Subjects. Keywords: acetylcholine, iontophoresis, laser Doppler fluxmetry, dermal microcirculation

Transcription

1 Br J Clin Pharmacol 1997; 43: An investigation into variability in microvascular skin blood flow and the responses to transdermal delivery of acetylcholine at different sites in the forearm and hand Janet M. Gardner-Medwin, J. Y. Taylor, I. A. Macdonald & R. J. Powell 1 Department of Physiology and Pharmacology, University of Nottingham Medical School and 1 Clinical Immunology Unit, Queen s Medical Centre, Nottingham, UK Aims Transdermal iontophoresis in combination with laser Doppler fluxmetry (LDF) are useful techniques for examining dermal microcirculatory responses to different vasodilators. Differences in skin and microcirculation structure could influence the recorded baseline flux, and the observed vasodilatation. To examine this we compared baseline flux and the response of microvascular blood flow to a single vasodilator, acetylcholine, at sites in the forearm and hand. Methods Baseline microcirculation flow was recorded by LDF in a temperature controlled laboratory. The change in flux with iontophoresis of identical doses of acetylcholine, 150 ma for 40 s, was recorded at 12 different sites in the forearm and hand in 10 female and 3 male subjects. Results Baseline flux patterns and the vasodilatation to identical periods of iontophoresis of acetylcholine were site dependent. Palmar sites showed a higher baseline flux, but no vasodilatation to iontophoresis of acetylcholine. In contrast the volar forearm, dorsal hand and finger sites showed lower site-dependent baseline flux, but did vasodilate. Conclusions Patterns of baseline flux are specific to sites on the hand and forearm reflecting differences in underlying microvascular structure. The vasodilatation to transdermal delivery of acetylcholine is also site dependent, but differences in skin structure may be more important than the underlying microvasculature in determining the response. Keywords: acetylcholine, iontophoresis, laser Doppler fluxmetry, dermal microcirculation Introduction Methods The techniques of iontophoresis in combination with laser Doppler fluxmetry (LDF) are increasingly used to investigate Subjects responses to transdermal delivery of charged vasoactive Thirteen healthy male and female subjects (10 female and 3 agents to the dermal microcirculation [1 3]. The skin has a male) aged years were recruited. Full ethical approval heterogeneous structure which can influence the transdermal was obtained from the local Ethics Committee. Subjects delivery of drugs [4 6]. Variation in the underlying microvascular refrained from alcohol, caffeine and tobacco for 12 h prior structure may also influence the baseline flux and to the experiment, and were asked not to eat for 2 h prior responses to iontophoresis. The palmar aspect of the hand is to the experiment, avoiding any heavy meals on that day. rich in arterio-venous anastomoses (AVAs) which are All subjects were asked not to use any detergents, hand important in thermoregulation [7, 8]. AVAs are fewer in creams or other substances on their hands on the day of the number on the dorsum of the fingers and absent from the experiment, and washed their hands in the same neutral ph dorsum of the hand, where the microvasculature s role is soap immediately prior to the experiment. predominantly nutritive. Here we have used LDF and iontophoresis to compare the characteristics of baseline flux and the microvascular response to transdermal administration Experimental outline of a single agent, acetylcholine, at different sites on the The effect of identical periods of iontophoresis of acetylchoforearm and hand. The sites were chosen for comparison of line at three sites on the non-dominant forearm and nine both nutritive and arterio-venous anastomoses rich areas. sites on the hand was compared in the 13 subjects. The sites However, differences in skin structure between the sites are shown in Figure 1, and are grouped according to the might be anticipated to alter the transdermal delivery of underlying vascular structures, and skin anatomy; nutritive acetylcholine. sites chosen were forearm (sites 1, 2 and 3), and dorsum of Correspondence Dr J. Gardner-Medwin, Department of Physiology and Pharmacology, University Hospital, University of Nottingham, NG7 2UH, UK hand (sites 8 and 9). Sites with high numbers of arteriovenous anastomoses were palmar sites (10, and 12), and the 1997 Blackwell Science Ltd 391

2 J. M. Gardner-Medwin et al * Figure 1 Sites at which laser Doppler fluxmetry were recorded: Anterior aspect of the forearm cm from wrist crease cm from wrist crease cm from wrist crease. Dorsum of finger 4. Proximal phalanx of the index finger. 5. Proximal phalanx of the middle finger, also used as the control site. 6. Proximal phalanx of the ring finger. 7. Middle phalanx of the middle finger. Dorsum of the hand 8. Central dorsum of the hand, proximal to the middle metacarpal phalangeal joint. 9. First web space. Palmar sites 10. Thenar eminence 11. Finger pulp of the index finger 12. Central palm designed to allow easy attachment to fingers, and were attached to the skin using adhesive tape with care not to occlude the vascular supply to the finger. A cathode, consisting of an electrocardiogram electrode (3M), was attached to the adjacent skin 4 cm away. A laser Doppler probe (Moor Instruments MBF 2, Axminster, Devon) was incorporated into the iontophoresis chamber recording flux from the 1 cm 2 area of skin to which the acetylcholine was administered. The signal, in volts, was recorded onto a chart recorder. Baseline flux was continuously monitored by LDF until it had remained stable for at least 60 s. A standard dose of 1% acetylcholine given with a constant current of 150 ma applied across the skin for 40 s was administered by iontophoresis. This dose was developed from a pilot study which had examined a range of dose-response curves for iontophoresis of 0.1 2% acetylcholine over the range 50 ma for 40 s to 200 ma for 60 s. Following iontophoresis, flux continued to be monitored for a minimum of 5 min or until the flux returned to the baseline value. A control period of iontophoresis was performed on the dorsum of the middle finger overlying the proximal phalanx, site 5. This consisted of the identical protocol, except that 0.9% normal saline replaced acetylcholine in the iontophoresis chamber. A separate chamber and laser Doppler probe were reserved for this control to ensure no contamination occurred. The control was performed prior to iontophoresis of acetylcholine at the start of each experiment, and iontophoresis of acetylcholine at site 5 was performed at the end of the protocol. Baseline flux was comparable at this site between the first and second occasions of iontophoresis, (see Table 1). Analysis of traces and statistics finger pulp (11). Sites on the dorsum of the fingers (DOF) The flux was recorded as an electrical signal expressed in (sites 4, 5, 6, and 7) were chosen as having fewer AVAs volts, and plotted against time. Figure 2 illustrates the than palmar sites. The control was at site 5. different elements of the signal that were measured. The In a single subject the correlation of the cardiac pulse and pulse amplitude (PA) is the amplitude of the signal variation the respiratory cycle with flux was formally examined with with the cardiac pulse, the flux amplitude (FA) is the LDF recorded at site 5. Simultaneously the cardiac cycle maximum amplitude of the whole flux signal, and was was recorded by a single lead electrocardiogram, and the calculated during stable periods of flux over a 10 s period. respiratory cycle by means of a chest wall strain gauge. The median flux was calculated from the mid-point of the flux amplitude. Materials The response to acetylcholine was calculated as the maximum change from baseline values, expressed as the Acetylcholine was obtained from Sigma, Poole, and stored ratio of (maximum value-baseline value)/baseline value) and at 20 C in a desiccator. A fresh 1% solution was made expressed as change in flux, PA or FA (Table 2). Paired t- immediately prior to the experiment, by dilution of the salt tests were used to assess the significance of the change in in 0.9% saline, filtered and kept at room temperature during flux. The Pearson product-moment correlation coefficient the experiment. A control solution of 0.9% saline was also between baseline flux and baseline flux amplitude, and prepared. between baseline flux and the change in flux in response to iontophoresis of acetylcholine, were determined. Technique The experiments took place in a temperature controlled environment (19 23 C). All subjects sat on a couch with their hands supported at the level of the right atrium. At each of 13 sites an identical protocol was followed: an iontophoresis chamber, 1 cm in diameter and 0.5 cm deep containing an inert platinum anode, was attached to the skin and filled with 1% acetylcholine. Chambers were Results Characteristics of baseline flux The upper part of Figure 3 illustrates normal sinus arrhythmia where inspiration precedes an increase in heart rate. It is clear from the laser Doppler recording from the dorsum of the hand that variation also occurs in the flux. It would Blackwell Science Ltd Br J Clin Pharmacol, 43 (4)

3 Variability in vasodilatation to transdermal acetylcholine Table 1 Baseline flux characteristics at the different sites for all subjects. Median flux Pulse amplitude Flux amplitude Area Site (volts) (volts) (volts) Forearm (0.7) 0.16 (0.08) 0.59 (0.39) (1.4) 0.09 (0.13) 0.42 (0.35) (1.3) 0.07 (0.08) 0.49 (0.49) Dorsum of finger (1.1) 0.06 (0.04) 0.68 (0.50) (1.0) 0.06 (0.03) 0.39 (0.17) (0.7) n= (0.13) 0.58 (0.33) (1.4) n= (0.20) 1.06 (1.26) Dorsum of hand (0.7) n= (0.04) 1.03 (0.79) (0.7) n= (0.14) 0.84 (0.71) Palmar (1.9) n= (0.09) 1.45 (1.39) (2.5) n= (0.13) 2.09 (2.18) (1.1) n= (0.08) 1.73 (1.09) Control 1.6 (0.8) n= (0.61) Results are expressed as means±s.d. of the mean. 2.8 Period of iontophoresis PA FA Median flux Volts PA FA Median flux Time (s) Figure 2 Increase in flux (median flux) following iontophoresis of acetylcholine to the dorsum of the middle finger. Changes in pulse amplitude (PA) and flux amplitude (FA) are also shown. appear that flux begins to increase as the heart rate rises prominent at the other sites, but, as can be seen in Figure 4, with inspiration. Similar variation was noted at the other sites. were also demonstrated at DOF sites. There was a good correlation between the magnitude of baseline flux and flux Variability of baseline flux amplitude (r=0.82, P<0.001). Baseline flux showed site-dependent differences in its size and characteristics. Baseline flux was greatest at palmar sites. Fingertip flux was approximately three times greater than forearm baseline flux, and thenar eminence baseline flux two times greater. The flux at the DOF was intermediate between forearm and palmar flux values. The mean baseline flux values are given in Table 1. Large (up to two fold) changes in flux occurred at palmar sites at slower rates than the respiratory cycle (Figure 4). These changes were less Response to iontophoresis A typical response to iontophoresis of acetylcholine on the dorsum of the finger is shown in Figure 2. An increase in flux in response to iontophoresis of the standard dose of acetylcholine was variably observed at the different sites as shown in Table 2, where a change in flux was defined as a change greater than the 95% confidence limits of the mean flux at the control site. Where a change in flux was noted 1997 Blackwell Science Ltd Br J Clin Pharmacol, 43 (4) 393

4 J. M. Gardner-Medwin et al. Table 2 Change in flux, pulse amplitude and flux amplitude in response to acetylcholine at the different sites. Number of occasions /maximum possible showing a change in Change in flux Change in pulse Change in flux Change in flux Change in flux flux greater than Area Site (all subjects) amplitude amplitude (men) (women) (sd) control Forearm (0.16) 1.37 (0.07) 0.41 (0.07) 3.03 (0.59) 4.27 (0.22) 13/ (0.24) 6.36 (0.23) 1.51 (0.12) 2.47 (0.28) 5.40 (0.33) 12/ (0.09) 7.25 (0.32) 1.34 (0.14) 1.83 (0.28) 1.89 (0.12) 12/13 Dorsum of finger (0.09) 2.65 (0.13) 0.97 (0.10) 0.83 (0.31) 1.03 (0.12) 9/ (0.20) 2.39 (0.14) 0.46 (0.07) 0.67 (0.16) 2.13 (0.13) 12/ (0.09) n= (0.19) 0.69 (0.12) 0.82 (0.33) 1.27 (0.11) n=9 9/ (0.07) n= (0.27) 0.70 (0.09) 0.56 (0.08) n= (0.15) n=8 6/10 Dorsum of hand (0.15) n= (0.21) 0.46 (0.08) 0.33 (0.33) n= (0.19) n=8 5/ (0.22) n= (0.27) 0.39 (0.09) 1.92 n= (0.25) n=9 8/10 Palmar (0.15) n= (0.08) 0.10 (0.05) 0.16 (0.22) 1.68 (0.22) n=9 8/ (0.05) n=12 NS 0.15 (0.06) 0.09 (0.06) 0.09 (0.19) 0.05 (0.06) n=9 3/ (0.10) n= (0.09) 0.08 (0.05) 0.77 (0.23) 0.70 (0.15) n=9 8/12 Control 0.06 (0.06) n=10 NS 0.01 (0.06) (0.07) 0.33 (0.26) 0.14 (0.06) Values are means±s.d., change=((max-base)/base). Statistical significance: NS, not significant Blackwell Science Ltd Br J Clin Pharmacol, 43 (4)

5 Variability in vasodilatation to transdermal acetylcholine Beatsmin 1 Resp Flux FA PA ECG Time (s) Figure 3 Simultaneous recording of the cardiac pulse (ECG), respiratory cycle (resp) and flux from the dorsum of the middle finger (flux). Beats min -1 indicates the beat by beat heart rate estimated by the R-R interval on the ECG. Pulse amplitude (PA), and flux amplitude (FA) are shown on the flux trace. 5 Finger pulp 4 Volts 3 2 Dorsum of finger 1 Forearm Time (s) Figure 4 Simultaneous recording of laser Doppler flux at three sites. Simultaneous recording of baseline flux (volts) from the forearm, dorsum of fingers and finger pulp over 60 s. the time from the start of iontophoresis to the first change acetylcholine was only weakly related to the magnitude of in flux was consistent, (mean 17.48±6.33 s) at all sites. increase in flux at different sites (Table 2 and Figure 2), (r= The magnitude of the change in response to iontophoresis 0.47, P+0.1). This was associated with an apparent decrease was site dependent, with the three forearm sites showing in the variability in flux due to other factors such as cardiac the greatest and most consistent response to acetylcholine and respiratory cycles. PA showed little variation at palmar (Table 2). Good responses to acetylcholine were also seen at sites in response to iontophoresis. DOF and DOH sites. Vasodilatation to acetylcholine was minimal at palmar sites (Table 2). At all sites, except the control and site 11, the change in flux was significantly different from baseline flux (P<0.001) and there was an Control studies inverse correlation between the baseline flux, and the change Flux was very sensitive to minute quantities of acetylcholine, in flux with iontophoresis of acetylcholine (r= 0.64, and a different iontophoresis chamber and laser Doppler P<0.02). probe were used during the control studies. Probes which had been at all contaminated with acetylcholine on previous Pulse amplitude and flow amplitude occasions produced a change in flux and pulse amplitude if the chamber was used as the anode. This change in flux Baseline pulse amplitude (PA) was similar at all sites. Baseline could be eliminated by reversing the direction of current flux amplitudes (FA) were greatest at palmar sites, particularly flow, so preventing the iontophoresis of positively charged the finger tip, as shown in Table 1 and Figure 4. acetylcholine. Using separate iontophoresis chambers no The magnitude of increase in PA in response to change in flux, or any other parameters was observed in the 1997 Blackwell Science Ltd Br J Clin Pharmacol, 43 (4) 395

6 J. M. Gardner-Medwin et al. control studies using 0.9% saline with the iontophoresis Surprisingly this was very consistent, and the time to the chamber acting as anode (Table 1). onset of flux and pulse amplitude changes was approximately 18 s at all sites, and between men and women. Discussion At the greatest values of flux which were achieved following iontophoresis there was loss of the typical baseline LDF is a technique which allows continuous monitoring of pattern of variation in flux. This re-emerged as the flux the dynamic variation in flux in the skin microcirculation returned towards baseline values. At the greatest flux values [9]. The recorded signals show spatial and temporal variation achieved neither vasomotion or the respiratory cycle could [10 14]. The results presented here confirm the variability be differentiated as causing variability to the flux. The and suggest that the characteristics of the flux are related to changes in flux mirrored changes in the PA, which also the site from which they are recorded. Site also influences increased following iontophoresis, and decreased as flux the observed vasodilatation in response to acetylcholine. returned towards baseline values, with loss of the flux The patterns of flux at all sites were altered by the cardiac pulsatility at high flux values, as described previously [11]. pulse, and respiratory cycle (Figure 3) as shown by others In conclusion, there are characteristic patterns of baseline [9]. Local changes in flux were prominent at palmar sites, flux in the skin of the hand and forearm reflecting the especially the finger tip, at rates which correlate with underlying microvascular structure. The responses to ion- previous descriptions of vasomotor function [13, 14], and tophoresis of acetylcholine are also site dependent, with this was also demonstrated on the DOF. palmar sites failing to respond, probably because of differences Local differences in patterns and magnitude of baseline in transdermal delivery due to differences in skin structure. flux may reflect differences in the anatomical arrangement Sites on the dorsum of the hand and fingers are the most of vessels and the local control of vascular tone [15]. At suitable for LDF recording, and for examining responses to palmar sites, particularly at the finger pulp, there are large iontophoresis of acetylcholine. numbers of AVAs, which reduce in number towards the palm [7]. The characteristics of flux at palmar sites would be consistent with the particularly high flows in these vessels, References especially at the finger tip. In contrast there are few, if any, 1 Kellogg DL, Johnson JM, Kosiba WA. Selective abolition of AVAs at the forearm sites or dorsal sites on the hand, where adrenergic vasoconstrictor responses in skin by local the flux was lower and showed smaller variability, consistent iontophoresis of bretylium. Am J Physiol 1989; 257: with the predominantly nutritive function of the microvascu- H1599 H1606. lature at these sites. The dorsum of the fingers, with fewer 2 Lindblad L, Ekenvall L. Alpha-adrenoceptors in the vessels of arterio-venous anastomoses than the palmar surface, displayed human skin. Acta Physiol Scand 1986; 128: values of flux and flux variability intermediate between 3 Noon J, Hand M, Jordan P, Simpson J, Walker B, Webb D. Iontophoresis of acetylcholine causes prostanoid-mediated palmar and forearm sites. Differential changes in skin vasodilatation of human skin microvessels. Br J Pharmacol microcirculation flux recorded by LDF at different sites on 1995; 114: 83P. the dorsum of fingers, and finger pulp have recently been 4 Garn S. Types and distribution of hair in man. Ann NY Acad described in response to intra-arterial infusion of nitric oxide Sci 1951; 53: 398. synthase inhibitors [16]. In the study presented here 5 Holbrook K, Odland G. Regional differences in the thickness comparison of the response to transdermal delivery of the (cell layers) of the human stratum corneum: an ultrastructural endothelium-dependent vasodilator, acetylcholine, was made analysis. J Invest Dermatol 1974; 62: which was associated with an increase in flux on the volar 6 Scheuplein R. Site variation in diffusion and permeability. In aspect of the forearm, the dorsum of hand and finger sites, The Physiology and Pathophysiology of the Skin, volume 5), but in contrast to the above no change in flux was observed Jarrett A, ed., New York: Academic Press at the palmar sites. There are a number of possible 7 Clark ER. Arterio-venous anastomoses. Physiol Rev 1938; 18: explanations for this. The skin structure may determine the amount and the rate at which transdermal drugs reach the 8 Grant R, Bland E. Observations on arteriovenous anastomoses microcirculation at any given site. The major barrier to in human skin and the bird s foot with special reference to iontophoresis is the stratum corneum, with the appendages, the reaction to the cold. Heart 1931; 15: particularly the sweat glands, of prime importance in the 9 Fagrell B. Dynamics of skin microcirculation in humans. transport of charged molecules [17 19]. The palmar surface J Cardiovasc Pharmacol 1985; 3(suppl): Bongard O, Fagrell B. Variations in laser Doppler flux and of the hand has a much thicker stratum corneum, at 400 flow motion patterns in the dorsal skin of the human foot. 600 mm [6], in contrast to the thinner stratum corneum at Microvasc Res 1990; 39: the other sites [5, 6]. The thick keratin layer at the palmar 11 Braverman I, Schechner J. Contour mapping of the cutaneous sites is likely to act as a barrier to the passage of charged, microvasculature by computerised laser Doppler velocimetry. hydrophilic molecules, and of an electrical current, and may J Invest Dermatol 1991; 97: be particularly rich in proteases and cholinesterases [20]. 12 Braverman IM, Schechner JS, Silverman DG, KehYen A. The size of the relative response was greatest at the Topographic mapping of the cutaneous microcirculation using forearm sites where the skin was thinnest. The time from two outputs of laser-doppler flowmetry: Flux and the the start of iontophoresis to the first increase in flux might concentration of moving blood cells. Microvasc Res 1992; 44: be expected to vary according to the distance from the surface of the skin to the underlying vessels, and the 13 Salured E. Rhythmical variation in human blood flow. Int structures through which the acetylcholine had to pass. J Microcirc Clin Exp 1983; 2: Blackwell Science Ltd Br J Clin Pharmacol, 43 (4)

7 Variability in vasodilatation to transdermal acetylcholine 14 Tenland T. Spatial and temporal variation in human blood hormone across excised nude mouse skin. J Pharm Sci 1986; flow. Int J Microcirc Clin Exp 1983; 2: : Braverman I, Keh A, Goldminz D. Correlations of laser 19 Burnette RR, Bagniefski TM. Influence of constant current Doppler wave patterns with underlying microvascular iontophoresis on the impedance and passive Na + permeability anatomy. J Invest Dermatol 1990; 95: of excised nude-mouse skin. J Pharm Sci 1988; 77: Noon J, Haynes W, Webb D, Shore A. Local inhibition of 20 Theiß U, Kuhn I, Lücker P. Iontophoresis Is there a future nitric oxide generation in man reduces blood flow in finger for clinial application? Meth Find Exp Clin Pharmacol 1991; 13: pulp but not hand dorsum skin. J Physiol 1996; 490: Grimnes S. Pathway of ionic flow through human skin in vivo. Acta Dermatol Venereol 1984; 64: Burnette R, Marrero D. Comparisons between the ( Received 5 March 1996, iontophoretic and passive transport of thyrotropin releasing accepted 12 November 1996) 1997 Blackwell Science Ltd Br J Clin Pharmacol, 43 (4) 397