Changes in regulators of the circulation in patients undergoing lung ...

3 downloads 0 Views 173KB Size Report
Ib, whereas they increased in pneumonectomy patients (from 2.63 (0.3) to 6.61 (1.01) pg ml91). Vasopressin plasma concentrations increased during OLV in all ...
British Journal of Anaesthesia 1997; 79: 733–739

Changes in regulators of the circulation in patients undergoing lung surgery

J. BOLDT, M. PAPSDORF, D. UPHUS, M. MÜLLER AND G. HEMPELMANN

Summary Several vasoactive substances are responsible for control of the circulation. During lung surgery these substances may be influenced either by the technique of anaesthesia or by altered production and elimination. We have studied plasma concentrations of important regulators of the circulation in patients undergoing lobectomy using two different regimens: propofol–low-dose fentanyl–nitrous oxide–vecuronium, tracheal extubation immediately after surgery (group la (n:15)); fentanyl–midazolam–pancuronium, delayed tracheal extubation in the intensive care unit (ICU) (group Ib (n:15)). We also studied patients undergoing pneumonectomy using fentanyl– midazolam–pancuronium anaesthesia (group II (n:15)). Plasma concentrations of endothelin-1 (ET-1), atrial natriuretic peptide (ANP), vasopressin, catecholamines (adrenaline, noradrenaline) and 6-keto-prostaglandin F1alpha were measured. Extensive haemodynamic monitoring was performed using a pulmonary artery catheter. All measurements were carried out after induction of anaesthesia (baseline values), 30 min after one-lung ventilation (OLV) was induced, at the end of surgery, 2 h after surgery in the ICU and on the first day after operation. CVP, PAP and PCWP increased in all groups during OLV but remained increased only in patients undergoing pneumonectomy. CI decreased significantly in the pneumonectomy group but did not differ between both lobectomy groups. SVRI was lowest in the propofol-treated patients and remained highest in the pneumonectomy group. Plasma concentrations of adrenaline and noradrenaline did not change in group Ia, but increased significantly in groups Ib (noradrenaline: from 267 (67) to 496 (91) pg ml91) and II (adrenaline: from 237 (59) to 681 (210) pg ml91). Plasma concentrations of ET-1 remained almost unchanged in groups Ia and Ib, whereas they increased in pneumonectomy patients (from 2.63 (0.3) to 6.61 (1.01) pg ml91). Vasopressin plasma concentrations increased during OLV in all groups; they were lowest in the propofol patients and highest in patients undergoing pneumonectomy (32.2 (10.2) pg ml91). Plasma concentrations of ANP and 6-keto-prostaglandin F1alpha increased in all groups during OLV and remained increased only in the pneumonectomy group. (Br. J. Anaesth. 1997; 79: 733–739).

Key words Surgery, thoracic. Cardiovascular system, responses. Sympathetic nervous system, catecholamines. Sympathetic nervous system. Ventilation, one lung.

During major surgery, circulatory stability is of major importance for patient outcome.1 Functional disturbances in the microcirculation must also be avoided because it has become increasingly clear that reduction in microcirculatory blood flow is an important trigger for the development of multiple organ dysfunction syndrome.2 Neural control mechanisms and various circulating vasoactive substances are involved in the control of organ perfusion in this situation. Humoral agents which are known to alter (micro-) circulatory blood flow include adrenaline, noradrenaline and vasopressin. Atrial natriuretic peptide (ANP) has been detected more recently and it participates in the regulation of vasomotor tone either directly by affecting vascular tone or indirectly by influencing hormone homeostasis (e.g. renin release, modifying catecholamine synthesis).3 4 In recent years there has been a large body of literature suggesting that the vascular endothelium is involved in the regulation of regional microcirculatory blood flow.5 Among the substances released from the endothelium to maintain haemostasis and circulatory homeostasis is the 21-amino acid endothelin-1.6 7 It appears to have a greater vasoconstrictor effect than any other substance, including vasopressin and noradrenaline. Its vasoconstrictive potency is approximately 10 times that of angiotensin II.6 In addition to the vascular properties of endothelin-1, interactions with other vasoactive substances may be important (e.g. inhibition of release of catecholamines and renin, stimulation of ANP release).8 The interactions between these (often opposing) vasoactive forces ultimately determine systemic and regional blood flow.9 During thoracic surgery this regulatory system J. BOLDT*, MD, Department of Anaesthesiology and Intensive Care Medicine, Klinikum der Stadt Ludwigshafen, Ludwigshafen, Germany. M. PAPSDORF, D. UPHUS, M. MÜLLER, MD, G. HEMPELMANN, MD, Department of Anaesthesiology and Intensive Care Medicine, Justus-Liebig-University Giessen, Giessen, Germany. Accepted for publication: July 30, 1997. *Address for correspondence: Department of Anaesthesiology and Intensive Care Medicine, Klinikum der Stadt Ludwigshafen, Bremserstr. 79, D-67063 Ludwigshafen, Germany.

734 may be altered by many mechanisms. Surgical stress, the lateral position in association with one-lung ventilation (OLV), clamping of the pulmonary artery and the anaesthetic regimen may result in changes in plasma concentrations of vasoactive substances. It is known that the lung serves not only gas exchange but is also involved in the regulation of circulatory homeostasis: some vasoactive substances are produced by the lung (e.g. angiotensin II, prostaglandins) and others are taken up from the circulation and metabolized (e.g. endothelin, catecholamines).10–12 Thus surgery of the lung may have a profound influence on regulators of the macro- and microperfusion. This study was designed to determine the influence of different anaesthetic management regimens and different types of lung surgery on plasma concentrations of important vasoactive substances.

Patients and methods After obtaining institutional review and written informed consent, we studied 30 patients undergoing lobectomy (group I) and 15 patients undergoing pneumonectomy (group II), both caused by primary lung cancer. Mediastinal lymph nodes were dissected extensively in all patients. Patients with renal insufficiency (creatinine 91.5 mg dl91), liver dysfunction (aspartate aminotransferase (AST) 940 iu litre91, alanine aminotransferase (ALT) 940 iu litre91) or preoperative pulmonary hypertension (mean pulmonary artery arterial pressure (PAP) 930 mm Hg) were excluded. No patient had a history of myocardial infarction in the previous 6 months or had documented valvular heart disease. Patients in group I were separated into one of two groups before operation according to a randomized sequence. In group Ia (n:15), propofol was given for induction of anaesthesia followed by a continuous infusion until the end of operation (range 0.05–0.15 mg kg91 min91). Fentanyl 2.5 ␮g kg91 was given for induction of anaesthesia followed by administration of additional fentanyl if necessary (maximum dose 0.5 mg). Propofol infusion was supplemented with 66% nitrous oxide in oxygen. Neuromuscular block was produced by vecuronium. In these patients the trachea was extubated immediately after surgery and the patient taken to the intensive care unit (ICU), breathing spontaneously. In group Ib (standard management, n:15), induction of anaesthesia was carried out with thiopentone 400–500 mg, fentanyl 0.1 mg and pancuronium 0.1 mg kg91. For additional anaesthesia, fentanyl, midazolam and pancuronium were given according to the patient’s clinical response. An oxygen–air mixture was used. Mechanical ventilation was continued in the ICU in these patients for at least 2 h after surgery. The trachea was extubated when haemodynamic and respiratory variables remained stable for at least 30 min. Group II (n:15) included patients undergoing pneumonectomy and fentanyl–midazolam–pancuronium were used for anaesthesia (same technique as in group Ib). Controlled mechanical ventilation was continued in the ICU until patients breathed

British Journal of Anaesthesia spontaneously and cardiorespiratory variables remained stable. A double-lumen endobronchial tube was used to carry out one-lung ventilation (OLV). Controlled mechanical ventilation was adjusted to maintain P aO2 at 15–20 kPa and PaCO2 at 5.3–5.7 kPa. When necessary, 100% oxygen was used during the period of one-lung ventilation (OLV) (aim SpO2 990%). In groups 1b and II the double-lumen tube was replaced by a standard tube at the end of surgery. Extensive intercostal nerve block was performed directly by the surgeon before closure of the chest (0.5% bupivacaine 1.5 mg kg91). In the ICU, additional intercostal nerve block was used when the patient felt pain and/or piritramide was given i.v. After tracheal extubation, all patients received oxygen 3 litre min91 by nasal tube. Fluid (low molecular weight hydroxyethylstarch solution (LMW-HES) or Ringer’s solution, or both, was given to maintain central venous pressure (CVP) at 10–14 mm Hg throughout the study, packed red blood cells (PRBC) were administered when haemoglobin was :9 g dl91. Respiratory complications were defined as the need for mechanical ventilation for more than 24 h and cardiocirculatory complications were defined as the need for inotropic support (adrenaline, noradrenaline). The therapeutic management of all patients (volume replacement, pharmacological support, ventilation patterns) was performed by physicians who were not involved in the study. MEASURED VARIABLES AND DATA POINTS

Heart rate (HR), mean arterial pressure (MAP), pulmonary artery pressure (PAP), pulmonary capillary wedge pressure (PCWP), central venous pressure (CVP) and cardiac output (CO, thermodilution technique) were monitored using a pulmonary artery catheter. From arterial blood samples, plasma concentrations of endothelin-1 (ET-1) were measured by radioimmunoassay (RIA) (Biocode, Sclessin, Belgium13). Blood samples were collected in tubes containing 5 mg of editic acid (ethylenedinitro tetraacetic acid) and plasma was separated immediately by centrifugation and stored at 935 ⬚C until analysis. As provided by the manufacturer, normal plasma values in healthy volunteers are 2–4 pg ml91, the sensitivity of the RIA was 0.5 pg ml91, and cross reactivity with ET-1 was 100%, with ET-2 34%, with ET-3 64%, with big-ET 2% and with other vasoactive peptides (e.g. angiotensin I and II, bradykinin, ANP) :0.1%. Atrial natriuretic peptide (ANP) (by RIA; Instar Corp., Stillwater, MN, USA14), vasopressin (by RIA; Instar Corp., Stillwater, MN, USA15), 6-keto-prostaglandin F1alpha (the stable but inactive end product of prostaglandin I2 (prostacyclin)), using a commercially available enzyme immunoassay kit (TiterZyme, PerSeptive Diagnostics, Cambridge, MA, USA) and catecholamines (adrenaline and noradrenaline; by a high-pressure liquid chromatographic (HPLC) technique with electrochemical detection) were also analysed from arterial blood samples. Standard laboratory variables were also monitored.

Regulators of the circulation in lung surgery

735

Measurements were performed after induction of anaesthesia (supine position, after haemodynamic stability was reached:baseline values), 30 min after atelectasis was induced (:one-lung ventilation (OLV), lateral position), at the end of surgery (supine position), 2 h after the end of surgery (in the ICU) and at the morning of the first day after operation (supine position, in the ICU). STATISTICAL ANALYSIS

All variables are expressed as mean (SD). One- and two-factorial analyses of variance for repeated measurements (ANOVA, including Scheffe’s test) were used for statistical analysis. Additionally, the Mann–Whitney U test and chi-square test were used when appropriate. P:0.05 was considered significant.

Results Patients undergoing pneumonectomy were at higher risk (higher ASA classification) than those undergoing lobectomy (table 1). Three patients undergoing pneumonectomy died during the intensive care period. In patients in group Ia the trachea was extubated at 31 (10) min, in group Ib, 345 (111) min and in patients undergoing pneumonectomy at 611 (219) min after surgery (table 1). Cumulative fluid balance was higher in propofol-treated patients than in the other two groups. HAEMODYNAMIC STATE

MAP increased significantly during and after OLV in groups Ib and II (table 2). CVP, PAP and PCWP remained increased after OLV only in patients

undergoing pneumonectomy, while they normalized in groups Ia and Ib (table 2). CI decreased significantly in patients undergoing pneumonectomy but did not differ in the two other groups. SVRI was lowest in the propofol-treated group and remained highest in patients undergoing pneumonectomy. Pa O2 /FI O2 ratio was comparable in both lobectomy groups (table 2). Pneumonectomy patients showed a significantly lower Pa O2 /FI O2 ratio than those in the groups undergoing lobectomy. REGULATORS OF THE CIRCULATION

Plasma concentrations of noradrenaline (fig. 1) did not change in propofol-treated patients (group Ia) but increased significantly in the two other groups (group Ib, from 267 (67) to 496 (91) pg ml91 at the end of surgery; group II, from 237 (59) to 681 (210) pg m91). Plasma concentrations of adrenaline (fig. 1) also increased only in patients in groups Ib and II. ET-1 plasma concentrations remained almost unchanged in groups Ia and Ib (fig. 2). It increased significantly in patients in group II (from 2.63 (0.3) to 6.61 (1.01) pg ml91 at the end of the study). In the perioperative period, vasopressin plasma concentrations (fig. 2) increased similarly in all groups. Two hours after surgery vasopressin concentrations were significantly lower in group Ia (increase from 2.23 (0.88) to 14.8 (3.1) pg ml91) and significantly higher in patients undergoing pneumonectomy (increase from 2.58 (0.79) to 32.2 (10.2) pg ml91). On the morning of the first day after operation, plasma concentrations of vasopressin were still significantly highest in the pneumonectomy group (16.9 (3.9) pg ml91). Plasma concentrations of ANP and 6-ketoprostaglandin F1alpha increased significantly in all groups during OLV (fig. 3). ANP concentrations

Table 1 Patient and perioperative data (mean (SD or range) or number) in group Ia (lobectomy patients using propofol), group IIb (lobectomy using fentanyl/midazolam) and group II (pneumonectomy patients). OLV:Onelung ventilation; po:postoperative; L:left side; R:right side; FEV1:forced expiratory volume in 1 s (% of vital capacity); VC:vital capacity; HES:hydroxyethylstarch solution; PRBC:packed red blood cells. *P:0.05 compared with other groups

Age (yr) weight (kg) ASA

Group Ia (n:15)

Group Ib (n:15)

Group II (n:15)

66.5 (55–72) 77.1 (12.2) I:3, II:6; III:4; IV:2

65.5 (52–75) 78.2 (10.2) I:2; II:7; III:4; IV:2

64.1 (54–71) 81.1 (10.7) I:1; II:4; III:6; IV:4*

— L:8/R:7 —

L:7/R:8 — 3

70 (10) 3.61 (0.9) 173 (71) 51 (20) 345 (111)

65 (8) 3.09 (0.6) 172 (66) — 611 (219)*

2.1 (0.3) 20.1 (2.1) — 8.4 (2.7)

2.3 (0.4) 22.9 (3.9) — 10.9 (2.5)

2.650 (550) 350 (250) 4

2.650 (550) 400 (250) 8

Surgery (n) Left/right pneumonectomy — Lobectomy L:9/R:6 Non-survivors — Preoperative lung function 67 (9) FEV1 (%) VC (litre) 3.83 (0.7) Time of surgery (min) 167 (56) OLV (min) 49 (21) Extubated after surgery (min) 31 (10) Total amount of drugs at the end of surgery Fentanyl (mg) 0.39 (0.04)* Midazolam (mg) — Vecuronium (mg) 4.1 (2.2) Pancuronium — Cumulative volume therapy (until 1st postoperative day) Ringer’s solution (ml) 2.350 (500) HES (ml) 850 (250)* Total units PRBC per group 3

736

British Journal of Anaesthesia Table 2 Changes in mean arterial pressure (MAP), heart rate (HR), pulmonary artery pressure (PAP), pulmonary capillary wedge pressure (PCWP), central venous pressure (CVP), cardiac index (CI), systemic vascular resistance index (SVRI) and PaO2 /FIO2 ratio (mean (SD)) in group Ia (lobectomy patients using propofol) IIb (lobectomy using fentanyl–midazolam) and group II (pneumonectomy patients). OLV:one-lung ventilation; po:postoperative. *P:0.05 compared with other groups; †P:0.05compared with baseline values Variable MAP (mm Hg)

Group

Ia Ib II HR Ia 91 Ib (min ) II PAP Ia (mm Hg) Ib II CVP Ia (mm Hg) Ib II PCWP Ia (mm Hg) Ib II CI Ia 91 92 (litre min m ) Ib II SVRI Ia (dyn s cm95 m92) Ib II Ia PaO2 /FIO2 (mm Hg) Ib II

After induction

30 min OLV

76.2 (9.5) 78.9 (7.9) 79.5 (10.1) 77 (11) 78 (12) 79 (12) 19.9 (3.2) 20.1 (2.5) 20.5 (3.0) 11.2 (2.2) 10.5 (3.1) 10.1 (2.7) 11.1 (3.1) 11.8 (2.1) 12.0 (2.0) 2.74 (0.5) 2.81 (0.4) 2.72 (0.5) 1895 (233) 1930 (256) 1978 (239) 384 (61) 377 (75) 341 (68)

80.1 (8.9) 78.9 (10.1) 91.2 (10.4)*† 90.2 (11.2)*† 92.3 (11.7)*† 94.1 (12.2)*† 70 (11) 71 (12) 92 (10)*† 90 (9)*† 94 (13)*† 92 (11)*† 24.4 (4.3)† 20.4 (3.4) 28.9 (3.8)*† 22.2 (3.5) 29.1 (3.4)*† 27.7 (3.5)*† 13.2 (2.2)† 10.9 (2.9) 14.1 (2.6)† 10.8 (2.3) 14.5 (3.0)† 13.9 (3.2)*† 14.7 (3.6)† 12.5 (3.3) 15.2 (4.0)† 12.1 (4.1) 17.0 (4.4)† 18.4 (3.2)*† 2.86 (0.6) 3.31 (0.6) 2.51 (0.5) 3.10 (0.5) 2.43 (0.4) 2.50 (0.5)* 2009 (311) 1537 (288) 2491 (317)*† 2176 (303)* 2449 (307)*† 2541 (317)*† 292 (88)† 410 (87) 268 (79)† 389 (92) 232 (77)† 321 (95)*

returned to baseline values in groups Ia and Ib thereafter, but were still significantly higher in the pneumonectomy group at the end of the study

Figure 1 Changes in plasma concentrations of noradrenaline (normal range 180–250 pg ml91) and adrenaline (normal range 30–85 pg ml91) in group Ia (propofol patients who underwent tracheal extubation immediately after surgery), group Ib (midazolam–fentanyl patients, extubation in the ICU) and group II (pneumonectomy patients). P:0.05 compared with: *group Ia; †baseline values; §groups Ia and Ib.

End of surgery

2 h after surgery

1st po day

80.1 (11.5) 90.9 (10.1)*† 92.1 (11.4)*† 80 (12) 91 (11)*† 94 (13)*† 20.1 (3.3) 21.6 (3.0) 27.5 (3.2)*† 9.9 (2.1) 10.1 (2.8) 13.2 (3.2)*† 11.5 (2.2) 11.6 (2.6) 15.3 (3.3)*† 3.20 (0.5) 3.28 (0.6) 2.81 (0.4)* 1739 (356) 2020 (310)* 2390 (293)*† 401 (67) 395 (58) 327 (65)

89.1 (9.8)† 91.0 (11.1)† 93.0 (11.4)† 82 (13) 85 (12) 85 (12) 20.2 (3.1) 21.1 (3.0) 25.4 (3.5)*† 9.2 (2.1) 10.1 (2.3) 13.9 (3.7)*† 10.9 (2.1) 10.8 (3.2) 13.9 (2.3)*† 3.50 (0.6) 3.42 (0.5) 2.80 (0.5)* 1820 (287) 1978 (319) 2392 (297)*† 413 (69) 391 (76) 303 (70)*

(115.5 (13.1) pg ml91). Plasma concentrations of 6-keto-prostaglandin F1alpha remained significantly increased until the end of the study in patients undergoing pneumonectomy (60.4 (8.8) pg ml91).

Figure 2 Changes in plasma concentrations of vasopressin (normal range :3.0 pg ml91) and endothelin-1 (normal range 2.0–5.0 pg ml91) in the three groups (see fig. 1 for details). P:0.05 compared with: *group Ia; †baseline values; §groups Ia and Ib.

Regulators of the circulation in lung surgery

737

Figure 3 Changes in plasma concentrations of 6-ketoprostaglandin F1alpha (normal values :30 pg ml91) and atrial natriuretic peptide (ANP) (normal values 20–80 pg ml91) in the three groups (see fig. 1 for details). P:0.05 compared with: *group Ia; †baseline values; §groups Ia and Ib.

Discussion We found that propofol-treated patients showed early co-operation, were orientated, could sigh and cough on command, and effective physiotherapy was possible earlier in the postoperative period than in the other lobectomy group. Haemodynamic variables in the propofol group were almost comparable with patients in whom an anaesthetic regimen with longer acting anaesthetic substances (midazolam–fentanyl–pancuronium (standard management technique) and subsequent delayed extubation in the ICU were used. CI was slightly higher in propofol-treated patients, possibly because of a lower SVR than in the other group. PAP during OLV was also lower in the propofol than in the other group indicating less increase in right ventricular work. MAP remained stable in propofol-treated patients, probably because of adequate volume replacement which is known to prevent unwanted hypotension during propofol infusion. Plasma concentrations of catecholamines were significantly higher in the anaesthetic regimen with delayed extubation than in propofol-treated patients who underwent extubation immediately after surgery. All other vasoactive substances (ET-1, vasopressin, ANP, 6-keto-prostaglandin F1alpha) were similar between the two groups during the study. PNEUMONECTOMY PATIENTS

Patients

undergoing

pneumonectomy

showed

significant differences both in haemodynamic state and vasoactive regulators of the circulation compared with patients in group Ib in whom the same anaesthetic technique was used. PAP increased during OLV and remained higher than baseline until the end of the study. In a study by Lewis and coworkers,16 including 20 patients with moderately severe chronic obstructive pulmonary disease undergoing pneumonectomy, only 10% had a normal PAP in the immediate postoperative period. Others also reported a marked increase in right ventricular afterload during OLV and after completion of resection.16–18 Increased right ventricular (RV) afterload in this situation most likely results from increased pulmonary vascular constriction secondary to hypoxic vasoconstriction (HPV), pulmonary microvascular plugging and decreased pulmonary vascular compliance.17 Plasma concentrations of potent vasoconstrictive substances (e.g. vasopressin, ET-1, noradrenaline) increased significantly in patients undergoing pneumonectomy. All are potent vasopressors which may act particularly in the intestinal circulation with detrimental consequences for the organism.19 Activation of the sympathetic system is caused by a complex interaction of different stimuli, for example hypotension, hypovolaemia and hypoperfusion. The increase in plasma concentrations of catecholamines reflects the magnitude of the (reflectory) sympathetic nervous system response in this situation. In recent years it has become obvious that in addition to “systemic” regulators of the circulation (e.g. catecholamines), substances released by the endothelium are also involved in the pathophysiology of circulatory abnormalities.20 21 ET-1 is one of these substances which can induce regional vasoconstriction followed by impaired organ blood flow. ET-1 is released by pathophysiological states that are characterized by limitation of cardiac output or by hypotension. Apart from the direct vasoconstrictive actions of ET-1, it may also influence the circulation by altering other regulator systems. Thus ET-1 was reported to stimulate the release of catecholamines and renin22 thus possibly contributing to vasoconstriction and reduced blood flow at the microcirculatory level. In contrast, ET-1 also influences ANP secretion increasing circulating ANP in dogs23 24: ANP may function as a negative feedback regulator of ET-1 and the vasoconstrictive actions of ET-1 are opposed by the release of other vasoactive substances, including ANP. Plasma concentrations of ANP were significantly increased in the pneumonectomy group. A linear relationship between ANP plasma concentrations and right atrial (RAP) and pulmonary capillary wedge pressure (PCWP) has been shown.25 However, this may not be the only mechanism by which plasma concentrations of ANP were increased. Endothelin has been shown to stimulate release of ANP in vivo and in vitro.26 Both substances have important interactions on the vascular level: ANP appears to counteract the contractile response to endothelin.27 Additionally, results from animal and human studies suggest that adrenaline (which was also increased in pneumonectomy patients)

738 stimulates ANP secretion.24 6-Keto-prostaglandin F1alpha was also increased in the pneumonectomy patients. Nakae and colleagues28 showed that it may serve as a marker of illness severity. Pneumonectomy patients showed delayed stabilization of haemodynamic state and pulmonary function resulting in delayed tracheal extubation in the ICU. Thus it can be assumed that these were at higher risk of developing cardiopulmonary complications. The significant changes in plasma concentrations of potent vasoactive substances may have important consequences for organ function. Altered microregional vascular tone results in some capillary regions being over perfused while others are under perfused.29 Although three of the 15 pneumonectomy patients and none of the lobectomy patients died during their stay in the intensive care unit, the population was too small to draw definite conclusions on whether or not the measured vasoactive substances were responsible for the difference in survival rates between the groups. The reasons for the more pronounced changes in important regulators of the circulation in the pneumonectomy patients can only be speculative. It is unlikely that the technique of anaesthesia was the major reason because lobectomy patients in group Ib underwent the same anaesthetic management (midazolam–fentanyl–pancuronium) but showed only moderate changes in plasma concentrations of the vasoactive substances. As indicated by the higher ASA classification, severity of illness in the pneumonectomy patients appeared to be higher than in the two lobectomy groups resulting in less cardiopulmonary reserve. Last but not least, altered production or uptake may be the reason for the increased plasma concentrations of some of the vasoactive substances. There is convincing evidence that the lung is a highly metabolic organ which contributes markedly to production and elimination of various substances (e.g. noradrenaline, prostaglandins, ANP). Endothelial cells appear to play a particular role in this process. The pulmonary endothelium has an immense surface area and actively participates in production and inactivation of vasoactive substances.30 As an example, the lung is apparently the most relevant organ in rapid clearance of endothelin.31 The pulmonary circulation also has the potential to generate endothelin from its own endothelium given an appropriate stimulus such as hypoxia.23 This may be one explanation for the higher plasma concentrations of ET-1 in our pneumonectomy patients. The lung may also participate in regulation of the action of ANP secreted by the right atrium by active metabolism.4 Interactions of ET-1 with other vasoactive substances may have further influenced their plasma concentrations (e.g. inhibition of release of catecholamines and renin by ET-1, stimulation of ANP release by ET-1).8 In summary, regulation of the macro- and microcirculation is dependent on both centrally originated and locally derived control mechanisms. The anaesthetic regimen in patients undergoing major thoracic surgery was only of limited value in affecting important regulators of the circulation. Thus an

British Journal of Anaesthesia anaesthetic regimen using more short acting substances with early tracheal extubation of the patient appeared to be without detrimental effects on the (micro-) circulation. More extensive lung surgery (e.g. pneumonectomy) was followed by more extensive changes in plasma concentrations of important regulators of organ perfusion. As the (micro-) circulatory flow abnormalities appear to be a major determinant of development of organ function further studies are needed to assess whether its manipulation may improve patient outcome.

References 1. Shoemaker WC, Appel PL, Kram HB. Tissue oxygen debt as a determinant of lethal and nonlethal postoperative organ failure. Critical Care Medicine 1988; 94: 1117–1120. 2. Intaglietta M. Objectives for the treatment of the microcirculation in ischemia, shock, and reperfusion. In: Vincent JL, ed. Update in Intensive Care and Emergency Medicine. Berlin: Springer, 1990; 293–298. 3. Atlas SA. Atrial natriuretic factor: a new hormone of cardiac origin. Recent Progress in Hormonal Research 1986; 42: 207–209. 4. Needham P, Greewald JE. Atriopeptin: a cardiac hormone intimately involved in fluid, electrolyte, and blood-pressure hemostasis. New England Journal of Medicine 1986; 314: 828–834. 5. Brenner BM, Troy JL, Ballermann B. Endotheliumdependent vascular responses. Journal of Clinical Investigation 1989; 84: 1373–1378. 6. Yanagisawa M, Kurihara H, Kimura S. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature (London) 1988; 322: 411–415. 7. Ganghi CR, Berkowitz DE, Watkins D. Endothelins. Biochemistry and pathophysiologic actions. Anesthesiology 1994; 80: 892–905. 8. Änggard EE, Botting RM, Vane JR. Endothelins. Blood and Vessels 1990; 27: 269–281. 9. Bersten AD, Sibbald W. Circulatory disturbances in multiple system organ failure. Critical Care Clinics 1989; 5: 233–254. 10. Bakhle YS. Pharmacokinetic and metabolic properties of lung. British Journal of Anaesthesia 1990; 65: 79–93. 11. Moncada S, Korbut R, Bunting S, Vane JR. Prostacyclin is a circulating hormone. Nature (London) 1978; 273: 767–768. 12. Daniele RP, Barnes PJ, Goetzl EJ, Nadel J, O’Dorisio S, Kiley J, Jacobs T. Neuroimmune interactions in the lung. American Review of Respiratory Disease 1992; 145: 1230–1235. 13. Xuan YT, Whorton AR, Shearer-Poor E, Boyd J, Watkins WD. Determination of immunoreactive endothelin in medium from cultured endothelial cells and human plasma. Biochemicial and Biophysical Research Communications 1989; 164: 326–332. 14. Kross J, Dries DJ, Kumar P. Atrial natriuretic peptide may not play a role in diuresis and natriuresis after cardiac operations. Journal of Thoracic Cardiovascular Surgery 1992; 103: 1168–1171. 15. Pullan PT, Clappison BH, Johnston CI. Plasma vasopressin and human neurophysins in physiological and pathophysiological states associated with changes in vasopressin secretion. Journal of Clinical Endocrinology and Metabolism 1979; 49: 580–587. 16. Lewis JJ, Bastanfar M, Gabriel F, Mascha E. Right heart function and prediction of respiratory morbidity in patients undergoing pneumonectomy with moderately severe cardiopulmonary dysfunction. Journal of Thoracic Cardiovascular Surgery 1994; 108: 169–175. 17. Cryer HG, Mavroudis C, Yu J, Roberts AM, Cue JI, Richardson JD, Polk HJ. Shock, transfusion, and pneumonectomy. Death is due to right heart failure and increased pulmonary vascular resistance. Annals of Surgery 1990; 212: 197–201. 18. Crouch JD, Lucas CL, Keagy BA, Wilcox BR, Ha B. The acute effects of pneumonectomy on pulmonary vascular impedance in the dog. Annals of Thoracic Surgery 1987; 43: 613–616.

Regulators of the circulation in lung surgery 19. McNeill JR, Stark RD, Greeway CV. Intestinal vasoconstriction after hemorrhage: roles of vasopressin and angiotensin. American Journal of Physiology 1970; 219: 1342–1347. 20. Vane JR, Betting RM. Endothelium-derived vasoactive factors and the control of circulation. Seminar in Perinatology 1991; 15: 4–10. 21. Lüscher TF. Endothelin: Systemic arterial and pulmonary effects of a new peptide with biologic properties. American Review of Respiratory Disease 1992; 146 (Suppl. 2): S56–S60. 22. Boarder MR, Marriott DB. Characterization of endothelin-1 stimulation of catecholamine release from adrenal chromaffin cells. Journal of Cardiovascular Pharmacology 1989; 13 (Suppl. 5): S223–224. 23. Rubanyi GM, Bothelho LH. Endothelins. Federation of American Society of Experimental Biology 1991; 5: 2713–2720. 24. Rushkoaho H, Lang RE, Toth M, Ganten D, Unger T. Release and regulation of atrial natriuretic peptide (ANP). European Heart Journal 1987; 8 (Suppl. B): 99–109. 25. Bates ER, Shenker Y, Grekin RJ. The relationship between plasma levels of immunoreactive atrial natriuretic hormone and hemodynamic function in man. Circulation 1986; 73: 1155–1161.

739 26. Goetz KL, Wang BC, Madwed JB, Zhu Jl, Leadly RJ. Cardiovascular, renal, and endocrine responses to intravenous endothelin in conscious dogs. American Journal of Physiology 1988; 255: R1064–1068. 27. Marsden PA, Danthuluri NR, Brenner BM, Ballermann BJ, Brock TA. Endothelin action on vascular smooth muscle involves inositol trisphosphate and calcium mobilization. Biochemical and Biophysical Research Communications 1989; 158: 86–93. 28. Nakae H, Endo S, Inada K, Watanabe M, Baba N, Yoshida M. Relationship between leukotriene B4 and prostaglandin IV-2V- in patients with sepsis. Research Communications in Molecular Pathology and Pharmacology 1994; 86: 37–42. 29. Vane JR, Änggard EE, Botting RM. Regulatory functions of the vascular endothelium. New England Journal of Medicine 1990; 323: 27–36. 30. Pitt B, Lister G. Interpretation of metabolic function of the lung. Clinics in Chest Medicine 1989; 10: 1–12. 31. Kanse SM, Ghatei MA, Bloom SR. Endothelin binding sites in porcine aortic and rat lung membranes. European Journal of Biochemistry 1989; 182: 175–179.