Activation of the Complement System During and After Cardiopulmonary Bypass Surgery
Postsurgery Activation Involves C-Reactive Protein and Is Associated With Postoperative Arrhythmia
Background Complement activation during cardiopulmonary bypass (CPB) surgery is considered to result from interaction of blood with the extracorporeal circuit. We investigated whether additional mechanisms may contribute to complement activation during and after CPB and, in particular, focused on a possible role of the acute-phase protein C–reactive protein (CRP).
Methods and Results In 19 patients enrolled for myocardial revascularization, perioperative and postoperative levels of complement activation products, interleukin-6 (IL-6), CRP, and complement-CRP complexes, reflecting CRP-mediated complement activation in vivo, were measured and related to clinical symptoms. A biphasic activation of complement was observed. The ratio between the areas under the curve of perioperative and postoperative C3b/c and C4b/c were 3:2 and 1:46, respectively. IL-6 levels reached a maximum at 6 hours post-surgery. CRP levels peaked on the second postoperative day. Each complement-CRP complex had peak levels on the second or third postoperative day. By multivariate analysis, maximum levels of CRP on the second postoperative day were mainly explained by C4b/c levels after protamine administration, leukocyte count on the second postoperative day, and preoperative levels of CRP. Peak levels of C4b/c after protamine administration (P=.0073) and on the second postoperative day correlated with the occurrence of arrhythmia on the same day (P=.0065).
Conclusions Cardiac surgery with CPB causes a biphasic complement activation. The first phase occurs during CPB and results from the interaction of blood with the extracorporeal circuit. The second phase, which occurs during the first 5 days after surgery, involves CRP, is related to baseline CRP levels, and is associated with clinical symptoms such as arrhythmia.
Cardiopulmonary bypass induces a systemic inflammatory response by triggering the production and release of a multitude of inflammatory mediators.1 During surgery and in the early postoperative stage, the extent of this inflammatory response is associated with clinical symptoms such as hemodynamic instability, fever, bleeding disorders, and organ failure in severe cases.2–4 These symptoms are generally well controlled by adequate postoperative treatment. On the first and second postoperative days, however, patients generally develop fever, leucocytosis, tissue edema, and sometimes arrhythmia (ie, atrial fibrillation), which may lead to a prolonged hospital stay.5,6
Many investigators reported on the activation of the complement system during or shortly after CPB and on the release of proinflammatory cytokines such as tumor necrosis factor-α and IL-6. Complement activation during CPB occurs mainly through the alternative pathway and is induced by the contact of blood with the surface of the extracorporeal circuit.2,7–9 On administration of protamine and the subsequent formation of heparin-protamine complexes, complement is further activated through the classic pathway.4,10 Although the extent of complement activation during CPB correlates with the severity of the operation and the development of complications,2 most clinical problems do not occur until the first or second postoperative days.11–13 This is intriguing, considering that in animal models, most complement-mediated biological effects develop within hours of activation of the system.14
Patients undergoing CPB experience a rise in body temperature and leucocytosis, starting 24 hours after surgery. These symptoms point to an ensuing APR, which is characterized biochemically by changes in levels of various acute-phase proteins. CRP is the prototypical acute-phase protein in humans. Although its function in vivo is poorly understood, one of the in vitro functions attributed to CRP is its potency to activate complement.15–18 It is unknown whether CRP contributes to complement activation in patients during or after CPB. Recently, we developed methods to assess CRP-mediated complement activation in vivo and observed activation of complement by CRP in patients receiving renal allografts.19
In the present study, we used these methods to assess the activation of complement during and after CPB and the participation of CRP therein. Activation of complement by CRP was measured by determining levels of complexes between CRP and activated C4 or C3.19 Additionally, these complement activation parameters were related to clinical symptoms such as the occurrence of arrhythmia, fever, and leucocytosis.
We prospectively studied 19 patients (16 men, 3 women) 60 years of age (median; range, 51 to 75 years) undergoing CPB for myocardial revascularization. All patients had good left ventricular function and were free of diseases other than coronary artery disease. Patients using corticosteroids or other anti-inflammatory drugs except for low doses of aspirin, ie, 80 mg until as late as 3 days before the operation, were excluded from the study. Patients gave written informed consent and were evaluated according to a protocol approved by the Medical Ethics Committee of the Academic Medical Center of Amsterdam.
On the morning of surgery, patients received their usual dose of antianginal drugs and 2 to 4 mg lorazepam as premedication. Anesthesia was induced intravenously with 0.2 mg/kg etomidate, 50 μg/kg fentanyl, 0.1 mg/kg pancuronium bromide, and 0.1 mg/kg midazolam and was maintained by supplemental doses. After endotracheal intubation, patients were ventilated to normocapnia with oxygen and air mixture. Cefamandol (2 g) was given intravenously for infection prophylaxis. A radial artery catheter and a flow-directed pulmonary artery catheter (Swan-Ganz, Baxter/American Edwards Laboratories) were inserted for hemodynamic measurements and collection of blood samples. Volume was supplemented and further deviations from pressures or vascular resistances were treated with appropriate vasoactive medication.
The extracorporeal circuit consisted of a soft shell, closed, venous reservoir (BMR 1900, Baxter Healthcare Corp), roller pump, hollow fiber oxygenator with integrated heat exchanger (Univox, Baxter), arterial filter, cardiotomy reservoir, and polyvinyl tubing system. The extracorporeal circuit was primed with 1500 mL Ringer’s lactate solution, 200 mL aprotinin (2×106 KIU Trasylol, Bayer), and 100 mL of 20% mannitol. Magnesium sulfate 0.1 g/kg and 5000 IE bovine heparin (Leo Pharmaceutical Products) were added to the priming solution. Total priming volume was 1800 mL.
After systemic heparinization (250 U/kg), CPB was initiated with cannulas placed in the ascending aorta and right atrium (two-stage venous canula). Activated clotting time was kept >500 seconds with additional heparin. A nonpulsatile roller pump was used for all operations. The nonpulsatile flow rate was maintained at 2.4 L · min−1 · m−2 during cooling and rewarming phases. Patients were cooled to 27°C to 30°C at a flow rate of 1.8 L · min−1 · m−2. For myocardial protection, patients received high-potassium cold crystalloid cardioplegia (800 to 1000 mL, potassium 20 mmol/L, 4°C, St Thomas, Academic Medical Center) during aortic cross clamping. During aortic cross clamping, mechanical ventilation was interrupted, and the lungs were at rest with static inflation (5 cm H2O, 21% oxygen). Disturbances in the acid-base balance were appropriately treated. The hematocrit during CPB was maintained at 18% to 25%. Distal anastomoses of the grafts were placed during aortic cross clamping, and proximal anastomoses were placed after cross-clamp removal and restoration of mechanical ventilation. After termination of CPB, heparin was antagonized with protamine sulfate at a 1:1 ratio (3 mg/kg). If necessary, inotropic support was given when patients were weaned from CPB. Autologous blood and residual volume from the extracorporeal circuit were infused into the patient when volume supplementation was necessary.
After surgery, patients were admitted to the ICU and treated according to a standardized protocol. Mean arterial blood pressure was kept at 65 to 80 mm Hg, heart rate at 70 to 80 bpm, pulmonary artery wedge pressure at 8 to 12 mm Hg, and cardiac index >2.5 L · min−1 · m−2. Cardiotonic support was administered when necessary. Fluid balance, rectal temperature, and hemodynamic parameters were recorded every hour. Patients were ventilated to normocapnia and an arterial oxygen tension >80 mm Hg with continuous positive-pressure ventilation and positive end-expiratory pressure of 5 cm H2O until extubation according to the ICU regimen. Basic fluid administration consisted of 0.9% NaCl and 4% modified fluid gelatin (Gelofusine, Vifor Medical SA). Packed erythrocytes were infused when the hematocrit was <26%. When their cardiorespiratory condition had stabilized, patients were transported to the ward for further recovery.
Collection of Blood Samples
Blood specimens for hemoglobin, hematocrit, white blood cell numbers, and platelet counts were collected in 5-mL glass Vacutainer tubes containing EDTA (Becton Dickinson). Blood samples for analysis of complement activation products, CRP, and interleukins were collected in 5-mL siliconized Venoject tubes (Terumo Europe NV) containing 10 mmol/L EDTA, 0.1 mg/mL Soybean Trypsin Inhibitor (Sigma Chemical Co), and 25 mmol/L benzamidin (Agros Organics). During surgery and in the ICU, all samples were taken through a radial artery catheter; the other blood specimens were obtained by venous puncture. Plasma was prepared by centrifugation of the blood for 20 minutes at 1500g immediately after collection. The samples were stored at −70°C. Blood samples were obtained at the following time points: after the induction of anesthesia (before median sternotomy); 30 minutes after the start of CPB; immediately after CPB but before protamine administration; during closure of the sternum; 1 and 6 hours after arrival at the ICU; and on the 1st, 2nd, 3rd, 4th, and 5th postoperative days.
C3a levels were determined by radioimmunoassay as described previously20 and expressed as nanomoles per liter. Normal values of C3a are <6 nmol/L. C3b/c (ie, C3b, C3bi, or C3c) and C4b/c (ie, C4b, C4bi, or C4c) concentrations were determined with ELISAs as described.21–23 Briefly, C3b/c was measured with mAb anti-C3-9 as capture antibody, which binds to C3b, iC3b, C3c, as well as iC3 (C3b-like C3).24 Biotinylated polyclonal rabbit antibodies to human C3c were used as detecting antibody. C4b/c was measured by use of mAb anti-C4-1, which binds to a neoepitope exposed on C4b, iC4b, C4c, as well as iC4, as capture antibody. Bound C4b/c was detected with biotinylated anti-C4c antibodies. Normal values of C3b/c and C4b/c are <50 nmol/L.
Total C3, C4, and IgG concentrations were determined by nephelometry (C3c, C4, and human IgG, Behring Diagnostics Benelux NV; Behring Nephelometer Analyzer, Behringwerke AG) according to manufacturer protocol.
Plasma IL-6 was determined by ELISA (CLB, Dept Immune Reagents) as previously described.25 Normal values of IL-6 are <10 ng/L. CRP levels were measured with a sandwich-type ELISA in which polyclonal rabbit anti-CRP antibodies were used as catching antibodies and a biotinylated mAb against CRP (CLB anti–CRP-2) as the detecting antibody. Results were related to a standard consisting of commercially available CRP (Behringwerke AG) and expressed as milligrams per liter. The detection limit of the assay was 10 ng/L. CRP levels in healthy persons are <3 mg/L.
Complement-CRP complexes were measured with sensitive ELISAs as described previously.19 Briefly, complement proteins fixed to CRP were separated from unbound complement proteins by absorption onto phosphorylcholine-Sepharose (Pharmacia Fine Chemicals). For quantification of C3b-CRP, purified mAb anti-C3-9 was used as a catching antibody; for that of C3d-CRP complexes, mAb anti-C3-19; for that of C4b-CRP mAb, anti-C4-1; and for that of C4d-CRP, mAb anti-C4-4. It is to be noted that the ELISAs for C3b-CRP and C4b-CRP complexes also detect C3bi-CRP and C4bi-CRP, respectively. Similarly, the ELISA for C3d-CRP complex detects C3d-CRP, C3b-CRP, and C3bi-CRP, and the ELISA for C4d-CRP complex detects C4d-CRP, C4b-CRP, and C4bi-CRP, respectively.19 Complement-CRP complex levels in healthy volunteers are <4 pmol/L.
Data were stored and analyzed with standard computer software (SPSS 6.1.3, SPSS Inc). To analyze changes in time, one-factor ANOVA for repeated measures was applied, supplemented with the Scheffé post hoc test. Regression analysis was used to assess correlation between parameters. A two-sided probability value of P<.05 was considered statistically significant. Values are presented as medians with interquartile range unless stated otherwise.
The demographic and surgical data of the patients included are shown in Table 1⇓. Each patient had effort-induced angina pectoris resistant to antianginal medication and multivessel coronary artery disease documented by coronary cineangiography. Ten patients had a history of previous myocardial infarction. All patients survived the procedure and had a normal recovery, although 2 patients required surgical exploration for postoperative bleeding, including 1 patient who had an intra-aortic balloon inserted because of poor hemodynamic performance.
In the first 18 hours after surgery, the body temperature of the patients normalized (median, 37.5°C; interquartile range, 37°C to 38°C). Postoperative fever occurred in all patients on the second and third postoperative days (38.4°C; range, 38.1°C to 38.8°C). This phenomenon was accompanied by a significant twofold increase in polymorphonuclear cells (P<.01). Maximum values of polymorphonuclear cells occurred after protamine administration (9.5×109 cells/L; range, 7.3 to 11.4 ×109 cells/L) and on day 2 (10.8×109 cells/L; range, 8.9 to 13.3×109 cells/L).
Postoperative arrhythmia, ie, supraventricular tachyarrhythmia or atrial fibrillation with fast ventricular rate (120 to 180 bpm), requiring anti-arrhythmic therapy occurred in 7 of 19 patients (ie, 2 patients on day 2, 4 patients on day 3, and 1 patient on day 4).
C3a levels after induction of anesthesia were low and increased during CPB, resulting in peak levels after protamine administration (58 nmol/L; range, 39 to 71 nmol/L) (Fig 1A⇓). C3a levels then declined within the first 24 hours to pre-CPB levels. Levels of C3b/c also increased after the initiation of CPB, resulting in maximum levels after protamine administration (461 nmol/L; range, 370 to 655 nmol/L), and began to decline thereafter to pre-CPB levels on the first postoperative day (day 1; Fig 1B⇓ and Table 2⇓). On day 2, however, C3b/c levels started to increase again to reach maximum levels on day 4 (70 nmol/L; range, 33 to 82 nmol/L) and remained increased on day 5. C4b/c levels decreased significantly during the first minutes of CPB, mainly because of hemodilution (Table 2⇓). Thereafter, levels slightly increased, particularly after protamine administration, at which point they were significantly higher than baseline levels (57 nmol/L; range, 45 to 88 nmol/L; Fig 1C⇓ and Table 2⇓). On days 2, 3, and 4, a secondary and significant increase was observed (67 nmol/L; range, 47 to 89 nmol/L). Levels of C4b/c declined thereafter to pre-CPB levels.
For each patient, we compared perioperative with postoperative levels by calculating area under the curves for these periods, representing the total amount of complement activation products of both the alternative and the classic pathway for both C3b/c and C4b/c levels. It appeared that the mean ratio of perioperative to postoperative values was 3:2 for C3b/c compared with 1:46 for C4b/c.
During the CPB procedure, the blood is considerably diluted, which should be taken into account when complement activation markers are measured. We determined the total concentrations of circulating C3 and C4 and corrected these concentrations for dilution by the levels of total IgG (Table 2⇑). The total amounts for C3 and C4 dropped after the start of the CPB procedure, which was explained completely by the hemodilution. After several days, this drop was completely converted into an increase in C3 and C4 production, which was induced by the APR. When we calculated the percentage of activated C3 and C4 of the total amounts present and corrected for hemodilution, we still found a highly significant increase in C3 activation during the procedure, whereas enhanced C4 activation was detected only after protamine administration.
Plasma IL-6 levels increased at the end of CPB, reaching a maximum after 6 hours in ICU (112 ng/L; range, 56 to 184 ng/L; Fig 2A⇓) and then gradually returning to pre-CPB levels within 4 days. CRP levels increased in all patients from 6 hours in ICU, reaching maximum levels on day 2 (60 mg/L; range, 47 to 72 mg/L; Fig 2B⇓). Thereafter, CRP levels gradually declined, although they were still elevated on day 5. Complement-CRP complexes increased significantly (P<.01) in all patients postoperatively, reaching maximum levels on days 2 and 3. C4d-CRP complex levels (591 pmol/L; range, 335 to 750 pmol/L) were significantly higher than C4b-CRP levels (34 pmol/L; range, 14 to 88 pmol/L), whereas C3b-CRP (103 pmol/L; range, 72 to 159 pmol/L) and C3d-CRP (81 pmol/L; range, 50 to 98 pmol/L) levels were within the same range (Fig 3A⇓ through 3D).
Correlation Between Complement Parameters and Clinical Symptoms
Multivariate analysis was used to assess correlations between various parameters. C4d-CRP on day 2 correlated with levels of C4b/c on the same day (R2=.36, P=.009). Maximum CRP levels (ie, days 1 to 3) correlated with peak temperature during the first 18 hours after surgery (R2=.31, P=.017). The following independent variables for explanation of the CRP level on the second postoperative day were entered in a forward stepwise regression analysis: leukocyte count after protamine administration and on day 2; C3b/c at baseline and after protamine infusion; C4b/c at baseline, after protamine, and on day 2; and CRP at baseline. The level of CRP on day 2 was explained by three variables: CRP day 2=97.3−5.1×(leukocyte count day 2)+0.26×(C4b/c after protamine)+11.8×(CRP baseline); the adjusted R2 for this model was .61 (P=.0022). The confidence interval for leukocyte count day 2 was −7.5 to −2.8; for that of C4b/c after protamine, 0.09 to 0.43; and for that of CRP at baseline, 0.6 to 22.9.
IL-6 did not contribute to the model; maximum IL-6 levels showed no correlation (R2=.1, P=.46) with maximum CRP levels.
Differences in complement activation and CRP levels between patients with and without arrhythmia are depicted in Table 3⇓. Significant differences between patient groups occurred in terms of the peaks of the classic pathway of complement activation, ie, after protamine administration, and on day 2. Additionally, C4b/c values at baseline were different in the patients with or without arrhythmia. Moreover, C4d-CRP complexes after protamine administration and on day 2 differed between each group. C4b/c at baseline and the C4d-CRP level on day 2 also correlated with the occurrence of arrhythmia (R2=.41, P=.0037; Table 4⇓). The occurrence of arrhythmia also correlated with the levels of C4b/c after protamine administration (R2=.35, P=.0073) and on day 2 (R2=.37, P=.0065).
Most studies have focused on the activation of the complement system during the operation. This study confirms that during extracorporeal circulation, the complement system is activated mainly through the alternative pathway, whereas classic pathway activation occurs after the administration of protamine.7,9,22 Yet most clinical events affecting the recuperation of the patient occur during the first three days after surgery, during which period patients develop an APR.26 In this prospective study, we showed a marked APR in patients after cardiac surgery, even though these patients preoperatively were considered to be at low risk for adverse events in the postoperative period. Remarkably, this APR coincided with renewed activation of the complement system; the ratio between area under the curve of C4 activation products during or shortly after CPB and that during the APR was 1:46, whereas that for C3 activation products was 1.6:1. Thus, the classic complement pathway apparently was activated during the APR, which proceeded for several days.
The APR was biochemically marked by the enhanced production of the prototypical acute-phase protein CRP and by the levels of total C3 and C4. CRP levels rose markedly starting at 6 hours after surgery, culminating on the day 2. This lag between stimulus and maximum CRP levels is consistent with values reported in the literature,26,27 as is the magnitude of the increase (ie, median, 234-fold up to 1000-fold).26 CRP is able to activate the complement system through the classic pathway as was shown previously in vitro15,16,28 and recently in vivo.19 During the APR induced by cardiac surgery, CRP mediated complement activation at least in part, as was evident from increasing C3- and C4-CRP complexes on days 1 to 5. C3d-CRP complex levels, which specifically are generated during CRP-mediated complement activation,19 started to rise already during the operation, indicating that baseline values of CRP, although within the normal range (ie, <3 mg/L), are involved in complement activation before CRP levels rose and that some classic pathway activation (ie, triggered by CRP) occurs during cardiac surgery in addition to the more prominent alternative pathway activation. Moreover, protamine administration induced a rise in C4-CRP complexes as well as C3-CRP complexes. Thus, in line with in vitro observations,28 protamine may also induce CRP-mediated complement activation in vivo.
All patients had increased complement-CRP complexes during the first postoperative days. C4d-CRP levels were approximately 10-fold higher than the C4b-CRP levels, whereas C3b-CRP and C3d-CRP levels were comparable. This supports results of in vitro studies showing that C4b fixed to CRP is rapidly inactivated to C4d, whereas C3b is not.19 Apparently, this differential breakdown of C4b and C3b also occurs in vivo.
IL-6 is well known for its capability to induce the production of acute-phase proteins by the liver.26 In agreement with our results, CRP levels in the patients did not increase until IL-6 was released. Yet no correlation between CRP and IL-6 levels was found, which may have been caused by differences in clearance rates.
Our study does not allow conclusions regarding the ligand for CRP. In the presence of Ca2+ ions, CRP binds specifically to phosphatidylcholine,15 in particular when some lysophosphatidylcholine is present.17 Hence, as we have discussed elsewhere,18 a supposed ligand for CRP in inflamed tissues may be the membranes of flip-flopped cells or microvesicles derived from these membranes by hydrolyzation by secretory PLA-2. Long-chain acylcarnitines and lysophosphatidylcholines are generated from phosphatidylcholine by PLA2 enzymes and can both contribute to membrane dysfunction by inhibiting the exchange of sodium and calcium ions in sarcolemmal vesicles and lead to development of arrhythmia.29 This may explain the association between raised levels of CRP, complement-CRP complexes, and activated complement and the occurrence of arrhythmia. Alternatively, this association reflected the generation of active complement fragments in the myocardium, which, through mechanisms not completely resolved, may induce arrhythmia.30,31
Changes in levels of CRP, complement, and complement-CRP complexes during the first days after surgery corresponded to clinical phenomena like leucocytosis, fever, and the occurrence of arrhythmia in the same period. Multiple regression analysis revealed that baseline levels of CRP were correlated with the levels of CRP on day 2, suggesting that “high-CRP responders” can be distinguished from “low-CRP responders” by baseline CRP levels. Therefore, preoperative CRP levels may indicate an increased risk for the occurrence of adverse events. Yet baseline CRP levels were within the normal range, ie, <3 mg/L, and did not discriminate between patients with or without arrhythmia. Currently, we are extending these studies to establish the extent to which baseline CRP levels may predict an increased risk for arrhythmia and other complications in the late postoperative period. The incidence of arrhythmia in the postoperative period was 37% in this small group, which is in accordance with the literature.13 The occurrence of arrhythmia during the postoperative period after cardiac surgery is influenced by several factors, ie, operation procedure, severity of coronary artery disease, and withdrawal of β-sympathicolytic drugs.5,32,33 Because postoperative arrhythmia in general extends the hospitalization of these patients by at least 1 day, preoperative administration of antiarrhythmic drugs seems attractive to prevent the development of arrhythmia. Our results not only suggest that preoperative C4b/c levels may be used to select patients for such a pretreatment but also imply that anti-inflammatory drugs may be useful to prevent arrhythmia.
In conclusion, we demonstrate that complement is activated not only during CPB but also during the first days thereafter, that increasing CRP levels contribute to this activation, and that this activation is associated with the occurrence of adverse events such as arrhythmia.
Selected Abbreviations and Acronyms
|CRP||=||protein C–reactive protein|
|ICU||=||intensive care unit|
We thank Yvonne Lubbers and Gerard van Mierlo for their excellent assistance with the assays.
- Received April 24, 1997.
- Revision received August 6, 1997.
- Accepted August 7, 1997.
- Copyright © 1997 by American Heart Association
Moat NE, Shore DF, Evans TW. Organ dysfunction and cardiopulmonary bypass: the role of complement and complement regulatory proteins. Eur J Cardiothorac Surg. 1993;7:563–573.
Jansen PGM, te Velthuis H, Huybregts MAJM, Paulus R, Bulder ER, van der Spoel HI, Bezemer PD, Slaats EH, Eijsman L, Wildevuur CRH. Reduced complement activation and improved postoperative performance after cardiopulmonary bypass with heparin-coated circuits. J Thorac Cardiovasc Surg. 1995;110:829–834.
De Boer JP, Creasey AA, Chang A, Roem D, Eerenberg AJM, Hack CE, Taylor FB Jr. Activation of the complement system in baboons challenged with live Escherichia coli: correlation with mortality and evidence for a biphasic activation pattern. Infect Immun. 1993;61:4293–4301.
Kaplan MH, Volanakis JE. Interaction of C-reactive protein complexes with the complement system. J Immunol. 1974;6:2135–2147.
Volanakis JE. Complement-induced solubilization of C-reactive protein-pneumococcal C-polysaccharide precipitates: evidence for covalent binding of complement proteins to C-reactive protein and to pneumococcal C-polysaccharide. J Immunol. 1982;128:2745–2750.
Wolbink GJ, Brouwer MC, Buysmann S, ten Berge IJM, Hack CE. CRP-mediated activation of complement in vivo: assessment by measuring circulating complement-CRP complexes. J Immunol. 1996;157:473–479.
Jansen PGM, te Velthuis H, Wildevuur WR, Huybregts MAJM, Bulder ER, van der Spoel HI, Sturk A, Eijsman L, Wildevuur CRH. Cardiopulmonary bypass with modified fluid gelatin and heparin-coated circuits. Br J Anaesth. 1996;76:13–19.
Hack CE, Paardekooper J, Smeenk RJT, Abbink JJ, Eerenberg AJM, NuijensJH. Disruption of the internal thioester bond in the third component of complement (C3) results in the exposure of neodeterminants also present on activation products of C3: an analysis with monoclonal antibodies. J Immunol. 1988;141:1602–1609.
Helle M, Boeije L, de Groot ER, de Vos A, Arden LA. Sensitive ELISA for interleukin 6: detection of IL-6 in biological fluids: synovial fluids and sera. J Imunnol Methods. 1991;138:47–56.
Mackiewicz A, Kushner I, Baumann H. Acute Phase Proteins: Molecular Biology, Biochemistry, and Clinical Applications. Boca Raton, Fla: CRC Press, Inc; 1993.
Redl H, Schlag G, Schiesser A, Davies J. Tumor necrosis factor is a mediator of phospholipase release during bacteremia in baboons. Am J Physiol. 1993;264:H2119–H2123.
Siegel J, Rent R, Gewurz H. Interactions of C-reactive protein with the complement system. J Exp Med. 1974;140:631–647.
Wu J, Corr PB. Influence of long chain acylcarnitines on the voltage-dependent calcium current in adult ventricular myocytes. Am J Physiol. 1992;263:H410–H417.
Del Balzo U, Polley MJ, Levi R. Cardiac anaphylaxis: complement activation as an amplification system. Circ Res. 1989;65:847–857.
Del Balzo UH, Levi R, Polley MJ. Cardiac dysfunction caused by purified human C3a anaphylatoxins. Proc Natl Acad Sci U S A. 1985;82:886–890.