Introduction
The stellate ganglion is a part of the cervical sympathetic nervous system. It's formed by the merging of the inferior cervical ganglion and the T1 ganglion, and it sits between the C6 and C7 vertebrae. It contains sympathetic preganglionic fibers that innervate the head and neck, and sympathetic postganglionic fibers that innervate the upper limbs and heart [1,2].
Stellate ganglion blocks (SGBs) are widely used in clinical practice to treat a wide range of painful and non-painful conditions, such as pain in the head, face, ear, nose, throat, mouth, neck, shoulder, and upper limbs, as well as refractory angina pectoris, arrhythmia, and depression [3–6]. In the past, SGB was often performed by using the body surface positioning method. However, in recent years, as ultrasound imaging technology matured and ultrasound equipment gained popularity in the clinic, ultrasound-guided SGB has gradually become a viable alternative for physicians [7]. Ultrasound imaging technology allows precise visualization of the SGB: under ultrasound guidance, thereby largely improving the safety and effectiveness of the technique. With ultrasound, the puncture process can be observed in real time, adjacent vital organs and blood vessels can be avoided, and the diffusion of drugs can be monitored [8,9].
While the use of ultrasonography has helped clinicians to ensure the safety of nerve blocks, gaps remain regarding the optimal doses (volume and concentration) of local anesthetic to use in nerve blocks. Determining the lowest effective dose of a local anesthetic for ultrasound-guided SGB, i.e., the dose that could minimize the adverse effects caused by a local anesthetic overdose while ensuring a good blocking effect, forms the basis for the clinically rational use of this drug treatment. Nevertheless, the minimum effective volume and concentration of local anesthetic to use for ultrasound-guided SGB have not yet been reported.
The biased-coin design (BCD), up-and-down sequential allocation method (UDM) allows determination of the optimal drug dose by minimizing the number of trials required as well as the resource consumption, and is the preferred study-design method for higher-percentile dose-effect estimation [10,11]. Furthermore, lidocaine belongs to the amide class of local anesthetics. It has a rapid onset of action, wide dispersion, strong penetration, lacks a significant vasodilator effect, and causes almost no irritation to tissues, making it one of the most commonly used local anesthetics for SGBs [12]. In this study, we aimed to determine the 90% minimum effective volume (MEV90) and concentration (MEC90) of lidocaine for use in ultrasound-guided SGBs.
Materials and Methods
This study was approved by the Medical Ethics Committee of Union Hospital of Tongji Medical College, Huazhong University of Science and Technology (No. UHCT230596) and adhered to the tenets of the Helsinki Declaration 2013. Written informed consent was obtained from all participants. The trial was registered prior to patient enrollment at https://www.chictr.org.cn (ChiCTR2300078277) on December 4, 2023. Patients were enrolled from December 4, 2023, to April 16, 2024.
Study population
Patients who underwent ultrasound-guided SGBs at the Department of Pain in Union Hospital of Tongji Medical College, Huazhong University of Science and Technology, due to head, face, or upper-limb pain, insomnia, or facial paralysis, who were aged 18–80 years, had an American Society of Anesthesiologists (ASA) class Ⅰ–Ⅲ, with a body mass index (BMI) of 18.0–27.9 kg/m2, of either sex, were included. The exclusion criteria were as follows: allergy to local anesthetics; contraindications to nerve block (coagulation disorders, sepsis, tumor at the puncture site, infection, and other related diseases); neck deformity; neck trauma or surgery with obvious scars; mental disorders or diseases of the central nervous system, or inability to cooperate; serious cardiopulmonary and other major organ insufficiencies; and pregnancy or breastfeeding.
Ultrasound-guided SGB
After entering the treatment room, intravenous access was routinely established; oxygen was administered by nasal cannula at 2 L/min; and electrocardiography, and noninvasive blood pressure and peripheral oxygen saturation (SpO2) monitoring were conducted. The ultrasound-guided SGB was administered by a pain physician familiar with ultrasound and nerve block operations (attending physicians with at least 3 years of work experience), and the operation process adhered strictly to the principles of asepsis. The patient was placed in the supine position with the head tilted slightly to the opposite side, with a thin pillow under the shoulder, and the head tilted slightly back to expose the neck fully.
A high-frequency linear array probe (Labat PG, Shenzhen Wisonic Medical Technology Co., Ltd.) was placed at approximately the level of the cricoid cartilage. This allowed localization of key anatomical landmarks for the SGB, including the common carotid artery, the internal jugular vein, the cervical transverse process and its protruding anterior and posterior nodules, as well as the long neck muscle covered by the prevertebral fascia that is located on the superficial surface of the transverse process. The ultrasound probe was then moved caudally until the anterior nodule of the C6 transverse process disappeared, the puncture path was not obstructed by bony structures, and no important neurovascular structures were present in the puncture path. After determining the skin-puncture point, applying local disinfection, and laying a sterile dressing, the operator proceeded with in-plane insertion of a 20-gauge short-beveled puncture needle. Under ultrasound guidance, the puncture needle was passed through the sternocleidomastoid muscle and the anterior scalene muscle, to the lower part of the common carotid artery, the surface of the longus colli, and the deep surface of the prevertebral fascia. The operator then injected lidocaine slowly after checking that no blood or gas was withdrawn [13,14]. After the drug injection was completed, the ultrasound probe was used in a sweeping motion to observe the diffusion of the drug, and to confirm that the drug had diffused between the longus colli and the prevertebral fascial space. The puncture needle was then removed, and the puncture point was sterilized and covered with aseptic dressing. The patient was returned to the ward if no obvious adverse reactions had occurred after 30 min of observation.
The MEV90 of lidocaine for ultrasound-guided SGB
The expert consensus for SGB therapy in China (2022) recommends using a lidocaine concentration of 0.5%–1.0% for ultrasound-guided SGBs [13]. To ensure adequate blockage of the stellate ganglion and to avoid block failures caused by a low concentration, we chose to use 1% lidocaine in the study of the MEV90. The present study used a BCD-UDM design in which the first patient was injected with 2 ml of 1% lidocaine, and the volume used for each subsequent patient depended on the block results of the previous patient, with an adjacent volume-step of 0.2 ml. If the block failed in the previous patient, the volume for the next patient was increased by 0.2 ml. However, if the block was successful, the volume received by the next patient was biased-coin randomized, with an 11% probability of a volume reduction by 0.2 ml and an 89% probability of the volume remaining unchanged. This part of the study was concluded once 45 successful block instances were achieved.
Sequential allocation in the BCD-UDM was performed using a computer-generated list of random numbers prepared in Excel (Microsoft Corp.) by a statistician. Investigator accessed the list and prepared the corresponding syringes for lidocaine in the drug-preparation room, and then transported the syringes to a treatment room. In this way, the patient, physician, and caregiver were blinded to the volume allocation.
The sign of a successful SGB was defined as the presence of Horner's syndrome on the blocked side, or a temperature increase of ≥ 1°C in the head, face, or upper limbs on the blocked side, as indicated by infrared thermography [15,16].
Finally, using R statistical software (https://www.r-project.org/), the MEV90 and 99% minimum effective volume (MEV99) of 1% lidocaine for ultrasound-guided SGBs were calculated by using centered isotonic regression. The 95% CI for the results was calculated using a bootstrap algorithm.
The MEC90 of lidocaine for ultrasound-guided SGB
Based on the results obtained with 1% lidocaine SGBs in the first part of the study, we extrapolated the MEV99 of lidocaine. This value was used to decide on the volume of lidocaine injected in the second part of the study. In the second part, we employed the same study design as described for the first part; thus, the concentration of lidocaine for the first patient was set to 0.2%. The concentration for each subsequent patient was adjusted based on the outcome of the block in the previous patient, with an adjacent concentration-step of 0.1%. If the block failed, the concentration for the next patient was increased by 0.1%. However, if the block succeeded, the concentration for the next patient was randomly selected using a biased coin, with an 11% probability of decreasing by 0.1% and an 89% probability of remaining unchanged. The study was terminated after 45 successful block instances had been achieved.
The same statistical method was used to calculate the MEC90, 99% minimum effective concentration (MEC99), and 95% CI of lidocaine for ultrasound-guided SGB.
The detailed flow of this study is depicted in Fig. 1.
Data collection and outcome assessment
General patient information, including age, sex, BMI, ASA class, and underlying diseases, were recorded. A separate investigator (a physician who did not perform the nerve block) recorded the occurrence of Horner's syndrome in patients within 10 min after SGB. Infrared thermography was used to capture the changes in skin temperature of the head, face, and upper limb on the blocked side before and 5 and 10 min after block administration, for comprehensive assessment of whether the SGB block was successful. The occurrence of adverse reactions, including hoarseness, dysphagia, upper-limb weakness on the block side, local hematoma, pneumothorax, and local anesthetic poisoning, after the block was recorded. The heart rate, SpO2, and mean arterial pressure were recorded before and 10 min after block administration.
Statistical analysis
A two-stage BCD-UDM design was used in this study. Statistically, at least 45 patients with successful blockade were require to calculate the MEV90 and MEC90 by isotonic regression [17,18]. The 95% CI of the results was calculated by using a bootstrapping algorithm, with 2000 repeated samples. Graphing was performed using GraphPad Prism 8 (GraphPad, Inc.). We used the pooled-adjacent-violators algorithm (PAVA) for isotonic regression (package source: isotone_1.1-1.tar.gz; https://CRAN.R-project.org/package=isotone) to obtained the adjusted positive rate. The PAVA has three dose estimators, of which μ3 is the most accurate [18]. The dose estimator μ3 was defined as the interpolated dose with an estimated toxicity equal to exactly 0.9 [18]. Hence, in this study, μ3 represents the dose of the drug when the target effect (0.9) was achieved. We then performed point and interval estimations via centered isotonic regression (package source: cir_2.5.0.tar.gz; https://CRAN.R-project.org/package=cir) to derive MEV99 and MEC99 and their 95% CIs from the data. Statistical analyses were performed using R v.4.3.1 (R Foundation for Statistical Computing). Continuous variables are presented as mean ± standard deviation. Categorical variables are presented as numbers and percentages.
Results
Overall, 110 patients were included in this study; 58 participated in the first part of the study (minimum effective volume) and 52 participated in the second part of the study (minimum effective concentration). The BCD sequences for lidocaine volume and concentration are shown in Fig. 2.
Effective volume of lidocaine
In the study on the effective volume of lidocaine, the mean age of the 58 patients was 52.3 years, and 67.2% of this group were female. SGBs were successful in 45 of the patients and failed in 13 patients. No significant differences were found between the successful and failed SGB groups in terms of age, sex, BMI, or ASA class. The general information of this study group is shown in Table 1.
Effective concentration of lidocaine
In the second part of the study, investigating the effective concentration of lidocaine for SGB, the results of the volumetric study were used. Fifty-two patients were included in this part of the study; in 45 of these, the SGB was successful, while it failed in seven. The mean age of the group was 46.5 years, while 67.3% were female. The descriptive statistics of the group are presented in Table 2.
The MEV90, MEV99, MEC90, and MEC99 of lidocaine for ultrasound-guided SGBs
The MEV90 of 1.0% lidocaine was 3.83 ml (95% CI [3.19–3.91]) and the MEC90 of 4.0 ml lidocaine was 0.38% (95% CI [0.32–0.41]). We further extrapolated that the MEV99 of 1.0% lidocaine is 3.97 ml (95% CI [3.95–5.29]) and that the MEC99 of 4.0 ml lidocaine is 0.47% (95% CI [0.46–2.55]).
Overall incidence of side effects after ultrasound-guided SGBs
In the 1% lidocaine ultrasound-guided SGB MEV90 study, 11 steps were used, including 2 ml, 2.2 ml, 2.4 ml, 2.6 ml, 2.8 ml, 3.0 ml, 3.2 ml, 3.4 ml, 3.6 ml, 3.8 ml, and 4.0 ml. The positive response rates for these different volumes were 0%, 0%, 0%, 0%, 0%, 0%, 66.7%, 70%, 88.9%, 87.5%, and 100%, respectively. Among the 58 patients included, four experienced hoarseness after blockage, but the symptoms disappeared within 1 h after block administration, and the volume of 1% lidocaine resulting in hoarseness in these individuals was 3.4 ml, 3.6 ml, 3.8 ml, and 4.0 ml, respectively. In the 4-ml lidocaine ultrasound-guided SGB MEC90 study, four steps were used (0.2%, 0.3%, 0.4%, and 0.5%), with corresponding positive response rates of 0%, 61.5%, 96.6%, and 100%, respectively. No adverse events occurred in any of the patients enrolled in the concentration study.
Infrared Thermography Findings
Infrared thermography demonstrated progressive temperature increases on the blocked side in successful SGB cases, as shown in Fig. 3A. In the MEV90 study (1% lidocaine volume), significant temperature elevations (P < 0.001) occurred at both 5 min and 10 min post-block across all measured sites (forehead, cheek, upper limb) among successful blocks (n = 45), with the cheek exhibiting the most pronounced change (∆T: +1.17℃ at 5 min, +1.41℃ at 10 min), as shown in Figs. 3B-D. Conversely, the failure group (n = 13) showed only marginal forehead warming at 10 min (∆T: +0.53℃, P = 0.019), which remained below the clinical success threshold (≥ 1℃). Baseline temperatures did not differ significantly between groups (P > 0.050). Similarly, in the MEC90 study (4 ml lidocaine concentration), successful blocks (n = 45) displayed significant temperature elevations (P < 0.010) at both the 5 min and 10 min time points, as illustrated in Figs. 3E-G, most notably in the forehead (∆T: +1.05℃) and cheek (∆T: +1.33℃) at 10 min. The failure group (n = 7) exhibited no statistically significant changes (∆T < 0.5℃, P > 0.050) at any site or timepoint, with comparable baseline temperatures between groups (P > 0.050).
Discussion
SGBs are widely used in clinical practice and has therapeutic value in many diseases, such as postherpetic neuralgia, insomnia, tinnitus, and facial paralysis. However, the dose of local anesthetics currently used in ultrasound-guided SGBs in clinical practice remains empirical. In the present study, the MEV90 and MEC90 of the SGB were revealed using the BCD-UDM method. The results showed that the MEV90 for 1% lidocaine in ultrasound-guided SGBs was 3.83 ml (95% CI [3.19–3.91]), while the MEC90 for 4.0 ml of lidocaine was 0.38% (95% CI [0.32–0.41]); these can be used as reference values in clinical practice. Of the 110 patients included in the present study, the SGB failed in 20 overall. Fig. 2 shows that the dose of lidocaine used in almost all of the failed cases was less than the 90% of the minimum effective dose, which reflects the reliability and trustworthiness of our results.
An SGB involves the injection of local anesthetics into the tissues around the stellate ganglion and its vicinity, including the cervical sympathetic trunk, cervical sympathetic ganglion, preganglionic and postganglionic nerves, and their innervation areas [19]. Thus, an SGB disrupts signaling of the sympathetic nerves that innervate the head, face, neck, scapula, upper limbs, chest, and back. It modulates the tension of the sympathetic nervous system and ultimately regulates the body's autonomic nervous, circulatory, endocrine, and immune systems to maintain a dynamic equilibrium [1,2,13]. It can be used to treat a variety of chronic pain conditions and non-painful diseases. SGBs are currently widely used in clinical work, particularly since the maturation of ultrasound technology, which has significantly improved the accuracy and safety of SGBs, making its application more promising [13,19]. However, in our clinical practice, we found that the optimal volume and concentration of local anesthetics used for ultrasound-guided SGBs remain empirical, with no expert consensus or guidelines providing precise, evidence-based recommendations for these parameters. In practice, patients often experience side effects, such as recurrent laryngeal nerve block, causing hoarseness; diaphragmatic nerve block, which causes difficulty breathing, due to a high volume or concentration; or block failure, due to a low volume or concentration [20]. Based on these results, the present study sought to determine the MEV90 and MEC90 for ultrasound-guided SGBs and objectively assessed the effects of the SGB by combining it with the advanced technology of infrared thermography. Our results are expected to provide guidance for clinical practice.
In pharmacology, the smallest dose of a drug that can cause pharmacological effects is called the lowest effective dose (minimum effective volume or concentration). The experimental designs commonly used to determine this dose include the fixed-dose grouping method, the up-and-down sequential method (UDM), and the BCD-UDM, among which using the UDM for MEV50 and MEC50 is more classical and accepted by most statisticians [21,22]. However, for clinicians, higher-percentile data (e.g., MEV90/MEC90 and MEV99/MEC99) are more meaningful clinically. Studies have confirmed that the extrapolation of MEV50/MEC50, derived by UDM, to MEV90/MEC90 has a high degree of error and is of limited clinical value [23,24]. The BCD-UDM is a method for estimating the effect of an intervention by introducing a coin bias, which allows for the simple and rapid random intervention assignment. By using a cyclical process of gradual dose adjustment and observation of effects, the optimal drug dose can be determined while minimizing the number of trials and resource consumption. This approach is more suitable than the conventional UDM for estimating dose-response at higher or lower percentiles (e.g., MEV90/MEC90 or MEV10/MEC10) [10,21,23].
Although the BCD-UDM design can reduce the required sample size in drug studies, a minimum of 45 successful responses (a value greater than 40 and a multiple of 9) are necessary to estimate the 90% effective dose (ED90) with sufficient precision [25,26]. To study the 95% effective dose (ED95) and 99% effective dose (ED99), the number of successful responses required would increase significantly to 95 and 495, respectively. Thus, increasing the response rate from 90% to 95% or 99% requires a substantial increase in sample size. In the typical S-shaped dose-response curve of a drug, the curve is flatter at the ends, while the steepest slope occurs at the ED50 response. The ED90 is usually near or at the plateau of the curve and is close to the ED99. Therefore, focusing on determining the 90% minimum effective dose of lidocaine required for ultrasound-guided SGBs was deemed more practical, as this provides a balance between statistical robustness and practical feasibility.
Lidocaine is a commonly used amide-type, intermediate-acting local anesthetic, with the characteristics of having a rapid onset of action, strong penetration, and wide dispersion, without causing obvious vasodilatation. Lidocaine is widely used in surface, local infiltration, epidural, and peripheral nerve block anesthesia, and is also one of the most commonly used local anesthetics in SGBs [27,28]. Nevertheless, optimization of the dosage of local anesthetics for achieving a satisfactory blockade by using the smallest possible dose warrants in-depth discussion and research. Hence, our study explored the MEV90/MEC90 of lidocaine in ultrasound-guided SGBs by using a BCD-UDM design, with a view to guiding clinical practice.
Several previous studies have explored the optimal dose of local anesthetics for SGB; however, the vast majority were randomized controlled studies using fixed gradients. Lee et al. [29] explored the minimum volume of 0.5% mepivacaine required for ultrasound-guided SGBs by setting three volume-steps (2, 3, and 4 ml), and showed that the stellate ganglion could be successfully blocked by using 2 ml of this anesthetic. Jung et al. [30] explored ropivacaine dosage by setting four volume-steps (6, 4, 3, and 2 ml) and found that the optimal volume of 0.2% ropivacaine for ultrasound-guided SGBs was 4 ml. Yoo et al. [31] found that 4 ml of 1% lidocaine for ultrasound-guided SGBs had the same effect as 6 and 8 ml, although 8 ml was associated with a higher incidence of adverse effects; these results were similar to our findings.
On the other hand, few studies have investigated the concentration of local anesthetics used for SGB, Hardy [32] found that the minimum concentration of bupivacaine required for successful SGBs was 325–406 µmol/L, but bupivacaine solution containing 5 mmol/L potassium reduced the minimum concentration to 81–162 µmol/L. In our study, the MEV99 (3.97 ml) was obtained by extrapolating the results of MEV90. In order to ensure that any SGB block failure observed was due to a low concentration of local anesthetics rather than to insufficient volume, and given the precision of the volume of the syringes used in the clinic, we chose to use 4 ml to explore the minimum concentration of lidocaine required for ultrasound-guided SGB.
The appearance of Horner syndrome on the blocked side indicates successful SGB. This manifests as pupil narrowing, eyelid ptosis, sunken eyes, little or no sweating on the ipsilateral side, conjunctival congestion, facial flushing, nasal congestion, and elevated skin temperature [33]. Notably, patients consciously perceive warmth in the head, face, and upper limbs after successful administration of an SGB. Moreover, vasodilatation can result in a 1–3℃ increase in the skin temperature of the ipsilateral upper limb; this change in temperature can be detected by infrared thermography or skin temperature detectors [13,34,35]. Toshniwal et al. [36] demonstrated an increase in the ipsilateral limb temperature after an SGB. Stevens et al. [37] found that, in most patients, a 2℃ increase in skin temperature over the contralateral limb implied complete sympathetic blockade. Usually, an average increase of 1°C–2°C in the ipsilateral limb implies a successful SGB, and the degree of skin temperature increase correlates with the pain reduction in patients; nevertheless, some studies have shown no correlation between the skin temperature changes and long-term outcomes [38,39]. In the present study, we recorded the skin temperature changes before and after SGB administration by using infrared thermography. We found that the ipsilateral frontal, facial, and upper extremities showed different degrees of skin temperature elevation after the block, with changes in the facial area being the most significant (Fig. 3). However, notably, among the 90 patients with successful block, we observed that 10 cases (11.11%) showed only elevated skin temperature (≥ 1℃), without presenting Horner's syndrome, while 5 patients (5.55%) showed manifestations of Horner's syndrome on the side of the block, without significant elevation of skin temperature (< 1℃). In this study, four patients developed hoarseness after block administration, which disappeared after half an hour and was considered to be due to recurrent laryngeal nerve block. No serious adverse events, such as local anesthetic toxicity and vascular nerve injury, occurred.
Our study had some limitations. First, the MEV90 and MEC90 in this study were limited to lidocaine, and did not include other types of local anesthetics. Second, the patient's age and BMI may influence the spatial structure of neck tissues [40]. The subjects in our study were aged 18–80 years and had a BMI of 18.0–27.9 kg/m2. Thus, the results of the study are not applicable to children or patients with obesity. Further subgroup analysis should be performed in future to explore dose differences among children, the youth, and older individuals. Third, although we measured the change in skin temperature on the side of the block by means of infrared thermography, we did not assess the relationship between the magnitude of the change in skin temperature after SGB administration and the prognosis. The purpose of our study was to determine the lowest effective dose of lidocaine for ultrasound-guided SGBs, and the underlying condition was not the same in all patients. In future, the effect of the degree of skin temperature elevation on prognosis under the same disease conditions will be a promising research focus. Finally, the primary observations in our study were changes related to Horner's syndrome and skin temperature by 10 min after SGB administration, and the time to resolution of Horner's syndrome was not tracked and recorded in detail. Therefore, our outcome metrics did not evaluate the effect of different volumes and concentrations of lidocaine used in the SGB on onset and maintenance time. Whether the initiation and maintenance time of the SGB and the severity of positive features affect its efficacy warrant further study. In addition, the morphology, innervation areas, and physiological roles of the stellate ganglion on both sides are not exactly the same; for example, the regulation of cardiac function by the bilateral stellate ganglia differ, and generally, a left SGB should be used for treating cardiac conditions. However, to date, no study has explored the dosage difference of drugs required for SGBs between the left and right sides. Thus, future studies should endeavor to refine the relevant details to ensure clinical treatment precision.
In conclusion, lidocaine can be safely used for ultrasound-guided SGB. For people aged 18–80 years with a BMI of 18.0–27.9 kg/m2 who are undergoing ultrasound-guided SGB, the MEV90 for 1% lidocaine is 3.83 ml (95% CI [3.19–3.91]), while the 4.0-ml lidocaine MEC90 for this population is 0.38% (95% CI [0.32–0.41]).
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
