- Email: [email protected]
Effects on Skin Blood Flow by Provocation during Local Analgesia
Microvascular Research 59, 122–130 (2000) doi:10.1006/mvre.1999.2205, available online at http://www.idealibrary.com on
Effects on Skin Blood Flow by Provocation during Local Analgesia M. Arildsson, G. E. Nilsson, and T. Stro¨mberg Department of Biomedical Engineering, Linko¨ping University, University Hospital, SE-581 85 Linko¨ping, Sweden Received May 18, 1999
Although topical analgesia cream has been used for several years, little is known about its effects on the microcirculation. Previous studies have shown a vasoconstrictive effect after short application times and a vasodilatation after longer application. It has also been shown that vasomotion does not occur in the analgesized skin. The present study was undertaken to investigate the alterations in skin blood perfusion following local cooling, local heating and pin-pricking after the establishment of analgesia. In 11 healthy volunteers, skin analgesia was attained by use of a eutectic mixture of lidocaine and prilocaine (EMLA, Astra Pain Control AB, Sweden) applied to the skin three hours prior to provocation. The changes in skin blood perfusion, after applying three different provocation methods, were studied using the laser Doppler technique. Local cooling and heating to temperatures of 110 and 145°C, respectively, were applied for 9 s by use of a copper probe (Ø12 mm). In the pin-prick provocation method, a combined effect of deflection and penetration of the skin to in total 3 mm was attained. Identical provocation methods were applied to placebo treated and untreated skin areas. After heat provocation, significant differences in the perfusion response between the treatments were seen (P < 0.0001). Skin areas treated with analgesia cream responded with a slow increase in perfusion that persisted beyond the four minute measurement period. Placebo and untreated areas decreased their perfusion over time. After cooling a significant reduction in skin perfusion was seen, irrespective of the
122
treatment. Similarly, after pin-pricking a perfusion increase was seen for all treatments. The findings indicate that topical analgesia influences the myogenic control of the blood flow in those vascular plexa measured by laser Doppler following heat provocation. No differences could be seen in the response to pin-pricking and cooling for the different treatments. © 2000 Academic Press Key Words: laser doppler flowmetry; LDPI; EMLA; analgesia; skin microcirculation; heat stimuli; pin-prick.
INTRODUCTION Analgesia cream has been used for a number of years for pain relief during superficial clinical and surgical procedures such as venopuncture, split skin grafting, skin biopsies, and treatment of port wine stains (Santacana et al., 1993; Manner et al., 1987; Juhlin et al., 1990). In particular, analgesia cream is used during needle insertion through intact skin in children. The predominantly used topical analgesia cream, EMLA (Astra Pain Control AB, Sweden), is a mixture of the substances lidocaine and prilocaine. EMLA causes a biphasic vascular response comprising initial blanching and vasoconstriction (maximal after 1.5 h of application) and late erythema and vasodilatation at application times longer than 3 h (Bjerring et al., 1989). These local circulatory changes have been suggested 0026-2862/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
123
Effects on Skin Blood Flow by Provocation
to be due to the direct effect of local anesthetics on vascular smooth muscle, producing vasoconstriction at low concentrations and vasodilatation at high ones (Tucker, 1986). The analgesia onset and vascular response are faster in patients with eczematous skin than in healthy subjects as a result of increased penetration through abnormal skin (Juhlin et al., 1989). Salerud et al. (1983) showed that the rhythmical variations in skin blood flow, i.e., vasomotion, are locally regulated and could not be reduced by afferent nerve blockade. Local analgesia by EMLA treatment for 1 h, however, abolished the vasomotion leaving mean skin blood flow unchanged. Wårdell et al. (1993a) showed that EMLA eliminated the normal axon reflex mediated vasodilatation during electrical skin stimulus. Apart from the latter study, little is known about the vascular response to local skin stimulation during topical application of analgesia. The aim of the present study is therefore to investigate the response pattern, and thereby the regulation, of skin blood flow after topical application of EMLA, using local heating, cooling, and pin-pricking as stimuli.
MATERIALS AND METHODS Laser Doppler Perfusion Imaging (LDPI) The laser Doppler technique was developed during the mid seventies (Stern, 1975; Nilsson et al., 1980a, 1980b) as a method for noninvasive measurement of tissue blood perfusion (defined as the concentration multiplied by the average velocity of the moving blood cells). The method is based on the frequency shift photons undergo when scattered by moving blood cells, i.e., the Doppler effect. Laser Doppler Perfusion Imaging (LDPI) uses a freely impinging laser beam to generate an image of the tissue perfusion (Wårdell et al., 1993b). This image describes the spatial variations in perfusion due to the vascular structure in the measurement area (Braverman et al., 1991, 1992; Tenland et al., 1983). By continuously recording small images (duplex recording), one may also assess the
temporal changes in the blood flow (Wårdell et al., 1996). In this study a modified laser Doppler imager (PIM version 1.0 from Lisca AB, Sweden) has been used. This scanner records images of a size up to 12.5*12.5 cm (with the scanner head positioned 15 cm above the measurement surface) using at the most 64*64 measurement sites. The scanner was modified to also include a signal processing filter that allows the concentration of moving blood cells (CMBC) to be recorded (Arildsson et al., 1997). By recording the CMBC signal one may also estimate the average velocity, by dividing the perfusion estimate by the CMBC.
The Dermal Sensitivity Tester The dermal sensitivity tester (desensor) (Cenova AB, Sweden) consists of a computerized system, which delivers standardized preset skin stimuli to a defined area of the skin. The stimuli delivered by the desensor are pin-prick, cold, heat, and touch. Pain is provoked by pricking the skin with a dental needle (27 G) using a standardized depth of 3.0 mm. Cold and heat stimuli are induced by the use of metal probes that are brought into contact with the skin. The application time, as well as the temperature of the probe, can be set by the operator. Each type of stimuli is applied on a separate skin location, which can be determined with high precision. Figure 1 shows the set-up of the desensor system.
Skin Temperature The skin temperature was recorded using an emissivity thermometer, THI-300 (Tasco, Japan).
Placebo Cream The placebo cream was provided by Astra Pain Control, and contains coconut cream instead of the active substances. The pH of the placebo is the same as the pH of the EMLA cream.
Statistical Analysis A statistical model suitable for explaining alterations in the duplex recordings of skin perfusion (P)
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
124
Arildsson, Nilsson, and Stro¨mberg
There will not be an effect due to interaction: H0, (tb)ij 5 0 for all i, j; H1, (tb)ij Þ 0 for some i, j. The recorded images were used to analyze the effects of the analgesia and placebo creams on the perfusion and concentration of moving blood cells. For this purpose baseline recordings and the recordings made after the selected application time and stimulus were compared. This comparison were made using Student’s t test on the calculated averages of the individual images.
Experimental Set-Up
FIG. 1. The set-up of the dermal sensitivity tester (desensor).
due to a stimulus (local heating, cooling, or pin-pricking) includes a time effect (b j, j 5 1, 2) accounting for the early and late perfusion response, and a treatment effect (t i, i 5 1, 2, 3, corresponding to EMLA, placebo, and reference). Also a time-treatment interaction effect (tb) ij was included in the statistical analysis. Hence a 2-way ANOVA mode repeated measurements design (Montgomery, 1997) was selected: P 5 m 1 t i 1 b j 1 ~ tb ! ij 1 g k 1 « ijk,
(1)
where m is the mean perfusion, g k denotes the effects of the subject (k 5 1, 2, . . . , 11) (random effect), and « ijk is the model error. The following three null hypotheses were put forth. The different treatments will not affect the perfusion: H0, tI 5 0 for all i; H1, tI Þ 0 for at least one i. There will not be an effect due to time: H0, bj 5 0 for all j; H1, bj Þ 0 for at least one j.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
All experiments were made with the subject’s informed consent and were also approved by the Ethics Committee for Human Research at Linko¨ping University Hospital, Sweden. Prior to each experiment, all subjects were acclimatized to the room conditions for 20 min. During the acclimatization period one of the subject’s forearms was randomly chosen for measurement. The forearm skin was divided into three different areas. These three areas were randomly chosen to be treated with EMLA, placebo, or to be used as a reference area. Further, each of the areas was divided into three subareas, in which the different stimuli were to be applied; these areas were marked using ordinary ink. Since the heat probe and the cooling probe are circular in shape, the markings were made circular with diameters of 12 mm. The subareas were assigned different stimuli including pin-prick (pin-prick depth 3.0 mm), local heating (45°C, 9 s) and local cooling (10°C, 9 s), at random. A typical layout of an experiment is shown in Fig. 2. The forearm of the subject was fixated firmly in the Astra desensor using a vacuum pillow. This fixation kept the artifacts, due to movements of the arm during the recording of laser Doppler images and signals, at a minimum. After fixation, the unprovoked baseline perfusion of the areas chosen to be used for application of EMLA and placebo were recorded. These baseline recordings
125
Effects on Skin Blood Flow by Provocation
FIG. 2. Layout of the experiments, comprising three main areas that are divided into three subareas each. Each main area was treated using either EMLA or placebo or was used as a reference area. The subareas were subjected to the different stimuli.
consisted of image scans (size 16*16 measurement sites, approximately 1 cm 2) of all subareas, and a duplex scan (4 min, 2*2 probe) in one subarea in each of the application areas. At this point no measurements were made in the reference area. Also, the EMLA and placebo cream (Astra Pain Control, Sweden) were applied and covered with Tegaderm (3M, Canada). The average amount of EMLA applied was 1.6 6 0.2 mg and the average amount of placebo 2.0 6 0.2 mg, covering approximately 10 cm 2. The application of EMLA and placebo was spaced in time approximately 25 min apart, which was sufficient to perform all recordings in one application area prior to removing the cream in the second area, thus maintaining a three hour application in both areas. When the creams had been applied, the measurements on the area selected as reference were made. First all subareas were investigated, two of the subareas using the standard method which records an image, size 16*16, and one subarea using a duplex scan (4 min, 2*2 probe), followed by an image recording, size 16*16 pixels. The subareas were then challenged using the different stimuli. After each provocation a 4-min duplex signal was measured, followed by a 16*16 scan. In order to record the blood flow response as accurately as possible, the laser beam was directed to the center of the stimulated area, i.e., the pin-prick point or the center of the area provoked by the two different probes. The time to complete a full subarea measurement was 25 min. The reason for not also using duplex recordings in all subareas before provocation was that this would not be feasible; the time to complete a measurement would be too long.
After 3 h of application, the measurements made on the reference area were repeated on the areas treated with EMLA and placebo, respectively. In addition, the skin temperature was recorded, both prior to application, prior to provocation, and after provocation in all nine different subareas. This study comprises 12 healthy, caucasian subjects, 7 male and 5 female, with ages ranging from 16 to 35 years. All subjects were nonsmokers and refrained from drinking coffee or using any type of caffeine prior to the experiment. This will ascertain that the local skin blood flow was not affected by these types of stimuli. One male subject was excluded from the evaluation set, because of illness during the test. A total of 11 subjects was therefore used in the statistical analysis. The temperature of the room was 24 –26°C.
RESULTS Typical duplex signals recorded immediately after heat provocation in a reference area are shown in Fig. 3, while Fig. 4 shows a typical perfusion response from an area on to which EMLA has been applied topically. The recorded signals in the placebo and the reference area generally display an initial increase in the blood perfusion followed by a rapid normalization of the flow. In the area treated with EMLA this temporary increase in perfusion following heat stimulus was not seen. Instead a delayed increase in perfusion was seen which remained until the end of the mea-
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
126
Arildsson, Nilsson, and Stro¨mberg
FIG. 3.
Typical concentration (a) and perfusion (b) responses from untreated skin after local heating (9 s, 45°C).
surement, approximately 4 min after the stimulus was applied. This type of reaction was seen in both the CMBC-signal and in the perfusion signal. The response was calculated as the average perfusion during approximately 30 s, within two time windows of which the first started 30 s after provocation, and the second started at 3 min after provocation. These time windows are shown in Fig. 4. The delay between the stimulus and the first time window was chosen in such a way that the initial increase seen in normal responses was given time to fade. By use of this approach it was also possible to minimize the influence of differences in the time gap between the stimulation and the start of the recording. Figure 5 shows the results of evaluating these responses in the model described by Eq. (1). Figure 5 shows the time and time-treatment effects for the three different provocations (pin-prick, cooling, and heating) when measuring perfusion. In all treatments pinprick (Fig. 5a) increased the perfusion (time effect, P , 0.01), while cooling (Fig. 5b) resulted in a decrease (time effect, P , 0.025). For heating (Fig. 5c) there is
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
a difference between the treatments in that the reference areas and the placebo treated areas showed a reduced perfusion, while in the areas treated with EMLA a perfusion increase (time-treatment effect, P , 0.0001) was observed. In the images recorded pre- and postheat challenge in the EMLA-treated areas, the backscattered light intensity was found not to have changed (preheat image average 8.94 6 0.38 A.U., postheat 9.06 6 0.27 A.U., P . 0.3). Similarly, the light intensity did not change due to the 3-h EMLA application (preapplication image average 9.10 6 0.19 A.U., preheat 8.94 6 0.38 A.U., P . 0.1). In Table 1 the mean of the recorded images for the different stimuli and treatments is presented along with the results obtained using Student’s t test, comparing the baseline recording with the recordings after delivered stimuli. No significant changes in perfusion, concentration, or velocity of blood cells due to local skin analgesia were found, i.e., no changes could be seen between the recorded baseline images and the preheat images in the EMLA area. Table 1 also shows
127
Effects on Skin Blood Flow by Provocation
FIG. 4. Typical concentration (a) and perfusion (b) responses from skin treated with EMLA (3-h application) after local heating (9 s, 45°). Also shown are the time windows used for evaluation.
the mean of the recorded images after applying a heat stimulus. These values yield similar results as from the duplex-recordings, showing a significant increase in perfusion (138%, P , 0.01), concentration (54%, P , 0.001), and velocity (42%, P , 0.05) after heat stimulus in areas under local anesthesia. In areas treated with the placebo cream a slight decrease in both perfusion (17%, P , 0.05) and average velocity (19%, P , 0.05) was seen (baseline vs preheat). No significant change was seen after heat stimulus. The individual results obtained for the 11 subjects are listed in Table 2 as perfusion changes between pre- and postheat images. The pin-prick stimuli resulted in increased perfusion, independent of treatment. Furthermore, the size of the area in which the increase occurred, defined as the area in which the perfusion increase was more than two standard deviations above the mean prestimulus perfusion, was not found to vary between the different treatments in the recorded post pin-prick images.
The measured skin temperatures ranged from 28.6 to 34.3°C (mean 31.8 6 1.2°C), supporting the assumption that the ambient temperature interval selected for the study did not cause any discomfort to the test subjects.
DISCUSSION In this study, the skin blood perfusion response to local cooling, heating, and pin-pricking was investigated in normal skin as well as in skin treated with EMLA and placebo cream for 3 h. Pin-pricking resulted in an increased perfusion and concentration of blood cells while cooling resulted in a reduced perfusion and concentration in the late time window compared to the early. Skin treated with placebo as well as untreated skin normally responded to heat stimulus with a rapid increase in perfusion and subsequent normalization within 30 s, while EMLA-treated skin
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
128
Arildsson, Nilsson, and Stro¨mberg
TABLE 1 Skin Blood Flow Parameters
Velocity CMBC Perfusion Velocity CMBC Perfusion
Treatment
Baseline
Preheat
Postheat
EMLA EMLA EMLA Placebo Placebo Placebo
0.69 (0.15) 197 (39) 134 (32) 0.78 (0.16) 176 (63) 131 (37)
0.64 (0.13) 227 (92) 142 (54) 0.63 (0.17)† 176 (80) 108 (48)†
0.92 (0.39)* 350 (137)*** 339 (215)** 0.64 (0.16) 190 (65) 125 (58)
Note. †P , 0.05 preheat vs baseline; *P , 0.05; **P , 0.01; ***P , 0.001 after heat vs preheat using Student’s t test for paired comparisons.
by the laser Doppler imager in EMLA-treated areas was accompanied by the development of a red mark in the area. This red mark developed during the period after the challenge and could be seen to match the area of the probe used. This effect was not observed in the placebo area, nor was it observed in the reference area. The color change of the skin may influence the Doppler signal if the intensity of backscattered light is altered, leading to a misinterpretation of the changes in the recorded Doppler signal. For the 11 subjects investigated in this study, however, no such changes could be found in the backscattered light intensity. Local heating in the reference and the placebotreated areas resulted in a decreased perfusion in the late time window, compared to the early window. This decrease is due to the initial perfusion increase response after the provocation, which may not yet have faded completely in the early time window. TABLE 2 Perfusion Changes (%) after Heating, LDPI Images (Averages Postheat vs Preheat) for Individual Subjects
FIG. 5. Results of the statistical analysis of the duplex signal using an ANOVA model, time effect, and treatment-time effect for EMLA (circle), placebo (square), and reference (triangle). (a) Pin-pricking, (b) cooling, and (c) heating.
responded with a delayed increase in perfusion that remained after the 4-min measurement period. The perfusion increase due to local heating recorded
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Subject
EMLA (%)
Placebo (%)
Reference (%)
1 2 3 4 5 6 7 8 9 10 11
127 1102 1148 141 111 157 1153 120 1622 182 1230
129 210 115 18 15 123 155 152 110 28 135
16 29 16 110 222 114 16 211 127 143 28
129
Effects on Skin Blood Flow by Provocation
While local heating showed a marked difference between the different treatments (EMLA, placebo, and reference), no difference was found when using pinprick. This stimulus induced a response that was not constrained to the actual needle insertion point, but rather to a somewhat larger area. The size of this area containing increased blood perfusion was found to be constant between the different treatments. Since this response is independent of analgesia, the most plausible explanation is that the trauma is responsible for the perfusion increase, by the release of substances connected to tissue damage, for instance histamine, causing vasodilatation. The responses recorded from areas in which cooling was applied seemed to be originating from the vasoconstriction, which is the normal reaction to a cold stimuli by the skin. During the experiments a certain level of systemic lidocaine and prilocaine can be expected (Engberg et al., 1987). However, due to the small amount of applied EMLA cream, the concentration of these substances can be regarded as negligible at the placebo and reference sites, thus not influencing the measurements made at these sites. By using different application times of EMLA, Bjerring et al. (1989) demonstrated that blood flow measured with laser Doppler correlated well with the erythema index (EI). EI showed the lowest values after 1.5-h treatment, indicating blanching, after which the EI increased with application time, indicating skin erythema. After 3 h of treatment the EI returned to the preapplication values. This behavior is in accordance with our findings, namely that the basal perfusion after 3 h of EMLA treatment had not changed when compared to prior to application. The mechanism behind the heat response as measured by laser Doppler in the analgesized area might include: An increased cell metabolism due to the temperature rise leading to an accumulation of metabolic waste products in the tissue. This will open up the precapillary sphincters in order to elevate the tissue blood flow so that the waste products can be transported away. When a nutritive equilibrium is reached, the myogenic control of the microcirculation would
normally adjust the blood flow back to the prechallenge levels. However, the myogenic control is very likely to be under the influence of the local analgesia, preventing it from returning the flow to normal conditions. • The normal tissue blood flow response to a temperature rise is to increase the flow in the thermoregulative compartment. This response is not instigated locally but rather by the central nervous system. However, the myogenic negative feedback control of the vessels in the challenged area is under the influence of the analgesia and cannot adjust the flow back to normal conditions. • A trauma is inflicted to the tissue, leading to the release of a vasodilating substance such as histamine. The temperature of the probe used, 45°C, is high enough to create such damage, but the rather short application time, 9 s, makes it unlikely that the temperature rise in the tissue would suffice to damage the tissue. Also, the high degree of localization of the area in which the flow increase occurs, points toward the possible trauma not being responsible for the blood perfusion increase in contrary to the response due to pin-prick. In conclusion, local analgesia due to 3 h treatment of EMLA alters forearm skin regulation associated with local heating, while the response to a pin-prick or cooling stimulus is not. A delayed perfusion increase which persists at least 4 min after the stimulus was observed. The results are compatible with an inhibition of the vasoconstrictive local myogenic feedback control of the vascular plexa measured by laser Doppler. However, skin blood flow regulation after local cooling and pin-pricking stimulus was not affected.
•
ACKNOWLEDGMENTS
The Swedish Competence Centre of Noninvasive Medical Measurements, NIMED, supported this study. We also acknowledge Astra Pain Control for their valuable cooperation.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
130
REFERENCES Arildsson, M., Wårdell, K., and Nilsson, G. E. (1997). Higher order moment processing of laser doppler perfusion signals. J. Biomed. Optics 10, 358 –363. Bjerring, P., Andersen, P. H., and Arendt-Nielsen, L. (1989). Vascular response of human skin after analgesia with EMLA cream. Br. J. Anaesth. 63, 655– 660. Braverman, I. M., and Schechner, J. S. (1991). Contour mapping of the cutaneous microvasculature by computerized Laser Doppler velocimetry. J. Invest. Dermatol. 97, 1013–1018. Braverman, I. M., Schechner, J. S., Silverman, D. G., and Keh-Yen A. (1992). Topographic mapping of the cutaneous microcirculation using two outputs of Laser Doppler flowmetry: Flux and the concentration of moving blood cells. Microvasc. Res. 44, 33– 48. Engberg, G., Danielsson, K., Henneberg, S., and Nilsson, A. (1987). Plasma concentrations of prilocaine and lidocaine and methaemoglobin formation in infants after epicutaneous application of a 5% lidocaine-prilocaine cream (EMLA). Acta Anaesthesiol. Scand. 31, 624 – 628. Juhlin, L., Ha¨gglund, G., and Evers, H. (1989). Absorption of lidocain and prilocain after application of a eutectic mixture of local anaesthetics (EMLA) on normal and diseased skin. Acta Derm. Venereol. 69, 18 –22. Juhlin, L., and Evers, H. (1990). EMLA, a new topical anaesthetic. Adv. Dermatol. 5, 76 –92. Manner, T., Kanto, J., Iisalo, E., Lindbert, R., Viinamo¨ki, O., and Scheinin, M. (1987). Reduction of pain at venous cannulation in children with a eutectic mixture of lidocaine and prilocaine (EMLA cream). Acta Anaesthesiol. Scand. 31, 735–739.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Arildsson, Nilsson, and Stro¨mberg
Montgomery, D. C. (1997). Design and analysis of experiments, 4th ed., Wiley. ¨ berg, P.Å. (1980a). A new instruNilsson, G. E., Tenland, T., and O ment for continuous measurement of tissue blood flow by light beating spectroscopy. IEEE Trans. Biomed. Eng. 27, 12–19. ¨ berg, P.Å. (1980b). Evaluation of a Nilsson, G. E., Tenland, T., and O laser Doppler flowmeter for measurement for tissue blood flow. IEEE Trans. Biomed. Eng. 27, 597– 604. ¨ berg, P.Å. (1983). Salerud, E. G., Tenland, T., Nilsson, G. E., and O Rhytmical variations in human skin blood flow. Int. J. Microcirc. Clin. Exp. 2, 91–102. Santacana, E., Aliaga, L., Bayo, M., and Villar Landeira, J. M. (1993). EMLA. A new topical anaesthetic. Rev. Esp. Anestesiol. Reanim. 40, 284 –291. Stern, M. D. (1975). In vivo evaluation of microcirculation by coherent light scattering. Nature 254, 56 –58. ¨ berg, P.Å. (1983). Tenland, T., Salerud, E. G., Nilsson, G. E., and O Spatial and temporal variations in human skin blood flow. Int. J. Microcirc. Clin. Exp. 2, 81–90. Tucker, G. T. (1986). Pharmacokinetics of local anaesthetics. Br. J. Anaesth. 58, 717–731. Wårdell, K., Naver, H. K., Nilsson, G. E., and Wallin, B. G. (1993a). The cutaneous vascular axon reflex in humans characterized by laser Doppler perfusion imaging. J. Physiol. 460, 185–199. Wårdell, K., Jakobsson, A., Nilsson, G. E., and Wallin, B. G. (1993b). Laser doppler perfusion imaging by dynamic light scattering. IEEE Trans. Biomed. Eng. 40, 309 –316. Wårdell, K., and Nilsson, G. E. (1996). Duplex laser doppler perfusion imaging. Microvasc. Res. 52, 171–182.
Effects on Skin Blood Flow by Provocation during Local Analgesia M. Arildsson, G. E. Nilsson, and T. Stro¨mberg Department of Biomedical Engineering, Linko¨ping University, University Hospital, SE-581 85 Linko¨ping, Sweden Received May 18, 1999
Although topical analgesia cream has been used for several years, little is known about its effects on the microcirculation. Previous studies have shown a vasoconstrictive effect after short application times and a vasodilatation after longer application. It has also been shown that vasomotion does not occur in the analgesized skin. The present study was undertaken to investigate the alterations in skin blood perfusion following local cooling, local heating and pin-pricking after the establishment of analgesia. In 11 healthy volunteers, skin analgesia was attained by use of a eutectic mixture of lidocaine and prilocaine (EMLA, Astra Pain Control AB, Sweden) applied to the skin three hours prior to provocation. The changes in skin blood perfusion, after applying three different provocation methods, were studied using the laser Doppler technique. Local cooling and heating to temperatures of 110 and 145°C, respectively, were applied for 9 s by use of a copper probe (Ø12 mm). In the pin-prick provocation method, a combined effect of deflection and penetration of the skin to in total 3 mm was attained. Identical provocation methods were applied to placebo treated and untreated skin areas. After heat provocation, significant differences in the perfusion response between the treatments were seen (P < 0.0001). Skin areas treated with analgesia cream responded with a slow increase in perfusion that persisted beyond the four minute measurement period. Placebo and untreated areas decreased their perfusion over time. After cooling a significant reduction in skin perfusion was seen, irrespective of the
122
treatment. Similarly, after pin-pricking a perfusion increase was seen for all treatments. The findings indicate that topical analgesia influences the myogenic control of the blood flow in those vascular plexa measured by laser Doppler following heat provocation. No differences could be seen in the response to pin-pricking and cooling for the different treatments. © 2000 Academic Press Key Words: laser doppler flowmetry; LDPI; EMLA; analgesia; skin microcirculation; heat stimuli; pin-prick.
INTRODUCTION Analgesia cream has been used for a number of years for pain relief during superficial clinical and surgical procedures such as venopuncture, split skin grafting, skin biopsies, and treatment of port wine stains (Santacana et al., 1993; Manner et al., 1987; Juhlin et al., 1990). In particular, analgesia cream is used during needle insertion through intact skin in children. The predominantly used topical analgesia cream, EMLA (Astra Pain Control AB, Sweden), is a mixture of the substances lidocaine and prilocaine. EMLA causes a biphasic vascular response comprising initial blanching and vasoconstriction (maximal after 1.5 h of application) and late erythema and vasodilatation at application times longer than 3 h (Bjerring et al., 1989). These local circulatory changes have been suggested 0026-2862/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
123
Effects on Skin Blood Flow by Provocation
to be due to the direct effect of local anesthetics on vascular smooth muscle, producing vasoconstriction at low concentrations and vasodilatation at high ones (Tucker, 1986). The analgesia onset and vascular response are faster in patients with eczematous skin than in healthy subjects as a result of increased penetration through abnormal skin (Juhlin et al., 1989). Salerud et al. (1983) showed that the rhythmical variations in skin blood flow, i.e., vasomotion, are locally regulated and could not be reduced by afferent nerve blockade. Local analgesia by EMLA treatment for 1 h, however, abolished the vasomotion leaving mean skin blood flow unchanged. Wårdell et al. (1993a) showed that EMLA eliminated the normal axon reflex mediated vasodilatation during electrical skin stimulus. Apart from the latter study, little is known about the vascular response to local skin stimulation during topical application of analgesia. The aim of the present study is therefore to investigate the response pattern, and thereby the regulation, of skin blood flow after topical application of EMLA, using local heating, cooling, and pin-pricking as stimuli.
MATERIALS AND METHODS Laser Doppler Perfusion Imaging (LDPI) The laser Doppler technique was developed during the mid seventies (Stern, 1975; Nilsson et al., 1980a, 1980b) as a method for noninvasive measurement of tissue blood perfusion (defined as the concentration multiplied by the average velocity of the moving blood cells). The method is based on the frequency shift photons undergo when scattered by moving blood cells, i.e., the Doppler effect. Laser Doppler Perfusion Imaging (LDPI) uses a freely impinging laser beam to generate an image of the tissue perfusion (Wårdell et al., 1993b). This image describes the spatial variations in perfusion due to the vascular structure in the measurement area (Braverman et al., 1991, 1992; Tenland et al., 1983). By continuously recording small images (duplex recording), one may also assess the
temporal changes in the blood flow (Wårdell et al., 1996). In this study a modified laser Doppler imager (PIM version 1.0 from Lisca AB, Sweden) has been used. This scanner records images of a size up to 12.5*12.5 cm (with the scanner head positioned 15 cm above the measurement surface) using at the most 64*64 measurement sites. The scanner was modified to also include a signal processing filter that allows the concentration of moving blood cells (CMBC) to be recorded (Arildsson et al., 1997). By recording the CMBC signal one may also estimate the average velocity, by dividing the perfusion estimate by the CMBC.
The Dermal Sensitivity Tester The dermal sensitivity tester (desensor) (Cenova AB, Sweden) consists of a computerized system, which delivers standardized preset skin stimuli to a defined area of the skin. The stimuli delivered by the desensor are pin-prick, cold, heat, and touch. Pain is provoked by pricking the skin with a dental needle (27 G) using a standardized depth of 3.0 mm. Cold and heat stimuli are induced by the use of metal probes that are brought into contact with the skin. The application time, as well as the temperature of the probe, can be set by the operator. Each type of stimuli is applied on a separate skin location, which can be determined with high precision. Figure 1 shows the set-up of the desensor system.
Skin Temperature The skin temperature was recorded using an emissivity thermometer, THI-300 (Tasco, Japan).
Placebo Cream The placebo cream was provided by Astra Pain Control, and contains coconut cream instead of the active substances. The pH of the placebo is the same as the pH of the EMLA cream.
Statistical Analysis A statistical model suitable for explaining alterations in the duplex recordings of skin perfusion (P)
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
124
Arildsson, Nilsson, and Stro¨mberg
There will not be an effect due to interaction: H0, (tb)ij 5 0 for all i, j; H1, (tb)ij Þ 0 for some i, j. The recorded images were used to analyze the effects of the analgesia and placebo creams on the perfusion and concentration of moving blood cells. For this purpose baseline recordings and the recordings made after the selected application time and stimulus were compared. This comparison were made using Student’s t test on the calculated averages of the individual images.
Experimental Set-Up
FIG. 1. The set-up of the dermal sensitivity tester (desensor).
due to a stimulus (local heating, cooling, or pin-pricking) includes a time effect (b j, j 5 1, 2) accounting for the early and late perfusion response, and a treatment effect (t i, i 5 1, 2, 3, corresponding to EMLA, placebo, and reference). Also a time-treatment interaction effect (tb) ij was included in the statistical analysis. Hence a 2-way ANOVA mode repeated measurements design (Montgomery, 1997) was selected: P 5 m 1 t i 1 b j 1 ~ tb ! ij 1 g k 1 « ijk,
(1)
where m is the mean perfusion, g k denotes the effects of the subject (k 5 1, 2, . . . , 11) (random effect), and « ijk is the model error. The following three null hypotheses were put forth. The different treatments will not affect the perfusion: H0, tI 5 0 for all i; H1, tI Þ 0 for at least one i. There will not be an effect due to time: H0, bj 5 0 for all j; H1, bj Þ 0 for at least one j.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
All experiments were made with the subject’s informed consent and were also approved by the Ethics Committee for Human Research at Linko¨ping University Hospital, Sweden. Prior to each experiment, all subjects were acclimatized to the room conditions for 20 min. During the acclimatization period one of the subject’s forearms was randomly chosen for measurement. The forearm skin was divided into three different areas. These three areas were randomly chosen to be treated with EMLA, placebo, or to be used as a reference area. Further, each of the areas was divided into three subareas, in which the different stimuli were to be applied; these areas were marked using ordinary ink. Since the heat probe and the cooling probe are circular in shape, the markings were made circular with diameters of 12 mm. The subareas were assigned different stimuli including pin-prick (pin-prick depth 3.0 mm), local heating (45°C, 9 s) and local cooling (10°C, 9 s), at random. A typical layout of an experiment is shown in Fig. 2. The forearm of the subject was fixated firmly in the Astra desensor using a vacuum pillow. This fixation kept the artifacts, due to movements of the arm during the recording of laser Doppler images and signals, at a minimum. After fixation, the unprovoked baseline perfusion of the areas chosen to be used for application of EMLA and placebo were recorded. These baseline recordings
125
Effects on Skin Blood Flow by Provocation
FIG. 2. Layout of the experiments, comprising three main areas that are divided into three subareas each. Each main area was treated using either EMLA or placebo or was used as a reference area. The subareas were subjected to the different stimuli.
consisted of image scans (size 16*16 measurement sites, approximately 1 cm 2) of all subareas, and a duplex scan (4 min, 2*2 probe) in one subarea in each of the application areas. At this point no measurements were made in the reference area. Also, the EMLA and placebo cream (Astra Pain Control, Sweden) were applied and covered with Tegaderm (3M, Canada). The average amount of EMLA applied was 1.6 6 0.2 mg and the average amount of placebo 2.0 6 0.2 mg, covering approximately 10 cm 2. The application of EMLA and placebo was spaced in time approximately 25 min apart, which was sufficient to perform all recordings in one application area prior to removing the cream in the second area, thus maintaining a three hour application in both areas. When the creams had been applied, the measurements on the area selected as reference were made. First all subareas were investigated, two of the subareas using the standard method which records an image, size 16*16, and one subarea using a duplex scan (4 min, 2*2 probe), followed by an image recording, size 16*16 pixels. The subareas were then challenged using the different stimuli. After each provocation a 4-min duplex signal was measured, followed by a 16*16 scan. In order to record the blood flow response as accurately as possible, the laser beam was directed to the center of the stimulated area, i.e., the pin-prick point or the center of the area provoked by the two different probes. The time to complete a full subarea measurement was 25 min. The reason for not also using duplex recordings in all subareas before provocation was that this would not be feasible; the time to complete a measurement would be too long.
After 3 h of application, the measurements made on the reference area were repeated on the areas treated with EMLA and placebo, respectively. In addition, the skin temperature was recorded, both prior to application, prior to provocation, and after provocation in all nine different subareas. This study comprises 12 healthy, caucasian subjects, 7 male and 5 female, with ages ranging from 16 to 35 years. All subjects were nonsmokers and refrained from drinking coffee or using any type of caffeine prior to the experiment. This will ascertain that the local skin blood flow was not affected by these types of stimuli. One male subject was excluded from the evaluation set, because of illness during the test. A total of 11 subjects was therefore used in the statistical analysis. The temperature of the room was 24 –26°C.
RESULTS Typical duplex signals recorded immediately after heat provocation in a reference area are shown in Fig. 3, while Fig. 4 shows a typical perfusion response from an area on to which EMLA has been applied topically. The recorded signals in the placebo and the reference area generally display an initial increase in the blood perfusion followed by a rapid normalization of the flow. In the area treated with EMLA this temporary increase in perfusion following heat stimulus was not seen. Instead a delayed increase in perfusion was seen which remained until the end of the mea-
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
126
Arildsson, Nilsson, and Stro¨mberg
FIG. 3.
Typical concentration (a) and perfusion (b) responses from untreated skin after local heating (9 s, 45°C).
surement, approximately 4 min after the stimulus was applied. This type of reaction was seen in both the CMBC-signal and in the perfusion signal. The response was calculated as the average perfusion during approximately 30 s, within two time windows of which the first started 30 s after provocation, and the second started at 3 min after provocation. These time windows are shown in Fig. 4. The delay between the stimulus and the first time window was chosen in such a way that the initial increase seen in normal responses was given time to fade. By use of this approach it was also possible to minimize the influence of differences in the time gap between the stimulation and the start of the recording. Figure 5 shows the results of evaluating these responses in the model described by Eq. (1). Figure 5 shows the time and time-treatment effects for the three different provocations (pin-prick, cooling, and heating) when measuring perfusion. In all treatments pinprick (Fig. 5a) increased the perfusion (time effect, P , 0.01), while cooling (Fig. 5b) resulted in a decrease (time effect, P , 0.025). For heating (Fig. 5c) there is
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
a difference between the treatments in that the reference areas and the placebo treated areas showed a reduced perfusion, while in the areas treated with EMLA a perfusion increase (time-treatment effect, P , 0.0001) was observed. In the images recorded pre- and postheat challenge in the EMLA-treated areas, the backscattered light intensity was found not to have changed (preheat image average 8.94 6 0.38 A.U., postheat 9.06 6 0.27 A.U., P . 0.3). Similarly, the light intensity did not change due to the 3-h EMLA application (preapplication image average 9.10 6 0.19 A.U., preheat 8.94 6 0.38 A.U., P . 0.1). In Table 1 the mean of the recorded images for the different stimuli and treatments is presented along with the results obtained using Student’s t test, comparing the baseline recording with the recordings after delivered stimuli. No significant changes in perfusion, concentration, or velocity of blood cells due to local skin analgesia were found, i.e., no changes could be seen between the recorded baseline images and the preheat images in the EMLA area. Table 1 also shows
127
Effects on Skin Blood Flow by Provocation
FIG. 4. Typical concentration (a) and perfusion (b) responses from skin treated with EMLA (3-h application) after local heating (9 s, 45°). Also shown are the time windows used for evaluation.
the mean of the recorded images after applying a heat stimulus. These values yield similar results as from the duplex-recordings, showing a significant increase in perfusion (138%, P , 0.01), concentration (54%, P , 0.001), and velocity (42%, P , 0.05) after heat stimulus in areas under local anesthesia. In areas treated with the placebo cream a slight decrease in both perfusion (17%, P , 0.05) and average velocity (19%, P , 0.05) was seen (baseline vs preheat). No significant change was seen after heat stimulus. The individual results obtained for the 11 subjects are listed in Table 2 as perfusion changes between pre- and postheat images. The pin-prick stimuli resulted in increased perfusion, independent of treatment. Furthermore, the size of the area in which the increase occurred, defined as the area in which the perfusion increase was more than two standard deviations above the mean prestimulus perfusion, was not found to vary between the different treatments in the recorded post pin-prick images.
The measured skin temperatures ranged from 28.6 to 34.3°C (mean 31.8 6 1.2°C), supporting the assumption that the ambient temperature interval selected for the study did not cause any discomfort to the test subjects.
DISCUSSION In this study, the skin blood perfusion response to local cooling, heating, and pin-pricking was investigated in normal skin as well as in skin treated with EMLA and placebo cream for 3 h. Pin-pricking resulted in an increased perfusion and concentration of blood cells while cooling resulted in a reduced perfusion and concentration in the late time window compared to the early. Skin treated with placebo as well as untreated skin normally responded to heat stimulus with a rapid increase in perfusion and subsequent normalization within 30 s, while EMLA-treated skin
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
128
Arildsson, Nilsson, and Stro¨mberg
TABLE 1 Skin Blood Flow Parameters
Velocity CMBC Perfusion Velocity CMBC Perfusion
Treatment
Baseline
Preheat
Postheat
EMLA EMLA EMLA Placebo Placebo Placebo
0.69 (0.15) 197 (39) 134 (32) 0.78 (0.16) 176 (63) 131 (37)
0.64 (0.13) 227 (92) 142 (54) 0.63 (0.17)† 176 (80) 108 (48)†
0.92 (0.39)* 350 (137)*** 339 (215)** 0.64 (0.16) 190 (65) 125 (58)
Note. †P , 0.05 preheat vs baseline; *P , 0.05; **P , 0.01; ***P , 0.001 after heat vs preheat using Student’s t test for paired comparisons.
by the laser Doppler imager in EMLA-treated areas was accompanied by the development of a red mark in the area. This red mark developed during the period after the challenge and could be seen to match the area of the probe used. This effect was not observed in the placebo area, nor was it observed in the reference area. The color change of the skin may influence the Doppler signal if the intensity of backscattered light is altered, leading to a misinterpretation of the changes in the recorded Doppler signal. For the 11 subjects investigated in this study, however, no such changes could be found in the backscattered light intensity. Local heating in the reference and the placebotreated areas resulted in a decreased perfusion in the late time window, compared to the early window. This decrease is due to the initial perfusion increase response after the provocation, which may not yet have faded completely in the early time window. TABLE 2 Perfusion Changes (%) after Heating, LDPI Images (Averages Postheat vs Preheat) for Individual Subjects
FIG. 5. Results of the statistical analysis of the duplex signal using an ANOVA model, time effect, and treatment-time effect for EMLA (circle), placebo (square), and reference (triangle). (a) Pin-pricking, (b) cooling, and (c) heating.
responded with a delayed increase in perfusion that remained after the 4-min measurement period. The perfusion increase due to local heating recorded
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Subject
EMLA (%)
Placebo (%)
Reference (%)
1 2 3 4 5 6 7 8 9 10 11
127 1102 1148 141 111 157 1153 120 1622 182 1230
129 210 115 18 15 123 155 152 110 28 135
16 29 16 110 222 114 16 211 127 143 28
129
Effects on Skin Blood Flow by Provocation
While local heating showed a marked difference between the different treatments (EMLA, placebo, and reference), no difference was found when using pinprick. This stimulus induced a response that was not constrained to the actual needle insertion point, but rather to a somewhat larger area. The size of this area containing increased blood perfusion was found to be constant between the different treatments. Since this response is independent of analgesia, the most plausible explanation is that the trauma is responsible for the perfusion increase, by the release of substances connected to tissue damage, for instance histamine, causing vasodilatation. The responses recorded from areas in which cooling was applied seemed to be originating from the vasoconstriction, which is the normal reaction to a cold stimuli by the skin. During the experiments a certain level of systemic lidocaine and prilocaine can be expected (Engberg et al., 1987). However, due to the small amount of applied EMLA cream, the concentration of these substances can be regarded as negligible at the placebo and reference sites, thus not influencing the measurements made at these sites. By using different application times of EMLA, Bjerring et al. (1989) demonstrated that blood flow measured with laser Doppler correlated well with the erythema index (EI). EI showed the lowest values after 1.5-h treatment, indicating blanching, after which the EI increased with application time, indicating skin erythema. After 3 h of treatment the EI returned to the preapplication values. This behavior is in accordance with our findings, namely that the basal perfusion after 3 h of EMLA treatment had not changed when compared to prior to application. The mechanism behind the heat response as measured by laser Doppler in the analgesized area might include: An increased cell metabolism due to the temperature rise leading to an accumulation of metabolic waste products in the tissue. This will open up the precapillary sphincters in order to elevate the tissue blood flow so that the waste products can be transported away. When a nutritive equilibrium is reached, the myogenic control of the microcirculation would
normally adjust the blood flow back to the prechallenge levels. However, the myogenic control is very likely to be under the influence of the local analgesia, preventing it from returning the flow to normal conditions. • The normal tissue blood flow response to a temperature rise is to increase the flow in the thermoregulative compartment. This response is not instigated locally but rather by the central nervous system. However, the myogenic negative feedback control of the vessels in the challenged area is under the influence of the analgesia and cannot adjust the flow back to normal conditions. • A trauma is inflicted to the tissue, leading to the release of a vasodilating substance such as histamine. The temperature of the probe used, 45°C, is high enough to create such damage, but the rather short application time, 9 s, makes it unlikely that the temperature rise in the tissue would suffice to damage the tissue. Also, the high degree of localization of the area in which the flow increase occurs, points toward the possible trauma not being responsible for the blood perfusion increase in contrary to the response due to pin-prick. In conclusion, local analgesia due to 3 h treatment of EMLA alters forearm skin regulation associated with local heating, while the response to a pin-prick or cooling stimulus is not. A delayed perfusion increase which persists at least 4 min after the stimulus was observed. The results are compatible with an inhibition of the vasoconstrictive local myogenic feedback control of the vascular plexa measured by laser Doppler. However, skin blood flow regulation after local cooling and pin-pricking stimulus was not affected.
•
ACKNOWLEDGMENTS
The Swedish Competence Centre of Noninvasive Medical Measurements, NIMED, supported this study. We also acknowledge Astra Pain Control for their valuable cooperation.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
130
REFERENCES Arildsson, M., Wårdell, K., and Nilsson, G. E. (1997). Higher order moment processing of laser doppler perfusion signals. J. Biomed. Optics 10, 358 –363. Bjerring, P., Andersen, P. H., and Arendt-Nielsen, L. (1989). Vascular response of human skin after analgesia with EMLA cream. Br. J. Anaesth. 63, 655– 660. Braverman, I. M., and Schechner, J. S. (1991). Contour mapping of the cutaneous microvasculature by computerized Laser Doppler velocimetry. J. Invest. Dermatol. 97, 1013–1018. Braverman, I. M., Schechner, J. S., Silverman, D. G., and Keh-Yen A. (1992). Topographic mapping of the cutaneous microcirculation using two outputs of Laser Doppler flowmetry: Flux and the concentration of moving blood cells. Microvasc. Res. 44, 33– 48. Engberg, G., Danielsson, K., Henneberg, S., and Nilsson, A. (1987). Plasma concentrations of prilocaine and lidocaine and methaemoglobin formation in infants after epicutaneous application of a 5% lidocaine-prilocaine cream (EMLA). Acta Anaesthesiol. Scand. 31, 624 – 628. Juhlin, L., Ha¨gglund, G., and Evers, H. (1989). Absorption of lidocain and prilocain after application of a eutectic mixture of local anaesthetics (EMLA) on normal and diseased skin. Acta Derm. Venereol. 69, 18 –22. Juhlin, L., and Evers, H. (1990). EMLA, a new topical anaesthetic. Adv. Dermatol. 5, 76 –92. Manner, T., Kanto, J., Iisalo, E., Lindbert, R., Viinamo¨ki, O., and Scheinin, M. (1987). Reduction of pain at venous cannulation in children with a eutectic mixture of lidocaine and prilocaine (EMLA cream). Acta Anaesthesiol. Scand. 31, 735–739.
Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Arildsson, Nilsson, and Stro¨mberg
Montgomery, D. C. (1997). Design and analysis of experiments, 4th ed., Wiley. ¨ berg, P.Å. (1980a). A new instruNilsson, G. E., Tenland, T., and O ment for continuous measurement of tissue blood flow by light beating spectroscopy. IEEE Trans. Biomed. Eng. 27, 12–19. ¨ berg, P.Å. (1980b). Evaluation of a Nilsson, G. E., Tenland, T., and O laser Doppler flowmeter for measurement for tissue blood flow. IEEE Trans. Biomed. Eng. 27, 597– 604. ¨ berg, P.Å. (1983). Salerud, E. G., Tenland, T., Nilsson, G. E., and O Rhytmical variations in human skin blood flow. Int. J. Microcirc. Clin. Exp. 2, 91–102. Santacana, E., Aliaga, L., Bayo, M., and Villar Landeira, J. M. (1993). EMLA. A new topical anaesthetic. Rev. Esp. Anestesiol. Reanim. 40, 284 –291. Stern, M. D. (1975). In vivo evaluation of microcirculation by coherent light scattering. Nature 254, 56 –58. ¨ berg, P.Å. (1983). Tenland, T., Salerud, E. G., Nilsson, G. E., and O Spatial and temporal variations in human skin blood flow. Int. J. Microcirc. Clin. Exp. 2, 81–90. Tucker, G. T. (1986). Pharmacokinetics of local anaesthetics. Br. J. Anaesth. 58, 717–731. Wårdell, K., Naver, H. K., Nilsson, G. E., and Wallin, B. G. (1993a). The cutaneous vascular axon reflex in humans characterized by laser Doppler perfusion imaging. J. Physiol. 460, 185–199. Wårdell, K., Jakobsson, A., Nilsson, G. E., and Wallin, B. G. (1993b). Laser doppler perfusion imaging by dynamic light scattering. IEEE Trans. Biomed. Eng. 40, 309 –316. Wårdell, K., and Nilsson, G. E. (1996). Duplex laser doppler perfusion imaging. Microvasc. Res. 52, 171–182.