fentanyl and the 0-endorphin, acth and glucoregulatory hormonal ...

Br.J. Anaesth. (1987), 59, 713-720

FENTANYL AND THE 0-ENDORPHIN, ACTH AND GLUCOREGULATORY HORMONAL RESPONSE TO SURGERY S. LACOUMENTA, T. H. YEO, J. M. BURRIN, S. R. BLOOM, J. L. PATERSON AND G. M. HALL The concomitant secretion of the peptides p1endorphin and adrenocorticotrophic hormone SUMMARY (ACTH) from the pituitary in response to acute The effect of the supplementation of nitrous stress was described initially in the rat by oxide-oxygen anaesthesia with either fentanyl Guillemin and colleagues (1977). Since then an 15ngkg~x or 0.5% halothane on the /*increase in circulating P-endorphin concentration endorphin, ACTH, glucoregulatory hormonal has been described in a variety of clinical and metabolic response to pelvic surgery was conditions such as labour (Akil et al., 1979; investigated. Fentanyl inhibited the increases in Fletcher, Thomas and Hill, 1980; Goland et al., circulating p-endorphin, ACTH, growth hor1981), hypoxia (Yanagida and Corssen, 1981), mone, cortisol and glucose concentrations found haemorrhagic shock (Shatney et al., 1985) and in the patients receiving halothane. Changes in surgery (Dubois et al., 1981; Mclntosh et al., circulating P-endorphin concentrations during 1985). However, the effects of different anaesthetic surgery probably reflect alterations in pituitary techniques on the changes in plasma P-endorphin secretion and appear to have no major metabolic concentration associated with surgery have re- effects. The suppression of pituitary secretion ceived little attention. Dubois and co-workers persisted for at least 4 h after the start of surgery. (1982) found that fentanyl 10-20 ug kg"1 attenuated the P-endorphin response to laparotomy during the surgical procedure compared with investigate the effects of fentanyl on the ppatients who received enflurane. On the contrary, endorphin, ACTH and glucoregulatory hormonal during cardiac surgery, Cork, Hameroff and Weiss response to surgery, together with the associated (1985) observed that fentanyl 50-100 ug kg"1 and changes in the circulating metabolites. halothane had similar effects on plasma Pendorphin values. In both these studies, however, the changes in P-endorphin concentration were PATIENTS AND METHODS described in isolation, with few data on the associated changes in the other pituitary hormones Fourteen healthy patients admitted for Fallopian commonly used to assess the endocrine response tubal surgery were investigated. The patients were allocated randomly to receive supplementation of to surgery. anaesthesia with either fentanyl 15 ug kg"1 or The purpose of the present study was to 0.5 % halothane. The nature of the study was explained to the patients and consent obtained for the collection of central venous blood. The study S. L-ACOUMENTA, M.D., F.F.A.R.C.S.I., J. L . PATERSON, B.SC., was approved by the local Ethical Committee. G. M. HALL, M.B.B.S., PH.D., C.BIOL., M.I. BIOL., F.F.A.R.C.S. All patients were premedicated with diazepam (Department of Anaesthetics); T. H. YEO, B.SC., PH.D., J. M . BORRIN, B.SC., PH.D., S. R. BLOOM, M.A., M.D., D.SC., 10 mg by mouth 2 h before surgery. On arrival in F.R.C.P. (Department of Medicine); Royal Postgraduate the anaesthetic room the duration of starvation Medical School, Hammersmith Hospital, Ducane Road, was determined and a central venous catheter London W12 0HS. Accepted for Publication: September 15, was inserted percutaneously from a vein in an 1986. antecubital fossa to permit blood sampling, and Correspondence to G. M. H.

BRITISH JOURNAL OF ANAESTHESIA

714

the administration of fluids. After the patient had rested for 10 min a control blood sample was collected, and the heart rate and arterial pressure were measured. Anaesthesia was induced in all patients with thiopentone 4-5 mg kg"1, the trachea was intubated with the aid of pancuronium 0.1 mg kg"1 and the lungs were ventilated with 70% nitrous oxide in oxygen. Patients in the fentanyl group received a bolus dose of fentanyl 15 ug kg"1 i.v. at the induction of anaesthesia, while patients in the control group received supplementation with 0.5% halothane throughout surgery. Ventilation was adjusted to maintain an end-tidal carbon dioxide concentration of 4.5%. Sodium chloride solution 150 mmol litre"1 was administered i.v. at a rate of 6 ml kg"1 h"1 during and 2 ml kg"1 h"1 after operation. Measured blood loss did not exceed 200 ml. Fifteen minutes after the induction of anaethesia another blood sample was collected and surgery commenced. Further samples were obtained 20 and 40 min and 1, 2, 4 and 6 h after the start of surgery. At the same time as the blood samples were collected, arterial pressure and heart rate were measured, ^ain was assessed 1, 2 and 4 h after operation using a linear analogue scale from 0 to 10 (Bond and Pilowsky, 1966), and analgesia was provided with papaveretum 15 mg i.m. on demand. On the 3rd day after operation the skinfold thicknesses of the patients were measured and the percentage of fat to body weight calculated (Dumin and Womersley, 1974). All blood samples were analysed in duplicate for glucose, lactate and glycerol concentrations, and haematocrit, by methods described previously (Hall et al., 1980). Plasma (i-endorphin, ACTH, cortisol, growth hormone (GH), insulin and glucagon values were determined in the preinduction sample and in samples collected 20 min, 1 h and 4 h after the commencement of surgery. Immunoreactive P-endorphin was measured by

a sensitive and specific radioimmunoassay (Immunouclear Corporation, Stillwater, Minnesota) which showed no cross-reactivity with p-lipotrophin or ACTH. The minimal detectable concentration was 5 pmol litre"1 and the interassay coefficient of variation was 11.7% at 19.3 pmol litre"1. Plasma ACTH was measured by a modification of the method of Rees and colleagues (1971), the interassay coefficient of variation was 10.9%. Cortisol (Seth and Brown, 1978), GH (W.H.O. reference preparation for human GH), insulin (Soeldner and Sloane, 1965) and glucagon (Bloom and Long, 1982) concentrations were measured by established radioimmunoassays. Intra- and interassay coefficients of variation were 6.5% and 9.8% for cortisol, 5.9% and 8.0% for GH, 5.6% and 8.9% for insulin and 6.6% and 7.8% for glucagon. The results for the hormones, metabolites and physiological variables are presented as mean values (± SEM). Statistical evaluation was undertaken using two-way and one-way analyses of variance, as appropriate. The ACTH values were subjected to logarithmic transformation before analysis because of a skewed distribution. Pain scores are presented as medians with ranges and were evaluated using the Wilcoxon two-sample test for unpaired data. RESULTS

Details of the patients studied are shown in table I. There was no difference between the groups with respect to age, body weight, body type, duration of preoperative starvation and duration of surgery. Plasma fi-endorphin and ACTH concentrations (fig. 1) In the halothane group, p-endorphin values increased significantly from 12.5 pmol litre"1 before the induction of anaesthesia to 49.9 pmol litre"1 after 20 min of surgery (P < 0.001).

TABLE I. Details ojpatients studied (nuan±SEM)

Age(yr) Weight (kg) Duration of starvation (h) Body fat (%) Duration of surgery (min)

Halothane group (n = 7)

Fentanyl group (n = 7)

29.5 ±1.7 56.5 ±2.2 13.4±1.0 30.1 ±0.6 101 ±10

29.1±1.6 49.4 + 2.7 12.5±0.8 30.5±0.4 108 ±4

FENTANYL AND HORMONAL RESPONSE TO SURGERY 20-

600 I

T

O g CD

~

I* a E a.

•

1510501000-1

60-i

1 ~ 750-1

2 40? if 01

715

O E 20-

S 250-

E

0J

-15 0 20

60

120

180

240

Time (min) FIG. 1. Mean ( ± SEM) plasma P-cndorphin (pmol litre"1) and plasma ACTH (ng litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min.

-15 0 20

60

120

180

240

Time (mm

FIG. 2. Mean ( ± S E M ) plasma cortisol (nmol litre"1) and plasma GH (mu. litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min.

This significant increase persisted until 4 h surgery. In the fentanyl group, there was no after the start of surgery (35.3 pmol litre"1) significant change in cortisol concentration (P < 0.01). In contrast, the fentanyl group throughout the study. There were significant showed no significant change from the pre- differences between the groups after 20 min, induction value either during or after surgery. 1 h a n d 4 h ( P < 0 . 0 0 1 ) . There were significant differences between Plasma GH concentration increased in the the groups after 20 min, 1 h and 4 h of halothane group from 1.6 mu. litre"1 to surgery (P < 0.01, P < 0.01 and P < 0.05, 13.4 mu. litre"1 after 1 h of surgery (P < 0.05) respectively). and was still increased after 4 h (10.9 mu. litre"1) Changes in plasma ACTH concentration were (P < 0.05). On the contrary, there was no similar in pattern to the changes noted in significant increase from the preinduction value P-endorphin values. In the halothane patients in the fentanyl patients. There was a significant there was a significant increase from 69 ng litre"1 difference between the groups after 1 h before induction of anaesthesia to 475 ng litre"1 (P < 0.05). after only 20 min of surgery (P < 0.001) and values were still significantly increased 4 h after Plasma insulin and glucagon concentrations (fig. 3) the start of surgery (237 ng litre"1) (P < 0.05). Plasma insulin concentrations did not change The ACTH response to surgery was abolished in significantly throughout the study in either group the fentanyl group and there was no significant of patients. increase from the preinduction value of Plasma glucagon values increased significantly 75 ng litre"1. Significant differences between the in the fentanyl group from 15.0 pmol litre"1 groups were found after 20 min and 1 h of surgery before induction of anaesthesia to 24.6 and (P < 0.001 and P < 0.05, respectively). 26.4 pmol litre"1 after 1 h and 4 h of surgery (P < 0.05 and P < 0.01), respectively. In the Plasma cortisol and GH concentrations (fig. 2) halothane group no significant change in plasma In the halothane patients plasma cortisol glucagon was found. There was no significant concentration increased two-fold, from the difference between the groups. preinduction value of 252 nmol litre"1 to 509 nmol litre"1 after 20 min of surgery Blood glucose concentration (fig. 4) In the patients receiving halothane, blood (P < 0.001). Cortisol values increased throughout the study to 877 nmol litre"1 (P < 0.001) after 4 h glucose concentration increased from 3.94 to

716

BRITISH JOURNAL OF ANAESTHESIA 5.03 mmol litre"1 after 20 min of surgery (P < 0.01) and a significant increase persisted for the remainder of the study (5.21 mol litre"1 at 6 h) (P < 0.05). There was no significant change in blood glucose concentration in the fentanyl group. There were significant differences between the groups 40 min, 1 h and 2 h after the start of surgery (P < 0.05).

40-.

& ~ 303 S - i t - 20H

100-

a a.

Blood lactate concentration (fig. 5) 20to

—

E a 100-

-15 0 20

60

180

120

240

Time (min) FIG. 3. Mean ( ± SEM) plasma insulin (mu. litre"1) and plasma glucagon (pmol litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min.

Blood lactate values increased significantly during surgery in both groups of patients. After 1 h of surgery blood lactate concentrations had increased from 0.79 to 1.03 nmol litre"1 in the halothane group (P < 0.01) and from 0.66 to 1.07 mmol litre"1 in the fentanyl group (P < 0.01). Blood lactate concentrations decreased after the completion of surgery. There was no significant difference between the groups.

76543-

0

1

2

4

6

Time (h) FIG. 4. Mean ( ± SEM) blood glucose (mmol litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery suited at 0 min. 1.4-1

r

1.2(D

|

1.0-

1> 0.800

ts = 0.6*

0.40

1

2 4 6 Time(h) FlG. 5. Mean ( ± SEM) blood lactate (mmol litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min.

FENTANYL AND HORMONAL RESPONSE TO SURGERY Plasma glycerol concentration (fig. 6)

There were no significant changes in plasma glycerol concentration during the study in the halothane group. In the fentanyl patients the induction of anaesthesia was associated with a decrease in glycerol concentration from 140 to 80 umol litre"1 (P < 0.05); thereafter, there were no significant changes from the preinduction value. There was no significant difference between the groups.

717

anaesthesia. During and after surgery there were no significant differences in heart rate and mean arterial pressure between the two groups. Haematocrit

Arterial pressure and heart rate (fig. 7)

The haematocrit decreased progressively throughout the study in both groups, from 40.7% to 36.9% ( P < 0.001) after 6 h in the halothane patients and from 39.4% to 36.3% (P < 0.001) after 6 h in the fentanyl patients. There was no significant difference between the groups.

In the halothane group the induction of anaesthesia was associated with significant increases in mean arterial pressure (P < 0.05) and heart rate (P < 0.01). In those patients who received fentanyl, however, there was a significant decrease in mean arterial pressure (P < 0.01), but no change in heart rate, after the induction of

Pain evaluation There was no significant difference in pain scores between the two groups 1, 2 and 4 h after the completion of surgery. In the halothane patients the scores were 4.2 (1.5-9.0), 3.4 (1.2-7.2) and 2.9 (0.7-8.3) and in the fentanyl

I

150-

•5 x x > -

i 01

50-

Time (h) FIG. 6. Mean ( ± S E M ) plasma glycerol (umol litre"1) concentrations in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min. 100-1 908070" c 100 n

'i 8 90H £1

!

8

I 70-1 Time (h) FIG. 7. Mean ( ± S E M ) heart rate (beat min"1) and arterial pressure (mm Hg) in patients receiving halothane (solid circles) or fentanyl (open circles). Surgery started at 0 min.

718

BRITISH JOURNAL OF ANAESTHESIA

patients 4.6 (2.5-8.3), 3.0 (1.4-6.3) and 2.8 (2.2-7.6), respectively. The total dose of papaveretum given in the 24 h after surgery was 36 + 3 mg in the halothane group and 42 ± 2 mg in the fentanyl group. DISCUSSION

The results show that fentanyl 15 ug kg"1 prevented the P-endorphin, ACTH, GH, cortisol and glucose responses to pelvic surgery. Furthermore, inhibition of the secretion of the pituitary hormones persisted until at least 4 h after the start of surgery, whereas Dubois and colleagues (1982) found an increase in plasma p-endorphin concentration on awakening from anaesthesia. However, a direct comparison between the two studies is not possible because these workers used laparotomy as the surgical model and did not standardize either the dose of fentanyl (10-20 ng kg"1) or the analysis of the data, which were grouped into "early" or "late" surgery. Our finding of a similar pattern of change in plasma P-endorphin and ACTH values is not surprising as both peptides are derived from the same precursor molecule proopiomelanocortin. Although it is commonly assumed that changes in plasma P-endorphin concentration reflect changes in pituitary secretion, a contribution from peripheral tissues cannot be excluded as P-endorphin is also found in small amounts in organs such as the pancreas, gut and kidney (Grossman and Clement-Jones, 1983). The effects of fentanyl in inhibiting the GH as well as the ACTH and p-endorphin responses to surgery suggest that the changes in the circulating proopiomelanocortinderived peptides found in this study are caused mainly by alterations in pituitary secretion. The attenuation of the glycaemic response during surgery in patients who received fentanyl 15 ng kg"1 was surprising, as previous work suggested that a much greater dose was required (Hall et al., 1978). The increase in blood glucose concentration found during surgery is usually considered to be mediated mainly by the action of catecholamines, particularly adrenaline, in increasing glucose production by the liver (Hall, 1985). Strong evidence for this mechanism is provided by the ability of extradural analgesia, to a level which blocks the innervation of the adrenal medulla, to prevent the glycaemic response to surgery in the presence of the continuing secretion of other catabolic hormones (Kehlet, 1982).

Although we did not measure plasma catecholamine concentrations in the present study, fentanyl had no effect on the intraoperative changes in the blood lactate, plasma glycerol, or plasma insulin concentrations, or on heart rate and arterial pressure—indices known to be influenced by increased sympathetic nervous system activity (Cooper et al., 1980; Walsh et al., 1982). It seems unlikely, therefore, that there was a major difference in catecholamine secretion between the groups and so other catabolic hormones may have been mediators, at least in part, of the glycaemic response. Suppression of cortisol secretion in the fentanyl patients probably contributed towards the decreased glycaemic response, as Lacoumenta, Paterson and co-workers (1986) have shown that etomidate, which inhibits cortisol synthesis by an effect on the adrenal cortex, decreased blood glucose concentration by around 0.5 mmol litre"1 during similar surgery. A role for GH in mediating the glycaemic response is less likely, as Metcalfe and colleagues (1981) and Sherwin and co-workers (1983) failed to demonstrate an acute effect of GH on blood glucose concentration even when the hormone was infused to give a 10-fold increase in circulating values. The administration of large doses of P-endorphin i.v. is known to influence blood glucose concentration, either indirectly by modifying glucagon and insulin secretion, or by a direct effect on the liver (Feldman et al., 1983; Radosevich et al., 1984). It is conceivable, therefore, that the increased plasma P-endorphin concentration found during surgery contributes to the glycaemic response. We have shown recently, however, that when human p-endorphin is infused during surgery to achieve plasma values 100-fold greater than those observed in the present study, there is only a transient increase in blood glucose concentration (Lacoumenta, Yeo et al., 1986). This short-lived effect was associated with an increase in plasma glucagon concentration. Thus, the increase in p-endorphin found during surgery probably does not contribute significantly to the mobilization of glucose. The increase in plasma glucagon concentration observed only in the fentanyl patients is of considerable interest. Morphine has a direct effect on the pancreas and stimulates the secretion of glucagon (Ipp, Dobbs and Unger, 1978; Ipp et al., 1980). It is probable, therefore, that fentanyl has a similar effect and while its use clinically is often associated with suppression of hypothalamic-

FENTANYL AND HORMONAL RESPONSE TO SURGERY pituitary function, it may, paradoxically, stimulate the secretion of glucagon and, hence, tend to increase blood glucose concentration. In the present study there was no difference in plasma glucagon concentration between the fentanyl and halothane groups, but an investigation of the effects of high-dose fentanyl ( > 50 ug kg"1) on pancreatic secretion may show a greater response. A recent Editorial (1985) has drawn attention to the relatively small part played by analgesia in determining the endocrine and metabolic response to surgery. Our hormonal data lend further support to this supposition. Pain scores were similar in both groups 2 h after completion of surgery (approximately 4 h after start of surgery), but P-endorphin and ACTH values were much higher in the halothane patients (35 v. 14 pmol litre"1, P < 0.05; 237 v. 129 ng litre-1, ns, respectively). Plasma cortisol, which is commonly used to assess the "stress" of surgery, showed an even greater difference than ACTH, with plasma concentrations of 877 nmol litre"1 in the halothane group and 331 nmol litre"1 in the fentanyl patients (P < 0.001). Thus the patient's perception of pain is a poor guide to the magnitude of the hormonal changes. In conclusion, fentanyl 15 ug kg"1 inhibited the p-endorphin, ACTH, GH, cortisol and glucose response to pelvic surgery. Changes in circulating P-endorphin during surgery probably reflect alterations in pituitary secretion and appear to have no major metabolic effects. Other physiological effects of the hyperendorphinaemia of surgery remain to be elucidated. ACKNOWLEDGEMENTS We thank Pamela Sealy for skilled secretarial assistance and Mr Raoul Margara for surgical co-operation. REFERENCES Akil, H., Watson, S. J., Barchas, J. D., and Li, C. H. (1979). fJ-Endorphin immunoreactivity in rat and human blood: radiommunoassay, comparative levels and physiological alterations. Life Sci., 24, 1659. Bloom, S. R., and Long, R. G. (1982). Radioimmunoaisay of Gut Regulatory Peptides, p. 82. Eastbourne, U K : W. B. Saunders. Bond, M. R., and Pilowsky, I. (1966). The subjective assessment of pain and its relationship to the administration of analgesics in patients with advanced carcinoma. J. Psychosomatic Res., 10, 203. Cooper, G. M., Paterson, J. L., Mashiter, K., and Hall, G. M. (1980). Beta-adrenergic blockade and the metabolic response to surgery. Br. J. Anaesth., 52, 1231.

719

Cork, R. C , Hameroff, S. R., and Weiss, J. L. (1985). Effects of halothane and fentanyl on plasma P-endorphin immunoreactivity during cardiac surgery. Anesth. Analg., 64, 677. Dubois, M., Pickar, D., Cohen, M., Gadde, P., Macnamara, T. E., and Bunney, W. E. (1982). Effects of fentanyl on the response of plasma beta-endorphin immunoreactivity to surgery. Ancsthesiology, 57, 468. Roth, Y. F., Macnamara, T., and Bunney, W. E. (1981). Surgical stress in humans is accompanied by an increase in plasma beta-endorphin immunoreactivity. Life Set., 29, 1249. Durnin, J. V. G. A., and Womerslcy, J. (1974). Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br. J. Nutr., 32, 77. Editorial (1985). Editorial: Analgesia and the metabolic response to surgery. Lancet, 1, 1018. Feldman, M., Kiser, R. S., Unger, R. H., and Li, C. H. (1983). Beta-endorphin and the endocrine pancreas. N. Engl. J. Med., 308, 349. Fletcher, J. E., Thomas, T. A., and Hill, R. G. (1980). P-Endorphin and parturition. Lancet, 1, 310. Goland, R. S., Wardlaw, S. L., Stark, R. I., and Frantz, A. G. (1981). Human plasma P-endorphin during pregnancy, labor and delivery. J. Clin. Endocrinol. Metab., 52, 74. Grossman, A., and Clement-Jones, V. (1983). Opiate receptors: enkephalins and endorphins. Clin. Endocrinol.

Metab., 12,31. Guillemin, R., Vargo, T., Rossier, J., Minick, S., Ling, N., River, C , Vale, W., and Bloom, F. (1977). P-Endorphin and adrenocorticotropin are secreted concomitantly by the pituitary gland. Science, 197, 1367. Hall, G. M. (1985). The anaesthetic modification of the endocrine and metabolic response to surgery. Arm. R. Coll. Surg. Engl., 67, 25. Lucke, J. N., Lovell, R., and Lister, D. (1980). Porcine malignant hyperthermia VII: Hepatic metabolism. Br. J. Anaesth., 52, 11. Young, C , Holdcroft, A., and Alaghband-Zadeh, J. (1978). Substrate mobilisation during surgery. A comparison between halothane and fentanyl anaesthesia. Anaesthesia, 33, 924. Ipp, E., Dobbs, R., and Unger, R. H. (1978). Morphine and P-endorphin influence the secretion of the endocrine pancreas. Nature (Lond.), 276, 190. Schusdziarra, V., Harris, V., and Unger, R. H. (1980). Morphine-induced hyperglycaemia: Role of insulin and glucagon. Endocrinology, 107, 461. Kehlet, H. (1982). The modifying effect of general and regional anaesthesia on the endocrine-metabolic response to surgery. Reg. Anaesth., 1, 538. Lacoumenta, S., Paterson, J. L., Myers, M. A., and Hall, G. M. (1986). Effects of cortisol suppression by etomidate on changes in circulating metabolites associated with pelvic surgery. Acta Anaesthesiol. Scand., 30, 101. Yeo, T., Burrin, J. M., and Hall, G. M. (1986). The role of p-endorphin in stress-induced hyperglycaemia. J. Endocrinol., 108, 302S. Mclntosh.T. K.,Bush,H. L., Palter, M., Hay, J. R.,Aun, F., Yeston, N. S., and Egdahl, R. H. (1985). Prolonged disruption of plasma p-endorphin dynamics following surgery. J. Surg. Res., 38, 210. Metcalfe, P., Johnson, D. G., Nosadini, R., Orskov, H., and Alberti, K. G. M. M. (1981). Metabolic effects of acute and

720 prolonged growth hormone excess in normal and insulindeficient man. Diabetologia, 20, 123. Radosevich, P. M., Williams, P. E., McRae, J. R., Lacy, W. W., Orth, D. N., and Abumrad, N. N. (1984). |3Endorphin inhibits glucose production in the conscious dog. J. Clm. Invest., 73, 1237. Rees, L. H., Cook, D. M., Kendall, J. W., Alien, C. F., Kramer, R. M., Ratcliffe, J. G., and Knight, R. A. (1971). A radioimmunoassay for rat plasma ACTH. Endocrinology, 89, 245. Scth, J., and Brown, L. M. (1978). A simple RIA for cortisol. Clin. Chem. Acta, 86, 109. Shatney, C. H., Cohen, R. M., Cohen, M. R., and Imagawa, D. K. (1985). Endogenous opioid activity in clinical haemorrhagic shock. Surg. Gynaecol. Obstet., 160, 547.

BRITISH JOURNAL OF ANAESTHESIA Sherwin, R. S., Shulman, G. A., Hendlcr, R., Walesky, M., Belous, A., and Tamborlane, W. (1983). Effect of growth hormone on oral glucose tolerance and circulating metabolic fuels in man. Diabetologia, 24, 155. Soeldner, J., and Sloane, D. (1965). Critical variables in the radioimmunoassay of serum insulin using the double antibody technique. Diabetes, 14, 771. Walsh, E. S., Paterson, J. L., Mashiter, K., and Hall, G. M. (1982). Effect of phentolamine on the metabolic response to gynaecological surgery. Br.J. Anaesth., 54, 517. Yanagida, H., and Corssen, G., (1981). Respiratory distress and beta-endorphin-like immunoreactivity in humans. Atesthcsiology, 55, 515.