Severe sepsis is both common and deadly. The pathophysiology of sepsis is complex and includes a nonlinear interplay between multiple cell types and soluble mediators.1,2 Over the past decade, enormous resources have been expended on sepsis trials, with more than 10 000 patients enrolled in more than 20 placebo-controlled, randomized phase 3 clinical trials. The vast majority of these therapies have failed to improve survival in patients with severe sepsis. A notable exception is activated protein C, which was shown in the Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) study to reduce 28-day all-cause mortality in this patient population.3
The biologic plausibility of the PROWESS results is mired in a maze of inflammatory and coagulation pathways. Intensivists and hematologists (not to mention investigators from many other disciplines) have found themselves in an unlikely partnership, struggling to get their arms around the problem. If any consensus has been reached, it goes something like this: (1) activation of both inflammation and coagulation contributes to the pathophysiology of severe sepsis and organ dysfunction, (2) therapies that target only the inflammatory cascade (eg, antimediator therapy) or only the coagulation pathway (eg, antithrombin agents) do not improve survival, and (3) therapies that target both inflammatory and coagulation pathways (eg, activated protein C) hold more promise.
Just as the dust is settling, a study by Kerlin and colleagues (page 3085) promises to kick up another storm. This study provides evidence that factor V (FV) Leiden confers a survival advantage in severe sepsis and in an animal model of endotoxemia. While the numbers of patients in the PROWESS trial with the FV Leiden mutation were relatively small, the inclusion of the animal data goes a long way toward supporting their conclusion. Although much more work will be required to elucidate the mechanism(s) by which FV Leiden protects against sepsis mortality, the results amount to a fascinating paradigm shift, summarized as follows: (1) thrombin generation may be protective in the setting of sepsis, (2) the effect of thrombin is dose-dependent—a little more thrombin (FV carrier) is good, while too much thrombin (FV homozygous) is not—and (3) the FV Leiden mutation may have evolved as a means of protecting not so much against the saber-toothed tiger, but rather as a defensive weapon in the host-pathogen arms race.
If these results hold true in large populations of patients with severe sepsis, and if the mechanism is one of increased thrombin generation (and perhaps secondary activation of endogenous protein C), then we will be forced to revise our model of sepsis pathophysiology yet again, this time assigning a protective role to the coagulation pathway (at least up to a certain level of activation). Does that mean we are back to a simplified version of sepsis pathogenesis in which one or another component of the inflammatory cascade is the “Darth Vader” of the host response? The answer is most certainly no. As with most, if not all, biologic systems, the sepsis network is likely to display small-world properties consisting of diverse nodes, hubs, links, and connection weights. Mapping the topology of the network will keep us busy well into the millennium. That said, the revised model in its present form raises interesting therapeutic questions. For example, would a re-engineered activated protein C molecule that lacks anti-Va and anti-VIIIa function yet retains nonanticoagulant activity yield an improved therapeutic window? Moreover, could one actually titrate the systemic infusion of a prothrombotic agent such as rVIIa to optimize thrombin generation while minimizing the deleterious effects of tissue factor/VIIa/activated factor X–mediated cell activation? Given the unpredictable turns in this field, I would not be surprised if the answer is yes on both counts.