Kimberger et al.  1and the editors2are to be commended for attempting to shed light on an important topic: What is the optimal intraoperative fluid and resuscitation target?

Many experienced physicians, including us, who provide anesthesia for major intraabdominal surgery have evolved over time from crystalloid-only, “show me the proof” physicians to those being in philosophical agreement with both the author and the editorial writers—goal-directed therapy with colloid is best in intestinal cases. We believe this produces less gut edema without compromising gut or other critical organ perfusion (not to mention reducing the anesthesiologist’s aural discomfort from the oft repeated surgeon lament that the anesthesia team is “drowning” the patient). Indeed, Victor Hugo once said, “All the forces in the world are not so powerful as an idea whose time has come.”*

Unfortunately, despite our hope to the contrary, all the forces in the world may have to wait a little longer, because this study does not provide hard evidence to support the idea that goal-directed colloid infusion is the best method of managing these cases. The methodology is critically flawed in at least four ways.

First, these anesthetized pigs were hypotensive (baseline blood pressure = 57–60 mmHg) and tachycardic (heart rate = 110–117 beats/min) in baseline conditions, relative to well-established normal values for either conscious or anesthetized animals.3,4 

Second, the resuscitation was disparate; 250 ml of colloid is not the same resuscitation strategy as 250 ml of crystalloid. An intravascular equivalent of 500–750 ml crystalloid bolus should have been the comparator.

Third, there is no justification for the intraoperative mixed venous oxygen saturation target of 60, given the baseline value of 48–50.

Fourth, neither the threshold microcirculatory blood flow nor the tissue oxygen tension associated with anastomotic breakdown is established, so the excess blood flow or oxygen in the goal-directed group could be good, bad, or indifferent.

This study only demonstrates that inadequate fluid resuscitation is worse than adequate fluid resuscitation. The crystalloid group virtually never achieved the “goal” of mixed venous oxygen saturation > 60%; as the authors note themselves, six of nine animals in the group never achieved the goal over the entire experiment. The average of 1,794 ml per animal in the goal-directed crystalloid group indicated that each animal received the 250-ml bolus every 30 min (the maximum allowed) over the entire 4-h experiment, in contrast to the colloid group, which got a bolus every hour on average; this was about twice the colloid volume infused over the experiment and yet was still inadequate. The inability to achieve the goal in the crystalloid group does shed light on another debate, though. It suggests that the correct conversion is indeed 3 ml crystalloid to 1 ml of colloid, not 2:1.

The unexpected finding that the wet/dry ratio was not different in colloid versus  crystalloid is also obviously related to the fact that in the goal-directed crystalloid group, fluid resuscitation was inadequate. Since, by the authors’ own primary measure of mixed venous oxygen saturation, fluid resuscitation was not achieved in most goal-directed therapy crystalloid animals, adrenergic tone was likely increased throughout the experiment, and the very sensitive intestinal vasculature had vasoconstriction-limited perfusion—consistent with the decreased Po2of the intestinal tissue noted in the study. On the other hand, if the resuscitation had been adequate, it is probable that the wet/dry ratio would have been greater in the crystalloid group. It is not clear what effect appropriate resuscitation might have had on the primary measure of intestinal and perianastomotic tissue Po2, as an appropriate crystalloid comparator would have had more edema counteracting the positive effect of more perfusion. Regardless, it is impossible to attribute the different Po2of the tissue in this study to fluid choice versus  resuscitation adequacy, especially since the baseline condition was abnormal.

Then there is the issue of the measurement taken: Trying to identify a single and infallible parameter that predicts outcome in resuscitation is the search for the holy grail of critical care. Can we use a single number as a crystal ball and if so, which one? For all bedside clinicians the quest goes on. While variations in microcirculatory parameters like perianastomotic Po2tension increases our body of knowledge, it does not explain by itself better clinical outcome. As the authors point out, the lactate level in all groups was no different, which represents payment of the oxygen debt without any systemic sequelae. Why was resuscitation adequacy not comparable, but the endpoint of lactate not different? Could the colon possess protective mechanisms similar to those in effect with ischemic preconditioning of the cardiac muscle? The assertion that the use of goal-directed therapy with colloids accounts for improved patient outcomes because of the mechanism described is again not supported by the findings.

Furthermore, we believe that the journal has done the anesthesia community a mild disservice by publishing an editorial highlighting and lauding this critically flawed, albeit well-intentioned article, as “evidence” of the benefit of colloid goal-directed therapy. We need the information it seeks to convey, and believe a well-done study will support both the editorial and the paper. We just need a much better protocol and more insight when interpreting the results. In any case, we can hopefully all agree with another famous philosopher who said, “It ain’t over ’till it’s over .”5 

†University of Miami Miller School of Medicine, Miami, Florida.

Kimberger O, Arnberger M, Brandt S, Plock J, Sigurdsson GH, Kurz A, Hiltebrand L: Goal-directed colloid administration improves the microcirculation of healthy and perianastomotic colon. Anesthesiology 2009; 110:496–504
Kehlet H, Bundgaard-Nielsen M: Goal-directed perioperative fluid management: Why, when, and how? Anesthesiology 2009; 110:453–5
Hannon JP, Bossone CA, Wade CE: Normal physiological values for conscious pigs used in biomedical research. Lab Anim Sci 1009;40:293–8
Swindle MM: Swine in the Laboratory (Surgery, Anesthesia, Imaging, and Experimental Techniques), 2nd edition. Boca Raton, CRC Press, 2007
Boca Raton
CRC Press
Berra Y: The Yogi Book: I really didn’t say everything I said. New York, Workman Publishing Company, 1998
New York
Workman Publishing Company