Air Bubbles Activate Complement and Trigger Hemostasis and C3-Dependent Cytokine Release Ex Vivo in Human Whole Blood

Abstract

Venous air embolism, which may complicate medical and surgical procedures, activates complement and triggers thromboinflammation. In lepirudin-anticoagulated human whole blood, we examined the effect of air bubbles on complement and its role in thromboinflammation. Whole blood from 16 donors was incubated with air bubbles without or with inhibitors of C3, C5, C5aR1, or CD14. Complement activation, hemostasis, and cytokine release were measured using ELISA and quantitative PCR. Compared with no air, incubating blood with air bubbles increased, on average, C3a 6.5-fold, C3bc 6-fold, C3bBbP 3.7-fold, C5a 4.6-fold, terminal complement complex sC5b9 3.6-fold, prothrombin fragments 1+2 (PTF1+2) 25-fold, tissue factor mRNA (TF-mRNA) 26-fold, microparticle tissue factor 6.1-fold, β-thromboglobulin 26-fold (all p < 0.05), and 25 cytokines 11-fold (range, 1.5–78-fold; all p < 0.0001). C3 inhibition attenuated complement and reduced PTF1+2 2-fold, TF-mRNA 5.4-fold, microparticle tissue factor 2-fold, and the 25 cytokines 2.7-fold (range, 1.4–4.9-fold; all p < 0.05). C5 inhibition reduced PTF1+2 2-fold and TF-mRNA 12-fold (all p < 0.05). C5 or CD14 inhibition alone reduced three cytokines, including IL-1β (p = 0.02 and p = 0.03). Combined C3 and CD14 inhibition reduced all cytokines 3.9-fold (range, 1.3–9.5-fold; p < 0.003) and was most pronounced for IL-1β (3.2- versus 6.4-fold), IL-6 (2.5- versus 9.3-fold), IL-8 (4.9- versus 8.6-fold), and IFN-γ (5- versus 9.5-fold). Antifoam activated complement and was avoided. PTF1+2 was generated in whole blood but not in plasma. In summary, air bubbles activated complement and triggered a C3-driven thromboinflammation. C3 inhibition reduced all mediators, whereas C5 inhibition reduced only TF-mRNA. Combined C5 and CD14 inhibition reduced IL-1β release. These data have implications for future mechanistic studies and possible pharmacological interventions in patients with air embolism.

Venous Air Embolism Activates Complement C3 Without Corresponding C5 Activation and Trigger Thromboinflammation in Pigs

“…Activated C3- and C5-split products have been shown to interact with thrombocytes and leukocytes ( 28 , 29 ) and may potentially trigger a thromboinflammation involving both leukocytes, platelets, and coagulation. Interestingly, and in line with our previous in vitro findings in human whole blood ( 15 ), we discovered that air embolism predominantly activated C3 with only a minor C5 activation, contrary to what is otherwise observed in general when complement is activated, for example, by bacteria or damage or pathogen-associated molecular patterns. The C5a-C5aR axis is known to play an important role in complement-mediated thromboinflammation associated with a wide array of diseases ( 27 , 30 ), and the monoclonal anti-C5 antibody eculizumab has been used clinically for many years to treat complement-driven diseases, such as paroxysmal nocturnal hemoglobinuria and atypical hemorrhagic uremic syndrome ( 31 ).…”

“…The C5a-C5aR axis is known to play an important role in complement-mediated thromboinflammation associated with a wide array of diseases (27,30), and the monoclonal anti-C5 antibody eculizumab has been used clinically for many years to treat complement-driven diseases, such as paroxysmal nocturnal hemoglobinuria and atypical hemorrhagic uremic syndrome (31). C3a receptors have been shown on both platelets and activated astrocytes, neutrophils, and monocytes (32,33), and we have previously shown how C3 inhibition attenuates air-induced thromboinflammation (15). Recently, a specific C3 inhibitor, pegcetacoplan (Empaveli, Apellis Pharmaceuticals, Waltham, MA), was approved to treat paroxysmal nocturnal hemoglobinuria (34), and further studies should evaluate if C3 inhibitors can be used to dampen thromboinflammation triggered by air embolism.…”

“…The system consists of several plasma peptides, which bind to antibodies, lectins, or foreign surfaces, resulting in classical, lectin, or alternative pathway activation, respectively ( 27 ). It has previously been shown that air activates the complement system mainly through the alternative pathway ( 14 , 15 ) by hydrolysis of C3 to generate C3(H 2 O), also termed C3b-like or iC3 ( 14 ). This initial alternative C3 activation generates the C3(H 2 O)B, where factor B is cleaved by factor D to form Ba, which is released, and Bb, which together with properdin (P) binds to C3(H 2 O) and forms the first alternative C3 convertase, C3(H 2 O)BbP.…”

“…Patient case reports have described prolonged severe inflammation after air embolism (6,10), and one case report described a suspected link between perioperative air embolism and disseminated intravascular coagulation, albeit with an unknown pathophysiological mechanism (11). In vitro experiments in human serum and heparinized blood have shown that air activates C3 to C3(H 2 0), also termed iC3 (12)(13)(14), and in vitro studies in lepirudin anticoagulated human whole blood have shown that air emboli trigger a complement C3-driven thromboinflammation (15), which is attenuated by C3 inhibition (15). Ex vivo rat studies have shown how air emboli trigger a pulmonary inflammation involving both monocytes and granulocytes and complement (16), recent in vivo studies of divers suffering from decompression sickness have shown how microbubbles trigger an acute inflammation (17), and human in vivo and in vitro studies of bubble-oxygenators has shown that bubbles activate C3 in fully heparinized blood (12).…”

“…However, these measures only address the immediate emergency treatment and do not address the management of air-induced thromboinflammation. Based on our recent in vitro human whole blood study ( 15 ), we hypothesize that C3 activation plays a key role and that C3 inhibition can attenuate the detrimental thromboinflammatory effects of air embolism. Unfortunately, no C3 inhibitor shown to work in swine is available for therapeutic use in porcine studies.…”