The efficacy and safety of perioperative glucocorticoid for total knee arthroplasty: a systematic review and meta-analysis

Total knee arthroplasty (TKA) is a highly effective surgical treatment for severe knee arthritis which could reduce pain and maintain motor function [1, 2]. However, many patients still experience moderate to severe pain following TKA and require increased analgesic use postoperatively [3]. Various analgesic methods, including nerve blocks, local anesthetic infiltration, and intravenous opioids, are utilized in these patients [4‐6]. Unfortunately, patients often report dissatisfaction with pain management and are prone to complications such as nausea and vomiting [7]. Besides, surgical intervention triggers a local inflammatory response, which further leads to systemic inflammatioty response through the release of inflammatory factors [8]. Inflammation can further exacerbate postoperative joint pain, restrict knee joint range of motion, impede rapid postoperative rehabilitation, prolong hospital stay [9].

Glucocorticoids are commonly employed in TKA surgery due to their anti-inflammatory and antiemetic properties [10]. Some studies have indicated that glucocorticoids can effectively alleviate postoperative pain, reduce opioid consumption and mitigate complications such as nausea and vomiting in TKA patients. However, other studies have raised concerns about glucocorticoids usage, including potential adverse effects such as elevated blood glucose levels, infection, and impaired wound healing, particularly in patients receiving long-term glucocorticoid therapy [11]. Therefore, further investigation is required to determine the efficacy and safety of glucocorticoids in total knee arthroplasty. Additionally, the optimal medication, types, and number of doses of glucocorticoids administered perioperatively in TKA remain uncertain. The objective of this systematic review is to evaluate the available randomized controlled trials that investigate the efficacy and safety associated with glucocorticoid use in total knee arthroplasty. The main purposes of this paper include the following: (1) To explore the effect of glucocorticoids on postoperative rest pain and morphine consumption in patients after TKA; (2) To explore the effect of glucocorticoids on range of knee’s motion and length of hospital stay in patients after TKA; (3) To explore the effect of glucocorticoids on peripheral inflammation in patients after TKA; (4) To explore the adverse complications of glucocorticoids in patients after TKA, including PONV, high blood glucose, wound infection and venous thrombosis. Meanwhile, to investigate the impact of glucocorticoid dosage on PONV and compare the effect of the glucocorticoid by periarticular injection and system administration.

This systematic review and meta-analysis was conducted according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [12]. And this systemic review has been registered (PROSPERO ID: CRD42023435063).

The most important finding of the study is that the glucocorticoids can significantly reduce acute pain and morphine consumption within 24 h, and decrease the postoperative level of CRP and IL-6 and the incidence of PONV, and achieve a better ROM and shorter LOS, without increasing continued high blood glucose and the risk of postoperative wound infection and venous thrombosis.

Many patients experience severe pain after TKA surgery, and a significant number continue to suffer from chronic pain even after 3 months [53, 54]. To address this issue, numerous studies have investigated various postoperative analgesia protocols, such as the use of gabapentin [55] or dexmedetomidine [56] as adjuncts, as well as different combinations of nerve blocks [57]. Glucocorticoids have gained increasing attention as adjuvants for acute postoperative pain treatment in TKA [58]. In the current study, both intravenous and periarticular injection of glucocorticoids resulted in a significant decrease of 0.59 mean difference (MD) in resting pain scores on POD1 after surgery, which was consistent with previous studies [59‐62]. While the MD of improvement of the resting VAS on POM3 was 0.09. However, similar MD in improvement of resting VAS scores showed no statistically significant difference in other studies [61, 62]. Therefore the effect of glucocorticoids on chronic pain need to be further explored. Morphine consumption was also reduced in the glucocorticoid group on POD1 which was similar to the morphine reduction of 3.4 mg in Wang et al. study [61]. These findings are consistent with a meta-analysis conducted by De Oliveira et al., which concluded that intermediate and high doses of perioperative dexamethasone can effectively reduce postoperative pain and the need for additional painkillers [63]. They also found that preoperative dexamethasone was more effective in pain control compared to intraoperative administration. Similarly, Waldron et al. demonstrated the pain-relieving effect of perioperative dexamethasone in various surgical procedures, showing that patients who received dexamethasone required fewer postoperative opioids, had a longer time to first analgesic dose, needed less rescue analgesia, and had shorter stays in the recovery room [64]. In conclusion, glucocorticoid applications are effective for short-term pain control but long-term control needs to be explored further.

In this study, the beneficial effects of glucocorticoid administration extended beyond pain relief. It was observed that perioperative glucocorticoids administration significantly improved ROM and decreased the LOS, which is consistent with previous studies [65]. Patients with total knee osteoarthritis often have decreased ROM due to joint stiffness and pain. ROM is a significant parameter evaluating postoperative functional recovery for patients after TKA [66]. Better ROM is more benefit to different activities of daily living [67]and has a positive impact on TKA outcomes [68]. In our study, the glucocorticoid group presented about 5.22 degrees on POD1 and 3.50 degrees on POD3 increased in the ROM after TKA. Though these results were also confirmed in some studies [40, 59], a study from Xu B et indicated that the ROM on POD3 was not significantly increasing after applying glucocorticoids [30]. Glucocorticoid was found no significant improvement in long-term keen function of TKA patients due to it short-lasting pain relief [59]. In general, Glucocorticoids could improve short-term ROM contributing to better functional recovery [69]. Because many other variables may influence postoperative ROM, including surgical technique, implant design, and preoperative training as well as postoperative care, further investigation is required, including follow up for long-term ROM.

LOS is deemed as an important outcome for its economics burden [70]. Our study has showed that glucocorticoids shorten postoperative LOS by 0.27 day, which is lessen than 1.2 days in the study from Backes et al. [35] LOS is affected by many factors, including Enhanced Recovery After Surgery (ERAS), persistent postoperative pain, nausea and vomiting, wound infection, as well as the administration times and dose of glucocorticoids. As studies attention intensifies on ERAS, encompassing routine education, smoking and alcohol cessation, optimized fasting periods, standardized multimodal anesthesia protocols, alongside the utilization of local infiltrative analgesia, tranexamic acid, and multimodal prophylactic treatment for postoperative nausea, vomiting, and analgesia, a notable reduction in LOS by 5.4 days following TKA is achievable [71]. Therefore, further studies are warranted to substantiate the efficacy of combining glucocorticoids with ERAS in reducing Length of Stay.

Our study also found that periarticular injection analgesia group had lower pain score on POD2 but higher CRP level on POD 1 and POD2 compared to the systemic administration group in the two comparative studies. And in a subgroup analysis of the route of glucocorticoid administration across included studies, PIA group indicated decreased resting VAS on POD2, while the SA group didn’t. Previous studies have used glucocorticoids via different routes, with intravenous or topical administration being commonly used and proven to accelerate recovery after TKA [72, 73]. The different effects of intravenous or topical glucocorticoids in TKA patients may be explained by the analgesic mechanism of glucocorticoids. Glucocorticoids exert their analgesic effect by inhibiting phospholipase, which blocks the cyclooxygenase and lipoxygenase pathways in the inflammatory chain reaction [74‐76]. This leads to a decrease in the release of pain-causing substances such as bradykinin [74] from tissues and neuropeptides from nerve endings [75]. From this perspective, topical glucocorticoids directly inhibit the production of inflammatory factors, resulting in a faster and more targeted reduction of pain-causing substances such as bradykinin, compared to intravenous glucocorticoids [45, 51]. However, the advantages of locally administered glucocorticoids following TKA have not been conclusively demonstrated. Some studies have indicated that the use of local glucocorticoids results in prolonged pain relief and improved active knee flexion after surgery [77‐79]. On the other hand, a study by Christensen et al. [50] did not observe improvements on pain scores, range of knee motion, or narcotic consumption with local glucocorticoid administration. As similar in our study, there were insignificant difference of the resting VAS on POD1 and POD2 between PIA and SA group across all studies. These conflicting results may be attributed to differences in disease severity among the study groups, variations in treatments prior to TKA, differences in the type of steroid used, concomitant analgesic drugs, and variations in postoperative physical therapy approaches, which can influence the range of motion outcomes.

Indeed, TKA is associated with a profound inflammatory response due to the deep surgical site, osteotomy, massive blood loss, soft tissue injury, and surgical-related stress response. This led to increased levels of various inflammatory cytokines including CRP, IL-1β, IL-6, and TNF-a [80, 81]. The systemic and local inflammatory responses contribute to hyperalgesia [82, 83] and pain sensitization after TKA, which can result in severe pain even with a slight stimuli [30, 84]. Furthermore, the persistent inflammatory response can contribute to the development of chronic pain in the postoperative period [85]. Glucocorticoid had been wildly used in TKA to enhance recovery due to their potent anti-inflammatory effects [30, 72, 86]. The use of glucocorticoids can reduce CRP and IL6 level in the blood, as observed in our study. The analgesic effects of preoperative glucocorticoids can be partly attributed to their reduction of inflammation [87]. Glucocorticoids exert their anti-inflammatory response by suppressing the transcription of AP-1, NF-kB, and STAT3, which leads to the inhibition of proinflammatory cytokine production [88]. However, topical administration of glucocorticoid group showed higher CRP level postoperatively compared to system administration of glucocorticoid group in our study. The administration route we chose were closely associated with local or systemic inflammatory response. Intravenous corticosteroids have been proved to reduce systemic CRP and IL-6 effectively [30, 72, 86], while topical corticosteroids predominantly improve local inflammation at the surgical site in arthroplasty [45] and have limited effects on system inflammation after TKA [73]. However, these findings need to be confirmed through additional clinical studies, as there are only two comparative studies presented in this paper. Study by Ugras et al. proposed that amelioration of local inflammation is more significant in the analgesia of glucocorticoid than system inflammation [80], which is consistent with our results that topical administration of glucocorticoid has lower pain score on POD2. However, more studies are required to support it.

PONV commonly occur in TKA patients and can hinder participation in physiotherapy and early mobilization [89], leading to prolonged hospital stays and increasing the dissatisfaction of patients and health care costs [90, 91]. Glucocorticoids have been utilized in the prevention of PONV in many studies [92]. We found that the glucocorticoids have anti-emetic function after TKA not only in system administration but in local usage. The anti-emetic effect of glucocorticoids may be mediated through the inhibition of prostaglandin synthesis or the inhibition of endogenous opioid release [93]. Meanwhile, surgical trauma, anesthesia, and analgesics [94] are all factors associated with PONV, and surgical trauma can trigger the release of systemic inflammatory biomarkers, thereby increasing the occurrence of PONV [95]. The potent anti-inflammatory effects of glucocorticoids may contribute to their anti-emetic function. Though the anti-emetic of glucocorticoid has been well documented and supported in our study, the application dosage is still not clarified clearly. Therefore, we performed subgroup analysis of glucocorticoid dose and concluded that low dosage (≤ 15 mg dexamethasone) and high dosage (> 15 mg dexamethasone) both reduced the incidence of PONV, with high dosage reducing PONV more significantly. And pre-emptive low dose (10 mg) dexamethasone has been proved to reduce postoperative emesis [36]. Though higher dosage of glucocorticoid is more beneficial in reducing PONV, we cannot ignore higher risk of complications with it. It has so far been proposed multiple low dosage of glucocorticoid is more effective on decrease PONV and early rehabilitation [59, 86]. However, a single high dose of 20 mg dexamethasone is more preferable to multiple low dose application by an opposite study [96]. It may be related to the type of glucocorticoid, as they have different times of action and should therefore be used specifically according to the type of glucocorticoid in clinical.

In addition, referring to the concern of hyperglycemia caused by the administration of corticosteroids, previous studies indicated that the fasting blood glucose of patients treated with topical and intravenous glucocorticoid was significantly increased after TKA [52, 97, 98], which was similar to significant elevation of blood glucose in the glucocorticoid group on POD1 in the study. Hyperglycemia, especially when levels exceed 200 mg/dl has been associated with an increased risk of surgical site infection and wound complications [99, 100]. However, other studies reported that the intravenous administration of dexamethasone did not increase fasting blood glucose levels postoperatively following TKA [35, 101], which were consistent with the insignificant increase blood glucose in patients of glucocorticoid group after TKA on POD2. Different results may be related to the determination time of blood glucose and Hormone dosage. It also indicated that preoperative intravenous administration of methylprednisolone resulted in a transient postoperative increase in plasma glucose and insulin resistance, which were normalized 48 h postoperatively [97].

The other complications in terms of wound drainage, delayed wound healing, intramuscular venous thrombosis, which were not significantly different between the two groups. To our knowledge, most important possible risks with steroid during post-operative period, including gastric ulcers [102], impaired wound healing [103] or wound infections [11] were exclusively associated with chronic glucocorticoid use not with single dose glucocorticoid use. And the side effects from glucocorticoid use are proportional to the duration and intensity of therapy [104]. Theses similar results were presented in many articles [105]. Even though, no studies have confirmed that a short term use of glucocorticoid has not harm effect on patients of TKA. In addition, it may be small samples and low evidenced articles that result in negative adverse outcomes in single dose of glucocorticoid use [106]. Therefore, further high-quality studies with larger sample sizes is needed to establish the safety of perioperative glucocorticoid administration in TKA patients.

We have investigated the clinical efficacy of most glucocorticoid, but the optimal corticosteroids administration regimen regarding of dosage, times, drug combination are still undetermined. In concern of the comparison between topical and systemic glucocorticoid, there need more studies to explore their different effect on pain and system inflammation. Additionally, the lack of long-term follow-up data highlights the need for further studies to explore and validate the results over an extended period of time.

This systematic review and meta-analysis suggest that glucocorticoids may exert pain relief within 24 h, although its clinical relevance to patients may be limited. The application of glucocorticoids provided antiemetic effects, as well as facilitate functional recovery, accompanied by the inhibition of inflammatory factors releasing after TKA, including plasma CRP and IL-6. Importantly, these effects are observed without elevating blood glucose levels persistently, increasing the risks of wound infection and venous thrombosis.