The Eye of the Beholder

Don’t confuse symmetry with balance.

―Tom Robbins, Even Cowgirls Get the Blues

It is not often that one is asked to write an editorial about a review article—usually an editorial is called for after a striking new scientific finding or clinical breakthrough, some paradigm-shifting publication, or to address public policy concerns that affect anesthesiologists. Luckily, the review article in this edition of Anesthesia & Analgesia, “The Coagulation Profile of End-Stage Liver Disease and Considerations for Intraoperative Management,” by Dr Katherine Forkin¹ and her University of Virginia colleagues gets an editorial spotlight. And maybe, just maybe, this highlighted review will give nonanesthesiologist colleagues insights into the coagulation abnormalities of end-stage liver disease (ESLD), insights that have been uniquely hard-won by anesthesiologists who care for liver transplant patients.

In recent years, the concept of a “balanced” or “rebalanced” coagulation status has dominated literature reviews and conference proceedings dealing with the coagulation system in patients with ESLD. This descriptor of “balanced” has been impossible to ignore because it has been featured in the most prominent journals in articles written by world-class experts,² and so has been carried forward in many publications without thought to the implications the word “balanced” evokes. We now know, of course, that the liver is the site of synthesis of procoagulant factors and anticoagulant factors, so both positive and negative coagulation patterns are impaired in ESLD—but to use the word “balanced” to describe this state is problematic and misleading from both a scientific and clinical perspective. (The authors of these articles do acknowledge that the balance is a fragile one, easily tipped to one side or the other.)

As physiologists, we know that physiologic balance is actively maintained by complex control mechanisms that adjust the functioning or output of a system. Healthy clotting and fibrinolysis are together a good example of a well-controlled, complex system. But with ESLD, not only are components of the coagulation system lost, the controllers are lost too, such that the system can oscillate wildly and unpredictably, especially if perturbed. If the coagulation system of ESLD was “balanced” in the physiologic sense, we would expect to see control elements activated to maintain that balance: As far as I know there are not feedback control mechanisms that activate anticoagulant transcription or translation when procoagulant genes are overexpressed (or vice versa). Systems failure (the coagulation system in ESLD) is not accurately described by “balance.” This is important from an engineering perspective of a physiologic system as well—since a system without its controllers is much more easily sent into cascading failure by a perturbation than by a system with its controllers intact.³

Any anesthesiologist with experience in liver transplantation quickly learns that the coagulation status of ESLD patients is far from balanced—rather it is 2 diseases, 2 imbalances, in 1. Patients can be very coagulopathic AND make pathologic clots at the very same moment. Skillful intraoperative care of liver transplant patients is walking the fine line between preventing hemorrhage (eg, with individualized transfusion strategies) and preventing pathologic clot formation (eg, knowing when to withhold antifibrinolytics). Anesthesiologists get to witness this “2 diseases in 1” regularly in the operating room in ways that hematologists or hepatologists do not. We are witnesses because the “perturbation” that can initiate worsening of both disease processes is big—liver transplant under general anesthesia—not the smaller perturbations of coagulation seen in outpatient clinical settings.

Interestingly, the very first article in English I could find on the coagulation status in liver disease patients coming to surgery came from the Anesthesia Laboratory of the Harvard Medical School in 1959.⁴ The study is hard to follow because the clotting factors do not have familiar modern names, but the article is remarkably insightful on several fronts. The authors noted that antihemophilic factor, now called factor VIII, was present in normal amounts in some patients with advanced liver disease. They also noted the importance of fresh whole blood transfusion if platelets were to be replenished during surgery, and commented on dilutional thrombocytopenia being a known problem after massive transfusion. These authors also note that general anesthesia and surgery are both known to worsen coagulation, and they make this statement with references. (We have done little to follow up this particular observation in the liver transplant population, with most studies looking at effects of anesthetics on coagulation done in normal surgical populations.) In other words, there is a lot of information in this article that was lost to time, and had to be relearned in other settings. The other very strong point made by the authors of this article is the extreme heterogeneity in the clotting factor abundance, function, and response to transfusion in patients with liver disease. Although not necessarily emphasized strongly in the Forkin et al¹ review, this functional heterogeneity in coagulation profiles of patients with ESLD should be a major take-home message from the review.

The heterogeneity (and complexity) of the coagulation status of ESLD is 1 reason why conclusive studies about managing the coagulation status of liver transplant patients have been hard to come by (and why coagulation management practices still vary considerably from center to center). Any intervention being studied is being imposed on multiple phenotypes. For example, Forkin et al¹ review the literature and conclude that “the degree of derangement in fibrinolysis (in ESLD patients) remains unclear”— I contend that quantification of the derangement is “unclear” because an average means nothing in this heterogeneous population. Some ESLD patients will have excessive fibrinolysis, some will have normal fibrinolysis, so that optimal anesthetic management implies detective work to determine the individual patient’s coagulation status, followed by a personalized intraoperative management plan.

Perhaps stratifying patients by coagulopathic profiles will help us design better clinical trials for managing coagulation during liver transplantation (but it is difficult to do). We also need new approaches. The mathematical modeling approaches to coagulation and its abnormalities in the setting of trauma taken by Mitch Cohen’s group are a good place to start.⁵ These models are particularly appealing because the coagulation status is perturbed similarly in trauma to the way it is perturbed in surgery, with trauma-induced contact with activated platelet secretions (think intrinsic pathway) and trauma-induced subendothelium exposure (extrinsic pathway). Though not yet ready for real-time monitoring, these kinds of models should guide further research into the coagulation status of ESLD patients and their management during surgery, because they are useful for understanding why changes in some parts of the coagulation system result in changes in overall behavior of the coagulation system. In other words, the models are predictive.

Forkin et al¹ have done a fine job reviewing what we know about coagulation in patients with liver disease, and certainly pointing out important gaps in knowledge. In the first draft of the manuscript submitted to Anesthesia & Analgesia, the “balanced” descriptor so in fashion was used quite often, carried forward from major articles in the field. As a result of the Anesthesia & Analgesia editorial process, Forkin et al¹ decided to rock the boat just a bit, and removed the adjective “balanced” from the review. The result is a more “fair and balanced” discussion of a complex issue. I hope this change in wording influences our colleagues who study this field, so that they do not automatically carry over the “balanced” label for describing the coagulation system in ESLD. “Balance” suggests a state under active control rather than a failing control system, or at least it is too wimpy a word to capture the complexity of ESLD coagulation dysfunction.

DISCLOSURES

Name: Marie E. Csete, MD, PhD.

Contribution: This author wrote the editorial.

This manuscript was handled by: Roman M. Sniecinski, MD.