Whether hormonal responses to surgical trauma should be controlled to obtain better surgical prognosis is controversial [1-3]. A relationship between the incidence of myocardial ischemia and sympathetic nervous overactivity has been reported [4,5]. Also a relationship between sympathetic nervous activity, pain, and myocardial ischemia has been demonstrated . Perhaps intra- and postoperative control of sympathetic or hormonal responses to surgical trauma would be beneficial for the prevention of intra- and postoperative cardiac complications. There are some anesthetic techniques which try to prevent or attenuate hormonal responses to surgical trauma [3,7-11]. Epidural anesthesia, using a local anesthetic, blocks afferent signals from the peripheral nociceptive fibers, and thereby suppresses hormonal responses to such stimuli. However, an epidural anesthetic technique using a local anesthetic may create the risk of systemic hypotension due to the sympathetic efferent block. Recently, epidural fentanyl (EP-F) has been used in obstetric and upper abdominal surgeries [12,13] to produce a segmental analgesic effect and reduce the volatile anesthetic requirements . Epidurally administered fentanyl attenuates hormonal, metabolic, and physiologic responses during the postoperative period . But the effect of EP-F on hormonal responses to surgical trauma is not well known, especially when compared with epidural anesthesia using a local anesthetic.
In this study, we prospectively compared the hemodynamic and hormonal responses to upper abdominal surgery, and the postoperative analgesic requirement among patients anesthetized with isoflurane combined with EP-F, epidural lidocaine (EP-L), and intravenous fentanyl (IV-F).
After approval of the protocol by our local ethics committee, 30 ASA physical status I to II patients undergoing subtotal gastrectomy in University of Tsukuba Hospital were included in this study. The patients ranged in age from 37 to 71 yr, and those with known systemic hypertension and/or diabetes mellitus were excluded from the study. After written, informed consent was obtained, patients were randomly allocated to one of the three groups, EP-F (n = 11), EP-L (n = 11), and IV-F (n = 8). Oral diazepam 10 mg was given as a premedication 90 min before patients entered the operating room. After a peripheral venous line was placed, an epidural catheter was placed via the T8-9 interspace approaching through the midline by the hanging drop technique. An epidural test dose was performed using a 3-mL solution containing 45 mg lidocaine and 15 micro gram epinephrine. Epidural anesthesia was accomplished using fentanyl 2 micro gram/kg in 10 mL saline in Group EP-F and 1.5% lidocaine without epinephrine, 10 mL, in Group EP-L. This was followed by a maintenance dose using half of the initial dose every hour. In Group IV-F, epidural anesthesia was not performed. Fentanyl 2 micro gram/kg was administered IV as an initial dose and fentanyl 1 micro gram centered dot kg-1 centered dot h-1 was infused as a maintenance dose.
After epidural anesthesia was initiated and its cephalad level determined 15 min later, general anesthesia was induced with thiamylal 4-5 mg/kg and the trachea was intubated after administration of vecuronium 0.2 mg/kg IV. Anesthesia was maintained by isoflurane (0.2%-2.5% by inspiratory concentration) and 67% nitrous oxide in oxygen. Patients were mechanically ventilated. Isoflurane concentration was adjusted by the attending anesthesiologists according to the hemodynamic and physical changes, and/or symptoms during surgery, to maintain an appropriate anesthetic depth. End-tidal CO (2) was maintained at 35-40 mm Hg, and an end-tidal isoflurane concentration was monitored and determined before and 1 h after the skin incision using a precalibrated gas monitor (Ultima Registered Trademark; Datex, Helsinki, Finland). Arterial pressure using an indwelling arterial line through the radial artery and lead II of the electrocardiogram were monitored continuously. Heart rate (HR) and mean arterial pressure (MAP) were recorded every 5 min during anesthesia.
During the study period, lactated Ringer's solution was infused at the rate of 15-25 mL centered dot kg-1 centered dot h-1. The surgery was started approximately 20-30 min after the induction of general anesthesia. Intraoperative urinary output was determined with a urine collector in all patients. No diuretics were supplemented during the surgery.
Arterial blood, 7 mL each, was withdrawn from the arterial line immediately before (BEFORE) and 1 h after the skin incision (AFTER) for determination of plasma epinephrine, norepinephrine, adrenocorticotropic hormone (ACTH), cortisol, plasma renin activity (PRA), and antidiuretic hormone (ADH) levels. The blood samples were collected in 10-mL EDTA tubes and centrifuged immediately, and their plasma was stored at -30 degrees C until determination.
Systemic hypotension, defined as MAP less than 60 mm Hg, was treated by IV ephedrine, 0.1 mg/kg. When MAP increased to more than 100 mm Hg, inspired isoflurane concentration was increased by 0.5%-2.5% and unless it decreased MAP to less than 100 mm Hg within the 10 min thereafter, diltiazem 0.05 mg/kg IV was given. Bradycardia, defined as HR less than 50 bpm, was treated by atropine 0.01 mg/kg IV.
Plasma epinephrine and norepinephrine concentrations were determined by high-pressure liquid chromatography. Plasma ACTH, cortisol, PRA, and ADH concentrations were measured by radioimmunoassay. The limits of detection and the interassay coefficients of variation were 1 pg/mL and 2.3% for epinephrine, 3 pg/mL and 2.3% for norepinephrine, 2 pg/mL and 2.6%-9.9% for ACTH, 0.3 pg/mL and 4.1%-5.8% for cortisol, 0.1 ng centered dot mL-1 centered dot h-1 and 5.1%-9.4% for PRA, and 0.2 pg/mL and 8.0%-8.3% for ADH.
Postoperatively, patients were tracheally extubated after an appropriate recovery from anesthesia was confirmed, and transferred to the surgical intensive care unit. Pentazocine, 30 mg intramuscularly (IM), or morphine, 2 mg epidurally, was given for pain relief on patient demand. The choice of analgesic was made by the patients' physicians. The interval from the end of surgery to the first use of an analgesic for wound pain relief, and the total analgesic uses during the first 48 h postoperatively were compared among the three groups.
Patients' demographic data, the duration of surgery, the infused crystalloid solution volume during the study period (from the start of transfusion to 1 h after the skin incision), the intraoperative urinary output, and the interval from the end of surgery to the first use of an analgesic for wound pain relief were compared among the three groups using analysis of variance factorial analysis. Hemodynamic data, end-tidal isoflurane concentration, and hormonal data were compared between BEFORE and AFTER in the three groups using one-way analysis of variance repeated measurement followed by a paired t-test with the Bonferroni's correction. The incidences of systemic hypotension and bradycardia, expressed as the number of patients who required IV ephedrine and atropine supplements, respectively, and the analgesic use during the first 48 h postoperatively for the wound pain relief were compared among the three groups using the Kruskal-Wallis test. P < 0.05 was considered statistically significant.
All patients underwent anesthesia and surgery uneventfully. There were no episodes of excessive blood loss, no blood transfusion requirements during surgery, and no cardiac complications postoperatively. There were no significant differences in age, height, weight, gender, the duration of surgery, the infused crystalloid volume, and the duration of stay in surgical intensive care unit after surgery among the three groups Table 1.
There were no significant differences in BEFORE values of MAP and HR among the three groups. In all groups, MAP at AFTER increased significantly compared with those at BEFORE (P < 0.05, Table 2), and the magnitude of MAP increase in Group IV-F was significantly larger than those in the other two groups (P < 0.05). Group EP-L required ephedrine supplements more often due to systemic hypotension than the other two groups (P < 0.05). HR increased significantly in Group IV-F, although it did not change in the other two groups during surgery. Three of the eight patients in Group IV-F (P < 0.05 vs Groups EP-F and EP-L) who developed bradycardia were treated successfully with IV atropine Table 3. Two patients in Group IV-F needed IV diltiazem 2.5 mg when blood pressure increased because treatment by increasing isoflurane concentration was not successful. The frequency of diltiazem use was not significant among the three groups. Group IV-F required significantly larger isoflurane concentrations in order to control blood pressure during surgery than was required by the other two groups (P < 0.01, Table 2). Urinary output during surgery was significantly larger in Group EP-F than in the other two groups (P < 0.05, Table 2).
There were no significant differences at BEFORE values among the three groups in all the hormones studied. In all groups, epinephrine, norepinephrine, ACTH, cortisol, and ADH levels increased significantly at AFTER compared with their BEFORE values, respectively (P < 0.05, Table 4). The PRA level did not change after the start of surgery in any group. The magnitude of increase in epinephrine level in Group EP-F was larger than those in the other two groups (P < 0.05, Table 4). The magnitudes of the increases in plasma ADH level in Groups EP-L and IV-F were significantly larger than that in Group EP-F, and those of plasma ACTH and cortisol levels in Groups EP-F and IV-F were significantly larger than those in Group EP-L, respectively (P < 0.05, Table 4).
The mean time intervals from the end of surgery to the first use of postoperative analgesics for the wound pain relief was not significantly different among the three groups Table 5. But the frequency of analgesic use during the first 48 h postoperatively was significantly less in the Group EP-F than the other two groups (P < 0.01, Table 5).
Our results showed that Groups EP-L and EP-F attenuated the hormonal responses to the upper abdominal surgical trauma more than Group IV-F. Group EP-F had less incidence of hypotension than Group EP-L, which may be due to an analgesic effect of epidural fentanyl without autonomic dennervation.
Inagaki et al.  reported that the same dose of EP-F produces a more potent analgesic effect than IV-F. They demonstrated that the significantly larger halothane minimum alveolar anesthetic concentration reducing effect of EP-F versus IV-F when more than 1 micro gram/kg fentanyl was given, may be attributed to the analgesic action of fentanyl at the spinal cord. They also reported that EP-F 2 micro gram/kg reduced the minimum alveolar anesthetic concentration of halothane by 58% . EP-F 250 micro gram significantly reduced the isoflurane requirement at 1 h after the start of thoracic surgery compared with the same dose of IV-F in humans.* In this study, the same dose of fentanyl was given epidurally and IV in Groups EP-F and IV-F, respectively. EP-F analgesia decreases the postoperative pain score in obstetric patients [14-16], and epidural morphine decreases postoperative hypertension by attenuating sympathetic nervous system hyperactivity . On the other hand, another report shows the same efficacy on the postoperative pain score of the same dose of EP-F and IV-F . The results of this study suggest that EP-F acts mainly at the spinal cord level rather than at the supraspinal or systemic level [7,19], and provides more potent analgesic effect than IV-F, not only intraoperatively but also postoperatively, when the same dosing protocol is used.
*Grant GJ, Ramanathan S, Turndorf H. Epidural fentanyl reduces isoflurane requirements during thoracotomy [abstract]. Anesthesiology 1989;71:A668.
Hormonal responses to surgical trauma are suppressed by EP-L [20,21], although EP-L produces a pronounced hemodynamic suppression and may induce hypoperfusion to organs such as the kidneys and the intestine . The plasma epinephrine level reflects adrenomedullary secretion, whereas the plasma norepinephrine level is used as an index of sympathetic nervous system activation , and their changes provide direct evidence of adrenergic activation during surgery in patients . Also, perioperative stimulation of the hypothalamic-pituitary-adrenal system is reflected by the increases of plasma cortisol and ACTH levels . In this study arterial blood for determining hormonal levels was collected immediately before the skin incision and 1 h thereafter, during which interval the peak of plasma hormonal responses to surgery were obtained [3,25,26]. Moreover, all the patients underwent gastrectomy, during which the stomach was resected in the first hour, which means that the patients were exposed to the strongest surgical stimuli during that period. And some report that the increased plasma hormonal levels 1 h after the skin incision remained increased until the postoperative period [23,27,28]. In this study, Group EP-L attenuated the magnitude of increases in epinephrine and norepinephrine levels during surgery compared with Group IV-F, and Group EP-F attenuated that in epinephrine compared with Group IV-F. Two patients in Group IV-F required diltiazem supplements during the study. Diltiazem increases plasma catecholamine levels . However, in our study, patients taking diltiazem did not have increased catecholamine levels. The results of this study may indicate that EP-F is more potent than IV-F in terms of analgesic effect, but EP-F is not enough to control sympathetic nervous activity during surgery at the fentanyl doses used in this study. Udelsman et al.  also reported that adequate ACTH secretion and moderate cortisol hypersecretion occurred during the perioperative period as a result of the existence of nonhypothalamic corticotropin-releasing hormone. Barton et al.  observed that within 2 h of injury, patients with moderate injury had plasma cortisol increases in proportion to the severity of the injury. The magnitudes of the increases in plasma cortisol and ACTH levels during surgery were less in Group EP-F than in Groups EP-L and IV-F. These results suggest that the increases of plasma cortisol and ACTH are good indicators for surgical trauma [27,31], and may be modulated by anesthetic techniques.
Renin is secreted in response to changes in perfusion pressure at the afferent arteriole and in solute delivery to the distal tubulus . Renal sympathetic nerve stimulation, or an infusion of catecholamine, increases renin secretion [32,33]. In Groups EP-F and IV-F, PRA levels did not change after the start of surgery, which may suggest that renal perfusion was maintained . The absence of profound hypotension and the less significant increases in plasma ADH levels during surgery suggests that Group EP-F preserved perfusion to the kidneys better than the other groups. In this study, urinary output during the surgery was significantly more in Group EP-F than the other two groups, which may support the difference of the effect of the three analgesic techniques on plasma ADH changes during surgery.
Although the analgesic potency of 2 mg epidural morphine may differ from 15 mg IM pentazocine, Group EP-F patients required less frequent administration of epidural morphine and IM pentazocine for postoperative pain relief than the other two groups Table 5. This mechanism may be related to the EP-F-induced preemptive analgesic effect [34,35]. But still it is not clear why Group EP-L patients required more frequent analgesics postoperatively than Group EP-F in spite of the intraoperative attenuation of plasma hormonal increases.
We did not use a double-blind approach to analyzing the intraoperative anesthetic management among the three groups. But the study protocol, including the doses and time intervals of epidural anesthesia, the use of vasoactive drugs, and control of isoflurane concentration, was identical in the three groups. Furthermore, the attending physicians were blinded to the intraoperative anesthetic technique during the postoperative period. We therefore believe that the lack of a complete double-blind would impose only a small influence, if any, on the significant findings in this study.
In summary, Group EP-F attenuated the hormonal responses to the major abdominal surgery similar to that of Group EP-L and was not accompanied by profound hypotension. Furthermore, Group EP-L had reduced postoperative analgesic requirements when compared with Groups EP-L and IV-F. But it is not clear whether the intraoperative surgical stress response itself may be related to the postoperative prognosis because larger sample protocols are needed to investigate such a correlation. In conclusion, EP-F analgesia may have more advantages than EP-L in patients undergoing upper abdominal surgery when combined with general anesthesia.
1. Kehlet H. Surgical stress. The role of pain and analgesia. Br J Anaesth 1989;63:189-95.
2. Kehlet H. The surgical stress response: should it be prevented? Can J Surg 1991;34:565-7.
3. Weissman C. The metabolic response to stress: an overview and update. Anesthesiology 1990;73:308-27.
4. Longhurst JC. Coronary arteriolar vasoconstriction in myocardial ischaemia: reflexes, sympathetic nervous system, catecholamines. Eur Heart J 1990;11(Suppl B):43-52.
5. Schomig A, Richardt G. Cardiac sympathetic activity in myocardial ischemia: release and effects of noradrenaline. Basic Res Cardiol 1990;1:9-30.
6. Thamer V, Deussen A, Schipke JD, et al. Pain and myocardial ischemia: the role of sympathetic activation. Basic Res Cardiol 1990;1:253-66.
7. Salomaki TE, Leppaluoto J, Laitinen J, et al. Epidural versus intravenous fentanyl for reducing hormonal, metabolic, and physiologic responses after thoracotomy. Anesthesiology 1993;79:672-9.
8. Hall GM. The anaesthetic modification of the endocrine and metabolic response to surgery. Ann R Coll Sur Engl 1985;67:25-9.
9. Yeager MP, Glass DD, Neff RK, Brinck-Johnsen T. Epidural anesthesia and analgesia in high-risk surgical patients. Anesthesiology 1987;66:729-36.
10. Kehlet H. Epidural analgesia and the endocrine-metabolic response to surgery. Update and perspectives. Acta Anaesthesiol Scand 1984;26:56-8.
11. Weissman C, Hollinger I. Modifying systemic responses with anesthetic techniques. Anesthesiol Clin North Am 1988;6:221-36.
12. Inagaki Y, Mashimo T, Yoshiya I. Segmental analgesic effect and reduction of halothane MAC from epidural fentanyl in humans. Anesth Analg 1992;74:856-64.
13. Silbert BS. Epidural opioids as anaesthesia for extracorporeal shock wave lithotripsy in two patients with cardiac disease. Can J Anaesth 1988;35:624-7.
14. Elliot RD, Stockwell M. Continuous infusion epidural analgesia or obsterics: bupivacaine versus bupivacaine-fentanyl mixture. Can J Anaesth 1991;38:303-10.
15. Naulty JS, Datta S, Ostheimer GW, et al. Epidural fentanyl for postcesarean delivery pain management. Anesthesiology 1985;63:694-8.
16. Chestnut DH, Owen CL, Bates JN, et al. Continuous infusion epidural analgesia during labor: a randomized, double-blind comparison of 0.0625% bupivacaine/0.0002% fentanyl versus 0.125% bupivacaine. Anesthesiology 1988;68:754-9.
17. Breslow MJ, Jordan DA, Christopherson R, et al. Epidural morphine decreases postoperative hypertension by attenuating sympathetic nervous system hyperactivity. JAMA 1989;261:3577-81.
18. Ellis DJ, Millar WL, Reisner LS. A randomized double-blind comparison of epidural versus intravenous fentanyl infusion for analgesia after cesarean section. Anesthesiology 1990;72:981-6.
19. Bormann B, Weidler B, Dennhardt R, et al. Influence of epidural fentanyl on stress-induced elevation of plasma vasopressin (ADH) after surgery. Anesth Analg 1983;62:727-32.
20. Kehlet H. Epidural analgesia and the endocrine-metabolic response to surgery. Update and perspectives. Acta Anaesthesiol Scand 1984;28:125-7.
21. Stelzner J, Reinhart K, Fohring U, et al. The effect of thoracic peridural analgesia on the cortisol and glucose response in surgery of the abdominal aorta. Reg Anaesth (German) 1988;11:16-20.
22. Lundberg J, Lundberg D, Norgren L, et al. Intestinal hemodynamics during laparotomy: effects of thoracic epidural anesthesia and dopamine in humans. Anesth Analg 1990;71:9-15.
23. Halter JB, Pflug AE, Porte D. Mechanism of plasma catecholamine increases during surgical stress in man. J Clin Endocrinol Metab 1977;45:936-44.
24. Udelsman R, Norton JA, Jelenich SE, et al. Response of the hypothalamic-pituitary-adrenal and renin-angiotensin axes and the sympathetic system during controlled surgical and anesthetic stress. J Clin Endocrinol Metab 1987;64:986-94.
25. Lachuer J, Gaillet S, Barbagli B, et al. Differential early time course activation of the brainstem catecholaminergic groups in response to various stresses. Neuroendocrinology 1991;53:589-96.
26. Hamberger B, Jarnberg P-O. Plasma catecholamines during surgical stress: differences between neurolept and enflurane anaesthesia. Acta Anaesthesiol Scand 1983;27:307-10.
27. Newsome NH, Rose JC. The response of human adrenocorticotrophic hormone and growth hormone to surgical stress. J Clin Endocrinol 1971;33:481-7.
28. Enquist A, Gog-Moller F, Christiansen C, et al. Influence of epidural analgesia on the catecholamine and cyclic AMP responses to surgery. Acta Anaesthesiol Scand 1980;24:17-21.
29. Morooka H, Gotoh S, Makita T, et al. Adrenal catecholamine releases during hypotension with continuous administration of calcium entry blockers and nitroglycerin. Jpn J Anesthesiol 1992;41:1070-5.
30. Barton RN, Stoner HB, Watson SM. Relationship among plasma cortisol, adrenocorticotropin and severity of injury in recently injured patients. J Trauma 1987;27:384-92.
31. Barton RN, Passingham BJ. Effect of binding to plasma proteins on the integration of plasma cortisol concentrations after accidental injury. Clin Sci 1981;61:399-405.
32. Landsberg L, Young JB. 10. Cathecolamines and the adrenal medulla. In: Wilson JD, Foster DW, eds. Williams textbook of endocrinology. 8th ed. Philadelphia: WB Saunders, 1992;621-705.
33. Stella A, Zanchetti A. Effects of renal denervation on renin release in response to tilting and furosemide. Am J Physiol 1977;232:H500-7.
34. Katz J, Kavanagh BP, Sandler AN, et al. Preemptive analgesia. Clinical evidence of neuroplasticity contributing to postoperative pain. Anesthesiology 1992;77:439-46.
© 1995 International Anesthesia Research Society
35. Shir Y, Raja SN, Frank SM. The effect of epidural versus general anesthesia on postoperative pain and analgesic requirements in patients undergoing radical prostatectomy. Anesthesiology 1994;80:49-56.