Disclosure OBS Medical funds a small amount of research not - PDF document

5/19/2014 Goal-Directed Fluid Resuscitation Christopher G. Choukalas, MD, MS Department of Anesthesia and Perioperative Care University of California, San Francisco Disclosure • OBS Medical funds a small amount of research not related to this talk Obligatory joke • Not much new this year • Horse with colic 1

5/19/2014 The case for why it matters • Fluid balance a common concern • Sepsis • ALI/ARDS • Sepsis PLUS ARDS! • Patients with Sepsis who developed ALI • 4 groups: – Adequate initial + Conservative late fluids – Adequate initial only – Conservative late only – Neither Murphry, CV, et al. 2009. Chest. 136(1) 2

5/19/2014 It matters • So how do we do it? I would posit two factors: • Hemodynamic: – Is the circulation adequate? • Metabolic – Are oxygen delivery and utilization adequate? • Both have their own goals. Hemodynamic Goals • Blood pressure • CVP • Dynamic respiratory indices: – Pulse pressure/systolic pressure/perfusion index variation 3

5/19/2014 Hemodynamic Goals • Blood pressure • CVP • Dynamic respiratory indices: – Pulse pressure/systolic pressure/perfusion index variation Blood pressure • A proxy for flow, end organ perfusion • Flow = pressure/resistance • Do we ever really KNOW resistance? 4

5/19/2014 Wax, et al. • Non-cardiac cases with both ABP and NIBP. • Compared SBP, DBP, and MAP btwn technologies: – A-line alone vs A-line + cuff Randomized trials • This used to be the 2 nd joke of the talk 5

5/19/2014 Interesting review • Reviewed 2 trials and 1 meta-analysis (13 studies) – Target BP – Actual BP • Dissociation – BPs invariably higher than goal – Higher goal ranges permitted higher actual ranges:  pressors 6

5/19/2014 The NEJM study • Randomized to MAP 65 vs 85 (800 total) • Norepinephrine • Mortality • AKI/RRT, stratified by HTN Blood pressure • Necessary but not sufficient • Goals are nebulous • Supra-normal levels common, not helpful Hemodynamic • Blood pressure • CVP • Dynamic respiratory indices: – Pulse pressure/systolic pressure/perfusion index variation 7

5/19/2014 Concept: assumptions Normal CVP Normal CVP Optimal actin-myosin match Optimal actin-myosin match Adequate contractility Adequate contractility Adequate DO 2 Adequate DO 2 The data • Critical target in EGDT for sepsis • Incorporated into SSC guidelines 8

5/19/2014 Sepsis +  CVP = Death • Retrospective analysis of VASST trial – 778 pts w/ septic shock on NE • Divided into quartiles based on total fluid in at 12 hrs, 4 days Dry Quartile Wet Quartile 12 hours +0.7L +8.2L 4 days +1.6L +20.5L Boyd, JH, et al. 2011. CCM. 39(2) Sepsis +  CVP = Death • Outcomes: Quartile x 28 d mortality • Early (12 hrs) and Late (4 d) “dry-ness” saved lives: – HR 0.57 and 0.47, respectively Dry Quartile Wet Quartile Survival 12 hours 81% 58% 4 days 83% 65% Boyd, JH, et al. 2011. CCM. 39(2) Marik, PE, et al. 2008. Chest. 134(1) 9

5/19/2014 Fluid responsiveness and total blood volume • Volume responsiveness • Cardiac output before and after fluid challenge • 19 evaluated CVP and volume responsiveness Fluid responsiveness • Calculated a Receiver Operating Characteristic curve • Likelihood that at any given point (CVP level, score, etc) the true positives will exceed false positives. • Higher = better discrimination Volume responsiveness Marik, PE, et al. 2008. Chest. 134(1) 10

5/19/2014 Deja vu • 43 studies, half ICU • Same design – AUC btwn CVP and ∆ SV • Same pooled AUC – 0.56 • Same aggressive conclusion CVP • Necessary? • Certainly not sufficient • Potentially misleading Hemodynamic • Blood pressure • CVP/wedge • Dynamic respiratory indices: – Pulse pressure/systolic pressure/perfusion index variation 11

5/19/2014 The Principles  LV Preload  LV Preload  LV SV  LV SV Decreased RV SV Decreased RV SV   RV Preload RV Preload   RV Afterload RV Afterload Applies to lots of measures • Systolic pressure variation • Pulse pressure variation • Plethysmogram variation • Outcome is “fluid responsiveness” Variations on a theme… • A waveform… • A peak and trough… • And a proprietary algorithm: 12

5/19/2014 The data • Small studies • Mostly OR SVV, Vigileo PVI, Masimo 40% MORE fluid 1/3 LESS fluid Lower lactate Lower lactate Fewer “complications” • 29 studies, 685 patients – 9 ICU – 20 OR (15 in cardiac surgery) • All included correlation/ROC between SPV, PPV, or SVV and ∆ SVI/CI after a fluid challenge. Measure r AUC for ROC Threshold PPV 0.78 0.94 12.5% SVV 0.72 0.84 15.3% SPV 0.72 0.86 CVP 0.56 13

5/19/2014 ECOM ECOM • ETT-based electrodes • Current generated by flow in ascending aorta • Current + Nomogram = SV • SV  CO, SVV • R 2 = 0.63 Wallace, AW, et al. Under Review. Now, keep in mind… • Regular HR • Sedated, mechanically ventilated • Vt = 8 mL/kg • Pressors? 14

5/19/2014 PVI + NE = NEB Monnet, et al Monnet, et al Biais, et al Population Population 35 ICU patients on NE 35 ICU patients on NE 35 ICU patients on NE Gold Standard Gold Standard TD TD PPV > 13% Sensitivity FR Sensitivity FR 43 43 58 100 Specificity FR Specificity FR 90 90 61 72 0.93 AUC ROC AUC ROC 0.68 0.68 0.69 Hemodynamic goals • Numerous • State of the art: Dynamic indices – PPV – SPV – PVI – VTI and esophageal doppler • Necessary but not sufficient Metabolic • Mental status, urine output • Lactate • S(c)vO2 15

5/19/2014 Metabolic • Mental status, urine output • Lactate • S(c)vO2 Physical exam • Evidence of end-organ perfusion and function • Slow to change • Numerous confounders • Summarily dismissed Metabolic • Mental status, urine output • Lactate • S(c)vO2 16

5/19/2014 Lactate • The product of anaerobic respiration • Presence implies inadequate oxygen utilization, shock • Easily, quickly measured in arterial blood Lactate: the data Two trials: • JAMA: 300 patients, EGDT vs lactate clearance – Non-inferiority • AJRCCM: 348 patients, EGDT vs lactate clearance – Improved mortality (multivariate) – Less time on vent, in ICU 17

5/19/2014 How did they do it? Jones, et al (JAMA) Jansen et al (AJRCCM) Monitoring interval 2 2 Goal 10% clearance 20% clearance Fluid totals (L) Control: 4.3 Control: 2.2 Intervention: 4.5 ns Intervention: 2.7 * Outcome Non-inferiority to EGDT Decreased time on vent, in ICU The underpinnings… Metabolic • Mental status, urine output • Lactate • S(c)vO2 18

5/19/2014 How it’s used:  ScvO 2 attributed to:  Supply (cardiac output)  Demand (hypermetabolism) • In either case, treat by increasing DO 2 – Volume, inotropes, RBCs • But does it work? ScvO 2 • The cornerstone of Early Goal-Directed Therapy. • And we know that targeting SvO 2  mortality. – Septic, cardiogenic shock in humans, dogs – ScvO 2 = SvO 2 ? SvO 2 ScvO 2 19

5/19/2014 DOGS Changes in SvO2 and ScvO2 Humans w/ shock Humans w/ sepsis But does it work? • Conflicting data A-V(CO 2 ) • To the obscure… 20

5/19/2014 The premise: • Pts admitted to ICU w/ sepsis frequently have NL ScvO2 • Many – Still have evidence of hypoperfusion – Still die Lactate Lactate NL NL ScvO2 > ScvO2 > A-V(CO2) A-V(CO2) 70 70 < 6 < 6 Lactate Lactate EGDT EGDT High High ScvO2 < ScvO2 < A-V(CO2) A-V(CO2) 70 70 >6 >6 • 50 such patients – 26 Low gap – 24 High gap • The same on many measures: – SOFA, APACHE II, pressor use • Over 24 hrs, High gap exhibited hypoperfusion: – Persistent lactate – Higher SOFA 21

5/19/2014 • Higher mortality – 34% – 54% (ns) • No specific therapy for gap • Authors recommend augmenting CO Metabolic goals • Lactate and ScvO 2 – Base deficit? – A-V (CO 2 ) gradient? – A-V (CO 2 ) cer gradient? • Physiological rationale meets objective data. Does any of this… • Save lives? • Save money? • Actually work? 22

5/19/2014 Subgroup Mortality Complications Jadad high --- +++ Jadad low +++ +++ 1980s-1990s +++ +++ 2000s --- +++ 23

5/19/2014 24

5/19/2014 • Randomized CT surgery patients to – CVP and MAP – SVV, EDV index (PiCCO), EVLW Counted up outcomes • Less norepinephrine • Higher CI, S cv O2 • Fewer post-op complications • Left the ICU and hospital a full day sooner • No difference in fluid totals, RBCs, UOP – Timing of fluid: More in the OR. A great review… • Alphabet soup – SVV, PPV, SPV, PVI – LidCO, PiCCO, esCCO, ProAQT • Some require external calibration • Some require specific catheters in specific locations. 25