Blocking the pericapsular nerve group against the lumbar erector spinae plane for pain management following total hip replacement: A randomized controlled trial

ABSTRACT

Background:

Both the pericapsular nerve group block (PENG) and the lumbar erector spinae plane block (L-ESPB) are emerging regional anesthetic techniques, yet comparative data remain limited.

Patients and Methods:

This randomized, controlled study included 60 patients undergoing THA, assigned to receive either a PENG block (n = 30) or an L-ESPB (n = 30). The primary outcome was the time to the first analgesic request. Secondary outcomes included total analgesic consumption, pain scores measured by the numeric rating scale (NRS) at rest and during movement, assessment of quadriceps motor strength, and any block-related complications. All participants were administered a standardized multimodal analgesic protocol.

Findings:

The PENG group demonstrated a significantly longer time to the first analgesic request compared to the L-ESPB group (12.88 ± 4.56 hour vs. 8.45 ± 3.49 hours; P < 0.001) and lower total analgesic consumption (25.84 mg ± 5.59 mg vs. 40.18 mg ± 7.89 mg; P < 0.001). Resting NRS scores were comparable early postoperatively but were significantly lower in the PENG group at 24 hours. Dynamic pain scores were consistently lower in the PENG group across all time points. Quadriceps motor weakness was more frequent in the PENG group at 6 hours (P = 0.04), but resolved completely by 12 hours. No major complications were observed in either group.

Conclusion:

For postoperative analgesia following THA, the PENG block is superior to L-ESPB, offering prolonged analgesia and reduced opioid requirements. Although transient quadriceps weakness may occur, it resolves early without clinical sequelae.

Introduction

Pain management is a challenging task following total hip arthroplasty (THA), as inadequate control impedes early mobilization, rehabilitation, and patient satisfaction. With advancements in ultrasound-guided techniques, peripheral nerve blocks have emerged as effective opioid-sparing alternatives, providing targeted analgesia with fewer systemic side effects.[1]

Pericapsular nerve group (PENG)[2] block, introduced in 2018, targets the articular branches of the femoral and obturator nerves. It has gained attention for its potential to provide significant analgesia while theoretically preserving motor function.[2] The technique has been utilized as an adjunct for surgical anesthesia and postoperative analgesia in hip arthroscopy and arthroplasty.[3,4]

Lumbar erector spinae plane block (L-ESPB) has become a widely used interfacial plane block for various trunk and extremity procedures. The lumbar approach (L-ESPB) is an established technique that provides analgesia for hip surgery by spreading local anesthetic into the paravertebral space, where it blocks the ventral rami of the lumbar spinal nerves (L1–L4). Although the exact mechanism remains under investigation, cadaveric and clinical studies suggest that the injectate spreads via the erector spinae fascia to the paravertebral space, thereby anesthetizing multiple nerve roots involved in hip innervation.[5] While both PENG and L-ESPB have shown efficacy in hip surgery, prospective head-to-head comparisons are lacking. Existing literature typically compares these blocks against placebo or older techniques, such as the fascia iliaca block, leaving a gap in evidence regarding their relative efficacy, duration, and impact on dynamic pain control, essential for early postoperative mobilization.

This study aims to compare time to first analgesic request between patients receiving a PENG block and those receiving a L-ESPB following THA, with time to first analgesic request as the primary outcome. Secondary objectives include comparing total analgesic consumption within the first 24 postoperative hours; pain scores on the numeric rating scale (NRS) at rest and during movement at 6, 12, and 24 hours; the presence and duration of quadriceps motor weakness; and any blockrelated complications. The null hypothesis of the study supposes that the PENG block does not provide superior postoperative analgesia compared to the L-ESPB, as measured by time to first analgesic request, reduced opioid consumption, or improved dynamic pain control, while preserving motor function.

Methods

Study setting and design

A randomized, blinded, controlled study was conducted from 2022 to 2024 in the orthopedic section of the operating theatre. The local research ethics committee of the Faculty of Medicine approved this randomized clinical study (approval number: 17200688; date of approval: February 2022). The trial was prospectively registered in ClinicalTrials.gov (registration number: NCT 05050487; date of registration: April 2022. Informed written consent has been gathered from all study participants. This research was conducted in full accordance in accordance with the ethical standards outlined in the Declaration of Helsinki. Patient enrollment ran from May 2022 to October 2024.

Selection criteria

Patients aged 18 years or older, classified as American Society of Anesthesiologists (ASA) physical status I–III, scheduled to undergo THA were eligible. Exclusion criteria included age under 18, pregnancy or breastfeeding, opioid dependence, multiple traumas, cognitive impairment preventing reliable pain assessment, anticoagulant therapy, and infection at the injection site.

Randomization and blinding

Participants were assigned in a 1:1 ratio to the L-ESPB or PENG group through a computer-generated randomization sequence. Group allocation was concealed by using opaque, sealed envelopes prepared by an independent staff member. On the day of surgery, the assigned anesthesiologist opened the envelope and performed the block.

Surgeons, postoperative care providers, and outcome assessors were blinded to their group assignment. Considering the nature of the interventions, the anesthesiologist administering the block could not be blinded but had no role in intraoperative management, postoperative assessment, or data analysis. To maintain blinding, blocks were performed before spinal anesthesia, and patients were mildly sedated during the procedure with midazolam (0.05 mg/kg I.V), minimizing awareness of positioning differences.

Preoperative assessment

All patients underwent a standard preoperative evaluation that included a thorough history and clinical examination. Demographic data such as age and gender, as well as relevant comorbidities, were recorded. The baseline hemodynamic parameters analyzed included noninvasive heart rate (HR), blood pressure, and oxygen saturation (SpO₂). Routine laboratory investigations, including coagulation profile and random blood glucose, were rechecked.

Intraoperative assessment and anesthesia technique

Intravenous access was established, and Ringer’s solution infusion began as soon as the patient entered the operating room. Standard monitoring included intermittent noninvasive blood pressure checks, pulse oximetry, and continuous electrocardiography. All patients received standardized premedication with intravenous metoclopramide (10 mg) and famotidine (20 mg). The type and duration of hip arthroplasty procedures were recorded. Mean arterial pressure (MAP), HR, SpO₂, systolic blood pressure (SBP), as well as diastolic blood pressure (DBP), were among the intraoperative hemodynamic parameters that were measured every 15 min till the procedure was finished. Patients were randomly allocated to undergo one of the following ultrasound-guided regional blocks:

Pericapsular nerve group

With the patient in the supine position, a PENG block was performed under ultrasound guidance. A curvilinear probe with low frequency (2–5 MHz) was positioned in line with the pubic ramus and covering the anterior inferior iliac spine.

The iliopsoas muscle and tendon, the pectineus muscle, and the femoral artery were recognized as important anatomical landmarks. The plane that connects the psoas tendon to the pubic ramus was reached by advancing a 22-gauge echogenic needle in-plane from lateral to medial. Following a negative aspiration, 20 milliliters of 0.25% bupivacaine were injected.

Lumbar erector spinae plane block group

The patient was positioned laterally for the L-ESPB that was performed under ultrasound guidance. The L4 vertebral level was identified, and the probe was placed in a sagittal orientation approximately 3.5–4 centimeters lateral to the midline. After locating the erector spinae muscle and transverse process, a needle was placed into the fascial plane. Twenty milliliters of 0.25% bupivacaine were administered in the fascial plane between the erector spinae muscle and the transverse process.

Experienced anesthesiologists performed all blocks. Following the block, standard spinal anesthesia was administered using 12–15 mg of 0.5% hyperbaric bupivacaine. No intrathecal opioids were used. All surgeries were performed as elective primary THA by the same orthopedic surgical team utilizing a standardized surgical method.

Postoperative assessment

Postoperative monitoring included hemodynamic parameters (HR, SpO₂, SBP, DBP, MAP) recorded at 2, 4, 6, 12, and 24 hours. Patients received a standardized multimodal regimen: intravenous paracetamol 1 g every 8 hours. Rescue analgesia (IV ketorolac 30 mg) was administered if the NRS score exceeded 3, with a 12-hour lockout interval. If pain remained uncontrolled (NRS >3), opioids (nalbuphine 4 mg) were administered. Within the first 24 hours, total analgesic consumption and time to 1^st analgesic request have been recorded. Quadriceps motor block has been evaluated at 6, 12, and 24 hours postoperatively using a 3-point scale: normal strength (0, full knee extension against resistance), paresis (1, reduced strength with partial extension), and paralysis (2, no active extension or contraction). Assessment was performed supine with the hip flexed 45° and knee 90°, testing extension against gravity and resistance. Pain was evaluated using the NRS (0–10) at rest and throughout hip flexion at 6, 12, and 24 hours. Block-related complications (local anesthetic toxicity, paresthesia, and vascular puncture) were documented. Perioperative management was standardized across both groups, including uniform intraoperative anesthesia, identical adjunct medications, consistent postoperative multimodal analgesia, and standardized mobilization and physiotherapy protocols, ensuring equal distribution of these factors between groups and minimizing confounding effects.

Outcomes of the study

The primary outcome was the time to the first analgesic request (from the block to the patient’s first request). Secondary outcomes included the total cumulative dose of ketorolac (in mg) administered within the first 24 hours after surgery. Additionally, several outcomes were examined on an exploratory, post hoc basis. These include postoperative pain scores (measured with the NRS at rest and during hip flexion at 6, 12, and 24 hours), the presence and severity of quadriceps motor block at the same intervals, hemodynamic parameters (HR, systolic/BBP, and MAP at 2, 4, 6, 12, and 24 hours), and blockrelated complications (e.g., local anesthetic toxicity, paresthesia, vascular puncture).

Sample size calculation

Sample size calculation was performed using G*Power version 3.1.9.2 (Heinrich Heine University, Düsseldorf, Germany). The primary outcome was the time to the first request for analgesia. A between-group difference of 4 hours was considered clinically significant. Variance estimates were derived from the randomized trial by Aliste et al.,[6] which reported times to first analgesic request of 13.0 ± 4.5 hours and 8.3 ± 3.8 hours, respectively, corresponding to a pooled standard deviation of 4.2 hours. Using a two-sided α of 0.05 and 85% power, 25 patients were required per group. To compensate for an anticipated dropout rate of 15%, 30 patients were enrolled in each group.

Statistical analysis

Statistical analyses were conducted using IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp., Armonk, NY, USA). The Shapiro-Wilk test was employed to assess normality. Continuous variables were reported as mean ± standard deviation (SD) and compared with Student’s t-test or Mann-Whitney U test, as appropriate. Categorical variables were summarized as frequencies and percentages and compared using the Chi-square test. Survival analysis for the time to first analgesic request was performed using the Kaplan-Meier method with the log-rank test. Statistical significance was defined as P < 0.05, and confidence intervals were set at 95%.

Results

Seventy cases were assessed for eligibility; 10 were excluded from the study (5 had multiple traumas, 3 were on anticoagulation, and 2 refused to participate). A total of 60 cases were enrolled in the study, with 30 patients assigned to each group [Figure 1].

Figure 1.

CONSORT flow chart of the current study. PENG: pericapsular nerve group; L-ESP: lumbar-erector spinae plane

Baseline characteristics and operative times were comparable between groups [Table 1]. The total number of patients who requested analgesia was similar in both groups (80% vs. 93.3%; P = 0.120). Meanwhile, the PENG group had a significantly longer time to the first analgesic request (12.88 ± 4.56 hours vs. 8.45 ± 3.49 hours; P < 0.001) and a lower cumulative dose of analgesic consumption (ketorolac in mg) (25.84 ± 5.59 mg vs. 40.18 ± 7.89 mg; P < 0.001) compared to the L-ESPB group [Table 2].

Table 1.

Baseline data of examined groups

	PENG group (number=30)	LESP group (number=30)	Difference (95% CI)
Age (years)	55.03±12.36	49.77±14.58	5.26 (−1.73 to 12.25)*
Sex
Male	16 (53.3%)	18 (60%)	−6.7% (−31.7 to 18.3)**
Female	14 (46.7%)	12 (40%)	6.7% (−18.3 to 31.7)**
ASA class
I	4 (13.3%)	3 (10%)
II	15 (50%)	15 (50%)
III	11 (36.7%)	12 (40%)
Operative time (minutes)	65.22±10.58	64.49±11.11	0.73 (−4.88 to 6.34)*
Affected side
Right	17 (56.7%)	15 (50%)	6.7% (−18.3 to 31.7)**
Left	13 (43.3%)	15 (50%)	−6.7% (−31.7 to 18.3)**
Type of implant
Cemented	12 (40%)	11 (36.7%)	3.3% (−21.2 to 27.8)**
Cementless	18 (60%)	19 (63.3%)	−3.3% (−27.8 to 21.2)**
Surgical approach
Posterior approach	20 (66.7%)	21 (70%)	−3.3% (−26.8 to 20.2)**
Lateral approach	10 (33.3%)	9 (30%)	3.3% (−20.2 to 26.8)**

Data were presented as mean (SD) or frequency (percentage). P-value was significant if under 0.05. ASA: American Society of Anesthesiologists; L-ESP: lumbar-erector spinae plane; PENG: pericapsular nerve group. * Mean difference (95% CI). ** Difference in proportions (95% CI)

Table 2.

Analgesic request, time to 1^st analgesia, and cumulative analgesic dose

	PENG group (number=30)	LESP group (number=30)	P
Analgesic request	24 (80%)	28 (93.3%)	0.120
Time to 1st analgesic request (hr)	12.88±4.56	8.45±3.49	<0.001
Cumulative dose of analgesia (ketorolac in mg)	25.84±5.59	40.18±7.89	<0.001

Kaplan-Meier analysis confirmed the prolonged analgesic duration in the PENG group compared to the L-ESP block (the hazard ratio = 2.41 (CI 95%), P value ≤ 0.001) during the critical early postoperative period [Figure 2].

Figure 2.

Kaplan-Meier testing for analgesic requesting

Initial pain scores at rest were comparable between the groups; however, at 24 hours, the PENG group exhibited significantly lower scores. During movement, the PENG group demonstrated significantly lower NRS scores at all time points [Figures 3 and 4]. At 6 hours postoperatively, quadriceps motor impairment was more frequent in the PENG group than in the L-ESP group (26.7% vs. 6.7%; P = 0.040). Among patients with motor impairment, most cases were classified as paresis (20.0% in the PENG group vs. 6.7% in the L-ESP group), and only two patients (6.7%) in the PENG group had complete paralysis. No cases of paralysis occurred in the L-ESP group. Importantly, all motor deficits were transient, with complete recovery (100% normal motor function) observed in both groups by 12 hours and maintained at 24 hours. Failed blocks occurred in one patient in each group (3.3% vs. 3.3%; P = 0.750). None of the patients in either group developed additional complications. The presence and severity of quadriceps motor block at each time point is presented in Table 3.

Figure 3.

Resting numeric rate scale in studied groups. PENG: pericapsular nerve group; L-ESP: lumbar- erector spinae plane

Figure 4.

Dynamic numeric rate scale in studied groups. PENG: pericapsular nerve group; L-ESP: lumbar- erector spinae plane

Table 3.

Presence of quadriceps motor block in studied groups

	Motor function	PENG group (number=30)	LESP group (number=30)	P
At 6 hours	Normal	22 (73.3%)	28 (93.3%)	0.040
	Paresis	6 (20.0%)	2 (6.7%)
	Paralysis	2 (6.7%)	0 (0%)
At 12 hours	Normal	30 (100%)	30 (100%)	—
	Paresis	0 (0%)	0 (0%)
	Paralysis	0 (0%)	0 (0%)
At 24 hours	Normal	30 (100%)	30 (100%)	—
	Paresis	0 (0%)	0 (0%)
	Paralysis	0 (0%)	0 (0%)

Discussion

This study aimed to evaluate the efficacy of the L-ESPB compared with the PENG block in reducing postoperative pain and analgesic use after hip replacement surgery. Our hypothesis was that the PENG block would provide better analgesia than L-ESPB.

The results showed that cases in the PENG group used fewer analgesics overall during the first 24 hours and took longer to request analgesics for the first time. Resting pain scores were comparable at early time points, but the PENG group had significantly lower scores at 24 hours.

Additionally, the PENG group consistently reported less discomfort during hip mobility at all assessed time points. Although transient quadriceps weakness was more frequent with PENG at 6 hours, it resolved completely by 12 hours. Both techniques were safe, with no major complications.

According to Reysner et al.,[7] PENG and ESPB both significantly reduced the need for postoperative analgesics and delayed the first rescue analgesic for a longer interval than controls (P < 0.001), with consistently lower pain scores in both groups.

Compared with conventional multimodal analgesia, another study found that the PENG block was significantly more effective at relieving pain. At every time point, the PENG group’s NRS scores were lower than those of the L-ESP block group. Simultaneously, the PENG group’s first morphine need was 9.8 hours (SD = 6.3), whereas the L-ESP group’s was 3 hours.[8]

In contrast to the lumbar erector spinae plane block, another study found that the PENG block significantly prolonged the duration of initial rescue opioid analgesia (up to 13.3 hours) and provided superior pain control.[9]

The PENG block at 24 hours after total hip replacement was associated with markedly lower opioid use, according to a meta-analysis of five RCTs comparing it with other analgesic methods, such as the L-ESPB block (95% CI: −0.64 to −0.08; normalized mean difference = −0.36).[10]

In a double-blind, randomized study, the addition of a PENG block to a standard multimodal analgesic regimen was compared with the regimen alone in 70 patients undergoing elective THA. The PENG group showed a marked reduction in pain in the recovery room during the immediate postoperative period, but no discernible difference in pain scores was observed after discharge.[11]

The enhanced analgesic efficacy of the PENG block demonstrated in our study is consistent with earlier findings and supports its ability to target the hip’s articular innervation more effectively. Unlike the lumbar erector spinae plane block, which relies on the lumbar plexus and ventral rami and receives variable amounts of local anesthetic, the PENG block is anatomically directed toward the capsular nerves. This targeted mechanism likely accounts for the prolonged analgesia and reduced analgesic requirements observed in our cohort.[6,7,11,12]

Consistent with our findings regarding quadriceps motor function, earlier research indicates that the lumbar erector spinae plane block better preserves quadriceps strength than the PENG block.[7] In that trial, a mild reduction in quadriceps strength persisted for up to 12 hours postoperatively. Interestingly, at six hours, the PENG group’s hip adduction muscle strength was considerably lower than that of the L-ESPB group (4.6 ± 0.6 vs. 5.0 ± 0.0, P = 0.001) and the control group (4.6 ± 0.6 vs. 4.9 ± 0.3, P = 0.017).

PENG block induced mild quadriceps weakness in prior investigations, findings consistent with those from other randomized trials.[13,14,15] However, compared to the fascia iliaca block or femoral nerve block, the quadriceps muscle weakness associated with the PENG block is significantly less. The PENG block appears to have a volume-dependent motor-sparing effect. While local anesthetic volumes of 15–20 mL are considered optimal for achieving effective analgesia, they may increase the risk of femoral nerve involvement. This is likely due to excessive spread of the anesthetic, potentially extending into the hip bursa and reaching the femoral nerve, thereby resulting in transient motor impairment.[16]

Our findings indicate that although the PENG block is generally motor-sparing, optimizing the technique and injectate volume may be important to minimize transient motor involvement. In our study, quadriceps weakness resolved within 12 hours and had no functional consequences. Both techniques were safe and well tolerated, with no major complications, supporting their use when performed under ultrasound guidance by experienced clinicians.[8]

Regional anesthesia may be associated with patient discomfort and anxiety during block placement. Previous studies have compared the risks of nerve damage, injection-related discomfort, and paranesthesia in patients undergoing peripheral nerve blocks under general anesthesia versus those who are awake or lightly sedated. Although paresthesia and nerve pain have been linked to potential neurological complications, such signs are more readily detected in awake patients. However, current evidence remains inconclusive, and no definitive data support a clear difference in safety between these approaches.[8,17]

There are several limitations to this study. Because it was conducted at a single tertiary facility, the results may not be as broadly applicable as they could be. In addition, outcomes were assessed only within the first 24 hours postoperatively, precluding evaluation of longer-term effects. A fixed volume and concentration of local anesthetic (20 mL of 0.25% bupivacaine) were used for both techniques, which may influence block characteristics and limit comparisons with other protocols.

Functional recovery outcomes, such as time to ambulation, participation in physiotherapy, and length of hospital stay, were not assessed despite their clinical relevance. Although no major complications were observed, the ability to detect uncommon adverse events is limited by the very small sample size, underscoring the need for larger multicenter studies.

Conclusion

In hip arthroplasty, this randomized, blinded trial showed that the PENG[2] block provides superior postoperative analgesia compared with the lumbar erector spinae plane block, with a longer time to the first analgesic request, reduced analgesic consumption, and improved pain control, particularly at 24 hours. Although transient quadriceps weakness was more frequent with PENG, it resolved within 12 hours without clinical consequences. Both techniques were relatively safe. Clinically, the PENG block may be considered a useful opioid-sparing option when performed with appropriate technique and monitoring. Longer follow-up periods, more extensive multicenter trials, functional outcomes, and optimization of dosing strategies to balance analgesia and motor preservation are needed.

Data availability

The information will be provided upon a reasonable request from the corresponding author.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.