Introduction

Periarticular infiltration is an effective adjunct to multimodal pain management following a joint replacement. Multiple drugs have been proposed to use in the periarticular cocktail. Unfortunately, there is no consensus on which medications to use. This review article summarizes the most recent evidence available for the most commonly used medicines in periarticular blocks. In addition, there are multiple sites proposed for injection. Ross et al. have summarized the best potential locations for periarticular injection based on nociceptor prevalence [1]. The highest concentrations of nociceptors in the knee were found in the medial and lateral retinacula, infrapatellar fat pad, pes anserine bursa, tibial, femoral, and patellar periosteum, and the bony insertions of the MCL, LCL and IT band [1]. In the hip articulation, there is limited research, but higher concentrations of nociceptors have been found in the labral base, ligamentum teres and the hip capsule [1]. No data exist regarding nociceptor concentration for tendon insertions around the hip joint.

We have conducted a literature search on PubMed and Embase. Our search criteria included randomized controlled trials (RCTs) and systematic reviews (SR). We tried only to include the most recent studies to keep the information current. We included only studies published in the English language. We aimed to have at least ten studies available for each drug. If insufficient RCTs and systematic reviews (SRs) were available, we broadened the search to include other specialties that also utilize local infiltration following surgery. Our search terms included: dexmedetomidine; Presedex, liposomal ketorolac, ketorolac, Toradol, tranexamic acid, morphine, glucocorticosteroids, triamcinolone acetate, methylprednisolone, betamethasone, ropivacaine, epinephrine, total knee arthroplasty, total knee replacement, periarticular infiltration, local anesthetic, peripheral block in combination with Boolean operators.

Dexmedetomidine (Presedex)

Using the search criteria, we identified 12 randomized controlled trials evaluating dexmedetomidine (Table 1) [2,3,4,5,6,7,8,9,10,11,12,13]. Dexmedetomidine (DM) is a highly selective alpha-2 adrenoceptor agonist with a half-life of 2 h [2]. It binds an alpha-2 receptor eight times more than clonidine [3]. Though multiple theories have been proposed, the exact mechanism of action is not completely understood. The first prevalent theory suggests that DM activates the alpha-2 adrenoreceptors on the peripheral smooth muscle cells, leading to vasoconstriction with subsequent delayed absorption of the local anesthetic [2]. Secondly, it was proposed that DM can block the activity-dependent cation current [2]. The activation of the Na–K pump causes hyperpolarization of the membrane in the peripheral fibers. DM could enhance hyperpolarization by blocking the Na–K currents, thus inhibiting the action potentials [7].

Table 1 Articles included in the review analyzing the use of dexmedetomidine in periarticular infiltrations

Because of the sedating effects, anesthesiologists routinely use intravenous (IV) DM to help sedate patients during a surgical procedure. If DM is used in the periarticular block, the sedating effects can improve sleep at night, a common problem after hip or knee replacements. Additional benefits include: reduced opioid usage, prolonged neuraxial analgesia, decreased postoperative delirium and reduced postoperative nausea [14]. In addition to the sedative effects described above, it also has anxiolytic, analgesic, anesthetic-sparing and sympatholytic effects [3].

Yang et al. [14] demonstrated a dose-dependent response using IV DM and postoperative hypotension and bradycardia. They suggested using the lowest possible dose (< 50ug) to minimize hemodynamic side effects. Even though there was a statistical difference in hemodynamic side effects between patients receiving DM compared to patients not receiving DM, they did not find a statistical difference comparing side effects between different doses of DM (< 50; 50–99; > 100ug). It is important to note that conclusions cannot be extrapolated from IV administration of DM to periarticular infiltration. Fritsch et al. [15] found that blood concentration was extremely low at 180 min after surgery. In this study, 150ug DM was added to the interscalene brachial plexus block [15].

The dosing of DM is not clearly defined, with different dose recommendations per route administration. Typically, IV dosing should be as minimal as possible (< 50ug), while intramuscular (2.5ug/kg) or periarticular dosing (1–2 ug/kg) can be more generous due to the slower absorption and low blood concentrations [2, 15, 16]. In the RCTs in Table 1, the dosages varied between 0.5 and 2 ug/kg. There was no correlation between dosage and duration of analgesia. DM showed a benefit in combination with longer-acting anesthetic drugs, such as ropivacaine and levobupivacaine, demonstrating longer-lasting pain relief [12].

There were minimal complications observed with the addition of DM. Only two studies from the twelve RCTs listed in Table 1 had observed side effects (bradycardia, hypotension and sedation) [6, 8]. Therefore, with careful screening, DM can be a valuable addition to the periarticular block. The most commonly used exclusion criteria in the studies in Table 1 include allergies to the medications [2, 4, 5, 7,8,9,10,11,12,13], diagnosis of severe cardiovascular disease [2, 5,6,7,8,9,10,11,12], diagnosis of hepatic or renal disease [2, 4], psychiatric disorders [4, 9,10,11, 13], pregnancy or lactation [5, 6, 10, 12], and history of neuromuscular disorders [2, 8,9,10]. Care should be exercised in these groups of patients. It is advisable to discuss patients with these conditions with your anesthesiologist and consider omitting or using smaller doses of DM (e.g., 0.5ug/kg).

Epinephrine

Epinephrine is commonly used in periarticular injections. It is a sympathomimetic catecholamine that affects alpha- and beta-adrenergic receptors. Its effect on Alpha-1 receptors produces increased vascular smooth muscle contraction. It is thought by this mechanism to create a synergistic effect when used with local anesthetics [17]. Epinephrine is believed to reduce the peripheral blood flow, prolonging the duration of local anesthetics and increasing the maximal dose that can be used without fear of systemic toxicity [17]. Its effect on the duration of local anesthetic is controversial, as it has not been well demonstrated. It is also thought to cause potentially harmful effects on wound healing through the local vasoconstriction at the level of the skin when blood flow is poor [17].

A literature review concerning the use of epinephrine in periarticular blocks does not reveal a clear consensus on its utility (Table 2) [18,19,20,21,22,23,24]. Some studies have reported statistically decreased visual analog pain scores when epinephrine is used in the periarticular block [19, 22], but this did not reach clinical significance. Contrarily, Kong et al. [18] showed no significant differences in pain scores postoperatively nor on opioid use after arthroplasty. There is some evidence that epinephrine can reduce postoperative bleeding when used in the periarticular block without increasing deep venous thrombosis (DVT) risk [21, 23]. However, it does not influence intraoperative blood loss, postoperative hemoglobin levels, or postoperative transfusion rates [21, 22].

Table 2 Articles included in the review analyzing the use of epinephrine in periarticular infiltrations

Dosing of epinephrine as part of a periarticular cocktail does not have clear recommendations for administration. In the reviewed studies, epinephrine was typically diluted at concentrations of 1:1000 for total dosages ranging between 25 and 60 mcg. No studies compared different dosages of epinephrine.

No complications were recorded in any of the reviewed studies regarding the use of epinephrine in periarticular infiltrations. Specifically, no wound-healing problems related to using epinephrine in the periarticular cocktail were recorded.

Therefore, the benefits and recommended use of epinephrine as part of a periarticular block still need to be determined.

Glucocorticosteroids

Triamcinolone acetonide is a synthetic corticosteroid with a half-life of 18–36 h [25]. The steroids' acetate form is insoluble and postulated to remain in the tissues to give anti-inflammatory properties. This can be potentiated by the vasoconstricting effects of epinephrine [25, 26]. Methylprednisolone is a synthetic glucocorticosteroid with a half-life of 1.8–2.6 h [27, 28]. Betamethasone and dexamethasone have half-lives of 36–54 h [29,30,31]. All the corticosteroids examined in this review paper are primarily metabolized in the liver and excreted by the kidneys.

Glucocorticosteroids (GCSs) are thought to reduce the stress response, decrease edema and blood loss, prevent nausea and vomiting and achieve a better permanent range of motion by reducing the inflammatory response [32]. Pain is induced by inflammation. Inflammatory markers (IL-1Beta, IL-6, TGF-alpha) are released after surgery, leading to a decrease in the nociceptor threshold with subsequent occurrence of pain. GCS inhibits phospholipase A-2, which reduces the proinflammatory derivatives of arachidonic acid, decreasing inflammation [33, 34]. Interestingly, only one study [26] evaluated the inflammatory markers in patients receiving GCS and found a reduction in the CRP values until postoperative day four compared to the control group.

GCS also leads to decreased production of prostaglandins with their vasodilatory effects and a subsequent diminution in blood loss [33, 34]. Again, only two studies from the RCTs listed evaluated blood loss in patients receiving GCS as part of the PAI and found a non-statistical decrease in blood loss in the GCS group [26, 33].

There is, however, hesitancy in administering GCS routinely after joint replacements because of a fear of poor wound healing, postoperative infections or ligament/tendon ruptures. When evaluating the injection sites in the PAI, most studies injected GCS as part of the cocktail into the MCL [25,26,27,28, 30, 31, 35, 36] and LCL [26,27,28, 30, 31, 35, 36]. Four studies injected GCS into the patella tendon and fat pad [27, 28, 31, 35], while others did not inject GCS into the fat pad or patella tendon in fear of late rupture [25, 32, 36]. Two studies did not inject GCS in the subcutaneous skin to avoid steroid-induced skin atrophy [31, 33].

Most studies excluded patients with a history of renal insufficiency, uncontrolled diabetes mellitus, local infection, history of cardiac arrhythmias or prolonged QT intervals, immunosuppression (e.g., inflammatory conditions), history of congestive heart failure, psychiatric illnesses or history of gastrointestinal bleeding [25,26,27,28, 30, 31, 33, 35, 36].

No clear conclusions can be derived from the results of the randomized controlled trials in Table 3 [25,26,27,28,29,30,31,32,33, 35, 36]. Multiple studies demonstrated lower VAS scores in the immediate postoperative period. Most studies showed an effect in the 12–24-h postoperative period [26,27,28, 30, 32, 33, 35], while longer duration up to 48 h is supported only by several studies [27, 28, 33]. 72 h or more of lower VAS scores were only observed in a few studies [28, 33]. Even though some studies did not report any difference in VAS scores with the addition of corticosteroids [25, 31, 36], it does seem plausible to assume GCS can improve pain in the early postoperative period.

Table 3 Articles included in the review analyzing the use of glucocorticosteroids in periarticular infiltrations

The increase in range of motion with the addition of GCS still needs to be clarified with an equal number of studies demonstrating benefits [28,29,30, 32, 33] compared to studies showing no apparent benefit [25,26,27, 31, 36].

Most studies demonstrated a decrease in length of stay in the cohort of patients receiving GCS in their PAI [29, 33, 36] compared to only one study, which did not show any benefit [31].

GCS added to the PAI is a safe option, with most studies not demonstrating an increase in adverse events [26, 28, 30,31,32,33, 35]. Two studies did show complications after GCS administration. The first study had one periprosthetic joint infection (PJI) with triamcinolone acetonide [25], and the second study demonstrated one PJI and two manipulations under anesthesia with methylprednisolone [30].

Ketorolac (Toradol)

We identified 12 articles exploring the effectiveness of ketorolac (Toradol; ketorolac tromethamine) as an element of periarticular injection (PAI) for total knee arthroplasty (TKA) (Table 4) [3, 37,38,39,40,41,42,43,44,45,46,47]. Ketorolac is a non-steroidal anti-inflammatory drug (NSAID) that has been an additive agent to various high-volume local infiltration analgesia cocktails, often including additional medications such as ropivacaine and epinephrine [48]. However, the efficacy of the individual elements of such regimens has limited evidence [37]. Ketorolac is a non-specific cyclooxygenase (COX) inhibitor, limiting the conversion of arachidonic acid to thromboxane, prostacyclin, and prostaglandins, thus decreasing sensitization of afferent nerves [49]. There is also evidence of an impact on the central hypothalamic prostaglandin response by inhibiting prostaglandin synthetase systems [50]. It was first approved for parenteral use in 1990 and thus has been available as an anti-inflammatory for over three decades. Ketorolac has a rapid onset of action following IM and IV administration, with peak analgesic effects at 75–150 min [49]. The half-life of ketorolac is 5–6 h [49]. Adverse effects of ketorolac are similar to those of other NSAIDs, including gastrointestinal bleeding, nausea/vomiting, peptic ulceration, renal failure and increased bleeding due to inhibited platelet function [49], which may limit its effectiveness in many patients receiving total knee arthroplasty.

Table 4 Articles included in the review analyzing the use of ketorolac in periarticular infiltrations

Ketorolac's recommended IM and IV dose is 30 mg as a one-time dose or 30 mg every 6 h up to a maximum total of 120 mg in 24 h [51]. Still, some studies also describe 60 mg as a dosage option[39, 40] with a possible dose-dependent response [49, 52]. Many studies included visual analog scale (VAS) scores and opioid usage as measures of effectiveness for postoperative pain management. All but two studies [39, 45] that explored subjective pain scores found improved postoperative pain when ketorolac was added to the PAI. Decreased opioid usage was reported by Nikhar [3], Tammachote [47] and Andersen [37], but opioid usage was otherwise not significantly reduced [38,39,40]. Improvements in postoperative function, including range of motion [39, 40, 42, 44,45,46] and time to home readiness [37], were also investigated and were generally favorable, with some exceptions [39, 45]. Complications in the studies with ketorolac infiltration were rare, with one episode of hematoma [37] and one episode of DVT [43] highlighted across the studies. There were no documented statistical increases in complication rates for PAIs containing ketorolac. In the studies examined, the duration of action of ketorolac was reliably improved in the first 4–8 h [3, 37, 41,42,43,44, 46, 47], and pain relief was seen up to 96 h [43].

Overall, 9 out of the 12 studies did show a benefit of using ketorolac as part of multimodal pain management compared to a management plan not containing ketorolac. Exclusion criteria included previous surgery on the joint in question [37, 40, 42, 43, 46], previous infection [3, 41,42,43], bilateral knee joint involvement [44, 45], cardiac disease [3, 42, 43, 47], history of venous thromboembolism [42, 43], coagulopathies [3, 37, 40, 44], hepatic [3, 40, 46, 47] or kidney disease [3, 40, 42, 43, 46, 47], gastrointestinal ulcer [42, 43] or hemorrhage [42, 43], cerebrovascular disease [40, 42,43,44], neurocognitive disorders [41, 44, 47], rheumatoid arthritis [37, 40, 46], or allergies to the medications [3, 37, 40,41,42,43,44, 46, 47]. Patients under these exclusion criteria should be given extended clinical consideration before Toradol is included in periarticular infiltration during TKA.

Liposomal bupivacaine

Local anesthetics have proven beneficial in intraoperative and postoperative pain relief. Using local drugs, pain relief is mediated by impairing the voltage-gated sodium channels, which leads to decreased depolarization and therefore decreased conduction of pain signals [53]. However, their use is limited by their relatively short duration of action. Bupivacaine is a hepatically metabolized amide-type local anesthetic with an expected analgesic effect of up to 8–10 h [54].

Formulations of bupivacaine encapsulated in liposomes (liposomal bupivacaine) have been developed with the proposed benefit of sustained analgesia. These liposomes are composed of a lipid bilayer which encapsulates the anesthetic. This results in a delayed release of drugs based on lipid permeability and lipid bilayer breakdown prolonging the duration of action [55]. Initial studies have suggested an improvement of up to 72–96 h of effect [55]. The first formulations gained FDA approval in 2011. Subsequently, they began to be studied for the possibility of opioid-sparing effects [55].

Although some studies have shown benefits to liposomal bupivacaine over standard formulations[56], many have failed to show statistically significant improvements in opioid usage, time to discharge, or functional status in long-term follow-up [57,58,59,60,61,62] (Table 5). Several studies demonstrated no difference in pain experienced with rest [57,58,59,60,61,62,63]. Three studies did demonstrate decreased pain with physiotherapy [60, 64, 65], while only one study demonstrated the opposite [66]. Most studies showed no difference in the total narcotic usage [57,58,59,60,61, 64,65,66]. There were similar satisfaction rates in three studies [57, 62, 63], while only one study demonstrated better satisfaction with LB [56]. This has raised the question of the role of liposomal bupivacaine, especially given the greater-than-average cost of the medication60.

Table 5 Articles included in the review analyzing the use of liposomal bupivacaine in periarticular infiltrations

The most commonly used dosage in the RCTs examined was 20 mL (266 mg) of liposomal bupivacaine mixed with 40 mL of normal saline solution. Additional variations to the periarticular cocktail include adding epinephrine, ketorolac, or standard bupivacaine solutions.

The most commonly used exclusion criteria in the tabled studies include allergies to amide anesthetics [57, 59, 62, 63, 65, 66], chronic pain or opioid dependence [57, 59, 62, 64,65,66], abnormal hepatic, renal or cardiac function [56, 57, 65, 66] and elevated BMI [56, 62, 65, 66].

No specific complications were associated with liposomal bupivacaine described in the reviewed studies. Most reported complications involved arthroplasty complications such as infection, wound dehiscence or periprosthetic fracture following a fall [67]. One study involved a patient exceeding bupivacaine's toxicity threshold in a continuous femoral nerve block. However, this patient had no symptoms of systemic local toxicity [57]. The most reported serious complications of bupivacaine are related to cardiac and neurologic toxicity. These are more likely to occur when exceeding the recommended safe dose (2–2.5 mg/kg) but have been described in individuals even at lower doses [53]. The potentially life-threatening side effects highlight the importance of coordination between care team members to avoid reaching toxic doses.

Morphine

Morphine is an opioid medication that primarily acts through μ opioid receptors [68]. Activation of μ opioid receptors in the central nervous system has a long-standing history of use in pain control and sedation. Opioid receptors in the peripheral nervous system have also been described. However, the mechanism and effects are less well known.

The evidence described in arthroscopy literature shows the benefit of pain control with intra-articular injections. In animal studies, there is evidence for the benefit of local infiltration of morphine [69]. However, our literature review shows limited data for this in human studies (Table 6). Previous systematic reviews on periarticular morphine have identified a scarcity in the number of studies examining peri/intra-articular morphine [92] evaluating the effect of TXA as an additive to periarticular infiltration (Table 8). Tranexamic acid (TXA) is a competitive inhibitor of plasminogen, a component of the fibrinolytic pathway necessary for hemostasis [91]. As such, TXA has shown promise in orthopedic surgery in reducing bleeding-related complications, such as hemarthrosis or postoperative bleeding. In most cases, TXA is administered intravenously, but TXA can also be administered intra-articularly [93,94,95], using drain clam** [92, 96,97,98,99], or via intraoperative TXA soak [91].

Table 8 Articles included in the review analyzing the use of tranexamic acid in periarticular infiltrations

Peng et al. [91] hypothesized that TXA might be better functionally used as an element of PAI to infiltrate damaged tissues locally, prolonging the drug’s effects. They proposed that local infiltration may also reduce the risk of adverse effects [91], which include nausea, intraoperative hypotension, deep venous thrombosis, and blood transfusion [88,89,90, 92]. Peng et al. showed a significant decrease in HBL and blood loss in the PAI TXA group compared to the TXA administered intravenously [91]. Kim et al. showed decreased bleeding when PAI and IV administration are combined [88]. Pinsornak et al. [89] also showed decreases in blood loss and transfusion rates compared to intra-articular TXA.

However, the effect of TXA administered in a PAI may be limited to reduced blood loss. All the studies included showed decreased [89,90,91] or equivalent bleeding [88, 92] compared to other routes of TXA administration. Only one study by Zhang et al. [90] showed improved VAS scores and range of motion; it was only displayed in the short term. No studies demonstrated increased complications, including venous thromboembolism [88,89,90,91] or the need for transfusions [88,89,90,91], with the administration of TXA in a PAI.

Exclusion criteria for the studies listed include age < 18 [92] or > 80 [91], allergy to TXA [89, 91], secondary osteoarthritis [88, 89], bilateral TKA [88], cruciate-retaining prostheses [88], renal dysfunction [88, 91, 92], ischemic heart disease [88, 89], hepatic disease [88], malignancy [90], respiratory disease [88, 91], cerebrovascular disease [89], subarachnoid hemorrhage [89], acquired color-blindness [89], coagulopathy [88,89,90,91], or anticoagulation [89, 90], thrombocytopenia [88, 91], history of a pro-thrombotic condition or previous venous thromboembolism [88, 89, 91], pregnancy [91], breastfeeding [91], donated preoperative autologous blood [91, 92], postoperative allogenic blood transfusion [92], use of an unexpected prosthesis [92], severe synovectomy during the procedure [92] or low preoperative hemoglobin [89, 91]. Special consideration should be given to patients in these categories before providing TXA in a periarticular infiltration.

Conclusion

The ideal cocktail for pain control has not been established yet. Multiple drugs can be administered safely in this “cocktail” to help with pain control following a total knee and hip replacement. The medications should be individualized to avoid administering the medications to high-risk patients. Risk factors should be weighed against the benefits of the medications included in the periarticular injection. Even though the surgeon is administering the drug, there should be communication with the other team members (anesthesiologist, Internal medicine, ward physician, and pharmacists) to collaborate regarding the medications used and dosages. Certainly, other drugs might also prove beneficial in the future to optimize periarticular injections, and hopefully, future research might identify the optimal combination. The senior author uses periarticular injections around total knee and total hip replacements consistently with great success and have done so for the past 5 years. During this time, the “cocktail” has evolved and currently consists of 0.5% Ropivicaine 200 mg, Epinephrine 0.3 ml (1:1000), Ketorolac 30 mg, Dexmedetomidine 50ug, Methylprednisolone 40 mg and 1 g of Tranexamic acid. The senior author tries to adjust the “cocktail” if certain risk factors are present. That includes omitting Dexmedetomidine if patients have a history of bradycardia, hypotension etc., or Methylprednisolone if patients are a higher risk for infections among other things or adjusting the dosage of ropivacaine if a peripheral block was performed by the anesthesiologist. The main injection points utilized in total hip and knee replacements are at the highest concentrations of the nociceptors as described above.