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

S. A. Bew2, A. E. Bryant1, J. P. Desborough3 and G. M. Hall*,1

1Department of Anaesthesia and Intensive Care Medicine, St George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK. 2Department of Anaesthesia, Leeds General Infirmary, Great George Street, Leeds LS1 3EX, UK. 3Department of Anaesthesia, Epsom General Hospital, Dorking Road, Epsom, Surrey KT18 7EG, UK*Corresponding author

Accepted for publication: August 11, 2000


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
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.

Br J Anaesth 2001; 86: 230–5

Keywords: surgery, vascular; anaesthetic techniques, epidural


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
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.13 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%.2 This improvement was attributed to enhanced fibrinolysis measured by changes in plasminogen activator inhibitior-1 (PAI-1).4 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,5 sympathetic block increases leg blood flow,6 and an i.v. fluid load decreases blood viscosity.7 Systemic effects of local anaesthetics include decreased plasma viscosity8 and platelet aggregability,9 and increased prostacyclin production.10 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 (ßTG). 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.11


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
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 index12 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 <30% when packed red cells were transfused. Anaesthesia and fluid therapy were adjusted to maintain mean arterial pressure within 20% of pre-induction values and a heart rate of 60–100 beats min–1 with optimal filling. All patients received i.v. heparin 5–7000 units during surgery which was undertaken by the same surgical team of experienced consultants. Vein grafts were reversed and grafts were either femoral popliteal or femoral-distal in site. The technical success of the graft was confirmed by an intraoperative arteriogram.

At the end of surgery, neuromuscular block was reversed with neostigmine 2.5 mg and glycopyrronium 0.5 mg, the trachea extubated and the patient transferred to a high dependency area. Oxygen was given postoperatively by facemask to maintain PaO2 >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 ßTG 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. ßTG was estimated by non-proprietary ELISA in which rabbit antiserum to human ßTG 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.2

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.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
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.


View this table:
[in this window]
[in a new window]
 
Table 1 Patient characteristics. Table indicates number of patients except for age, weight, ABPI and Goldman risk score which are shown as mean (range) [SD]
 

View this table:
[in this window]
[in a new window]
 
Table 2 Perioperative management. Values shown are mean (range) [SD]
 

View this table:
[in this window]
[in a new window]
 
Table 3 Graft-related and systemic complications
 
Changes in circulating variables are shown in Table 4. There were no significant differences between groups in cortisol, IL-6, PAI-1 ag and ßTG concentrations. There were no significant changes from pre-induction values in both groups for cortisol, PAI-1 ag and ßTG values. However, IL-6 concentrations increased significantly in both groups after 4 h and remained significantly elevated until day 3 (P<0.05).


View this table:
[in this window]
[in a new window]
 
Table 4 Changes in mean (SEM) circulating cortisol (nmol litre–1), PAI-1 ag (ng ml–1), IL-6 (pg ml–1) and ßTG (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)
 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
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 ßTG values were increased (normal range 10–40 ng ml–1), reflecting the enhanced platelet activation found in elderly patients with vascular disease.13 14 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.15 Silent ischaemia and perioperative myocardial infarction occur frequently, with coronary artery disease the major cause of early and late mortality.16 17 Previous studies failed to show any benefit from regional anaesthesia in decreasing perioperative mortality and morbidity.13 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.18 19 Christopherson and colleagues found an impressive decrease in early occlusion of the graft in patients receiving epidural analgesia, from 22 to 4%.2 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 <2 h in the present study and >5 h in Christopherson et al.,2 but our low rates of graft occlusion are similar to those reported recently by other groups using GA and regional anaesthesia.20 21 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.22 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.4 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 analgesia23 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.24

The pattern of circulating cortisol response to arterial reconstructive surgery to the leg has been poorly defined.5 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.25 IL-6 induces acute phase protein synthesis in the liver.11 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,26 and this acute phase reactant also stimulates an increase in tissue factor which initiates the extrinsic pathway of coagulation.27 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.28

Patients with vascular disease have platelets with increased aggregability because of factors such as age, hypertension, and hyperlipidaemia.13 14 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.9 Thromboelastograph studies in major vascular surgery showed a decrease in platelet-fibrinogen interactions with epidural analgesia.29 These early stages of platelet activation are reversible, whereas degranulation occurs later and is irreversible. We used ßTG 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 ßTG data.30

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 ßTG responses.


    Acknowledgements
 
This study was funded by a grant from the Special Trustees of St. George’s Hospital.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Rivers SP, Scher LA, Sheehan E, Veith FJ. Epidural versus general anaesthesia for infrainguinal arterial reconstruction. J Vasc Surg 1991; 14: 764–70[ISI][Medline]

2 Christopherson 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: 422–34[ISI][Medline]

3 Bode RH, Lewis KP, Zarich SW, et al. Cardiac outcome after peripheral vascular surgery. Comparison of general and regional anesthesia. Anesthesiology 1996; 84: 3–13[ISI][Medline]

4 Rosenfeld 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: 435–43[ISI][Medline]

5 Naesh 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: 395–400

6 Cousins MJ, Wright CJ. Graft, muscle, skin blood flow after epidural block in vascular surgical procedures. Surg Gynaecol Obstet 1971; 133: 59–64

7 Korosue 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; 576–84

8 Orr JE, Lowe GDO, Nimmo WS, Watson R, Forbes CD. A haemorheological study of lignocaine. Br J Anaesth 1986; 58: 306–9[Abstract]

9 Borg T, Modig J. Potential anti-thrombotic effects of local anaesthetics due to their inhibition of platelet aggregation. Acta Anaesthesiol Scand 1985; 29: 739–42[ISI][Medline]

10 Casey LC, Armstrong MC, Fletcher JR, Ramwell PW. Lidocaine increases prostacyclin in the rat. Prostaglandins 1980; 19: 977–84[Medline]

11 Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J 1990; 265: 621–36[ISI][Medline]

12 Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977; 297: 845–50[Abstract]

13 Kieldsen 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: 151–5[ISI][Medline]

14 Zahavi J, Jones NAG, Leyton J, Dubiel M, Kakkar VV. Enhanced in vivo platelet "release reaction" in old healthy individuals. Thromb Res 1980; 17: 329–336[ISI][Medline]

15 Hertzer NR. Basic data concerning associated coronary disease in peripheral vascular patients. Ann Vasc Surg 1987; 1: 616–20[Medline]

16 Taylor LM, Yeager RA, Moneta GL, McConnell DB, Porter JM. The incidence of perioperative myocardial infarction in general vascular surgery. J Vasc Surg 1992; 15: 52–61[ISI][Medline]

17 Yeager 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: 598–606[ISI][Medline]

18 Bandyk 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: 286–96[ISI][Medline]

19 Whittemore AD. Failure of peripheral arterial reconstruction. Acta Chir Scand Suppl 1988; 550: 74–80

20 Dodds TM, Yeager MP, Walsh DB, Mandel D. General versus regional anaesthesia for peripheral vascular surgery. Anesth Analg 1997; 84: SCA86

21 Pierce ET, Pomposelli FB Jr, 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: 226–32[ISI][Medline]

22 Páramo 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: 713–6[ISI][Medline]

23 Parker SD, Breslow MJ, Frank SM et al. Catecholamine and cortisol responses to lower extremity revascularization: correlation with outcome variables. Crit Care Med 1995; 23: 1954–61[ISI][Medline]

24 Sharrock 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: 29–34[Abstract/Free Full Text]

25 Seekamp 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: 453–63[Abstract]

26 Heinrich 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: 374–9[ISI][Medline]

27 Cermak 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: 513–20[Abstract]

28 Rosenfeld BA, Nguyen ND, Sung I, Faraday N. Neuroendocrine stress hormones do not recreate the postoperative hypercoagulable state. Anesth Analg 1998; 86: 640–5[Abstract]

29 Tuman 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: 696–704[Abstract]

30 Lip 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: 425–31[Abstract/Free Full Text]





This Article
Abstract
Full Text (PDF)
E-Letters: Submit a response to the article
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (1)
Disclaimer
Request Permissions
Google Scholar
Articles by Bew, S. A.
Articles by Hall, G. M.
PubMed
PubMed Citation
Articles by Bew, S. A.
Articles by Hall, G. M.