- Email: [email protected]
Effects of extradural anaesthesia on interleukin-6 and acute phase response to surgery
British Journal of Anaesthesia 1994; 72: 272-279
Effects of extradural anaesthesia on interleukin-6 and acute phase response to surgery C. M. MOORE, J. P. DESBOROUGH, H. POWELL, J. M. BURRIN AND G. M. HALL
SUMMARY
KEY WORDS Hormones growth. Hormones- glucocorticoid. protein acute phase. Complications: trauma.
Metabolism,
The main causative factors in evoking the metabolic response to surgery are afferent neuronal input from the operative site and release of cytokines from tissue damaged by surgical trauma. The major cytokine synthesized after routine surgery is interleukin-6 (IL-6) and this polypeptide stimulates acute phase protein synthesis in the liver and affects lymphocyte function and haematopoiesis also [1]. There are data from in vitro and animal studies to show that cytokines, including IL-6, may stimulate pituitary hormonal secretion [2^1]. This has been interpreted as evidence for important links between inflammatory mediators and endocrine control and the
PATIENTS AND METHODS
We studied 16 healthy females, aged 20-50 yr, undergoing total abdominal hysterectomy for benign disease; the patients were allocated randomly to receive either general anaesthesia and extensive extradural block with local anaesthetic from T4 to S5 dermatomal segments (extradural group) or general anaesthesia alone (control group). All patients gave written informed consent and local Ethics Committee approval (registration No. 90/3497) was obtained. Patients with metabolic or hormonal disorders were excluded. After overnight fasting, each patient received papaveretum 0.3 mg kg""1 i.m. and hyoscine 1 0.006 mg kg" as premedication, 90 min before operation. On arrival in the anaesthetic room, electrocardiograph (ECG), arterial oxygen saturation (SpO2) and non-invasive arterial pressure monitoring were commenced. A central vein was catheterized via the antecubital fossa for administration of fluids and collection of blood samples and the position of the catheter confirmed by haemodynamic measurement. Mean arterial pressure (MAP) and heart rate were
C. M. MOORE, M.B., CH.B, F.R.C.A., J. P. DESBOROUGH*, M.B., CH.B., F.R.C.A., H . POWELLt, M.B., B.S., F.R.C.A., J. M . BURRIN^:, B.SC, PH.D., M.R.C.PATH., G. M . HALL*, M.B., B.S., PH.D.,
F.I.BIOL., F.R.C.A., Departments of Anaesthetics and Medicine,
Royal Postgraduate Medical School, Hammersmith Hospital, London W12 0NN. Accepted for Publication: October 21, 1993. Present addresses: * Department of Anaesthesia, St George's Hospital Medical School, Cranmer Terrace, London SW17 ORE. t Department of Anaesthesia, Freeman Hospital, High Heaton, Newcastle upon Tyne NE7 7DN. % Department of Clinical Biochemistry, London Hospital Medical College, Turner Street, London El 2AD. Correspondence to G.M.H. at St George's Hospital.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
Serum concentrations of the cytokine, interleukin-6 (IL-6), increase after surgical trauma. IL-6 mediates the synthesis of acute phase proteins and stimulates secretion of pituitary hormones. We have examined the time course of circulating IL-6, and cortisol and growth hormone responses in patients undergoing hysterectomy to determine if IL-6 contributes to the early pituitary hormone changes found during surgery. One group (n = 8) received a standardized general anaesthetic while the remaining patients (n = 8) received extradural analgesia to T4-S5 in addition to a similar general anaesthetic. In the general anaesthesia group, there was a significant increase in serum cortisol and growth hormone concentrations before any changes in IL-6 were detected. Furthermore, in the extradural group, in whom these hormonal responses were attenuated, circulating IL-6 concentrations did not differ significantly from the general anaesthesia group. There were no significant differences between the groups in the acute phase response, as measured by circulating concentrations of C-reactive protein and zinc, but the expected effects of extradural block on circulating metabolites and white cell count were demonstrated. We conclude that IL-6 is unlikely to contribute to the initial increases in secretion of pituitary hormones found during surgery, but a later effect of the cytokine on endocrine responses cannot be excluded. (Br. J. Anaesth. 1994; 72: 272-279)
term "endocrine immunology" has been used to describe the association. However, there are few clinical data, at present, to support this proposal. The purpose of the present study was to determine if IL-6 contributes to the early pituitary hormonal changes observed in the perioperative period. We have studied the time course of the endocrine and cytokine changes in patients undergoing pelvic surgery in whom it is known that catabolic hormonal secretion can be obtunded for several hours by extradural analgesia. Preliminary results of this study have been presented previously [5].
INTERLEUKIN-6 RESPONSE TO SURGERY
Blood sample collection
Venous blood samples were collected at the following times: a baseline sample 45 min before surgical incision, at the start of surgery (0 min), and at 1, 2,4, 6, 12, 24 and 48 h after the start of surgery. For serum assays, blood was placed in plain tubes and centrifuged. For plasma samples, blood was placed in lithium heparin tubes and centrifuged. The supernatant samples were collected and stored at - 2 0 °C until assay. All samples were measured in duplicate. Serum concentrations of C-reactive protein (CRP) and albumin, and plasma zinc concentrations were measured in samples obtained at baseline and 0, 6, 12, 24 and 48 h. Blood was placed in EDTA tubes for differential white cell counts from the baseline sample and at 0, 1, 2, 6, 12, 24 and 48 h. Serum cortisol, GH, insulin and IL-6 concentrations were analysed in the baseline sample and at 0, 1, 2, 4, 6, 12, 24 and 48 h. Blood glucose and lactate, and plasma non-esterified fatty acids (NEFA) concentrations were measured in all samples, and heart rate and MAP were measured at all times. Investigators were unaware of patient identity during sample analysis. Haematological and biochemical measurements
Serum concentration of IL-6 was measured by a two-site enzyme immunometric assay (Quantikine, RD systems, British Bio-technology Ltd, Abingdon, Oxon). Intra- and interassay precision were 7 % and
9%, respectively, at an IL-6 concentration of 93 pg ml"1. The lower limit of sensitivity of the assay was 3.5 pg ml"1. Serum concentrations of CRP and albumin were measured using latex-enhanced immunoturbidimetry [6,7]. The intra-assay coefficient of variation was 5 % for both assays at all concentrations, and all samples were measured in the same assay. Plasma zinc concentration was measured using a simple two-step colorimetric method with trichloracetic acid deproteinization and the specific chelating agent 2-(5-bromo-2-pyridylazo)-5-(iV-propyl-iVsulphopropylamino)-phenol (Wako, Alpha Laboratories, Eastleigh, Hants, U.K.) [8]. Results obtained from this assay correlated well (r = 0.963) with those by atomic absorption spectrophotometry. All samples were measured in the same assay and the within-assay coefficient of variation was 4.5 % at a zinc concentration of 20 umol litre"1. Packed cell volume was measured using a microcapillary haematocrit centrifuge with correction for trapped plasma. A full blood count was performed on EDTA blood using a Sysmek E-5000 (TOA Medical Electronics (U.K.) Ltd) which also performed a differential leucocyte count presenting these data as neutrophil percentage and absolute number, lymphocyte percentage and absolute number, and (monocytes, eosinophils and basophils) percentage and absolute number. Serum cortisol concentration was measured using the Farmos Diagnostica (Orion Corporation, Turku, Finland) 125I-direct-RIA kit. This assay uses a cortisol antiserum which has negligible crossreactivity with other endogenous corticosteroid and cortisol metabolites. Intra- and interassay coefficients of variation were 3.2% and 4.5% at 280 mmol litre"1 and 3.9 % and 6.2 % at 680 nmol litre"1, respectively. Serum growth hormone (GH) concentration was measured using a solid phase two-site immunoradiometric assay (Netria Labs, London, U.K.). The lower limit of sensitivity was 1.0 mu. litre"1. Within-batch variation at 4.2 mu. litre"1 was 9% and at 19.5 mu. litre"1 8%. Betweenbatch variation at 3.9 mu. litre"1 was 11% and at 20.7 mu. litre"1 10%. Serum insulin concentration was measured by a sensitive and specific double antibody radioimmunoassay described previously [9]. Intra- and interassay precision at 8.6 mu. litre"1 were 5.6% and 8.9%, respectively. Blood glucose and lactate values, and plasma NEFA concentration were determined enzymatically by standard methods [10]. Statistical analyses Parametric data are expressed as mean (SEM) and analysed using two-way analysis of variance (ANOVA) with Dunnett's test for within-group changes with respect to the baseline sample [11]. Between-group changes were analysed by ANOVA. P < 0.05 was accepted as significant. For GH and IL-6, concentrations less than the sensitivity of the assay were assigned the value of that limit. GH and IL-6 data were not normally distributed and therefore expressed as median (range). Differences within each group with respect to the baseline sample were
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
recorded and a baseline blood sample obtained: 14 ml kg"1 of sodium chloride solution 150mmol litre"1 was infused in all patients. All studies commenced between 08:00 and 10:00. In the extradural group, during fluid loading, a catheter was inserted into the extradural space at the L2-3 or L3—4 intervertebral level. The patient was then placed supine, a test dose of 0.5 % plain bupivacaine 3 ml given, followed 5 min later by increments of 5 ml of 0.5 % plain bupivacaine to produce bilateral analgesia to pin prick and loss of sensation to ethyl chloride spray from S5 to T4 or a higher dermatomal level. The block was maintained by an extradural infusion of 0.5 % plain bupivacaine at a rate of 6—8 ml h"1; this was discontinued 6 h after surgical incision. The extent of the block was tested before induction of anaesthesia and at 2, 4 and 6 h after surgical incision. MAP was maintained at or above 60 mm Hg by i.v. bolus doses of methoxamine 1 mg as required during operation in the extradural group. In both groups of patients, anaesthesia was induced with a dose of thiopentone sufficient to obtund the eyelash reflex. Vecuronium 0.1 mgkg"1 was given to facilitate tracheal intubation and the lungs ventilated with 0-1 % halothane and 66 % nitrous oxide in oxygen. End-tidal carbon dioxide partial pressure was maintained at 4—5 kPa. I.v. fluids (sodium chloride solution 150 mmol litre"1) were given at a rate of 6 ml kg"1 h"1 during operation and at 2 ml kg"1 h"1 after operation. Postoperative analgesia in the control group was provided by papaveretum 5 mg i.v. on demand.
273
BRITISH JOURNAL OF ANAESTHESIA
274
analysed by the Wilcoxon's matched pairs signed rank test, and between-group changes by the Wilcoxon rank sum test. RESULTS
TABLE I. Patient data {mean {SEM or range))
Age (yr) Weight (kg) Duration of surgery (min)
Control group (« = 8)
Extradural group (« = 7)
41.9(36-45) 66.9 (3.8) 79.9(11.5)
43.3 (39-47) 62.7 (2 2) 97.0(12.9)
1000 n
CRP (fig. 2) In both groups there was a significant increase in CRP 24 h after surgery, from 2.3 (0.5) to 79.7 (11.1) mg litre"1 (P < 0.01) in the control group and from 4.4(1.0) to 67.2 (8.5) mg litre"1 (P < 0.01) in the extradural group. After 48 h, CRP had increased further, to 130.8 (16.2) and 122.0 (15.7) mg litre"1 in the control and extradural groups, respectively. There were no significant differences in CRP concentrations between the two groups. Plasma zinc and serum albumin (fig. 3)
7. 100 -
£ a
10 -
6 8 10 Time (h)
12 24 48
FIG. 1. Median (range) serum concentration of interleukin-6 (IL-6) in the control (O) and extradural (#) groups. Interleukin-6 values are plotted on a logarithmic scale.
150 -i
100 CT>
£
a. 50 u
0
J 1
6
i
8 Time (h)
Circulating white blood cells (fig. 4)
i,
10
12
In both groups of patients, plasma zinc concentration decreased significantly after surgery, from 12.6(0.5) to 6.6 (0.5) umol litre"1 (P < 0.05) and from 11.8 (0.4) to 6.9 (0.6) umol litre"1 (P < 0.05) at 6 h in the control and extradural groups, respectively. Zinc concentration remained significantly decreased at 48 h compared with the concentration before operation (P < 0.05). There were no significant differences between the two groups. Serum albumin concentration decreased after fluid loading and induction of anaesthesia from 36.1 (0.5) to 31.7 (0.7) g litre"1 (P < 0.01) in the control group. After surgery, albumin concentrations in this group were stable until 12 h, but decreased significantly compared with baseline sample, with concentrations of 29.4 (1.1) and 30.1 (0.7) g litre"1 at 24 and 48 h, respectively. In the extradural group, there was no significant change in serum albumin concentrations during the study and there were no significant differences between the two groups. The zinc: albumin ratio was calculated to examine changes in divalent cation that were independent of concurrent alterations in its major binding protein. The results are shown as a percentage change from the value at the start of surgery. This ratio decreased after surgery in both groups, with no significant difference between groups.
24 48
FIG. 2. Mean (SEM) serum concentration of C-reactive protein (CRP) in the control (O) and extradural ( # ) groups.
The white cell count was increased significantly in both groups after surgery. In the control group the count increased from 6.1 (0.7) to 13.4 (1.3) x 109 litre"1 at 2 h (P < 0.01) and increased
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
There were no significant differences between groups in age, weight or duration of surgery (table I). One patient in the extradural group was withdrawn from the study because she had marked systolic hypotension because of intraoperative bleeding and extradural block was discontinued. The dermatomal levels of the extradural block were T1-T3-S5 at 2 h (w = 6), T2-T5-S5 at 4 h (« = 7) and T4-T6-S5 (« = 7) at 6 h. Only two patients in the extradural group needed methoxamine, both during operation (3 and 5 mg).
IL-6 (fig. 1) In the control group, serum IL-6 concentration increased from a baseline value of 3.5 to 28.4 (7.8-60.0) pg ml"1,4 h after the start of surgery (P < 0.01). This increase was maintained, with IL-6 values remaining significantly greater than baseline concentrations at 6 and 12 h (P < 0.01), 24 h (P < 0.02) and 48 h (P < 0.05). In contrast, in the extradural group, IL-6 concentrations were not increased significantly compared with the baseline* value of 3.5 pg ml"1 until 12 h, when concentrations reached 25.3 (3.5-218.0) pg ml"1 (P < 0.05). A significant increase in IL-6 in the extradural group was then sustained for up to 48 h (P < 0.05). Although the IL-6 response was delayed in the extradural group, there was no significant difference in IL-6 concentrations between the two groups during the study.
INTERLEUKIN-6 RESPONSE TO SURGERY
275
15 -i
20 -i
-
15 -
£ 10 o
° 10 -
E
a. 5 -
0
o m
J
0 20
110 -i |
90 -
i
80 -
JO
70 -
I5
60 -
2
15
— 10 -
c
!E a
N
0
50 -> 0
2
4
6 8 10 Time (h)
12
24
J
20 -i
48
FIG. 3. Mean (SEM) plasma zinc concentration and zinc: albumin ratio, as percentage of baseline sample, in the control (O) and extradural ( # ) groups. CD
9
S io H a
1
further to 16.4 (1.7) x 10 litre" after 6 h. In the extradural group, the increase in cell count was slower, from 5.8 (0.6) to only 9.0 (1.0) x 109 litre"1 at 2 h (P < 0.05) with a further increase to 14.6 (1.2) x 109 litre"1 after 6 h. The white cell count was significantly smaller in the extradural group compared with the control group at 2 h (P < 0.05). Neutrophil leucocytosis occurred in both groups, with a change from 3.5 (0.4) to 10.5 (1.0) x 109 litre"1 (P < 0.01) in the control group, but from 3.4 (0.5) to only 6.2 (1.0) x 109 litre"1 in the extradural group at 2 h. This difference was statistically significant (P < 0.05). In contrast, the lymphocyte count in both groups decreased in the postoperative period, although the decrease reached statistical significance only in the control group at 24 h (P < 0.05). There were no significant differences between groups in lymphocyte count during the study. Hormones
Cortisol (fig. 5). In the control group, serum cortisol concentration increased from 247 (28) to 725 (60) nmol litre"1 (P < 0.05) after 1 h of surgery and continued to increase to 1226 (154) nmol litre"1 (P < 0.01) at 4 h. Concentrations then declined progressively to baseline values after 48 h. In contrast, however, in the extradural group, serum cortisol concentration had not changed significantly after 1 h of surgery compared with the baseline value. Four hours after surgery started, cortisol concentrations had increased to only 672
E 0 -1 I
l
l
6 8 10 Time(h)
12
24
48
FIG. 4. Mean (SEM) total circulating white blood cells (WBC), neutrophils and lymphocytes in the control (O) and extradural ( • ) groups. *P < 0.05, **P < 0.01 between groups. 1500 -i
CD
£ 1000 "5 E _c § 500 eo
o
0
J
6 8 Time (h)
10
12
24
48
FIG. 5. Mean (SEM) serum concentration of cortisol in the control (O) and extradural ( # ) groups. *P < 0.05, **P < 0.01 between groups.
(101) nmol litre"1. This value was maintained without further change until 12 h, after which cortisol concentrations declined.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
100 -
BRITISH JOURNAL OF ANAESTHESIA
276
istically significant. After surgery started, insulin concentrations increased to baseline values, with no additional change. Differences in serum insulin concentrations between groups were not significant.
50 -i 40 -
I 30 -
Metabolites
1 20 -| I
10 0 -
t
u 0
2
4
6 8 10 Time (h)
12
24 48
Serum cortisol concentrations were significantly different between the two groups at 1, 2, 4 and 6 h (P < 0.05). Growth hormone (fig. 6). In the control group, GH increased from 1.0(1.0-1.0) to 16.8(11.2-39.3) mu. litre"1 after 1 h of surgery (P < 0.01), with a further increase to 22.8 (2.8-44.4) mu. litre"1 after 2 h. GH values then decreased. In the extradural group there was no statistically significant change from the baseline value after 1 or 2 h of surgery, hence GH concentrations were significantly different between control and extradural groups at 1 (P < 0.001) and 2 h (P < 0.05). Insulin (table II). Serum insulin concentrations decreased after induction of anaesthesia, from 9.7 (2.4) to 2.7 (0.5) mu. litre"1 and from 7.5 (1.6) to 3.5 (1.2) mu. litre"1 in the control and extradural groups, respectively, although this change was not stat-
TABLE II. Mean (SEM) blood concentrations of glucose and lactate, plasma concentrations of NEFA, serum concentrations of insulin and packed cell volume {PCV). Significant differences within groups: * P < 0.05, ** P < 0.01 compared with baseline sample Time after start of surgery (h)
Glucose (mmol litre"') Control Extradural Lactate (mmol litre"1) Control Extradural NEFA (mmol litre"1) Control Extradural PCV (%) Control Extradural Insulin (mu. litre"1) Control Extradural
Baseline
0 min
1
2
4
6
12
24
48
4.30 (0.19) 4.11 (0.23)
4.35 (0.19) 3.93 (0.24)
6.03* (0.47) 4.43 (0.14)
6.05* (0.17) 4.82 (0.40)
5.70* (0.28) 5.03 (0.47)
5 92* (0.44) 4.78 (0.26)
5.37 (0.40) 4.61 (0.11)
5.10 (0.36) 4.25 (0 33)
4.39 (0.33) 4.98 (0.83)
0.81 (0.09) 0.73 (0.17)
0.91 (0.17) 0.67 (0.15)
1.04 (0.21) 0.99 (0.24)
0.97 (0.20) 0.57 (0.11)
0.83 (0.16) 0.55 (0.11)
0.89 (0.13) 0.62 (0.13)
0.83 (0.16) 0.50 (0.07)
0.79 (0.12) 0.65 (0.11)
0.68 (0.20) 0.59 (0.09)
0.74 (0.20) 0.58 (0.10)
0.54 (0.11) 0.73 (0.12)
0.89 (0.13) 0.58 (0.07)
0.84 (0.23) 0.45 (0.07)
1.16 (0 14) 0.51 (0.10)
1 18 (0.08) 0.77 (0.09)
0.88 (0.10) 1.01** (0.20)
0.62 (0.07) 0.73 (0.06)
0 69 (0.17) 0.69 (0.09)
38(1) 37(1)
34(1)* 33(1)*
36(1) 33(1)*
36(1) 34(1)*
36(1) 33(1)*
36(1) 34(1)*
34 (1)* 31(1)*
32(1)* 31 (1)*
31 (2)* 31 (1)*
9.7 (2.4) 7.5 (1.6)
2.7 (0.5) 3.5 (1.2)
11.5 (2 9) 5.5 (0.8)
12.0 (2.9) 10.1 (1.5)
9.7 (2 0) 6.1 (0.9)
8.2 (1.1) 5.8 (0.4)
11.1 (2 7) 77 (1.7)
16.1 (5.9) 7.6 (1.3)
139 (5.0) 21 2 (8.1)*
p (between groups) < 0.05 at 1 h and 2 h
< 0 05 at 4 h < 0.01 at 6 h
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
FIG. 6. Median (range) serum concentrations of growth hormone (GH) in the control (O) and extradural ( # ) groups. *P < 0.05, ***P < 0.001 between groups.
Glucose (table II). In the control group, blood glucose concentrations increased significantly from the baseline value of 4.30(0.19) to 6.03(0.47) mmol litre"1 after 1 h of surgery (P < 0.05) and remained significantly increased for 6 h. In contrast, in the extradural group, blood glucose concentrations did not change significantly and they remained smaller than those in the control group throughout the study (except for the 48 h sample). Concentrations of blood glucose were significantly smaller in the extradural group compared with the control group up to 2 h after surgery started. NEFA (table II). In the control group there were no significant changes in plasma NEFA concentrations during the study. In the extradural group, NEFA concentrations were less than those of the control group during surgery and the early postoperative period. There was a significant difference between the two groups at 4 (P < 0.05) and 6 h after surgery (P < 0.01). Lactate (table II). There were no significant differences in blood lactate concentrations either within or between groups. Packed cell volume (table II). Packed cell volume (PCV) decreased significantly in response to fluid loading in both groups. PCV decreased from 38 (1) to 34 (1)% in the control group (P < 0.05) and from 37 (1) to 33 (1)% in the extradural group (P < 0.05) from the baseline sample to the start of surgery. There was no significant difference in PCV values between the two groups of patients.
INTERLEUKIN-6 RESPONSE TO SURGERY izu 100 o
a. < 60 -
40 -I -1
0
1
2
4
8
12 24
48
4 6 Time (h)
8
12 24 48
80 -
I
60-
40 - 1 -1
FIG. 7. Mean (SEM) mean arterial pressure (MAP) and heart rate (HR) in the control ( • ) and extradural ( • ) groups. *P < 0.05 between groups.
Mean arterial pressure and heart rate (fig. 7)
In patients receiving general anaesthesia alone, MAP and heart rate did not change significantly during the study. In the extradural group, MAP decreased significantly from 89 (4) to 63 (4) mm Hg at the start of surgery (P < 0.05) and this decrease was maintained throughout surgery. Thus in the extradural group MAP was significantly smaller than the control group at 0 and 1 h (P < 0.05) and also after operation from 4 to 12 h (P < 0.05). In the extradural group, heart rate decreased significantly from 70 (9) to 58 (2) beat min'1 at the start of surgery (P < 0.05) and heart rate was significantly slower than the control group at 0 and 1 h (P < 0.05). DISCUSSION
We have found that the circulating IL-6 response to pelvic surgery was unaffected by extradural analgesia and was unrelated temporally to changes in pituitary hormone secretion. Therefore, it is improbable that IL-6 contributes to the significant increases in cortisol and GH which occur within 1 h of the start of surgery, but a later effect of the cytokine on pituitary function cannot be excluded. The similarity of the acute phase response (CRP and zinc concentrations) in the two groups is compatible with IL-6 changes and confirms previous work which found no effect of extradural analgesia on acute phase proteins after hysterectomy [12]. We observed a marked between-patient variability in the circulating IL-6 response to the standardized surgical procedure in
both groups. This could be accounted for, in part, by the inconsistent delay before the cytokine was detected in some patients (2-6 h), particularly in the extradural group. Similar variability in IL-6 was reported by Cruickshank and colleagues [1]. Naito and colleagues showed recently that the addition of thoracic extradural analgesia to general anaesthesia did not change IL-6 and cortisol responses to major upper abdominal surgery [13]. They also observed that the increase in ACTHcortisol concentrations preceded the increase in circulating IL-6. However, interpretation of the data is complicated by failure to show an increase in IL-6 after total hip replacement which is known to be a major stimulus to the synthesis of this cytokine [1]. A combined prednisolone, extradural and spinal analgesia, and indomethacin regimen resulted in a significant decrease in IL-6 and CRP responses to colonic surgery [14]. Previous work by this group had failed to show an effect of extradural analgesia and indomethacin on the acute phase response, and by inference IL-6, to cholecystectomy [15]. It was suggested that high-dose prednisolone (30 mg kg"1) i.v. was responsible for the decrease in IL-6 after colonic surgery, although neural block was also more intense. Unfortunately, no hormonal measurements were obtained so that it is not possible to determine if a reduction in IL-6 resulted in a decline in pituitary secretion. There is some biochemical evidence to support the suppressive effects of highdose steroids on cytokine synthesis [16], but this technique is unlikely to be adopted clinically because in 11 patients undergoing abdominal surgery there were two instances of wound dehiscence and one anastomotic leak [14]. There are increasing in vitro and animal data to show that IL-6 alters pituitary function [2-4]. However, there are several problems that must be considered before this work can be applied to humans. IL-6 is a large glycoprotein, 23-30 kDa, that does not pass the blood-brain barrier. It is possible that IL-6 may exert its effects at areas where the blood-brain barrier is weak or deficient, such as the ventral part of the median eminence of the hypo thalamus. All these areas have fenestrated capillaries and circulating IL-6 may either gain direct access to the cerebral circulation or stimulate central secretion of IL-6. Corticotrophs are abundant in the median eminence of the hypothalamus and can release corticotrophin releasing factor which stimulates the pituitary to secrete ACTH. Experimental data show that the effects of IL-6 on anterior pituitary function occur rapidly. Administration of human recombinant IL-6 1.25 pmol into the third ventricle of freely moving rats produced an increase in ACTH secretion within 15 min [17], and within 30 min when given i.v. [2]. Spangelo and colleagues showed that the addition of large concentrations of murine recombinant IL-6 to anterior pituitary glands stimulated the release of prolactin, GH and LH during a 30-min incubation period [3]. Any effects of IL-6 released during surgery on pituitary function should occur promptly, although some in vitro work has suggested a more prolonged time course of 4 h [18].
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
6
100 -i
277
278
secretion in intact humans requires careful investigation and extrapolation of existing laboratory data has many difficulties. The use of anti-IL-6 antibodies in oncology patients, such as those with multiple myeloma, may be a valuable investigatory tool. ACKNOWLEDGEMENTS Financial support for the study was provided by Hammersmith and Queen Charlotte's Health Authority and we are grateful to Dr D. J. Newman, Department of Clinical Chemistry, London Hospital Medical School, for the albumin and CRP assays. REFERENCES 1. Cruickshank AM, Fraser WD, Burns HJG, Van Damme J, Shenkin A. Response of serum interleukin-6 in patients undergoing elective surgery of varying severity. Clinical Science 1990; 79: 161-165. 2. Naito Y, Fukata J, Tominaga T, Nakcin Y, Tamai S, Mori K, Imura H. Interleukin-6 stimulates the secretion of adrenocorticotropic hormone in conscious, freely moving rats. Biochemical and Biophysical Research Communications 1988;
155: 1459-1463. 3. Spangelo BL, Judd AM, Isakson PC, MacLeod RM. Interleukin-6 stimulates anterior pituitary hormone release in vitro. Endocrinology 1989; 125: 575-577. 4. Naito Y, Fukata J, Tominaga T, Masui Y, Hirai Y, Murakami N, Tamai S, Mori K, Imura H. Adrenocorticotropic hormone-releasing activities of interleukins in a homologous in vivo system. Biochemical and Biophysical Research Communications 1989; 164: 1262-1267. 5. Moore CM, Desborough JP, Burrin JM, Hall GM. IL-6 and the pituitary hormone response to surgery. Journal of Endocrinology 1992; 132S: 207. 6. Price CP, Trull AK, Berry D, Gorman EG. Development and validation of a particle-enhanced immunoassay for Creactive protein. Journal of Immunological Methods 1987; 99: 205-211. 7. Medcalf EA, Newman DJ, Gorman EG, Price CP. Rapid robust method for measuring low concentrations of albumin in the urine. Clinical Chemistry 1990; 36: 446-449. 8. Makino T, Saito M, Horiguchi D, Kina K. A highly sensitive colorimetric determination of serum zinc using water-soluble pyndylazo dye. Chnica Chimica Acta 1982; 120: 127-135. 9. Desborough JP, Edlin SA, Burrin JM, Bloom SR, Morgan M, Hall GM. Hormonal and metabolic responses to cholecystectomy. Comparison of extradural somatostatin and diamorphine. British Journal of Anaesthesia 1989; 63: 508-515. 10. Hall GM, Young C, Holdcroft A, Alaghband-Zadeh J. Substrate mobilisation during surgery. A comparison between halothane and fentanyl anaesthesia. Anaesthesia 1978; 33: 924-930. 11. Dunnett CW. New tables for multiple comparisons with control. Biometrics 1964; 20: 482^91. 12. Rem J, Saxtrup Nielsen O, Brandt MR, Kehlet H. Release mechanisms of postoperative changes in various acute phase proteins and immunoglobulins. Ada Chirugica Scandinavica 1980; 502S: 51-56. 13. Naito Y, Tamai S, Shingu K, Shindo K, Matsui T, Segawa H, Nakai Y, Mori K. Responses of plasma adrenocorticotrophic hormone, cortisol and cytokines during and after upper abdominal surgery. Anesthesiology 1992; 77: 426-431. 14. Schulze S, Sommer P, Bigler D, Honnens M, Shenkin A, Cruickshank AM, Bukhave K, Kehlet H. Effect of combined prednisolone, epidural analgesia and indomethacin on the systemic response after colonic surgery. Archives of Surgery 1992; 127: 325-331. 15. Schulze S, Roikjaer O, Hasselstrom L, Jensen NH, Kehlet H. Epidural bupivacaine and morphine plus systemic indomethacin eliminates pain but not the systemic response and convalescence after cholecystectomy. Surgery 1988; 103: 321-327. 16. Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochemical Journal 1990; 265: 621-636.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
The doses of IL-6 used in most experiments were much greater than those found during surgery. For example, Spangelo and colleagues used 200-2000 u. ml-1 (200-2000 pg ml"1) of IL-6 on pituitary glands in vitro [3] and Naito and colleagues administered 1 and 5 ug i.v. per rat [2]. In contrast, during surgery, peak serum IL-6 concentrations are usually less than 500 pg ml"1 (500 u. ml"1) [1] and the greatest concentration measured in the present study was 380 pg ml"1 (fig. 1). Caution must be exercised, therefore, before laboratory data are assumed to apply clinically. Indeed, in a critical assessment of the hypothalamic—pituitary—adrenal-immune axis, it was concluded that in vitro experiments, and experiments using large doses of cytokines, provide information only on what the system under study can do, not about what it does physiologically in the whole animal [19]. It would be unwise to dismiss this endocrinological work as having no clinical relevance. In severe cases of endotoxaemia, peak IL-6 concentrations have been shown to be in the ngml"1 range (> 1000 u. ml"') [20] and we speculate that in burns patients, who have persistent hypercortisolaemia lasting several weeks, this may be mediated in part by the cytokine response to massive tissue damage. A further complication to the link between IL-6 and pituitary function is provided by the observation that rat anterior pituitary cells in culture secrete spontaneously large amounts of IL-6 [21]. The cytokine is derived from the folliculo—stellate cells in the pituitary which envelop the hormone secreting cells. It has been suggested that IL-6 acts as a paracrine regulator, via humoral factors, of the hormone secreting cells. We found no evidence of an increase in circulating IL-6 at a time when there was marked activation of pituitary secretion. IL-6 is unlikely to be co-released with pituitary hormones in humans so that changes in this cytokine during surgery reflect predominantly tissue damage. Extradural block resulted in the expected attenuation of glucose, NEFA, cortisol and GH responses to surgery, together with intraoperative hypotension and bradycardia. The increased neutrophil count found 2 h after surgery was prevented also by extradural analgesia, although neural block had no effect on lymphopenia. It is known that extradural block inhibits granulocytosis after pelvic surgery [22], but not after abdominal surgery [15]. A beneficial effect of this technique on lymphopenia found after pelvic surgery [22] was not confirmed by other workers [23], or by this study. The neutrophilia of surgery is similar to that found during exercise [24] or after the infusion of adrenaline and is probably mediated therefore by the sympathetic nervous system and cortisol. The discrepancies in the literature on the effects of neural block on haematological indices may be explained by the variability in the abolition of the catecholamine and cortisol responses to surgery. A possible effect of IL-6 on the later changes in white cell counts cannot be excluded. We conclude that IL-6 is unlikely to be implicated in the early pituitary hormonal responses to surgery. The possible role of IL-6 in controlling pituitary
BRITISH JOURNAL OF ANAESTHESIA
INTERLEUKIN-6 RESPONSE TO SURGERY 17. Lyson K, McCann SM. The effect of interleukin-6 on pituitary hormone release in vivo and in vitro. Neuroendocrinology 1991; 54: 262-266. 18. Harbuz MS, Stephanou A, Sarlis N, Lightman SL. The effects of recombinant human interleukin (IL)-la, IL-lp or IL-6 on hypothalamo—pituitary—adrenal axis activation. Journal of Endocrinology 1992; 133: 349-355. 19. Lilly MP, Gann DS. The hypothalamic-pituitary-adrenalimmune axis. A critical assessment. Archives of Surgery 1992; 127: 1463-1474. 20. Fong Y, Moldawer LL, Marano M, Wei H, Tatter SB, Clarick RH, Santhanam U, Sherris D, May LT, Sehgal PB, Lowry SF. Endotoxaemia elicits increased circulating P2IFN/IL-6 in man. Journal of Immunology 1989; 142: 2321-2324.
279 21. Spangelo BL, MacLeod RM, Isakson PC. Production of interleukin-6 by anterior pituitary cells in vitro. Endocrinology 1990; 126: 582-586. 22. Rem J, Brandt MR, Kehlet H. Prevention of postoperative lymphopenia and granulocytosis by epidural analgesia. Lancet 1980; 1: 283-285. 23. Tonnesen E, Wahlgreen C. Influence of extradural and general anaesthesia on natural killer cell activity and lymphocyte subpopulations in patients undergoing hysterectomy. British Journal of Anaesthesia 1988; 60: 500-507. 24. Peters AM, Allsop P, Stuttle AWJ, Arnot RN, Gwilliam ME, Hall GM. Granulocyte margination in the human lung and its response to strenuous exercise. Clinical Science 1992; 82: 237-244.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
Effects of extradural anaesthesia on interleukin-6 and acute phase response to surgery C. M. MOORE, J. P. DESBOROUGH, H. POWELL, J. M. BURRIN AND G. M. HALL
SUMMARY
KEY WORDS Hormones growth. Hormones- glucocorticoid. protein acute phase. Complications: trauma.
Metabolism,
The main causative factors in evoking the metabolic response to surgery are afferent neuronal input from the operative site and release of cytokines from tissue damaged by surgical trauma. The major cytokine synthesized after routine surgery is interleukin-6 (IL-6) and this polypeptide stimulates acute phase protein synthesis in the liver and affects lymphocyte function and haematopoiesis also [1]. There are data from in vitro and animal studies to show that cytokines, including IL-6, may stimulate pituitary hormonal secretion [2^1]. This has been interpreted as evidence for important links between inflammatory mediators and endocrine control and the
PATIENTS AND METHODS
We studied 16 healthy females, aged 20-50 yr, undergoing total abdominal hysterectomy for benign disease; the patients were allocated randomly to receive either general anaesthesia and extensive extradural block with local anaesthetic from T4 to S5 dermatomal segments (extradural group) or general anaesthesia alone (control group). All patients gave written informed consent and local Ethics Committee approval (registration No. 90/3497) was obtained. Patients with metabolic or hormonal disorders were excluded. After overnight fasting, each patient received papaveretum 0.3 mg kg""1 i.m. and hyoscine 1 0.006 mg kg" as premedication, 90 min before operation. On arrival in the anaesthetic room, electrocardiograph (ECG), arterial oxygen saturation (SpO2) and non-invasive arterial pressure monitoring were commenced. A central vein was catheterized via the antecubital fossa for administration of fluids and collection of blood samples and the position of the catheter confirmed by haemodynamic measurement. Mean arterial pressure (MAP) and heart rate were
C. M. MOORE, M.B., CH.B, F.R.C.A., J. P. DESBOROUGH*, M.B., CH.B., F.R.C.A., H . POWELLt, M.B., B.S., F.R.C.A., J. M . BURRIN^:, B.SC, PH.D., M.R.C.PATH., G. M . HALL*, M.B., B.S., PH.D.,
F.I.BIOL., F.R.C.A., Departments of Anaesthetics and Medicine,
Royal Postgraduate Medical School, Hammersmith Hospital, London W12 0NN. Accepted for Publication: October 21, 1993. Present addresses: * Department of Anaesthesia, St George's Hospital Medical School, Cranmer Terrace, London SW17 ORE. t Department of Anaesthesia, Freeman Hospital, High Heaton, Newcastle upon Tyne NE7 7DN. % Department of Clinical Biochemistry, London Hospital Medical College, Turner Street, London El 2AD. Correspondence to G.M.H. at St George's Hospital.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
Serum concentrations of the cytokine, interleukin-6 (IL-6), increase after surgical trauma. IL-6 mediates the synthesis of acute phase proteins and stimulates secretion of pituitary hormones. We have examined the time course of circulating IL-6, and cortisol and growth hormone responses in patients undergoing hysterectomy to determine if IL-6 contributes to the early pituitary hormone changes found during surgery. One group (n = 8) received a standardized general anaesthetic while the remaining patients (n = 8) received extradural analgesia to T4-S5 in addition to a similar general anaesthetic. In the general anaesthesia group, there was a significant increase in serum cortisol and growth hormone concentrations before any changes in IL-6 were detected. Furthermore, in the extradural group, in whom these hormonal responses were attenuated, circulating IL-6 concentrations did not differ significantly from the general anaesthesia group. There were no significant differences between the groups in the acute phase response, as measured by circulating concentrations of C-reactive protein and zinc, but the expected effects of extradural block on circulating metabolites and white cell count were demonstrated. We conclude that IL-6 is unlikely to contribute to the initial increases in secretion of pituitary hormones found during surgery, but a later effect of the cytokine on endocrine responses cannot be excluded. (Br. J. Anaesth. 1994; 72: 272-279)
term "endocrine immunology" has been used to describe the association. However, there are few clinical data, at present, to support this proposal. The purpose of the present study was to determine if IL-6 contributes to the early pituitary hormonal changes observed in the perioperative period. We have studied the time course of the endocrine and cytokine changes in patients undergoing pelvic surgery in whom it is known that catabolic hormonal secretion can be obtunded for several hours by extradural analgesia. Preliminary results of this study have been presented previously [5].
INTERLEUKIN-6 RESPONSE TO SURGERY
Blood sample collection
Venous blood samples were collected at the following times: a baseline sample 45 min before surgical incision, at the start of surgery (0 min), and at 1, 2,4, 6, 12, 24 and 48 h after the start of surgery. For serum assays, blood was placed in plain tubes and centrifuged. For plasma samples, blood was placed in lithium heparin tubes and centrifuged. The supernatant samples were collected and stored at - 2 0 °C until assay. All samples were measured in duplicate. Serum concentrations of C-reactive protein (CRP) and albumin, and plasma zinc concentrations were measured in samples obtained at baseline and 0, 6, 12, 24 and 48 h. Blood was placed in EDTA tubes for differential white cell counts from the baseline sample and at 0, 1, 2, 6, 12, 24 and 48 h. Serum cortisol, GH, insulin and IL-6 concentrations were analysed in the baseline sample and at 0, 1, 2, 4, 6, 12, 24 and 48 h. Blood glucose and lactate, and plasma non-esterified fatty acids (NEFA) concentrations were measured in all samples, and heart rate and MAP were measured at all times. Investigators were unaware of patient identity during sample analysis. Haematological and biochemical measurements
Serum concentration of IL-6 was measured by a two-site enzyme immunometric assay (Quantikine, RD systems, British Bio-technology Ltd, Abingdon, Oxon). Intra- and interassay precision were 7 % and
9%, respectively, at an IL-6 concentration of 93 pg ml"1. The lower limit of sensitivity of the assay was 3.5 pg ml"1. Serum concentrations of CRP and albumin were measured using latex-enhanced immunoturbidimetry [6,7]. The intra-assay coefficient of variation was 5 % for both assays at all concentrations, and all samples were measured in the same assay. Plasma zinc concentration was measured using a simple two-step colorimetric method with trichloracetic acid deproteinization and the specific chelating agent 2-(5-bromo-2-pyridylazo)-5-(iV-propyl-iVsulphopropylamino)-phenol (Wako, Alpha Laboratories, Eastleigh, Hants, U.K.) [8]. Results obtained from this assay correlated well (r = 0.963) with those by atomic absorption spectrophotometry. All samples were measured in the same assay and the within-assay coefficient of variation was 4.5 % at a zinc concentration of 20 umol litre"1. Packed cell volume was measured using a microcapillary haematocrit centrifuge with correction for trapped plasma. A full blood count was performed on EDTA blood using a Sysmek E-5000 (TOA Medical Electronics (U.K.) Ltd) which also performed a differential leucocyte count presenting these data as neutrophil percentage and absolute number, lymphocyte percentage and absolute number, and (monocytes, eosinophils and basophils) percentage and absolute number. Serum cortisol concentration was measured using the Farmos Diagnostica (Orion Corporation, Turku, Finland) 125I-direct-RIA kit. This assay uses a cortisol antiserum which has negligible crossreactivity with other endogenous corticosteroid and cortisol metabolites. Intra- and interassay coefficients of variation were 3.2% and 4.5% at 280 mmol litre"1 and 3.9 % and 6.2 % at 680 nmol litre"1, respectively. Serum growth hormone (GH) concentration was measured using a solid phase two-site immunoradiometric assay (Netria Labs, London, U.K.). The lower limit of sensitivity was 1.0 mu. litre"1. Within-batch variation at 4.2 mu. litre"1 was 9% and at 19.5 mu. litre"1 8%. Betweenbatch variation at 3.9 mu. litre"1 was 11% and at 20.7 mu. litre"1 10%. Serum insulin concentration was measured by a sensitive and specific double antibody radioimmunoassay described previously [9]. Intra- and interassay precision at 8.6 mu. litre"1 were 5.6% and 8.9%, respectively. Blood glucose and lactate values, and plasma NEFA concentration were determined enzymatically by standard methods [10]. Statistical analyses Parametric data are expressed as mean (SEM) and analysed using two-way analysis of variance (ANOVA) with Dunnett's test for within-group changes with respect to the baseline sample [11]. Between-group changes were analysed by ANOVA. P < 0.05 was accepted as significant. For GH and IL-6, concentrations less than the sensitivity of the assay were assigned the value of that limit. GH and IL-6 data were not normally distributed and therefore expressed as median (range). Differences within each group with respect to the baseline sample were
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
recorded and a baseline blood sample obtained: 14 ml kg"1 of sodium chloride solution 150mmol litre"1 was infused in all patients. All studies commenced between 08:00 and 10:00. In the extradural group, during fluid loading, a catheter was inserted into the extradural space at the L2-3 or L3—4 intervertebral level. The patient was then placed supine, a test dose of 0.5 % plain bupivacaine 3 ml given, followed 5 min later by increments of 5 ml of 0.5 % plain bupivacaine to produce bilateral analgesia to pin prick and loss of sensation to ethyl chloride spray from S5 to T4 or a higher dermatomal level. The block was maintained by an extradural infusion of 0.5 % plain bupivacaine at a rate of 6—8 ml h"1; this was discontinued 6 h after surgical incision. The extent of the block was tested before induction of anaesthesia and at 2, 4 and 6 h after surgical incision. MAP was maintained at or above 60 mm Hg by i.v. bolus doses of methoxamine 1 mg as required during operation in the extradural group. In both groups of patients, anaesthesia was induced with a dose of thiopentone sufficient to obtund the eyelash reflex. Vecuronium 0.1 mgkg"1 was given to facilitate tracheal intubation and the lungs ventilated with 0-1 % halothane and 66 % nitrous oxide in oxygen. End-tidal carbon dioxide partial pressure was maintained at 4—5 kPa. I.v. fluids (sodium chloride solution 150 mmol litre"1) were given at a rate of 6 ml kg"1 h"1 during operation and at 2 ml kg"1 h"1 after operation. Postoperative analgesia in the control group was provided by papaveretum 5 mg i.v. on demand.
273
BRITISH JOURNAL OF ANAESTHESIA
274
analysed by the Wilcoxon's matched pairs signed rank test, and between-group changes by the Wilcoxon rank sum test. RESULTS
TABLE I. Patient data {mean {SEM or range))
Age (yr) Weight (kg) Duration of surgery (min)
Control group (« = 8)
Extradural group (« = 7)
41.9(36-45) 66.9 (3.8) 79.9(11.5)
43.3 (39-47) 62.7 (2 2) 97.0(12.9)
1000 n
CRP (fig. 2) In both groups there was a significant increase in CRP 24 h after surgery, from 2.3 (0.5) to 79.7 (11.1) mg litre"1 (P < 0.01) in the control group and from 4.4(1.0) to 67.2 (8.5) mg litre"1 (P < 0.01) in the extradural group. After 48 h, CRP had increased further, to 130.8 (16.2) and 122.0 (15.7) mg litre"1 in the control and extradural groups, respectively. There were no significant differences in CRP concentrations between the two groups. Plasma zinc and serum albumin (fig. 3)
7. 100 -
£ a
10 -
6 8 10 Time (h)
12 24 48
FIG. 1. Median (range) serum concentration of interleukin-6 (IL-6) in the control (O) and extradural (#) groups. Interleukin-6 values are plotted on a logarithmic scale.
150 -i
100 CT>
£
a. 50 u
0
J 1
6
i
8 Time (h)
Circulating white blood cells (fig. 4)
i,
10
12
In both groups of patients, plasma zinc concentration decreased significantly after surgery, from 12.6(0.5) to 6.6 (0.5) umol litre"1 (P < 0.05) and from 11.8 (0.4) to 6.9 (0.6) umol litre"1 (P < 0.05) at 6 h in the control and extradural groups, respectively. Zinc concentration remained significantly decreased at 48 h compared with the concentration before operation (P < 0.05). There were no significant differences between the two groups. Serum albumin concentration decreased after fluid loading and induction of anaesthesia from 36.1 (0.5) to 31.7 (0.7) g litre"1 (P < 0.01) in the control group. After surgery, albumin concentrations in this group were stable until 12 h, but decreased significantly compared with baseline sample, with concentrations of 29.4 (1.1) and 30.1 (0.7) g litre"1 at 24 and 48 h, respectively. In the extradural group, there was no significant change in serum albumin concentrations during the study and there were no significant differences between the two groups. The zinc: albumin ratio was calculated to examine changes in divalent cation that were independent of concurrent alterations in its major binding protein. The results are shown as a percentage change from the value at the start of surgery. This ratio decreased after surgery in both groups, with no significant difference between groups.
24 48
FIG. 2. Mean (SEM) serum concentration of C-reactive protein (CRP) in the control (O) and extradural ( # ) groups.
The white cell count was increased significantly in both groups after surgery. In the control group the count increased from 6.1 (0.7) to 13.4 (1.3) x 109 litre"1 at 2 h (P < 0.01) and increased
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
There were no significant differences between groups in age, weight or duration of surgery (table I). One patient in the extradural group was withdrawn from the study because she had marked systolic hypotension because of intraoperative bleeding and extradural block was discontinued. The dermatomal levels of the extradural block were T1-T3-S5 at 2 h (w = 6), T2-T5-S5 at 4 h (« = 7) and T4-T6-S5 (« = 7) at 6 h. Only two patients in the extradural group needed methoxamine, both during operation (3 and 5 mg).
IL-6 (fig. 1) In the control group, serum IL-6 concentration increased from a baseline value of 3.5 to 28.4 (7.8-60.0) pg ml"1,4 h after the start of surgery (P < 0.01). This increase was maintained, with IL-6 values remaining significantly greater than baseline concentrations at 6 and 12 h (P < 0.01), 24 h (P < 0.02) and 48 h (P < 0.05). In contrast, in the extradural group, IL-6 concentrations were not increased significantly compared with the baseline* value of 3.5 pg ml"1 until 12 h, when concentrations reached 25.3 (3.5-218.0) pg ml"1 (P < 0.05). A significant increase in IL-6 in the extradural group was then sustained for up to 48 h (P < 0.05). Although the IL-6 response was delayed in the extradural group, there was no significant difference in IL-6 concentrations between the two groups during the study.
INTERLEUKIN-6 RESPONSE TO SURGERY
275
15 -i
20 -i
-
15 -
£ 10 o
° 10 -
E
a. 5 -
0
o m
J
0 20
110 -i |
90 -
i
80 -
JO
70 -
I5
60 -
2
15
— 10 -
c
!E a
N
0
50 -> 0
2
4
6 8 10 Time (h)
12
24
J
20 -i
48
FIG. 3. Mean (SEM) plasma zinc concentration and zinc: albumin ratio, as percentage of baseline sample, in the control (O) and extradural ( # ) groups. CD
9
S io H a
1
further to 16.4 (1.7) x 10 litre" after 6 h. In the extradural group, the increase in cell count was slower, from 5.8 (0.6) to only 9.0 (1.0) x 109 litre"1 at 2 h (P < 0.05) with a further increase to 14.6 (1.2) x 109 litre"1 after 6 h. The white cell count was significantly smaller in the extradural group compared with the control group at 2 h (P < 0.05). Neutrophil leucocytosis occurred in both groups, with a change from 3.5 (0.4) to 10.5 (1.0) x 109 litre"1 (P < 0.01) in the control group, but from 3.4 (0.5) to only 6.2 (1.0) x 109 litre"1 in the extradural group at 2 h. This difference was statistically significant (P < 0.05). In contrast, the lymphocyte count in both groups decreased in the postoperative period, although the decrease reached statistical significance only in the control group at 24 h (P < 0.05). There were no significant differences between groups in lymphocyte count during the study. Hormones
Cortisol (fig. 5). In the control group, serum cortisol concentration increased from 247 (28) to 725 (60) nmol litre"1 (P < 0.05) after 1 h of surgery and continued to increase to 1226 (154) nmol litre"1 (P < 0.01) at 4 h. Concentrations then declined progressively to baseline values after 48 h. In contrast, however, in the extradural group, serum cortisol concentration had not changed significantly after 1 h of surgery compared with the baseline value. Four hours after surgery started, cortisol concentrations had increased to only 672
E 0 -1 I
l
l
6 8 10 Time(h)
12
24
48
FIG. 4. Mean (SEM) total circulating white blood cells (WBC), neutrophils and lymphocytes in the control (O) and extradural ( • ) groups. *P < 0.05, **P < 0.01 between groups. 1500 -i
CD
£ 1000 "5 E _c § 500 eo
o
0
J
6 8 Time (h)
10
12
24
48
FIG. 5. Mean (SEM) serum concentration of cortisol in the control (O) and extradural ( # ) groups. *P < 0.05, **P < 0.01 between groups.
(101) nmol litre"1. This value was maintained without further change until 12 h, after which cortisol concentrations declined.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
100 -
BRITISH JOURNAL OF ANAESTHESIA
276
istically significant. After surgery started, insulin concentrations increased to baseline values, with no additional change. Differences in serum insulin concentrations between groups were not significant.
50 -i 40 -
I 30 -
Metabolites
1 20 -| I
10 0 -
t
u 0
2
4
6 8 10 Time (h)
12
24 48
Serum cortisol concentrations were significantly different between the two groups at 1, 2, 4 and 6 h (P < 0.05). Growth hormone (fig. 6). In the control group, GH increased from 1.0(1.0-1.0) to 16.8(11.2-39.3) mu. litre"1 after 1 h of surgery (P < 0.01), with a further increase to 22.8 (2.8-44.4) mu. litre"1 after 2 h. GH values then decreased. In the extradural group there was no statistically significant change from the baseline value after 1 or 2 h of surgery, hence GH concentrations were significantly different between control and extradural groups at 1 (P < 0.001) and 2 h (P < 0.05). Insulin (table II). Serum insulin concentrations decreased after induction of anaesthesia, from 9.7 (2.4) to 2.7 (0.5) mu. litre"1 and from 7.5 (1.6) to 3.5 (1.2) mu. litre"1 in the control and extradural groups, respectively, although this change was not stat-
TABLE II. Mean (SEM) blood concentrations of glucose and lactate, plasma concentrations of NEFA, serum concentrations of insulin and packed cell volume {PCV). Significant differences within groups: * P < 0.05, ** P < 0.01 compared with baseline sample Time after start of surgery (h)
Glucose (mmol litre"') Control Extradural Lactate (mmol litre"1) Control Extradural NEFA (mmol litre"1) Control Extradural PCV (%) Control Extradural Insulin (mu. litre"1) Control Extradural
Baseline
0 min
1
2
4
6
12
24
48
4.30 (0.19) 4.11 (0.23)
4.35 (0.19) 3.93 (0.24)
6.03* (0.47) 4.43 (0.14)
6.05* (0.17) 4.82 (0.40)
5.70* (0.28) 5.03 (0.47)
5 92* (0.44) 4.78 (0.26)
5.37 (0.40) 4.61 (0.11)
5.10 (0.36) 4.25 (0 33)
4.39 (0.33) 4.98 (0.83)
0.81 (0.09) 0.73 (0.17)
0.91 (0.17) 0.67 (0.15)
1.04 (0.21) 0.99 (0.24)
0.97 (0.20) 0.57 (0.11)
0.83 (0.16) 0.55 (0.11)
0.89 (0.13) 0.62 (0.13)
0.83 (0.16) 0.50 (0.07)
0.79 (0.12) 0.65 (0.11)
0.68 (0.20) 0.59 (0.09)
0.74 (0.20) 0.58 (0.10)
0.54 (0.11) 0.73 (0.12)
0.89 (0.13) 0.58 (0.07)
0.84 (0.23) 0.45 (0.07)
1.16 (0 14) 0.51 (0.10)
1 18 (0.08) 0.77 (0.09)
0.88 (0.10) 1.01** (0.20)
0.62 (0.07) 0.73 (0.06)
0 69 (0.17) 0.69 (0.09)
38(1) 37(1)
34(1)* 33(1)*
36(1) 33(1)*
36(1) 34(1)*
36(1) 33(1)*
36(1) 34(1)*
34 (1)* 31(1)*
32(1)* 31 (1)*
31 (2)* 31 (1)*
9.7 (2.4) 7.5 (1.6)
2.7 (0.5) 3.5 (1.2)
11.5 (2 9) 5.5 (0.8)
12.0 (2.9) 10.1 (1.5)
9.7 (2 0) 6.1 (0.9)
8.2 (1.1) 5.8 (0.4)
11.1 (2 7) 77 (1.7)
16.1 (5.9) 7.6 (1.3)
139 (5.0) 21 2 (8.1)*
p (between groups) < 0.05 at 1 h and 2 h
< 0 05 at 4 h < 0.01 at 6 h
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
FIG. 6. Median (range) serum concentrations of growth hormone (GH) in the control (O) and extradural ( # ) groups. *P < 0.05, ***P < 0.001 between groups.
Glucose (table II). In the control group, blood glucose concentrations increased significantly from the baseline value of 4.30(0.19) to 6.03(0.47) mmol litre"1 after 1 h of surgery (P < 0.05) and remained significantly increased for 6 h. In contrast, in the extradural group, blood glucose concentrations did not change significantly and they remained smaller than those in the control group throughout the study (except for the 48 h sample). Concentrations of blood glucose were significantly smaller in the extradural group compared with the control group up to 2 h after surgery started. NEFA (table II). In the control group there were no significant changes in plasma NEFA concentrations during the study. In the extradural group, NEFA concentrations were less than those of the control group during surgery and the early postoperative period. There was a significant difference between the two groups at 4 (P < 0.05) and 6 h after surgery (P < 0.01). Lactate (table II). There were no significant differences in blood lactate concentrations either within or between groups. Packed cell volume (table II). Packed cell volume (PCV) decreased significantly in response to fluid loading in both groups. PCV decreased from 38 (1) to 34 (1)% in the control group (P < 0.05) and from 37 (1) to 33 (1)% in the extradural group (P < 0.05) from the baseline sample to the start of surgery. There was no significant difference in PCV values between the two groups of patients.
INTERLEUKIN-6 RESPONSE TO SURGERY izu 100 o
a. < 60 -
40 -I -1
0
1
2
4
8
12 24
48
4 6 Time (h)
8
12 24 48
80 -
I
60-
40 - 1 -1
FIG. 7. Mean (SEM) mean arterial pressure (MAP) and heart rate (HR) in the control ( • ) and extradural ( • ) groups. *P < 0.05 between groups.
Mean arterial pressure and heart rate (fig. 7)
In patients receiving general anaesthesia alone, MAP and heart rate did not change significantly during the study. In the extradural group, MAP decreased significantly from 89 (4) to 63 (4) mm Hg at the start of surgery (P < 0.05) and this decrease was maintained throughout surgery. Thus in the extradural group MAP was significantly smaller than the control group at 0 and 1 h (P < 0.05) and also after operation from 4 to 12 h (P < 0.05). In the extradural group, heart rate decreased significantly from 70 (9) to 58 (2) beat min'1 at the start of surgery (P < 0.05) and heart rate was significantly slower than the control group at 0 and 1 h (P < 0.05). DISCUSSION
We have found that the circulating IL-6 response to pelvic surgery was unaffected by extradural analgesia and was unrelated temporally to changes in pituitary hormone secretion. Therefore, it is improbable that IL-6 contributes to the significant increases in cortisol and GH which occur within 1 h of the start of surgery, but a later effect of the cytokine on pituitary function cannot be excluded. The similarity of the acute phase response (CRP and zinc concentrations) in the two groups is compatible with IL-6 changes and confirms previous work which found no effect of extradural analgesia on acute phase proteins after hysterectomy [12]. We observed a marked between-patient variability in the circulating IL-6 response to the standardized surgical procedure in
both groups. This could be accounted for, in part, by the inconsistent delay before the cytokine was detected in some patients (2-6 h), particularly in the extradural group. Similar variability in IL-6 was reported by Cruickshank and colleagues [1]. Naito and colleagues showed recently that the addition of thoracic extradural analgesia to general anaesthesia did not change IL-6 and cortisol responses to major upper abdominal surgery [13]. They also observed that the increase in ACTHcortisol concentrations preceded the increase in circulating IL-6. However, interpretation of the data is complicated by failure to show an increase in IL-6 after total hip replacement which is known to be a major stimulus to the synthesis of this cytokine [1]. A combined prednisolone, extradural and spinal analgesia, and indomethacin regimen resulted in a significant decrease in IL-6 and CRP responses to colonic surgery [14]. Previous work by this group had failed to show an effect of extradural analgesia and indomethacin on the acute phase response, and by inference IL-6, to cholecystectomy [15]. It was suggested that high-dose prednisolone (30 mg kg"1) i.v. was responsible for the decrease in IL-6 after colonic surgery, although neural block was also more intense. Unfortunately, no hormonal measurements were obtained so that it is not possible to determine if a reduction in IL-6 resulted in a decline in pituitary secretion. There is some biochemical evidence to support the suppressive effects of highdose steroids on cytokine synthesis [16], but this technique is unlikely to be adopted clinically because in 11 patients undergoing abdominal surgery there were two instances of wound dehiscence and one anastomotic leak [14]. There are increasing in vitro and animal data to show that IL-6 alters pituitary function [2-4]. However, there are several problems that must be considered before this work can be applied to humans. IL-6 is a large glycoprotein, 23-30 kDa, that does not pass the blood-brain barrier. It is possible that IL-6 may exert its effects at areas where the blood-brain barrier is weak or deficient, such as the ventral part of the median eminence of the hypo thalamus. All these areas have fenestrated capillaries and circulating IL-6 may either gain direct access to the cerebral circulation or stimulate central secretion of IL-6. Corticotrophs are abundant in the median eminence of the hypothalamus and can release corticotrophin releasing factor which stimulates the pituitary to secrete ACTH. Experimental data show that the effects of IL-6 on anterior pituitary function occur rapidly. Administration of human recombinant IL-6 1.25 pmol into the third ventricle of freely moving rats produced an increase in ACTH secretion within 15 min [17], and within 30 min when given i.v. [2]. Spangelo and colleagues showed that the addition of large concentrations of murine recombinant IL-6 to anterior pituitary glands stimulated the release of prolactin, GH and LH during a 30-min incubation period [3]. Any effects of IL-6 released during surgery on pituitary function should occur promptly, although some in vitro work has suggested a more prolonged time course of 4 h [18].
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
6
100 -i
277
278
secretion in intact humans requires careful investigation and extrapolation of existing laboratory data has many difficulties. The use of anti-IL-6 antibodies in oncology patients, such as those with multiple myeloma, may be a valuable investigatory tool. ACKNOWLEDGEMENTS Financial support for the study was provided by Hammersmith and Queen Charlotte's Health Authority and we are grateful to Dr D. J. Newman, Department of Clinical Chemistry, London Hospital Medical School, for the albumin and CRP assays. REFERENCES 1. Cruickshank AM, Fraser WD, Burns HJG, Van Damme J, Shenkin A. Response of serum interleukin-6 in patients undergoing elective surgery of varying severity. Clinical Science 1990; 79: 161-165. 2. Naito Y, Fukata J, Tominaga T, Nakcin Y, Tamai S, Mori K, Imura H. Interleukin-6 stimulates the secretion of adrenocorticotropic hormone in conscious, freely moving rats. Biochemical and Biophysical Research Communications 1988;
155: 1459-1463. 3. Spangelo BL, Judd AM, Isakson PC, MacLeod RM. Interleukin-6 stimulates anterior pituitary hormone release in vitro. Endocrinology 1989; 125: 575-577. 4. Naito Y, Fukata J, Tominaga T, Masui Y, Hirai Y, Murakami N, Tamai S, Mori K, Imura H. Adrenocorticotropic hormone-releasing activities of interleukins in a homologous in vivo system. Biochemical and Biophysical Research Communications 1989; 164: 1262-1267. 5. Moore CM, Desborough JP, Burrin JM, Hall GM. IL-6 and the pituitary hormone response to surgery. Journal of Endocrinology 1992; 132S: 207. 6. Price CP, Trull AK, Berry D, Gorman EG. Development and validation of a particle-enhanced immunoassay for Creactive protein. Journal of Immunological Methods 1987; 99: 205-211. 7. Medcalf EA, Newman DJ, Gorman EG, Price CP. Rapid robust method for measuring low concentrations of albumin in the urine. Clinical Chemistry 1990; 36: 446-449. 8. Makino T, Saito M, Horiguchi D, Kina K. A highly sensitive colorimetric determination of serum zinc using water-soluble pyndylazo dye. Chnica Chimica Acta 1982; 120: 127-135. 9. Desborough JP, Edlin SA, Burrin JM, Bloom SR, Morgan M, Hall GM. Hormonal and metabolic responses to cholecystectomy. Comparison of extradural somatostatin and diamorphine. British Journal of Anaesthesia 1989; 63: 508-515. 10. Hall GM, Young C, Holdcroft A, Alaghband-Zadeh J. Substrate mobilisation during surgery. A comparison between halothane and fentanyl anaesthesia. Anaesthesia 1978; 33: 924-930. 11. Dunnett CW. New tables for multiple comparisons with control. Biometrics 1964; 20: 482^91. 12. Rem J, Saxtrup Nielsen O, Brandt MR, Kehlet H. Release mechanisms of postoperative changes in various acute phase proteins and immunoglobulins. Ada Chirugica Scandinavica 1980; 502S: 51-56. 13. Naito Y, Tamai S, Shingu K, Shindo K, Matsui T, Segawa H, Nakai Y, Mori K. Responses of plasma adrenocorticotrophic hormone, cortisol and cytokines during and after upper abdominal surgery. Anesthesiology 1992; 77: 426-431. 14. Schulze S, Sommer P, Bigler D, Honnens M, Shenkin A, Cruickshank AM, Bukhave K, Kehlet H. Effect of combined prednisolone, epidural analgesia and indomethacin on the systemic response after colonic surgery. Archives of Surgery 1992; 127: 325-331. 15. Schulze S, Roikjaer O, Hasselstrom L, Jensen NH, Kehlet H. Epidural bupivacaine and morphine plus systemic indomethacin eliminates pain but not the systemic response and convalescence after cholecystectomy. Surgery 1988; 103: 321-327. 16. Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochemical Journal 1990; 265: 621-636.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015
The doses of IL-6 used in most experiments were much greater than those found during surgery. For example, Spangelo and colleagues used 200-2000 u. ml-1 (200-2000 pg ml"1) of IL-6 on pituitary glands in vitro [3] and Naito and colleagues administered 1 and 5 ug i.v. per rat [2]. In contrast, during surgery, peak serum IL-6 concentrations are usually less than 500 pg ml"1 (500 u. ml"1) [1] and the greatest concentration measured in the present study was 380 pg ml"1 (fig. 1). Caution must be exercised, therefore, before laboratory data are assumed to apply clinically. Indeed, in a critical assessment of the hypothalamic—pituitary—adrenal-immune axis, it was concluded that in vitro experiments, and experiments using large doses of cytokines, provide information only on what the system under study can do, not about what it does physiologically in the whole animal [19]. It would be unwise to dismiss this endocrinological work as having no clinical relevance. In severe cases of endotoxaemia, peak IL-6 concentrations have been shown to be in the ngml"1 range (> 1000 u. ml"') [20] and we speculate that in burns patients, who have persistent hypercortisolaemia lasting several weeks, this may be mediated in part by the cytokine response to massive tissue damage. A further complication to the link between IL-6 and pituitary function is provided by the observation that rat anterior pituitary cells in culture secrete spontaneously large amounts of IL-6 [21]. The cytokine is derived from the folliculo—stellate cells in the pituitary which envelop the hormone secreting cells. It has been suggested that IL-6 acts as a paracrine regulator, via humoral factors, of the hormone secreting cells. We found no evidence of an increase in circulating IL-6 at a time when there was marked activation of pituitary secretion. IL-6 is unlikely to be co-released with pituitary hormones in humans so that changes in this cytokine during surgery reflect predominantly tissue damage. Extradural block resulted in the expected attenuation of glucose, NEFA, cortisol and GH responses to surgery, together with intraoperative hypotension and bradycardia. The increased neutrophil count found 2 h after surgery was prevented also by extradural analgesia, although neural block had no effect on lymphopenia. It is known that extradural block inhibits granulocytosis after pelvic surgery [22], but not after abdominal surgery [15]. A beneficial effect of this technique on lymphopenia found after pelvic surgery [22] was not confirmed by other workers [23], or by this study. The neutrophilia of surgery is similar to that found during exercise [24] or after the infusion of adrenaline and is probably mediated therefore by the sympathetic nervous system and cortisol. The discrepancies in the literature on the effects of neural block on haematological indices may be explained by the variability in the abolition of the catecholamine and cortisol responses to surgery. A possible effect of IL-6 on the later changes in white cell counts cannot be excluded. We conclude that IL-6 is unlikely to be implicated in the early pituitary hormonal responses to surgery. The possible role of IL-6 in controlling pituitary
BRITISH JOURNAL OF ANAESTHESIA
INTERLEUKIN-6 RESPONSE TO SURGERY 17. Lyson K, McCann SM. The effect of interleukin-6 on pituitary hormone release in vivo and in vitro. Neuroendocrinology 1991; 54: 262-266. 18. Harbuz MS, Stephanou A, Sarlis N, Lightman SL. The effects of recombinant human interleukin (IL)-la, IL-lp or IL-6 on hypothalamo—pituitary—adrenal axis activation. Journal of Endocrinology 1992; 133: 349-355. 19. Lilly MP, Gann DS. The hypothalamic-pituitary-adrenalimmune axis. A critical assessment. Archives of Surgery 1992; 127: 1463-1474. 20. Fong Y, Moldawer LL, Marano M, Wei H, Tatter SB, Clarick RH, Santhanam U, Sherris D, May LT, Sehgal PB, Lowry SF. Endotoxaemia elicits increased circulating P2IFN/IL-6 in man. Journal of Immunology 1989; 142: 2321-2324.
279 21. Spangelo BL, MacLeod RM, Isakson PC. Production of interleukin-6 by anterior pituitary cells in vitro. Endocrinology 1990; 126: 582-586. 22. Rem J, Brandt MR, Kehlet H. Prevention of postoperative lymphopenia and granulocytosis by epidural analgesia. Lancet 1980; 1: 283-285. 23. Tonnesen E, Wahlgreen C. Influence of extradural and general anaesthesia on natural killer cell activity and lymphocyte subpopulations in patients undergoing hysterectomy. British Journal of Anaesthesia 1988; 60: 500-507. 24. Peters AM, Allsop P, Stuttle AWJ, Arnot RN, Gwilliam ME, Hall GM. Granulocyte margination in the human lung and its response to strenuous exercise. Clinical Science 1992; 82: 237-244.
Downloaded from http://bja.oxfordjournals.org/ at University of Bath Library & Learning Centre on June 22, 2015