The splanchnic region has become a region of growing interest because of its possible key function in the development of systemic inflammatory response syndrome and sepsis . The rational for the ‘gut hypothesis’ are experimental data, which suggest, that impaired mucosal perfusion and oxygenation might contribute to bacterial translocation and endotoxaemia . This splanchnic hypoperfusion can be caused by enhanced sympathetic nervous activity as a result of surgical manipulation, stress or pain .
Recently methods to reduce regional or systemic sympathetic activity like β-receptor blockade, epidural anaesthesia and application of α2-agonists have gained importance, because they might be able to reduce perioperative mortality [4,5] and morbidity . However, because of imminent reduction of cardiac output, heart rate (HR) or blood pressure (BP), detailed knowledge of the effects of reduced sympathetic activity on intestinal perfusion and oxygenation is still lacking. For a reduction of regional sympatholytic activity with epidural anaesthesia, we could show that despite induction of a marked decrease in mean arterial pressure, the intestinal oxygenation was not affected .
For clonidine (2-(2,6-dichlorphenylamino)-2-imidazoline-HCl), which is used in clinical practice to reduce systemic sympathetic activity and as an adjuvant sedating drug in intensive care medicine, no data exist about the effects on intestinal oxygenation during general anaesthesia and laparotomy. It is well known, that clonidine induces a long lasting reduction in mean arterial pressure and cardiac output . Side-effects are disturbances of gastric emptying and intestinal motility . Clonidine is said to induce intestinal mucosal hypoperfusion, necrosis and perforation of the gut [10,11]. The discussed mechanism of pathogenesis is as follows: dysmotility of the gut wall may lead to microvascular ischaemia without vascular occlusion and as a second mechanism mucosal ischaemia may result from redistribution of blood away from the gut and within the gut wall due to reduction of mean arterial pressure and cardiac output.
Therefore, the present study was performed to characterize intestinal perfusion, the oxygen supply-uptake relationship of the gut and mucosal tissue oxygenation during systemic reduction of sympathetic nervous activity induced by intravenously (i.v.) applied clonidine during general anaesthesia in an intact animal model .
After approval of the study by the local Ethics Committee on Animal Research and in concordance with the Helsinki convention for the care and use in animals, 17 female German domestic pigs, were premedicated with intramuscular flunitrazepam 0.2 mg kg−1 body weight (Rohypnol®; Hoffmann-La Roche, Grenzach-Wyhlen, Germany) and ketamine 15 mg kg−1 (Ketanest®; Parke-Davis, Freiburg, Germany) after overnight fasting and receiving water ad libitum. Anaesthesia was induced i.v. via an ear vein with ketamine 1.6-3.3 mg kg−1 and fentanyl 3 μg kg−1 (Fentanyl-Janssen®; Janssen-Cilag, Neuss, Germany). The trachea was intubated after injection of vecuronium 0.3 mg kg−1 (Norcuron®; Organon Teknika, Eppelheim, Germany). Anaesthesia was maintained by continuous infusion of flunitrazepam 0.07-0.1 mg kg−1 h−1, ketamine 7-10 mg kg−1 h−1 and vecuronium 0.5-0.7 mg kg−1 h−1. Volume- controlled mechanical ventilation was provided with a Bennett® ventilator (Puritan-Bennett Corporation, Galway, Ireland).
Respiratory rate and tidal volume were adjusted to maintain arterial carbon dioxide tension (PaCO2) between 5.1 and 5.6 kPa for the whole experiment. Inspired oxygen fraction was adjusted to maintain arterial oxygen partial pressure (PaO2) of 12.6-15.3 kPa. No positive endexpiratory pressure (PEEP) was applied.
Instrumentation has been previously described . In brief, after induction of anaesthesia, the animals were placed in supine position on a heating pad to keep body temperature constant. A double-lumen catheter (7F Two-Lumen Central Catheterization Set; Arrow, Reading, PA, USA) and an 8.5F introducer (Arrow Percutaneous Sheath Introducer Set; Arrow) were inserted into the right internal jugular vein. Both catheters were advanced 11-13 cm to guarantee correct position of the tip in the superior vena cava. A Swan-Ganz thermodilution catheter (model 93A-131-7F, 7F Swan-Ganz Thermodilutions Catheter; American Edwards Laboratories, Irvine, CA, USA), was introduced into the pulmonary artery. Body temperature was monitored continuously with a thermistor in this flow-directed catheter.
The right femoral artery was cannulated with a 4.5F introducer set (Arrow Percutaneous Sheath Introducer Set). After median laparotomy, the superior mesenteric vein was cannulated with a 16-G single-lumen catheter. A Trip® gastric tube (Baxter, Unterschleissheim, Germany) was placed intraluminally into the jejunum. An ultrasonic perivascular transit-time flow probe (Transonic Systems Inc., Ithaca, NY, USA) of appropriate seize was placed around the superior mesenteric artery. The vessel had to fill 75-100% of the probe's acoustic window. An improved signal quality was provided with perivascular ultrasonic gel. Care was taken to preserve the periarterial nerve plexus. The flow probe consists of two ultrasonic transducers and a fixed acoustic reflector, and measures the ultrasound transit-time. Hence, it does not produce heat and the measurement is independent from the corpuscular content of the blood. The presence of the flow probe has no effect on the flow.
Intermittently a multiwire surface electrode to measure tissue surface PO2 was placed on the intestinal serosa and mucosa of a jejunal segment.
For this purpose, a 1 cm transmural antimesenteric incision was made and a spacer was introduced to obtain uncompromised access to the mucosa. After each measurement the incision in the gut was sutured.
At the end of the surgical preparation the abdomen was sutured except for a cleft of 10 cm to allow intermittent measurements of surface oxygen partial pressure (PO2) on mucosa and serosa tissue of the small intestine. The gap was covered with wet, warm swabs.
For maintenance of normovolaemia, all animals received i.v. full-electrolyte solution 12 mL kg−1 h−1 (Jonosteril®; Fresenius-Klinik, Bad Homburg, Germany) before and 15-20 mL kg−1 h−1 during laparotomy to maintain BP, central venous pressure (CVP) and haematocrit values at the level measured shortly after insertion of the femoral artery and central venous catheters.
Before laparotomy, and during the experimental procedure arterial blood gas values were taken to adjust inspired oxygen concentration and ventilation to maintain predetermined PaO2 and PaCO2 values, and to measure initial haematocrit values. Blood gases, and serum electrolytes were measured using an ABL615 Autoanalyzer (Radiometer, Copenhagen, Denmark).
All intravascular catheters were connected to pressure transducers. A multichannel recorder (Hugo-Sachs, March, Germany) and PO-NE-MAH® (Plugsys®; Digital Acquisition Analysis and Archive Systems, Simsbury, USA) was used for online recording. Cardiac output was determined by thermodilution technique (Baxter CO-computer, Unterschleissheim, Germany). The mean value of three injections of 10 mL ice-cold saline was considered to estimate actual cardiac output if the measurements were within a range of ±10% of the calculated mean. Systemic and regional vascular resistances were calculated from the equations listed in appendix. HR was derived from the spike-interval of the continuous arterial pressure measurement.
Systemic and intestinal oxygen supply (i.e. oxygen supply via the superior mesenteric artery to the gut), oxygen uptake and lactate production were calculated as shown in appendix. Tissue surface PO2 was measured using a polarographical eight-channel multiwire platinum surface electrode (Eschweiler, Kiel, Germany) as described previously [12,13], which was placed on the serosa or mucosa of the gut. This electrode is a further development of Clark's electrode with a weight of 1.2 g and is able to determine the tissue PO2 by measuring electricity which is generated by reduction of oxygen. The low weight of the electrode reduces the risk of capillary compression and hence measuring lower PO2 values. To exclude interference of the measurement with air, the tissue surface and the electrode were ‘sealed’ with a thin rubber of 1 cm2 and a film of water. During each measurement a total of approximately 200 individual PO2 values were registered at 10 different electrode locations. Data were processed by a microprocessor-supported system (Ingenieurbüro für Mess- und Datentechnik, Dipl. Ing. K. Mussler, Aachen, Germany) and downloaded to a personal computer as an ASCII file. Mean values of these data reflect tissue oxygenation, which is the net result of nutritive blood flow and tissue oxygen consumption . Data are presented in Table 2. Mucosal carbon dioxide pressure (PCO2) was measured intermittently by air tonometry (Trip® catheter) as described previously . The TRIP® catheter was connected early during laparotomy to a calibrated tonometry monitor (Tonocap TC-200; Datex, Helsinki, Finland).
To demonstrate the sympatholytic effect of systemic applied clonidine, we measured the arterial norepinephrine concentrations. Analysis followed standard procedures, which were described in detail earlier . The accuracy of the determination of the plasma concentrations of norepinephrine was controlled regularly. The intra assay coefficient of variation was <5%, whereas the between assay coefficient of variation was <10%.
At the end of surgical preparation at least 90 min was allowed for stabilization before baseline readings were obtained in all animals (t 0). Subsequently, the animals were randomly divided into two groups:
Eight animals received 2 μg kg−1 body weight clonidine as a bolus i.v. (administered over 2 min) and a continuous i.v. infusion of 2 μg kg−1 h−1, which started 45 min after bolus application. Nine animals served as control group. No further intervention was applied. During the whole procedure in all animals i.v. fluid administration was continued in amounts, which were necessary to keep filling pressures constant during the stabilization period.
Measurements were made 90 min (t 1) and 250 min (t 2) after clonidine bolus application or the corresponding time in the control group. Tissue serosal and mucosal surface PO2 was only measured at baseline (t 0) and at the end of the experiment (t 2). At the end of the experiments all animals were killed in deep anaesthesia with a potassium chloride overdose, according to German laws for animal studies.
Statistical analysis was performed with the JMP® software package (SAS, Cary, NC, USA). Medians are given throughout with interquartile range (25-75th percentiles) [15,16]. Differences between experimental periods within the two groups were analysed using Friedman's statistic followed by the Wilcoxon signed rank sum test. Differences between the two groups were analysed with the U-test. The level of significance was set at P ≤ 0.05. All results were indexed to compensate for differences in body weight.
Biometric data of the 17 pigs (3-4-months old) used were comparable. The median weight (25th/75th percentile) in the clonidine group was 32.0 kg (30.5/36.5) and 31.0 kg (29.8/33.8) in the control group. The body temperature of the animals (median, 25th/75th percentile) was stable during the experiments: clonidine group: t 0: 36.9°C (35.8; 37.8), t 1: 37.6°C (36.5; 38.6), t 2: 38.0°C (37.0; 38.8); control group: t 0: 36.4°C (35.7; 36.9), t 1: 37.0°C (35.8; 37.8), t 2: 37.7°C (36.6; 39.1).
Systemic haemodynamics and oxygen transport
The systemic application of clonidine induced a reduction of mean arterial pressure by 10%, of HR by 20% and of cardiac output by 30%. Systemic vascular resistance did not change. Central venous and pulmonary capillary wedge pressure remained unchanged (Table 1).
Arterial haemoglobin concentration did not change, systemic oxygen delivery decreased by 30% compared to baseline values, systemic oxygen consumption was maintained.
In the control group, mean arterial pressure, HR, cardiac output, CVP and pulmonary capillary wedge pressure did not change. Systemic haemodynamics and oxygen delivery were not affected.
The superior mesenteric arterial blood flow, the intestinal oxygen delivery and the intestinal oxygen uptake were not affected by the reduction of mean arterial pressure and/or cardiac output. In the control group there was no change in superior mesenteric arterial blood flow, small intestinal oxygen delivery or small intestinal oxygen uptake either (Table 2).
Intestinal tissue oxygenation
Tonometric PCO2 values were not influenced by clonidine (Table 2). The serosal and the mucosal tissue surface PO2 were in a normal range (serosa: 60 mmHg, mucosa: 30 mmHg) throughout the whole experiment. The application of clonidine did not affect these parameters.
Plasma norepinephrine concentrations
The plasma concentrations of arterial norepinephrine are presented in Table 3. Already at baseline the values were very low. They did not change in the control group, while they decreased in each animal of the clonidine group.
The principal findings of this study are as follows:
- The systemic reduction of sympathetic nervous activity by application of clonidine induces a reduction of cardiac output, mean arterial pressure and systemic oxygen delivery.
- The systemic reduction of sympathetic nervous activity by i.v. application of clonidine does not affect the intestinal macroperfusion and oxygenation of the mucosa and serosa.
Clonidine is an α-adrenoreceptor agonist with a 200-fold higher affinity to α2-receptors than to α1-receptors. It reduces sympathetic activity and induces parasympathetic stimulation . The main point of action is localized in the central nervous system, peripheral effects are side-effects. The decrease of sympathetic activity has been attributed to an inhibition of norepinephrine release by binding to α2-receptors for a long time . However, recently it became clear, that clonidine acts on central and peripheral imidazoline receptors as well and induces reduction of arterial pressure . The mechanism of peripheral vasodilatation is the same as with presynaptic α2-receptors . Additionally stimulation of imidazoline receptors in the adrenal glands reduced catecholamine release and thereby affect the BP . However, the effects of clonidine on α2-receptors and on imidazoline receptors are considered equally .
Besides its action on the sympathetic nervous system clonidine modulates pain and has sedative properties. Therefore in anaesthesia clonidine is used as a supportive drug and in intensive care medicine it is used for prophylaxis or treatment of delirium tremens, for example in alcohol withdrawal  with infusion rates of up to 180 μg h−1. However hypotension, bradycardia and reduction of cardiac output as a side-effect are observed quite often.
As low cardiac output and hypotension are main risk factors for splanchnic ischaemia, clonidine might induce necrosis, rupture and pseudo-obstruction of the gut due to hypoperfusion and reduced gut motility [10,11]. Thollander and colleagues  and De Pontin and colleagues  describe inconsistent effects of clonidine on mucosal blood flow, and Holzer and Holzer  describe a decreased hyperaemic defence mechanism of the gastric mucosa in rats.
The clonidine effect of an inhibited motility on the intestinal microcirculation must be addressed in this context. Herbert and colleagues  could prove in an excellent experimental in vitro set-up that clonidine elicits a profound inhibition of gastrointestinal motility which might contribute to pseudo-obstruction of the gut, intestinal necrosis and perforation. This effect on intestinal peristalsis is independent from effects on blood vessels and consecutively on blood flow. But this does not exclude possible detrimental effects of clonidine on intestinal macro- and microcirculation and oxygenation. However, to our knowledge, there has not yet been a study to investigate the effects of clonidine on intestinal oxygenation under physiological conditions.
The systemic haemodynamic effects in our study are in agreement with known effects. However, despite a reduction of mean arterial pressure, cardiac output and consecutively systemic oxygen delivery, the blood flow of the superior mesenteric artery and the intestinal oxygen delivery were not affected. This supports the hypothesis, that clonidine vasodilates the mesenteric artery and reduces regional vascular resistance by reduction of sympathetic nervous activity. This vasodilatation is able to compensate for the reduction of mean arterial pressure and cardiac output. Although clonidine was administered systemically, the local or regional reduction of sympathetic nervous activity by presynaptic effects on norepinephrine outflow seems to contribute more to vasodilatation than central reduction of sympathetic activity. However, the vasodilating effect is only possible because at rest the superior mesenteric artery has an increased sympathetic nervous activity .
We could not detect any effect of clonidine on mucosal and serosal oxygenation and on mucosal PCO2 either, indicating a maintained regional microvascular perfusion in this two layers of the gut wall. However, the maintained mucosal oxygenation might be due to redistribution from muscularis blood flow to mucosal blood flow inside the gut wall . This in turn might explain the development of paralytic ileus under application of clonidine [10,11].
Advantages and disadvantages of the animal model, and the methods for measuring surface PO2 of the gut and PCO2 of the mucosa used in this setting have been discussed extensively earlier [7,24]. We have chosen clonidine for systemic reduction of sympathetic nervous activity, because it is widely used in clinical practice as an analgesic or a sedative.
The measurement of norepinephrine plasma concentration is an accepted method to evaluate effects on the sympathetic nervous system, because there is a strong correlation between sympathetic activity and norepinephrine concentration in the plasma . The norepinephrine circulating in the plasma, is about 5-10% of the amount, which was synthesized in the sympathetic nervous system . In our study, the plasma norepinephrine concentration declined after application of clonidine, while it was constant in the control group. This proves a reduction of sympathetic activity in the clonidine group and is in agreement with previous studies . However a limitation of the study might be, that due to a low sympathetic nervous activity at baseline, indicated by very low arterial plasma norepinephrine concentrations, an additional effect of clonidine on sympathetic activity might be questioned. But from our point of view, this objection can be discarded, because clonidine did affect HR and cardiac output.
Another limitation of our study might be the small number of animals in each group. With tribute to a balance of statistical necessities and ethical issues a sample size of 7-10 animals per group is widely accepted for animal experiments. The statistical methods suggested for are non-parametric test and presentation of medians and interquartile ranges [15,16].
In summary we found that systemic application of clonidine under physiological conditions did not affect intestinal macroperfusion and mucosal and serosal oxygenation in pigs with i.v. baseline anaesthesia and laparotomy.
1. Pastores SM, Katz DP, Kvetan V. Splanchnic ischemia and gut mucosal injury in sepsis and the multiple organ dysfunction syndrome. Am J Gastroenterol
2. Meakins JL, Marshall JC. The gut as the motor of multiple system organ failure. In: Marston A, Bulkley GB, Fiddian-Green RG et al
., eds. Splanchnic Ischemia and Multiple Organ Failure.
London, UK: Edwart Arnold, 1989: 339-348.
3. Gelman SI. Disturbances in hepatic blood flow during anaesthesia and surgery. Arch Surg
4. Mangano DT, Layung EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. New Engl J Med
5. Wallace AW, Galindez D, Salahieh A et al
. Effect of clonidine
on cardiovascular morbidity and mortality after noncardiac surgery. Anesthesiology
6. Nishina K, Mikawa K, Uesugi T et al
. Efficacy of clonidine
for prevention of perioperative myocaedial ischemia. Anesthesiology
7. Vagts DA, Iber T, Szabo B et al
. Effects of epidural anaesthesia on intestinal oxygenation
. Br J Anaesthesiol
8. Kobinger W, Walland A. Investigations into the mechanism of hypotensive effect of 2-(2,6-dichlorphenylamino)-2-imidazoline-HCl. Eur J Pharmacol
9. Herbert MK, Roth-Goldbrunner S, Holzer P, Roewer N. Clonidine
and dexmedetomidine potently inhibit peristalsis in the guinea pig ileum in vitro
10. Stieger D, Cantieni R, Frutiger A. Acute colonic pseudoobstruction (Ogilvie's syndrome) in two patients receiving high dose clonidine
for delirium tremens. Intens Care Med
11. Frey C, Takala J, Krähenbühl Ö. Non-occlusive small bowel necrosis during gastric tube feeding: a case report. Intens Care Med
12. Kessler M, Hoeper J, Krumme BA. Monitoring of tissue perfusion and cellular function. Anesthesiology
13. Spiegel HU, Hauss J, Bünte H. Theory and methods of local tissue pO2
-monitoring in experimental angiology. Angiology
14. Nylander E, Lund N, Wranne B. Effect of increased blood oxygen affinity on skeletal muscle surface oxygen pressure fields. J Appl Physiol
15. Bulpitt CJ. Confidence intervals. Lancet
16. Altman DG, Gardener MJ. Confidence intervals for research findings. Br J Obstet Gynaec
17. Houston MC. Clonidine
hydrochloride. South Med J
18. Ernsberger P, Giuliano R, Willette RN, Reis DJ. Role of imidazole receptors in the vasopressor response to clonidine
analogs in the rostral ventrolaterla medulla. J Pharmacol Exp Ther
19. Göthert M, Molderings GJ. Involvement of presynaptic imidazoline receptors in the α1-adrenoceptor-independent inhibition of noradrenaline release by imidazoline derivatives. Naunyn-Schmiedeberg's Arch Pharmacol
20. Molderings GJ, Ruppert K, Bönisch H, Göthert M. No relationship of I1
- and I2
-imidazoline binding sites to inhibitory effects of imidazolines on ligand-gated ion channels: an investigation in the adrenal medulla and in neuroblastoma cells. Ann NY Acad Sci
21. Ip Yam PC, Forbes A, Kox WJ. Clonidine
in the treatment of alcohol withdrawal in the intensive care unit. Br J Anaesthesiol
22. Thollander M, Gazelius B, Hellström PM. Adrenergic modulation of small bowel hemodynamics in interdigestive motility state of man. Eur J Gastroenterol Hepatol
23. Ponti F, Giaroni C, Casentino M, Lecchini S, Frigo G. Adrenergic mechanisms in the control of gastrointestinal motility: from basic science to clinical application. Pharmacol Ther
24. Holzer P, Holzer PE. Differential effects of clonidine
, dopamine dobutamine and dopexamine on basal and acid-stimulated mucosal blood flow in the rat stomach. Crit Care Med
25. Thomaides TN, Chaudhuri KR, Maule S, Mathias CJ. Differential responses in superior mesenteric artery blood flow may explain the variant pressor responses to clonidine
in two groups with sympathetic denervation. Clin Sci
26. Hiltebrand LB, Krejci V, ten Hoevel ME, Banic A, Sigurdsson GH. Redistribution of microcirculatory blood flow within the intestinal wall during sepsis and general anesthesia. Anesthesiology
27. Vagts DA, Hecker K, Iber T et al
. Effects of xenon anaesthesia on intestinal oxygenation
in acutely instrumented pigs. Br J Anaesth
28. Grassi G, Esler M. How to assess sympathetic activity in humans. J Hypertens
29. Esler M, Jennings G, Lambert G, Meredith I, Hjorne M, Eisenhofer G. Overflow of catecholamines neurotransmitters to the circulation: source, fate and functions. Physiol Rev
Equation used for calculation of vascular resistance:
MAP: mean arterial pressure; CVP: central venous pressure; PVP: portal venous pressure; CO: cardiac output; SMABF: superior mesenteric artery blood flow; bw: body weight.
Equations used for calculation of oxygen supply/uptake:
DO2SMA: superior mesenteric arterial oxygen delivery; CO2A: systemic arterial oxygen contents; CO2SMV: superior mesenteric venous oxygen contents; VO2SI: total small intestinal oxygen uptake.
1Presented in part at the 16th Annual Congress ESICM, Amsterdam, NL, 5-8 October 2003.