INTRODUCTION
Regional anesthesia using brachial plexus blockades is a commonly employed technique for upper limb surgeries. When performed under ultrasonographic guidance, this approach may enhance outcomes by enabling direct visualization of nerves and surrounding anatomical structures, thereby increasing effectiveness in clinical practice [1].
The combination of local anesthetics with differing onset times and durations has been extensively studied in regional anesthesia, yielding inconsistent results [2]. Adding lidocaine to bupivacaine may shorten the onset of anesthesia but could also reduce the overall duration of the sensory block [3]. Despite the controversy in the literature and limited pharmacological and clinical evidence supporting this combination, lidocaine-bupivacaine mixtures are frequently used in various healthcare settings [4]. The goal is to balance the rapid onset of lidocaine with the prolonged duration of bupivacaine. While some studies have reported no clinical benefit from combining these agents in regional anesthesia [5], others have shown reduced onset times and extended postoperative analgesia [6].
We conducted this systematic review and meta-analysis to evaluate the clinical benefits of combining bupivacaine and lidocaine in ultrasound-guided brachial plexus blocks for upper limb surgeries. The primary outcome was the onset time of the sensory block. Secondary outcomes included the onset time of the motor block, the duration of sensory and motor blocks, and the number of procedures requiring conversion to general anesthesia during surgery.
MATERIALS AND METHODS
Eligibility criteria
This systematic review included studies that met the following criteria: 1) patients scheduled for upper limb surgeries, 2) patients receiving a brachial plexus block with bupivacaine or a mixture of lidocaine and bupivacaine in any proportion as a single injection, 3) only randomized controlled trials (RCTs) were considered, 4) studies reporting any comparable outcome of interest, and 5) blocks performed under ultrasound guidance. Studies were excluded if they lacked a relevant comparison group or if patients were under 18 years of age.
Search strategy and data extraction
The study protocol was registered and published on July 11, 2024, in the International Prospective Register of Systematic Reviews of the National Institute for Health Research, under ID [CRD42024564072]. This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [7].
We systematically searched the PubMed, EMBASE, and Cochrane Library databases for RCTs meeting the inclusion criteria, with the search conducted up to May 2025. The search was independently performed by two investigators (JMC and VSB), with disagreements resolved by a third investigator (LVB) with greater expertise. The search strategy was developed using the PICO framework: the Population included patients undergoing brachial plexus block (interscalene, supraclavicular, axillary, or infraclavicular approaches); the Intervention was a combination of lidocaine (or lignocaine) and bupivacaine; the Comparison was bupivacaine alone; and the Outcomes included any comparable outcome (e.g., onset time, duration of motor/sensory block, duration of analgesia, or conversion to general anesthesia). The search string included the terms (“brachial plexus” OR “interscalene” OR “supraclavicular” OR “axillary” OR “infraclavicular”) AND ((“lidocaine” OR “lignocaine”) AND (bupivacaine)) AND (bupivacaine). The detailed search strategy is provided in Appendix 1.
For RCTs with more than one intervention group using different dosages of local anesthetics, we selected comparison arms with “bupivacaine alone,” preferably at a concentration of 0.5%, vs. “bupivacaine with lidocaine” at concentrations of 0.5% and 2.0%, respectively, in a 1:1 ratio. If the RCT included only groups with different proportions or concentrations of bupivacaine-lidocaine mixtures, the available data were considered, provided all other eligibility criteria were strictly met.
Endpoints
The primary endpoint was the onset time of the sensory block, defined slightly differently across the selected studies but generally as the time from the end of the local anesthetic injection to the loss of touch sensation, evaluated by pinprick tests. All time-related endpoints were measured in minutes.
Secondary endpoints included: the onset time of motor block (time from the end of the local anesthetic injection to muscle weakness in brachial plexus-innervated areas or complete inability to move the arm, forearm, and hand); duration of sensory block (time from the onset of sensory block to the return of sensation or pain); duration of motor block (time from the onset of motor block to the return of movement in the arm, forearm, or hand); and the number of procedures requiring conversion to general anesthesia.
Subgroup analyses
Given the variations in local anesthetic dosages and proportions among studies, and the use of epinephrine or other adjuvants, a subgroup analysis was conducted for the primary endpoint. To analyze heterogeneity, intervention groups were divided and combined into four subgroups: 1) lidocaine dose in the intervention group < 200 mg vs. 200-300 mg; 2) bupivacaine dose in the control group ≤ 75 mg vs. ≥ 100 mg; 3) presence vs. absence of epinephrine as an adjuvant; and 4) premedication with midazolam, midazolam and fentanyl, or no premedication. These categories were chosen based on the possibility of grouping despite minor variations in each study. These small differences were not considered clinically significant.
Quality assessment and risk of bias
The quality of included RCTs was assessed using the Cochrane risk-of-bias tool for randomized trials (RoB2) [8]. Studies were categorized as having high, low, or moderate risk of bias across five domains: selection, performance, detection, attrition, and reporting bias. Publication bias was evaluated using funnel plot analysis, with estimates based on study weights.
The GRADEpro GDT software (McMaster University and Evidence Prime, https://gdt.gradepro.org/app/) [9] was used to assess the quality of evidence for each primary outcome across studies, in accordance with the Grading of Recommendation, Assessment, Development, and Evaluation (GRADE) criteria [10].
Statistical analysis
Statistical analysis for this systematic review and meta-analysis followed Cochrane Collaboration and PRISMA guidelines. Risk ratios (RRs) with 95% confidence intervals (CIs) were used to compare treatment effects for categorical and continuous outcomes. A random-effects model was applied for all outcomes. Heterogeneity was assessed using the Cochrane Q test and I² statistic, with P-values < 0.10 and I² > 50% indicating significant heterogeneity. Statistical analyses were performed using Review Manager (RevMan 5.4, Cochrane Collaboration, Denmark).
RESULTS
Study selection and characteristics
The initial search yielded 1,490 records (210 from PubMed, 970 from EMBASE, and 310 from the Cochrane Library). After removing 754 duplicates and 57 animal studies, and excluding 671 records through title and abstract screening, 65 articles were retrieved for full-text review. Additionally, citations and references of included articles were manually reviewed, but no further studies were identified. The study by Gunjiyal et al. [11] was excluded because both comparison groups used bupivacaine combined with lidocaine, albeit in different proportions. Thus, a total of 7 RCTs comprising 358 patients were included in this systematic review and meta-analysis. Fig. 1 depicts the study flow diagram.
Of the seven selected studies, three [2,3,12] utilized epinephrine 1:200,000, and one study [3] used dexamethasone 4 mg as an adjuvant to local anesthetics. The type of brachial plexus block varied across studies, including axillary [12,13], axillary or supraclavicular [14], supraclavicular only [2], and infraclavicular [3,6,15] approaches. Most studies administered premedication just before the blockade, consisting of midazolam [6,13] or a combination of midazolam and fentanyl [2,3,12,15], with dosages ranging from 1 mg to 2 mg of midazolam and 50 μg to 100 μg of fentanyl. Only one study [14] did not use any premedication.
The dosing of local anesthetics varied among the included studies. In the bupivacaine-only groups, volumes ranged from 20 ml to 35 ml, doses from 75 mg to 175 mg, and concentrations from 0.375% to 0.5%. In the bupivacaine-lidocaine combination groups, there was greater variability: injection volumes ranged from 20 ml to 35 ml, bupivacaine doses from 43.75 mg to 75 mg, and lidocaine doses from 150 mg to 300 mg. Detailed information on volume, concentration, dose, and proportions of local anesthetics used is provided in Table 1.
Pooled analysis of all studies
Compared to bupivacaine alone, the addition of lidocaine to the bupivacaine solution was not associated with a significant reduction in sensory block onset time (MD -1.81, 95% CI -3.92 to 0.29; P = 0.09; I² = 98%; Fig. 2A). Similarly, there were no significant differences between groups in motor block onset time (MD 0.02, 95% CI -2.34 to 2.39; P = 0.99; I² = 95%; Fig. 2B) or the number of procedures converted to general anesthesia (MD 1.81, 95% CI 0.39 to 8.48; P = 0.45; I² = 0%; Fig. 2C). Conversely, the duration of both sensory (MD -172.88, 95% CI -215.18 to -130.59; P < 0.001; I² = 90%; Fig. 2D) and motor blocks (MD -212.13, 95% CI -374.99 to -49.28; P = 0.01; I² = 93%; Fig. 2E) was significantly reduced in the bupivacaine-lidocaine groups, favoring bupivacaine alone.
Subgroup analyses and heterogeneity
In the subgroup analysis of the primary endpoint (sensory block onset time), no statistically significant difference was observed between groups receiving less than 200 mg of lidocaine versus those receiving 200-300 mg (test for subgroup differences, P = 0.29; I² = 10.4%; Fig. 3A). Similarly, when analyzing bupivacaine dosage in the combination groups, no difference was found between those receiving 75 mg or less and those receiving 100 mg or more (test for subgroup differences, P = 0.39; I² = 0%; Fig. 3B). The presence of epinephrine 1:200,000 (test for subgroup differences, P = 0.73; I² = 0%; Fig. 3C) and the use of premedication with midazolam or a midazolam-fentanyl combination (test for subgroup differences, P = 0.56; I² = 0%; Fig. 3D) were also not associated with significant subgroup differences.
Meta-regression analysis to explore endpoint heterogeneity was not feasible, as fewer than ten studies were included. In line with the existing literature, the relatively small number of RCTs also precluded the use of Egger’s test for assessing funnel plot asymmetry.
Quality assessment and risk of bias
Individual RCT appraisal is presented in Table 2. Of the seven included studies, four were judged to have a low risk of bias according to RoB2. Two studies [6,15] were classified as having some concerns. Study [6] had potential bias related to deviations from intended interventions and outcome measurement, while study [15] showed potential bias in the selection of reported results. One study [14] was assessed as having a high risk of bias due to concerns related to the randomization process.
Tables 3, 4 present the GRADE evidence profiles for continuous outcomes, including onset and duration of sensory and motor block, which were rated as very low in quality. The quality of evidence for the categorical outcome (conversion to general anesthesia) was rated as low. These lower ratings were primarily due to methodological limitations, small sample sizes, and CIs crossing the null value.
Visual inspection of the funnel plots suggested potential publication bias for the outcomes of depression, global functioning, and pain catastrophizing. Similarly, possible publication bias was noted for sensory and motor block onset times and durations (Supplementary Fig. 1).
DISCUSSION
This systematic review and meta-analysis demonstrated no clinical benefits from adding lidocaine to bupivacaine solutions for brachial plexus blocks. Furthermore, to the best of our knowledge, this is the first meta-analysis to compare bupivacaine alone with a combination of lidocaine and bupivacaine specifically for ultrasound-guided brachial plexus blocks. This study focused on the effects of combining lidocaine and bupivacaine for ultrasound-guided brachial plexus blocks to provide a more precise and internally valid assessment. While the inclusion of all types of peripheral nerve blocks could have broadened the scope of this study, we deliberately chose to focus on upper limb blocks to reduce heterogeneity and ensure a more robust comparison. Peripheral nerve blocks vary widely in terms of technique, anatomy, and patient populations, which could have introduced significant variability if multiple block types were analyzed together. By narrowing our focus, we aimed to provide specific and clinically actionable insights into the practice of mixing lidocaine with bupivacaine in a clearly defined setting.
No single local anesthetic offers both rapid onset and prolonged duration [16]. Combining lidocaine (faster onset, pKa ≈ 7) with bupivacaine (longer duration, pKa ≈ 8.1) aims to leverage their complementary properties [17], despite lacking manufacturer endorsement [18] and pharmacological support [4]. The onset of local anesthetics depends on pKa, which governs the balance between ionized and non-ionized forms [5]. Lidocaine’s lower pKa enables faster tissue diffusion compared to bupivacaine [19]. However, our findings show no significant difference in sensory block onset time between bupivacaine alone and lidocaine-bupivacaine mixtures. This may result from a reduced bupivacaine concentration (e.g., halved from 0.5% to 0.25% in 1:1 mixtures), which lowers its sodium channel affinity and efficacy [19]. Dilution effects likely also shortened the durations of sensory and motor blocks by approximately 3 and 3.5 h, respectively, consistent with prior studies [5,11]. Only one study [15], rated as having “some concerns” by the RoB2 tool, reported longer block durations with the mixture.
The lack of clinical benefit from combining lidocaine with bupivacaine may largely be attributed to the dilution effect on bupivacaine’s effective concentration and dose. In the combination groups, bupivacaine’s concentration was typically halved (e.g., from 0.5 to 0.25% in a 1:1 mixture), reducing its pharmacological efficacy, particularly its prolonged duration of action. For example, studies using 75 mg or less of bupivacaine in the mixture consistently reported shorter sensory and motor block durations compared to bupivacaine alone at 100-175 mg [6,18]. This dilution likely diminished bupivacaine’s high affinity for sodium channels, a key factor in its long-acting profile [19]. While a formal meta-regression was not feasible due to the small number of studies, this qualitative analysis underscores the critical role of maintaining an adequate bupivacaine dose to achieve the desired block longevity, supporting the preference for single-agent bupivacaine in brachial plexus blocks.
Occasionally, failure to achieve a high-quality blockade may necessitate conversion from regional anesthesia to general anesthesia. This change in the planned anesthetic approach can be particularly problematic if the patient has conditions such as a predictably difficult airway. Additionally, the choice of anesthetic modality can influence surgical outcomes. For instance, a retrospective study comparing anesthesia methods for arteriovenous fistula surgery found that selecting general anesthesia over regional techniques may result in higher failure rates after 2 months [20]. In orthopedic procedures, such as shoulder arthroplasties, the use of regional instead of general anesthesia has been associated with safer outcomes, although no clear superiority of regional anesthesia was observed regarding pulmonary complications and length of hospital stay [21]. Furthermore, Kalthoff et al. [22], in a meta-analysis involving 851 patients undergoing arthroscopic rotator cuff repair, concluded that nerve blocks are highly effective in attenuating postoperative pain. However, our results indicate that adding lidocaine to bupivacaine in blockade solutions for upper limb surgeries did not reduce the rates of conversion to general anesthesia compared to using plain bupivacaine alone.
Mixing lidocaine and bupivacaine introduces potential safety concerns, including neurotoxicity, local anesthetic systemic toxicity (LAST), and solution stability. Zhao et al. [23] investigated intrathecal lidocaine neurotoxicity in rats. They observed that mixtures of lidocaine and bupivacaine at a 1:1 ratio could be safely used, although the combination of lidocaine with ropivacaine increased the risk of neurotoxicity. Other studies have highlighted that the neurotoxicity of local anesthetic mixtures may be influenced by concentration, pH, and the specific agents used, with some mixtures exacerbating neurotoxic potential through additive or synergistic mechanisms [4]. Furthermore, the use of off-label adjuvants, such as clonidine, dexmedetomidine, or corticosteroids, in regional anesthesia has raised concerns due to reports of increased neurotoxicity when combined with local anesthetics, particularly at higher concentrations or with prolonged exposure [24]. One of the potential complications when using local anesthetics for regional blocks is LAST, as overdoses can lead to life-threatening situations [4]. Mixing different local anesthetics can make it challenging to calculate safe dosages [25]. It is recommended to adhere to the maximum dosage of each combined local anesthetic by not exceeding 100% of the combined maximum dose [18]. None of the studies included in this meta-analysis found an association between the combination of bupivacaine and lidocaine and an increased risk of developing LAST, but clinicians must remain vigilant when using such mixtures, particularly in patients with comorbidities that may predispose them to toxicity.
We performed a subgroup analysis of the primary outcome to investigate trends in the onset time of sensory block based on local anesthetic doses, the use of epinephrine as an adjuvant, and the presence of premedication before blockade procedures. As noted in Table 1, the dosages of lidocaine and bupivacaine varied significantly. A 2014 review by Eng et al. [26] identified anesthetic dose as the most critical determinant of the onset and duration of peripheral nerve blockades, rather than concentration or volume. Therefore, we divided the results into two categories: less than 200 mg of lidocaine [14] and dosages between 200 mg and 300 mg of lidocaine [6,12,13,15]. However, the difference between these subgroups was not statistically significant. We also hypothesized that different bupivacaine doses, specifically less than 75 mg [6,13,14] and 100 mg or more [12,15], might influence onset times for sensory block, suggesting a possible dose-response relationship. However, the subgroup analysis did not reveal a significant association between dose and onset time. Epinephrine, which prolongs block duration through vasoconstriction and receptor effects, did not affect onset time—possibly due to inconsistent dosing or overriding pharmacokinetic factors [27]. Premedication with fentanyl or midazolam also showed no effect [28,29].
Several additional aspects related to the mixing of local anesthetics warrant discussion. First, the off-label use of such combinations raises concerns about potential drug preparation errors, contamination, and infection risk, particularly when adjuvants are added. Solution stability is another critical consideration. Watkins et al. [30] reported crystallization when ropivacaine and dexamethasone were mixed, raising questions about the physical and chemical stability of local anesthetic mixtures. Although no studies in our meta-analysis explicitly reported issues with solution stability, this remains an area requiring further investigation to ensure patient safety. Healthcare providers must be aware of these risks and follow strict protocols during drug preparation to minimize potential complications.
Despite the historical popularity of combining lidocaine with bupivacaine in clinical practice, the results of this systematic review raise important concerns regarding the clinical appropriateness of this approach. From a pharmacological standpoint, mixing lidocaine and bupivacaine predictably dilutes the more potent, long-acting agent—bupivacaine—without offering clear advantages in onset time, as demonstrated by our findings. Moreover, the addition of lidocaine introduces potential safety concerns, including neurotoxicity, systemic toxicity, and solution instability, as previously discussed.
Although the practice of mixing local anesthetics persists in some institutions—particularly where rapid turnover is prioritized—our findings suggest that this strategy lacks strong evidence of clinical benefit and may even compromise the desired anesthetic effect. Therefore, while lidocaine-bupivacaine mixtures may still be used under certain circumstances, such as specific institutional protocols or individualized patient considerations, the overall clinical appropriateness of this practice appears limited when assessed against current evidence. These findings emphasize the importance of adhering to evidence-based practices in regional anesthesia and support the use of single-agent formulations, particularly when aiming for consistent block quality and patient safety.
This meta-analysis has several limitations that may have contributed to the observed heterogeneity. Although outcome definitions across the source studies were similar, minor variations in continuous outcome measures may have introduced variability. Standardizing these outcomes is essential for more accurate comparisons in regional anesthesia research. Additionally, endpoints such as onset and duration of blockade are inherently subjective, as they often rely on patient-reported outcomes [3]. Moreover, aside from conversion to general anesthesia, most outcomes—including the primary outcome of sensory block onset time—are vulnerable to publication bias, as indicated by funnel plot asymmetry (Supplementary Fig. 1). Potential sources of this bias include selection bias, the file drawer effect, and selective reporting. The limited number of RCTs available for inclusion may also have influenced the results.
This systematic review and meta-analysis did not find evidence of a clinical advantage in adding lidocaine to bupivacaine solutions for brachial plexus blocks. Although combining these local anesthetics is a common practice intended to leverage lidocaine’s rapid onset and bupivacaine’s prolonged duration, our results did not show significant improvements in the onset times of sensory or motor blocks with this strategy. Additionally, the combination may reduce block duration, potentially impacting patient satisfaction.
It is important to note that the overall certainty of evidence supporting these findings is very low, as assessed by the GRADE methodology. Therefore, these conclusions should be interpreted with caution. Further high-quality RCTs are needed to better clarify the potential advantages, risks, and pharmacodynamic interactions of different local anesthetic combinations in brachial plexus blockade. Addressing these gaps through future research may help optimize anesthetic techniques and improve patient outcomes in regional anesthesia.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
