Oxford Lipid Metabolism Group, Oxford Centre for Diabetes, Endocrinology and Metabolism, Radcliffe Infirmary, Oxford, UK

Correspondence to: F Karpe, Oxford Lipid Metabolism Group, Oxford Centre for Diabetes, Endocrinology and Metabolism, Radcliffe Infirmary, Oxford OX2 6HE, UK. E-mail: fredrik.karpe@oxlip.ox.ac.uk

INTRODUCTION: Adipose tissue blood flow (ATBF) increases after meal intake and a failure to regulate ATBF in the postprandial period seems to be a feature of insulin resistance and obesity. ATBF can be measured quantitatively by the ¹³³Xe washout technique, but the microdialysis ethanol escape method has also been employed to detect relative changes in ATBF.

METHODS: We compared ¹³³Xe washout and the recovery of exogenous ethanol and endogenous urea by microdialysis in abdominal subcutaneous adipose tissue, after physiological stimulation of ATBF by ingestion of oral glucose (75 g) in eight healthy people (age 23-52 y, body mass index (BMI) 19.4-29.6 kg/m²).

RESULTS: The ATBF response was heterogeneous. In subjects responding vigorously to the stimulus as measured by ¹³³Xe washout, the microdialysis ethanol escape was increased (indicating an increase in ATBF). An increased recovery of urea was observed, also indicating an increase in ATBF. The recovery of both small molecules was delayed compared with increased blood flow and failed to return to baseline in response to a rapid decline in ATBF.

CONCLUSION: We conclude that the ¹³³Xe washout technique is more responsive to physiological change in ATBF than ethanol escape or urea recovery by microdialysis.

International Journal of Obesity (2002) 26, 1-5. DOI: 10.1038/sj/ijo/0801843

blood flow; insulin; postprandial; glucose; subcutaneous

Introduction

Adipose tissue blood flow (ATBF) increases after meal intake. Human in vivo studies have suggested that ATBF may be involved in regulating biological processes such as triacylglycerol clearance. By increasing the rate of supply of substrate (triacyglycerols) to lipoprotein lipase, the rate of hydrolysis is increased.¹ Experiments in dogs indicate that an increase in ATBF may facilitate the removal of non esterified fatty acids (NEFA) produced during lipolysis² and in accordance with that finding, an increase in ATBF brought about by physical exercise was attenuated when lipolysis was blocked by nicotinic acid. This suggests that NEFAs may regulate ATBF.³ Conversely, a vasoconstrictor effect has been reported for NEFAs.⁴ Another role for modulation of ATBF could be to regulate the transport of endocrine molecules from the tissue.⁵

A failure to regulate ATBF in the postprandial period seems be a feature of insulin resistance and obesity.⁶ Quantification of ATBF may therefore be of interest to understand mechanisms of dyslipidaemia and insulin resistance. ATBF can be measured quantitatively by the ¹³³Xe washout technique.^7,8 Alternatively, the microdialysis ethanol escape method has been employed to detect relative changes in ATBF after local pharmacological delivery or external manipulation such as heating of the subcutaneous tissue.^9,10 In addition, increased ethanol escape has been reported after glucose ingestion¹¹ and after institution of hyperinsulinaemic normoglycaemia.¹² In this study, we compared ¹³³Xe washout and the recovery of exogenous ethanol and endogenous urea by microdialysis in abdominal subcutaneous adipose tissue after physiological stimulation of ATBF by ingestion of oral glucose. Urea was used as an endogenous marker with the assumption that there is no production in adipose tissue and that the background level was in steady state. Hypothetically, the microdialysis probe would drain the tissue of urea and any change in urea recovery would depend on an altered delivery to the tissue, presumably by an altered blood flow. Some of the results have been published previously in abstract form.¹³

Subjects and methods

Adipose tissue ethanol microdialysis

Studies were conducted on eight healthy subjects (six male, two female), median age 42 (23-52) y, with mean BMI 24.0 (19.4-29.6) kg/m² at rest, following an overnight fast. Subjects were asked to refrain from strenuous exercise, smoking or alcohol for 24 h beforehand. A CMA 60 microdialysis catheter (Biotech Instruments Limited, Kimpton, UK) was perfused overnight with sterile Ringer's solution. In the morning of the study, the catheter was inserted into the subcutaneous adipose tissue in the para-umbilical area. The probe was then perfused with Ringer's solution containing 30 mmol/l ethanol at 3 µl/min and allowed to equilibrate for 0.5 h. The effluent dialysate was collected every 10 min during the experiment. In order to minimise evaporation, the dialysate was collected into 500 µl microcentrifuge tubes and immediately placed on ice and subsequently kept at 4°C until analysed. The collected volume was determined by weighing and was diluted with an equal volume of phosphate buffered saline. Ethanol was measured, within 24 h of sampling, in the diluted dialysate using an enzymatic method (Sigma, Poole, UK), adapted for use on an IL Monarch centrifugal analyser (Instrumentation Laboratory, Warrington, UK). The intra-assay coefficient of variation was 3.8% for 30 duplicate (diluted) samples with a mean ethanol concentration of 4.9 mmol/l (1.9-10.1). Results were expressed as the ratio of the concentration of ethanol in the effluent dialysate to that in the perfusate (the outflow: inflow (O/I) ratio).

Adipose tissue ¹³³xenon washout

A dose of 2 MBq of ¹³³Xe in 0.9% saline was injected into the contralateral side, at the same depth as the microdialysis probe, 30 min prior to recording the tissue radioactivity. ATBF_Xe was monitored by collecting continuous 20 s readings from a -counter probe placed over the exact site of injection and taped firmly in place.⁸ Blood flow was calculated from a semilog plot of disappearance of counts vs time in 10 min intervals. ATBF was then calculated according to the equation ATBF=slope of semilog plot´partition coefficient´100.^7,8

Blood sampling

A cannula was inserted retrogradely into a distal forearm vein and kept patent by a continuous slow infusion of saline (NaCl 9 g/l). The lower part of the forearm was heated to provide arterialised blood samples. Samples were taken at 10 min intervals throughout the study into heparinised syringes. At time zero, 75 g glucose, dissolved in 200 ml water and flavoured with fresh lemon, was ingested. Plasma glucose was measured the same day on samples stored at 4°C using an enzymatic method.¹⁴ Plasma insulin was measured by radioimmunoassay (Pharmacia and Upjohn, Milton Keynes, UK).

Endogenous urea recovery by microdialysis

In six of the above studies, dialysate urea concentrations were measured in addition to ethanol. Dialysate urea concentrations were measured using a kit method (66-UV; Sigma, Poole, UK) which was adapted for use on the IL Monarch centrifugal analyser. The method was linear to 10 mmol/l and used 3 µl sample. The intra-assay coefficient of variation was 3.2% for 30 duplicate samples with a mean urea concentration of 1.15 (0.46-1.49) mmol/l.

The Central Oxford Research Ethics Committee approved the study and all subjects gave informed consent.

Statistics

Changes in concentrations with time were assessed by repeated-measures analysis of variance (RM-ANOVA) using time and treatment as within-subjects factors. Calculations were done with SPSS for Windows Release 7.1 (SPSS Inc., Chicago, IL, USA).

Peak values for ATBF_Xe, were calculated as the mean of three points; the maximum value and the two highest adjacent points between 0 and 100 min. Absolute change in ATBF_Xe were calculated as peak minus mean baseline. Peak and trough values for urea concentrations and ethanol O/I ratios respectively, were taken between 0 and 150 min. Relative changes for urea and ethanol were calculated as peak or trough values/mean baseline.

Results

ATBF

Mean ATBF_Xe increased in response to the oral glucose load. The mean time to peak was 57.5 min (25-85). ATBF_Xe peak values were attained between 35 and 55 min after ingestion of glucose and this coincided with the peak plasma glucose concentration of (9.3 mmol/l at 50 min) and of plasma insulin (39.0 mU/ml at 40 min). There was significant variation in ATBF between subjects (P<0.005) with a range of peak values from 2.4 to 23.0 ml/min per 100 g tissue. In fact some people showed a very limited response, if any, whereas others showed a several-fold increased ATBF from baseline (range of 0-4.1).

Microdialysis

Overall, the mean dialysate weight was 28.2 mg (s.e. 0.5). The mean ranged from 26.2 to 30.1 mg for each person.

Ethanol escape

The mean ethanol O/I ratio at baseline was 0.49 (s.e. 0.05). There was no statistically significant change in ethanol O/I ratio with time as determined by RM ANOVA. However, the change in ethanol O/I ratio and change in ATBF_Xe were correlated (r_s=-0.79, P<0.05), consistent with increased removal of ethanol by increased blood flow. Due to the heterogeneous ATBF response the subjects were divided into two groups, according to the magnitude of the change in ATBF_Xe (Figure 1). This created two groups with distinctly different blood flow responses; one group had a mean peak ATBF_Xe that was two-fold higher than baseline (Figure 1, panel A), whereas the four subjects with the lowest change in ATBF_Xe showed no obvious mean response to the ingestion of glucose (Figure 1, panel B). A decrease of ethanol O/I ratio, was, however only observed in the four subjects with the greater response (Figure 1, panel A). There was no subsequent trend towards return to baseline despite the fact that ATBF_Xe rapidly declined after attaining its peak. This suggests that the decrease in ethanol ratio was delayed in comparison with the increase in ATBF_Xe, and that it was either not sufficiently sensitive, or sufficiently rapid, to respond to the subsequent decrease in ATBF_Xe.

Urea

There was a two-fold increase in ATBF_Xe in the six subjects studied (Figure 2), and the dialysate urea concentration tended to increase from 30 to 80 min after the mean peak ATBF_Xe. However, a reversal to baseline was not observed with decreasing ATBF_Xe. The change in urea concentration was correlated with the change in ethanol O/I ratio (r_s=-0.886, P<0.05), but not with change in ATBF_Xe. These results are consistent with the hypothesis; an increased blood flow would lead to increased delivery of urea to the microdialysis area. The result is also consistent with the ethanol escape experiment in the sense that there was no return to baseline indicating the same kind of delayed response or insufficient sensitivity as with ethanol microdialysis.

Five additional microdialysis studies were performed before and after the ingestion of high carbohydrate test meals in order to assess the performance of alternative small molecules. Sorbitol and sucrose were selected because they are not likely to be taken up intracellularly and are therefore more likely to reflect the distribution volume of the extracellular/interstitial space. However, these experiments did not show anything different from the studies using ethanol microdialysis (data not shown).

Discussion

We compared the performance of two methods to estimate physiological changes in ATBF; the ¹³³Xe washout technique and microdialysis of small molecules such as ethanol and urea. The ethanol O/I ratio showed a delayed response (approximately 40 min) to the rapid increase in ATBF_Xe elicited by the ingestion of glucose. In addition, late in the experiment, when ATBF_Xe had shown a sharp decline, the ethanol O/I ratio showed no tendency to return to baseline.

Urea microdialysis recovery is based on the principle that microdialysis creates a condition of local depletion of endogenous urea, with the assumption that urea is not produced in the tissue¹⁵ or, if produced,¹⁶ it is produced at a constant rate. We hypothesised that this local loss might be more rapidly replenished by increased blood flow. To the best of our knowledge quantification of endogenous urea by microdialysis in adipose tissue has not been employed for this purpose before, but the same principle has been advocated for endogenous glucose in estimating muscle blood flow by microdialysis.¹⁷ The urea results paralleled those of ethanol, which suggest that recovery of small molecules by microdialysis is generally delayed compared with changes in blood flow. The reason for the delayed and different response of microdialysis compared with the xenon washout technique in quantifying changes in ATBF should be considered. One obvious difference is the magnitude or strength of the stimulus needed comparing physiological and pharmacological regulation of ATBF. Enocksson and colleagues reported the effect of microdialysis delivery of a beta-2-adrenoreceptor agonist (terbutaline) and simultaneous quantification of ethanol escape.¹⁸ They used two doses, 10^-6 and 10^-5 M of terbutaline, and the ethanol escape was increased at the higher concentration, whereas a lipolytic response was observed at both concentrations. A similar experimental approach with a corresponding effect has been reported by Millet and colleagues¹⁹ at concentrations ranging from 10^-7 to 10^-5 M. These concentrations can be compared with those achieved by systemic delivery of epinephrine and simultaneous quantification of ATBF by xenon washout. Samra et al¹ reported adipose tissue concentrations of 10^-9 or 10^-10 M after a steady-state infusion of epinephrine of 25 ng/kg/min. This infusion provoked a strong lipolytic response and an increase in blood flow from 3 to 17 ml/100 g/min, which is close to a maximal physiological response. Although terbutaline and epinephrine are not absolutely equipotent as beta-2-agonists, there is not likely to be a 100-10000-fold difference in activity. Accordingly, when beta-2-adrenergic agonists have been delivered to adipose tissue by microdialysis it has been done at pharmacological levels to enable detection of a change in blood flow by the microdialysis ethanol escape technique.

The time taken to reach equilibrium has been shown to be longer with low flow rates in muscle,²⁰ and this is likely to apply to adipose tissue as well. This may provide an additional explanation for the lack of coherence between and microdialysis based methods and the ¹³³Xe washout method for the quantification of ATBF.

The microdialysis system was perfused overnight and a 30 min equilibration period was allowed before sampling; this period was chosen in accordance with the literature.^18,19 However, a steady-state ethanol outflow:inflow ratio was not reached until 30 min (Figure 1B) into the experiment. We do not believe that this affects the interpretation of our results, since the subgroup who responded to the stimulus with an increase in ATBF_Xe showed the expected decline in outflow:inflow ratio. We have performed additional experiments with a 60 min equilibration period which show an absolutely stable baseline with ethanol outflow:inflow ratios of 0.566±0.051, 0.569±0.058 and 0.567±0.052 at time points 0, 10 and 20 min. Future studies should allow a 60 min equilibration period to allow for the initial trauma due to insertion of the probe.

Our findings correspond well with those from another tissue bed, skeletal muscle, in which it has recently been concluded that rapid changes in blood flow are poorly reflected by changes in microdialysis ethanol escape.^21,22

Although the ethanol escape method has been shown to be comparable to the ¹³³Xe washout technique after external warming,⁹ we conclude that the ¹³³Xe washout technique seems to be more discriminative to monitor physiological changes in ATBF. However, the microdialysis ethanol escape method may still be a useful tool to detect changes in response to local, pharmacological stimuli.

Acknowledgements

This work was funded by the Wellcome Trust.

1 Samra JS, Simpson EJ, Clark ML, Forster CD, Humphreys SM, Macdonald IA, Frayn KN. Effects of epinephrine infusion on adipose tissue: interactions between blood flow and lipid metabolism. Am J Physiol 1996; 271: E834-839. MEDLINE

2 Bülow J, Madsen J. Influence of blood flow on fatty acid mobilization from lipolytically active adipose tissue. Pflügers Archiv 1981; 390: 169-174. MEDLINE

3 Bu_low J. Human adipose tissue blood flow during prolonged exercise, III. Effect of beta-adrenergic blockade, nicotinic acid and glucose infusion. Scand J Clin Lab Invest 1981; 41: 415-424. MEDLINE

4 Bülow J, Madsen J, Astrup A, Christensen NJ. Vasoconstrictor effect of high FFA/albumin ratios in adipose tissue in vivo. Acta Physiol Scand 1985; 125: 661-667. MEDLINE

5 Ailhaud G. Adipose tissue as an endocrine organ. Int J Obes Relat Metab Disord 2000; 24: (Suppl 2) S1-S3. MEDLINE

6 Summers LKM, Samra JS, Frayn KN. Impaired postprandial tissue regulation of blood flow in insulin resistance: a determinant of cardiovascular risk? Atherosclerosis 1999; 147: 11-15. Article MEDLINE

7 Larsen OA, Lassen NA, Quaade F. Blood flow through human adipose tissue determined with radioactive xenon. Acta Physiol Scand 1966; 66: 337-345. MEDLINE

8 Samra JS, Frayn KN, Giddings JA, Clark ML, Macdonald IA. Modification and validation of a commercially available portable detector for measurement of adipose tissue blood flow. Clin Physiol 1995; 15: 241-248. MEDLINE

9 Felländer G, Linde B, Bolinder J. Evaluation of the microdialysis ethanol technique for monitoring of subcutaneous adipose tissue blood flow in humans. Int J Obes Relat Metab Disord 1996; 20: 220-226. MEDLINE

10 Lafontan M, Arner P. Application of in situ microdialysis to measure metabolic and vascular responses in adipose tissue. Trends Pharmac Sci 1996; 17: 309-313.

11 Kerckhoffs D, Arner P, Bolinder J. Lipolysis and lactate production in human skeletal muscle and adipose tissue following glucose ingestion. Clin Sci 1998; 94: 71-77. MEDLINE

12 Rosdahl H, Lind L, Millgard J, Lithell H, Ungerstedt U, Henriksson J. Effect of physiological hyperinsulinemia on blood flow and interstitial glucose concentration in human skeletal muscle and adipose tissue studied by microdialysis. Diabetes 1998; 47: 1296-1301. MEDLINE

13 Karpe F, Fielding BA, Ilic V, Humphreys SM, Frayn KN. Xenon washout and microdialysis 'ethanol escape' techniques give discrepant results when monitoring changes in adipose tissue blood flow after oral glucose. Proc Nutr Soc 2001; 60: 106A.

14 Clark ML, Humphreys SM, Frayn KN. Stability of plasma glucose during storage. Ann Clin Biochem 1990; 27: 373-377. MEDLINE

15 Felig P. Amino acid metabolism in man. A Rev Biochem 1975; 44: 933-955.

16 Strindberg L, Lonnroth P. Validation of an endogenous reference technique for the calibration of microdialysis catheters. Scand J Clin Lab Invest 2000; 60: 205-212. MEDLINE

17 Hickner RC, Rosdahl H, Borg I, Ungerstedt U, Jorfeldt L, Henriksson J. Ethanol may be used with the microdialysis technique to monitor blood flow changes in skeletal muscle: dialysate glucose concentration is blood-flow-dependent. Acta Physiol Scand 1991; 143: 355-356. MEDLINE

18 Enocksson S, Shimizu M, Lönnqvist F, Nordenström J, Arner P. Demonstration of an in vivo functional beta 3-adrenoceptor in man. J Clin Invest 1995; 95: 2239-2245. MEDLINE

19 Millet L, Barbe P, Lafontan M, Berlan M, Galitzky J. Catecholamine effects on lipolysis and blood flow in human abdominal and femoral adipose tissue. J Appl Physiol 1998; 85: 181-188. MEDLINE

20 Wallgren F, Amberg G, Hickner RC, Ekelund U, Jorfeldt L, Henriksson J. A mathematical model for measuring blood flow in skeletal muscle with the microdialysis ethanol technique. J Appl Physiol 1995; 79: 648-659. MEDLINE

21 Stallknecht B, Donsmark M, Enevoldsen LH, Fluckey JD, Galbo H. Estimation of rat muscle blood flow by microdialysis probes perfused with ethanol, [¹⁴C] ethanol, and ³H₂O. J Appl Physiol 1999; 86: 1054-1061. MEDLINE

22 Rådegran G, Pilegaard H, Nielsen JJ, Bangsbo J. Microdialysis ethanol removal reflects probe recovery rather than local blood flow in skeletal muscle. J Appl Physiol 1998; 85: 751-757. MEDLINE

Figure 1 ATBF_Xe and ethanol microdialysis O/I ratio measured in eight subjects before and after the ingestion of glucose. Subjects were divided into two groups according to change in ATBF_Xe (see text for details) and mean values are shown without error bars for clarity. (A) High responders (n=4); (B) low responders (n=4). Values shown are mean and s.e.m.

Figure 2 ATBF_Xe and microdialysis urea recovery in six subjects before and after the ingestion of glucose. Values shown are mean and s.e.m.