- Email: [email protected]
ACID-BASE MANAGEMENT DURING HYPOTHERMIC CARDIOPULMONARY BYPASS DOES NOT AFFECT CEREBRAL METABOLISM BUT DOES AFFECT BLOOD FLOW AND NEUROLOGICAL OUTCOME
British Journal of Anaesthesia 1992; 69: 51-57
ACID-BASE MANAGEMENT DURING HYPOTHERMIC CARDIOPULMONARY BYPASS DOES NOT AFFECT CEREBRAL METABOLISM BUT DOES AFFECT BLOOD FLOW AND NEUROLOGICAL OUTCOME H. STEPHAN, A. WEYLAND, S. KAZMAIER, T. HENZE, S. MENCK AND H. SONNTAG SUMMARY
PATIENTS AND METHODS KEY WORDS Acid-base equilibrium: intraoperative management. Brain: blood flow, metabolism. Hypothermia. Surgery: cardiopu/monary bypass.
We studied 65 male patients (ages 38-68 yr, weight 51-110 kg) undergoing elective coronary artery bypass surgery. The study was approved by the
Optimal acid—base management during hypothermic cardiopulmonary bypass (CPB) has been a matter of discussion for many years. According to the pH-stat theory, arterial pH is maintained constant at 7.40 at
H. STEPHAN, M.D., A . WEYLAND, M.D., S. KAZMAIER, M.D., H. SONNTAG, M.D. (Department of Anaesthesiology); T. HENZE,
M.D., S. MENCK, M.D. (Department of Neurology); University of Gottingen, Robert Koch-Str. 40, 3400 GSttingen, Germany. Accepted for Publication: January 3, 1992.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
In order to compare the effects of blood-gas management on cerebral blood flow, metabolism and neurological outcome after hypothermic cardiopulmonary bypass (CPB) ,-we have studied 65 patients undergoing aorto-coronary bypass surgery allocated randomly to either a pH-stat (temperature-corrected blood-gas management) or an a-stat (temperature-uncorrected blood-gas management) group. All patients were examined neurologically on the day before and the 7th day after operation. In 20 patients of the pH-stat group and in 15 patients of the a-stat group we measured cerebral blood flow (CBF), using the argon washin technique, and also cerebral oxygen (CMRoJ and glucose (CMRg) uptake. Measurements were performed in awake patients, after induction of anaesthesia with fentanyl, midazo/am and pancuronium under normothermic conditions, during CPB at a venous blood temperature of 26 °C and at the end of surgery. Compared with postinduction values, hypothermia was associated with an 18% reduction in CBF and decreases in CMRO2 and CMRg of 61 % and 60%, respectively, in the a-stat group. In the pH-stat group, CMRo2 and CMRg decreased also, by 58% and 74%, respectively, whereas CBF increased by 191%, indicating uncoupling of flow and metabolism. As there were no statistically significant differences between the metabolic variables in both groups, we conclude that acidbase management did not affect cerebral metabolism, despite its influence on blood flow. After rewarming, CBF and cerebral metabolism normalized independently of acid-base management during hypothermia. Nevertheless, neurological dysfunction occurred more often in the pH-stat group (P = 0.036).
all temperatures. Currently, the a-stat approach is favoured, which allows arterial pH to increase with decreasing temperature according to the theory of Rahn [1]. With this regulation, arterial pH stays close to the value that represents the biological neutrality of blood. Hence, cell function should be better preserved with the a-stat concept [2]. With pH-stat management, enzymatic activity should be depressed by acidosis, in addition to the metabolic depression caused by hypothermia. On the other hand, acidosis may partly counteract the leftward shift in the oxygen dissociation curve which occurs in hypothermia, such that tissue oxygenation might be better preserved with the pH-stat approach. Studies in man and animals on the effect of acid—base management on myocardial function, systemic and myocardial metabolism have yielded conflicting results [3-8]. Although cerebral hyperperfusion associated with pH-stat management results in loss of CBF autoregulation and uncoupling of flow and metabolism [9], and may deliver more microemboli to the brain [10] or cause a steal phenomenon in patients with regional cerebral ischaemia [9], Bashein and colleagues [11] did not find any clinically significant effects of acid-base management on neurobehavioural outcome of patients who had undergone hypothermic cardiopulmonary bypass. However, an important objection against this study was that there was no correlation between neurological outcome and measured cerebral circulatory and metabolic changes during CPB [12]. The present study was designed to investigate the effects of both acid-base managements on cerebral blood flow, metabolism and neurological outcome after hypothermic cardiopulmonary bypass.
BRITISH JOURNAL OF ANAESTHESIA
52 TABLE I. Patient data (mean (range or SD) or number)
Management pH-stat No. patients Age (yr) Weight (kg) Height (cm) Body surface area (ml) Diabetes mellitus (n) Hypertension (n) Cold (n) Ejection fraction < 0.4 (n) CPB time (min) Aortic cross-clamp time (min) Balloon pump (n) Re-exploration (n) Time to tracheal extubation (h) Time to ICU discharge (h)
35
55(38-68) 82 (12) 175(7) 1.98(0.17) 6 20
a-stat 30
56 ((42-66) 80(11) 174(7) 1.94(0.16) 5 19
5
2
9
14
109 (32) 62 (20) 1 1
18(7) 41 (22)
109(34) 58(21) 1 1
19(18) 42(32)
Catheterization procedure
In the anaesthetic room, ECG leads and five-lead EEG (Lifescan, Neurometrics Inc.) were attached. The following catheters were inserted percutaneously under local anaesthesia: a 20-gauge catheter in the radial artery of the non-dominant hand to monitor arterial pressure and for blood sampling; a flow-directed pulmonary artery catheter (Edwards quadruple thermodilution model no. 93A 131-7F) via an antecubital vein into the pulmonary artery for measurement of pulmonary artery and wedge pressures and cardiac output; and a polyethylene catheter into the superior vena cava for administration of drugs and infusions. In patients in whom CBF was measured, the pressure line of the arterial catheter was replaced by a gas-tight Goodale-Lubin catheter (6-French gauge, USCI) and a second Goodale—Lubin catheter of the same size was positioned retrogradely into the superior bulb of the right internal jugular vein for measurement of CBF and jugular bulb pressure and withdrawal of blood samples. The position of this catheter was confirmed radiologically. Expired car-
Neurological examination
Neurological examination was performed by two trained neurologists who were blinded to the study. This conformed to the standard neurological examination performed in clinical practice and included evaluation of cranial nerve function, co-ordination, pain, temperature and light touch sensation, muscle strength, reflexes, wakefulness, recent and remote memory and presence of hallucinations or disorientation. Study periods and anaesthesia
Measurements were performed in the awake patient after a period of rest following catheterization (I), after induction of anaesthesia under normothermic conditions (II), 30 min after start of bypass at stable venous and arterial blood temperatures of 26 °C (III) and at the end of surgery (IV). Anaesthesia was induced with fentanyl 6 ug kg"1 and midazolam 0.2 mg kg"1 and maintained with fentanyl 0.15 ug kg"1 min"1 and midazolam 3 ug kg"1 min"1. Pancuronium 8 mg was administered to facilitate tracheal intubation and the lungs were ventilated with oxygen in air using a constant volume respirator (Engstrom ER 300). Incremental doses of fentanyl 0.25 mg and midazolam 7.5 mg were given immediately before sternotomy and when lightening of anaesthesia was apparent in the EEG. Cardiopulmonary bypass
The extracorporeal circulation technique consisted of a Polystan heart-lung machine with a membrane oxygenator (Maxima). Forty-micrometre filters (Sartorius, Gottingen) were used in the cardiotomy suction and the arterial line. Priming comprised Ringer's lactate 1000 ml, 5% glucose 500 ml, 20 % human albumin solution 400 ml, and sodium bicarbonate 100 mmol. Flow was maintained at 1.7-2.4 litre min"1 m"1 during bypass. Bretschneider's cardioplegic solution, at 4 °C, was infused into the aortic root after the aorta had been cross-clamped and was removed from the right atrium to prevent uptake into systemic circulation. In the pH-stat group, carbon dioxide was added to the fresh gas flow to maintain a temperaturecorrected PacOt of approximately 5.3 kPa, while in the a-stat group, no carbon dioxide was used and the fresh gas flow was varied to keep the Paco, near 5.3 kPa measured at 37 °C (not temperature-corrected). Paco was monitored continuously using an in-line blood-gas analysis sensor (Cardiovascular Devices Inc., Irvine, CA). CBF measurement and variables
CBF was measured using the argon washin technique. This method includes simultaneous blood sampling from the jugular bulb and the radial artery
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
Gottingen University Human Subjects Review Committee and written informed consent was obtained from all patients. Patients were allocated randomly to one of two groups managed with either the pHstat (group 1, n = 35) or the a-stat approach (group 2, n = 30) during hypothermic CPB. All patients were examined neurologically on the day before and the 7th day after operation. Cerebral blood flow and metabolism were studied in 20 patients of group 1 and in 15 patients of group 2. None of the patients suffered from pre-existing neurological disorders and there were no significant differences between the characteristics of both groups (table I). No patient had a history of valvular heart disease, liver disease or renal impairment. All patients were receiving maintenance doses of calcium channel blocking drugs, beta adrenoceptor antagonists and nitrates. Premedication consisted of fiunitrazepam 1-2 mg orally and piritramide 7.5—15 mg and promethazine 25-50 mg i.m., 1 h before arrival in the anaesthetic room.
bon dioxide concentration was measured with a Normocap CO2 Analyzer (Datex CD 102/02 Helsinki). The ECG and all pressures were monitored continuously and recorded simultaneously on a 10-channel chart recorder (Hellige, Freiburg). The EEG was also monitored continuously and recorded using the Lifescan printer.
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA TABLE II. Incidence of neurological sequelae
Management pH-stat (n = 35) Cerebellar symptoms (nystagmus, ataxia, dysdiadochokinesia etc.) Disorientation, confusion Right hemiparesis Illrd and Vlth nerve paresis Xllth nerve paresis
a-stat (n = 30)
5 1 1 2 1
The cerebral metabolic rates (CMR) of oxygen and glucose were calculated by multiplying the arterial-cerebral venous blood oxygen content and substrate concentration differences, respectively, by CBF. Statistics All results are expressed as mean (SD). Statistical analyses of the data obtained in each group were performed using the Friedman two-way ANOVA and the Wilcoxon matched-pairs signed-ranks test. The Mann-Whitney U test was used for intergroup comparisons. Factors that might predict neurological dysfunction were subjected to a multivariate analysis by a stepwise logistic regression procedure. P < 0.05 was assigned statistical significance. RESULTS
The characteristics of the patients in the pH-stat and in the a-stat group were similar (table I). Ten of 35 patients in the pH-stat group and two of 30 patients in the a-stat group were found to suffer from neurological sequelae on the 7th day after operation (table II); these had not existed before operation. Most of these sequelae were the result of focal infarcts in regions supplied by the basilar artery. One patient in each group was disoriented and confused at the time of the neurological examination— symptoms of global cerebral damage (encephalopathy). Multivariate analysis, by stepwise logistic regression procedure, revealed that the occurrence of neurological dysfunction did not relate to age (P =
TABLE III. Haemodynamic variables, PaCOf values and arterial hydrogen ion concentrations {mean {SD)). I = Awake; II = after induction of anaesthesia; / / / = during CPB at 26 "C; IV = end of surgery. CPP = Cerebral perfusion pressure; CVR = cerebral vascular resistance; CBF = cerebral blood flow; CI = cardiac index; [H+a] = arterial hydrogen ion concentration. Uncorr. = Values not corrected for temperature. P < 0.05: * compared with pH- slat; + /vs / / ; + //vs / / / ; S//vs/K;$///vs/K Time of measurement
CPP (mmHg) pH-stat a-stat CVR (mm Hg (ml/100 g min"1)"1) pH-stat a-stat CBF (ml/100 gmin"1) pH-stat a-stat PCV (%) pH-stat a-stat CI (litre min"1 m"1) pH-stat a-stat Pa~ (kPa) (Uncorr.) pH-stat a-stat [H+a] (nmol litre"1) (Uncorr.) pH-stat a-stat
I
II
III
91 (12) 92(17)
70(11)+ 89 (17)*
71 (15) 69 (14)+.
72 (12) 64(12)*f$
1.66(0.35) 1.96(0.50)
2.17(0.45)+ 2.75 (0.76)*+
0.89(0.51)$ 2.53(0.81)+.*
1.69(0.45)«$ 1.56(0.43)«
IV
56(10) 49(9)*
33 (4)+ 34(8)+
96 (39)+ 28 (5)+*
44 (8)« 43(9)«
42(4) 41(3)
38(4)+ 38(4)+
26(2)+ 25 (4)+.
29(2)i$ 27(5)f
3.3 (0.6) 2.7 (0.5)*
2.4(0.3>t 2.5 (0.6)+
2.0 (0.2)+. 2.1(0.2)+.
2.7(0.4*$ 2.9(0.5)§S
5.9 (0.4) 5.5 (0.4)*
5.5(0.3>t 5.1(0.4)*+
8.4(0.7)+. 5.2 (0.3)*
5.5(0.5* 5.3 (0.4)
57.9(4.7)+. 40.8 (2.7)*
40.0(3.9)5$ 42.4(3.7)*
44.3(2.1) 42.4(3.1)
41.9(2.8)+ 40.9 (2.9)+
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
during a 10-min period of inhalation of a standard concentration of argon. Argon was administered to the awake patient via a facemask, to the anaesthetized patient via the tracheal tube, and directly into the oxygenator during bypass. Immediately before and after each measurement of CBF, blood samples were taken from the jugular bulb and the radial artery and analysed for haemoglobin concentration, oxygen saturation (COOximeter IL 282), blood-gas tensions (standard electrodes, Radiometer), glucose concentration (standard test combination, Boehringer Mannheim) and electrolyte concentrations (absorption spectrometry, Perkin Elmer 303). Cardiac output was measured by thermodilution (cardiac output computer : Fischer BN 7206). Cerebral perfusion pressure (CPP) was calculated as mean arterial pressure minus jugular bulb pressure, and cerebral vascular resistance (CVR) as CPP divided by CBF. Cardiac index was calculated by dividing cardiac output by the body surface area.
53
BRITISH JOURNAL OF ANAESTHESIA
54
9-
710
O O
5-
I
3-
0.5599), diabetes mellitus (P = 0.4470), hypertension (P = 0.2006), cold (P = 0.6892), ejection fraction (P = 0.2481), CPB time (P = 0.644), aortic cross-clamp time (P = 0.0624) or use of an intra-aortic balloon pump (P = 0.4789), but only to the acid-base management (P = 0.0360). Haemodynamic and blood-gas data of those 35 patients in whom CBF and cerebral metabolism were studied are presented in table III. According to the design of the study, PaCOi and arterial hydrogen ion concentrations remained nearly constant over the course of the study in the a-stat group (n = 15) while in the pH-stat group (n = 20), P a ^ increased by 53 % at 26 °C venous blood temperature (not corrected for temperature). The individual PaCOj values of both groups were distinctly different at 26 °C (fig. 1) Anaesthesia decreased CBF by 41 % and 31 % in the pH-stat and the a-stat groups, respectively, the greater reduction in the pH-stat group being the result of a greater control value of P a ^ . CVR values increased concomitantly by 31% and 40%, respectively. During bypass at a venous blood temperature of 26 °C, there was a further 18 % reduction in CBF in the a-stat group, whereas in the pH-stat group CBF increased by 191 %, accompanied by a 59 % decrease in CVR. At the end of surgery, CBF values in both groups were identical again, although greater than those measured after induction of anaesthesia because of smaller PCV values. Individual CBF values at 26 °C in the a-stat group did not vary markedly, but those of the pH-stat group varied from 39 to 180 ml/100 g min"1 (fig. 2), which is most likely a result of loss of autoregulation of cerebral blood flow in this group. After induction of anaesthesia, CPP first decreased, but afterwards remained unchanged in the pH-state group, whereas in the a-stat group it decreased progressively, but not below the range of autoregulation. Table IV shows the control values and the responses of cerebral metabolism to anaesthesia under
DISCUSSION
Cerebral blood flow and metabolism
The mean control values for CBF (table III) in diis study are widiin die range of normal values obtained widi die Kety-Schmidt technique. Mean values for CMRo2 and CMRg (table IV) also agree widi awake values reported by odier investigators [13]. The effects of fentanyl-midazolam anaesthesia on cerebral haemodynamics and metabolism, which have been discussed in detail elsewhere [14], were characterized by decreases in CBF and, to a lesser degree, in CMRo2 and in CMRg. CMRos and CMRg changed to die same extent, indicating diat aerobic metabolism was not impaired by anaesdiesia. Hypodiermia (26 °C venous blood temperature) led to a further reduction in CBF in the a-stat group, whereas in die pH-stat group it increased markedly. The latter was the result of carbon dioxide mediated vasodilatation caused by addition of carbon dioxide to die oxygenator to maintain a temperaturecorrected PaCOf of 5.3 kPa ( P a ^ was increased to 8.4 kPa when measured in die blood-gas analyser at 37 °C). In die a-stat group, to whom carbon dioxide was not given (non-temperature-corrected P a ^ of
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
a-stat pH-stat FIG. 1. Individual Pacc. values of the a-stat and the pH-stat groups during CPB at 26 °C (not corrected for temperature).
normothermic and hypothermic conditions and after rewarming at die end of surgery. The control values of cerebral oxygen and glucose uptake were similar in both groups although arterial-cerebral venous oxygen content and glucose concentration differences ((Ca^ — CcvOt) and (Ca, —Ccvg), respectively) were smaller in the pH-stat group as a result of die greater control CBF. Anaesthesia led to a reduction of CMROj, by 29 % in the pH-stat and by 20 % in the a-stat group, followed by a further 58% and 61 % decrease, respectively, under die influence of hypothermic CPB. This is equivalent to Q10 values of 2.6 and 2.45, respectively (not significantly different). After rewarming, CMROj normalized in both groups. CMRg changed similarly: it decreased by 32% and 27% in the pH-stat and a-stat groups, respectively, after induction of anaesthesia, decreased by another 74% and 60%, respectively, at 26 °C venous blood temperature and regained its postinduction value after rewarming (no statistically significant differences between groups). (Ca^ — CcvOi) and (Ca, — Ccvf) reflect the changes in blood flow and metabolism. As CBF decreased more than CMRo2 and CMRg, (Ca^-Ccv,^) and (Cag — Ccvg) increased after induction of anaesthesia. In contrast, during hypothermia, when there was a stronger depression of metabolism than of CBF, bodi concentration differences became smaller in die a-stat group. This effect was even more pronounced in the pH-stat group. At die end of surgery, CBF was increased compared widi postinduction values, while metabolic parameters had normalized such tiat (Ca^ — CcvOj) and (Ca, — Ccvf) were still smaller tian before bypass. As cerebral venous Poz was greater than 3 kPa in bodi groups over die whole course of the study, it may be concluded diat cerebral metabolism was not affected by inadequacy of oxygen supply to demand.
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA
150 -
c E 100 -
CD O
50 -
5.2 kPa was maintained), hyperperfusion did not occur. The differences in CBF between the groups cannot be attributed to differences in cerebral perfusion pressure, perfusion flow rate or PCV during bypass. Several studies have been published on the relationship between P a ^ and CBF during hypothermic CPB [9, 10, 15-17]. They indicate that pHstat management produces passive cerebral vaso-
dilatation and overrides autoregulatory responses of cerebral vessels toflowand pressure changes and to metabolic demands of the brain. Under a-stat conditions, in contrast, flow correlates with CMRo t and is independent of cerebral perfusion pressure and perfusion flow rate in the ranges 30-100 mm Hg and 1-2 litre min"1 m"1,respectively. Recently, Hindman and colleagues [18] demonstrated that the differences between a-stat and pH-stat management in both CBF and CBF dynamics were eliminated at bypass flow rates of less than 70 ml kg"1 min"1 in rabbits. However, it is doubtful if bypass flow rates used in animal studies can be compared to those clinically used. In our study, perfusion flow rates were about 2 litre min"1 m~£ during hypothermia, which is equivalent to 50 ml kg"1 min"1. Nevertheless, mean global CBF values of our groups were distinctly different. Moreover, although we cannot separate the effects of pressure from those of flow, because we varied flow to maintain pressure between 60 and 100 mm Hg, the wide range of CBF values in the pH-stat group compared with those in the a-stat group (fig. 2) clearly demonstrates that autoregulation was better preserved in the latter group. We also observed uncoupling between flow and metabolism in the pH-stat group. At 26 °C venous blood temperature, CMRo2 was reduced to 39 % and 42 % of its postinduction value in the a-stat and pHstat groups, respectively. These values were not significantly different, although, theoretically, metabolic depression should be more distinct in the pHstat group, because enzymatic activity should be depressed not only by hypothermia but also by acidosis [19]. However, this conclusion was drawn from observations made in hibernators and is not necessarily valid in humans. Thus, in contrast with its effect on CBF, acid-base management did not affect cerebral metabolism.
TABLE IV. Cerebral metabolic variables (mean (SD)). / = Awake; II = after induction of anaesthesia; / / / = during CPB at 26 "C; IV = end of surgery. (Ca Of — CcvOj) = arterial—cerebral venous oxygen content difference; CMRot=* cerebral metabolic rate of oxygen; (Ca f — Ccv,) = arterial-cerebral venous glucose concentration difference; CMRg =• cerebral metabolic rate of glucose. Uncorr. = Values not corrected for temperature. P < 0.05: * compared with pH-stat; + / vs / / ; + / / vs III; 9II vs 1V;%1U vu IV Time of measurement
Pcvo (kPa) (Uncorr.) pH-stat a-stat (Cao-Ccv 0 i ) (ml dl-') pH-stat a-stat CMRo, (ml/lOOgmin"1) pH-stat a-stat (Ca,-Ccv,) (ml dl"1) pH-stat a-stat CMRg (mg/100 grain"1) pH-stat a-stat
I
II
III
IV
5.1(0.3) 5.0 (0.4)
4.5 (0.5)+ 4.7 (0.6)
10.6(1.7)$ 5.5 (0.4)*+.
4.5(0.6)$ 4.5(0.5)5$
5.90 (0.64) 6.69 (0.86)*
7.29(1.48)+ 7.63(1.33)+
1.09(0.38)+ 3.65 (0.78)*+
5.44(1.33)5$ 5.59(1.0O)f$
3.32 (0.67) 3.24 (0.65)
2.37 (0.55>t 2.59 (0.73)t
0.99 (0.46)+ 1.01 (0.25)+
2.35 (0.58)$ 2.34 (0.37)$
7.73(1.34) 8.87 (2.39)*
9.07 (2.79) 9.20 (2.60)
1.53(2.33)+ 4.53(1.91)*+
5.57(2.46)f$ 6.70 (2.08)f$
4.36(1.19) 4.24(1.08)
2.97 (0.99)+ 3.09 (0.97)+
0.78 (2.56)+ 1.25(0.51)+
2.35(0.76)«$ 3.00(1.84)$
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
a-stat pH-stat FIG. 2. Individual CBF values of the a-stat and the pH-stat groups during CPB at 26 °C.
55
56
Neurological outcome
The reported incidence of neurological or neuropsychological dysfunction after hypothermic bypass procedures ranges from 0% to 40% [24,25]. Possible causes during extracorporeal circulation may be: microembolization, macroembolization and ischaemia secondary to impaired oxygen delivery to the tissue. Mismatch between oxygen demand and supply during hypothermia may result from hypoperfusion, the hypothermia-induced leftward shift of the oxyhaemoglobin dissociation curve, an increased diffusion distance because of total cessation of blood flow in selected blood vessels, and arteriovenous shunting as a result of increased catecholamine concentrations and vasoconstriction [26]. Finally, various other factors not related to bypass may also affect neurological outcome, such as advanced age, preoperative neurological deficit and low cardiac output state. Theoretically, there are several mechanisms by which acid-base management might increase the risk of cerebral damage after hypothermic cardiopulmonary bypass: hyperperfusion associated with the pH-stat strategy may result in distribution of an increased proportion of microemboli to the brain rather than the systemic circulation [10] or may cause a steal phenomenon [15]. Alternatively, the astat approach may lead to hypoperfusion in selected patients and may limit tissue oxygen delivery by further shift of the oxyhaemoglobin dissociation curve to the left, although this is counterbalanced by the fact that at lower temperatures, diminished metabolic needs are met increasingly by dissolved oxygen, if flow is adequate [23]. As was shown above, metabolism was not flowlimited in either of the groups in the present study and there was no oxygen demand—supply mismatch during or after cardiopulmonary bypass. Nevertheless, hyperperfusion during hypothermia was extreme in the pH-stat group. Hence, it is not surprising that neurological disorders were significantly more frequent in the pH-stat group, although our results are contradictory to those of
Bashein and colleagues [11], who did not find any difference in neuropsychological outcome between patients assigned to a-stat or pH-stat management. Several factors may have contributed to these conflicting results. Neuropsychological outcome was assessed with series of tests in their study, while our patients were subjected to neurological examinations. We varied bypass flow to maintain perfusion pressure within certain limits while Bashein's group treated haemodynamic changes with phenylephrine or sodium nitroprusside, both of which may affect the cerebral pressure-flow relationship [27, 28]. The same authors studied patients undergoing both coronary artery bypass surgery and intracardiac procedures; they used bubble oxygenators and not arterial line filtration. These differences in study design may have accounted for a greater risk of cerebral embolism in both of Bashein's groups and thus may have masked the differential effects of acid-base management. Unfortunately, they did not measure CBF and cerebral metabolism and so any discussion about what may have caused these conflicting results remains speculative. In summary, although the acid—base management did not affect cerebral metabolism, the influence on CBF was striking: while pH-stat management resulted in luxury cerebral perfusion and loss of cerebral autoregulation, the a-stat approach was characterized by concomitant reductions in blood flow and metabolism during hypothermia. Moreover, the present study demonstrated that neurological damage was related to the pH-stat rather than the a-stat strategy. We conclude that the a-stat management should be favoured for conduct of hypothermic cardiopulmonary bypass procedures. ACKNOWLEDGEMENT This study was supported by the B. Braun Foundation, Melsungen, Germany. REFERENCES 1. Rahn H. Body temperature and acid base regulation. Pneumonologu 1974; 151: 87-94. 2. Swan H. The importance of acid-base management for cardiac and cerebral preservation during open heart operations. Surgery, Gymcology & Obstetrics 1984; 158: 391-414. 3. Swain JA, McDonald TJ jr, Robbins RC, Hampshire VA. Hemodynamics and metabolism during surface-induced hypothermia in the dog: a comparison of pH management strategies. Journal of Surgical Research 1990; 48: 217-222. 4. Willford DC, Moores WY, Ji S, Tung Chen Z, Palencia A, Daily PO. Importance of acid-base strategy in reducing myocardial and whole body oxygen consumption during perfusion hypothermia. Journal of Thoracic and Cardiovascular Surgery 1990; 100: 699-707. 5. Alston RP, Singh M, McLaren AD. Systemic oxygen uptake during hypothermic cardiopulmonary bypass: effects of flow rate, flow character, and arterial pH. Journal of Thoracic and Cardiovascular Surgery 1989; 98: 757-768. 6. Tuppurainen T, Settergren G, Stensved P. The effect of arterial pH on whole body oxygen uptake during hypothermic cardiopulmonary bypass in man. Journal of Thoracic and Cardiovascular Surgery 1989; 98: 769-773. 7. Sinet M, Muffat-Joly M, Bendaace T, Pocidalo JJ. Maintaining blood pH at 7.4 during hypothermia has no significant effect on work of the isolated rat heart. Anesthesiology 1985; 62: 582-587.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
Similar results have been obtained by Murkin and colleagues [9], although in their study cerebral metabolic activity at 26 °C was reduced to 25 % of its normothermic value. According to our investigation, the mean Q10 value of the anaesthetized human brain was 2.5, which implies that each 10 °C change in temperature altered cerebral oxygen demand by a factor of 2.5. This is in close agreement with Q lo values found by other investigators in brain and whole body [20-22]. In the present study, the metabolic rate of the brain was matched with both acid-base regimens. This can be inferred not only from an equal reduction in CMROj and CMRg during hypothermia in both groups, but also from the fact that cerebral venous Pot never reached a critical value, which has been found to decrease to less than 2.6 kPa with lesser temperatures [23]. This indicates that metabolism was not flow limited. Moreover, after rewarming, cerebral metabolic activity regained its prebypass value, irrespective of the acid-base strategy.
BRITISH JOURNAL OF ANAESTHESIA
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA
19. 20. 21.
22.
23. 24.
25.
26.
27.
28.
between alpha-stat and pH-stat management are eliminated during periods of decreased systemic flow and pressure. Anesthesiology 1991; 74: 1096-1102. White FN. A comparative physiological approach to hypothermia. Journal of Thoracic and Cardiovascular Surgery 1981; 82: 821-831. Michcnfelder JD, Theye RA. Hypothermia: effect on canine brain and whole-body metabolism. Anesthesiology 1968; 29: 1107-1112. Fox LS, Blackstone EH, Kirklin JW, Stewart RW, Samuelson PN. Relationship of whole body oxygen consumption to perfusion flow rate during hypothermic cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery 1982; 83: 239-248. Mundemann A, Stephan H, Weyland A, Wellhausen A, Sonntag H. Effect of acid-base management on whole-body oxygen uptake during hypothermic cardiopulmonary bypass in man. Anaesthesist 1991; 40: 530-536. Willford DC, Hill EP, Moore WY. Theoretical analysis of oxygen transport during hypothermia. Journal of Clinical Monitoring 1986; 2: 3CM3. Ellis RJ, Wisniewski A, Potts R, Calhoun C, Loucks P, Wells MR. Reduction offlowrate and arterial pressure at moderate hypothermia does not result in cerebral dysfunction. Journal of Thoracic and Cardiovascular Surgery 1980; 79: 173-180. Kolkka R, Hilberman M. Neurologic dysfunction following cardiac operation with low-flow, low-pressure cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery 1980; 79: 432-^37. Bridges KG, Reichard GA, Macraugh H, Kues JR, Cevallos WH, Lechmann MJ, Hoffman WS, Donahoo JS. Effect of phentolamine in controlling temperature and acidosis associated with cardiopulmonary bypass. Critical Care Medicine 1985; 13: 72-76. Patel PM, Mutch WAC. The cerebral pressure-flow relationship during 1.0 MAC isoflurane anesthesia in the rabbit: the effect of different vasopressors. Anesthesiology 1990; 72: 118-124. Michenfelder JD, Milde JH. The interaction of sodium nitroprusside, hypotension, and isoflurane in determining cerebral vasculature effects. Anesthesiology 1988; 69:870-875.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
8. Becker H, Vinten-Johansen J, Buckberg GD, Robertson JM, Leaf JD, Lazar HL, Manganaro AJ. Myocardial damage caused by keeping pH 7.40 during deep hypothermia. Journal of Thoracic and Cardiovascular Surgery 1981; 82: 810-820. 9. Murkin JM, Farrar JK, Tweed WA, McKenzie FN, Guiraudon GM, Guiraudon G. Cerebral autoregulation and flow/metabolism coupling during hypothermic cardiopulmonary bypass: the influence of P a ^ . Anesthesia and Analgesia 1987; 66: 825-832. 10. Henriksen L. Brain luxury perfusion during cardiopulmonary bypass in humans. A study of the cerebral blood flow response to changes in CO,, Ov and blood pressure. Journal of Cerebral Blood Flow and Metabolism 1986; 6: 366-378. 11. Bashein G, Townes BD, Nessly ML, Bledsoe SW, Horabein TF, Davis KB, Goldstein DE, Coppel DB. A randomized study of carbon dioxide management during hypothermic cardiopulmonary bypass. Anesthesiology 1990; 72: 7-15. 12. Prough DS, Stump DA, Todd Troost B. P a ^ management during cardiopulmonary bypass; intriguing physiologic rationale, convincing clinical data, evolving hypothesis. Anesthesiology 1990; 72: 3-6. 13. Smith AL, Wollman H. Cerebral bloodflowand metabolism: effects of anesthetic drugs and techniques. Anesthesiology 1972; 36: 378-^100. 14. Stephan H, Sonntag H, Lange H, Rieke H. Cerebral effects of anaesthesia and hypothermia. Anaesthesia 1989; 44: 310-316. 15. Prough DS, Stump DA, Roy RC, Gravlee GP, Williams T, Mills SA, Hinshelwood L, Howard G. Response of cerebral blood flow to changes in carbon dioxide tension during hypothermic cardiopulmonary bypass. Anesthesiology 1986; 64: 576-581. 16. Govier AV, Reves JG, McKay RD, Karp RB, Zorn GL, Morawetz RB, Smith LR, Adams M, Freeman AM. Factors and their influence on regional cerebral blood flow during nonpulsatile cardiopulmonary bypass. Annals of Thoracic Surgery 1984;' 38: 592-600. 17. Soma Y, Hirotani T, Yozu R, Onoguchi K, Misumi T, Kawada K, Inoue T, Mohri H. A clinical study of cerebral circulation during extracorporeal circulation. Journal of Thoracic and Cardiovascular Surgery 1989; 97: 187-193. 18. Hindman BJ, Funatsu N, Harrington J, Cutkomp J, Miller T, Todd MM, Tinker JH. Differences in cerebral blood flow
57
ACID-BASE MANAGEMENT DURING HYPOTHERMIC CARDIOPULMONARY BYPASS DOES NOT AFFECT CEREBRAL METABOLISM BUT DOES AFFECT BLOOD FLOW AND NEUROLOGICAL OUTCOME H. STEPHAN, A. WEYLAND, S. KAZMAIER, T. HENZE, S. MENCK AND H. SONNTAG SUMMARY
PATIENTS AND METHODS KEY WORDS Acid-base equilibrium: intraoperative management. Brain: blood flow, metabolism. Hypothermia. Surgery: cardiopu/monary bypass.
We studied 65 male patients (ages 38-68 yr, weight 51-110 kg) undergoing elective coronary artery bypass surgery. The study was approved by the
Optimal acid—base management during hypothermic cardiopulmonary bypass (CPB) has been a matter of discussion for many years. According to the pH-stat theory, arterial pH is maintained constant at 7.40 at
H. STEPHAN, M.D., A . WEYLAND, M.D., S. KAZMAIER, M.D., H. SONNTAG, M.D. (Department of Anaesthesiology); T. HENZE,
M.D., S. MENCK, M.D. (Department of Neurology); University of Gottingen, Robert Koch-Str. 40, 3400 GSttingen, Germany. Accepted for Publication: January 3, 1992.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
In order to compare the effects of blood-gas management on cerebral blood flow, metabolism and neurological outcome after hypothermic cardiopulmonary bypass (CPB) ,-we have studied 65 patients undergoing aorto-coronary bypass surgery allocated randomly to either a pH-stat (temperature-corrected blood-gas management) or an a-stat (temperature-uncorrected blood-gas management) group. All patients were examined neurologically on the day before and the 7th day after operation. In 20 patients of the pH-stat group and in 15 patients of the a-stat group we measured cerebral blood flow (CBF), using the argon washin technique, and also cerebral oxygen (CMRoJ and glucose (CMRg) uptake. Measurements were performed in awake patients, after induction of anaesthesia with fentanyl, midazo/am and pancuronium under normothermic conditions, during CPB at a venous blood temperature of 26 °C and at the end of surgery. Compared with postinduction values, hypothermia was associated with an 18% reduction in CBF and decreases in CMRO2 and CMRg of 61 % and 60%, respectively, in the a-stat group. In the pH-stat group, CMRo2 and CMRg decreased also, by 58% and 74%, respectively, whereas CBF increased by 191%, indicating uncoupling of flow and metabolism. As there were no statistically significant differences between the metabolic variables in both groups, we conclude that acidbase management did not affect cerebral metabolism, despite its influence on blood flow. After rewarming, CBF and cerebral metabolism normalized independently of acid-base management during hypothermia. Nevertheless, neurological dysfunction occurred more often in the pH-stat group (P = 0.036).
all temperatures. Currently, the a-stat approach is favoured, which allows arterial pH to increase with decreasing temperature according to the theory of Rahn [1]. With this regulation, arterial pH stays close to the value that represents the biological neutrality of blood. Hence, cell function should be better preserved with the a-stat concept [2]. With pH-stat management, enzymatic activity should be depressed by acidosis, in addition to the metabolic depression caused by hypothermia. On the other hand, acidosis may partly counteract the leftward shift in the oxygen dissociation curve which occurs in hypothermia, such that tissue oxygenation might be better preserved with the pH-stat approach. Studies in man and animals on the effect of acid—base management on myocardial function, systemic and myocardial metabolism have yielded conflicting results [3-8]. Although cerebral hyperperfusion associated with pH-stat management results in loss of CBF autoregulation and uncoupling of flow and metabolism [9], and may deliver more microemboli to the brain [10] or cause a steal phenomenon in patients with regional cerebral ischaemia [9], Bashein and colleagues [11] did not find any clinically significant effects of acid-base management on neurobehavioural outcome of patients who had undergone hypothermic cardiopulmonary bypass. However, an important objection against this study was that there was no correlation between neurological outcome and measured cerebral circulatory and metabolic changes during CPB [12]. The present study was designed to investigate the effects of both acid-base managements on cerebral blood flow, metabolism and neurological outcome after hypothermic cardiopulmonary bypass.
BRITISH JOURNAL OF ANAESTHESIA
52 TABLE I. Patient data (mean (range or SD) or number)
Management pH-stat No. patients Age (yr) Weight (kg) Height (cm) Body surface area (ml) Diabetes mellitus (n) Hypertension (n) Cold (n) Ejection fraction < 0.4 (n) CPB time (min) Aortic cross-clamp time (min) Balloon pump (n) Re-exploration (n) Time to tracheal extubation (h) Time to ICU discharge (h)
35
55(38-68) 82 (12) 175(7) 1.98(0.17) 6 20
a-stat 30
56 ((42-66) 80(11) 174(7) 1.94(0.16) 5 19
5
2
9
14
109 (32) 62 (20) 1 1
18(7) 41 (22)
109(34) 58(21) 1 1
19(18) 42(32)
Catheterization procedure
In the anaesthetic room, ECG leads and five-lead EEG (Lifescan, Neurometrics Inc.) were attached. The following catheters were inserted percutaneously under local anaesthesia: a 20-gauge catheter in the radial artery of the non-dominant hand to monitor arterial pressure and for blood sampling; a flow-directed pulmonary artery catheter (Edwards quadruple thermodilution model no. 93A 131-7F) via an antecubital vein into the pulmonary artery for measurement of pulmonary artery and wedge pressures and cardiac output; and a polyethylene catheter into the superior vena cava for administration of drugs and infusions. In patients in whom CBF was measured, the pressure line of the arterial catheter was replaced by a gas-tight Goodale-Lubin catheter (6-French gauge, USCI) and a second Goodale—Lubin catheter of the same size was positioned retrogradely into the superior bulb of the right internal jugular vein for measurement of CBF and jugular bulb pressure and withdrawal of blood samples. The position of this catheter was confirmed radiologically. Expired car-
Neurological examination
Neurological examination was performed by two trained neurologists who were blinded to the study. This conformed to the standard neurological examination performed in clinical practice and included evaluation of cranial nerve function, co-ordination, pain, temperature and light touch sensation, muscle strength, reflexes, wakefulness, recent and remote memory and presence of hallucinations or disorientation. Study periods and anaesthesia
Measurements were performed in the awake patient after a period of rest following catheterization (I), after induction of anaesthesia under normothermic conditions (II), 30 min after start of bypass at stable venous and arterial blood temperatures of 26 °C (III) and at the end of surgery (IV). Anaesthesia was induced with fentanyl 6 ug kg"1 and midazolam 0.2 mg kg"1 and maintained with fentanyl 0.15 ug kg"1 min"1 and midazolam 3 ug kg"1 min"1. Pancuronium 8 mg was administered to facilitate tracheal intubation and the lungs were ventilated with oxygen in air using a constant volume respirator (Engstrom ER 300). Incremental doses of fentanyl 0.25 mg and midazolam 7.5 mg were given immediately before sternotomy and when lightening of anaesthesia was apparent in the EEG. Cardiopulmonary bypass
The extracorporeal circulation technique consisted of a Polystan heart-lung machine with a membrane oxygenator (Maxima). Forty-micrometre filters (Sartorius, Gottingen) were used in the cardiotomy suction and the arterial line. Priming comprised Ringer's lactate 1000 ml, 5% glucose 500 ml, 20 % human albumin solution 400 ml, and sodium bicarbonate 100 mmol. Flow was maintained at 1.7-2.4 litre min"1 m"1 during bypass. Bretschneider's cardioplegic solution, at 4 °C, was infused into the aortic root after the aorta had been cross-clamped and was removed from the right atrium to prevent uptake into systemic circulation. In the pH-stat group, carbon dioxide was added to the fresh gas flow to maintain a temperaturecorrected PacOt of approximately 5.3 kPa, while in the a-stat group, no carbon dioxide was used and the fresh gas flow was varied to keep the Paco, near 5.3 kPa measured at 37 °C (not temperature-corrected). Paco was monitored continuously using an in-line blood-gas analysis sensor (Cardiovascular Devices Inc., Irvine, CA). CBF measurement and variables
CBF was measured using the argon washin technique. This method includes simultaneous blood sampling from the jugular bulb and the radial artery
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
Gottingen University Human Subjects Review Committee and written informed consent was obtained from all patients. Patients were allocated randomly to one of two groups managed with either the pHstat (group 1, n = 35) or the a-stat approach (group 2, n = 30) during hypothermic CPB. All patients were examined neurologically on the day before and the 7th day after operation. Cerebral blood flow and metabolism were studied in 20 patients of group 1 and in 15 patients of group 2. None of the patients suffered from pre-existing neurological disorders and there were no significant differences between the characteristics of both groups (table I). No patient had a history of valvular heart disease, liver disease or renal impairment. All patients were receiving maintenance doses of calcium channel blocking drugs, beta adrenoceptor antagonists and nitrates. Premedication consisted of fiunitrazepam 1-2 mg orally and piritramide 7.5—15 mg and promethazine 25-50 mg i.m., 1 h before arrival in the anaesthetic room.
bon dioxide concentration was measured with a Normocap CO2 Analyzer (Datex CD 102/02 Helsinki). The ECG and all pressures were monitored continuously and recorded simultaneously on a 10-channel chart recorder (Hellige, Freiburg). The EEG was also monitored continuously and recorded using the Lifescan printer.
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA TABLE II. Incidence of neurological sequelae
Management pH-stat (n = 35) Cerebellar symptoms (nystagmus, ataxia, dysdiadochokinesia etc.) Disorientation, confusion Right hemiparesis Illrd and Vlth nerve paresis Xllth nerve paresis
a-stat (n = 30)
5 1 1 2 1
The cerebral metabolic rates (CMR) of oxygen and glucose were calculated by multiplying the arterial-cerebral venous blood oxygen content and substrate concentration differences, respectively, by CBF. Statistics All results are expressed as mean (SD). Statistical analyses of the data obtained in each group were performed using the Friedman two-way ANOVA and the Wilcoxon matched-pairs signed-ranks test. The Mann-Whitney U test was used for intergroup comparisons. Factors that might predict neurological dysfunction were subjected to a multivariate analysis by a stepwise logistic regression procedure. P < 0.05 was assigned statistical significance. RESULTS
The characteristics of the patients in the pH-stat and in the a-stat group were similar (table I). Ten of 35 patients in the pH-stat group and two of 30 patients in the a-stat group were found to suffer from neurological sequelae on the 7th day after operation (table II); these had not existed before operation. Most of these sequelae were the result of focal infarcts in regions supplied by the basilar artery. One patient in each group was disoriented and confused at the time of the neurological examination— symptoms of global cerebral damage (encephalopathy). Multivariate analysis, by stepwise logistic regression procedure, revealed that the occurrence of neurological dysfunction did not relate to age (P =
TABLE III. Haemodynamic variables, PaCOf values and arterial hydrogen ion concentrations {mean {SD)). I = Awake; II = after induction of anaesthesia; / / / = during CPB at 26 "C; IV = end of surgery. CPP = Cerebral perfusion pressure; CVR = cerebral vascular resistance; CBF = cerebral blood flow; CI = cardiac index; [H+a] = arterial hydrogen ion concentration. Uncorr. = Values not corrected for temperature. P < 0.05: * compared with pH- slat; + /vs / / ; + //vs / / / ; S//vs/K;$///vs/K Time of measurement
CPP (mmHg) pH-stat a-stat CVR (mm Hg (ml/100 g min"1)"1) pH-stat a-stat CBF (ml/100 gmin"1) pH-stat a-stat PCV (%) pH-stat a-stat CI (litre min"1 m"1) pH-stat a-stat Pa~ (kPa) (Uncorr.) pH-stat a-stat [H+a] (nmol litre"1) (Uncorr.) pH-stat a-stat
I
II
III
91 (12) 92(17)
70(11)+ 89 (17)*
71 (15) 69 (14)+.
72 (12) 64(12)*f$
1.66(0.35) 1.96(0.50)
2.17(0.45)+ 2.75 (0.76)*+
0.89(0.51)$ 2.53(0.81)+.*
1.69(0.45)«$ 1.56(0.43)«
IV
56(10) 49(9)*
33 (4)+ 34(8)+
96 (39)+ 28 (5)+*
44 (8)« 43(9)«
42(4) 41(3)
38(4)+ 38(4)+
26(2)+ 25 (4)+.
29(2)i$ 27(5)f
3.3 (0.6) 2.7 (0.5)*
2.4(0.3>t 2.5 (0.6)+
2.0 (0.2)+. 2.1(0.2)+.
2.7(0.4*$ 2.9(0.5)§S
5.9 (0.4) 5.5 (0.4)*
5.5(0.3>t 5.1(0.4)*+
8.4(0.7)+. 5.2 (0.3)*
5.5(0.5* 5.3 (0.4)
57.9(4.7)+. 40.8 (2.7)*
40.0(3.9)5$ 42.4(3.7)*
44.3(2.1) 42.4(3.1)
41.9(2.8)+ 40.9 (2.9)+
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
during a 10-min period of inhalation of a standard concentration of argon. Argon was administered to the awake patient via a facemask, to the anaesthetized patient via the tracheal tube, and directly into the oxygenator during bypass. Immediately before and after each measurement of CBF, blood samples were taken from the jugular bulb and the radial artery and analysed for haemoglobin concentration, oxygen saturation (COOximeter IL 282), blood-gas tensions (standard electrodes, Radiometer), glucose concentration (standard test combination, Boehringer Mannheim) and electrolyte concentrations (absorption spectrometry, Perkin Elmer 303). Cardiac output was measured by thermodilution (cardiac output computer : Fischer BN 7206). Cerebral perfusion pressure (CPP) was calculated as mean arterial pressure minus jugular bulb pressure, and cerebral vascular resistance (CVR) as CPP divided by CBF. Cardiac index was calculated by dividing cardiac output by the body surface area.
53
BRITISH JOURNAL OF ANAESTHESIA
54
9-
710
O O
5-
I
3-
0.5599), diabetes mellitus (P = 0.4470), hypertension (P = 0.2006), cold (P = 0.6892), ejection fraction (P = 0.2481), CPB time (P = 0.644), aortic cross-clamp time (P = 0.0624) or use of an intra-aortic balloon pump (P = 0.4789), but only to the acid-base management (P = 0.0360). Haemodynamic and blood-gas data of those 35 patients in whom CBF and cerebral metabolism were studied are presented in table III. According to the design of the study, PaCOi and arterial hydrogen ion concentrations remained nearly constant over the course of the study in the a-stat group (n = 15) while in the pH-stat group (n = 20), P a ^ increased by 53 % at 26 °C venous blood temperature (not corrected for temperature). The individual PaCOj values of both groups were distinctly different at 26 °C (fig. 1) Anaesthesia decreased CBF by 41 % and 31 % in the pH-stat and the a-stat groups, respectively, the greater reduction in the pH-stat group being the result of a greater control value of P a ^ . CVR values increased concomitantly by 31% and 40%, respectively. During bypass at a venous blood temperature of 26 °C, there was a further 18 % reduction in CBF in the a-stat group, whereas in the pH-stat group CBF increased by 191 %, accompanied by a 59 % decrease in CVR. At the end of surgery, CBF values in both groups were identical again, although greater than those measured after induction of anaesthesia because of smaller PCV values. Individual CBF values at 26 °C in the a-stat group did not vary markedly, but those of the pH-stat group varied from 39 to 180 ml/100 g min"1 (fig. 2), which is most likely a result of loss of autoregulation of cerebral blood flow in this group. After induction of anaesthesia, CPP first decreased, but afterwards remained unchanged in the pH-state group, whereas in the a-stat group it decreased progressively, but not below the range of autoregulation. Table IV shows the control values and the responses of cerebral metabolism to anaesthesia under
DISCUSSION
Cerebral blood flow and metabolism
The mean control values for CBF (table III) in diis study are widiin die range of normal values obtained widi die Kety-Schmidt technique. Mean values for CMRo2 and CMRg (table IV) also agree widi awake values reported by odier investigators [13]. The effects of fentanyl-midazolam anaesthesia on cerebral haemodynamics and metabolism, which have been discussed in detail elsewhere [14], were characterized by decreases in CBF and, to a lesser degree, in CMRo2 and in CMRg. CMRos and CMRg changed to die same extent, indicating diat aerobic metabolism was not impaired by anaesdiesia. Hypodiermia (26 °C venous blood temperature) led to a further reduction in CBF in the a-stat group, whereas in die pH-stat group it increased markedly. The latter was the result of carbon dioxide mediated vasodilatation caused by addition of carbon dioxide to die oxygenator to maintain a temperaturecorrected PaCOf of 5.3 kPa ( P a ^ was increased to 8.4 kPa when measured in die blood-gas analyser at 37 °C). In die a-stat group, to whom carbon dioxide was not given (non-temperature-corrected P a ^ of
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
a-stat pH-stat FIG. 1. Individual Pacc. values of the a-stat and the pH-stat groups during CPB at 26 °C (not corrected for temperature).
normothermic and hypothermic conditions and after rewarming at die end of surgery. The control values of cerebral oxygen and glucose uptake were similar in both groups although arterial-cerebral venous oxygen content and glucose concentration differences ((Ca^ — CcvOt) and (Ca, —Ccvg), respectively) were smaller in the pH-stat group as a result of die greater control CBF. Anaesthesia led to a reduction of CMROj, by 29 % in the pH-stat and by 20 % in the a-stat group, followed by a further 58% and 61 % decrease, respectively, under die influence of hypothermic CPB. This is equivalent to Q10 values of 2.6 and 2.45, respectively (not significantly different). After rewarming, CMROj normalized in both groups. CMRg changed similarly: it decreased by 32% and 27% in the pH-stat and a-stat groups, respectively, after induction of anaesthesia, decreased by another 74% and 60%, respectively, at 26 °C venous blood temperature and regained its postinduction value after rewarming (no statistically significant differences between groups). (Ca^ — CcvOi) and (Ca, — Ccvf) reflect the changes in blood flow and metabolism. As CBF decreased more than CMRo2 and CMRg, (Ca^-Ccv,^) and (Cag — Ccvg) increased after induction of anaesthesia. In contrast, during hypothermia, when there was a stronger depression of metabolism than of CBF, bodi concentration differences became smaller in die a-stat group. This effect was even more pronounced in the pH-stat group. At die end of surgery, CBF was increased compared widi postinduction values, while metabolic parameters had normalized such tiat (Ca^ — CcvOj) and (Ca, — Ccvf) were still smaller tian before bypass. As cerebral venous Poz was greater than 3 kPa in bodi groups over die whole course of the study, it may be concluded diat cerebral metabolism was not affected by inadequacy of oxygen supply to demand.
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA
150 -
c E 100 -
CD O
50 -
5.2 kPa was maintained), hyperperfusion did not occur. The differences in CBF between the groups cannot be attributed to differences in cerebral perfusion pressure, perfusion flow rate or PCV during bypass. Several studies have been published on the relationship between P a ^ and CBF during hypothermic CPB [9, 10, 15-17]. They indicate that pHstat management produces passive cerebral vaso-
dilatation and overrides autoregulatory responses of cerebral vessels toflowand pressure changes and to metabolic demands of the brain. Under a-stat conditions, in contrast, flow correlates with CMRo t and is independent of cerebral perfusion pressure and perfusion flow rate in the ranges 30-100 mm Hg and 1-2 litre min"1 m"1,respectively. Recently, Hindman and colleagues [18] demonstrated that the differences between a-stat and pH-stat management in both CBF and CBF dynamics were eliminated at bypass flow rates of less than 70 ml kg"1 min"1 in rabbits. However, it is doubtful if bypass flow rates used in animal studies can be compared to those clinically used. In our study, perfusion flow rates were about 2 litre min"1 m~£ during hypothermia, which is equivalent to 50 ml kg"1 min"1. Nevertheless, mean global CBF values of our groups were distinctly different. Moreover, although we cannot separate the effects of pressure from those of flow, because we varied flow to maintain pressure between 60 and 100 mm Hg, the wide range of CBF values in the pH-stat group compared with those in the a-stat group (fig. 2) clearly demonstrates that autoregulation was better preserved in the latter group. We also observed uncoupling between flow and metabolism in the pH-stat group. At 26 °C venous blood temperature, CMRo2 was reduced to 39 % and 42 % of its postinduction value in the a-stat and pHstat groups, respectively. These values were not significantly different, although, theoretically, metabolic depression should be more distinct in the pHstat group, because enzymatic activity should be depressed not only by hypothermia but also by acidosis [19]. However, this conclusion was drawn from observations made in hibernators and is not necessarily valid in humans. Thus, in contrast with its effect on CBF, acid-base management did not affect cerebral metabolism.
TABLE IV. Cerebral metabolic variables (mean (SD)). / = Awake; II = after induction of anaesthesia; / / / = during CPB at 26 "C; IV = end of surgery. (Ca Of — CcvOj) = arterial—cerebral venous oxygen content difference; CMRot=* cerebral metabolic rate of oxygen; (Ca f — Ccv,) = arterial-cerebral venous glucose concentration difference; CMRg =• cerebral metabolic rate of glucose. Uncorr. = Values not corrected for temperature. P < 0.05: * compared with pH-stat; + / vs / / ; + / / vs III; 9II vs 1V;%1U vu IV Time of measurement
Pcvo (kPa) (Uncorr.) pH-stat a-stat (Cao-Ccv 0 i ) (ml dl-') pH-stat a-stat CMRo, (ml/lOOgmin"1) pH-stat a-stat (Ca,-Ccv,) (ml dl"1) pH-stat a-stat CMRg (mg/100 grain"1) pH-stat a-stat
I
II
III
IV
5.1(0.3) 5.0 (0.4)
4.5 (0.5)+ 4.7 (0.6)
10.6(1.7)$ 5.5 (0.4)*+.
4.5(0.6)$ 4.5(0.5)5$
5.90 (0.64) 6.69 (0.86)*
7.29(1.48)+ 7.63(1.33)+
1.09(0.38)+ 3.65 (0.78)*+
5.44(1.33)5$ 5.59(1.0O)f$
3.32 (0.67) 3.24 (0.65)
2.37 (0.55>t 2.59 (0.73)t
0.99 (0.46)+ 1.01 (0.25)+
2.35 (0.58)$ 2.34 (0.37)$
7.73(1.34) 8.87 (2.39)*
9.07 (2.79) 9.20 (2.60)
1.53(2.33)+ 4.53(1.91)*+
5.57(2.46)f$ 6.70 (2.08)f$
4.36(1.19) 4.24(1.08)
2.97 (0.99)+ 3.09 (0.97)+
0.78 (2.56)+ 1.25(0.51)+
2.35(0.76)«$ 3.00(1.84)$
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
a-stat pH-stat FIG. 2. Individual CBF values of the a-stat and the pH-stat groups during CPB at 26 °C.
55
56
Neurological outcome
The reported incidence of neurological or neuropsychological dysfunction after hypothermic bypass procedures ranges from 0% to 40% [24,25]. Possible causes during extracorporeal circulation may be: microembolization, macroembolization and ischaemia secondary to impaired oxygen delivery to the tissue. Mismatch between oxygen demand and supply during hypothermia may result from hypoperfusion, the hypothermia-induced leftward shift of the oxyhaemoglobin dissociation curve, an increased diffusion distance because of total cessation of blood flow in selected blood vessels, and arteriovenous shunting as a result of increased catecholamine concentrations and vasoconstriction [26]. Finally, various other factors not related to bypass may also affect neurological outcome, such as advanced age, preoperative neurological deficit and low cardiac output state. Theoretically, there are several mechanisms by which acid-base management might increase the risk of cerebral damage after hypothermic cardiopulmonary bypass: hyperperfusion associated with the pH-stat strategy may result in distribution of an increased proportion of microemboli to the brain rather than the systemic circulation [10] or may cause a steal phenomenon [15]. Alternatively, the astat approach may lead to hypoperfusion in selected patients and may limit tissue oxygen delivery by further shift of the oxyhaemoglobin dissociation curve to the left, although this is counterbalanced by the fact that at lower temperatures, diminished metabolic needs are met increasingly by dissolved oxygen, if flow is adequate [23]. As was shown above, metabolism was not flowlimited in either of the groups in the present study and there was no oxygen demand—supply mismatch during or after cardiopulmonary bypass. Nevertheless, hyperperfusion during hypothermia was extreme in the pH-stat group. Hence, it is not surprising that neurological disorders were significantly more frequent in the pH-stat group, although our results are contradictory to those of
Bashein and colleagues [11], who did not find any difference in neuropsychological outcome between patients assigned to a-stat or pH-stat management. Several factors may have contributed to these conflicting results. Neuropsychological outcome was assessed with series of tests in their study, while our patients were subjected to neurological examinations. We varied bypass flow to maintain perfusion pressure within certain limits while Bashein's group treated haemodynamic changes with phenylephrine or sodium nitroprusside, both of which may affect the cerebral pressure-flow relationship [27, 28]. The same authors studied patients undergoing both coronary artery bypass surgery and intracardiac procedures; they used bubble oxygenators and not arterial line filtration. These differences in study design may have accounted for a greater risk of cerebral embolism in both of Bashein's groups and thus may have masked the differential effects of acid-base management. Unfortunately, they did not measure CBF and cerebral metabolism and so any discussion about what may have caused these conflicting results remains speculative. In summary, although the acid—base management did not affect cerebral metabolism, the influence on CBF was striking: while pH-stat management resulted in luxury cerebral perfusion and loss of cerebral autoregulation, the a-stat approach was characterized by concomitant reductions in blood flow and metabolism during hypothermia. Moreover, the present study demonstrated that neurological damage was related to the pH-stat rather than the a-stat strategy. We conclude that the a-stat management should be favoured for conduct of hypothermic cardiopulmonary bypass procedures. ACKNOWLEDGEMENT This study was supported by the B. Braun Foundation, Melsungen, Germany. REFERENCES 1. Rahn H. Body temperature and acid base regulation. Pneumonologu 1974; 151: 87-94. 2. Swan H. The importance of acid-base management for cardiac and cerebral preservation during open heart operations. Surgery, Gymcology & Obstetrics 1984; 158: 391-414. 3. Swain JA, McDonald TJ jr, Robbins RC, Hampshire VA. Hemodynamics and metabolism during surface-induced hypothermia in the dog: a comparison of pH management strategies. Journal of Surgical Research 1990; 48: 217-222. 4. Willford DC, Moores WY, Ji S, Tung Chen Z, Palencia A, Daily PO. Importance of acid-base strategy in reducing myocardial and whole body oxygen consumption during perfusion hypothermia. Journal of Thoracic and Cardiovascular Surgery 1990; 100: 699-707. 5. Alston RP, Singh M, McLaren AD. Systemic oxygen uptake during hypothermic cardiopulmonary bypass: effects of flow rate, flow character, and arterial pH. Journal of Thoracic and Cardiovascular Surgery 1989; 98: 757-768. 6. Tuppurainen T, Settergren G, Stensved P. The effect of arterial pH on whole body oxygen uptake during hypothermic cardiopulmonary bypass in man. Journal of Thoracic and Cardiovascular Surgery 1989; 98: 769-773. 7. Sinet M, Muffat-Joly M, Bendaace T, Pocidalo JJ. Maintaining blood pH at 7.4 during hypothermia has no significant effect on work of the isolated rat heart. Anesthesiology 1985; 62: 582-587.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
Similar results have been obtained by Murkin and colleagues [9], although in their study cerebral metabolic activity at 26 °C was reduced to 25 % of its normothermic value. According to our investigation, the mean Q10 value of the anaesthetized human brain was 2.5, which implies that each 10 °C change in temperature altered cerebral oxygen demand by a factor of 2.5. This is in close agreement with Q lo values found by other investigators in brain and whole body [20-22]. In the present study, the metabolic rate of the brain was matched with both acid-base regimens. This can be inferred not only from an equal reduction in CMROj and CMRg during hypothermia in both groups, but also from the fact that cerebral venous Pot never reached a critical value, which has been found to decrease to less than 2.6 kPa with lesser temperatures [23]. This indicates that metabolism was not flow limited. Moreover, after rewarming, cerebral metabolic activity regained its prebypass value, irrespective of the acid-base strategy.
BRITISH JOURNAL OF ANAESTHESIA
CEREBRAL EFFECTS OF ACID-BASE MANAGEMENT IN HYPOTHERMIA
19. 20. 21.
22.
23. 24.
25.
26.
27.
28.
between alpha-stat and pH-stat management are eliminated during periods of decreased systemic flow and pressure. Anesthesiology 1991; 74: 1096-1102. White FN. A comparative physiological approach to hypothermia. Journal of Thoracic and Cardiovascular Surgery 1981; 82: 821-831. Michcnfelder JD, Theye RA. Hypothermia: effect on canine brain and whole-body metabolism. Anesthesiology 1968; 29: 1107-1112. Fox LS, Blackstone EH, Kirklin JW, Stewart RW, Samuelson PN. Relationship of whole body oxygen consumption to perfusion flow rate during hypothermic cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery 1982; 83: 239-248. Mundemann A, Stephan H, Weyland A, Wellhausen A, Sonntag H. Effect of acid-base management on whole-body oxygen uptake during hypothermic cardiopulmonary bypass in man. Anaesthesist 1991; 40: 530-536. Willford DC, Hill EP, Moore WY. Theoretical analysis of oxygen transport during hypothermia. Journal of Clinical Monitoring 1986; 2: 3CM3. Ellis RJ, Wisniewski A, Potts R, Calhoun C, Loucks P, Wells MR. Reduction offlowrate and arterial pressure at moderate hypothermia does not result in cerebral dysfunction. Journal of Thoracic and Cardiovascular Surgery 1980; 79: 173-180. Kolkka R, Hilberman M. Neurologic dysfunction following cardiac operation with low-flow, low-pressure cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery 1980; 79: 432-^37. Bridges KG, Reichard GA, Macraugh H, Kues JR, Cevallos WH, Lechmann MJ, Hoffman WS, Donahoo JS. Effect of phentolamine in controlling temperature and acidosis associated with cardiopulmonary bypass. Critical Care Medicine 1985; 13: 72-76. Patel PM, Mutch WAC. The cerebral pressure-flow relationship during 1.0 MAC isoflurane anesthesia in the rabbit: the effect of different vasopressors. Anesthesiology 1990; 72: 118-124. Michenfelder JD, Milde JH. The interaction of sodium nitroprusside, hypotension, and isoflurane in determining cerebral vasculature effects. Anesthesiology 1988; 69:870-875.
Downloaded from http://bja.oxfordjournals.org/ at Mount Royal University on June 10, 2015
8. Becker H, Vinten-Johansen J, Buckberg GD, Robertson JM, Leaf JD, Lazar HL, Manganaro AJ. Myocardial damage caused by keeping pH 7.40 during deep hypothermia. Journal of Thoracic and Cardiovascular Surgery 1981; 82: 810-820. 9. Murkin JM, Farrar JK, Tweed WA, McKenzie FN, Guiraudon GM, Guiraudon G. Cerebral autoregulation and flow/metabolism coupling during hypothermic cardiopulmonary bypass: the influence of P a ^ . Anesthesia and Analgesia 1987; 66: 825-832. 10. Henriksen L. Brain luxury perfusion during cardiopulmonary bypass in humans. A study of the cerebral blood flow response to changes in CO,, Ov and blood pressure. Journal of Cerebral Blood Flow and Metabolism 1986; 6: 366-378. 11. Bashein G, Townes BD, Nessly ML, Bledsoe SW, Horabein TF, Davis KB, Goldstein DE, Coppel DB. A randomized study of carbon dioxide management during hypothermic cardiopulmonary bypass. Anesthesiology 1990; 72: 7-15. 12. Prough DS, Stump DA, Todd Troost B. P a ^ management during cardiopulmonary bypass; intriguing physiologic rationale, convincing clinical data, evolving hypothesis. Anesthesiology 1990; 72: 3-6. 13. Smith AL, Wollman H. Cerebral bloodflowand metabolism: effects of anesthetic drugs and techniques. Anesthesiology 1972; 36: 378-^100. 14. Stephan H, Sonntag H, Lange H, Rieke H. Cerebral effects of anaesthesia and hypothermia. Anaesthesia 1989; 44: 310-316. 15. Prough DS, Stump DA, Roy RC, Gravlee GP, Williams T, Mills SA, Hinshelwood L, Howard G. Response of cerebral blood flow to changes in carbon dioxide tension during hypothermic cardiopulmonary bypass. Anesthesiology 1986; 64: 576-581. 16. Govier AV, Reves JG, McKay RD, Karp RB, Zorn GL, Morawetz RB, Smith LR, Adams M, Freeman AM. Factors and their influence on regional cerebral blood flow during nonpulsatile cardiopulmonary bypass. Annals of Thoracic Surgery 1984;' 38: 592-600. 17. Soma Y, Hirotani T, Yozu R, Onoguchi K, Misumi T, Kawada K, Inoue T, Mohri H. A clinical study of cerebral circulation during extracorporeal circulation. Journal of Thoracic and Cardiovascular Surgery 1989; 97: 187-193. 18. Hindman BJ, Funatsu N, Harrington J, Cutkomp J, Miller T, Todd MM, Tinker JH. Differences in cerebral blood flow
57