevolume responsive. All patients increased their cardiac output, as expected, in response to the fluid challenge, but only thosepatients with normal or increased Ea also increased their arterial pressure. Importantly, they could not predict who wouldincrease their arterial pressure based on pre-challenge measures of systemic vascular resistance, mean arterial pressure or theratio of arterial pulse pressure to stroke volume, only the PPV/SVV slope defining Ea predicted responders from nonresponders.They found that Ea <0.9 reflected a severely vasodilator state. In support of this study, Hadian et al.［30］demonstrated that when post-operative cardiac surgery patients were given vasodilator therapy they significantly decreased Eafrom 1.44 to 1.13. Collectively, these data strongly support the use of PPV/SVV estimates of Ea as part of the overallassessment strategy of critically ill patients. Indeed if arterial tone is markedly decreased PPV may not reliably track SVV at all［31］.Identification of cardiovascular insufficiencyCardiovascular insufficiency is characterized by an inadequate O2 delivery relative to the metabolic demands. Shock can be, inthe early stages, compensated by autonomic mechanisms, such regional vasoconstriction, in an attempt to maintain central bloodpressure and vital organ perfusion above an anaerobic threshold. In this stage of compensated shock, microcirculatory measureslike arterial pressure or cardiac output are often inside the range of values defined as normal and, therefore, insensitive as earlypredictors of subsequent decompensation due to the increased risk of tissue ischemia and subsequent development of multi