Improvement in Autonomic Function with Rosiglitazone in

Improvement in Autonomic Function with Rosiglitazone in Type 2 Diabetes

Jerrold Petrofsky, PhD, JD*

Scott Lee, MD*

Salameh Bweir, PT, MPH*

Michael Laymon, DPT, Sc†

*The Departments of Endocrinology and Physical Therapy, Loma Linda University, Loma Linda, California

†The Department of Physical Therapy, Azusa Pacific University

This work was supported in part by a grant from GlaxoSmithKline, 0313416800.

KEY WORDS: Diabetes, heat exposure, cardiovascular, thermoregulation, orthostatic, sweat, thiazolidinedione, rosiglitazone, endothelial, autonomic, dermal vascular flow, C-reactive protein, tumor necrosis
factor-a

Abstract

The insulin sensitizer rosiglitazone (RSG) has shown positive effects on the endothelial vasculature, but no study has evaluated its effect on the autonomic nervous system. We hypothesized that the RSG would improve autonomic neuropathy in diabetic subjects through a positive effect on the endothelium of the nervous system microvasculature. We therefore studied the autonomic component of two organ systems: the cardiovascular system and dermal vascular system. Five insulin-resistant diabetics underwent dynamic testing for cardiovascular reactivity, dermal vascular flow, and thermoregulation. Cardiovascular reactivity was assessed with tilt table testing performed in a heated environmental room (42˚C). Dermal vascular flow was assessed with laser Doppler flowmetry. Three different parameters of thermal regulation were analyzed: lower limb cutaneous microcirculatory flow on hairy skin, hairy skin sweat rates (by sweat hygrometry), and core temperature. We correlated our autonomic findings with postocclusion forearm hyperemia as an index of endothelial cell function. RSG, 4 mg was given daily for four weeks to 5 insulin resistant diabetics (4 male, 65 + 14 years). Baseline evaluation was performed along with 4-week evaluation after initiation of RSG. All subjects experienced impairment in the compensatory blood pressure response and redistribution of cardiac output when placed in a heated environment at base line. The fall in systolic blood pressure (reduction of 21 mm Hg when tilted P < .03) was accompanied by corresponding reduction in stroke volume and cardiac output (from 5.06 L/min to 4.75 L/min), but no change was demonstrated in heart rate. Four weeks after the initiation of RSG, orthostatic tolerance significantly improved. Systolic blood pressure did not fall, but remained relatively constant (increase of 2.4 mm HG when tilted P = NS). Improvements in thermoregulation were noted: increased sweat rate, cutaneous flow and appropriate reductions in core temperature. Autonomic improvements correlated with improvements in endothelial function, and reductions in tumor necrosis factor-a (TNF-a) and C-reactive protein (CRP).

Introduction

Thiazolidinedione (TZD) therapy, rosiglitazone and pioglitazone, are high affinity ligands for peroxisome proliferator-activated receptor (PPAR)-l, a regulatory protein of the nuclear hormone receptor superfamily.1,2 One of the primary mechanisms of action of TZDs in type 2 diabetes is increased insulin sensitivity in adipose tissue with resultant enhanced glucose uptake in the cells and subsequent lowering of ambient glucose levels.3 A large body of scientific evidence, however, supports the role of TZDs in improving markers of insulin resistance syndrome, tumor necrosis factor-a (TNF-a)4,5 and nontraditional inflammatory markers of cardiovascular disease, C–reactive protein (CRP).6

Expression of PPAR-l is thought to occur in adipose tissue,2 endothelium,7 smooth muscle cells,8 and monocytes and macrophages.9 Thiazolidinedione therapy has also been shown to have positive effects on endothelial cell function, suggesting a possible anti-atherogenic action.10 Markers for endothelial cell function such as asymmetric dimethylarginine (ADMA), reactive oxygen species generation, myocardial blood flow, and vascular activity have been shown to improve with rosiglitazone.11–14

When the endothelium is healthy, nitric oxide is present and appropriate vasodilation occurs in response to the endothelium. A healthy endothelium acts as a protective barrier to plaque formation in the vasculature and is resistant to clotting and penetration by monocytes to form an atheroma.15 Troglitazone and rosiglitazone have shown a positive effect on endothelial function by demonstrating improvement in ischemia induced flow mediated vasodilation of the forearm.16

Although the insulin sensitizer rosiglitazone (RSG) has demonstrated positive effects on the endothelial vasculature, no study has assessed its effect on the autonomic nervous system. We hypothesized that RSG would improve autonomic neuropathy in diabetic subjects through its effect on the endothelium of the nervous system microvasculature. We, therefore, studied the autonomic component of two organ systems: the cardiovascular system and dermal vascular system. Five insulin resistant diabetics underwent dynamic autonomic testing for cardiovascular reactivity, dermal vascular flow, and thermoregulation. Global heat exposure was performed on subjects to observe centrally mediated neurogenic alterations of heat induced vasodilation in the hairy skin and centrally mediated sudomotor baroreceptor receptor function during cardiovascular tilt table testing. Thermal regulation was analyzed using a series of three different parameters: lower limb cutaneous microcirculatory flow on hairy skin, hairy skin sweat rates (by sweat hygrometry) and corresponding core temperature. To assess the contribution of the endothelium, we correlated our autonomic findings with upper limb postischemic hyperemia as an index of endothelial cell function. Blood volume was measured in the arm with Whitney plethysmography. We further correlated our findings with nontraditional markers for the insulin resistance syndrome (TNF-a) and cardiovascular disease (CRP).

Subjects

The subjects in this study were 5 patients. The average height was 71.6 ± 7.9cm, average weight was 93.6 ± 27.6kg, and average age was 65.2 ± 13.8 years. The Hb_Alc of the subjects at study onset was 8.5 ± 1.5. Body mass index was 29.6 ± 6.8. All procedures were approved by the committee on human experimentation, and all subjects signed a statement of informed consent.

Methods

Skin Blood Flow

Skin blood flow was measured by a laser Doppler flowmeter produced by Moor Instruments, Inc. (LDV 304, Oxford, England). The device sat on a stand 35 cm above the left lower leg of the subject while the patient was lying supine. The device scanned the body and produced a picture of blood flow to the skin. The scanned area was 117 x 149 pixels, and the scan rate was 4 ms/pixel. This device was completely noninvasive and had no physical contact with the body. The error on repeat measurements is less than 5% from day to day. An area of 25 cm2 was scanned over a 2-minute period. An area of 10 cm2 was chosen in the center, and markers placed on the skin allowed repeat measurements in this same area of interest under different experimental conditions. The laser was warmed for 30 minutes before flow measurements to increase stability. After the subject was tilted to a 45˚ head-up position, care was taken to move the laser to the same distance and angle (90˚) relative to the leg of the subject. In preliminary testing on 4 subjects, the laser head was moved ± 30˚ to see if the flows would change due to minor misalignments of the head during tilt; no differences were seen in flows when the head was tilted in this range. To assure that the same area of the leg was scanned, marks were placed on the outside of the scan area, and the laser was positioned to scan the same area at each body position. The units of blood flow stated in the results are in “flux” units, the measure of flow generated by laser Doppler systems.

Cytokine Methods

Tumor necrosis factor-a and CRP levels were determined from peripheral serum samples. Baseline (before intervention) and weekly serum level determinations were acquired for the duration of the study.

Standard venipuncture techniques were used to extract blood samples, which were allowed to clot for 1 hour at room temperature, centrifuged for 10 minutes (4˚ C), and the serum extracted. Grossly hemolytic, lipidic, or turbid samples were not used. Samples assayed within 48 hours were stored at 2˚ to 8˚C. Samples not assayed within 48 hours were stored at –20˚C to avoid loss of bioactivity and contamination.

Samples were assayed using an enzyme linked immunosorbent assay (ELISA). Procedures listed in ELISA kits from Anitgenix America to analyze TNF-a and CRP were followed. Briefly, 100 or 200 µl of serum was added to an antibody-coated microtiter plate. Samples where incubated and washed, conjugate added and incubated, substrate added and incubated, and stop solution added with samples read within 30 minutes at O.D. 450 nm using an MTX model MCC 340 microtiter plate reader.

Forearm Blood Flows

Forearm blood flows were measured by Whitney strain gauge plethysmography. Whitney strain gauge plethysmography is a technique of measuring limb blood flow by volume plethysmography. Briefly, a mercury in rubber strain gauge is placed around the forearm. The gauge is made of silastic and is prestretched of 20 g to eliminate hysteresis during the measurements. The gauge is first calibrated by stretching the gauge a set distance and measuring the output on a chart recorder. An arterial occlusion cuff is then placed around the wrist and a venous cuff on the upper arm. Thirty seconds before flow measurements, the occlusion cuff is inflated to 200 mm Hg on the wrist to remove the hand circulation from measurements of forearm flow. During flow measurements, the upper arm cuff placed approximately 4 cm above the elbow is inflated rapidly to a pressure of 55 mm Hg. The cuff is inflated for a period of 5 seconds. During this period the change in arm size is transduced by the Whitney string gauge to an electrical output and displayed on a chart recorder. After 5 seconds, a vacuum pump rapidly deflates the cuff for a period of 7 seconds. During measurements of flow, the arm is placed as close to the level of the heart as possible while the arm was placed in a controlled temperature water bath at 37˚C. The arm was immersed in the water bath for a period of 15 minutes to allow the entire arm to come to equilibrium at central core temperature. A detailed description of this technique is published elsewhere.17,18

Body Temperature

Central core temperature was measured by a thermocouple placed aurally under the back of the tongue (Yellow Springs Instruments, Yellow Springs, OH). The mouth was kept closed for 1 minute during temperature measurements and subjects breathed through their noses.

Blood Pressure

Blood pressure was measured by auscultation of the inactive limb with a sphygmomanometer. The systolic blood pressure was determined as the first sound as the pressure is reduced in the cuff while the diastolic was identified as the change from a sharp sound to a muffle. The cuff was deflated at 3 mm Hg/s as per the American Heart Association Standards.

Sweat

Sweat was measured by a sweat hygrometry system. The system is composed of a source of compressed dry air, a flow meter, sweat capsules, and a hygrometer. Compressed air at a pressure of approximately 1,000 mm Hg was stored in a 30-L pressure reservoir. The output of the air was regulated through a regulator and flow meter such that an output of 50 mL per minute flowed through individual sweat capsules applied to the skin. Each capsule was 2 cm2 in diameter and was round. The capsule had a cavity inside and was held to the skin by an elastic strap. The internal capsule had connections to two hoses. The air inlet provided a source of dry air, and the air outlet was connected to a manifold and into a hydrometer system. The hydrometer consisted of a sensor using calcium chloride to measure the humidity in the air. The greater the humidity, the lower the electrical resistance of the calcium chloride. Therefore, an electrical output was provided proportional to relative humidity. By knowing the humidity, the gas temperature and flow rate of air across the sweat capsules, and the cross sectional area of the capsule, we could calculate the sweat rate per square centimeter of the skin. This technique has been described previously.19

Orthostatic Tolerance

A tilt table will be used to change the body position of the subject. The tilt table is a commercially available tilt table used as a standing frame for people with osteoporosis or orthostatic intolerance. A motor drive operated through a series of buttons allows the table to go from the horizontal to the vertical position. Pads and straps maintain the subjects body position at all times if the subject becomes dizzy. The subject cannot fall off the table. The table was fully padded.

Cardiac Output

Cardiac output was measured by impedance plethysmography. This technique involves placing electrodes on the neck, top of chest, abdomen, and leg. The most cephalic and distal electrodes pass a 1 mA alternating current at a frequency of 50 KH. The electrical impedance is measured from the inside electrodes, and cardiac output is calculated and displayed.

Procedures

Two series of experiments were performed as described below. At the beginning of each experimental day, subjects started by resting comfortably in a thermally neutral room for about 15 minutes. The first series of experiments evaluated simply any impairment of metabolic control of blood flow by examining the effects of vascular occlusion on the microcirculation. Here blood flow is influenced by build-up of metabolites altering activity of the endothelial cells of the arterioles and not under the influence of neurogenic control of the autonomic nervous system. In the second series of experiments, various levels of neurogenic control were evaluated by increasing the complexity of the autonomic response from local segmental reflexes to whole body responses to stress by a combination of postural changes and whole body heating.

Series 1: Metabolic Control of
the Circulation

In this first series of experiments, the reactivity of the arm vascular bed was assessed. Muscle temperature varies in the forearm between 27˚ and 42˚C. This is due to the fact that the forearm is a shell tissue and temperature varies to help gain or loose heat from the core of the body.20 Limb tissue temperature varies with room temperature, clothing, body fat content, and the phase of the menstrual cycle.21 Therefore, a thin person may have resting arm metabolism less than 20% of that of a person with a high body fat content due to the high Q10 of the tissues.22 Therefore, to remove some of the variability in previous studies, a water bath was used to elevate all forearm temperatures to that of the core, 37ºC. Subjects placed their arms in a bath with the arm held dependent and the elbow at an angle of 90º so that their arms were submerged to the belly of the biceps muscle. The bath was well stirred. After 15 minutes, resting arm flows were recorded. A 4-minute period of arterial occlusion induced by a cuff under the axilla and inflated to 200 mm Hg was then used. After this period, flows were measured for 3 minute.

Series 2

The subject entered a thermoregulatory room where temperature was controlled at 32.2˚C ± 2˚, and humidity was controlled at 35% ± 5%. The subject laid comfortably in the horizontal posture wearing shorts and a thin T-shirt on a tilt table. At the end of the 30-minute period, an area of 10 cm2 just below the knee was scanned with a laser Doppler flow meter, as described under methods. Core temperature, cardiac output, blood pressure, and sweat rates were measured as described above. After the flows were measured, the table was tilted to the 45˚ vertical position over a period of 15 seconds. The laser Doppler flow meter was moved so that the flow meter was maintained 35 cm above the table with the laser at 90˚ perpendicular to the leg and flows were measured in exactly the same location with the subject tilted into the 45˚ vertical position. The laser then scanned the body and flows were measured again. This procedure was repeated before subjects took rosiglitazone, and then at 2 and 4 weeks after taking a dose of 4 mg per day of rosiglitazone.

Statistical Analysis

Statistical analysis involved the calculation of means, standard deviations, and 2-tailed ANOVA. The level of significance was P < .05. All data in “results” is expressed ± standard deviation.

Results

The resting flows in the forearm (Figure 1) and the flows during the 2-minute period after occlusion before taking and at 2 and 4 weeks after taking rosiglitazone are shown in Figure 2. Resting flows for the 5 subjects averaged 0.26 ± 0.13 mL per 100g tissue/min. After occlusion, flow increased to an average of 2.97 mL per 100g tissue/min and then exponentially decreased to rest over a period of 2 minutes. After 2 weeks on rosiglitazone, the average flows, as shown in this figure immediately after the release of the occlusion, averaged 3.9 ± 1.4 mL/100 g tissue/min. After 4 weeks on rosiglitazone, the flows increased to 6.1 ± 1.9 mL/100 g tissue/min. The increase at 2 and 4 weeks was significant. For basis of comparison, similar data are shown collected under the same experimental circumstances on age-matched control subjects from another study.18 As can be seen in this figure, subjects without diabetes had substantially higher resting flows and higher exercising flows than this group.

When exposed to a warm room for a period of 30 minutes, the average blood flow of the group is shown in Figure 3. For ease of presentation, the flows had been normalized in terms of the flow during the period prior to taking rosiglitazone. After 2 and 4 weeks of taking rosiglitazone, flows increased. For example, the average flow after taking rosiglitazone for 2 weeks measured in the skin after 30 minutes of exposure to heat was 120% ± 9% of the flow during the control period. In contrast, after 4 weeks of taking rosiglitazone the flows had increase to 136% ± 13% of those of the control period. These increases were significant.

Much more dramatic is the data shown in Figure 4. Normally, when posture is changed from the horizontal to the vertical position, to maintain blood pressure, the sympathetic nervous system constricts blood flow to the legs by as much as 50% or more.18 Figure 4 compares the flows measured in the horizontal posture to the flows measured in the vertical posture for subjects during the control period and after 2 and 4 weeks of taking rosiglitazone. Although flows would have been twice as high in the horizontal versus the vertical posture in control subjects, the subjects with diabetes, as has been shown previously in studies with diabetes,18 only showed a small reduction in flows (horizontal versus vertical flow ratio) when changed in position from the horizontal to the vertical posture. Figure 4 shows that the average ratio of horizontal to vertical flows when being tilted was 1.58 in the subjects before taking rosiglitazone. However, after 2 weeks of taking rosiglitazone, the ratio of horizontal to vertical flows increased to 1.66 and finally after 4 weeks of taking rosiglitazone, the flow ration increased to 1.79.

During heat exposure, subjects sweat very little before administration of rosiglitazone. For example, shown in Figure 5 are the sweat rates from the forehead, forearm, chest, and calf for the 5 subjects with type 2 diabetes involved in these studies. Sweat rate on the calf was the lowest, averaging only 0.05 mg/cm2/min for the group of 5 subjects. The highest sweat rate was on the forehead, averaging 0.34 mg/cm2/min. These flow rates, given the heat exposure to 90ºF, were lower than normally seen in control subjects.18 After rosiglitazone was administered to the subjects for 2 and 4 weeks, there was a sharp increase in sweating in all areas, as shown in Figure 5. As can be seen in this figure, for example, the sweat rate on the forehead increased from an initial sweat rate of 0.33 mg/cm2/min to 0.48 mg/cm2/min, an increase of approximately 50% over a period of 4 weeks. The same phenomenon was seen for the forearm, chest, and calf. In fact, in all areas, as can be shown in the average data in this figure and the appropriate standard deviations, there was an increase in sweat at both 2 and 4 weeks. This increase was significant (ANOVA).

These increases in sweat rates matched the reduction in core temperature. Before rosiglitazone was administered to these subjects, the core temperature increased by 0.52˚ ± 0.23˚C throughout the 30-minute heat exposure. As seen in Figure 6, by week 2, there was still an increase in core temperature but only 0.34˚ ± 0.31˚C and by week 4 the average increase was even less and was reduced to only 0.31˚ ± 0.31˚C. The reduction in temperature after 2 weeks and 4 weeks on rosiglitazone was significant in this group of subjects (P < .05).

Concerning the cardiovascular system, all subjects experienced impairment of the compensatory blood pressure response and redistribution of cardiac output when placed in a heated environment at the baseline data. The fall in systolic blood pressure averaged 21 mm Hg and was accompanied by a corresponding reduction in stroke volume and cardiac output from 5.06 to 4.75 L/min with no change in heart rate. As has been demonstrated previously,18 all subjects showed marked orthostatic intolerance when body position was changed after they were placed in a warm environment. Four weeks after the initiation of rosiglitazone, orthostatic intolerance improved even further. Systolic blood pressure did not fall when tilted in the heat, but remained relatively constant, in fact increasing by 2.4 mm Hg during tilt, and cardiac output was maintained constant.

The improvement in cardiovascular function was mirrored by a reduction in C-reactive protein and TNF-a as shown in Figure 7. The absolute values of CRP in the cytokines were very variable. For example, looking at TNF-a, one subject had a TNF-a of 48.4, whereas 3 others were approximately at 100, and the final 2 were over 130. Therefore, with this large degree of variability, both in TNF-a and the cytokines, in figure 7 the change in TNF-a and CRP were expressed as a percent change. When looking at a percent change during week 0 (baseline data), all data would show as 100%. As can be seen in this figure, the average reduction in TNF-a and cytokines averaged approximately a 25% reduction by week 4 and about half that reduction in TNF-a and CRP by week 2.

Discussion

At baseline, we showed autonomic neuropathy in our diabetic cohort. Global heat exposure was performed on subjects to observe centrally mediated neurogenic alterations with heat-induced vasodilation in the hairy skin and cardiovascular reactivity. All subjects experienced orthostatic intolerance, diminished cutaneous flow on the hairy skin of the lower extremity, and impairments in thermoregulation during heat exposure. Follow-up 1 month later, after initiation of rosiglitazone, revealed significant improvements in all affected systems. These centrally mediated gains in autonomic function were mirrored by improvements in postocclusive endothelial mediated hyperemia through Whitney plethysmography and other nontraditional markers of the insulin resistance syndrome (TNF-a) and cardiovascular disease (CRP).

Endothelial mediated vasodilation through the release of nitric oxide has been shown to be altered by diabetes and the insulin resistance syndrome.3,23 The thiazolidinediones and metformin have been shown to have antiatherogenic action through their positive effect on the endothelium.15,24,25 Stansberry et al.26 has postulated that a possible common etiologic factor between sudomotor disruptions seen in the dermovascular bed and subsequent alterations in dermal microcirculation is underlying endothelial dysfunction. Troglitazone in one study was found to enhance dermal vascular blood flow, and its mechanism of action was attributed in part to PPAR-l mediated improvements in endothelial function.27

The presence of microalbuminuria in the urine is considered an early indicator of not only renal disease but of cardiovascular disease.28,29 and all cause mortality.30,31

Intuitively, it would seem contradictory that a marker for microvascular complications of the kidney is also a strong marker for macrovascular risk. The link, however, between these seemingly separate problems is the widely held notion that the presence of microalbuminuria is a strong indicator for the impairment of vascular integrity.32 Improvements on the vasculature mediated by the endothelial action of PPAR-l activity of TZDs, therefore, may have broader implications than simply reduction of possible macrovascular risk. TZDs may also provide microvascular protection through their action on the endothelium independent of their effect on glycemic control. In fact, rosiglitazone has been shown to confer kidney protection and has been associated with lowering urinary albumin excretion.32,33 Our findings support this notion of microvascular benefit, and additionally suggest protection on another organ system susceptible to microcirculatory damage, the nervous system.

Autonomic neuropathy can have far reaching effects on a multitude of organ systems in individuals with diabetes: gastrointestinal, genitourinary, cardiovascular, papillary, dermal and counter-regulatory hormonal systems.34 The autonomic nerves are small fragile myelinated and unmyelinated fibers that are easily damaged. Such disruptions can damage can occur early in diabetes.35 Several studies in diabetic patients show that damage from autonomic neuropathy is often insidious in nature and can therefore slip by unnoticed by both patient and physician prior to any clinical symptoms such as overt gastroparesis or hypoglycemic unawareness.36

Our limited findings show autonomic neuropathy affecting the cardiovascular system, dermal vascular system, and thermoregulation. Postural instability as assessed by traditional tilt table testing at room temperature is a clear example of a subtle cardiovascular autonomic manifestation which can be missed.37 In patients with diabetes, sympathetic dysfunction can cause orthostasis (a drop of over 20 mm Hg in blood pressure) on standing. In a recent study observing cardiovascular reactivity in diabetics, only about 25% of diabetic patients at room temperature were found to have impaired orthostais on standing.38 In contrast, we enhanced the sensitivity of this test by exposing patients to a heated environment. We showed in this and previous experiments that nearly all diabetic patients experience orthostatic intolerance when placed under heat stress.18,39 Post initiation of rosiglitazone revealed restoration of cardiovascular reactivity with no orthostasis detected.

Sudomotor dysfunction in the dermal vascular bed can also be easily missed. Prior data on the prevalence of sympathetic dermal dysfunction were often contradictory at best.40 The development of the noninvasive laser Doppler has improved the ability to assess dermal vascular function.40 Abnormalities in the vasomotor reactivity of the skin microcirculation have been observed in patients with diabetes and are, in part, a result of damage to the peripheral sympathetic nervous system.26,27,40,41 Sympathetic regulation of vascular tone in the dermal microcirculation is complex involving central, short arc, and local reflex control mechanisms.40,42-44 Our preliminary baseline data support prior experiments in which defects in both vasoconstriction26,40,45,46 and vasodilation of dermal vasomotor function26,40,41 were observed.

Enhanced dermal vascular flow in the hairy skin of the lower extremity during heat exposure was shown after initiation of rosiglitazone and also corresponded to improvements in sweat rate. Damage to small blood vessels in the skin and to sympathetic cholinergic nerve fibers that innervate sweat glands47 cause a specific type of sympathetic neuropathy, resulting in death of sweat glands.48 At baseline, local and regional sweat rates were diminished in the diabetic cohort and were associated with an inappropriate rise in skin and core temperatures. Taken together, the positive changes seen in cutaneous microcirculatory flow and sweat rate on hairy skin after RSG resulted in overall improved thermoregulation with corresponding reductions in core temperature.

Our preliminary findings in this pilot study, though novel, are clearly limited by the small number of subjects and lack of controls. Larger, prospective studies with carefully matched controls are needed to draw stronger and more definitive conclusions.

References

1. Lehmann J, Moore L, Smith-Oliver T, et al. An antidiabetic thiazolidinedione is a high affinity ligand for PPAR- l. J Biol Chem 270:12953–12956, 1995.

2. Diep Q, Mabrouk M, Cohn J, et al. Structure, endothelial function, cell growth, and inflammation in blood vessels of angiotensin II-infused rats. Circulation 205:2296–2302, 2002.

3. Hsueh WA, Law RD. Cardiovascular risk continuum: implications of insulin resistance and diabetes. Am J Med 105:4S–14S 1998.

4. Peraldi P, Xu M, Spiegelman B. Thiazolidinediones block factor-a induced inhibition of insulin signaling. J Clin Invest 100:1863–1869, 1997.

5. Szalkowski D, White-Carrington S, Berger J, Zhang B. Antidiabetic thiazolidinediones block inhibitory effect of factor-a on differentiation, insulin-stimulated glucose uptake, and gene expression in 3T3-L1 cells. Endocrinology 136:1474–1481, 1995.

6. Haffner et al

7. Marx N, Bourcier T, Sukhova G, et al. PPAR-l activation in human endothelial cells increases plasminogen activator inhibitor type-1 expression: PPAR-l as a potential mediator in vascular disease. Aterioscler Thromb Vasc Biol 19:546–551, 1999.

8. Marx N, Schonbeck U, Lazar M, et al. PPAR-l activators inhibit gene expression and migration in human vascular smooth muscle cells. Circ Res 83:1097–1103, 1998.

9. Marx N, Sukhova G, Murphy C, et al. Macrophage in human atheroma contain PPAR-l differentiation-dependent PPAR-l expression and reduction of MMP-9 activity through PPAR-l activation in mononuclear phagocytes in vitro. Am J Pathol 153:17–23, 1998.

10. Viberti GC. Rosiglitazone: potential beneficial impact on cardiovascular disease. IJCP 57:128–134, 2003.

11. Mohanty P, Ghanim H, et al. (2002) Evidence for a potent anti-inflammatory effect or rosiglitazone in obese diabetes patients. Diabetes 51:A109, 2002.

12. Stuhlinger MC, Abbasi F, Chu JW, et al. Relationship between insulin resistance and an endogenous nitric oxide synthase inhibitor. JAMA 287:1420–1426, 2002.

13. Plummer EV, Lawson M, Domb A, et al. Effects of treatment with rosiglitazone on myocardial blood flow (MBF) in type 2 diabetes. (DM2). Diabetes 51:A161, 2002.

14. Natali A, Baldeweg S, Toschi E, et al. Rosiglitazone directly improves endothelial function in type 2 diabetic patients. Diabetes 51:A142, 2002.

15. Bell DS. Why I initiate insulin therapy with two insulin sensitizers in patients with type 2 diabetes. Endocr Pract 9:98–100, 2003.

16. Martens et al.

17. Whitney RJ. The measurement of volume changes in human limbs. J Physiol 121:1–17, 1953.

18. Petrofsky JS, Besonis C, Rivera D, et al. Does local heating really help diabetic patients increase circulation. J Orthop Neurol Surg In press, 2003.

19. Mitchell D, Senay LC, Wyndham CH, et al. Acclimatization in a hot, humid environment: energy exchange, body temperature, and sweating. J Appl Physiol 40:768–778, 1976.

20. Petrofsky JS, Le Donne D, Reinhart JS, Lind AR. The influence of the menstrual cycle on blood flow through muscle during isometric contractions. Ohio J Sci 1:236–238, 1981.

21. Petrofsky JS, Lind AR. The relationship of body fat content to deep muscle temperature and isometric endurance in man. Clin Sci Mol Med 48:405–412, 1975.

22. Wang G, Kawai M. Effect of temperature on elementary steps of the cross–bridge cycle in rabbit soleus slow-twitch muscle fibres. J Physiol Feb 531:219–34, 2001.

23. Henry RR. Type 2 diabetes care: The role of insulin-sensitizing agents and practical implications for cardiovascular disease prevention. Am J Med 105:20–26, 1998.

24. Walker AB, Naderali EK, Chattington PD, Buckingham RE, et al. Differential vasoactive effects of the insulin sensitizers rosiglitazone (BRL 49653) and trolitazone on human small arteries in vetro. Diabetes 47:810-814, 1998.

25. Walker AB, Chattington P, Buckingham R, et al. The thiazolidinedione rosiglitazone (BRL – 49653) lowers blood pressure and protects against impairment of endothelial function in Zucker fatty rats. Diabetes 48:1448–1453, 1999.

26. Stansberry KB, Peppard HR, Babyak LM, et al. Primary nociceptive afferents mediate the blood flow dysfunction in non-glabrous (hairy) skin of type 2 diabetes: A new model for the pathogenesis of microvascular dysfunction. Diabetes Care 22:1549–1554, 1999.

27. Vinik AI, Erbas T. Recognizing and treating diabetic autonomic neuropathy. Cleveland Clin J Med 68:928–30, 932, 934–944, 2001.

28. Mogensen C. Microalbuminuria, blood pressure and diabetic renal disease: Origin and development of ideas. Diabetologia 42:263–285, 1999.

29. Savage S, Estacio RO, Jeffers B, Schrier R. Urinary albumin excretion as a predictor of diabetic retinopathy, neuropathy, and cardiovascular disease in NIDDM. Diabetes Care 19: 1243–1248, 1996.

30. Gall MA, et al. Albuminuria and poor glycemic control predict mortality in NIDDM. Diabetes 44:1303–1309, 1999.

31. Neil A, et al. A prospective population-based study of microalbuminuria as a predictor of mortality in NIDDM. Diabetes Care 16: 996–1003, 1993.

32. Bakris G, Viberti G, Weston W, et al. Rosiglitazone reduces microalbumin excretion in type II diabetes. J Hum Hypertens 17:7–12, 2003.

33. Buckingham RE, et al. Peroxisome proliferator-activated receptor-gamma agonist, rosiglitazone, protects against nephropathy and pancreatic islet abnormalities in Zucker fatty rats. Diabetes 47:1326–1334, 1998.

34. Accurso V, Shamsuzzaman AS, Somers VK. Rhythms,rhymes,and reasons—spectral oscillations in neural cardiovascular control. Auton Neurosci 20;90:41–46, 2001.

35. Martine M. Involvement of autonomic nerve fibers in diabetes. Lancet 1:560–565, 1953.

36. Sagliocco L, Sartucci F, Giampietro O, Murri L. Amplitude loss of electrically and magnetically evoked sympathetic skin responses in early stages of type 1 (insulin-dependent) diabetes mellitus without signs of dysautonomia. Clin Auton Res 9:5–10, 1999.

37. Netea RT, Elving LD, Lutterman JA, Thien T. Body position and blood pressure measurement in patients with diabetes mellitus. J Intern Med 251:393–399, 2002.

38. Agrawal A, Saran R, Khanna R. Management of orthostatic hypotension from autonomic dysfunction in diabetics on peritoneal dialysis. Perit Dial Int 19:415–417, 1999.

39. Petrofsky JS, Besonis C, Rivera D, Schwab E, Lee S. Heat Tolerance in patients with diabetes. J Appl Res 3:28–34, 2003.

40. Stansberry KB, Hill MA, Shapiro SA, et al. Impairment of peripheral blood flow responses in diabetes resembles an enhanced aging effect. Diabetes Care 20:1711–1716, 1997.

41. Rendell M, Bergman T, O’Donnell G, et al. Microvascular blood flow, volume, and velocity measured by laser Doppler techniques in IDDM. Diabetes 38:819–824, 1998.

42. Coffman JD, Cohen RA. Adrenergic and serotonergic mechanisms in the human digit. J Cardiovasc Pharmacol 11:S49–S53, 1998.

43. Coffman JD, Cohen RA. Cholinergic vasodilator mechanism in the human finger. Am J Physiol 252:H594–H597, 1997.

44. Henriksen 1991.

45. Rendell M, Bamisedun O. Skin blood flow and current perception in pentoxifylline-treated diabetic neuropathy. Angiology 43:843–851, 1992.

46. McDaid EA, Monaghan B, Parker AI, et al. Peripheral autonomic impairment in patients newly diagnosed with type 2 diabetes. Diabetes Care 17:1422–1427, 1999.

47. Robertshaw D. Neuroendocrine control of sweat glands. J Invest Dermatol 69:121–129, 1977.

48. Fealey RD, Low PA, Thomas JE. Thermoregulatory sweating abnormalities in diabetes mellitus. Mayo Clin Proc 64:617–628, 1998.