“[As] evidence accumulates, we find that the biologic plausibility of liposomal bupivacaine does not translate into clinical relevance.”

For over a century, anesthesiologists have sought to extend the utility, duration, and safety of local anesthetics, from the original topical instillation of cocaine onto the eye to more targeted injections of long-acting local anesthetics upon nerve bundles. This progression included the modification of cocaine to produce nonaddictive local anesthetics (e.g., procaine) and the addition of an amide linkage that avoids degradation by esterase. Certain innovations, like neuraxial catheters, produced dramatic results and improved the pain experience for millions of parturients. Other biologically plausible interventions failed to produce clinical benefits compared to the existing standard of care, with examples including etidocaine and interpleural catheters.

In this issue of Anesthesiology, Hussain et al. address two biologically plausible interventions intended to extend the effectiveness of local anesthetics: liposomal bupivacaine and truncal fascial plane blocks. Liposomal bupivacaine was originally described as early as 1986 as a method “… to obtain a longer duration of action or a reduction of circulating plasma level after locoregional administration …” based on pharmacokinetic release of bupivacaine from liposomal particles. Abdominal fascial plane blocks, with particular emphasis on the transversus abdominis plane block were identified as a tool to provide cutaneous analgesia without the complications and contraindications of neuraxial techniques. In brief, the authors tested the question of whether liposomal bupivacaine reduces pain scores compared to plain bupivacaine in abdominal fascial plane blocks. They assessed pain using area under the curve (AUC) with visual analog scale pain score measured in 0 to 10 cm (10 cm being the worst) and integrated from 24 to 72 h using a trapezoidal method (unit = cm*h). They set a clinical threshold of 2.0 cm*h over the 48-h period so a visual analog difference of 0.042 cm (out of 10 cm) each hour would provide clinical significance; taken another way, a consistent 1.0 cm visual analog difference over the 48-h time period equates to a 48.0 cm*h difference. They focus on the 24- to 72-h interval based on previous assertions that liposomal bupivacaine will prolong pain control over conventional bupivacaine.They identified 16 trials (15 transversus abdominis plane trials and a single quadratus lumborum trial) including 1,287 patients and found no statistically significant or clinically relevant difference in pain score AUC (–0.21 cm*h; 95% CI, –0.43 to 0.01; P = 0.058). The authors conducted sensitivity analyses accounting for trial design (e.g., adjuvants in the control group and others) and only found a statistically significant but clinically unimportant effect (–0.25 cm*h; 95% CI, –0.47 to –0.03; P = 0.031) when excluding the lone quadratus lumborum study. They went on to compare pain scores at multiple time points (1, 6, 12, 24, 48, 72 h), morphine consumption by 24-h intervals, time to first analgesic request, length of stay, opioid-related side effects, and liposomal side effects. The authors only identified a statistically significant but clinically unimportant difference in visual analog scale pain score at 24 h (–0.31 cm; 95% CI, –0.61 to –0.01; P = 0.045) but no difference between groups for any other metric. In support of the manuscript, the authors went to great lengths to obtain data from unpublished trials listed on clinicaltrials.gov. They also incorporated new statistical techniques including the Hartung-Knapp-Sidik-Jonkman random effects model, specifically to “… account for expected heterogeneity, small number of included studies, and reduce the risk of error and false-positive results” along with Doi plots and the LFK index (an abbreviation for the inventor’s name: Luis Furuya-Kanamori). These are two recently described metrics that have a higher sensitivity for assessing small study effects than funnel plots and Egger’s test.

As with previous studies, this meta-analysis exhibits several limitations including heterogeneity of operations. The analysis incorporated three studies with abdominally based reconstruction of the breast, six open abdominal trials, and seven laparoscopic or robotic trials. Transversus abdominis plane blocks only cover cutaneous abdominal pain and do not provide pain coverage for thoracic sites of surgery or visceral resection. In the case of breast reconstruction or open abdominal surgery, opioids provided the primary pain control with median 72-h oral morphine of up to 350 mg. In comparison, laparoscopic or robotic cases required median 72-h oral morphine equivalent doses of less than 45 mg. Given the difference in opioid use, the pain scores reflect less about bupivacaine and more about access to opioids. The authors investigated morphine use but only for 24-h periods and set a clinically significant threshold of 30 mg morphine: for laparoscopic surgeries with low total opioid consumption, this threshold is potentially impossible to identify. This heterogeneity brings up a point of concern that is quickly quashed when reading the individual studies: not a single published open abdominal paper provided a median reduction in 72-h opioids of more than 30 mg oral morphine equivalents.

Overall, the data suggest that liposomal bupivacaine offers no clinical superiority over conventional bupivacaine in abdominal fascial plane blocks (i.e., transversus abdominis plane blocks given the inclusion of only a single quadratus lumborum paper). This adds to a collection of high-quality meta-analyses from Abdallah’s group and others investigating the clinical effect of liposomal bupivacaine. The group previously identified a clinically unimportant decrease in 24- to 72-h pain AUC of 1.0 cm*h in peripheral nerve blocks (95% CI, 0.5 to 1.6; P = 0.003; n = 619 patients), an effect that lost significance (0.7 cm*h; 95% CI, −0.1 to 1.5; P = 0.100) after the exclusion of the lone industry-sponsored trial. Subsequently they found a null effect in surgical site infiltration (0- to 72-h AUC, 28 cm*h; 95% CI, –1.4 to 57.4 cm*h; n = 712) with a positive skew driven by early industry-sponsored trials.  Finally, in a study on periarticular infiltration (n = 1,836 patients), liposomal bupivacaine provided no reduction in morphine consumption on postoperative day 2 (0.54 mg; 95% CI, –5.09 to 6.18 mg; P = 0.85) or pain score AUC for 24 to 48 h (0.08 cm*h; 95% CI, –0.19 to 0.35; P = 0.56).  To date, Abdallah et al. have studied liposomal bupivacaine at four different locations (perineural, fascial plane, periarticular, local infiltration) and summarized data from more than 4,400 patients. The accumulated meta-analyses corroborate a lack of clinical difference between liposomal bupivacaine and plain bupivacaine.

Hence, as evidence accumulates, we find that the biologic plausibility of liposomal bupivacaine does not translate into clinical relevance. Liposomes may prolong the release of bupivacaine, but this does not change a patient’s pain experience. Despite comparable pain control, liposomal bupivacaine differs from plain bupivacaine in cost to patients and providers, with prices up to 100 times more per milligram of bupivacaine. Liposomal bupivacaine also exerts an opportunity cost on the anesthesia community, given the 77 ongoing trials listed on clinicaltrials.gov including seven transversus abdominis plane and two rectus sheath trials. In the face of multiple high-quality meta-analyses, the marginal utility of such studies is diminished, and the anesthesia community should consider less tested alternatives or a large scale pragmatic trial.

Notwithstanding the underwhelming clinical effect of liposomal bupivacaine, slow-release preparations might still play a role in the local anesthetic story. Alternative formulations exist including the Food and Drug Administration (Silver Spring, Maryland)–approved bupivacaine and meloxicam in Biochronopolymer technology (Heron Therapeutics, USA). Novel compounds may offer a bridge to prolonged pain relief with the development of nontoxic sodium channel blockers and capsaicinoids. Finally, other technologies are emerging with recent advances in cryoneurolysis and peripheral nerve stimulators along with peripheral nerve stimulators integrated into local anesthetic catheters. Overall, the future appears bright for long-acting, nonopioid pain control, but as evidence-based practitioners, we should evaluate our investment of time, money, and energy wisely to advance pain control for our patients.