Introduction
An ideal analgesic should provide pain relief locally at the site of trauma without any systemic or local side effects, preferably as long as the pain persists. Scientists have searched for this ideal analgesic for several centuries, but as yet, no such drug exists. Local anesthetics (LAs) provide excellent analgesia with minimal toxicity when they are used in safe doses. In addition, they have anti-inflammatory and antithrombotic effects. However, LAs have a short duration of effect, which limits their usefulness to the immediate postoperative period unless injected intermittently or continuously into traumatized tissues. Therefore, the search for LA enclosed in microspheres or administered through catheters to prolong the duration of their effect continues. Opiates are efficacious, specifically following major surgery, but have several disadvantages, including nausea and vomiting, pruritus, constipation, and rarely, respiratory depression, when administered systemically. Following the discovery and isolation of opioid receptors on peripheral nerves and joints in the late 1980s, intensive investigation into the use of opiates peripherally to obtain pain relief without significant side effects ensued. Thus, analgesia with intra-articular (IA) morphine for the relief of postoperative pain following arthroscopy and arthroscopic knee surgery has been the focus of more than 60 publications involving humans during the last 20 years, when the first reports described the clinically beneficial effects of morphine. Recently, there has been increasing focus on the use of large-volume LA injections, with or without other adjuvants, into periarticular tissue during surgery to achieve sustained analgesia following knee and hip arthroplasty. Although this method has been reported in several publications, consensus on its applicability has not yet been attained.
Most authors have used the knee joint as a model for research. However, morphine and other drugs have also been injected into other joints, including the ankle, shoulder, and elbow, for assessment of clinical effects. Among the drugs and drug combinations that have been studied, the following are the most common: LAs, opioids, nonsteroidal anti-inflammatory drugs (NSAIDs), α 2 -adrenergic agonists, anticholinesterase drugs, steroids, and ketamine ( Table 57.1 ). Despite the relatively large number of publications on these issues, a well-designed study recruiting a sufficient number of patients and asking the relevant questions has yet to be performed. This review of the literature summarizes the findings of some of the studies published and focuses on the drawbacks, limitations, and problems that surround this complex but interesting question of the efficacy of IA analgesics in the clinical setting.
| Drug | Doses Used | Number of Studies Published | Number of Systematic Reviews Published |
|---|---|---|---|
| Local anesthetics | >20 | 1 | |
| Opioids | 4 | ||
| Morphine ∗ | 0.5, 1, 2, 3, 4, 5, 10 mg | >50 | |
| Pethidine | 10, 50, 100, 200 mg | 5 | |
| Fentanyl | 10, 50 µg | <5 | |
| Sufentanil | 5, 10 µg | <5 | |
| NSAIDs | 0 | ||
| Ketorolac | 5, 30, 60 mg | 5-10 | |
| Tenoxicam | 20 mg | 5-10 | |
| α 2 -Agonists | 0 | ||
| Clonidine | 150, 1 µg/kg | 5-10 | |
| Others | 0 | ||
| Neostigmine | 500 µg | <5 | |
| Ketamine | 0.5 mg/kg | ||
| Oxycodone | 5 mg | 1 | |
| Diamorphine | 5 mg | 1 |
∗ For studies on intra-articular morphine, the reader is referred to already published systematic reviews with extensive reference lists.
Variations in Pain Intensity and Methods of Measurement
Intensity of Postoperative Pain
Studying the efficacy of analgesics following procedures associated with only mild pain is likely to either be unsuccessful (i.e., not show analgesic efficacy) or require forbiddingly large numbers of patients to achieve statistical significance. In addition, even when statistical difference is seen, it may not be clinically meaningful since minor differences in pain intensity are questionable. Although a difference in pain intensity of 30% and higher is thought to be relevant, a 30% reduction in pain from 1.5 to 1.0 cm on a visual analog scale (VAS) is not necessarily of clinical importance. Therefore, when assessing the efficacy of drugs injected intra-articularly, it is likely to be more meaningful to study either patients or procedures associated with moderate to severe postoperative pain. In this way it is estimated that 20% to 30% fewer patients may need to be studied, as well as to provide meaningful assessment of drug efficacy. Because of the variation in pain intensity between operative procedures, international experts have suggested evidence-based guidelines for procedure-specific pain management, which are frequently updated with new evidence ( www.postoppain.org ).
Biologic Variations in Pain Intensity
Recent evidence suggests that women have more pain following arthroscopic surgery than men do, which is important when managing postoperative pain after arthroscopic procedures. In one study, Taenzer and colleagues studied 736 patients undergoing arthroscopic anterior cruciate ligament reconstruction (ACLR) and found that women experience greater pain intensity, which is associated with a worsened functional outcome. They concluded that these differences might result from variations in either response to analgesics or neuron processing. In another study, Rosseland and Stubhaug showed that gender might account for the differences in analgesia experienced by women after arthroscopic knee surgery, with women experiencing more pain than men do. Cepeda and Carr found that women have more intense pain and require 30% more morphine to achieve a similar degree of analgesia than do men. In addition to sex differences, variation in pain intensity during the menstrual cycle should also be considered in studies on analgesic efficacy since analgesic consumption is greater in women during the luteal phase of the menstrual cycle. There is also a clinically significant reduction in the intensity of pain perception or symptoms with increasing age, which could be related to the decrease in Aδ- and C-fiber nociceptive function, delay in central sensitization, or an increase in pain thresholds.
Assessment of Pain
Pain is a subjective sensation and varies between individuals. Several methods to measure pain have been described in the literature. Although the VAS and numeric rating scale (NRS) are well described and validated in the literature, other scales have also been used. When presenting the results, it is important to specify whether pain has been assessed at rest (static pain) or during provocation (dynamic pain) since mobilization often aggravates the intensity of pain. Dynamic pain can be measured as “pain on coughing,” “pain on movement,” or pain during “knee flexion” or “leg elevation” and should be described appropriately. Measuring pain intensity at fixed time intervals after the operation may sometimes give inaccurate results, depending on the last intake of analgesics, as well as the time of the day that the pain is being measured. Thus, both pain intensity and total analgesic consumption are important when drawing conclusions. Sometimes, the “area under the curve” for pain intensity may be more relevant than the pain intensity at fixed time points. Finally, in studies on postoperative pain and analgesic consumption, it may be equally important to measure pain relief as pain intensity. However, the former is more difficult to measure since VASs or NRSs may not correctly describe pain relief, as they do for pain intensity.
Placebo in Clinical Studies
In a randomized controlled trial, Rosseland and colleagues showed that pain after knee arthroscopy is modest and short-lived and can be treated successfully with IA saline (placebo). They also showed that the addition of 2 mg of morphine to 10 mL of saline does not reduce pain following arthroscopic knee surgery in patients with moderate to severe pain. IA saline may produce analgesia by cooling (cryoanalgesia) or by diluting IA algogenic substances. Indeed, this may represent a true therapeutic effect of saline attributable to the removal or dilution of pain-mediating substances (histamine, potassium, or vasoactive polypeptides) in the wound. Alford and Fadale also found that IA saline infusions provide similar pain relief as bupivacaine infusion following anterior cruciate ligament repair, thus suggesting that local pain mediators may be washed away by the infusion of saline. The placebo effect is well defined in the literature in studies on analgesics, and therefore it is important to include this group in randomized, double-blind trials, specifically when studying new analgesics or techniques. However, the use of placebo groups in pain trials is controversial, and this has been highlighted in the updated Helsinki declaration of 2001.
Variations in Pain Intensity and Methods of Measurement
Intensity of Postoperative Pain
Studying the efficacy of analgesics following procedures associated with only mild pain is likely to either be unsuccessful (i.e., not show analgesic efficacy) or require forbiddingly large numbers of patients to achieve statistical significance. In addition, even when statistical difference is seen, it may not be clinically meaningful since minor differences in pain intensity are questionable. Although a difference in pain intensity of 30% and higher is thought to be relevant, a 30% reduction in pain from 1.5 to 1.0 cm on a visual analog scale (VAS) is not necessarily of clinical importance. Therefore, when assessing the efficacy of drugs injected intra-articularly, it is likely to be more meaningful to study either patients or procedures associated with moderate to severe postoperative pain. In this way it is estimated that 20% to 30% fewer patients may need to be studied, as well as to provide meaningful assessment of drug efficacy. Because of the variation in pain intensity between operative procedures, international experts have suggested evidence-based guidelines for procedure-specific pain management, which are frequently updated with new evidence ( www.postoppain.org ).
Biologic Variations in Pain Intensity
Recent evidence suggests that women have more pain following arthroscopic surgery than men do, which is important when managing postoperative pain after arthroscopic procedures. In one study, Taenzer and colleagues studied 736 patients undergoing arthroscopic anterior cruciate ligament reconstruction (ACLR) and found that women experience greater pain intensity, which is associated with a worsened functional outcome. They concluded that these differences might result from variations in either response to analgesics or neuron processing. In another study, Rosseland and Stubhaug showed that gender might account for the differences in analgesia experienced by women after arthroscopic knee surgery, with women experiencing more pain than men do. Cepeda and Carr found that women have more intense pain and require 30% more morphine to achieve a similar degree of analgesia than do men. In addition to sex differences, variation in pain intensity during the menstrual cycle should also be considered in studies on analgesic efficacy since analgesic consumption is greater in women during the luteal phase of the menstrual cycle. There is also a clinically significant reduction in the intensity of pain perception or symptoms with increasing age, which could be related to the decrease in Aδ- and C-fiber nociceptive function, delay in central sensitization, or an increase in pain thresholds.
Assessment of Pain
Pain is a subjective sensation and varies between individuals. Several methods to measure pain have been described in the literature. Although the VAS and numeric rating scale (NRS) are well described and validated in the literature, other scales have also been used. When presenting the results, it is important to specify whether pain has been assessed at rest (static pain) or during provocation (dynamic pain) since mobilization often aggravates the intensity of pain. Dynamic pain can be measured as “pain on coughing,” “pain on movement,” or pain during “knee flexion” or “leg elevation” and should be described appropriately. Measuring pain intensity at fixed time intervals after the operation may sometimes give inaccurate results, depending on the last intake of analgesics, as well as the time of the day that the pain is being measured. Thus, both pain intensity and total analgesic consumption are important when drawing conclusions. Sometimes, the “area under the curve” for pain intensity may be more relevant than the pain intensity at fixed time points. Finally, in studies on postoperative pain and analgesic consumption, it may be equally important to measure pain relief as pain intensity. However, the former is more difficult to measure since VASs or NRSs may not correctly describe pain relief, as they do for pain intensity.
Placebo in Clinical Studies
In a randomized controlled trial, Rosseland and colleagues showed that pain after knee arthroscopy is modest and short-lived and can be treated successfully with IA saline (placebo). They also showed that the addition of 2 mg of morphine to 10 mL of saline does not reduce pain following arthroscopic knee surgery in patients with moderate to severe pain. IA saline may produce analgesia by cooling (cryoanalgesia) or by diluting IA algogenic substances. Indeed, this may represent a true therapeutic effect of saline attributable to the removal or dilution of pain-mediating substances (histamine, potassium, or vasoactive polypeptides) in the wound. Alford and Fadale also found that IA saline infusions provide similar pain relief as bupivacaine infusion following anterior cruciate ligament repair, thus suggesting that local pain mediators may be washed away by the infusion of saline. The placebo effect is well defined in the literature in studies on analgesics, and therefore it is important to include this group in randomized, double-blind trials, specifically when studying new analgesics or techniques. However, the use of placebo groups in pain trials is controversial, and this has been highlighted in the updated Helsinki declaration of 2001.
Drugs Used Intra-Articularly
Local Anesthetics
Single doses of LAs administered intra-articularly are used frequently. LAs have been used successfully as the sole anesthetic for minor arthroscopic procedures ( Fig. 57.1 ) and, when used in this way, provide adequate analgesia of short duration. In the only systematic review published in the literature on the efficacy of LA injected intra-articularly, Møiniche and associates evaluated randomized, double-blind controlled, controlled trials comparing LA with placebo or no treatment for the relief of postoperative pain following arthroscopic knee surgery. They found a significant prolongation of pain relief lasting between 30 and 50 minutes in only 2 of 6 studies; in addition, these authors also found that in 9 (of 20) studies, consumption of supplementary analgesics was reduced by 10% to 50% during observation periods of up to 4 hours ( Fig. 57.2 ). However, in most cases, the analgesic requirements were small to moderate, thus suggesting mild pain intensity in these studies. They concluded that the pain relief obtained with LA injected intra-articularly is mild to moderate and of short duration. This may, however, be of clinical significance in day-case surgery. In another recent systematic review and meta-analysis of the literature, the efficacy of LA injection via catheters was assessed following non–orthopedic-related surgery. The authors found very little benefit of this technique, except for short-lasting effects in patients undergoing obstetric surgery. LAs have been administered via catheters as intermittent injections or as continuous infusions into the knee joint intra-articularly, subacromially, intra-abdominally following hysterectomy, and subcutaneously following cesarean section, as well as during peripheral nerve blocks, with variable success.
Major Knee Surgery
A summary of the articles published and the conclusions drawn by the authors during knee surgery is presented in Table 57.2 . The anterior synovium, infrapatellar fat pad, and joint capsule are very sensitive to pain stimuli. Chew and coworkers tested the analgesic efficacy of 0.25% or 0.5% bupivacaine administered into the intrapatellar fat pad via a catheter after ACLR. A self-administered infusion pump (50 mL) allowed the patients to administer 4-mL doses of bupivacaine. The authors found no significant difference between 0.5% and 0.25% bupivacaine, thus suggesting absence of correlation between the dose of bupivacaine and pain intensity. It is likely that there is no direct relationship between the concentration and volume of LAs injected intra-articularly. Continuous catheter techniques were used for pain relief after knee replacement in three studies. In one study, DeWeese and associates compared continuous infusion of 0.5% bupivacaine at 2 mL/hr intra-articularly with a historical group in which patients were given controlled epidural analgesia with 0.125% bupivacaine plus fentanyl, 2 µg/mL. The IA infusion was less efficient, and higher analgesic consumption was registered during the 24-hour test period. In a double-blind study, 0.25% bupivacaine, 5 mL/hr, was compared with saline during a 48-hour IA infusion. The authors noted a significant reduction in opioid consumption, which resulted in less nausea, fatigue, and malaise and even enhanced rehabilitation and increased satisfaction. In another study, Hoenecke and colleagues found that patients undergoing knee surgery and receiving an LA infusion postoperatively experience less pain and require fewer doses of narcotic. They also found that the disposable pump allows administration of the medication on an outpatient basis. The site of LA injection (IA vs. intracapsular) was studied. In this study no differences were found between the groups, and the authors concluded that intracapsular LA has similar analgesic efficacy as IA LA after total knee arthroplasty. Warming of lidocaine appears to improve intraoperative anesthetic and postoperative analgesic conditions. Several published studies have focused on a new technique of high-volume LA (with or without adjuvants) infiltration periarticularly, often called local infiltration analgesia, for relief of postoperative pain. Although the majority of these studies appear to be promising, some have not been equally positive. It is possible that a combination of drugs rather than simply LAs injected during surgery will offer better analgesia without significant side effects. Further studies on short- and long-term outcomes following this technique are awaited.
| Drug/Concentration | Method of Drug Delivery | Type of Surgery | Type of Study | Effect (VAS or Analgesic Intake) | Reference |
|---|---|---|---|---|---|
| Bupivacaine, 0.25% and 0.5% | Intermittent injection via catheter | ACLR | Open | No difference between 0.25% and 0.5% | |
| Bupivacaine, 0.5% | Continuous infusion | TKR | Open | Epidural better than IA LA | |
| Bupivacaine, 0.25% | Continuous infusion | TKR | Double blind | Lower morphine requirement in IA LA group | |
| Bupivacaine, 0.25% | Continuous infusion | ACLR | Double blind | Lower VAS and analgesic needs | |
| Bupivacaine, 0.25% | Continuous infusion | TKR | Double blind | No difference between saline and LA |
In summary, IA injection of LA after total knee replacement, specifically, the recently used technique of local infiltration analgesia, offers an efficacious and inexpensive method for achieving good pain relief following total knee arthroplasty.
Total Hip Arthroplasty
Several studies have assessed the analgesic effects of LAs, with or without adjuvants, following IA and periarticular infiltration during surgery. A majority of these studies have shown better analgesia, a shorter duration of hospital stay, and improved mobilization, with some exceptions. Ropivacaine combined with ketorolac and adrenaline and infiltrated periarticularly reduced rescue analgesic consumption and decreased postoperative hospital stay when compared with epidural analgesia. Busch and colleagues found a reduction in pain during rest and on mobilization with LA infiltration and multimodal pain management in comparison to saline. In another study, no benefit of continuous infusion of 0.5% bupivacaine was noted when compared with saline for relief of pain after arthroplasty. Prolonging analgesia with postoperative infusion of drugs following hip arthroplasty does not seem to reduce pain intensity. Finally, adding LA infiltration to a multimodal pain management protocol involving the use of several analgesics does not add to the pain relief already achieved with these drugs. In summary, local infiltration analgesia using a combination of drugs is probably beneficial, but not the use of LAs alone. When using multimodal analgesia, adding LA infiltration does not add to the analgesia, and postoperative infusion of LA is not recommended for pain relief following total hip arthroplasty.
Shoulder Surgery
A summary of articles published on continuous LA infusion during shoulder surgery and the conclusions drawn by the authors is presented in Table 57.3 . To treat postoperative pain after acromioplasty, LA administered subacromially has recently been found to be effective. In seven studies, continuous infusion was used, and in four studies, patient-controlled regional analgesia (PCRA) was used. Savoie and colleagues found that 0.25% bupivacaine infused at a rate of 2 mL/hr decreases not only VAS scores during 48 hours of LA administration but also analgesic consumption during the first 5 postoperative days. Others reported significantly decreased VAS scores in comparison to saline when 0.5% bupivacaine was administered at an infusion rate of 2 mL/hr. In contrast to studies following knee surgery, the analgesic effect was dose dependent when 0.2% and 0.375% concentrations of ropivacaine were compared ( Fig. 57.3 ). In two other studies, however, continuous IA bupivacaine infusion resulted in only minor analgesic effects. The use of chilled compressive dressings and NSAID medication may have contributed to the small difference seen between the test drug and placebo.
| Drug/Concentration | Method of Drug Delivery | Type of Surgery | Type of Study | Effect (VAS or Analgesic Intake) | Reference |
|---|---|---|---|---|---|
| 0.25% bupivacaine, 2 mL/hr | Continuous infusion | Subacromial decompression | Double blind | Lower VAS scores and analgesic consumption | |
| 0.5% ropivacaine, 2 mL/hr | Continuous infusion | Unilateral shoulder arthroscopy | Double blind | Lower VAS scores at rest and with movement | |
| 0.2% and 0.375% ropivacaine | Continuous infusion | Cuff repair | Double blind | Lower VAS scores and analgesic consumption | |
| 0.5% bupivacaine, 2 mL/hr | Continuous infusion | Rotator cuff repair | Double blind | Lower VAS scores | |
| 0.25% bupivacaine, 2 mL/hr | Continuous infusion | Subacromial surgery | Double blind | Mild analgesic effect | |
| Ropivacaine, 6 mL/hr | Continuous infusion | Acromioplasty | Double blind | No benefit | |
| 0.2% ropivacaine | Continuous infusion and bolus | Rotator cuff repair | Open | Interscalene block better than IA infusion of LA | |
| 0.5% ropivacaine | Intermittent injection | Subacromial decompression | Double blind | Lower VAS score and analgesic consumption | |
| 0.125% bupivacaine, 2.5-10 mL/hr | Intermittent injection | Subacromial decompression | Open | Good analgesia | |
| 2% lidocaine, 2 mL/hr, plus bolus | Continuous infusion plus intermittent injection | Subacromial decompression | Open | Better pain relief than with placebo | |
| 0.2% ropivacaine plus bolus | Continuous infusion plus intermittent injection | Subacromial decompression | Double blind | Lower VAS score by 44% |
∗ Effect of IA LA on postoperative pain following shoulder surgery. The list of studies shown is not exhaustive.
Klein and coworkers evaluated postoperative pain relief when a long-acting preoperative interscalene block (0.5% ropivacaine) was compared with continuous IA infusion of 0.5% ropivacaine, 2 mL/hr for 24 hours. IA ropivacaine resulted in lower VAS scores both at rest and during movement. Delaunay and associates compared interscalene block with subacromial LA infusion following arthroscopic rotator cuff repair. A continuous interscalene block provides better analgesia than does continuous subacromial infusion, but with an increased incidence of minor side effects. In an open study, Mallon and Thomas described a patient-controlled catheter technique based on both continuous subacromial infusion of 2% lidocaine at 2 mL/hr and self-administration of 1-mL lidocaine boluses at 15-minute intervals as needed. During the 72-hour infusion, pain relief was significantly increased in the lidocaine relative to the placebo group. In another PCRA study, 0.2% ropivacaine infused at 5 mL/hr was combined with 2-mL boluses of 0.2% ropivacaine with a 15-minute lockout time. The authors reported a significant 44% decrease in VAS scores during 48 hours. We have tested the combined analgesic effect of LA administered both into the subacromial bursa and intra-articularly via a PCRA technique ( Fig. 57.4 ). Patients given LAs both in the bursa and subacromially had significantly lower VAS scores early postoperatively and lower morphine consumption than did those administered saline.
Pitfalls in Studies On Intra-Articular Local Anesthetics
Intermittent versus Continuous Techniques
Intermittent injection of drugs offers an advantage in that it prevents “overdosing” since pain is not continuous but exacerbated during certain maneuvers, such as movement, and can be either prevented or treated by the self-injection of fixed amounts of drugs intra-articularly through a catheter and an infusion pump. This technique, called PCRA and originally described by Rawal and colleagues, has now been used effectively in several studies for postoperative analgesia following shoulder surgery, hand surgery, laparoscopic cholecystectomy, and cesarean section. In contrast, continuous infusion of LA or other analgesic combinations has an advantage in that the patient may be pain free during an unplanned movement, but at the cost of drug overdose. This method also has the disadvantage that pain relief may be inadequate during maximal pain intensity. Perhaps a combination of low-dose infusion and intermittent injections as needed may be best.
Site of Injection of Local Anesthetic
Injections of LA at portal sites may reduce postoperative pain for short periods in a similar way as has been shown for IA injections. However, the anterior synovium, infrapatellar fat pad, and joint capsule are very sensitive to pain stimuli, and injections into these areas may provide better pain relief. Subacromial injections of LA have been found to provide good pain relief in comparison to placebo following shoulder surgery. No differences were seen in one study in which the authors injected LA intracapsularly versus extracapsularly. However, the study was small in size, and this issue has not been resolved to date.
Use of Adrenaline
Although adrenaline has been known to have analgesic efficacy for more than 100 years and recent evidence suggests that its pharmacodynamic effect is exerted via α 2 adrenoceptors in the substantia gelatinosa in the dorsal horn of the spinal cord, controlled studies on the use of adrenaline intra-articularly as an analgesic are lacking. Certainly, epidurally administered LA with fentanyl added to adrenaline provides better pain relief than do LA and fentanyl alone. Whether adrenaline has a preventive, pharmacologic (α 2 adrenoceptors), or physiologic effect (decrease in blood flow) remains uncertain. More studies on this important subject are warranted.
Volume and Dose of Local Anesthetic Injected Intra-articularly
Both the volume and the dose (milligrams) of LA injected intra-articularly may play a role in the analgesic efficacy of LA postoperatively. Small volumes of LA injected intra-articularly may theoretically leak out of the IA space, thus limiting its usefulness. It is difficult to fill the entire joint with LA because of the sensitive joint capsule. Similarly, small doses of LA may not have the desired effects. In one study, Gottschalk and coworkers found a reduction in pain intensity postoperatively at some time points in patients receiving 3.75 mg/mL versus 2 mg/mL ropivacaine (see Fig. 57.3 ). In a systemic analysis of the effect of IA LA on postoperative pain by Møiniche and colleagues, a mean dose of 90 mg bupivacaine was recommended with an injection volume of between 20 and 40 mL. Recent studies have found no relationship between the volume, concentration, or dose of LA and postoperative pain, except perhaps following shoulder surgery.
Morphine
The presence of peripheral morphine receptors has been noted. The first clinical study on the efficacy of morphine injected intra-articularly by Stein in 1991 found a reduction in pain intensity during the first 6 postoperative hours following minor arthroscopic surgery. This analgesic effect could be confirmed due to morphine via peripheral receptors since IA injection of naloxone reversed the analgesic effect. Since then, interest in the IA effects of morphine has grown by leaps and bounds, with more than 50 published studies assessing the efficacy of morphine injected intra-articularly into the knee joint. The results and conclusions have, however, been contradictory in these studies. In a systematic review of the literature in 1997, Kalso and coauthors concluded that “intra-articular morphine may have some effect in reducing postoperative pain intensity and consumption of analgesics.” Gupta and colleagues evaluated 45 studies in a meta-analysis of the literature and found “a definite but mild analgesic effect, which could be dose-dependent” ( Fig. 57.5 ). However, these authors could not completely exclude a systemic effect of IA morphine. Another systematic review by Kalso and coworkers found that 5 mg of morphine injected intra-articularly into the knee joint provides relief of postoperative pain for up to 24 hours. They also concluded that when there is “no pain, there is no gain,” thus suggesting that pain intensity must be at least moderate to detect any significant analgesic effects of morphine administered intra-articularly. Rosseland and colleagues confirmed that a postoperative analgesic effect of IA morphine is found only in a subgroup of patients with greater pain intensity in the immediate postoperative period. They also found that women perceive greater pain than men do, and therefore it is important to consider sex differences in studies on pain, thus adding another dimension to this complex problem. In one published systematic review, the authors showed that when only high-quality studies were considered in which pain intensity was moderate to severe, 5 mg IA morphine provided no significant analgesia postoperatively. This was in contrast to the findings of Kalso and associates presented earlier.
Pitfalls in Studies On Intra-Articular Morphine
Use of Systemic Opioids Intraoperatively
A substantial number of studies have used intraoperative analgesics such as fentanyl, which complicates the issue since analgesic efficacy (when seen) could be due to the preemptive or pharmacologic effect of these opiates in the early postoperative period. Studies in which only LAs are used intraoperatively without general anesthesia or in which general anesthesia is given without opioids may be more important to exclude this possibility. In two studies, the authors assessed postoperative pain following arthroscopy performed under LA without intraoperative opioids. In one study, Gupta and colleagues found no benefit of morphine, 3 mg intra-articularly, over saline. In another published study, Ng and coworkers found improved analgesia when morphine and ketorolac are combined with ropivacaine versus ropivacaine or bupivacaine alone. Whether this effect was due to ketorolac alone or its combination with morphine remains unclear.
Method of Injection of Drugs Intra-articularly
Variations in techniques of injection of drugs intra-articularly may account for some of the differences seen between studies. For instance, drugs have been injected via the arthroscope under direct vision or at the end of surgery intra-articularly via a needle. Injection of drugs through the arthroscope before it is removed may result in some of the drug either “running out” from the site of injection into tissue planes or, in the worst scenario, exiting through the incision sites. This depends naturally on the volume of the injectant and the meticulousness of the operator. A small volume of injectant that leaks out of the sore would not have any meaningful effect, whereas large volumes may not remain in the limited IA space. Similarly, a quick injection through the arthroscope and immediate removal may not ensure that the drugs remain where they are intended to be. Thus, the volume of drug remaining in the IA space may vary and account for some of these differences. Injections of drugs postoperatively through a catheter placed intra-articularly via an arthroscope may enable the entire volume of the drug to remain intra-articularly, but very few studies have used this simple but effective method. Finally, intermittent or continuous infusion of drugs via a catheter may prolong the analgesic effects and has been used recently in many studies. This technique, as opposed to a single injection of the drug via the arthroscope or a needle, needs to be further evaluated.
Systemic Effect of Intra-articular Morphine
It remains unclear whether the analgesic effects of IA morphine are due to its systemic absorption. Therefore, a similar dose of morphine was injected by the intramuscular (IM), intravenous (IV), and IA routes in several studies. The results were equivocal, with some studies documenting an equianalgesic effect and others documenting better analgesia with IA morphine. No clear relationship appears to exist between the dose of morphine and analgesia when comparing IA with IM/IV morphine. Plasma concentrations of morphine were measured after IA injection of 1 and 5 mg morphine in one study and after IV and IA injection (5 mg) in another study. In the first study, two patients had spuriously large concentrations and two others had undetectable levels (<1 ng/mL). In the remaining 6 of 10 patients, the authors found lower concentrations of morphine than usually described after parenteral morphine administration. In the second study, the plasma concentration measured 2 hours after IA morphine administration was approximately 50% of the concentration achieved after the same dose administered by the IV route. In a third study, although the maximum plasma concentration following IA injection was lower than after IV injection, the area under the curve during hours 0 to 6 was similar, thus suggesting that substantial amounts of morphine are absorbed into the systemic circulation over time. In conclusion, whether there is a systemic effect of morphine following IA injection remains unclear.
Dose-Response Effect
The effects of increasing IA doses of morphine on postoperative analgesia have also been studied. Better postoperative analgesia was obtained with 5 mg than with lower doses. Denti and coworkers found that 2 mg of morphine is adequate for minor arthroscopic procedures but 5 mg is necessary for anterior cruciate ligament surgery, thus suggesting that the dose of IA morphine is dependent on pain intensity. Although low doses of IA morphine (<1 mg) should theoretically produce a high concentration of IA morphine, a dose-response effect was demonstrated by Likar and associates with IA doses of 1 to 4 mg. Kalso and colleagues also demonstrated in their systematic review of the literature that all studies in which 5 mg of morphine was injected intra-articularly showed a positive result at all time periods. Thus, clinical data seem to suggest that larger IA doses of morphine result in better pain relief.
Use of a Tourniquet
Use of a tourniquet on the thigh has been proposed to reduce pain intensity and may account for the differences in results. The mechanism for this effect is unclear, and therefore the role of the tourniquet remains controversial. Whitford and coauthors suggested that retaining the tourniquet for 10 minutes after the injection of morphine postoperatively provides better pain relief than if it is released immediately, a finding that has not been confirmed by others. Many surgeons do not use a tourniquet while performing arthroscopic procedures on the knee, and some joints are not accessible to a tourniquet, so this method may not always be applicable.
Role of Inflammation
Marchal and colleagues classified procedures a priori into low inflammatory (diagnostic arthroscopy, partial meniscectomy) or high inflammatory (synovial plicae removal, patellar shaving, anterior cruciate ligament repair) and found that operative procedures associated with low-inflammatory states respond best to bupivacaine whereas those associated with high-inflammatory states respond better to IA morphine, thus supporting the theory that inflammation is a prerequisite for the peripheral analgesic effect of opioids ( Fig. 57.6 ). Others have also suggested that upgrading of morphine receptors during states of inflammation may account for its improved efficacy, a phenomenon that has been well documented in a rat model. However, this was not found in one study in which the authors concluded that the dose-response relationship of IA morphine analgesia is not shifted by enhanced inflammation in synovial tissue. Further clinical studies in this area are urgently needed.
Pethidine
In addition to being an opioid, pethidine is unique in having LA effects and has been used as the sole anesthetic during spinal anesthesia and analgesia. Therefore, the analgesia following IA injection of pethidine may have an effect as an LA, as well as via peripheral morphine receptors. IA administration of 5% pethidine with adrenaline was found to be comparable to 5% lidocaine with adrenaline during arthroscopy of the ankle. However, lower pain scores at rest were found postoperatively than with IA prilocaine. In another study, 1 mg morphine, 10 mg pethidine, or 10 µg fentanyl resulted in similar pain intensity and analgesic consumption regardless of whether the drugs were administered intra-articularly or systemically. The authors concluded that the analgesia seen following IA pethidine may be identical to that following its systemic administration. Lyons and coworkers compared the effects of morphine, pethidine, and placebo and concluded that the LA effect of pethidine may be responsible for the improved early analgesia but that its duration of action was shorter than that of morphine. Finally, Ekblom and colleagues found that pethidine, 200 mg intra-articularly, provides the best analgesia and that the effect is not potentiated by the addition of adrenaline and prilocaine. However, in the dose range of 50 to 200 mg of pethidine intra-articularly, analgesia is due to both peripheral and central mechanisms. A summary of the studies on IA administration of pethidine is presented in Table 57.4 .
