chest point/counterpoint editorials...1408 point/counterpoint editorials references 1 . shoemaker wc...

CHEST Point/Counterpoint Editorials

1406 Point/Counterpoint Editorials

Point: Should Lactate Clearance Be Substituted for Central Venous Oxygen Saturation as Goals of Early Severe Sepsis and Septic Shock Therapy? Yes Abbreviations: CVP 5 central venous pressure ; D o 2 5 systemic oxygen delivery ; EGDT 5 early goal-directed therapy ; GTH 5 global tissue hypoxia ; MAP 5 mean arterial pressure ; OER 5 systemic oxygen extraction ; PDH 5 pyruvate dehydrogenase ; Scv o 2 5 central venous oxygen saturation ; Sv o 2 5 mixed venous oxygen saturation ; o 2 5 oxygen consumption

Quantitative resuscitation in critically ill patients consists of structured cardiovascular interven-

tions, such as intravascular volume expansion and vasoactive agent support, to achieve explicit pre-defi ned physiologic parameters or goals. The con-cept of quantitative resuscitation (also referred to as hemodynamic optimization or goal-directed therapy) as a treatment strategy to improve clinical outcome was fi rst reported in high-risk surgery patients. 1 A recent meta-analysis of randomized clinical trials that compared quantitative resuscitation with stan-dard resuscitation in septic shock found that when therapy was initiated within 24 h of the onset of sep sis (six trials, 740 patients), resuscitation targeting specifi c physiologic end points improved mortality compared with standard resuscitation (39% vs 57%: OR, 0.50; 95% CI, 0.37-0.69). 2 In contrast, when therapy was initiated . 24 h after the onset of sep-sis (three trials, 261 patients), resuscitation target-ing specifi c physiologic end points did not improve mortality (64% vs 58% for standard resuscitation; OR, 1.16; 95% CI, 0.60-2.22). Although the data supporting the use of early quantitative resuscitation are robust, the optimal end points or goals of such therapy are controversial.

Currently, consensus guidelines recommend the use of central venous pressure (CVP), mean arterial pressure (MAP), urine output, and central venous oxygen saturation (Scv o 2 ) as resuscitation goals. 3 These recommendations are based largely on an ED-based clinical trial of quantitative resuscitation for septic

shock, an approach termed “early goal-directed therapy,” which was a single-center study published by Rivers et al 4 in 2001. In this trial, 263 patients with severe sepsis or septic shock were randomly assigned to therapy targeting an Scv o 2 of � 70% or to conven-tional therapy that did not target an Scv o 2 . In both groups, therapy targeted CVP, MAP, and urine out-put. Mortality was signifi cantly lower in the group that targeted an Scv o 2 of � 70% (31% vs 47%). Given that the only difference in the treatment proto-cols in this trial was the Scv o 2 target, the observed treatment effect appears to hinge on achieving this node of the algorithm. In contrast, earlier studies of critically ill patients that targeted mixed venous oxygen saturation (Sv o 2 ) of � 70% found no mor-tality benefi t. 5

Multiple studies have unfortunately documented important barriers to implementing and maintain-ing an ED-based quantitative resuscitation protocol for septic shock. 6 - 8 Among these, the use of a central venous catheter and the need for specialty equip-ment such as a continuous central venous oxygen spectrophotometer, and the training required for it, were major barriers that limited generalizability. To begin to address these barriers, the Lactate Assess-ment in the Treatment of Early Sepsis (LACTATES) randomized multicenter noninferiority trial, the larg-est ED-based early sepsis resuscitation trial com-pleted to date, was designed to compare the use of lactate clearance to Scv o 2 as the fi nal goal of early sepsis resuscitation. 9 In the study, enrolled patients were randomly assigned to one of two groups. Each group received structured quantitative resuscitation while in the ED. The Scv o 2 group (n 5 150) was resuscitated by sequentially providing the therapy needed to meet thresholds of CVP, followed by MAP, and then Scv o 2 of � 70%. The lactate clear-ance group (n 5 150) had similarly targeted thresh-olds in CVP and MAP, and then lactate clearance of � 10% or more. The study protocol was continued until all end points were achieved or for a maximum of 6 h. The published results of this study showed a 6% (95% CI, 2 3% to 14%) in-hospital mortality difference between the two study groups (17% in the lactate clearance group vs 23% in Scv o 2 group), con-fi rming the primary hypothesis of noninferiority.

© 2011 American College of Chest Physicians at Mount Sinai School of Medicine on December 9, 2011chestjournal.chestpubs.orgDownloaded from

http://chestjournal.chestpubs.org/

CHEST / 140 / 6 / DECEMBER, 2011 1407www.chestpubs.org

There are many evidence-based, data-driven, and log-ical arguments as to why lactate clearance monitor-ing is a superior therapeutic target to oxygen-derived variables such as Scv o 2 . First, the published exper-imental (randomized trial) evidence supporting the use of lactate clearance as a therapeutic target is more robust in terms of the number of multicenter studies. 9 , 10 Similar published experimental evidence supporting Sc vo 2 is derived only from single-center studies. 4 , 11 Furthermore, multicenter studies have failed to show the use of Sv o 2 as a resuscitation goal 5 ; however, unlike Scv o 2 or other oxygen-derived vari-ables, the ability to clear lactate has consistently pre-dicted better survival in published studies of sepsis resuscitation. 12 - 15

Second, elevated lactate levels refl ect the total picture of energy metabolism in the acutely stressed patient with sepsis. Elevated blood lactate has long been known to refl ect anaerobic metabolism from tissue hypoxia in critically ill patients. 16 However, besides these anaerobic processes, aerobic (meta-bolic) mechanisms that affect the host’s effi ciency of energy transfer contribute to lactate production in sepsis. Cytokine-mediated glucose uptake and catecholamine-stimulated Na-K pump overactivity can both result in increased pyruvate production that eventually will overwhelm the catalytic capacity of pyruvate dehydrogenase (PDH) and result in increased lactate because of either mass effect, sepsis-induced PDH dysfunction, or both. This mechanism may explain part of the lactate production observed from the lungs and WBC in response to the infl am-matory stress, rather than tissue hypoxia of sepsis. 17 Additionally, reduced lactate clearance may refl ect globally impaired metabolic function by the liver and kidney, both of which normally contribute to systemic lactate disposal through anaplerosis, a mechanism that carboxylates lactate and delivers it to the tricarboxylic acid cycle , independent of the action of PDH. 18 Recent studies have shown that early lactate clearance is associated with improve-ment in the biomarkers of infl ammation and organ dysfunction. 19 Thus, as opposed to Scv o 2 , which is a rudimentary indicator of only the balance between oxygen supply and demand, lactate clearance bio-logically refl ects more of the general homeostasis of the host and provides more meaningful data about the overall adequacy of the resuscitative processes.

Third, in some circumstances the use of Scv o 2 might erroneously lead a clinician to believe that the physi-ologic status of the patient has improved, when in fact it may not have improved. A recent multicenter study of 619 patients demonstrated that venous hyperoxia (Scv o 2 . 89%) is present in 36% of ED patients with septic shock and is associated with an increased risk of death, and, when adjusted for confounders, venous

hyperoxia was actually associated with a higher risk of death than venous hypoxia (Scv o 2 , 70%). 20 In this situation, high Scv o 2 values represent either an inability to exchange oxygen because of impaired fl ow in the small vessels from dysfunctional vascular autoregulatory mechanisms and functional shunt-ing of oxygen or the inability of cells to use the oxygen because of derangement of cellular respi-ration, so-called “cytopathic hypoxia.” 21 Although the Rivers et al 4 protocol focuses on the correction of a low Scv o 2 level signifying impairment in macrovas-cular oxygen delivery, the algorithm treats venous hyperoxia the same as normoxia (Scv o 2 70%-90%). The finding that a high Scv o 2 is associated with increased mortality reminds us that tissue dysoxia may occur despite adequate global oxygen delivery and that this situation is not identifi ed by the pres ence of normal venous oxygen levels. However, impaired oxygen transfer at any point from the lungs to the nicotinamide adenine dinucleotide dehydrogenase enzyme will cause lactic acidosis, and clearing lac-tate levels almost always signifi es improvement in host oxygen use. 6

Finally, a recently reported secondary analysis of the LACTATES study 9 reported no signifi cant con-cordance in achieving lactate clearance and Scv o 2 goals when measured simultaneously in the same subject, suggesting that these tests may be measuring and/or providing data about physiologically distinct processes. If lactate clearance was , 10%, the mor-tality was 40%, but if the Scv o 2 was , 70%, the mor-tality was 11% (proportion difference 29%; 95% CI, 6%-50%). 22

In conclusion, early sepsis resuscitation remains a dynamic topic of research interest, with many impor-tant questions that have yet to be answered. As sum-marized in this report, the best available evidence suggests that if a clinician has to choose a single goal of early sepsis resuscitation, lactate clearance, as opposed to Scv o 2 , is the more appropriate goal to choose.

Alan E. Jones , MD Jackson, MS

Affi liations: From the Department of Emergency Medicine, University of Mississippi Medical Center . Financial/nonfi nancial disclosures: The author has reported to CHEST the following confl icts of interest: Dr Jones has received funding from the National Institutes of Health to study lactate clearance in sepsis resuscitation. Dr Jones has never been assigned patents, nor has he received patent royalties, honoraria, consulting fees, or other monetary or nonmonetary payments at any time related to the use of lactate or lactate clearance . Correspondence to: Alan E. Jones, MD, Department of Emer-gency Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216; e-mail: [email protected] © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( http :// www . chestpubs . org / site / misc / reprints . xhtml ). DOI: 10.1378/chest.11-2560


mailto:[email protected]

http://www.chestpubs.org/site/misc/reprints.xhtml


www.chestpubs.org



References 1 . Shoemaker WC , Appel PL , Kram HB , Waxman K , Lee TS . Pro-

spective trial of supranormal values of survivors as therapeu-tic goals in high-risk surgical patients . Chest . 1988 ; 94 ( 6 ): 1176 - 1186 .

2 . Jones AE , Brown MD , Trzeciak S , et al ; Emergency Medi-cine Shock Research Network investigators . The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta-analysis . Crit Care Med . 2008 ; 36 ( 10 ): 2734 - 2739 .

3 . Dellinger RP , Levy MM , Carlet JM , et al ; International Sur-viving Sepsis Campaign Guidelines Committee ; American Association of Critical-Care Nurses ; American College of Chest Physicians ; American College of Emergency Physi-cians ; Canadian Critical Care Society ; European Society of Clinical Microbiology and Infectious Diseases ; European Society of Intensive Care Medicine ; European Respiratory Society ; International Sepsis Forum ; Japanese Association for Acute Medicine ; Japanese Society of Intensive Care Medi-cine ; Society of Critical Care Medicine ; Society of Hospital Medicine ; Surgical Infection Society ; World Federation of Societies of Intensive and Critical Care Medicine . Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008 . Crit Care Med . 2008 ; 36 ( 1 ): 296 - 327 .

4 . Rivers E , Nguyen B , Havstad S , et al ; Early Goal-Directed Therapy Collaborative Group . Early goal-directed therapy in the treatment of severe sepsis and septic shock . N Engl J Med . 2001 ; 345 ( 19 ): 1368 - 1377 .

5 . Gattinoni L , Brazzi L , Pelosi P , et al . A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Col-laborative Group . N Engl J Med . 1995 ; 333 ( 16 ): 1025 - 1032 .

6 . Carlbom DJ , Rubenfeld GD . Barriers to implementing protocol-based sepsis resuscitation in the emergency department—results of a national survey . Crit Care Med . 2007 ; 35 ( 11 ): 2525 - 2532 .

7 . Jones AE , Kline JA . Use of goal-directed therapy for severe sepsis and septic shock in academic emergency departments . Crit Care Med . 2005 ; 33 ( 8 ): 1888 - 1889 .

8 . Jones AE , Shapiro NI , Roshon M . Implementing early goal-directed therapy in the emergency setting: the challenges and experiences of translating research innovations into clin-ical reality in academic and community settings . Acad Emerg Med . 2007 ; 14 ( 11 ): 1072 - 1078 .

9 . Jones AE , Shapiro NI , Trzeciak S , Arnold RC , Claremont HA , Kline JA ; Emergency Medicine Shock Research Network (EMShockNet) Investigators . Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial . JAMA . 2010 ; 303 ( 8 ): 739 - 746 .

10 . Jansen TC , van Bommel J , Schoonderbeek FJ , et al ; LACTATE study group . Early lactate-guided therapy in intensive care unit patients: a multicenter, open-label, randomized controlled trial . Am J Respir Crit Care Med . 2010 ; 182 ( 6 ): 752 - 761 .

11 . Wang XC , Lü CJ , Gao FQ , Li XH , Yan WF , Ning Fw . Effi cacy of goal-directed therapy in the treatment of septic shock [in Chinese] . Zhongguo Wei Zhong Bing Ji Jiu Yi Xue . 2006 ; 18 ( 11 ): 661 - 664 .

12 . Arnold RC , Shapiro NI , Jones AE , et al ; on behalf of the Emer-gency Medicine Shock Research Network (EMShockNet) Investigators . Multi-center study of early lactate clearance as a determinant of survival in patients with presumed sepsis . Shock . 2008 ; 32 : 36 - 39 .

13 . Nguyen HB , Rivers EP , Knoblich BP , et al . Early lactate clearance is associated with improved outcome in severe sep-sis and septic shock . Crit Care Med . 2004 ; 32 ( 8 ): 1637 - 1642 .

14 . Bakker J , Coffernils M , Leon M , Gris P , Vincent J-L . Blood lactate levels are superior to oxygen-derived variables in pre-dicting outcome in human septic shock . Chest . 1991 ; 99 ( 4 ): 956 - 962 .

15 . Bakker J , Gris P , Coffernils M , Kahn RJ , Vincent J-L . Serial blood lactate levels can predict the development of multiple organ failure following septic shock . Am J Surg . 1996 ; 171 ( 2 ): 221 - 226 .

16 . Weil MH , Afi fi AA . Experimental and clinical studies on lactate and pyruvate as indicators of the severity of acute circulatory failure (shock) . Circulation . 1970 ; 41 ( 6 ): 989 - 1001 .

17 . De Backer D . Lactic acidosis . Intensive Care Med . 2003 ; 29 ( 5 ): 699 - 702 .

18 . Russell RR III , Taegtmeyer H . Changes in citric acid cycle fl ux and anaplerosis antedate the functional decline in isolated rat hearts utilizing acetoacetate . J Clin Invest . 1991 ; 87 ( 2 ): 384 - 390 .

19 . Nguyen HB , Loomba M , Yang JJ , et al . Early lactate clear-ance is associated with biomarkers of infl ammation, coagu-lation, apoptosis, organ dysfunction and mortality in severe sepsis and septic shock . J Infl amm (Lond) . 2010 ; 7 : 6 .

20 . Pope JV , Jones AE , Gaieski DF , Arnold RC , Trzeciak S , Shapiro NI ; Emergency Medicine Shock Research Network (EMShockNet) Investigators . Multicenter study of central venous oxygen saturation (ScvO 2 ) as a predictor of mor-tality in patients with sepsis . Ann Emerg Med . 2010 ; 55 ( 1 ): 40 - 46, e1 .

21 . Fink MP . Bench-to-bedside review: cytopathic hypoxia . Crit Care . 2002 ; 6 ( 6 ): 491 - 499 .

22 . Puskarich M , Trzeciak S , Shapiro N , Kline J , Jones AE . Con-cordance and prognostic value of central venous oxygen saturation and lactate clearance in emergency department patients with septic shock . Acad Emerg Med . 2011 ; 18 ( 5 ): S159 - S160 .

Counterpoint: Should Lactate Clearance Be Substituted for Central Venous Oxygen Saturation as Goals of Early Severe Sepsis and Septic Shock Therapy? No

In 2001, early goal-directed therapy (EGDT ) resulted in a 16% reduction in hospital mortality and,

post hoc, a higher lactate clearance in severe sepsis and septic shock. 1 Multiple studies have confi rmed the validity and generalizability of EGDT, resulting in its adoption into the Surviving Sepsis Campaign Guidelines. 2 , 3 Nguyen et al 4 , 5 examined early lactate clearance and found a signifi cant retrospective associa-tion with infl ammation, apoptosis, coagulation, organ dysfunction, and mortality. Following this rationale, Jones et al 6 modifi ed the EGDT protocol in 2010 using a noninferiority study design and concluded that lactate clearance is equivalent to central venous oxygen saturation (Scv o 2 ) in the management of indi-vidual patients.

Before applying the fi ndings of Jones et al 6 to one’s next patient, compare the baseline characteristics,




accompany venous hyperoxia. The phase of tissue dysoxia can be the result of microcirculatory dys-function causing maldistribution of blood fl ow or mitochondrial dysfunction with defects in substrate utilization. In this phase, improvement in D o 2 may not result in improvement in o 2 .

Sepsis may consist of four hemodynamic phases where a decreased Scv o 2 /Sv o 2 always precedes the appearance of lactate, making them complementary and nonexclusive end points, ( Fig 1, Table 1 ). These hemodynamic phases are not always distinct and may overlap depending on the timing and quality of the resuscitation. By characterizing these phases in hemodynamic outcome studies, future trials can be conducted with the appropriate research design and interpreted with clarity, facilitating generalizability and external validation in clinical management. 15

Lactate Kinetics Are Complex and Limit the Interpretation of Lactate Levels and Lactate Clearance in the

Individual Patient

Lactate elevation may indicate stress-induced upreg-ulation in epinephrine-stimulated sodium-potassium adenosine triphosphatase activity in skeletal muscle and inhibition of pyruvate metabolism rather than, or in addition to, the traditionally implicated cellular hypoxia. Other confounding infl uences may include exogenous lactate sources (Ringers lactate or packed RBC transfusions), lactate shuttles and transport, delayed washout from underperfused tissue, variable lactate clearance by a number of organs, and dilution (large-volume resuscitations) ( Fig 2 ). These inter-actions are not in a steady state and depend on the pathophysiology, timing, and quality of the resuscita-tion in the individual case. 16

Normal lactate levels occur in up to 45% of cases of septic shock, and although there is signifi cant vari-ability, the associated mortality can be up to 52%. 9 , 17 - 19 In fact, many patients develop multisystem organ fail-ure and die without ever having increased lactate levels. 9 Thus, lactate has limitations as a tool for risk stratifi cation and as a guide for resuscitation in individual patients. In the Jones et al 6 study, the lactate clearance goal was at least 10% at � 2 h or normality of both initial and subsequent lactates. Nguyen et al, 20 however , found an optimal lactate clearance cutoff of , 10% after 6 h of interven-tion to have a sensitivity of 44.7%, specifi city of 84.4%, and accuracy of only 67.6% for predicting in-hospital mortality. Additionally, lactate clearance was less predictive of outcome in septic shock, the predominant feature of the patients in the Jones et al 6 study. Because of the variable expression of lactate, its complicated kinetics, and the limited accuracy of

early hemodynamic patterns, and therapeutic inter-ventions between those of Jones et al 6 and the EGDT study. 1 Further, review the complexities of lactate kinetics and the weaknesses of a noninferiority study design. 7 Based on these facts, it is clear that lactate clearance and Scv o 2 are not equivalent, but comple-mentary goals for the individual patient.

The Hemodynamic Phases of Severe Sepsis and Septic Shock

The early stages of sepsis are accompanied by cir-culatory insuffi ciency that results from hypovolemia, vasomotor dysfunction, myocardial depression, and increased metabolic demands. In the systemic oxygen delivery (D o 2 )-dependent (hypodynamic) phase, a decrease in D o 2 results in a decrease in Scv o 2 /mixed venous oxygen saturation (Sv o 2 ) and usually an increase in systemic oxygen extraction (OER) or 1 2 Scv o 2 /Sv o 2 ( Fig 1, 8 Table 1 ). When the limits of the OER (anaerobic threshold) are reached, lactate is produced, signifying the development of global tissue hypoxia (GTH). There is signifi cant individual variation in the anaer-obic threshold leading to variable lactate production. 9 This gives rise to why some patients may require nor-mal or elevated D o 2 in order to resolve GTH (decreased Scv o 2 /Sv o 2 and increased lactate) ( Fig 1, Table 1 ). GTH is associated with increased morbidity and mortality if not adequately treated. 10 , 11 Because GTH can occur with normal vital signs, it has been termed “cryptic shock.” 12 GTH or cardiovascular insuffi ciency is a signifi cant part of the natural history of sepsis and responsible for the sudden cardiopulmo-nary deterioration seen in 12% to 21% of patients. 13 , 14 EGDT is associated with a 50% reduction in this adverse event, an issue not addressed by Jones et al. 6

With adequate volume therapy and myocardial reserve, a hyperdynamic or compensated phase fol-lows. During this compensated phase, D o 2 is in the normal or elevated range, systemic oxygen consump-tion ( o 2 ) is increased, and vascular resistance is gen-erally decreased. In contrast to the hypodynamic phase (patients in the Rivers et al 1 study), Jones et al 6 enrolled patients in this phase with a lower systolic BP, normal central venous pressure (CVP), normal Scv o 2 , lower lactate levels, and triple the frequency of vasopressor dependence ( Fig 1 ; Tables 1, 2 ). These patients also had corresponding Simplifi ed Acute Physiology Score II scores and predicted mortality that was nearly 14% lower than that in patients receiv-ing EGDT (34.8% vs 48.4%) and other studies. 2

Pathological D o 2 dependency is a result of a progres-sive impairment of OER, which is accompanied by a markedly increased Scv o 2 /Sv o 2 (venous hyperoxia) and a hyperdynamic circulation. When D o 2 is insuf-fi cient, o 2 decreases, and increased lactate levels


www.chestpubs.org



Table 1 —Hemodynamic Phases of Sepsis

Stage Hemodynamic Picture SBP CVP Treatment and Comments

B Hypovolemia Variable ↓ Volume Myocardial suppression Variable ↑ Correct anemia, inotropic therapy A Resuscitated, compensated, and vasodilatory Variable Normal Vasopressors, low-dose corticosteroids C Supranormal D o 2 dependency Variable ↑ to normal Increased o 2 after augmentation of D o 2 D Impairment of tissue O 2 utilization Variable Normal r-APC Decreased o 2 Variable Normal Resuscitated

CVP 5 central venous pressure; D o 2 5 systemic oxygen delivery; r-APC 5 recombinant activated protein C; SBP 5 systolic BP; SvO 2 5 mixed venous oxygen saturation; o 2 5 oxygen consumption.

lactate clearance, Nguyen et al 20 did not recommend lactate clearance as a sole therapeutic end point. Serum lactate levels may rise or fl uctuate during therapy. Of patients with increased initial lactate levels, 41% have delayed peak values (20 � 12 h) after the initial presentation. 11 , 17 Of patients with normal initial lactates, 15% will later demonstrate elevations. These patients have abnormal Scv o 2 (66.7% � 8.6%) at baseline compared with their coun-terparts with normal levels. 17 Lactate levels over time can increase (negative clearance), stay the same, or decrease (positive clearance) after intervention ( Fig 2 ). Not only is the direction of clearance impor tant but also the magnitude of change. There are signifi cantly different clinical and outcome implications in patients whose lactate levels decrease from 10 to 9 mmol/L vs 4 to 3.6 mmol/L. Although both represent clear-ance of 10%, the implications for illness severity and prognostic signifi cance are much different.

What if the Patient Requires More Than Fluid and Vasopressors and the Lactate

Is Still High?

Optimization of preload (CVP) and afterload (mean arterial pressure) were addressed by Jones et al 6 ; however, the remaining components of EGDT, includ-ing optimizing D o 2 (oxygen carrying capacity [sup-plemental oxygen and hemoglobin], cardiac output) and decreasing o 2 (mechanical ventilation and sedation) to prevent delayed cardiopulmonary com-plications, were not elicited or examined. 15 Over the past decade, numerous studies have validated the clinical utility of Scv o 2 in recognizing supply depen-dency, need for a transfusion, detection of myocardial dysfunction, response to oxygen and mechanical ventilation, early cardiopulmonary complications, and overall infl uence on mortality. To establish non-inferiority, lactate clearance has to be appropriately

Figure 1. The hemodynamic phases of sepsis. DO 2 5 systemic oxygen delivery; NI 5 normal; SvO 2 5 mixed venous oxygen saturation; VO 2 5 oxygen consumption. Reprinted with permission from Kruse. 8




Table 2 —Comparison of Clinical Variables Between Jones et al 6 and Rivers et al 1

Jones et al 6 Rivers et al 1

Lactate Clearance Scv o 2 Guided Standard Therapy EGDT

Comorbidities Congestive heart failure … … 30.2 36.7 Coronary artery disease … … 23.5 26.5 Liver disease … … 23.5 23.1 Alcohol use … … 38.7 38.5 Baseline hemodynamics Lactate, a mmol/L 3.9 � 3.1 4.2 � 3.1 6.9 � 4.5 7.7 � 4.7 CVP, a mm Hg 11 � 6.5 11 � 6.2 6.1 � 7.7 5.3 � 9.3 Systolic BP, a mm Hg 91 � 24.6 92 � 21.0 109 � 34 106 � 36 Scv o 2 a … 74 � 12.3 49.2 � 13.3 48.6 � 11.2 Treatments (0-6-h ranges) RBC transfusions 7 3 18.5 64.1 P Value .2 … , .001 … Vasopressors a 72 75 30.3 27.4 Inotropes 3 5 0.8 13.7 P Value .57 … , .001 … Mechanical ventilation a 27 26 53.8 53.0 6 h hemodynamics Lactate, mmol/L … … 4.9 � 4.7 4.3 � 4.2 P Value .01 Lactate clearance (0-6 h) … … 29 44.2 P Value .01 CVP, mm Hg … … 11.8 � 6.8 13.8 � 4.4 P Value .007 Scv o 2 … … 66.0 � 15.5 77.3 � 10.0 P Value , .001 Sudden cardiopulmonary collapse … … 21.0 10.3 P Value .02 SAPS II scores (baseline) a 44.8 � 18.4 44.1 � 17.3 48.8 � 11.1 51.2 � 11.1 Predicted mortality (approximate) 34.8 32.8 48.4 48.7 Actual in-hospital mortality 23 17 46.5 30.5 P Value NS … .009 …

Data are presented as mean � SD or %. EGDT 5 early goal-directed therapy; NS 5 not signifi cant; SAPS 5 Simplifi ed Acute Physiology Score; Scv o 2 5 central venous oxygen saturation. See Table 1 legend for expansion of other abbreviations. a No statistical signifi cance.

examined in these scenarios in order to be gener-alizable to all hemodynamic phases of sepsis and these facets of care. 7 The discrepancy between Scv o 2 -triggered interventions in the Rivers et al 1 study vs the 30 interventions (10% of patients) guided by lactate clearance refl ects signifi cant differences in hemodynamic phases, patient populations, and frequency and timing of interventions ( Fig 2 ). This undermines the conclusion of equivalency from a noninferiority research design. 7 Patients more likely to require inotropes (congestive heart failure or coronary artery disease) or patients with reduced lactate clearance (liver failure) were not described by Jones et al 6 ( Table 2 ). This lower number of inter-ventions refl ects a lower illness sever ity compared with other studies, 2 the possibility of poor compliance to the protocol, or a study design that is not equiv-alent to EGDT. The threefold greater use of vaso-pressors by Jones et al 6 may have resulted in higher lactate levels (catecholamines), CVP (increased after-

load and venous tone), and Scv o 2 (decreased OER). As a result, triggers for more fl uid administration, RBC transfusion, inotropes, and mechanical ventila-tion may have been obscured by catecholamines. In this vasodilatory phase of sepsis, one would expect a higher use of corticosteroids 3 ; however, they were only used in 37% and 35% of eligible patients in the lactate clearance and Scv o 2 groups, respectively.

Real-World Clinical Practice

Central venous catheterization is recommended for patients with septic shock, and this was indeed the practice in the Jones et al 6 study. However, this study often is misinterpreted to imply that lactate clearance precludes the need for central venous catheterization altogether. This could result in a delay in a safer route for administration of vasopressors and achievement of EGDT goals within 6 h. The Surviving Sepsis Campaign recommendations include intermittent


www.chestpubs.org



Figure 2. The kinetics, diagnostic, and therapeutic clinical scenarios of lactate. EGDT 5 early goal-directed therapy; PRBC 5 packed RBC; ScvO 2 5 central venous oxygen saturation. See Figure 1 legend for expansion of other abbreviation.

or continuous Scv o 2 sampling. 3 It is a simple matter to add intermittent Scv o 2 to lactate measurements in the absence of continuous monitoring. Bundle compliance and socioeconomic costs improve signifi -cantly with continuous monitoring. 21

Conclusions

Scv o 2 provides immediate feedback to the o 2 /D o 2 relationship but requires interpretation that depends on the phase of sepsis. Lactate is a delayed indicator of tissue perfusion and is subject to complex kinetics that are never clear in the individual case. Lactate levels may be normal or fl uctuate, leading to inappro-priate risk stratifi cation and therapy. Lactate clear-ance and Scv o 2 , therefore, are complementary and not mutually exclusive end points.

Emanuel P. Rivers , MD, MPH, FCCP Detroit, MI

Ronald Elkin , MD San Francisco, CA

Chad M. Cannon , MD Kansas City, KS

Affi liations: From the Department of Emergency Medicine and Surgery (Dr Rivers), Henry Ford Hospital, Wayne State University; Department of Medicine (Dr Elkin), Pulmonary and Critical Care Medicine, California Pacifi c Medical Center; and Department of Emergency Medicine (Dr Cannon), University of Kansas Hospital. Financial/nonfi nancial disclosures: The authors have reported to CHEST the following confl icts of interest: In the past 3 years, Dr Rivers has received funding from the National Institutes of Health, Aggennix AG, and Alere Corporation. He has been a one-time consultant for Aggennix AG; Eisai Co, Ltd; Idaho Technologies Inc; AstraZeneca; Massimo; and Sangard. He is a consultant to the Institute of Medicine, National Academies. The EGDT study was performed without external industry support or funding of any kind. Any intellectual properties associated with Dr Rivers’ research are exclusively owned by Henry Ford Hospital. Dr Rivers holds no past or present intellectual proper-ties and has never received royalties or stock interest related to technologies in EGDT research and practice. Dr Elkin has received funding from the Gordon and Betty Moore Foundation, been a one-time consultant for Eisai Co, Ltd, and participated on the speaker’s bureau for Edwards Lifesciences LLC on three occasions. Dr Cannon has been a one-time consultant for Aggennix AG and Eisai Co, Ltd. Correspondence to: Emanuel P. Rivers, MD, MPH, FCCP, Department of Emergency Medicine, Wayne State University, 270 Clara Ford Pavilion, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202; e-mail: [email protected] © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( http :// www . chestpubs . org / site / misc / reprints . xhtml ). DOI: 10.1378/chest.11-2563







Rebuttal From Dr Jones

In their counterpoint, Dr Rivers and colleagues 1 present the theoretical view that patients with

septic shock present in very distinct “hemodynamic phases” and that Jones et al 2 enrolled patients in a different phase of septic shock than did Rivers et al. 3 According to their theory, decreased central venous oxygen saturation (Scv o 2 ) always precedes the appear-ance of lactate—a concept not observed in my clin-ical practice. Clinicians who routinely care for the critically ill encounter patients with elevated lac-tate and normal Scv o 2 . Furthermore, as shown in Table 1 , 2 - 10 the hemodynamic patterns of the sub-jects enrolled by Rivers et al 3 are markedly different from any other reported populations of patients with septic shock treated with quantitative resuscitation. The study by Rivers et al 3 patients had much higher lactate, much lower Scv o 2 , and much higher mor-tality than described elsewhere. Possible explana-tions for this discrepancy may include that patients with septic shock in Detroit between 1997 and 2000 were markedly different than any other septic shock population reported in the world’s literature and/or that systematic selection bias was a signifi cant problem in the their study. In such a scenario, their results have questionable external validity. Support-ing either of these assertions is the fact that mor-tality in the control group of the Rivers et al 3 study was 20% higher than any septic shock mortality reported in the recent literature, leaving one to question exactly what care they received. 3 Little evidence supports the contention that Jones et al 2 enrolled patients in a different phase of septic shock

References 1 . Rivers E , Nguyen B , Havstad S , et al ; Early Goal-Directed

Therapy Collaborative Group . Early goal-directed therapy in the treatment of severe sepsis and septic shock . N Engl J Med . 2001 ; 345 ( 19 ): 1368 - 1377 .

2 . Rivers EP . Point: adherence to early goal-directed therapy: does it really matter? Yes. After a decade, the scientifi c proof speaks for itself . Chest . 2010 ; 138 ( 3 ): 476 - 485 .

3 . Dellinger RP , Levy MM , Carlet JM , et al ; International Surviving Sepsis Campaign Guidelines Committee ; American Association of Critical-Care Nurses ; American College of Chest Physicians ; American College of Emergency Physicians ; Canadian Critical Care Society ; European Society of Clinical Microbiology and Infectious Diseases ; European Society of Intensive Care Medicine ; European Respiratory Society ; Inter-national Sepsis Forum ; Japanese Association for Acute Med-icine ; Japanese Society of Intensive Care Medicine ; Society of Critical Care Medicine ; Society of Hospital Medicine ; Surgical Infection Society ; World Federation of Societies of Intensive and Critical Care Medicine . Surviving Sepsis Cam-paign: international guidelines for management of severe sepsis and septic shock: 2008 . Crit Care Med . 2008 ; 36 ( 1 ): 296 - 327 .

4 . Nguyen HB , Rivers EP , Knoblich BP , et al . Early lactate clearance is associated with improved outcome in severe sep-sis and septic shock . Crit Care Med . 2004 ; 32 ( 8 ): 1637 - 1642 .

5 . Nguyen HB , Loomba M , Yang JJ , et al . Early lactate clear-ance is associated with biomarkers of infl ammation, coagu-lation, apoptosis, organ dysfunction and mortality in severe sepsis and septic shock . J Infl amm (Lond) . 2010 ; 7 : 6 .


7 . Piaggio G , Elbourne DR , Altman DG , Pocock SJ , Evans SJ ; CONSORT Group . Reporting of noninferiority and equiv-alence randomized trials: an extension of the CONSORT statement . JAMA . 2006 ; 295 ( 10 ): 1152 - 1160 .

8 . Kruse J . Lactic acidosis, part 1: understanding pathogenesis and causes. As mechanisms are elucidated, some assumptions are being revised. J Crit Illness . 1999 ; 14 ( 8 ): 456 - 466 .

9 . Levraut J , Ichai C , Petit I , Ciebiera JP , Perus O , Grimaud D . Low exogenous lactate clearance as an early predictor of mortality in normolactatemic critically ill septic patients . Crit Care Med . 2003 ; 31 ( 3 ): 705 - 710 .

10 . Astiz ME , Rackow EC , Kaufman B , Falk JL , Weil MH . Relationship of oxygen delivery and mixed venous oxygena-tion to lactic acidosis in patients with sepsis and acute myo-cardial infarction . Crit Care Med . 1988 ; 16 ( 7 ): 655 - 658 .

11 . Rivers EP , Kruse JA , Jacobsen G , et al . The infl uence of early hemodynamic optimization on biomarker patterns of severe sepsis and septic shock . Crit Care Med . 2007 ; 35 ( 9 ): 2016 - 2024 .

12 . Rady MY , Rivers EP , Nowak RM . Resuscitation of the crit-ically ill in the ED: responses of blood pressure, heart rate, shock index, central venous oxygen saturation, and lactate . Am J Emerg Med . 1996 ; 14 ( 2 ): 218 - 225 .

13 . Brun-Buisson C , Doyon F , Carlet J , et al ; French ICU Group for Severe Sepsis . Incidence, risk factors, and out-come of severe sepsis and septic shock in adults. A multi-center prospective study in intensive care units . JAMA . 1995 ; 274 ( 12 ): 968 - 974 .

14 . Carr GE , Edelson DP , Yuen TC , et al . In-hospital cardiac arrest among patients with coexisting pneumonia: a report from the American Heart Association’s Get With The Guidelines–Resuscitation Program . Am J Respir Crit Care Med . 2011 ; 183 : A6339 .

15 . Sevransky JE , Nour S , Susla GM , Needham DM , Hollenberg S , Pronovost P . Hemodynamic goals in randomized clinical trials in patients with sepsis: a systematic review of the literature . Crit Care . 2007 ; 11 ( 3 ): R67 .

16 . Mizock BA . The hepatosplanchnic area and hyperlactatemia: A tale of two lactates . Crit Care Med . 2001 ; 29 ( 2 ): 447 - 449 .

17 . Hernandez G , Castro R , Romero C , et al . Persistent sepsis-induced hypotension without hyperlactatemia: is it really sep-tic shock? J Crit Care . 2010 ; 26 ( 4 ): 435.e9 - e14 .

18 . Dugas A , Mackenhauer J , Joyce N , et al . Prevalence and char-acteristics of non-lactate and lactate expressors in septic shock [abstract] . Crit Care Med . 2009 ; 37:A227 .

19 . Cannon CM ; Multicenter Severe Sepsis and Septic Shock Collaborative Group . The GENESIS Project (GENeralization of Early Sepsis InterventionS): a multicenter quality improve-ment collaborative . Acad Emerg Med . 2010 ; 17 ( 11 ): 1258 .

20 . Nguyen HB , Rivers EP , Havstad S , et al . Critical care in the emergency department: a physiologic assessment and out-come evaluation . Acad Emerg Med . 2000 ; 7 ( 12 ): 1354 - 1361 .

21 . Ising P , Smith TW , Simpson SQ . Effect of intermittent vs con-tinuous ScvO2 monitoring on sepsis bundle compliance and mortality [abstract] . Chest . 2009 ; 136 ( 4 ): 21S .


www.chestpubs.org



Table 1 —Summary of Initial Clinical Characteristics and Mortality Rates of Recent Studies of Quantitative

Resuscitation for Septic Shock

Study No. Lactate, mmol/L Scv o 2 ,% Mortality, %

Rivers et al 3 EGDT group 130 7.7 48.6 30.5 Control group 133 6.9 49.2 46.5 Jones et al 2 Scv o 2 group 150 4.2 74 23 LC group 150 3.9 … 17 Shapiro et al 4 116 4.4 72 20 Jones et al 5 77 3.6 69-81 18 Nguyen et al 6 197 3.4-4.4 62-69 … Kortgen et al 7 30 2.6 … 27 van Beest et al 8 125 2.7 74 26 Peake et al 9 324 6.9 a … 23 Xiao-Zhi et al 10 16 5.3 66 18

The most abnormal values in each column are present in bold and italic text. EGDT 5 early goal-directed therapy; LC 5 lactate clearance; Scv o 2 5 central venous oxygen saturation. a Only 98 had lactate measured.

than did Rivers et al, 3 because the population enrolled by Jones et al 2 used inclusion cri teria identical to the Rivers et al 3 study. Furthermore, the population in the Jones et al 3 study appears to be an accurate con-temporaneous population from three US EDs; it has characteristics, including mortality rates, nearly iden-tical to those described from other studies ( Table 1 ).

Rivers et al 3 have raised the issue of potential methodologic limitations. They state that one must consider “weaknesses of the non inferiority study design” in the Jones et al 2 study. However, our study did not show equivalence (not worse than and not better than) but rather noninferiority (not worse than and maybe better than). This study design is the only design that could test the hypothesis at hand and, furthermore, the methodologic quality of a noninferiority and superiority randomized con-trolled trial is the same if properly conducted. So it is unclear exactly what “weaknesses” in our study are being criticized. On the other hand, there appear to be real methodologic concerns in the Rivers et al 3 trial. They failed to follow recommendations by not reporting whether any patients were randomized but not analyzed or performing an intention-to-treat analysis that could easily change the statistically sig-nifi cant results of the trial.

Finally, the counterpoint 1 states that 45% of patients with septic shock present with a normal lactate and suggest that because of this, lactate is an incon-sistent guide of resuscitation. 1 In fact, . 65% of patients in contemporaneous studies present with an Scv o 2 . 70%, and using their argument, these data make Scv o 2 an even more inconsistent resuscitative guide, particularly given that there is no therapy for markedly elevated and pathologic Scv o 2 (ie, . 90%). 2 , 5

Because data from an experimental clinical trial are the only way to scientifi cally deduce the clinical effi -cacy of lactate clearance vs Scv o 2 and because data from a large multicenter clinical trial demonstrated that lactate clearance is not inferior to Scv o 2 as an end point of early sepsis resuscitation, as described herein, lactate clearance has principles that may make it the more appropriate end point to choose.

Alan E. Jones , MD Jackson, MS

Affi liations: From the Department of Emergency Medicine, University of Mississippi Medical Center . Financial/nonfi nancial disclosures: The author has reported to CHEST the following confl icts of interest: Dr Jones has received funding from the National Institutes of Health to study lactate clearance in sepsis resuscitation. Dr Jones has never been assigned patents, received patent royalties, honoraria, consulting fees, or other monetary or nonmonetary payments at any time related to the use of lactate or lactate clearance . Correspondence to: Alan E. Jones, MD, Department of Emer-gency Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216; e-mail: [email protected] © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( http :// www . chestpubs . org / site / misc / reprints . xhtml ). DOI: 10.1378/chest.11-2597

References 1 . Rivers EP , Elkin R , Cannon CM . Counterpoint: should lac-

tate clearance be substituted for central venous oxygen satura-tion as goals of early severe sepsis and septic shock therapy? No . Chest . 2011 ; 140 ( 6 ): 1408 - 1413 .


3 . Rivers E , Nguyen B , Havstad S , et al ; Early Goal-Directed Therapy Collaborative Group . Early goal-directed therapy in the treatment of severe sepsis and septic shock . N Engl J Med . 2001 ; 345 ( 19 ): 1368 - 1377 .

4 . Shapiro NI , Howell MD , Talmor D , et al . Implementation and outcomes of the Multiple Urgent Sepsis Therapies (MUST) protocol . Crit Care Med . 2006 ; 34 ( 4 ): 1025 - 1032 .

5 . Jones AE , Focht A , Horton JM , Kline JA . Prospective exter-nal validation of the clinical effectiveness of an emergency department-based early goal-directed therapy protocol for severe sepsis and septic shock . Chest . 2007 ; 132 ( 2 ): 425 - 432 .

6 . Nguyen HB , Corbett SW , Steele R , et al . Implementation of a bundle of quality indicators for the early management of severe sepsis and septic shock is associated with decreased mortality . Crit Care Med . 2007 ; 35 ( 4 ): 1105 - 1112 .

7 . Kortgen A , Niederprüm P , Bauer M . Implementation of an evidence-based “standard operating procedure” and outcome in septic shock . Crit Care Med . 2006 ; 34 ( 4 ): 943 - 949 .

8 . van Beest PA , Hofstra JJ , Schultz MJ , Boerma EC , Spronk PE , Kuiper MA . The incidence of low venous oxygen satura tion on admission to the intensive care unit: a multi-center observa-tional study in The Netherlands . Crit Care . 2008 ; 12 ( 2 ): R33 .

9 . Peake SL , Bailey M , Bellomo R , et al ; ARISE Investigators, for the Australian and New Zealand Intensive Care Society Clinical Trials Group . Australasian resuscitation of sepsis evaluation (ARISE): a multi-centre, prospective, inception cohort study . Resuscitation . 2009 ; 80 ( 7 ): 811 - 818 .







Rebuttal From Dr Rivers et al

An End Point Must Be Consistently Present to Be Generalized

Levraut et al 1 stated, “It is common knowledge that many septic patients develop multiple organ

failure and die despite normal blood lactate levels.” Below the critical systemic oxygen delivery, central venous oxygen saturation (Scv o 2 ) decreases; however, lactate level elevation may not occur. More impor-tantly, an Scv o 2 � 70% is not the only goal, as achieve-ment of all early goal-directed therapy (EGDT) goals actually resulted in an Scv o 2 � 77.3%. According to Dr Jones, 2 a study by Gattinoni et al 3 did not show an outcome benefi t of reaching a mixed venous oxygen saturation of 70% up to 48 h after ICU admission. However, Chamberlain et al 4 found in a meta-analysis that patients resuscitated to this end point within a more reasonable 6 h were twice as likely to survive than those without it.

A Repeat Look at a Previously Quoted Study

Dr Jones 2 cites the study by Jansen et al 5 as sup-portive of lactate clearance, but not all patients in that study were septic. Furthermore, the reduction of lac-tate was no faster when the control group therapy was compared with the lactate-guided aggressive resus-citation group. In fact, these authors concluded that “this observation might actually argue against lactate level as a target of hemodynamic therapy. However, given that Scv o 2 monitoring was mandatory in the lactate group and control group, we cannot exclude the possibility that this had an impact on the observed outcome difference.” 5

Does Noninferiority Mean Equivalency? Be Careful What You Read

Noninferiority is a double negative that may confuse clinicians because of the complexity of study design. Noninferiority trials are contro versial and diffi cult to design, conduct, analyze, and interpret for trialists, clinicians, reviewers, and editors. 6 The low number of interventions observed by Jones et al 7 bias toward the conclusion of noninferiority. Thus, for appropriate interpretation, one must be aware of and apply the CONSORT (Consolidated Standards of Reporting Trials) recommendations on noninferiority and equiva-lence trials ( Table 1 ). 6 - 10

Responding to Perceived Barriers

Barriers specifi ed by Dr Jones 2 are unacceptable as excuses for our failure to save lives. We do not avoid complex interventions for trauma, stroke, or myocar-dial infarction. Severe sepsis carries a mortality risk far in excess of these acknowledged emergencies. Surely, placement of central lines, as well as contin-uous or intermittent venous saturation measurement, should be well within the capabilities of competent emergency and critical care practitioners.

Conclusion

Today’s clinical tools for assessing tissue perfusion, including Scv o 2 and lactate level, have benefi ts and limitations. Scv o 2 has a half-life of seconds, providing value as an early goal of resuscitation with interpre-tation potentially confounded by changes in systemic oxygen delivery, tissue extraction, and distribution of blood fl ow at both the macrocirculatory and microcir-culatory levels. Serum lactate levels may remain nor-mal before and throughout resuscitation or fl uctuate due to the complexities of lactate kinetics, causing one to question the clinical usefulness of lactate clearance. Moreover, a 10% drop in serum lactate level has different implications if the initial value is 12 mmol/L rather than 4 mmol/L . The concept of lac-tate clearance as the single goal of resuscitation is, therefore, fl awed and potentially dangerous. Today’s prudent clinician will use both normalization of Scv o 2 and lactate levels to guide resuscitation rather than rely on one parameter alone.

Emanuel P. Rivers , MD, MPH, FCCP Detroit, MI

Ronald Elkin , MD San Francisco, CA

Chad M. Cannon , MD Kansas City, KS

Affi liations: From the Department of Emergency Medicine and Surgery (Dr Rivers), Henry Ford Hospital, Wayne State University; Department of Medicine (Dr Elkin), Pulmonary and Critical Care Medicine, California Pacifi c Medical Center; and Department of Emergency Medicine (Dr Cannon), University of Kansas Hospital. Financial/nonfi nancial disclosures: The authors have reported to CHEST the following confl icts of interest: In the past 3 years, Dr Rivers has received funding from the National Institutes of Health, Aggennix AG, and Alere Corporation. He has been a one-time consultant for Aggennix AG; Eisai Co, Ltd; Idaho Technologies Inc; AstraZeneca; Massimo; and Sangard. He is a consultant to the Institute of Medicine, National Academies. The Early Goal-Directed Therapy (EGDT) study was performed without external industry support or funding of any kind. Any intellectual properties associated with Dr Rivers’ research are exclusively owned by Henry Ford Hospital. Dr Rivers holds no past or present intellectual properties and has never received royalties or stock interest related to technologies in EGDT research and practice. Dr Elkin has received funding from the Gordon and Betty Moore Foundation, has been a one-time consultant for Eisai Co, Ltd, and participated on the speaker’s bureau for Edwards Lifesciences LLC on three occasions. Dr Cannon has been a one-time consultant for Aggennix AG and Eisai Co, Ltd.

10 . Wang XZ , Lü CJ , Gao FQ Li XH , Yan WF , Ning FY . Effi cacy of goal-directed therapy in the treatment of septic shock [in Chinese] . Zhongguo Wei Zhong Bing Ji Jiu Yi Xue . 2006 ; 18 ( 11 ): 661 - 664 .


www.chestpubs.org



(Con

tinu

ed)

Tabl

e 1 —

CO

NSO

RT

Che

ckli

st f

or R

epor

ting

Non

infe

rior

ity

or E

quiv

alen

ce T

rial

s in

Riv

ers

et a

l 8 a

nd J

ones

et

al 7

C

omm

enta

ry

Topi

cR

equi

rem

ents

and

Con

side

ratio

nsJo

nes

et a

l 7,10

R

iver

s et

al 8

Bac

kgro

und

Scie

ntifi

c ba

ckgr

ound

and

exp

lana

tion

of

ra

tiona

le, i

nclu

ding

the

ratio

nale

for

usin

g a

noni

nfer

iori

ty o

r eq

uiva

lenc

e de

sign

.

Inte

rmitt

ent o

r co

ntin

uous

Scv

o 2 a

s pa

rt o

f EG

DT

and

reco

mm

ende

d by

the

SSC

9 is

not g

ener

aliz

able

bec

ause

of

the

follo

win

g is

sues

:

•

The

nee

d fo

r sp

ecia

lty e

quip

men

t and

trai

ning

, su

ch a

s co

ntin

uous

Scv

o 2 s

pect

roph

otom

eter

an

d ca

thet

er.

• T

his

tech

nolo

gy is

des

crib

ed a

s “t

echn

ical

ly d

iffi c

ult,

unav

aila

ble

in m

any

tert

iary

car

e E

Ds,

too

com

plex

, co

stly

and

take

s tim

e aw

ay fr

om th

e pa

tient

.”

•

Scv o

2 is

a “c

ontr

over

sial

” m

etho

d of

det

erm

inin

g tis

sue

oxyg

en d

eliv

ery.

Ass

umpt

ion,

LC

is e

quiv

alen

t to

Scv o

2 :

•

Incr

ease

d la

ctat

e le

vels

are

ubi

quito

us in

all

patie

nts

with

sev

ere

seps

is a

nd s

eptic

sho

ck.

Bac

kgro

und

of E

GD

T 8 :

• It

has

bee

n re

plic

ated

in v

ario

us m

ultip

le

stud

ies

com

pris

ing

. 4

0 st

udie

s to

talin

g .

10,

000

patie

nts.

• R

educ

es a

bsol

ute

mor

talit

y by

16%

-18%

in

patie

nts

of e

qual

illn

ess

seve

rity

. •

It is

cos

t eff

ectiv

e an

d de

crea

ses h

ealth

-car

e re

sour

ce c

onsu

mpt

ion.

• T

he S

cv o 2 t

echn

olog

y is

. 4

0 y

old,

requ

ires

se

cond

s to

cal

ibra

te, a

nd is

opt

iona

l with

in

term

itten

t sam

plin

g ac

cord

ing

to th

e SS

C

reco

mm

enda

tions

.

Rat

iona

le fo

r a

noni

nfer

iori

ty d

esig

n:

•

To e

stab

lish

equi

pois

e be

caus

e co

mpa

ring

EG

DT

to

a c

ontr

ol g

roup

in a

pro

spec

tive

rand

omiz

ed

tria

l with

info

rmed

con

sent

wou

ld b

e un

ethi

cal

base

d on

rob

ust e

vide

nce

of p

revi

ous

stud

ies.

Met

hods

Inte

rven

tions

Prec

ise

deta

ils o

f the

inte

rven

tions

inte

nded

for

each

gro

up, d

etai

ling

whe

ther

the

refe

renc

e tr

eatm

ent i

n th

e no

ninf

erio

rity

or

equi

vale

nce

tria

l is

iden

tical

(or

very

sim

ilar)

to th

at in

any

tr

ial(s

) tha

t est

ablis

hed

effi c

acy

and

how

and

w

hen

they

wer

e ac

tual

ly a

dmin

iste

red.

LC

trig

gere

d in

terv

entio

ns:

EG

DT

trig

gere

d in

terv

entio

ns o

ver

6 h:

•

Supp

lem

enta

l oxy

gen

• B

lood

tran

sfus

ions

to h

emat

ocri

t of 3

0%

• D

obut

amin

e to

a 1

0% d

ecre

ase

in L

C

• F

luid

ther

apy

• Va

sopr

esso

r th

erap

y

• B

lood

tran

sfus

ions

•

Inot

rope

s us

e

•

Mec

hani

cal v

entil

atio

n

•

Prev

entio

n of

sud

den

card

iopu

lmon

ary

com

plic

atio

ns

O

bjec

tives

Spec

ifi c

obje

ctiv

es a

nd h

ypot

hese

s, in

clud

ing

the

hy

poth

esis

con

cern

ing

noni

nfer

iori

ty o

r eq

uiva

lenc

e.O

bjec

tive:

• To

add

ress

the

pote

ntia

l util

ity o

f LC

as

a su

bstit

ute

for

Scv o

2 .

Hyp

othe

sis:

•

Targ

etin

g L

C a

s th

e m

arke

r of

ade

quac

y of

D o

2 was

no

ninf

erio

r to

the

curr

ently

rec

omm

ende

d Sc

v o 2

mon

itori

ng fo

r th

e ou

tcom

e of

in-h

ospi

tal m

orta

lity.




Tabl

e 1—

Con

tinu

ed

(Con

tinu

ed)

C

omm

enta

ry

Topi

cR

equi

rem

ents

and

Con

side

ratio

nsJo

nes

et a

l 7,10

R

iver

s et

al 8

Oth

er o

bjec

tives

:

•

Disa

ssem

ble

a pr

otoc

ol “t

oo c

ompl

ex” f

or th

e pr

ovid

ers o

f car

e.

• Pr

ove

evid

ence

that

che

ckin

g se

rial

lact

ates

is e

asie

r an

d a

relia

ble

equi

vale

nt to

Scv

o 2 .

• F

ind

a si

mpl

er a

nd m

ore

gene

raliz

able

met

hod

to m

onito

r th

e ad

equa

cy o

f D o

2 as

a re

sear

ch im

pera

tive

in th

e tr

eatm

ent

of p

atie

nts

with

sev

ere

infe

ctio

n.

O

utco

mes

Cle

arly

defi

ned

prim

ary

and

seco

ndar

y ou

tcom

e m

easu

res,

de

taili

ng w

heth

er th

e ou

tcom

es in

the

noni

nfer

iori

ty

or e

quiv

alen

ce tr

ial a

re id

entic

al (o

r ve

ry s

imila

r) to

th

ose

in a

ny tr

ial(s

) tha

t est

ablis

hed

effi c

acy

of th

e re

fere

nce

trea

tmen

t and

, whe

n ap

plic

able

, any

m

etho

ds u

sed

to e

nhan

ce th

e qu

ality

of m

easu

rem

ents

(e

g, m

ultip

le o

bser

vatio

ns, t

rain

ing

of a

sses

sors

).

• T

he p

rimar

y en

d po

int w

as a

bsol

ute

in-h

ospi

tal m

orta

lity

rate

.

• Se

cond

ary

end

poin

ts w

ere

ICU

leng

th o

f sta

y, h

ospi

tal

leng

th o

f sta

y, v

entil

ator

-fre

e da

ys, a

nd n

ew o

nset

mul

tiple

or

gan

failu

re.

• O

ther

end

poi

nts

asse

ssed

wer

e th

e nu

mbe

r of

res

usci

tativ

e go

als

achi

eved

, adm

inis

tere

d tr

eatm

ents

, and

pre

defi n

ed

prot

ocol

-rel

ated

ser

ious

adv

erse

eve

nts.

Sam

ple

size

an

d st

atis

tical

m

etho

ds

How

sam

ple

size

was

det

erm

ined

, det

ailin

g w

heth

er it

was

cal

cula

ted

usin

g a

noni

nfer

iorit

y or

equ

ival

ence

cr

iteri

on a

nd s

peci

fyin

g th

e m

argi

n of

equ

ival

ence

with

th

e ra

tiona

le fo

r its

cho

ice.

Whe

n ap

plic

able

, exp

lana

tion

of a

ny in

teri

m a

naly

ses a

nd st

oppi

ng ru

les (

and

whe

ther

re

late

d to

a n

onin

feri

ority

or

equi

vale

nce

hypo

thes

is).

• U

sing

a on

e-sid

ed te

st o

f non

infe

riorit

y, as

sum

ing

a co

ntro

l gr

oup

mor

talit

y ra

te o

f 25%

and

a 5

.05,

a sa

mpl

e siz

e of

15

0 pe

r gr

oup

gave

71%

pow

er to

det

erm

ine

the

inte

rven

tion

did

not i

ncre

ase

mor

talit

y by

. 1

0%.

Prev

ailin

g m

orta

lity

in p

revi

ous b

efor

e an

d af

ter

in

terv

entio

n st

udie

s sho

w a

hig

her m

orta

lity. 8

• B

asel

ine

mor

talit

y of

47.

6% �

5%

•

Aft

er in

terv

entio

n m

orta

lity

of

27.5

% �

13%

•

The

req

uire

d si

ze o

f non

infe

rior

ity tr

ials

is u

sual

ly

larg

er th

an th

at fo

r su

peri

ority

tria

ls. 6

B

lindi

ng

(mas

king

)W

heth

er p

artic

ipan

ts, t

hose

adm

inist

erin

g th

e

inte

rven

tions

, and

thos

e as

sess

ing

the

outc

omes

wer

e bl

inde

d to

gro

up a

ssig

nmen

t. W

hen

rele

vant

, how

the

succ

ess

of b

lindi

ng w

as e

valu

ated

.

• B

lindi

ng d

oes

not p

rote

ct a

gain

st b

ias

near

ly a

s w

ell

in a

non

infe

rior

ity tr

ial a

s it

does

in a

sup

erio

rity

tria

l.

Res

ults

Part

icip

ant

fl ow

and

re

crui

tmen

t

Flo

w o

f par

ticip

ants

thro

ugh

each

sta

ge (a

dia

gram

is

st

rong

ly r

ecom

men

ded)

. Spe

cifi c

ally

, for

eac

h gr

oup

repo

rt th

e nu

mbe

rs o

f par

ticip

ants

rand

omly

ass

igne

d,

rece

ivin

g in

tend

ed tr

eatm

ent,

com

plet

ing

the

tria

l pr

otoc

ol, a

nd a

naly

zed

for

the

prim

ary

outc

ome.

D

escr

ibe

prot

ocol

dev

iatio

ns fr

om tr

ial a

s pl

anne

d,

toge

ther

with

rea

sons

.

One

sho

uld

avoi

d fe

atur

es th

at m

ight

dilu

te tr

ue d

iffer

ence

s be

twee

n

EG

DT

and

LC

, the

reby

enh

anci

ng th

e ri

sk o

f err

oneo

usly

co

nclu

ding

non

infe

rior

ity:

• Po

or a

dher

ence

to th

e pr

otoc

ol

• D

ropo

uts

• R

ecru

itmen

t of p

atie

nts u

nlik

ely

to n

eed

or re

spon

d to

inte

rven

tions

(lo

w il

lnes

s se

veri

ty a

nd m

orta

lity)

trea

tmen

t cro

ssov

ers.

Bas

elin

e

da

ta o

f the

pa

rtic

ipan

ts

Bas

elin

e de

mog

raph

ic a

nd c

linic

al c

hara

cter

istic

s

of e

ach

grou

p. E

ligib

ility

cri

teri

a fo

r pa

rtic

ipan

ts

(det

ailin

g w

heth

er p

artic

ipan

ts in

the

noni

nfer

iorit

y or

equ

ival

ence

tria

l are

sim

ilar

to th

ose

in a

ny

tria

l[s] t

hat e

stab

lishe

d ef

fi cac

y of

the

refe

renc

e tr

eatm

ent)

and

the

sett

ings

and

loca

tions

whe

re

the

data

wer

e co

llect

ed.

At b

asel

ine

com

pare

d to

the

Riv

ers

et a

l 8 stu

dy, J

ones

et a

l 7 pat

ient

s

wer

e ch

arac

teri

zed

by:

• Yo

unge

r ag

e

•

Few

er c

omor

bidi

ties

• L

ower

illn

ess

seve

rity

•

Low

er m

orta

lity

risk

•

Low

er la

ctat

e le

vels

•

Hig

her

Scv o

2 (no

rmal

ran

ge)

• H

ighe

r C

VP

(nor

mal

ran

ge)


www.chestpubs.org



Tabl

e 1—

Con

tinu

ed

C

omm

enta

ry

Topi

cR

equi

rem

ents

and

Con

side

ratio

nsJo

nes

et a

l 7,10

R

iver

s et

al 8

N

umbe

rs

anal

yzed

Num

ber

of p

artic

ipan

ts (d

enom

inat

or) i

n ea

ch g

roup

incl

uded

in e

ach

anal

ysis

and

whe

ther

“in

tent

ion-

to-t

reat

” an

d/or

alte

rnat

ive

anal

yses

wer

e co

nduc

ted.

St

ate

the

resu

lts in

abs

olut

e nu

mbe

rs w

hen

feas

ible

(e

.g.,

10 o

f 20,

not

50%

).

Jone

s et

al 7 r

epor

ted

30 in

terv

entio

ns in

300

pat

ient

s ra

ndom

ized

:

• 10

% (n

5 2

9) g

ot to

the

poin

t in

the

prot

ocol

whe

re o

ne

of th

e tw

o tr

eatm

ents

bei

ng e

valu

ated

wer

e ad

min

iste

red

(Scv

o 2 ;

n 5

13)

and

(LC

; n 5

16)

.

O

utco

mes

and

es

timat

ion

For

eac

h pr

imar

y an

d se

cond

ary

outc

ome,

a s

umm

ary

of

res

ults

for

each

gro

up a

nd th

e es

timat

ed e

ffec

t siz

e an

d its

pre

cisi

on (e

g, 9

5% C

I). F

or th

e ou

tcom

e(s)

for

whi

ch n

onin

feri

ority

or

equi

vale

nce

is h

ypot

hesi

zed,

a

fi gur

e sh

owin

g co

nfi d

ence

inte

rval

s an

d m

argi

ns

of e

quiv

alen

ce m

ay b

e us

eful

.

Patie

nts

wer

e tr

eate

d in

the

ED

from

the

time

of r

ando

miz

atio

n un

til a

ll

trea

tmen

t goa

ls w

ere

achi

eved

or

6 h

had

elap

sed:

6-h

end

poin

ts a

re p

rovi

ded

in th

e R

iver

s et

al 8

st

udy.

•

No

6-h

hem

odyn

amic

end

poi

nts

are

repo

rted

.

•

Alth

ough

hem

atoc

rit w

as a

ther

apeu

tic e

nd p

oint

, no

data

w

ere

repo

rted

in r

each

ing

this

end

poi

nt in

eith

er g

roup

.

•

The

re w

as n

o m

etho

d to

ass

ess

whe

ther

an

indi

cate

d th

erap

eutic

act

ion

was

per

form

ed in

res

pons

e to

a

para

met

er b

elow

the

inte

nded

goa

l.

Adv

erse

eve

nts

All

impo

rtan

t adv

erse

eve

nts

or s

ide

effe

cts

in e

ach

in

terv

entio

n gr

oup.

Del

ayed

hem

odyn

amic

det

erio

ratio

n w

as n

ot r

epor

ted.

Del

ayed

hem

odyn

amic

det

erio

ratio

n is

an

im

port

ant f

eatu

re in

up

to 2

0% o

f pat

ient

s w

hich

was

red

uced

to 1

0% w

ith E

GD

T.

In

terp

reta

tion

and

gene

raliz

abili

ty

Gen

eral

izab

ility

(ext

erna

l val

idity

) of t

he tr

ial fi

ndi

ngs.

• L

C c

anno

t be

used

in p

atie

nts

with

non

elev

ated

lact

ate

leve

ls.

•

LC

is u

npro

ven

in p

atie

nts

with

mor

e co

mpl

ex p

rese

ntat

ions

w

ho r

equi

re m

ultip

le in

terv

entio

ns.

O

vera

ll ev

iden

ceG

ener

al in

terp

reta

tion

of th

e re

sults

in th

e

cont

ext o

f cur

rent

evi

denc

e.

•

Thi

s st

udy

is u

nder

pow

ered

to p

rovi

de e

noug

h ev

iden

ce

to u

se L

C a

s a

repl

acem

ent o

f Scv

o 2 .

Add

ition

al

cont

ribu

tions

U

ntes

ted

hypo

thes

isC

linic

ians

mus

t be

confi

den

t tha

t the

new

trea

tmen

t

wou

ld h

ave

been

sho

wn

to b

e ef

fi cac

ious

if a

pl

aceb

o-co

ntro

lled

tria

l had

bee

n pe

rfor

med

.

• Is

ther

e an

out

com

e di

ffer

ence

bet

wee

n a

LC

from

10

to

9 vs

4 to

3.6

, bot

h 10

% c

lear

ance

but

diff

eren

t out

com

es?

• H

ow m

any

lact

ate

leve

ls w

ere

requ

ired

per

pat

ient

to

com

plet

e th

erap

y an

d th

e tim

e du

ratio

n?

• Is

LC

cos

t eff

ectiv

e as

sho

wn

in E

GD

T?

• W

hat i

s th

e te

chno

logy

ass

essm

ent c

ompa

ring

the

clin

ical

ut

ility

of a

pre

calib

rate

d Sc

v o 2 a

nd h

ourl

y or

eve

n m

ore

freq

uent

lact

ate

leve

ls?

C

ompe

ting

inte

rest

sT

his

may

pot

entia

lly in

fl uen

ce th

e ed

itori

al

pr

oces

s an

d in

terp

reta

tion

of th

e re

sults

.

•

The

sec

tion

edito

r of

the

jour

nal w

here

Jon

es e

t al 7

is p

ublis

hed

is a

pri

ncip

le in

vest

igat

or fo

r co

mpe

ting

rese

arch

on

EG

DT.

• T

he a

utho

r of

the

acco

mpa

nyin

g ed

itori

al is

par

ticip

atin

g in

com

petin

g re

sear

ch o

n E

GD

T. 10

CO

NSO

RT

5 C

onso

lidat

ed S

tand

ards

of

Rep

ortin

g Tr

ials

; CV

P 5

cent

ral v

enou

s pr

essu

re; D

o 2 5

syst

emic

oxy

gen

deliv

ery;

EG

DT

5 e

arly

goa

l dir

ecte

d th

erap

y; L

C 5

lact

ate

clea

ranc

e; S

cv o

2 5 ce

ntra

l ve

nous

oxy

gen

satu

ratio

n; S

SC 5

surv

ivin

g se

psis

cam

paig

n.




Correspondence to: Emanuel P. Rivers, MD, MPH, FCCP, Department of Emergency Medicine, Wayne State University, 270 Clara Ford Pavilion, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202; e-mail: [email protected] © 2011 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( http :// www . chestpubs . org / site / misc / reprints . xhtml ). DOI: 10.1378/chest.11-2599

References 1 . Levraut J , Ichai C , Petit I , Ciebiera JP , Perus O , Grimaud D .

Low exogenous lactate clearance as an early predictor of mortality in normolactatemic critically ill septic patients . Crit Care Med . 2003 ; 31 ( 3 ): 705 - 710 .

2 . Jones AE . Point: should lactate clearance be substituted for central venous oxygen saturation as goals of early severe sepsis and septic shock therapy? Yes . Chest . 2011 ; 140 ( 6 ): 1406 - 1408 .

3 . Gattinoni L , Brazzi L , Pelosi P , et al . A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collabo-rative Group . N Engl J Med . 1995 ; 333 ( 16 ): 1025 - 1032 .

4 . Chamberlain DJ , Willis EM , Bersten AB . The severe sepsis bundles as processes of care: a meta-analysis [published online ahead of print February 14, 2011] . Aust Crit Care .

5 . Jansen TC , van Bommel J , Schoonderbeek FJ , et al ; LACTATE study group . Early lactate-guided therapy in intensive care unit patients: a multicenter, open-label, randomized con-trolled trial . Am J Respir Crit Care Med . 2010 ; 182 ( 6 ): 752 - 761 .

6 . Piaggio G , Elbourne DR , Altman DG , Pocock SJ , Evans SJ ; CONSORT Group . Reporting of noninferiority and equiva-lence randomized trials: an extension of the CONSORT statement . JAMA . 2006 ; 295 ( 10 ): 1152 - 1160 .


8 . Rivers EP . Point: adherence to early goal-directed therapy: does it really matter? Yes. After a decade, the scientifi c proof speaks for itself . Chest . 2010 ; 138 ( 3 ): 476 - 480 .

9 . Dellinger RP , Levy MM , Carlet JM , et al ; International Sur-viving Sepsis Campaign Guidelines Committee ; American As-sociation of Critical-Care Nurses ; American College of Chest Physicians ; American College of Emergency Physicians ; Cana-dian Critical Care Society ; European Society of Clinical Micro-biology and Infectious Diseases ; European Society of Intensive Care Medicine ; European Respiratory Society ; International Sepsis Forum ; Japanese Association for Acute Medicine ; Japanese Society of Intensive Care Medicine ; Society of Crit-ical Care Medicine ; Society of Hospital Medicine ; Surgical Infection Society ; World Federation of Societies of Intensive and Critical Care Medicine . Surviving Sepsis Campaign: inter-national guidelines for management of severe sepsis and septic shock: 2008 . Crit Care Med . 2008 ; 36 ( 1 ): 296 - 327 .

10 . Lewis RJ . Disassembling goal-directed therapy for sepsis: a fi rst step . JAMA . 2010 ; 303 ( 8 ): 777 - 779 .





www.chestpubs.org


chest point/counterpoint editorials...1408 point/counterpoint editorials references 1 . shoemaker wc...

Documents