Splanchnic circulation and regional sympathetic outflow during peroperative PEEP ventilation in humans

British Journal of Anaesthesia 82 (6): 838–42 (1999)

Splanchnic circulation and regional sympathetic outflow during peroperative PEEP ventilation in humans A. Åneman1*, G. Eisenhofer5, L. Fa¨ndriks4, L. Olbe2, J. Dalenba¨ck2, P. Nitescu1 and P. Friberg3 1Department

of Anaesthesiology and Intensive Care, 2Department of Surgery, 3Department of Physiology for Gastroenterological Research, Sahlgren’s University Hospital, S-413 45 Go¨teborg, Sweden. and 5Clinical Neuroscience Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892, USA 4Centre

*Corresponding author The splanchnic organs represent a major target for sympathetic outflow and an important region for haemodynamic effects on cardiovascular homeostasis. We have studied regional haemodynamic and sympathetic changes in the splanchnic bed during standardized circulatory stress from positive end-expiratory pressure ventilation (PEEP). We investigated eight patients undergoing major upper abdominal surgery using a radiotracer method to measure plasma spillover of norepinephrine as an index of sympathetic nerve activity using arterial, portal and hepatic venous blood sampling. Mesenteric and hepatic perfusion were measured by ultrasound transit time flowmetry and blood-gas analyses. Steady state measurements were performed before and during PEEP ventilation at 10 cm H2O. Plasma spillover of norepinephrine in the mesenteric and hepatic organs represented mean 49 (SEM 8)% and 7 (2)%, respectively, of systemic norepinephrine spillover at baseline, and PEEP ventilation did not cause any significant changes. However, PEEP ventilation significantly decreased portal venous blood flow while hepatic blood flow was preserved by a compensatory increase in hepatic arterial blood flow. Mesenteric and hepatic oxygen delivery changed according to blood flow, and there were no changes in regional oxygen consumption. Thus PEEP ventilation altered mesenteric and hepatic perfusion, independent of any change in corresponding sympathetic nerve activity. Regulation of hepatic blood supply, not related to sympathetic activity, maintained liver oxygenation during PEEP ventilation despite a simultaneous decrease in mesenteric perfusion. Br J Anaesth 1999; 82: 838–42 Keywords: ventilation, positive end-expiratory pressure; sympathetic nervous system, norepinephrine; cardiovascular system, effects; liver, blood flow Accepted for publication: November 19, 1998

The splanchnic organs receive approximately 25% of resting cardiac output and hold approximately 30% of the total blood volume.1 Recent studies have demonstrated that the splanchnic organs in humans receive close to 50% of total sympathetic outflow, as measured by plasma spillover of norepinephrine,2 indicating the vasoregulatory importance of the sympathetic system. General anaesthesia is associated commonly with depression of sympathetic activity which may predispose to cardiovascular instability. This study was designed to investigate, in the mesenteric organs and liver, the sympathetic and haemodynamic responses to a standardized haemodynamic challenge in patients undergoing major upper abdominal surgery. Sympathetic nerve activity was studied using a kinetic radiotracer method to measure plasma spillover of norepinephrine.2–4 Positive end-expiratory pressure (PEEP) was

used as circulatory stress, associated with sympathoexcitation,5 decreased cardiac output6 and reduction of hepatomesenteric blood flow and oxygenation.7–9

Patients and methods We studied eight patients (four females; mean age 61.4 (range 47–77) yr) after obtaining informed, written consent and approval from the Human Ethics Committee of Go¨teborg University. Data on this method in similar subjects have been reported previously.2 Preoperative diagnoses were non-metastatic, medium to well differentiated gastric adenocarcinoma (n57) and highly differentiated pancreatic neoplasm (n51). Routine blood chemistry variables, including liver enzymes, were normal in all patients except one, who presented with increased alanine aminotransferase concentrations of unknown aetiology.

© British Journal of Anaesthesia

Splanchnic organs and PEEP ventilation

All patients were given omeprazole 40–80 mg daily. One patient was receiving levothyroxin for hypothyroidism and atenolol after a myocardial infarction more than 24 months before the study, and another patient was receiving metoprolol and spironolactone for hypertension. All medication was withdrawn and patients were fasted from midnight on the day before surgery. Patients were premedicated with lorazepam 1 mg and anaesthesia was induced with thiopental 3–5 mg kg–1 , fentanyl 2–3 µg kg–1 and vecuronium 0.1 mg kg–1. After tracheal intubation, the lungs were ventilated mechanically (Servo 900C, Siemens-Engstro¨m, Stockholm, Sweden) with enflurane (0.5–0.7 MAC) and 70% nitrous oxide in oxygen, using a low-flow, closed circuit. Atropine, catecholamines and β-adrenergic blockers were avoided during the study. Patients were not given a volume load and central venous pressures were within normal limits. Radial arterial, central venous and hepatic venous catheters were placed to sample blood and/or monitor blood pressures. After upper midline laparotomy, the portal vein and hepatic artery distal to the gastroduodenal artery were isolated with special care not to interfere with perivascular nerve fibres. A HT207 dual channel ultrasound transit-time flowmeter (Transonic Systems Inc., Ithaca, NY, USA) was connected to the portal venous (diameter 16–18 mm) and hepatic arterial (diameter 6–8 mm) probes to record blood flows continuously.

endogenous norepinephrine and 3.9% for [3H]norepinephrine. Intra-assay coefficients of variation were 1.9% for norepinephrine and 1.2% for [3H]norepinephrine.

Calculations Systemic, mesenteric and hepatic arterial vascular resistances were calculated based on blood flows and arterial and central venous pressures. Oxygen delivery and consumption were derived using standard equations based on blood flows and blood-gas analyses, including haemoglobin concentrations. All blood flow and oxygenation variables are expressed per kg body weight. Calculations of total body, mesenteric and hepatic norepinephrine spillover have been described in detail elsewhere.2–4 Total body plasma spillover (SOTB) of norepinephrine (NE) was calculated according to the equation3: SOTB 5 I3NEA /[3H]NEA

where I5infusion rate of [3H]NE (dpm min–1), [3H]NEA5 arterial plasma concentration of [3H]NE (dpm ml–1) and NEA5arterial plasma concentration of endogenous norepinephrine (pmol ml–1). Fractional extractions (F) of norepinephrine (i.e. fractions of norepinephrine removed from plasma during passage through hepatomesenteric organs) were estimated using the equation: F5([3H]NEI –[3H]NEO)/[3H]NEI

(levo-[2,5,6–3H]norepi-

Ringlabelled norepinephrine nephrine 40–60 Ci mmol–1; New England Nuclear, Boston, MA, USA) was diluted in 0.9% saline containing ascorbic acid 1 mmol litre–1 and infused continuously into an antecubital vein at 1.0 µCi min–1 (0.4 ml min–1), beginning at least 30 min before collecting the first blood samples. Baseline measurements of arterial pressure (AP), heart rate ˙ PV) (HR), central venous pressure (CVP), portal venous (Q ˙ HA) blood flows were made. Blood and hepatic arterial (Q (20 ml) was simultaneously obtained into ice-chilled plastic syringes from the radial artery, hepatic vein and portal vein. The portal venous sample was obtained using a fine calibre needle. Blood samples were transferred to ice-chilled plastic tubes containing EDTA and reduced glutathione. One millilitre of each blood sample was used for blood-gas analyses (ABL510, Radiometer, Copenhagen, Denmark). PEEP 10 cm H2O was then applied, as measured at the respiratory flowmeter piece (D-Lite and Capnomac, Datex, Helsinki, Finland) connected to the tracheal tube. Haemodynamic recordings and blood sampling were performed at steadystate 20 min later. Blood samples were centrifuged at 14°C and the plasma separated and stored at –80°C until assayed. All samples were assayed within 1 month by the same laboratory. Plasma concentrations of endogenous and tritium-labelled norepinephrine were measured by highpressure liquid chromatography with electrochemical detection. Inter-assay coefficients of variation were 6.5% for

(2)

where I and O5 inflow and outflow plasma concentrations of [3H]NE (dpm ml–1), respectively. Inflow plasma concentration of [3H]NE for the liver was estimated from arterial and portal venous plasma concentrations weighted according to hepatic arterial and portal venous blood flow.2 4 Mesenteric and hepatic spillovers of norepinephrine (SOMES pmol min–1 and SOHEP pmol min–1) were calculated according to the equation3: ˙ 3(1–Hct) (3) regional SO5[(NEO–NEI)1(NEI3F)]3Q [3H]NE

Radiotracer infusion and blood sampling

(1)

[3H]NE

where NEI and NEO5concentrations of norepinephrine in inflow and outflow plasma (pmol ml–1), respectively, ˙ 5regional blood flow (ml min–1), Hct5haematocrit and Q F5fractional extraction of norepinephrine described in equation (2).

Statistical analysis All values are given as mean (SEM) (n58). Norepinephrine kinetics were not normally distributed and were analysed using the Wilcoxon paired signed rank sum test. Haemodynamic and oxygenation variables were analysed using the Student’s t test.10

Results Haemodynamics, oxygen delivery and consumption Systolic and diastolic arterial pressures did not change when PEEP 10 cm H2O was applied (Table 1). In addition, heart

839

Åneman et al.

Fig 1 Oxygen delivery (DO2) in mesenteric organs (top left) and the liver (top right) and oxygen consumption (V˙ O2) in the mesenteric organs (bottom left) and the liver (bottom right). Lines illustrate changes in individual patients from baseline to application of positive end-expiratory pressure (PEEP) 10 cm H2O. Lines in the margins indicate mean values during baseline and PEEP. *Significant difference (P,0.05) in PEEP vs baseline; ns5not significant. Table 1 Haemodynamic variables at baseline and during positive end-expiratory pressure ventilation (PEEP) at 10 cm H2O. SAP5Systolic arterial pressure, DAP5diastolic arterial pressure, CVP5central venous pressure, HR5heart rate, RMES5mesenteric vascular resistance, RHA5hepatic arterial vascular resistance. All data are presented as mean (SEM) (n57)

Baseline PEEP P

SAP (mm Hg)

DAP (mm Hg)

97 (4) 88 (6) ns

54 (2) 52 (3) ns

CVP (mm Hg) 9 (1) 15 (2) ,0.05

rate remained unchanged after application of PEEP. Portal blood flow decreased and mesenteric vascular resistance increased (P,0.05) during PEEP ventilation. In contrast, hepatic arterial blood flow increased and hepatic arterial vascular resistance decreased (P,0.05) after establishing PEEP 10 cm H2O. The proportion of total hepatic blood flow derived from the hepatic artery increased during PEEP from 31 (2)% to 49 (3)% (P,0.05). Haemodynamic changes were reflected in regional oxygen delivery; while mesenteric oxygen delivery decreased (P,0.05) during PEEP, hepatic oxygen delivery remained unchanged (Fig. 1). Neither mesenteric nor hepatic oxygen consumption changed significantly (Fig. 1).

Plasma spillover of norepinephrine Mean arterial, portal and hepatic venous plasma concentrations of endogenous and tritiated norepinephrine are given in Table 2. Total body norepinephrine spillover was 3.37 (0.64) nmol min–1 and did not change during PEEP (3.19 (0.54) nmol min–1) (Fig. 2). Mesenteric norepinephrine spillover was 1.57 (0.39) nmol min–1 at baseline and ventilation with PEEP 10 cm H2O did not cause any significant change (1.51 (0.39) nmol min–1). Hepatic norepi-

HR (beat min–1)

RMES (mm Hg min–1ml–1)

RHA (mm Hg min–1 ml–1)

59 (3) 61 (4) ns

0.091 (0.004) 0.151 (0.011) ,0.05

0.226 (0.035) 0.181 (0.025) ,0.05

nephrine spillover also did not change significantly from 0.21 (0.06) nmol min–1 at baseline to 0.20 (0.07) nmol min–1 during PEEP ventilation. Hence, the regional haemodynamic changes observed during PEEP ventilation occurred without any accompanying changes in regional spillover of norepinephrine.

Discussion We found that the haemodynamic changes observed during ventilation with PEEP 10 cm H2O occurred without corresponding changes in regional norepinephrine spillover, suggesting that alterations in splanchnic haemodynamics at this level of PEEP are independent of sympathetic discharge. The neurohumoral changes underlying the haemodynamic response to PEEP in humans are not understood fully. Reductions in cardiac output and regional blood flow during PEEP are well recognized and may be attributed to decreased venous return and end-diastolic ventricular filling.5 The level of PEEP (10 cm H2O) chosen for our study is in accordance with clinical practice. This mode of PEEP ventilation has been reported to cause sympathoexcitation, shown by increased muscle sympathetic nerve activity in

840

Splanchnic organs and PEEP ventilation Table 2 Endogenous and exogenous, tritiated ([3H]NE) concentrations of norepinephrine (NE) at baseline and during positive end-expiratory pressure ventilation (PEEP) of 10 cm H2O. Subscripts: A5arterial, PV5portal venous, HV5hepatic venous. No significant differences

Baseline PEEP

NEA (pmol ml–1)

NEPV (pmol ml–1)

NEHV (pmol ml–1)

[3H]NEA (dpm ml–1)

[3H]NEPV (dpm ml–1)

[3H]NEHV (dpm ml–1)

2.09 (0.53) 2.24 (0. 74)

4.72 (1.21) 4.42 (1.41)

0.73 (0.12) 0.70 (0.12)

1306 (118) 1401 (185)

802 (95) 888 (99)

120 (32) 148 (40)

Fig 2 Plasma spillover of norepinephrine (NE) in the systemic circulation (top), mesenteric circulation (middle) and hepatic circulation (bottom). Lines illustrate changes in individual patients from baseline to the application of positive end-expiratory pressure (PEEP) 10 cm H2O. Lines in the margins indicate mean values during baseline and PEEP. ns5Not significant.

awake humans,5 and to increase sympathetic nerve activity in rats and dogs by unloading of arterial baro- and volume receptors.11 12 However, these findings are not uniform and studies in humans during general anaesthesia have failed to demonstrate any increase in muscle sympathetic nerve activity during ventilation with PEEP 10 cm H2O.13 Experimental studies in pigs have yielded conflicting results largely dependent on the choice of anaesthesia.14 Portal venous and hepatic arterial blood flows in our

study (approximately 750 ml min–1 and 350 ml min–1 at baseline) are consistent with previous studies in anaesthetized patients.8 15 Although cardiac output was not measured, data were obtained in other patients undergoing gastrectomy during similar anaesthetic conditions at our institution. Descending aortic blood flow (ABF), reflecting cardiac output, was semi-invasively assessed in five patients by transoesophageal Doppler echocardiography (Sometec). Baseline ABF was 4.75 (SEM 0.33) litre min–1 which decreased to 3.65 (0.19) litre min–1 (n55) during PEEP ventilation applied as in our study (Dr Ola Stenqvist, personal communication). Extrapolation of these values would indicate that the splanchnic circulation received approximately 20–25% of cardiac output at baseline, which supports previous observations.1 In our study, PEEP ventilation provided a means of changing cardiac output in a standardized manner. However, the study showed the integrated cardiovascular effects of PEEP which is not well reported in the literature. Winso¨ and colleagues8 reported a reduction in portal blood flow during PEEP 10 cm H2O and a concomitant increase in mesenteric vascular resistance, which might be secondary to increased sympathetic discharge. Lack of increased norepinephrine spillover during PEEP 10 cm H2O in our study, despite a 40% decrease in portal blood flow, suggests that sympathetic nerves are not responsible for the response to this level of PEEP. In addition, systolic and diastolic arterial pressures did not change significantly, making sympathoexcitation by unloading of baroreceptors unlikely. Moreover, heart rate did not increase, suggesting that there was no increase in general sympathetic outflow during PEEP, as shown by unchanged total body norepinephrine spillover and supported by previous peroperative studies.13 Other vasoconstrictors, such as angiotensin II or vasopressin, may explain the increased mesenteric vascular resistance. In an experimental study in pigs, PEEP ventilation during isoflurane anaesthesia markedly increased plasma angiotensin II concentrations while no support for sympathoexcitation was observed.14 Increased angiotensin II concentrations might have contributed to the observed mesenteric vasoconstriction in the present study. It remains possible that higher levels of peroperative PEEP might have resulted in sympathetic activation. However, further increasing PEEP solely for investigational purposes was not compatible with ethical standards. It might be that portal blood flow decreased during PEEP ventilation secondary to an increase in central venous pressure that was transmitted to the liver, thereby increasing

841

Åneman et al.

hepatic venous outflow resistance. However, the significant hepatic venous capacitance makes this explanation less likely.16 The hepatic haemodynamic response to PEEP occurred independent of any change in hepatic sympathetic activity. The counterbalancing changes in portal and hepatic arterial blood flows during PEEP ventilation maintained total hepatic blood flow. This reciprocal relationship of portal and hepatic arterial blood flows, referred to as the hepatic arterial buffer response, has been proposed to be independent of sympathetic activity, possibly involving adenosine.16 The hepatic arterial buffer response has been observed in patients after orthotopic liver transplantation,17 further indicating how this response is independent of sympathetic activity. Mesenteric oxygen delivery was reduced during PEEP by the decrease in portal blood flow, as arterial oxygenation did not change. The concomitant, reciprocal increase in well oxygenated hepatic arterial blood prevented any reduction in hepatic oxygen delivery. These results illustrate the greater susceptibility of mesenteric organs, compared with the liver, to reductions in blood flow which might remain undetected when splanchnic perfusion is assessed as one vascular unit. Both surgery and anaesthesia may have affected our results. Thiopental can depress sympathetic activity18 while ventilation with oxygen in nitrous oxide may increase sympathetic tone.19 Using enflurane results in maintenance of hepatomesenteric blood flow by reductions in vascular resistance,20 partly caused by sympathoinhibition. Fentanyl and lorazepam are not known to interfere with sympathetic transmission21 or splanchnic haemodynamics at the doses used. Malignant disease and the associated cachexia may be related to increased sympathetic activity.22 However, baseline arterial plasma concentrations of endogenous norepinephrine and total body norepinephrine spillover observed in our study were comparable with previously reported values in awake, healthy humans.3 23

4

5

6

7 8

9

10 11

12

13

14

15

16 17

Acknowledgements This study was made possible by the staff at the Department of Surgery, Team 19, and the Department of Anaesthesia and Intensive Care, Sahlgrens’ Hospital. We thank Staff Nurses Lena Henriksson and Annika Henningsson, and Ms AnneLie Ambring for technical assistance. Financial support was provided by the Go¨teborg Medical Society, the Swedish Society of Medicine, the Swedish Society of Medical Sciences, the Swedish Medical Research Council (projects 760, 2855, 8663, 9047, 11133) and the Knut and Alice Wallenberg Foundation.

18

19

20 21

References 1 Donald DE. Splanchnic circulation. In: Handbook of Physiology, section 6, volume 1 part 2. Bethesda, Maryland: American Physiological Society, 1983; 219–40 2 Åneman A, Eisenhofer G, Olbe L, et al. Sympathetic discharge to mesenteric organs and liver in humans. Evidence for substantial mesenteric norepinephrine spillover. J Clin Invest 1996; 97: 1–7 3 Esler M, Jennings G, Lambert G, Meredith I, Horne M, Eisenhofer

22

23

842

G. Overflow of catecholamine neurotransmitters to the circulation: source, fate and functions. Physiol Rev 1990; 70: 963–85 Åneman A, Eisenhofer G, Fa¨ndriks L, Friberg P. Hepatomesenteric release and removal of norepinephrine in swine. Am J Physiol 1995; 268: R924–30 Tanaka S, Sagawa S, Miki K, Claybaugh JR, Shiraki K Changes in muscle sympathetic nerve activity and renal function during positive-pressure breathing in humans. Am J Physiol 1994; 266: R1220–8 Johnston WE, Vinten-Johansen J, Santamore WP, Case DL, Little WC. Mechanism of reduced cardiac output during positive endexpiratory pressure in the dog. Am Rev Respir Dis 1989; 140: 1257–64 Fujita Y. Effects of PEEP on splanchnic haemodynamics and blood volume. Acta Anaesthesiol Scand 1993; 37: 427–31 Winso¨ O, Biber B, Gustavsson B, Holm C, Milsom I, Niemand D. Portal blood flow in man during graded positive end-expiraotry pressure ventilation. Intensive Care Med 1986; 12: 80–5 Berendes E, Lippert G, Loick HM, Brussel T. Effects of positive end-expiratory pressure ventilation on splanchnic oxygenation in humans. J Cardiothorac Vasc Anesth 1996; 10: 598–602 Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall, 1994 Sellde´n H, Sjo¨vall H, Ricksten S-E. Sympathetic nerve activity and central hemodynamics during mechanical ventilation with positive end-expiratory pressure in rats. Acta Physiol Scand 1986; 127: 51–60 Chernow B, Soldano S, Cook D, et al. Positive end-expiratory pressure increases plasma catecholamine levels in non-volume loaded dogs. Anaesth Intensive Care 1986; 14: 421–5 Sellgren J, Ponte´n J, Wallin BG. Characteristics of muscle nerve sympathetic activity during general anaesthesia in humans. Acta Anaesthesiol Scand 1992; 36: 336–45 Åneman A, Ponte´n J, Fa¨ndriks L, Friberg P, Biber B. Haemodynamic, sympathetic and angiotensin II responses to PEEP ventilation before and during isoflurane. Acta Anaesthesiol Scand 1997; 41: 41–8 Biber B, Holm C, Winso¨ O, Gustavsson B. Portal blood flow in man during surgery, measured by a modification of the continuous thermodilution method. Scand J Gastroenterol 1983; 18: 233–9 Lautt WW, Greenway C. Conceptual review of the hepatic vascular bed. Hepatology 1987; 7: 952–63 Hendersson JM, Gilmore GT, Mackay GJ, Galloway JR, Dodson TF, Kutner MH. Hemodynamics during liver transplantation: the interactions between cardiac output and portal venous and hepatic arterial flows. Hepatology 1992; 16: 715–18 Ebert TJ, Kanitz DD, Kampine JP. Inhibition of sympathetic neural outflow during thiopental anesthesia in humans. Anesth Analg 1990; 71: 319–26 Sellgren J, Ponte´n J, Wallin BG. Percutaneous recording of muscle nerve sympathetic activity during propofol, nitrous oxide, and isoflurane anesthesia in humans. Anesthesiology 1990; 73: 20–7 Andreen M, Irestedt L. Effects of enflurane on splanchnic circulation. Acta Anaesthesiol Scand 1979; 71 (Suppl.): 48–51 Marshall BE, Longnecker DE. General anesthetics. In: Gilman AG, Rall TW, Nies AS, Taylor P, eds. The Pharmacological Basis of Therapeutics. New York: Pergamon Press, 1990; 303–6 Hyltander A, Ko¨rner U, Lundholm K. Evaluation of mechanisms behind elevated energy expenditure in cancer patients with solid tumors. Eur J Clin Invest 1993; 23: 46–52 Dousa MK, Tyce GM. Free and conjugated plasma catecholamines, DOPA and 3-O-methyldopa in humans and in various animal species. Proc Soc Exp Biol Med 1988; 188: 427–34