Portal de Periódicos da CAPES

Percutaneous Penetration of Local Anesthetic Bases: Pharmacodynamic Measurements

1999; Elsevier BV; Volume: 113; Issue: 3 Linguagem: Inglês

10.1046/j.1523-1747.1999.00691.x

1523-1747

Claudia Leopold, Howard I. Maibach,

Advanced Chemical Sensor Technologies

Local anesthetics do not penetrate readily through human skin if applied in their salt form; however, if applied in their base form various effects may be observed, such as a decrease in pricking pain and a change in burning, itch, and thermal sensations. These effects occur after skin penetration and may be attributed to the action of the anesthetics on nociceptors and thermoreceptors, i.e., on C and Aδ nerve fiber respectively. As there is little known about the time course of the pharmacodynamic response of cutaneously applied local anesthetic bases, this study was conducted to characterize various local anesthetics pharmacodynamically by measuring thermal thresholds over time with a thermal sensory analyzer. The results show that the investigated local anesthetics affect thermal thresholds to a different extent, with tetracaine and lidocaine being most efficient. From the response versus time profiles of all eight study subjects various response parameters were obtained: only the cold sensation parameters proved suitable for characterization of the local anesthetics, possibly because cold receptors are located in the epidermis and can easily be reached. Lag times of onset are short and the maximum anesthetic effect is reached within 2–3 h. Cold sensation parameters correlate linearly with the solubility of the local anesthetic bases in medium chain triglycerides and with the drug flux of 50% saturation, indicating that medium chain triglycerides may have similar properties with regard to the local anesthetics solubility as the stratum corneum lipids. Local anesthetics do not penetrate readily through human skin if applied in their salt form; however, if applied in their base form various effects may be observed, such as a decrease in pricking pain and a change in burning, itch, and thermal sensations. These effects occur after skin penetration and may be attributed to the action of the anesthetics on nociceptors and thermoreceptors, i.e., on C and Aδ nerve fiber respectively. As there is little known about the time course of the pharmacodynamic response of cutaneously applied local anesthetic bases, this study was conducted to characterize various local anesthetics pharmacodynamically by measuring thermal thresholds over time with a thermal sensory analyzer. The results show that the investigated local anesthetics affect thermal thresholds to a different extent, with tetracaine and lidocaine being most efficient. From the response versus time profiles of all eight study subjects various response parameters were obtained: only the cold sensation parameters proved suitable for characterization of the local anesthetics, possibly because cold receptors are located in the epidermis and can easily be reached. Lag times of onset are short and the maximum anesthetic effect is reached within 2–3 h. Cold sensation parameters correlate linearly with the solubility of the local anesthetic bases in medium chain triglycerides and with the drug flux of 50% saturation, indicating that medium chain triglycerides may have similar properties with regard to the local anesthetics solubility as the stratum corneum lipids. cold pain threshold cold sensation heat pain threshold LA flux of 50% saturation local anesthetic bases medium chain triglycerides warm sensation Topical local anesthetic preparations are widely available as over-the-counter remedies. Such products are usually intended for anesthesia of mucosal epithelia and of broken or abraded skin. They are almost without effect on healthy skin where the chemical barrier function of the stratum corneum remains intact; however, if certain anesthetics are applied to healthy skin in their base form rather than the more usual, water soluble salts, effects may be observed, such as a decrease in pricking pain (McCafferty et al., 1988McCafferty D.F. Woolfson A.D. McClelland K.H. Boston V. Comparative in vivo and in vitro assessment of the percutaneous absorption of local anaesthetics.Br J Anaesth. 1988; 60: 64-69Crossref PubMed Scopus (39) Google Scholar;Woolfson and McCafferty, 1993Woolfson D. McCafferty D. Percutaneous Local Anaesthesia. Chichester, Ellis Horwood1993Google Scholar), a reduction of the flare response to histamine (Pipkorn and Andersson, 1987Pipkorn U. Andersson M. Topical dermal anaesthesia inhibits the flare but not the weal response to allergen and histamine in the skin-prick test.Clinical Allergy. 1987; 17: 307-311Crossref PubMed Scopus (32) Google Scholar), and a change in burning, itch, and thermal sensations (Adriani and Dalili, 1971Adriani J. Dalili H. Penetration of local anaesthetics through epithelial barriers.Anesth Analg. 1971; 50: 834-841Crossref PubMed Scopus (55) Google Scholar;Yosipovitch and Maibach, 1997Yosipovitch G. Maibach H.I. Effect of topical pramoxine on experimentally induced pruritus in humans.J Am Acad Dermatol. 1997; 37: 278-280Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar). These effects are due to the action of the anesthetics on nociceptors and thermoreceptors, i.e., on C and Aδ nerve fibers, respectively. With the introduction of the eutectic mixture of lidocaine and prilocaine (EMLA) and the phase change system consisting of tetracaine in a xanthan gum gel (Ametop), effective percutaneous local anesthesia is feasible. As the skin is richly provided with sensory nerve fibers, quantitative sensory testing such as the assessment of vibration, light touch, thermal sensations, and pain thresholds is a useful procedure not only for diagnostic purposes but also for the investigation of the effect of topical drugs on sensory perception (Yosipovitch et al., 1996aYosipovitch G. Szolar C. Hui X.Y. Maibach H. Effect of topically applied menthol on thermal, pain and itch sensations and biophysical properties of the skin.Arch Dermatol Res. 1996; 288: 245-248Crossref PubMed Scopus (118) Google Scholar,Yosipovitch et al., 1996bYosipovitch G. Szolar C. Hui X.Y. Maibach H. High-potency topical corticosteroid rapidly decreases histamine-induced itch but not thermal sensation and pain in human beings.J Am Acad Dermatol. 1996; 35: 118-120Abstract Full Text PDF PubMed Scopus (39) Google Scholar,Yosipovitch et al., 1997Yosipovitch G. Ademola J. Lui P. Amin S. Maibach H.I. Topically applied aspirin rapidly decreases histamine-induced itch.Acta Derm Venereol. 1997; 77: 46-48PubMed Google Scholar;Yosipovitch and Maibach, 1997Yosipovitch G. Maibach H.I. Effect of topical pramoxine on experimentally induced pruritus in humans.J Am Acad Dermatol. 1997; 37: 278-280Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar). Thermal sensory testing is becoming an important quantitative sensory testing technique because it allows the investigation of small nerve fiber function, which cannot be evaluated by nerve conduction velocity tests (Yosipovitch and Yarnitsky, 1996Yosipovitch G. Yarnitsky D. Quantitative sensory testing.in: Marzulli FN Maibach HI Dermatotoxicology Methods. Washington DC, Hemisphere1996: 313-318Google Scholar). Thermal sensory testing allows the measurement of the effect of drugs on both thermal sensations and thermal pain thresholds. If a peltier device such as a thermal sensory analyzer is used as a testing instrument, the effect of topically applied local anesthetics on cold and warm sensations as well as on thermal pain thresholds may be determined (Yosipovitch and Yarnitsky, 1996Yosipovitch G. Yarnitsky D. Quantitative sensory testing.in: Marzulli FN Maibach HI Dermatotoxicology Methods. Washington DC, Hemisphere1996: 313-318Google Scholar). As there is little known about the time course of the pharmacodynamic response of cutaneously applied local anesthetic bases, this study was conducted to characterize various anesthetics pharmacodynamically by measuring thermal thresholds over time with a thermal sensory analyzer. Fomocaine-HCl (Engelhard, Frankfurt am Main, Germany), benzocaine (Ritsert, Eberbach, Germany), tetracaine-HCl (Mann, Berlin, Germany), lidocaine-HCl (Astra, Wedel, Germany), etidocaine-HCl (Astra, Wedel, Germany), and carticaine-HCl (Hoechst, Frankfurt am Main, Germany) were used to prepare the LA. The bases were obtained by dissolving the hydrochlorides in distilled water and adjusting the pH to 10 with 3 M sodium hydroxide. Subsequently, the LA were extracted three times with diethyl ether. The combined fractions were dried with anhydrous sodium sulfate, evaporated to dryness after filtration, and stored in a desiccator under a vacuum for 2 d. All six LA are solid at room temperature. Only one melting peak was detected by DSC and in HPLC chromatograms one sharp peak was found with each of the LA even after 2.5 y of storage. The study was approved by the Ethics Committee of the School of Medicine of the University of California at San Francisco. Eight healthy volunteers (age 25–65 y) provided written informed consent to participate. They received all six LA, one at a time and 7 d apart. According to their solubility the LA were dissolved in either methylene chloride (fomocaine, benzocaine, tetracaine, carticaine) or ethanol (lidocaine, etidocaine) at a concentration of 100 mg per ml and 1 ml of each solution was applied to an area of 3.5 × 3.5 cm2 on the volar aspect of the forearm with an insulin syringe. To guarantee the maximum thermodynamic LA activity and to avoid a barrier-modifying action of the solution media the solvents were allowed to evaporate. Cold and warm sensations (CS, WS) as well as cold and heat pain thresholds (CP, HP) were measured non-invasively on the uncovered application sites with a thermal sensory analyzer (TSA 2001, Medoc U.S., Minneapolis, MN). The instrument was equipped with a thermode (size 3 × 3 cm2) adjusted to an adaptation temperature of 33°C, which caused occlusion of the application site as it remained on the skin for 4–5 h. Control measurements were done immediately before application of the LA and thresholds were recorded starting 15 min after LA application using the method of limits as test algorithm (Yosipovitch and Yarnitsky, 1996Yosipovitch G. Yarnitsky D. Quantitative sensory testing.in: Marzulli FN Maibach HI Dermatotoxicology Methods. Washington DC, Hemisphere1996: 313-318Google Scholar). With this method, stimuli increase continuously in intensity until the requested sensation is perceived, at which moment the stimulus is halted by the subject and the thermode temperature returns to adaptation temperature. A reaction time artifact is built in this measurement. All subjects were trained in CS, WS, CP, and HP perception before starting the experiment to minimize the intersession bias and to achieve acceptable repeatability (Yarnitsky and Sprecher, 1994Yarnitsky D. Sprecher E. Thermal testing: Normative data and repeatability for various test algorithms.J Neurol Sci. 1994; 125: 39-45Abstract Full Text PDF PubMed Scopus (250) Google Scholar). Thermal thresholds were obtained in the following manner. In session I, only WS and CS were measured every 5 min over 4–5 h until constant thresholds were achieved. Six oscillating stimuli starting with WS were given at a constant temperature rate of 0.3°C and with a time interval of 5 s between stimuli. In session II, 1 d later, HP and CP were measured on the contralateral side every 10 min. The six oscillating stimuli were given with a temperature rate of 0.5°C, a return rate to adaptation temperature of 10°C, and time intervals between the stimuli of 10 s. Sessions were held in a sound-proof air-conditioned room, with distractions minimized. Subjects did not have visual access to the computer screen; no visual or auditory cues were given to signal stimulus onset. WS and CS as well as HP and CP means were calculated automatically by the Medoc software for each cluster of stimuli. In preliminary experiments no significant effect of stimulus repetition on thermal thresholds measured over 5 h under the above-mentioned conditions could be detected. From the thermal thresholds versus time profiles the following response parameters were obtained: the maximum change in CS and CP (Δ CS, Δ CP) as intensity-related parameters, the CS slope as a combined time- and intensity-related parameter, and the lag time of onset of the anesthetic action. Δ CS and Δ CP were calculated as the differences between the baseline and the minimum thresholds. Statistical significance (p = 0.05) of these differences was confirmed with a paired t test. Minimum thresholds correspond to the means of the last five data points whereas the baseline thresholds represent the means of the first four data points including the control value. As the standard deviations of these data points are small in comparison with the intersubject variations, they were assumed to be negligible. The slope of the curves was determined by linear regression analysis, as the decrease in CS over time appeared to be linear (correlation coefficients between 0.95 and 0.98). All data points that exceeded the range of variation of the first four and the last five data points were included in the regression analysis. The lag times of onset were determined as the time points of intersection of the mean baseline plateau and the CS regression lines. Drug penetration measurements were done with a glass cell system (Leopold and Lippold, 1992Leopold C.S. Lippold B.C. A new application chamber for skin penetration studies in vivo with liquid preparations.Pharm Res. 1992; 9: 1215-1218Crossref PubMed Google Scholar,Leopold and Lippold, 1995Leopold C.S. Lippold B.C. Enhancing effect of lipophilic vehicles on skin penetration of methyl nicotinate in vivo.J Pharm Sci. 1995; 84: 195-198Crossref PubMed Scopus (40) Google Scholar). This method allows the measurement of even small drug fluxes by the difference method (Stricker et al., 1987Stricker H. Winer G. Leber A. Galenische Faktoren der kutanen Wirkstoffpenetration in vivo.Acta Pharm Technol. 1987; 33: 80-87Google Scholar). Etidocaine, benzocaine, lidocaine, and tetracaine were dissolved in the inert non-irritating vehicle purified light mineral oil (Parafluid Mineralölgesellschaft, Hamburg, Germany) at concentrations that correspond to 50% of the LA solubility in this vehicle (etidocaine, 852.8 mg per 100 ml; benzocaine, 60 mg per 100 ml; lidocaine, 2419.6 mg per 100 ml; tetracaine, 1037.7 mg per 100 ml). The four drug solutions were applied to the upper arms of six healthy subjects and examined under occlusive conditions because of the closed system over 6 h. After a 1 h pretreatment period with the vehicle, each glass cell was filled with one of the drug solutions, emptied after 1 or 2 h and refilled with the initial drug solution. Zero order kinetics were assumed because the concentration decrease in each 2 h time interval was <10%, with the exception of the tetracaine solution which for this reason had to be replaced at 1 h time intervals. The concentration of the donor phase samples was measured spectrophotometrically (Lambda 2, Perkin-Elmer, Überlingen, Germany), in 0.1 cm quartz cells at wavelengths that provided an UV absorption between 0.9 and 1.0 of the initial solutions. LA disappearance rates per area unit were calculated from the concentration differences between the initial solution and the samples obtained after every 1 or 2 h, multiplied by the volume of the respective cells, and divided by the application area and the time interval. Steady state, indicated by the constant drug disappearance rates, was reached after approximately 2–4 h with all subjects. For further data analysis only the six h values were used. Steady-state drug disappearance rates per area unit are referred to as "Drug flux of 50% saturation (Jmax 50%)" throughout this paper. Typical thermal thresholds versus time profiles for the investigated LA are shown in Figure 1. It is obvious from these curves that the investigated LA affect thermal thresholds to a different extent, with tetracaine and lidocaine being the most efficient LA. Interestingly, a similar observation has been made after topical application of several LA in an o/w cream base at an equimolar concentration instead of equal thermodynamic drug activity (McCafferty et al., 1988McCafferty D.F. Woolfson A.D. McClelland K.H. Boston V. Comparative in vivo and in vitro assessment of the percutaneous absorption of local anaesthetics.Br J Anaesth. 1988; 60: 64-69Crossref PubMed Scopus (39) Google Scholar). Lidocaine and tetracaine were more effective at producing local anesthesia to the challenge of insertion of a needle into the ventral forearm skin than benzocaine and fomocaine. The same tendency was observed with the steady-state flux values and permeability coefficients through Silastic membrane. According to the thermal thresholds versus time profiles in Figure 1, CS and WS appear to change linearly over time, a relationship also found with argon laser-induced cutaneous pain after application of EMLA cream (Bjerring and Arendt-Nielsen, 1990Bjerring P. Arendt-Nielsen L. A quantitative comparison of the effect of local analgesics on argon laser induced cutaneous pain and on histamine induced wheal, flare and itch.Acta Derm Venereol. 1990; 70: 126-131PubMed Google Scholar). Note that 30–60 min after tetracaine application six of the eight subjects developed erythema in combination with itching and edema at the application site; however, these known adverse effects of tetracaine, caused by vasodilation and histamine release (Willats and Reynolds, 1985Willats D.G. Reynolds F. Comparison of the vasoactivity of amide and ester local anaesthetics.Br J Anaesth. 1985; 57: 1006-1011Crossref PubMed Scopus (66) Google Scholar;Lawson et al., 1995Lawson R.A. Smart N.G. Gudgeon A.C. Morton N.S. Evaluation of an amethocaine gel preparation for percutaneous analgesia before venous cannulation in children.Br J Anaesth. 1995; 75: 282-285Crossref PubMed Scopus (96) Google Scholar), disappeared within 2–3 h after application and did not affect thermal thresholds. Table 1 provides an overview of the response parameters obtained from the individual curves of all eight subjects. Δ CS and CS slope appeared suitable parameters for the pharmacodynamic characterization of the LA, possibly because cold receptors are located in the epidermis (Bazett et al., 1930Bazett H.C. McGlone B. Brocklehurst R.J. The temperature in tissues which accompany temperature sensations.J Physiol. 1930; 69: 88-112Crossref PubMed Scopus (24) Google Scholar) and can easily be reached by LA. In contrast, warm receptors are located deeper in the dermis (Zotterman, 1959Zotterman Y. Thermal sensations.in: Field J Magoun HW Hall VE Handbook of Physiology. Washington DC, American Physiological Society1959: 431-458Google Scholar) and warmth- and heat-related response parameters such as the maximum change in WS and HP, as well as the WS slope, were not useful for characterization of the LA. HP appeared to be unaffected and the maximum WS in most cases correspond to the HP threshold plateaus, i.e., Δ WS does not differ significantly among the LA. A similar observation has recently been made with pramoxine lotion 1%, where no effect on WS and HP but a significant change in CP thresholds was found after application for 30 min (Yosipovitch and Maibach, 1997Yosipovitch G. Maibach H.I. Effect of topical pramoxine on experimentally induced pruritus in humans.J Am Acad Dermatol. 1997; 37: 278-280Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar). The fact that CS is mediated by myelinated Aδ fibers in contrast to WS, which is transmitted by unmyelinated and therefore better accessible C fibers, does not appear to affect the pharmacodynamic response.Table 1Overview of response parametersaLag time of onset of local anesthesia, rate of change in cold sensation (CS slope, displayed as positive values), and maximum change in cold sensation and cold pain (Δ CS, Δ CP) determined from the thermal thresholds versus time profiles of eight subjects. LA were applied to forearm skin at their maximum thermodynamic activity. Means ± SD, n = 8 subjects.FomocaineBenzocaineTetracaineLidocaineEtidocaineCarticaineLag time of onset [h]0.52 ± 0.070.61 ± 0.080.86 ± 0.111.01 ± 0.162.32 ± 0.222.30 ± 0.29CS slope [°C per min]0.136 ± 0.0310.051 ± 0.0080.341 ± 0.0410.189 ± 0.0190.006 ± 0.0040.087 ± 0.016Δ CS [°C]12.7 ± 2.34.5 ± 1.027.0 ± 3.516.1 ± 2.21.7 ± 0.46.9 ± 1.6Δ CP [°C]20.2 ± 2.27.7 ± 0.928.7 ± 2.623.4 ± 0.97.8 ± 0.55.5 ± 1.0a Lag time of onset of local anesthesia, rate of change in cold sensation (CS slope, displayed as positive values), and maximum change in cold sensation and cold pain (Δ CS, Δ CP) determined from the thermal thresholds versus time profiles of eight subjects. LA were applied to forearm skin at their maximum thermodynamic activity. Means ± SD, n = 8 subjects. Open table in a new tab CP thresholds are difficult to evaluate because with potent LA such as lidocaine and tetracaine the minimum instrument temperature of 0°C is reached, which makes it impossible to distinguish between these compounds. Moreover, instead of feeling cold pain many subjects felt a light pricking sensation after onset of anesthesia, which explains the high intrasubject variability in CP thresholds (Figure 1). The lag times of onset, differing between 0.5 and 2.3 h, have in the past been shown to be suitable response parameters only if one model drug is looked at (Le, 1993Le V.H. Einfluβ von Substanzeigenschaften auf Permeabilität und maximalen Flux von homologen Nicotinsäureestern in vitro und an der Haut in vivo. Heinrich Heine University, Düsseldorf1993Google Scholar); however, the values for lidocaine and tetracaine correspond to the recommended pretreatment periods for EMLA cream and Ametop gel, respectively. The maximum anesthetic effect is reached within 2–3 h with all LA. If the CS response data are correlated with physicochemical parameters such as LA lipophilicity, solubility in various solvents, degree of dissociation, protein binding, relative potency, or the ability to change the packing order of the stratum corneum lipids, a correlation may be observed only with the LA solubility in medium chain triglycerides (MCT), consisting of caprylic/capric acid triglycerides (Figure 2). Surprisingly, no correlation was found with the relative LA potency, even if a combined parameter such as the product of the LA potency and Jmax 50% was used. Etidocaine, a highly potent LA, affects thermal sensations only marginally, which may be attributed to the low drug flux through stratum corneum. Obviously LA penetration through the barrier stratum corneum is the rate-limiting step and determines the pharmacodynamic effect of percutaneously applied LA. A possible explanation for the linear relationship between the CS response data and the LA solubility in MCT is that MCT represents a solvent with properties similar to the stratum corneum lipids. If this is true, the response should also correlate linearly with Jmax 50%. In Figure 3, CS response data are plotted versus Jmax 50% of eticocaine, benzocaine, lidocaine, and tetracaine from light mineral oil as vehicle. The relationship between the CS response data and Jmax 50% is indeed linear, which supports the above-mentioned theory. The x axis intercepts of the regression lines in Figure 3 do not significantly differ from zero, in contrast to the plots in Figure 2, where x axis intercepts of about 0.3 mol per litre are obtained, indicating that apparently minimum solubility is required to observe an effect. A possible explanation for these intercepts is that the response data as well as the drug flux data not only depend on the solubility of the drug in the barrier but also on the drug diffusion coefficient. LA affect the structure of the lipids in biomembranes (Seeman, 1972Seeman P. The membrane actions of anesthetics and tranquilizers.Pharmacol Rev. 1972; 24: 583-655PubMed Google Scholar). Tetracaine and lidocaine are able to fluidize the lipid bilayers in the stratum corneum (Wolfson et al., 1991Wolfson A.D. McCafferty D.F. McGowan K.E. Differential scanning calorimetry of a novel amethocaine preparation and its effect on human stratum corneum.Proceedings of the 10th Pharm Technol Conf. 1991; 10: 405-423Google Scholar;Römmen et al., 1998Römmen C. Leopold C.S. Lippold B.C. Do local anesthetics have an influence on the structure of isolated human stratum corneum?.Perspectives Percutaneous Penetration. 1998; 6A: 113Google Scholar), a process that may lead to an increase of the drug diffusion in the barrier. Other anesthetics such as ketamine increase the packing order of the stratum corneum lipids, leading to a decrease in the diffusion coefficient (Römmen et al., 1998Römmen C. Leopold C.S. Lippold B.C. Do local anesthetics have an influence on the structure of isolated human stratum corneum?.Perspectives Percutaneous Penetration. 1998; 6A: 113Google Scholar). It may, therefore, be possible that the x axis intercepts in Figure 2 are the result of an increase in slope and a shift of the regression lines due to the different effects of LA on the packing order of the stratum corneum lipids and thus the diffusion coefficient in the barrier.Figure 3Pharmacodynamic response versus LA flux. Maximum change in cold sensation (Δ CS) and the rate of change in cold sensation (CS slope) versus Jmax 50%. Means ± SD, n = 6 subjectsView Large Image Figure ViewerDownload (PPT) Taken together it may be concluded that cutaneously applied LA are able to penetrate through human skin and affect thermal thresholds. Lag times of onset are short and the maximum anesthetic effect is reached within 2–3 h. CS parameters are the most suitable response parameters and they may be correlated with the solubility of the LA in MCT and Jmax 50%, indicating that MCT has similar properties with regard to LA solubility as stratum corneum lipids.