Journal Logo


Fluid Resuscitation and Vasopressors in Septic Shock: The Importance of Filling the Tank While Squeezing the Pipes*

Schellenberg, Morgan MD, MPH, FRCSC; Cobb, J. Perren MD, FCCM

Author Information
doi: 10.1097/CCM.0000000000004539
  • Free
  • Editor's Choice

Treatment of shock requires an increase in preload, afterload, cardiac contractility, and/or afterload, which is a physiologic application of Ohm’s law. Fluid resuscitation increases preload; vasoactive agents increase the rest. For patients in septic shock, optimal resuscitation strategies remain controversial and are a frequent focus of investigation. Publication of the study by Rivers et al (1) of early goal-directed therapy (EGDT) in the treatment of severe sepsis and septic shock emphasized protocolized, emergency department (ED) care using fluid resuscitation, vasopressors, blood transfusion, and inotropes to maintain central venous pressure greater than 8 mm Hg, mean arterial pressure (MAP) greater than 65 mm Hg, and central venous oxygen saturation greater than or equal to 70%. Patients in the EGDT protocolized arm received on average almost 5 L of fluid in the first 6 hours (mean ± sd 4,981 ± 2,984 mL), significantly more than patients in the nonprotocolized arm. EGDT was reported to improve mortality when compared to standard, nonprotocolized care, which led to its widespread adoption in both EDs and ICUs. However, the enthusiasm for EGDT was tempered by a series of reports demonstrating the negative consequences of a liberal crystalloid resuscitation strategy in critically ill patients. For example, when conservative fluid resuscitation was compared against liberal fluid resuscitation among patients with acute lung injury, conservative resuscitation was associated with shortened duration of mechanical ventilation and ICU length of stay, without an associated increase in extrapulmonary organ failure (2). The mean ± se cumulative fluid balance during the first 7 days in the study by Roberts et al (3) was –136 ± 491 mL for the conservative-strategy group and 6,992 ± 502 mL for the liberal-strategy group (p < 0.001). Three subsequent harmonized, international, randomized control trials in 2014 and 2015 (Protocolised Management in Sepsis, Australasian Resuscitation in Sepsis Evaluation, and Protocolized Care for Early Septic Shock) further shifted the focus of ICU sepsis care away from EGDT with findings that continuous central venous oxygen saturation and central venous pressure monitoring did not improve outcomes and potentially led to over-resuscitation (4–6). Contemporary clinical practice recognizes the importance of the Surviving Sepsis Guidelines to improving patient outcomes globally, including a recommended initial crystalloid bolus of 30 mL/kg (2,100 mL in a 70 kg patient) and titration of vasopressors within the first 3 hours of severe sepsis or septic shock (7). Thus, it is clearly critical to balance “filling the tank” with “squeezing the pipes”… but how?

In this issue of Critical Care Medicine, Roberts et al (3) report on the results of CHASERS, a substudy of VOLUME-CHASERS, the Observation of Variation in Fluids Administered and Characterization of Vasopressor Requirements in Shock, conducted by the Society’s Discovery Clinical Research Network. The investigators report on the impact of vasopressor administration on mortality among patients in the initial stages of septic shock. Importantly, they determine the impacts of fluid resuscitation and vasopressor dosing strategies on this association. To accomplish this, the study included consecutive patients greater than or equal to 18 years who presented with shock to one of 33 centers (32 in the United States and one in Jordan) between September 2017 and February 2018. In VOLUME-CHASERS, patients were defined to be presenting with shock if they required any amount of vasopressor to achieve and maintain MAP greater than 65 mm Hg or systolic blood pressure less than 90 mm Hg. In the current CHASERS substudy, only patients in septic shock who were treated with appropriate initial antibiotics and greater than or equal to 1 vasopressor within 24 hours of shock onset were included. The study objectives were threefold. The first was to determine the association between vasopressor dosing intensity (VDI) in the first 6 hours and 24 hours of septic shock and 30-day in-hospital mortality. The second and third objectives were to see if these associations were dependent upon volumes of fluid resuscitation or vasopressor dosing titration pattern. VDI, which was the primary exposure, was specified as the total vasopressor dose given in µg/min and expressed in norepinephrine dosing equivalents. This calculation included any administered vasopressor and was tabulated by converting doses of non-norepinephrine vasopressors into norepinephrine dosing equivalents. The cumulative average VDI between 0 and 6 hours and 0–24 hours after diagnosis of shock were used for analysis. Cumulative volumes of fluid administration (crystalloid, colloid, packed RBCs, fresh frozen plasma, or platelets) were also recorded over the same time periods. Last, the vasopressor dosing titration patterns were defined over these time periods based both on absolute dose and fluctuations in dose over the two time periods. The study enrolled 616 patients, with a mean age of 64 years (sd, 15 yr), of whom 54% were male. Mean Acute Physiology and Chronic Health Evaluation III score was 97 (sd 28), and median Sequential Organ Failure Assessment score was 9 (interquartile range, 7–12). Most patients were diagnosed with shock in the ED (n = 273, 44%) and admitted to a medical ICU (n = 342, 56%). Overall in-hospital mortality at 30 days was 31%. By far, the most common vasopressor administered was norepinephrine (93%). Mean volume of fluid administered was 1,600 mL in the first 6 hours and 3,400 mL by 24 hours.

The primary study finding was that increasing VDI was associated with increasing mortality over the first 24 hours after the diagnosis of shock. Specifically, every increase in average VDI of 10 µg/min was associated with a 33% increase in the odds of in-hospital mortality at 30 days, regardless of volume resuscitation. However, in the first 6 hours, this effect was dependent upon the volume of fluid administered. During the first 6 hours, as volume of fluid administration increased, the association between VDI and mortality weakened, to the point that if greater than 2,000 mL of fluids were administered between 0–6 hours, increasing VDI was no longer associated with increased mortality. In terms of VDI dosing and titration pattern, patients who required high (≥ 15 µg/min) doses during the first 6 hours but low (< 15 µg/min) doses by the end of the 24-hour period had a lower mortality rate than patients who required high VDI dosing throughout or those who initially required low doses but needed high doses by 24 hours. This is relatively intuitive as this deescalating VDI pattern would seem to reflect marked clinical improvement over the first 24 hours of septic shock.

How does this report inform treating patients with septic shock, that is, answer how best to balance “filling the tank” with “squeezing the pipes”? The study findings that increasing doses of vasopressors during the first 6 hours are associated with increased mortality unless they are paired with at least 2,000 mL of crystalloid resuscitation corresponds nicely with the Surviving Sepsis Campaign Hour-1 bundle recommendation to resuscitate with 30 mL/kg IV crystalloid, equating to 2,100 mL for a typical 70 kg adult (7). Put differently, the study by Roberts et al (3) adds substantial support to the concept that immediate, adequate fluid resuscitation to maintain circulating volume should be the initial focus of care for patients with severe sepsis or septic shock (8). This may seem obvious, but it is a topic for which there is current clinical equipoise given recent publications encouraging fluid minimization with preferential use of vasopressors to maintain MAPs for septic patients (9,10).

The limitations of the current study by Roberts et al (3) are well delineated in their report; the investigators are wise to consider their results as hypothesis-generating only. The principal study limitation relates to its observational design, wherein identified associations must not be interpreted as causations. For example, the finding that increased VDI was associated with increased mortality may simply be a reflection of a higher baseline risk of mortality in these patients, and not the alternate explanation that increasing VDI imparts a mortality risk. Further, a risk of confounding bias exists. Although the authors did a statistically thorough adjustment for known confounders, if two populations differ at baseline, there is the possibility of unrecognized and thereby unaccounted for confounding differences between the populations. Information about cardiac function would be clinically relevant as we seek to understand optimal resuscitation, which will almost certainly be related to underlying cardiac physiology, although this was not examined by the current study by Roberts et al (3). Last, the calculation of VDI used in the study by Roberts et al (3) involved significant averaging over time periods, resulting in the possibility of measurement error in the dosage reporting. In the future, we hope the investigators will prospectively test their hypothesis that early aggressive vasopressor titration, coupled with adequate fluid resuscitation in the early hours after shock, may be superior when compared to slower vasopressor titration to high doses. In addition, VDI could be captured more accurately with careful hourly measurement of this variable. Studies also are needed to guide individualizing the volume of fluid resuscitation for septic shock, complicated by the presence of cardiac, pulmonary, and/or renal disease.

Finally, the study by Roberts et al (3) is an important model for investigator-initiated clinical studies in the United States and serves as a remarkable milestone. Based upon the successful collaborative research approaches used in Canada and Australia/New Zealand, the Society’s Discovery Clinical Research Network was created in 2017 as a network of networks, including the U.S. Critical Illness and Injury Trials (USCIIT) group (formed in 2008) and the Critical Care Pharmacotherapy Trials Network (CCPTN, formed in 2010) (11,12). VOLUME-CHASERS served as a primary mechanism for multidisciplinary, multicenter collaboration under the new Discovery umbrella, with USCIIT and CCPTN each contributing sites (13 and 16, respectively). A remarkable rate of patient enrollment followed as VOLUME-CHASERS hit full stride: 1,639 patients enrolled across 33 hospitals in 6 months, with no external funding. We commend the investigators on their impressive achievement, rigorous analysis of the results, and what we expect to be an impactful series of reports.


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; 345:1368–1377
2. Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med 2006; 354:2564–2575
3. Roberts RJ, Miano TA, Hammond DA, et al. Observation of VariatiOn in fLUids adMinistEred in shock-CHaracterizAtion of vaSoprEssor Requirements in Shock (VOLUME-CHASERS) Study Group and SCCM Discovery Network: Evaluation of Vasopressor Exposure and Mortality in Patients With Septic Shock2020; 48:1445–1453
4. Mouncey PR, Osborn TM, Power GS, et al.; ProMISe Trial Investigators: Trial of early, goal-directed resuscitation for septic shock. N Engl J Med 2015; 372:1301–1311
5. Peake SL, Delaney A, Bailey M, et al.; ARISE Investigators, ANZICS Clinical Trials Group: Goal-directed resuscitation for patients with early septic shock. N Engl J Med 2014; 371:1496–1506
6. Yealy DM, Kellum JA, et al.; ProCESS Investigators: A randomized trial of protocol-based care for early septic shock. N Engl J Med 2014; 370:1683–1693
7. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: International guidelines for management of sepsis and septic shock, 2016. Crit Care Med2017; 45:486–552
8. Chang DW, Huynh R, Sandoval E, et al. Volume of fluids administered during resuscitation for severe sepsis and septic shock and the development of the acute respiratory distress syndrome. J Crit Care 2014; 29:1011–1015
9. Permpikul C, Tongyoo S, Viarasilpa T, et al. Early use of norepinephrine in septic shock resuscitation (CENSER). A randomized trial. Am J Respir Crit Care Med 2019; 199:1097–1105
10. Self WH, Semler MW, Bellomo R, et al.; CLOVERS Protocol Committee and NHLBI Prevention and Early Treatment of Acute Lung Injury (PETAL) Network Investigators: Liberal versus restrictive intravenous fluid therapy for early septic shock: Rationale for a randomized trial. Ann Emerg Med 2018; 72:457–466
11. Cobb JP, Ognibene FP, Ingbar DH, et al. Forging a critical alliance: Addressing the research needs of the United States critical illness and injury community. Crit Care Med 2009; 37:3158–3160
12. Bauer SR, Abraham PE, Barletta JF, et al. Development of the critical care pharmacotherapy trials network. Am J Health Syst Pharm 2017; 74:287–293

crystalloids; early goal-directed therapy; intensive care unit; sepsis; septic shock; volume resuscitation

Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.