| ||||||||||||||
|
|
|||||||||||||


*Department of Anesthesia, University of California, San Francisco, California; and
Omeros Medical Systems, Inc., Seattle, Washington
Address correspondence and reprint requests to Dr. Pamela Pierce Palmer, Department of Anesthesia, S-455, University of California, San Francisco, 513 Parnassus Ave., San Francisco, CA 94143-0464.
| Abstract |
|---|
|
|
|---|
Implications: Preemptive, intraarticular irrigation of a combination of multiple antiinflammatory drugs is a novel and potentially effective method for reducing the synovial inflammatory response, such as that during arthroscopy. In this study, a three-drug combination infusion was statistically superior to one- or two-drug infusions in a rat model.
| Introduction |
|---|
|
|
|---|
Selecting a combination of drugs that target known inflammatory mediators is crucial to inhibit the myriad inflammatory pathways known to exist. The antiinflammatory and antihyperalgesic effects of cyclooxygenase (COX) inhibitors and lipoxygenase inhibitors are attributed to the inhibition of prostaglandins, prostacyclins, and leukotrienes, all of which produce inflammation and sensitize nociceptor terminals in the periphery (2,3). For example, the inflammatory and hyperalgesic effects of bradykinin are mediated by prostaglandins (1).
Antagonists to specific 5-HT receptors decrease pain and inflammation in animals. Serotonin is thought to produce pain by stimulating 5-HT receptors on nociceptive neurons in the periphery. 5-HT2A and 5-HT3 receptors both activate peripheral nociceptors, producing hyperalgesia and neurogenic inflammation (47). In addition to 5-HT receptors mediating pain and hyperalgesia, activation of a subset of inhibitory 5-HT receptors (the 5-HT1B and 5-HT1D receptors) decreases pain and inflammation. Drugs that are clinically used for migraine headache pain display high affinity for the 5-HT1B and 5-HT1D receptors (8). These drugs inhibit trigeminal C-fiber-induced pain and neurogenic inflammation in dural vessels, presumably by activating 5-HT1B and 5-HT1D receptor subtypes on trigeminal neuron terminals (9,10). In the periphery, a 5-HT1B and 5-HT1D agonist inhibits neurogenic inflammation of sciatic epineurium in rats (11). Therefore, these drugs possess both antinociceptive and antiinflammatory properties.
Histamine receptors, which are generally divided into H1, H2, and H3 subtypes, are also involved in pain and inflammation pathways. The classic inflammatory response to the peripheral administration of histamine is mediated via the H1 receptor and is inhibited by selective H1 antagonists in many species (12). In addition to the involvement of H1 receptors in inflammation, evidence from the H1 knock-out mouse supports a direct role for H1 receptors in nociception (13).
In this study, plasma extravasation in a rat knee joint is used as a measure of inflammation, and an inflammatory soup consisting of bradykinin, 5-HT, and mustard oil perfused into the knee joint simulated the inflammation produced by surgical tissue injury. At sites of tissue injury, the nine-amino-acid peptide bradykinin is cleaved from a larger circulating protein, and 5-HT is locally released in large amounts from platelets, producing inflammation and activating peripheral C-fiber terminals (1,3,12). Mustard oil produces inflammation and chemically excites C fibers; it is often used to simulate the actions of endogenous mediators in laboratory studies (14).
In an attempt to inhibit the inflammatory soup-induced synovial inflammation, three drugs were chosen with the following rationale. Ketoprofen was selected for inhibition of the COX enzymes (both COX-1 and COX-2), as well as lipoxygenase enzymes, thereby inhibiting prostaglandin, prostacyclin, and leukotriene synthesis (1517). Amitriptyline, in nanomolar concentrations, is a 5-HT2A receptor antagonist (18) as well as an H1 receptor antagonist (19). Oxymetazoline potently activates 5-HT1B and 5-HT1D receptors, and this results in inhibition of neurogenic inflammation (20,21). This study attempts to determine whether the intraarticular irrigation of one or any combination of these antiinflammatory/antihyperalgesic drugs is an effective way to inhibit synovial inflammation.
| Methods |
|---|
|
|
|---|
Rats were anesthetized with sodium pentobarbital (65 mg/kg intraperitoneally; Abbott, Chicago, CA). The animals then received a tail vein injection of Evans blue dye (50 mg/kg in a concentration of 20 mg/mL; Sigma, St. Louis, MO), which was used as a marker for plasma protein extravasation. The knee joint capsule was exposed by excising the overlying skin, and a 30-gauge needle, which was used for the infusion of fluid, was inserted into the joint. After perfusion of 100200 µL of fluid, a 25-gauge outflow needle was also placed into the joint space approximately 3 mm from the inflow needle to extract fluid. The infusion and extraction rate (200 µL/min) was controlled by an SP120p push-pull syringe pump (WPI, Sarasota, FL). Perfusate samples were collected over 5-min intervals for 50 min. The samples were immediately centrifuged to determine whether red blood cells were present; only blood-free samples were acceptable. Samples were then analyzed for Evans blue dye concentration by spectrophotometric measurement of absorbance at 620 nm (Spectronic 21D; Spectronic Instruments, Inc., Rochester, NY). Absorbance is linearly related to dye concentration (22).
All drugs and chemicals, with the exception of mustard oil, are from Sigma, St. Louis, MO. Plasma extravasation was activated by an inflammatory soup consisting of 1 µM 5-HT, 200 nM bradykinin, and 1% mustard oil (allyl isothiocyanate; Aldrich Chemical, Milwaukee, WI). These concentrations were chosen on the basis of previous synovial plasma extravasation studies with these compounds (14,23,24). Bradykinin and 5-HT were dissolved in saline, and mustard oil was dissolved first in 40% ethanol and 20% Tween 80, with a final concentration of ethanol 1% and Tween 80 0.5%. Amitriptyline and oxymetazoline were dissolved in saline, and ketoprofen was dissolved first in 40% ethanol and then diluted to a final ethanol concentration of <1%.
To determine the optimal antiinflammatory drug concentrations, each of the three drugsamitriptyline, oxymetazoline, and ketoprofenwas tested for its ability to dose-dependently inhibit plasma extravasation produced by a single inflammatory mediator (5-HT, mustard oil, and bradykinin, respectively) against which the drug is targeted. An initial 5-min intraarticular baseline perfusion with 0.9% saline was followed by a 10-min perfusion with the drug, followed by a 35-min perfusion with the drug and the inflammatory mediator.
In studies that tested the ability of the inflammatory soup to stimulate plasma extravasation, the inflammatory soup was perfused for 35 min, beginning immediately after a 15-min baseline saline perfusion. In experiments testing preemptive administration of the single drugs against the inflammatory soup, amitriptyline, oxymetazoline, or ketoprofen was added to the saline perfusion after 5 min for a period of 10 min, then the drugs were perfused together with the inflammatory soup for an additional 35 min. In addition to perfusing the individual drugs before, and then together with, the inflammatory soup, two-drug and three-drug combinations were also tested for their ability to inhibit inflammatory soup-induced plasma extravasation. Each of the four possible combinations (amitriptyline + ketoprofen, amitriptyline + oxymetazoline, ketoprofen + oxymetazoline, and amitriptyline + ketoprofen + oxymetazoline) were perfused 10 min before, and then together with, the inflammatory soup solution, similar to the single-drug studies. In studies testing the postinflammatory administration of drugs, the inflammatory soup alone was perfused for 10 min after the baseline 15-min saline perfusion, followed by all three antiinflammatory drugs together with the inflammatory soup for an additional 25 min.
A total of 68 rat knees were excluded from the study because of physical damage of the knee joint or inflow and outflow mismatch (detectable by presence of blood in perfusate and high baseline plasma extravasation levels or knee joint swelling caused by improper needle placement). After the procedure, rats were killed by a lethal injection of pentobarbital and bilateral thoracotomy.
Data are expressed as mean ± SEM for the area under the curve (AUC). The AUC is a measure of the cumulative inflammation produced in the knee joint and therefore was chosen for statistical comparison among groups. Statistical evaluations were performed with modified t-statistics corrected for multiple comparisons by the Bonferroni method.
| Results |
|---|
|
|
|---|
Similarly, oxymetazoline 20 nM (n = 6) has no significant effect on mustard oil-induced plasma extravasation (1%; n = 6) (1.555 ± 0.266 vs 1.139 ± 0.123; P > 0.05) (data not shown). Oxymetazoline 200 nM (n = 6) produces a 50% reduction of plasma extravasation (0.712 ± 0.156; P = 0.01). Oxymetazoline 2000 nM (n = 6) does not result in any further significant decrease in plasma extravasation compared with the 200 nM concentration (0.683 ± 0.056; P > 0.05, 2000 vs 200 nM). The inhibitory activity of ketoprofen was tested against bradykinin (200 nM)-induced (n = 6) plasma extravasation at larger concentrations (75 to 7500 nM), given the extensive protein binding of ketoprofen in synovial fluid (25). Ketoprofen 75 nM (n = 6) has no significant effect on bradykinin-induced plasma extravasation (1.070 ± 0.063 vs 0.981 ± 0.094; P > 0.05), yet ketoprofen 750 nM (n = 6) reduces extravasation by 40% (0.673 ± 0.079; P < 0.005). Ketoprofen 7500 nM (n = 6) does not produce any additional inhibition compared with the 750 nM concentration (0.602 ± 0.033; P > 0.05, 7500 nM vs 750 nM) (data not shown). These dose-response data suggest that amitriptyline 200 nM, oxymetazoline 200 nM, and ketoprofen 750 nM are concentrations that are maximally effective in inhibiting plasma extravasation while avoiding unnecessarily large concentrations that may produce undesirable side effects. These concentrations will be used in the following experiments.
Intraarticular perfusion of the inflammatory soup produces a marked increase in plasma extravasation in the rat knee joint after 10 min and reaches a plateau by 15 min that continues for the remainder of the perfusion time (a total of 35 min) ( Fig. 1). Use of the absorbance measurement at each time point results in a calculated AUC of 4140 ± 420 for the inflammatory soup alone. The perfusion of a single drugamitriptyline 200 nM (3430 ± 140), oxymetazoline 200 nM (3330 ± 170), or ketoprofen 750 nM (3080 ± 260)10 min before, and then in combination with, the inflammatory soup does not significantly inhibit plasma extravasation.
|
|
|
| Discussion |
|---|
|
|
|---|
Only physiologic saline solutions are used as irri-gants during arthroscopy; therefore, peripheral inflammation and sensitization is only treated intraoperatively by the anesthesiologists use of IV COX inhibitors, opioids, or both. However, these drugs have systemic side effects that are well known, and, as single drugs, they can inhibit only part of the inflammatory cascade, as previously mentioned. Often only an IV benzodiazepine combined with an epidural or spinal anesthetic is used for arthroscopic cases; none of these has any inhibitory effect on peripheral inflammation or peripheral sensitization.
A combination of antiinflammatory drugs delivered locally throughout the duration of the inflammatory process appears to be the optimal method for inhibiting the effects of peripherally acting inflammatory mediators, as demonstrated in this study. The three-drug combination is clearly the most effective at inhibiting plasma extravasation. The single-drug solutions and the two-drug solutions do not significantly inhibit inflammatory soup-induced plasma extravasation compared with control (saline). Furthermore, only the three-drug solution delivered before the inflammatory soup was effective in inhibiting plasma extravasation. Waiting as little as 10 minutes after the inflammatory insult to the synovium to begin perfusion with the antiinflammatory drugs is already too late to inhibit the effects of the inflammatory mediators. Most of these mediators rapidly activate receptors coupled to second-messenger systems (e.g., mobilization of intracellular calcium, increasing phospholipid metabolism, activation of protein kinase pathways, etc.). Once the receptors have been activated, the intracellular events resulting in inflammation and peripheral sensitization begin. Inhibiting the receptor or enzyme (in the case of COX inhibitors) after the second-messenger systems have been activated is dramatically less effective, as shown in Figure 3. As an example, the currently used technique of postoperative injection of intraarticular opioids is performed well after the inflammatory pathways have been activated. This may be the reason for the debated efficacy of postoperative intraarticular morphine in the literature (2629). Interestingly, intraarticular morphine administered before arthroscopy appears to be more effective than after arthroscopy (30), which supports the importance of preemptive inhibition of peripheral inflammatory and hyperalgesic pathways.
One of the advantages of the specific antiinflammatory drugs chosen in this study is that each individual drug inhibits the actions of multiple proinflammatory mediators; therefore, the combination of the three drugs inhibits a significant portion of the inflammatory cascade. For example, amitriptyline is a 5-HT2A receptor antagonist, a receptor that is involved in inflammation and hyperalgesia (6,7,24). Amitriptyline is also a histamine H1 receptor antagonist and an N-methyl-D-aspartate receptor antagonist (19,31,32); both receptors are located in the periphery and are involved in inflammation and hyperalgesia (12,14,33). Ketoprofen is both a COX and lipoxygenase inhibitor, thereby decreasing the formation of multiple inflammatory and hyperalgesic mediators, such as prostaglandins, prostacyclins, and leukotrienes (15,17). Oxymetazoline is an agonist at the inhibitory 5-HT receptor subtypes, 5-HT1B and 5-HT1D receptors, which are located on nociceptors and inhibit neurogenic inflammation produced by a variety of mediators (9,11,34). Furthermore, oxymetazoline is an agonist at
1-adrenergic receptors, causing local vasoconstriction (35). This activity may result in reduced bleeding, improving the surgeons visibility, and it may also reduce the local hyperemic inflammatory response.
Recently a prospective, randomized, placebo-controlled human clinical pilot study was performed by using a combination of drugs perfused intraarticularly during arthroscopy in 32 patients (36). The drugs (amitriptyline, sumatriptan, and metoclopramide) were selected to target inflammatory/hyperalgesia pathways similar to those targeted by the three-drug combination previously discussed. Sumatriptan is a 5-HT1B and 5-HT1D receptor agonist (8) similar to oxymetazoline, and metoclopramide is a 5-HT3 receptor antagonist (37). Postoperative visual analog scale scores and fentanyl use in the recovery room were assessed and were significantly lower in patients who received the intraarticular three-drug irrigation compared with those who received intraarticular saline irrigation. In this animal study, the effect of drug irrigation on synovial inflammation was assessed in contrast to the human trial in which pain-related variables (visual analog scale and fentanyl use) were used as the outcome measurements. Therefore, intraarticular irrigation with a combination of drugs may be effective at inhibiting both inflammation and pain/hyperalgesia, which is not surprising because the targeted mediators activate these multiple processes.
In summary, inflammation is caused by the local release of multiple mediators, and the effects of these mediators can be inhibited if a preemptive approach targeting multiple inflammatory pathways is used. Surgical procedures (such as most endoscopic procedures) that require irrigation solutions can possibly take advantage of this method of inhibiting intraoperative and postoperative inflammation. Phase I/II clinical trials are continuing to determine the clinical efficacy of this arthroscopic irrigation solution.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J.-W. Shin, K.-S. Hwang, Y.-K. Kim, J.-G. Leem, and C. Lee Nonsteroidal Antiinflammatory Drugs Suppress Pain-Related Behaviors, but Not Referred Hyperalgesia of Visceral Pain in Mice Anesth. Analg., January 1, 2006; 102(1): 195 - 200. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|