Portal de Periódicos da CAPES

Epidural analgesia and arterial reconstructive surgery to the leg: effects on fibrinolysis and platelet degranulation

2001; Elsevier BV; Volume: 86; Issue: 2 Linguagem: Inglês

10.1093/bja/86.2.230

1471-6771

Stephanie Bew, Allison Bryant, J.P. DESBOROUGH, G. M. Hall,

Peripheral Artery Disease Management

It has been suggested that the incidence of early graft occlusion after arterial reconstructive surgery to the leg may be decreased by epidural analgesia. This effect may be mediated by the suppression of the usual cortisol response to surgery, which results in increased circulating plasminogen activator inhibitor-1 with consequent adverse effects on fibrinolysis. To investigate this and other potential mechanisms, 30 patients undergoing arterial reconstructive surgery to the leg were randomized to receive either general anaesthesia or general anaesthesia plus epidural analgesia. Post-operative analgesia was provided by morphine infusion or epidural analgesia, respectively. Blood samples were collected at 0, 2, 4, 6, 12 and 24 h, and 2, 3 and 5 days and analysed for cortisol, plasminogen activator inhibitor-1 antigen, interleukin-6 and beta thromboglobulin. The incidence of graft-related and systemic complications was recorded for 30 days. Only one patient developed early graft occlusion that required embolectomy and eventually amputation. There were no significant changes from control values in either group of patients in circulating cortisol, plasminogen activator inhibitor-1 and beta thrombogobulin (a marker for platelet degranulation). Interleukin-6 values increased significantly in both groups after 4 h and remained elevated until day 3. There were no significant differences between the groups in any variable measured. We conclude that any effect of epidural analgesia on early graft patency is unlikely to be mediated by fibrinolysis or platetlet degranulation. It has been suggested that the incidence of early graft occlusion after arterial reconstructive surgery to the leg may be decreased by epidural analgesia. This effect may be mediated by the suppression of the usual cortisol response to surgery, which results in increased circulating plasminogen activator inhibitor-1 with consequent adverse effects on fibrinolysis. To investigate this and other potential mechanisms, 30 patients undergoing arterial reconstructive surgery to the leg were randomized to receive either general anaesthesia or general anaesthesia plus epidural analgesia. Post-operative analgesia was provided by morphine infusion or epidural analgesia, respectively. Blood samples were collected at 0, 2, 4, 6, 12 and 24 h, and 2, 3 and 5 days and analysed for cortisol, plasminogen activator inhibitor-1 antigen, interleukin-6 and beta thromboglobulin. The incidence of graft-related and systemic complications was recorded for 30 days. Only one patient developed early graft occlusion that required embolectomy and eventually amputation. There were no significant changes from control values in either group of patients in circulating cortisol, plasminogen activator inhibitor-1 and beta thrombogobulin (a marker for platelet degranulation). Interleukin-6 values increased significantly in both groups after 4 h and remained elevated until day 3. There were no significant differences between the groups in any variable measured. We conclude that any effect of epidural analgesia on early graft patency is unlikely to be mediated by fibrinolysis or platetlet degranulation. Patients undergoing arterial reconstructive surgery to the leg are at high risk of perioperative cardiovascular and respiratory complications, early occlusion of their graft requiring urgent embolectomy, further surgery or amputation of the limb. Studies, which attempted to decrease early morbidity and mortality by the use of regional anaesthesia, concluded that the choice of anaesthetic, provided it was carefully conducted, had no influence on overall morbidity or mortality.1Rivers SP Scher LA Sheehan E Veith FJ Epidural versus general anaesthesia for infrainguinal arterial reconstruction.J Vasc Surg. 1991; 14: 764-770Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, 2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar, 3Bode RH Lewis KP Zarich SW et al.Cardiac outcome after peripheral vascular surgery. Comparison of general and regional anesthesia.Anesthesiology. 1996; 84: 3-13Crossref PubMed Scopus (279) Google Scholar However, it is possible that anaesthesia may affect the success of the graft. Christopherson and colleagues found that epidural analgesia decreased early graft occlusion from 22 to 4%.2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar This improvement was attributed to enhanced fibrinolysis measured by changes in plasminogen activator inhibitior-1 (PAI-1).4Rosenfeld BA Beattie C Christopherson R et al.The effects of different anesthetic regimens on fibrinolysis and the development of postoperative arterial thrombosis.Anesthesiology. 1993; 79: 435-443Crossref PubMed Scopus (266) Google Scholar The authors postulated that epidural analgesia prevented the usual cortisol response to surgery, and that the cortisol response results in an increase in PAI-1, with adverse effects on fibrinolysis. If epidural analgesia improves the success of arterial reconstructive grafts, it is important to determine the mechanism(s). There are other potentially beneficial effects of epidural analgesia in addition to the postulated influence on fibrinolysis. For example, platelet aggregation is inhibited,5Naesh O Haljamäe H Hindberg I Holm J Jivegård L Wennmalm A Epidural anaesthesia prolonged into the postoperative period prevents stress response and platelet hyperaggregability after peripheral vascular surgery. Eur.J Vasc Surg. 1994; 8: 395-400Abstract Full Text PDF Scopus (19) Google Scholar sympathetic block increases leg blood flow,6Cousins MJ Wright CJ Graft, muscle, skin blood flow after epidural block in vascular surgical procedures.Surg Gynaecol Obstet. 1971; 133: 59-64Google Scholar and an i.v. fluid load decreases blood viscosity.7Korosue K Heros RC Ogilvy CS Hyodo A Tu Y-K Graichen R Comparison of crystalloids and colloids for hemodilution in a model of focal cerebral ischaemia.J Neurosurg. 1990; 73: 576-584Crossref PubMed Scopus (46) Google Scholar Systemic effects of local anaesthetics include decreased plasma viscosity8Orr JE Lowe GDO Nimmo WS Watson R Forbes CD A haemorheological study of lignocaine.Br J Anaesth. 1986; 58: 306-309Crossref PubMed Scopus (14) Google Scholar and platelet aggregability,9Borg T Modig J Potential anti-thrombotic effects of local anaesthetics due to their inhibition of platelet aggregation.Acta Anaesthesiol Scand. 1985; 29: 739-742Crossref PubMed Scopus (87) Google Scholar and increased prostacyclin production.10Casey LC Armstrong MC Fletcher JR Ramwell PW Lidocaine increases prostacyclin in the rat.Prostaglandins. 1980; 19: 977-984Crossref PubMed Scopus (18) Google Scholar Good analgesia after surgery, with the prevention of pain-mediated vasoconstriction, may also be important. To investigate some of these potential mechanisms we examined the effects of epidural analgesia on changes in serum cortisol, fibrinolysis, platelet degranulation and cytokine release for 5 days after arterial reconstructive surgery to the leg. Circulating cortisol concentrations were measured to establish the usual endocrine response to this surgery and its relationship to plasminogen activator inhibitor-1 antigen (PAI-1 ag). Changes in platelet degranulation were assessed by measurement of circulating beta thromboglobulin (βTG). The interleukin-6 (IL-6) response to surgery was determined, not only to indicate the severity of the inflammatory response to surgical trauma, but also because IL-6 initiates acute phase protein synthesis thus increasing several prothrombotic factors.11Heinrich PC Castell JV Andus T Interleukin-6 and the acute phase response.Biochem J. 1990; 265: 621-636Crossref PubMed Scopus (2303) Google Scholar With the approval of the Local Research Ethics Committee and after written informed consent, 30 patients were randomized to receive either general anaesthesia (GA) or general anaesthesia plus epidural analgesia (GAE) for arterial reconstructive surgery to the leg. Patients were included in the study if they were scheduled for vascular reconstruction below the inguinal ligament and excluded if there were contraindications to epidural analgesia such as a coagulopathy and local or systemic sepsis. Pre-operative assessment of the patients included the Goldman risk index12Goldman L Caldera DL Nussbaum SR et al.Multifactorial index of cardiac risk in noncardiac surgical procedures.N Engl J Med. 1977; 297: 845-850Crossref PubMed Scopus (2075) Google Scholar the ankle brachial pressure index (ABPI) and the indications for surgery. Patients were randomized, using random number tables, to receive either GA or GAE. All operations started between 08.00 and 11.00 h and routine cardiac medications were continued until 2 h before surgery. Insulin-dependent diabetic patients had an i.v. infusion of glucose and insulin from 06.00 h on the morning of surgery maintaining blood glucose at 5–10 mmol litre−1. The infusion was continued for 24 h after surgery when oral intake and the usual insulin regimen were resumed. Non-insulin-dependent diabetic patients had oral hypoglycaemic drugs stopped 12 h before surgery and were also managed with a glucose-insulin infusion during and after surgery. Patients taking aspirin continued with the drug until hospital admission. Premedication was not given and, in the anaesthetic room after routine non-invasive monitoring was established, a wide bore, peripheral venous cannula and a radial artery cannula were inserted using local anaesthesia. All patients received an i.v. fluid load of 0.9% sodium chloride solution 10–12 ml kg−1 and a sleep dose of thiopentone. Neuromuscular block was established with vecuronium 0.1 mg kg−1, the trachea intubated and anaesthesia maintained with nitrous oxide in 40% oxygen and isoflurane. During surgery bolus doses of vecuronium and fentanyl up to 5 μg kg−1 were given, as needed. Epidural analgesia was established in the GAE group at the L2-3 interspace and a test dose of 0.5% bupivacaine 3 ml was given followed by increments of 0.25% bupivacaine to establish a block to at least T10. Anaesthesia was induced as in the GA group but with a maximum of fentanyl 50 μg. During surgery epidural analgesia was maintained with an infusion of 0.25% bupivacaine 6 ml h−1. In all patients a central venous cannula and urinary catheter were inserted after induction of anaesthesia. During surgery, 0.9% sodium chloride solution was infused at 5–8 ml kg−1 h−1 and gelatin solution given to replace blood loss unless the haematocrit was 10 kPa and 0.9% sodium chloride solution infused at 1–2 ml kg−1 h−1. Analgesia was provided by an i.v. morphine infusion in the GA group and by the epidural infusion of 0.125% bupivacaine+fentanyl 10 μg ml−1 in the GAE group. In addition to routine haemodynamic monitoring, pulses, perfusion and temperature in the operated leg were recorded hourly and pain measured every 4 h. Epidural analgesia or a morphine infusion were continued for at least 18 h after surgery and followed by oral analgesics. Blood samples were collected before induction of anaesthesia, at 0, 2, 4, 6, 12 and 24 h after the start of surgery and at 08.00 h on days 2, 3 and 5. Samples for the measurement of cortisol and IL-6 were centrifuged and the serum stored at –70°C. Samples for the determination of PAI-1 ag were collected in refrigerated stabilyte tubes (Biopool) containing 1/10 volume strong acid citrate, transferred on ice, centrifuged immediately and the top third of supernatant plasma stored at –70°C. Blood for βTG analysis was collected in tubes containing 1/10 volume anticoagulant (Dipotassium EDTA 5.42 g, theophylline 0.36 g, adenosine 0.267 g, 0.19% saline 100 ml), and immediately double centrifuged to ensure removal of all platelets before storage at –70°C. If venepuncture was necessary to obtain the sample, the first 2–3 ml were discarded. Serum cortisol concentration was measured by ELISA (Milennia Cortisol EAI), which had a sensitivity of 7.3 nmol litre−1, intra-assay coefficient of variation (CV) of 3.2% and inter-assay CV of 8.0%. IL-6 values were determined with a sandwich ELISA (Quantikine R and D systems) which had a sensitivity of 0.7 pg ml−1, intra-assay CV of 4.3% and inter-assay CV of 6.3%. PAI-1 antigen was measured by ELISA (Biopool TintElize) with a sensitivity of 2 ng ml−1 and an inter-assay CV of 2.5% at 40 ng ml−1. βTG was estimated by non-proprietary ELISA in which rabbit antiserum to human βTG was purified with n-octanoic acid and used to coat microtitre plates (Maxisorb, Nunc, Denmark). Standards were calibrated against the 1st International Standard (83/501, National Institute for Biological Standards and Controls, Potters Bar, UK). Samples and standards were added to the plates and incubated for 1 h, washed and labelled with tag antibody (diluted biotin antibody/horse-radish peroxidase/streptavidin), and incubated for an 1 h. After further washing, enzyme activity was measured by the addition of substrate solution (o-phenylenediamine hydrochloride 10 mg and 30% hydrogen peroxide 7 μl in 0.1 M citrate phosphate buffer 15 ml, pH 5.0) and stopped after 10 min with 2 M sulphuric acid. Absorbance was read at 492 nm. Intra-assay CV was 3.7% and inter-assay CV was 4.3%. All assays were undertaken in duplicate. Patients were assessed until discharge to monitor the outcome of surgery and the incidence of complications. Failure of the arterial graft was defined as a decrease >0.3 in the ABPI, need for anticoagulation, embolectomy, regrafting and amputation. Cardiovascular, respiratory, infectious and renal complications were noted as defined by Christopherson and colleagues.2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar The sample size of 30 patients was based on the expected effect of epidural analgesia on the cortisol response, and had the power of 80% to detect a standardized difference of 1.0 with a significance level of 0.05. Statistical analysis was undertaken with SPSS version 6.1. The results were analysed with chi-squared test, one-way analysis of variance and two-way analysis of variance for parametric and non-parametric data as appropriate. P<0.05 was considered statistically significant. The details of the patients studied including pre-operative characteristics, previous vascular surgery, current medication, reason for surgery and type of graft were similar in both groups (Table 1). Perioperative management of the patients was not significantly different between the groups, in particular the fluid preload, duration of surgery and intraoperative fluid management (Table 2). Graft-related and systemic complications were uncommon and there were no significant differences between the groups (Table 3). The patient who died had an intra-operative blood loss of 5600 ml and succumbed in the intensive care unit to multiple organ failure on the sixth postoperative day. Four patients returned to theatre within 24 h of surgery, three with bleeding and one for an embolectomy. The latter patient required further embolectomies and finally a below knee amputation.Table 1Patient characteristics. Table indicates number of patients except for age, weight, ABPI and Goldman risk score which are shown as mean (range) [sd]General anaesthesiaGeneral anaesthesiaplus epidural anaglesiaM/F12/311/4Age (yr)71 (54–87) [8.6]69 (53–88) [10.2]Weight (kg)73 (50–97) [13.0]68 (45–92) [13.9]IDDM/NIDDM3/30/3Never smoked32Ex-smoker82Currently smoking411IHD54Previous MI23Hypertension106ABPI0.29 (0–0.83) [0.29]0.33 (0–0.79) [0.29]Goldman risk score4.0 (0–12) [3.9]5.5 (0–12) [4.9]Previous vascular surgeryNone65Limb angioplasty34Peripheral vascular34Carotid surgery11Coronary artery graft21MedicationCalcium antagonist54Digoxin13ACE inhibitors43Aspirin57Reason for surgeryCritical ischaemia67Claudication65Rest pain23Gangrene10Graft typePTFE1010Vein45No graft10 Open table in a new tab Table 2Perioperative management. Values shown are mean (range) [sd]General anaesthesiaGeneral anaesthesia plusepidural analgesiaPre-operativeHaematocrit (%)39 (31–45) [4.5]38 (33–48) [4.5]Base excess1.0 (–4.8 to 4.3) [2.5]1.6 (–1.4 to 5.5) [1.9]IntraoperativePreload (ml)1010 (700–1500) [302]987 (700–1200) [155]Duration of surgery (min)117 (80–260) [49]117 (80–360) [81]Blood loss (ml)934 (200–5600) [1359]519 (200–1500) [363]Colloid infused (ml)1030 (0–3500) [828]953 (0–4000) [1018]Blood transfused (ml)495 (0–1400) [144]235 (0–1400) [413]PostoperativeHaematocrit (%)32 (29–34) [3.5]31 (28–34) [2.3]Base excess–4.2 (–11.8 to 1.1) [3]–3.3 (–7.0 to –0.1) [2]Core temp (°C)35.7 (33.5–36.7) [0.9]35.5 (34.5–36.2) [0.5]Lactate (mmol litre−1)1.0 (0.1-1.7) [0.6]0.9 (0.2–1.7) [0.5]ABPI0.83 (0.43–1.0) [0.2]0.86 (0.48–1.0) [0.2] Open table in a new tab Table 3Graft-related and systemic complicationsGeneral anaesthesiaGeneral anaesthesiaplus epidural analgesiaReoperation in 24 h forBleeding12Embolectomy01Early complicationsRespiratory11Renal10Wound infection03Death1030-day complicationsAmputation01 Open table in a new tab Changes in circulating variables are shown in Table 4. There were no significant differences between groups in cortisol, IL-6, PAI-1 ag and βTG concentrations. There were no significant changes from pre-induction values in both groups for cortisol, PAI-1 ag and βTG values. However, IL-6 concentrations increased significantly in both groups after 4 h and remained significantly elevated until day 3 (P<0.05).Table 4Changes in mean (SEM) circulating cortisol (nmol litre−1), PAI-1 ag (ng ml−1), IL-6 (pg ml−1) and βTG (ng ml−1) concentrations in GA patients and GAE patients. There were no significant differences between the groups. IL-6 concentrations increased significantly in both groups from 4 h to day 3 (P<0.05)TimePre02 h4 h6 h12 h24 hDay 2Day 3Day 5Cortisol (nmol litre−1)GA605469342604652628699580659624(65)(75)(57)(72)(60)(57)(66)(80)(44)(65)GAE567480296524591729701541559702(76)(64)(57)(79)(80)(111)(109)(93)(72)(71)PAI-1 ag (ng ml−1)GA44353850474742455145(5)(4)(4)(5)(4)(5)(4)(4)(2)(5)GAE40313850474742455145(5)(3)(4)(5)(4)(5)(4)(4)(2)(5)IL-6 (pg ml−1)GA111012729689102643814(2)(2)(2)(20)(20)(14)(17)(10)(11)(4)GAE98124072113121663519(3)(3)(3)(6)(12)(33)(20)(8)(9)(4)βTG (ng ml−1)GA125133215235213253136155159155(16)(18)(56)(58)(54)(44)(16)(22)(27)(27)GAE11490109175152233123201144161(13)(10)(19)(13)(14)(57)(20)(43)(21)(26) Open table in a new tab No effect of epidural analgesia was found on the circulating variables chosen to represent the hormonal, inflammatory, fibrinolytic and platelet degranulation responses during and after arterial reconstructive surgery to the leg. Furthermore, the only variable to show a significant increase during the study was IL-6. Baseline βTG values were increased (normal range 10–40 ng ml−1), reflecting the enhanced platelet activation found in elderly patients with vascular disease.13Kieldsen SE Lande K Gjesdal K et al.Increased platelet release reaction in 50-year-old men with essential hypertension: correlation with atherogenic cholesterol fractions.Am Heart J. 1987; 113: 151-155Abstract Full Text PDF PubMed Scopus (23) Google Scholar 14Zahavi J Jones NAG Leyton J Dubiel M Kakkar VV Enhanced in vivo platelet “release reaction” in old healthy individuals.Thromb Res. 1980; 17: 329-336Abstract Full Text PDF PubMed Scopus (100) Google Scholar We chose to compare GA with GAE to ensure that any differences observed were the result of regional block. An anaesthetic technique of epidural analgesia and sedation would have introduced other variables, but the benefits of avoiding GA merit investigation. Perioperative morbidity is common after arterial reconstructive surgery as the patients are usually elderly, often hypertensive and diabetic, most are current or ex-smokers and widespread arterial disease is common.15Hertzer NR Basic data concerning associated coronary disease in peripheral vascular patients.Ann Vasc Surg. 1987; 1: 616-620Abstract Full Text PDF PubMed Scopus (149) Google Scholar Silent ischaemia and perioperative myocardial infarction occur frequently, with coronary artery disease the major cause of early and late mortality.16Taylor LM Yeager RA Moneta GL McConnell DB Porter JM The incidence of perioperative myocardial infarction in general vascular surgery.J Vasc Surg. 1992; 15: 52-61Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar 17Yeager RA Moneta GL Edwards JM Taylor LM McConnell DB Porter JM Late survival after perioperative mycocardial infarction complicating vascular surgery.J Vasc Surg. 1994; 20: 598-606Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar Previous studies failed to show any benefit from regional anaesthesia in decreasing perioperative mortality and morbidity.1Rivers SP Scher LA Sheehan E Veith FJ Epidural versus general anaesthesia for infrainguinal arterial reconstruction.J Vasc Surg. 1991; 14: 764-770Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, 2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar, 3Bode RH Lewis KP Zarich SW et al.Cardiac outcome after peripheral vascular surgery. Comparison of general and regional anesthesia.Anesthesiology. 1996; 84: 3-13Crossref PubMed Scopus (279) Google Scholar Early graft occlusion is an important cause of morbidity and occurs in 6–10% of patients within 30 days of surgery and results in anticoagulation, embolectomy, or further surgery.18Bandyk DF Schmitt DD Seabrook GR Adams MB Towne JB Monitoring functional patency of in situ saphenous vein bypasses. The impact of a surveillance protocol and elective revision.J Vasc Surg. 1989; 9: 286-296Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar 19Whittemore AD Failure of peripheral arterial reconstruction.Acta Chir Scand Suppl. 1988; 550: 74-80Google Scholar Christopherson and colleagues found an impressive decrease in early occlusion of the graft in patients receiving epidural analgesia, from 22 to 4%.2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar Our results failed to replicate these findings and early graft occlusion occurred in only one patient who required embolectomy and eventually amputation (Table 3). There are many possible reasons for this difference, for example the mean duration of surgery was 5 h in Christopherson et al.,2Christopherson R Beattie C Frank SM et al.Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery.Anesthesiology. 1993; 79: 422-434Crossref PubMed Scopus (477) Google Scholar but our low rates of graft occlusion are similar to those reported recently by other groups using GA and regional anaesthesia.20Dodds TM Yeager MP Walsh DB Mandel D General versus regional anaesthesia for peripheral vascular surgery.Anesth Analg. 1997; 84: SCA86Google Scholar 21Pierce ET Pomposelli Jr, FB Stanley GD et al.Anaesthesia type does not influence early graft patency or limb salvage rates of lower extremity arterial bypass.J Vasc Surg. 1997; 25: 226-232Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar It is possible, therefore, that this influential study of Christopherson and colleagues is not representative of arterial reconstructive surgery to the leg in other major centres. The mechanism proposed for the beneficial effects of epidural analgesia on graft patency was an attenuation of the perioperative increase in PAI-1, thus decreasing the inhibition of fibrinolysis. PAI-1 is a serine protease inhibitor produced by endothelial cells, and increased concentrations are associated with thrombosis.22Páramo JA Alfaro MJ Rocha E Postoperative changes in the plasmatic levels of tissue-type plasminogen activator inhibitor and its fast acting inhibitor – relationship to deep vein thrombosis and influence of prophylaxis.Thromb Haemost. 1985; 54: 713-716PubMed Google Scholar Rosenfeld and colleagues only measured PAI-1 activity on three occasions within 72 h of surgery, and found significantly lower values in epidural patients at the end of surgery.4Rosenfeld BA Beattie C Christopherson R et al.The effects of different anesthetic regimens on fibrinolysis and the development of postoperative arterial thrombosis.Anesthesiology. 1993; 79: 435-443Crossref PubMed Scopus (266) Google Scholar The authors suggested that the increase in PAI-1 was secondary to increased cortisol secretion, but circulating values were not measured. A later report of urinary cortisol excretion in a sub-group of 60 patients in this study failed to show an effect of epidural analgesia23Parker SD Breslow MJ Frank SM et al.Catecholamine and cortisol responses to lower extremity revascularization: correlation with outcome variables.Crit Care Med. 1995; 23: 1954-1961Crossref PubMed Scopus (114) Google Scholar suggesting that any link between cortisol and PAI-1 was tenuous. The inability of epidural analgesia to alter fibrinolysis was also found in a study of patients undergoing knee arthroplasty.24Sharrock NE Go G Williams-Russo P Haas SB Harpel PC Comparison of extradural and general anaesthesia on the fibrinolytic response to total knee arthroplasty.Br J Anaesthesia. 1997; 79: 29-34Crossref PubMed Scopus (54) Google Scholar The pattern of circulating cortisol response to arterial reconstructive surgery to the leg has been poorly defined.5Naesh O Haljamäe H Hindberg I Holm J Jivegård L Wennmalm A Epidural anaesthesia prolonged into the postoperative period prevents stress response and platelet hyperaggregability after peripheral vascular surgery. Eur.J Vasc Surg. 1994; 8: 395-400Abstract Full Text PDF Scopus (19) Google Scholar We observed no difference between the GA and GAE groups for up to 5 days, but also no change from baseline values indicating that, with appropriate anaesthesia and postoperative analgesia, peripheral arterial surgery causes little stimulation of the hypothalamic-pituitary adrenal axis. We cannot exclude the possibility that the low incidence of early graft occlusion in this study was related to the absence of a cortisol response. Surgery evoked a rapid increase in IL-6 that was sustained until day 3 and was unaffected by epidural analgesia. Ischaemia and reperfusion injury also stimulate IL-6 production and this may have contributed to the response, which was greater than that found after routine abdominal surgery.25Seekamp A Warren JS Remik DG Till GO Ward PA Requirements for tumour necrosis factor-a and interleukin-1 in limb ischaemia/reperfusion injury and associated lung injury.Am J Pathol. 1993; 143: 453-463PubMed Google Scholar IL-6 induces acute phase protein synthesis in the liver.11Heinrich PC Castell JV Andus T Interleukin-6 and the acute phase response.Biochem J. 1990; 265: 621-636Crossref PubMed Scopus (2303) Google Scholar Some of these proteins, fibrinogen and C reactive protein (CRP), are prothrombotic, and the binding of fibrinogen to platelets is one of the early events in arterial thrombosis. Plasma CRP concentrations correlate with the presence and severity of peripheral arterial atherosclerosis,26Heinrich J Schulte H Schönfeld R Köhler E Assmann G Association of variables of coagulation, fibrinolysis and acute-phase with athersclerosis in coronary and peripheral arteries and those arteries supplying the brain.Thromb Haemost. 1995; 73: 374-379PubMed Google Scholar and this acute phase reactant also stimulates an increase in tissue factor which initiates the extrinsic pathway of coagulation.27Cermak J Key NS Bach RR Balla J Jacob HS Vercellotti GM C-reactive protein induces human peripheral blood monocytes to synthesize tissue factor.Blood. 1993; 82: 513-520Crossref PubMed Google Scholar Many of these prothrombotic factors are produced in response to tissue damage and so will be unaffected by neural block. The importance of surgical trauma in the development of the prothrombotic state is shown by the failure of stress hormone infusions to mimic postoperative hypercoagulability.28Rosenfeld BA Nguyen ND Sung I Faraday N Neuroendocrine stress hormones do not recreate the postoperative hypercoagulable state.Anesth Analg. 1998; 86: 640-645Crossref PubMed Google Scholar Patients with vascular disease have platelets with increased aggregability because of factors such as age, hypertension, and hyperlipidaemia.13Kieldsen SE Lande K Gjesdal K et al.Increased platelet release reaction in 50-year-old men with essential hypertension: correlation with atherogenic cholesterol fractions.Am Heart J. 1987; 113: 151-155Abstract Full Text PDF PubMed Scopus (23) Google Scholar 14Zahavi J Jones NAG Leyton J Dubiel M Kakkar VV Enhanced in vivo platelet “release reaction” in old healthy individuals.Thromb Res. 1980; 17: 329-336Abstract Full Text PDF PubMed Scopus (100) Google Scholar A further increase in platelet activation during surgery results from exposure to damaged endothelium and thrombin formation, but catecholamines, angiotensin and IL-6 may also be implicated. Epidural analgesia might decrease platelet activation by effects on leg blood flow, a reduction in catecholamine secretion and even a direct effect of bupivacaine itself.9Borg T Modig J Potential anti-thrombotic effects of local anaesthetics due to their inhibition of platelet aggregation.Acta Anaesthesiol Scand. 1985; 29: 739-742Crossref PubMed Scopus (87) Google Scholar Thromboelastograph studies in major vascular surgery showed a decrease in platelet-fibrinogen interactions with epidural analgesia.29Tuman KJ McCarthy RJ March RJ DeLaria GA Patel RV Ivankovich AD Effects of epidural anesthesia and analgesia on coagulation and outcome after major vascular surgery.Anesth Analg. 1991; 73: 696-704Crossref PubMed Scopus (525) Google Scholar These early stages of platelet activation are reversible, whereas degranulation occurs later and is irreversible. We used βTG to assess platelet degranulation and found no effect of epidural analgesia. These results may have been influenced by the marked activation that was already present in the patients before surgery, so that any beneficial effects of epidural analgesia were not apparent. Although patients continued to take aspirin until the day before surgery, this was unlikely to have affected the βTG data.30Lip GY Lip PL Zarifis J et al.Fibrin D-dimer and beta-thromboglobulin as markers of thrombogenesis and platelet activation in atrial fibrillation. Effects of introducing ultra-low-dose warfarin and aspirin.Circulation. 1996; 94: 425-431Crossref PubMed Scopus (224) Google Scholar We conclude that the supplementation of GA with epidural analgesia in patients undergoing arterial reconstructive surgery to the leg is not associated with changes in cortisol, PAI-1, IL-6 and βTG responses. This study was funded by a grant from the Special Trustees of St. George's Hospital.