Cardiac surgery with cardiopulmonary bypass, often attendant with periods of myocardial ischemia-reperfusion (I/R), can induce a coagulopathy, a systemic and local inflammatory response, and transient myocardial dysfunction. The serine protease inhibitor aprotinin (APRO) was used in this clinical context to minimize blood loss.1,2 Although APRO was considered as a mainstay for improving hemostasis in the context of I/R and cardiac surgery, recent retrospective analysis has identified significant concerns regarding APRO.3 Moreover, a prospective study using the clinical full Hammersmith protocol was terminated prematurely.4 Thus, although APRO favorably affects hemostasis, which is an important determinant of morbidity and mortality in I/R and cardiac surgery, APRO likely influences other biological processes which impart deleterious effects. Studies have shown a strong association between the use of APRO and the reduction of various inflammatory markers, particularly tumor necrosis factor-α (TNF) and interleukin-6 (IL-6).5,6 Accordingly, the overall goal of the present study was to examine the potential relationship between left ventricular (LV) contractility and cytokine release with varying doses of APRO in the context of I/R. It has been shown that TNF levels are elevated in the setting of myocardial I/R, leading to a reduction of LV function through several pathways.7,8 IL-10 was demonstrated to have antiinflammatory effects which can induce TNF.9 Some studies have shown that cytokine release can be attenuated by APRO in the setting of myocardial I/R.10,11 However, little is known about the overall mechanisms by which APRO attenuates the response to I/R, whether these effects of APRO are dose dependent, and whether they lead to a preservation LV function after I/R. Accordingly, this study examined whether and to what degree APRO imparts a dose-dependent effect on LV contractility and on the release of specific cytokines (TNF, IL-6, IL-10).
Instrumentation and Animal Model
Adult mice (FVB strain, 10–16 wk, 24–30 g) were anesthetized, tracheally intubated with a 20-gauge Jelco needle (Medex Medical Ltd., Rossendale, UK), and maintained under isoflurane anesthesia (2%) using a MiniVent Type 845 ventilator (Hugo Sachs Elektronik, March-Hugstetten, Germany) with tidal volumes of 250 μL, at a rate of 250 strokes/min, and a Fio2 of 27%. Temperature was monitored via a rectal probe during the length of the procedure and maintained by a feedback loop to a heating pad within the operating table, as well as a heating lamp. It was previously determined by conducting baseline studies that the ventilator settings gave a pH of 7.35 ± 0.01, Pco2 of 29 ± 2, Po2 of 453 ± 34, and an O2 saturation of 100%. The right carotid was exposed and a precalibrated four-electrode-pressure sensor catheter (1.4F, SPR-839, Millar Instruments, Houston, TX) was placed in the LV with pressure tracing confirmation.
A left thoracotomy was then performed, the posterolateral aspect of the LV-free wall visualized, and a purse-string placed around the left anterior descending artery just distal to the bifurcation of the left main coronary artery using 6.0 Prolene and an atraumatic needle (Ethicon, Somerville, NJ). The suture was exteriorized and the wound was closed in layers. The ligature was tightened to induce ischemia (30 min) and then released for reperfusion (60 min). In a preliminary set of studies (n = 6), fluorescent microspheres (F-8838, Molecular Probes, 15 μm diameter, 7.5 × 104) of different emission spectra were injected at baseline, at 30 min of ischemia, and at 60 min of reperfusion by LV injection methods described previously.12 LV regional myocardial blood flow decreased to approximately 50% of baseline values with peak ischemia and returned to within baseline values with reperfusion. Thus, this murine model provided a transient period of low myocardial blood flow followed by a restoration of blood flow and, therefore, allowed for the study of LV function in the context of I/R. Furthermore, it was documented using this coronary occlusion model that the area at risk for the LV was 50%. At the end of the reperfusion period, the pressure sensor catheter was removed, the mouse was injected with 0.2 mL of heparin, killed, and blood was collected from the right carotid artery for cytokine and APRO analysis. Total procedure time was 120 min. The mice received a one-time intraperitoneal normal saline bolus of 0.5 mL. All animals were treated and cared for in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (National Research Council, Washington, DC, 1996). This protocol was reviewed and approved by the MUSC Institutional Animal Care & Use Committee (AR# 2345).
APRO doses of 2, 4, and 8 mL/kg were modeled to achieve plasma concentrations corresponding to half-Hammersmith, Hammersmith, and double-Hammersmith doses, respectively, while using the standard concentration APRO of 10,000 kallikrein-inhibiting units (KIU)/mL.13,14 Royston et al.15 reported a model in which an APRO 4 mL/kg IV bolus reached plasma concentrations (>250 KIU/mL) comparable with the Hammersmith, or high-dose, protocol often used clinically. Our approach was to use this clinical dosing regimen using the weight-based initial bolus. Furthermore, this is similar to weight-based dosing regimens used in previous animal models.16,17 For clarity and uniformity, APRO doses are reported as 2 × 104 KIU/kg for the half-Hammersmith, 4 × 104 KIU/kg for the Hammersmith, and 8 × 104 KIU/kg for the double-Hammersmith doses.
To measure relative plasma levels of APRO, a fluorogenic substrate cleaved by the serine protease plasmin was used in an ex vivo assay system. Specifically, the peptide sequence d-ala-leu-lys-7-amido-4-methylcomarin (Sigma, A8171), at a fixed concentration of 10 nM, was mixed in a reaction buffer containing a 1:33 dilution of normal mouse plasma and incubated at 37°C for 15 min in the presence and absence of 7 μg/mL of plasmin (Sigma, P1876, 3 U/mg). The fluorescence of this reaction was detected in continuous fashion (Fluostar Galaxy, BMG Labtech, NC) at an excitation/emission wavelength of 365/440 nm. The plasmin substrate, plasmin concentrations, and the incubation conditions were determined from preliminary dilution studies to yield peak performance as defined as that which yielded a consistent and stable fluorescence signal. This reaction solution was then incubated in the presence and absence of increasing concentrations of APRO (range, 0–560 KIU/mL) to generate a standardized enzyme activity-inhibition curve. The standard APRO inhibition curve, which was generated in triplicate, along with the 95% confidence interval for this standard curve is shown in Figure 1. This inhibition curve demonstrated the classical exponential decay and was subjected to regression analysis yielding a significant relationship between the reduction in fluorescence to APRO concentrations (r = 0.99, P < 0.001). The intraassay coefficient of variation was 5%, and an interassay coefficient was 9%. This APRO enzyme inhibition assay was used to extrapolate relative APRO plasma concentrations. Specifically, at the completion of the studies described in the subsequent paragraphs, plasma was prepared and incubated in the substrate/plasmin substrate, the relative fluorescence obtained, input into the standardized APRO inhibition curve, and an APRO concentration computed.
Fifty mice were randomized to one of four treatment groups: Vehicle (0.9% saline), APRO 2 × 104 KIU/kg, APRO 4 × 104 KIU/kg, or APRO 8 × 104 KIU/kg. Randomization was accomplished by drawing group assignments from an envelope. After baseline hemodynamic measurements were taken, the assigned group dose of saline or APRO was given by intraperitoneal injection using an equivalent final volume (0.5 mL). Ischemia was initiated 5–10 min after the injection.
LV Contractility Measurements
After stabilization from the instrumentation, LV pressures (peak, diastolic) were measured along with relative volume units which were converted to true volumes using a calibration curve. For this purpose, the pressure sensor catheter was interfaced to a pressure-conductance unit (ARIA, MPCU-200, Millar) in which electrical excitation was performed under digital control (DAQ, PV Analysis Software, Millar). The continuous digital pressure and conductance signals were integrated with an electrocardiograph signal (PowerLab, AD instruments, NSW Australia) and displayed in real-time using dual-display flat screen monitors (Sony Electronics, New York, NY). This allowed for the optimal placement of the LV catheter with respect to the LV conductance signal. The LV conductance signal was converted to a LV volume using an algorithm described previously.18 This included correction for parallel conductance and continuous in vitro calibration. Steady-state LV pressures and conductance volumetry were determined by digitizing a minimum of 12 consecutive cardiac cycles. LV peak pressure, end-diastolic pressure, and maximal rate of LV peak pressure (dP/dt) were recorded. The ventilator was then suspended during continuous recording of the LV pressure-conductance volume signal to alter venous return. This maneuver resulted in a family of LV pressure-volume loops in which definable points of the LV end-systolic pressure-volume relationship (ESPVR) could be determined. Representative LV pressure-volume loops at baseline, peak ischemia, and reperfusion are shown in Figure 2a and demonstrate the linearity of the end ESPVR. Isochronal points from the recorded LV pressure-volume loops were used to compute Emax, an index of LV contractility defined as maximal LV elastance, and ESPVR.19–21 These indices were calculated for each mouse at each time point from pressure volume loops, adjusted for parallel conductance, generated from the readings obtained. Emax was indexed to LV mass (Emax/mg), which was obtained gravimetrically at autopsy and had a mean value of 81.1 ± 9.8 mg.
Diluted mouse plasma samples taken after 60 min of reperfusion were diluted appropriately to bring the analyte into a measurable range and assayed according to manufacturer’s instructions (Cat. # LUM410, LUM406, LUM417, R&D Systems, Minneapolis, MN) for TNF, IL-6, and IL-10. The relative fluorescence detected was compared with a 5-parameter logistic calibration curve generated independently for each analyte tested (Bio-Plex Manager 4.1.1). From the calibration curve, the concentration of each analyte was determined, and the plasma concentration was calculated. The sensitivity for each of the measured plasma cytokines was 0.42 pg/mL for TNF, 0.71 pg/mL for IL-6, and 0.59 pg/mL for IL-10.
To examine whether and to what degree anesthesia and/or instrumentation affected LV contractility in the absence of I/R, Emax was assessed at baseline and at 90 min, which was equivalent to the end point of the I/R study. Specifically, saline was administered intraperitoneally in mice (n = 9), and Emax was measured. Furthermore, in a second control study, Emax was measured in a similar fashion in mice (n = 8) which received intraperitoneal APRO (8 × 104 KIU/kg) without I/R. To further address whether APRO affected plasma cytokine levels, plasma cytokine concentrations were measured at the 90-min time point in subsets of mice injected with normal saline (n = 6), 2 × 104 KIU/kg of APRO (n = 6), 4 × 104 KIU/kg of APRO 9 (n = 6), and 8 × 104 KIU/kg of APRO (n = 5), and were used as referent controls.
Investigators performing LV function and biochemical analyses were blinded to group assignments. All values were entered in a collective database and coded based on treatment group to maintain blinding. LV contractility, plasma cytokine levels, and the effects of APRO dosing on these parameters were first compared among the groups using analysis of variance (ANOVA). If the ANOVA revealed significant differences, post hoc mean separation was performed using Tukey-adjusted mean square differences (Module PR comp, STATA Intercooled, v8, College Station, TX). Following this multiway ANOVA approach, data transformation was also performed in which changes with I/R on the indices of LV contractility from baseline were computed and expressed as a percentage. In addition, plasma cytokine levels determined in the reference control mice for each cytokine and percent changes from these reference values were computed after I/R. These transformed computations were then examined using an adjusted t-score (STATA). Results are presented as mean ± sd unless otherwise indicated. Values of P < 0.05 were considered to be statistically significant.
Fifty mice were used in the I/R protocol, with nine mice dying before the final set of measurements. These mice died of arrhythmias during reperfusion and were equally distributed among the three treatment groups (vehicle: n = 3, 2 × 104 KIU/kg APRO: n = 2, 4 × 104 KIU/kg APRO: n = 2, 8 × 104 KIU/kg APRO: n = 2). Thus, the final sample sizes were vehicle: n = 10, 2 × 104 KIU/kg APRO: n = 11, 4 × 104 KIU/kg APRO: n = 10, APRO 8 × 104 KIU/kg: n = 10). Using the APRO plasma assay and standard curve shown in Figure 1, the computed APRO plasma concentrations for the 2 × 104 KIU/kg APRO group was 180 ± 33, for the 4 × 104 APRO group APRO levels were significantly higher 242 ± 25 KIU/mL (P < 0.05), and increased again in the 8 × 104 APRO group (315 ± 7 KIU/mL, P < 0.05). However, it must be recognized that at high-APRO doses, the assay became nonlinear, which may have underestimated the higher APRO concentrations.
Steady-State Hemodynamics and LV Contractility
Baseline LV function and hemodynamic measurements before randomization and vehicle/APRO infusions are summarized in Table 1. Baseline values were identical across groups after randomization. LV function values at peak ischemia and during reperfusion are shown in Table 1. At 30 min of ischemia, LV peak systolic pressure decreased from baseline in all groups except in the APRO 2 × 104 KIU/kg group. At 60 min of reperfusion, LV peak systolic pressure recovered to within baseline values in the vehicle, APRO 2 × 104 KIU/kg, and APRO 4 × 104 KIU/kg groups, but remained reduced in the APRO 8 × 104 KIU/kg group. LV dP/dt was decreased versus baseline was decreased in all the treatment groups at peak ischemia except for the APRO 2 × 104 KIU/kg group. LV dP/dt returned within baseline values after 60 min of reperfusion for both the APRO 2 × 104 KIU/kg and the APRO 4 × 104 KIU/kg treatment groups.
LV ESPVR values are summarized in Table 1. ESPVR decreased at peak ischemia in all four groups and remained reduced at 60 min of reperfusion except for the APRO 4 × 104 KIU/kg group. Representative LV pressure-volume loops at reperfusion for a vehicle-treated mouse and an APRO 2 × 104 KIU/kg-treated mouse at 60 min of reperfusion are shown in Figure 2b. Clear changes in the slope of ESPVR depict reduced LV contractility with ischemia and some recovery of function after reperfusion. Emax decreased from baseline values at peak ischemia in all groups and remained decreased in all groups at 30 min of reperfusion. Emax was not reduced after 60 min of reperfusion in the APRO 2 × 104 KIU/kg group but remained reduced in the vehicle, APRO 4 × 104 KIU/kg, and APRO 8 × 104 KIU/kg groups. The absolute values for this index of contractility are summarized in Table 1. The relative changes from baseline are depicted in Figure 2c. Emax was measured in the absence of I/R and remained constant from a baseline value of 0.8 ± 0.1 mm Hg·μL−1·mg−1 to a 90 min value of 0.7 ± 0.1 mm Hg·μL−1·mg−1 in both vehicle and APRO 8 × 104 KIU/kg groups (P = 0.45). Thus, the instrumentation alone, or APRO alone, in the absence of I/R had no effect on this index of LV contractility.
Plasma Cytokine Levels
Plasma cytokine concentrations were unaffected by APRO in the absence of I/R. For example, plasma concentrations of TNF in the vehicle group were 3.2 ± 0.33 pg/mL and were 2.8 ± 0.2 pg/mL in the 8 × 104 KIU/kg group (P = 0.40) in the absence of I/R. Accordingly, plasma cytokine concentrations for all of the non-I/R mice were pooled to construct referent control values as shown in Table 2.
Absolute values for plasma cytokine concentrations for all of the I/R groups are presented in Table 2. Plasma concentrations of TNF significantly increased in all treatment groups compared with reference controls but were decreased in the 4 × 104 KIU/kg and 8 × 104 KIU/kg APRO groups. IL-6 and IL-10 plasma levels were significantly increased in all treatment groups versus referent controls, without any significant difference among groups.
Using a unique murine model of I/R, the present study provided two distinct observations. First, indices of LV contractility were improved by APRO at a dose that corresponds to a low or half-Hammersmith dose but not at higher doses. Selective inhibition of TNF-α release only occurred at high doses (corresponding to the Hammersmith and twice-Hammersmith dosing). Taken together, these findings suggest that there are potentially different mechanisms of action of APRO in the context of I/R which are dose dependent. The present study demonstrates that a low dose of APRO (2 × 104 KIU/kg) increased LV contractility while having a “sparing” effect on cytokine release, whereas the high-dose APRO groups did not impart a protective effect. The current study directly measured plasma APRO levels which allowed direct correlation to the Hammersmith dosing protocols. Therefore, in terms of APRO concentrations, the findings of the present study have relevance to the clinical context. For example, the blood conservation using antifibrinolytics (BART) trial used a full-Hammersmith APRO dose in patients undergoing cardiac surgery and reported a higher mortality rate when compared with lysine analogs.4 In the present study, using an APRO dose which was reflective of a full-Hammersmith dose did not impart any protective effects on LV contractility and blunted IL-10 plasma concentrations, a known antiinflammatory molecule.9 However, extrapolating these acute findings in this acute I/R animal model to the observations regarding the long-term clinical effects of a full-Hammersmith APRO dose is problematic.
Myocardial I/R causes activation of an inflammatory response involving numerous cytokines.9 TNF is the prototypical proinflammatory cytokine, and IL-10 has been shown to be strongly antiinflammatory.7,8,22–24 Increased TNF levels have been documented to have time-dependent effects on LV performance, including a positive inotropic effect in the first 3 h after I/R, followed by a cardiodepressant effect and an increase in the extent of reperfusion injury in subsequent hours to days.8,23 Serine proteases play a central role in the amplification of the inflammatory response to I/R through numerous pathways, including contact activation, coagulation, cytokine release, and complement cascades, all of which are modulated by APRO.5,6,10,11,13,14,23–26 In the present study, higher APRO doses attenuate TNF release. It may be that with a longer post-I/R period, the reduction in TNF at higher doses may eventually translate into improvements in myocardial function.23 Another observation regarding cytokine release from the present study was that IL-6, another proinflammatory molecule, was unaffected by higher doses of APRO. This suggests that APRO may selectively affect cytokine release in the context of I/R.
Although the present study demonstrated diverse effects on myocardial contractility and cytokine release, it must be placed in context and its limitations recognized. First, the murine I/R model does not necessarily recapitulate the transient myocardial I/R that may occur in the context of cardiac surgery. Nevertheless, the present study did cause regional ischemia accompanied by complete reperfusion, as confirmed by initial microsphere blood flow studies. Thus, the biological milieu that would be attendant with I/R injury was likely operative. Second, the APRO dosing was based upon a clinical weight-based algorithm that may not be translatable to a mouse model. However, the present study directly measured plasma APRO levels at the end of the I/R period, which is a unique aspect of this study, since plasma APRO measurements were directly assayed in an experimental model of I/R, rather than inferred. In a clinical study, Beath et al.27 used an ex vivo approach to compute relative plasma APRO concentrations in patients undergoing cardiac surgery requiring cardiopulmonary bypass. In this past study, the steady-state APRO plasma concentrations obtained in patients receiving either the full or half-Hammersmith dose were very similar to those obtained in the present study. However, it must be recognized that the volume of distribution, pharmacokinetics, and serine protease inhibitory profiles are likely to be different in the murine system than that of humans. These limitations notwithstanding the unique findings of the present study demonstrated differential dose-dependent effects of APRO on both LV contractility and cytokine release in an intact model of I/R. Moreover, the experimental approaches defined in the present study could be used to examine the effects of lysine analogs with respect to LV contractility and cytokine release. In light of the recent clinical findings, which have resulted in the removal of APRO from clinical application, these future mechanistic investigations would be appropriate.
1. Sedrakyan A, Treasure T, Elefteriades JA. Effect of aprotinin on clinical outcomes in a coronary arterty bypass graft surgery: a systematic review and meta-analysis of randomized clinical trials. J Thorac Cardiovasc Surg 2004;128:442–8
2. Levi M, Cromheecke ME, de Jonge E, Prins MH, de Mol BJ, Briet E, Buller HR. Pharmacological strategies to decrease excessive blood loss in cardiac surgery: a meta-analysis of clinically relevant endpoints. Lancet 1999;354:1940–7
3. Mangano DT, Tudor IC, Dietzel C. The risk associated with aprotinin in cardiac surgery. NEJM 2006;354:353–65
4. Fergusson DA, Hébert PC, Mazer CD, Fremes S, MacAdams C, Murkin JM, Teoh K, Duke PC, Arellano R, Blajchman MA, Bussières JS, Côté D, Karski J, Martineau R, Robblee JA, Rodger M, Wells G, Clinch J, Pretorius R, BART Investigators. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med 2008;358:2319–31
5. Bull D, Connors R, Albanil A, Reid B, Neumayer LA, Nelson R, Stringham JC, Karwande SV. Cardiopulmonary Support and Physiology. Aprotinin preserves myocardial biochemical function during cold storage through suppression of tumor necrosis factor. J Thorac Cardiovasc Surg 2000;119:242–50
6. Bull D, Maurer J. Aprotinin and preservation of myocardial function after ischemia-reperfusion injury. Ann Thorac Surg 2003;75:S735–S739
7. Meldrum DR. Tumor necrosis factor in the heart. Am J Physiol 1998;274:R577–R595
8. Prabhu SD. Cytokine-induced modulation of cardiac function. Circ Res 2004;95:1140–53
9. Ren G, Dewald O, Frangogiannis NG. Inflammatory mechanisms in myocardial infarction. Curr Drug Targets Inflamm Allergy 2003;2:242–56
10. Greilich PE, Okada K, Latham P, Kumar RR, Jessen ME. Aprotinin but not epsilon-aminocaproic acid decreases interleukin-10 after cardiac surgery with extracorporeal circulation: randomized, double-blind, placebo-controlled study in patients receiving aprotinin and epsilon-aminocaproic acid. Circulation 2001;104:I265–I269
11. Turkoz A, Cigli A, But K, Sezgin N, Turkoz R, Gulcan O, Ersoy MO. The effects of aprotinin and steroids on generation of cytokines during coronary artery surgery. J Cardiothorac Vasc Anesth 2001;15:603–10
12. Krombach RS, Clair MJ, Hendrick JW, Houck WV, Zellner JL, Kribbs SB, Whitebread S, Mukherjee R, de Gasparo M, Spinale FG. Angiotensin converting enzyme inhibition, AT1 receptor inhibition, and combination therapy with pacing induced heart failure: effects on left ventricular performance and regional blood flow patterns. Cardiovasc Res 1998;38:631–45
13. van Oeveren W, Jansen NJ, Bidstrup BP, Royston D, Westaby S, Neuhof H, Wildevuur CR. Effects of aprotinin on hemostatic mechanisms during cardiopulmonary bypass. Ann Thorac Surg 1987;44:640–5
14. Royston D, Bidstrup BP, Taylor KM, Sapsford RN. Effect of aprotinin on need for blood transfusion after repeat open-heart surgery. Lancet 1987;2:1289–91
15. Royston D, Cardigan R, Gippner-Steppert C, Jochum M. Is perioperative plasma aprotinin concentration more predictable and constant after a weight-related dose regimen? Anesth Analg 2001;92:830–6
16. McCarthy RJ, Tuman K, O’Connor C, Ivankovish AD. Aprotinin pretreatment diminished postischemic myocardial contractile dysfunction in dogs. Anesth Analg 1999;89:1096–101
17. Khan TA, Bianchi C, Araujo E, Voisine P, Xu SH, Feng J, Li J, Sellke FW. Aprotinin preserves cellular junctions and reduces myocardial edema after regional ischemia and cardioplegic arrest. Circulation 2005;112:I196–I201
18. Ikonomidis JS, Hendrick JW, Parkhurst AM, Herron AR, Escobar PG, Dowdy KB, Stroud RE, Hapke E, Zile MR, Spinale FG. Accelerated LV remodeling after myocardial infarction in TIMP-1-deficient mice: effects of exogenous MMP inhibition. Am J Physiol Heart Circ Physiol 2005; 288:H149–H158
19. Suga H, Sagawa K. Mathematical interrelationship between instantaneous ventricular pressure-volume ratio and myocardial force-velocity relation. Ann Biomed Engl 1972;1:160–81
20. Feldman MD, Erikson JM, Mao Y, Korcarz CE, Lang RM, Freeman GL. Validation of a mouse conductance system to determine LV volume: comparison to echocardiography and crystals. Am J Physiol Heart Circ Physiol 2000;279:H1698– H1707
21. Chang KC, Peng YI, Lee FC, Tseng YZ. Effects of food restriction on systolic mechanical behavior of the ventricular pump in middle-aged and senescent rats. J Gerontol A Biol Sci Med Sci 2001;56:B108–B114
22. Giomarelli P, Scolletta S, Borrelli E, Biagioli B. Myocardial and lung injury after cardiopulmonary bypass: role of interleukin (IL)-10. Ann Thorac Surg 2003;76:117–23
23. Blancke F, Claeys MJ, Jorens P, Vermeiren G, Bosmans J, Wuyts FL, Vrints CJ. Systemic inflammation and reperfusion injury in patients with acute myocardial infarction. Mediators Inflamm 2005;2005:385–9
24. Harmon D, Lan W, Shorten G. The effect of aprotinin on hypoxia-reoxygenation-induced changes in neutrophil and endothelial function. Eur J Anaesthes 2004;21:973–9
25. Gott JP, Cooper WA, Schmidt FE, Brown WM, Wright CE, Merlino JD, Fortenberry JD, Clark WS, Guyton RA. Modifying risk for extracorporeal circulation: trial of four antiinflammatory strategies. Ann Thorac Surg 1998;66:747–54
26. Hill GE, Pohorecki R, Alonso A, Rennard S, Robbins R. Aprotinin reduces interleukin-8 production and lung neutrophil accumulation after cardiopulmonary bypass. Anesth Analg 1996;83:696–700
© 2009 International Anesthesia Research Society
27. Beath SM, Nuttall GA, Fass DN, Oliver WC Jr, Ereth MG, Oyen LJ. Plasma aprotinin concentrations during cardiac surgery: full-versus half-dose regimens. Anesth Analg 2001;91:257–64