Secondary Logo

Journal Logo

Pain and Analgesic Mechanisms: Research Report

Inhibition by Local Bupivacaine-Releasing Microspheres of Acute Postoperative Pain from Hairy Skin Incision

Ohri, Rachit PhD*; Wang, Jeffrey Chi-Fei MD; Blaskovich, Phillip D. MS*; Pham, Lan N. MSc*; Costa, Daniel S. BS*; Nichols, Gary A. PhD*; Hildebrand, William P. PhD*; Scarborough, Nelson L. PhD*; Herman, Clifford J. PhD*; Strichartz, Gary R. PhD

Author Information
doi: 10.1213/ANE.0b013e3182a00851
  • Free

Acute postoperative pain is physiologically debilitating, emotionally stressful, and a major factor in delaying hospital discharge. In addition, its intensity is correlated positively with the development of chronic postoperative pain.1,2 Traditional local anesthetics, such as bupivacaine, can provide anesthesia over the surgical field for 8 to 12 hours,3–5 but hyperalgesia often sets in once the anesthesia regresses, suggesting that the degree or the duration of analgesia was inadequate to eliminate the postoperative afferent hypersensitivity or its critical input to the central nervous system (CNS).6,7

Acute postoperative pain has been treated by a variety of methods, including epidural catheters delivering local anesthetics or opiates, systemic opiates, and catheters placed in the incision at closing.8,9 Control of acute pain by these methods is often adequate but requires the insertion of the various needles, catheters, and delivery from external pumps, and with opiates is accompanied by nausea, pruritus, and drowsiness. In previous work from this laboratory, we have shown that a simple skin incision plus suture on the rat’s back causes mechanical hypersensitivity for 4 to 5 days which could be largely obtunded by the subcutaneous injection of systemically distributed bupivacaine before the incision.10 Although small animals, like rats, are not very susceptible to the well-known cardiotoxic actions of bupivacaine,11 the drug levels causing such a systemic action would likely be lethal to humans.12

We, therefore, sought a means to suppress postoperative pain by the delivery of local anesthetic, restricted to the surgical site, over a period of several days. Local delivery of local anesthetics by a variety of slow-release formulations can provide prolonged blockade while reducing the circulating levels of the drug, and previous publications report promising results from this approach.13–19 In the past, bupivacaine has been successfully encapsulated in poly-lactide-co-glycolide (PLGA) polymers to yield microsphere formulations that proved effective for prolonged peripheral nerve block in animal models13,19 and were subsequently used in human experimental studies.20–23 However, the inclusion of anti-inflammatory drugs such as dexamethasone were necessary in these formulations to produce a nerve blocking action longer than a day.24,25 For the studies reported here, in contrast, we used a proprietary formulation of microspheres containing only bupivacaine as the active drug. We previously demonstrated that this formulation, when injected around the innervating sciatic nerve, blocks motor and sensory functions and suppresses postincisional pain from a paw incision.16 In the studies reported here, this formulation was injected directly into the surgical field on the back of the rat, and the results analyzed in terms of the duration and the dose dependence of its pain-mitigating actions.

To more closely resemble the situation of clinical surgery, where untreated postoperative pain can persist for many days,10,26 we have developed a novel surgical model in which the skin and subcutaneous fat were dissected away from the deep fascia and underlying muscle by about 0.5 cm using blunt dissection. This tissue manipulation, here called “skin incision plus extension,” resulted in a mechanical hypersensitivity adjacent to the incision that was detected at the first few hours after surgery and persisted for at least 2 weeks. The aim of this study was to show that subcutaneous delivery of the bupivacaine-releasing microspheres, injected at the surgical site before the procedure, resulted in a profound, long-lasting, and dose-dependent suppression of the postoperative mechanical pain.


Preparation of Microspheres

Bupivacaine microsphere (MS-Bupi) was prepared using a modified oil-in-water (o/w) emulsion/solvent evaporation technique. Methylene chloride (Spectrum Chemicals, New Brunswick, NJ) was the solvent used for the bupivacaine (from BASF, Chicago, IL). The encapsulating PLGA polymer (75:25 PLGA Lactel® polymer) was obtained from Durect Corp. (Pelham, AL). PLGA was the choice for the encapsulating polymer based on the large number of formulation options offered by the PLGA polymer family (with differing lactide:glycolide ratios) and its successful track record in a wide variety of controlled-release systems.27–29 Polyvinyl alcohol was used as the emulsifying agent. The target percent drug loading was 60% (w/w), and the typical loading achieved was in the range of 55% to 60%. After wet sieving, microspheres of the size fraction in the range of 45 to 105 μm were used (mean particle size was in the range of 65–75 μm). The in vitro release kinetics fit well with the Higuchi model of diffusion-based controlled-release systems.30 The release rate is near constant for the first 12 hours. By approximately 30 hours, 50% of the total releasable bupivacaine is delivered, and by approximately 72 hours, 75% of the total releasable bupivacaine has been delivered16 (data shown in Ref. 16). At each time point for which release data were collected, the first derivative of the cumulative release profile was generated by averaging the slope of the release curve in comparison with the preceding and the subsequent time points.

Microsphere Characterization Using Raman Spectroscopy

Raman maps (chemical images) were acquired for cross-sectioned MS-Bupi using a Renishaw inVia Raman Microscope. The cross-sectioned samples were adhered to a glass slide with carbon adhesive tabs for these experiments. Raman maps were collected using a 785 nm source (50% power, 1–30 seconds accumulation), 1200 L/mm grating (840–1900 cm−1), and either a 1.4 µm (50×) or 0.7 µm (100×) step size. Raman spectra of the individual components (PLGA and bupivacaine) were also collected (785 nm, 1200 L/mm, 4000 cm−1). These component spectra were used to generate chemical images via spectral correlation (Renishaw, Gloucestershire, UK).

Microsphere Characterization Using Gas Physisorption

Gas physisorption—surface area measurements were obtained using a Micromeritics ASAP 2020 Gas Physisorption Analyzer (Micrometrics, Norcross, GA). Samples were degassed under vacuum at 25°C for 4 hours before analysis. An 11-point isotherm was collected from P/Po = 0.03 to 0.30 using Krypton gas at approximately 77°K. The resultant Brunauer-Emmett-Teller surface area was calculated using the accompanying ASAP 2020 software. True density was evaluated via helium pycnometer (Ultrapycnometer 1000, Quantachrome, Boynton Beach, FL). The samples (microspheres) were loaded into the pycnometer cell, subjected to a 5-minute helium purge pretreatment, and analyzed.


All procedures using animals have been approved by the Harvard Committee on Animals and are consonant with the international guidelines for the use of laboratory animals. Male Sprague-Dawley rats (Charles River, Wilmington, MA), weighing 275 to 300 g, were handled for 5 to 7 days for familiarization with the test site,31 the containing apparatus, and the experimenter (JC-FW). During the 2 to 3 days before surgery, each individual rat’s response to the test protocol (see below) was measured and averaged to give the baseline, preoperative response level.


A modification of the method of Duarte et al.10 was used, wherein under barbiturate anesthesia (sodium pentobarbital, Sigma-Aldrich, St. Louis, MO; 60 mg/kg, intraperitoneally), a 1.2-cm incision was made through skin (no. 10 surgical blade) that had been clipped 24 hours before and had been swabbed twice with alcohol plus povidone–iodine solution (Clinidine®; Clinipad Corp., Rocky Hill, CT) and draped with a sterile towel. After the incision, the skin and underlying fat were dissected approximately 0.5 cm away from the underlying fascia with the blunt end of closed tissue scissors, using the manipulation, skin incision plus extension. The incision was then closed with two 3-0 silk sutures (Look®; Surgical Specialties Corp, Reading, PA). Test substances were injected subcutaneously at the center of the skin location where the incision would later be made and at preoperative times dependent on the drug (see Drug Delivery). After surgery, the rats were placed individually in recovery cages, and behavioral testing begun 4 hours after surgery when rats appeared to be alert and responsive.

Drug Delivery

Subcutaneous injections of 0.4 mL volumes, delivered from a 1-mL tuberculin syringe through a 25-gauge, 16-mm needle (Becton Dickinson and Co., Franklin Lakes, NJ), inserted to its full length, were made in lightly sevoflurane (2%)-anesthetized rats before the incision. Either MS-Bupi containing 5, 10, 20, or 40 mg bupivacaine-free base (BASF) or no drug (MS-Placebo), the microsphere vehicle control, were injected 2 hours before surgery using thin-walled needles. A 0.75% (w/v) solution of carboxymethyl cellulose in phosphate-buffered saline was used as the suspending medium for the microspheres (both drug containing and placebos). Alternatively, a solution of 0.5% bupivacaine HCl (2 mg in 0.4 mL; Sigma, St. Louis, MO) was injected 1 hour before surgery, as was its vehicle control, sterile-buffered saline (Hospira, Inc., Lake Forest, IL).

Dose Accuracy

A suspension of MS-Bupi that contained 40 mg in 0.4 mL injectate was diluted sequentially 1:1 in the carboxymethyl cellulose solution (see above) thrice so that the same injected volume would contain 20, 10, and 5 mg drug. The range of doses was chosen because the effective dose for optimal suppression of postincisional pain needed to be determined and cannot be assessed from the in vitro release kinetics alone. After the subcutaneous injections, the syringes used individually for each animal were carefully rinsed with acetonitrile, after which the bupivacaine concentration in this acetonitrile rinse was determined by high-performance liquid chromatography using methodology described in our previous study.16 The bupivacaine content in the acetonitrile rinse was subtracted from the total bupivacaine dose in the syringe to estimate the actual bupivacaine dose delivered for that individual animal. The syringe contained doses of bupivacaine of 40, 20, 10, and 5 mg per animal that were, respectively, adjusted to be 32.9 ± 1.5 mg (mean ± SD), 15.0 ± 1.1 mg, 7.3 ± 0.3 mg, and 3.1 ± 1.4 mg. However, this determination of actual doses delivered is based on a conservative approach, since we did not specifically account for the dead-space volume in the Luer lock of the syringe–needle system used. Given that dead-space volume is usually about 20% of actual delivered volume (as indicated by the syringe markings), we estimate that our true delivered doses were at 95%+ of the hypothetical ones. In this article, the hypothetical, intended, rather than the actual, delivered doses are referred to, although we believe these differ by <5%.

Effect Location

To establish that any effects resulted from a local action, at the injection site, rather than a systemic action due to drug absorbed by the local vasculature and systemically distributed, injections were made at a contralateral site opposite the side of the incision (ipsilateral site), at the same time before the surgery. Testing was then conducted at and around the incision site, comparing effects from ipsilateral and contralateral drug injections (see Behavioral Testing).

Behavioral Testing

One hundred sixty rats were used in this study. Eight animals that were included in each cohort were used to test each drug formulation’s effectiveness. Anesthesia of intact skin was tested with 1 cohort for bupivacaine HCl and another for MS-Bupi (40 mg). For each of the 4 doses of MS-Bupi tested, 4 cohorts were used, 2 for ipsilateral (relative to the incision) drug or placebo and 2 for contralateral drug or placebo (drug-free microspheres). Of the 8 rats in each cohort, 4 were tested at 1 session, 2 sessions per cohort/group. Testing time for every individual rat was about 15 minutes at each session, which assessed responsiveness on a particular pre- or postoperative day.

For all procedures, the hair over the skin to be tested was clipped with electric shears 24 hours before testing; local irritation produced by this clipping resolved overnight. Behavioral testing occurred when the rat was restrained in a loose-fitting plastic device that prevented it from escaping but allowed twisting or turning. To establish the maximum possible duration of anesthesia from the test materials, subcutaneous MS-Bupi containing nominal “40 mg” local anesthetic base was injected in intact skin, under transient, light sevoflurane anesthesia (approximately 0.5%–1%). The skin was then probed with nylon monofilaments (von Frey hairs, VFHs), as described below, and also with a 19-gauge hypodermic needle affixed to the end of a 26 g VFH. Mechanical stimulation by VFH with a bending force of 4 g was used to assess tactile allodynia, and stimulation by a VFH with a force of 15 g was used to assess tactile hyperalgesia. The hypodermic needle on the even stiffer 26 g VFH produced a sensation on the human volar forearm that was unanimously considered painful to 4 incidental human subjects and gave a rapid and strong withdrawal reaction plus squealing when applied to the shaved back of rats. Reactions to innocuous stroking and to relatively thin diameter VFHs were intentionally extinguished by repeated handling, as previously reported.31

Each VFH was pressed sequentially against the skin until it bent, 6 times in each test period, at a frequency of about once per 5 seconds. “Responsiveness” was scored by the number of times in 1 session that the rat responded immediately with a robust contraction of the subdermal muscles (the cutaneous trunci muscle reflex [CTMR]) under the test VFH location, per 6 probes on the skin, N/6, for example, a “normalized” responsiveness of 0.5 would result from 3 such contractions during the 6 test probings. This CTMR has been identified as mediated by cutaneous nociceptive afferents,32 and its increase after surgery is morphine-sensitive, indicating the involvement of pain pathways.10 Muscle contractions lasted only 1 to 2 seconds in intact animals, before surgery, but after incision and blunt extension these contractions sometimes continued for 5 to 6 seconds, so the time between skin probes was lengthened to allow the muscles to fully relax before the next application. Possible changes in responsiveness that developed over the course of a full sequence of 6 probes were not evident.

Responsiveness was assessed on the area of the skin closely adjacent to an incision injury (primary hypersensitivity; probed at 0.5 cm from the wound) and also at 1 and 2 cm from the incision, as well as on the contralateral side, 2 cm away (secondary hypersensitivity). Previous publications trace the anatomy of this innervation, showing that peripheral innervation of the skin and the underlying muscles is unilateral32 such that changes measured contralateral to an injury site cannot result directly from injury of nerves in the tested region. Such secondary contralateral changes are considered to arise from changes in the sensitivity of pathways in the spinal cord and brain that result from the incision injury-induced afferent input, so called “central sensitization.”33

Data Analysis

The results, describing behavioral responsiveness to VFH or pinprick stimulation that indicates mechanical sensitivity, are given as number of responses per each 6 stimuli. The values from each test group (usually with n = 8 rats per group) are graphed as mean ± SEM. Mixed model analyses, which incorporate the main effects for group and time, a group–time interaction term, and autoregressive variance–covariance AR(1) structure to account for correlation among observations, were used to compare treatment groups in terms of their pattern of change from baseline. Differences in mean changes between assessment/responses at baseline and follow-ups at day 1, day 2, and up to day 14, or at 1 hour, 3 hours, and until 48 hours were evaluated individually for the postoperative assessment of drug effects on hypersensitivity. For the measures of local anesthesia/analgesia in intact skin produced by subcutaneous injections of bupivacaine (BUP) solution or MS-Bupi, responses were compared with their baseline, preinjection values. Difference values with P < 0.05 were considered significant.

The differences between all treatments and the appropriate vehicle/placebo group were estimated using the LSMEANS statement from the mixed model. The area under curve (AUC) data between different treatments and MS-Placebo were compared with multigroup analysis of variance analysis with Bonferroni corrections. Shapiro–Wilk and Bartlett tests were used to check the normality and homogeneity of variance of the AUC data, respectively. Residual plots and Q-Q plots of residuals were also plotted to ensure our data meet the assumptions of analysis of variance. The areas were determined by the trapezoidal approximation using GraphPad software (GraphPad, LaJolla, CA). Almost all of the comparisons among treatments after postoperative day (POD) 5 are nonsignificant, and since the focus of this study is on acute postoperative pain, the AUC analysis was limited to the first 5 days.


Microsphere Characterization

Further characterization of the MS-Bupi, beyond that described previously, is shown here by the first derivative of the previously published in vitro cumulative release profile (Ref. 16, Fig. 1). The release rate is expressed in 2 different units, on the y-axes (%/h and mg/h). The milligrams per hour values were calculated based on the actual total dose delivered per injection for the MS-Bupi, which was an average of 34.0 ± 1.0 mg (SD) per injection. The early rate of release varies from 0.6 to 0.75 mg/h for the first 7 to 13 hours.

Figure 1
Figure 1:
The first-order derivative curve (expressed both in %/h and mg/h) for the in vitro kinetic drug release profile of the 60% (w/w) bupivacaine-loaded 75:25 poly-lactide-co-glycolide microspheres. Error bars are based on standard error mean of n = 5. The original representation of the data as % cumulative release versus time was included in an earlier publication.16

A scanning electron micrograph (FEI Quanta 600FEG, FEI, Hillsboro, OR) of the bupivacaine-loaded microspheres is shown in Figure 2A, revealing the uniformly structured matrix within the microsphere. Raman microscopy chemical images, acquired for cross-sectioned microspheres, indicate a uniform distribution of both bupivacaine and PLGA throughout the matrix (Fig. 2B). From the gas physisorption isotherms (not shown), a Brunauer-Emmett-Teller surface area of 0.0834 ± 0.0019 (SD) m2/g was calculated for the MS-Bupi. For comparison, the theoretical surface area for 70 µm diameter microspheres with an assumed smooth surface would be 0.0750 m2/g (based on the measured true density of 1.138 ± 0.0009 [SD] g/mL), a value not significantly below the measured surface area.

Figure 2
Figure 2:
A cross-section of 60% (w/w) bupivacaine-loaded poly-lactide-co-glycolide (PLGA) microsphere examined via (A) scanning electron microscopy and (B) Raman spectroscopy. The chemical mapping via Raman shows a uniform distribution of bupivacaine and PLGA. The PLGA and bupivacaine is identified by the colors cyan and magenta, respectively.

Prolonged Local Anesthesia of Intact Skin by Subcutaneous MS-Bupi

Before the actions on postincisional hypersensitivity after back skin incisions were tested, the time course of bupivacaine’s local anesthetic effect in intact skin was profiled. The hypothesis for these experiments is that anesthesia from the slow-release formulation will outlast that from standard solutions of bupivacaine. Microsphere-encapsulated bupivacaine or bupivacaine 0.5% solution was injected subcutaneously, and the skin’s mechanical sensitivity tested over the following 2 days or 9 hours, respectively, until responses recovered to their baseline, preinjection levels. For each test force and distance, the responsiveness at each time was compared with the preinjection baseline. Anesthesia by the BUP solution, tested at 0.5 and 1 cm distances, shows a significant group–time effect. Anesthesia had a rapid onset and recovery; at the first test time, 30 minutes after the injection, the skin 0.5 cm from the injection site, and under the wheal raised by the injected solution, was unresponsive and remained so for 1 hour (Fig. 3A). For the 15 g force and for the 26-gauge needle (pinprick), the responsiveness differed significantly from baseline for 3 and 2 hours, respectively; no significant anesthesia was detected by the weak, 4 g force, probably due to the limited dynamic range of the weak baseline responsiveness. A similar pattern occurred at 1 cm from the injection on the ipsilateral side, with anesthesia to the 15 g force lasting 3 hours and that to pinprick lasting 2 hours (Fig. 3B). However, at a 2 cm test distance, ipsilateral or contralateral (not shown) to the injection, locations beyond the region of the wheal where the bulk solution had spread, no change in response was detectable (Fig. 3C). Again, the 4 g VFH gave such a weak baseline response that significant local anesthesia was impossible to assess.

Figure 3
Figure 3:
Local anesthesia from 0.4 mL of 0.5% bupivacaine (Bupi) injected subcutaneously and measured at distances of (A) 0.5 cm, (B) 1 cm, and (C) 2 cm from the center of the injection. Responses of cutaneous trunci muscle contractions were elicited by 6 probings each with a 4 g von Frey filament (squares), a 15 g von Frey filament (circles), or a 19-gauge hypodermic needle mounted on a 26 g von Frey filament (triangles). Differences between the preinjection, baseline responsiveness, and those after injection were analyzed by mixed effect modeling (see Methods), for each stimulus mode:*P < 0.0001; +P < 0.002.

In contrast to the brief local anesthesia from the 0.5% BUP solution, cutaneous anesthesia from microsphere-released bupivacaine was much slower to resolve. Group–time analysis showed significant effects for both the 15 g VFH and for pinprick stimulation. Responsiveness measured near the injection site by the 15 g VFH was significantly below baseline from 30 minutes until 6 hours after injection and decreased to its minimum value, 15% to 25% of preinjection values, at 3 hours (Fig. 4). By comparison, when pinprick was used at this distance, responsiveness remained significantly below baseline from 30 minutes to 36 hours after the injection (Fig. 4A). Tests with a 4 g VFH showed significant responsiveness reduction at 1 and 3 hours. The course of anesthesia at the 1 cm distance was nearly identical to this pattern (Fig. 4B), with the 15 g responsiveness reduced for 6 hours and the pinprick reduced for 24 hours, while the responsiveness to 4 g was lower than baseline only at 1 and 3 hours. At the 2 cm distance, beyond the area of skin where the bulk of injected suspension had spread, responsiveness was unchanged except to the 15 g force, where it increased continuously above baseline from 9 to 48 hours after injection (Fig. 4C). We have no explanation for this anomaly. Other than this single result, it generally appears from the decline of effect with distance that the bupivacaine that was contained and slowly released from microspheres was restricted in its subcutaneous spread. Thus, local anesthesia in intact skin lasts longer after an injection of the bupivacaine-releasing microspheres than after injection of this drug in solution.

Figure 4
Figure 4:
Local anesthesia in intact skin after the injection of bupivacaine microspheres loaded with 40 mg drug. Responses were measured using the same stimuli as for liquid bupivacaine (cf. Fig. 1), probing the skin with von Frey filaments (VFHs) of 4 g (squares), 15 g (circles) and a 19-gauge hypodermic needle mounted on a 26 g VFH (triangles). Differences between the preinjection, baseline responsiveness and those after injection were analyzed by mixed effect modeling (see Methods), for each stimulus mode: *P < 0.0001;+P < 0.002, #P < 0.05. Complete loss of response was not achieved for the 2 stronger stimuli, unlike the effect of 0.5% bupivacaine (Fig. 3).

The Long-Time Course of Postoperative Mechano-Hypersensitivity

Untreated, control postoperative mechano-hypersensitivity, measured as the increase over baseline in responsiveness to VFH stimulation, persisted for the entire 14 days of testing (Figs. 5 and 6). Postoperative allodynia was identical when tested in uninjected skin (data not shown) and in skin injected with drug-free microspheres (MS-Placebo) or saline (see below), appearing at the locus nearest the wound as a 5-fold increased response frequency to the 4 g VFH, from approximately 0.08 (0.5 responses/6 stimuli) at baseline to a maximum of approximately 0.4(2.8 responses/6 stimuli) by 48 hours (Fig. 5A). Mechano-allodynia, so tested at all 3 distances (Fig. 5, A–C), remained at a constant level, significantly above the preoperative baseline, for the remainder of the 14-day postoperative testing period. The same incision also resulted in an increase in the response to the 15 g VFH (mechano-hyperalgesia), from a baseline of 3.5 of 6 to the maximum value of 6 of 6, where it also remained, significantly increased, for 14 days (Fig. 6). There was a significant group–time interaction for tests with both the 4 and 15 g VFHs, and the overall time course of hyperalgesia was the same at all test distances. This profile of mechano-hypersensitivity was the same in skin that was not preinjected with any substances (data not shown) and is substantially longer lasting than the 5- to 7-day long mechano-hypersensitivity that follows incision of the skin, with or without incision of the underlying muscle but without the blunt dissection procedure.10,26

Figure 5
Figure 5:
Suppression of postoperative tactile allodynia by microsphere bupivacaine is a local phenomenon. Allodynia appears shortly after the incision + blunt extension surgery and reaches a maximum value by 1 to 2 postoperative days. This is shown by the change in response after surgery, after drug-free microsphere (MS-Placebo) injection on either the side, ipsilateral (open circles) or contralateral (open squares) to the incision. Local anesthesia and a reduction of postoperative allodynia occur after ipsilateral injection of microsphere bupivacaine (MS-Bupi, closed circles) but not after contralateral injection of bupivacaine microspheres (MS-Bupi; closed squares). A, Tests close to the wound show the local anesthesia as a reduction below baseline of the evoked responsiveness. B, The reduction in response is just sufficient to delay the development of allodynia for several days when tested at 1 cm from the incision for ipsilateral MS-Bupi. C, Secondary allodynia, tested at 2 cm from the incision, is only delayed by a few hours by the ipsilateral MS-Bupi, while the effects of contralateral MS-Bupi are indistinguishable from the drug-free control (MS-Placebo). Significant differences determined by mixed effect modeling, *P < 0.0001; #P < 0.05.
Figure 6
Figure 6:
Suppression of postoperative mechano-hyperalgesia is exclusively accomplished by local microsphere bupivacaine. Whether measured next to (A) or 1 cm away (B) from the incision, tactile hyperalgesia is only reduced by subcutaneous bupivacaine microspheres (MS-Bupi) injected on the ipsilateral side (closed circles). Microsphere bupivacaine injected on the contralateral side (closed squares) has the same noneffect as the drug-free control microspheres, injected at the same locus (open circles) or on the side contralateral to the incision (open squares). C, Smaller antihyperalgesic actions are detected for mechano-sensitivity at 2 cm from the incision site after ipsilateral MS-Bupi injection. Significant differences determined by mixed effect modeling, *P < 0.0001;#P < 0.05. MS-Placebo = drug-free microspheres.

Reduction of Postincisional Pain by MS-Bupi Is due to Local Anesthesia

Subsequent experiments tested the hypothesis that the prolonged anesthesia/analgesia from the slow-release formulation of MS-Bupi effectively reduces postoperative pain (mechano-hypersensitivity) and more so than the brief block by solution BUP. First, the possibility of a systemic versus a local action of microsphere-released bupivacaine was investigated by comparing the effects of a local, ipsilateral versus a distant, contralateral injection of MS-Bupi. Ipsilateral injection of MS-Bupi (40 mg drug) followed by incision and extension surgery resulted in a transient hyporesponsiveness that had returned to baseline after 1 to 2 days and by POD 5 converged to the hypersensitivity observed for MS-Placebo–injected skin (Fig. 6, A and B). When compared with the responsiveness of MS-Placebo–injected skin, there was a significantly lower allodynic responsiveness of the skin injected ipsilaterally with MS-Bupi up to POD 14 (with the single exception of POD 11) when tested at 0.5 cm (Fig. 5A), and until POD 9 (with the single exception of POD 7) when tested at 1 cm (Fig. 5B). In contrast, at no time was there a significant reduction of primary postincisional mechano-allodynia (Fig. 5) or mechano-hyperalgesia (Fig. 6) from contralaterally injected MS-Bupi. This ineffectiveness of MS-Bupi injected at a distant location from the incision is in accord with the hypothesis that any antihypersensitivity is due to a local and not a systemic action.

When compared with the baseline responsiveness at 4 g VFH, the skin tested at 0.5 cm reached a significantly elevated level at POD 9, that tested at 1 cm at POD 7, and that tested at 2 cm at POD 2. These delays in the time of appearance of allodynia were shorter at increasing distance from the injection site, mirroring the distance dependence of the allodynic reduction when compared with the other conditions, as noted previously.

The overall pattern of effect appears to depend on the test stimulus. Although mechano-allodynia near the wound remained suppressed by ipsilateral MS-Bupi throughout the 14-day testing period (Fig. 5A), when compared with MS-Placebo–injected rats, mechano-hyperalgesia at this same location was significantly reduced by the drug only up to POD 3 (Fig. 6A). These were also the times when the responsiveness to 15 g VFH increased significantly above baseline, that is., when hyperalgesia first appeared.

Bupivacaine Microspheres Suppress Postincisional Hypersensitivity in a Dose-Dependent Manner

The responsiveness of postincisional skin preinjected with different doses of MS-Bupi was each compared with the postincisional responsiveness after bupivacaine-free microspheres, using the mixed model group–time interaction analysis (see Methods). Such comparisons were used to designate those postoperative times during which mechano-hyperalgesia was significantly suppressed by MS-Bupi at the different doses. The longest times for such suppressions, equivalent to the duration of postoperative antihypersensitivity, are presented in Table 1.

Table 1
Table 1:
Duration of Antiallodynia—Latest Postoperative Day for Continuous Significant Suppression of Allodynia

Preinjection with the MS-Bupi at the 2 higher doses suppressed primary allodynia significantly for up to 2 days by 40 mg and for 1 day by 20 mg, when compared with MS-Placebo (Fig. 7A, Table 1). The antihyperalgesic actions of MS-Bupi lasted longer than the antiallodynic actions by about 1 day (compare Tables 1 and 2). Microsphere bupivacaine doses were generally less effective, for allodynia and hyperalgesia, at distances further than 1 cm from the incision/injection site (Figs. 7, C and D and 8, C and D; Table 1). The 2 lower doses of MS-Bupi, containing 10 and 5 mg anesthetic, were mostly ineffective at all locations and times, except for the 10 mg dose that suppressed postoperative hyperalgesia for 2 days (Fig. 7, Table 1). Injections of bupivacaine HCl solutions were also largely without effect, except for 1 day of antihyperalgesic action tested at 1 cm distance (Table 2).

Figure 7
Figure 7:
Dose dependence of the antiallodynic actions of microsphere bupivacaine. Preoperative subcutaneous injections of microspheres containing the designated dose of bupivacaine produce local unresponsiveness (cf. Fig. 2) and reduce the response increase (A) next to the incision and (B) at 1 cm distance, but only for the 40 mg (closed squares) and 20 mg (closed circles) doses. An insignificant reduction occurs with the 10 mg (closed triangles) and 5 mg doses. Injection of drug-free microspheres (MS-placebo, open squares) or saline (open triangles) are also ineffective. Preoperative injection of the 0.5% bupivacaine solution (left-pointing closed triangles), although it produces complete loss of responsiveness (cf, Fig. 1), does not affect postoperative allodynia. Secondary tactile hyperalgesia, measured at 2 cm on the ipsilateral (C) or the contralateral (D) side, is not changed by any of the treatments. Significant differences determined by mixed effect modeling, *P < 0.0001; #P < 0.05
Figure 8
Figure 8:
Dose dependence of the antihyperalgesic actions of microsphere bupivacaine. Subcutaneous injections of microspheres containing the designated dose of bupivacaine, given 2 hours before the incision + blunt extension, reduces the response increase (A) next to the incision and (B) at 1 cm distance, but only for the 40 mg (closed squares) and 20 mg (closed circles) doses. An insignificant reduction in hyperalgesia occurs from the 10 mg dose, although this affected brief local anesthesia (closed triangles), but the 5 mg dose is without any discernible effect. Injection of drug-free microspheres (MS-placebo, open squares) has the same effect as saline injection (open triangles), whereas 0.5% bupivacaine solution (left-pointing closed triangles) results in a very transient but complete loss of responsiveness, yet has no effect of postoperative hyperalgesia. None of the treatments, including the highest dose of bupivacaine, affect secondary tactile hyperalgesia measured at 2 cm on the ipsilateral (C) or the contralateral (D) side. Significant differences determined by mixed effect modeling, *P < 0.0001; #P < 0.05.
Table 2
Table 2:
Duration of Antihyperalgesia—Latest Postoperative Day for Continuous Significant Suppression of Hyperalgesia

The integrated suppression of allodynia and hyperalgesia was quantitated by AUC calculations from the response versus time curves over the first 5 PODs. Table 3 lists the raw data, in units of response × time product, and also the “normalized” data when the area in a drug-treated group is divided by the area when drug-free microspheres were injected (rightmost column) before surgery. Microspheres with 40 mg bupivacaine reduced the primary allodynia’s AUC (tested at 0.5 cm) by 78% ± 12%, and that at the 1 cm distance by 60% ± 14%. The 20 mg dose reduced the AUC at 0.5 and 1 cm distance by approximately 50%, but no significant reductions occurred at the further distances. Hyperalgesia was less suppressed, but still significantly, by 64% ± 11% at 0.5 and 1 cm and by <20% (P > 0.05) at test distances of 2 cm ipsilateral and contralateral. The 0.5% BUP solution reduced the AUC by only 10% to 15% of the saline control values (P > 0.05), for all test distances.

Table 3
Table 3:
Area Under the Curve of Normalized Response Versus Time for Postoperative Allodynia and Hyperalgesia


The long-term motivation behind the work reported here is the development of an injectable local anesthetic formulation that could provide acute postoperative pain relief for the first 1 to 2 days after surgery. For this first step, we have used a modified model of postsurgical pain in rats that provides at least 2 weeks of mechanical hypersensitivity and have tested the effects of a slow-release bupivacaine formulation over a range of doses to define the optimal dosing conditions. Inasmuch as the release characteristics are critical for an effective steady state of drug with minimal possibility of systemic accumulation, we have reported further analysis of the release kinetics, beyond an earlier report,16 and have also included a physicochemical characterization of the drug-loaded microspheres.

The results clearly show that the higher doses of bupivacaine-containing microspheres used here are able to suppress postincisional mechano-hypersensitivity for 2 to 3 days. These effects appear to be local, since injection of the microsphere-releasing bupivacaine at a distant, contralateral site did not reduce hypersensitivity at the incision locus, unlike the previously reported actions of BUP solutions in a similar model.10 Preoperative injections of 0.5% BUP solutions at the incision site gave much briefer periods of antihypersensitivity, never lasting >1 day.

What is the relationship of postincisional mechano-hypersensitivity in this model to the presence of pain? Postoperative changes in mechanical hypersensitivity have been confirmed as evidence for increased pain by their transient suppression by systemic morphine (2.5 mg/kg, intraperitoneally),10 analogous to the presence of CTMR in swine in response to noxious heating and its reduction by systemic opiates.34 A similar increase in local response of muscles to tactile stimulation in human neonates and infants after surgery has also been interpreted as an index of increased pain.35 In the current studies, the initial handling of the rats led to an extinction of the CTMR to non-noxious tactile stimuli, and this, taken together with the aforementioned observations, supports the interpretation of the surgery-induced responses as heightened pain.

Although many publications have reported antinociceptive actions from extended-release formulations of local anesthetics,14,17,19–24,36,37 few have described the use of these formulations for preventing postoperative pain.38 Recent work in our laboratory has shown that lidocaine released from a bone wax–like putty produced a sciatic nerve block able to suppress the pain from incision of the innervated paw15 and that a similar effect was possible when applying the present microsphere formulation to the same situation.16 However, direct delivery of slow-release formulations to the area of surgical incision has rarely been reported, although a recent report of liposomal bupivacaine for posthemorrhoidectomy is promising for pain reduction.18

Certain aspects of postoperative pain induced by skin incision plus blunt extension, such as tactile allodynia, were suppressed by bupivacaine-releasing microspheres for a week or longer, although the antinociceptive actions of this formulation, delivered identically into intact skin, had fully regressed within 24 to 72 hours. The discrepancy between these 2 phenomena can be explained by the observations that early afferent fiber discharge,39 and perhaps other acute effects (see below) after surgical tissue injury that occur during the period of bupivacaine release by microspheres, are critical for the development of long-lasting hyperalgesia.40 An alternative, but not totally separate, explanation is that the bupivacaine-sensitive processes that are activated in response to tissue injury and that contribute to postincisional hypersensitivity, such as the sensitization of primary afferent fibers per se, are unrelated to the mechanisms of impulse conduction that underlies nociception in intact skin or nerve.41,42

We also observed a difference between the duration of antiallodynic (approximately 1 day) and antihyperalgesic (2–3 days) actions of MS-Bupi. This discrepancy might have resulted from the difference in local anesthetic sensitivity of mechano-receptive A-δ and C-fiber afferents43 that are respectively activated by weak and strong forces44,45 and afferents that are sensitized by skin and skin plus deep tissue incision.46

What mechanisms might account for the antihyperalgesic actions of microsphere-delivered bupivacaine? One likely mechanism is a blockade of impulses in afferent fibers that innervate the region around the incised and extended skin. Local anesthetics, including bupivacaine, are primarily known for this physiological activity. However, other molecular and cellular mechanisms of local anesthetics may also come into play, including the blockade of various receptors for peripheral substances that contribute to injury-induced pain47–49 and the inhibition of peripheral inflammatory processes, especially neutrophil priming.50,51

Although the antihyperalgesic effects reported here appear to be mediated through exclusively local actions, since distant injections are without detectable effect, it is possible that some of the actions do occur from bupivacaine that has entered the CNS. To detect such effects behaviorally, however, may require a combination of local direct and indirect CNS actions. Two types of evidence support such an idea. First, very low plasma concentrations of local anesthetics are able to alter electrophysiological signals in peripheral nerve,52,53 reducing trains of “abnormal” action potentials at concentrations that are totally ineffective on normally conducting impulses.54 Such low plasma concentrations of lidocaine, for example, are used to treat persistent neuropathic pain55 and are effective in preventive analgesia when infused IV in the perioperative period.56,57 Second, an analogous situation has been documented in the cardiotoxic actions of bupivacaine, where this drug in the CNS makes a critical contribution to arrhythmias induced by its direct application to the heart.58 There may be a contribution of systemic bupivacaine, still well below toxic levels, to the therapeutic benefit from direct neural application.

Some previously developed controlled-release PLGA–MS-Bupi achieved 24 hours or more of functional blockade in rats only on the incorporation of dexamethasone in the formulation as the second active pharmaceutical ingredient.13,25,36 In contrast, the results reported here demonstrate a significant reduction of pain in rats lasting several days (i.e., well above 24 hours) with the use of bupivacaine as the only active pharmaceutical ingredient in the microsphere formulation. While it is beyond the scope of this report to investigate specific mechanistic differences between past PLGA–MS-Bupi formulations and those reported here, we speculate below on possible hypotheses that might explain these differences.

The drug-loading levels achieved in past studies with PLGA–MS-Bupi13,25,40 and those reported here were both high, being 75% (w/w) for the past studies and 60% (w/w) reported in this article. In addition, the total drug dosage administered per animal (without accounting for small losses during the delivery) was similar between past studies (150 mg drug/kg animal) and the current studies at the highest dosing (133.3 mg drug/kg animal). Perhaps the difference in performance is based on the differences in kinetics of release, impacting the bioavailability of bupivacaine and subsequently the duration of in vivo efficacy. A comparison of the release kinetics for the past studies with PLGA–MS-Bupi suggests that 75% of the release of bupivacaine in vitro occurred over about 7 days. In contrast, in our studies, 75% of bupivacaine was released in vitro in about 3 days. Therefore, the effective rate of release differs by a factor of >2. However, this average difference contrasts with the first 6 to 12 hours of bupivacaine release; in the previously published formulations, a “burst effect” was observed in vitro, whereas the formulation reported here had no pronounced burst effect but rather a relatively constant release rate of approximately 0.6 to 0.75 mg/h (Fig. 1). Consistent with this observation, a recent study suggests that reducing the burst effect might prolong in vivo efficacy for bupivacaine.59

Experimental studies in humans of similar slow-release bupivacaine formulations using PLGA microspheres report much lower dosing, 0.9 to approximately 7 mg/kg for minor peripheral nerve block,22 and 0.6 to 3 mg/kg for intercostal nerve block,21 neither of which was followed by surgery. A similar dose range was used for liposomal bupivacaine infiltrated at the wound for knee arthroscopy, 2.5 to 10 mg/kg.60 None of these studies reported any adverse events, and the last examined cardiac electrophysiological signs as an end point. Our animal study, using much higher doses, was intended to anesthetize a surgical field that is relatively large compared with the rat’s total skin area, so we used much larger doses than would be needed for peripheral nerve blocks. The systemic levels of bupivacaine after such injections are now being evaluated.

For the MS-Bupi formulation used here, the uniform distribution of both polymer and bupivacaine throughout the microsphere cross-section (Raman spectroscopy data, Fig. 2) and the low porosity for these microspheres (gas physisorption data) probably contributed to achieving a relatively consistent release rate for bupivacaine. The Raman spectroscopy and porosity data are also consistent with a surface erosion and diffusion-based mechanism for bupivacaine release, an inference supported by the relatively good fit for our formulation with the Higuchi model (R2 = 0.985) for diffusion-based controlled release systems.16

Regardless of the mechanisms, the presence of this slow-release microsphere formulation for bupivacaine at the incision site provides substantial reduction of pain-like symptoms in a surgical model for prolonged postoperative mechanical hypersensitivity. Whether given before the surgical manipulations or after closing,61 this formulation may be a safe and effective means for postoperative pain control. Our ongoing studies continue to further evaluate the efficacy and safety of our MS-Bupi formulation, including an evaluation of the inflammatory response.


Name: Rachit Ohri, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Rachit Ohri has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: Rachit Ohri worked for Covidien.

Name: Jeffrey Chi-Fei Wang, MD.

Contribution: This author helped conduct the study, analyze the data, and write the manuscript.

Attestation: Jeffrey Chi-Fei Wang has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Phillip D. Blaskovich.

Contribution: This author helped conduct the study and analyze the data.

Attestation: Phillip D. Blaskovich has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: Phillip D. Blaskovich worked for Covidien.

Name: Lan N. Pham, MS.

Contribution: This author helped conduct the study, analyze the data, and write the manuscript.

Attestation: Lan N. Pham has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Daniel S. Costa, BS.

Contribution: This author helped analyze the data.

Attestation: Daniel S. Costa has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Gary A. Nichols, PhD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Gary A. Nichols has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: Gary A. Nichols worked for Covidien and is an employee of Covidien Pharmaceuticals, Inc.

Name: William P. Hildebrand.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: William P. Hildebrand has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: William P. Hildebrand worked for Covidien Pharmaceuticals is an employee of Covidien.

Name: Nelson L. Scarborough, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Nelson L. Scarborough has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Conflicts of Interest: The author has no conflicts of interest to declare.

Name: Clifford J. Herman, PhD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Clifford J. Herman has seen the original study data and approved the final manuscript.

Conflicts of Interest: Clifford J. Herman is an employee of Covidien.

Name: Gary R. Strichartz, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Gary R. Strichartz has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Gary R. Strichartz received research funding from Covidien, is the Principal Investigator on this project, funded by Covidien, and derives some salary support from Covidien through this funding.

This manuscript was handled by: Martin S. Angst, MD.


The authors thank Mr. Jamie Bell, Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women’s Hospital, Boston, MA, for help with the figures, and Ms. Xiaoxia Liu, MS, for substantial aid with statistical analyses. David Brooks, MD, Department of Surgery, Brigham and Women’s Hospital, helped with the surgical descriptors. The authors also wish to acknowledge the counsel and support received from the following individuals at Covidien: Dr. Jeffrey Zaruby, Marcella Szwaja, Valentino Tramontano, Steve Bennett, Jason Fortier, Kreg Howk, Kenneth Martin, Daniel Broom, Brian Donley and Donald Hazuka.


1. Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet. 2006;367:1618–25
2. Althaus A, Hinrichs-Rocker A, Chapman R, Arránz Becker O, Lefering R, Simanski C, Weber F, Moser KH, Joppich R, Trojan S, Gutzeit N, Neugebauer E. Development of a risk index for the prediction of chronic post-surgical pain. Eur J Pain. 2012;16:901–10
3. Abramov Y, Sand PK, Gandhi S, Botros SM, Miller JJ, Koh EK, Goldberg RP. The effect of preemptive pudendal nerve blockade on pain after transvaginal pelvic reconstructive surgery. Obstet Gynecol. 2005;106:782–8
4. Chester JF, Stanford BJ, Gazet JC. Analgesic benefit of locally injected bupivacaine after hemorrhoidectomy. Dis Colon Rectum. 1990;33:487–9
5. Vinson-Bonnet B, Coltat JC, Fingerhut A, Bonnet F. Local infiltration with ropivacaine improves immediate postoperative pain control after hemorrhoidal surgery. Dis Colon Rectum. 2002;45:104–8
6. Apfelbaum JL, Chen C, Mehta SS, Gan TJ. Postoperative pain experience: results from a national survey suggest postoperative pain continues to be undermanaged. Anesth Analg. 2003;97:534–40
7. Kissin I. A call to reassess the clinical value of preventive (preemptive) analgesia. Anesth Analg. 2011;113:977–8
8. White PF. Multimodal analgesia: its role in preventing postoperative pain. Curr Opin Investig Drugs. 2008;9:76–82
9. Barreveld A, Witte J, Chahal H, Durieux ME, Strichartz G. Preventive analgesia by local anesthetics: the reduction of postoperative pain by peripheral nerve blocks and intravenous drugs. Anesth Analg. 2013;116:1141–61
10. Duarte AM, Pospisilova E, Reilly E, Mujenda F, Hamaya Y, Strichartz GR. Reduction of postincisional allodynia by subcutaneous bupivacaine: findings with a new model in the hairy skin of the rat. Anesthesiology. 2005;103:113–25
11. Ohmura S, Kawada M, Ohta T, Yamamoto K, Kobayashi T. Systemic toxicity and resuscitation in bupivacaine-, levobupivacaine-, or ropivacaine-infused rats. Anesth Analg. 2001;93:743–8
12. McLeod GA, Butterworth JE, Wildsmith AWCousins MJ, Bridenbaugh PO, Carr DB, Horlocker TT. Local anesthetic systemic toxicity. Neural Blockade in Clinical Anesthesia and Pain Medicine. 20094th ed Philadelphia, PA Lippincott Williams & Wilkins,:114–32
13. Curley J, Castillo J, Hotz J, Uezono M, Hernandez S, Lim JO, Tigner J, Chasin M, Langer R, Berde C. Prolonged regional nerve blockade. Injectable biodegradable bupivacaine/polyester microspheres. Anesthesiology. 1996;84:1401–10
14. Grant GJ, Barenholz Y, Bolotin EM, Bansinath M, Turndorf H, Piskoun B, Davidson EM. A novel liposomal bupivacaine formulation to produce ultralong-acting analgesia. Anesthesiology. 2004;101:133–7
15. Wang CF, Pancaro C, Gerner P, Strichartz G. Prolonged suppression of postincisional pain by a slow-release formulation of lidocaine. Anesthesiology. 2011;114:135–49
16. Ohri R, Blaskovich P, Wang JC, Pham L, Nichols G, Hildebrand W, Costa D, Scarborough N, Herman C, Strichartz G. Prolonged nerve block by microencapsulated bupivacaine prevents acute postoperative pain in rats. Reg Anesth Pain Med. 2012;37:607–15
17. Wang CF, Djalali AG, Gandhi A, Knaack D, De Girolami U, Strichartz G, Gerner P. An absorbable local anesthetic matrix provides several days of functional sciatic nerve blockade. Anesth Analg. 2009;108:1027–33
18. Haas E, Onel E, Miller H, Ragupathi M, White PF. A double-blind, randomized, active-controlled study for post-hemorrhoidectomy pain management with liposome bupivacaine, a novel local analgesic formulation. Am Surg. 2012;78:574–81
19. Viscusi ER, Candiotti KA, Onel E, Morren M, Ludbrook GL. The pharmacokinetics and pharmacodynamics of liposome bupivacaine administered via a single epidural injection to healthy volunteers. Reg Anesth Pain Med. 2012;37:616–22
20. Kopacz DJ, Lacouture PG, Wu D, Nandy P, Swanton R, Landau C. The dose response and effects of dexamethasone on bupivacaine microcapsules for intercostal blockade (T9 to T11) in healthy volunteers. Anesth Analg. 2003;96:576–82
21. Kopacz DJ, Bernards CM, Allen HW, Landau C, Nandy P, Wu D, Lacouture PG. A model to evaluate the pharmacokinetic and pharmacodynamic variables of extended-release products using in vivo tissue microdialysis in humans: bupivacaine-loaded microcapsules. Anesth Analg. 2003;97:124–31
22. Holte K, Werner MU, Lacouture PG, Kehlet H. Dexamethasone prolongs local analgesia after subcutaneous infiltration of bupivacaine microcapsules in human volunteers. Anesthesiology. 2002;96:1331–5
23. Pedersen JL, Lillesø J, Hammer NA, Werner MU, Holte K, Lacouture PG, Kehlet H. Bupivacaine in microcapsules prolongs analgesia after subcutaneous infiltration in humans: a dose-finding study. Anesth Analg. 2004;99:912–8
24. Dräger C, Benziger D, Gao F, Berde CB. Prolonged intercostal nerve blockade in sheep using controlled-release of bupivacaine and dexamethasone from polymer microspheres. Anesthesiology. 1998;89:969–79
25. Castillo J, Curley J, Hotz J, Uezono M, Tigner J, Chasin M, Wilder R, Langer R, Berde C. Glucocorticoids prolong rat sciatic nerve blockade in vivo from bupivacaine microspheres. Anesthesiology. 1996;85:1157–66
26. Brennan TJ, Vandermeulen EP, Gebhart GF. Characterization of a rat model of incisional pain. Pain. 1996;64:493–501
27. Uchida T, Hazekawa M, Morisaki T, Yoshida M, Sakai Y. Effect of antioxidants on the stability of ONO-1301, a novel long-acting prostacyclin agonist, loaded in PLGA microspheres. J Microencapsul. 2013;30:245–56
28. Zhang XJ, Ke LM, Yang J, Lin LW, Xue ES, Wang Y, Yu LY, Chen ZK. Development, characterization and anti-tumor effect of a sequential sustained-release preparation containing ricin and cobra venom cytotoxin. Pharmazie. 2012;67:618–21
29. Rong X, Yang S, Miao H, Guo T, Wang Z, Shi W, Mo X, Yuan W, Jin T. Effects of erythropoietin-dextran microparticle-based PLGA/PLA microspheres on RGCs. Invest Ophthalmol Vis Sci. 2012;53:6025–34
30. Siepmann J, Peppas NA. Higuchi equation: derivation, applications, use and misuse. Int J Pharm. 2011;418:6–12
31. Thalhammer JG, Vladimirova M, Bershadsky B, Strichartz GR. Neurologic evaluation of the rat during sciatic nerve block with lidocaine. Anesthesiology. 1995;82:1013–25
32. Theriault E, Diamond J. Nociceptive cutaneous stimuli evoke localized contractions in a skeletal muscle. J Neurophysiol. 1988;60:446–62
33. Treede RD, Meyer RA, Raja SN, Campbell JN. Peripheral and central mechanisms of cutaneous hyperalgesia. Prog Neurobiol. 1992;38:397–421
34. Rodriguez NA, Cooper DM, Risdahl JM. Antinociceptive activity of and clinical experience with buprenorphine in swine. Contemp Top Lab Anim Sci. 2001;40:17–20
35. Andrews K, Fitzgerald M. Wound sensitivity as a measure of analgesic effects following surgery in human neonates and infants. Pain. 2002;99:185–95
36. Kohane DS, Smith SE, Louis DN, Colombo G, Ghoroghchian P, Hunfeld NG, Berde CB, Langer R. Prolonged duration local anesthesia from tetrodotoxin-enhanced local anesthetic microspheres. Pain. 2003;104:415–21
37. Gerner P, Wang CF, Lee BS, Suzuki S, Degirolami U, Gandhi A, Knaack D, Strichartz G. The relationship between functional sciatic nerve block duration and the rate of release of lidocaine from a controlled-release matrix. Anesth Analg. 2010;111:221–9
38. Tobe M, Obata H, Suto T, Yokoo H, Nakazato Y, Tabata Y, Saito S. Long-term effect of sciatic nerve block with slow-release lidocaine in a rat model of postoperative pain. Anesthesiology. 2010;112:1473–81
39. Xie W, Strong JA, Meij JT, Zhang JM, Yu L. Neuropathic pain: early spontaneous afferent activity is the trigger. Pain. 2005;116:243–56
40. Aida S, Baba H, Yamakura T, Taga K, Fukuda S, Shimoji K. The effectiveness of preemptive analgesia varies according to the type of surgery: a randomized, double-blind study. Anesth Analg. 1999;89:711–6
41. Peters CM, Eisenach JC. Contribution of the chemokine (C-C motif) ligand 2 (CCL2) to mechanical hypersensitivity after surgical incision in rats. Anesthesiology. 2010;112:1250–8
42. Yanagidate F, Strichartz GR. Bupivacaine inhibits activation of neuronal spinal extracellular receptor-activated kinase through selective effects on ionotropic receptors. Anesthesiology. 2006;104:805–14
43. Gokin AP, Philip B, Strichartz GR. Preferential block of small myelinated sensory and motor fibers by lidocaine: in vivo electrophysiology in the rat sciatic nerve. Anesthesiology. 2001;95:1441–54
44. Burgess PR, Perl ER. Myelinated afferent fibres responding specifically to noxious stimulation of the skin. J Physiol. 1967;190:541–62
45. Bessou P, Perl ER. Response of cutaneous sensory units with unmyelinated fibers to noxious stimuli. J Neurophysiol. 1969;32:1025–43
46. Xu J, Brennan TJ. Guarding pain and spontaneous activity of nociceptors after skin versus skin plus deep tissue incision. Anesthesiology. 2010;112:153–64
47. Hitosugi H, Kashiwazaki T, Ohsawa M, Kamei J. Effects of mexiletine on algogenic mediator-induced nociceptive responses in mice. Methods Find Exp Clin Pharmacol. 1999;21:409–13
48. Li YM, Wingrove DE, Too HP, Marnerakis M, Stimson ER, Strichartz GR, Maggio JE. Local anesthetics inhibit substance P binding and evoked increases in intracellular Ca2+. Anesthesiology. 1995;82:166–73
49. Hollmann MW, Fischer LG, Byford AM, Durieux ME. Local anesthetic inhibition of m1 muscarinic acetylcholine signaling. Anesthesiology. 2000;93:497–509
50. Hollmann MW, Durieux ME. Local anesthetics and the inflammatory response: a new therapeutic indication? Anesthesiology. 2000;93:858–75
51. Ploppa A, Kiefer RT, Haverstick DM, Groves DS, Unertl KE, Durieux ME. Local anesthetic effects on human neutrophil priming and activation. Reg Anesth Pain Med. 2010;35:45–50
52. Xiao WH, Bennett GJ. C-fiber spontaneous discharge evoked by chronic inflammation is suppressed by a long-term infusion of lidocaine yielding nanogram per milliliter plasma levels. Pain. 2008;137:218–28
53. Kirillova I, Teliban A, Gorodetskaya N, Grossmann L, Bartsch F, Rausch VH, Struck M, Tode J, Baron R, Jänig W. Effect of local and intravenous lidocaine on ongoing activity in injured afferent nerve fibers. Pain. 2011;152:1562–71
54. Persaud N, Strichartz GR. Micromolar lidocaine selectively blocks propagating ectopic impulses at a distance from their site of origin. Pain. 2002;99:333–40
55. Attal N, Bouhassira D. Translating basic research on sodium channels in human neuropathic pain. J Pain. 2006;7:S31–7
56. Marret E, Rolin M, Beaussier M, Bonnet F. Meta-analysis of intravenous lidocaine and postoperative recovery after abdominal surgery. Br J Surg. 2008;95:1331–8
57. De Oliveira GS Jr, Fitzgerald P, Streicher LF, Marcus RJ, McCarthy RJ. Systemic lidocaine to improve postoperative quality of recovery after ambulatory laparoscopic surgery. Anesth Analg. 2012;115:262–7
58. Bernards CM, Carpenter RL, Rupp SM, Brown DL, Morse BV, Morell RC, Thompson GE. Effect of midazolam and diazepam premedication on central nervous system and cardiovascular toxicity of bupivacaine in pigs. Anesthesiology. 1989;70:318–23
59. Sokolsky-Papkov M, Golovanevski L, Domb AJ, Weiniger CF. Poly(DL:lactic acid-castor oil) 3:7-bupivacaine formulation: reducing burst effect prolongs efficacy in vivo. J Pharm Sci. 2010;99:2732–8
60. Bergese SD, Onel E, Morren M, Morganroth J. Bupivacaine extended-release liposome injection exhibits a favorable cardiac safety profile. Reg Anesth Pain Med. 2012;37:145–51
61. Møiniche S, Kehlet H, Dahl JB. A qualitative and quantitative systematic review of preemptive analgesia for postoperative pain relief: the role of timing of analgesia. Anesthesiology. 2002;96:725–41
© 2013 International Anesthesia Research Society