Efficacy of Local Bupivacaine Injection of Postoperative Pain in Endoscopic Sinus Surgery

Authors: Wiracha Leewannapasai, Navarat Tangbumrungtham, Kanokporn Sarsitthithum, Kangsadarn Tanjararak, Boonsam Roongpuvapaht

Abstract

Background

Regional anesthesia is effective in alleviating postoperative pain and reducing the requirement for systemic pain medications. However, optimal postoperative pain management in endoscopic sinus surgery (ESS) remains challenging. This study investigated the potential pain‐reducing outcomes of general anesthesia combined with a bupivacaine injection at the sphenopalatine ganglion (SPG) and an anterior ethmoid nerve block following ESS.

Methods

A double‐blind randomized controlled trial that included 80 patients (18–75 years old) with chronic rhinosinusitis who underwent ESS was conducted. Either 0.5% bupivacaine or normal saline was injected into the anterior ethmoid nerve and SPG. The visual analog scale (VAS) pain score was assessed at 0, 1, 2, 4, 6, 8, 12, and 24 h following the surgical procedure. The timing of analgesic administration and any postoperative adverse effects of 0.5% bupivacaine were recorded.

Results

The VAS scores were significantly improved in the bupivacaine group at 1 h (95% confidence interval [CI] = 0.77–3.81, p = 0.004), 2 h (95% CI = 1.00–3.47, p = 0.001), 4 h (95% CI = 0.59–2.38, p = 0.002), and 6 h (95% CI = 0.22–2.28, p = 0.0019). Overall, the results showed a significant difference (95% CI = 0.51–1.74, p < 0.001).

Conclusions

Postoperative endoscopic SPG and anterior ethmoid nerve blocks with bupivacaine effectively reduce postoperative pain and minimize analgesic requirements after ESS.

Introduction

Chronic rhinosinusitis (CRS) is a prevalent sinonasal condition. As per the EPOS guideline of 2020, endoscopic sinus surgery (ESS) emerges as one of the most effective treatment modalities for patients who receive adequate medical interventions yet fail to attain satisfactory clinical outcomes.

ESS has improved in the last decade due to surgical techniques, instruments, and new technology. However, patients often experience mild to moderate postoperative pain, especially in the first 24 h [1, 2, 3]. Even minor postoperative discomfort can hinder recovery and prolong the hospital stay. Effective postoperative pain management is paramount for minimizing the need for analgesics, which may cause gastrointestinal, neurological, and hematological adverse effects, as well as expediting recovery and enhancing patient satisfaction [4]. Effective postoperative pain management involves local anesthesia injections and intraoperative systemic analgesics. Regional anesthesia injections offer the advantage of prolonged and enhanced pain management in various surgical specialties, including arthroplasty, spine surgery, and gynecology.

Local anesthesia can be administered via injections into the nasal cavity, sinus mucosa, or specific nerve locations. In ESS, the sphenopalatine ganglion (SPG) and anterior ethmoid nerve are the primary targets, especially with the transnasal approach [2]. Lidocaine is the most commonly used analgesic drug in ESS due to its favorable safety profile. Bupivacaine, another amide‐class anesthetic, offers a longer duration of action with an effective half‐life of up to 7 h, compared to lidocaine's shorter duration [2, 5]. Currently, there's no consensus on optimal strategies for managing postoperative pain after ESS. This study examined the effect of postoperative bupivacaine administration on reducing post‐ESS pain.

Methods

The research was conducted within the parameters of a double‐blind randomized controlled trial. Participants were recruited from May 2023 to January 2025 at the Department of Otolaryngology Head and Neck Surgery, Ramathibodi Hospital. This study was approved by the Human Research Ethics Committee, Faculty of Medicine, Ramathibodi Hospital, Mahidol University (approval ID: MURA2023/262 Ref.2024/642). All participants in the study provided written informed consent and agreed to abstain from participating in any additional research studies.

Patients

The inclusion criteria were as (1) patient age between 18 and 75 years, (2) meeting the diagnostic criteria for CRS in the EPOS 2020 guidelines and underwent bilateral ESS, and (3) an American Society of Anesthesiology physical status of I–III. The exclusion criteria were as (1) cardiovascular, pulmonary, hepatic, or neurologic conditions of clinical significance; (2) known allergies to bupivacaine, lidocaine, or epinephrine; (3) the possibility of increased anesthetic or analgesic requirements in individuals with persistent alcohol or narcotic consumption; (4) administration of non‐prescription analgesics within a 48‐h timeframe before a planned surgical procedure; (5) the inability to comprehend the questionnaires or visual analog scale (VAS) pain assessments; and (6) pregnancy (Figure 1).

Study Design

The patients were randomly assigned in a 1 ratio to two groups using block randomization. Group A received an injection of 0.5% bupivacaine, whereas Group B received NSS. The participants and the surgeon were blinded to the injection substance.

Surgical Procedure

The patients were provided with guidance on the use of the VAS to assess their pain levels, and the preoperative VAS score was recorded. The patients then underwent general anesthesia. One percent xylocaine with adrenaline (1:200,000) was injected at both the anterior ethmoid nerve and SPG area. Bilateral functional ESS was performed. After finishing ESS, 0.5% bupivacaine was injected into the SPG area and the anterior ethmoid nerve. Each injection was 2.5 mL (a total of 10 mL). The control group performed the same procedure with normal saline (NSS). Both bupivacaine and NSS have the same viscosity and transparency. The pain VAS score was assessed at 0, 1, 2, 4, 6, 8, 12, and 24 h following the surgical procedure. Patients with VAS scores > 2 received 500 mg of acetaminophen. If the VAS score remained unchanged between 2 and 6 after 30 min, an additional 500 mg was given. If the VAS score exceeded 7, a 3‐mg morphine injection was given. The time of acetaminophen and morphine administration, and postoperative adverse effects were recorded.

Outcomes Measurement

Patients' postoperative pain was evaluated using VAS scores as the primary outcome. The secondary outcomes were the time to first analgesic drug use and the adverse effects of treatment substance. The patients’ demographic data, the sinonasal outcome test‐22 (SNOT‐22), and Lund–Mackay computed tomography scores were collected.

Sample Size

The determination of the sample size was derived from a prior investigation conducted by Al‐Qudah [5]. The sample size was calculated by two independent group formulas. The sample size was determined with a 5% chance of a Type I error and a 90% chance of detecting a true difference to test the difference in means (testing for two independent means), resulting in a sample size of 38 participants in each group for a total of 72 participants. When we added a dropout of 10%, the total sample size was 80.

Statistical Analysis

Data were collected, manually assessed, and transcribed. The data analysis was conducted using Stata version 18.0 software (StataCorp). The patients’ demographics were analyzed using descriptive statistical measures. Categorical data are shown as frequency and percentage, while continuous data are shown as mean and standard deviation, median, maximum, and minimum. The t‐test was used to compare normally distributed data, and the Mann–Whitney U test was used for imbalance distributed data.

The postoperative VAS scores at various time points (1, 2, 4, 6, 8, 12, and 24 h) were analyzed using an analysis of covariance (ANCOVA) reported in mean difference with a 95% confidence interval (CI). Additionally, the overall effects of VAS scores were examined using generalized estimating equations.

We also examined the use of postoperative analgesics by using logistic regression to evaluate the association between treatment types (NSS and bupivacaine) and the analgesics administered (acetaminophen, morphine, or none). The results are presented as odds ratios (ORs), accompanied by CI and p values. Cox regression was used to investigate the relationship between the treatment type and the time until a patient required additional analgesia (rescue analgesia). The findings are reported as hazard ratios (HR), along with 95% CI and p values, which < 0.05 means a statistically significant difference.

Results

80 patients were participated in this study, with a mean age of 48.3 (SD = 16.3) in the control group and 56.4 (SD = 13.3) in the bupivacaine group. The cohort included 44 male and 36 female patients who showed a relatively balanced distribution, except for factors related to sex and weight, with p values of 0.017 and 0.028, respectively. The imbalanced baseline characteristic data were statistically adjusted in this research. Baseline demographic data and intraoperative parameters, such as the operative time and estimated blood loss, were similar in the two groups (Table 1). There were no statistically significant differences in the SNOT‐22 or Lund–Mackay computed tomography scores, which were used to evaluate the severity of CRS. Among the participants, 61 were diagnosed with CRS with nasal polyps, while 19 had CRS without nasal polyps. One postoperative complication, specifically postoperative bleeding, was documented; however, none of the patients experienced adverse effects from bupivacaine.

Postoperative pain scores as measured using the VAS at 1, 2, 4, 6, 8, 12, and 24 h (designated T1, T2, T4, T6, T8, T12, and T24) after surgery were compared between the NSS and bupivacaine groups. The bupivacaine group showed significantly less postoperative pain than the NSS group at T1 (95% CI = 0.77–3.81, p = 0.004), T2 (95% CI = 1.00–3.47, p = 0.001), T4 (95% CI = 0.59–2.38, p = 0.002), and T6 (95% CI = 0.22–2.28, p = 0.002) (Table 2 and Figure 2). Furthermore, the overall postoperative pain score at 24 h was lower in the bupivacaine group than in the NSS group (95% CI = 0.51–1.74, p < 0.001). The effect size (Cohen's d) for the overall VAS pain score was 0.45, indicating a moderate magnitude of clinical benefit.

Table 3 shows the distribution of the analgesic types used. There was no significant difference in acetaminophen use between the two groups (95% CI = 0.23–1.61), with an adjusted OR of 0.61, which suggested a slightly lower likelihood of acetaminophen consumption in the bupivacaine group than in the NSS group. Similarly, morphine use did not show a significant difference between the groups (95% CI = 0.05–1.4), with an adjusted OR of 0.26. However, the proportion of patients who did not require any analgesia in the bupivacaine group was significantly higher than that in the NSS group (95% CI = 1.04–6.85, p = 0.041). The adjusted OR of 2.67 indicated that participants in the bupivacaine group had significantly higher odds of forgoing additional analgesic treatment than those in the NSS group.

Table 4 shows the relationship between the treatment type and the time until the patients required rescue analgesia. A significant difference was found in the time to first rescue analgesia between the two groups (95% CI = 0.26–0.99, p = 0.049). The adjusted HR of 0.51 indicated that patients in the bupivacaine group had a lower risk of requiring rescue analgesia than those in the NSS group, which suggested that they took longer to require further pain relief.

The subgroup analysis in this study focused on three main the type of CRS, the packing material used, and the intraoperative analgesic drugs administered. All were prespecified based on baseline population characteristics prior to data analysis. There was no significant difference in postoperative pain between patients with CRS with nasal polyps and those without nasal polyps (Table 5). Specifically, the 95% CI for CRS with nasal polyps was 0.25–1.62 (p = 0.007), while that for CRS without nasal polyps was 0.35–1.74 (p = 0.03). Regarding nasal packing, both types, PosiSep (95% CI = 0.14–2.45, p = 0.029) and WoundClot (95% CI = 0.09–1.63, p = 0.029), showed significant differences in postoperative pain between the NSS and bupivacaine groups. Regarding intraoperative analgesic drugs, the use of Parecoxib showed a nonsignificant difference in VAS scores between the groups (95% CI = −0.45 to 1.61). However, the use of acetaminophen significantly affected the VAS scores (95% CI = 1.29–3.14, p < 0.001).

Discussion

This prospective, double‐blinded randomized controlled trial was conducted to investigate the effects of injecting 0.5% bupivacaine at the SPG and anterior ethmoid nerve area after completing ESS. The study showed that this intervention decreased postoperative pain at 1, 2, 4, and 6 h and the overall postoperative 24 h, and reduced the requirement for analgesic medication compared with the administration of NSS.

The primary cause of postoperative pain is nociceptive stimulation resulting from surgical trauma. There are two mechanisms by which surgical trauma might induce central sensitization. First, this trauma can cause the release of inflammatory mediators in the periphery, which leads to peripheral sensitization and sets off the central pathway. Second, partial peripheral nociceptive stimulation by itself may be able to activate the central route. Therefore, early and adequate postoperative pain control is critical because it prevents altered central processing of afferent input from injuries [6].

During the initial 24 h following surgery, patients who have undergone ESS may experience mild to moderate levels of pain. This pain typically reaches its peak a few hours after surgery, necessitating the administration of additional analgesic medication. These analgesics commonly include opioids and nonsteroidal anti‐inflammatory drugs. However, these medications may be associated with adverse effects on the gastrointestinal and neurological systems [3, 4]. The effective management of postoperative pain through early and proactive interventions poses a considerable challenge for surgeons.

The effectiveness of pain management is based on various factors, such as the specific anatomical site, the chosen approach for regional nerve blockade, and the administration of appropriate medications. The SPG plays a crucial role in sinonasal functions, receiving sensory input from the trigeminal and facial nerves as well as sympathetic and parasympathetic effects through various nerve connections. An anterior ethmoid block can offer additional pain relief, ad previous studies have indicated that this regional nerve block provides better pain relief than injected into nonspecific parts of the nasal mucosa [1, 2, 3, 4, 5, 7, 8, 9, 10, 11].

There are three main techniques for blocking the SPG, namely the transnasal, transoral, and lateral infratemporal approaches. The transnasal approach has shown substantial improvements in postoperative pain scores and can provide better visualization of the injection site, reduce intraoperative bleeding, and result in improved postoperative pain management compared with the transoral injection method [4]. Additionally, the transnasal endoscope technique prevents side effects, such as tooth numbness and elevate3d intraocular pressure, which may occur with the palatal SPG block technique. Consequently, this approach has become the standard technique in ESS [4].

Lidocaine is commonly used as the primary analgesic in ESS because of its safety, while bupivacaine, which has a similar profile, acts more quickly (2–10 min) and has a longer‐lasting effect (up to 7 h). The effectiveness of bupivacaine in controlling postoperative pain following ESS has been studied extensively. That's why this study prefers transnasal endoscopic bupivacaine injection at SPG after ESS to improve postoperative pain score.

Four studies evaluated the efficacy of SPG block with bupivacaine before ESS and showed that bupivacaine reduced postoperative pain compared with controls. Ismail et al. reported lower postoperative pain scores and less time to first rescue analgesia with ropivacaine than with NSS [4]. Additionally, Bhattacharyya et al. found an improvement in pain scores and a reduced time to the first analgesic dose with levobupivacaine [8]. These two studies employed a transoral approach for local analgesic injection, which differs from our studwe prefer transnasal endoscopic approach as Kim et al. [3].

In two additional studies that employed the transnasal endoscopic approach, Kesimci et al. observed that bupivaciane and levobupivacaine improved pain scores at various intervals compared with NSS but did not reduced overall pain score 24 h after administration [2]. Sethi et al. showed differences in pain relief for bupivacaine and ropivacaine at multiple time points compared with NSS but the maximum postoperative time in their study was 8 h [12].

However, Friedman et al. did not find that bupivacaine was superior to lidocaine for postoperative analgesia following ESS [1]. These authors acknowledged that their study had a limited sample size, and they noted that the bupivacaine group showed a higher baseline VAS postoperative score than the lidocaine group. Al‐Qudah et al. conducted a lidocaine block at the end of ESS and found that lidocaine administration reduced postoperative pain extensively within the initial PACU, 6 h, and overall 24 h [5].

Based on these previous studies, we used bupivacaine as an alternative because of its prompt onset of action and longer half‐life, which is approximately two to three times greater than that of lidocaine. Our study showed that patients who received bupivacaine injections had reduced VAS pain scores at 1, 2, 4, and 6 h, and overall 24 h after surgery compared with those who received NSS. The use of analgesic medications was lower in patients who received bupivacaine than in patients who received NSS. In addition to statistical significance, the moderate effect size (Cohen's d = 0.45) highlights the clinical benefit of this intervention. Cohen's d is a standardized measure that quantifies the difference between two groups relative to the variability within the data. A value of 0.45 indicates a moderate effect, meaning the treatment leads to a noticeable improvement in outcomes. Specifically, it suggests that patients receiving bupivacaine had better pain control than approximately 67% of those without it—an effect both noticeable and valuable in practice, supporting its use to enhance postoperative comfort and reduce analgesic needs.

Although there is currently no established minimal clinically important difference (MCID) for VAS scores specific to sinus surgery, Todd et al. reported that a change of approximately 13 mm (95% CI, 10–17 mm) on a 100‐mm VAS represented the MCID in patients with acute traumatic pain [11]. In our study, the adjusted mean difference at 1–6 h postoperatively achieved the MCID, and the overall postoperative pain score was 1.13 (on a 10‐point scale), corresponding to 11.3 mm on a 100‐mm VAS. This value approaches the MCID threshold described by Todd et al., suggesting that it approaches the threshold of clinically significant pain reduction.

Our results indicate that bupivacaine use significantly reduces the need for rescue analgesia and increases the likelihood of avoiding additional postoperative analgesic treatment compared with NSS. No significant difference was observed in acetaminophen consumption between groups. However, patients administered bupivacaine had higher odds of requiring no further analgesia than those in the NSS group (adjusted OR, 2.67; 95% CI, 1.04–6.85). Additionally, the time to first rescue analgesia was significantly prolonged in the bupivacaine group, with patients receiving their initial dose approximately twice as late as controls (adjusted HR, 0.51). These findings underscore the beneficial role of bupivacaine in effective postoperative pain management.

The subgroup analysis separately examined the type of CRS, nasal packing, and intraoperative analgesic drugs. We found that the bupivacaine group had significantly reduced postoperative VAS scores in patients with CRS with and without nasal polyps.

Friedman et al. found that patients who received nasal packing experienced a decrease in pain, while those who did not receive packing reported an increase in pain [1]. They found no significant differences in changes in pain scores between Vaseline gauze and Merocel packing. Our study also examined the effect of nasal packing on postoperative pain scores. We found that while the type of nasal packing did not significantly affect pain levels, the bupivacaine group experienced greater pain relief than the NSS group.

Our investigation faced challenges, especially in managing analgesic sedation during intraoperative care. The subgroup analysis showed that patients who had bupivacaine and received acetaminophen had lower postoperative VAS scores than those who received NSS. There was no difference in VAS scores with Parecoxib use between the groups.

This study reported one postoperative complication, specifically postoperative bleeding, which required cessation of procedures in the operating room. The septal branch of the sphenopalatine artery was cauterized using bipolar instruments. Importantly, no adverse effects were associated with bupivacaine administration.

Our study has some limitations, such as being a single‐center study, a lack of long‐term follow‐up, the absence of active comparators like lidocaine or ropivacaine, and an inability to control intraoperative analgesic, which was considered by the anesthesiologists. Additionally, there was limited control over the dosage of lidocaine with adrenaline (1:200,000), which was frequently administered before ESS. Strengths of the study include the blinding of surgeons using a saline placebo and the assessment of multiple outcomes, such as pain on a VAS scale and postoperative pain medication requirements, and the statistical analysis, which uses ANCOVA and generalized estimating equations that are more suitable for repeated measurement data.

Conclusion

Postoperative administration of bupivacaine through transnasal endoscopic SPG and anterior ethmoid nerve blocks significantly alleviates pain following ESS. This innovative approach further contributes to a reduced requirement for postoperative analgesics.

Funding

This work was supported by the Department of Otolaryngology, Faculty of Medicine Ramathibodi Hospital, Mahidol University.

Ethics Statement

This work received approval from the Human Research Ethics Committee, Faculty of Medicine Ramathibodi Hospital, Mahidol University, with approval ID MURA2023/262 Ref.2024/642.

Conflicts of Interest

The authors declare no conflicts of interest.