Recombinant human activated protein C inhibits local and systemic activation of coagulation without influencing inflammation during Pseudomonas aeruginosa pneumonia in rats Goda Choi, MD; Jorrit-Jan H. Hofstra, MD; Joris J. T. H. Roelofs, MD; Sandrine Florquin, MD, PhD; Paul Bresser, MD, PhD; Marcel Levi, MD, PhD; Tom van der Poll, MD, PhD; Marcus J. Schultz, MD, PhD A lveolar fibrin deposition is a hallmark of the acute respira- tory distress syndrome (1) and pneumonia (2– 4). Excessive fi- brin deposition within the airways results from severe inflammation and conse- quent activation of bronchoalveolar coag- ulation, which are not compensated by the anticoagulant and fibrinolytic sys- tems (5). In severe sepsis, which is character- ized by a strong systemic inflammatory response and coagulation disturbances, a reduction in patient mortality was shown with administration of recombinant hu- man activated protein C (rhAPC), one of the major natural anticoagulants (6). In- terestingly, in this trial, rhAPC treatment led to more rapid resolution of respira- tory failure (7), and patients with com- munity-acquired pneumonia as a primary source of infection seemed to benefit most from treatment (8). The beneficial effects of rhAPC are not only attributed to its sys- temic anticoagulant properties (9, 10), but also to its anticoagulant and anti-inflam- matory effects in the lungs (11, 12). In the current study, crucial compo- nents of the protein C system (i.e., levels of protein C [PC], activated protein C [APC], and soluble thrombomodulin) were measured in bronchoalveolar lavage fluid (BALF) of human patients with uni- lateral ventilator-associated pneumonia (VAP). Furthermore, rhAPC was adminis- tered to rats infected with Pseudomonas aeruginosa, a Gram-negative microor- ganism that is most frequently involved in VAP (13). We hypothesized that the protein C pathway would become insuffi- cient to maintain the hemostatic balance in pulmonary infection and, in addition, that rhAPC would have significant anti- coagulant and anti-inflammatory effects in an in vivo model of Gram-negative bacterial pneumonia. MATERIALS AND METHODS Patient Study Design. A bilateral bronchoalveolar lavage was performed in five patients with a unilat- eral VAP. Ten mechanically ventilated patients From the Department of Intensive Care Medicine (GC, JJHH, MJS), Department of Internal Medicine (GC, JJHH, ML, TvdP), Department of Pathology (JJTHR, SF), Depart- ment of Pulmonology (PB), Center for Experimental and Molecular Medicine (GC, TvdP), Center for Infection and Immunity Amsterdam (GC, TvdP), and Laboratory of Ex- perimental Intensive Care and Anesthesiology (GC, JJHH, MJS), Academic Medical Center, University of Amster- dam, Amsterdam, The Netherlands. The authors have not disclosed any potential con- flicts of interest. Supported, in part, by the Netherlands Organiza- tion for Scientific Research (NWO), VENI grant 2004, to Dr. Schultz (project 016.056.001). For information regarding this article, E-mail: godachoi@mail.com Copyright © 2007 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000261888.32654.6D Objective: Alveolar fibrin deposition is a hallmark of pneumonia. It has been proposed that recombinant human activated protein C exerts lung-protective effects via anticoagulant and anti-inflamma- tory pathways. We investigated the role of the protein C system in pneumonia caused by Pseudomonas aeruginosa, the organism that is predominantly involved in ventilator-associated pneumonia. Design: An observational clinical study and a controlled, in vivo laboratory study. Setting: Multidisciplinary intensive care unit and a research laboratory of a university hospital. Patients and Subjects: Patients with unilateral ventilator-as- sociated pneumonia and male Sprague-Dawley rats. Interventions: Bilateral bronchoalveolar lavage was performed in five patients with unilateral ventilator-associated pneumonia. A total of 62 rats were challenged with intratracheal P. aeruginosa (10 8 colony-forming units), inducing pneumonia. Rats were ran- domized to treatment with normal saline, recombinant human activated protein C, heparin, or recombinant tissue plasminogen activator. Measurements and Main Results: Patients with pneumonia demonstrated suppressed levels of protein C and activated pro- tein C in bronchoalveolar lavage fluid obtained from the infected site compared with the contralateral uninfected site. Intravenous administration of recombinant human activated protein C in rats with P. aeruginosa pneumonia limited bronchoalveolar generation of thrombin–antithrombin complexes, largely preserving local an- tithrombin activity. However, recombinant human activated pro- tein C did not have effects on neutrophil influx and activity, expression of pulmonary cytokines, or bacterial clearance. Conclusions: In patients with ventilator-associated pneumo- nia, the pulmonary protein C pathway is impaired at the site of infection, and local anticoagulant activity may be insufficient. Recombinant human activated protein C prevents procoagulant changes in the lung; however, it does not seem to alter the pulmonary host defense against P. aeruginosa pneumonia. (Crit Care Med 2007; 35:1362–1368) KEY WORDS: coagulation; fibrinolysis; inflammation; mechanical ventilation; pneumonia; protein C 1362 Crit Care Med 2007 Vol. 35, No. 5