Current issues of ACP Journal Club are published in Annals of Internal Medicine


Boosting oxygen delivery was not effective and increased mortality in critically ill patients

ACP J Club. 1994 Nov-Dec;121:63. doi:10.7326/ACPJC-1994-121-3-063

Source Citation

Hayes MA, Timmins AC, Yau EH, et al. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med. 1994 Jun 16;330:1717-22.



To evaluate the effectiveness of boosting oxygen delivery by infusing dobutamine in critically ill patients.


Randomized controlled trial with follow-up to hospital discharge. The trial was stopped after the second interim analysis because the intensive care unit (ICU) and in-hospital mortality differed significantly between groups.


2 ICUs in the United Kingdom.


100 patients (mean age 63 y) with life-threatening cardiorespiratory illness or those classified as high-risk surgery patients in whom the cardiac index, oxygen consumption, and oxygen delivery failed to reach target values (cardiac index > 4.5 L/min/m2 of body surface area, oxygen delivery > 600 mL/min/m2, and oxygen consumption > 170 mL/min/m2) after fluid expansion to achieve optimal left arterial filing pressure. Exclusion criteria were age < 16 years, pregnancy, neurosurgical procedure, preexisting cardiac disease, or hematologic cancer.


50 patients were assigned to dobutamine, 5 to 200 µg/kg/min, until all 3 targets for cardiac index, oxygen consumption, and oxygen delivery were achieved, unless sinus tachycardia, tachyarrhythmia, or electrocardiographic evidence of myocardial ischemia developed, in which case the dose of dobutamine was immediately reduced or discontinued and then titrated to achieve the highest possible values for cardiac index and oxygen consumption and delivery. 50 patients were assigned to the control group and received dobutamine only if the cardiac index was < 2.8 L/min/m2.

Main outcome measures

In-hospital mortality; duration of ICU stay, ventilation, and hospitalization; mean arterial blood pressure; and oxygen consumption.

Main results

21 patients in the control group and all patients in the treatment group received dobutamine. No difference existed between the treatment and control groups for oxygen consumption (P = 0.12) or mean arterial blood pressure (P = 0.42), despite a higher cardiac index and level of oxygen delivery in the treatment group (P < 0.05). Groups were similar for duration of ICU stay (10 d for each group), days of ventilation (8 d in both groups), and hospitalization (24 in control group vs 19 d in treatment). Although the predicted risk for death (from the APACHE II score) was 34% for both groups, in-hospital mortality was higher in the treatment group (54%) when compared with the control group (34%) (P = 0.04) and ICU mortality showed a similar pattern (30% for control group vs 50% for treatment group, P = 0.04) (Table).


Boosting oxygen delivery in critically ill patients failed to influence oxygen consumption but was associated with increased mortality.

Source of funding: North East Thames Regional Health Authority.

For article reprint:Dr. D. Watson, Department of Anesthesia, St. Bartholomew's Hospital at Smithfield, West Smithfield, London, England EC1A 7BE, United Kingdom. FAX 44-71-601-7520.

Table. Routine vs as-needed dobutamine to increase delivery in critically ill patients

Outcomes Routine dobutamine As-needed dobutamine RRI (95% CI) NNH (CI)
ICU mortality 50% 30% 67% (2 to 180) 5 (3 to 142)
In-hospital mortality 54% 34% 59% (1 to 156) 5 (3 to 225)

*Abbreviations defined in Glossary; RRI, NNH, and CI calculated from data in article.


In this unblinded study of a heterogeneous group of critically ill patients, Hayes and colleagues evaluated the use of dobutamine to elevate systemic oxygen delivery. Patients were comparable at baseline, and all randomized patients were included in the analysis. A unique feature of this study is that patients were only randomized if they did not achieve supraphysiologic treatment goals after fluid resuscitation. Thus, the group of patients selected were those not responsive to fluid challenge. Also, although the intervention in the active group was based on previous recommendations that were associated with improved outcome, the study protocol may have resulted in higher doses of inotropes and vasoactive drugs. Patients in the experimental arm received high-dose dobutamine and 68% received norepinephrine (the latter, though, only to elevate systemic vascular resistance to low normal values). This may have exacerbated tissue ischemia and resulted in the high incidence of death attributable to organ failure.

The Hayes study, taken in the context of other experiments, does not conclude the ongoing heated debate of whether pharmacologically increasing oxygen delivery in critically ill patients is beneficial. This study does, however, alert us to the fact that delayed (even by a few hours after admission), aggressive use of inotropes to maximize oxygen delivery may be harmful. To date, 6 other randomized controlled trials have been published; 2 reported a significant survival benefit, 2 reported a favorable trend, and 2 showed no difference between groups. It is interesting to note that the 2 studies showing a significant reduction in mortality after maximizing tissue oxygen delivery aimed preoperatively for these supraphysiologic targets. It is also important to note that inferences based on these previous studies are plagued by problems, such as pseudorandomization, selection bias, multiple cointerventions, and crossovers. The Hayes study confirms previous findings that ability to achieve supraphysiologic cardiac index and oxygen delivery and consumption (either spontaneously or pharmacologically) is a marker for survival. This is also, however, the first randomized trial showing that attempting to maximize cardiac output and oxygen delivery may actually do more harm than good.

Overall, the evidence concerning enhancing tissue oxygen delivery for critically ill patients remains inconclusive. A recent systematic review (1) of 7 randomized controlled trials (RCTs) of interventions designed to achieve supraphysiologic values of cardiac index and oxygen delivery showed that targetting therapy to supraphysiologic endpoints in critically ill patients had no significant effect on mortality (relative risk 0.86, 95% CI 0.62 to 1.20). The 2 trials in which these goals were targetted preoperatively (2, 3), however, showed potential benefit (relative risk 0.20, CI 0.07 to 0.55) in this subgroup. One recent RCT from the United Kingdom (4) also showed that in major elective surgery, preoperative increase in oxygen delivery has a favorable effect on mortality, whereas a recent RCT in septic patients (5) did not find any benefit.

Daren K. Heyland, MD
Deborah J. Cook, MDMcMaster UniversityHamilton, Ontario, Canada


1. Heyland DK, Cook DJ, King D, Kernerman P, Brun-Buisson C. Maximizing oxygen delivery in critically ill patients: a methodologic appraisal of the evidence. Crit Care Med. 1996;24:517-24.

2. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA. 1993;270:2699-707.

3. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest. 1988;94:1176-86.

4. Wilson J, Woods I, Fawcett J, et al. Reducing the risk of major elective surgery: randomised controlled trial of preoperative optimisation of oxygen delivery. BMJ. 1999;318:1099-103.

5. Alia I, Esteban A, Gordo F, et al. A randomized and controlled trial of the effect of treatment aimed at maximizing oxygen delivery in patients with severe sepsis or septic shock. Chest. 1999;115:453-61.