Comparative effects of intracapsular corticosteroid injection on pain control after total hip arthroplasty: a prospective, randomized controlled trial

Abstract

Effective pain control after total hip arthroplasty (THA) is essential for postoperative recovery; however, limited data exist comparing different corticosteroid agents using a three-arm design that includes a saline control, administered as a single-site intracapsular injection with extended postoperative assessment. This prospective, single-center, randomized controlled trial compared the analgesic effects of two intracapsular corticosteroids with saline. One hundred fifty patients undergoing primary THA were randomized to receive an injection into the hip joint capsule at wound closure with triamcinolone acetonide plus bupivacaine, betamethasone plus bupivacaine, or saline. After protocol deviations, 138 patients were included in the final analysis. Postoperative pain was assessed using the visual analog scale (VAS) at rest and during motion on postoperative days 3, 5, and 7. Baseline demographics and perioperative variables were comparable among groups. At rest, triamcinolone acetonide resulted in significantly lower VAS scores than saline across all assessment points. During motion, both corticosteroid groups demonstrated a significantly higher proportion of patients achieving sufficient pain control on postoperative day 3 compared with saline. Rescue analgesic use and frequency were significantly reduced in the corticosteroid groups. No major injection-related complications were observed. These findings indicate differential analgesic effects among corticosteroid agents and suggest that triamcinolone acetonide provides more effective postoperative pain control after THA.

Introduction

Total hip arthroplasty (THA) has long been regarded as one of the most successful surgical procedures of the 20th century, providing excellent pain relief, restoration of function, and high levels of patient satisfaction for individuals suffering from end-stage hip osteoarthritis¹. Despite these favorable outcomes, it is well recognized that persistent postoperative pain can negatively affect long-term functional recovery and patient satisfaction after THA^2,3.

Postoperative pain management following joint arthroplasty has therefore been extensively investigated, especially for total knee arthroplasty (TKA). Numerous studies have reported the use of opioids, corticosteroids, local anesthetic injections, as well as intravenous and oral analgesics to achieve effective pain control in TKA patients^4,5. In contrast, fewer studies have focused on pain control after THA, partly because early mobilization and rehabilitation are generally perceived to be less painful than after TKA⁴.

Previous approaches to THA pain control have frequently involved multimodal periarticular injections (“cocktail” formulations) administered into the joint capsule, fascia, and subcutaneous tissues^6,7. Several studies have also included opioids such as morphine in these mixtures, which can cause well-known side effects, such as nausea or vomiting, constipation, sedation, dizziness, urinary retention, respiratory depression, delirium, and potential dependence, which limit their desirability in postoperative protocols^5,8.

Given these limitations, we examined the efficacy of an opioid-free, corticosteroid-enhanced injection administered solely into the hip joint capsule. This focused approach stems from evidence demonstrating a high density of nociceptors in the joint capsule⁹. Additionally, we compared two different steroids to determine which offers superior pain control. Recent studies suggest that soft tissue infiltration with saline can itself reduce pain in procedures like fascial release, prompting us to include a saline-only control group¹⁰. Consequently, we conducted a three-arm, prospective, randomized trial to evaluate pain outcomes following THA.

Materials and methods

Ethics

This study was approved by the institutional ethics committee and conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent. Trial registration: UMIN000042302.

Study design

This study was designed as a single-center, prospective, single-blind randomized controlled trial. Due to visible differences in the injectates, the operating surgeons were aware of the group allocation; however, patients and outcome assessors were blinded throughout the study period.

Participants

Eligible participants were patients aged 20–79 years undergoing primary THA under spinal anesthesia for primary or secondary osteoarthritis of the hip or osteonecrosis of the femoral head. All patients provided written informed consent.

Inclusion criteria included the ability to provide informed consent and compliance with scheduled visits during the study period. Exclusion criteria were difficulty with spinal anesthesia, complete capsular excision required during surgery, poorly controlled diabetes mellitus or immunocompromised status, neuromuscular disease in the affected limb, psychiatric conditions interfering with consent or evaluation, allergy to study drugs, planned lower limb surgery within 6 months before or after enrollment, and other conditions deemed inappropriate by the investigators.

Randomization and blinding

After consent, patients were randomized into three groups (n = 50 per group) using block randomization stratified by age and sex by an independent third party not involved in patient care (Fig. 1). Group allocation was concealed until surgery. Surgeons could not be blinded due to visible differences in the injectates; however, patients and evaluators collecting outcome data were blinded.

Fig. 1.

Participant enrollment flow diagram. Group A, Triamcinolone acetonide 40 mg + 20 mL of 0.5% Levobupivacaine (100 mg); Group B, Betamethasone 4 mg + 20 mL of 0.5% Levobupivacaine (100 mg); Group C, Saline 20 mg of intracapsular injection at the time of wound closure.

Interventions

At the time of wound closure, an intracapsular injection was administered according to randomization:

• Group A: Triamcinolone acetonide 40 mg + 20 mL of 0.5% levobupivacaine (100 mg).

• Group B: Betamethasone 4 mg + 20 mL of 0.5% levobupivacaine (100 mg).

• Group C (control): Saline 20 mL.

Six experienced hip surgeons performed the operations, each with > 50 THAs/year of experience. The solution was injected into the capsular tissue at four sites: two in the anterior capsule and two in the posterior capsule (5 mL per site) via a posterior approach using an 18-gauge needle with an injection depth of approximately 3 cm. After completion of the injection, the posterior capsule was repaired using interrupted sutures with 0 polyglactin 910 antibacterial absorbable sutures (Fig. 2).

Fig. 2.

Diagram showing the method of intracapsular injection (posterior approach in right hip joint). At the time of wound closure, a total of four injections with 5 mL at each site are administered into the anterior and posterior joint capsules. After completion of the injection, the posterior capsule was repaired using interrupted sutures with 0 polyglactin 910 antibacterial absorbable sutures.

Perioperative pain management protocol

Postoperatively, all patients received acetaminophen 15 mg/kg intravenously at 3, 6, 12, and 24 h after surgery. From postoperative day 1, acetaminophen 400 mg was administered orally three times daily, and celecoxib was prescribed (400 mg initially, then 200 mg twice daily) for 7 days. Additional analgesia with tramadol sustained-release tablets (100 mg) was permitted as rescue medication when needed. The use and frequency of rescue analgesics were recorded.

Outcome measures

The primary outcome was postoperative pain assessed using the visual analog scale (VAS) at rest and during motion on postoperative days 3, 5, and 7. The VAS ranged from 0 mm (no pain) to 100 mm (worst pain imaginable). Patients recorded VAS scores independently on paper forms distributed by a physician not involved in their surgical care. These forms were collected at discharge, ensuring that treating surgeons were not aware of the individual pain scores during hospitalization. Secondary outcomes included the total use and frequency of rescue analgesics and perioperative complications. Postoperative inflammatory responses were assessed using C-reactive protein (CRP) levels measured on postoperative days 1, 4, and 7. In addition, early postoperative functional recovery was assessed using a 4-meter walk test¹¹ and walking status. Patients were instructed to walk 4 m at their usual pace, and the walking time was recorded. Walking status was categorized into three levels: walker-assisted ambulation, cane-assisted ambulation, and independent ambulation. Measurements were obtained preoperatively and at postoperative day 10.

Sample size and statistical analysis

The required sample size was calculated to detect a minimum clinically important difference of 10 mm in VAS score (SD = 30 mm) with a significance level of α = 0.05 and a power of 80%, resulting in a total of 134 patients; to account for possible dropouts, 150 patients were enrolled.

Patient demographics and baseline characteristics were compared among the three groups using one-way analysis of variance (ANOVA) for continuous variables (age, height, weight, body mass index (BMI), surgical time, and blood loss) and the chi-squared test for categorical variables (sex and diagnosis). Repeated measures ANOVA with the Tukey–Kramer method for multiple comparisons was used to analyze VAS scores at each time point. The percentage of patients achieving sufficient pain control (VAS < 20) was compared among groups using the chi-squared test. A VAS score < 20 mm was defined as sufficient pain control, in line with previous studies^12,13. The rate and number of patients requiring rescue analgesics were compared among groups using ANOVA with the Tukey–Kramer method. Postoperative walking time (4-meter walk test), change from baseline, CRP levels at each time point, and length of hospital stay were compared among groups using one-way ANOVA with the Tukey–Kramer method for multiple comparisons. Preoperative walking status, postoperative walking status at postoperative day 10 (walker-assisted ambulation, cane-assisted ambulation, or independent ambulation), and changes in walking status from baseline were compared among groups using the Kruskal–Wallis test. P value of < 0.05 was considered statistically significant. All statistical analyses were performed using JMP software (version 9.0.1; SAS Institute Inc., Cary, NC, USA).

Results

A total of 150 patients were enrolled and randomized into three groups of 50 patients each. Twelve patients were excluded intraoperatively due to conversion to general anesthesia, resulting in 138 patients included in the final analysis (Fig. 1).

Baseline characteristics

Table 1 summarizes patient demographics and perioperative findings. There were no significant differences among the groups in age, sex distribution, diagnosis (osteoarthritis vs. osteonecrosis), height, body weight, BMI, surgical time, or intraoperative blood loss.

Table 1.

Baseline demographic and perioperative characteristics of the study groups.

Parameters	Groups			P value
	Group A n = 45	Group B n = 47	Group C n = 46
Age at surgery (years old)	64.9 ± 8.3 (46–76)	64.3 ± 8.8 (45–78)	65.5 ± 6.7 (46–76)	0.79
Sex; men/women, n (%)	8 (17.8)/37 (82.2)	2 (4.3)/45 (95.7)	3 (6.5)/43 (93.5)	0.95
Diagnosis (OA/ONFH), n (%)	42 (93.4)/3 (6.7)	45 (95.7)/2 (4.3)	43 (93.5)/3 (6.5)	0.86
Height (cm)	157.6 ± 7.8 (141.1–182.0)	158.3 ± 6.7 (147.5–173.3)	155.7 ± 7.8 (140.4–168.7)	0.18
Body weight (kg)	63.2 ± 9.0 (39.6–79.9)	61.5 ± 9.5 (44.0–82.6)	59.1 ± 9.4 (41.5–80.9)	0.12
BMI (kg/m2)	25.5 ± 3.5 (18.3–31.6)	24.5 ± 3.2 (18.2–30.8)	24.4 ± 3.5 (18.3–36.6)	0.23
Surgical time (minutes)	76.9 ± 11.8 (59–107)	79.0 ± 15.3 (52–120)	78.9 ± 17.3 (58–154)	0.76
Intraoperative blood loss (g)	142.1 ± 93.2 (10–358)	101.1 ± 83.1 (20–442)	122.0 ± 91.5 (10–400)	0.10

Group A, Triamcinolone acetonide 40 mg + 0.5% Levobupivacaine 20 mL; Group B, Betamethasone 4 mg + 0.5% Levobupivacaine 20 mL; Group C, Saline 20 mg of intracapsular injection at the time of wound closure. Continuous values are expressed as mean ± standard deviation (range). OA: osteoarthritis of the hip, ONFH: osteonecrosis of the femoral head, BMI: body mass index.

Pain outcomes

Figure 3a and b show the changes in VAS scores at rest and during motion. At rest, mean VAS scores decreased over time in all groups; however, Group A consistently demonstrated significantly lower mean VAS scores than Group C on postoperative days 3, 5, and 7 (P < 0.05). For example, on day 3, the mean VAS was 14.6 ± 13.4 in Group A compared with 23.5 ± 20.9 in Group C (P < 0.05). On day 7, the proportion of patients with sufficient pain control (VAS < 20) was higher in Group A (86.7%) than in Group C (63.0%) (P < 0.05).

Fig. 3.

Changes in visual analog scale (VAS) scores (a and b) and postoperative C-reactive protein (CRP) levels (c) (※P < 0.05). (a) At rest, the VAS scores in group A are significantly lower than those in group C on postoperative days 3, 5, and 7. (b) During exercise, there are no significant differences in the VAS scores between the three groups, although the VAS scores in group A are lower than those in group B and group C on postoperative days 3, 5, and 7. (c) On postoperative day 1, CRP levels in group B are significantly lower than those in group C. On postoperative day 4, CRP levels in groups A and B are significantly lower than those in group C. On postoperative day 7, CRP levels in group A are significantly lower than those in groups B and C.

During motion, mean VAS scores also decreased in a time-dependent manner in all groups. Although there were no significant differences in mean VAS between groups at each time point, the proportion of patients with VAS < 20 was significantly lower in Group C than in Groups A and B on day 3 (P < 0.05).

Functional outcomes

The mean postoperative 4-meter walk time at week 2 was 4.45 ± 1.41 s in Group A, 4.56 ± 1.92 s in Group B, and 4.84 ± 1.92 s in Group C, with no significant difference among groups (P = 0.559). However, the improvement in 4-meter walk time from baseline was significantly greater in Group A than in Group C (P < 0.05). Preoperative walking status did not significantly differ among the three groups (P = 0.872). Postoperative walking status at postoperative day 10 also did not significantly differ among the three groups (P = 0.532). In addition, changes in walking status from preoperative assessment to postoperative day 10 did not significantly differ among the groups (P = 0.831).

Inflammatory markers

Figure 3c shows the changes in postoperative CRP levels among the three groups. CRP levels differed significantly among the groups at several time points. On postoperative day 1, mean CRP levels were significantly lower in Group B than in the control group (P < 0.05). On postoperative day 4, both steroid groups (Groups A and B) showed significantly lower CRP levels than the control group (P < 0.05). On postoperative day 7, CRP levels were significantly lower in Group A than in Groups B and C (P < 0.05).

Use of additional analgesics

Figure 4 illustrates the use and frequency of additional analgesics for pain control. The proportion of patients requiring rescue analgesics and the average number of doses were significantly lower in Groups A and B than in Group C (P < 0.05).

Fig. 4.

Additional medications for pain control (※P < 0.05). (a) The usage rates of additional medications for pain control in Group A and Group B are significantly lower compared to Group C. (b) The average numbers of additional medications used in Group A and Group B are significantly lower compared to Group C.

Length of hospital stay

The mean length of hospital stay was 15.51 ± 3.50 days in Group A, 15.40 ± 2.37 days in Group B, and 15.93 ± 3.05 days in Group C, with no significant difference among groups (P = 0.668).

Complications

Superficial surgical site infections were reported in one patient each in Groups B and C. No deep infections or other major complications occurred during hospitalization.

Discussion

Effective postoperative pain management is essential for patient satisfaction and functional recovery after THA⁵. In this study, we demonstrated that a single intracapsular injection of local anesthetic with triamcinolone acetonide significantly reduced postoperative pain at rest compared to placebo, despite all groups receiving the same standardized multimodal analgesic regimen including intravenous acetaminophen and oral celecoxib. Unlike many previous studies that included opioids such as morphine within periarticular injections^6,7, our injection protocol was opioid-free, thereby eliminating concerns about well-known opioid-related side effects such as nausea, constipation, sedation, urinary retention, and dependence⁸.

Notably, even though all groups received sufficient baseline analgesia, the control group’s VAS scores fell below 20 mm by postoperative day 5, which is generally considered adequate pain control^12,13. Nonetheless, the triamcinolone group showed significantly better pain relief than placebo, suggesting that the single-site intracapsular steroid injection provides an additive benefit, especially in the early postoperative period. Moreover, the proportion analysis further demonstrated that at rest on postoperative day 7, significantly more patients in the triamcinolone group achieved sufficient pain control (VAS < 20) compared to the control group. This shows that the additive effect of the single-site steroid injection was maintained even when overall pain levels had declined in all groups.

Additionally, although the mean VAS during motion did not show a significant difference as continuous data, the proportion of patients with VAS < 20 during motion on day 3 was significantly higher in both steroid groups compared to the control group. This underlines the potential benefit of assessing pain outcomes using categorical thresholds. Furthermore, measuring pain during motion in addition to at rest adds novelty to our study design, as prior studies have mainly focused on rest pain only^14–17.

Another unique feature of this study is that the injection was administered solely into capsular tissue of the hip joint capsule⁵. This approach is anatomically justified, as nociceptors are widely distributed throughout the hip capsule, with particularly high concentrations at the labral base and ligamentum teres⁹. By targeting this specific structure, we aimed to maximize local pain control while avoiding unnecessary injection into surrounding soft tissues. While previous studies have reported the use of rescue analgesics, few have quantified the frequency or total number of rescue doses^14,17. In this study, both the rate and mean number of rescue analgesic administrations were significantly lower in the steroid groups (Groups A and B) than in the control group, further supporting the analgesic effect of intracapsular steroid injection.

Interestingly, the effect was more pronounced with triamcinolone acetonide than with betamethasone. Although direct comparative studies are limited, it has been suggested that triamcinolone’s prolonged local anti-inflammatory action and slower release profile may contribute to superior pain control compared to other corticosteroids¹⁸. This difference may be due to its more crystalline structure and lower solubility, which allow for slower absorption and sustained local effects. Similar findings have been reported for intra-articular injections in other joints, such as the knee and shoulder, where triamcinolone has provided longer-lasting pain relief than other corticosteroids^19,20. In the present study, postoperative CRP levels were also significantly lower in the steroid groups than in the control group at several time points, with a more sustained reduction observed in the triamcinolone group. These findings suggest that intracapsular corticosteroid injection suppressed postoperative inflammatory responses, with patterns consistent with the pharmacologic duration of the steroids.

Another strength of this study is the use of a saline-only control group, which allowed us to exclude the possibility that the mechanical effect of soft tissue infiltration alone (as in hydrodissection or fascial plane block) was responsible for pain relief¹⁰. This enhances the internal validity of our findings.

This study has several limitations. First, we did not evaluate the association between VAS scores and objective functional outcomes such as hip-specific clinical scores. Although early functional recovery was partially evaluated using the 4-meter walk test, other functional parameters and the length of hospital stay did not significantly differ among the groups. This may be partly explained by the standardized clinical pathway at our institution, in which patients are generally discharged approximately 2 weeks after THA, potentially limiting the ability to detect differences in recovery speed among treatment groups. Second, postoperative pain outcomes were evaluated only up to postoperative day 7. Although our study focused on early postoperative pain control after THA, longer follow-up would be helpful to clarify the duration of the analgesic effect and to evaluate potential complications such as infection or wound healing problems. Third, the procedures were performed by multiple surgeons rather than a single surgeon, which may introduce surgeon-related variability. Although the injection technique was standardized among the participating surgeons using a shared instructional video prior to the start of the study, some variability in surgical technique cannot be completely excluded. Fourth, the doses of triamcinolone acetonide (40 mg) and betamethasone (4 mg) were selected based on commonly used clinical doses rather than strict pharmacologic equivalency. Therefore, the exact anti-inflammatory potency of the two agents may not have been identical, which could potentially influence the observed differences between groups. Fifth, the single-center design and relatively small sample size may limit the generalizability of our findings. In addition, all procedures were performed using a posterior approach under spinal anesthesia at a high-volume academic center in Japan, which may further limit the applicability of these results to other surgical approaches or anesthetic protocols. Finally, the open-label design for surgeons due to visible injectate differences may introduce a potential source of bias, although patients and outcome assessors were blinded.

In conclusion, a single-site intracapsular injection of local anesthetic with triamcinolone acetonide effectively reduced postoperative pain and additional analgesic use without the need for opioids following THA. This simple technique may offer a safe and practical option for enhancing early recovery after hip arthroplasty.

Abbreviations

THA

Total hip arthroplasty

RCT

Randomized controlled trial

VAS

Visual analog scale

TKA

Total knee arthroplasty

CRP

C-reactive protein

ANOVA

Analysis of variance

BMI

Body mass index

Author contributions

SI performed the literature review and drafted the manuscript. GM and YN supervised writing of the manuscript. SI, TU, KS, GM, SH, SK, TS, RY, and DH contributed to the data collection. SI, TU, and KS analyzed the data. All authors read and approved the final manuscript.

Funding

This work was supported in part by a Grant-in-Aid in Scientific Research (No. JP19K09628) from the Japan Society for the Promotion of Science.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval and consent to participate

This prospective randomized controlled trial was approved by the Institutional Review Board of Kyushu University (Registration number: 20202011, Approved date: October 23, 2020). Written informed consent was obtained from all participants prior to enrollment. The trial was registered with the UMIN Clinical Trials Registry (UMIN000042302, registered on November 2, 2020).

Consent for publication

Not applicable.