Programmed Intermittent Bolus for Erector Spinae Plane Block Versus Intercostal Nerve Block With Patient-controlled Intravenous Analgesia in Video-assisted Thoracoscopic Surgery

Abstract

Objective:

Postoperative analgesia is crucial after video-assisted thoracoscopic surgery (VATS). This study was designed to investigate whether the analgesic effect of programmed intermittent bolus (PIB) erector spinae plane block (ESPB) is noninferior to that of intercostal nerve block with patient-controlled intravenous analgesia (ICNB-PCIA) for VATS.

Methods:

The study was a single-center, open labeled, randomized noninferiority trial. A total of 80 patients (American Society of Anesthesiologists I to III) undergoing elective video-assisted thoracoscopic lobectomy or bulla resection were randomly allocated to the ICNB-PCIA (n=40) or the ESPB (n=40) group using a PIB injection. The primary outcome was pain intensity at movement at 4 hours postoperatively using the Numeric Rating Scale (NRS). Secondary outcomes included pain scores at rest and movement in the recovery room, at 8, 24, and 48 hours postoperatively, perioperative analgesics, adverse effects, hospital stay, and patient satisfaction.

Results:

The mean difference in NRS scores at movement at 4 hours postoperatively between the ESPB (n=39) and the ICNB-PCIA (n=37) groups was under the noninferiority margin. NRS scores were significantly higher in the ICNB-PCIA group than the ESPB group at movement postoperatively. At rest, NRS scores were significantly elevated in the ICNB-PCIA at 4, 8, and 24 hours. The postoperative opioids consumption was decreased in the ESPB group. No difference was found in rescue analgesics, hospital stay, and patient satisfaction.

Discussion:

ESPB using a PIB injection offers noninferior analgesia to ICNB-PCIA after VATS.

Video-assisted thoracoscopic surgery (VATS) has been accepted as a treatment of choice for patients undergoing lung surgery all over the world. Although VATS causes less trauma than traditional thoracotomy, it is still associated with moderate to severe acute postoperative pain.¹ Opioid-based patient-controlled intravenous analgesia (PCIA) and systemic nonsteroidal anti-inflammatory drug or other analgesics, combined with neuraxial blocks or regional nerve blocks are used as part of multimodal analgesia. While thoracic epidural analgesia (EA)² and paravertebral blocks (PVBs)³ are considered as the “gold standard” for pain treatment after thoracotomy,⁴ they are not widely used after VATS. A survey of Canadian thoracic anesthesiologists⁵ showed that up to 78% felt that routine EA was not necessary in VATS procedures, and only 14% of the respondents used PVB. Lack of experience and fear of the risks and the technique difficulty were the main reasons for the anesthesiologist avoiding EA and PVB. Less invasive intercostal nerve block (ICNB), often performed under direct vision by surgeons, has become one of the most frequently used techniques for post-thoracoscopic analgesia.⁶ Since the duration of single-shot ICNB is limited, patients after thoracic surgery could experience “rebound pain,” defined as mechanical surgical pain that are brutally uncovered after the nerve block wear off.⁷ Therefore, opioid-based PCIA combined with ICNB regimens are used to attenuate rebound pain⁸ and lower pain scores.⁹ However, opioids can be addictive and are associated with adverse effects such as respiratory depression, nausea, and vomiting, and increase in mortality and length of hospital stay.^10,11 Focus on opioid-sparing analgesics and the development of regional anesthesia techniques have greatly expanded the options for acute pain management.

Erector spinae plane block (ESPB) was first described by Forero et al¹² in 2016. This block is clinically safe and technically easier to perform and is shown to reduce thoracic pain effectively.^13,14 However, the efficacy of programmed intermittent bolus (PIB) ESPB has not been compared with ICNB combined with PCIA in randomized controlled trial (RCT) of VATS.

We anticipated that adequate pain relief could be achieved with both PIB for ESPB and ICNB combined with PCIA in multimodal analgesia regimen after VATS. This study was designed to investigate whether the analgesic effect of PIB ESPB is noninferior to that of ICNB-PCIA for VATS.

METHODS

The present study is a prospective RCT registered at: https://clinicaltrials.gov (NCT03682354) on September 24, 2018. The experimental protocol was approved by the Institutional Ethical Committee of Peking University Third Hospital (approval number: 2018-263-01). Written informed consent was obtained from all patients involved in the study.

From September 2018 to May 2019, patients with American Society of Anesthesiologists status I to III, aged 18 to 70 years, who were scheduled for a unilateral elective VATS, including bullectomy, lobectomy, segmentectomy, and wedge resection, were screened for eligibility into the study. The exclusion criteria included coagulation disorders, infection at the injection site, known allergy to study drugs, chronic consumption of opioid or other analgesics, inability to understand pain score, and PCIA pump. On the day before surgery, the residents explained the pain Numeric Rating Scale (NRS) from 0 (no pain) to 10 (worst pain) and the use of PCIA pump to the patient.

Using a computer-generated random allocation sequence with block size of 5, 80 patients were randomly assigned into 1 of the 2 groups with 1:1 allocation ratio. Group allocation numbers were concealed in sealed, opaque envelopes by the institutional statistician and were revealed to the anesthesia team 30 minutes before the patients entered the operation room.

Perioperative Management

Intraoperative monitoring included 5-lead electrocardiography, pulse oximetry, noninvasive blood pressure, and bispectral index monitoring. Patients were premedicated with intravenous dexamethasone 5 mg and flurbiprofen axetil 50 mg. General anesthesia was induced with sufentanil and propofol, and tracheal intubation with 35- to 39-Fr left-side double-lumen tube was facilitated by rocuronium. Anesthesia was maintained with propofol and remifentanil. The infusion rate of propofol and remifentanil was adjusted to maintain the target bispectral index between 40 and 60, heart rate and blood pressure within 20% from basal measurements. One-lung ventilation was initiated with a tidal volume of 6 mL/kg and a respiration rate of 12 per minute, and these parameters were adjusted to 35 to 40 mm Hg of end-tidal carbon dioxide.

All surgeries were performed using 3 ports by one team of surgeons. A chest tube was inserted and lung expansion was confirmed directly by thoracoscopy at the end of the operation. All patients received intravenous ondansetron (4 mg) and butorphanol (1 mg) before skin closure. In the recovery room, bolus of 5 μg of sufentanil was given if NRS ≥4/10. The postoperative pain regimen in the ward included 100 mg of intravenous flurbiprofen axetil every 12 hours during the first 3 postoperative days. Oral Tylox (oxycodone 5 mg and acetaminophen 325 mg per tablet) was used as rescue analgesic for pain not effectively managed.

ICNB-PCIA Group

At the end of the operation, just before chest closure, the surgeon injected 0.5% ropivacaine into the intercostal spaces from T4 to T9, 4 mL at each interspace, under direct vision in the vicinity of the intercostal bundle, which is the standard of practice at our institution. The surgeons performing the ICNB all had >10 years of experience.

PCIA regimen was conducted with background infusion of sufentanil 2 μg/h and a patient-controlled bolus of sufentanil 1 μg with a 15-minute lockout period. The PCA pump was started after extubation and maintained for 48 hours. Nurses evaluated postoperative pain at rest and movement.

Ultrasound-guided ESPB Group

Ultrasound-guided ESPB was performed before anesthesia induction according to the method described by Forero et al.¹⁵ All blocks were performed by 2 attendings. Patients were placed in lateral position. After skin disinfection and draping, a liner probe (6 to 13 MHz, X-porte ultrasound system; Fujifilm Sonosite) was placed longitudinal at the level of the T5 spinous process and then placed laterally from the midline to the side involved in the surgery. The transverse process, trapezius, rhomboideus, and erector spinae muscles were identified. An 80-mm 22-G block needle (Stimuplex D; B Braun Melsungen AG) was inserted in-plane in the cranial-to-caudal direction until the tip contacted the T5 transverse process. After the hydrodissection with 2 to 5 mL of isotonic saline solution confirming the needle tip position, 20 mL of 0.5% ropivacaine was injected in the anatomical plane between the erector spinae muscles and transverse process. Local anesthetic spread in a fascial longitudinal pattern was visualized using ultrasound. A catheter was inserted and secured in place. In the postanesthesia care unit, patient-controlled perineural analgesia regimen was started with an hourly PIB of 0.1 mL/kg ropivacaine 0.2% plus a patient-controlled bolus of 5 mL with a 30-minute lockout period. Anesthetized dermatomes were assessed at the anterior and posterior axillary line by the cold tests using an ice pack at 24 hours after surgery.

Outcomes

The primary noninferiority outcome was postoperative pain score at 4 hours at movement after surgery, as measured by an 11-point NRS. Secondary outcome measures included postoperative pain score at rest and movement at other time points, intraoperative opioids consumption, and analgesic requirements during the first 48 hours after surgery. Postoperative pain was assessed in the postanesthesia care unit and at 4, 8, 24, and 48 hours after surgery by nurses. Secondary outcome measures also included adverse events, patient satisfaction with analgesia, and duration of hospital stay. As part of the postoperative assessments during the first 48 hours, patients were asked to report any opioid-related or block-related adverse events, such as nausea and vomiting, bleeding or hematoma, drug leakage or accidental catheter removal, infection or pain at the catheter site. At 48 hours, patients were asked to assess overall satisfaction with analgesia, as measured by an 11-point NRS (where 0=not satisfied at all and 10=extremely satisfied).

Statistical Analysis

The sample size was determined from a power analysis on the basis of noninferiority hypothesis. According to the pilot study of 10 patients, the mean NRS scores at movement 4 hours postoperatively after ESPB was 2.5, and the SD of the NRS scores at movement was 1.5 (2.0±1.5 in the ICNB-PCIA group). For noinferiority of the ESPB versus ICNB with PCIA, a maximum difference of 0.5 (margin of noninferiority) on the NRS scale was considered as acceptable.¹⁶ The required sample size for each group was thus determined to be at least 36 (α=0.025, power=0.80). Accounting for a possible 10% dropout rate, 40 participants per group were recruited.

All primary and secondary endpoints of pain scores were analyzed by an independent statistician on a per-protocol basis according to a noninferiority design. Preoperative measurements and secondary endpoints were compared using the t test or Mann-Whitney U test as appropriate for continuous variables and the χ² test for categorical variables. A P-value <0.05 was considered statistically significant. The difference of primary endpoint was compared using the t test for independent samples, with the noninferiority margin of 0.5 as the null. The pain scores over time between 2 intervention groups were compared by repeated-measures analysis of variance.

RESULTS

The study flowchart is shown in Figure 1. From September 2018 to March 2019, 98 patients were assessed for eligibility. After exclusion of 18 patients (14 patients refused and 4 patients because of chronic use of analgesics), 80 patients were randomized into 2 groups. Three patients were excluded for conversion to thoracotomy and 1 for operation canceled. Thus, the final analyses were based on 37 patients in the ICNB-PCIA group and 39 patients in the ESPB group.

FIGURE 1.

Consolidated Standards of Reporting Trials flowchart. ESPB indicates erector spinae plane block; ICNB, intercostal nerve block; PCIA, patient-controlled intravenous analgesia.

As seen in Table 1, the 2 groups were comparable in demographic and baseline clinical variables. Length of surgery did not differ between the 2 groups. NRS scores at rest and movement are shown in Figure 2. NRS scores at movement were significantly higher in the ICNB-PCIA group than the ESPB group at 4 time points (Fig. 2A): at 4 hours, the estimated mean difference was 0.4 (95% CI: −0.0 to 0.7, P=0.040); at 8 hours, the estimated mean difference was 0.5 (95% CI: 0.1-0.9, P=0.011); at 24 hours, the estimated mean difference was 0.6 (95% CI: 0.2-0.9, P<0.001); at 48 hours, the estimated mean difference was 0.8 (95% CI: 0.3-1.3, P=0.001). At rest, NRS scores were significantly elevated in the ICNB-PCIA group than the ESPB group at 3 time points (Fig. 2B): at 4 hours, the estimated mean difference was 0.3 (95% CI: 0.0-0.7, P=0.040); at 8 hours, the estimated mean difference was 0.4 (95% CI: 0.0-0.8, P=0.017); at 24 hours, the estimated mean difference was 0.6 (95% CI: 0.2-0.9, P<0.001).

TABLE 1.

Patient Characteristics

	ICNB-PCIA group (N=37)	ESPB group (N=39)	P
Age (y)	52.9±15.1	48.2±14.4	—
Male	20 (54)	14 (36)	—
BMI (kg/m2)	24.0±3.76	22.9±3.72	—
ASA I/II/III	15/21/1	8/29/2	—
Surgical procedure
Bullectomy	8 (22)	9 (23)	—
Lobectomy	4 (11)	5 (13)
Wedge resection	25 (67)	25 (64)	—
Length of surgery (min)	138±59	124±47	0.263

Values are shown as mean±SD, median (range), and n (%).

ASA indicates American Society of Anesthesiologists classification; BMI, body mass index; ESPB, erector spinae plane block; ICNB, intercostal nerve block; PCIA, patient-controlled intravenous analgesia.

FIGURE 2.

Pain course at movement (A), at rest (B). Data given as mean and 95% CI of the mean. *P<0.05; **P<0.01: comparison between the ICNB-PCIA and ESPB groups (repeated-measures analysis of variance). ESPB indicates erector spinae plane block; ICNB, intercostal nerve block; NRS, Numeric Rating Scale; PACU, postanesthesia care unit; PCIA, patient-controlled intravenous analgesia.

Noninferiority of analgesic efficacy at movement (Fig. 3A) and at rest (Fig. 3B) of ESPB to ICNB-PCIA was evaluated. The 95% CI of the median differences of the primary outcome (NRS scores at movement at 4 h postoperatively) was under 0.5, which indicates noninferiority of the ESPB group.

FIGURE 3.

Evaluation of noninferiority of analgesic efficacy of ESPB to ICNB-PCIA at movement (A), at rest (B). The difference in mean NRS scores for pain between ICNB-PCIA and ESPB (NRS values of ESPB group minus NRS values of ICNB-PCIA group) and the resulting 95% CIs are shown for the different time points. A difference in mean NRS of <0.5 point is considered noninferior. ESPB indicates erector spinae plane block; ICNB, intercostal nerve block; NRS, Numeric Rating Scale; PACU, postanesthesia care unit; PCIA, patient-controlled intravenous analgesia.

As shown in Table 2, the consumption of intraoperative opioids and postoperative rescue analgesics were comparable between both groups. The postoperative opioids consumption was significantly decreased in the ESPB group (oral morphine milligram equivalents, 343 [300, 343] vs. 43 [5, 43], P<0.001). The ESPB group was not significantly different from the ICNB-PCIA group in incidence of postoperative nausea and vomiting (2.6% vs. 8.1%, P=0.279). Two patients in the ESPB group described transient tingling inside the block catheter site when they changed to a specific body position. No local anesthetic toxicity, infection, bleeding, or hematoma was observed in both groups. No difference was found regarding the hospital stay (P=0.817) or overall satisfaction score (P=0.644).

TABLE 2.

Secondary Outcomes Following Surgery

	ICNB-PCIA group (N=37)	ESPB group (N=39)	P
Intraoperative opioids
Sufentanil (μg)	20 (15–20)	20 (15–20)	0.366
Remifentanil (mg)	1.1±0.5	0.9±0.4	0.194
Postoperative sufentanil (μg)	100 (99.5–100)	0	—
Rescue analgesics
24 h, dose (tablet)	2 (0–2)	2 (0–2)	0.721
24 h	27 (72.9)	27 (69.2)	0.719
48 h, dose (tablet)	4 (0–4)	4 (0–4)	0.828
48 h	27 (72.9)	29 (74.4)	0.891
Postoperative oral MME (mg)	343 (300, 343)	43 (5, 43)	<0.001
Adverse events
Nausea	3 (8.1)	1 (2.6)	0.279
Pain at the block catheter site	—	2 (5.1)
Hospital stay (d)	12.9±4.9	13.1±3.7	0.817
Satisfaction score	9 (8–10)	9 (8–10)	0.644

Values are shown as mean±SD, median (range), and n (%).

Oral Tylox (oxycodone 5 mg and acetaminophen 325 mg per tablet) are used as rescue analgesic.

Postoperative opioids: reported in oral MME.

ESPB indicates erector spinae plane block; ICNB, intercostal nerve block; MME, morphine milligram equivalents; PCIA, patient-controlled intravenous analgesia.

Table 3 shows frequency of sensory segments blockade to cold sensation in the ESPB group at 24 hours after surgery. The number of segments blocked to cold sensation at the posterior axillary line was more than at the anterior axillary line (5 [4, 5] and 5 [4, 5], P=0.044). The highest dermatome blocked to cold sensation 24 hours after surgery was 4 (4, 5) at the anterior axillary line and 4 (3, 4) at the posterior. The lowest dermatome blocked to cold sensation was 8 (7, 8) at both lines.

TABLE 3.

Dermatome of Blocked Segments in the Erector Spinae Plane Block Group

Dermatome of blocked segments	Anterior axillary line	Posterior axillary line
T3	4 (10.3)	12 (30.8)
T4	27 (69.2)	33 (84.6)
T5-T7	37 (94.9)	37 (94.9)
T8	21 (53.8)	22 (56.4)
T9	1 (2.6)	2 (5.1)

Values are shown as n (%).

T3 to T9 indicates thoracic.

DISCUSSION

In this study, the mean difference in NRS scores at movement at 4 hours postoperatively between the ESPB and the ICNB-PCIA groups was under the noninferiority margin, which demonstrated that the analgesic effects of a continuous ESPB were noninferior to that of ICNB combined with PCIA during the early period after VATS. Comparisons of the NRS values at movement demonstrated effective analgesia in the ESPB group during the first and second postoperative days. No difference was found in adverse effects, hospital stay, and overall patient satisfaction.

There is no consensus for a gold-standard analgesic strategy for VATS. In clinical practice, especially in an environment with heavy workload, simple and convenient analgesic methods with low risk can be more readily accepted as routine. Easily performed ICNB is limited by short duration of action of local anesthetics, which may increase the risk of rebound pain. The exact incidence of rebound pain after single-shot ICNB is unknown. Strategies to mitigate the impact of rebound pain include routine prescribing of a systemic multimodal analgesic regimen such as PCIA and oral analgesics.⁸ In the control group, ICNB-PCIA regimen was shown to provide effective and prolonged analgesia for patients. Based on Williams et al’s⁷ computations, each hour of additional femoral nerve block duration was predictive of a 0.03-unit reduction in rebound pain scores. Continuous nerve block could also attenuate rebound pain and lower pain scores. Patients of continuous ESPB group described low NRS scores postoperatively (at movement: 2.3±0.3 at 4 h, 2.5±0.2 at 8 h, 2.2±0.2 at 24 h, 1.7±0.3 at 48 h) and no patients reported rebound pain. Except for the advantage of easily placement of catheter, ESPB is reported to be associated with lower risk of blood vessel and neural damage.^17,18

A recent RCT of mini-thoracotomy indicated the ESPB could provide superior analgesia, lower perioperative analgesics, and less respiratory muscle strength impairment than ICNB,¹⁴ but the surgical approach was not routine 3 ports of VATS. In Chen et al’s¹⁹ study of VATS, no significant difference was found in the Visual Analog Scale scores between ESPB and ICNB group, and multiple-injection PVB provided superior analgesia to ICNB and single ESPB, but multiple injections could expose the patients to more risks associate with puncture. In Zhang et al’s²⁰ literature, Visual Analog Scale score in the ESPB group was higher than the PVB group after surgery. It is worth noting that single injection ESPB was conducted in all of these above studies. In Taketa et al’s¹³ literature, the analgesic effect of continuous ESPB was noninferior to that of PVB 24 hours postoperatively. Continuous ESPB may provide adequate and prolonged analgesia for patients undergoing VATS.

Focus on opioid-sparing analgesics and the development of regional anesthesia techniques have greatly expanded the options for acute pain management after thoracic surgery.²¹ Opioid-sparing effect and reduced opioid-related adverse effects were reported in ESPB of pectus excavatum surgery²² and thoracic surgery.²³ In this study, the mean difference of NRS at movement at 4, 8, and 24 hours postoperatively is 0.4, 0.5, and 0.6, statistically significant, but they are not clinically significant. Both groups seem to have had excellent analgesia as shown in Figure 2. It is worth noting that the postoperative opioid usage significantly decreased in the ESPB group. Due to the small sample size, the incidence of nausea and vomiting was not significantly different between the 2 groups. Moreover, reduction of intraoperative remifentanil was observed in the ESPB group. We consider that central sensitization of neurons could be prevented by preblocking the neuronal transmission of nociceptive pain with preoperative ESPB.

The direct evidence of nerve block is examining the anesthetized dermatomal range, but few clinical studies reported sensory changes after ESPB. In a previous study examining the sensory changes of 6 volunteers after single-shot ESPB,²⁴ only 3 volunteers reported sensory change that extending beyond the posterior chest wall to the lateral. However, in Taketa et al’s¹³ study that continuous infusion was conducted, the number of anesthetized dermatomes was 3.9 (1.6) at the anterior axillary line and 3.3 (1.2) at the midclavicular. Wider spread toward the lateral cutaneous branch region is probably attributed to the advantage of continuous infusion over single shot.²⁵ Moreover, in another study of continuous ESPB using a PIB injection,²⁶ mean (SD) dermatomes with pinprick test were 4.2 (1.6) at the anterior axillary line and 3.8 (1.1) at the midclavicular line, wider than the former continuous infusion regimen. In this study, the spread is wide, 5 (4, 5) at the anterior axillary line, which also indicates the advantage of intermittent bolus techniques. Wide dermatomal spread of sensory block after PIB infusion were reported in former literatures of ESPB.^27,28

This study has several limitations. First, because of the different connection area of patient-controlled analgesia pump, patients and nurses could not be blinded to group allocation. Second, plasma ropivacaine concentrations are not monitored, but the initial dose is not exceeding the maximum recommended dose²⁹ and continuous infusion is conducted with a low dose, and no patient report local anesthetic toxicity symptoms. Third, all blocks are performed by experienced anesthesiologists and surgeons, and other operators of different level may achieve different results.

CONCLUSIONS

PIB ESPB offers noninferior analgesia to ICNB-PCIA analgesia after VATS. These findings may provide clinicians a new method for analgesia selection without compromising efficacy.