- Email: [email protected]
Effect of iodixanol on in vitro bleeding time
Effect of Iodixanol on In Vitro Bleeding Time Laura G. Melton, PhD, Kathleen M Muga, BS, Don A. Gabriel, MD, PhD
Rationale and Objectives. Both monomeric and dimeric ionic radio comrast media (CM) have b e e n shown to be more anticoagulant than nonionic monomeric CM. Iodixanol is a relatively new nonioriic dimeric contrast medium. We investigated whether the dimeric structure of iodixanol would produce the same level of anticoagulation that has b e e n observed using nonionic monomeric CM. Methods. We used a global screening device that operates under physiologic flow conditions to monitor the effects of iodixanol on in vitro bleeding time (IVBT). This flow dynamic technique perfuses nonanticoagulated whole blood through a hollow fiber device. A leak in the fiber is created by a precision needle, and the resulting pressure fluctuations within the fiber are monitored to examine the ability of the hemostatic system to close the leak by forming a stable platelet plug. Results. In 20 donors, iodixanol (25% CM/blood, v/v) was shown to lengthen the m e a n IVBT (18.74 rain) c o m p a r e d with the normal blood m e a n IVBT (4.24 min). C o n c l u s i o n . The addition of dimeric iodixanol to normal blood affects the IVBT in the same manner as nonionic monomeric CM (ioversol, iopamidol, and iohexol). K e y W o r d s . Iodixanol; ionic contrast media; nonionic contrast media; in vitro bleeding time.
From the Center for Thrombosis and Hemostasis, Department of Medicine, Division of Hematology, University of North Carolina School of Medicine, Chapel Hill, NC. This research was supported by the Xylum Corporation (Scarsdale, NY) and Mallinckrodt Medical, Inc. (St. Louis, MO). Address reprint requests to D. A. Gabriel, MD, PhD,
Division of Hematology, CB#7035, University of North Carolina School of Medicine, Chapel Hill, NC 27599. Received October 17, 1995, and accepted for publication after revision February 12, 1996.
Acad Radio11996;3:407-411 © 1996, Association of University Radiologists
I odixanol is an iso-osmolal contrast medium that is nonionic, dimeric, and electrolyte balanced with a ratio of sodium and calcium equivalent to blood. Preclinical studies of iodixanol have shown g o o d physicochemical, pharmacologic, and toxicologic properties [1-5]. Clinical trials have shown that iodixanol is effective, is well tolerated, and has few adverse events [4-17]. However, relatively few studies have b e e n conducted to assess the effect of iodixanol on the coagulation system. Both nonionic and ionic contrast media (CM) have b e e n s h o w n to be anticoagulant, although nonionic CM generally have a smaller anticoagulant effect. Using a flow device, we [18] previously examined the effect of five CM (an ionic monomer, an ionic dimer, and three nonionic monomers)
407
MELTON
ET AL.
on in vitro bleeding time (IVBT) in nonanticoagulated whole blood. The m e a n IVBT for all five CM tested were prolonged relative to normal blood, The addition of a nonionic contrast medium to normal blood produced individual IVBTs ranging from normal to completely anticoagulated, unlike ionic CM, which consistently produced a complete inhibition of coagulation. Nonionic dimeric CM are a relatively new class of contrast agents. We investigated whether the dimeric structure of iodixanol would affect coagulation to a greater extent than the nonionic monomeric CM. Therefore, w e examined the effect of the addition of iodixanol to normal blood samples on IVBT using a flow device. MATERIALS AND METHODS Contrast Medium
IodLxanol (Visipaque) was obtained from Nycomed (Collegeville, PA). Blood Samples
Blood samples were obtained from 20 healthy volunteers (staff and students at the University of North Carolina [UNC], Chapel Hill). A normal range for the bleeding time was established from the blood taken directly from these healthy volunteers: 1.25-9.13 min (34 = 4.24 min, SD = 2.56 min). Ethical committee approval was given by the UNC School of Medicine for the use of normal volunteer blood. All blood samples were collected using a 19-gauge butterry needle. The tourniquet was removed before the blood sample was aspirated to minimize platelet activation. The standard two-syringe veniptmcture technique was used to draw the blood, in which the first syringe containing 5 ml of blood was discarded to avoid possible sample contamination with tissue thromboplastin [19]. Two milliliters of blood was then collected into a second synnge and immediately attached to the polypropylene fiber assembly by way of a luer lock and inserted into the heating block (37°C). Blood samples that were drawn into syringes conmining the contrast medium were gently inverted 10 times prior to being attached to the fiber assembly. IVBT Measurements
The hollow fiber flow device, the Clot Signature Analyzer (CSA; Xylum, Scarsdale, NY), is an integrated system based on Gorog's haemostatometer and consists of the CSA instrument, an AST Bravo/286 computer (AST
408
vol. 3, No. 5, May 1996
Research, Irvine, CA), an NEC MultiSync GS2A monitor (NEC Home Products, Wood Dale, IL), and an Epson LX810 printer (Epson America, Torrance, CA) [20J. Blood flow through the 0.5-n'n'n fiber (Xylum) began exactly 2 min after venipuncture. Sixty millimeters of mercury (8 KPa) and a flow rate of 0.1 ml/min were maintained by pumping paraffin oil into the sample syringe, which produced a shear rate of 135 sec -1 [21, 22]. The shear rate was calculated by assuming that blood was a Newtonian fluid in the following equation: 4Q/nR 3, where Q is the flow rate (ml/ sec) and R is the radius of the flow pathway (in centimeters). Because the volume in the sample syringe was kept constant, blood flowed through the fiber as oil was p u m p e d into the syringe. Thus, oil served not only to displace the blood but also to minimize red cell sedimentation as it bubbled through the blood to the top of the inverted syringe [21, 22]. After a 4-min initialization period that established a constant flow rate, a machine-operated, spring-loaded needle assembly (0.i6-mm needle) pierced a hole (R = 0.005 cm) through the top and bottom of the fiber, which was directly measured by the manufacturer (Xytum) using an inverted microscope. The shear rate at the punch site was estimated to be greater than 10,000 sec -1 by the manufacturer. The pressure across the fiber was monitored by a pressure transducer located downstream from the holes in the fiber. An additional monitor of the blood leak was provided by a photodiode (Xylum) that detected blood dripping from the fiber and w a s s h o w n as a solid black line below the x-axis of the flow spectra (Fig. 1). As the aggregate grew by recruitment of platelets, the plug was stabilized and the pressure rapidly returned to the baseline pressure (point D in Fig. 1). The IVBT was the difference from the time of the punch in the hollow fiber
80 N aO
, , , , , , , , , ,
0
4
8
12
16 20 TIME (rain)
24
,,
28
30
FIGURE 1. Schematic representation of the flow profile for a normal blood sample. A fine precision needle pierced the walls of the punch blood flow pathway, producing a drop in pressure (A). Leakage of blood was optically detected and recorded as a solid black line (B) until the platelet plug had successfully sealed the holes created by the needle wound (D). Instability of initial formation of the platelet aggregates was observed as an erratic fluctuation in the pressure (C). As the platelet thrombus was formed, it eventually occluded the inner lumen of the flow pathway and the pressure was reduced to zero (E).
Vol. 3, No. 5, May 1996
I O D I X A N O L A N D IN V I T R O B L E E D I N G T I M E
(point A in Fig. 1) to the time a stable plug was formed (point D in Fig. 1). 1VBTs in excess of 30 min were recorded as 30.00 min in the data analysis and were considered to be an infinite bleeding time. Nonanticoagulated blood from normal human donors was drawn into syringes containing iodixanol, resulting in 25% CM/blood (v/v), which was the same volume ratio used for a similar study on ionic and nonionic CM [18]. In the current study, each donor's blood was drawn two times within an 8-hr period. RESULTS
Table 1 shows a comparison of normal IVBTs and the IVBTs obtained by the addition of iodixanol to nonanticoagulated blood. The m e a n IVBT for iodixanol was prolonged compared with normal blood (n = 20; M = 18.75 min, SD = 8.70 min). There was a significant statistical difference using the paired t test (p < .05) between the normal blood and the blood containing iodixanol. DISCUSSION
Various laboratory tests have b e e n used to assess the effect of CM on hemostasis; however, these tests have
TABLE 1: Normal IVBTs and IVBTs Resulting from the Addition of Iodixanol to the Blood Samples of the 20 Donors (Arranged in Descending Order) Donor 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M IVBT = in vitro bleeding time.
IVBT (min) Normal
iodixanol
3.67 2.05 2.00 8,63 2.00 8.98 5.45 2.95 1.25 4.08 9.13 2.20 3.16 3.00 5.05 2.87 3.83 8.58 2.55 3.37 4,24
30.00 30.00 30.00 30.00 30.00 30.00 20.33 18.17 18.05 17.47 17.16 15.37 15.26 15.22 14.87 13.00 12.42 11.78 3.40 2.48 18.75
limitations. Both the activated partial thromboplastin time (aPTF) and the activated clotting time (ACT) evaluate only the soluble phase of coagulation. The ACT is performed on fresh whole blood using an activating agent. The aP'IT requires a citrated plasma sample, activating agents, and substrate reagents. Sodium citrate removes most of the calcium ions from bloo d , resulting in a nonphysiologic situation in which the response to the platelet agonist is altered [23]. The platelet isolation process used for platelet aggregation tests alters the size, density, and behavior of the platelets [24]. Wholeblood aggregometry eliminates this problem but introduces others, including the need for anticoagulation and dilution of blood [25]. In addition, all these static tests exclude the important effect of flow on platele t activation and subsequent hemostasis. There are several advantages to using a hollow fiber flow device to evaluate the effect of CM on coagulation. First, the test uses nonanticoagulated whole blood to test the entire coagulation system without the use of additiona! reagents. Thus, the system is sensitive to reagent additives such as CM. Second, the physiologic flow system permits shear activation of platelets under a stable thermal environment. Third, platelet activation and recruitment is reflected in a qualitative flow profile, which automatically produces a quantitative bleeding parameter (IVBT). Studies using similar flow devices have shown the validity of this technique as a measure of both platelet activation and recruitment [18, 20-22, 26-33]. Finally, the hollow fiber flow device is a relatively quick test (30 min) that is both sensitive and reliable. The effects of CM (ioxaglate, diatrizoate, iohexol, ioversol, and iopamidol) on 1VBT assessed by the flow device have been investigated previously [18]. The normal mean IVBT from that study (3.62 -+ 1.64 min) was comparable to the normal mean IVBT (4.24 _+ 2.56 min) from this study. There was no statistically significant difference in the IVBTs obtained from the normal blood samples with the two studies (paired t test, p > .05). The addition of ionic CM to normal blood in the previous study always produced an IVBT greater than 30 min. The nonionic CM produced mean IVBTs of 16.43, 17.63, and 19.84 min for iopamidol, iohexol, and ioversol, respectively. The mean IVBT for iodixanol (18.75 min) was similar to the other three nonionic CM. A large range of IVBTs (2.48-30 min) was observed with iodixanol. This resul!c also was observed with the other nonionic CM. Addition of the same nonionic contrast medium to three separate blood samples from the same donor was con-
409
MELTON
ET AL.
ducted using three different donors and two nonionic CM (iopamidol and ioversol) [18]. Two of the three donors produced variable IVBTs on addition of either nonionic CM to their blood sample. In both studies, there were occasions on which nonionic CM produced no anticoagulation protection, which can increase the risk of a thrombotic complication. Because the normal control samples as well as the blood samples containing ionic CM produced highly reproducible results, the variable inhibition of coagulation observed with the nonionic CM could be related to the physical properties of nonionic CM. Therefore, in this study, we observed that the dimeric structure of iodixanol produced a nonionic contrast agent with anticoagulant effects similar to other nonionic CM. In 1995, n e w data supporting past evidence that ionic CM are m o r e anticoagulant than nonionic CM were reported [34]. A study monitoring platelet counts documented almost-immediate platelet aggregation in heparinized blood with nonionic CM (ioversol, iohexol, and iopamidol) and no aggregation at any concentration for the ionic CM (diatrizoate and ioxaglate). Hay et al. [34] also used flow cytometry to evaluate both platelet aggregation and activation and found no or minimal platelet activation or aggregation with ionic CM and high levels of platelet activation and aggregation with nonionic CM. Similarly, in a flow cytometry study, Koza et al. [35] reported low-to-background levels of platelet activation with ionic CM (diatrizoate and ioxaglate) and dose- and time-dependent platelet activation with nonionic CM (iohexol and iopamidol). Nonionic CM (loversol, iohexol, and iopamid01) induced adhesion of leukocytes on h u m a n endothelium to a greater extent than did ionic CM (diatrizoate and ioxaglate) [36J. Thrombogenic effects tested in a laser-induced rat thrombosis model were observed for both nonionic CM (iohexol and iopamidol) and ionic CM (diatrizoate and ioxaglate), but the thrombogenicity was the greatest for nonionic CM [37]. Additionally, an in vitro study by Corot et al. [38] showed that the ionic ioxaglate inhibited coagulation and platelet activation to a greater extent than did the nonionic iodixanol. The weaker anticoagulant effect of nonionic CM m a y be clinically relevant, especially in the realm of interventional cardiac procedures in which the important side effects are acute thrombotic vessel closure, acute myocardial infarction, thromboembolism, stroke, and death.
410
Vol. 3, No. 5, May 1996
REFERENCES 1. Karstoft J, Baath L, Jansen I, Edvinsson L. Vasoconstriction of isolated arteries induced by angiographic contrast media. Acta Radiol 1995; 36:312-316. 2. Jorgensen NP, Nossen JO, Borch KW, Kristiansen AB, Lundby B, Theodorsen L. Safety and tolerability of iodixano[ in healthy volunteers with reference to two monomeric X-ray contrast media. Eur J Radiol 1992; 15:252-257. 3. Svaland MG, Haider T, Langseth-Manrique K, Andrew E, Hafs PA. Human pharmokinetics of iodixanol. Invest Radio11992;27:130-133. 4. Nossen JO, Aakhus T, Berg KJ, Jorgensen NP, Andrew E. Experience with iodixanol, a new nonionic dimeric contrast medium: preliminary results from the human phase I study. Invest Radio11990;25[suppl]:S113-$114. 5. Bolstad B, Borch KW, Grynne BH, et al. Safety and tolerability of iodixanol: a dimeric, nonionic contrast medium. An emphasis on European clinical phase I and II. Invest Radio11991 ;26[suppl]:S201-S204. 6. Anderson PE, Bolstad B, Berg KJ, Justesen P, Thayssen E Kloster YF. Iodixanol and ioxaglate in cardioangiography: a double-blind randomized phase Ill study. Clin Radio11993;48:268-272. 7. Hill JA, Cohen MB, Kou WH, et al. Iodixanol, a new isosmotic nonionic contrast agent compared with iohexol in cardiac angiography. Am J Cardio11994;74:57-63. 8. Klew NE, Levorstad K, Berg KJ, et al. Iodixanol in c&dioangiography in patients with coronary artery disease. Acta Radio11993;34:72-76. 9. Harnish PP, Feuntaine H, Ebrahimi R. Experience in 1,259 patients in the United States. Invest Radio11994 ;29[suppl]:S236-S237. 10. Tveit K, Bolz KD, Bolstad B, et al..todixanol in cardioangiography: a double-blind parallel comparison between iodixanol-320 mg I/ml and ioxaglate 320 mg I/ml. Acta Radio11994;35:614-618. 11. Gavant ML, Siegle R L. ]odixanol in excretory urography: initial clinical experience with a nononic, dimeric (ratio 6:1) contrast medium. Radiology
1992;183:515-518. 12. Conroy RM, Bjartveit K, Sheppick A, Long U, Masterson J. Iodixanol in intravenous urography: a comparison of iodixanol 270 mg I/ml, iodixanol 320 mg I/m] and iopamidol 300 mg l/ml (NIOPAM). Clin Radiol 1994; 49:337-340. 13. Palmers Y, Greef DD, Grynne BH, Smits J, Put E. A double-blind study comparing safety, tolerability and efficacy of iodixanol 320-mg I/ml and ioxaglate 320 mg I/ml in cerebral arteriography. Eur J Radiol 1993; 17:203-209. 14. Poirier VC, Monsein LH, Newberry PD, Kreps BJ. Double-blind, randomized comparison of iodixanol 320 and iohexol 300 for cerebral angiography. Invest Radio11994;29[suppl]:S43-S44. 15. Pugh ND, Sissons GRJ, Ruttley MST, Berg KJ, Nossen JO, Eide H. Iodixanol in femoral arteriography (phase III): a comparative double-blind parallel trial between iodixanol and iopromide. Clin Radio11993;47:96-99. 16. Afbrechtsson U, Larusdottir H, Nergren L, Lundby B. Iodixanol a new nonionic dimer in aortofemora] angiography. Acta Radio11992;33:611-613. 17. Thorstensen O, Albrechtsson U, Calissendorff B, et al. Iodixanol in femoral arteriography: a randomized, double-blind, phase III, parallel study with iodixanol 270 mg I/m] and iohexol 300 mg I/ml. Acta Radiol 1994; 35:629-631. 18. Melton LG, Muga KM, Gabriel DA. Effect of contrast media on in vitro bleeding time: assessment by a hollow fiber instrument. Aead Radiol
1995;2:239-243. 19. Lotspeich-Steininger CA, Stiene-Martin EA, Koepke JA, eds. Clinical hematology: principles, procedures, correlations. Philadelphia: Lippincott, 1992:599-604. 20. Muga KM, Melton LG, Gabriel DA. A flow dynamic technique used to assess global hemostasis. Blood Coag Fibrinolysis 1995;6:73-78. 21. Gorog E Ahmed A. Haemostatometer: a new in vitro technique for assessing haemostatic activity of blood. Thromb Res 1984;34:341-357. 22. Gorog R A new, ideal technique to monjtor thrombolytic therapy. Angiology 1986;37:99-105. 23. Harfenist EJ, Packman MA, Kinglough-Rathbone RL, Cataneo M, Mustard JF. Effect of calcium ion concentration on the ability of fibrinogen and
Vol. 3, No. 5, May 1996
24. 25. 26. 27.
28. 29. 30.
31.
IODIXANOL
von Willebrand factor to support the ADP-induced aggregation of human platelets. Blood 1987;70:827-831. Born GVR, Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 1962;194:272-276. Cardinal DL, Flower RJ. The electronic aggregometer: a novel device for assessing platelet behaviour in blood. J Pharmacol Methods 1980;3:135-158, John LCH, Rees GM, Kovacs lB. Inhibition of platelet function by heparin. J Thorac Cardiovasc Surg 1993;105:816-822. Kovacs IB, Hutton RA, Kernoff PBA, Hemostatic evaluation in bleeding disorders from native blood: clinical experience with the haemostatometer. Am J Patho11989;91:271-279. Gorog P, Kakkar VV. Transient effect of aspirin on collagen-induced platelet accumulation. Am J Clin Pathol 1988;86:311-316. Gorog P, Kovacs lB. Coagulation of flowing native blood: advantages over stagnant (tube) clotting tests. Thromb Res 1991 ;64:611-619. John LCH, Rees GM, Kovacs lB. Effect of heparin on in vitro platelet reactivity in cardiac surgical patients: a comparative assessment by whole blood platelet aggregometry and haemostatometry. Thromb Res 1992; 66:649-656. Kovacs IB, Gorog P. Simultaneous measurement of all thrombosis parameters from native human blood: usefulness in monitoring efficacy and complications of thrombolytic therapy. Angiology 1990;41:829-835.
A N D IN V I T R O
BLEEDING
TIME
32. Peters NS, Abrams LSM, Dymond DS, Kovacs lB. Platelet hyperreactivity and inefficient spontaneous thrombolysis in patients at high risk from an acute coronary event. Cardiovasc Res 1989;23:567-572. 33. Ratnatunga CP, Edmondson SF, Rees GM, Kovacs lB. High-dose aspirin inhibits shear-induced platelet reaction involving thrombin generation. Circulation 1992;85:1077-1082. 34. Hay KL, Bull BS. An analysis of platelet activation and aggregation produced by three classes of contrast media. J Vasc Intervent Radio11996; 6:211-217. 35. Koza MJ, Shankey TV, Walenga JM, Moncada R, Fareed J, Pifarre R. Flow cytometric evaluation of platelet activation by ionic or nonionic contrast media and modulation by heparin and recombinant hiruden. Invest Radio11998;30:90-97. 36. Abeyama K, Oh S, Kawano K, et al. Nonionic contrast agents produce thrombotic effect by inducing adhesion of leukocytes on human endothelium. Biochem Biophys Res Commun 1995;212:776-783. 37. Aguejouf O, Doutremepuich F, Azougagh OF, Doutremepuich C. Thrombogenicity of ionic and nonionic contrast media tested in a laser induced rat thrombosis model. Thromb Res 1995;77:259-269. 38. Corot C, Chronos N, Sabattuer V, Gruentzig A. Iodinated contrast media structure activity relationship based on in vitro coagulation and platelet activation investigations. Cardiovasc Intervent Radio11995;18[suppl] :S107,
Announcements The Johns Hopkins University School of Medicine and the Eberhard-Karls-Universit~tt are sponsoring Abdominal Radiology: Practical Approach and Recent Advances, which will be held June 13-15, 1996, at Kongresshaus in Salzburg, Austria. Credit will be announced. The fee is $375. For more information, contact the Program Coordinator, Johns Hopkins Medical Institutions, Office of Continuing Medical Education, Turner Bldg., 720 Rutland Ave., Baltimore, MD 21205; (410) 955-2959, fax (410) 955-0807.
The Department of Radiology of the Johns Hopkins Medical Institutions is sponsoring Advanced Topics in CT w i t h Emphasis o n Spiral CT, which will be held July 13-16, 1996, at the Eldorado Hotel in Santa Fe, NM. Thirteen hours of AMA Category 1 and 13 hours of Category 1 ACE-ASRT credit will be awarded. The fee is $500 for physicians and $450 for residents, fellows, and CRNAs. The program directors are Elliot-K. Fishman, MD, and Stanley S. Siegelman, MD. For more information, contact the Program Coordinator, Office of Continuing Medical Education, Johns Hopkins Medical Institutions, Turner 20, 720 Rutland Ave., Baltimore, MD 21205-2195; (410) 955-2959; fax (410) 955-08O7.
41 1
Rationale and Objectives. Both monomeric and dimeric ionic radio comrast media (CM) have b e e n shown to be more anticoagulant than nonionic monomeric CM. Iodixanol is a relatively new nonioriic dimeric contrast medium. We investigated whether the dimeric structure of iodixanol would produce the same level of anticoagulation that has b e e n observed using nonionic monomeric CM. Methods. We used a global screening device that operates under physiologic flow conditions to monitor the effects of iodixanol on in vitro bleeding time (IVBT). This flow dynamic technique perfuses nonanticoagulated whole blood through a hollow fiber device. A leak in the fiber is created by a precision needle, and the resulting pressure fluctuations within the fiber are monitored to examine the ability of the hemostatic system to close the leak by forming a stable platelet plug. Results. In 20 donors, iodixanol (25% CM/blood, v/v) was shown to lengthen the m e a n IVBT (18.74 rain) c o m p a r e d with the normal blood m e a n IVBT (4.24 min). C o n c l u s i o n . The addition of dimeric iodixanol to normal blood affects the IVBT in the same manner as nonionic monomeric CM (ioversol, iopamidol, and iohexol). K e y W o r d s . Iodixanol; ionic contrast media; nonionic contrast media; in vitro bleeding time.
From the Center for Thrombosis and Hemostasis, Department of Medicine, Division of Hematology, University of North Carolina School of Medicine, Chapel Hill, NC. This research was supported by the Xylum Corporation (Scarsdale, NY) and Mallinckrodt Medical, Inc. (St. Louis, MO). Address reprint requests to D. A. Gabriel, MD, PhD,
Division of Hematology, CB#7035, University of North Carolina School of Medicine, Chapel Hill, NC 27599. Received October 17, 1995, and accepted for publication after revision February 12, 1996.
Acad Radio11996;3:407-411 © 1996, Association of University Radiologists
I odixanol is an iso-osmolal contrast medium that is nonionic, dimeric, and electrolyte balanced with a ratio of sodium and calcium equivalent to blood. Preclinical studies of iodixanol have shown g o o d physicochemical, pharmacologic, and toxicologic properties [1-5]. Clinical trials have shown that iodixanol is effective, is well tolerated, and has few adverse events [4-17]. However, relatively few studies have b e e n conducted to assess the effect of iodixanol on the coagulation system. Both nonionic and ionic contrast media (CM) have b e e n s h o w n to be anticoagulant, although nonionic CM generally have a smaller anticoagulant effect. Using a flow device, we [18] previously examined the effect of five CM (an ionic monomer, an ionic dimer, and three nonionic monomers)
407
MELTON
ET AL.
on in vitro bleeding time (IVBT) in nonanticoagulated whole blood. The m e a n IVBT for all five CM tested were prolonged relative to normal blood, The addition of a nonionic contrast medium to normal blood produced individual IVBTs ranging from normal to completely anticoagulated, unlike ionic CM, which consistently produced a complete inhibition of coagulation. Nonionic dimeric CM are a relatively new class of contrast agents. We investigated whether the dimeric structure of iodixanol would affect coagulation to a greater extent than the nonionic monomeric CM. Therefore, w e examined the effect of the addition of iodixanol to normal blood samples on IVBT using a flow device. MATERIALS AND METHODS Contrast Medium
IodLxanol (Visipaque) was obtained from Nycomed (Collegeville, PA). Blood Samples
Blood samples were obtained from 20 healthy volunteers (staff and students at the University of North Carolina [UNC], Chapel Hill). A normal range for the bleeding time was established from the blood taken directly from these healthy volunteers: 1.25-9.13 min (34 = 4.24 min, SD = 2.56 min). Ethical committee approval was given by the UNC School of Medicine for the use of normal volunteer blood. All blood samples were collected using a 19-gauge butterry needle. The tourniquet was removed before the blood sample was aspirated to minimize platelet activation. The standard two-syringe veniptmcture technique was used to draw the blood, in which the first syringe containing 5 ml of blood was discarded to avoid possible sample contamination with tissue thromboplastin [19]. Two milliliters of blood was then collected into a second synnge and immediately attached to the polypropylene fiber assembly by way of a luer lock and inserted into the heating block (37°C). Blood samples that were drawn into syringes conmining the contrast medium were gently inverted 10 times prior to being attached to the fiber assembly. IVBT Measurements
The hollow fiber flow device, the Clot Signature Analyzer (CSA; Xylum, Scarsdale, NY), is an integrated system based on Gorog's haemostatometer and consists of the CSA instrument, an AST Bravo/286 computer (AST
408
vol. 3, No. 5, May 1996
Research, Irvine, CA), an NEC MultiSync GS2A monitor (NEC Home Products, Wood Dale, IL), and an Epson LX810 printer (Epson America, Torrance, CA) [20J. Blood flow through the 0.5-n'n'n fiber (Xylum) began exactly 2 min after venipuncture. Sixty millimeters of mercury (8 KPa) and a flow rate of 0.1 ml/min were maintained by pumping paraffin oil into the sample syringe, which produced a shear rate of 135 sec -1 [21, 22]. The shear rate was calculated by assuming that blood was a Newtonian fluid in the following equation: 4Q/nR 3, where Q is the flow rate (ml/ sec) and R is the radius of the flow pathway (in centimeters). Because the volume in the sample syringe was kept constant, blood flowed through the fiber as oil was p u m p e d into the syringe. Thus, oil served not only to displace the blood but also to minimize red cell sedimentation as it bubbled through the blood to the top of the inverted syringe [21, 22]. After a 4-min initialization period that established a constant flow rate, a machine-operated, spring-loaded needle assembly (0.i6-mm needle) pierced a hole (R = 0.005 cm) through the top and bottom of the fiber, which was directly measured by the manufacturer (Xytum) using an inverted microscope. The shear rate at the punch site was estimated to be greater than 10,000 sec -1 by the manufacturer. The pressure across the fiber was monitored by a pressure transducer located downstream from the holes in the fiber. An additional monitor of the blood leak was provided by a photodiode (Xylum) that detected blood dripping from the fiber and w a s s h o w n as a solid black line below the x-axis of the flow spectra (Fig. 1). As the aggregate grew by recruitment of platelets, the plug was stabilized and the pressure rapidly returned to the baseline pressure (point D in Fig. 1). The IVBT was the difference from the time of the punch in the hollow fiber
80 N aO
, , , , , , , , , ,
0
4
8
12
16 20 TIME (rain)
24
,,
28
30
FIGURE 1. Schematic representation of the flow profile for a normal blood sample. A fine precision needle pierced the walls of the punch blood flow pathway, producing a drop in pressure (A). Leakage of blood was optically detected and recorded as a solid black line (B) until the platelet plug had successfully sealed the holes created by the needle wound (D). Instability of initial formation of the platelet aggregates was observed as an erratic fluctuation in the pressure (C). As the platelet thrombus was formed, it eventually occluded the inner lumen of the flow pathway and the pressure was reduced to zero (E).
Vol. 3, No. 5, May 1996
I O D I X A N O L A N D IN V I T R O B L E E D I N G T I M E
(point A in Fig. 1) to the time a stable plug was formed (point D in Fig. 1). 1VBTs in excess of 30 min were recorded as 30.00 min in the data analysis and were considered to be an infinite bleeding time. Nonanticoagulated blood from normal human donors was drawn into syringes containing iodixanol, resulting in 25% CM/blood (v/v), which was the same volume ratio used for a similar study on ionic and nonionic CM [18]. In the current study, each donor's blood was drawn two times within an 8-hr period. RESULTS
Table 1 shows a comparison of normal IVBTs and the IVBTs obtained by the addition of iodixanol to nonanticoagulated blood. The m e a n IVBT for iodixanol was prolonged compared with normal blood (n = 20; M = 18.75 min, SD = 8.70 min). There was a significant statistical difference using the paired t test (p < .05) between the normal blood and the blood containing iodixanol. DISCUSSION
Various laboratory tests have b e e n used to assess the effect of CM on hemostasis; however, these tests have
TABLE 1: Normal IVBTs and IVBTs Resulting from the Addition of Iodixanol to the Blood Samples of the 20 Donors (Arranged in Descending Order) Donor 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M IVBT = in vitro bleeding time.
IVBT (min) Normal
iodixanol
3.67 2.05 2.00 8,63 2.00 8.98 5.45 2.95 1.25 4.08 9.13 2.20 3.16 3.00 5.05 2.87 3.83 8.58 2.55 3.37 4,24
30.00 30.00 30.00 30.00 30.00 30.00 20.33 18.17 18.05 17.47 17.16 15.37 15.26 15.22 14.87 13.00 12.42 11.78 3.40 2.48 18.75
limitations. Both the activated partial thromboplastin time (aPTF) and the activated clotting time (ACT) evaluate only the soluble phase of coagulation. The ACT is performed on fresh whole blood using an activating agent. The aP'IT requires a citrated plasma sample, activating agents, and substrate reagents. Sodium citrate removes most of the calcium ions from bloo d , resulting in a nonphysiologic situation in which the response to the platelet agonist is altered [23]. The platelet isolation process used for platelet aggregation tests alters the size, density, and behavior of the platelets [24]. Wholeblood aggregometry eliminates this problem but introduces others, including the need for anticoagulation and dilution of blood [25]. In addition, all these static tests exclude the important effect of flow on platele t activation and subsequent hemostasis. There are several advantages to using a hollow fiber flow device to evaluate the effect of CM on coagulation. First, the test uses nonanticoagulated whole blood to test the entire coagulation system without the use of additiona! reagents. Thus, the system is sensitive to reagent additives such as CM. Second, the physiologic flow system permits shear activation of platelets under a stable thermal environment. Third, platelet activation and recruitment is reflected in a qualitative flow profile, which automatically produces a quantitative bleeding parameter (IVBT). Studies using similar flow devices have shown the validity of this technique as a measure of both platelet activation and recruitment [18, 20-22, 26-33]. Finally, the hollow fiber flow device is a relatively quick test (30 min) that is both sensitive and reliable. The effects of CM (ioxaglate, diatrizoate, iohexol, ioversol, and iopamidol) on 1VBT assessed by the flow device have been investigated previously [18]. The normal mean IVBT from that study (3.62 -+ 1.64 min) was comparable to the normal mean IVBT (4.24 _+ 2.56 min) from this study. There was no statistically significant difference in the IVBTs obtained from the normal blood samples with the two studies (paired t test, p > .05). The addition of ionic CM to normal blood in the previous study always produced an IVBT greater than 30 min. The nonionic CM produced mean IVBTs of 16.43, 17.63, and 19.84 min for iopamidol, iohexol, and ioversol, respectively. The mean IVBT for iodixanol (18.75 min) was similar to the other three nonionic CM. A large range of IVBTs (2.48-30 min) was observed with iodixanol. This resul!c also was observed with the other nonionic CM. Addition of the same nonionic contrast medium to three separate blood samples from the same donor was con-
409
MELTON
ET AL.
ducted using three different donors and two nonionic CM (iopamidol and ioversol) [18]. Two of the three donors produced variable IVBTs on addition of either nonionic CM to their blood sample. In both studies, there were occasions on which nonionic CM produced no anticoagulation protection, which can increase the risk of a thrombotic complication. Because the normal control samples as well as the blood samples containing ionic CM produced highly reproducible results, the variable inhibition of coagulation observed with the nonionic CM could be related to the physical properties of nonionic CM. Therefore, in this study, we observed that the dimeric structure of iodixanol produced a nonionic contrast agent with anticoagulant effects similar to other nonionic CM. In 1995, n e w data supporting past evidence that ionic CM are m o r e anticoagulant than nonionic CM were reported [34]. A study monitoring platelet counts documented almost-immediate platelet aggregation in heparinized blood with nonionic CM (ioversol, iohexol, and iopamidol) and no aggregation at any concentration for the ionic CM (diatrizoate and ioxaglate). Hay et al. [34] also used flow cytometry to evaluate both platelet aggregation and activation and found no or minimal platelet activation or aggregation with ionic CM and high levels of platelet activation and aggregation with nonionic CM. Similarly, in a flow cytometry study, Koza et al. [35] reported low-to-background levels of platelet activation with ionic CM (diatrizoate and ioxaglate) and dose- and time-dependent platelet activation with nonionic CM (iohexol and iopamidol). Nonionic CM (loversol, iohexol, and iopamid01) induced adhesion of leukocytes on h u m a n endothelium to a greater extent than did ionic CM (diatrizoate and ioxaglate) [36J. Thrombogenic effects tested in a laser-induced rat thrombosis model were observed for both nonionic CM (iohexol and iopamidol) and ionic CM (diatrizoate and ioxaglate), but the thrombogenicity was the greatest for nonionic CM [37]. Additionally, an in vitro study by Corot et al. [38] showed that the ionic ioxaglate inhibited coagulation and platelet activation to a greater extent than did the nonionic iodixanol. The weaker anticoagulant effect of nonionic CM m a y be clinically relevant, especially in the realm of interventional cardiac procedures in which the important side effects are acute thrombotic vessel closure, acute myocardial infarction, thromboembolism, stroke, and death.
410
Vol. 3, No. 5, May 1996
REFERENCES 1. Karstoft J, Baath L, Jansen I, Edvinsson L. Vasoconstriction of isolated arteries induced by angiographic contrast media. Acta Radiol 1995; 36:312-316. 2. Jorgensen NP, Nossen JO, Borch KW, Kristiansen AB, Lundby B, Theodorsen L. Safety and tolerability of iodixano[ in healthy volunteers with reference to two monomeric X-ray contrast media. Eur J Radiol 1992; 15:252-257. 3. Svaland MG, Haider T, Langseth-Manrique K, Andrew E, Hafs PA. Human pharmokinetics of iodixanol. Invest Radio11992;27:130-133. 4. Nossen JO, Aakhus T, Berg KJ, Jorgensen NP, Andrew E. Experience with iodixanol, a new nonionic dimeric contrast medium: preliminary results from the human phase I study. Invest Radio11990;25[suppl]:S113-$114. 5. Bolstad B, Borch KW, Grynne BH, et al. Safety and tolerability of iodixanol: a dimeric, nonionic contrast medium. An emphasis on European clinical phase I and II. Invest Radio11991 ;26[suppl]:S201-S204. 6. Anderson PE, Bolstad B, Berg KJ, Justesen P, Thayssen E Kloster YF. Iodixanol and ioxaglate in cardioangiography: a double-blind randomized phase Ill study. Clin Radio11993;48:268-272. 7. Hill JA, Cohen MB, Kou WH, et al. Iodixanol, a new isosmotic nonionic contrast agent compared with iohexol in cardiac angiography. Am J Cardio11994;74:57-63. 8. Klew NE, Levorstad K, Berg KJ, et al. Iodixanol in c&dioangiography in patients with coronary artery disease. Acta Radio11993;34:72-76. 9. Harnish PP, Feuntaine H, Ebrahimi R. Experience in 1,259 patients in the United States. Invest Radio11994 ;29[suppl]:S236-S237. 10. Tveit K, Bolz KD, Bolstad B, et al..todixanol in cardioangiography: a double-blind parallel comparison between iodixanol-320 mg I/ml and ioxaglate 320 mg I/ml. Acta Radio11994;35:614-618. 11. Gavant ML, Siegle R L. ]odixanol in excretory urography: initial clinical experience with a nononic, dimeric (ratio 6:1) contrast medium. Radiology
1992;183:515-518. 12. Conroy RM, Bjartveit K, Sheppick A, Long U, Masterson J. Iodixanol in intravenous urography: a comparison of iodixanol 270 mg I/ml, iodixanol 320 mg I/m] and iopamidol 300 mg l/ml (NIOPAM). Clin Radiol 1994; 49:337-340. 13. Palmers Y, Greef DD, Grynne BH, Smits J, Put E. A double-blind study comparing safety, tolerability and efficacy of iodixanol 320-mg I/ml and ioxaglate 320 mg I/ml in cerebral arteriography. Eur J Radiol 1993; 17:203-209. 14. Poirier VC, Monsein LH, Newberry PD, Kreps BJ. Double-blind, randomized comparison of iodixanol 320 and iohexol 300 for cerebral angiography. Invest Radio11994;29[suppl]:S43-S44. 15. Pugh ND, Sissons GRJ, Ruttley MST, Berg KJ, Nossen JO, Eide H. Iodixanol in femoral arteriography (phase III): a comparative double-blind parallel trial between iodixanol and iopromide. Clin Radio11993;47:96-99. 16. Afbrechtsson U, Larusdottir H, Nergren L, Lundby B. Iodixanol a new nonionic dimer in aortofemora] angiography. Acta Radio11992;33:611-613. 17. Thorstensen O, Albrechtsson U, Calissendorff B, et al. Iodixanol in femoral arteriography: a randomized, double-blind, phase III, parallel study with iodixanol 270 mg I/m] and iohexol 300 mg I/ml. Acta Radiol 1994; 35:629-631. 18. Melton LG, Muga KM, Gabriel DA. Effect of contrast media on in vitro bleeding time: assessment by a hollow fiber instrument. Aead Radiol
1995;2:239-243. 19. Lotspeich-Steininger CA, Stiene-Martin EA, Koepke JA, eds. Clinical hematology: principles, procedures, correlations. Philadelphia: Lippincott, 1992:599-604. 20. Muga KM, Melton LG, Gabriel DA. A flow dynamic technique used to assess global hemostasis. Blood Coag Fibrinolysis 1995;6:73-78. 21. Gorog E Ahmed A. Haemostatometer: a new in vitro technique for assessing haemostatic activity of blood. Thromb Res 1984;34:341-357. 22. Gorog R A new, ideal technique to monjtor thrombolytic therapy. Angiology 1986;37:99-105. 23. Harfenist EJ, Packman MA, Kinglough-Rathbone RL, Cataneo M, Mustard JF. Effect of calcium ion concentration on the ability of fibrinogen and
Vol. 3, No. 5, May 1996
24. 25. 26. 27.
28. 29. 30.
31.
IODIXANOL
von Willebrand factor to support the ADP-induced aggregation of human platelets. Blood 1987;70:827-831. Born GVR, Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 1962;194:272-276. Cardinal DL, Flower RJ. The electronic aggregometer: a novel device for assessing platelet behaviour in blood. J Pharmacol Methods 1980;3:135-158, John LCH, Rees GM, Kovacs lB. Inhibition of platelet function by heparin. J Thorac Cardiovasc Surg 1993;105:816-822. Kovacs IB, Hutton RA, Kernoff PBA, Hemostatic evaluation in bleeding disorders from native blood: clinical experience with the haemostatometer. Am J Patho11989;91:271-279. Gorog P, Kakkar VV. Transient effect of aspirin on collagen-induced platelet accumulation. Am J Clin Pathol 1988;86:311-316. Gorog P, Kovacs lB. Coagulation of flowing native blood: advantages over stagnant (tube) clotting tests. Thromb Res 1991 ;64:611-619. John LCH, Rees GM, Kovacs lB. Effect of heparin on in vitro platelet reactivity in cardiac surgical patients: a comparative assessment by whole blood platelet aggregometry and haemostatometry. Thromb Res 1992; 66:649-656. Kovacs IB, Gorog P. Simultaneous measurement of all thrombosis parameters from native human blood: usefulness in monitoring efficacy and complications of thrombolytic therapy. Angiology 1990;41:829-835.
A N D IN V I T R O
BLEEDING
TIME
32. Peters NS, Abrams LSM, Dymond DS, Kovacs lB. Platelet hyperreactivity and inefficient spontaneous thrombolysis in patients at high risk from an acute coronary event. Cardiovasc Res 1989;23:567-572. 33. Ratnatunga CP, Edmondson SF, Rees GM, Kovacs lB. High-dose aspirin inhibits shear-induced platelet reaction involving thrombin generation. Circulation 1992;85:1077-1082. 34. Hay KL, Bull BS. An analysis of platelet activation and aggregation produced by three classes of contrast media. J Vasc Intervent Radio11996; 6:211-217. 35. Koza MJ, Shankey TV, Walenga JM, Moncada R, Fareed J, Pifarre R. Flow cytometric evaluation of platelet activation by ionic or nonionic contrast media and modulation by heparin and recombinant hiruden. Invest Radio11998;30:90-97. 36. Abeyama K, Oh S, Kawano K, et al. Nonionic contrast agents produce thrombotic effect by inducing adhesion of leukocytes on human endothelium. Biochem Biophys Res Commun 1995;212:776-783. 37. Aguejouf O, Doutremepuich F, Azougagh OF, Doutremepuich C. Thrombogenicity of ionic and nonionic contrast media tested in a laser induced rat thrombosis model. Thromb Res 1995;77:259-269. 38. Corot C, Chronos N, Sabattuer V, Gruentzig A. Iodinated contrast media structure activity relationship based on in vitro coagulation and platelet activation investigations. Cardiovasc Intervent Radio11995;18[suppl] :S107,
Announcements The Johns Hopkins University School of Medicine and the Eberhard-Karls-Universit~tt are sponsoring Abdominal Radiology: Practical Approach and Recent Advances, which will be held June 13-15, 1996, at Kongresshaus in Salzburg, Austria. Credit will be announced. The fee is $375. For more information, contact the Program Coordinator, Johns Hopkins Medical Institutions, Office of Continuing Medical Education, Turner Bldg., 720 Rutland Ave., Baltimore, MD 21205; (410) 955-2959, fax (410) 955-0807.
The Department of Radiology of the Johns Hopkins Medical Institutions is sponsoring Advanced Topics in CT w i t h Emphasis o n Spiral CT, which will be held July 13-16, 1996, at the Eldorado Hotel in Santa Fe, NM. Thirteen hours of AMA Category 1 and 13 hours of Category 1 ACE-ASRT credit will be awarded. The fee is $500 for physicians and $450 for residents, fellows, and CRNAs. The program directors are Elliot-K. Fishman, MD, and Stanley S. Siegelman, MD. For more information, contact the Program Coordinator, Office of Continuing Medical Education, Johns Hopkins Medical Institutions, Turner 20, 720 Rutland Ave., Baltimore, MD 21205-2195; (410) 955-2959; fax (410) 955-08O7.
41 1