Local application of volatile anesthetics attenuates the response to a mechanical stimulus in humans

Transcription

1 GENERAL ANESTHESIA 951 Local application of volatile anesthetics attenuates the response to a mechanical stimulus in humans [L application locale d anesthésiques volatils diminue la réaction au stimulus mécanique chez les humains] Argyro Fassoulaki MD PhD DEAA,* Ioanna Skouteri MD,* Ioanna Siafaka MD,* Constantine Sarantopoulos MD MSc PhD DEAA Purpose: Volatile anesthetics may cause local hyperalgesia and/ or analgesia. This double-blind randomized study investigated the effect of these medications when applied locally on the response to a mechanical stimulus. Methods: In experiment 1, standard commercial preparations of halothane 1 ml, isoflurane 1.5 ml and sevoflurane 2.7 ml were randomly applied on the forearm of 30 volunteers for 30 min, after which the response to a mechanical stimulus was recorded. The other forearm received water as control. The next day, the experiment for each anesthetic was repeated in a reverse fashion. Thirty minutes after the application, the response to a standardized mechanical stimulus was recorded. In experiments 2 and 3, the response to the same mechanical stimulus was tested after local applications of 2, 4, and 6 ml of halothane or after a local application of 5 ml sevoflurane respectively. Results: Low doses of the three anesthetics did not alter the response to the mechanical stimulus (F = 3.055, df = 1,174, P = 0.082). Two, 4, and 6 ml of halothane attenuated the response to the mechanical stimulus by 36%, 27% and 29% respectively (F = 9.586, df = 1,114, P = 0.002). Five millilitres of sevoflurane attenuated the response to the mechanical stimulus by 36% (F = 5.111, df = 19, P < 0.001). Conclusion: Low liquid volumes of volatile anesthetics, when applied locally to the skin, did not influence the response to a mechanical stimulus, but higher volumes attenuated this response. Objectif : Les anesthésiques volatils peuvent causer une hyperalgésie et/ou analgésie locale. L étude randomisée à double insu a recherché l effet de ces médicaments, appliqués localement, sur la réaction au stimulus mécanique. Méthode : Pour la première expérience, des préparations régulières de 1 ml d halothane, 1,5 ml d isoflurane ou de 2,7 ml de sévoflurane ont été appliquées au hasard sur l avant-bras de 30 volontaires pendant 30 min, après quoi la réaction à un stimulus mécanique a été enregistrée. On a appliqué de l eau sur l autre avant-bras. Le jour suivant, l expérience a été répétée pour chaque anesthésique de manière inversée. Trente minutes après l application, la réaction au stimulus mécanique normalisé a été notée. Dans les expériences 2 et 3, la réaction au même stimulus mécanique a été testée après l application locale de 2, 4, et 6 ml d halothane ou de 5 ml de sévoflurane, respectivement. Résultats : Les doses faibles des trois anesthésiques n ont pas modifié la réaction au stimulus mécanique (F = 3,055, df = 1,174, P = 0,082). Deux, 4, et 6 ml d halothane ont diminué la réaction au stimulus mécanique de 36 %, 27 % et 29 % respectivement (F = 9,586, df = 1,114,P = 0,002). Cinq millilitres de sévoflurane ont diminué la réaction au stimulus mécanique de 36 % (F = 5,111, df = 19, P < 0,001). Conclusion : De faibles volumes liquides d anesthésiques volatils, appliqués localement sur la peau, n ont pas d influence sur la réaction à un stimulus mécanique, mais de plus grands volumes diminuent cette réaction. From the Department of Anesthesiology,* Aretaieio Hospital, Medical School, University of Athens, Athens; and the Department of Anesthesia, Medical College of Wisconsin, Milwaukee, USA. Address correspondence to: Dr. Argyro Fassoulaki, Raftopoulou street, Athens, Greece. Phone: ; ; ; Fax: ; fassoula@aretaieio.uoa.gr; afassou1@otenet.gr Support was provided solely from institutional sources and no conflict of interest is involved. Accepted for publication May 6, Revision accepted June 10, CAN J ANESTH 2005 / 52: 9 / pp

2 952 CANADIAN JOURNAL OF ANESTHESIA THE analgesic effects of volatile anesthetics at various concentrations have been investigated extensively. Methoxyflurane, trichloroethylene and isoflurane have all been used to provide inhalational analgesia in obstetrics. 1,2 The effect of methoxyflurane and halothane at subanesthetic concentrations on the pain threshold has also been documented in awake volunteers. 3 In rats, halothane inhibits the inhibitory and excitatory responses of the somatosympathetic reflex, suggesting a potential analgesic effect. 4 On the other hand, very low concentrations of halothane increase the analgesic effect of intrathecal morphine, but increase the sensitivity to a noxious thermal stimulus and reduce the analgesic effect of intraventricular morphine. 5 In general, studies investigating the analgesic effects of volatile anesthetics provide controversial results. When these anesthetics are administered via the lungs, their potential analgesic effect is difficult to distinguish from the concomitant sedation which they produce. In a previous study, we investigated in volunteers the analgesic effects of isoflurane when applied locally in the forearm. 6 This methodology distinguishes between a decreased response due to sedation, and a decreased response due to analgesia. The aim of the present study was to investigate the effects of topically administered volatile anesthetics halothane, isoflurane and sevoflurane to a standardized mechanical stimulus, after skin application in three different doses. Materials and methods Subjects The study was approved by the local Ethics Committee, and written informed consent was obtained from all participants. Healthy volunteer male and female subjects were recruited for the study. Participants were residents and nurses working in departments other than anesthesia, and were not aware of the purpose of the study. Subjects with body weight exceeding the 20% of their ideal, allergic to cotton, metal or plastic or allergic to previous anesthetics, with acute or chronic pain condition, and consuming analgesics, calcium channel blockers or beta-blockers within one month of recruitment were excluded from the study. To exclude pregnant participants, women were specifically asked about contraception and menstrual cycle history. Experiments EXPERIMENT 1 Thirty volunteers were exposed to the anesthetic liquid or to an equal volume of water in a crossover randomized, double-blinded manner. Three volatile anesthetics, halothane (Zeneca Ltd, Maccfield, UK), isoflurane (Abbott, Kent, UK) and sevoflurane (Abbott Kent, England, UK), at equipotent doses 1 ml, 1.5 ml and 2.7 ml were tested for analgesia produced after application on the skin. On the contralateral forearm, equal volumes of water were applied. The volumes of the three different volatile agents were chosen to be equipotent in terms of minimal alveolar concentration (MAC). Since MAC of halothane is 0.75 and isoflurane is 1.15, it appears that isoflurane is less potent by 1.15: 0.75 = 1.5 times, so we applied locally 1 ml of halothane and 1.5 ml of isoflurane. Similarly, sevoflurane with a MAC 2 is 2: 0.75 = 2.7 times less potent than halothane (MAC = 0.75). Therefore the equipotent volumes of isoflurane and sevoflurane corresponding to 1 ml of halothane are 1.5 and 2.7 ml respectively. Allocation sequence was determined by use of sealed envelopes, to specify which anesthetic should be tested first, second or last. The envelopes were opened, and the order of exposure to each anesthetic was recorded by an independent anesthesiologist who prepared the syringes with the liquid anesthetic and water. The tests for each anesthetic were conducted at two day intervals to ensure complete washout between interventions. EXPERIMENT 2 Twenty volunteers were recruited. Their forearms were exposed in a cross-over random double-blind manner to receive 2, 4 and 6 ml of liquid halothane using equal volumes of water as control for the other forearm. Randomization was done using a sealed envelope method. Again, randomization and preparation of the syringes were done by an independent anesthesiologist. The experiments for the different doses were conducted two days apart. EXPERIMENT 3 Twenty volunteers were scheduled to receive on their forearms 5 ml of liquid sevoflurane or 5 ml of water in a cross-over manner. Mechanical stimulus The effect of application of volatile anesthetics vs water on the volar surface of the forearms was tested against a mechanical stimulus. As in previous studies, 6,7 we used a pressure palpator with a stimulation area similar to the distal cross sectional area of a thick ball-pen (pressure FEELER 650 g Sedatelec; Irigny, France). The pressure probe, when applied firmly on a surface, exerts a standardized mechanical painful

3 Fassoulaki et al.: VOLATILE ANESTHETICS AND NOCICEPTION 953 stimulus of 650 g. The palpator was applied firmly to the middle volar aspect of each forearm for 20 sec, and the visual analogue scale (VAS) scores for the right and left forearm were recorded. Procedure All subjects who participated in experiments 1, 2, or 3 were exposed to the test without anesthetic one to two days earlier, to limit the influence of learning the test. Subjects used in one experiment did not participate in either of the other two experiments. Experiments were performed in a quiet room with the volunteer in a sitting position, and with the forearms on a table exposing the volar surface. The order of the right vs the left arm to be tested was determined after tossing a coin, heads for the right, and tails for the left forearm to be tested first. The forearms were wrapped with a cotton band and covered with drapes. An independent anesthesiologist prepared the syringes with the anesthetic or water, and after tossing a coin marked each syringe as left or right accordingly. The investigator who applied the liquid anesthetic or water, and the volunteer were wearing masks during the liquid application to prevent them from recognizing which arm was treated with the anesthetic. In experiments 1 and 2 both the investigator and the test subject were blinded to the anesthetic to be tested and to the anesthetic vs water. Regarding the different volumes of fluids the volunteer but not the investigator was blinded to the volume of the fluid injected. The assessment of the response to the stimulus 30 min after fluid injection was done by another investigator who participated in the study. Ten-millilitre syringes were used for all the experiments. The quantity of the liquid anesthetic or water was emptied simultaneously in the space between the volar surface of the forearms and the cotton band, and both forearms were wrapped with aluminum foil from the wrist to the elbow joint. The aluminum foil was covered with transparent film and all layers were fixed with tape around the elbow and wrist, with the elbow joint left free. Treatment lasted for 30 min, after which the plastic film, the aluminum foil and the cotton were removed. The forearms were then exposed to the mechanical stimulus and the VAS scores were recorded. Temperature probes were then applied on the surface of each forearm, and skin temperatures were recorded. The experiment for the same dose of the same anesthetic was repeated the next day in a cross-over manner. Thus, the forearm treated previously with the anesthetic was now exposed to water and vice-versa. Different anesthetics, or different doses of the same TABLE I Demographics of volunteers in experiments 1, 2, and 3 Demographics Experiment 1 Experiment 2 Experiment 3 Age (yr) 33 ± ± ± 4.2 Body Weight (kg) 66 ± ± ± 13.2 Height (cm) 170 ± ± ± 7.9 Males/females 15/15 11/9 10/10 Values are mean ± standard deviation. anesthetic were performed two days apart from the previous drug or dose. In each experiment the VAS scores obtained from the right and the left forearm treated with anesthetic vs water were averaged and compared. Statistical analysis Sample size estimation for two-way analysis of variance was based on the instructions provided by the Data analysis and biostatistics software and resources, GraphPad.com, GraphPad Software (San Diego, CA, USA). This estimation showed that approximately 30 subjects were required in experiment 1 and 20 subjects in experiment 2, in order to ensure a power of 0.80 of detecting 40 to 50% changes of the mean VAS scores after mechanical stimulation. An alpha error was assumed 0.05 and standard deviation (SD) of approximately 18 to 22 were estimated from initial pilot observations and a previous publication. 6 Sample size estimation for the third experiment, performed using the GB-StatTM PPC software for Macintosh, showed that approximately 20 subjects were required in order to detect at least a 40% change from the mean VAS scores after mechanical stimulation. An alpha error of 0.05 was assumed, and a SD of approximately 20 was estimated from initial pilot observations. The Kolmogorov-Smirnov test indicated that all variables followed normal distribution. VAS and temperature values were compared using two-way ANOVA in experiments 1 and 2, and paired Student s t tests in the third experiment. Paired Student s t tests were also employed for pair-wise comparisons of VAS values between each anesthetic treatment and its water-control measurements wherever appropriate. Student s t and ANOVA have been shown to be appropriate tests for statistical comparisons of VAS scores. 8 P < 0.05 was considered significant. We used the Microsoft Excel X, SPSS 11, and GB-Stat TM PPC software for Macintosh.

4 954 CANADIAN JOURNAL OF ANESTHESIA FIGURE The flow diagram for the three experiments. Results Demographics of the volunteers who participated in each experiment, as well as the male/female ratio are shown in Table I. In experiment 2, two volunteers had some local irritation of the skin after application of 6 ml of halothane. However, this was the last dose for them, so they completed the tests for the three doses of halothane. A flow diagram of the three experiments is shown in the Figure. Experiment 1 Local application of 1 ml of halothane was associated with a numerical decrease in the VAS response to the mechanical stimulus of 29%, vs a 10% reduction recorded after local application of 1.5 ml of isoflurane or 2.7 ml of sevoflurane (Table II). We found no statistically significant differences between the three anesthetics, (multifactorial ANOVA corrected model F = 0.977, df = 5,174, P = 0.433) neither significant differences between the three anesthetics and water applications (F = 3.055, df = 1,174, P = 0.082) nor between the three anesthetics themselves (F = 0.543, df = 2,174, P = 0.582). The VAS values for each anesthetic and the corresponding water volume are shown in Table II. The temperatures measured on the forearms immediately after recording the responses to the mechanical stimuli did not differ between the exposures to the three anesthetics and water (Table V). Experiment 2 Local application of 2, 4 or 6 ml of halothane as described by the study protocol produced 36%, 27% and 29% decrease in the VAS response to the mechanical stimulus when compared with water-control appli-

5 Fassoulaki et al.: VOLATILE ANESTHETICS AND NOCICEPTION 955 TABLE II VAS values (mm) obtained after application of the mechanical stimulus when forearms were treated with 1 ml of HAL, 1.5 ml of ISO and 2.7 ml of SEVO vs 1, 1.5 or 2.7 ml of water (experiment 1) VAS HAL Water ISO Water SEVO Water (mm) 1 ml 1 ml 1.5 ml 1.5 ml 2.7 ml 2.7 ml Mean ± SD 27 ± ± ± ± ± ± % CI % vs water 29% 10% 10% VAS = visual analogue scale, HAL = halothane, ISO = isoflurane, SEVO = sevoflurane, CI = confidence intervals; % = % difference in the VAS values between the anesthetic and equal volume of water; F = 3.055, df = 1,174, P = TABLE III VAS values (mm) obtained after application of the mechanical stimulus when forearms were treated with 2, 4, and 6 ml of HAL, vs 2, 4, and 6 ml of water (experiment 2) VAS HAL Water HAL Water HAL Water (mm) 2 ml 2 ml 4 ml 4 ml 6 ml 6 ml Mean ± SD 34 ± 19.2* 47 ± 19.6* 38 ± ± ± ± % CI % vs water 36% 27% 29% VAS = visual analogue scale, HAL = halothane, CI = confidence intervals; % = % difference in the VAS values between the anesthetic and equal volume of water; F = 9.586, df = 1,114, P = 0.002; *t = 3.753, df = 19, P = ; t = 3.959, df = 19, P < 0.001; t = 4.062, df = 19, P = < TABLE IV VAS (mm) obtained after application of the mechanical stimulus when forearms were treated with 5 ml of SEVO, vs 5 ml of water (experiment 3) VAS (mm) SEVO 5 ml Water 5 ml Mean ± SD 26 ± ± % CI % vs water 36% VAS = visual analogue scale, SEVO = sevoflurane, CI = confidence intervals; % = % difference in the VAS values between the anesthetic and equal volume of water; t = 5.111, df = 19, P < cations. Multifactorial ANOVA (corrected model F = 2.090, df = 5,114, P = 0.072) showed a significant difference regarding halothane effect vs water (F = 9.586, df = 1,114, P = 0.002) but this effect was not dose-dependent (F = 0.404, df = 2,114, P = 0.668). The VAS values were 34 ± 19.2 vs 47 ± 19.2 (P = ), 38 ± 17.8 vs 48 ± 24.2 (P < 0.001) and 39 ± 17.1 vs 50 ± 20.9 (P < 0.001) for the 2, 4, and 6 ml of halothane vs equal volumes of water (Table III). The forearm temperatures recorded after completion of VAS score measurements did not differ (Table V). Experiment 3 Application of 5 ml sevoflurane, as described previously, decreased the VAS values after the mechanical stimulus by 36% (t = 5.111, df = 19, P < 0.001). VAS values recorded were 26 ± 15.7 for the sevoflurane and 36 ± 16.1 for water application (Table IV). Temperatures recorded on the forearms were higher for the sevoflurane applications vs water (t = 3.805, df = 19, P = 0.001), (Table V). Discussion Our results indicate that local application of higher liquid volumes of halothane and sevoflurane but not low volumes of halothane, isoflurane and sevoflurane attenuate the response to a mechanical stimulus. Also, the high dose of sevoflurane application increased the skin temperature when compared with water application. Various studies have examined the antinociceptive effects of the volatile anesthetics when administered via the lungs. In rats, 0.1 MAC of halothane, isoflurane, nitrous oxide or diethyl ether produced hyperalgesia, but at 0.4 to 0.8 MAC produced analgesia to a thermal stimulus. 9 The antanalgesic effect of subanesthetic concentrations of isoflurane in rats has not been reconfirmed in humans. 10 On the contrary, the results of the present study suggest that various doses of halothane and the higher dose of sevoflurane, when applied on the skin as liquid, attenuate the response to the mechanical stimulus. The suppression of sodium and activation of potassium channels by the volatile agents may be involved in their analgesic effect. In fact, halothane and isoflurane suppress central and peripheral mammalian sodium channels. 11

6 956 CANADIAN JOURNAL OF ANESTHESIA TABLE V Skin temperatures ( C) recorded in each arm after visual analogue scale recording and average for both arms Experiment 1 HAL Water ISO Water SEVO Water 1 ml 1 ml 1.5 ml 1.5 ml 2.7 ml 2.7 ml Temp ( C)* 27.9 ± ± ± ± ± ± 1.9 Experiment 2 HAL Water HAL Water HAL Water 2 ml 2 ml 4 ml 4 ml 6 ml 6 ml Temp ( C) 27.5 ± ± ± ± ± ± 1.9 Experiment 3 SEVO Water 5 ml 5 ml Temp ( C) 29.0 ± ± 1.5 HAL = halothane, ISO = isoflurane, SEVO = sevoflurane. Values are mean ± SD. *F = 1.142, df = 1,174, P = 0.287; F = 0.660, df = 1,114, P = 0.418; t = 3.805, df = 19, P = Calcium channels may also play a role. Isoflurane and halothane suppress the current of T-calcium channels in the dorsal root ganglion neurons and in the adrenal glomerulosa cells. 12 Excitatory synaptic transmission involves calcium T-channels, which may be suppressed by volatile anesthetics. Inhalational anesthetics also activate subunits of voltage dependent potassium channels. Increased conductance of potassium channels results in less excitable neurons and may provoke changes in voltage gated sodium channels. 13 Most of the research is orientated towards the effect of volatile anesthetics on the central nervous system voltage gated sodium or potassium channels. However, possible effects of these anesthetics on the peripheral nervous system may also result in antinociception. Recently, adenosine triphosphate sensitive potassium channels have been discovered in mammalian primary afferent neurons. 14 The attenuated response to a mechanical stimulus produced by halothane and sevoflurane in this study may involve an action on the sodium, potassium, and/or calcium channels. In our study, local application of 1 ml halothane was associated with a lower response to the mechanical stimulus by 29%, vs water, almost three times the attenuation obtained after equipotent doses of isoflurane or sevoflurane. However, as the effect of the other two anesthetics, isoflurane and sevoflurane tested in the same volunteers was small, the difference between the liquid anesthetics and water controls using ANOVA was not significant. In experiment 2, the three doses of locally applied liquid halothane significantly decreased the response to mechanical stimulus by 27 to 36% when compared with equal volumes of water. This attenuation is not dose-dependent. A possible explanation is the higher solubility of halothane, therefore its better penetration through the tissues for all the doses we tested. Similar attenuation (36%) to the mechanical stimulus was observed after application of 5 ml sevoflurane, which is roughly equipotent to 2 ml halothane. In a previous study, we found that 10 ml of isoflurane, which when normalized for MAC, is roughly equipotent to 6 ml of halothane, produced a significant decrease in the response to the same mechanical stimulus by 19% vs water. This was lower than that produced by an equipotent dose of halothane (29%). 6 However, the mechanical stimulus was applied for 30 instead of 20 sec, as occurred in the present study. Either isoflurane exerts a less intense local analgesic effect compared with halothane, or the longer duration of stimulation may have caused temporary hyperalgesia, attenuating the response to the stimulus after both anesthetic and water application. Our results regarding the response to the mechanical stimulus are different from those of a previous study, where subanesthetic concentrations of isoflurane administered via the lungs did not alter the response to mechanical pressure. 10 Such differences may be due to a different study design. Prolonged pressure application may result in temporary hyperalgesia in the area. A fixed predetermined mechanical stimulus as used in our study may be more appropriate as it applies equally to all subjects, although it does not allow pain threshold determination. Also, we are unable to compare the skin concentration of isoflurane after local vs systemic application. 15 The significantly higher skin temperature measured in the high volume sevoflurane might suggest a vasodilating effect of the anesthetic when applied locally. The VAS values obtained after water application were higher in experiment 2 than in experiments 1 and 3. The difference in VAS scores obtained after water application may be due to individual variabil-

7 Fassoulaki et al.: VOLATILE ANESTHETICS AND NOCICEPTION 957 ity. Differences in solubility of the volatile anesthetics most likely contributed to a more pronounced effect of the low volume of halothane vs equipotent volumes of isoflurane and sevoflurane. The percutaneous losses of halothane have been found six and two times greater than those of desflurane and isoflurane respectively. 16 In conclusion, local application of various doses of halothane, and sevoflurane attenuated the response to a mechanical stimulus with higher drug volumes. The most soluble agent halothane exhibited the most potent local anesthetic effect. References 1 Major V, Rosen M, Mushin WW. Methoxyflurane as an obstetric analgesic: a comparison with trichloroethylene. Br Med J 1966; 2: Abboud TK, Gangolly J, Mosaad P, Crowell D. Isoflurane in obstetrics. Anesth Analg 1989; 68: Siker ES, Wolfson B, Ciccarelli HE, Telan RA. Effect of subanesthetic concentrations of halothane and methoxyflurane on pain threshold in conscious volunteers. Anesthesiology 1967; 28: Samso E, Farber NE, Kampine JP, Schmeling WT. The effects of halothane on pressor and depressor responses elicited via the somatosympathetic reflex: a potential antinociceptive action. Anesth Analg 1994; 79: Drasner K. Low concentrations of halothane increase response to a noxious thermal stimulus and attenuate the antinociceptive effect of intraventricular but not intrathecal morphine. Anesthesiology 2001; 94: Fassoulaki A, Sarantopoulos C, Karabinis G, Derveniotis C. Skin application of isoflurane attenuates the responses to a mechanical and an electrical stimulation. Can J Anaesth 1998; 45: Fassoulaki A, Sarantopoulos C, Zotou M, Karabinis G. Assessment of the level of sensory block after subarachnoid anesthesia using a pressure palpator. Anesth Analg 1999; 88: Dexter F, Chestnut DH. Analysis of statistical tests to compare visual analog scale measurements among groups. Anesthesiology 1995; 82: Zhang Y, Eger EI II, Dutton RC, Sonner JM. Inhaled anesthetics have hyperalgesic effects at 0.1 minimum alveolar anesthetic concentration. Anesth Analg 2000; 91: Petersen-Felix S, Arendt- Nielsen L, Bak P, et al. Analgesic effect in humans of subanaesthetic isoflurane concentrations evaluated by experimentally induced pain. Br J Anaesth 1995; 75: Duch DS, Rehberg B, Vysotskaya TN. Volatile anesthetics significantly suppress central and peripheral mammalian sodium channels. Toxicol Lett 1998; : McDowell TS, Pacrazio JJ, Barrettt PQ, Lynch C III. Volatile anesthetic sensitivity of T-type calcium currents in various cell types. Anesth Analg 1999; 88: Franks NP, Lieb WR. Volatile general anaesthetics activate a novel neuronal K+ current. Nature 1988; 333: Sarantopoulos C, McCallum B, Sapunar D, Kwok WM, Hogan Q. ATP-sensitive potassium channels in rat primary afferent neurons: the effect of neuropathic injury and gabapentin. Neurosci Lett 2003; 343: Eckenhoff RG, Johansson JS. On the relevance of clinically relevant concentrations of inhaled anesthetics in in vitro experiments. Anesthesiology 1999; 91: Fassoulaki A, Lockhart SH, Freire BA, et al. Percutaneous loss of desflurane, isoflurane, and halothane in humans. Anesthesiology 1991; 74: