Analgesic efficacy of triple-level erector spinae plane block versus triple-level costotransverse foramen block in patients undergoing percutaneous nephrolithotomy surgery: A randomised non-inferiority trial

Abstract

Background and Aims:

Percutaneous nephrolithotomy (PCNL), although minimally invasive, causes moderate to severe pain after surgery. In regional anaesthesia techniques, such as the erector spinae plane block (ESPB) and costotransverse foramen block (CTFB), local anaesthetic reaches the paravertebral space, focussing on the dorsal rami and intercostal nerves. The aim of this study was to assess the postoperative analgesic efficacy of triple level ultrasound guided ESPB versus triple level ultrasound guided CTFB in patients undergoing PCNL surgery under general anaesthesia. In this study, three injections of study drug at three consecutive levels (triple-level) were given, anticipating adequate craniocaudal spread.

Methods:

In this randomised, double-blind, parallel group trial, 50 adults set to undergo PCNL with general anaesthesia were randomised to two groups: Group I (ESPB, n = 25) and Group II (CTFB, n = 25). Before anaesthesia, patients received an ultrasound guided triple level block with 0.375% ropivacaine (7 mL at each of the T10, T11, and T12 levels, for a total of 21 mL) on the operative side. The statistical analyses were performed using SPSS version 23.0, employing the independent t-test, Mann–Whitney U test, paired t-test, and Chi-square test.

Results:

The median cumulative morphine use over 24 hours was 7 mg in both groups {ESPB: interquartile range (IQR) 4–11.75 mg; CTFB: IQR 3–11 mg} (P = 0.267).

Conclusion:

Postoperative analgesia provided by triple level ESPB was not inferior to triple level CTFB and thus can be used as an effective analgesic technique for PCNL surgeries.

INTRODUCTION

Percutaneous nephrolithotomy (PCNL), though a minimally invasive procedure, is frequently associated with moderate to severe post-operative pain, which is primarily visceral in origin. Afferent fibres conveying visceral pain from the kidney travel along the sympathetic pathway (T10 to L1), and the cutaneous innervation of the incision site is supplied by T10-11.

Opioids are conventionally employed for post-operative analgesia but are often limited by side effects. Ultrasound (US)-guided thoracic paravertebral block (PVB), though effective, requires significant technical expertise. To mitigate these risks, alternative approaches adjacent to the paravertebral space have been proposed.[1]

Several studies involving erector spinae plane have demonstrated that dye injected into the space also spreads to the intercostal space and efficacy of single shot erector spinae plane block (ESPB) in providing postoperative analgesia for PCNL procedures, thus validating its spread to the paravertebral space.[2,3] In contrast, the costotransverse foramen block (CTFB) and costotransverse block (CTB) involve administration of the local anaesthetic (LA) superficial to the SCTL and near the CTF, promoting direct spread into the paravertebral space (PVS).[4] The PVS does not constitute a true anatomical compartment and the SCTL does not prevent the spread of injectate, contrary to earlier assumptions.[4] In ESPB, the needle is farther away from the pleura and neural structures than in CTFB. In a magnetic resonance imaging (MRI) study in healthy volunteers, the LA spreads in the intercostal space, the paravertebral space, and the neural foramina after an ESPB and even in epidural space in some volunteers.[5] Thus, the present study aimed to evaluate the spread and analgesic efficacy of LA administered at three levels in the erector spinae plane compared to the costotransverse level.

We hypothesised that ultrasound guided triple level ESPB provides postoperative analgesia comparable to that of ultrasound guided triple level CTFB in patients undergoing PCNL under general anaesthesia (GA). The primary objective of the study is to compare the total post-operative analgesic requirement in the first 24 h. Secondary objectives include assessment of analgesia duration, numeric rating scale (NRS) scores at rest and during movement at pre-defined intervals, incidence of adverse events, and patient satisfaction scores within 24 h after surgery.

METHODS

The study was conducted in a tertiary care hospital. Institutional Ethics Committee approval (vide approval number IECPG-460/25.08.2021, dated 27-01-2022) was obtained. It was registered in the Clinical Trials Registry-India (CTRI/2021/10/037579; http://www.ctri.nic.in). It was a randomised, parallel-group, double-blinded, non-inferiority study. The study was carried out as per the principles of the Declaration of Helsinki, 2013 and Good Clinical Practice guidelines.

Patients undergoing PCNL surgery under GA, aged between 18 and 65 years, were included in this study. Written and informed consent was obtained prior to participation in the study and for the use of the data for research and educational purposes. The patients who refused to participate, had known hypersensitivity to LAs and opioids, infection at the site of the block, spinal deformities, or a previous history of spine surgeries; were on anticoagulation therapy or had untreated psychiatric illness were excluded from the study. Recruitment of the sample size was completed from February to September 2022. Those who met the inclusion criteria (n = 50) and consented to participate in the study were randomised using a computer-generated sequence (1:1 ratio) and then placed into sequentially numbered, opaque sealed envelopes (SNOSEs) by an independent statistician. On the day of surgery, an independent staff member opened the envelope, and the patient was allocated to either of the two groups. Patients were not informed about the group they were allotted to. However, the anaesthesiologist who was supposed to perform the block was informed about the allotment. He prepared the patient in the pre-operative room, where he had also performed the block. Both patients and assessor were blinded to the allotment. After the block had been performed by one of the authors, an anaesthesiologist (assessor) blinded to allotment was present in the operation theatre who documented the intra-operative findings, while the duty resident (assessor) who was also blinded to the allotment (group name or type of block was not mentioned anywhere in the proforma until assessment was completed) documented the post-operative findings up to 24 h.

Patients were taught to express pain by using an 11-point NRS and were trained about patient controlled analgesia (PCA) pump. The intervention was done prior to induction in the pre-anaesthesia room. Standard American Society of Anesthesiologists (ASA) monitors were attached, and the patient was asked to lie prone. Skin was sterilised using chlorhexidine and draped using strict aseptic precautions. Both blocks were performed at three levels. A total of 21 mL of 0.375% LA ropivacaine was used for each intervention (7 mL at each level) at T10, T11, and T12.[3,6]

Triple level ESPB was performed similar to a single shot ESPB, but the drug was administered at three consecutive levels. A linear or a curvilinear US probe was placed parallel to the vertebral axis at the level of the 10^th rib and was moved from the lateral to the medial side till the tip of T10 transverse process (TP) was visualised {Figure 1(a)}. A 22G 100 mm echogenic block needle was placed in-plane in a caudad to cephalad direction, traversing the trapezius and erector spinae, with the needle tip just above the TP and 7 ml of 0.375% ropivacaine was injected after hydrodissection. The procedure was repeated at the tips of the TPs of the T11 and T12 levels, for a total of 21 mL of study drug. Correct block placement was ensured by a linear spread of the injectate separating the erector spinae muscle from the TP, both cranially and caudally, at least two vertebral levels beyond the levels of injection on either side. For the CTFB group, a curvilinear US probe was placed in the parasagittal plane. The transducer was moved from lateral to medial while maintaining a parasagittal orientation and observing for the transition from ribs to TPs till their bases were identified [Figure 1b]. The injection was performed using a 22 G 100 mm echogenic needle inserted in-plane in a caudad to cephalad direction under real-time US guidance until the needle tip contacts the anteroinferior part of the base of the T10 TP. The correct needle tip position was confirmed by hydro-dissection to visualise the spread from posterior to anterior, following which 7 ml of 0.375% ropivacaine was injected over 1–2 minutes under real-time US guidance. The procedure was repeated to further inject 7 mL of 0.375% ropivacaine at the T11 and T12 levels, for a total of 21 mL of study drug.

Figure 1.

Image (a) Ultrasonographic image of needle trajectory for erector spinae plane block. Yellow arrows show the needle trajectory inserted to contact the tip of the transverse process. (b) Needle trajectory for the costotransverse foramen block to target the base of the transverse process. TP = Transverse process; SCTL = Superior costotransverse ligament; PVS = Paravertebral space; ESPB = Erector spinae plane block; CTFB = Costotransverse foramen block

The patient was then shifted to the operation theatre (OT), and anaesthesia was induced with intravenous (IV) fentanyl 2 µg/kg, propofol, and atracurium and intubated with an endotracheal tube. Intraoperatively, 0.5 µg/kg IV bolus of fentanyl was given if the heart rate (HR) increased by >20% of baseline, or the mean arterial pressure (MAP) increased by >20% of baseline, after ensuring an adequate plane of anaesthesia and muscle relaxation.

Patients were shifted to the postanaesthesia care unit, wherein the NRS scores were noted at 2, 4, 6, 12, and 24 h; at rest, on coughing; and on change of posture. A 20% change in NRS between two time points of an assessment was regarded as being clinically significant. Postoperative analgesia was provided with morphine IV via PCA pump (in the first 24 h), which was set to deliver a bolus of 1 mg with a lockout interval of 10 minutes and a 10 mg 4 h limit. The time to the first bolus dose was recorded as the first analgesic requirement. Duration of analgesia was calculated as the time elapsed since placement of the block till the first use of the PCA pump (rescue analgesia). Total opioid consumption at regular time intervals was recorded. Patients were observed for any adverse events, and patient satisfaction with the technique was assessed 24 h after the operation.

The primary outcome was to assess cumulative morphine consumption (PCA pump) in the first 24 h postoperatively. The secondary outcome measures were to measure the duration of analgesia (first requirement of PCA bolus/rescue analgesia); NRS scores at rest and at movement at 2, 4, 6, 12, and 24 h; adverse events {postoperative nausea and vomiting (PONV), respiratory depression and pruritus}; and patient satisfaction scores up to 24 h (using 11-point patient satisfaction score, 0, unsatisfied; 10, most satisfied).

The sample size was calculated from previous studies after TPVB intervention, which is anatomically similar to the active-control intervention in this study.[7] A non-inferiority margin of 0.25% was selected based on the effect of the active control and its associated variance, with the margin conservatively reduced to appropriately reflect the non-inferiority framework. Power analysis showed that 23 patients in each group will be required to detect non-inferiority with a significance level of α = 0.01 and a power of 1-β = 80%. To compensate for dropouts, 25 patients were included in each group. Sample size was calculated with http://clincalc.com.

Statistical analyses were performed using statistical package for the social sciences (SPSS) v23.0 software (International Business Machines, Armonk, New York, United States of America). Continuous variables such as age, height, weight, HR, blood pressure (BP), saturation, respiratory rate, time for first PCA use, and patient satisfaction score were described as mean, and variation of each observation from the mean value was represented as mean [standard deviation (SD)]. It was analysed using an independent t-test. The variables which failed to follow a normal distribution, such as cumulative morphine consumption, NRS scores for pain, and PONV scores, were analysed using the Mann–Whitney U test. Paired continuous data HR, systolic BP, and diastolic BP following normal distribution were analysed using paired T test. Categorical variables like gender and ASA physical status were described by taking percentages and were analysed using the Chi-square test. Variables with P value <0.05 were considered as statistically significant.

RESULTS

A total of 68 patients were recruited in the study; 18 patients were excluded and the remaining 50 patients (25 each group) were included and analysed [Figure 2]. The demographic details are shown in Table 1. The median cumulative morphine consumption in 24 h in Group I was 7 mg { interquartile range (IQR) 4–11.75}, range 1–21, and in Group II, it was 7 mg (IQR 3–11), range 1–21 (P = 0.267) [Figure 3, Table 2]. The mean time for the first analgesic requirement in the postoperative period/first use of PCA pump in Group I was 189.8 (SD: 80.2) {95% confidence interval (CI): 156.7, 222.9} minutes and 199.6 (SD: 79.8) (95% CI: 166.7,232.6) minutes in Group II after the end of duration of anaesthesia (P = 0.667). Mean intraoperative fentanyl consumption in Group I was observed to be significantly lower than that in Group II [139.4 (SD: 28.44) (95% CI: 127.7, 151.1) µg in Group I and 154.6 (SD: 20.91) (95% CI: 145.97, 163.23) µg in Group II (P = 0.037)]; however, it was not considered as a pre-specified outcome.

Figure 2.

CONSORT diagram. CONSORT = Consolidated Standard of Reporting Trials; n = Number; Group 1 (ESPB), Group 2 (CTFB); ESPB = Erector spinae plane block; CTFB = Costotransverse foramen block

Table 1.

Baseline demographic and pain characteristics

Parameters	Group 1 (n=25)	Group 2 (n=25)	P
Age (years)	37.64 (13.25)(32.17, 43.11)	39.2 (12.48) (34.05, 44.35)	0.67
Height (cm)	158.92 (4.59) (157.02, 160.82)	160.68 (4.12) (158.98, 162.38)	0.16
Weight (kg)	63.84 (8.84) (60.19, 67.49)	68 (8.53) (64.48, 71.52)	0.097
Gender: Male/Female	18/7	20/5	0.508
ASA Physical Status: I/II/III	21/3/1	16/7/2	0.271
Duration of block performance (min)	189.8 (80.2) (156.7, 222.9)	199.6 (79.8) (166.7,232.6)	0.667
Duration of surgery (min)	124.8 (40.32) (108.16, 141.44)	127.2 (45.23) (108.53, 145.87)	0.84
Time for First Analgesic Use (min)	189.8 (80.2) (156.7, 222.9)	199.6 (79.8) (166.7,232.6)	0.667
Total Morphine consumption in 24 h (mg)	7 mg (4 –11.75); 1-21	7 mg (3 – 11); 1-21	0.267

Data are expressed as mean (SD) (95% CI); for total morphine consumption, data are expressed as Median [(first quartile–third quartile) IQR]; Range (min-max); Group I=ESPB, Group II=CTFB, n=Number of patients; SD=Standard deviation; CI=Confidence interval; ESPB=Erector spinae plane block; CTFB=Costotransverse foramen block; IQR=Interquartile range; ASA=American Society of Anesthesiologists

Figure 3.

Cumulative morphine consumption: Box plots represent the cumulative morphine consumption in milligrams for Group 1 (triple-level ESPB) shown in blue and Group-2 (triple-level CTFB) shown in green at 2 h, 4 h, 6 h, 12 h, and 24 h, as well as the total consumption across 24 h. The central line within each box denotes the median, the box boundaries represent the interquartile range (IQR), and the whiskers indicate the range excluding outliers. Outliers are displayed as circles and stars. ESPB = Erector Spinae Plane Block; CTFB = Costotransverse Foramen Block

Table 2.

Cumulative morphine consumption and at regular intervals

PCA (mg)	Group 1 (n=25)	Group 2 (n=25)	P
Total	7 (4–11.75); 1-21	7 (3–11); 1-21	0.267
2 h	0 (0–1.75); 0-5	0 (0–2); 0-2	0.367
4 h	2 (0–3); 0-6	1 (0–2); 0-3	0.265
6 h	2 (2–3.75); 1-6	3 (1–4); 0-5	0.591
12 h	2 (1–3); 0-7	1 (1–3); 0-5	0.234
24 h	0 (0–1); 0-7	0 (0–2); 0-7	0.758

Data are expressed as median (first quartile–third quartile) (IQR); Range (min-max) for morphine consumption via PCA. Group I=ESPB, Group II=CTFB. n=Number of patients; IQR=Interquartile range; PCA=Patient controlled analgesia; ESPB=Erector spinae plane block; CTFB=Costotransverse foramen block

The median of 11-point NRS scores at rest and movement and the postoperative vitals at regular time intervals revealed no significant differences between the two groups [Table 3]. No adverse events were noted in any of the patients due to the intervention in either of the groups. The mean (SD) of the patient satisfaction score in Group I was 7.8 (SD: 1.07) (95% CI: 7.36, 8.24) out of 10, and that in Group II was 8.12 (SD: 1.09) (95% CI: 7.67, 8.57) out of 10, (P = 0.364).

Table 3.

NRS scores at rest and at movement

NRS R [Median (IQR); Range]				NRS (M) [Median (IQR); Range]
Time interval (hr)	Group 1 (n=25)	Group 2 (n=25)	P	Group 1 (n=25)	Group 2 (n=25)	P
2	2 (0.25–4); 0-9	1 (0–3); 0-7	0.136	5.5 (4–7); 1-9	4 (4–7); 0-7	0.33
4	4 (2–4.75); 0-8	3 (2–5); 0-6	0.477	6 (4–6); 0-7	4 (3–5); 2-7	0.121
6	3 (2–4); 0-7	2 (2–4); 0-5	0.252	4 (4–6); 2-7	4 (3–5); 0-7	0.479
12	2 (0–2); 0-8	1 (0–3); 0-4	0.712	3 (3–4.75); 0-7	4 (2–5); 0-5	0.594
24	0 (0–0.75); 0-6	0 (0–1); 0-5	0.846	2 (2–4); 0-7	2 (0–3); 0-7	0.409

Data are expressed as Median (IQR)] Range; Group I=ESPB, Group II=CTFB, n=Number of patients; IQR=Interquartile range; NRS R=Numeric Rating Scale at rest; NRS M=Numeric Rating Scale at movement. ESPB=Erector Spine Plane Block; CTFB=Costotransverse Foramen Block

DISCUSSION

Our study demonstrated that triple-level ESPB provided postoperative analgesia that was noninferior to triple-level CTFB in terms of opioid consumption, time to first analgesic requirement, postoperative pain scores, and patient satisfaction. However, it was observed that there was significantly lower intraoperative fentanyl consumption in the ESPB group.

In an MRI study, an ESPB block was performed with LA in ten volunteers at the thoracic seven vertebra level. The spread of LA was observed in erector spinae muscles and towards intercostal space. More than 80% volunteers have spread to PVS and neuroforamina. It confirms the hypothesis that an injectate passing through the costotransverse foramen under US guidance would result in the successful spread of to the paravertebral space. Four volunteers also had spread of LA to the epidural space. Sensory testing was performed after 30–50 minutes, which showed variable results after ESPB.[5]

The effect of single-shot ESPB was comparable to single-shot CTF despite differences in dermatomal spread.[7] In CTFB, the target for injectate lies medial to the SCTL and close to dorsal ramus and spinal nerves. It serves as conduit for spread of injectate between retro-SCTL and TPVS.[8,9] Both ESPBN and CTFB are propagated in clinical use to ease the technical difficulties and to decrease the potential associated risks associated with thoracic PVB. Better and easy identification of various structures with use of ultrasound technique and less chances of vascular injury and pneumothorax is more advantageous compared to thoracic PVB.[8,9] In a study that compared CTF block with thoracic PVB and ESPB, the authors found that CTF block provided non-inferior analgesia.[10] Moreover, the clinical efficacy of single-shot ESP block has been observed widely in PCNL surgeries.[3,4,11,12]

Due to differences in the dermatomal spread and despite injection at single level in ESPB in the aforementioned case studies, triple-level ESPB was considered the active arm of our study to compare its analgesic efficacy with triple-level CTFB. 7 mL of 0.375% ropivacaine was placed at 3 levels to achieve a total volume of 21 mL. No significant differences were observed between the groups in post-operative NRS scores or time to first analgesic requirement.

In thoracic PVB, the injectate is deposited just anterior to the superior costotransverse ligament (SCTL), whereas in CTFB, it is placed above, medial, and posterior to the SCTL. Given the anatomical proximity and lack of clinical studies on CTFB, our sample size calculation was based on studies assessing the postoperative analgesic efficacy of TPVB in PCNL.[6]

To quantify patient satisfaction, the 11-point scale offers a more precise and broader range of responses compared to the 5-point Likert scale. Although it is limited by its reliance on a single-item measure and is susceptible to ceiling effects, these effects are generally less pronounced than those observed with 4- or 5-point scales.

The CTFB is a superficial and safe technique as the needle tip need not penetrate superior costotransverse ligament and so will not be closer to the pleura, thus decreasing the chances of pleural puncture. Unlike ESPB, where injectate spread can be visualised under US, CTFB does not allow real-time imaging of spread, necessitating fluoroscopic dye studies for anatomical validation. Future cadaveric and clinical investigations are needed to determine the ideal dosing and spread characteristics of triple-level ESPB and CTFB.

CONCLUSION

Triple level ESPB offers a safety advantage as the needle trajectory is farther from the pleura and nerve roots compared to CTFB, potentially making it easier to perform. Our study has shown that triple level ESPB will have postoperative analgesic efficacy not inferior to triple level CTFB. However, studies with larger sample sizes are warranted to validate and strengthen our findings.

Presentation at conferences/CMEs and abstract publication

Abstract accepted for e-poster presentation at Koreanesthesia 2025 and for e-poster presentation at European Society of Regional Anesthesia (ESRA congress), Paris, 2023. Das N, Mohan VK, Bhoi D, Kashyap L, Seth A, Darlong V. EP050 Analgesic efficacy of ultrasound-guided triple-level erector spinae plane block versus triple-level costotransverse foramen block in patients undergoing percutaneous nephrolithotomy: a randomized, double-blind, non-inferiority trial (Abstract). Regional Anesthesia and Pain Medicine 2023;48:A67-A68.

Study data availability

De-identified data may be requested with reasonable justification from the authors (email to the corresponding author) and shall be shared after approval as per the authors’ institution’s policy.

Disclosure of use of artificial intelligence (AI)-assistive or generative tools

The AI tools or language models (LLMs) have not been utilised in the manuscript, except that software has been used for grammar corrections.

Declaration of use of permitted tools

The scales, scores, figures, and tables are self-made and not copyrighted.

Supplementary material

None.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.