- Email: [email protected]
Influence of plasma expansion on plasma protein binding of ketorolac
Original Contributions Influence of Plasma Expansion on Plasma Protein Binding of Ketorolac Dietrich Gravenstein, MD,* Ajit Suri, PhD,† Hartmut C. Derendorf, PhD,‡ A. Jay Koska, MD, PhD§ Department of Anesthesiology, University of Florida College of Medicine; and Department of Pharmaceutics, College of Pharmacy, University of Florida, Gainesville, FL
*Assistant Professor of Anesthesiology, University of Florida †Research Scientist, University of Florida ‡Professor of Pharmaceutics, University of Florida §Assistant Professor of Anesthesiology, University of Texas Medical Branch at Galveston Address correspondence to Dr. Gravenstein at the Department of Anesthesiology, University of Florida College of Medicine, P.O. Box 100254, Gainesville, FL 32610 – 0254, USA. Received for publication September 25, 1997; revised manuscript accepted for publication May 13, 1998.
Study Objective: To determine the effect of dilution with intravascular volume expanders commonly used by anesthesiologists on clinically relevant levels of free serum ketorolac. Design: In vitro study. Setting: Pharmaceutics laboratory of a medical college. Interventions: The effect of 6% hydroxyethylstarch, 5% albumin, 6% dextran 60, and lactated Ringer’s solution on in vitro plasma protein binding of ketorolac was investigated by ultrafiltration. The binding was studied at three different drug concentrations: low therapeutic (0.3 mg/ml), high therapeutic (3 mg/ml), and toxic (10 mg/ml), and at two or more volume expander dilutions. Measurements and Main Results: The effect of plasma dilution on free ketorolac was consistent across all volume expanders tested and for each ketorolac concentration studied. As the plasma dilution with albumin, hydroxyethylstarch, dextran 60, or lactated Ringer’s solution increased, the unbound ketorolac also increased from 3.2% to 3.3% in undiluted plasma to 5.0% to 8.7% in 50% dilution of the plasma with the investigated expanders. Dilution of plasma by only 10% resulted in a significant, but relatively minor, increase of unbound ketorolac to 3.2% to 3.8%. Conclusion: Because of the pharmacokinetic properties of ketorolac, this pharmacokinetic interaction can be expected to have only minor effects on unbound ketorolac concentrations when ketorolac is administered after the plasma expander. When ketorolac administration is followed by rapid plasma expander infusion, a transient increase of unbound ketorolac in plasma can be expected. © 1998 by Elsevier Science Inc. Keywords: Albumin; dextran 60; hydroxyethylstarch; ketorolac; plasma; protein binding; Ringer’s solution, lactated.
Introduction Ketorolac is a potent analgesic and member of the nonsteroidal anti-inflammatory drug (NSAID) class. Its therapeutic advantages over opioid analgesic drugs are that it is nonsedating, it is not associated with pruritis, nausea, respiratory depression or alteration of the carbon dioxide (CO2)–minute ventilation response curve, urinary retention or ileus, and it has opioid-sparing qualities. Like other NSAIDs, however, ketorolac has been implicated as a contributing
Journal of Clinical Anesthesia 10:464 – 468, 1998 © 1998 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
0952-8180/98/$19.00 PII S0952-8180(98)00064-6
Ketorolac binding to volume expanders: Gravenstein et al.
cause of increased postoperative bleeding,1 renal failure, and gastritis.2 The severity of these side effects is probably dose-related.3,4 Ketorolac has been administered prophylactically prior to surgery and found to reduce postoperative pain and frequency of supplemental analgesic therapy.5,6 Ketorolac also has been demonstrated to be an effective analgesic drug when administered postoperatively, reducing opioid requirements by up to 40%.7–9 Ketorolac is highly bound to proteins in human plasma. Patients who undergo acute normovolemic hemodilution, or who need significant intravascular volume resuscitation, may experience alterations of plasma ketorolac levels. Because albumin and other colloids are commonly used for volume expansion, it was the goal of this study to determine how clinically relevant serum dilution with commonly used volume expanders influenced the free fraction of ketorolac in plasma. Heightened appreciation of the effects of volume expanders on unbound ketorolac levels allows the risk of side effects to be minimized while maintaining analgesia.
unbound (fu), which, when multiplied by 100, results in the percentage of free ketorolac. Statistical analysis of data was performed using Microsoft EXCEL (Microsoft, Redmond, WA) and Sigma Stat (Jandel, San Rafael, CA). The effect of dilution, concentration, and expander on protein binding was analyzed using analysis of variance for repeated measures followed by Tukey’s multiple comparison test. P-values of 0.05 or less were considered to show statistically significant differences.
Results The effect of plasma dilution on free ketorolac was consistent across all volume expanders tested and for each ketorolac concentration studied. As the plasma dilution with ALB, HES, DEX, or LR increased, the percent of free ketorolac also increased (Table 1).
Undiluted Plasma Percent unbound of ketorolac in plasma was found to be 3.31, 3.17, and 3.33 at 0.3, 3, and 10 mg/ml, respectively.
Materials and Methods Binding of ketorolac was studied with four commonly used intravascular volume expanders: 6% hydroxyethylstarch [6% Hetastarch; Abbott Laboratories, North Chicago, IL (HES)], 5% albumin [Baxter, Glendale, CA (ALB)], 6% dextran 60 [Macrodex 6%; Schiwa, Glandorf, Germany (DEX)], and lactated Ringer’s solution [Baxter, Deerfield, IL (LR)]. The binding was studied at three different drug concentrations: low therapeutic (0.3 mg/ml), high therapeutic (3 mg/ml), and toxic (10 mg/ml).10,11 The binding was studied at four different plasma dilutions: undiluted plasma, a 10% dilution and 50% dilution of the plasma with volume expander, and in 100% volume expander. In the case of lactated Ringer’s solution, 20%, 30%, and 40% dilutions also were studied. The experiments were performed in triplicate at each of the four dilutions for each of the three ketorolac concentrations. Appropriate volume of plasma and each volume expander were mixed to prepare a 10 ml solution of each dilution. Three 2.4 ml aliquots were prepared from each of these dilutions; 100 ml of appropriate solution (7.5 mg/ml, 75 mg/ml, or 250 mg/ml) of ketorolac was then added to each aliquot to obtain the desired concentrations (0.3 mg/ml, 3 mg/ml, and 10 mg/ml, respectively). Each one of these concentrations for each of the dilutions was divided again into triplicate volumes of 700 ml each. All of these samples were spiked with 100 ml of tritium-labeled ketorolac (Syntex Laboratories, Palo Alto, CA). Each sample was vortexed and 100 ml withdrawn for analysis by liquid scintillation counter for measuring activity due to total concentration. A 500 ml aliquot was also taken for ultrafiltration (Centricon 30 Filter, Amicon Inc., Beverly, MA) performed at 6,000 rpm at 10°C for 15 minutes; 100 ml of filtrate was taken for liquid scintillation counting to measure activity due to free concentration. The ratio of free to total counts was used to determine the fraction
Dilution With Albumin No trends of change in protein binding of ketorolac with increasing concentrations of ketorolac was observed at 10% dilution. However, there was a slight trend toward decrease in fraction unbound on increasing concentration of ketorolac at 50% dilution. However, the trend was not found to be statistically significant. The percent of unbound ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 1).
Dilution With Hydroxyethylstarch HES: There was no apparent trend of change in protein binding at various concentrations of ketorolac, at each of two dilution levels. However, the percent unbound of ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 2).
Dilution With 6% Dextran 60 The fraction unbound of ketorolac did not show any trend at various concentrations of ketorolac at 10% dilution of plasma. However, the fraction unbound did show an upward trend on increasing concentration of ketorolac. The percent unbound of ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 3).
Dilution With Lactated Ringer’s Solution Protein binding of ketorolac was determined at five dilution levels (from 10% to 50%) with lactated Ringer’s J. Clin. Anesth., vol. 10, September 1998
465
Original Contributions
Table 1. Percentage of Free Ketorolac in Plasma in the Presence of Plasma Expanders in Different Dilutions 0.3 mg/ml
Plasma 100% ALB 10%* 50%† 100%‡ HES 10% 50%† 100%‡ DEX 10%* 50%† 100%‡ LR 10%* 20%* 30%* 40%† 50%† 100%‡
3 mg/ml
10 mg/ml
Mean
SD
Mean
SD
Mean
SD
3.31
0.10
3.17
0.25
3.33
0.41
3.89 7.49 19.07
0.18 0.04 0.21
3.71 7.34 19.29
0.11 0.24 0.79
3.72 6.84 17.69
0.29 0.23 0.48
3.29 4.97 78.84
0.11 0.15 2.41
3.24 5.14 80.34
0.39 0.31 0.92
3.48 5.48 77.41
0.16 0.09 1.05
3.66 5.81 84.40
0.08 0.22 0.45
3.60 5.85 85.20
0.06 0.16 1.94
3.66 8.73 85.28
0.08 0.13 0.89
3.80 4.73 4.34 4.80 5.31 98.81
0.24 0.02 0.14 0.09 0.15 0.46
3.65 4.31 4.40 4.75 5.19 99.75
0.25 0.12 0.17 0.13 0.25 2.21
3.65 4.10 4.20 4.67 5.50 97.68
0.22 0.17 0.17 0.30 0.14 0.91
Note: Data are means and SD. ALB 5 5% albumin; HES 5 6% hydroxyethylstarch; DEX 5 6% dextran 60; LR 5 lactated Ringer’s solution. *p , 0.05, difference at all concentrations to plasma. †p , 0.05, difference at 10% dilution. ‡p , 0.05, difference at 50% dilution.
Figure 1. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 5% albumin. Ketorolac binding in the 50% dilution was different from the 10% dilution and plasma; the 10% dilution was also different from plasma (p , 0.05). 466
J. Clin. Anesth., vol. 10, September 1998
Figure 2. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 6% hydroxyethylstarch. Ketorolac binding in the 50% dilution was different from 10% dilution and plasma (p , 0.05).
Ketorolac binding to volume expanders: Gravenstein et al.
Figure 3. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 6% dextran 60. Ketorolac binding in the 50% dilution was different from the 10% dilution and plasma; the 10% dilution was also different from plasma (p , 0.05).
solution. No clear trend of change in ketorolac free fraction with increasing ketorolac concentration was observed at any of five dilutions studied. However, percent free fraction of ketorolac did show an increasing trend on increasing the level of dilution with LR (Figure 4).
Figure 5. Percentage of free ketorolac (means and SD) in undiluted plasma expander studied at three different concentrations (0.3, 3, and 10 mg/ml). PLA 5 plasma, ALB 5 albumin (5%), HES 5 6% hydroxyethylstarch, DEX 5 6% dextran 60, LR 5 lactated Ringer’s solution. All the expanders showed significant differences in binding when compared with any other expander (p , 0.05).
Protein binding of ketorolac in plasma did not change significantly with increasing concentrations of ketorolac from 0.3 mg/ml to 10 mg/ml. However, for all plasma expanders studied, percent unbound was increased in diluted plasma and was significantly higher at 50% dilution than at 10% dilution. The increase in percent free ketorolac compared with undiluted plasma was significant at plasma dilutions of 10% with ALB, DEX and with LR. It was not statistically significant for a 10% dilution with HES, but was significant at 50% HES dilution. When the undiluted volume expanders were compared, a consistent binding effect was observed across the three ketorolac concentrations examined. Undiluted plasma had the least percent free ketorolac, followed by ALB, HES, DEX, and finally LR solution (Figure 5). All of these differences between the expanders were statistically significant.
Discussion
Figure 4. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and six different dilutions (0, 10%, 20%, 30%, 40%, and 50%) with lactated Ringer’s solution. Ketorolac binding in the 40% and 50% dilutions was different from the 10% dilution and plasma; the 10%, 20% and 30% dilutions also were different from plasma (p , 0.05).
Physicians’ concern about the safety of ketorolac and other NSAID class drugs has led some to reduce or eliminate the use of this drug class in their patients. Our research shows one possible mechanism by which some patients who have been treated prophylactically with ketorolac prior to volume expansion may become exposed to toxic concentrations of ketorolac. Understanding the pharmacokinetics of volume expanders on ketorolac free drug concentrations may allow dosing to be altered in such a manner that the risk of drug-related complications is decreased. It should be kept in mind that our results are based on in vitro plasma protein binding studies, and that it may be difficult to predict precisely which effect this change in J. Clin. Anesth., vol. 10, September 1998
467
Original Contributions
protein binding will have in vivo. Ketorolac has a total body clearance of 0.35 ml/kg-min21,10 which translates into approximately 1.5 L/hr for a 70 kg patient. Hence, the drug represents a low-extraction drug for which the total body clearance approximates the product of fraction unbound and intrinsic clearance (CLtot 5 fu 3 CLint). The average total plasma concentration at steady state is the ratio of infusion rate (R0) and total body clearance (Cpss 5 R0/Cltot). Therefore, a decrease in plasma protein binding such as that observed in this study induced by the plasma expanders will increase the total body clearance and decrease steady-state plasma concentrations proportionally. On the other hand, the respective unbound steady-state concentration (fu 3 Cpss), which is responsible for pharmacologic activity, will be the same (fu 3 Cpss 5 R0/CLint). However, there may be an instantaneous release of protein bound drug when the plasma expander dilutes the plasma, which will lead to transient high unbound concentrations of ketorolac. This is of particular importance when the expander is administered rapidly. A similar effect has been described when sulfisoxazole displaces bilirubin from protein binding sites.12 The resulting high concentrations can be potentially harmful to the patient. Another effect of the changing protein binding is an increase of the drug’s volume of distribution (Vd). Ketorolac has a small Vd of 0.11 L/kg.10 In a 70 kg patient, this value is approximately 8 L. This small Vd is caused by the high plasma protein binding, which ensures that the drug remains in the vascular space. A decrease in plasma protein binding caused by plasma expanders will increase Vd; a change from 3% to 7% may approximately double Vd. Hence, the decreased plasma protein binding will be compensated by a redistribution of ketorolac from the intravascular into the extravascular space. Because clearance and Vd will change proportionally, there will be no apparent change in the drug’s half-life of approximately 5 hours. It should be kept in mind that all of these extrapolations are valid only if there is no saturation of binding sites or any other nonlinearity present. Risk factors for ketorolac toxicity include increased age, recent major surgery, and underlying cardiovascular disease.2 In the elderly, ketorolac elimination is decreased 30%, although its absorption and plasma protein binding are not substantially altered compared with that seen in young adults.11 Current dosing recommendations for ketorolac advise cautious dosing in patients with renal failure, low cardiac output states, bleeding disorders, gastritis, and allergies to NSAIDs (Package insert, Roche Pharmacueticals, Nutley, NJ). The rise in free ketorolac observed with plasma dilution by volume expanders may result in an increased risk for sustaining a drug-related complication. Therefore, it may be wise to include hemodilution as another indication for cautious dosing. This action would be specifically pertinent when ketorolac has been administered prophylactically prior to major volume resuscita-
468
J. Clin. Anesth., vol. 10, September 1998
tion, where the plasma expander will lead to displacement of ketorolac from binding sites. In these situations, the plasma expanders would be best administered by slow infusion, if possible, to allow sufficient time for reequilibration of ketorolac. It can be concluded from this work that minor hemodilution of 10% by any of the tested volume expanders leads to relatively minor and consistent, but significant, increases in the free ketorolac fraction. However, a dramatic increase of free ketorolac fraction occurs when plasma dilution approaches 50%. From a theoretical point of view, this interaction will result in transient increases in unbound, active ketorolac concentrations. If these results obtained in the laboratory are verified in a clinical study, the implications are clear: the risk of drug toxicity can be reduced without compromising analgesia by adjusting current dosing recommendations.
References 1. Rusy LM, Houck CS, Sullivan LJ, et al: A double-blind evaluation of ketorolac tromethamine versus acetaminophen in pediatric tonsillectomy: analgesia and bleeding. Anesth Analg 1995;80: 226 –9. 2. Corelli RL, Gericke KR: Renal insufficiency associated with intramuscular administration of ketorolac tromethamine. Ann Pharmacother 1993;27:1055–7. 3. Green D, Klement P, Liao P, Weitz J: Interaction of low molecular weight heparin with ketorolac. J Lab Clin Med 1996;127:583–7. 4. Hyers TM, Tricomi SM, Liao JJ: Inhibition of some human neutrophil functions by the cyclooxygenase inhibitor ketorolac tromethamine. J Leukoc Biol 1992;51:490 –5. 5. Peduto VA, Toscano A, D’Uva R, Piga M: Ketorolac for prevention of acute postoperative pain. Minerva Anestesiol 1995;61:367– 72. 6. Watcha MF, Ramirez-Ruiz M, White PF, Jones MB, Lagueruela RG, Terkonda RP: Perioperative effects of oral ketorolac and acetaminophen in children undergoing bilateral myringotomy. Can J Anaesth 1992;39:649 –54. 7. Grass JA, Sakima NT, Valley M, et al: Assessment of ketorolac as an adjuvant to fentanyl patient-controlled epidural analgesia after radical retropubic prostatectomy. Anesthesiology 1993;78: 642– 8. 8. Kinsella J, Moffat AC, Patrick JA, Prentice JW, McArdle CS, Kenny GN: Ketorolac trometamol for postoperative analgesia after orthopaedic surgery. Br J Anaesth 1992;69:19 –22. 9. Cataldo PA, Senagore AJ, Kilbride MJ: Ketorolac and patient controlled analgesia in the treatment of postoperative pain. Surg Gynecol Obstet 1993;176:435– 8. 10. Jung D, Mroszczak EJ, Bynum L: Pharmackinetics of ketorolac tromethamine in humans after intravenous, intramuscular and oral administration. Eur J Clin Pharmacol 1988;35:423–5. 11. Jallad NS, Garg DC, Martinez JJ, Mroszczak EJ, Weidler DJ: Pharmacokinetics of the single-dose oral and intramuscular ketorolac tromethamine in the young and the elderly. J Clin Pharmacol 1990;30:76 – 81. 12. Oie S, Levy G: Effect of sulfisoxazole on pharmacokinetics of free and plasma protein-bound bilirubin in experimental unconjugated hyperbilirubinemia. J Pharm Sci 1979;68:6 –9.
*Assistant Professor of Anesthesiology, University of Florida †Research Scientist, University of Florida ‡Professor of Pharmaceutics, University of Florida §Assistant Professor of Anesthesiology, University of Texas Medical Branch at Galveston Address correspondence to Dr. Gravenstein at the Department of Anesthesiology, University of Florida College of Medicine, P.O. Box 100254, Gainesville, FL 32610 – 0254, USA. Received for publication September 25, 1997; revised manuscript accepted for publication May 13, 1998.
Study Objective: To determine the effect of dilution with intravascular volume expanders commonly used by anesthesiologists on clinically relevant levels of free serum ketorolac. Design: In vitro study. Setting: Pharmaceutics laboratory of a medical college. Interventions: The effect of 6% hydroxyethylstarch, 5% albumin, 6% dextran 60, and lactated Ringer’s solution on in vitro plasma protein binding of ketorolac was investigated by ultrafiltration. The binding was studied at three different drug concentrations: low therapeutic (0.3 mg/ml), high therapeutic (3 mg/ml), and toxic (10 mg/ml), and at two or more volume expander dilutions. Measurements and Main Results: The effect of plasma dilution on free ketorolac was consistent across all volume expanders tested and for each ketorolac concentration studied. As the plasma dilution with albumin, hydroxyethylstarch, dextran 60, or lactated Ringer’s solution increased, the unbound ketorolac also increased from 3.2% to 3.3% in undiluted plasma to 5.0% to 8.7% in 50% dilution of the plasma with the investigated expanders. Dilution of plasma by only 10% resulted in a significant, but relatively minor, increase of unbound ketorolac to 3.2% to 3.8%. Conclusion: Because of the pharmacokinetic properties of ketorolac, this pharmacokinetic interaction can be expected to have only minor effects on unbound ketorolac concentrations when ketorolac is administered after the plasma expander. When ketorolac administration is followed by rapid plasma expander infusion, a transient increase of unbound ketorolac in plasma can be expected. © 1998 by Elsevier Science Inc. Keywords: Albumin; dextran 60; hydroxyethylstarch; ketorolac; plasma; protein binding; Ringer’s solution, lactated.
Introduction Ketorolac is a potent analgesic and member of the nonsteroidal anti-inflammatory drug (NSAID) class. Its therapeutic advantages over opioid analgesic drugs are that it is nonsedating, it is not associated with pruritis, nausea, respiratory depression or alteration of the carbon dioxide (CO2)–minute ventilation response curve, urinary retention or ileus, and it has opioid-sparing qualities. Like other NSAIDs, however, ketorolac has been implicated as a contributing
Journal of Clinical Anesthesia 10:464 – 468, 1998 © 1998 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
0952-8180/98/$19.00 PII S0952-8180(98)00064-6
Ketorolac binding to volume expanders: Gravenstein et al.
cause of increased postoperative bleeding,1 renal failure, and gastritis.2 The severity of these side effects is probably dose-related.3,4 Ketorolac has been administered prophylactically prior to surgery and found to reduce postoperative pain and frequency of supplemental analgesic therapy.5,6 Ketorolac also has been demonstrated to be an effective analgesic drug when administered postoperatively, reducing opioid requirements by up to 40%.7–9 Ketorolac is highly bound to proteins in human plasma. Patients who undergo acute normovolemic hemodilution, or who need significant intravascular volume resuscitation, may experience alterations of plasma ketorolac levels. Because albumin and other colloids are commonly used for volume expansion, it was the goal of this study to determine how clinically relevant serum dilution with commonly used volume expanders influenced the free fraction of ketorolac in plasma. Heightened appreciation of the effects of volume expanders on unbound ketorolac levels allows the risk of side effects to be minimized while maintaining analgesia.
unbound (fu), which, when multiplied by 100, results in the percentage of free ketorolac. Statistical analysis of data was performed using Microsoft EXCEL (Microsoft, Redmond, WA) and Sigma Stat (Jandel, San Rafael, CA). The effect of dilution, concentration, and expander on protein binding was analyzed using analysis of variance for repeated measures followed by Tukey’s multiple comparison test. P-values of 0.05 or less were considered to show statistically significant differences.
Results The effect of plasma dilution on free ketorolac was consistent across all volume expanders tested and for each ketorolac concentration studied. As the plasma dilution with ALB, HES, DEX, or LR increased, the percent of free ketorolac also increased (Table 1).
Undiluted Plasma Percent unbound of ketorolac in plasma was found to be 3.31, 3.17, and 3.33 at 0.3, 3, and 10 mg/ml, respectively.
Materials and Methods Binding of ketorolac was studied with four commonly used intravascular volume expanders: 6% hydroxyethylstarch [6% Hetastarch; Abbott Laboratories, North Chicago, IL (HES)], 5% albumin [Baxter, Glendale, CA (ALB)], 6% dextran 60 [Macrodex 6%; Schiwa, Glandorf, Germany (DEX)], and lactated Ringer’s solution [Baxter, Deerfield, IL (LR)]. The binding was studied at three different drug concentrations: low therapeutic (0.3 mg/ml), high therapeutic (3 mg/ml), and toxic (10 mg/ml).10,11 The binding was studied at four different plasma dilutions: undiluted plasma, a 10% dilution and 50% dilution of the plasma with volume expander, and in 100% volume expander. In the case of lactated Ringer’s solution, 20%, 30%, and 40% dilutions also were studied. The experiments were performed in triplicate at each of the four dilutions for each of the three ketorolac concentrations. Appropriate volume of plasma and each volume expander were mixed to prepare a 10 ml solution of each dilution. Three 2.4 ml aliquots were prepared from each of these dilutions; 100 ml of appropriate solution (7.5 mg/ml, 75 mg/ml, or 250 mg/ml) of ketorolac was then added to each aliquot to obtain the desired concentrations (0.3 mg/ml, 3 mg/ml, and 10 mg/ml, respectively). Each one of these concentrations for each of the dilutions was divided again into triplicate volumes of 700 ml each. All of these samples were spiked with 100 ml of tritium-labeled ketorolac (Syntex Laboratories, Palo Alto, CA). Each sample was vortexed and 100 ml withdrawn for analysis by liquid scintillation counter for measuring activity due to total concentration. A 500 ml aliquot was also taken for ultrafiltration (Centricon 30 Filter, Amicon Inc., Beverly, MA) performed at 6,000 rpm at 10°C for 15 minutes; 100 ml of filtrate was taken for liquid scintillation counting to measure activity due to free concentration. The ratio of free to total counts was used to determine the fraction
Dilution With Albumin No trends of change in protein binding of ketorolac with increasing concentrations of ketorolac was observed at 10% dilution. However, there was a slight trend toward decrease in fraction unbound on increasing concentration of ketorolac at 50% dilution. However, the trend was not found to be statistically significant. The percent of unbound ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 1).
Dilution With Hydroxyethylstarch HES: There was no apparent trend of change in protein binding at various concentrations of ketorolac, at each of two dilution levels. However, the percent unbound of ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 2).
Dilution With 6% Dextran 60 The fraction unbound of ketorolac did not show any trend at various concentrations of ketorolac at 10% dilution of plasma. However, the fraction unbound did show an upward trend on increasing concentration of ketorolac. The percent unbound of ketorolac was found to be significantly higher at 50% dilution compared with 10% dilution, and it was also higher than that in undiluted plasma (Figure 3).
Dilution With Lactated Ringer’s Solution Protein binding of ketorolac was determined at five dilution levels (from 10% to 50%) with lactated Ringer’s J. Clin. Anesth., vol. 10, September 1998
465
Original Contributions
Table 1. Percentage of Free Ketorolac in Plasma in the Presence of Plasma Expanders in Different Dilutions 0.3 mg/ml
Plasma 100% ALB 10%* 50%† 100%‡ HES 10% 50%† 100%‡ DEX 10%* 50%† 100%‡ LR 10%* 20%* 30%* 40%† 50%† 100%‡
3 mg/ml
10 mg/ml
Mean
SD
Mean
SD
Mean
SD
3.31
0.10
3.17
0.25
3.33
0.41
3.89 7.49 19.07
0.18 0.04 0.21
3.71 7.34 19.29
0.11 0.24 0.79
3.72 6.84 17.69
0.29 0.23 0.48
3.29 4.97 78.84
0.11 0.15 2.41
3.24 5.14 80.34
0.39 0.31 0.92
3.48 5.48 77.41
0.16 0.09 1.05
3.66 5.81 84.40
0.08 0.22 0.45
3.60 5.85 85.20
0.06 0.16 1.94
3.66 8.73 85.28
0.08 0.13 0.89
3.80 4.73 4.34 4.80 5.31 98.81
0.24 0.02 0.14 0.09 0.15 0.46
3.65 4.31 4.40 4.75 5.19 99.75
0.25 0.12 0.17 0.13 0.25 2.21
3.65 4.10 4.20 4.67 5.50 97.68
0.22 0.17 0.17 0.30 0.14 0.91
Note: Data are means and SD. ALB 5 5% albumin; HES 5 6% hydroxyethylstarch; DEX 5 6% dextran 60; LR 5 lactated Ringer’s solution. *p , 0.05, difference at all concentrations to plasma. †p , 0.05, difference at 10% dilution. ‡p , 0.05, difference at 50% dilution.
Figure 1. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 5% albumin. Ketorolac binding in the 50% dilution was different from the 10% dilution and plasma; the 10% dilution was also different from plasma (p , 0.05). 466
J. Clin. Anesth., vol. 10, September 1998
Figure 2. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 6% hydroxyethylstarch. Ketorolac binding in the 50% dilution was different from 10% dilution and plasma (p , 0.05).
Ketorolac binding to volume expanders: Gravenstein et al.
Figure 3. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and three different dilutions (0, 10%, and 50%) with 6% dextran 60. Ketorolac binding in the 50% dilution was different from the 10% dilution and plasma; the 10% dilution was also different from plasma (p , 0.05).
solution. No clear trend of change in ketorolac free fraction with increasing ketorolac concentration was observed at any of five dilutions studied. However, percent free fraction of ketorolac did show an increasing trend on increasing the level of dilution with LR (Figure 4).
Figure 5. Percentage of free ketorolac (means and SD) in undiluted plasma expander studied at three different concentrations (0.3, 3, and 10 mg/ml). PLA 5 plasma, ALB 5 albumin (5%), HES 5 6% hydroxyethylstarch, DEX 5 6% dextran 60, LR 5 lactated Ringer’s solution. All the expanders showed significant differences in binding when compared with any other expander (p , 0.05).
Protein binding of ketorolac in plasma did not change significantly with increasing concentrations of ketorolac from 0.3 mg/ml to 10 mg/ml. However, for all plasma expanders studied, percent unbound was increased in diluted plasma and was significantly higher at 50% dilution than at 10% dilution. The increase in percent free ketorolac compared with undiluted plasma was significant at plasma dilutions of 10% with ALB, DEX and with LR. It was not statistically significant for a 10% dilution with HES, but was significant at 50% HES dilution. When the undiluted volume expanders were compared, a consistent binding effect was observed across the three ketorolac concentrations examined. Undiluted plasma had the least percent free ketorolac, followed by ALB, HES, DEX, and finally LR solution (Figure 5). All of these differences between the expanders were statistically significant.
Discussion
Figure 4. Percentage of free ketorolac (means and SD) in plasma studied at three different concentrations (0.3, 3, and 10 mg/ml) and six different dilutions (0, 10%, 20%, 30%, 40%, and 50%) with lactated Ringer’s solution. Ketorolac binding in the 40% and 50% dilutions was different from the 10% dilution and plasma; the 10%, 20% and 30% dilutions also were different from plasma (p , 0.05).
Physicians’ concern about the safety of ketorolac and other NSAID class drugs has led some to reduce or eliminate the use of this drug class in their patients. Our research shows one possible mechanism by which some patients who have been treated prophylactically with ketorolac prior to volume expansion may become exposed to toxic concentrations of ketorolac. Understanding the pharmacokinetics of volume expanders on ketorolac free drug concentrations may allow dosing to be altered in such a manner that the risk of drug-related complications is decreased. It should be kept in mind that our results are based on in vitro plasma protein binding studies, and that it may be difficult to predict precisely which effect this change in J. Clin. Anesth., vol. 10, September 1998
467
Original Contributions
protein binding will have in vivo. Ketorolac has a total body clearance of 0.35 ml/kg-min21,10 which translates into approximately 1.5 L/hr for a 70 kg patient. Hence, the drug represents a low-extraction drug for which the total body clearance approximates the product of fraction unbound and intrinsic clearance (CLtot 5 fu 3 CLint). The average total plasma concentration at steady state is the ratio of infusion rate (R0) and total body clearance (Cpss 5 R0/Cltot). Therefore, a decrease in plasma protein binding such as that observed in this study induced by the plasma expanders will increase the total body clearance and decrease steady-state plasma concentrations proportionally. On the other hand, the respective unbound steady-state concentration (fu 3 Cpss), which is responsible for pharmacologic activity, will be the same (fu 3 Cpss 5 R0/CLint). However, there may be an instantaneous release of protein bound drug when the plasma expander dilutes the plasma, which will lead to transient high unbound concentrations of ketorolac. This is of particular importance when the expander is administered rapidly. A similar effect has been described when sulfisoxazole displaces bilirubin from protein binding sites.12 The resulting high concentrations can be potentially harmful to the patient. Another effect of the changing protein binding is an increase of the drug’s volume of distribution (Vd). Ketorolac has a small Vd of 0.11 L/kg.10 In a 70 kg patient, this value is approximately 8 L. This small Vd is caused by the high plasma protein binding, which ensures that the drug remains in the vascular space. A decrease in plasma protein binding caused by plasma expanders will increase Vd; a change from 3% to 7% may approximately double Vd. Hence, the decreased plasma protein binding will be compensated by a redistribution of ketorolac from the intravascular into the extravascular space. Because clearance and Vd will change proportionally, there will be no apparent change in the drug’s half-life of approximately 5 hours. It should be kept in mind that all of these extrapolations are valid only if there is no saturation of binding sites or any other nonlinearity present. Risk factors for ketorolac toxicity include increased age, recent major surgery, and underlying cardiovascular disease.2 In the elderly, ketorolac elimination is decreased 30%, although its absorption and plasma protein binding are not substantially altered compared with that seen in young adults.11 Current dosing recommendations for ketorolac advise cautious dosing in patients with renal failure, low cardiac output states, bleeding disorders, gastritis, and allergies to NSAIDs (Package insert, Roche Pharmacueticals, Nutley, NJ). The rise in free ketorolac observed with plasma dilution by volume expanders may result in an increased risk for sustaining a drug-related complication. Therefore, it may be wise to include hemodilution as another indication for cautious dosing. This action would be specifically pertinent when ketorolac has been administered prophylactically prior to major volume resuscita-
468
J. Clin. Anesth., vol. 10, September 1998
tion, where the plasma expander will lead to displacement of ketorolac from binding sites. In these situations, the plasma expanders would be best administered by slow infusion, if possible, to allow sufficient time for reequilibration of ketorolac. It can be concluded from this work that minor hemodilution of 10% by any of the tested volume expanders leads to relatively minor and consistent, but significant, increases in the free ketorolac fraction. However, a dramatic increase of free ketorolac fraction occurs when plasma dilution approaches 50%. From a theoretical point of view, this interaction will result in transient increases in unbound, active ketorolac concentrations. If these results obtained in the laboratory are verified in a clinical study, the implications are clear: the risk of drug toxicity can be reduced without compromising analgesia by adjusting current dosing recommendations.
References 1. Rusy LM, Houck CS, Sullivan LJ, et al: A double-blind evaluation of ketorolac tromethamine versus acetaminophen in pediatric tonsillectomy: analgesia and bleeding. Anesth Analg 1995;80: 226 –9. 2. Corelli RL, Gericke KR: Renal insufficiency associated with intramuscular administration of ketorolac tromethamine. Ann Pharmacother 1993;27:1055–7. 3. Green D, Klement P, Liao P, Weitz J: Interaction of low molecular weight heparin with ketorolac. J Lab Clin Med 1996;127:583–7. 4. Hyers TM, Tricomi SM, Liao JJ: Inhibition of some human neutrophil functions by the cyclooxygenase inhibitor ketorolac tromethamine. J Leukoc Biol 1992;51:490 –5. 5. Peduto VA, Toscano A, D’Uva R, Piga M: Ketorolac for prevention of acute postoperative pain. Minerva Anestesiol 1995;61:367– 72. 6. Watcha MF, Ramirez-Ruiz M, White PF, Jones MB, Lagueruela RG, Terkonda RP: Perioperative effects of oral ketorolac and acetaminophen in children undergoing bilateral myringotomy. Can J Anaesth 1992;39:649 –54. 7. Grass JA, Sakima NT, Valley M, et al: Assessment of ketorolac as an adjuvant to fentanyl patient-controlled epidural analgesia after radical retropubic prostatectomy. Anesthesiology 1993;78: 642– 8. 8. Kinsella J, Moffat AC, Patrick JA, Prentice JW, McArdle CS, Kenny GN: Ketorolac trometamol for postoperative analgesia after orthopaedic surgery. Br J Anaesth 1992;69:19 –22. 9. Cataldo PA, Senagore AJ, Kilbride MJ: Ketorolac and patient controlled analgesia in the treatment of postoperative pain. Surg Gynecol Obstet 1993;176:435– 8. 10. Jung D, Mroszczak EJ, Bynum L: Pharmackinetics of ketorolac tromethamine in humans after intravenous, intramuscular and oral administration. Eur J Clin Pharmacol 1988;35:423–5. 11. Jallad NS, Garg DC, Martinez JJ, Mroszczak EJ, Weidler DJ: Pharmacokinetics of the single-dose oral and intramuscular ketorolac tromethamine in the young and the elderly. J Clin Pharmacol 1990;30:76 – 81. 12. Oie S, Levy G: Effect of sulfisoxazole on pharmacokinetics of free and plasma protein-bound bilirubin in experimental unconjugated hyperbilirubinemia. J Pharm Sci 1979;68:6 –9.