Drug delivery to the peritoneum is hampered by rapid clearance, and could be improved by application of controlled release technology. We investigated the suitability for peritoneal use of micro- and nanoparticles of poly(lactic-co-glycolic) acid (PLGA), a biodegradable polymer with generally excellent biocompatibility commonly used for controlled drug release. We injected 90 kDa PLGA microparticles, 5-250 microm in diameter, into the murine peritoneum, in dosages of 10-100 mg (n=3-5 per group). We found a high incidence of polymeric residue and adhesions 2 weeks after injection (e.g., 50 mg of 5-microm microparticles caused adhesions in 83% of animals). Histology revealed chronic inflammation, with foreign body giant cells prominent with particles>5 microm in diameter. Five micrometer microspheres made from 54, 57, and 10 kDa PLGA (gamma irradiated) caused fewer adhesions (16.7%) with a similar incidence of residue. Nanoparticles (265 nm) of 90 kDa PLGA also caused much fewer adhesions (6.3% of animals), possibly because they were cleared from the peritoneum within 2 days, and sequestered in the spleen and liver, where foamy macrophages were noted. The effect of sterilization technique on the incidence of adhesion formation is also studied.

Background: Sustained release of local anesthetics is frequently associated with myotoxicity. The authors investigated the role of particulate delivery systems and of the pattern of drug release in causing myotoxicity.Methods: Rats were given sciatic nerve blocks with bupivacaine solutions, two types of bupivacaine-containing microparticles (polymeric microspheres and lipid-protein-sugar particles), or blank particles with or without bupivacaine in the carrier fluid. Myotoxicity was scored in histologic sections of the injection sites. Bupivacaine release kinetics from the particles were measured. Myotoxicity of a range of bupivacaine concentrations from exposures up to 3 weeks was assessed in C2C12 myotubes, with or without microparticles.Results: Both types of bupivacaine-loaded microparticles, but not blank particles, were associated with myotoxicity. Whereas 0.5% bupivacaine solution caused little myotoxicity, a concentration of bupivacaine that mimicked the amount of bupivacaine released initially from particles caused myotoxicity. Local anesthetics showed both concentration and time-dependent myotoxicity in C2C12s. Importantly, even very low concentrations that were nontoxic over brief exposures became highly toxic after days or weeks of exposure. The presence of particles did not increase bupivacaine myotoxicity in vitro but did in vivo. Findings applied to both particle types.Conclusions: Whereas the release vehicles themselves were not myotoxic, both burst and extended release of bupivacaine were. A possible implication of the latter finding is that myotoxicity is an inevitable concomitant of sustained release of local anesthetics. Particles, and perhaps other vehicles, may enhance local toxicity through indirect mechanisms.

Conventional local anesthetics such as bupivacaine cause considerable myotoxicity and neurotoxicity, whereas tetrodotoxin (TTX) does not. Tetrodotoxin combined with bupivacaine or vasoconstrictors produces long-duration nerve blockade. To assess whether these prolonged blocks can be produced without increased myotoxicity, Sprague-Dawley rats were injected with bupivacaine, TTX, and both, or TTX plus epinephrine. Median durations of thermal nociceptive blockade were, respectively, 188, 401, 882, and 972 min. On dissection 4 days later, all tissues appeared macroscopically pristine. Muscle injury was at most mild to moderate in all animals, and the muscle injury scores for the combination formulations were not higher than for bupivacaine alone. Similarly, in differentiated cells from a myoblast cell line (C2C12), TTX caused either no or minimal worsening of cell viability from bupivacaine at 2 or 7 days. Epinephrine did not worsen TTX's relatively minimal cytotoxicity. Tetrodotoxin may thus be useful in producing prolonged nerve block with minimal myotoxicity and perhaps neurotoxicity.

Many recent reports in the literature address the local anesthetics efficacy of tricyclic antidepressants (TCAs). Here we investigated whether nerve block from TCAs is prolonged by site 1 sodium channel blockers such as tetrodotoxin and saxitoxin, which are known to prolong block from conventional local anesthetics. Tetrodotoxin and saxitoxin greatly prolonged block from TCAs. For example, the median duration of thermal nociceptive blocks for 10 mM amitriptyline, nortriptyline and doxepin were 0, 0, and 124 min; co-injection with 20 microM TTX (median block duration=0), yielded blocks lasting 404, 325, and 697 min, respectively. Co-injection of 12 microM saxitoxin (median block duration=0) with 10 mM amitriptyline resulted in a thermal nociceptive block duration of 373 min. Co-injection of 7.7 mM bupivacaine and 7.7 mM amiptriptyline did not result in block prolongation. Systemic (subcutaneous) delivery of tetrodotoxin or amitriptyline did not result in prolongation of block from the other class of drug injected at the sciatic nerve. In TCA-containing formulations, motor blockade was consistently longer than thermal nociceptive block; motor blockade was also prolonged by tetrodotoxin and saxitoxin. In summary site 1 sodium channel blockers prolong the duration of TCAs via a locally mediated mechanism.

Application of controlled release technology to the peritoneum would allow for sustained drug levels. However, some polymeric systems either create adhesions, or rapidly exit the peritoneum; neither result is desirable. Here we have produced particles based on sphyngomyelin, a phospholipid that occurs naturally in the peritoneum, along with hyaluronic acid and the polymethacrylate Eudragit E100 (to modulate drug release). Particles with a low proportion of E100 (5% (w/w); "high SPM") release albumin rapidly over 2 days, then more slowly; increasing the E100 to 20% (w/w; high "E100") slowed drug release markedly. When injected in the murine peritoneum, high SPM particles were disseminated as free particles, without forming collections. There was a mild inflammatory response but no formation of adhesions. High E100 particles formed collections in all animals, with an intense inflammatory response. Even so, there were very few adhesions. These results suggest that microparticulate formulations can be produced that have acceptable drug-releasing properties and are suitable for use in the peritoneum from the standpoint of biocompatibility.