- Email: [email protected]
Skin vasoreactivity to insulin iontophoresis is reduced in elderly subjects and is absent in treated non-insulin-dependent diabetes patients
Biomedicine & Pharmacotherapy 58 (2004) 560–565 http://france.elsevier.com/direct/BIOPHA/
Dossier: Diabetes: Basic research and clinical approach
Skin vasoreactivity to insulin iontophoresis is reduced in elderly subjects and is absent in treated non-insulin-dependent diabetes patients Marco Rossi a,*, Adamasco Cupisti a, Roberto Ricco a, Gino Santoro a, Ferdinando Pentimone a, Angelo Carpi b b
a Department of Internal Medicine, University of Pisa, Pisa, Italy Department of Reproduction and Ageing, University of Pisa, Pisa, Italy
Received 16 October 2003 Available online 12 October 2004
Abstract We investigated the skin vasoreactivity to insulin in normal subjects and in treated non-insulin-dependent diabetes mellitus (NIDDM) patients. We measured cutaneous perfusion by laser-Doppler flowmetry (LDF) at rest and during skin cathodal iontophoresis (six pulses of 0.1 mA each for 20 s, with 40 s interval between stimulations) of insulin (0.1 ml Humulin R 100 IU/ml diluted 1/10 with of 0.9% saline solution) in 45 healthy subjects (HS), (25 males, 20 females, aged 45 ± 18 years), and in 15 treated NIDDM patients (13 males), aged 66 ± 8 years. Fifteen of the HS were used as controls. In these 15 sex- and age-matched HS and in the patients, we assessed also the skin postischemic hyperemia by LDF. In HS cutaneous blood flux response (CBF) to iontophoresis of insulin in saline (expressed as percent changes from baseline) was significantly higher than CBF response to iontophoresis of pure saline (maximum response: 360 ± 51% versus 172 ± 42%, respectively; P < 0.001, ANOVA for repeated measures). The maximum “net” CBF response to insulin (response to insulin minus response to saline) showed a negative correlation (r = –0.361; P < 0.01) with age in HS, and resulted significantly lower in the oldest than in the youngest HS (105 ± 40% versus 307 ± 45%, respectively; P < 0.01). No significant correlation was observed between the maximum CBF response to saline and the age of subjects. In NIDDM patients the “net” CBF response to insulin iontophoresis resulted significantly lower than in 15 sex- and age-matched control subjects (maximum response: –50 ± 89% versus 201 ± 81%, respectively; P < 0.001, ANOVA for repeated measures). No significant difference was observed between diabetics and controls, nor in basal perfusion (6.5 ± 1.3 IU versus 6.8 ± 1.7 IU, respectively) neither in the skin postischemic hyperemia (250 ±14% versus 258 ± 27%, respectively). These results confirm that insulin iontophoresis induces a skin vasodilatatory effect in normal subjects and show that this effect is reduced by aging and is absent in treated NIDDM patients. The local skin vasodilatatory effect induced by insulin seems to involve mechanisms different from those underlying the skin postischemic hyperemia. © 2004 Elsevier SAS. All rights reserved. Keywords: Insulin; Iontophoresis; Laser Doppler flowmetry; Skin; Diabetes mellitus
1. Introduction Recent findings suggest that insulin plays an important role as a vasoactive hormone, reducing pre-capillary arteriolar tone and increasing the number of perfused capillaries [15,21]. It has also been suggested that this vasodilatory action is functionally linked to insulin’s metabolic activity since it can shift the blood flow from non-nutritive vessels to nutritive capillaries, enhancing insulin and glucose disposal * Corresponding author. Dipartimento di Medicina Interna, Università degli Studi di Pisa, Via Roma 67, 56100 Pisa, Italy. E-mail address: [email protected] (M. Rossi). 0753-3322/$ - see front matter © 2004 Elsevier SAS. All rights reserved. doi:10.1016/j.biopha.2004.09.002
to the cells [1,4]. Recently the concept of “insulin vascular resistance” has been proposed to indicate a reduced vascular responsiveness to insulin [11]. Namely, in patients affected by non-insulin-dependent diabetes mellitus (NIDDM), a blunted insulin-mediated blood flow response in skeletal muscle has been observed [9] and recent findings in subjects with insulin resistance and arterial hypertension suggest that vascular insulin resistance may be the link between elevated blood pressure and the metabolic disorder [20]. On the other hand, a redistribution of skin blood flow favouring the nutritive microcirculation has been observed during systemic insulin administration in
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
561
patients with NIDDM [23]. However, these studies were not able to ascertain between a direct or an indirect effect of insulin on the vascular bed. More recently, a direct vasodilatory effect of insulin on skin microcirculation has been observed in 10 normal subjects by means of a laser-Doppler flowmeter coupled with cutaneous iontophoresis delivery [21], a method that allows to investigate the pure local effects of a substance on the skin perfusion [5,6,10,12,17,18]. We evaluated the effect of insulin iontophoresis on skin perfusion in normal subjects and in treated NIDDM patients with these following aims: (1) to confirm and extend the preliminary observation in normal subjects; (2) to perform a preliminary investigation in treated NIDDM patients.
Patients suffering from familial hyperlipaemia or chronic renal failure as well as heavy smokers (more than five cigarettes per day) were excluded. Two patients had a previous myocardial infarction; no patients presented a clinically evident cerebral or lower extremity arterial disease.
2. Methods
2.4. Methods
2.1. Normal subjects Forty-five healthy subjects (25 males, 20 females, aged 45 ± 18 years) were recruited among students, nurses, physicians, technical staff or visitors to our department. All subjects were considered healthy based on their medical history, physical examination, common laboratory tests, nondiabetic according to American Diabetes Association [7] and normotensive, following three office arterial blood pressure measurements. 2.2. NIDDM patients Fifteen patients affected by NIDDM (13 males, 2 females, aged 66 ± 8 years) were recruited for the study from our outpatients clinic and their main clinical features are shown in Table 1. All NIDDM patients were on oral antidiabetic therapy with a good control as assessed by a mean ± S.D. glycohemoglobin level of 7.1 ± 1.4% (normal range 4–6%). The duration of disease was 10 ± 4 years. No patients were on insulin therapy. All patients developed mild arterial hypertension, following the onset of diabetes, and were on antihypertensive treatment, showing normal blood pressure values at the study. The antihypertensive drugs were calcium channel blockers in six, angiotensin II receptor antagonists in two and ACE-inhibitors in seven cases. Table 1 Principal clinical parameters (mean ± S.E.M.) in 15 NIDDM patients and in 15 healthy age-matched controls Parameter M/F Age, years BMI Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Glycohemoglobin, % Total cholesterol, mg/dl Triglycerides, mg/dl
Diabetics (n = 15) 13/2 66 ± 2.1 30.6 ± 0.82 133 ± 3.4 83 ± 1.9 129 ± 3 7 ± 0.4 197 ± 6.6 142 ± 15
Controls (n = 15) 13/2 64 ± 2.1 24.5 ± 0.55 129 ± 3.1 81 ± 2.4 90 ± 7 5.4 ± 0.1 186 ± 4.2 158 ± 4.2
2.3. Control subjects Fifteen healthy subjects sex- and age-matched with patients (13 males, 2 females, aged 64 ± 8 years) were selected from the group of normal subjects (see Table 1). The study was approved by the local Ethic Committee of the Pisa University Hospital. All the participants to the study gave their informed consent to the study.
Skin vasoreactivity to insulin was evaluated by laserDoppler flowmetry coupled with iontophoresis, a method already used for the evaluation of skin vasodilatory response to acetylcholine and to sodium nitroprussiate [5,6,10,12, 17,18]. A battery-powered iontophoresis controller (Perijont 328, Power Supply), equipped with a drug delivery electrode (PF 383, Perimed, Jarfallan, Sweden) and with an indifferent electrode (PF 384, Perimed, Jarfallan, Sweden), was used to provide the current needed for insulin delivery, as previously reported [19,21]. The skin blood flux was measured by a laser-Doppler flowmeter (Periflux PF4001, standard probe PF408, Perimed, Jarfallan, Sweden), a method which allows evaluation of microvascular skin perfusion in real time [3,13]. The technique is based on this principle: a beam of laser light, carried by a fibre-optic probe, is widely scattered and partly absorbed by the tissue being studied. The light hitting the circulating blood cells changes the wavelength and is picked up by a returning fibre. The instrument converts the information into an electronic signal which is recorded continuously by an interfaced personal computer (Acer, Travelmate 202 T), equipped with Perisoft dedicated software, which allows the measurement of laser-Doppler flux, expressed as Perfusion Unit (PU). The probe calibration was performed by a device composed by a colloidal latex particles whose Brownian motion provides the standard value. Examinations were performed in a temperature controlled room (22 ± 1 °C), in the morning, with the subjects lying in supine position from 20 min, fasted for at least 12 h. Insulin (0.1 ml Humulin R 100 IU/ml diluted 1/10 with of 0.9% saline) and the control substance (0.1 ml of 0.9% saline) were delivered, after 5 min of baseline blood flux registration, on the medial surface of each arm, in a singleblind randomized order, with the same cathodal iontophoresis protocol: six pulses (0.1 mA) for 20 s each, followed by 40 s interval before the next stimulation. Cathodal iontophoresis is the preferable procedure, respect to anodal ionto-
562
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
phoresis, to deliver insulin to the skin. In fact, as regular insulin is in an anionic form [19] and the pH of the diluting medium for soluble insulin approximates 7.40, the regular insulin preparation remains mainly negatively charged and hence can be easily delivered by cathodal iontophoresis. The blood flux response to iontophoresis of insulin and saline were expressed as maximum percentage change from the baseline. The difference between the blood flux response to insulin and the blood flux response to control substance was assumed as the “net” blood flux response to insulin [21]. In the 15 NIDDM patients and in the 15 control subjects, the skin perfusion was assessed also before and during postischemic hyperemia, using a procedure previously reported [24]. This test was performed 30’ after the iontophoresis test, placing the laser-Doppler probe, fixed by a double-sided adhesive disk, to the dorsal surface of right foot, between the second and the third metatarsal radius, and positioning a blood pressure cuff loosely placed around the right thigh. Stable baseline perfusions were recorded for 5 min. Then the cuff was inflated to suprasystolic pressure for 3 min and the laser-Doppler signal was recorded during reactive hyperemia until the recovery. The postischemic hyperemia was expressed both as absolute maximal value during hyperemia and as percent maximal increase from baseline (medium value during 5 min before ischemia) [24]. The reproducibility of laser-Doppler measurement was estimated by repeated examination of six subjects during three consecutive days; the day-to-day variation resulted below 10%.
Fig. 1. Skin blood flux responses (mean ± S.E.M.) to cathodal iontophoresis of insulin and to cathodal iontophoresis of saline (expressed as percent change from baseline) in 45 normal healthy subjects. * P < 0.01. * * P < 0.001.
2.5. Statistical analysis All the data are expressed as mean ± S.E.M. Statistical analysis was performed using: ANOVA for repeated measures, Scheffè’s for multiple comparison testing, one way ANOVA, Student’s t-test for unpaired data and the Pearson’s linear correlation test.
3. Results
Fig. 2. Correlation (r = –0,35; P < 0.01) between the “net” cutaneous blood flux response to insulin (response to insulin minus response to saline) and age in the 45 normal healthy subjects studied.
3.1. Normal subjects In healthy subjects cathodal iontophoresis of insulin in saline induced a significantly higher cutaneous blood flux response than pure saline, with the maximum response of 360 ± 51% and 172 ± 42%, respectively (P < 0.001, ANOVA for repeated measures) (Fig. 1). The “net” maximal blood flux response to cathodal iontophoresis of insulin was negatively related to age (r = –0.35; P < 0.01) in the whole healthy population studied (Fig. 2). No significant correlation was observed between the maximal blood flux response to cathodal iontophoresis of saline and the age of subjects. The subdivision of healthy subjects into tertiles by age showed a significant (P < 0.01) lower “net” skin blood flux
response to insulin in the holder subjects than in the youngest subjects (105 ± 40% versus 307 ± 45%, respectively; P < 0.01; medium value: 200 ± 33%) (Fig. 3). 3.2. NIDDM patients No significant difference was observed in basal cutaneous perfusion between NIDDM patients and the sex- and agematched control subjects (6.8 ±1.7 UI versus 6.5 ± 1.2 UI, respectively). A “net” skin blood flux increase to cathodal iontophoresis of insulin occurred in controls (maximum response: 201 ± 81%) while it was absent in NIDDM patients (maximum response: –50 ± 89%; P < 0.001) (Fig. 4).
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
563
Fig. 5. Basal cutaneous blood flux and maximal cutaneous postischemic blood flux (expressed in P.U.) in 15 treated NIDDM patients and in 15 healthy subjects sex- and age-matched (mean ± S.E.M.). Fig. 3. “Net” cutaneous blood flux response to insulin (response to insulin minus response to saline) in the 45 healthy subjects subdivided into tertiles by age. * P < 0.01 between the oldest and the youngest subjects.
Fig. 4. “Net” cutaneous blood flux responses (mean ± S.E.M.) to cathodal iontophoresis delivery of insulin in 15 treated NIDDM patients and in 15 healthy subjects sex- and age-matched. * P < 0.001, diabetic patients versus controls.
No significant difference was observed between NIDDM patients and controls in maximal skin blood flux response to saline (181 ± 43.3% versus 313 ± 86.9%, respectively). As to the postischemic hyperemia test in NIDDM patients and controls, we did not observe any difference either in basal foot skin perfusion or in maximal skin postischemic hyperemia both as absolute values (21.6 ± 3 UI versus 21.2 ± 2.8 26.7 UI, respectively) (Fig. 5) and as percentage change from baseline (220 ± 14.7% versus 258 ± 26.7%, respectively).
4. Discussion and conclusion In recent years iontophoresis coupled with the laserDoppler flowmetry has been used for studying skin blood flow changes following local delivery of acetylcholine and sodium nitroprusside. Cutaneous endothelial function in normal subjects and in patients with various diseases was investigated by this method [5,6,10,12,17,18]. More recently iontophoresis has been used by Serné et al. [21] to study the skin vasodilatory response to insulin in 10 healthy subjects. We used an insulin iontophoresis technique very similar to that reported by Serné et al. [21]. Serné found a mean net cutaneous blood flux increment of 82 ± 28% in response to insulin iontophoresis, in 10 normal young subjects, while we observed in our 45 normal subjects, a mean net cutaneous blood flux increment of 200 ± 33% to iontophoresis of insulin. Our results confirm in a larger group of healthy subjects that cathodal iontophoresis allows insulin to enter the skin and elicit a local hyperaemic response. The smaller increment in cutaneous blood flux in response to insulin iontophoresis observed by Serné is probably related to the thermostatic probe set to 30 °C used by this author, that recorded a basal cutaneous blood flux (33 ± 7.9 IU) higher than in our patients (6.5 ± 1.2 IU), studied with a nonthermostatic probe. As some cutaneous vasodilatation occurring also in response to cathodal iontophoresis of saline alone (172 ± 42%), we considered, as response to insulin, the difference between the absolute response to the hormone minus the response to control substance (namely the “net” response), in accordance with Serné et al. [21]. The skin hyperemic response to saline is attributed to different substances involved in cutaneous vasodidation, as substance P, calcitonine gene-related peptide and neurokinin induced by the depolarisation of nociceptive nerves of the
564
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
skin [25]. Therefore the skin blood flux response to cathodal iontophoresis of saline can be considered for evaluating nonspecific skin vasodilation mechanisms and the integrity of the axon reflex in this vascular response. The negative relationship between the net cutaneous vasodilatory response to insulin and age, we observed in healthy subjects, can be attributed to reduced skin vessel compliance by aging. However, we have not observed this age dependent response to saline in the same subjects, suggesting that the tested non-specific vasodilation capacity did not influence the reduced skin blood flux response to insulin in elderly subjects. This study with iontophoresis coupled with laser-Doppler flowmeter allowed to investigate the local effect of insulin on total skin blood flow while the previous studies explored the effect of systemic administration of this hormone in diabetic patients [9,22]. Tooke et al. [22] reported a reduction of resting finger blood flow, measured by venous occlusion plethysmography, in type 1 diabetic patients after some days of continuous subcutaneous insulin infusion, and a significant increase of nailfold capillary red cell velocity, indicating a redistribution of skin blood flow. On the contrary, Laakso [9] observed a reduced skeletal muscle blood flow in response to venous insulin infusion in obese NIDDM patients. Our study did not aim to explore the possible mechanisms for the impaired skin vasodilatory response to insulin observed in treated NIDDM patients; however, our results lead to some considerations. The impaired hyperemic response to insulin in our NIDDM patients may be the consequence of the reported entothelial dysfunction in NIDDM [12]. Arterial hypertension is another pathologic condition known to be associated to endothelial dysfunction [14,16]. Our patients became affected by mild arterial hypertension following the onset of diabetes and they did not show an impaired skin hyperemic response to ischemia compared to controls. Increasing evidence indicates that the skin postischemic hyperemia represents a suitable method for the assessment of endothelial function in microcirculatory vessels [2,8], and recently it as been shown that this effect is related to flow-mediated release of prostaglandins from endothelium [2]. The preserved postischemic hyperemia observed in NIDDM patients suggests that different mechanisms are involved in the impaired skin blood flux response to insulin observed in the same patients. Moreover, the preserved skin blood flux response to saline solution observed in NIDDM patients suggests that the possible bias related to the hypothetical presence of diabetic neuropathy is negligible, taking into account the role of the axon reflex in this response. These results, as a whole, suggest that the blunted cutaneous vasodilatory response observed in NIDDM is related to a specific vascular resistance to insulin, which may be implicated in the vascular risk profile of these patients. Our data are the first observation on direct skin vasoreactivity to insulin in NIDDM patients. Our patients were in good metabolic control. Therefore the difference in skin
vasoreactivity to insulin observed from normal subjects cannot be attributed to the chronic metabolic disorder. In conclusion our results confirm that cathodal iontophoresis coupled with laser Doppler is a useful method in evaluating the cutaneous hyperemic response to insulin and show that this response is impaired in elderly normal subjects and absent in treated NIDDM patients. This last result coupled with the preserved skin postischemic hyperemia in our NIDDM patients suggests that local skin vasodilatory effect induced by insulin involves mechanisms different from the endothelium dependent mechanisms underlying the skin postischemic hyperemia.
References [1]
[2]
[3] [4]
[5]
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13]
[14]
Baron AD, Tarshoby M, Hook G, Lazaridis EN, Cronin J, Johonson A, Steinberg HO. Interaction between insulin sensitivity and muscle perfusion on glucose uptake in human skeletal muscle: evidence of capillary recruitment. Diabetes 2002;49:768–74. Binggeli C, Spieker LE, Corti R, Sudano I, Stojanovic V, Hayoz D, Luscher TF, Noll G. Statins enhance postischemic hyperemia in the skin circulation of hypercholesterolemic patients. E monitoring test of endothelial dysfunction for clinical practise? JACC 2003;42:71–7. Bonner R, Nossal R. Model of laser Doppler measurement of blood flow in tissue. Appl Ptic 1981;20:2097–3007. Cleland S, Petrie JR, Ueda S, Elliott HL, Connell JM. Insulinmediated vasodilation and glucose uptake are functionally linked in humans. Hypertension 1999;33:554–8. Cupisti A, Rossi M, Fabbri A, Placidi A, Morelli E, Vagheggini G, et al. Responses of skin microcirculation to acetylcholine in patients with essential hypertension and in normotensive patients with chronic renal failure. Nephron 2000;85:114–9. Cupisti A, Rossi M, Placidi S, Caprioli R, Morelli E, Vagheggini G, et al. Response of the skin microcirculation to acetylcholine and sodium nitroprusside in chronic uremic patients. Int J Clin Lab Res 2000;30:157–62. Harris MI, Eastman RC, Cowie CC, Flegal KM, Eberhardt MS. Comparison of diabetes diagnostic categories in the U.S. population according to the 1997 American Diabetes Association and 1980– 1985 World Health Organisation diagnostic criteria. Diabetes Care 1997;20:1859–63. Kubli S, Waeber B, Dalle-Ave A, Feihl F. Reproducibilty of laser Doppler imaging of skin blood flow as a tool to assess endothelial function. J Cardiovasc Pharmacol 2000;36:640–8. Laakso M, Edelman SV, Brechtel G, Baron AD. Impaired insulinmediated skeletal muscle blood flow in patients with NDDM. Diabetes 1992;41:1076–83. La Civita U, Rossi M, Vagheggini G, Credidio L, Fabbri A, Giusti C, et al. Microcirculatory effects of acetylcholine and sodium nitroprusside in patients with systemic sclerosis: a preliminary investigation. Arch Rheum Dis 1998;6:231–4. Mather K, Anderson TJ, Verma S. Insulin action in the vasculature: physiology pathophysiology. J Vasc Res 2001;38:415–22. Morris SJ, Shore AC, Tooke JE. Responses of the skin microcirculation to acetylcholine and sodium nitroprusside in patients with NIDDM. Diabetologia 1995;38:1337–44. Nilsson GE, Tenland T, Oeberg PA. Evaluation of laser Doppler flowmeter for measurement of tissue blood flow. IEEE Trans Biomed Eng 1980;27:597–604. Panza JA, Quyumi AA, Brush JE, Epstein SE. Abnormal endothelium dependent vascular relaxation in patients with essential hypertension. N Engl J Med 1990;323:22–7.
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565 [15] Rattigan S, Clark MG, Barrett E. J Hemodynamic action of insulin in rat skeletal muscle: evidence for capillary recruitment. Diabetes 1997; 46:1381–8. [16] Rossi M, Taddei S, Fabbri A, Tintori G, Credidio L, Virdis A, et al. Cutaneous vasodilation to acetylcholine in patients with essential hypertension. J Cardiovasc Pharmacol 1997;29:406–11. [17] Rossi MA, Cupisti A, Mariani S, Santoro G, Pentimone F. Endothelium-dependent and endothelium-independent skin vasoreactivity in the elderly. Aging Clin Exp Res 2002;14:343–6. [18] Rossi M, Cupisti A, Perrone L, Mariani S, Santoro G. Acute effect of exercise-induced leg ischemia on cutaneous vasoreactivity in patients with stage II peripheral artery disease. Microvasc Res 2002;64:14–20. [19] Sage BH. Insulin iontophoresis. Pharm Biotechnol 1997;10:319–41. [20] Serné EH, Gans RO, Maaten JC, Wee PM, Donker AJ, Stehouver CD. Capillary recruitment is impaired in essential hypertension and relates to insulin’s metabolic and vascular actions. Cardiovasc Res 2001;49: 161–8.
565
[21] Serné EH, Ijezerman RG, Gans RO, Nijveldt R, De Vries G, Evertz R, Donker JM, Stehouwer CD. Direct evidence for insulin-induced capillary recruitment in skin of healthy subjects during physiological hyperinsulinemia. Diabetes 2002;51:1515–22. [22] Tooke JE, Lins PE, Ostergren J, Adamson U, Fagrell B. The effects of intravenous insulin infusion on skin microcirculatory flow in Type 1 diabetes. Int J Microcirc Clin Exp 1985;4:69–83. [23] Tymms DJ, Tooke JE. The effects of continuous subcutaneous insulin infusion (CSII) on microvascular blood flow in diabetes mellitus. Int J MicrocicClin Exp 1988;7:347–56. [24] Wahlberg E, Olofson P, Takolander R, Swedemborg J. Postocclusive reactive hyperemia estimating peripheral vascular resistance: a noninvasive method to predict outcome of infrainguinal vascular reconstruction? J Cardiovasc Surg (Torino) 1996;37:545–52. [25] Wardell K, Naver HK, Nilsson GE, Wallin BG. The cutaneous axon reflex in humans characterized by laser Doppler perfusion imaging. J Physiol 1993;460:185–9.
Dossier: Diabetes: Basic research and clinical approach
Skin vasoreactivity to insulin iontophoresis is reduced in elderly subjects and is absent in treated non-insulin-dependent diabetes patients Marco Rossi a,*, Adamasco Cupisti a, Roberto Ricco a, Gino Santoro a, Ferdinando Pentimone a, Angelo Carpi b b
a Department of Internal Medicine, University of Pisa, Pisa, Italy Department of Reproduction and Ageing, University of Pisa, Pisa, Italy
Received 16 October 2003 Available online 12 October 2004
Abstract We investigated the skin vasoreactivity to insulin in normal subjects and in treated non-insulin-dependent diabetes mellitus (NIDDM) patients. We measured cutaneous perfusion by laser-Doppler flowmetry (LDF) at rest and during skin cathodal iontophoresis (six pulses of 0.1 mA each for 20 s, with 40 s interval between stimulations) of insulin (0.1 ml Humulin R 100 IU/ml diluted 1/10 with of 0.9% saline solution) in 45 healthy subjects (HS), (25 males, 20 females, aged 45 ± 18 years), and in 15 treated NIDDM patients (13 males), aged 66 ± 8 years. Fifteen of the HS were used as controls. In these 15 sex- and age-matched HS and in the patients, we assessed also the skin postischemic hyperemia by LDF. In HS cutaneous blood flux response (CBF) to iontophoresis of insulin in saline (expressed as percent changes from baseline) was significantly higher than CBF response to iontophoresis of pure saline (maximum response: 360 ± 51% versus 172 ± 42%, respectively; P < 0.001, ANOVA for repeated measures). The maximum “net” CBF response to insulin (response to insulin minus response to saline) showed a negative correlation (r = –0.361; P < 0.01) with age in HS, and resulted significantly lower in the oldest than in the youngest HS (105 ± 40% versus 307 ± 45%, respectively; P < 0.01). No significant correlation was observed between the maximum CBF response to saline and the age of subjects. In NIDDM patients the “net” CBF response to insulin iontophoresis resulted significantly lower than in 15 sex- and age-matched control subjects (maximum response: –50 ± 89% versus 201 ± 81%, respectively; P < 0.001, ANOVA for repeated measures). No significant difference was observed between diabetics and controls, nor in basal perfusion (6.5 ± 1.3 IU versus 6.8 ± 1.7 IU, respectively) neither in the skin postischemic hyperemia (250 ±14% versus 258 ± 27%, respectively). These results confirm that insulin iontophoresis induces a skin vasodilatatory effect in normal subjects and show that this effect is reduced by aging and is absent in treated NIDDM patients. The local skin vasodilatatory effect induced by insulin seems to involve mechanisms different from those underlying the skin postischemic hyperemia. © 2004 Elsevier SAS. All rights reserved. Keywords: Insulin; Iontophoresis; Laser Doppler flowmetry; Skin; Diabetes mellitus
1. Introduction Recent findings suggest that insulin plays an important role as a vasoactive hormone, reducing pre-capillary arteriolar tone and increasing the number of perfused capillaries [15,21]. It has also been suggested that this vasodilatory action is functionally linked to insulin’s metabolic activity since it can shift the blood flow from non-nutritive vessels to nutritive capillaries, enhancing insulin and glucose disposal * Corresponding author. Dipartimento di Medicina Interna, Università degli Studi di Pisa, Via Roma 67, 56100 Pisa, Italy. E-mail address: [email protected] (M. Rossi). 0753-3322/$ - see front matter © 2004 Elsevier SAS. All rights reserved. doi:10.1016/j.biopha.2004.09.002
to the cells [1,4]. Recently the concept of “insulin vascular resistance” has been proposed to indicate a reduced vascular responsiveness to insulin [11]. Namely, in patients affected by non-insulin-dependent diabetes mellitus (NIDDM), a blunted insulin-mediated blood flow response in skeletal muscle has been observed [9] and recent findings in subjects with insulin resistance and arterial hypertension suggest that vascular insulin resistance may be the link between elevated blood pressure and the metabolic disorder [20]. On the other hand, a redistribution of skin blood flow favouring the nutritive microcirculation has been observed during systemic insulin administration in
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
561
patients with NIDDM [23]. However, these studies were not able to ascertain between a direct or an indirect effect of insulin on the vascular bed. More recently, a direct vasodilatory effect of insulin on skin microcirculation has been observed in 10 normal subjects by means of a laser-Doppler flowmeter coupled with cutaneous iontophoresis delivery [21], a method that allows to investigate the pure local effects of a substance on the skin perfusion [5,6,10,12,17,18]. We evaluated the effect of insulin iontophoresis on skin perfusion in normal subjects and in treated NIDDM patients with these following aims: (1) to confirm and extend the preliminary observation in normal subjects; (2) to perform a preliminary investigation in treated NIDDM patients.
Patients suffering from familial hyperlipaemia or chronic renal failure as well as heavy smokers (more than five cigarettes per day) were excluded. Two patients had a previous myocardial infarction; no patients presented a clinically evident cerebral or lower extremity arterial disease.
2. Methods
2.4. Methods
2.1. Normal subjects Forty-five healthy subjects (25 males, 20 females, aged 45 ± 18 years) were recruited among students, nurses, physicians, technical staff or visitors to our department. All subjects were considered healthy based on their medical history, physical examination, common laboratory tests, nondiabetic according to American Diabetes Association [7] and normotensive, following three office arterial blood pressure measurements. 2.2. NIDDM patients Fifteen patients affected by NIDDM (13 males, 2 females, aged 66 ± 8 years) were recruited for the study from our outpatients clinic and their main clinical features are shown in Table 1. All NIDDM patients were on oral antidiabetic therapy with a good control as assessed by a mean ± S.D. glycohemoglobin level of 7.1 ± 1.4% (normal range 4–6%). The duration of disease was 10 ± 4 years. No patients were on insulin therapy. All patients developed mild arterial hypertension, following the onset of diabetes, and were on antihypertensive treatment, showing normal blood pressure values at the study. The antihypertensive drugs were calcium channel blockers in six, angiotensin II receptor antagonists in two and ACE-inhibitors in seven cases. Table 1 Principal clinical parameters (mean ± S.E.M.) in 15 NIDDM patients and in 15 healthy age-matched controls Parameter M/F Age, years BMI Systolic BP, mmHg Diastolic BP, mmHg Fasting blood glucose, mg/dl Glycohemoglobin, % Total cholesterol, mg/dl Triglycerides, mg/dl
Diabetics (n = 15) 13/2 66 ± 2.1 30.6 ± 0.82 133 ± 3.4 83 ± 1.9 129 ± 3 7 ± 0.4 197 ± 6.6 142 ± 15
Controls (n = 15) 13/2 64 ± 2.1 24.5 ± 0.55 129 ± 3.1 81 ± 2.4 90 ± 7 5.4 ± 0.1 186 ± 4.2 158 ± 4.2
2.3. Control subjects Fifteen healthy subjects sex- and age-matched with patients (13 males, 2 females, aged 64 ± 8 years) were selected from the group of normal subjects (see Table 1). The study was approved by the local Ethic Committee of the Pisa University Hospital. All the participants to the study gave their informed consent to the study.
Skin vasoreactivity to insulin was evaluated by laserDoppler flowmetry coupled with iontophoresis, a method already used for the evaluation of skin vasodilatory response to acetylcholine and to sodium nitroprussiate [5,6,10,12, 17,18]. A battery-powered iontophoresis controller (Perijont 328, Power Supply), equipped with a drug delivery electrode (PF 383, Perimed, Jarfallan, Sweden) and with an indifferent electrode (PF 384, Perimed, Jarfallan, Sweden), was used to provide the current needed for insulin delivery, as previously reported [19,21]. The skin blood flux was measured by a laser-Doppler flowmeter (Periflux PF4001, standard probe PF408, Perimed, Jarfallan, Sweden), a method which allows evaluation of microvascular skin perfusion in real time [3,13]. The technique is based on this principle: a beam of laser light, carried by a fibre-optic probe, is widely scattered and partly absorbed by the tissue being studied. The light hitting the circulating blood cells changes the wavelength and is picked up by a returning fibre. The instrument converts the information into an electronic signal which is recorded continuously by an interfaced personal computer (Acer, Travelmate 202 T), equipped with Perisoft dedicated software, which allows the measurement of laser-Doppler flux, expressed as Perfusion Unit (PU). The probe calibration was performed by a device composed by a colloidal latex particles whose Brownian motion provides the standard value. Examinations were performed in a temperature controlled room (22 ± 1 °C), in the morning, with the subjects lying in supine position from 20 min, fasted for at least 12 h. Insulin (0.1 ml Humulin R 100 IU/ml diluted 1/10 with of 0.9% saline) and the control substance (0.1 ml of 0.9% saline) were delivered, after 5 min of baseline blood flux registration, on the medial surface of each arm, in a singleblind randomized order, with the same cathodal iontophoresis protocol: six pulses (0.1 mA) for 20 s each, followed by 40 s interval before the next stimulation. Cathodal iontophoresis is the preferable procedure, respect to anodal ionto-
562
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
phoresis, to deliver insulin to the skin. In fact, as regular insulin is in an anionic form [19] and the pH of the diluting medium for soluble insulin approximates 7.40, the regular insulin preparation remains mainly negatively charged and hence can be easily delivered by cathodal iontophoresis. The blood flux response to iontophoresis of insulin and saline were expressed as maximum percentage change from the baseline. The difference between the blood flux response to insulin and the blood flux response to control substance was assumed as the “net” blood flux response to insulin [21]. In the 15 NIDDM patients and in the 15 control subjects, the skin perfusion was assessed also before and during postischemic hyperemia, using a procedure previously reported [24]. This test was performed 30’ after the iontophoresis test, placing the laser-Doppler probe, fixed by a double-sided adhesive disk, to the dorsal surface of right foot, between the second and the third metatarsal radius, and positioning a blood pressure cuff loosely placed around the right thigh. Stable baseline perfusions were recorded for 5 min. Then the cuff was inflated to suprasystolic pressure for 3 min and the laser-Doppler signal was recorded during reactive hyperemia until the recovery. The postischemic hyperemia was expressed both as absolute maximal value during hyperemia and as percent maximal increase from baseline (medium value during 5 min before ischemia) [24]. The reproducibility of laser-Doppler measurement was estimated by repeated examination of six subjects during three consecutive days; the day-to-day variation resulted below 10%.
Fig. 1. Skin blood flux responses (mean ± S.E.M.) to cathodal iontophoresis of insulin and to cathodal iontophoresis of saline (expressed as percent change from baseline) in 45 normal healthy subjects. * P < 0.01. * * P < 0.001.
2.5. Statistical analysis All the data are expressed as mean ± S.E.M. Statistical analysis was performed using: ANOVA for repeated measures, Scheffè’s for multiple comparison testing, one way ANOVA, Student’s t-test for unpaired data and the Pearson’s linear correlation test.
3. Results
Fig. 2. Correlation (r = –0,35; P < 0.01) between the “net” cutaneous blood flux response to insulin (response to insulin minus response to saline) and age in the 45 normal healthy subjects studied.
3.1. Normal subjects In healthy subjects cathodal iontophoresis of insulin in saline induced a significantly higher cutaneous blood flux response than pure saline, with the maximum response of 360 ± 51% and 172 ± 42%, respectively (P < 0.001, ANOVA for repeated measures) (Fig. 1). The “net” maximal blood flux response to cathodal iontophoresis of insulin was negatively related to age (r = –0.35; P < 0.01) in the whole healthy population studied (Fig. 2). No significant correlation was observed between the maximal blood flux response to cathodal iontophoresis of saline and the age of subjects. The subdivision of healthy subjects into tertiles by age showed a significant (P < 0.01) lower “net” skin blood flux
response to insulin in the holder subjects than in the youngest subjects (105 ± 40% versus 307 ± 45%, respectively; P < 0.01; medium value: 200 ± 33%) (Fig. 3). 3.2. NIDDM patients No significant difference was observed in basal cutaneous perfusion between NIDDM patients and the sex- and agematched control subjects (6.8 ±1.7 UI versus 6.5 ± 1.2 UI, respectively). A “net” skin blood flux increase to cathodal iontophoresis of insulin occurred in controls (maximum response: 201 ± 81%) while it was absent in NIDDM patients (maximum response: –50 ± 89%; P < 0.001) (Fig. 4).
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
563
Fig. 5. Basal cutaneous blood flux and maximal cutaneous postischemic blood flux (expressed in P.U.) in 15 treated NIDDM patients and in 15 healthy subjects sex- and age-matched (mean ± S.E.M.). Fig. 3. “Net” cutaneous blood flux response to insulin (response to insulin minus response to saline) in the 45 healthy subjects subdivided into tertiles by age. * P < 0.01 between the oldest and the youngest subjects.
Fig. 4. “Net” cutaneous blood flux responses (mean ± S.E.M.) to cathodal iontophoresis delivery of insulin in 15 treated NIDDM patients and in 15 healthy subjects sex- and age-matched. * P < 0.001, diabetic patients versus controls.
No significant difference was observed between NIDDM patients and controls in maximal skin blood flux response to saline (181 ± 43.3% versus 313 ± 86.9%, respectively). As to the postischemic hyperemia test in NIDDM patients and controls, we did not observe any difference either in basal foot skin perfusion or in maximal skin postischemic hyperemia both as absolute values (21.6 ± 3 UI versus 21.2 ± 2.8 26.7 UI, respectively) (Fig. 5) and as percentage change from baseline (220 ± 14.7% versus 258 ± 26.7%, respectively).
4. Discussion and conclusion In recent years iontophoresis coupled with the laserDoppler flowmetry has been used for studying skin blood flow changes following local delivery of acetylcholine and sodium nitroprusside. Cutaneous endothelial function in normal subjects and in patients with various diseases was investigated by this method [5,6,10,12,17,18]. More recently iontophoresis has been used by Serné et al. [21] to study the skin vasodilatory response to insulin in 10 healthy subjects. We used an insulin iontophoresis technique very similar to that reported by Serné et al. [21]. Serné found a mean net cutaneous blood flux increment of 82 ± 28% in response to insulin iontophoresis, in 10 normal young subjects, while we observed in our 45 normal subjects, a mean net cutaneous blood flux increment of 200 ± 33% to iontophoresis of insulin. Our results confirm in a larger group of healthy subjects that cathodal iontophoresis allows insulin to enter the skin and elicit a local hyperaemic response. The smaller increment in cutaneous blood flux in response to insulin iontophoresis observed by Serné is probably related to the thermostatic probe set to 30 °C used by this author, that recorded a basal cutaneous blood flux (33 ± 7.9 IU) higher than in our patients (6.5 ± 1.2 IU), studied with a nonthermostatic probe. As some cutaneous vasodilatation occurring also in response to cathodal iontophoresis of saline alone (172 ± 42%), we considered, as response to insulin, the difference between the absolute response to the hormone minus the response to control substance (namely the “net” response), in accordance with Serné et al. [21]. The skin hyperemic response to saline is attributed to different substances involved in cutaneous vasodidation, as substance P, calcitonine gene-related peptide and neurokinin induced by the depolarisation of nociceptive nerves of the
564
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565
skin [25]. Therefore the skin blood flux response to cathodal iontophoresis of saline can be considered for evaluating nonspecific skin vasodilation mechanisms and the integrity of the axon reflex in this vascular response. The negative relationship between the net cutaneous vasodilatory response to insulin and age, we observed in healthy subjects, can be attributed to reduced skin vessel compliance by aging. However, we have not observed this age dependent response to saline in the same subjects, suggesting that the tested non-specific vasodilation capacity did not influence the reduced skin blood flux response to insulin in elderly subjects. This study with iontophoresis coupled with laser-Doppler flowmeter allowed to investigate the local effect of insulin on total skin blood flow while the previous studies explored the effect of systemic administration of this hormone in diabetic patients [9,22]. Tooke et al. [22] reported a reduction of resting finger blood flow, measured by venous occlusion plethysmography, in type 1 diabetic patients after some days of continuous subcutaneous insulin infusion, and a significant increase of nailfold capillary red cell velocity, indicating a redistribution of skin blood flow. On the contrary, Laakso [9] observed a reduced skeletal muscle blood flow in response to venous insulin infusion in obese NIDDM patients. Our study did not aim to explore the possible mechanisms for the impaired skin vasodilatory response to insulin observed in treated NIDDM patients; however, our results lead to some considerations. The impaired hyperemic response to insulin in our NIDDM patients may be the consequence of the reported entothelial dysfunction in NIDDM [12]. Arterial hypertension is another pathologic condition known to be associated to endothelial dysfunction [14,16]. Our patients became affected by mild arterial hypertension following the onset of diabetes and they did not show an impaired skin hyperemic response to ischemia compared to controls. Increasing evidence indicates that the skin postischemic hyperemia represents a suitable method for the assessment of endothelial function in microcirculatory vessels [2,8], and recently it as been shown that this effect is related to flow-mediated release of prostaglandins from endothelium [2]. The preserved postischemic hyperemia observed in NIDDM patients suggests that different mechanisms are involved in the impaired skin blood flux response to insulin observed in the same patients. Moreover, the preserved skin blood flux response to saline solution observed in NIDDM patients suggests that the possible bias related to the hypothetical presence of diabetic neuropathy is negligible, taking into account the role of the axon reflex in this response. These results, as a whole, suggest that the blunted cutaneous vasodilatory response observed in NIDDM is related to a specific vascular resistance to insulin, which may be implicated in the vascular risk profile of these patients. Our data are the first observation on direct skin vasoreactivity to insulin in NIDDM patients. Our patients were in good metabolic control. Therefore the difference in skin
vasoreactivity to insulin observed from normal subjects cannot be attributed to the chronic metabolic disorder. In conclusion our results confirm that cathodal iontophoresis coupled with laser Doppler is a useful method in evaluating the cutaneous hyperemic response to insulin and show that this response is impaired in elderly normal subjects and absent in treated NIDDM patients. This last result coupled with the preserved skin postischemic hyperemia in our NIDDM patients suggests that local skin vasodilatory effect induced by insulin involves mechanisms different from the endothelium dependent mechanisms underlying the skin postischemic hyperemia.
References [1]
[2]
[3] [4]
[5]
[6]
[7]
[8]
[9]
[10]
[11] [12]
[13]
[14]
Baron AD, Tarshoby M, Hook G, Lazaridis EN, Cronin J, Johonson A, Steinberg HO. Interaction between insulin sensitivity and muscle perfusion on glucose uptake in human skeletal muscle: evidence of capillary recruitment. Diabetes 2002;49:768–74. Binggeli C, Spieker LE, Corti R, Sudano I, Stojanovic V, Hayoz D, Luscher TF, Noll G. Statins enhance postischemic hyperemia in the skin circulation of hypercholesterolemic patients. E monitoring test of endothelial dysfunction for clinical practise? JACC 2003;42:71–7. Bonner R, Nossal R. Model of laser Doppler measurement of blood flow in tissue. Appl Ptic 1981;20:2097–3007. Cleland S, Petrie JR, Ueda S, Elliott HL, Connell JM. Insulinmediated vasodilation and glucose uptake are functionally linked in humans. Hypertension 1999;33:554–8. Cupisti A, Rossi M, Fabbri A, Placidi A, Morelli E, Vagheggini G, et al. Responses of skin microcirculation to acetylcholine in patients with essential hypertension and in normotensive patients with chronic renal failure. Nephron 2000;85:114–9. Cupisti A, Rossi M, Placidi S, Caprioli R, Morelli E, Vagheggini G, et al. Response of the skin microcirculation to acetylcholine and sodium nitroprusside in chronic uremic patients. Int J Clin Lab Res 2000;30:157–62. Harris MI, Eastman RC, Cowie CC, Flegal KM, Eberhardt MS. Comparison of diabetes diagnostic categories in the U.S. population according to the 1997 American Diabetes Association and 1980– 1985 World Health Organisation diagnostic criteria. Diabetes Care 1997;20:1859–63. Kubli S, Waeber B, Dalle-Ave A, Feihl F. Reproducibilty of laser Doppler imaging of skin blood flow as a tool to assess endothelial function. J Cardiovasc Pharmacol 2000;36:640–8. Laakso M, Edelman SV, Brechtel G, Baron AD. Impaired insulinmediated skeletal muscle blood flow in patients with NDDM. Diabetes 1992;41:1076–83. La Civita U, Rossi M, Vagheggini G, Credidio L, Fabbri A, Giusti C, et al. Microcirculatory effects of acetylcholine and sodium nitroprusside in patients with systemic sclerosis: a preliminary investigation. Arch Rheum Dis 1998;6:231–4. Mather K, Anderson TJ, Verma S. Insulin action in the vasculature: physiology pathophysiology. J Vasc Res 2001;38:415–22. Morris SJ, Shore AC, Tooke JE. Responses of the skin microcirculation to acetylcholine and sodium nitroprusside in patients with NIDDM. Diabetologia 1995;38:1337–44. Nilsson GE, Tenland T, Oeberg PA. Evaluation of laser Doppler flowmeter for measurement of tissue blood flow. IEEE Trans Biomed Eng 1980;27:597–604. Panza JA, Quyumi AA, Brush JE, Epstein SE. Abnormal endothelium dependent vascular relaxation in patients with essential hypertension. N Engl J Med 1990;323:22–7.
M. Rossi et al. / Biomedicine & Pharmacotherapy 58 (2004) 560–565 [15] Rattigan S, Clark MG, Barrett E. J Hemodynamic action of insulin in rat skeletal muscle: evidence for capillary recruitment. Diabetes 1997; 46:1381–8. [16] Rossi M, Taddei S, Fabbri A, Tintori G, Credidio L, Virdis A, et al. Cutaneous vasodilation to acetylcholine in patients with essential hypertension. J Cardiovasc Pharmacol 1997;29:406–11. [17] Rossi MA, Cupisti A, Mariani S, Santoro G, Pentimone F. Endothelium-dependent and endothelium-independent skin vasoreactivity in the elderly. Aging Clin Exp Res 2002;14:343–6. [18] Rossi M, Cupisti A, Perrone L, Mariani S, Santoro G. Acute effect of exercise-induced leg ischemia on cutaneous vasoreactivity in patients with stage II peripheral artery disease. Microvasc Res 2002;64:14–20. [19] Sage BH. Insulin iontophoresis. Pharm Biotechnol 1997;10:319–41. [20] Serné EH, Gans RO, Maaten JC, Wee PM, Donker AJ, Stehouver CD. Capillary recruitment is impaired in essential hypertension and relates to insulin’s metabolic and vascular actions. Cardiovasc Res 2001;49: 161–8.
565
[21] Serné EH, Ijezerman RG, Gans RO, Nijveldt R, De Vries G, Evertz R, Donker JM, Stehouwer CD. Direct evidence for insulin-induced capillary recruitment in skin of healthy subjects during physiological hyperinsulinemia. Diabetes 2002;51:1515–22. [22] Tooke JE, Lins PE, Ostergren J, Adamson U, Fagrell B. The effects of intravenous insulin infusion on skin microcirculatory flow in Type 1 diabetes. Int J Microcirc Clin Exp 1985;4:69–83. [23] Tymms DJ, Tooke JE. The effects of continuous subcutaneous insulin infusion (CSII) on microvascular blood flow in diabetes mellitus. Int J MicrocicClin Exp 1988;7:347–56. [24] Wahlberg E, Olofson P, Takolander R, Swedemborg J. Postocclusive reactive hyperemia estimating peripheral vascular resistance: a noninvasive method to predict outcome of infrainguinal vascular reconstruction? J Cardiovasc Surg (Torino) 1996;37:545–52. [25] Wardell K, Naver HK, Nilsson GE, Wallin BG. The cutaneous axon reflex in humans characterized by laser Doppler perfusion imaging. J Physiol 1993;460:185–9.